University of Southern California Dissertations and Theses

Conflict mediation: a comparative study of the effects that mediators have on dyadic conflicts

(USC Thesis Other)

Conflict mediation: a comparative study of the effects that mediators have on dyadic conflicts

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content CONFLICT MEDIATION: A COMPARATIVE STUDY OF THE EFFECTS THAT MEDIATORS HAVE ON DYADIC CONFLICTS by Yamit Gavrieli A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (POLITICS AND INTERNATIONAL RELATIONS) May 2010 Copyright 2010 Yamit Gavrieli ii TABLE OF CONTENTS LIST OF TABLES iv LIST OF FIGURES v ABSTRACT vi ACRONYM GLOSSARY viii CHAPTER ONE: Introduction 1 Purpose 17 CHAPTER TWO: Literature Review 18 Requisites for Success 22 The Prisoner’s Dilemma 24 Motives 30 Acceptance 36 Timing 38 Leverage and Stalemate 41 CHAPTER THREE: Theory and Methods 45 Data Collection 45 Framework and Application 47 The Prisoner’s Dilemma 51 Case Studies 52 Purpose of the Study 53 Hypotheses 54 CHAPTER FOUR: Historical Context of the 1966-1998 Conflict in Northern Ireland 56 Parties to the Northern Ireland Conflict 63 Unionists 63 Nationalists 64 Timeline of the Conflict in Northern Ireland 67 CHAPTER FIVE: The Mediation Process in Northern Ireland 70 The IRA: Key Interventions and Major Actors 70 The Anglo-Irish Agreement of 1985 70 The Brooke-Mayhew Initiative 72 The Downing Street Declaration and the 1994 Cease-fires 74 An Opportunity for Conflict Resolution 77 The Military Stalemate and the Emergence of a Republican Political iii Strategy 78 The Clinton Administration as Mediator 79 Mechanisms of Intervention 83 The International Body of Decommissioning 84 The Report 93 International Involvement 98 Overview of the Agreement 106 International Support for the Peace Process 109 The Role of the European Union 114 CHAPTER SIX: Historical Context of the 1994 Conflict in Rwanda 118 Background 121 Indirect Rule through Tutsi 123 Dictatorship and Exploitation of Ethnicity 127 Civil War and Multiparty Politics 132 Timeline of the Conflict in Rwanda 139 CHAPTER SEVEN: The Mediation Process in Rwanda 142 Aims of the Arusha Negotiations 154 RPF: Armed and Determined 155 GoR: 156 Internal Opposition to GoR (Habyarimana) 158 New Mediation in Arusha, Tanzania 159 Tanzania: The Arusha Facilitator 161 France: Playing Both Sides 164 United States: Reluctant Intervention 168 Actual Negotiations 169 Implementation 178 United Nations: A Vague Mandate 179 CHAPTER EIGHT: Applying Moore’s 12 Stages to the Two Cases 186 Analysis: Northern Ireland and the 12 Stages 190 Analysis: Rwanda and the 12 Stages 200 CHAPTER NINE: Applying the Prisoner’s Dilemma to the Two Cases 216 The Prisoner’s Dilemma and the IRA: First Attempt 218 Second Attempt 220 The Prisoner’s Dilemma and Rwanda 228 CHAPTER TEN: Conclusion 244 BIBLIOGRAPHY 263 iv LIST OF TABLES Table 1.1 - The Prisoner’s Dilemma Game in “Win-Lose” Terminology 30 Table 1.2 - Moore’s 12 Stages to Mediation 188 Table 1.3 – Breakdown of Conflict Resolution in Northern Ireland 228 Table 1.4 – Breakdown of Conflict Resolution in Rwanda 241 v LIST OF FIGURES Figure 1.1 - Map of Ireland 55 Figure 1.2 – Map of Rwanda 117 vi ABSTRACT The nature of global conflict has transformed dramatically in the past few decades thanks to changes in international relations. Many attempts at successful conflict mediation have failed due to the dramatic scope of the situation; still, however, some of the successes in mediation reveal the possibility of finding common ground between two fiercely opposed sides. The purpose of this study is to investigate the importance of leverage, in the form of power, tools and resources, and interests, of the mediator, as individual or governmental entity, in influencing the outcomes of the mediation process in international conflicts. The Prisoner’s Dilemma will be used as a theoretical tool, while Christopher Moore’s 12-step framework will be used as an organizing principle to better understand the mediations. This study comparatively analyzes two cases using two methods: Christopher Moore’s twelve stages to mediation and The Prisoner’s Dilemma. The cost-benefit analysis will be based on a power-politics analysis of both the mediators’ and parties’ interests as a key to understanding their decisions and motivations. The first case study examines the conflict between the IRA and Britain; the second looks at the civil conflict between the Tutsi and Hutu of Rwanda. The decision to study these two cases was based on the need to have one successful and one unsuccessful mediation case, and to understand the reasoning behind why one failed and one vii succeeded the respective success and failure. 1 The findings suggest that even the most skilled individual mediators can only achieve success when they have leverage. Based upon conclusions from an extensive literature review of this field, I chose to analyze two different hypotheses, they include: (1) that mediation is unlikely to be successful unless the mediator has some form of leverage (i.e., power, interests, resources), (2) finding the right measure of interest in dispute resolution is central to a mediator’s success. Both of these hypotheses will be proven true for the mediation in Northern Ireland and the mediation in Rwanda. By undertaking a critical analysis of these hypotheses, I seek to gain insight into conflict mediation and to offer a productive understanding of the mediator’s role in negotiations. 1 This is also a common perception by the international community. viii ACRONYM GLOSSARY BBTG Broad Based Transitional Government CDR Coalition of the Defense of the Republic CEPL Communauté Economique de Pays des Grand Lacs DUP Democratic Unionist Party EU European Union FAR Force Armees Rwandaises GoR Governor of Rwanda HC1 Low Cost Armament Strategy HC2 High Cost Armament Strategy IBD International Body of Decommissioning INLA Irish National Liberation Army IRA Irish Republican Army JFD Joint Framework Document JPMC Joint Political Military Committee LC1 Low Cost Armament Strategy LC2 High Cost Armament Strategy MDR Mouvement Démocratique Républicain MRND Mouvement Revolutionnare National pour le Development NGO Non-governmental Organizations NMOG Neutral Military Group ix OAU Organization of African Unity PL Liberal Party PSD Social Democratic Party RGF Rwanda Government Forces RPD Regional Program Director RPF Rwandese Patriotic Front RSF Republican Sinn Fein RTLMC Radio Television Libre des Mille Collines SDLP Social Democratic and Labor Party UDA Ulster Defense Association UDP Ulster Democratic Party UFF Ulster Freedom Fighters UN United Nations UNAMIR United Nations Assistance Mission in Rwanda UUP Ulster Unionist Party 1 CHAPTER ONE: INTRODUCTION The end of the Cold War and the changes that have taken place in international relations over the past two decades have fundamentally altered the character and occurrence of conflict. Most pronounced has been a reduction in the involvement of the United States and Russia in many of the conflicts. The end of Cold War superpower polarity has not, however, removed the causes of global conflict or affected their intensity and the need to deal with them effectively. Indeed, many would argue that the new international system is as conflict-prone, if not more so, as the previous one. Global relations in the first decade of the twenty-first century have been fraught with ethnic, religious, territorial, and nationalist conflicts–-as serious, costly, and intense as any in the past—with high-stakes and an exigent need for careful management and resolution. Throughout history, individuals, groups, communities, and more recently, states, have endeavored to avoid violence by seeking methods of dealing with conflict in constructive, peaceful, and mutually satisfying ways. Though there may be no definitive answers, scholars such as William Ury (1988), Stephen Goldberg (1995 and 2003), Howard Raiffa (1985), Jacob Bercovitch (1992, 1996 and 2003), Martin Shapiro (1986), William Zartman (1985, 1991, 1992, 2000 and 2008) R. Hrair Dekmejian (1980 and 2007), and others from disciplines as diverse as anthropology, psychology, political science, sociology, law, and communication have attempted to further our understanding of how the “art” of mediation works and under 2 what conditions mediation can be most effective. 2 However, their collective efforts have also been bedeviled by the immense scope of mediation, the secrecy that often enshrouds it, and the inherent challenges to the objective study of mediation, especially at an international level. Mediation has a rich and varied history. References to it abound in the Bible, Homer’s Iliad, and Sophocles’s Ajax. Its application is found in arenas as diverse as community, marital, and professional disputes, and in organizational, environmental, and policy conflicts. In the present international system, where recourse to weapons makes the violent pursuit of conflict both costly and mutually destructive and where there is no adherence to a generally accepted set of rules or to a central authority regulating international behavior, mediation offers an invaluable way of dealing with differences and settling conflicts between antagonistic and fiercely independent states. My approach to mediation recognizes its practice in numerous arenas; that it is, in many ways, a continuation of the parties’ own conflict-management efforts and that it involves the non-coercive intervention of a third party. “Mediator” may refer to an individual, an ad hoc group, an organization, a sponsor, or a state seeking to influence or resolve the conflict; in my case studies, “mediator” in Northern Ireland refers specifically to George Mitchell (and more generally to the support of the United States and President Bill Clinton) and “mediator” in Rwanda refers to the Tanzanian mediators, most particularly the Tanzanian diplomat Ami Mpungwe. 2 Howard Raiffa (1985) refers to negotiation as an “art.” 3 Mediators seek to fulfill their primary objectives through persuasion, utilization of leverage, appeals to logic, the use of information, and the application of social-influence strategies. As such, the mediators’ objective to change, reduce, or resolve a conflict legitimizes their intervention, while the material, political, or other resources they invest in the process may compel the disputants to compromise on their own objectives and demands. As Martin Shapiro cautions in his study of the courts, “many judges are not entirely ‘neutral’ third parties, but instead bring to the triad distinct policy preferences, which they seek to implement through their decisions” (Shapiro, 1986: 29). While a judge must present findings in favor of one side or another, a mediator works to bring the two parties to a compromise; thus, in method and objective the work of the judge and the mediator differs substantially. When, however, the mediator has considerable power based on the backing of a powerful state, he or she may act more like a judge. Because the balance of power is at the heart of so many conflicts, the mediator must enter into the relationship with a keen sense of how he or she will affect the equilibrium between stakeholders. The various perceptions of the mediator can influence the parties’ willingness to enter into dialogue with a third party. Disputants’ sense of their own influence relative to each other and to the mediator factor into the equation as well. When a mediator enters, compromise immediately becomes a likelihood, meaning that “victory” as imagined and fought for yields options that either or both parties are likely to regard unfavorably. A third party entering into the scenario will inevitably affect the power symmetry, creating ample opportunity for resistance on either side. According to 4 Richmond (1998) there must be clarity about the mediator’s role from the beginning. He explains, An initial starting point must be to identify what the mediator’s role and objective is perceived as being by the disputants—facilitator, diplomat, manager, ally, agent of empowerment or legitimation—and if success is equated with a compromise solution. (p. 709) The reputation of the mediator is paramount, particularly as it pertains to past successes and failures. The disputants must have faith in the mediator as well as a belief in his or her objectivity. The mediator’s assets and resources (Richmond, 1998) also factor in to how the parties will regard him, and can even serve as a greater incentive than reaching a compromise solution. The leverage tied up in a mediator’s associations with a nation such as the United States can go far in legitimating his role and can lead to different expectations and responses (Bercovitch, 1996). It is uncommon that a mediator is not backed by a state or nongovernmental organization, each of which carries different implications and triggers different responses. Richmond cites the position of UN-backed mediators who have particular constraints, concerns, and obligations and bring with them a high degree of moral authority, legitimization, and recognition (1998). But as shown in the case of Rwanda, there was a lack of power to enforce the results. Entering into a situation already fraught with conflict, misunderstanding, and dueling interests, the mediator can often himself become the object of contempt, as displaced frustration and resentment between the two sides can be projected onto the newly entered third party. Particularly when negotiations are not successful, blame can 5 fall on the mediator himself. Where conflict was once directed between two entities, the mediator now must diffuse and deflect tensions along the way to resolving the issues. To be sure, the model of the triangular relationship is much more complicated than two opposing parties with one neutral go-between. Not all mediation scenarios fit what Zartman calls “the classic model’s triad of government, opposition, and host- neighbor” (1995: 27), as the conflict can be intragovermental as well as intergovernmental. The dynamic can include more than two disputants, or there may be a situation, as in Rwanda, where the opposing parties within themselves are at odds and have not yet reached a consensus position to bring to the dyadic discussion. In such instances, the triangular model fails to reflect the fragmentation and dissent within the opposing sides themselves. Thus if what Richmond asserts is true, that “all conflict is basically asymmetric” (p. 709), the mediator must negotiate with the skewed dynamics so that while there will certainly be compromise on both sides, there will also be the perception by both sides that the resolution can be “win/win.” 3 This study undertakes a comparative analysis of the mediator’s role in conflict mediation by applying the Prisoner’s Dilemma to conflict resolution in two case studies. Specifically, I will explore the mediator’s role in the relationship, and how it works, or fails to work, in offering a settlement that satisfies both parties involved in the conflict. This analysis corresponds to previous research in assuming that mediators, as well as the other actors, are motivated by self-interest of varying degrees and natures. I posit that the mediator’s role, with its range of actions, incentives, and leverage, directly affects the 3 This expression will be explained in the following chapter. 6 mediation process and that by learning more about that role, and its leverage or lack thereof, we may gain insight into the larger dynamics at play in mediation itself. The approach is based on cost-benefit and power politics analysis, which understands the mediators’ and the parties’ interests as a key to motives, roles, and leverage. Since mediators have a vested interest in finding a resolution, I will apply a modified Prisoner’s Dilemma assessment to investigate what elements are necessary for a successful mediation outcome. By removing mediation from the realm of idealism, and by bridging the artificial distinction between the disinterested facilitator and the interested manipulator, this analysis will seek to illuminate the ways that mediators operate, depending upon the resources at their disposal. As a theoretical tool, the Prisoner’s Dilemma helps to assess the risk management of negotiation scenarios as the mediator seeks equilibrium by exchanging information and offering incentives that serve the interests of both parties. This research is intended to contribute to the scholarship on mediation through a detailed examination of two case studies that apply game theory as a productive logic for dealing with dyadic conflicts. The case studies are the conflict between the IRA and Britain beginning in 1966, 4 and the internal conflict between the exiled Tutsi guerilla group (the Rwandese Patriotic Front) and the Hutu government of President Juvenal Habyarimana. Selection of these two cases was based on the important role mediation played in them, extending over a long period of time and thus allowing for the consideration of many remarkable and complex issues. Importantly, these two cases 4 The chapter on the historical background of the Northern Ireland conflict will actually trace back to 1688. 7 complement one another, together covering many different situations that arise during mediation, with similarities and differences that produce a comprehensive study with the potential to yield significant findings. The decision to use two cases was based on the need to have one successful and one unsuccessful case to compare. The successful conflict resolution case is that of the IRA and Britain and the unsuccessful conflict resolution case is the conflict between the Tutsi and Hutu in Rwanda. To be sure, this valuation begs the question—how exactly is success in mediation measured? Existing literature does not provide conclusive answers. Analysts of mediation often avoid altogether employing the terms success and failure around mediation (Assefa, 1987). Others still craft their own definitions, such as mutual acceptance of mediation in the first place (Frei, 1976) or achieving an end to violence. “By successful outcomes we mean producing a cease-fire, a partial settlement or a full settlement” (Bercovitch, Anagnoson, and Wille, 1991: 8). Measures of success can also be goal-based—using the objectives of either the parties or the mediator as a tangible indicator (Kleiboer, 1996; Smith, 1985; Touval and Zartman, 1985). Kleiboer sums up the challenges of measuring success by asserting: At present, we seem to lack a widely agreed upon Archimedean point for evaluating attempts at international mediation. Success and failure are construed rather than discovered by the analyst: they are a matter of idiosyncratic values, interpretations, and labeling. (1996, 362) Knowing that definitions of success and failure are not hard and fast, I nonetheless apply these terms in the present research. For purposes of my argument, I categorize Ireland as a success and Rwanda as a failure based upon how events unfolded 8 following the culmination of talks, in the case of Ireland with The Good Friday Agreement, signed on April 10, 1998 and, in the case of Rwanda, with the Arusha Accords, signed on August 4, 1993. In doing so, I seek not to dismiss the complexity of the respective situations, but rather to show their effectiveness as case studies that can impart valuable lessons about the mediator’s role in the peace process. By all critical accounts Rwanda’s negotiated settlement is not considered a success. Tutsi rebels and the Rwandan government agreed to follow the lead of highly regarded Tanzanian mediators, ultimately signing a peace settlement that afforded them both representation in the military and legislature, with a UN “guarantee” of security for the two sides. Despite these favorable elements, the Arusha accords were followed by genocide, an example of failed mediation, to be sure. The 1993 Arusha Peace Accords, which ostensibly ended the ongoing conflict in Rwanda between the exiled Tutsi guerilla group (the Rwandese Patriotic Front) and the Hutu government of President Juvenal Habyarimana, serve as an example of a negotiated agreement failing miserably in its implementation. Almost immediately following the signing of the Arusha Accords, Hutu extremists who felt threatened by the results organized a horrendous massacre of Tutsi and Hutu moderates. Between 800,000 and one million people died between April and July 1994 (Greenberg and Barton, 2000: 222). Prior to the genocide, “the vast majority of Hutu and Tutsi lived side-by-side, dispersed throughout the country” (Greenberg and Barton, 2000: 222). Menaced by the high poverty rate and history of tensions, “a small band of Hutu elites killed Tutsi and Hutu 9 peasants in a desperate bid to preserve their power and economic privilege” (Greenberg and Barton, 2000: 222). Questions about the quality of the mediation remain—how effective could mediation be in the face of a centuries-old internecine dispute between Tutsis, emboldened by privileged treatment under German and Belgian colonialists, and the majority Hutus, the dominant force in the government at that time? Innumerable efforts to explain the failure cite the defects of the third-party peacemaking efforts (Jones, 1995b), the treaty terms, government concessions, implementation of the accords, coherent post- treaty strategies, follow-though and response time when genocide first reared its ugly head. Those involved in the peacemaking—in particular the United Nations-- made failure inevitable when they neither included nor planned to control and contain potential “spoilers.” The oppression of the Irish Catholics by the British originally led to the creation of the IRA, which wanted Irish independence from Britain and equal civil rights. In their drive to achieve territorial sovereignty and self-determination there was very little that the IRA wouldn’t do. After the “Bloody Sunday” attacks of 1972, in which the British used tanks to attack Irish Catholic protestors marching for equal rights, the IRA converted into a terrorist organization. After many deadly attacks, which included the detonation of nail bombs and the murder of Earl Mountbatten (cousin to the Queen of England), it became clear that the civil war had escalated to such a point that outside help was necessary. Tony Blair’s back door diplomacy with the IRA combined with Bill Clinton’s assistance, led to creation and approval of the Good Friday Agreement of 1998. The resolution 10 between the IRA and Britain clearly illustrates successful mediation that led to peace that endures to the present day. Many analysts assert that President Clinton’s entrance into the Northern Ireland conflict turned the tide that led to the favorable conclusions. His arrival served as a changed condition to the arduous and unsuccessful efforts to negotiate a treaty between the parties. From 1970-1995 numerous attempts to resolve the conflict had ultimately been ineffective because the parties did not feel their needs were being met and/or certain parties were excluded from or refused to participate in the talks. That Clinton made the conflict a foreign policy priority, came from the United States, with its attendant resources, influence, assets, and had a general reputation as a trustworthy individual boded well for negotiations as both parties knew they would receive just treatment. The Clinton Administration augmented the circumstances by creating the economic conference, traveling to Belfast, granting Gerry Adams two visas to travel to the United States, sponsoring trade missions, funding the International Relief of Ireland, and most importantly appointing George Mitchell to mediate a peace initiative in Northern Ireland. Retrospectively considering how the peace process concluded, United States mediator George Mitchell is generally recognized as having “achieved a substantial measure of success” in his efforts to broker peace in Northern Ireland (Curran, Sebenius, and Watkins, 2004: 513). Mitchell brought with him 14 years of Senate experience (1989-1995 as Democratic Majority Leader) and a reputation for patience, fairness, and meticulousness in his work. Mitchell’s characteristic persistence helped move the 11 decommissioning 5 effort ahead. His objective was clear: to end the conflict by facilitating dialogue between parties whose mutual dislike and distrust had long impeded effective communication. He served as a neutral mediator who unrelentingly worked for almost two years on a conflict that ranked relatively low among the U.S.’s foreign policy threats. Though Mitchell had “no formal power and only had a few carrots and sticks with which to influence negotiators and their constituents” (Curran, Sebenius, and Watkins, 2004: 513), he eventually forged a mutually satisfying peace agreement. As detailed below, his skills as a mediator contributed significantly to creating conditions of success for the negotiations. Though pronouncements about success and failure of peacekeeping missions can never be unequivocal, it is safe to assert that the conflict in Northern Ireland fundamentally changed as a result of Mitchell’s efforts. To exemplify: the aftermath and response to the recent killings of two soldiers and a policeman in March 2009 (with responsibility claimed by two IRA splinter groups—“Continuity IRA” and “The Real IRA” 6 ) confirm that although peace is precarious, attitudes toward violence have changed significantly. According to John Darby, an expert on the Irish peace process, the response of respective party leaders clearly demonstrates “unprecedented” progress, as, “Northern Ireland’s First Minister, Peter Robinson, and Martin McGuinness, his Sinn Féin Deputy First Minister, condemned the killings unequivocally.” He continues, “Such unanimity is rare in Northern Ireland. All the other parties have presented a united front….The loyalist 5 The process of removing weapons from use by disposing of them or putting them on standby. 6 The Real IRA was a breakaway group from the main Republican Party (Dingley, 2008: 240). 12 Protestant parties, which had conducted a violent struggle with the IRA, praised McGuinness’s stand.” 7 Close examination of these cases can increase our knowledge about specific instances of conflict and the mediation processes that took place. It is hoped that the concepts and case studies presented here will allow for the formulation of better theories and applications of mediation and will contribute to a more effective response to future conflicts in international relations. The following chapter presents a review of pertinent theoretical work that addresses conflict mediation and game theory. A thorough analysis of the available literature will provide a comprehensive foundation for a comparative study of the two cases. In his effort “to move toward a more general theory of the nature of judicial institutions” (vii), political scientist Martin Shapiro provides an overview of the basic function of courts, which, among other things, serve as arenas for conflict resolution. Shapiro characterizes the move to seek a third party for conflict resolution as transcultural common sense. He explains: “Cutting quite across cultural lines, it appears that whenever two persons come into a conflict that they cannot themselves solve, one solution appealing to common sense is to call upon a third for assistance in achieving a resolution” (Shapiro, 1986: 1). Importantly, according to Shapiro, conflict resolution is effective when the two disputing parties recognize the authority of the mediator’s role. The parties’ explicit acceptance of the decision maker’s role must last until the dispute is resolved so that they see the end-result as an extension of their initial agreement. 7 (http://kroc.nd.edu/, Accessed 10/1/09) 13 Moreover, the disputants’ ongoing support of the mediator prevents the triad from disintegrating into a scenario in which the loser perceives two-against-one favoritism by the judge and “winner” of the case causing legitimacy effectively to vanish. The disputing parties’ affirmation of the decision maker’s role and their commitment to being present at the talks signifies mediation’s fundamental difference from court (and thus the mediator from the judge). Judges must favor one side, often functioning with a zero-sum logic, whereas mediators must bring both sides to agreement and find a happy medium. Because of its unavoidable judgment in favor of one disputant, courts by nature destabilize “the logic of the triad.” Shapiro insists, “The most fundamental device for maintaining the triad is consent” (Shapiro, 1986: 2). By way of example, he invokes the procedure in Early Roman law in which disputants not only chose the “norm” of the dispute (without which, the case would not continue) but they also agreed upon a judge. As such, both parties have a vested interest in the procedures and know that when the judge finally issues his decision, their consent is seen as an integral part of that process. As Shapiro explains it, agreeing to the norm and the judge means implicitly consenting to the final judgment (this is very similar to mediation today). The disputing parties are not the only ones with a vested interest in mediation, however. By virtue of being between two entities, the mediator must relay and interpret information for the respective parties and, in doing so, implicitly exerts influence. Though he may propose solutions, he may not require or enforce them. Mediation works when the integrity of the “logic of the triad” is upheld and the consent of the parties 14 persists. Parties in the both the cases of Rwanda and of Ireland at times perceived partiality in the mediation process. The difference between the two cases is that with few exceptions the disputants in Northern Ireland understood the importance of their participation and respected the methods and processes of the mediation. Deep divisions plagued the Rwandan mediations and the necessary commitment to the “logic of the triad” simply could not be upheld. As Shapiro’s work suggests, then, the individual mediator’s own intentions play a significant role in the ultimate effectiveness of the mission. Various motivations can come into play, “ranging from a genuine impulse toward peace-making, to pressure from his constituency and personal ambition” (Richmond, 1998: 713). Further, an individual mediator’s intentions may not be one and the same with the nation or organization he or she represents; though degrees of difference are acceptable, these interests certainly cannot be at odds. A close look at how mediator self-interest was demonstrated in the cases of Ireland and Rwanda foretells a lot about the ultimate success of the mediation. While some scholars argue that absolute neutrality is essential to eliciting trust from the disputing parties, still others cite the importance that close alliances can have, believing that if a mediator is a friend to one side, he will have a greater chance of convincing that party of his good intentions and be able to apply pressure that will in fact work to the opposition’s advantage. According to Touval and Zartman, the additional motivation of an interested mediator can actually be a benefit to both parties as they may be more 15 willing to commit resources and maximize leverage to bring about results (Touval and Zartman, 1985). The relationship between mediator self-interest and mediator neutrality is a complex one. That a mediating entity has self-interest does not mean that it cannot be neutral in approaching the negotiations. Self-interest (political, territorial, and diplomatic) does not necessarily mean there will be bias in the arbitration, so that where fairness is critical—and indeed the perception as such by the disputants is indispensable—mediator self-interest does not inevitably contaminate the negotiating circumstances. What emerges from a comparative study of the Rwanda and Ireland cases is that while the mediator’s self-interest is indeed important, too much or too little can derail the process. President Clinton’s self-interest in seeking to negotiate peace was productively complemented by “George Mitchell’s selflessness” (Mayer, 2004: 92). As Ed Moloney, an Irish journalist, explains, Gerry Adams’s allies in the United States had tirelessly “wooed” President Clinton, so that when seeking the Democratic nomination for the upcoming presidential election, Clinton promised that, if re-elected, he would grant Gerry Adams a visa to visit the United States. 8 Clinton also astutely realized that there were votes, Irish-American votes, to be won, and at a relatively low cost (Moloney, 2003: 420). Though by no means exclusively self-interested, Clinton had a steadfast determination to aid the IRA case, which may in part be attributed to its desire to craft his image as a world leader and at the same time to leave behind a legacy. 8 At the time, Gerry Adam was the leader of Sinn Fein’s political wing. 16 Presenting something of an extreme example, Mitchell started as “economic envoy,” but his unabashed affection for the Irish people and genuine desire to bring an end to “the troubles” ultimately transformed him into a successful de facto peace envoy. Rwanda, by extreme contrast, had too many parties to the negotiation without any one caring enough to see the peacekeeping through to the bitter end. Where Mitchell was a strong central figure who focused on sustaining the momentum and achieving the objectives of decommissioning, the many mediators in Rwanda failed to create, implement, or establish guidelines, and the peacekeeping entities—the United Nations and international intervention—were just not invested or interested in a long-term, unwavering commitment to the process. Moreover, the role of France in the Rwandan crisis, the ambivalent authority of its troops, and its questionable alliances with the Hutu- led government presents anything but the stable and impartial international leadership seen with the United States’ role in Northern Ireland. Moreover, the post-negotiation monitoring and enforcement failed egregiously and virtually assured the crisis would continue, resulting ultimately in genocide. Thus, the two hypotheses are: 1. Mediation is unlikely to be successful unless the mediator has some form of leverage/power. 2. Finding the right measure of interest in dispute resolution is central to the mediator’s success. 17 Purpose The purpose of this research is to investigate the importance of leverage, in the form of power, tools and resources, and interests, on the mediator, as individual or entity, in influencing the outcomes of mediation process in international conflicts. To this end, I engage the Prisoner’s Dilemma as a theoretical tool to provide analysis that may prove useful to future mediators and scholars of mediation. I also use Christopher Moore’s 12 steps of mediation-- often used as a training device-- as an organizing principle to provide clarity to the often complicated mediation processes in these cases. 18 CHAPTER TWO: LITERATURE REVIEW The term “mediation” is popularly employed in international relations to describe almost any third-party activity short of armed intervention. But its broad application also means that its merits are not clearly understood. Rather than serving to define mediation, descriptors such as “neutrality,” “voluntarily,” “concessions,” and “separately,” represent a particular quality of the intervention. Mediation is sometimes just as broadly defined by practitioners and researchers as by the general public, as in these two definitions by researchers: “Mediation is the assistance of a neutral third party to a negotiation” (Bingham, 1986: 5), or “third party assistance to two or more disputing parties who are trying to reach agreements” (Pruitt and Kressel, 1985: 1). Some practitioners are more specific: “Mediation involves the intervention of a third party who first investigates and defines the problem and then usually approaches each group separately with recommendations designed to provide a mutually acceptable solution” (Blake and Mouton, 1985: 15). The definition of mediation chosen for this research is a form of third-party intervention into a conflict, with the stated purpose of contributing to its abatement or resolution, through negotiation. 9 Like other forms of peacemaking or conflict resolution, mediation is an intervention acceptable to the adversaries, who cooperate diplomatically with the intervener (Bolduc, 1989: 43). 10 Mediation differs from other forms of third 9 This is the definition that will be used for this study. 10 This will prove to be true in the case of Rwanda and Northern Ireland. 19 party intervention in two ways: it is not based on the direct use of force, and it is not aimed at helping only one of the participants “win” (Greenberg, 2008: 20). Mediators attempt to help the adversaries communicate and, as in conciliation, emphasize changing the parties’ images and attitudes toward one another. In addition, mediators suggest ideas for a compromise, and they negotiate and bargain directly with the adversaries. According to Wallenstein, mediation is not arbitration; arbitration uses judicial procedure and issues a verdict that the parties are committed to accepting; mediation is basically a political process, without advance commitment of the parties to accept the mediator’s ideas (Wallenstein, 2007: 79). A number of researchers have tried to make sense of the wide range of mediation processes by categorizing mediator(s) roles. Dealing with different circumstances for each case, the mediator must feel his or her way through the obstacles posed by personalities, issues, and political and economic limitations. Some researchers attempt to understand mediation by the kinds of strategies and tactics used. Zartman and Touval (1985) argue that mediators adopt one of three roles: communicator, formulator, or manipulator. Pruitt (1981) distinguishes between mediators who emphasize the importance of the process and those who focus on content. Bolduc (1989) suggests four mediator styles: “fair treatment,” “mutual self interest,” “just-outcome” and “just relationship.” In his analysis of mediation, Wall (1981) provides an exhaustive survey of various mediator strategies, techniques, and functions. These procedures include not only such commonly accepted tactics as identifying the real issues and summarizing the 20 agreement, but also much more controversial ones such as exploiting the weaker side, obfuscating the negotiator’s position, publicly criticizing the negotiators for being too tough, and misrepresenting and distorting information. A number of scholars have divided the mediation process into stages. Keltner (1987) suggests seven phases of mediation: setting the stage; opening and development; exploration of the issues; identification of alternatives; evaluation, negotiation, and bargaining; decision making and testing; and terminating the process. Paul Milgrom (1991) argues that for any form of cooperative behavior to evolve into the dominant behavior and persist in that role, it must meet three requirements. First, it must satisfy initial viability, that is, it must be able to gain a foothold in a predominantly non- cooperative environment. Second, it must display robustness, the ability to thrive and grow in an environment in competition with a varied mix of alternative strategies. Finally, once fully established, it must have stability, the ability to fend off invasion by other forms of behavior. Among the most successful efforts to present a practical guide to successful mediation is found in The Mediation Process: Practical Strategies for Resolving Conflict (1986) by Christopher W. Moore, a partner in the Denver Center of Dispute Resolution. Moore formulated his theory based on the history and range of effective mediation techniques across various disciplines, having researched the mediator’s tasks prior to joint negotiations, including mediator entry, data collection techniques, and designing a mediation plan. Trained as a mediator by the US Federal Mediation and Conciliation Service, Moore has focused on facilitating or mediating issues in which preserving 21 relationships is critical for reaching and implementing agreements. Moore reviewed the history of mediation and its contemporary practice and then compiled a thorough list of 12 stages that he believed to be crucial to successful mediation. In his book, Moore (1987) proposes the following 12 stages: 1. Making initial contacts with the disputing parties 2. Selecting a strategy to guide mediation 3. Collecting and analyzing background information 4. Designing a detailed plan for mediation 5. Building trust and cooperation 6. Beginning the mediation session 7. Defining issues and setting the agenda 8. Uncovering hidden interests of the disputing parties 9. Generating options for settlement 10. Assessing option for settlement 11. Reaching final bargaining 12. Achieving formal settlement Moore explains that if a mediation process successfully abides by all of the 12 steps, mediation will be successful. 11 Similarly, mediation that cannot successfully complete all 12 steps, Moore explains, may therefore be considered unsuccessful. 12 As an organizing principle, his 12-steps offer a superlative means of anatomizing mediation 11 The steps can be followed in any order, but the order Moore suggests in his framework depicts the logical progression of the steps. 12 This will be shown when applied to the Rwandan mediation process. 22 in order to pinpoint where the process went awry. Moreover, he explains that it is possible to study failed mediations of the past by applying his framework to see where the mediation went wrong. This framework will prove to be very useful when analyzing the mediations in Rwanda and Northern Ireland. Requisites for Success What constitutes successful mediation is, as stated earlier, disputable. If we take success to mean the final resolution of all conflict and the reconciliation of the parties, then there would be very few successful mediations. Since one of the motives for mediation is the protection or advancement of the mediator’s interests, a successful exercise will result in an agreement that accords with the mediator’s interests. Its terms should not provide opportunities for rival powers to gain influence, but rather should augment the mediator’s own authority. There is wide agreement that the personal qualities of the mediator are relevant to the success of the undertaking (Mayer, 2004: 103). Intelligence, specific knowledge of and expertise in the problems at hand, imagination, persuasiveness, perseverance, commitment, tact, and drafting skills are some of the qualities that contribute to success. Bercovitch and Rubin explain that the rank of the mediator(s) and of his or her interlocutors is also of consequence. A mid-level official negotiating with officials of similar rank may succeed only on issues of secondary or technical importance. The higher the rank of the mediator(s), the more influential he is likely to be (Bercovitch and Rubin, 1992: 89). Presumably, “a high-level official is better able to commit his country 23 (or organization) and its resources to the mediation effort, and will therefore be more persuasive than a mid-level official whose ability to affect such a commitment is uncertain” (Bercovitch and Rubin, 1992: 89). The case study of Ireland clearly demonstrates the difference that a mediator’s position and personality can make as well as the importance that the sides share this positive regard. Not only did Mitchell have his years of success as a United States senator to back him, but he also had a productive and amicable relationship with the Irish people—one he had developed during his extended tenure in the country (Jacobsen, 2003). Still, Mitchell’s case is unique because he had an enormous superpower and an ambitious and determined president sponsoring his efforts, while he himself was a retired senator who had nothing direct or material to gain from advancing the peace efforts. The mediator is obligated both to talk and to listen, as so often stalemate and conflict are based upon the collapse of dialogue between the two parties. The breakdown of communication may not only hinder the delivery of messages, but also keep the parties from finding a solution to meet the needs of both sides. According to McKittrick, in such a situation, a formulator is needed, a third party more capable of innovative thinking than the parties to the conflict (McKittrick, 1996: 211). Redefining the issues in conflict or finding a formula for its resolution or management is key, with the parties frequently needing help to find a solution hidden in the morass of bad relationships or to generate a solution from the pieces of the conflict itself. McKittrick explains that the mediator as a formulator helps the two parties help themselves through tactful, sympathetic, accurate, straightforward prodding and suggestion (McKittrick, 1995: 218). 24 The Prisoner’s Dilemma The inclusion of a mediator in any arena turns a dyadic relationship into a triadic one, leading to, what can be thought of as, a “game” of Prisoner’s Dilemma. This transformation may be entirely planned or totally accidental. At its heart, however, is the mediator’s attempt to shape the behavior, perception, and choice of outcome of the adversaries conflict. Mediators enter a conflict thus becoming part of a unique system comprising two parties at odds with each other’s concerns and interests, who have a stake in a set of issues (that may or may not be clear to all), opposing attitudes, and positions that are at best incompatible and at worst downright antagonistic. According to Marieke Kleiboer, the challenge of evaluating a highly complex phenomenon such as international mediations, “begins with conceptualizing the phenomenon and developing standards for the assessment of mediation outcomes” (Kleiboer, 1998: 12). Whether as individuals, organizations, or states, mediators enter a conflict system to effectuate change and to promote or protect any interests they may have. In this sense, a mediator is not unlike another party in the conflict-management process, as his or her approach and performance are conditioned by context and circumstances as well as the behavior of the adversaries themselves. The peculiarities and distinctive features of mediation relate to its imprecise classification: it is neither a quasi-legal process nor an effort extraneous to the parties’ own conflict management efforts. To be successful, mediation must complement a given conflict in its context (i.e., it must align with the needs of the parties in conflict and work along with them to appease any fears or concerns). Kolb explains that to treat mediation 25 as either a legal process or a disinterested undertaking is to miss important attributes that explain the relationship between a mediator and the parties (Kolb, 1983: 48). Mediation begins with the interaction of two conflicting parties within a unique context. These parties’ reaction to their conflict and to the act of mediation is a result of their particular experience, society, culture, and structure. These features in turn affect how a mediator intervenes, what strategies and outcomes are pursued, and the mediation’s impact on the parties and the outcome. In game theory, as developed by Von Neumann (1928), the Prisoner’s Dilemma is a type of non-zero-sum game in which two players may “cooperate” or “defect” (i.e., betray) the other player. In this game, as in all game theory, the only concern of each individual player (“prisoner”) is to maximize self-interest, without concern for the other player’s payoff. In the classic form of this game, cooperating is dominated by defecting, so that the players will only reach equilibrium in the game if they defect (Simon, 1996: 37). In short, no matter what the other player does, one player will usually achieve a greater payoff by defecting. Since in any situation, as postulated by the theory, defecting is more beneficial than cooperating, all rational players will defect, all things being equal. 13 The unique equilibrium for this game is known as a Pareto-suboptimal solution – that is when rational choice leads both players to defect even though each player’s reward would be greater if they both cooperated-- hence the dilemma (Rubinstein, 1991: 42). 13 This is not necessarily true in an empirical situation where people must continue to interact over time. 26 The Prisoner’s Dilemma game has been widely studied by economists, psychologists, game theorists, sociologists, and other researchers. Rubenstein states that the main findings are: (a) the unique Nash equilibrium 14 of the game involves mutual defection, (b) cooperation can be supported as an equilibrium phenomenon if the game is repeated many times without too much discounting, or (c) if the game is repeated finitely many times but the players are uncertain about each other’s motivation or strategies (Rubinstein, 1991: 43). Rather than being guard over the prisoners, the mediator is in a position to help the disputing parties find the “win-win” solution. Indeed, the dynamic between mediator and disputants differs significantly from that between guard and prisoner. Though like the guard, the mediator represents a larger power, organization, or institution, the latter seeks to serve all interests in an impartial manner. To the extent that it is trusted by both sides, a neutral third party may be able to induce the parties to reveal information about their underlying interests, needs, priorities, and aspirations that they would not otherwise disclose to their adversary. This information can permit a trusted mediator to help the parties formulate the most efficient “trades” possible, or even to aid the parties in combining resources and opportunities not previously considered by either. Moreover, a skilled mediator can reduce the adversaries’ incentives, and hence their temptation, to engage in the various hardball tactics that waste resources and breed ill-will. Mediators can also do much to overcome or attenuate psychological barriers. Perhaps most importantly, they can create a problem-solving atmosphere, one that 14 A Nash equilibrium occurs when each player is assumed to know the equilibrium strategies of the other players, and no player can gain anything by unilaterally changing his or her strategy. 27 persuades the parties to move beyond political posturing and recriminations about past wrongs to recognize the gains from an efficient and fair resolution of the dispute and work to achieve them. Jacobsen asserts that in a sense, the mediator can turn the parties’ attention away from the direct pursuit of equity to the pursuit of enlightened self-interest. While divergent views of the past are inevitable, the mediator must endeavor to help each side understand the case from the other’s perspective (Jacobsen, 2003). The mediator can also endeavor to reframe the status quo, the impact of particular resolutions, and the costs of continued struggle in a manner that minimizes (or even takes advantage of) the participants’ aversion to accepting losses, especially losses that are definite. The phenomenon of reactive devaluation 15 sometimes can be avoided (and almost always can be reduced) if the settlement proposal in question derives, or appear to derive, from the third party or from some collaborative effort rather than from one of the principals acting unilaterally (Ross and Ward, 1996). Such blurring of authorship may occur when, after talking separately to each side about what is or is not acceptable, the mediator takes accountability for creating a proposal that he or she presents as a direct response to the principals’ own equally considered and valued concerns. Fisher and Ury assert that the mediator can also help the parties to understand why a particular concession is being offered, and why it is being offered at that time, lest attribution be made that would encourage devaluation rather than reciprocation of those concessions (Fisher and Ury, 1983: 40). Such explanations may have the effect of lessening some of 15 “Reactive devaluation” means that a party will view a proposal less positively if their opponents have proposed it rather than a third party, and often view alternative proposals not put forward more positively than those "on the table." 28 the discord associated with making concessions and accepting the many issues associated with ending a protracted struggle. When the antagonists are roughly matched in terms of resources, capabilities, and drive to maximize their self-interest, war (or the worst case scenario) can occur. Dekmejian and Doron explain that once they realize that they cannot arrive at their preferred outcome unilaterally, they may be led to a “satisficing” mode of behavior where they decide to decrease the cost or intensity of the conflict and arrive at equilibrium (Dekmejian and Doron, 1980: 42). This recognition is the first step towards conflict resolution. The second step towards conflict resolution occurs when a third party pursuing its own interests enters the scenario and imposes its own demands on the two parties in conflict. If conflict resolution is no longer a priority, one of the rivals may calculate that war will unilaterally advance its own interests, leading to disequilibrium in the mediation process. For the two principal parties engaged in conflict, any solution other than peace is a “non-Pareto” optimal point. According to Dekmejian and Doron, (1980) this means that. . . one situation, one social state, or one alternative is better than another if every individual (country) feels it is better according to its own values. Since a non- Pareto outcome is inherent in the Prisoner’s Dilemma game, it is possible to utilize its logic to clarify certain properties of the two case studies. (Dekmejian and Doron, 1980: 42) The inability of the two “prisoners” to communicate effectively is one of the key aspects of the Prisoner’s Dilemma game. Therefore panic about miscalculation of the opponent’s intent and its destructive consequences leads them to choose a non-Pareto alternative. 29 The second feature of the game, according to Dekmejian and Doron, is the existence of two alternative decisions that confront each “prisoner.” If territories are at stake, one player can maintain a low-cost armament strategy (LC1) or a high-cost strategy (HC1) and the second player can utilize a low-cost armament strategy (LC2) or a high-cost strategy (HC2). “The interest of the first player will be served when it maintains the low-cost armament strategy, if the second player follows suit. Consequently, they would both maintain their respective territorial interests and pursue other national interests not related to the conflict, if necessary” (Dekmejian and Doron, 1980: 43). However, one party could possibly capture its enemy’s territories if its rival follows a low-cost strategy. Simon explains that because both parties are aware of these possibilities, and because they cannot communicate their strategies to each other due to the low level of mutual trust, they are locked in a Prisoner’s Dilemma (Simon, 1996: 37). Therefore, if player 1 maintains the low-cost strategy, player 2 may maintain the high-cost strategy, thereby increasing its ability to capture its rival’s territory. If player 1 maintains a high-cost strategy (HC1), than player 2 should again retain a high-cost strategy (HC2), therefore securing its existing area. It therefore follows that regardless of the strategy followed by player 1, player 2 maximizes its utility by choosing a high-cost strategy. Once set-up has been established, player 1 must also retain its expensive strategy. Dekmejian and Doron explain that, an arms race could therefore emerge, and a costly equilibrium of terror (HC1, HC2) could prevail, although both would retain the same territories as they did in the LC1, LC2 equilibrium of peace (Dekmejian and Doron, 1980: 43). 30 Table 1.1 The Prisoner’s Dilemma Game in “win-lose” Terminology: Cooperate Defect Cooperate win-win lose a lot-win a lot Defect win a lot -lose a lot lose-lose Red = player 1; Blue = player 2 The imposition of a solution by an outsider becomes attractive as both parties may seek to return to the more pleasing low-cost equilibrium and a Pareto optimal point. Notably, however, a sufficient condition for a solution by the mediator is the ability to impose its will on the antagonists. The imposition should be perceived as leading to the LC1, LC2 position, the peace equilibrium. Therefore there must be a high level of trust in the mediator(s) by both parties in order for it to fulfill its role as mediator (Dekmejian and Doron, 1980: 44). Motives The intervention by mediators is usually legitimized by the aim to resolve conflict. Often, however, their desire to mediate involves other motives that play out in the larger context of power politics. According to Graham Allison (1999), a rational- actor approach is most useful for understanding these motives, as it employs cost-benefit considerations which reflect the reality that mediators are have interest in the conflict’s 31 outcome— otherwise they would not mediate . In view of the considerable investment of political, moral, and material resources that mediation requires, and the risks to which mediators expose themselves, the motives for mediation may be a combination of self- interest and humanitarian impulses. Mediators may engage in mediation for expansionist motives: the desire to extend and increase their resources, influence, and power. Moreover, Kleiboer asserts that a state “may seek the mediating role mainly because it wants to improve relations with one or all of the conflicting parties and thus strengthen its regional position” (Kleiboer, 1998: 45). Any mediating effort guided exclusively by self-interest is doomed to failure, just as one that does not have any investment at all in the outcome may fail to follow through appropriately. A parallel statement can be made about the conflicting parties’ attitude toward mediation. It is unlikely that they invite or accept mediation because they are interested only in peace. Each party usually expects the mediator’s intervention to work in its interest. Thus the term “self-interest” as it applies to the motives of mediator or disputants is not necessarily pejorative (Crocker et al., 2007: 442). A certain measure of self-interest on the part of all entities involved can mean that the process receives the attention, concern, and follow through necessary for success. From the mediator’s point of view as a player, two kinds of interests can be promoted through mediation. One is essentially defensive: the continued conflict between two actors threatens the mediator’s own interests. In this case, a solution to the conflict is imperative to the mediator because of the conflict’s effects on relations 32 between the parties. For example, having two of the mediator’s own allies or friends involved in a conflict can disrupt and weaken the alliance. Moreover, Kleiboer explains that a conflict between two states may be seen as disturbing a regional stability, or may provide opportunities for a rival power to increase its influence by intervening on one side of the conflict (Kleiboer, 1998: 49). Kolb provides a second self-interested motive for mediation as the aspiration to extend and increase influence. Here the solution of the conflict has no direct importance to the mediator, it is only a vehicle for establishing closer relations with one or both parties. A third party may hope to win the gratitude of one or both parties in a conflict, either by getting them out of the conflict or by helping one achieve better terms. To be sure, the mediator cannot throw its full weight behind one party, but can boost influence by rendering its involvement essential to the negotiations and by making each party reliant on him or her (Kolb, 1983: 47). Mediators can also increase their influence by becoming guarantors of whatever agreement is reached. It follows, then, that mediators are rarely impartial to the terms being negotiated. Even when they seek peace in the abstract, they try to circumvent terms not in line with their own interests, although that usually allows for a wider range of acceptable outcomes than the immediate interests of the parties (Kolb, 1983: 52). They can also allow themselves more flexibility in bargaining because they have acquired fewer commitments and have less invested in the conflict. Kolb writes that, ”mediators are likely to seek terms that will improve the prospects of stability, deny their rivals opportunities for 33 intervention, earn them the appreciation of one or both parties, or enable them to have a continued say in relations between the two adversaries” (Kolb, 1983: 52). The propositions can be illustrated by a number of historical examples. Both the US mediation in the Rhodesia-Zimbabwe conflict and the Soviet mediation between India and Pakistan were motivated by a mix of defensive and expansionist motives (Kleiboer, 1998: 15). The United States feared the Rhodesian conflict would provide opportunities for the Soviet Union to gain influence by supporting the African nationalists. But since the African groups concerned were already politically close to the Soviet Union and China, Kleiboer argues that one can also interpret the US mediation as an attempt to win over, or at least to better relations with these groups, and not to abandon them to the exclusive influence of its political rivals (Kleiboer, 1998: 17). Soviet mediation between India and Pakistan was partly inspired by the need to improve its relations with Pakistan, which had previously been on far better terms with the United States and China (Kleiboer, 1998: 15). The Soviet Union also sought to build its reputation and establish a precedent that might help to explain future involvement in the affairs of the region. At the same time, Furley and Roy explain that there were vital defensive motives for its intervention (Furley and Roy, 2006: 90). The Indo-Pakistan conflict provided an opportunity for China to expand its influence in Pakistan, and thus to establish its presence close to the southern borders of the Soviet Union. The reduction of the conflict would make China’s task more difficult (Furley and Roy, 2006: 90). Mediation by middle-sized powers may be motivated by a desire to enhance their influence and prestige. Egypt’s and Algeria’s 1975 mediation between Iran and Iraq, 34 compelled by Iraq’s war against the Kurds, resulted from their desire to prove their usefulness to both belligerents, as well as to reduce intra-Islamic conflict. 16 Furley and Roy explain that the 1982 Algerian mediation between the United States and Iran “seems to have been inspired by the desire to create goodwill toward Algeria among the US public and thus to help enhance relations between Algeria and the United States. This hope was presumably related to US support for Algeria’s adversary, Morocco, in the Western Sahara war against the Algerian-supported Polsario movement” (Furley and Roy, 2006: 90). In The Multiple Realities of International Mediation, Marieke Kleiboer, explains that superpowers have dual incentives to become a mediator: “to serve their national interests but also to promote the stability of the system” (Kleiboer, 1996: 45). Furthermore, the national interest motive itself is twofold, Kleiboer explains, as it contains both defensive and expansionist elements. Defensive motives may emerge when a conflict between two states directly threatens a mediator’s interest. Likewise, a conflict between one’s allies may provide opportunities for a rival power to increase its influence by intervening in the dispute. Mediators may also engage in mediation for expansionist motives: the desire to extend and increase their resources, influence, and power. A state may seek the mediating role mainly because it wants to “improve relations with one or all of the conflicting parties and thus strengthen its regional position” (Kleiboer, 1998: 45). 16 This is not to be mistaken with the Iran-Iraq war of 1980. The war in reference involved the Algerians mediating between the Shah of Iran and Iraq’s Ba’ith government. The Shah and the US stopped aid to the Kurds in exchange for Saddam making a territorial concession to the Shah at the Shatt al Arab waterway. The result was the collapse of the Kurdish revolt and another massacre of the Kurds in the north. 35 Importantly, Kleiboer’s work elucidates that the motivation of the respective parties can be highly complex and often not entirely transparent to the other entities involved (Kleiboer, 1998: 45). Still, that a mediator carries an agenda for becoming involved in a conflict does not necessarily compromise the process, but rather may be the very carrot needed to sustain interest in guiding the talks to a successful resolution. For their part, Kleiboer explains, disputants must have motives at the same time that they must be willing to participate in give and take with the opponent (Kleiboer, 1998: 45). Furley and Roy suggest several motives that lead conflicting parties to seek or accept mediation. The most obvious is the expectation that mediation will help gain an outcome more favorable in the balance than continued conflict. There is also the hope that mediation will provide a more favorable settlement than could be achieved by direct negotiation. Although the opposing party may not agree that the proposal is preferable, it may nonetheless cooperate with the mediator(s) and thus avoid prolonging a costly conflict. The parties may also accept mediation in the hope that the intermediary will “help them reduce some of the risks entailed in concession making, protecting their image and reputation as they move toward a compromise” (Furley and Roy, 2006: 93). They may also believe a mediator’s involvement implies a guarantee for the final agreement, thus reducing the risk of violation by the adversary. Kleiboer (1998) suggests additional motives to help explain the acceptance of mediation by international organizations. It appears to be premised more on the ability of these organizations to bestow normative approval than on their capacity to influence the adversary or arrange for a satisfactory compromise. This point is clearest in the case of 36 the International Committee of the Red Cross (ICRC)., an agency capable of offering an improved image to a fighting or detaining authority, which can be a powerful incentive for the parties to accept its presence and services, and accede to its proposals (Kleiboer, 1998: 52). Acceptance Studying mediation based on a cost-benefit calculation of obtainable outcomes provides an understanding of the mediator’s position and methods. The acceptability of a mediator to the adversaries in a conflict is not, as is sometimes believed, exclusively determined by their perceptions of the mediator’s neutrality. What is important to the parties’ decision is their consideration of the consequences of acceptance or rejection – for achieving a favorable outcome to the conflict above all, but also for their relations with the would-be mediator. Thus, interfering by third parties is tolerated because they are already part of the relationship, but third parties are accepted as mediators only to the extent that they are considered capable of bringing about acceptable outcomes. Although there is no necessary relation between past partiality and future usefulness, good relations between the mediator and one party may, in fact, help foster communication, develop creative proposals, and bring the two parties’ positions into alignment. Wallenstein explains that closeness to one party implies the possibility of “delivering” that party and hence can stimulate the other party’s cooperativeness (Wallenstein, 2007: 46). For example, Wallenstein asserts that the Africans’ belief that 37 British and US sympathies were with the white Rhodesians rendered British and US mediation promising for them and stimulated their cooperation (Wallenstein, 2007: 46). A mediator who is seen as motivated and able to deliver the party to which it is closest, may turn its partiality into a positive feature. It may foster acceptability to both parties: the “ally” of the mediator may feel confident to have him or her on the same side, whereas the party less close to the mediator may concentrate on the mediator’s ability to coerce the “ally.” In the latter case, though, the party in question needs to be convinced of the mediator’s intention to curb the conflict and prevent further escalation rather than to help his or her ally. “This is why Egyptian leaders agreed to Kissinger’s, and later Carter’s, mediation efforts between Egypt and Israel, despite the US bias in favor of the Israelis” (Kleiboer, 1998: 45). Moreover, Algeria was permitted by the US to be the mediator with Iran not because it was considered impartial, but “because its access to people close to Khomeini held promise that it might help to liberate the hostages” (Crocker et al., 2007: 444). The mediator may have two different partialities that balance each other only in the aggregate but that render it acceptable to both sides. Often a mediator motivated by its own concern for position in the area needs to use its intervention to maintain close ties with one side, at the same time improving ties with the other. According to much of the literature, the Soviet Union was neutral in its mediation between India and Pakistan (Crocker et al., 2007: 443). However, Russia and India had positive relations neither wanted to risk, and “the Soviet motive of slowing the growth of Chinese-Pakistan relations meant that the Soviets wanted an outcome that could induce Pakistan to lose 38 interest in improving relations with China” (Crocker et al., 2007: 443). If one of the Egyptian motives in mediating between Iran and Iraq was to improve its relations with the Shah, Wallenstein asserts that, then clearly the fulfillment of its goals was dependent upon how much it supported the Iranian position (Wallenstein, 2007: 50). At the same time, “both parties must perceive the mediators as concerned with achieving an outcome acceptable to both sides, not so partial to one side as to preclude that outcome” (Crocker et al., 2007: 444). Again, the question for the parties is not whether the mediator is unbiased, but whether it can come up with an acceptable solution. Timing From my premise that mediators are motivated by self-interest, it follows that they will not intervene on impulse, but rather only when they think a conflict threatens their interests, or when they recognize an opportunity. But a threat or an opportunity is unlikely to be noticed when there is only a mild disagreement between the parties. It is usually only after the conflict worsens that its implications are perceived. After the conflict has escalated, the parties are likely to become entrenched in their positions and committed to a confrontational policy, thus decreasing the possibility of successful mediation. In order for mediation to thrive, the parties must reevaluate their policies (Wallenstein, 2007: 49). The case studies in this research suggest two situations especially conducive to such reevaluations: hurting stalemates, and crises bounded by a deadline; or to use a metaphor, plateaus and precipices. 39 A plateau and its hurting stalemate begin when one side is unable to achieve its aims, resolve the problem, or win the conflict by itself and is completed when the other side arrives at a similar perception. Both parties must feel uneasy about the costly dead end they have reached. Bennett explains that both must perceive a plateau not as a momentary resting ground but as a flat, displeasing terrain stretching into the future, providing no later possibilities for critical escalation or for graceful escape (Bennett, 1995: 34). Mediation plays on the perception of an intolerable situation. Without this perception already established, the mediator must persuade the parties that escalation to break out of deadlock is impossible. Indeed, “the mediator may even be obliged to make it impossible. Thus, deadlock cannot be seen merely as a brief stalemate to be easily resolved in one’s favor by a little effort or even a big offensive, a gamble, or foreign assistance” (Wallenstein, 2007: 53). Rather, it is important that each party acknowledges its opponent’s strength and its own inability to overcome it. These perceptions require the mediator to emphasize the dangers of stalemate as each party comes to acknowledge the other’s strength. Each party’s unilateral policy option (the action it can take alone without negotiation) must be seen as a more costly and less likely way of achieving an acceptable outcome than the policy of negotiation. A plateau is thus as much a matter of perception as of reality for the parties and as much a subject of persuasion as of timing for the mediator. According to Crocker et al., successful exploitation of a plateau produces a “shift from a winning mentality to a conciliating mentality on both sides” (Crocker et al., 2007: 445). 40 A precipice is the conceptual opposite of a plateau, representing a realization that matters will rapidly get worse (Crocker et al., 2007: 445). As the term implies, a precipice may be an impending catastrophe such as a probable military defeat or economic collapse. It may be accompanied by a policy dilemma: to engage in a major escalation, the outcome of which is unpredictable, or to seek cooperation, with the first choice a matter of desperate uncertainty that threatens one side as much as the other. Or, it may be a catastrophe that has already taken place, the stolen horse that drives both parties to lock the stable. Whatever its tense (since parties are bound to disagree about the inevitability of an impending event), Wallenstein explains that, “it marks a time limit to the judgment that the situation has become unfeasible” (Wallenstein, 2007: 59). For the mediator, the precipice merely reinforces the nature of the plateau, lest the parties become accustomed to their uncomfortable deadlock. Mediators can respond two ways to precipices: they can use them and they can make them. If the parties recognize impending danger, Crocker, et al., clarify that mediators can hold up the plateau as a warning and an objectionable contrast that can make the negotiated alternative appear more favorable (Crocker, et al., 2007: 445). If there is no agreement on what a precipice would constitute for both sides, mediators can work to craft a common perception of its existence, knowing that its presence can be a valuable tool in their efforts. In his/her most manipulative role, a mediator may have to manufacture some reality to the perception of a precipice– that is, create a precipice for example through citing pressure by a fourth party. According to Crocker et al., such a tactic was undertaken by the US in 2002 to restart Sudanese negotiations, citing pressure from domestic Christian groups in 41 the United States and antiterrorist policy needs that would force it to adopt a harsher policy if serious negotiations did not begin (Crocker, et al., 2007: 445). Leverage and Stalemate In some situations, position and communication may not be enough to diffuse conflict or to further progress in negotiations. The mediator may need to draw upon other available resources to manipulate or even strong-arm parties into agreement. The mediator as manipulator requires leverage – resources of power, influence, and persuasion-- that can be brought to bear on the parties. Leverage remains, however, one of the most elusive factors of mediation: even where the need for a facilitator is glaringly apparent and the would-be mediator’s interest in peace is very clear, the mediator may simply be unable to move the parties in a productive direction. Yet leverage is the key to mediation – third parties are only accepted as mediators if they are likely to produce an agreement or resolve a predicament, and they can only aid the parties or create an agreement if they have leverage. Understandably great-power states and well-funded international organizations are more likely to have a range of substantial resources at their disposal than are private individuals who only have their skills of persuasion at hand. The use of carrot and sticks in the form of economic, military, political inducements can go far in influencing the outcome of a crisis, though they are far more effective when presented in conjunction with an effective mediator. There are many sources of leverage, beginning with the two already discussed: the need of the parties for help in getting out of a problem, and the position and interest 42 of the mediator and the parties in mediation. Rubinstein makes clear that one of the most important sources of leverage is found in the so-called “ripe moment” of intolerable stalemate weighing on each party, threatening a worse outcome in the future while offering no present way out of the deadlock (Rubinstein, 1991: 11). The stalemate must be seen by both parties as permanent unless a bilateral agreement is reached with the help of the mediator (Wallenstein, 2007: 66). The conceptual components of leverage are both simple and fragile. They are used to either make the present situation more uncomfortable for the conflicting parties (thus inducing them to come to terms with each other under the aid of the mediator) or to make the future situation of mediated peace more satisfying. Theoretically, the more ties the mediator has to a party, according to Rubinstein, the more disposable goods it possesses that the parties value—the greater the effects of pressing the parties by suspending ties and withholding value. This insight forms the basis for effective use of leverage in mediation (Rubinstein, 1991: 14). Unfortunately, suspended ties also mean reduced influence, as certain elements of communication are thereby interrupted. Moreover, feelings of resentment can ensue and the party may decide to forego mediation altogether. One challenge that has not received as much attention as it deserves, is having mediators stay away from eliciting concessions from one party while simultaneously disinclining the other party to concede. For example, Sandole and Van der Merwe assert that “assurances offered to one party by the mediator as an incentive for concessions might solidify the adversary’s attitude toward bargaining with the mediator” (Sandole 43 and Van der Merwe, 1993: 78). If one side’s interests in a mediated settlement is the prospect that the process will strain relations between the mediator and the adversary, their rapprochement “may lessen the first party’s interest in an agreement” (Sandole and Van der Merwe, 1993: 79). Still another problem concerns the mediator’s own proposals, which may be acceptable to one party but not to the other. In such situations, Wall explains that the party to which the proposals were acceptable is likely to be unwilling thereafter to accept less, making it difficult for the mediator to submit any new proposals to break the deadlock (Wall, 1985: 149). With the benefit of hindsight and a surplus of international conflicts in need of critical analysis, several newly produced dissertations are taking up the subject of mediation. Work on the advantages of a close alliance between mediator and disputants takes up Italo-Yugoslav clash (1945) and Serbo-Albanian conflict (1995—1999) as its case studies (Favretto, 2009). Another dissertation (Nuamah, 2008) considers the leverage employed to resolve conflicts of Ghana, Peru-Ecuador, and Ethiopia-Eritrea, tracking the use of six different sources of leverage—limitation, termination, gratification, deprivation, persuasion, and extraction. Mediation around conflict in former Yugoslavia is studied in a dissertation that applies a utility model to understanding the management efforts that resulted in the Dayton Peace Accords (Melin, 2008). One last, less recent dissertation, warrants mentioning, as this work, The Difference Makers: The Role of the Mediator in Peace Negotiations (2003), by Diane DeMell Jacobson. It offers case studies of five successful mediation efforts through which she generates a list of five 44 essential qualities for mediator success, which are respect, impartiality, management and leadership skills, sufficient and significant political resources, and “the ripe moment.” The existing research on leverage, power and interests that has been presented in this chapter has lead to the following two hypotheses: 1. Mediation is unlikely to be successful unless the mediator has some form of leverage/power; 2. Finding the right measure of interest in dispute resolution is central to the mediator’s success. It is hoped that the application of these two hypotheses to the two cases will yield a more thorough and enhanced understanding to conflict mediation. This chapter examined the pertinent literature in the field of conflict mediation and The Prisoner’s Dilemma, specifically explicating, issues pertaining to the mediator’s role in the conflict. These issues included leverage, stalemate, timing, acceptance, and motives. Understanding these important aspects of the mediator’s role in the Prisoner’s Dilemma game is critical to applying and understanding the successes and failures of particular cases. In later chapters, these key points will be applied to the cases in Rwanda and Northern Ireland in order to better understand the mediation process. The following chapter elucidates the theoretical and methodological underpinnings of the dissertation and their implications for the selected case studies of Northern Ireland and Rwanda. 45 CHAPTER THREE: THEORY AND METHODS Data Collection The researcher pursued several different avenues when gathering the historical data for this dissertation, cataloguing and investigating the distinguishing characteristics of mediation as discussed in the disciplines of history, political science, and international relations. Of particular importance was the work of William Ury (1988), Stephen Goldberg (1995 and 2003), Howard Raiffa (1985), Jacob Bercovitch (1992, 1996 and 2003), Martin Shapiro (1986), William Zartman (1985, 1991, 1992, 2000 and 2008), and R. Hrair Dekmejian (1980 and 2007), whose findings serve as a solid foundation from which the purpose and practice of mediator(s) may be examined and explained. While much of the information overlapped in the various narratives and discussions of the respective conflicts, remaining mindful of the insights of Carl Becker in his essay, “What Are Historical Facts?”(1955). Becker’s characterization of the historical fact” as intangible and shape-shifting bears a particular relevance to historical analyses of mediation and conflict. The nature of the past event, as he explains, is that it is no longer in front of us; thus the historian “cannot deal directly with this event himself, since the event itself has disappeared. What he can deal with directly is a statement about the event” (p. 330). As Becker would suggest, then, the various critical accounts of these conflicts are versions, not direct windows onto the events; they are told by diverse participants, scholars and critics across time, each with his or her own interpretation and investment in certain presentations of the information. 46 Becker states that each generation (and researcher) writes history based on the facts relevant to his or her time and interests and remains ever under the influence of “present purposes, desires, prepossessions, and prejudices, all of which enter into the process of knowing [an event]” (Becker, 1955: 336). Though perhaps not intending to be deceptive, no one researcher tells the whole story, indeed, knows the whole story, “[B]y no possibility,” Becker emphasizes, “can the historian present in its entirety any actual event, even the simplest” (p. 334). Becker’s caveats about history and historiography are germane to the concerns of mediation precisely because the efforts involve considering so many different versions and opinions, interpretations and facts. Translation, if not of language, but of opinion and emotion is central to the mediator’s mission. The researcher, like Becker’s careful historian, must take into consideration that each party will have its own desires, prejudices, and so forth. With these cautions, I not only analyzed the various interpretations of the conflicts but also sought to understand the motivations and meanings of the historians and scholars themselves. As such synthesis of the source material, primarily derived from secondary sources, involved consulting a variety of interdisciplinary sources, as well as engaging in a comparative study of them. In the interest of gathering material old and new, I consulted published books and articles from peer-reviewed journals and more popular publications as well theses and dissertations that addressed topics similar to mine. This material was found through ProQuest, JSTOR, and other databases with both discipline-specific and transdisciplinary publication inventories. Useful sources included scholarly journals such as The Journal of 47 Conflict Resolution, Journal of Social Issues, Journal of Peace Research, Journal of International Affairs, Journal of Policy, Analysis, and Management, Journal of Foreign Affairs, Journal of African History, The American Political Science Review, The RAND Journal of Economics, and The Journal of Econometrics, Nonacademic material included The Irish Times and PBS Television Broadcast for the IRA interviews and specials. Framework and Application The theoretical framework of this dissertation derives from a number of sources: William Ury, Stephen Goldberg, Howard Raiffa, J. Bercovitch, William Zartman, and R. Hrair Dekmejian. The framework and methodology in this research, however, are most explicitly drawn from Moore and Von Neumann’s version of the Prisoner’s Dilemma. For its part, game theory has a long and complex history with respect to its use in international relations, particularly “the theory’s association with the politics of the early Cold War” (Bennett 1995). Where some see Game Theory as unrealistic, 17 others regard it as very usefully 18 applied to explicating conflictual and competitive decision-making (Bennett 1995). It is in the spirit of the latter that I chose to use Game Theory as a productive theoretical model; in doing so, I seek to fill gaps in the research and offer a 17 For example in Institutions and the Path to the Modern Economy, Avner Greif explains, “The empirical usefulness of the analytical framework of classical game theory is puzzling, however, because this theory rests on seemingly unrealistic assumptions about cognition, information, and rationality” (2006: 124). 18 For example, Robert J. Aumann, a professor emeritus of mathematics at the Hebrew University of Jerusalem, and Thomas C. Schelling, a professor emeritus of economics and public policy at the University of Maryland at College Park, were awarded the Nobel Prize in Economics in 2005 for separate studies showing how game theory provides a pervasive mode of analysis of human interaction. Their models have been applied to such questions as price wars between businesses and nuclear wars between nations. 48 new means by which to consider mediation so that other practitioners and scholars of the subject can predict or enhance the outcome of their own conflict resolution efforts. After an extensive literature review in this topic area, it became apparent that the use of game theory in the context of mediation had not yet been addressed in the fields I examined. Thus my application of the Prisoner’s Dilemma to these case studies offers a fresh approach to understanding the intricacies of mediation, one that may be usefully consulted and applied to other negotiation scenarios. Notably, I will apply an extended form of the Prisoner’s Dilemma where the model of win-win and lose-lose is still valid. As Roger Fisher and William Ury detail in Getting to Yes: Negotiating Agreement Without Giving In (1981), their best-selling book about non-adversarial bargaining, a mediator must guide disputants toward understanding that “win-win” may require some very real compromises. By way of example, they detail Fisher’s work in the Middle East Peace process. Seeking a mutually satisfying agreement between Israeli Prime Minister Golda Meier and Egyptian President Gamal Abdel Nasser, Fisher found that the parties had to come to an understanding of the relative meanings of “win” (1981). After Nasser enumerated a list of demands for Meier, Fisher provoked Nasser’s own admission that the proffered list would be problematic and even dangerous for Meier to accept. The problematic nature of the demands, as Nasser himself recognized, meant that Meier would never accept them; to offer something intolerable to the opposition, he realized, was to preclude his own success. Only when the possibility emerges that the opposition will say yes does the real work of securing a win-win situation begin. 49 This anecdote underscores that parties need to consider what they are willing to give up (and still “win”) as much as they need to acknowledge what they are asking their adversary to sacrifice (and still “win”). Such an approach provides clarity to the mediation process, helping the mediator see possibilities for a successful resolution to the conflict. This dissertation’s central objective is to explicate the dynamics of the two case studies by applying the deductive knowledge employed in game theory. Ultimately, this study will show that greater insight into conflict resolution can be gained when the factors of the Prisoner’s Dilemma game are clearly understood. 19 The game allows the mediator(s) to simplify the issues by getting to the key factors that have caused the conflict to continue. Before applying game theory, I will relate Moore’s 12 stages of mediation to the two cases. 20 The selection of Moore’s study as the framework for this research was based upon its comprehensive 12-step process, which facilitates careful analysis of the mediation process and can even serve as a template with which to analyze mediation scenarios retrospectively. According to Moore, if the case studies fulfill the 12 steps, one can determine with a fair degree of certainty that the process was a success. The incremental nature of the framework, breaking down as it does the mediation process into 12 steps, allows a clear understanding of where and why mediation failed. Never before applied to a scholarly study of mediation, Moore’s steps will be used to provide insight into how agreements are reached (or not) even in the most complex 19 These specific factors are discussed in chapter eight. 20 It is important to note that these are two western constructs being applied to the two cases. 50 conflicts. After close consultation of the literature on conflict management and resolution, I found Moore’s processes to be the most productive for my purposes. Applying Moore’s typology to the Rwandan case will help pinpoint and clarify where and why the mediation process failed. Likewise, applying Moore’s steps to the Northern Ireland case will help clarify why it was able to succeed. Once this application is completed, I will analyze the cases using the stages of a modified Prisoner’s Dilemma setting, considering issues such as payoffs, costs, and players’ roles. Using Moore’s theory to illuminate where the mediation went wrong will help shed some light on how the Prisoner’s Dilemma game was played and why it succeeded or failed. Even though Moore’s theory and the Prisoner’s Dilemma are two separate methodologies, they effectively complement one another and help lead to a more balanced understanding of the mediator’s role in the negotiation process. To comprehend the scope, purpose, and influence of mediation, it is crucial to understand the context and the issues of the involved parties. I will demonstrate that applying game theory to the capability of the mediator(s) affords a better understanding of how and why the mediation succeeds or fails. For this reason, Moore’s 12 steps of mediation will be applied to the two case studies. After careful research, I also found Moore’s framework to be the most cohesive and thorough. Its 12 steps will help pinpoint where and why mediation succeeded or failed. Use of this framework will help make the results of the Prisoner’s Dilemma game more comprehensible. 51 The Prisoner’s Dilemma I chose to use the Prisoner’s Dilemma (PD) as a lens through which to understand mediation because of what the model can tell us about how to manage the deadlock that so often results from two-party disputes. One of the mediator’s most important jobs is to convince the disputing parties that the instinct to pursue rational self-interest will not arrive at a common good and therefore may not ultimately serve either one of them. Like contestants in a Prisoner’s Dilemma scenario, disputants want to achieve their maximum payoff, but the game is often zero-sum so that neither can have their maximum payoff. The mediator must then present alternative incentives and convince the parties that these “carrots” are better recompense than remaining in deadlock and getting nothing at all. As an interlocutor and guide, the mediator seeks cooperative resolution between two parties who may not see the personal, economic, political, or moral advantage of compromise. In the literature examined, there were no studies on the significance of the Prisoner’s Dilemma to mediation. Game Theory, however, has served a popular application in disciplines as wide-ranging as economics, 21 law, 22 biology, 23 criminology, 24 animal behavior, 25 business, 26 mathematics, 27 computer science, 28 21 For example, the work of Ananish Chaudhuri, Robert J. Aumann, and Sergiu Hart (see bibliography for titles). 22 For example, the work Douglas Baird, Robert H. Gertner, and Randal C. Picker (see bibliography for titles). 23 For example, the work of John Maynard Smith and Ronald Fisher (see bibliography for titles). 24 For example, the work of Andrew Colmon and Clare Wilson (see bibliography for titles). 25 For example, the work of Lee Alan Dugatkinand Hudson Kern Reeve (see bibliography for titles). 26 For example, the work of Pankaj Ghemawat (see bibliography for titles). 27 For example, the work of Steve Heims, John Von Neumann and Norbert Wiener (see bibliography for titles). 28 For example, the work of Erich Grädel (see bibliography for titles). 52 philosophy, 29 ethics, 30 psychology, 31 and “everyday life 32 . For political science, PD specifically serves as a productive model for understanding the behavior of two states in an arm’s race, its relevance to mediation and the pivotal role of the mediator has still not been fully examined. Few researchers have attempted to understand whether a relationship exists both between them even though case studies suggest that there is much to learn about how to craft systematic approaches to some of mediation’s thornier issues. In applying this model to the cases of Northern Ireland and Rwanda, I seek to map out an “extended version” of the Prisoner’s Dilemma, one that specifically aids the efforts of the mediator and enhances our understanding of the strategies of mediation. I label my adaptation an “extended version” because there are clear variations on the model: The parties are not prisoners, there is no jail; in mediation, generally the parties know what the opposition is being offered; and the parties are not being held in separate rooms. Case Studies The researcher chose the Northern Ireland and Rwanda case studies for their complementary nature and exceptional qualities—one of exemplary success and the other of egregious failure. These case studies allowed the researcher to draw conclusions about the impact of each characteristic on the mediation efforts. As detailed in the Introduction, in the main, scholars consider the Northern Ireland negotiations a success and Rwanda’s, 29 For example, the work of K.G. Binmore (see bibliography for titles). 30 For example, the work of David Gauthier (see bibliography for titles). 31 For example, the work of Herbert Gintis and Steven Brams (see bibliography for titles). 32 For example, the work of Avinash Dixit (see bibliography for titles). 53 a failure. The change in attitudes by the respective disputants in Ireland, as demonstrated by recent violence, which was strongly condemned by both sides. This is one indication that Mitchell’s efforts brought about fundamental change to the balance of power between the parties. As for Rwanda, the staggering statistics that surround the post-peace accord genocide speak for themselves. With its population reduced by several million, Rwanda was inexorably transformed and implementation of the peace treaty recognized internationally as a tragic failure. As with all qualitative research, generalizing these results to other case studies is not necessarily the end goal but transferability may be achieved; in other words, the theories and conclusions found herein may be transferred to other cases but the results cannot always be extended/generalized to all similar sites. Purpose of the Study The purpose of this study is to investigate the importance of leverage, in the form of power, tools and resources, and interests, on the mediator, as individual or governmental entity, in influencing the outcomes of the mediation process in international conflicts. The use of the Prisoner’s Dilemma will be utilized to show future mediators how it is possible to predict the way the game will play out by understanding the different motives and interests of each party and helping to predict the steps each one will take. While the Prisoner’s Dilemma shows the mediator why and how the mediation fell apart, 54 Christopher Moore’s 12-step framework will help the mediator understand when and where the mediation went wrong, if indeed that is the case. These methods will be applied with the main consideration of the mediator’s use of leverage and neutrality and how they affect the outcome of the process. Hypotheses 1. Mediation is unlikely to be successful unless the mediator has some form of leverage/power. 2. Finding the right measure of interest in dispute resolution is central to the mediator’s success. 55 Figure1.1: Map of Northern Ireland Source: http://en.wikipedia.org/wiki/File:Northern-irland-religions-1991.png 56 CHAPTER FOUR: HISTORICAL CONTEXT OF THE 1966-1998 CONFLICT IN NORTHERN IRELAND Northern Ireland’s religious and communal structure dates back to 1602, when the English defeated Hugh O’Neill’s uprising. Since the uprising was supported by Spain, it was part of the broader conflict between Catholics and Protestants in Europe. The English declared their land “forfeited to the Crown” when O’Neill and his supporters fled (Alonso, 2003: 8). This move led to a systematic pattern of English and Scottish settlement in the “plantation” of Ulster. 33 Plans were put in place to completely eliminate the Irish from the region but were improperly executed, leaving a blend of religions and ethnic communities in the area (Foster, 1988: 26). This conflict fueled resentment in Ireland’s native Catholic population since the community’s close ties with, and fidelity to, London turned into economic and political power (Foster, 1988: 26). Tensions deepened when James II, a Catholic, fled to Ireland after England’s “Glorious Revolution” of 1688 and, despite Ulster’s help, was defeated by William of Orange at the Battle of the Boyne in 1690 (O’Brien, 1993: 83). According to Foster, the English severely mistreated the native Irish on the island, regarding them as traitors, discriminating against them, and denying many of them the right to vote and the right to inherit property. By the middle of the eighteenth century, Catholics owned “less than five percent” of the land in all of Ireland (Foster, 1988: 154-55). 33 The term “plantation” did not mean colony at this time. It was the term used to refer to a specific area. 57 When the ideas of the French Revolution spread to Ireland by the end of the eighteenth century, another bloody revolution supported by the French occurred. Though the revolution was successfully suppressed, the English quickly realized the need for reform. They created a plan for both Catholic emancipation and incorporated governance of both England and Ireland through one Parliament in Westminster. Ranelagh explains that Ireland was formally declared a part of the United Kingdom in 1801, but Catholic emancipation would not be achieved until 1829 (Ranelagh, 1995: 264). The history of Ireland took a drastic turn during the great famine of the late 1840s. Dingley explains that nearly one-eighth of the population was killed and another million were forced to immigrate, primarily to the United States (Dingley, 2008: 86). The famine was a reflection of the tremendous poverty of Catholic Ireland and the relative economic prosperity of Ulster. Eventually, large-scale action was taken in the Catholic areas, as exemplified by the Land League, which selected specific (typically English) oppressive landlords and ostracized them through mass meetings and the refusal to provide services (Boyle and Hadden, 1994: 134). This period also witnessed the rise of nationalist societies, such as the Irish Republican Brotherhood (the forerunner of the IRA), that were “committed to overthrowing the Union by force” (Boyle and Hadden, 1994: 134). In an attempt to deal with these issues, the British governments created the 1881 Land Act, reinforcing the position of tenant farmers against their landlords. However, radicalization did not cease. Dingley explains that the success in getting these reforms through Parliament encouraged the founding of the Home Rule League in 1874, which “aimed to achieve its objectives by peaceful constitutional means” (Dingley, 2008: 247). 58 During the late 1880s, Home Rule legislation was considered in Westminster, stirring up the opposition of Ulster Protestants (Dingley, 2008: 247). On the Catholic side, Irish nationalism grew among both intellectuals and the working people, with the spread of organizations such as the Gaelic Movement (Townshend, 1988: 4). Due to internal political pressures, a new Home Rule Bill was put forward in 1912 by the British government. It would have provided for home rule through a single minister (as were others in the Commonwealth), but the bill was initially blocked by the House of Lords and went through several readings (Townshend, 1988: 7). This interval provided Ulster Oppositionists the time to threaten an uprising of their own. Townshend explains that they organized volunteer forces and obtained support from some of the military as well as from the police who stood by while arms were illegally consigned to the volunteers (Townshend, 1988: 6). The Home Rule Bill did pass in 1914, following talk of arrangements for Ulster to opt out. However, these arrangements were not resolved, and formal implementation of home rule was to be delayed until the end of World War I (Townshend, 1988: 7). These discussions ultimately contributed to the postwar division and establishment of two separate parliaments: one for the Catholic south in Dublin and one for Ulster in Belfast. Disappointed by the Home Rule Bill, Irish leaders began to seek out more complete independence. Townshend explains that they took advantage of Britain’s attention to the war by organizing an Easter rebellion in Dublin in 1916. British forces put down the uprising after a week and executed the leaders, further inflaming popular resentment against them. Guerrilla conflict continued between the newly named Irish 59 Republican Army (IRA) on the one hand, and the Royal Irish Constabulary, the infamous “Black and Tans,” on the other (Townshend, 1988: 77). Finally, the guerilla effort succeeded, leading the British and Dublin leaders to negotiate a settlement in 1921, which culminated in the creation of the Irish Free State (Ranelagh, 1995: 133). This state became a republic and left the Commonwealth in 1949 and Northern Ireland remained part of the United Kingdom, but was given its own parliament in Belfast with certain limited powers, including policing, education, and social services (Townshend, 1988: 80). Partitioning Ireland was not meant to be a permanent solution to the Anglo-Irish conflict. While it successfully shifted the locus of the conflict, it did nothing to address the intercommunal tensions and resentments at its core. The discrimination and subordination of Catholics continued to intensify after Northern Ireland became a distinct political entity. Townshend asserts that Northern Ireland’s Protestant majority used the newly acquired instruments of government to entrench itself and hold down the Catholic minority. Even though Britain had considerable financial power over Northern Ireland, it did not make use of this leverage in order to implement reforms (Townshend, 1988: 80). Northern Ireland began to destabilize due to the widespread inequalities. By the 1960s, the Catholic community in Northern Ireland began to organize and demand an end to the Protestant abuse of political and economic power. By 1968, “a nonviolent civil rights movement modeled after its US counterpart emerged from Northern Ireland’s Catholic community” (Ruane and Todd, 1996: 224). Attempts by the Catholic community to bring about peaceful change, however, resulted in deadly clashes with 60 Northern Ireland authorities and parts of the Protestant community. Elliott declares that these clashes in turn led to the resurgence of the IRA (Elliott, 2002: 169). In 1969, in the face of a rapidly deteriorating security situation, British military forces were called in to maintain law and order. This marked the end of Northern Ireland’s relative independence and established the present context for the conflict. Ruane and Todd emphasize that, as the number of British troops in Northern Ireland grew from initially a few hundred to 30,000 at one point in 1972, so too did British involvement in the daily affairs of Northern Ireland (Ruane and Todd, 1996: 224). While the British implemented reform policies, the presence of the army created several new problems. The summer of 1971 reflected the implementation of new policies intended to destroy the IRA. English affirms that the British and Northern Ireland authorities instituted a policy of internment under which Catholics suspected of involvement in the IRA were arrested and held without trial. By the spring of 1972, 900 individuals were being held, many of them innocent (English, 2003: 203). Many of these detainees were put through a grueling process of harsh and demeaning interrogation. Amid escalating violence and British involvement, it became increasingly difficult for the British to prop up the local government. Bendan Hughes, leader of the 1980 hunger strike, later argued with reference to the prison protests that, “What took place between 1976 and 1981 was a war. The British government tried to defeat the republican struggle through a policy of degradation, isolation, beatings and criminsalisation. They did not succeed…" (English, 2003: 204). In March 1972, the United Kingdom imposed direct rule and disbanded the Northern Ireland Parliament. 61 While it was originally intended to be a temporary measure, Cox, Guelke and Stephen state that both communities possessed the power to thwart any attempt at reform. Northern Ireland thus entered a state of political deadlock. With the exception of a three month period in 1974, Northern Ireland was governed directly from Westminster until the summer of 1998 (Cox, Guelke and Stephen, 2000: 29). The modern legacy of Northern Ireland’s tragic history is a society severed into two distinct communities – one Irish Catholic and Nationalist, the other settler (i.e., Scottish, Welsh, or English) – Protestant and Unionist. The divide consists of a complex layering of differences that transcends religion and encompasses ethnicity, culture, historical experience, and social class. The divide that exists between Northern Ireland’s two communities had traditionally presented a formidable obstacle to peace and reconciliation. Not only does it help perpetuate cross community animosities and intolerance, it can also, during times of inter-communal tension, generate a siege mentality and radicalization. The root of this mentality is insecurity – for Catholics arising from their minority status and their history of persecution and subordination, for Protestants arising from concerns about their future ability to preserve their cultural identity and remain independent from the Republic of Ireland. This dynamic is particularly evident during Northern Ireland’s “marching season” when Unionist hard liners insisted on parading through Catholic neighborhoods in tribute to the Protestant victory over the Catholics in July 1690 (Elliott, 2002: 40). 62 The communal partition in Northern Ireland is exemplified by the self-imposed segregation in schools, living areas, and work settings (Boyle and Hadden, 1994: 33-48). According to demographics and studies, the people in Northern Ireland lived in areas populated primarily by members of their own community (Gosling 2007). Some contend that this choice is influenced by the need for security. As Kevin Boyle and Tom Hadden observe, “it takes only one brick through a window or one scribbled graffiti to make a family move out of a mixed into a safer area” (Boyle and Hadden, 1994: 35). Boyle and Hadden note, the rate of unemployment amongst the Catholic community in Northern Ireland has always been higher than that of the Protestant community. In 1991, Northern Ireland experienced an overall unemployment rate for male workers of 20% (Boyle and Hadden, 1994: 44). Broken down by community, the male unemployment rate among Protestants was 13.9%, compared to male unemployment among Catholics of 28% (Ranelagh, 1995: 264). The downward-focused wealth profile compared to that of the Protestant community helps explain the uneven unemployment rate for the Catholic community. Boyle and Hadden observe that the average income for Protestant households tends to be higher than the income of Catholic households. In 1992, for example, the average income for Protestant households in Northern Ireland was ₤290 a week compared to ₤264 for Catholic households, roughly a 10% disparity. These statistics belie an even greater disparity, as the average Catholic household of 3.2 persons is larger than the average Protestant household of 2.5 persons (Boyle and Hadden, 1994: 50). 63 Parties to the Northern Ireland Conflict Many attempts to secure peace have been made by members from both the Protestant and Catholic communities. Hundreds of grassroots meetings, often sponsored by churches, religious groups, and nongovernmental organizations (NGOs), have sought common ground in the conflict. The conflict in Northern Ireland has been an inter- communal struggle over the political future of the province. The principal antagonists have been the province’s Protestant and Catholic communities, each represented by political movements that embody their political aspirations. These movements, referred to generally as Unionism and Nationalism, were not monolithic, but rather include numerous factions and subgroups. The roles of the British and Irish governments have been historical in origin and place them at the center of the peace process. Unionists The primary objectives for Northern Ireland’s Unionists were a Northern Ireland independent from the Irish Republic, under Protestant control, and part of the United Kingdom. The two primary Unionist political parties in Northern Ireland are the Ulster Unionist Party (UUP) and the Democratic Unionist Party (DUP), the latter the more antinationalist of the two. Although they differ on domestic social and political issues, Hennessey stresses that the two parties align in their rejection of unification with the Republic of Ireland and in their support of Northern Ireland’s continued status as a province within the United Kingdom (Hennessey, 2001). 64 The common position among the Northern Ireland’s Unionists with respect to the United Kingdom, however, belies complex motives. As Joseph Ruane and Jennifer Todd note, the reasons for valuing ties with the United Kingdom vary. Some do so because they feel a deep sense of loyalty and affinity with the British world or fellow British subjects, others because of a commitment to what Arthur Aughey has called the “idea of the Union.” Others value particular British institutions or traditions … or the British economic subventions. For still others the union serves as a purely defensive function – it is a defense of Protestant interests against Roman Catholicism and united Ireland. For many years, there is a strong conditional quality to the support; if the benefits ceased, loyalty would no longer be assured. (Ruane and Todd, 1996: 54) The various motivations produce a complex relationship between London and Northern Ireland – characterized by a mixture loyalty and distrust. Many Unionist paramilitary groups exist, the largest of which is the Ulster Defense Association (UDA). Traditionally, Unionist paramilitary violence has been reactionary, shadowing Republican violence (O’Brien, 1993: 35). Nationalists The unification of Ireland was the primary objective of Northern Ireland’s Nationalists (or Republicans). Although 80% of Northern Ireland’s Catholics supported the goal of Irish unification, the community – and the Nationalist movement – was divided over how and when unification should be achieved (O’Brien, 1993: 20). The two primary Nationalist political parties were: the Social Democratic and Labor Party (SDLP), and Sinn Fein, the political wing of the IRA. The SDLP was the largest Catholic political party in Northern Ireland, holding approximately two-thirds of the Catholic vote 65 (Cox, Guelke, and Stephen, 2000:16). 34 It preferred a united Ireland achieved through constitutional means that secured Protestant participation. For the SDLP, the goal of a united Ireland was a long term aspiration – the major obstacle being the intransigence of Northern Ireland’s Protestant community. Sinn Fein, SDLP’s principal rival for Catholic votes in Northern Ireland, favored the immediate unification of Ireland and the ejection of British forces from the island (Cox, Guelke, ad Stephen, 2000: 19). As its ties with the IRA suggest, Sinn Fein traditionally viewed the use of political violence, in the form of bombings and assassinations, a strategic inevitability. The IRA was a well-financed and well-equipped force of about 400 principal members, the largest of the Republican paramilitary groups 35 . For almost three decades, the IRA waged an armed campaign against Northern Ireland’s security forces and the civilian Protestant community. During that time, Moloney affirms that the IRA is believed to be responsible for the deaths of approximately 1,800 people (Moloney, 2003: 73). The British military maintained a large military presence in Northern Ireland beginning in 1969. While the official British position was that Northern Ireland was a part of the United Kingdom, the British government stated that it had no economic or strategic self-interests or desires in Northern Ireland and agreed to uphold the democratic wishes of the majority of Northern Ireland’s citizens. Moreover, O’Ballance emphasizes 34 Sinn Fein held about 10% of the vote (Cox, Guelke, and Stephen, 2000:16). 35 According to W.A. Tupman it is difficult to pin down the exact sources of IRA income, but evidence indicates that various illegal undertakings finance the IRA, among them money laundering, drug dealing, illegal drinking clubs, and counterfeit computer software. Contributions also come from overseas subsidiaries (1998). 66 that the government claimed that it was willing to support the people if they preferred a Union or a sovereign United Ireland (O’Ballance, 1981: 209). The British government traditionally viewed the conflict as an internal matter and resisted efforts to “internationalize” the conflict. However, in a major change in policy in 1985, the British government recognized the Republic of Ireland’s “special interest” in the affairs of Northern Ireland (O’Ballance, 1981, 209). The Republic of Ireland, which is approximately 95% Catholic, never relinquished its claim to the entire island following partition in 1921,as indicated in the Republic’s Constitution, Articles 2 and 3, which affirm jurisdiction over all of Ireland’s original thirty-two counties (O’Brien, 1993: 85). Since its acceptance of the Anglo-Irish agreement in 1985, however, the Republic’s claim to all of Ireland has given way to the recognition that unification will require the approval of a majority in Northern Ireland (O’Brien, 1993: 86). Understanding the historical background of the conflict in Northern Ireland is essential to grasping how the steps of the mediation process were carried out. Motives and preferences are understood better when the historical background of the party is included in the justification. Without a clear understanding of what happened in the past, it would be impossible to recognize why the Prisoner’s Dilemma game concluded as it did. The following chapter maps out the various efforts to resolve the conflict; chapter seven considers mediation in the context of Moore’s 12-setps and chapter eight explicates the mediation efforts through the lens of the Prisoner’s Dilemma. 67 Timeline of the Conflict in Northern Ireland 36 1801 Ireland is formally declared a part of the United Kingdom. 1912 The British government puts forward a new Home Rule Bill that would have provided for home rule through a single parliament for a unified Ireland. Home Rule Bill passes in 1914. World War I forced the postponement of further consideration of home rule. 1916 Irish leaders, disappointed with the failure of the Home Rule Bill and seeking more complete independence, organize an Easter uprising in Dublin. The uprising fails to receive wide support and British forced put it down. Following the uprising, guerilla fighting continues between the newly named Irish Republican Army (IRA) and the Royal Ulster Constabulary. 1920 The Government of Ireland Act provides for the partition of Ireland into a six-county north and a twenty-six county south, each with its own government and parliament. 1949 The Irish Free State becomes a republic and leaves the British Commonwealth. The British Parliament adopts the Ireland Act, which confirms the status of Northern Ireland as part of the United Kingdom of Great Britain. 1968 The civil rights campaign begins in Northern Ireland. 1969 British troops deployed in Derry and Belfast. 1972 The Northern Ireland Parliament is suspended and direct rule is introduced. Henceforth, Northern Ireland is ruled form Westminster. 1981 The Anglo-Irish Intergovernmental Council is established at the Thatcher- Haughey Summit. 1985 The Anglo-Irish Agreement establishes the Anglo-Irish Ministerial Conference and Joint Secretariat. 36 Greenberg, Melanie C., John H. Barton and Margaret E. McGuinness, eds. Words over War: Mediation and Arbitration to Prevent Deadly Conflict. Lanham: Rowman and Littlefield, 2000. 68 1986 Sinn Fein drops abstentionism from its constitution. If elected, Sinn Fein candidates will take their seat in the Dail, the Republic of Ireland’s Parliament. 1988 The first round of secret talks are held between John Hume, leader so the Social Democrat and Labour Party, and Gerry Adams, leader of Sinn Fein. 1990 Northern Ireland Secretary Peter Brooke initiates the three-strand formula for talks between the Northern Ireland’s primary political parties and British and Irish governments. Strand 1 of the talks break down after 10 weeks. 1992 The three-strand talks resume. An official Unionist delegation meets an Irish delegation in Dublin under Strand 2. Negotiations end inconclusively. 1993 Renewed Hume-Adams talks deliver peace proposals to Irish government. The British and Irish governments issue the Downing Street Declaration. 1994 August: The IRA announces a unilateral cease-fire. October: The Combined Loyalist Military Commands declares a cease-fore on behalf of Protestant paramilitaries. 1995 February: The Irish and British Governments publish a New Framework Document. November: The Irish and UK governments issue a communiqué announcing the creation of an International Body to provide an independent assessment of the decommissioning issue. 1996 January: The International Body issues its report on the decommissioning issue recommending disarmament in stages during substantive negotiations. Shortly after the report’s release, the British government proposes an election in Northern Ireland to establish a constituent assembly from which teams of negotiators would be chosen to represent the Republican and Unionist parties in multiparty negotiations. Feb. 9: The IRA terminates its 17 month cease-fire by exploding a bomb in the Docklands area of London. May 30: Elections are held in Northern Ireland to select representatives to all-party negotiations, which are scheduled to begin on June 10. Although Sinn Fein has its best electoral showing ever, it is banned from the opening of negotiations because of the IRA’s continuing military campaign. June 10: Delegates from Northern Ireland’s political parties gather in Belfast for the Opening Plenary Session of negotiations. Shortly after negotiations begin they deadlock over the issue of decommissioning. 69 1997 March: After eight months of stalemate over the issue of decommissioning the Opening Plenary Session is suspended pending the approaching British general election and local elections in Northern Ireland. May1: British voters elect the Labour government of Tony Blair to replace the paralyzed Conservative government of John Major. June: The British government offers Sinn Fein admission to the peace process if the IRA declares a “genuine” cease-fire and observes it for six weeks. July 19: After a joint appeal from Gerry Adams and John Hume, the IRA accepts the offer and renews its cease-fire. July 24: Both the Democratic Unionist Party and the UK Unionists walk out of the negotiations. August: The British and Irish governments sign an agreement establishing the Independent International Commission on decommissioning. Sept. 15: Sinn Fein negotiators pledge their adherence to the Mitchell Principles and take their seats at the multiparty talks. Dec. 22 - Jan. 20: Loyalist extremists detonate three car bombs in Northern Ireland. In possible retaliation for the first of these attacks, the Irish National Liberation Army, a Republican paramilitary group not bound by the IRA cease-fore, murders the leader of the Loyalist Volunteer Force in the Maze prison in Belfast. 1998 Jan. 26: The British and Irish governments expel the Ulster Democratic Party (UDP) from the negotiations for four weeks, after the Ulster Freedom Fighters – a Unionist paramilitary group with whom the UDP is affiliated – is tied to a string of sectarian killings. February: Sinn Fein is expelled for two weeks from the negotiations in response to IRA killings of a Loyalist paramilitary and a “punishment” killing of a Catholic drug dealer. George Mitchell decides to move up the deadline for completion of the negotiations from May to April 9. April 10: After weeks of intensive negotiations, the parties to multiparty talks declare that they have reached an agreement. May 22: The people of Ireland, north and south, approve the Northern Ireland peace agreement in separate referendums. Aug. 15: Republican extremists opposed to the peace talks detonate a bomb in the Markey town of Omagh, in Northern Ireland. The blast kills twenty eight and injures over 200. Instead of halting the peace process, however, the attack appears to strengthen the momentum toward peace and reconciliation in both the Protestant and Catholic communities. 70 CHAPTER FIVE: THE MEDIATION PROCESS IN NORTHERN IRELAND The IRA: Key Interventions and Major Actors The origins of the multiparty negotiations that culminated in the Good Friday Agreement can be traced back to the early 1980s and improved Anglo-Irish relations. Beginning with the first Anglo-Irish summit between British Prime Minister Margaret Thatcher and Irish Taoiseach (Prime Minister) Charles Haughey in 1980, bilateral relations that had for decades been characterized by conflict were transformed and came to be defined by a relatively high degree of cooperation and mutual support (Alonso, 2007: 34). Over the next 15 years, British and Irish officials worked collectively to produce a series of joint agreements, declarations, and initiatives that set the structure of negotiations and identified fundamental principles upon which a political resolution to the conflict could be based. These developments preceded multiparty negotiations, incrementally pushing the peace process forward, though not without significant obstacles. The key developments during this period included the Anglo-Irish Agreement of 1985, the Brooke-Mayhew Initiative of 1992, and the Downing Street Declaration of 1993. The Anglo-Irish Agreement of 1985 The 1985 Anglo-Irish Agreement was the first solid demonstration of the new relationship between Dublin and London. It was expected to bolster constitutional 71 nationalism against the IRA – an idea that gained support not only among British officials, but also within the Republic. Kane explains that the Agreement was able to accomplish this objective by recognizing the rightful interest of the Republic of Ireland in the affairs of Northern Ireland and by institutionalizing the emerging relationship between Dublin and London (O’Kane, 2007). The momentous nature of the Agreement arose from its creation of a standing Intergovernmental Conference served by a permanent joint Secretariat composed of Irish and British civil servants. Within the conference, the Republic of Ireland was allowed to make proposals on British policy toward the province, “which would give the Nationalists a voice in the formulation of British policy in Northern Ireland for the first time” (O’Brien, 1993: 24-25). The process was not without its sharp critics, however, as the London Times of November 23, 1985 reported on plans for an enormous rally of loyalists to demonstrate against the Agreement. Figures such as Rev. Ian Paisley, leader of the Democratic Union Party, and James Molyneaux, leader of the Official Unionist Party, declared “all out opposition to the deal” (Ford, 1985:1) Though the Agreement was never fully implemented, the relationships that came about as a result of the Conference had notably improved (Mitchell, 2001: 209). These relationships of trust helped provide reinforcement and resilience to the peace process and ultimately proved essential to the success of the mediation. 72 The Brooke-Mayhew Initiative In January 1990, Peter Brooke, the newly appointed British Secretary for Northern Ireland, took the first step toward bringing the major political parties of Northern Ireland to the negotiating table. On January 11, The Independent’s “Cautious Optimism in Ireland” reported Brooke’s words to an audience in Bangor, Co Down that, “Much work needs to be done . . . But there may now be in my judgment enough common ground to make worthwhile the start of talks soon on new arrangements for exercising political power within Northern Ireland” (1990: 28). Due to the central role played by the Republic of Ireland, the Brooke initiative differed from earlier efforts to create a dialogue between the parties to the conflict. According to Elliott, the Intergovernmental Conference and the Joint Secretariat established under the 1985 Anglo-Irish Agreement not only guaranteed formal Irish government support for the effort, but also brought them into the process as a cosponsor of the effort (Elliott, 2002: 111). After more than a year of relentless effort, Brooke was able to put together a three-strand framework for the talks, designed to take account of the diverse facets of the conflict in Northern Ireland (Fisher and Keashly, 1991: 33). Fisher and Keashly assert that, Strand 1 intended to address the relationships between the parties in Northern Ireland; Strand 2 meant to take account of the relationship between Northern Ireland and the Republic of Ireland; and, finally, Strand 3 sought to address the relationship between the Republic of Ireland and the United Kingdom (Fisher and Keashly, 1991: 33). According to Fisher and Keashly, despite the many attempts to initiate the talks, they 73 collapsed at the outset. The Strand 1 talks conducted under Brooke’s supervision were bogged down over procedural matters and were not effectively concluded. Strands 2 and 3 were initiated by Brooke’s successor, Patrick Mayhew, and were also ineffectively executed (Fisher and Keashly, 1991: 33). Nevertheless, the initiative marked a significant contribution to the peace process. The three-strand formula devised by Brooke remained the main framework for negotiations throughout the remainder of the mediation process (Fisher and Keashly, 1991: 33) and the failed initiative provided a constructive lesson regarding the limits of Britain’s role in the process (O’Brien, 1993: 266). Not surprisingly, Gerry Adams, president of Sinn Fein, was a highly vocal critic of the initiative. As The Irish Times reported Adams retrospectively accused the British Government of responsibility for the inherent flaws in the process: It is important that we all realize the difficulties facing the British because of their failure to resolve the conflict—even temporarily—on their terms. For as long as they fail to so this and for so long as we succeed in exhausting and surviving their offenses—then the irreversible thrust of this struggle moves relentlessly toward the end of British rule in our country . . . It is important that we always keep our eyes on this big picture and that we are mindful of the strength of our struggle and the weakness of the British position. Our political position has integrity which the British one lacks. (O’Reagan, 1993: 8) Notably, however, Adams praised Bill Clinton’s pre-election assurance that Ireland would rank high on his presidential agenda, but made a pointed remark of the “abject failure” of Brooke-Mayhew to re-establish talks. While Brooke’s determination and personal attributes are credited with making the talks possible, his position as British secretary for Northern Ireland hopelessly 74 complicated his role in the negotiations. Bloomfield explains that Brooke used reasoning, persuasion, the control of information, and the suggestion of alternatives to bring about a negotiated settlement, but as a representative of a party to the negotiations, Brooke could not be impartial (Bloomfield, 1997: 123). Thus despite his persistence and admirable intentions, Brooke’s associations meant that neither the perception nor the reality of neutrality could ever be achieved. By November 1992 the talks had completely fallen apart, with the Unionists withdrawing— but Brooke’s legacy was not completely without distinction. As one critic put it, Brooke’s role would play its own tacit part in later aspects of the Mitchell negotiations: When historians finally write the list of heroes of the Mitchell talks, Peter Brooke’s name and that of his successor, Sir Patrick Mayhew, may appear first. The troubles Senator Mitchell faced in reaching his agreement had all appeared in the Brooke-Mayhew effort. He also used their mode of operation: he “forced” closure to issues whenever necessary. (O’Grady, 2001: 92) The Downing Street Declaration and the 1994 Cease-fires Despite the failure of the Brooke-Mayhew Initiative, the peace process progressed (Elliott, 2002: 207). On December 15, 1993, Irish Taoiseach Albert Reynolds and British Prime Minister John Major issued a Joint Declaration, setting out both governments’ understanding of the origins of the conflict, and the principles upon which they believed the peace process could be founded (Elliott, 2002: 209). The professed goal of the Irish and British governments was the creation, through political dialogue, of “institutions and structures that, while respecting the diversity of the people of Ireland, would enable them 75 to work together in all areas of common interests” (Joint Declaration of Irish Taoiseach, 1993: 9). At the heart of the Declaration were the principles of democracy and respect for the consent of the majority (Guelke, 1996: 525). In the Declaration, Britain stressed that it possessed no selfish interests in Northern Ireland and affirmed that it would “uphold the democratic wish of a greater number of people of Northern Ireland on the issue of whether they prefer to support the Union or a sovereign United Ireland” (Joint Declaration of Irish Taoiseach, 1993: 4). The Irish government, for its part, noted that “it would be wrong to attempt to impose a united Ireland, in the absence of a freely given consent of a majority of the people in Northern Ireland,” and accepted that: [the] democratic right of self-determination by the people of Ireland as a whole must be achieved and exercised with and subject to the agreement and consent of the majority of the people of Northern Ireland, and must, consistent with justice and equity, respect the democratic dignity and the civil rights of both communities. (Joint Declaration of Irish Taoiseach, 1993:5) While it was mandated that the political objectives must be met exclusively through peaceful and democratic means, the Declaration did hold out an open invitation to paramilitaries positioned outside of the legitimate dialogue. Paragraph 10 of the Declaration stated that: democratically mandated parties which establish a commitment to exclusively peaceful methods and which have shown that they abide by the democratic process, are free to participate fully in democratic politics and to join in dialogue in due course between the Governments and the political parties on the way ahead. (Joint Declaration of Irish Taoiseach, 1993: 9) The Declaration thus presented paramilitaries with an option: renounce violence and participate in the peace process or risk political marginalization. 76 Gerry Adams offered that while the Downing Street Declaration marked progress toward peace, it certainly did not present a viable solution. Its lack of clarity and “negative and contradictory elements,” reported a Deutsche Presse-Agentur article, “Sinn Fein Criticizes Downing Street Declaration,” made Adams highly wary of the Declaration (1). Thus, between the December 1992 and the end of July, 1993, there was no reduction of violence; in fact, 49 people were murdered during that period. Finally, on August 31, 1994, the IRA announced “a complete cessation of military activities.” In its words: We note that the Downing Street Declaration is not a solution, nor was it presented as such by its authors. A solution will only be found as a result of inclusive negotiations . . . It is our desire to significantly contribute to the creation of a climate which will encourage this. (Irish Republican Army Ceasefire Statement; Qtd. in CAIN) Republican splinter groups, including the Irish National Liberation Army (INLA) and Republican Sinn Fein (RSF), declined to participate in the cease-fire declared by the Combined Loyalist Military Command on behalf of Protestant paramilitaries (Harkness, 1996: 60). However, with the IRA agreeing to a cease-fire, genuine progress was still taking place; the desire to bring about change through political means instead of through violent methods indicated a hopeful departure from the many years of military stalemate. An Opportunity for Conflict Resolution According to William Zartman (1997), the success of mediation relies on the recognition and creation of a “ripe moment” (272). Zartman explains that three dimensions determine the suitability of a moment for conflict resolution: the intensity of 77 the conflict, the alternatives available to the parties, and the power relations among the parties (Zartman, 1997: 272). By some accounts Northern Ireland’s “ripe moment” came with the IRA cease-fire of August 1994. It was an explicit sign of willingness on behalf of Republican paramilitaries to engage in serious political dialogue – an essential ingredient that had been missing from the peace process since its slow beginning in the early 1980s (Elliott, 2002: 162). The negotiations offered little or no hope of yielding a political settlement that would end Republican paramilitary violence without Republican cooperation. The Republican decision to partake in negotiations toward a political settlement was driven by the existing military stalemate in Northern Ireland and was actively encouraged and reinforced by the Clinton administration (Zartman, 1991: 280). With Sinn Fein and the IRA on board, however, the stage was set for achieving meaningful advancement. British Prime Minister John Major echoed this sentiment when he stated following the IRA cease-fire that the situation in Northern Ireland offered the “best opportunity for peace in twenty-five years” (Elliott, 2002: 120). Other reports were not so sanguine, as demonstrated by an article that appeared four months later in The Irish Times. Recounting an assessment by Sir Hugh Annesley, the RUC Chief Constable, the piece announces that the “The IRA cease-fire has a 60 per cent chance of lasting until Easter.” He goes on to describe the limitations of the IRA decommissioning given that many of its explosives are home made. If not violating the letter of the cease- fire the IRA was certainly violating its spirit, as Ulster Unionists and UDP security 78 spokesman reported that, “the IRA was carrying out dummy runs and targeting individuals for possible assassination should the cease-fire collapse” (Moriarty, 1994: 8). The Military Stalemate and the Emergence of a Republican Political Strategy Neither Nationalists nor Unionists possessed the military or political resources necessary to guarantee fulfillment of their political aspirations. Whereas other political parties were engaged in the political process, Sinn Fein and the IRA remained politically marginalized because of their continued reliance on violence. Without the participation of Sinn Fein and the approval of the IRA, however, forging a meaningful peace agreement seemed hopeless. For the IRA, the stalemate depicted the malfunction of its military campaign: 25 years of violence had failed to achieve the unification of Ireland or force the withdrawal of British troops. Moreover, the continuous cycle of violence was undermining support for Republican paramilitaries within the Catholic community. The Catholic community – despite its Nationalist tendencies – “agreed with the need for a peaceful change in Northern Ireland and disapproved of the violent methods employed by the IRA and other Republican paramilitaries” (Elliott, 2002: 159). The stalemate thus left the IRA and Sinn Fein no choice but to turn a new strategy and a new forum for pursuing their goals. In the words of Richard McAuly, press secretary for Gerry Adams: Republicans had long accepted that this was a war that nobody could win. And in that context, what do you do? The obvious answer is you try to find some way of bringing the different enemies together, preferably around a table, and work out some way of negotiating a settlement. (Frontline, 1997) 79 The hunger strikes of 1980 and 1981 had attracted much attention and sympathy for the Republican cause. The politicization of the Republican movement began only after these hunger strikes. From a standing start, within three years, Sinn Fein was capturing over 13% of the vote in Northern Ireland (representing 40% of Northern Ireland’s Catholic vote) (Moloney, 2003: 189). According to Dingley, the political component of the Republican strategy accelerated in the second half of the 1980s both with the first-time decision to permit Sinn Fein representatives to take their seats in the Republic of Ireland’s parliament and with Gerry Adams’s secret negotiations with John Hume, head of the SDLP, regarding the future of Northern Ireland (Dingley, 2008, 247). With the IRA cease-fire, however, the transformation of Republican strategy was nearly complete, indicating that for the first time “political participation was more prominent than military campaigns” (Dingley, 2008: 249). The Clinton Administration as Mediator The Clinton Administration recognized and encouraged the changes taking place within the Republican movement, helping facilitate the transformation of Republican strategy. Having promised on the 1992 campaign trail to appoint a peace envoy to Northern Ireland, Clinton made good on his word (Tran, 1992: 6). Clinton’s bold entry into the political fray offered encouraging and hopeful new circumstances making change feel possible. Before Clinton had even taken office, the Herald Sun reported positive responses to the new administration: “IRA’s political wing says the election of Bill Clinton as US President could help break the deadlock in Northern Ireland” (Tran, 1993). 80 Not everyone expressed so much confidence; indeed, one commentator in the Mail on Sunday (London) characterized the Clinton Administration involvement as “a mission to meddle in British internal affairs” (Dobble, 1993). Nonetheless, on January 1994, not long after the election and seven months before the IRA declared its cease-fire, the Clinton Administration granted Gerry Adams, president of Sinn Fein, a one-day visa to address a peace conference at the Waldorf-Astoria in New York (Dingley, 2008: 251). In his own words, President Clinton explains, The NSC determined that we should grant the visa, because it would boost Adam’s leverage within Sinn Fein and the IRA, while increasing American influence with him. That was important because unless the IRA renounced violence and Sinn Fein became a part of the peace process, the Irish problem could not be resolved (Clinton, 2004: 580) President Clinton granted Gerry Adams a very narrow visa with the express purpose of coming to the conference in the hope that it would advance the peace process (Frontline, 1997). In Moloney’s words, Adams scrupulously honored the terms of the Clinton visa. There were no public references to the armed struggle, much to the disappointment of some of the hundreds of IRA supporters who had turned up to hear him speak. “I come here with a message of peace,” he told the crowds at the Sheraton hotel” (Moloney, 2003: 61). The decision to issue the visa was one of the Administration’s boldest foreign policy decisions, but also one of its politically safest. The White House made the decision against the advice of the State and Justice Departments, and in the face of significant resistance from the British government (O’Brien, 1993: 235). In My Life, Clinton writes: 81 The British were furious. They thought Adams was just a fast-talking deceiver who had no intention of giving up the violence that had included an attempt to assassinate Margret Thatcher and had already claimed the lives of thousands of British citizens, including innocent children, government officials, and a member of the royal family, Lord Mountbatten. . . For days John Major refused to take my phone calls. The British press was filled with articles and columns saying I had damaged the special relationship between our countries. (Clinton, 2004: 580) In spite of this formidable resistance, President Clinton explained in his book My Life, that he knew that the visa would prove extremely beneficial for the mediation process, thus encouraging him to ignore the criticism (Clinton, 2004: 580). Characteristic of this opposition was an article in the Times (London)—“US Visa for Adams Angers No. 10”— reporting that the Ulster Unionist MP, John T. Taylor, considered it “despicable . . . that the Clinton Administration had ‘given in’ to pressure from the Irish lobby in Washington” (Fletcher and Leathley, 1994: 1). Most importantly, granting the visa represented a clean break with the traditional “hands off” policy of earlier administrations, which viewed Northern Ireland as an internal British matter. America’s strong Irish-American lobby, however, provided plenty of support (Moloney, 2003: 437). While the initial visa was influential in its own right, it was part of a bigger plan to draw Sinn Fein into the political mainstream (Clinton, 2004: 580). The strategy connected Administration overtures to Sinn Fein to that organization’s commitment to the peace process. After the IRA’s cease-fire declaration, President Clinton allowed Adams into the United States once again (O’Brien, 1993: 249). This time The Washington Times, offered a venomous assessment, equating the act with ones by previous administrations in which 82 visas had been granted to Fidel Castro and Yasser Arafat, quipping, “Clearly, one man’s hero is another man’s terrorist” (B5). Days before the visit took place, The Daily Mirror featured a piece on page one—“Fury as Adams Gets Visa; Unionists Slam Clinton; Gerry Adams Granted New Visitors Visa for the United States.” A Clinton spokesman assured reporters that, “There will be no meetings at the White House or at other government departments until the cease-fire is restored.” (Dowdney, 1996: 1) According to the article, fund-raising would be banned as well. Nonetheless, in March, 1995 Adams was granted another visa permitting him to take a two-week tour of the United States, which included a personal visit with the president and fund raising opportunities netting $1.3 million for Sinn Fein (Moloney, 2003: 66). This second visit was longer than the first and included a warm reception from Senator Edward Kennedy, whose lobbying among Irish-Americans had been integral to the granting of the visa. Lashing out at the United States for allowing the visit, a DUP official asked, “What would America think if Britain gave a visa to friends of the Oklahoma bombers?” (Dowdney, 1996: 1) According to Clinton, the visas and the subsequent successes of Adam’s visits were crucial factors in securing the IRA’s cease-fire and the eventual participation of Sinn Fein in the peace process (Clinton, 2004: 579). First, the visits – which attracted a great deal of media attention – provided remarkable examples to Sinn Fein and the IRA of the political rewards that could be gained by becoming legitimate players in the political process. Second, they gave the IRA and Sinn Fein confidence and access to a much larger audience (O’Brien and O’Brien, 1981: 189). 83 Finally, receiving the visas convinced Sinn Fein and the IRA that the United States was dedicated to seeing the peace process through to a successful conclusion and was willing to take their views seriously (Clinton, 2004: 581). Speaking upon his arrival at New York’s Kennedy Airport, Gerry Adams exclaimed: There’s a clear signal that progressive public opinion in the US wants peace. This conference gives us an opportunity to move the peace process forward . . . For the last 20 years, US public opinion has been misinformed about Northern Ireland. Most of that news came from Britain. (McGoldrick, 1994: 6) As Adams saw it, the visa presented an opportunity to secure his political objectives. Although political participation, with its emphasis on process and compromise, meant that Republican aspirations would not be achieved in the short term, it offered the IRA a better chance for eventual success than the alternative strategy of continued violence and military stalemate (Townsend, 1988: 178). Mechanisms of Intervention The task of moving from the 1994 cease-fire to comprehensive all-inclusive political negotiations was daunting. The divide between Northern Ireland’s communities remained enormous (Moloney, 2003: 82). In February 1995, the Irish and British governments published A New Framework for Agreement, more commonly referred to as the Joint Framework Document. According to Moloney, it represented “a shared understanding reached between the [Irish and British governments] on the parameters of a possible outcome to the Talks” and was intended to give impetus and direction to the process (Moloney, 2003: 83). The Joint Framework Document proposed structuring 84 negotiations according to the three-strand formula of the Brooke-Mayhew Initiative and applying the principles espoused in the Joint Declaration--namely self-determination, democracy, and nonviolence (The Agreement, 1998: Strand 1, par. 1). The Joint Framework Document and the previous peace initiatives had moved the process along by establishing a workable framework for negotiations and by identifying fundamental principles that would have to be incorporated into any ultimate resolution. However, the initiatives had not created a solid foundation on which all-inclusive political talks could be based. Moloney explains that comprehensive negotiations had never occurred between the major political parties of Northern Ireland (Moloney, 2003: 86). Moreover, a number of troublesome issues needed resolution before the parties would even consider participating in negotiations. The most pressing was how and when to decommission the stockpiles of weapons amassed by Northern Ireland’s paramilitaries. The International Body of Decommissioning For both Unionists and Republicans, the decommissioning issue took on a symbolic significance. Concerned about the ease with which paramilitary cease-fires could be terminated, and unwilling to negotiate under the threat of violence, several political parties hinged their participation in multiparty negotiations on the decommissioning of weapons prior to the start of negotiations. The British also felt this way and therefore “turned paramilitary decommissioning into a prerequisite for the participation of political parties linked to paramilitary groups” (Bew and Gillespie, 1996: 85 91), including Sinn Fein as well as the Ulster Democratic Party, which is affiliated with the Ulster Freedom Fighters. The IRA, however, had been operating under the understanding that its cease-fire alone would be sufficient to guarantee Sinn Fein’s participation. It perceived this new demand as a delaying tactic and a sign of British bad faith (McKittrick, 1996: 208). The IRA thought that decommissioning before a final political settlement would represent defeat. In the words of Gerry Adams, “The British government was not simply interested in a gesture. It was, in reality, demanding the start of a surrender process as a precondition to multiparty talks” (Bew and Gillespie, 1996: 105). To break this deadlock, on November 28, 1995, the Irish and British governments issued a communiqué announcing the creation of an International Body to provide an independent assessment of the decommissioning issue (Report of the International Body, 1996: 2). The communiqué signified a severing of the decommissioning issue from the pre-negotiation process. From this point forward, decommissioning would be handled separately from the preliminary discussions concerning the basis, participation, structure, and agenda of multiparty negotiations. The International Body’s mandate concerned solely the narrow issue of decommissioning (Report of the International Body, 1996: 3). Specifically, the International Body was requested to do the following: • identify and advise on a suitable and acceptable method for full and verifiable decommissioning • report whether there was a clear commitment on the part of those in possession of such arms to work constructively to achieve full and verifiable decommissioning (Bew and Gillespie, 1996: 105). 86 Former US Senator George Mitchell, then serving as a US envoy to Northern Ireland for economic development, was asked to head the International Body. Mitchell was an extremely able and experienced negotiator (Dingley, 2008: 254). He was also joined on the commission by Harri Holkeri, former prime minister of Finland, and General John de Chastelain, a Canadian diplomat and the former head of the International Body. Mitchell was familiar with both the conflicts and the prominent figures involved in the process, including British Prime Minister John Major. Mitchell’s appointment helped to uphold US support for the peace process, and his reputation as a skilled mediator and politician with extensive first-hand knowledge of Irish politics made him an obvious choice. During his tenure as senator, it was said of him that, “There is not a man, woman or child in the Capitol who does not trust George Mitchell” (Spectrum Lecture Series). The International Body on Decommissioning was a traditional consultative body without authority to impose its views on the parties. The International Body’s strategy was to involve a neutral third party to facilitate problem-solving through communication and analysis—a formidable effort given that there were two governments and 10 political parties involved, many of whom refused to talk to each other. According to Bew and Gillespie, the communiqué provided the International Body with authority to determine its own procedures (Bew and Gillespie, 1996: 107). However, both the Irish and British governments requested the International Body to meet with the relevant parties and extract their views on decommissioning (Report of the International Body, 1996: 8). 87 Two series of meetings were held by the International Body: the first in Belfast and Dublin, December 15-18, 1995, and the second in London, January 11-22, 1996. During the course of these meetings, the International Body received oral and written submission from the various parties to the conflict, including testimony from government officials, political leaders, and church officials (Report of the International Body, 1996: 6). Hundreds of letters and telephone calls were also received from members of the general public (Report of the International Body, 1996: 6). This was a topic of great importance in Ireland and it therefore received a lot of attention. It became apparent that the conflict might finally be progressing towards a resolution and that the International Body was going to lead the way. Another significant development in the Northern Ireland question occurred in November and December of 1995, when President Clinton visited Northern Ireland and the Irish Republic. The President received what a New York Times reporter called a “hero’s welcome,” as he traveled around the North preaching reconciliation and injecting vital momentum into the peace process. Dingley explains the visit to the Province by an incumbent president was unprecedented (Dingley, 2008: 253). Unionists and Nationalists alike praised the President’s visit; and Clinton told those who would use violence that, “You are the past: your day is over” (Moloney, 2003: 390). President Clinton’s visit convinced the British and Irish governments to find a way of resolving their differences about decommissioning at least some terrorist arms through the so-called twin-track approach, whereby moves toward talks between the 88 political parties in Northern Ireland could take place at the same time as an international commission under the chairmanship of the American Senator George Mitchell. The Irish Taoiseach, John Burton, declared that the visit of the “most powerful politician in the world” would give the people of Northern Ireland and the politicians the “confidence that they need to make the compromises, to give the ground that is essential if the potential of the twin-track process is to be realized” (Moloney, 2003: 390). Both preparatory talks on the future of Northern Ireland, and the report of the Commission, would thus be completed in time for all party negotiations to begin in earnest by the end of February 1996. President Clinton was taking the lead in revitalizing the peace process. However, it was not long before the old realities surfaced once again. Gerry Adams was less than forthcoming in supporting the decommissioning body led by Senator Mitchell and the IRA made it clear that they were hostile to the whole idea, asserting that it was “ludicrous” to suggest decommissioning arms either through what it called the “front” or “back” door (Adams, 2005: 243). Adams sought to play down the meaning of the IRA statement by declaring that it was “reiterating and stating the obvious” (Adams, 2005: 243); the Reverend Ian Paisley for the Democratic Unionists sought to play up its significance by insisting that the “Clinton bubble, blown up by Burton and Major, has burst” (Cox, Guelke and Stephen, 2000: 254-55). The Ulster Unionist Deputy leader, John Taylor, warned that the IRA still wanted to keep its arms and that there was no change in their situation. For the SDLP, Dr. Joseph Hendron spoke of his disappointment at the statement, but claimed that, “all is not 89 lost” (Cox, Guelke and Stephen, 2000: 254-55). Equally important was the paramilitary attack by members of the IRA upon petty criminals in Roman Catholic areas, resulting in fatalities and intimidation, which dismayed the British government—but not to the point that they wished to acknowledge that the cease-fire was crumbling (Adams, 2005: 245). If these attacks had indeed been directed at members of the Protestant community, then the cease-fire would have been put in serious jeopardy (Adams, 2005: 245). These events took place against a background of increasing disarray in the British Conservative Party. The party remained deeply divided, and the tensions between Left and Right surfaced in the defection of the Conservative MP Emma Nicholson to the Liberal Democrats on December 19, 1995 (Adams, 2005: 386). This switch made the support of the nine Unionist MPs vital to the survival of a government with a working majority of five, and appeared to raise once again the possibility that an Irish political party would play a key role at Westminster, with all the consequences that would flow from Conservative dependence on Unionist votes. Adams asserts that Cardinal Cahal Daly, leader of Ireland’s Roman Catholics, warned that it would be very “unfortunate” if the Conservative’s shrinking majority at Westminster influenced the attempt for peace in Ulster (Adams, 2005: 246). New tempting possibilities arose for the Unionists when it seemed as if the Conservative Government was weakening and that a sudden general election was imminent. However, this was to overstate the possibility for a new lease of life for the Irish Question in British politics. The Ulster Unionists would be cautious about replacing a government party whose policy they were familiar with one whose statements on 90 Northern Ireland were hardly controversial but whose sympathies with the idea of the Union had not yet been put to the test. There was no forthcoming legislation that might provoke the Unionists to withdraw their support. The Ulster Unionists were inclined to take a cautious line and disclaim any possibilities of special deals from the Government (Ranelagh, 1995: 211). In any event, Major’s close contact with the former Unionist leader, James Molyneux Hardly, suggested that when key policy decisions on Northern Ireland were to be made, they would be substantially altered by perceptions of what might please Unionists most. Nevertheless, Unionists were, as their security spokesman, Ken Maginnis, put it, likely “to make the most of it” (Moloney, 2003: 99). These developments helped put the Clinton visit into perspective. It had encouraged the British and Irish Governments to obtain a reasonable agreement on the arms issue; but there were those in the IRA who would not let it discourage them. According to Mitchell, another result of the Clinton visit was the fact that the Northern Ireland political parties were able to have more peaceful negotiations, but they would not surrender what they regarded as their critical interests, and Sinn Fein was still hardly a model of a democratic and constitutional political party (Mitchell, 2001: 50). Each government had its own political position, and its own political culture, and the prospects of a weakening and perhaps disintegrating Conservative government in Westminster was bound to create unstable conditions from the point of view of Northern Irish political parties. Marchak stresses that the state of the parties at Westminster, the government’s need to steer a careful course through the conflicting wishes of the 91 Northern Ireland parties, and its need to keep a watchful eye on international, and especially American official opinion made the resolution of the question as complex and as difficult in 1996 as it had been when Prime Minister Gladstone introduced the first Home Rule Bill in 1886 (Marchak, 2008: 73). George Mitchell issued a report on the decommissioning process on January 24, 1996. The Mitchell Report stated that there was no hope of the IRA starting to decommission weapons before all-party talks took place. Mitchell offered a way around the issue: arms decommissioning could take place during all-party talks, providing that the participants agreed to six conditions before committing to the talks (Mitchell, 2001: 40). They included: 1) committing to a democratic and exclusively peaceful means of resolving political issues; 2) totally disarming all paramilitary organizations; 3) agreeing to verifiable disarmament to the satisfaction of an independent commission; 4) renouncing the use of force or the threat of force to influence the outcome of negotiations; 5) agreeing to abide by the terms of any settlement reached in all- party talks and; 6) ending punishment beatings (CAIN Web Service) Since the British government felt that this report undermined its policy of demanding decommissioning before talks, it responded to this setback with an announcement by Prime Minister Major that there would be elections for a new assembly in Northern Ireland. 37 While it is easy to attribute this political maneuvering to Ulster Unionist members in a House of Commons with a diminishing government majority, it is more likely to be a result of the Prime Minister’s much underrated political skills, and his ability to balance 37 A suggestion first put forward by the Ulster Unionists some time ago. 92 competing interests – experience honed in the great Conservative controversy over continuing membership of the European Union (Cox, Guelke and Stephen, 2000: 88). Northern Ireland Nationalists strongly opposed the dual approach to the next stage in the peace process. John Hume accused Major of “playing politics” with people’s lives (Mitchell, 2001: 89). However, Major’s balancing act was swept aside on February 9, 1996—some 17 days after the Mitchell report was published-- when the IRA ended the cease-fire by detonating a bomb in the Docklands area of London. Though some critics suggest that the return to violence was largely a reflection of the internal stresses and strains in the IRA military command, which was loathe to split over the question of peace and war, even if a renewal of violence might prove counterproductive (Cox, Guelke and Stephen, 2000: 89). According to Elliott, the Dublin government, in its initial reaction to the end of the cease-fire, almost surrendered to the idea that in some sense the British were responsible for the renewed crisis (Elliott, 2002: 165). Its anger and shock at the bombing in London on February 9, 1996 was real and profound. But the British political parties did not break ranks over Northern Ireland, with the partial exception of the marginalized former Northern Ireland shadow secretary, Kevin Macnamara (Elliott, 2002: 165). The opposition leader, Tony Blair, had continuously echoed John Major’s condemnation of the IRA, his criticism of Sinn Fein, and his wish to continue with the peace process. 93 The Report On January 22, 1996, the International Body issued its independent assessment of the decommissioning issue. Although the International Body found “a clear commitment on the part of those in possession of arms to work constructively to achieve full and verifiable decommissioning,” it concluded that this commitment did not extend to decommissioning prior to multiparty negotiations (Report of the International Body, 1996: 25-26). To address that reality, the International Body allowed a compromise position calling for decommissioning in phases during the negotiations. The International Body argued that decommissioning during the negotiations would contribute momentum to the peace process and could serve as a useful confidence- building measure (Report of the International Body, 1996: 35). Notwithstanding its compromise recommendation, the International Body understood the need to protect the negotiations from violence and the threat of violence. To achieve this objective, it recommended that all parties “affirm their total and absolute commitment” to six principles of democracy and nonviolence (Report of the International Body, 1996: 20). These six commitments, which later came to be known as the Mitchell Principles, connected political participation in multiparty negotiations to the renunciation of violence. These principles were created in order to ascertain that parties approached the negotiations in good faith. For the Mitchell Principles to be effective as a code of conduct, two conditions had to exist. First, the advantages from participating in the negotiations had to outweigh those presented by violence. If parties believed that they could gain more from continued 94 violence, there would be little incentive to participate in negotiations. Second, the Mitchell Principles had to be backed by a plausible threat of sanctions. In the Northern Ireland context, these preconditions were met. The prevailing military stalemate essentially nullified the military option and consequently enhanced the relative attractiveness of negotiations. Moreover, the threat of expulsion for violation of the Mitchell Principles was real as would subsequently become clear from Sinn Fein’s exclusion from the peace process (Mitchell, 2001: 98). In August, the International Body pushed for the creation of an independent commission to ensure that decommissioning would proceed in an orderly and acceptable manner. It was appointed by the British and Irish governments after the consultations with the parties, to supervise the process (Report of the International Body, 1996: 40). In order for it to perform adequately, the International Body argued that the independent commission would need to be authorized to act independently of the government and be given appropriate legal immunities. With respect to the method of decommissioning, the International Body recommended that the decommissioning process should (i) be applied in a manner that suggests neither victory nor defeat; (ii) be verifiable; (iii) result in complete destruction of weapons; and (iv) not expose individuals to prosecution (i.e., the weapons turned over during the decommissioning process should be immune from forensic examination). (Report of the International Body, 1996: 38-50) In an implied rejection of the International Body’s recommendation on the timing of decommissioning, the British government proposed another option to multiparty talks that was based on a confidence-building measure identified in the International Body’s 95 report (Report of the International Body, 1996: 56). The proposal called for an election in Northern Ireland to establish a constituent assembly from which teams of negotiators would be chosen to represent the Republican and Unionist parties. Under the British suggestion, the only purpose of the elections would be to supply an alternative to the prior decommissioning of weapons. The elected body would not possess legislative or executive powers (Report of the International Body, 1996: 56). In the short term, the British proposal was a major setback to the peace process, as well as to Anglo-Irish relations. Watt explains that the Irish government, which had cordially embraced the International Body’s recommendations, was caught off guard by the British action (Watt, 1996). Irish officials were enraged that they had not been previously consulted on the proposal and criticized the electoral scheme on the basis that it would provide the Unionist majority with a veto over political developments in Ireland (Mitchell, 2001: 4). This criticism was also shared by Northern Ireland’s Nationalist community (Watt, 1996). Because there are no public documents, the British government’s rationale for not implementing the International Body’s recommendations remains unknown. The electoral scheme would effectively evade the decommissioning issue – an obstacle that the British had helped to construct – by providing an alternative route to multiparty negotiations. To the extent that the proposal decoupled decommissioning and involvement in negotiations, the electoral scheme represented a clear win for Republicans and a concession by the British: it would grant the IRA a way into the negotiations without having to turn over any weapons. However, in a similar example of the level of 96 distrust and hostility between the parties to the conflict, Republicans cynically believed the British proposal was not a victory, but rather an obstacle to negotiations and further proof of British bad faith. On January 25, 1996 an article in The Irish Times described Sinn Fein as feeling “ambushed” by the new preconditions, which it felt “exposed a serious disregard on the British part for the Government” (Tynan, 1996: 12). In the words of one source, “It seems now that there is no hope of negotiations starting by the end of the February deadline. John Major has provided another device to ensure that all party talks won’t start.” (Tynan, 1996: 12). On February 9, 1996, 17 months into the cease-fire, the IRA was responsible for exploding a 100-pound bomb in the Docklands area east of London. The explosion caused $150 million in damages, killed 2 people and injured 43 more (Moloney, 2003: 441). By February 18 th , two additional bombings were staged by the IRA. In the weeks preceding the explosion, the IRA and Sinn Fein had repeatedly expressed frustration over the slow pace of the peace process. According to President Clinton, “The IRA was a tough nut to crack, full of hard men who had built a life on hatred of the British and the Ulster Unionists, and for whom the idea of peaceful coexistence and continuing to be a part of the UK was anathema” (Clinton, 2004: 579). The IRA had declared its cease-fire with the expectation that substantive negotiations would begin within three months. An IRA statement issued one hour prior to the Docklands blast echoed frustration: “the cessation [in hostilities] presented an historic challenge for everyone and the IRA commends the leadership of the nationalist Ireland at home and abroad. They rose to the challenge. The British Prime Minister did 97 not” (Moloney, 2003: 441). John Major responded with dismay but not discouragement; indeed, the bombing emboldened his commitment to decommissioning, as he asserted in the Times of London, “This atrocity confirms again the urgent need to remove illegal arms from the equation. For my part, I remain committed to the search for peace in Northern Ireland and will not be distracted by terrorism” (Watt, Wood, Pierce, and Fresco, 1996: 1) Likewise, Clinton denounced the violence in a phone call to Major in which he declared, “I condemn in the strongest possible terms this cowardly action and hope those responsible are swiftly brought to justice” (Watt, Wood, Pierce, and Fresco, 1996: 1). Notably, the International Body had labeled parallel decommissioning as a possible confidence-building measure. Nevertheless, the measure of the International Body’s role in the peace process is not the implementation of its recommendations for resolving the decommissioning issue, but its significant impact on the larger process (Elliott, 2002: 131). First, the intervention permitted the British government to elude the dual role, namely that of sponsor of and party to the peace process, that disabled the earlier Brooke-Mayhew initiative. Second, the International Body brought impartiality and legitimacy to the process. Third, consenting to the Mitchell Principles of democracy and nonviolence paved the way into the peace process and afforded the security needed for the success of the talks 98 International Involvement Throughout the 15-year process, one of the lowest points was the period immediately following publication of the International Body’s report. Hopes for a quick settlement were shattered throughout Ireland and the United Kingdom due to the deadly bombing in the Dockland’s. According to Elliott, accusations of blame for collapse of the negotiations were exchanged between the Irish and British governments (Elliott, 2002: 150). However, within several weeks, the situation began to move ahead when a compromise blueprint regarding the path to all party talks was worked out between the Irish and British governments. Elliott explains that it established a two step plan. First, it called for Northern Ireland’s major political parties to participate in Dayton-style proximity talks, in which parties would gather at a single location, but not face-to-face (put forward by the Irish government) (Elliott, 2002: 133). Rather negotiations would be conducted through intermediaries – in this case, representatives of the Irish and British governments. It was the hope of the Irish government that proximity talks would restore confidence in the overall process. The purpose of the proximity talks was to figure out the procedures for step two, the Northern Ireland elections called for by the British. Under the blueprint, the elections would “lead directly to all party talks, which were scheduled to begin on June 10, 1996” (Elliott, 2002: 138). Delegates from Northern Ireland’s political parties gathered in Belfast on June 10, 1996 to open negotiations. There was a failure to assemble all of Northern Ireland’s major political parties, as Sinn Fein was denied entrance to the talks due to the IRA’s continuing campaign of violence. Despite the provocation planned by the IRA, the 99 Unionist cease-fire had remained in effect. Consequently, the UDP and the Progressive Unionist Party, both of which had affiliations with Unionist paramilitaries, were allowed to participate in the talks (Graham, 1996). From the first day, the discussion was mired in deep acrimonious debate (Graham, 1996). Led by David Trimble, leader of the UUP, the Unionists bitterly opposed Senator Mitchell’s appointment as independent chairman of the Plenary Session, arguing that he was too closely tied to the Irish-American lobby to be impartial (Graham, 1996). On June 12, after two days of intense negotiations, the UUP split from their fellow Unionists, dropped their objections, and allowed Senator Mitchell to assume the post of independent chairman. With that initial crisis behind them, the parties (Alliance, Labour, NIWC, PUP, SDLP, UDP, and UUP) formally declared their commitment to the International Body’s principles of democracy and nonviolence (Graham, 1996). In a 2002 interview, Mitchell reflected upon his role at that period during the negotiations as well as the responses of the various disputants in the following way: I tried hard to establish an atmosphere in which there could be full and fine discussion, but to keep it civil. I was only partially successful. For one thing, never once . . . have I been able to get all the parties in the same room at the same time. Secondly, there were a lot of insults… So I tried using my experience in the Senate and as a Federal Judge to create a situation where they felt they could have a civil discussion . . . I never said “this is the way we’re going to talk.” I tried to create an atmosphere that was conducive to genuine, frank discussion. (Qtd. In Jacobsen, 2003: 201). However, progress throughout the opening Plenary Session was extremely slow. After agreeing on revised procedural rules in late June 1996, the negotiations derailed over the priority to be given to the issue of decommissioning on the agenda. The UUP 100 restated its desire for decommissioning of paramilitary weapons prior to the start of substantive negotiations. The SDLP opposed the UUP plan, arguing that the parties should carry on with parallel decommissioning as called for in the International Body’s Report (Moloney, 2003: 457). After eight months of stalemate over the issue of decommissioning, the opening Plenary session was suspended pending the approaching British general election and local elections in Northern Ireland (Moloney, 2000: 457-58). The British elections would prove to be a turning point in the negotiation process. After eight months of stagnation, another chance at progress was created by the British general elections, held on May 1, 1997 (Moloney, 2003: 458). British voters decisively elected to replace the paralyzed Conservative government of John Major with the vibrant Labour government of Tony Blair and two Sinn Fein MPs were elected to British Parliament. A Dublin government spokesman responded to the elections with the following statement, “We could be on the brink of something historic. There are strong grounds for optimism,” while a top IRA source explained, “Violence has paid up until now, but we feel that, if we finally lay down our arms, Tony Blair will make Northern Ireland his Number One priority. We feel we can do business with him” (Klerans and Ryan, 1997: 1). It had become increasingly clear that John Major’s political party lacked the political support necessary for active leadership of the process. Tony Blair’s Labour government was free to kick start the negotiations because unlike its predecessor, it enjoyed broad support and was not indebted to Northern Ireland’s Unionists. In order to break the deadlock over decommissioning, Blair initiated efforts to pull Sinn Fein back into the political process. In early June, the British government took the controversial 101 step of extending an offer to Sinn Fein and the IRA – Sinn Fein would be allowed to participate in negotiations if the IRA declared a “genuine” cease-fire and observed it for six weeks (Moloney, 2003: 455). Following public disclosure of the offer, Northern Ireland’s Unionists demanded that it be withdrawn, promising to walk out of negotiations if Sinn Fein representatives were allowed to participate. Nevertheless, the Blair government left its offer intact and on July 19, after a joint appeal from Gerry Adams (Sinn Fein) and John Hume (SDLP), the IRA accepted the offer and renewed its cease- fire (Moloney, 2003: 455). Despite the lack of consensus among the political parties of Northern Ireland an agreement was signed on August 27, 1997, establishing the independent International Commission on Decommissioning (Cox, Guelke and Stephen, 2000: 266). Two days later, with the preconditions expressed by Tony Blair nearly met, Sinn Fein was formerly invited to participate in the substantive “strand” negotiations. On September 15, 1997, Sinn Fein negotiators guaranteed their adherence to the Mitchell Principles and took their seats at the negotiations. Richard English explains that this commitment in itself implied that they were prepared to compromise. Sinn Fein’s leading figure observed that “negotiations are negotiations, you can’t go in and dictate them and have a take it or leave it position. So we have to go in and listen” (English, 2003: 296). The first item for discussion was an indictment by the Ulster Unionist Party (UUP) calling for the expulsion of Sinn Fein from the peace process for defiance of the Mitchell Principles. Although it was clear that the governments would not expel Sinn Fein after so much effort was put into including them in the process, the indictment was 102 leveled to provide the UUP with political cover (Cox, Guelke and Stephen, 2000: 270). Following rejection of the UUP’s challenge and the beginning of negotiations, the UUP could claim that it was forced into negotiations. The indictment to reject Sinn Fein was rejected and on September 25, the political parties of Northern Ireland consented to address the decommissioning issue alongside the strand of negotiations and proceed to substantive talks (Cox, Guelke and Stephen, 2000: 270). The UUP’s strategy was to negotiate with all of the other parties involved in the process except Sinn Fein. This strategy, which was pursued throughout the remainder of the process, was feasible because Sinn Fein’s vote was not required to reach sufficient consensus under the peace process’s procedural rules. As the three-strand negotiations proceeded throughout 1997 and into 1998, a surge in violence put the peace process back in doubt. Between December 22, 1997 and January 20, 1998, Loyalist extremists detonated three car bombs in Northern Ireland. Cox, Guelke and Stephen explain that in possible retaliation for the first of these attacks, the Irish National Liberation Army, a Republican paramilitary group not bound by the IRA cease-fire, murdered the leader of the Loyalist Volunteer Force in the Maze prison in Belfast (Cox, Guelke and Stephen, 2000: 266). These and other incidents had a direct impact on the faltering negotiations. On January 26, 1998, the British and Irish governments expelled the UDP from negotiations, after the Ulster Freedom Fighters – a Unionist paramilitary group with whom the UDP was affiliated – was tied to a string of sectarian killings. At the time of the UDP’s expulsion, the government made clear that the party would be allowed to reenter the talks after an unspecified period of time (which 103 turned out to be four weeks) if it remained committed to the Mitchell Principles. In late February 1998, Sinn Fein was “expelled from the negotiations for two weeks in response to IRA killings of a Loyalist paramilitary and a punishment killing of a Catholic drug dealer” (Cox, Guelke and Stephen, 2000: 267). According to Senator Mitchell, enforcement of the principles – though necessary – came at considerable cost to the peace process in terms of party distraction and lost negotiation time (Mitchell, 2001: 60). When Clinton began planning another trip in April, 1998, Unionist politicians considered it poor timing. With a referendum upcoming, they felt that Clinton’s presence could complicate the democratic process. According to an article in Belfast News Letter, Politicians Split Over Clinton Trip Timing,”(1998) the UDP secretary saw Clinton’s previous work as a past facilitator for nationalism meant that he could not be an honest broker in the peace process . Still, according to Dingley, Clinton’s visit on September 3, 1998, undoubtedly helped to mobilize more popular support for the peace process in Northern Ireland and the street demonstrations against the IRA for breaking its cease-fire played a part in pushing the IRA to restore the cease-fire and pursue the peace process the public so yearned for (Dingley, 1995: 254). In Armed Struggle: The History of the IRA, Richard English, an historian from Northern Ireland, explains that President Clinton was “an important factor” in the political movement towards the Good Friday Agreement (English, 2003: 304). “Here was someone involved sympathetically, who was more powerful than the British, who might act against British preferences (as with the 1994 Adams visa) and who might serve 104 as a kind of guarantor of fair dealing, an international referee with muscle” (English, 2003: 304). In 1998, Gerry Adams issued the following statement on the eve of President Clinton’s second visit to Northern Ireland: Sinn Fein is committed to exclusively peaceful and democratic means to achieve a way forward. We have to work politically to make the Omagh bombing the last violent incident in our country, the last incident of this kind. We are committed to making conflict a thing of the past. There is a shared responsibility to removing the causes and to achieving an end to all conflict. Sinn Fein believe the violence we have seen must be for all of us now a thing of the past, over, done with and gone. (Adams, 2005: 368) Due to the increased threat that the violence would deter the peace process, Mitchell decided to move up the deadline for concluding the peace process from May to April 9, 1998, Good Friday. This date, which marks the start of the Easter holiday weekend, is important to both of Northern Ireland’s communities and was strategically chosen by the independent chairman to strengthen the talks (Clinton, 2008: 785). Negotiations intensified as the April 9 deadline approached. The independent chairman presented the parties with a draft agreement representing the culmination of all three strands of negotiations. As the political parties and governments focused their efforts on the draft document, tensions flared and rumors circulated that the UUP and Sinn Fein were close to walking out on the negotiations (Cox, Guelke and Stephen, 2000: 273). While the April 9 deadline passed without an agreement, the parties and governments continued to negotiate for a settlement. Excruciatingly close to an agreement, the talks were again almost derailed by the decommissioning issue (Clinton, 2008: 786). A senior member of the UUP informed David Trimble that they could not 105 accept an agreement that would allow Sinn Fein into a government without the IRA having decommissioned a single bullet (Cox, Guelke and Stephen, 2000: 272). The threat was averted by assurances from Tony Blair, who provided political cover for David Trimble and the UUP. Prime Minister Blair assured the UUP that if the Agreement proved ineffective in dealing with office holders no longer committed to peaceful means, the British government would support changes to make it so (Breadun, 1998). Blair also reaffirmed that he agreed with the UUP’s view that decommissioning measures should begin immediately upon implementation of the Agreement. With these final concerns addressed, the parties presented the Agreement to the people of Ireland for their acceptance in separate referendums, North and South (English, 2003: 301). Tony Blair was crucial to securing Unionist support. He had helped to shift his party’s policy on Northern Ireland away from a commitment to Irish unity and towards a respect for northern consent. In his words, he believed that “the important thing is not that the government takes up the role of pushing people in one direction or another, but that they allow the wishes of those in Northern Ireland to be paramount” (English, 2003: 301). On May 22, 1998, Blair made public pledges and pleas trying to reassure nervous unionist opinion: I believe the Agreement can work because it is just and it is based on principle. The principle of consent is clear – there can be no change in the status of Northern Ireland without the express consent of the people here. . . There can be no accelerated prisoner releases unless the organizations and individuals concerned have clearly given up violence for good – and there is no amnesty in any event. Representatives of parties intimately linked to paramilitary groups can only be in a future Northern Ireland government if it is clear that there will be no more violence and the threat of violence has gone. That doesn’t just mean 106 decommissioning, but all bombings, killings, beatings, and an end to targeting, recruiting and all the structures of terrorism. (English, 2003: 301) The people of Ireland overwhelmingly accepted the Agreement. In the Republic, almost 95% of the votes were in favor of the Agreement. In the north – where a record of 81% of the voters turned out to register their opinion – 71% cast their votes in favor of the Agreement (Breadun, 1998). Taylor Branch writes that, “the president said this milestone was the equivalent of our adoption of the 1787 Constitution. Now they must breathe life into its new government against die-hard resistance from both extremes” (Branch, 2009: 501). However, this large margin of victory did little to ease concerns about Northern Ireland’s future as nearly half of the province’s Unionist voters chose to reject the Agreement. Branch explains that “Progress in Northern Ireland came slowly and in fits – maddening and inspirational, silly and mundane – leaving Clinton upbeat but tentative about its long term miracle” (Branch, 2009: 604). Overview of the Agreement The Agreement presented to the people of Ireland is in essence a joint awareness on the principles and processes that will establish Northern Ireland’s political future. Under the Agreement, Northern Ireland’s political status as part of the United Kingdom remains unchanged (Adams, 2005: 328). However, the two governments and the political parties of Northern Ireland agree that any future change to the status of Northern Ireland must be based on the democratically expressed wish of the majority of the people residing there (Adams, 2005: 328). In accordance with these principles, both the British 107 and Irish governments relinquished claim to all of Ireland: the British by repealing the Government of Ireland Act of 1920 and the Irish government by modifying Articles 2 and 3 of its constitution (The Agreement, 1998: strand 1, par. 5). Under the Agreement, local government is restored to the province. The Agreement calls for the establishment of a 108 member Northern Ireland Assembly to exercise legislative authority on local matters (i.e., those matters previously within the competency of the United Kingdom’s six Northern Ireland Government Departments). “Executive authority will be held by a first minister, a deputy first minister, and additional ministers with the departmental responsibilities, with all ministerial posts allocated in proportion to party strengths. Legal and institutional safeguards will be employed to ensure that all sections of the community can take part and work together successfully, and that all sections of the community are protected” (The Agreement, 1998: strand 1, par. 5). Negotiations dealing with the relationship between Northern Ireland and the Republic resulted in the establishment of a North/South Ministerial Council intended to develop consultation, cooperation and action within the island of Ireland. This measure included thorough implementation on an all-island and cross-border basis – on matters of joint interest within the competence of the administrations, North and South (The Agreement, 1998: strand 2, par. 1). At the international level, the Agreement calls for a new British-Irish Agreement to replace the Anglo-Irish Agreement of 1985. According to this new agreement, both governments would have to honor the political aspirations of the majority of citizens in 108 Northern Ireland. Two new institutions would also be created: A British Irish Council will be comprised of representatives of the Irish and British governments, including the devolved institutions in Northern Ireland, Scotland and Wales, aiming to “promote the harmonious and mutually beneficial development of the totality of relationships among the people of these islands” (The Agreement, 1998: strand 3, par. 1). Under the terms of the Agreement, the British-Irish Intergovernmental Conference would subsume both the Anglo-Irish Intergovernmental Council and the Intergovernmental Conference established under the Anglo-Irish Agreement. Its mission was to promote bilateral cooperation on matters of mutual interest. In recognition of the Irish government’s special interest in Northern Ireland, it would continue to be allowed to put forward views and proposals on non-devolved Northern Ireland matters (i.e., those matters that are not within the competency of the Northern Ireland Assembly) (English, 2003: 305). The issue of decommissioning was one of the weaknesses of the Northern Ireland Peace Agreement. The Agreement provided a two-year deadline for decommissioning, but failed to establish a solid time line for disarming paramilitaries, or a set of detailed procedures of how to do so. Therefore, in some sense the success of the Agreement was determined by the actions of paramilitary groups not directly bound by the text. Reflecting on the tenuous conditions of decommissioning in an interview several years later, Mitchell acknowledged, “Well, I felt that they were the best that could be done under the circumstances” (Jacobson, 2003: 210). Not long after the adoption of the 109 agreement, representatives from the IRA continuously expressed their unwillingness to undertake even minimal disarmament (Adams, 2005: 370). The Agreement was without a doubt a historic moment in Ireland’s history—one for which Hume and Trimble received the Nobel peace prize-- and was hailed by the media and political world alike. The long, slow, and complex journey to the cessation had been watched with a mixture of fascination and hope by millions in Ireland and abroad (Moloney, 2003: 429). International Support for the Peace Process By most accounts, Clinton’s role in the peace negotiations was a resounding success and a feather in the cap of his diplomatic legacy. Concurrent as these events were with the heated Al Gore-George W. Bush election battle, Clinton’s work perhaps did not receive as much media attention as it should have, given the formidable accomplishment. The Belfast Agreement took place at the end of Clinton’s presidency and, as such, become part of media sentimentality about the end of an era. Where The Boston Globe and New York Daily News expressed an admixture of commendation and nostalgia, relief and deep admiration, The Washington Post, a more right-leaning publication, saw Clinton’s work and future role in Ireland differently (Markey, 2000: 11). "President Clinton is in Northern Ireland today for a 'thank you' tour. Unfortunately, the peace process he has sponsored there, like so many others, is currently in shambles" (Qtd. In Markey 2000: 11). 110 In this third and final “farewell” visit—on December 12, 2000— Clinton visited Dublin to make his final plea for Protestants and Catholics to come together for common cause. With an unusually informal and almost jocular manner, Clinton addressed a crowd at the Guinness Beer Brewery. His speech touched upon his particular affection for the Irish people and his own Irish ancestry, and even gestures toward an account for what motivated his passionate interest in the Northern Ireland, quipping: “Maybe it's because there are 45 million Irish-Americans, and I was trying to make a few votes at home” (Shinsho, 2000). Despite the obvious attempt at humor, Clinton’s comment reveals what many scholars and critics take quite seriously: the President’s political self-interest in furthering the peace process in Northern Ireland. Had he not felt a connection through personal and national identity, it is unlikely, by his own admission, that he would have persisted and followed through as he did. He, indeed, left behind a highly regarded legacy of peace, and certainly braved the odds in his deep commitment as he himself remarks, “when I started my involvement with the Irish peace process, to put it charitably, half of the political experts in my county thought that I lost my mind” (Shinsho, 2000). The enduring stalemate in Northern Ireland had presented a seemingly insurmountable challenge, one that Clinton never failed to recognize. As reported in a Belfast News Letter article, “Peace: It’s a People Thing,” he told a crowd at Dundalk on December 13, 2000, "You cannot win by making your neighbour lose." Here Clinton unintentionally and yet poignantly evinces the importance of the Prisoner’s Dilemma to 111 the mediation process. Both sides must feel they have achieved something substantial, something more alluring than perpetuating the stalemate of “the Troubles.” Around the time of Clinton’s final visit, the overt praise of op-ed piece in the The Irish Times reflected the general sentiment of the Irish people and the Irish media. Appreciative but measured, the author explains that “the peace process in Northern Ireland is seen as one, perhaps the only one, of Bill Clinton's foreign policy successes,” describing the president as a “nanny” whose interest in Ireland dates back to his days at Oxford when the conflicts between Derry and Belfast were heating up (Holland, 2000: 18). As President, the author explains, Clinton brought his unrivaled power to bear on the negotiations, citing “his ability to pick up the phone to harass or cajole each of the parties to the conflict to take another step forward” (Holland, 2000: 18). Notably, commentators and journalists at home and abroad recognized that personal interest and significant political clout had together established the grounds for success. Admiration was not unanimous, however, as Clinton faced opposition from hard-line Protestant politicians even on his final visit. Clinton’s refusal to meet with Democratic Unionists raised the ire of Ian Paisley, Jr., who verbally attacked Clinton for including Sinn Fein in Northern Ireland’s coalition government, accusing him of “encouraging terrorism” (Freeman, 2000: A16). On this visit, Clinton basked in the praise and unflappably accepted the criticism. When Clinton again visited after his second term had ended, Ireland feted him by holding numerous galas and a fundraiser for the Northern Ireland Fund for Reconciliation, established a WJ Clinton scholarship, granted him an honorary doctorate from Belfast University, and established a William Jefferson Clinton 112 International Peace Centre and William Jefferson Clinton Centre in American Studies. Reflecting on his work, and again uncannily echoing the logic of the Prisoner’s Dilemma, Clinton defined the solution as one, "in which, in order for me to win, you have to win too" (McKittrick, 2001: 12). The steady support of the Northern Ireland peace process by the Clinton Administration merits considerable notice. US support was a positive and valued influence on the process and the participants. In 2003 John Bruton, Taoiseach of Ireland, reflected on the peace process as aided by George Mitchell, expressing that the Senator “was very helpful in finding formulae that would enable people to move forward on the political front” (Jacobsen, 2003: 250). The Administration adopted a facilitating role involving “passive” economic support for Northern Ireland and “active” encouragement and support for parties committed to the peaceful resolution of the conflict. US economic aid to Northern Ireland was meant to promote peace by ameliorating the economic conditions that exacerbated the conflict, namely discrimination, wealth disparity, and unemployment. While the US also sponsored trade missions and conferences to promote US foreign direct investment in Northern Ireland, it also supplied direct funding through the International Fund for Ireland. With an annual budget of $10-15 million, O’Brien explains that the Fund made capital available for projects that both supported economic development and brought members of Northern Ireland’s two communities together (O’Brien, 1993: 343). The Clinton Administration’s active support and encouragement of the peace process was the foremost component of US facilitation. His 1998 visit marked the first 113 time a US president undertook a trip to the province while in office and the audience that received him in Northern Ireland was one of the largest of any of his presidential trips. In My Life, Bill Clinton explains that he: flew to Belfast as the first American President ever to visit Northern Ireland. It was the beginning of two of the best days of my presidency. On the road in from the airport, there were people waving flags and thanking me for working for peace (686). The agreement was a fine piece of work, calling for majority rule and minority rights; shared political decision making and shared economic benefits; continued ties to the United Kingdom and new ties to Ireland. . . My first visa to Adams and the subsequent intense engagement of the White House made a difference, and George Mitchell handled the negotiations brilliantly. (Clinton, 2004: 784) According to Ed, the Clinton White House had its own reasons for getting involved in the Northern Ireland peace process. It was popular with Irish-American voters, a not inconsiderable factor, given that Clinton would run for the presidency again in 1996. Northern Ireland also “offered Clinton the prospect of at least one foreign policy success during his time in the White House” (Moloney, 2003: 424). Moreover, English explains that the reason for President Clinton’s active involvement in the politics of the north of Ireland is more likely to be found in the complexities “of his own political trajectory and interests – such as his response to a new kind of Irish-American lobbying, or his desire to be associated with international success stories in conflict resolution – than in the macropolitics of interstate power relations” (English, 2003: 307). These motives could have served as the force behind the continued support of the Clinton administration. The administration’s continued support for the peace process extended throughout the negotiations and was felt in their final hours. In the early 114 morning hours of the April 10 deadline, as the parties struggled to hammer out the final details of their historic agreement, President Clinton made personal phone calls to the various party leaders urging them to go forward with the agreement (Clinton, 2004: 784). Senator Mitchell explained that these phone calls were not cold calls – President Clinton, through his unrelenting allegiance to the process, knew the parties, knew the issues, and the individuals. Each of these aspects added considerable weight to his last minute appeals for peace (Mitchell, 2001: 113). The Role of the European Union The European Union’s influence in Northern Ireland derives from the conflict’s presence within its political and economic framework. The matters at the heart of the conflict – territorial sovereignty and self-determination – in fact run against the established European trends of economic and political integration. Although it possesses the political resources to influence the conflict, the EU as an institution has been disinclined to undertake continual and significant action. Both the Council of Ministers and the European Commission embraced the view that the conflict was an internal affair of the United Kingdom and that direct intervention by the EU could endanger Europe’s move toward regional assimilation. However, though the EU did not apply direct influence in the peace process, it had indirectly influenced events in Northern Ireland by providing a political forum for the British and Irish governments and economic aid to the region. 115 While European integration did little to diffuse the conflicting positions of Northern Ireland’s communities, it clearly impacted relations between the British and Irish governments. By providing the United Kingdom and the Republic of Ireland with a neutral political forum, the EU played a role in improving relations between the two governments. As far as the EU is concerned, the government representatives of the United Kingdom and the Republic are formally equals, a dramatic departure from the traditional perception of Irish subordination. Moreover, the EU provided representatives of both countries with the opportunity to work together and build relations of trust and understanding. Moloney asserts that the Republic of Ireland’s voice in the formulation of British policy in Northern Ireland, as effectuated through the Intergovernmental Conference and Joint Secretariat established under the 1985 Anglo-Irish Agreement, represented a degree of joint sovereignty over Northern Ireland that would have been unthinkable outside the context of European integration (Moloney, 2003: 199). The EU, like the United States, was also a source of economic assistance for the province. The EU’s Peace and Reconciliation Fund for Northern Ireland was budgeted to provide Northern Ireland with $550 million between 1995 and 1999 (Moloney, 2003: 201). Like its US counterpart, the EU economic assistance program was intended to ease the economic conditions that aggravated the conflict. The EU played a subtler role as well. For much of the 1990s, the Republic of Ireland prospered financially as a member of the EU. Jobs were created at a strong rate, including a thriving technology sector, and young people who might at one time have left Ireland, stayed there to live and work. This prosperity in Ireland was tied at many levels to its participation in the EU and was in 116 stark contrast to the high levels of unemployment and relative poverty of Northern Ireland (when compared to Britain or the Republic of Ireland). The conflict of Northern Ireland became increasingly anachronistic in an era of a strong Europe, as the opportunity costs of the conflict in terms of productivity and prosperity became abundantly clear. A united Northern Ireland, even with ambiguous sovereignty, stood a greater chance of benefiting from European economic growth. Once the mediation process is understood, it becomes possible to apply the Prisoner’s Dilemma game. This chapter explained the role of the Clinton administration and the many different tactics that were used to get the mediation process going. Other factors that heavily influenced the mediation process and the decisions of the parties include the 1994 cease-fire, the emergence of the Republican political strategy, and the report by the International Body of Decommissioning. Once these critical elements are understood, it now becomes possible to lay out the Prisoner’s Dilemma game and better analyze how the mediation process in Northern Ireland managed to succeed. 117 Figure 1.2: Map of Rwanda Source: http://www.historywiz.com/rwanda.htm 118 CHAPTER SIX: HISTORICAL CONTEXT OF THE CONFLICT IN RWANDA 38 The effort to document and make sense of the history of Rwandan genocide reflects the multiple, contradictory, and complexly motivated parties to the conflict itself. That recreating a timeline or coherent account of what lead up to the horrific events presents such a formidable challenge doubtless involves many of the same problems that stymied mediation efforts: too many disputants, too many interests, unclear identities and loyalties. Theories on the cause of the genocide are legion. How to make sense of the enormity of the slaughter in a post-Holocaust era of “never again”? How to bring the innumerable participants to justice? How to explain the international community’s passive response to the culminating horror? Why did the promise of the Arusha accords fail so bitterly? Racism, colonialism, hatred, power disparity, collaboration, and negligence all serve as partial explanations of why the peace accords failed, why genocide occurred, and why even now the history of Rwanda is so difficult to recount. Signed by the Rwandan government and Rwandan exiles on August 4, 1993, the Arusha Peace Accords appeared to offer an end to the long history of violent conflict in Rwanda. But the "agreement" was not accepted by all disputants and the mediator(s) had little or no leverage to enforce them—though both sides requested the presence of United Nations peacekeepers. The end result was the massacre of up to one million Tutsis in just 38 It is important to note that the historic information on Rwanda is controversial. This includes considerable ignorance about pre-colonial times, and also differences in perspective among the various African groups, nations, western observers and colonial governments' histories. 119 a few months in 1994. To understand the negotiations and their failure, it is essential to understand the history of Rwanda, as far as it can be known. In reflecting upon Rwanda, it is helpful to recall what Carl Becker said about the nature of historian’s work. As Becker (1955) notes, there are differences in the choices historians make as to what "facts" are relevant, and in Rwanda there are significant differences between the various reports and many "facts" are still very much in dispute. As to the cause of the genocide, historians fall into several distinct camps, exemplifying in a very real sense Becker’s problematization of history as a “clear, persistent outline” (1955:328). Rwanda’s highly contested history is anything but clear, but the effort to detail and explicate it has been undertaken by scholars with vastly different backgrounds, disciplines, motivations, and theories. As such, a brief review of the ideological camps is warranted. Prominent scholars include Philip Gourevitch, author of We Wish to Inform You That Tomorrow We Will Be Killed With Our Families: Stories From Rwanda (Farrar, Straus and Giroux, 1998); Mahmood Mamdani, a political scientist at Columbia University and author of When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda (Princeton University Press, 2001); David and Catharine Newbury, a historian and a political scientist who both teach at Smith College and have published widely on Rwanda (and lived there as well); Filip Reyntjens, a Belgian scholar and author of The War of the Great Lakes (Cambridge University Press, 2002) and American historian Alison Des Forges, a consultant for Human Rights Watch/Africa. According to many of her peers, Des Forges’s work is the most respected—perhaps because of its 120 extensive archival work (she had eight researchers with her)— despite her unwillingness to assert one conclusive cause for the genocide. Notably, what critics and journalists (and even observers and participants) knew at the time of the negotiations would greatly differ from the revelations exposed after the genocide. The Mucyo Commission report of 2008 and the Gacaca courts, which began in 2004 began the painstaking process of bringing perpetrators to justice. The Mucyo Commission, which is discussed in further detail in Chapter Seven, among other things took aim at the international negligence such as that of France. Rwandan courts had to bring the individual genocide suspects to justice and thus drew upon a traditional system of justice known as “gacaca.” The Global Policy Forum explains that, Gacaca hearings “are traditionally held outdoors (the word loosely translates as "justice on the grass"), with household heads serving as judges in the resolution of community disputes. The system is based on voluntary confessions and apologies by wrongdoers” (Rinaldo, 2004). To discuss Rwanda’s history and genocide is to negotiate with the vastly different agendas of these respective figures (and many others), to wade through the conflicting viewpoints and interpretations not in search of a clear cause and effect but rather to recognize the complexity of the narrative and to gain insight about how the innumerable convolutions would inevitably become a formidable obstacle to mediation. Along the way, scholars stake out territory, pit area studies against transdisciplinary work, stage direct confrontations between history and theory, and perpetuate or challenge the ethnic mythology. As one example, Gerard Prunier, author of The Rwanda Crisis: History of a Genocide (1998: xii), takes aim at ingrained beliefs about the continent: “Africa is a place 121 of darkness, where furious savages clobber each other on the head to assuage their dark ancestral bloodlusts.” He thus seeks to “dispel this feeling, (which sneaks even into the recesses of many ‘liberal minds’” and, as even a brief survey reveals, became the default argument of much of the Western media. Background Prior to the genocide, the vast majority of Hutu and Tutsi lived side-by-side, dispersed throughout the country (Melvern, 2000: 170). Nonetheless, the high poverty rate among the Hutu and history of tensions between Hutu and Tutsi, made the majority of Hutu peasants susceptible to manipulation by a small band of Hutu elites willing to destroy Rwanda in a desperate bid to protect their power and economic privilege. As Melvern explains, the Hutu and Tutsi ethnic classifications have a plethora of historical antecedents, including colonial classifications, clan affiliation, occupational status, “feudal” relations, and ancestral origin (Melvern, 2000: 9). As in many enduring conflicts, a range of pre-genocidal beliefs about ethnicity and identity played a considerable role in provoking discord among the parties. Although by the middle of the eighteenth century the Tutsi dominated the local power structure, they themselves had been assimilated by the Hutu and the ethnic classifications were highly flexible. The Tutsi adopted the Hutu language, Kinyarwanda, and incorporated many of their traditions and rituals (Melvern, 2000: 7). During this period, the Hutu/Tutsi classification broke down according to occupational status and wealth. Cattle-herders, soldiers and administrators were generally Tutsi, and the majority 122 of the farmers were Hutu (Sellstrom & Wohlgemuth, 1996: 21-23). According to Melvern, “the word Hutu means subject or servant and the word Tutsi means those rich in cattle” (Melvern: 2000: 8). As symbols of wealth, power and good breeding, cattle provided a form of social mobility in which a Hutu, and his lineage, could become “Tutsified” if he were able to acquire a sizeable enough herd. Conversely, a Tutsi who lost all his wealth and needed to cultivate the land would be “Hutufied” (Prunier, 1995: 11-12). Clans, which were multi-class and multi-ethnic, were arguably a more central form of identification at this time than “ethnicity.” Once the Tutsi extended their control in the nineteenth century, “ethnic” classification began to play a more important role. 39 From the fifteenth to the nineteenth centuries, the Tutsi rulers of the eastern Nyiginya region expanded their dominance over the neighboring areas (Sellstrom & Wohlgemuth, 1996: 23-24). According to Sellstrom and Wohlgemuth, King Rwabugiri, who reigned from 1860 until his death in 1895, was the most influential of the Tutsi kings. He increased the centralization of power and formed a highly competent administrative state that extracted obedience from even remote areas. However, he was incapable of overpowering the northern and southwestern regions (Sellstrom & Wohlgemuth, 1996: 23-24)–a crucial factor that would shape subsequent intra-Hutu and Hutu/Tutsi relations. The feudal system of patronage and the hierarchical system of social change, ubuhake, was central to the development of Hutu/Tutsi relations. Between two men of unequal status, this arrangement involved bonds of loyalty and an exchange of goods and 39 There is no consensus among historians or anthropologists on the origins of these divisions so crucial to Rwanda’s history (Melvern, 2000: 8). 123 services. Tutsi were generally the patrons with subordinate Tutsi or Hutu as clients. Prunier explains that Ubuhake institutionalized socioeconomic disparities between the Hutu and Tutsi and leant legitimacy to bogus “ethnic” classifications, particularly among the Hutu (Prunier 1995:13). For Tutsi ideologues, the system was a mutually advantageous partnership; for Hutu ideologues the system was a method of subservience. Prunier asserts that the Tutsi elites who held all the highest positions of power began to correlate their status and privilege with their Tutsi “ethnicity.” At the middle-level and lower ranks of the social structure, however, “ethnicity” remained but one factor in determining an individual’s social stature (Prunier, 1995: 15). Indirect Rule through Tutsi Like much of Africa, Rwanda has a long, grim history of European colonization that begins in the late nineteenth century. Rwanda’s first colonizers were Germans who, eagerly importing the racist pseudo-science so popular back home, promoted the notion of a naturalized and superior Tutsi “race” (Sharlet 2001). Not surprisingly, the differences were largely based on the aesthetics of appearance—as Sharlet describes them: “for Hutus, stocky and round-faced, dark-skinned, flat-nosed, thick-lipped, and square-jawed; for Tutsis, lanky and long-faced, light-skinned, narrow-nosed, thin-lipped, and narrow-chinned” (A17). Despite these surface differences, the Hutus and Tutsis are neither separate tribes nor distinct ethnic groups. As Sharlet asserts, “Both speak the same language, practice the same religion, and live by the same customs” (2001, A17). As such, the notion of 124 “tribal warfare” and “ancient ethnic hatred” to which so much of the media ascribes the genocide is actually unfounded. Hence, Mamdani’s thesis, which challenges this so- called “heart of darkness” notion promoted and perpetuated through Western culture; the problem, Mamdani asserts, begins with the introduction of racist ideology by the Belgians and carries through to its wholesale embrace by the average Rwandan. First came the Germans in 1894 and established a hierarchy enabling proxy governance through the Tutsi. The Germans ruled through the existing power structure of the ubuhake. As Melvern explains, Count von Gotzen believed the German policy must be to support the chiefs in such a manner that they would be convinced that their own salvation and that of their supporters depended on their faithfulness to the Germans. But the Germans also favored expansion and in 1912 helped the Tutsi monarchy to subjugate the areas to the north. (Melvern, 2000: 9) After the First World War, the western provinces of German East Africa, Rwanda and neighboring Burundi (Urundi), were given to Belgium to administer as “Ruanda-Urundi” under a League of Nations mandate (Melvern, 2000: 9). What in fact happened was that a policy of indirect rule under the Germans was gradually changed by the Belgians to one of direct rule, and in the years to come the power of the king was eroded by Belgian administrators (Melvern, 2000: 10). In 1922, the king was “obliged to be assisted by Belgian representatives of the colonial resident, and in 1923 he was forbidden to appoint regional chiefs” (Melvern, 2000: 10). Melvern explains that in November 1931, the king, Mwami Musinga, who was against colonization, was deposed by the Belgian administration with little objection from the 125 League of Nations. The king was replaced with one of his more pliant sons, Mutara Rudahigwa, who became known as “the king of the whites because he wore western suites and drove his own car” (Melvern, 2000: 10). Much later, Melvern explains, “His conversion to Christianity in 1943 was part of a Belgian policy for encouraging mass conversions. Christianity became a prerequisite for membership of the Tutsi elite” (2000: 10). The Belgian colonial presence meant that racism was further systematized through the formation of de jure Tutsi authority in the form of official cards classifying identity as either Tutsi or Hutu. Furthermore, Belgians, according to Mamdani, classified the Hutu ethnic majority as “natives” and the Tutsi minority as “settlers,” “aliens,” and nonindigenous invaders, a dichotomy that would linger and later serve as official history during the genocide (2002). With the Belgians, then, protean social differences were transformed into stark racial and political differences. According to Mamdani, the myth of Tutsi superiority—based upon spurious scientific logic and fantasies around race and Europeanized features—thus became legitimized as law and reconceived as historical fact. As Mamdani argues, the idea of genocide was already percolating in the colonizer’s notions of racial pre-eminence and in their transformation of spurious racial claims into a hierarchy of political haves and have-nots (2002). Thus since the 1920s Rwanda was governed by a highly organized hierarchical social and political structure with a colonial government administered indirectly through the Tutsi (Lemarchand, 2002: 307). Hutu and Tutsi classification was turned from a multifaceted social description--often second to local clan affiliation--to an unyielding 126 social and political order with dominant overtones of intellectual superiority and political and economic dominance for the Tutsi, and by extension, the Belgians. Simply put, as in many other African countries, colonization meant European rule through cultural ideology and political manipulation. With limited resources, the Europeans usurped the royal governing structure of Rwanda and ruled indirectly through the Tutsi. The Europeans actively pursued a policy of “ethnogenesis,” a politically motivated creation of ethnic identities “based on socially constituted categories of the pre-colonial past” (Grunfeld, 2007: 44). The policy intended to intensify the economic and political differences between the two groups and to present a theoretical basis for the divisions 40 . The organizing principle for colonial rule was the racist Hamitic thesis stating that everything of value in Africa could be traced back to Caucasian origin (Sellstrom & Wohlgemuth, 1996: 30). Early anthropologists identified the Tutsi as Hamites. According to this classification, colonial administrators came to regard and promote the Tutsi as the intellectually superior, naturally aristocratic race (Sellstrom & Wohlgemuth, 1996: 27-34). Acting under this view, the Germans and, especially, the Belgians, in 1923, changed the royal governing structure to fit their administrative needs. 41 They created a monopoly Tutsi rule and an inflexible ethnic classification system. From approximately 1926 to 1931, the Belgians instituted an administrative reform replacing all Hutu chiefs and deputy chiefs with Tutsi. By 1933, the Belgians introduced compulsory ethnic 40 This was a divide and rule method used in many colonial empires. The idea was to keep potential opponents divided and they cannot unite against you. 41 In 1923 Belgium was granted a League of Nations mandate to govern Rwanda, which it ruled indirectly through Tutsi kings. By 1946 Rwanda became a UN trust territory governed by Belgium. 127 identity cards that were based on fairly arbitrary criteria. Ethnicity became the primary factor for access to education and economic and political rights (Sellstrom & Wohlgemuth, 1996: 27). The effect of 60 years of colonial rule (approximately from 1899 to 1959), was to inflate the egos and fortunes of the Tutsi and foment a sense of resentment and inferiority among the Hutu (Prunier, 1995: 17). Dictatorship and the Exploitation of Ethnic Difference Four key factors that influenced Rwanda’s transition to independence and the first and second Hutu dominated republics are important to understanding the 1990 civil war and the dynamics of the Arusha Accords. First, the struggle for power had become defined in ethnic terms in the late 1950s. Second, with the ascension of the first Hutu republic in the 1960s a cycle began in which power was transferred from one political elite to another in a zero-sum game of winner takes all (Sellstrom & Wohlgemuth, 1996: 37). Third, the Tutsi refugee population displaced in the 1950s and 1960s by fighting and repression became the basis of the Rwandan Patriotic Front (RPF) guerilla army. Fourth, intra-Hutu tensions that developed during the first and second republics (1962- 90), mainly between northern/western and southern groups, dominated Rwanda’s multiparty democracy and the government of Rwanda (GoR) during the Arusha Accords negotiations (Eriksson, 1996: 4). In 1957 nine Hutu intellectuals signed the Bahutu Manifesto, including future president Gregoire Kayibanda, an act clearly illustrating that the colonial conception of Rwandan society as defined by the Hutu/Tutsi divide had taken hold (Prunier, 1995: 45). 128 The point of this document was to create a general awareness of ethnic identity among the Hutu masses, an important statement of “principle” for the coming Hutu “revolution.” It explained that the basis of Rwanda’s problems was the ethnic conflict between the Hutu and the foreign Hamites, primarily the Tutsi. In summary, the prejudicial terms of the document sets a firm set of rules for all Hutu to follow: 1. Hutu men should desist from any contact with Tutsi women; 2. Hutu men should only marry Hutu women; 3. Hutu women should be vigilant and bring their husbands, brothers and sons back to reason; 4. Hutu should refrain from any form of business dealings with Tutsi because they are distrustful, only seeking supremacy for Tutsi; 5. Hutu should hold all strategic positions – political, administrative, economic, military and security; 6. The majority of teachers and students in school should be Hutu; 7. All military forces should be exclusively Hutu; 8. Hutu should stop having mercy on Tutsi; 9. Hutu must maintain unity and solidarity against Tutsi within and outside Rwanda and remain vigilant against the common Tutsi enemy; 10. The Hutu ideology must be taught to every Hutu, and any Hutu opposing the propagation of this ideology is a traitor (Dekmejian, 2007: 282-283). In 1959, revolution brought the Hutu to power and marked an end to the colonial reign, but not to the racist and discriminatory mindset it had promoted. As Mohammed Omar Maundi described events, “[I]f there was anything revolutionary about the 1959 revolution, it was the elitist role reversal whereby Hutu elites succeeded in taking the reins of power from the Tutsis” (31). Severe repression of Tutsi during the early 1960s, including a massacre that killed 5,000 to 8,000, led to a massive exodus of refugees to Burundi, Zaire, Tanzania, and Uganda 42 (Prunier, 1995: 63). Violence in Rwanda 42 It is important to note that the UN recognized Rwanda and Burundi as independent countries in July 1962. 129 continued against the Tutsi, particularly in 1963-65 and 1973, leading to successive waves of mass departure. By 1990, the refugee population, including its descendants, numbered almost 600,000 (Prunier, 1995: 63). The deadly coup by Juvenal Habyarimana in 1973 shifted power to the northern Hutu, stimulating vengeance for the subjugation suffered by the northerners under colonial powers and the favored Tutsi early in the century. During the 17 years of the Habyarimana regime, members of Habyarimana’s immediate power circle, the akazu, consolidated power at the expense of southern Hutu groups from other clans. Jones explains that by the mid 1980s, members of Habyarimana’s Bagogwe clan held 80% of the top military posts and took over the ruling Mouvement Revolutionnare National pour le Development (MRND), Rwanda’s only political party, many of whose leaders were among the main organizers of the 1994 genocide (Jones, 1995b: 5). Habyarimana’s wife, brother-in-law, and other akazu, represented a unique Hutu subculture whose predecessors had remained independent until they were defeated by combined German and Tutsi-led southern forces in the first decade of the twentieth century. Extremely proud of its pre-Tutsi past, Erikkson explains that this group was very wary of any reconciliation with the RPF and unsympathetic to moderate Hutu who supported dialogue (Erikkson, 1996: 6-11). Having felt like second-class citizens first to the Tutsi and then to other Hutu, the northerners led by Habyarimana were unwilling to give up power once they gained control. 130 According to Prunier (1998) severe population pressure on available land exacerbated tensions. The terrain of Rwanda is primarily mountainous with high plateaus, and though the country is blessed with generous rainfall and moderate average temperatures—many regions are capable of up to three agricultural growing cycles per year (Sellstrom & Wohlgemuth, 1996: 1)—the demands on the land are extreme. A country of only 26,338 square kilometers (a bit smaller than Massachusetts), Rwanda had a population of 7.15 million people and an annual growth rate of 3.1% (1991 census). With the national parks and forests excluded, the average pre-genocide population density was 405 people per square kilometer, the highest in Africa (Sellstrom & Wohlgemuth, 1996: 2). The northern Rhuengeri region, the most populated, had 820 people per square kilometer of productive land. Most people were poor subsistence farmers. In 1984, 57% of the rural households farmed less than one hectare 43 and 25% has access to less than half a hectare to support a family of five (Sellstrom & Wohlgemuth, 1996: 3). Rwanda’s economy was severely affected by the collapse of the international coffee market in the 1980s, as coffee was Rwanda’s main export. Further, the “structural adjustment program” imposed by the World Bank and the International Monetary Fund in 1990 were particularly hard on Rwanda’s peasants. The combined effects resulted in the average per capita income falling from US $330 in 1989 to US $200 in 1993, a 40% decrease in only four years (Sellstrom & Wohlgemuth, 1996: 3) (still another source cites the currency devaluation as even higher at 67% [Sharlet 2001]). Notably, however, 43 One hectare equals 10,000 square meters and approximately 2.5 acres. 131 despite this history of conflict, Rwanda was undergoing a renaissance of sorts in the early 1980s and progress was being made to redress the deep poverty while reforestation was helping bring the land back to life (Sharlet 2001). Habyarimana’s regime also brought about dramatic economic development: From 1976 to 1990, Rwanda moved from the bottom to the top of a per capita GNP ranking of countries in that region. The World Bank considered Rwanda a relative success during the 1980s. In 1987 Rwanda’s debt was 28% of its GNP, one of the lowest rates in Africa (Sellstrom & Wohlgemuth, 1996: 49). Although Rwanda’s human rights record was still an issue, the situation was considered to be improving (Sellstrom & Wohlgemuth, 1996). Before 1990, no major ethnic violence had occurred under Habyarimana’s regime, and, to some extent, he was preferred by the Tutsi living in Rwanda (Sellstrom & Wohlgemuth, 1996: 52-54). Regional violence, however, was significant; indeed, some critics believe it played a part in stirring the genocide in Rwanda. One prominent example is found in the overthrow of the Ugandan dictatorship in 1986, in which many Rwandans took part. Among the many exiled refugee Rwandan Tutsis who aided Yoweri Museveni’s guerilla forces (the National Resistance Army) in toppling the Ugandan dictatorship under President Milton Obote and General Tito Okello was Paul Kagame. Kagame rose to prominence in Museveni’s new Ugandan army, becoming head of military intelligence, but later going on to found the RPF (“Rwanda: Ugandan Forces Accused of Deployment in Northern Rwanda,” 1993). 132 Hutu-Tutsi violence in the small country of Burundi, just south of Rwanda, may also have contributed to conflict in Rwanda, according to some critics (Gettleman, 2008: A7). A 1993 coup consisting of “mostly Tutsi army officers who assassinated the country’s first Hutu president” led to a “violent free-for-all involving warring militias, rival politicians, criminal gangs and child soldiers. More than 200,000 people died.” One journalist described the relationship between Burundi’s civil war and Rwanda’s genocide as a mutually inciting cycle of hate. Burundi’s Hutu-Tutsi violence of 1993 would be ominously revisited in Rwanda’s genocidal violence the following year. In October 1993, even as the Arusha accords were being negotiated in Rwanda, 50,000 people died in Burundian clashes between Hutu and Tutsis. A year later, the Rwandan genocide would once again stir these hatreds in Burundi. Noting “chilling similarities” between the two conflicts, one reporter explained that during the genocide, one radio station in southwest Rwanda broadcast its messages of anti-Tutsi hatred to Burundi as well. Civil War and Multiparty Politics Serious transformations began to occur in Rwanda itself in 1990. First, the exiled Tutsi began their guerilla war with an invasion in October 1990. Second, as mentioned earlier, the economic crises placed extensive strain on the government. Third, due to internal and external pressure, Habyarimana’s regime began instituting multiparty politics. The Rwandan Patriotic Front, consisting of exiled Tutsi, invaded northern Rwanda on October 1, 1990 (Gribbin, 2005: 39). After initial success, it was pushed 133 back into Uganda by the internationally reinforced Hutu-dominated Force Armees Rwandaises (FAR) 44 , an organization benefiting from substantial French support in the form of military supplies. A formidable campaign of strident racist propaganda was undertaken by Hutu Power extremists, which helped to galvanize the Hutu masses against the Tutsi “settlers” (Mamdani, 2000). The fighting settled into a low-level guerilla war over the next few years. A stark forewarning of genocide came in Kinigi in January 1993 when a state- sponsored massacre of 1,000 members of the Bagogwe subgroup of Tutsi was undertaken by Rwandan authorities—but one outstanding example of the violence perpetrated against the Tutsi (Norris, 1993: A3). Significantly motivated by the refugee crisis and their increasingly hostile treatment in Uganda, the RPF had an eight-point political plan that was directed at changing the Rwandan government (Moghalu, 2005: 125). According to Alan Destexhe, a prominent scholar on the Rwandan genocide, the initial effect of the October offensive was to aggravate ethnic tension and motivate the loyalty of many Hutus around the President (Destexhe, 1995: 46). Though much of the media, validated by the scholarship of Philip Gourevitch, saw the RFP as heroes, still other critics disagree with such Manichean interpretations (Sharlet, 2001). Still, the effort to vilify the opposition was fierce, as the Rwandan government referred to the RFP as “inyenzi” (cockroaches) and anyone who opposed Hutu power as “ibyitso” (accomplices) (Gourevitch, 1995: 1). The effort to recruit Hutu youth militia was undertaken in full force. 44 The army of the ethnic Hutu-dominated Rwandan regime. 134 With the menace of Tutsi forces, “Hutu Power” extremists redoubled their efforts to retain power. According to Mamdani, genocide required the efforts of both masses and governing elites, a contention at odds with the belief of most scholars that there was a “core group of planners” (Sharlet, 2001, A18). Summarizing Mamdani’s viewpoint, Sharlet (2001) explains: By April 1994, he argues, a propaganda campaign had succeeded in convincing Hutus that killing Tutsis was not murder, but moral duty. When the killing commenced, it was referred to as “communal work.” Chopping up male victims was known as "bush clearing"; slaughtering women and children was referred to as "pulling out the roots of bad weeds" (A18). To Mamdani, the racist ideas of colonizers had infected the average Rwandan, driving them to willingly participate in the massacre led by regime leaders. Not so, say other authorities in the field, such as David and Catherine Newbury who, having lived in Rwanda, sharply contest the notion that the animosity was so widely and popularly embraced (Sharlet, 2001). Mamdani’s theory, as they characterize it, resembles that of Daniel Jonah Goldhagen, who in Hitler's Willing Executioners: Ordinary Germans and the Holocaust (Alfred A. Knopf, 1996) makes average citizens partly accountable for Nazism’s success. Instead, the Newburys look to disputes among the Hutu moderates and Hutu Power extremists and argue that the military and militias were mostly responsible; as such, the race argument becomes less relevant as does the idea of internecine warfare and popular participation. As Sharlet explains, then, “The idea of neighbor against neighbor that made such a gripping story was just an excuse for the West not to intervene” (2001: A18). 135 Not long after the 1990 RPF invasion, however, France, Belgium, and Zaire sent troops to aid the Habyarimana regime. According to Melvern, a specialist in African history, France sent at least six million dollars worth of arms to Rwanda, including mortars, light artillery, armored cases, and helicopters (48). She asserts that by 1993, “Rwanda was receiving US$4 million military aid from France”(48). All this support was under the terms of an amended 1975 military technical assistance accord, which had entailed “an initial modest annual transfer of arms” (48). President Giscard wanted to expand France’s African sphere to include Belgian and French speaking colonial territories, including neighboring Zaire. “France justified its intervention in Rwanda to be no more than honouring a treaty obligation to an ally. In 1992, the agreement was amended to an ‘Accord de defense’” (Melvern, 2000: 48). 45 By October 4, 1990, France, an early supporter of the Habyarimana regime, had sent 150 members of the 2eme Regiment Etranger Parachutiste to Kigali, Rwanda’s capital. Brussels followed with over 400 paratroopers (Odom, 2005: 76). Mobutu Sese Seko, president of Zaire, sensing an opportunity to gain esteem from France and Belgium, sent several hundred of his top Presidential Guard to aid his longtime friend Habyarimana. That one such as Mobuto would be an ally to the Rwandan president casts significant doubt on Habyarimana, as Zairean president exemplifies the corrupt leadership and kleptocracy that plagues so many postcolonial African governments. The involvement of Zaire and France, however, —as well as, along the way 45 The full story of French involvement will never be known. Five days after the 1994 genocide began, the French embassy in Kigali was abandoned. Left behind was a huge pile of shredded documents (Melvern, 2000: 4 20) 136 Uganda, Congo, Angola, the United States, France, and Britain—substantiates for the Belgian scholar Filip Reyntjens that most interpretations of the Rwandan genocide are too local, too focused on tribes, nativism, and ethnicity at the expense of larger regional and international considerations (Sharlet, 2001, A18). Indeed, Reyntjens contends that work such as Mamdani’s reiterates and validates the misconceptions (and hence apathy) of most journalists, who believed that the genocide reflected inscrutable local issues (typical of the dark continent) that the larger international community could not understand much less solve. According to the theory known as the Fashoda Syndrome, 46 French intervention in African affairs is motivated by the desire to stem the extension of British colonial interests in the area, essentially to maintain Francophone areas. Whatever the precise reason, French interest in Rwanda was formidable. In addition to logistical assistance, Paris purchased $10 million in military supplies to aid the Habyarimana regime. (Prunier, 1995: 148-149). Melvern explains: there were also two companies of parachutists and paramilitaries from the French secret service, with combat helicopters. Secretly, the French had operational control of the counter-insurgency campaign. Later on in 1990, a pro-Hutu journal, Kangura, published a full-page photograph of President Mitterand. The caption, in Kinyarwanda read: ‘Great friends, they stand by you during difficult times.’ (2000: 31) Due to its desperate economic situation in 1990, Rwanda was sensitive to pressure from donor nations. Habyarimana’s regime instituted political reforms that were 46 Fashoda Syndrome is the name given to a tendency within French foreign policy in Africa, giving importance to asserting French influence in areas which may be becoming susceptible to British influence. It is named for the Fashoda incident (1898) which is judged to have given rise to it. 137 initiated by the French military attaché, Lt. Col. Galinie. Prunier explains that in March 1991, the dissident and highly racist Mouvement Démocratique Républicain (MDR) was launched and identified itself as the successor to Kayibanada’s MDR/PARMEHUTU party (Prunier, 1995: 144). In addition, the socialist party, or PSD, which enjoyed principally southern support, and the liberal party, or PL, whose base came primarily from the capital, were formed. Under strong internal and external pressure, Habyarimana formed a new transitional government on April 6, 1992, that included all the major opposition parties (Prunier, 1995: 148-149). The government was led by Habyarimana, as president, and Dismas Nsengiyaremye, an MDR member, as prime minister. As the following chapter will detail, Rwanda provides an example of negotiations that appeared to succeed with the Arusha Peace Accord, but the "agreement" was not accepted by all parties involved, and the mediator(s) had little or no leverage to enforce them. From August 1992 to August 1993 Tutsiphobic extremists were excluded from the negotiations and from the transitional government designated by the Accords. With the assassination of President Habyarimana and his Burundi counterpart, Cyprien Ntaryamira, on April 6, 1994 the path of violence was set. Parties to the genocide endeavored to eliminate Tutsi and Tutsi sympathizers alike. UN assistance—as part of the Arusha Accords-- pulled out, international embassies closed and foreign nationals were evacuated. “You cockroaches must know you are made of flesh,” a broadcaster at Radio Mille Collines proclaimed. “We won’t let you kill. We will kill you” (Gourevitch, 2000: 10). 138 Three crucial interrelated factors enabled a small group of Hutu political elites in April 1994 to slaughter their neighbors with such chilling efficiency (Furley and May, 2006: 33). The occurrence of this horrific event is even more tragic because it occurred after the signing of the Arusha Accords, which were meant to result in power sharing and peace between the Hutu and Tutsi. The three factors are: the poverty and large demographic of the country, Rwanda’s history of authoritarian rule, and the “ethnic” conflict between the Hutu and Tutsi. Though, as Sharlet importantly asserts, the precise reasons for the genocide are unclear, still: “Theories of the genocide in Rwanda tend inevitably toward the monocausal: Mr. Mamdani's explanation of race, Mr. Gourevitch's homicidal masses, the Newburys' treacherous elites” (2001, p. A18). As I have sought to demonstrate in this chapter, one issue cannot account for the genocide. The historical context of the case in Rwanda makes clear the critical issues that needed to be addressed during the mediation process. Numerous international and regional actors participated in the negotiations, each with its own motivation and interpretation. Understanding where each party is coming from is crucial to grasping their motives and preferences when playing the Prisoner’s Dilemma game. The mediator(s) must recognize the many intricate details of the historical conflict and use that knowledge to steer the mediation process towards a successful resolution. 47 The next chapter presents history and analysis of the mediation in Rwanda; the following two chapters consider the case studies of Ireland and Rwanda from the theoretical perspective of the Prisoner’s Dilemma and Moore’s 12-step framework. 47 This is also true in the case of Northern Ireland. 139 Timeline of the Conflict in Rwanda 48 1860 Tutsi King Rwabugiri (1860-95) expands his power and creates a centralized state. 1880 The first European explorers arrive in Rwanda. 1899 Germany begins its colonial rule. 1911 An uprising in northern Rwanda is repressed by combined German and Tutsi-led southern forces, creating significant bitterness on the part of the northern Hutu. 1916 Belgium takes over control (under a League of Nations mandate). 1920s-30s A policy of “ethnogenesis” exaggerates the “ethnic” differences between the Hutu and Tutsi and creates a monopoly Tutsi rule. Ethnic identity cards are introduced. 1959-61 A Hutu “revolution” replaces the Tutsi monarch with an independent Hutu republic controlled by President Kayibanda and southern Hutu political elites. 1973 A coup by Juvenal Habyarimana transfers power to northern Hutu. 1960s-70s Attacks and repression against the Tutsi result in a refugee population that numbers approximately 600,000. 1990 Oct. 10: The Rwandese Patriotic Front (RPF) invades Rwanda. 1990-91 A series of regional mediation efforts are undertaken. The Communaute Economique de Pays des Grand Lacs (CEPL) hosts summits that include the participation of Zaire, Uganda, Burundi, and Tanzania. 1991 Feb. 19: The Dar-es-Salaam Declaration produces a refugee agreement that is later used as the basis for the Arusha refugee talks. Mar. 29: The N’Sele Cease-fire, although short-lived, formally creates the Neutral Military Observer Group (NMOG). Sept. 16: The Gabolite Cease-fire deploys the NMOG. In addition, the control of mediation switches from Zaire to Tanzania. October: low intensity traditional guerilla warfare by 48 Source: Greenberg, Melanie C., John H. Barton and Margaret E. McGuinness, eds. Words over War: Mediation and Arbitration to Prevent Deadly Conflict. Lanham: Rowman and Littlefield, 2000. 140 the Force Armees Rwandaises (FAR) sees initial success and is followed by a resurgence of the RPF. 1992 May 8: US Assistant Secretary of State for Africa Herman Cohen meets with the RPF and Yoweri Museveni in Kampala, Uganda. May 10: Cohen meets with Habyarimana and opposition members, Prime Minister Nsengiyaremye and Foreign Minister Boniface Ngulinzira, in Kigali, Rwanda. May 24: PM Nsengiyaremye and FM Ngulinzira meet in Kampala with the RPF to explore peace talks. June 6: the RPF and the Rwandan Government hold preliminary talks in Paris. June 20: Government of France’s Direction for Africa and Maghreb Affairs Dijoud and Assistant Secretary Cohen meet with Ugandan Foreign Minister Ssemmogere and RPF representatives in Paris to strongly encourage the talks. July 12: The Arusha talks open under the auspices of the Organization of African Unity (OAU). The primary facilitation is delegated to Tanzanian President Ali Hassan Mwinyi and his ambassador, Mpungwe. Aug 18: The Rule of Law Protocol signed. September: The power-sharing talks begin. Oct. 6-13: The talks on the Framework for a Broad Based Transitional Government (BBTG) continue. An agreement that significantly reduces the scope of presidential powers is reached. The Transitional National Assembly is also created to replace the Conseil National de Developpement (CDN). FM Ngulinzira initials the agreement without the authority of Habyarimana. November – December: Discussions on the composition of the transitional institutions continue. Nov. 15: Habyarimana calls the Arusha Accords “pieces of paper.” 1993 Jan. 5-10: The distribution of the seat in the BBTG is finalized. The extremists Coalition pour la Defense de la Republique (CDR) is excluded. Feb. 8: The RPF launches an offensive. June 9: The protocol on the Rights of Refugees and Displaced Persons is signed. June 24: The Protocol on the Integration of the Armed Forced of the Two Parties is signed. July 8: The extremist Radio-Television Libre des Milles Collines starts broadcasting. July 23-24: A meeting of the MDR expels Faustin Twagiramungu, signaling the beginning of the splintering of the opposition parties. Aug. 4: President Habyarimana and the RPF chairman Alexis Kanyarengwe sign the Arusha Accords. Oct. 5: The UN Security Council approved the United Nations Assistance Mission for Rwanda (UNAMIR), a 2,500 – strong peacekeeping force. Nov. 1: The UNAMIR forces start arriving. Nov. 21: Tutsi military extremists assassinate Burundi Hutu President Ndadaye. The subsequent repression and reprisal killings result in tens of thousands of deaths and over 600,000 refugees. Dec. 28: 600 RPF soldiers arrive in Kigali as part of the Arusha Accords. 141 1994: Jan. 8: Deadlocks within the Mouvement Democratique Republicain (MDR) and Parti Liberal (PL) block the convening of the BBTG. Feb. 10: Further attempts to convene the government are aborted by the assassination of the Parti Social Democrate (PSD) leader Felicien Gatabazi and the reprisal killing of Martin Bucyana of the CDR. Apr. 6: Habyarimana and Burundi President Ntaryamira are killed when their plane is shot down while approaching the Kigali airport. Apr. 7: Interim Prime Minister Agathe Uqilingiyimana is killed along with ten Belgian peacekeepers assigned to protect her. Belgium withdraws its forces. Apr. 8: The RPF renews its fighting. Apr. 21: UN Security Council resolution reduces UNAMIR from 2,500 to 270. Apr. 30: By the end of April, at least 100,000 people have been killed, and 1.3 million have become refugees. July 18: The RPF declares that the war is over. Almost one million people have lost their lives and in October the United Nations estimated the war, through death and displacement, has reduced the population of Rwanda from 7.9 million to 5 million. 142 CHAPTER SEVEN: THE MEDIATION PROCESS IN RWANDA Chapter seven presents the various negotiations that led to the Arusha Peace Accords between the RPF and the government of Rwanda (GoR). The Arusha Accords are best analyzed along a timeline that begins with the RPF’s invasion on October 10, 1990 through the launching of the peace process in 1992—with several efforts at negotiation along the way— and culminating with the egregious failure of the Accords as starkly indicated by the genocide that took place in the Spring of 1994. Not surprisingly, interpretations of the Arusha Accord failure vary—with some accounts citing dubious ancient tribal warfare theories and racialized “dark continent” explanations, and still others assessing that the negotiation terms themselves were highly flawed, offering few incentives to the key segments of Habyarimana’s government (Lemarchand, 1994) and thus fomenting reaction from the heavily armed Hutu elite akazu minority group which had no stake in honoring the successful implementation of the Accords. In fact, as this chapter will explain, many factors must be taken into consideration, among them the disparate and disconnected efforts at resolution— mediation efforts (official and unofficial, collective and individual), peacekeeping missions, and humanitarian aid—as well as the obduracy of the disputants’ respective positions, failures in intelligence, over-ambitious mediation efforts, inconsistent strategies for managing spoilers to the plan, and lukewarm support from the international 143 community. Add to these issues that the conflict was not in fact dyadic as it appeared on the face. Unlike the Northern Ireland case, which indeed had its share of fragmentation and splinter groups, Rwanda was far more than a binary dynamic, Hutu versus Tutsi, but rather involved numerous parties from moderate to extreme, many with sharply incompatible interests and motivations. In considering all of these questions, this chapter will attempt to identify the respective positions and perceptions of the central parties in the conflict, including the regional organizations and international actors in the pre-Arusha mediations and the role of the Tanzanian mediators, specifically, the diplomat, Ami Mpungwe, to the Accord itself. Finally, this chapter seeks to explain how a series of failed cease-fires, the uncompromising self-interest of certain parties, peacekeeping forces that suffered from a feeble mandate, and catalytic events such as the suspicious death of President Habyarimana resulted in genocide. RPF’s invasion of Rwanda on October 10, 1990 immediately prompted significant diplomatic and military activity. A few days after the invasion, Belgium, a former colonial power in Rwanda, sent a high-level delegation consisting of its prime minister, foreign minister, and defense minister to Kigali in an effort to settle the conflict. The group met in Nairobi, Kenya, with President Habyarimana on October 14 and then held talks with the governments of Kenya, Tanzania, and Uganda, and the Organization of African Unity (OAU). The efforts by the Belgian group culminated in the regional Mwanza summit on October 17, which laid the groundwork for the involvement of neighboring countries and the OAU (Jones, 1995a: 2). 144 Regional efforts integrated the sponsoring of cease-fire talks and a series of ad hoc summits. The next summit, held on October 27, 1990, at Gbadolite, Zaire, was hosted by the heads of state of the Communauté Economique de Pays des Grand Lacs (CEPL), among the members were Rwanda, Zaire, and Burundi; also present were representatives from Uganda, Tanzania, the OAU, and the United Nations. The third gathering, the Goma, Zaire, summit, was attended by presidents of Zaire, Rwanda, and Burundi, and a deputy prime minister from Uganda. Notably, however, rather than addressing the actual nature of the conflict, much of the time at these summits was spent on the less politically inflammatory Ugandan refugee problem (Maundi, 2006: 43). Mobutu Sese Seko, the former Congolese dictator, was in charge of the mediation despite his country’s military involvement in Rwanda. The decision to appoint Mobutu as a mediator came not in spite but because of his role as a regional leader and his membership in the CEPL, which were both seen as increasing his potential to influence Habyarimana. Given his well-earned reputation as a supremely corrupt kleptocrat, Mobutu as mediator was nothing more than a politicized pose through which he hoped to curry favor with the West. Mobutu thus presents the ultimate example of a thoroughly self-interested and overtly biased mediator whose failure to resolve the entrenched problems was inevitable. Nonetheless, when the Rwandan government proposed him as mediator, the RPF accepted him in spite of these obvious problems. Further moves toward a refugee conference began with a meeting of regional foreign ministers in Kinshasa, Zaire (January 17-19, 1991), attended by the OAU and the 145 UN High Commissioner for Refugees (UNHCR). At this meeting, Habyarimana confirmed that the Mwinyi would host the refugee conference and a date for it was set. At the next meeting on February 17, 1991, Ugandan President Yoweri Museveni, Tanzanian President Al Hassan Mwinyi, and Habyarimana, the Rwandese president, were persuaded to sign the Zanzibar Communiqué. It restated a commitment to finding a peaceful solution to the conflict between the GoR and RPF through a cease-fire agreement, a dialogue with the internal and external opposition, and a conference on the refugee problem in Rwanda (Adelman and Suhrke, 2000: 133). In a speech presented the evening before the next meeting on February 19, a conciliatory Habyarimana expressed his dedication to nonviolent solutions to the refugee problem: Rwanda has always considered the solution to the problem of its refugees as being exclusively of a humanitarian nature. Rwanda condemns the use of arms, which will never bring about a solution to the problem of its refugees. It believes that any political claim can only be achieved through the framework already open, of the democratic process of political pluralism which has been launched.(“Rwandan President’s Address to the Nation on Refugee and Rebel Problems, 1991 ) The conference held in Dar-es-Salaam (February 19, 1991) was successful insofar as it resulted in the Dar-es-Salaam Declaration on Rwandese Refugees Problem. The Declaration committed the Government of Rwanda to finding a “definitive and durable solution” to the refugee problem. 49 This agreement was the first noteworthy achievement from these summits because it later formed the foundation for refugee negotiations in Arusha (Adelman and Suhrke, 2000: 133). The Declaration also expressed gratitude to 49 This included both Hutu and Tutsi refugees. 146 Mobutu “for his efforts in instituting a dialogue between the Rwandan government and the armed opposition” (“Declaration on Rwandese Refugees Problem,”1991). By many accounts, these initial talks failed to deal with the on-going civil war because they over-focused on the refugee problem and misconceived of the larger issue as an interstate clash between Uganda and Rwanda, to the neglect of the military conflict among the various parties. So very little success was actually made. Despite its renegade status, RPF—an irrefutably formidable military entity—would have to be included in any talks to end the conflict. Neglecting to do so would preclude any genuine dialogue and allow violence to continue, but Habyarimana still needed convincing of that. Moreover, the mediators, led by Mobutu, had thus far presented no compelling incentives to cease the violence; if he did have leverage, critical interpretations of his mediation do not mention them. Maundi details the problems thusly: regional leaders realized that their mediation efforts had failed to stop the military conflict, and it must have dawned on them that the failure had much to do with their mediation approach. This realization necessitated changes in both their perceptions of the conflict and the approach to its mediation. (2006: 44) Whatever success mediation had had so far, it was negligible and temporary. Though negotiation by regional actors led to an agreement on paper, it did not bring about any long-term end to violence or any commitment to conciliation. On March 29, 1991 in Zaire, the N’Sele Cease-fire Agreement was signed and the first formal cease-fire of the civil war was agreed. Article Two of the Agreement articulated an immediate end to violence and a commitment to negotiations: “The cease- fire implies the cessation of all hostilities in order to start a dialogue and serious 147 negotiations between the two parties under the auspices of the mediator” (“Rwanda Details of Cease-Fire Agreement Signed with Rebels,”1991). The N’Sele text established the terms of the cease-fire, set part of the agenda for continuing talks, and formalized the creation of the Neutral Military Observer Group (NMOG). Arriving two weeks after the cease-fire was signed, NMOG was to police the cease-fire and report to the OAU on violations. The NMOG was made up of 45 officers from three countries—Burundi, Uganda, and Zaire. Before the NMOG had even deployed, the Force Armees Rwandaises (FAR) broke the cease-fire by shelling RPF positions in north-eastern Rwanda. Meanwhile, Radio Rwanda in Kigali presented a steady stream of reports about the RPF’s own violations of the cease-fire (“Rwanda Details of Cease-Fire Agreement Signed with Rebels, 1991).” As Radio Rwanda reported, The NMOG officially launched its efforts on May 19, 1991 under Commander General Hashim Mbita: Gen Mbita called on journalists to show honesty in their reporting and to report information based on facts and not speculation or the search for sensationalism. He also took the opportunity to call upon the two conflicting parties to give the neutral NMOG the necessary assistance and cooperation in order to help it carry out its mission. ( “Rwanda Observer Group to be Deployed in Combat Areas, 1991) Of these events, Dingley explains that, “the first cease fire [March 1991] was a step in the dark, with unpredictable consequences, but at the very least, proved a positive kick start to an otherwise stalemate condition” (Dingley, 2008: 239). Consequences were “unpredictable,” as Dingley says, because of the history of the conflict, the depth of the animosity, the number of splinter Tutsi and Hutu rebel groups that could become 148 potential spoilers, and, according to several accounts, the fact that Habyarimana never had a genuine commitment to the terms of the agreement (Maundi, 2006, 46). Whatever the reasons, the military conflict continued. Despite the failure of the first cease-fire, a second cease-fire was signed by the GoR and the RPF on September 16, 1991 in Zaire. In this agreement, the military leadership of the NMOG shifted from Zaire to Nigeria, and the objectives of a peaceful solution were reconfirmed. At this point, mediation switched from Zairean President Mobutu to Tanzanian President Mwinyi. According to both Rwandese and western sources, by this time all the parties recognized the incompetence of Mobutu’s mediation role. Mobuto retained the formal title of “mediator,” but from that point on had little to do with the peace process. Adelman and Suhrke explain that the need to bypass Mobuto as mediator shifted the locus of the peace process, minimized the role played by the CEPGL, and increased the participations of non-regional states and actors. An intensification of American and French diplomatic activity following the Gabolite 50 talks marked a significant change in the direction of the peace process. (Adelman and Suhrke, 2000: 135) Though France sent in its troops to protect its interests, international response was tepid at best. Scholars such as Maundi attribute this disregard to, among other reasons, post-Cold War standoffishness, foreign policy distractions (the Persian Gulf, and Yugoslavia) and disasters (such as Somalia), and a lack of national self-interest. Nonetheless, the last resolution efforts prior to the Arusha negotiations involved the partially coordinated, low-level involvement of the United States. Until that point, the 50 Also spelled “Gbadolite.” 149 US had been only moderately attentive to the Rwandan problem. As a matter of course, the intermittent violence by Hutu against Tutsi, and Tutsi against Hutu, between 1990 and 1993 was noted by the State Department in its human-rights reports and the US duly recognized Habyarimana’s complicity in this violence. President Clinton responded by dramatically reducing aid to Rwanda and “redirecting much of the remainder away from the government and into nongovernmental organizations” (Callaghan, 1997: 111). Still, the United States undertook the next mediation effort. In mid-January, 1992, Carol Fuller, the Rwandan desk officer at the US State Department, was communicating with RPF representatives in Washington and these discussions finally matured into the involvement of U.S. Assistant Secretary of State for Africa, Herman Cohen. Cohen began a series of interagency meetings that for the next two years oversaw US involvement in Rwanda (Jones, 1995a: 4). Due to the lack of progress in the peace talks throughout 1991, US Ambassador Bruce Flatin in Kigali sought a more active US role. In response, Cohen arranged for a parallel series of meetings with Ugandan President Museveni and representatives of the RPF in Kampala, Uganda on May 8, 1992 where he encouraged negotiations and offered US technical assistance (Gribbin, 2005: 18). At these meetings, Cohen pointed out that the war was gradually ruining Rwanda’s economy, which was true, and that a military solution was not possible given that the RPF could not hope to gain more than its small northern enclave (Gribbin, 2005: 20). (Notably, the latter assertion was a gross underestimate of RPF military potential). Cohen suggested to Museveni that a negotiated settlement of the refugee problem would 150 serve the dual purposes of easing the internal burden on Uganda and weakening his longtime opponent, Habyarimana. According to Adelman and Suhrke: Museveni was pleased that the US did not agree with France’s view that Uganda was the source of conflict and pledged to use his influence with the RPF. For its part, the RPF expressed an interest in negotiations but asserted that the GoR was refusing their right to return. (Adelman and Suhrke, 2000: 136) Cohen met separately in Kampala on May 8, 1992 with Habyarimana and representatives of internal Hutu-led opposition parties with whom he was sharing power. There were five predominant opposition parties: the MDR (Mouvement Démocratique Républicain), Social Democratic Party (PSD), the liberal party (PL), the Christian Democrat Party (PDC), and the Coalition for the Defense of the Republic. Though they espoused dramatically different ideologies, the opposition parties were bound primarily by a dislike of the president and a wariness of RPF intentions. At this gathering Habyarimana again refused to acknowledge the RPF as anything but Ugandans or to characterize the invasion as anything but a Ugandan invasion. He did, however, express a willingness to consider the right of return issue. Meanwhile, Habyarimana and some of the opposition parties harbored deep fears that the rebels would force a throwback to previous aristocratic Tutsi rule and were concerned that the Tutsi would be able to fool them at the negotiating table. Though Cohen left Kigali without any guarantees of negotiations, he still felt certain that talks were forthcoming (Jones, 1995a: 5). Indeed, in an open discussion between the assistant secretary of state and Habyarimana, the former appeared sanguine that the resumed 151 negotiations were promising and imminent. As Radio Rwanda reported, Cohen announced: From my talks with the RPF and the Rwandan government both are eager to end the war which benefits no one, and the best way to end the war is through negotiations . . . I now have the impression that the two parties are ready to have serious negotiations. (“Rwanda Herman Cohen on Talks with the Rebels,” 1992) At this point French involvement became more substantial. Unlike the United States, France had deep ties to and a history of interest (and self-interest) in Rwandan politics. Like Zaire, Rwanda was Francophone without being a former French colony and its government had long benefited from French support mostly in the form of military and monetary support. Precisely what support France provided remains unknown, however, as Christopher William explains, “Most of the details of French military involvement in Rwanda are secret, and likely to remain so for the foreseeable future” (287). What is known is that when the RPF attacked Habyarimana’s Hutu-led government in 1990, France launched Operation Noroît, which dispatched 350 men to Kigali to protect the government (William, 287). Williams reiterates, “Much about this operation remains secret, but it seems that a large part of the operation was aimed at arming and training the Rwandan government forces to fight the RPF” (287). Soon collective pressure was exerted by the US and France on the RPF to eliminate its precondition that Habyarimana step down in favor of an interim government, as the demand was presenting a severe impediment to the start of substantive negotiations. The joint effort occurred when the Director of African Affairs for the French Foreign 152 Ministry, Paul Dijoud, asked Cohen to join him on June 20, 1992 at the Foreign Ministry Conference Center in Kampala for a meeting. The French role in the negotiations and its close attention to Rwanda had numerous motivations, most of which were self-serving. Not until the findings of the Mucyo Commission were revealed some 14 years later would the full scope of France’s complicity in the genocide begin to come to light. Led by a former Justice Minister by the name of Jean de Dieu Mucyo, the commission—which first convened in 2006— produced a 500-page report that implicated 33 French political and military leaders in the genocide. Among those cited for their direct involvement in the genocide were Former President Francois Mitterand, former Prime Minister Dominique de Villepin, and Mitterands's son, Jean Christophe Mitterrand (“Rwanda Prosecutor General Studies Mucyo Report,” 2008). Based upon the witness testimony of genocide survivors, as well as journalists and researchers, the report claimed that French soldiers gave direct material and diplomatic support to the Hutus. An article that appeared a few days after the report became public reported the following of the commission’s findings: The French allegedly created a secret command of the Rwandan Army through what [Jean Christophe Mitterrand] called a "légion présidentielle." The most shocking allegations come from survivors who allege that French soldiers participated in a raping spree, aiding the facilitating genocidal forces to run into the then Zaire. These soldiers were a part of “Operation Turquoise,” a French military intervention in June 1994, an ostensibly humanitarian mission that had the backing of the UN Security Council. (“Rwanda Country Report Accuses France of Role in Killings,” 2008). But back in 1992 Dijoud euphemistically characterized the alliance with Rwanda by explaining that the two countries were linked through “friendly cooperation,” with 153 France “interested in ensuring that the states and peoples of the subregion were close to each other, and that they were brotherly and friendly” (“East and Horn of Africa in Brief,” 1991). Herman Cohen forcefully told Ugandan Foreign Minister, Ssemmogere, and his military entourage, whom he assumed were RPF members, that the US considered the RPF demands to be irrational and expected Uganda to use its considerable leverage with the RPF to create an atmosphere conducive to negotiations (Furley & May, 2006: 41). To back up the demand, Cohen made an implicit threat of withdrawing any remaining aid to the region. He argued that the refugee situation was draining limited Western resources, and if the situation were not resolved, then the assistance currently provided to Uganda would most likely need to be diverted (Furley & May, 2006: 44). After the June 1992 talks, the RPF and the GoR agreed to meet in Arusha under the auspices of the OAU for comprehensive negotiations. Habyarimana assigned Tanzanian President Mwinyi to be in charge of the talks. The first session commenced on July 12, 1992 (Furley & May, 2006: 44). The leverage introduced by the Western powers was critical to getting an agreement to enter negotiations where African efforts had not succeeded. The achievement was also attributable to the ripeness of the conflict. From 1990 until 1992, both parties thought there was an advantage to be gained from continued fighting. The government had hopes that it could defeat the RPF militarily due to its initial victory and the support of the French military forces. Jones, however, explains that by 1992 military victory for the government seemed highly doubtful, and the economic crisis was placing 154 considerable strain on the government (Jones, 1995a: 5), which by many accounts was now facing a “hurting stalemate.” On the other side, as Adelman and Suhrke explain, RPF leaders realized that “winning on the battlefield was not possible and they would therefore need to proceed with the negotiations” (2000: 133). Agreement to meet in Arusha, Tanzania was tenuous at best. For self-interested reasons, the government wanted the talks to be held in Paris, with French mediators; the RPF wanted African mediators and an African meeting place. Aims of the Arusha Negotiations The internal dynamics of the parties greatly complicated the Arusha negotiations and the attempt to bring peace. Still, in theory, Arusha was a textbook mediation for several reasons: (1) Both the GoR and the RPF were present at the table; (2) the conflict had reached a window of opportunity for conflict resolution; (3) the root causes were addressed, and (4) a neutral mediator guided the negotiations (Kuperman 1996: 22). Even with the promising new mediation leadership, the talks were difficult because the GoR was forced to negotiate not only with the RPF, but also with internal opposition (Cook, 2006: 30). While the RPF was disciplined and well-organized, the GoR was highly fragmented with moderates and extremists pulling in opposite directions. Indeed, to gain a real perspective of the conflict, according to Maundi, consideration must be given not to two but to three parties. The polarity between the PRF and the Rwandan government omits the important role of the internal opposition to Habyarimana, which actually presents a third party, thus making the conflict in essence a 155 triad. To provide an understanding of the negotiating dynamics behind the Arusha Accords, it is necessary to analyze the role, objectives, motivations, and strategies of the various parties to the talks. RPF: Armed and Determined Consisting of Rwandan refugees from Uganda, mostly Tutsi with a handful of Hutu exiles, the RPF was a regimented and highly effective guerilla army, as reflected in its negotiating approach. When the RPF arrived in Arusha, it already had a theoretical structure for its demands and detailed position papers for many of the finer points. A former RPF official described the basis of the party’s success at the table as four-fold: 1) they were highly motivated because they felt that they were fighting for a just cause, 2) their strong organization and discipline allowed them to speak unfailingly with one voice, 3) they were in a strong negotiating position because of their military successes, and 4) they were able to more effectively foster support from the observer group (Sorbo & Vale, 1997: 105). The RPF had several major issues with which it was concerned: -First, the establishment of the rule of law; -second, a power-sharing arrangement that included veto authority over essential government functions; -third, integration of the national army; -fourth, the right of return for refugees (Sorbo & Vale, 1997: 105). Security concerns were at the core of most of the RPF’s positions. The final settlement 156 gave the RPF almost all of its demands, and indeed observers noted that in some key instances the RPF would probably have settled for less if forced to do so. Throughout the negotiations, the RPF tried to earn the support of the government’s internal opposition in creating a post-Habyarimana regime—and it used both enticements and implicit threats to try to reach it. In addition, all of the parties involved knew that the RPF was willing to return to the battlefield if an acceptable agreement was not reached. This option thus loomed over the proceedings GoR Primary opposition to the RPF was the Habyarimana’s Rwandan ruling party, the Hutu-dominated Mouvement Revolutionnaire National pour le Developpement (MRND) and its FAR military wing. According to Maundi (2006), the government was emboldened by its majority presence and was willing to fight the RPF and internal opposition to retain power. In order to draw international attention, Habyarimana promoted the RPF’s connections to Uganda and presented the rebel offensive as an interstate conflict (Maundi 2006: 41). France was quick to respond in support of these claims and the GoR’s alliance with France meant a steady supply of arms, special training for its armed forces including the interahamwe militia, as well as substantial diplomatic and military support (38). In turn, France protected its interests against Anglo-Saxon interference and, according to Prunier, maintained opportunities for money-laundering and a set of interests at once nostalgic and venal. Prunier explains: 157 There is a high degree of symbiosis between French and francophone African political elites. It is a mixture of many things: old memories, shared material interests, delusions of grandeur, gossip, sexual peccadilloes—in short, a common culture…(1995: 103). Despite this formidable support, the GoR delegation was still disjointed and contentious within, putting it at a distinct disadvantage as a negotiating entity, as the RPF was able to exploit these noticeable divisions. To Habyarimana and even more to the political extremists, ethnic difference was ideologically fundamental and politically irreconcilable, serving above all as the basis for maintaining power. Ethnic hierarchy in the form of identity cards and an interpretation of the events of 1990 as a Tutsi-led Ugandan invasion legitimized the efforts of those who would soon perpetrate genocide. Indeed, promoting notions of the Tutsi’s ethnic- based inferiority was central to the genocide and with the aid of media and official pronouncements, a steady stream of virulently racist propaganda won over many converts (Gourevitch, 1995). The ethnic superiority that the Tutsi had enjoyed under European colonial rule was thus inverted as the Hutus fought back by promoting their own system of ethnic-based privilege. Hutu elitists justified their superiority by invoking what is called the “Hamitic hypothesis,” developed by the Nile explorer John Hanning Speke, which figures Tutsi as foreigners from Ethiopia, interlopers and strangers to Rwanda. Ironically, Europeans used this same argument at the turn of the century to justify Tutsi superiority. As Gourevitch explains: “Speke’s idea was that all culture and civilization in the region had been introduced by the taller, fairer people, whom he declared a Caucasoid tribe of Ethiopian origin, and therefore a race superior to the native 158 Negroids [Hutu]” (1995: 9). Rwanda’s German and Belgian colonizers adapted these ideas to their own racist pseudoscience, promoting the more European looking Tutsi as superior. But 1959 marked the turning point from Tutsi ethnic-based superiority to Hutu ethnic-based dominance with the Hutu power movement leading the charge. However, as previously noted, because of “mixing, ethnographers and historians agree that Hutus and Tutsis cannot properly be called distinct groups” (Gourevitch, 1995: 9). Nonetheless, the Tutsi went from elite to denigrated all from their associations with the tribe of Ethiopian origin. Internal Opposition to GoR (Habyarimana) The negotiating triad was completed by five opposition parties, MDR (Mouvement Démocratique Républicain), Social Democratic Party (PSD), and the liberal party (PL), Christian Democrat Party (PDC), and the Coalition for the Defense of the Republic, all of which were essentially negotiating the intra-Hutu power dynamics of a future government while trying to resolve issues with the RPF. Notably, the RPF threat fortified the will of the various internal opposition groups, for whom the invasion served as a catalyst to their own uprising (Maundi, 2006: 39). The leader of the delegation, Foreign Minister Ngulinzira, an MDR member, commented that it was often more difficult to reach an agreement within the GoR than with the RPF. As Herman Cohen had believed, many of the opposition parties 51 were genuine reformers who had a common purpose with the RPF on numerous issues, especially the 51 This included MDR (Mouvement Démocratique Républicain), Social Democratic Party (PSD), and the liberal party (PL). 159 removal of Habyarimana. However, they were caught between Habyarimana, the hardliners—especially the CDR, the Coalition for the Defense of the Republic, an extremist and racist anti-Tutsi party—and the RPF. Some of the opposition parties did not have clear and consistent objectives throughout the negotiations, whereas others were deeply suspicious of the RPF and uncertain about the rebels’ true commitment to power sharing. The RPF’s offensive in February 1992 significantly deepened this concern. Opposition groups were afraid of being taken advantage of by the RPF, which some viewed as very cunning and intellectually superior. In addition, the members of the opposition were entangled in complicated power struggles of their own. New Mediation in Arusha, Tanzania With regional mediation and Mobutu clearly unsuccessful in reconciling the parties, fresh guidance was necessary. When the RPF requested that Tanzanian official Ami Mpungwe serve as mediator and the Rwandan government agreed, it appeared that a new tone of conciliation had been set. Curiously, Mobutu was out of the negotiations but retained the title of mediator. On July 12, 1992 talks began under the aegis of the OAU, with Mpungwe as primary facilitator. Radically different from his predecessor, Mpungwe had none of the taint of Mobutu (Greenberg, Barton & McGuinness, 2000: 226). Reflecting on his predecessor, Mpungwe criticized Mobutu’s “summitry approach,” feeling that the format afforded too much opportunity for grandstanding (“Nature of the Mediation,” 2002). Mpungwe began 160 by seeking to establish trust among the entities and to institute a cease-fire and then move on to the more contentious issues of integrated government, armed forces, and an end to ethnic apartheid. Throughout the negotiations Habyarimana vacillated between the moderates who supported some reconciliation and the hard-liners in the MRND and CDR who adamantly opposed it. The CDR in particular “laid plans to abort the peace process and exterminate their political opponents” (Burkhalter, 1994: 44). Several participants felt that Habyarimana himself never intended to abide by the Accords and that he anticipated that the French would bail him out of any agreement. In November 1992, Habyarimana gave a speech in which he maligned the Arusha Accords as mere “pieces of paper” (Sorbo & Vale, 1997: 103). While Habyarimana retracted the claim later, his lack of allegiance was made obvious with his decision to place an opposition figure in charge of the GoR delegation. Some claim that Habyarimana expected the talks to fail and that he intended to blame the failure on his political opponents (Sorbo & Vale, 1997: 103). Burkhalter (1994) expresses this same sentiment, explaining that the president “had no principled commitment to the accords and temporized on their implementation at every step” (44). Ultimately, however, the negotiations reached a settlement on August 4, 1993 in Arusha, Tanzania by the government of Rwanda and the RPF. Whatever Habyarimana’s true stance, once his plane was shot down on April 6, 1994, power was rapidly seized by extremists critical of what they saw as conciliation on the President’s part. The extent of Habyarimana’s awareness of the genocide and speculation around its inevitability had he 161 not been assassinated is a fool’s game, but a compelling one nonetheless; as Kuperman writes, interest remains about whether Habyarimana “was an extremist or a reluctant moderate.” He continues, If an extremist, then his commitments during the Arusha process were but a cynical effort to buy sufficient time to organize and implement a “final solution” for the Tutsi. If a moderate, his repeated backpedaling from those commitments was a desperate, and ultimately failed, attempt to mollify extremists in his government while reluctantly steering Rwanda toward a pluralist future. (1996: 232). Tanzania: The Arusha Facilitator In its efforts, Tanzania was motivated by a variety of interests compatible with its role; three are of note: (1) the mutually compatible goals of humanitarian concern; (2) a self-interest in resolving the long-standing refugee problem, and (3) the promotion of regional stability through the development of a sustainable Rwandan government (Sorbo & Vale, 1997: 108). As reported by Track Two the official publication of the Centre for Conflict Resolution and Media Peace Centre, Mpungwe conceded to putting pressure on the parties, but maintained that he did not use political leverage on behalf of Tanzania. As he said, “From day one I put it clearly to them that they have the responsibility to resolve their own problem and it was nothing to do with Tanzania…We are not a big power, we couldn’t threaten much” (“Nature of the Mediation,” 2002). Beginning in July 1992, Ami M. Mpungwe, the primary facilitator, sought to exhibit a mature and practical approach to the negotiations. It was his belief that his team of mediators more than adequately met the criteria for the knowledge, time, and persistence needed for successful mediation (Mthembu-Salter, 2002). Mpungwe directed 162 the talks in a prescribed sequence, developing and employing various mechanisms and leverage to resolve sticking points (Cook, 2006: 100). He believed that he was there to “referee” and encourage the parties, sometimes forcefully, to address their problems by developing a coherent, reasonable agreement for the future of Rwanda. The need to establish trust between the two parties was Mpungwe’s primary objective at the outset of the negotiations. First, as previously mentioned, both parties had agreed to a cease-fire on July 12, 1992. A next round of talks, which culminated on August 18, 1992, defined the new political order. “The agreement stated that no one should be above the rule of law, advocated the separation of powers, condemned ethnic discrimination and stated that the return of Rwanda's refugees was their ‘inalienable right’" (Mthembu-Salter, 2002: 10). Momentum built throughout October, as a third step steered the talks towards the power-sharing arrangements in the Broad-Based Transitional Government (BBTG) and the transitional institutions (Grunfeld, 2007: 67). At this phase the RPF and opposition members worked together stripping away Habyarimana’s presidential powers. Signed on October 30, 1992, this phase of the Agreement divested the president of his power to veto bills passed by the national assembly as well as his right to appoint Supreme Court judges; in fact, these judges would have jurisdiction over the president (Mthembu-Salter, 2002: 10). Over the next few months cabinet positions were apportioned in such a way that if the opposition parties and the RPF were in concert, they would always outnumber Habyarimana. While the MRND won the position of president and minister of defense, it failed to win its bid to include the extremist CDR in the government. According to Mthembu-Salter (2002), 163 integration of the armed forces became a major sticking point, stalling talks in April and May. Even as the disputants managed to generate a plan for the return of refugees, they continued to accuse each other of participating in negotiations while covertly preparing for continued fighting. Both in fact were doing just that: RPF was working on its battle plan and strategizing for the clashes against FAR, while the akazu (literally translated as “little house,” akazu was a clique of Hutu elite) trained the notorious youth militias of MRND and CDR. Nonetheless, on August 3, 1993, a “highly technical and detailed agreement was finally signed on the integration of the RPF and FAR into a new national defense force” (Mthembu-Salter, 2002: 10). Mpungwe drew upon several strategic measures whenever problematic situations arose. First, whenever the talks became contentious, Mpungwe would separate the groups and hold proximity discussions, functioning as an intermediary. To apply leverage, Mpungwe would threaten to remove Tanzania as the mediator. He also sought to pressure the parties by threatening to identify one party to the international community as blameworthy. As Tanzania was the only neutral regional party, and a negotiated settlement significantly depended on its involvement, the former warning carried considerable weight. Mpungwe specified deadlines and appealed to the international and regional community who provided substantial and persuasive diplomatic and international pressure. (As Chapter Nine will detail further, the leverage applied in the Rwanda case differs substantially from that levied by Mitchell in the case of Northern Ireland. In Rwanda, negative leverage prevailed, whereas in Northern Ireland numerous economic incentives presented a package of positive leverage). Finally, when an 164 agreement had been reached and finalized on August 4, 1993, Mpungwe would strengthen commitment by coordinating, prior to its announcement, its immediate recognition by the international community. The statements of support from the regional community lent the new, sometimes tense, and tentative agreement an aura of international authority (Grunfeld, 2007: 74). Though the Organization of African Unity (OAU) previously refrained from involvement in the internal politics of its member countries, from 1992 to 1993 it worked in concert with Tanzanian mediation efforts, helping to facilitate talks and working particularly on maintaining and renewing commitments to ceasefires. Their effort was helped by the fact that Ugandan President Museveni was chair of the OAU at the time and supported a diplomatic resolution of the conflict; the new secretary-general of the OAU, Salim Ahmed Salim, a Tanzanian, was extremely interested in the organization taking a lead in resolving African conflicts; RPF and FAR both regarded the OAU as a neutral, nonpartisan entity. Moreover, the United States was supportive of these efforts and provided an initial $3-4 million for the Neutral Military Observer Group (NMOG) force that was under OAU control (Gribbin, 2005: 76). France: Playing Both Sides Symptomatic of the Fashoda Syndrome, the French government perceived the English speaking RPF (and Anglophone Uganda) as a threat to its Francophone hegemony (Barnett, 2003: 55). Along these lines, historian Gerard Prunier suggests that such dedicated French involvement was due to a long-standing fear of being out- 165 numbered by the “Anglo-Saxons.” Since before the Fashoda crisis in 1898, when the British eliminated the French presence in the upper Nile, French anxiety about losing status and power to the Anglo-Saxons had been high; any slip in French status or power (particularly in Africa) was seen as part of an Anglo-Saxon conspiracy. Therefore, Barnett explains that France saw the English speaking Uganda government as supporting the RPF in one more attempt to topple a French-speaking country (Barnett, 2003: 56). Though there is some critical divergence from this reading, France’s involvement is generally attributed to quasi-romantic ideas relating to colonialism and civilizing natives long-rooted in the French republic and were used to justify a continued military activity, much to France’s benefit. Despite the fact that Rwanda had never been a French colony, Francois Mitterrand was supportive of the Habyarimana regime, pursuing a dual policy of supporting the government militarily, at the same time promoting a negotiated settlement (Adelman & Suhrke, 1996: 29). In his book The Rwanda Crisis: History of a Genocide, Prunier, a French academic and historian, elaborates upon the French involvement, writing that, “The French line of argument in defending renewed intervention in Rwanda was to insist that France has supported the Arusha negotiations which led to an agreement between the government and the opposition to create a transition cabinet” (Prunier, 1995: 177). The dual policy was principally the result of divisions within the French government, with the Africa Unit at the Elysee Palace taking the military approach by providing Habyarimana with troops, while the Ministry of Foreign Affairs preferred the 166 political effort of negotiations. President Mitterand’s son, in charge of the Africa Unit, 52 said immediately following the RPF invasion in 1990 that France would “send a few boys to help old man Habyarimana” (Prunier, 1995: 100). He expected the conflict to be short-lived, but the Africa Unit continued its military assistance as the conflict continued. The Ministry of Foreign Affairs was increasingly critical of this approach; as indicated by the Paris talks discussed above, it actively supported negotiations as the best solution. While it encouraged the negotiations, France was hardly a disinterested observer. The French response to the Rwandan conflict was to a large degree conditioned by the long-standing friendly relationship between French president Mitterand and Rwandan president Habyarimana. Totten et al. explain that through a secret consultative body located within the French president’s administrative domain, the so-called “Africa cell,” 53 President Mitterand was able to keep a close eye on the Francophone countries of Africa, providing economic and military assistance when they required it and thus helping to ensure the predominance of the French language (and, by extension, French influence) in these states (Totten et al., 2007: 380). Totten et al. assert that Habyarimana’s Rwanda had long cultivated close links to France, notwithstanding the fact that Rwanda had been a Belgian colony prior to independence in 1962. “It was through the Africa Cell, headed by Mitterand’s son Jean- Christopher, that French military advisers, technical experts, and large amounts of military equipment arrived in Rwanda throughout the 1990s” (Totten, Bartrop et al., 2007: 381). Such an investment in, and commitment to the future of a single, small 52 The Africa Unit was a part of the military. 53 Also referred to as the “Africa Unit.” 167 French-speaking country indicated the degree to which France was prepared to go “to ensure that the country in question remained within the French sphere of influence” (Totten et al., 2007: 381). As such, throughout the talks, the French government championed the GoR position, especially regarding the composition of the government and the integration of the military. Barnett explains that the French were ready to prop up a faltering dictatorship while asking for very little in return and very few questions in the process. France became intimately involved with almost all facets of Rwanda’s military expansion and training. Rwanda was becoming rapidly militarized; the army expanded several fold and the government became the third largest arms importer in Africa. (Barnett, 2003: 56) After the genocide, France would be brutally criticized for the material and psychological assistance it had provided to Habyarimana’s regime (Furley & May, 2006, 203); interest, as that of a powerful nation, served to destabilize the equilibrium among the members to the talks. Since France had propped up the government with military and economic aid, critics argued, France had in essence encouraged intransigence on the part of the GoR and provided a shield behind which the Hutu extremists were able to advance their vicious plan of genocide. Even more egregious, according to Wallice (2007), France was keen to defend its influence in Africa, even if it meant complicity in the genocide. France had considerable investment in the balance of power in Rwanda, with practically direct control through various native generals and admirals who worked with French political and military elite to protect French interests. According to Wallice, President Francois Mitterand once said, “in countries like that, genocide is not so important” (Wallice, 2007, xi). 168 United States: Reluctant Intervention The United States participated in the Arusha process only as a facilitator that provided technical assistance and, through US Assistant Secretary of State Herman Cohen, lent authority to the talks (Stettenheim, 228). Early US involvement focused on the development of a workable cease-fire and on helping the parties determine the issues that would need to be addressed in later rounds. During the N’Sele talks in March 1991, American state department representatives with observer status offered advice on the creation and implementation of the cease-fire and the structure of the Neutral Military Observer Group (NMOG). The creation of the latter was intended to ensure strict adherence by the two parties to the cease-fire and to monitor the implementation of the Peace Agreement (Mays, 2003: 95). From May 8-May 9, 1992 Cohen met with both the RRF and Yoweri Museveni in Kampala, Uganda at which time he underscored the conflict’s dire economic impact and the futility of continued military violence. Two days later he met with Habyarimana and opposition members in Kigali, Rwanda where he discussed the refugees’ right of return (Mays, 2003: 95). By 1992, US commitment was dramatically reduced due to significant involvement in several other ongoing negotiations in Angola, Mozambique, Ethiopia, Liberia, and Namibia. These distractions as well as a lack of strategic interest in the area led the US to significantly reduce its involvement. Despite its awareness of the military violence—as indicated by the human rights report—the US still played a small (but significant) role though Cohen, who focused efforts on maintaining an effective process, that is, on helping the parties reach a come to 169 an agreement about the specific points of contention. As suggested above, the United States generally functioned as an information resource in the time leading up to the talks, although it occasionally took a more active role. For example when the State Department’s Bureau of African Affairs deployed a military advisor, Lt. Col. Tony Marley, 1992 to salvage the clearly declining transition process and work toward a plan for integrating the RPF and GoR’s officer ranks. By the time the actual talks began on July 12, 1992 the role of the United States was limited to observer status along with other regional and international communities. Thus during the talks themselves, the United States played a very hands-off role, encouraging the RPF and GoR “to reach a mutually agreed settlement,” whereas, “A higher priority would also have meant more continuity, as facilitators would not have been switched from Arusha to other conflicts.” Finally, according to Stettenheim, U.S. participants “noted that the United States purposefully took a less active role so as not to undermine the Tanzanian and OAU mediators” (235). Actual Negotiations The Arusha talks lasted from July 12, 1992, until the signing of the Accords in August 1993. Under the sponsorship of the OAU, Tanzanian President Ali Hassan Mwinyi and his OAU Ambassador Ami Mpungwe, mediation took place between the RPF and the GoR. To serve as an additional source of leverage, representatives from the international community (France, Belgium, Germany, the United States) served as observers at a number of sessions. Regional countries such as Senegal, Burundi, Uganda, and Zaire were also present to express concerns about regional stability (Khadiagala, 170 2004: 72). During the course of the talks, the US extended support to the OAU, including one million dollars to go toward improving mediation and peacekeeping (Callaghan,1997: 112). A primary issue during the talks was the lack of agreement between the Habyarimana regime and the opposition parties within the delegation. Sellstrom and Wohlgemuth explain that Habyarimana continuously vetoed breakthroughs agreed to by Foreign Minister Ngulinzira that significantly lowered the former’s influence in the ruling MRND (Sellstrom & Wohlgemuth, 1996: 50). In the end, however, the final package of protocols included: o an amended version of the N’Sele cease-fire between the Government of Rwanda and the Rwandese Patriotic Front (RPF) (7/12/92) o principle and creation of the rule of law (8/18/92) o power sharing, the enlargement of the transition government (i.e., inclusion of the RPF), and the creation of a transition parliament (10/30/92 and 1/9/93) o an agreement for the reintegration of refugees and internally dislocated persons (6/9/93) o creation of a nationally unified army (i.e., merger of RPF and the FAR (8/3/93) (Sellstrom & Wohlgemuth, 1996: 6). The first session in July 1992 not only established a cease-fire but also specified the agenda for the talks and the implementation of the Joint Political Military Commission (JPMC). Under Article III of the final agreement, it was accepted that “the verification and control of the cease-fire shall be conducted by the NMOG under the supervision of the Secretary General of the OAU,” while section (3) of the same article provided that the “report any violations of the cease-fire to the Secretary General of the OAU and a Joint Political Military Commission (JPMC)” (Adelman and Suhrke, 2000: 171 118). The JPMC consisted of a Rwandan government delegation headed by a Rwandan ambassador and an RPF delegation under the leadership of the RPF general coordinator. The JPMC was a forum that ran parallel to the Arusha talks to address military issues, especially cease-fire violations, and was attended by observers from Burundi, Tanzania, Uganda, Zaire, Belgium, France, and USA. Although the Commission had no arbitration authority, it established itself as an important safety valve for releasing tensions. It did this by serving as a place to meet but not to negotiate, a place where complaints could be aired, where the neutral military observer group could lodge complaints or accusations of violations, and in general where problems could be addressed without interfering with the process of negotiations. According to Adelman and Suhrke, the JPMC became effectively a “second channel through which the thorny issues of the Arusha process could be hammered out” (137). The session on the rule of law, which was held on August 18, 1992, dealt with background conditions in Rwanda, which the RPF and the opposition saw as the roots of the country’s instability. In particular, the protocol sought to define the fundamental rights of the people, including the freedom of expression and the right to life as articulated by the UN Charter and other internationally recognized fundamental documents. For the RPF, pluralism represented not simply multipartyism, but also a political culture capable of sustaining debate and dialogue. Along these lines, the RPF stressed that the Rwandan people should be indivisible and not labeled by ethnicity. The RPF viewed the principles articulated in the rule of law protocol as the theoretical 172 underpinnings on which all subsequent agreements should be built and which the parties should reinforce (Jones, 1995a: 4-7). Negotiations on the creation of the new government and the transitional institutions proved extremely contentious. They lasted from October 30, 1992 to January 9, 1993 and involved close to 60 days of negotiations. They ultimately resulted in an agreement with three key ingredients: one, the extremist Coalition for the Defense of the Republic (CDR) party was excluded from participation; two, the power of the government shifted from the presidency to the parliament, with the MRND relegated to a minority position; and three, a majority vote was defined as requiring the concurrence of four parties despite the mention of consensus rule. (Sellstrom & Wohlgemuth 1996: 49) Not surprisingly, Habyarimana reacted strongly to his party’s reduction of power and responded by issuing his notorious “pieces of paper” remark about the protocols. Facilitators worked hard to bring the president back to the table by seeking to convince him that he would in fact have a chance to regain power through free elections. Perhaps seeing an opportunity to temporize, Habyarimana returned to negotiations. Free elections were to occur at the end of the transitional period, which was to last five years and a commission was to draft a new constitution that would have to be approved by a referendum (“RPF Statement Names New President,” 1994). The new constitution, which would not be drafted for another eight years, was the work of a commission headed by Titus Rutaremara (“Rwanda: Plans Under Way to Draft Historic New Constitution,” 2001). In 2003 Rwandans finally voted on the referendum to the constitution, which stipulated that one party could not “control more than half the Cabinet posts, even if it wins an absolute majority in parliamentary elections -- and the president, 173 prime minister and president of the lower house of the two-chamber parliament cannot belong to the same political party” (Ngowi, 2003: A16). The most significant barrier that stalled talks for weeks toward the end of 1992 was the RPF’s demand that the CDR be excluded from the government. To deal with this issue, Mpungwe separated the parties, and along with members of the observer group, strenuously lobbied the RPF to allow the CDR to participate in the government. As articulated by one Western diplomat, “the logic was that it was better to have the CDR inside the tent than outside, threatening to burn it down” (Jones , 1995a: 6). The RPF, however, remained extremely determined because it justifiably believed that the extremists were bent on destroying any power-sharing arrangement. This reasoning ultimately prevailed over Mpungwe, with the agreement of the Hutu opposition to Habyarimana. Several factors can be attributed to the government’s eventual decision to sign the Arusha Accords. First, the opposition was sympathetic to the RPF viewpoint. Second, the RPF had made it clear that it would readily resume hostilities. Third, in the eyes of the international community, Habyarimana’s commitment to peace would be dubious if he scuttled talks by insisting on the inclusion of a sectarian extremist group like CDR. Prunier explains that President Habyarimana had consented to sign the Arusha peace agreement not as a genuine gesture marking the beginning of democratization in Rwanda, but as a “tactical move calculated to buy time, shore up the contradictions of the various segments of the opposition and look good in the eyes of foreign donors” (Prunier, 1995: 195). Not surprisingly, Habyarimana was dissatisfied with the terms and while publically 174 agreeing to them, he privately began elaborate plans for an anti-Tutsi propaganda campaign to be executed via radio broadcasts (Callahan, 2007: 113). During and after the signing of the agreement, the Tanzanian facilitators continuously attempted to encourage the supporters of Habyarimana by expressing that he would be guaranteed a fair shot in free elections. To some extent Habyarimana took to heart the facilitator’s advice about the possibilities of maintaining power through the new political process. Adelman and Suhrke explain that he began to manipulate the opposition to change the distribution of power in the new transitional government. In late January, 1993, however, a foreboding event took place in which over 300 Tutsi were massacred in the community of Bagogwe, home to Habyarimana’s clan (Adelman & Suhrke, 1996: 40). (Notably, Bagogwe had been the site of reprisal killings of hundreds of Tutsi in January 1991; during inquiries several years later, mass graves were found in the mayor’s garden) (“Rwanda Justice Files,” 2001). The negotiations on the integration of the armed forces were the last significant obstacle to the conclusion of the Accords and were completed in the summer of 1993. The main difficulty of the negotiations focused on the proportions each side would gain in the new integrated army. While the government entered the negotiations proposing approximately a 20% share for the RPF as commensurate with their ethnic proportion of the population, the RPF was committed to an equal fifty-fifty split (Adelman & Suhrke, 1996: 43). Mpungwe’s mediation was guided by a desire to avoid the situation that had occurred in Angola where each faction had retained its own military force (Adelman and Suhrke, 2000: 134). He desired a reconciliation of the security concerns of the two sides. 175 But as Jones (1995) describes them, the talks became needlessly “democratized,” with too many voices and opinions weighing in; had the conflict been dyadic in nature (as with the case of Ireland) it would have had a greater chance of being resolved. Certainly, however, the exclusion of extremists would have presented its own set of problems—but their influence was far greater than Mpungwe appreciated. Mpungwe was not sympathetic to the government’s initial offer of 20%, as it was not in line with his dedication to the rule of law and de-emphasis on ethnicity (Jones, 1995a: 7). He separated the parties and declined to express the GoR position to the RPF because he felt it was unreasonable and would exacerbate the differences between the GoR and the RPF. Eventually, the formula developed by Mpungwe for settlement involved a horizontal and vertical integration of the security forces. The horizontal balance was achieved by giving the MRND control over the military and the RPF control over the gendarmerie. Vertical checks were instituted by requiring that the two most senior officers at each command level be from the opposing sides. For example, if the commander was from the FAR, then the deputy commander would be an RPF officer (Jones, 1995a: 7). The settlement split the command and control functions 50-50 and gave the GoR a 60-40 advantage in troop composition (Jones, 1995a: 7). When Foreign Minister Ngulinzira agreed to the above formula, Habyarimana flatly rejected it and in fact recalled Ngulinzira. Ngulinzira refused to leave and the negotiations continued for a period during which he had no authority. Ultimately, Habyarimana reshuffled the cabinet 176 and Former Defense Minister and MRND party member, James Gasana, was put in charge. Frustrated by Habyarimana’s obstinance, the RPF turned around and demanded an even greater increase of representation in the troop levels. After considerable lobbying by Mpungwe and the observers, the RPF returned to its earlier position and a final agreement was reached—command and control functions were split fifty-fifty (Dallaire, 2004: 123). According to Jones (1995), the agreement was in many ways a “victory deal” for the RPF but still diplomats present to the deal-making characterized the Accords as one of the best deals ever brokered in Africa (Melvern, 2000: 52). For his part, Mpungwe described them in the following way: In spite of its unfortunate collapse, largely caused by the prolonged delays in its implementation, the Arusha Peace Process for Rwanda was a great success, a good model … The conflict was deep-rooted . . . and was heavily characterized by profound fear, mistrust, contempt, insecurity and a strong desire for recognition and self-esteem by ethnic sections of the Rwandese population… Our facilitation team had put together a comprehensive mediation model that was specific to the deep-rooted conflict that characterized Rwanda. (“Nature of the Mediation,” 2002) Mpungwe’s congratulatory characterization is peculiar in light of Habyarimana’s lack of genuine commitment to the process. Though its efforts were sincere and dedicated, the Tanzanian mediation team appears to have lacked adequate insight and understanding that the MRND, as Mthembu-Salter put it, “could ever be induced to move from a zero sum conception of the conflict, according to which the only alternatives were absolute victory or absolute defeat, towards an acceptance of the possibility of compromise” (“Nature of the Mediation,” 2002). 177 A formidable media campaign conducted by the extremists throughout Rwanda, many of whom were supported by the president behind closed doors, effectively exploited events to provoke hatred and fear and sow ethnic division between the Hutu and Tutsi. As early as July 8, 1993, Radio Television Libre des Mille Collines (RTLMC) began broadcasting rhetoric vehemently opposed to the Arusha Accords that were not yet complete, and espousing a rabid “Hutu Power” ideology. It soon broadcast a “kill or be killed” message and encouraged ethnic cleansing and recleansing areas (Burkhalter 1994). This Hutu Power message would come to epitomize and galvanize rabid anti- Tutsi thinking and manifested as a dissident movement within each of the main opposition parties, mostly Mouvement Démocratique Républicain (MDR) and Parti Libéral (PL). Some of the most extreme anti-Tutsi rallying was in the Convention pour la Défense de la République (CDR), which had been excluded from the Arusha talks and the broad-based transitional government itself. In July of 1993, the opposition parties became splintered by personal power struggles that dissolved into ideological and regional divisions. Habyarimana exploited these divisions among the opposition and used them to justify delaying the implementation of the transitional government. Nonetheless, in August, 1993—13 months after negotiations began— the Accords were signed by Habyarimana and RPF chairman, Alexis Kanyarengwe, and called for the inauguration of the transitional government to be led by Hutu politician, Faustin Twagiramumngu, within the next 37 days. Moreover, the caretaker government of Agathe Uwilingiymana, a Hutu political figure, was to maintain control until the arrival 178 of the UN peacekeeping and implementation force in November 1993 (Sellstrom & Wohlgemuth 1996: 51). Implementation Numerous factors internal and external to Rwanda led to the devastating failure of the implementation of the Arusha Accords. Even during negotiations, Habyarimana was still trying to have it both ways: presenting an image of compliance with the Accords while undertaking all necessary efforts to hold on to his power. To this end, he vilified the RPF through propaganda, stirring animosity and underscoring the old ethnic divisions. Habyarimana developed a rural militia and also promoted anti-Tutsi ideology to create a climate of fear that would allow him to maintain power. Although the Accords prescribed a dramatic reduction in the extremists’ power, these groups still held control of the military and administrative machinery of the government during the negotiations and the crucial implementation phase. The akazu formed the core of the hard-liners and had members in top positions in the military, the CDR, and the MRND. They organized and trained local militias, the infamous interahamwe, which consisted of disenfranchised youth inculcated with racist ideology and ready to use any means to defend the power position of Habyarimana’s party, the MRND (Mugabowineza, 2005: 52). According to Adelman and Suhrke, the extremists targeted unemployed young men, hoping to arm one in ten peasants (1996: 31). The extremists, including Habyarimana, also developed extermination lists of Tutsi and moderate Hutu. Although 179 these efforts intensified during late 1993 and early 1994, the groundwork for genocide had begun even while the Arusha talks were advancing (Adelman & Suhrke, 1996: 31). Disastrous events in neighboring Burundi created a trigger that led to the failure of implementation of the Arusha Accords. On November 21, 1993, Tutsi extremist military officers assassinated the newly elected Hutu president, Melchior Ndadaye of Burundi, and began a campaign of repression that resulted in approximately 50,000 Hutu and Tutsi deaths and about 300,000 Hutu refugees fleeing to Rwanda (Prunier, 1995: 199). When combined with the installation of a 600-man RPF battalion in Kigali, the Hutu extremists were able to paint a dark picture of an imminent Tutsi usurpation of power. Since the transitional government was not yet in place in Rwanda, the only governmental institution with real power was Habyarimana’s presidency. Although the president had clearly been hesitant about the Accords, Mpungwe believed that the assassination of Ndadaye allowed the extremists to decisively commit Habyarimana to oppose the Accords (Prunier, 1995: 199). Regardless of Habyarimana’s position, numerous observers believe the assassination of Ndadaye allowed the extremists to develop a broad anti-Tutsi constituency of Hutus that otherwise might have still supported the transitional government. United Nations: A Vague Mandate The United Nations had been involved in the Rwanda conflict beginning with the various summits that took place during pre-negotiation phase. Its most important role, however, was peacekeeping during the implementation of the Accords. On October 5, 180 1993—two months after the Accords had been signed— the UN Security Council approved the United Nations Assistance Mission for Rwanda (UNAMIR), a peace- keeping force of 2,500 members, who began arriving on November 1, 1993. According to General Romeo Dallaire, force commander of UNAMIR, the best means of controlling the extremists was the quick implementation of the transitional government and the integrated security forces that could check their agitation (Dallaire, 2004: 89). This plan did not come to fruition. Extensive critical analysis has taken up the issue of the UN’s dismal failure at peacekeeping, citing among other reasons, a “disjuncture” with the OAU. In Stettenheim’s words: Under pressure from the French, the United Nations had rebuffed efforts by the OAU to take an active role in the implementation and monitoring of the Arusha Accords and had assumed primary responsibility for what would be a completely ineffective peacekeeping force. (233) The UN thus bypassed the opportunity to coordinate efforts with the OAU, an organization that in fact had significant political will to bring peace to the region. From his first-hand experience, Dallaire, a Canadian General in charge of the UN peace- keeping force in Rwanda, writes that, “There was a void of leadership in New York. We had sent a deluge of paper and received nothing in return; no supplies, no reinforcement, no decisions… no one was interested in Rwanda, and now, because of the escalating risks, they were even less interested” (Dallaire, 2004: 290). UNAMIR’S responsibilities were: to contribute to the security of the city of Kigali inter alia within a weapons’ secure area established by the parties in and around the city; to monitor 181 observance of the ceasefire agreement;…to monitor the security situation during the final period of the transitional government’s mandate, leading up to elections. . . [and] to investigate and report on incidents regarding the activities of the gendarmerie and police. (qtd. In Bulkhalter 1994: 45). According to Bulkhalter (1994), government sympathy on the part of Jaques-Roger Booh Booh, special envoy to Rwanda and head of UNAMIR, led to deliberate misrepresentation of the violence and to the failure to fulfill many of the above-named responsibilities. Thus, an ineffective peacekeeping force of troops from the United Nations and the OAU failed to support the peace process by providing the needed security and stability. One critic sums up the prevailing opinion: “Given the lack of a clear mandate, the UN forces, present in Rwanda to monitor the implementation of the Accords, did not do anything to help stop the massacre” (Mugabowineza, 2005: 60). Furthermore, while numerous sources of diplomatic, military, and economic forms of leverage were applied leading up to the Accord, no such pressure was established to guarantee follow through and compliance. The plan was that the UN would oversee the transition. With hostilities resumed between Tutsi and Hutu, the UN saw the violence as an extension of the earlier civil war and not a situation in which there was actual peace to keep—thus there was no justification for a peacekeeping mission as such (Mugabowineza 2005). According to Prunier, the UN soldiers were “not even in a position to protect themselves, and the massacres were taking place right before their eyes while they had neither a mandate nor equipment enabling them to do anything” (1995: 204, 234). 182 On January 8, 1994, the convening of the Broad Based Transitional Government (BBTG) was blocked by deadlocks within MDR and the Liberal Party (PL). Attempts to proceed on February 10 without the PL (due to its troubled relationship with the MDR), were aborted by the assassination of the PSD leader Felicien Gatabazi on February 21, 1994 and the reprisal killing of Martin Bucyana of the CDR on February 23, 1994 (Jones (b), 1995: 12). General Dallaire explains that “the next morning, [Hutu] moderates and extremists both took to the streets of Kigali. The extremist media, controlled by Habyarimana, immediately spun headlines celebrating the killing of Gatabazi as a victory against Hutu traitors; the Interahamwe was much in evidence” (Dallaire, 2004: 99). Although the United Nations was able to reestablish order after riots in Kigali, Rwanda was hurtling towards the precipice. UNAMIR lacked much of its essential equipment and was short on even basic supplies such as ammunition (Adelman & Suhrke, 2000: 36). General Dallaire explains that he had “soldiers and officers die in theatre, and had several others evacuated because of injuries sustained in the line of duty, or because of sickness due to lack of medical supplies” (Dallaire, 2004: 404). Furthermore, the UN mission was undermined by a ineffectual mandate that prohibited any involvement should hostilities erupt. A special representative of the UN Secretary General met with the RPF on March 1, 1994 to try to put the peace process back on track. Tanzania, the Arusha facilitator, also kept up its efforts to persuade the two sides to implement the agreement. Even Russia, which had never been involved in the situation, was called to assist in the attempt at diplomatic pressure, and issued a statement 183 on March 29, “regretting the failure of the transitional institutions to assume office” and promising to “support all efforts towards that end” (Prunier, 1995: 209). On April 3, 1994, a group of Western ambassadors and General Dallaire, met with Habyarimana and stressed that the Accords must be implemented (Kuperman, 1996: 230). Three days later at a regional summit, the secretary-general of the OAU harshly reprimanded Habyarimana and reiterated the same point. The difficulty was that Habyarimana may have been a prisoner of his own party and the extremists he has fostered and used. He too would become a victim of the very Hutu extremists that had supported him: As he returned from the summit, on April 6, 1994 the plane carrying Habyarimana and Burundi President Cyprien Ntaryamira was shot down by a missile as it approached the Kigali airport (Dallaire, 2004: 221). This event is generally characterized as the tipping point, after which anti-Tutsi hatred was fully unleashed. In Shake Hands with the Devil: The Failure of Humanity in Rwanda, Dallaire recounts the moments after the plane went down. He writes that “A plane had crashed but no one could confirm that it was Habyarimana’s. Presidential Guards and members of Rwanda’s Government Forces (RGF) Para-Commando Battalion from Camp Kanombe were running wild at the airport, threatening people with their weapons, and the UN observers had gone into hiding” (221). Even before the missile was fired, the interahamwe militias had set up roadblocks around the city and had begun their killing. Genocide began at once, as Hutu extremists undertook a door-to-door effort to seek and kill members of the transitional government, RPF collaborators, and anyone who was pro-Tutsi (Mugabowineza 2005: 70). According 184 to an unclassified memo sent to the Clinton Administration, the American embassy in Kigali immediately sent a dispatch to Secretary of State, Warren Christopher: “If, as it appears, both presidents have been killed, there is a strong likelihood that widespread violence could break out in either or both countries…” (qtd. in Mugabowineza, 2005: 71). On April 7, Agathe Uwilingiymana, the appointed prime minister, was murdered in her home along with then Belgian peacekeepers who were tortured before being killed (Jones, 1995b: 13). This event led to the withdrawal of the European component of UNAMIR two weeks later, which meant a reduction of troops from 2,500 to 270. Prunier characterizes this evacuation as a “disgrace,” as the Tutsi knew they were left to face “certain death” (Prunier, 1995: 235). Over the next three months, the Hutu extremists cut a genocidal swath through the countryside (Prunier, 1995: 236). Before the RPF forces were able to take control of the country, nearly one million people had been brutally murdered, and approximately two million had escaped over Rwanda’s borders in the largest and fastest exodus of displaced persons in modern history (Jones, 1995b: 14). As the events of the genocide unfolded, response from the United States was negligible at best. Addressing the senate on April 10, 1994, Republican Senate Minority Leader Bob Dole summed up the US response in declaring: “I don’t think we have any national interest there. The Americans are out, and as far as I am concerned, in Rwanda, that ought to be the end of it” (qtd, in Mugabowineza. 2005:66). 185 What happened in Rwanda provides an egregious example of the coexistence of different social groups or castes metamorphosing into an ethnic problem leading to extreme violence. Alain Destexhe explains that, “archaic political divisions were progressively transformed into racial ideologies and repeated outbreaks of violence resulting from the colonial heritage which was absorbed by local elites who then brought it into the political arena” (Destexhe, 1995: 47). He elaborates that it was the “ethnic classification registered on identity cards introduced by the Belgians that served as the basic instrument for the genocide of the Tutsi people who were ‘guilty’ on three counts: they were a minority, they were a reminder of a feudal system and they were regarded as colonizers in their own country” (Destexhe, 1995: 47). This chapter has partially described the mediation process in Rwanda and what factors affected the different parties involved. Considering the motivations and biases of each party creates a clearer picture of why the mediation process had so much trouble succeeding. Moreover, the comprehensive analysis on the roles and influences of the mediators involved helps clarify why the implementation of the Arusha Accords failed so miserably. The next chapter will provide a more detailed understanding of where exactly the mediation process went wrong and why. Applying this case study to Moore’s 12 stages will clarify what point in the mediation process accounts for the gradual decline in the talks and ultimately to the disintegration of the peace process. 186 CHAPTER EIGHT: APPLYING MOORE’S 12 STAGES TO THE TWO CASES The preceding chapters have shown that mediation is intricate and complicated; it can stretch over many years and involve many parties and people. Christopher Moore provides a valuable tool for understanding and organizing the many details and intricacies of negotiation, explaining that a mediation process can be deemed successful if it successfully completes all of the 12 stages 54 ( Moore, 2001). Using these stages as a lens through which to examine the mediation process in Ireland and Rwanda, I seek to show how Moore’s 12 stages—including the degree to which each was completed reveals early indicators of the ultimate outcome of each. As such, application of the 12 stages will show why mediation in Northern Ireland was successful, while in Rwanda it was not. In doing so, I hope to contribute to a body of knowledge that will result in better approaches to conflict management and improved future mediation. In his book The Mediation Process: Practical Strategies for Resolving Conflict (1996), Moore defines mediation as "the intervention in a standard negotiation or conflict of an acceptable third party who has limited or no authoritative decision-making power but who assists the involved parties in voluntarily reaching a mutually acceptable settlement of issues in dispute" (1996:15). He differentiates among three different mediators: social network mediators, authoritative mediators, and independent mediators. 54 Moore asserts that the 12 steps can be completed in any order, but all 12 have to be completed in order to deem a mediation process successful. If all 12 steps are successfully completed the outcome is always positive. 187 In the case of Rwanda, Zairean president Mobutu and even the Tanzanian mediator, Mpungwe, would be most clearly categorized as social network mediators, who are generally part of the community of the disputants, in this case the regional community represented by the Communauté Economique de Pays des Grand Lacs (CEPGL). Though not necessarily neutral, the social network mediator must be fair and often sees his role as part of a long-term commitment to stabilizing social relations, which is precisely how Mpungwe regarded his position (Moore, 1996). In contrast, George Mitchell adhered to a role much more like that of an independent mediator, a neutral and impartial entity who guides the parties to mutually acceptable solutions. According to Moore, a mediator exercises power by pressuring the parties to arrive at a solution about which he or she expresses a neutral opinion. Moore analyzes the mediation process that follows the designation of a mediator, breaking the process down into 12 stages, each with its own set of substages (Table 1.4). According to Moore, mediators may enter disputes either at the request of the parties, or through appointment by an authoritative third-party. Once appointed, the mediator should abide by the following stages: (1) Making Initial contracts with the disputing parties; (2) Selecting a strategy to guide mediation; (3) Collecting and analyzing background information; (4) Designing a detailed plan for mediation; (5) Building trust and cooperation; (6) Beginning the mediation session; (7) Defining issues and setting the agenda; (8) Uncovering hidden interests of the disputing parties; (9) Generating options for settlement; (10) Assessing options for settlement; (11) Overseeing Final Bargaining and (12) Achieving formal settlement. 188 Table 1.2 Moore’s 12 Stages to Mediation Stage 1: Initial Contacts with the Disputing Parties • Making initial contacts with the parties • Building credibility • Promoting rapport • Educating the parties about the process • Increasing commitment to the procedure Stage 2: Selecting a Strategy to Guide Mediation • Assisting the parties to assess various approaches to conflict management and resolution • Assisting the parties to select an approach • Coordinating the approaches of the parties Stage 3: Collecting and Analyzing Background Information • Collecting and analyzing relevant data about the people, dynamics and substance of a conflict • Verifying accuracy of data • Minimizing the impact of inaccurate or unavailable data Stage 4: Designing a Detailed Plan for Mediation • Identifying strategies and consequent non-contingent moves that will enable the parties to move toward agreement • Identifying contingent moves to respond to situations peculiar to the specific conflict 189 Table 1.2: Continued Stage 5: Building Trust and Cooperation • Preparing disputants psychologically to participate in negotiations • Handling strong emotions • Checking perceptions and minimizing effects of stereotypes • Building recognition of the legitimacy of the parties and issues • Building trust • Clarifying communications Stage 6: Beginning the Mediation Session • Opening negotiation between the parties • Establishing an open and positive tone • Establishing ground rules and behavior guidelines • Assisting the parties in venting emotions • Delimiting topic areas and issues for discussion • Assisting the parties in exploring commitments, salience, and influence Stage 7: Defining Issues and Setting an Agenda • Identifying broad topic areas of concern to the parties • Obtaining agreement on the issues to be discussed • Determining the sequence for handling the issues Stage 8: Uncovering Hidden Interests of the Disputing Parties • Identifying the substantive, procedural and psychological interests of the parties • Educating the parties about each other’s interests Stage 9: Generating Options for Settlement • Developing and awareness among the parties of the need for options • Lowering commitment to positions or sole alternatives • Generating options using either positional or interest-based bargaining 190 Table 1.2: Continued Stage 10: Assessing Options for Settlement • Reviewing the interests of the parties • Assessing how interests can be met by available options • Assessing the costs and benefits of selecting options Stage 11: Final Bargaining • Reaching agreement through either incremental convergence of positions, final leaps to package settlements, development of a consensual formula, or establishment of a procedural means to reach a substantive agreement Stage 12: Achieving Formal Settlement • Identifying procedural steps to operationalize the agreement • Establishing an evaluation and monitoring procedure • Formalizing the settlement and creating an enforcement and commitment mechanism Analysis: Northern Ireland and the 12 Stages This section will apply the 12 stages to the IRA case, showing that all 12 stages were fulfilled. Stage One, “Initial Contacts with the Disputing Parties,” played a critical role in the case of the IRA. The process of initializing contacts and building credibility began four years before the culmination of the Agreement, in 1994 when the Clinton administration intrepidly welcomed Gerry Adams to the United States, granting him a one-day visa to address a peace conference in New York. Again, in March 1995, Gerry Adams was permitted to take a two-week tour of the US, including a personal visit with 191 the president. This trip raised Sinn Fein over $1.3 million and furthered the process of successful mediation (Moloney, 2002: 437). In welcoming Adams to the United States, Clinton already fulfilled the requirements of Stage One, not only making contact with a key disputant but also “promoting rapport” and acknowledging Adams as a trustworthy and accountable player. With this move, Clinton also showed U.S. commitment to the peace process, which went a long way toward earning Northern Ireland’s confidence. While it tremendously angered the British government and the Protestants of Northern Ireland, Clinton’s invitation was part of a larger plan to draw Sinn Fein into the political mainstream and thus bolster its own credibility (Moloney, 2002: 437). (As the case of Rwanda demonstrates, isolating extremists—as with the CDR— can be hazardous). Very diplomatically, Clinton’s strategy tied administration overtures to Sinn Fein to that organization’s commitment to the peace process. The subsequent successes of Gerry Adam’s visits would prove to be important factors in securing the IRA cease-fire and the eventual participation of Sinn Fein in the peace process. The visits attracted a great deal of media attention, and provided examples to Sinn Fein and the IRA of the political rewards that could be gained by becoming legitimate players in the political process. It also gave the IRA and Sinn Fein confidence and access to a much wider audience. Lastly, the visas convinced them that the United States was committed to seeing the peace process through to a successful conclusion and was willing to seriously consider their views (Moloney, 2002: 438). 192 Clinton’s effort with Adams represented a new hands-on approach that was a far cry from traditional hands-off policies of earlier administrations, which considered problems in Northern Ireland to be an internal British matter. By directly involving the United States, Clinton internationalized the issue. A cease-fire signed by the IRA in August, 1994 and by the Loyalists in October 1994, built the disputants’ faith in moving away a military stalemate toward finding a political solution to the Troubles. While there were many opponents to President Clinton’s new strategy, the administration moved forward with the hope of finding success and to this end, began to coordinate getting the parties to the negotiation table. Clinton’s designation of US Senator George Mitchell to chair the international body to decommission illegal arms in Northern Ireland went a long way toward building further rapport, as Mitchell had earned credibility in Ireland for his work as Special Advisor to the President and Secretary of State on Economic Initiatives. Mitchell was a central part of Clinton’s strategy to guide the mediation. As an American, Mitchell helped to sustain US support for the peace process through his appointment, and his familiarity with both the conflict and its prominent figures earned their trust. His genuine enthusiasm for the undertaking furthered the requirements of Stage One—to build rapport and increase commitment to the mediation—and his demeanor and experience helped to establish an amenable approach to the process of confidence building. Stage Two, “Selecting a Strategy to Guide Mediation,” was duly met with the establishment of The International Body of Decommissioning in 1995. The plan was supported by British Prime Minister John Major and Republic of Ireland’s John Bruton 193 and had the financial support of the European Union and the United States. Mitchell chaired the group of mediators that included Harri Holkeri, former Prime Minister of Finland, and General John de Chastelain, Canada’s Chief of Defense Staff. In order to better understand the issues and to continue to show his dedication to resolving the conflict, President Clinton traveled to Northern Ireland and the Irish Republic in November of 1995. The fact that the most powerful world leader traveled to Northern Ireland gave a big push to the peace process. While this gesture definitely showed President Clinton’s commitment, it also gave him the chance to “collect and analyze information,” Moore’s third stage of mediation. John Burton claimed that President Clinton’s visit provided the confidence to begin compromising on a solution. The visit prompted the British and Irish governments to begin the process of decommissioning some terrorist arms through the so called “twin-track” approach. While it proved extremely important, Clinton’s trip was not the only method used to collect information. In order to better analyze the situation, President Clinton “spent many nights on the phone with major players in the peace process. He prepared economic aid packets and private funding opportunities from American firms in order to push the mediation process forward” (Mitchell, 2001: 49). Due to his continued quest for information and data to make the peace process successful, President Clinton was able to successfully influence both sides. At the same time, Mitchell’s group was preparing the Independent Body Report for the British and Irish governments, which would play a major role in fulfilling the 194 requisites of Moore’s second stage—coordinating approaches and crafting an approach; the Report also made progress toward Stage Three of mediation—collecting information. The report issued by the International Body on January 22, 1996 suggested a compromise position of decommissioning in phases during the negotiations. Mitchell described his work in this phase as consulting with as many people with as many different points of view as possible” with six weeks spent “shuttling between Belfast, London and Dublin. . . [meeting with] government officials, political leaders, business and religious leaders. . . listening, asking questions, taking notes.” (Qtd. In Jacobsen 193) Mitchell’s description of this process resembles Moore’s own characterization that "The mediator's central task during this stage is to integrate and understand the elements of the dispute: people, dynamics, issues, and interests" (1996: 132). In order to generate mutually agreed upon settlement options, Mitchell had to do extensive work with each party. Identifying strategies and consequent non-contingent moves that would enable the parties to move toward agreement was needed to complete Moore’s fourth stage. It had become obvious that each government had its own political position and political culture and the prospects of a disintegrating Conservative government in Westminster was bound to create an unstable environment from the point of view of the various Northern Irish political parties. Moreover, it was important to be careful with the conflicting wishes of the Northern Ireland parties. As Moore points out, however, the critical aspect here is that these issues were being articulated and the likelihood of success was higher because of that. 195 The following stages involved cooperation and beginning the mediation sessions. Stage Five—“Building Trust and Cooperation” was absolutely critical to Mitchell’s effort because it involved the decommissioning of weapons. To get paramilitary groups to relinquish their weapons was, in their eyes, asking them to give up their leverage and safety. But Britain held fast to decommissioning as prerequisite to peace negotiations. With such strongly held but opposed positions “handling strong emotions,” a subcategory to Stage Five, was critical to Mitchell’s work. Moore describes this activity as “conciliation.” He explains, "Conciliation is essentially an applied psychological tactic aimed at correcting perceptions, reducing unreasonable fears, and improving communication to an extent that permits reasonable discussion to take place and, in fact, makes rational bargaining possible" (1996: 161). Because the disputants held such heightened distrust of each other, Mitchell’s solution was to have decommissioning and talks occur at the same time, but John Major still resisted. In response to this opposition, Mitchell drafted a set of six commitments to democracy and nonviolence—the so-called Mitchell Principles, a code of conduct that sought to clarify goals, build confidence, and, most importantly establish much-needed trust. The six rules struck the right the tone for the negotiations. In brief, they included a commitment to “peaceful means” and disarming paramilitaries, consent to supervision by an independent commission, renunciation of force as a form of leverage, agreement to abide by terms, and a pledge to cease “punishment” killings (Jacobsen 2003: 195). 196 Through these six principles, Mitchell strove to distinguish among acceptable options, necessary elements, and trade-offs. What, then, was the incentive for these parties to move into negotiations with this set of commitments? According to King, “First, the advantages from participating in the negotiations had to outweigh those presented by violence . . . Second, the Mitchell Principles had to be backed with a plausible threat of sanctions” (2000: 200). With a military stalemate occurring, these commitments helped to ensure that the parties approached the negotiations in good faith. These principles were backed with a credible threat of a complete withdrawal of public support. Agreeing to the Mitchell Principles provided the necessary security for the success of the talks (King, 2000: 202). Another part of the report, entitled “Further Confidence Building,” directly addressed the issue of trust. Submitted on January 22, 1996, the report represented a workable framework with rules and guidelines for productive mediation. With the Report, Stage 6, “Beginning the Mediation Session,” along with its indispensable qualities—“Establishing an open and positive tone” and “Establishing ground rules and behavior guidelines”—seemed secure. Despite the disappointing setback of the IRA’s Canary Wharf bombing in early February 1996, all party negotiations were set to begin in July with Mitchell as the official mediator. Mitchell describes his role in the following way: “I was in a position where people felt that I could be helpful and everyone has an obligation to be of help to others where possible (qtd. In Jacobsen, 2003: 197). By way of preparation, Mitchell drew upon his many years of experience as a senator and a judge and read widely about the history and culture of 197 Ireland. “Establishing ground rules and behavior guidelines” took two months, during which time Three Strands were designated. Strand One addressed the concerns of Northern Ireland’s internal parties; Strand Two was dedicated to North and South issues; Strand Three was concerned with negotiations between England and Ireland. Mitchell’s methodical approach to the mediation was matched by his positive attitude and the encouraging tone he set for the talks. Chairman de Chastelain recalls the culture of respect established through his colleague’s approach: It included calling on all participants to extend to others the same courtesies they expected to receive themselves; it allowed participants the maximum leeway in making their point, without interruption, and subject only to time limits imposed by the participants . . . The objective manner in which he handled his role, and the wise way in which he proposed solutions to contentious issues, quickly impressed all the participants, including those with earlier reservations about outside involvement. (qtd. in Jacobsen 2003: 202) Under Mitchell’s careful guidance, then, the International Body had already managed to establish an environment conducive to productive exchange. Though parallel decommissioning had met with insurmountable resistance from both sides, Mitchell managed to present the International Body as an objective and legitimate entity resoundingly committed to the peace process while adherence to his six principles “became the ticket of admission to the peace process and provided the security necessary for the success of the talks” (Greenberg, Barton & McGuinn, 2000: 201). In the seventh stage, Moore points to the importance of highlighting issues of concern, which he calls “Defining Issues and Setting an Agenda.” A Plenary Committee with Mitchell, Holkeri, and de Chastelain led this effort by convening seven of the negotiating parties. (The Sinn Fein Canary Wharf bombing had violated the cease-fire 198 thus precluding the group’s participation in the negotiation). Because of the struggles to establish rules for decommissioning, the Committee was not able to put the finishing touches on the agenda until mid-October, 1996. Talks continued through to the Spring of 1997 when Tony Blair was elected Prime Minister. The election of Tony Blair’s Labour government was the last move to ensure that negotiations would find a successful solution. Blair’s government enjoyed broad support and was not tied to Northern Ireland’s Unionists. The talks were making good headway. Moore’s eighth stage, “Uncovering Hidden Interests,” means uncovering information through direct and indirect procedures. Examples of direct procedures include testing, in which mediators listen to disputants and echo back what they have heard, and “hypothetical modeling” (243), in which the mediator makes suggestions that the parties then rank for satisfaction level. Direct procedures are more about questioning and brainstorming with the parties. Because “hidden interests” may be intentional or unintentional, this phase is one of discovery for disputants and mediator alike. According to Moore the mediator may guide the parties to articulate interests in a way that clarifies their position for others as well as for themselves. As violence by both the UUP and Sinn Fein interrupted the peace process several times throughout 1997 and into 1998, Mitchell had to redouble negotiation efforts and work around the various exclusions required by violations of the cease-fire. In the first months of 1998, Mitchell worked assiduously on Stage Nine, “Generating Options for Settlement,” during which there was constant back and forth assessment of the respective interests as he hoped to reach a conclusion by Easter. Each side had to make concessions and still see that they 199 would “win” by agreeing to the proposal. The decommissioning issue continued to stymie negotiations but Mitchell worked round the clock (Jacobsen, 2003: 209) Stage Ten, “Assessment of the Options for Settlement,” was met as Mitchell worked in direct consultation with Blair and Trimble on the terms of decommissioning, which he assured them would be articulated as soon as the Agreement was reached. Very quickly Stages 11 and 12 were met with as the historic Agreement was reached on April 10, 1998; six weeks later the parties presented the Agreement to the people of Ireland for their acceptance. On May 22, 1998, the people of Ireland overwhelmingly accepted the Agreement, officially ending the conflict. In the proposal were plans for a Northern Ireland assembly that would have a: (1) power-sharing executive, (2) new cross-border institutions involving the Irish Republic and, (3) a body linking devolved assemblies across the UK with Westminster and Dublin. Moreover, the Irish Republic no longer claimed the six counties that formed Northern Ireland. Strand one of the agreement established a democratically elected power-sharing assembly in Northern Ireland, including safeguards for cross-community participation. Under the second strand of the agreement, a Ministerial Council was responsible for bringing together whoever had executive responsibilities in Northern Ireland and the Irish government “on matters of mutual interest” (The Agreement, Strand 2). Under the third strand, a British Irish Council was created to promote bilateral cooperation between the UK and Ireland. The completion of each stage allowed for the positive progression of the peace process, ultimately culminating in a successful resolution. As an organizing tool. Moore’s stages allow close exegesis of why the mediation helped to resolve the conflict. 200 Analysis: Rwanda and the 12 Stages Even if a figure like George Mitchell had mediated the Arusha Accords, the events would not likely have unfolded as they did in Ireland. From the very beginning of the process, Habyarimana and the MRND elite were unwilling to renounce the decades- long prejudices and divisions that had menaced Hutu-Tutsi relations and thus corrupted the very spirit and letter of mediation. As I will enumerate, Habyarimana’s duplicity is but one (albeit significant) reason for Arusha’s failure; the RPF, the mediation and its final apportioning of power, and the negligence of the international community also played a part. Notably, what distinguishes the Rwandan case is that even as the mediation process proceeded from one stage to the next, most of the “successes” were in fact deceptions, so that the “triumph” of achieving GoR and RPF signatures on the Accords was a travesty of agreement. Since there was no real agreement, the Accords could not stop the fatal march to genocide. In Ireland, an agreement was reached and agreed to by the disputing parties— with a fair share of conflicts and setbacks along the way; with the exception of extremist positions and some lapses in the pledge to nonviolence, both sides were committed to the process and were successfully influenced by leverage and Mitchell’s confidence-building tactics. In Rwanda, the circumstances and attitudes were very different. In Rwanda, each stage was always already compromised because Habyarimana and the MRND came to the table in bad faith. Commitment to the process was not sincere and data was inaccurate, thus Mpungwe, the Tanzanian mediator, could never 201 really have designed a strategy adequate to the situation, despite his very real dedication to the cause. Thus, even though at the surface, an analysis of the Rwandan situation reveals a degree of “textbook” compliance at each stage, Habyarimana’s disingenuousness means that all of the stages resonate with a subtext of blatant artificiality. Words and actions simply did not match; so, for example, as I enumerate below, continued participation is really a delay tactic to reinforce military strength and thus to avert Tutsi military takeover. In the first round of mediation attempts, Zairean president Mobutu Sese Seko served as mediator between GoR and RPF, a designation that, given his long history of corruption, was problematic to say the least. Notably, Mobutu remained the official mediator of the peace process in name until the signing of the Accords, whereas Tanzania was dubbed its “facilitator.” But when it essentially replaced Mobutu with Mpungwe, Tanzania acted as mediator and thus is referred to here as such. Though Habyarimana certainly sabotaged the mediation in his own way, other factors also contributed to the failure; whether they could have effectively changed the outcome had they not played a role is debatable. According to Mthembu-Salter, too many aspects would have to have been different—the MRND’s stalwart ideological beliefs in Hutu supremacy would need to be eradicated, to be sure, but also the established political structure would need changing, more substantial leverage would have to be presented by the Tanzanian mediator, and the French government would need to assist in blocking the FAR and MRND militia. 202 To Habyarimana, that Mpangwe made such substantial cuts to the President’s power served to justify the double-dealing; the more mediation conceded to the RPF, the less reason Habyarimana had to participate in a genuine way. What follows is a closer look at the fissures and weaknesses of the mediation effort; using Moore’s 12 steps as a heuristic tool, I seek to elucidate where mediation met with success and failure in spite of Habyarimana’s covert but stanch unwillingness to cede power and to join a multiparty democracy without regard to ethnic origin. Stage one, “Initializing Contacts with the Disputing Parties,” began with Mpungwe’s designation as mediator. With the backing of GoR and RPF, Mpungwe brought a fresh purpose and commitment to the process as well as established credibility, especially after the rocky negotiations under the ill-chosen Mobutu. Mpungwe undertook the new challenge with a genuine desire to facilitate the process and to find a compromise solution satisfactory to both. In the first round of talks, beginning on July 12, 1992, Mpungwe met with RPF leadership and government delegation headed by Ngulinzira. The foundation for the talks was laid by amending the N’Sele cease-fire, which though unsuccessful, had formally established the NMOG. However, while it may seem like there was successful compliance with Stage One, insofar as the parties came to the table and agreed to cease their violent tactics, once the other 11 stages are analyzed, it becomes obvious that the lack of true commitment from MRND meant that investment in the success of the process was really a way to prepare for more violence. Moore’s second stage is “Selecting a Strategy to Guide the Mediation.” When RPF requested that Tanzania lead mediations and the GoR agreed, the second stage 203 seemed to have been successfully initiated. It was now up to Mpungwe to develop a method and tactics for getting concessions from both sides. The July 12 cease-fire established the new strategy of the mediation; unlike Mobutu’s N’Sele Agreement, Mpungwe’s plan was to guide the peace agreement strictly through political means (Mthembu-Salter, 2002). MRND clearly signaled its opposition via boycotts and angry demonstrations (Mthembu-Salter, 2002). Whatever victory gained in earning both parties’ consent to Tanzania as mediator quickly dissipated as the MRND jealously protected its power and soon interpreted the mediator’s efforts to establish power sharing as partisan favoritism toward the RPF. Collecting information and designing a detailed plan for mediation, Stages Three and Four, were carefully conducted by Mpungwe. Because Mobutu’s mediation had been ineffective, Mpungwe essentially had to begin the process from Stage One, which came about relatively easily considering that both parties consented to his intervention. After the establishment of the “cease-fire,” 55 discussions on the rule of law defining the new political order begun. Mpungwe was also able to specify deadlines as well as appeal to the international and regional community to apply diplomatic pressure. By August 18, Mpungwe had detailed the important negotiating points. As Mthembu-Salter describes it, “The agreement stated that no one should be above the rule of law, advocated the separation of powers, condemned ethnic discrimination and stated that the return of Rwanda's refugees was their ‘inalienable right’” (2002 website). Notably, MRND 55 In reality a cease-fire was never established, only the words were accepted. 204 quickly agreed to these terms despite its patent lack of commitment to the actions they would require. Building trust, Stage Five, was Mpungwe’s primary objective. However, realistically, there was no acceptance of Habyarimana. Belgium, France, the US, Burundi, and Uganda were in attendance to try to provide assurance to the disputants that there was international support and accountability (“Nature of the Mediation,” 2002). Tanzania’s neutrality and genuine interest in regional peace and security was crucial to establishing trust. As Mpungwe’s describes it, Tanzanian self-interest was strong but not designing: “[W]e had a vested interest because we are neighbours. Whatever happens in those countries effects us directly. I think this reaffirmed our commitment to wanting a sustainable solution, because it was in our interests too” (“Nature of the Mediation,” 2002). Mpungwe considered his role to be essentially that of a referee who encouraged the parties, yet showed no bias. Notably, a spokesperson for the RPF did not experience the Tanzanian mediation as bias-free, but this lack of objectivity did not deter his party, whose reputation as “trouble makers” meant it expected a certain lack of support. It was Tanzania’s dedication to finding a solution that mattered to the RPF, “They were not forcing us into positions. So that was neutral” (“Nature of the Mediation,” 2002). Indeed, the guiding principle for Mpungwe in the talks was the need to establish trust and confidence between the parties (“Nature of the Mediation,” 2002). To this end, the talks were disaggregated and moved from the least difficult to the most contentious issues. As momentum built, the talks next addressed how the new order would be 205 reached. When the process had gained significant authority, respect, and drive, especially in the eyes of the observers, the negotiations would move on to thornier issues. But as many critics indicate, substantial trust was never built and indeed even literally bringing the parties to the same table often proved impossible. In a 2001 interview, Mpungwe described the intricate balance he had to strike and the constant need to reassess attitudes and positions: When the mood was good and they [the GoR and RPF] were talking to each other then they would meet directly . . . When they were not talking to each other we would do the shuttle and proximity talks to narrow the gap and put them directly again. That was one of the strategies. As far as possible we avoided the plenary, and only used the plenary for announcing progress and good results. A lot of negotiations were in small groups. Where the plenary model fails, the caucus model often succeeds because it affords the mediator and disputant privacy; Mpungwe found this format to be much more effective Stage Six, “Beginning the Mediation Session,” began when an agenda was set and the JPMC was implemented. Although the JPMC had no adjudication authority, it proved to be an important safety valve for releasing tensions. A recurring issue during the talks was the tension between the Habyarimana regime and the opposition parties. Interestingly, the GoR party members were fighting more among themselves than they were engaging in the negotiations. Habyarimana would continuously veto breakthroughs agreed to by the Foreign Minister Ngulinzira, thus significantly reducing the former’s power and the influence in the ruling MRND. Though both the GoR and RPF had agreed to the terms, as enumerated in Stage Five, tensions were rising. The RPF had proposed most of the issues and thus gained a significant victory in getting the GoR to agree. 206 Though the RPF’s demand to have Habyarimana step down was not met with approval, the RPF was still gaining ground. Moore’s Stages Seven and Eight, defining the issues and uncovering hidden interests, proved to be among the most challenging because any problem that had previously been bypassed in the interest of progressing the talks, soon became more conspicuous. The sessions sought to negotiate for the basic rights of stability, specifically to establish the fundamental rights of the people, including the freedom of expression and the right to life as articulated by the UN Charter. Along these lines, the RPF explained that the Rwandan people, in its view, should be indivisible and not labeled in terms of their ethnicity. It viewed these principles as the theoretical underpinnings on which all subsequent agreement should be built and which the parties should reinforce. This issue, the RPF felt, must be explicitly addressed. Still, the GoR refused to address the validity of these concerns, causing negotiations on the creation of the new government and the transitional institutions to be extremely contentious. While the mediation process continued, it was undeniable that the failure of the prior stages had created a weak chance at successful implementation. Even as mediation was well underway, Habyarimana and the MRND never seemed entirely committed to its fulfillment and the power sharing it would inevitably entail. According to Mthembu-Salter, their continued participation was likely just part of a larger plan to temporize while planning and stocking up for an armed response. The more the Tanzanian mediators urged on the idea of power sharing, the more the MRND felt threatened by inevitable loss of power. 207 To the MRND, power sharing simply meant losing power, not gaining peace (“Nature of the Mediation,” 2002). In retrospect, negotiators from the RPF and internal opposition parties insist that the president was, in fact, never interested in compromise or genuinely committed to a settlement. Mpungwe underscores the lopsided commitment to the mediation, explaining that “If the MRND had its way it would not have attended the negotiations, and continued on the military option,” which is in essence what it did following the signing of the Accords. But because the RPF presented such a formidable military force, with several impressive defeats of FAR between 1991 and 1993, MRND continued negotiating. The leverage represented by the RPF military power was thus useful insofar as it kept MRND at the table whether or not it believed in resolution. As one member of the Tanzanian mediation team put it, RPF’s military prowess had achieved inroads into MRND territory. Seeing the threat of further loss of ground, MRND “didn’t have any other alternative” but to appear committed to mediation (“Nature of the Mediation,” 2002). The RPF’s military force (as compared to MRND’s weak forces) was in many ways the most compelling form of leverage—far beyond any other internal or external pressures. Thus, the MRND thwarted progress through what Richmond (1998) calls “devious objectives,” which he defines as “any involvement in a mediation or peace- making process on the part of the disputant that is not committed to a compromise.” In this case, “devious objectives” include the Rwandan president’s effort to prolong the process without the intention of abiding by the final terms. Habyarimana’s “hidden 208 interest” in retaining majority power was entrenched and non-negotiable, despite whatever impressions of good faith he faked (Richmond, 1998: 707). So whereas Stage 3 makes clear the important effort mediators must undertake in learning about the dispute, the parties, the circumstances, issues, and motivations, certain limits will always exist to the mediator’s ability to gain an accurate and relevant assessment of the disputants. Without good intelligence (often from third parties), getting the whole picture can be close to impossible. Thus in applying Moore’s steps, trouble at Stage 3 inevitably portends problems at Stage 8 and elsewhere, as defining issues and uncovering hidden motives is impossible without accurate data. According to critics of the peace process, the Tanzanian mediation suffered most for its inadequate understanding that “the MRND could ever be induced to move from a zero sum conception of the conflict, according to which the only alternatives were absolute victory or absolute defeat, towards an acceptance of the possibility of compromise” (“Nature of the Mediation,” 2002). Quite simply, mediation’s premise of give and take, compromise instead of victory, of achieving a “win-win” situation was simply not acceptable to Habyarimana. Even as negotiations took place, the president was making a back-door arms deal with the French company DYL Investments for 12 million dollars worth of grenades, bombs, ammunition, truncheons, pistols, and AK 47s (Melvern, 2000: 55). Hard-liners such as the CDR were also galvanizing forces against the peace-process so that whatever knowledge Mpungwe had of the motives of the various parties was rapidly diminishing as more and more deals and plans were undertaken covertly. 209 Despite the president’s obvious duplicity, some critics place a substantial portion of blame for the Accords’ failure on mediator bias and the RPF’s own intransigence. Lemarchand (1994) and Jones (2001) believe that few incentives were given to the Hutu elite to implement its side of the bargain thus inciting the anger of the Azaku group, who had no stake in implementing the Accords. As such, significant failure took place in the ninth stage, “Generating Options for Settlement,” where apportioning and assessing power sharing should have been more equitable (Lemarchand, 1994). With MRND feeling that it was doing the lion’s share of the sacrificing, resentment redoubled. Years after the Accords, a leading member of the RPF reflected on the hazards of his own group’s gains: "The most difficult thing, I must say, was the reduction of the powers of the president ... that was a very strong limitation of his powers ... It was when we finished that chapter, that is, when the government of Habyarimana decided to go ahead and plan the genocide” (Qtd. In Mthembu-Salter). Stage Nine’s requirement of “Lowering commitment to positions or sole alternatives” was flawed, then, insofar as the GoR was offered such a dramatic decrease in its power share. Precisely what motivated Mpungwe to effect such a dramatic reduction in MRND’s powers (and make such strong concessions to the RPF)—including cutting the president’s veto power, distributing cabinet positions so that the MRND would always be outnumbered, excluding CDR from government—is unknown, but doing so added fuel to the situation. Reacting to MRND’s diminishing powers, Hutu supremacist opposition began to flare, as Habyarimana was increasingly seen by Hutu hard-liners as betraying their cause. 210 While critics report with certainty that Tanzania knew that “major killings were being planned,” Mpungwe still sought to include all parties in the negotiations thus indicating that perhaps he did not appreciate “the gruesome specificity of Rwanda’s case” (“Nature of the Mediation,” 2002). To get through these challenges, Mpungwe separated the parties and, along with members of the observer groups, strenuously lobbied the RPF to allow the CDR to participate. The RPF, however, remained extremely adamant because it believed that the extremists were in fact the very source of Rwanda’s problems and that it was bent on destroying any power-sharing arrangements. As the RPF put it, “The CDR represented the cancer of our society and you cannot negotiate with a cancer. The CDR was simply the spearhead of the hatred that was at the origins of the conflict. Habyarimana had created this party to front that ideology [of] hatred” (“Nature of the Mediation,” 2002). In generating options to settlement, Mpungwe finally had to concede to the RPF and exclude the CDR. Thus the “success” of this stage led to a weakening of Stage 10. Against the wishes of Habyarimana and French diplomats, Mpungwe finally had to bar the extremist party and deny it power in the transitional institution. This move was a victory for the RPF, which did not see any benefit to including the group in the new government—despite Tanzania’s and the United States’ opinion to the contrary (Melvern p. 54) Eliminating extremists was a short-term solution and a sure recipe for inciting spoilers to the process: “Western governments saw the exclusion of the CDR as a departure from constructive negotiations, insisting that a more substantive role should be given to those who stood to lose power” (Melvern, 2000: 54). 211 Assessing how interests could be met (Stage Ten) led to the government’s eventual accommodation. This move occurred because the government opposition parties were sympathetic to the RPF viewpoint and the RPF also made it clear that it would readily resume violence. However, hard-liners reacted violently to their exclusion, threatening to bring about a genocide –and, in a foreshadow of atrocities to come, in late January, 1997, over 300 Tutsi were killed (Adelman & Suhrke, 1996: 43). Stage Ten was complicated by Tanzania’s lack of truly compelling leverage—costs and benefits could not and did not play a significant part. Though he explains that his role involved “putting pressure” on the parties, Mpungwe concedes that Tanzania exercised no political leverage whatsoever. As he explains it, One thing that I made clear, and this is my experience in this area, they sometimes feel that when they come to negotiate they are doing you a favor. You owe them a solution. From day one I put it clearly that they have the responsibility to resolve their own problem and it was nothing to do with Tanzania…We are not a big power, we couldn’t threaten much. (“Nature of the Mediation,” 2002) Just as it could not employ leverage in the form of incentives, Tanzania itself didn’t have the clout or resources to threaten embargoes, limitations to travel, or withdrawal of aid. Nonetheless, manipulation played an important role and leverage was actualized by making recourse to previous relationships between Tanzania and Rwanda and by drawing upon the involvement of regional and international parties: Mwinyi used his personal relationship with Habyarimana, for example, to prevail on the Rwandan president to drop his objections to the RPF constituting half the new Rwandan armed forces. Tanzania also requested influential countries, such as France in the Rwandan government's case and Uganda in the RPF's, to pressure the parties to be more accommodating. According to Mazimpaka, the mediation 212 team also made extensive use of leaks to the media to embarrass the RPF into compromise. The eleventh and twelfth stages, “Final Bargaining” and “Achieving Formal Settlement,” proved to be the most difficult. To the extent that the signatures of both parties were inked on the Accords, the process was successful. However, the fourth requisite of success explains that the perceptions expressed by the parties and mediator(s) have to be in conformity. Since Habyarimana signed the Accords but never meant to follow through, his signature became meaningless. There was no realistic conformity of perceptions. Almost all accounts consent on the part of the MRND was a lie from the beginning—its signature in no way reflected an agreement to the terms or to their implementation. Outside, extremists such as the CDR continued to focus on the development of a rural militia and the manipulation of ethnicity to create a climate of fear while inside, political infighting contributed to undermining of the Accords. An ineffective peacekeeping force, in part a result of disjuncture between the United Nations and OAU, failed to provide the needed security and stability. Moreover, pressure by the international community ultimately proved ineffective. In light of all of these complexities, even with the requisite signatures on paper, the successful implementation of the Arusha Accords became an impossibility. Reflecting on his extensive efforts, Mpungwe explains that he had left no stone unturned in attempting to create a viable and transparent agreement: “Nothing was swept under the table… in the end, every aspect and detail of the Arusha Agreement was 213 thoroughly negotiated and ultimately agreed upon by both parties” (“Nature of the Mediation,” 2002). What the Rwanda peace process demonstrably indicates then is that an agreement is only as good as the intentions of its signatories. That Mpungwe and the Tanzanian mediators were unable to see through Habyarimana’s duplicity also indicates the utter failure of their intelligence and, perhaps, a bit too much wishful thinking on their part. Notably, the international community looked on, seeing success in the progressing mediation, so that even as Mpungwe shepherded the parties through the process and fulfilled the stages in good faith, the MRND’s ulterior motives and other missteps undermined the entire process. A comparative analysis of the case studies of Rwanda and Northern Ireland reveal stark contrasts. Everything from disputant commitment to the process, international support, and mediator leverage, to clarity of communication and confidence building were radically different. That Clinton himself made the initial contacts through the Gerry Adams visas put Mitchell’s work to great advantage. Where Northern Ireland had the passionate commitment of the Clinton Administration, with the President himself visiting Belfast, Rwanda barely garnered any attention from a United States still gun-shy over recent foreign policy debacles, especially Somalia. Where Clinton sought to cultivate economic confidence through trade missions and the International Fund of Ireland, Herman Cohen’s dismay that the Rwandan economy was on the brink fell on deaf ears despite its inevitable contribution to the degeneration of Hutu-Tutsi relations as resources became scarcer. While the events of Northern Ireland and Rwanda were internationalized, the former received much more substantial support for resolution and 214 became about more than Northern Ireland and Britain. Though Habyarimana sought to internationalize the issue by presenting the Hutu-Tutsi conflict as one between Rwanda and Uganda, this interpretation did not convince the international community. A final note about balance of power: According to Moore, mediation is most effective when power relations are symmetrical. But often disputes occur precisely because of power imbalances and the struggle for the disadvantaged party to even the playing field. In the Rwanda example, power was asymmetrical for both sides. While the MRND had the advantage of government control (as well as government-owned media) and French support, the RPF had a formidable military advantage that worked to intimidate the president’s party and keep it at the negotiating table despite its covert intention to retain power at any cost. Where power imbalances are dramatic, Moore (1996) explains, the mediator must work toward equilibrium. To this end, "the mediator provides the necessary power underpinnings to the weaker negotiator: information, advise, friendship, she may reduce those of the stronger" (p. 337). It appears that Mpungwe lacked sufficient data and adequate leverage so that the Arusha Agreement awarded the RPF with conspicuous powers, agitating the MRND and extremists alike. Using this analysis will make the Prisoner’s Dilemma game much more comprehensible. Moore’s 12 stages of mediation enables a systematic breakdown of the mediation process, providing a clearer understanding of what went wrong in the case of Rwanda and what was worked out well in the case of Northern Ireland. The following chapter offers analysis of the two cases, using the principles of the Prisoner’s Dilemma 215 game so that the roles, motives, preferences, weaknesses, and strengths of each party can be identified in the mediation process. 216 CHAPTER NINE: APPLYING THE PRISONER’S DILEMMA TO THE TWO CASES Mediators enter a conflict situation in an attempt to affect the perceptions, behaviors, and outcome of the parties in conflict. Usually, mediators become involved in conflicts to resolve them and to protect any interests they might have. But can a mediator predict and produce certain responses in the disputants? By applying the theoretical model of prisoners’ dilemma (PD), a paradigmatic scenario in game theory, mediators may be able to improve cooperation and make productive assumptions about player motivation. The premise of PD is that people, and in this case organizations, function directly out of self-interest. In game theory, each player (“prisoner”) is, above all, concerned with maximizing his or her own payoff, without concern for the other player’s payoffs. 56 Ultimately, only two decisions confront each player. One player can maintain a low-cost armament strategy (LC1) or a high-cost strategy (HC1) and the second player can utilize a low-cost armament strategy (LC2) or a high-cost strategy (HC2). The participation of an outsider may become desirable as both parties seek to maintain a low- cost equilibrium and a Pareto optimal point (at which point no further Pareto improvements are possible). According to game theory, players are rational decision makers ever considering what is at stake, what the opposition wants from them, and what 56 That is only true in a single iteration case. If the parties must continue to interact, taking account of the other’s position is essential. 217 the incentives and payoffs are for ending deadlock (and, in conflict, violence). There is no assumption that the opposition seeks anything but its own desired outcome. Still, both parties must have a high level of trust in the mediator in order for the latter to fulfill his or her role. As mentioned in the literature review, of the sizable body of work on mediation, a significant portion is dedicated to the argument that mediators must be impartial (Touval 1982; Berkovitch and Houston, 1996; Carneval and Arad 1996; Zartman and Touval, 1996). However, as Touval and Zartman explain, “mediators are seldom indifferent to the terms being negotiated. Even when they seek peace in the abstract, they try to avoid terms not in accord with their own interests” (1986:118). The mediator often derives these ends through leverage, as “leverage is the ticket to mediation” (129). So while trust in the mediator is crucial, mediator partiality is not necessarily deleterious, indeed it may even be critical, as Kydd (2003) explains, “While impartiality is appropriate for the weak mediator, bias is seen as acceptable, perhaps, inevitable for the powerful mediator” (18). The theoretical framework of prisoners’ dilemma then, involves among other factors, player payoff, cost of fighting versus cost of status quo, and the mediator’s degree of bias (and perceived degree of bias) as well as their level of power/leverage. According to Levy, depending on the dynamic and circumstances in the prisoner’s dilemma scenario, the mediator may begin by “making the parties more fully aware of the consequences of their actions,” in which case the “goal is to make them see that there is a certain outcome beneficial to both, that is, that movement to a Nash 218 solution is a practical possibility” (1985: 590). Where bringing further awareness to the parties proves ineffective, incentives and leverage enter the picture; in other words, “influencing the costs of standing firm or benefits of cooperating” (1985:590). As I discuss below, leverage plays a significant role in the mediation efforts in Northern Ireland and Rwanda to dramatically different ends. What the PD model reveals about the mediation efforts in these respective cases is also startlingly divergent. Prisoners’ Dilemma and the IRA: First Attempt In early attempts to resolve the conflict, Britain intervened as peacekeepers and mediators of a dispute in which it clearly had a stake. Through the Northern Ireland Office, developed in the early 1970s, the British brokered the Sunnydale Agreement in 1973, which sought to give Nationalists from the Republic of Ireland representation in the politics of Northern Ireland. In 1985, the Anglo-Irish Agreement, the Republic of Ireland gained “a formal consultative role for the first time” (qtd. in Jacobsen 2003: 189). The multiparty talks in Belfast, London, and Dublin in 1992 were led by the Northern Ireland Secretary, Peter Brooke. As I have sought to elucidate throughout this dissertation, mediation involves compromise and willingness of the parties to redefine what it means to win. Throughout the various mediation efforts in the case of Northern Ireland, disputants were required to revise and reconsider demands, and to take political risks in order to break out of the cycle of violence. The challenge along the way was to keep each party convinced that the 219 benefits to mediation exceeded those of continued violence. The many efforts at mediation before George Mitchell had some successes, though each time recourse to violence indicated that the incentives to negotiating compromise were not substantial enough. Though Brooke’s three-tiered talks did not succeed, they established some of the terms for future talks, such as requiring a permanent end to the support of and use of paramilitary violence and the stipulation that parties seeking to participate in talks must, in fact, be genuinely committed to peace (Jacobsen, 2003: 190). Though Brooke made these impressive advances, his mediation efforts often ended in deadlock because of Sinn Fein’s demand to have talks without preconditions (Bowcott, 1991). According to a contemporary article in The Guardian, Sinn Fein made its share of conciliatory gestures under Brooke’s mediation, but as long as Margaret Thatcher was in power the organization saw stalemate as inevitable. Moreover, Sinn Fein itself could not come to a consensus; where Sinn Fein President, Gerry Adams commented, “We are prepared to take political risks in the search for peace. We are prepared to give and take” (Bowcott, 1991:1), Martin McGuiness, the organization’s leading executive member, took a harder line, insisting, “[R]epublicans want peace. We want an honourable peace, no papering over the cracks or brushing under the carpet the humiliations, degradation and injustices inflicted on us as a foreign power (Bowcott, 1991:1). Such strong and deeply held beliefs meant that one of Brooke’s primary challenges was to manage political and military strategies at once, to build trust even as armed struggle continued. At first, the British were welcomed by both sides, but 220 ultimately, they failed to meet the demands for a peaceful resolution. As long as mediation was led by an entity with such direct interest in the outcome, in this case the British government, success would be elusive. According to Richard H. Dekmejian (2007), the failure in its peacekeeping mission had two consequences: the reversion of the tripartite game to a two-party interactive struggle between the Catholics and the Unionist/British coalition; and the change of Catholic preferences from self-defense and civil rights, to the IRA’s ultimate preference to achieve a unified Irish state (Dekmejian, 2007: 44). Second Attempt The second attempt at negotiation was made when the United States entered the picture. The peace process was reignited when President Clinton granted Gerry Adams, a visa to visit the US in February 1994. In return Adams promised a “complete cessation of military activities” –marking the first such promise since 1969 (Kelly, 1998: A19). According to Pillar (1983) a cease-fire is one of the biggest challenges in the disengagement process, which is why disputants will usually continue to fight until all negotiations are complete. But because a cease-fire was agreed upon from the outset, progress seemed possible with the new US dedication to mediation; as previously discussed, however, the cease-fire was violated several times and the possibility of violence loomed over the mediation throughout. Still, with British and IRA consent, the US was welcomed into the process, creating a new prisoners’ dilemma scenario. The US 221 strategy was to press both the IRA and the British to negotiate. Again, in the words of Dekmejian: Prime Minister John Major’s relatively hard line position on the IRA disarmament was shaped by a two-level strategy: first, to force the IRA to give up its threat power, making it impossible for the IRA to revert to terrorism if the talks proved unsatisfactory; and second, to assuage Conservative Party hardliners, particularly its traditional Unionist allies in Parliament, as Britain prepared for the 1997 elections. (Dekmejian, 2007: 45) It was not, however, until the election of Tony Blair as prime minister that a path to resolution opened up. Among Clinton, Blair, and Mitchell, the process and the outcome of the talks were closely watched with full appreciation by all of the high stakes. Incentives played a crucial part throughout the process, as Clinton linked progress in the talks to increased US aid and to encouraging American companies to invest in Belfast (Kelly 1998: A19). Thus the payoff for productive participation in the talks would not only be peace, but also much-needed economic development as investment was promised to flow into the province. With the United States applying such leverage, the prospects of successful mediation were improved. Between the options for negative sanctions (sticks) and positive sanctions (carrots), the US chose the latter, bringing its crucial power and influence to bear on the parties (Kleiboer, 1996: 371). Nonetheless, when Mitchell was appointed to chair the Northern Ireland peace talks in 1996, he encountered significant barriers to consensus. Side talks already underway on the subject of IRA disarmament were deadlocked, positions were entrenched, the conflict had, in many ways, been going on for 300 years. The disputant positions he faced are perhaps best summarized by one journalist’s observation that 222 Mitchell, “discovered that Catholics and Protestants are routinely willing to cast aside progress to revive bloody history and family trials” (Marquand, 1998: 1). During the course of his mediation, Mitchell built substantial credibility and trust among the pro-Irish and pro-British parties. But this support did not mean that the parties willingly complied throughout the mediation process. Indeed, as Mitchell himself described the attitudes, there was a hardened unwillingness to see the situation through the eyes of the opposition: “There was no ideal of compromise, no sense that there was such a thing as magnanimity in victory. They never heard of the American phrase win- win” (“Northern Ireland’s Impasse,” 1999). Getting Sinn Fein to give up its weapons, to see that the incentives to disarmament were greater than continued sectarian violence was one of Mitchell’s biggest challenges. Consistent with one prisoners’ dilemma scenario, the IRA was loathe to give up its weapons before talks, seeing such a move as akin to an admission of defeat—importantly, and a renunciation of leverage. The Unionists, on the other hand, believed that unless the opposition relinquished its weapons before a settlement, they would become part of the planned bi-sectarian government with their arsenal—and, the IRA was projected to be the majority. Over 22 months, Mitchell labored to move recalcitrant parties toward a cooperation threshold, a place where it would be possible to see the advantages of compromise—but the sides did not trust each other and thus neither side would budge on the decommissioning issue. While the idea of decommissioning before talks was rejected by the IRA, so too was the viable alternative of “parallel decommissioning.” The 223 deadlock that Mitchell faced when first taking the role of mediator was holding fast. Players, as PD would characterize them, still sought their maximum payoff without making concessions. At this point, the alternatives were violence or stalemate—incentives for cooperation were not sufficiently compelling. Notably, however, Mitchell’s work differed from the predominant PD paradigm in which players choose their strategies simultaneously with no knowledge of the other’s choice. Mitchell went constantly back and forth and a certain degree of transparency exemplified his mediation. Still, Mitchell confronted many of the archetypal qualities of PD as Levy describes it: “Both parties see more to gain from standing firm than cooperating; consequently there is no incentive to negotiate and the conflict persists” (1985: 582). To the IRA the reward for cooperation (decommissioning) was not clearly superior to what Levy calls the “sucker’s payoff for cooperating while the opponent does not” (1985: 582). For his part, John Major would not consent to parallel decommissioning, as he saw the reward (beginning productive peace talks in earnest) as less attractive than the “sucker’s payoff” of continued reprisals by the IRA. In this sense, both players were “high threshold” (Levy, 1985: 587) in being loathe to accept any risk in cooperating. In one important respect, however, the conundrum Mitchell faced differed significantly from the PD archetype, as Levy describes it: “Simultaneous play is at the heart of the dilemma, for it is the absence of communication that prevents movement toward the mutually beneficial” outcome (1986: 588). As I have discussed, 224 communication (talking and a lot of listening) was at the crux of Mitchell’s mediation style; shuttling between parties and keeping them apprised of the opponents’ “moves” was part of what made his work so time-consuming and yet ultimately successful. As such, Mitchell’s mediation more closely resembled the PD model of “incremental play,” which, as it sounds, “allows for step-by-step building of trust by the two parties, allows for a cooperative approach, and leads players to the strategies that yield the Nash solution” (1986: 588). In Northern Ireland conflict, deadlock, and violence had become the norm; because the costs of maintaining the status quo were not necessarily increasing (although certainly more violence was leading to more deaths), the promise of collaboration did not seem alluring enough to the IRA. Meanwhile, Britain was not prepared to begin all-party talks about power-sharing or granting self- government to Northern Ireland unless the IRA decommissioned its weapons. The Mitchell Principles went a long way toward promoting cooperation and breaking the deadlock, but Britain had to accept that decommissioning would not occur before negotiations. As Mitchell explained, “In the real world of Northern Ireland, prior decommissioning simply was not a practical solution” (Qtd. in Jacobsen, 2003: 194). Perhaps in a the theoretical model of PD, the failure of prior decommissioning would have led to deadlock, as Britain had made its insistence on this stipulation clear. Indeed Trimble’s obstinance that “actual decommissioning occur” before talks ran the risk of provoking deadlock in that it simply asked too much of the opposition. As Mitchell himself very valuably pointed out, such hard and fast positions do not necessarily apply to “real world” scenarios. It is very much to Mitchell’s credit that he had the insight to 225 determine that forcing the issue would have done more harm than good. That tangible progress was not privileged over genuine understanding and cooperation—as well as patience and moderation—was fundamental to Mitchell’s success as a mediator. Mitchell’s perseverance and skill, and the psychological attrition of sustained conflict, finally produced Zartman’s so-called “ripe moment” of negotiation. The military option was no longer attractive or viable; the parties were ready to talk, as Mitchell himself described it, by June 1996: “[T]he British and Irish governments . . . were totally committed to getting it resolved” (qtd. in Jacobsen 2003: 197) and thus a series of quid pro quos took place (Freedland, 2001: 13). Though the “game” of negotiations was incremental rather than simultaneous, parties were kept separate because bringing them together resulted in too much hostility (Jacobsen, 2003). Instead of using the plenary format, with the two governments and seven parties involved (Alliance, Labour, NIWC, PUP, SDLP, UDP, and UUP), Mitchell helped control the players (and thus their cooperation level) by mediating through a bilateral form with two or three parties and a chairman. Though it was not necessarily by design that Mitchell’s negotiations went slowly, the unhurried pace had its distinct advantages— indeed, one journalist described them as “sensibly gradual, advanced by a series of confidence building measures” (Freedland, 2001: 13). Groups must not only find agreement with each other, as Freedland points out, but also present a cohesive position to their the respective leaders, or “players,” as PD would have it. He explains this further challenge to the Unionists and Republicans: 226 David Trimble has …trouble with the hardline unionists who deny him a mandate to say yes. On this score, Sinn Fein is the luckiest kind of peacemaker. Republicans are a tightly disciplined (some would say undemocratic) and relatively small movement: what their leaders decide at the table comes with a copper-bottom guarantee, partly because they tend to promise only what they can deliver. (2001: 13) Bombings and murders by the IRA, Ulster Democratic Party (UDP), and splinter groups continued to prevent full participation, as backsliding into tried (and in many ways, effective) strategies of violence never entirely lost their appeal for the disputants, despite Mitchell’s intense dedication to the talks and resolute presentation of cooperation’s advantages. Finally Mitchell brokered a power-sharing agreement that produced the Northern Ireland Assembly with decommissioning to take place afterwards, a contentious conclusion to be sure. As the day Mitchell had hoped to secure an agreement approached, The Christian Science Monitor reported he was “dropping his poker-faced demeanor . . . and resorting to pressure,” or as one White House source put it: “Mitchell is going to trade on all that credibility by the end of Good Friday. He’s in the end game, and he is telling the Irish, ‘Go back to conflict, or have peace, but I won’t do this forever.’ Given his role, these are high stakes” (Marquand, 1998: 1). In his final week of negotiations before the Good Friday deadline, Mitchell presented a 65-page document articulating the Agreement, but Trimble, head of the pro- British Ulster Unionist Party objected to the plan believing it made the province too inextricable with Ireland, which would then smooth the progress toward a united Ireland 227 (Marquand, 1998:1). Mitchell made modifications, persisted, and “For the first time, external intervention had succeeded in imposing a consociational regime on Northern Ireland based on shared power between the Unionist Protestants and the Republican Catholics” (Dekmejian, 2007: 50). Both the IRA and the Protestants were required to give up their first preferences in order to reach a settlement. Ireland was forced to change its constitution, specifically Articles 2 and 3, which laid a territorial claim on Northern Ireland and called for the unification of both parts of the island. This act was seen as a good will gesture to establish trust between the parties. The definition of Ireland’s nationhood was also altered to include all people in all the diversity of their identities and traditions, regardless of their place of territorial habitat. In return for the revisions of the constitution, the Irish Republic gained an institutionalized voice in governing the province through cross border bodies. In 2000, the IRA agreed to permanently renounce political violence and turn in its weapons in return for Democratic Unionists to recognize Sinn Fein’s political legitimacy. For a successful settlement to be reached, the interests of all parties needed to be minimally met. The two party game model of prisoners’ dilemma takes into consideration the inevitability of deadlock on the way to cooperation. It is up to the mediator to use strategies that move the “players” out of their entrenched positions toward alternative outcomes that may not be commensurate with their initial perceptions of victory, but that under the circumstances, and in the interest of compromise, hold a recognized appeal. 228 In the case of Northern Ireland, hardliners on each side represented a persistent threat to the moderates by vetoing any concessions and disrupting the process through terrorist attacks. On the Unionist side, the demands around decommissioning became, as one journalist described it, a “fetish”: “they may well get their way on IRA arms, but at what price? In return for bowing to the unionist demand for weapons, republicans seem to have got their way on almost everything else, from demilitarisation to policing” (Freedland, 2001: 13). Once the two sides reached Nash equilibrium, it became unacceptable to upset the equilibrium by changing their preferences to violence due to the high cost that such behavior would entail. Table 1.3: Breakdown of Conflict Resolution in Northern Ireland Low Cost (LC) or High Cost (HC) Armament Cooperate (C) or Defect (D) Preferences Met? (Yes or No) The IRA LC C Yes Great Britain LC C Yes The Unionists LC C Yes Republic of Ireland LC C Yes The Prisoner’s Dilemma and Rwanda The implementation of the Arusha Accords, backed by growing pressure from the UN and the RPF, was seen by Habyarimana and his associates as a political and existential threat to their leadership position and to Hutu dominance. This perception led the president to pursue a dual strategy–stalling the implementation of the accords, while strengthening his domestic following through military training, illegal import and 229 distribution of weapons, and anti-Tutsi propaganda. By postponing negotiations with the RPF and mobilizing the home front for war and genocide, Dekmejian argues, Habyarimana may have hoped to achieve one of the following objectives: 1. Resist and reverse RPF’s military gains by military means as a first preference; and if that failed, 2. Use the threat of the Tutsi genocide as a bargaining tool to force RPF to retreat and negotiate a new settlement more favorable to the Hutu elite than Arusha, that would accord the president’s circle a major role in the coalition regime or at least protection from persecution; or as a last resort 3. Proceed with the planned genocide as a last act of mass revenge against the Tutsi as his regime collapses or he suffers an uncertain fate. (Dekmejian, 2007: 282) Like the Northern Ireland negotiations, the Rwanda peace talks ended when the two primary parties signed the accord; and surprisingly, many external accounts characterized the mediation as “a virtual textbook case of modern conflict management. The conflict was ripe for settlement because the civil war had reached a mutually hurting stalemate” (“The Other Lesson of Rwanda,” 1996: 221). Through the course of the mediation, Mpungwe addressed all of the hot-button topics: power-sharing in the government and military, economic consequences of continued violence, the growing refugee crisis, and the brutal effects of continued apartheid. The international community promised peacekeeping forces that were to guarantee security and to ease the transition that would follow the settlement. To the MRND the “game” of negotiations was zero-sum: to concede or lower the cooperation threshold was to lose power, an unacceptable option. Thus, as I argued in Chapter Eight, there was never a real agreement since one side participated in the letter but not the reality of the mediation and did not intend to stand by the Accords despite 230 assertions to the contrary. Nonetheless, several genuine efforts by Mpungwe were undertaken to bring the parties to a cooperative resolution. Because the MRND covertly regarded power-sharing as a nonnegotiable issue, it participated in the talks in bad faith. As Habyarimana saw it, the benefits of cooperating were far less than the costs of holding out; stalling was not costly and afforded him the time and opportunity to build his power base, acquire weapons, and promote fears of the RPF that galvanized Hutu across party lines (Beardsley, 2006). That Tanzania did not fully appreciate the degree of resistance—and consequently that productive mediation was even impossible—represents a “blind spot” in its mediation effort. As Kuperman explains, They failed 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 to the breaking point. To protect the privileges they saw the international community trying to wrest from them, extremists proved willing to massacre fellow countrymen with whom they had been living in relative calm for more than a decade. (1996: 222, italics mine) Needless to say, characterizing genocide as part of any rational decision making is impossible. Even the highly rational, self-interest driven logic of PD takes into consideration a certain degree of human morality 57 so that in many ways the Rwanda case calls into question the very capacities of logical parsing through conflict management theory. 57 The work of David Gauthier explains the role of morality in prisoner’s dilemma, insofar as players may make certain assumptions about fundamental human morality in order to gain trust and cooperate successfully. <http://plato.stanford.edu/entries/contractarianism/#3> 231 Following on the heels of the Mobutu’s failed efforts to mediate, Tanzania faced mounting hostilities as the MRND began to see what was at stake in participating in talks. For the RPF’s part, the prospects of mediation were strictly about gain, as it had very little to lose. By participating in the talks, the Tutsi-led RPF stood to earn representation in a government in which it had long been marginalized and even excluded. Notably, the RPF sought not to take control of the government, but rather to participate in a plural format, a genuine power-sharing regime. Meanwhile, the Rwandan Patriotic Army, the RPF’s military wing, presented imposing force and served as a highly compelling source of leverage. Coming into Rwanda from neighboring Uganda (many members had lived there as refugees since the 1960s), the RPF quickly made advances in territory, small triumphs that intimidated Habyarimana’s regime. Since the Hutu manifesto of 1959, anti-Tutsi sentiment had grown significantly so that with the 1973 coup d’etat of Army Chief of Staff, General Juvenal Habyarimana, the ethnic-driven ideology was solidified through ethnic and regional quotas. Under the Rwanda’s new president, the constitution was single-party: Habyarimana’s MRND. Once in power, Habyarimana bestowed his fellow Northern Hutu with the privilege of political patronage and dominance in the civil service, “the key source of wealth in Rwanda as in many developing countries” (Kuperman, 1996: 224-225). With its institutionalized superiority, the Hutu had no reason to promote anti-Tutsi discrimination, though seeds of discord from pre-Revolution colonialist anti-Hutu prejudice were ready to be sown. With the RPF invasion in October 1990, Habyarimana’s unchecked power was being challenged politically, ideologically, and militarily. Because it still had the 232 formidable military force (supported as it was by the French), Habyarimana violated the cease-fire achieved by the Zairian mediation in March 1991, citing RPF violations. Still, there was forward movement as Habyarimana made agreed to modify the constitution: by June 1991, the constitution was modified to accommodate a multi-party democracy; as a result several opposition parties formed, such as the MDR, PSD, PDC, and PL. To counter these changes and to shore up his own power, Habyarimana established the extreme right-wing anti-Tutsi CDR and the Interahamwe militia, a paramilitary organization that would become a key player in the genocide. Though there was certainly dissent among these various groups (and much of it played out during the mediation), the various Hutu moderate and extremist parties recognized that the considerable threat of the RPF meant dropping differences and banding together to actualize an anti-Tutsi platform using violent means (Kuperman, 1996). With his fierce attachment to his position of power, Habyarimana consented to participate in talks in order to delay Tutsi military action and to give himself time to prepare for more war.. Lacking any genuine commitment to mediation’s objectives, the president saw the RPF’s military progress as a reason to participate—therein lay the leverage. In the language of PD, he saw important benefits to cooperating—they just weren’t the same benefits the mediator and opposition saw. Where peace and power sharing were the ostensible reasons for mediation, Habyarimana had his own set of motivations: stalling and fortifying his power. Throughout the negotiations of the Arusha Accords, the RPF led offensives and the MRND and CDR militia undertook violent demonstrations and attacks against civilians. 233 International aid donors and trading partners undertook cursory efforts to apply leverage in an effort to compel government reforms. (Kuperman, 1996: 224). Conflicts within the Rwandan government presented further roadblocks and point to an even more fragmented model of the prisoners’ dilemma as innumerable side disputes prevented it from formulating a unified position. The July 1992 cease-fire agreement between the GoR and the RPF was followed shortly by militia massacres of Tutsi and Hutu opposition members, another indication that the cease-fire existed on paper, not in reality. At this point, Habyarimana, recognizing the so-called ripeness of the moment as it served his ends, agreed to participate in mediation in Arusha, Tanzania. The mediation seemed straightforward with both players adopting cooperative strategies on issues that previously seemed nonnegotiable. By then, the uncertain capacity of RPF military might compelled Habyarimana to lower his threshold and consent to Tanzania’s proposals on legal authority and power sharing protocol. By October, 1992, however, he issued his infamous “scraps of paper” denouncement of the agreement, most likely to appease growing extremist criticism of his compromises. The negotiations continued like this for several months, alternating among cooperation, violence, and deadlock, much like the model of the iterative prisoners’ dilemma game in which several simultaneous games occur because cooperative behavior cannot be sustained. In February 1993 with the Arusha protocols on a transitional government already signed, RPF made military advances while MRND militias led violent attacks throughout the countryside (Mugabowineza, 2005). Still unable to present adequate troop strength, 234 Habyarimana returned to the mediation table. In March, 1993 again cease-fires were signed while the UN Security Council authorized intervention (UNAMIR) and the Intervention Commission made the important move of characterizing the violence as genocide. Power sharing in the army presented one of the biggest stalling points. Habyarimana, already anxious about the RPF’s military capabilities, saw no payoffs in allowing the RPF 50% representation as proposed by the Accords (“Rwanda Peace Agreement Postponed Indefinitely, 1993). In response to the president’s rejection of this distribution, however, the RPF upped its demand to 60%. Much to the MRND’s disadvantage, the RPF knew its own strength was superior. Indeed, one reporter noted of his travels with the RPF around Rwanda that April: “All along the way, we saw young guerrillas weighed down with supplies walking the hills with apparent ease. ‘These kids are tough,’ says Frank Bamugambagye, the RPF's political commissar. ‘That's why we beat the government soldiers, who are used to a comfortable life.’” The journalist’s further observations warrant extended quoting: By any measure, the scale of the RPF victories in the February fighting was phenomenal. In just a week of skirmishes, the rebels doubled the territory under their control. The Rwandan government, skeptical the RPF could advance so far without outside help, accused Uganda of sending ground troops, a charge the Ugandans have denied. The RPF counters with its own charges that Rwanda is receiving help from French forces. If the French did help in the February fighting, the Rwandan army gained little apparent benefit. John Bagire, the regional RPF commander, says his men easily overran ten government outposts…The rebels say they are confident a political settlement will be reached, although they seem ready to pursue the fighting if the government refuses to compromise. With a fighting force now approaching 40,000, the movement might well be able to continue the conflict if peace efforts fail 235 To Habyarimana, the RPF’s sheer military might presented a redoubtable form of leverage. With French troops set to pull out in December, 1993, the president’s forces would soon be even less potent. Caught between diplomatic and economic “carrots and sticks” on the one side and impending loss of territory to the fierce RPA on the other, Habyarimana saw the impending precipice (the situation would soon worsen without a settlement) and returned to the talks both to stall and to posture as compliant (Kuperman, 1996: 228). Going from being president with absolute power to but one entity in a plural government presented an intolerable alternative to Habyarimana. That genocide occurred in spite of the efforts and “success” of the Arusha Accords is devastating. While the consequences of deadlock were continued violence and dissent, the aftermath of the peace accords was genocide. The Arusha process demonstrates that signing papers means nothing unless those who sign are actually in agreement. Did mediation itself push the conflict to a breaking point? In his study of the prisoner’s dilemma conflicts, Levy (1985) explains that Certain games . . . offer no way out of stalemate, and the players remain deadlocked until the game evolves into something else. In serious conflicts the empirical outcome is war. The most commonly discussed game of this sort is, literally, deadlock. Both parties see more to gain from standing firm than cooperating; consequently there is no incentive to negotiate and the conflict persists. (1985: 582) While applying this assertion to the Rwanda negotiations is imprecise, it does raise a few important questions about the problems Tanzania faced in seeking to negotiate its way through such recalcitrance. Most importantly, what Levy’s observation implies is that some prisoners’ dilemmas scenarios prove irresolvable as “games” of rational exchange, 236 of give and take, of cooperation and anticipation of self-interest. In some ways, the Rwanda situation exceeds the limits of the sort of predictions PD allows because the problems there progressed far beyond what could be predicted of “rational decision makers” with “desired outcomes.” What is difficult to determine is whether (and the extent to which) the continuing negotiations under Mpungwe actually contributed to Habyarimana’s increasing dissatisfaction with the power-sharing deal as it unfolded. As such the model of the prisoners’ dilemma as it applies to Mpungwe’s mediation is highly complex because, as the discussion of Moore’s 12 steps indicated in the previous chapter, the duplicitous motivations, amenability to cooperation, and incentives to negotiation versus continued conflict were ambiguous at best. Tanzania’s self-interest in bringing resolution to the two parties was primarily centered around bringing peace to the region and working out the increasingly dire refugee problem that was affecting Zaire, Uganda, Burundi, and Tanzania. As mentioned, Tanzania itself had little in the way of leverage in order to manipulate the players to cooperate and to see their way beyond the violent tactics but it was still able to bring the parties to the table and get them eventually to sign the Accord; but at what cost? As a theoretical model, the prisoners’ dilemma game indicates the conundrum of mediation in which each player rationally pursues goals in the form of maximum payoff. The job of the intermediary is to bring the players toward a cooperation and compromise in order to arrive at a “win-win” situation; again, at what cost? 237 Though the Arusha Accords appeared to be “win-win,” they turned out to be a sham that led to a “lose-lose” situation. Thus the role of the mediator in bringing the two players to cooperation bears scrutinizing. For, as Alan J. Kuperman, in his unique and provocative interpretation of the mediation that culminated with the Accord, posits “that mediators sometimes inadvertently provoke the very tragedies they seek to prevent. Indeed the events leading up to the tragedy in Rwanda demonstrate that mediation is an inexact science” (1996: 221). Though by its own admission Tanzania’s leverage was not substantial, along with RPF leverage in the form of military threat, and international diplomatic and economic pressure, it compelled Habyarimana to sign the Accord (although he never intended to follow it), demonstrating as Kuperman says “that leverage is a double-edged sword, equally capable of driving contending parties to the most extreme measures” (221). In some conflicts, the level of suffering rises to an intolerable level at which point “escalation” is no longer offering returns and disputants are propelled toward mediation. This condition is “mutually hurting stalemate” (Kuperman, 1996: 227). If this doesn’t occur on its own, the mediator introduces leverage in forms that are positive, “carrots,” (aid, trade) or negative, “sticks,” (sanctions, embargoes). In many prisoners’ dilemma scenarios, the effective mediator must influence the perceptions of the players and work toward crafting the very structure of the game” (Levy, 1985: 590). Such work can be actualized by altering cooperation thresholds by affecting the player’s costs of standing firm or cooperating (Levy, 1985). Leverage is also commonly applied, as in the case of Rwanda. According to Kuperman, however, this leverage 238 provoked its own form of dilemma. Leverage came in several forms. First, international and economic: in 1993 “the United States conditioned Rwanda’s economic aid of $20 million on continued democratization, respect for human rights and the rule of law, and reduced violence, therefore siding with the RPF. When the Rwandan government failed to comply, the US applied the stick, terminating all but humanitarian assistance and capping aid at $6 million” (1996: 227). IMF and World Bank suspended all aid. This negative leverage served its purpose by forcing Habyarimana to the table. After extended talks, an agreement that included “a comprehensive roadmap for transition to a genuine multiparty democracy” (Kuperman, 1996: 5) was reached on August 4, 1993. Nonetheless, many issues remained unresolved and the threat of violence from both sides continued to simmer. By the time the parties signed the comprehensive version of the Accords in August, 1993, Radio-television Libre de Milles Colline (RTLM) was undertaking vituperative anti-Tutsi radio broadcasts in earnest. The transitional government to be established within 37 days, under the terms of the agreement, did not manifest. Over several months Habyarimana continued to haggle over the distribution of power in the transitional government, most forebodingly lobbying for inclusion of the anti-Tutsi extremist party, CDR. On April 3, 1994 ambassadors “from key western nations met with the president, insisting that he immediately install the diplomatic institutions” (Kuperman, 1996: 230). He finally consented to sign on April 6 in the face of mounting diplomatic pressure. One minister in attendance at the historic summit in Dar es Salaam succinctly remarked, “[T]here was a great deal of international pressure to 239 proceed with the implementation of the agreements. The Azaku [the ruling northern-Hutu elite] realized the game was up” (qtd. in Kuperman 1996: 230). That same day the plane carrying Habyarimana and the president of Burundi was shot down and the genocide began. What Mpungwe offered in the negotiations was a dramatic change in balance of power and an opportunity to delay a RPF victory, a sacrifice too great for Habyarimana to make. And yet simultaneous economic, diplomatic, and military pressures transformed a compromise and path to peace into an escape route. With Habyarimana’s death it is impossible to know how events would have unfolded if he had returned to participate in the transitional government to which he had just agreed. Extremist forces were themselves becoming much more vocal and were highly critical of the president’s gestures of conciliation, however meaningless they were. Azaku privilege was hard-won and not willingly relinquished. Was genocide an inevitable culmination of this trajectory? To what degree did negotiations with leverage applied from all sides create a “perfect storm” of extreme circumstances? Kuperman convincingly exposes the flaws in the mediation: the dearth of good intelligence, the failure of critical understanding, of deciphering the elaborate campaign of duplicity; as he bluntly, but accurately, puts it, even as Habyarimana “negotiated and signed an elaborate plan to share power with the Tutsi, he simultaneously established a military and communications infrastructure to annihilate them” (1996: 232). Though his assassination (at whose hands is still unknown, though theories abound) prevented his overseeing the genocide, its systematic and immediate instigation signifies 240 both that planning had been long underway and that the mediators failed to appreciate the degree to which extremists were driving policy. UNAMIR’s mandate to monitor peacekeeping efforts following the agreement did not include the authority to use force, nor could it have under international laws regarding sovereign states (Kuperman, 1996: 236). Still the UN played an indecisive role and lost opportunities to strengthen UNAMIR, which had been permitted to deploy thousands of soldiers as peacekeepers. UN Secretary General, Butros Ghali, repeatedly denied requests by General Dallaire to disrupt arms distribution and deploy more soldiers. As a result, UNAMIR and the UN lost credibility as they could no longer be effective mediators or use their power to prevent escalation of the conflict. It became evident that Rwanda was not a priority for the members of the UN Security Council. Furthermore, as Kuperman explains, “[T]he Security Council simply did not have the will for a peace enforcement operation so soon after the Somalia and Bosnia debacles” (1996: 236). For similar reasons, the Clinton Administration did not want any other external entanglements, even low-risk humanitarian intervention (Dekmejian, 2007: 283) and thus “used Rwanda as the test case of its new, restrictive UN peacekeeping policy contained in Presidential Decision Directive 25” (Kuperman, 1996, 236). Samantha Power (2003) also comments on the reasons for the failure of intervention by the United States and others, adding that after the Somalian expedition, the Clinton Administration lost interest in humanitarian intervention, and faced a 241 formidable Republican challenge in the November 1994 elections. Furthermore, Rwanda had no strategic value in anyone’s calculations of self-interest. Andrew Wallis writes that when General Dallaire returned from Rwanda, he gave a highly charged speech where “he admitted his own failure to save those who were killed – but went further, attacking the very mindset of the West and its politicians in allowing the slaughter to take place without question or concern” (Wallis, 2007: xi). In his speech, Dallaire cynically declared, “Rwanda is black. It is in the middle of Africa. It has no strategic value. And all that’s there, they told me, are people, and there are too many anyway” (Wallis, 2007: xi). Though France and Belgium were more deeply involved in Rwandan affairs, when the death toll of their peacekeepers rose, even their commitment declined. Wallis explains that France, was the only Western nation to take an interest in Rwanda. Unfortunately, for its people, this interest was borne out in supporting a government that was intent on solving its political problems by mass murder. Ultimately, “stopping the butchery of thousands of black Africans in a remote land was not a priority in the UN’s global agenda and irrelevant to the self-interests of its member states” (Dekmejian, 2007: 287). Table 1.4: Breakdown of Conflict Resolution in Rwanda Low Cost (LC) or High Cost (HC) Armament Cooperate (C) or Defect (D) Preferences Met? (Yes or No) Hutu’s HC D No Tutsi’s LC C No 242 Table 1.6 helps clarify the reasons for failed mediation in Rwanda: The Hutu’s choice to maintain a high-cost armament strategy made compromise impossible, preventing either side from achieving its preferences and thus further diminishing the possibility of a successful resolution. Prisoners’ dilemma can serve as a valuable heuristic tool for considering the rewards of cooperation versus the so-called “sucker’s payoff” for compliance when the other disputant refuses, or the appeal of competing in the face of the opposition’s cooperation (Kuperman 1986: 584). When the disputants have no vested interest in resolving the conflict, however, but in fact have interest in maintaining the status quo, as in Rwanda, the usefulness of the prisoners’ dilemma is extremely limited. As Levy explains, “[T]he more ambivalent one or more parties are about the relative merits of a negotiated solution over continued deadlock, the more closely the dynamics of the prisoners’ dilemma game will approach deadlock” (1986: 584). When faced with deadlock, mediators apply various forms of leverage. But what the case of Rwanda demonstrates most dismally is that leverage very well may have pushed Habyarimana to concede to terms utterly antithetical to himself and, possibly even more, to the extremists in his party. Even if the president signed the agreement without any intention of following through, the conciliatory measures were still unacceptable to the extremists. With a steady stream of anti-Tutsi propaganda, CDR gained ideological domination over the MDNR. 243 The other option would have been a protracted negotiation, a continued pattern of violence, deadlock, and resumed mediation, and perhaps genocide would have occurred anyway. With benefit of hindsight, the Rwanda mediation reveals astonishing failures in intelligence and judgment. Sadly, as Kuperman asserts, “[I]f Rwanda teaches nothing else about conflict management it is that, sometimes, doing the wrong thing is far worse than doing no thing at all” (1986: 238). Rather than headlines announcing reconciliation, a nonviolent transition, or a final repudiation of racist beliefs, perpetrators initiated the now infamous genocide commenced without hesitation. By July 1994, international headlines sought to put the horror into words, as in the case of this article in The Scotsman: entitled “Nightmare for Rwanda,” the piece bluntly reported, “Rwanda is in the midst of a terrible humanitarian disaster.” 244 CHAPTER TEN: CONCLUSION If the peace brokering in Northern Ireland showed how mediation can go fundamentally right, that of Rwanda demonstrated how it can go dismally wrong. But even the most ideal mediation process will have its flaws. Mediation will never become an “exact science,” but it can improve the chances of bearing an effective outcome to a conflict by becoming more attentive to the measured and careful work that it requires. In this dissertation, I have applied a modified prisoners’ dilemma assessment to explicate how mediators can move disputants from fighting and stalemate to negotiated settlements. By removing mediation from the realm of idealism and considering these cases in their complexity, this analysis illuminated some of the ways that mediators move parties past seemingly insurmountable problems. The application of the two hypotheses asserted in the introduction to the case studies of Rwanda and Northern Ireland have served the study of mediation’s complex and often thorny issues. To recall, the two hypotheses are: 1. Mediation is unlikely to be successful unless the mediator has some form of leverage/power. 2. Finding the right measure of interest in dispute resolution is central to the mediator’s success. The first hypothesis that states that mediation is unlikely to be successful unless the mediator has some form of leverage proved to be true for the conflict in Ireland. President Clinton and George Mitchell were both successful mediators because they had 245 the leverage to entice both parties to cooperate. In Rwanda, the mediation process was not a success in part because the mediators did not hold the substantial or compelling positive leverage to bring both parties to a productive cooperation threshold. The first hypothesis is therefore also true for the case of Rwanda. The second hypothesis-- that having the right measure of vested interest in the dispute’s resolution is central to a mediator’s success-- proved true for the mediators in Northern Ireland. President Clinton’s interests revolved around his political aspirations for more votes and his desire to leave behind a legacy. While George Mitchell had a deep desire and connection to the Irish people, he was also working on behalf of President Clinton who clearly had invested interests. In the case of Rwanda, the second hypothesis also proved true. Since none of the mediators had enough of an interest in resolving the conflict in Rwanda, the mediation process never accumulated enough momentum to make it succeed. The conflict in Rwanda was not a high enough priority for the mediators involved nor for the international community, who asserted economic leverage but no significant military, political, or diplomatic concern. It can therefore be asserted that the two hypotheses proposed at the beginning of this study can be declared true in both cases. As a theoretical tool, prisoners’ dilemma has proven useful in evaluating the risk management of negotiation scenarios, as the mediator seeks equilibrium by exchanging information and offering incentives that serve the interests of both parties. A careful consideration of how Moore's stages apply to the cases of Northern Ireland and Rwanda 246 has revealed how the incremental framework can organize and breakdown mediation processes so they can be better understood and thus serve to improve future negotiations. By applying the stages to each case, I demonstrated how George Mitchell achieved success in Northern Ireland, while Ami Mpungwe failed to do so in Rwanda in spite of outwardly reaching an agreement. This study has revealed that the concepts of Moore’s framework and PD reveal valuable information about conflict scenarios but that that even these heuristics have their limits. Indeed, an application of Moore’s stages to Rwanda revealed the defects and oversights to the mediation process, however it did not and could not predict the duplicity and volatility of human behavior. Likewise, though PD served to reveal the aids and obstacles to resolution, like all game theory, it depends upon rational actors. That Habyarimana acted in bad faith (although he was entirely rational from his own perspective), tainted the mediation process from beginning to end; its souring within hours of being finalized signaled that the genocide had long been in the offing. In the case of Northern Ireland, the success of the Good Friday Agreement was due in large part to the extensive international mediation that preceded it. The failed Brooke-Mayhew initiative demonstrated that it was virtually impossible for the British and Irish governments to oversee the process themselves, directly invested as they were in its outcome. The troubled history of Northern Ireland and the animosity and distrust between the antagonists made peace unattainable without outside help. To be viewed as impartial and just by all parties, the peace process required external intervention by a neutral third party, someone who would break the patterns of thinking and present viable 247 and compelling alternatives to violence. Though backed by a powerful nation, Mitchell approached the parties in a characteristically personable and dutiful manner. Whatever Clinton’s motives were—and, as I have discussed, there were probably several— Mitchell’s dedication to negotiating peace had its own reasons. According to Ruth Weiss, Mitchell was a man who appeared totally calm in public, and in private listened to all sides over long periods, showing understanding in the face of rival recriminations and accusations. He was thus able eventually to bring together the two men that mattered, David Trimble and Gerry Adams – no mean feat for any mediator. (Weiss, 2001: 10) His warmth and genuine concern about the people of Ireland came across in his mediation style, an approach that helped to break down the barriers and set a new tone in a conflict long characterized by hostility and belligerence. During the negotiations, Mitchell increased levels of trust between himself and the respective parties. At first, the intervention took the form of a consultative body, the International Body of Decommissioning. Since it lacked the authority to enforce its views and was limited in scope, the International Body posed little risk to individual parties. The lack of authority (as opposed to potential influence) accounted for the absence of resistance to international involvement generally, and the International Body’s composition in particular. Not until the international mediators assumed positions of influence within the framework of multiparty talks did various parties raise objections, but by this time the peace process had gained significant momentum. For his part, Mitchell remarked that he was able to build trust through a series of small, procedural decisions and was ultimately trusted enough by all parties to make 248 significant procedural and substantive suggestions. The vital test of Mitchell’s influence came in the final months of the negotiations in his effort to manipulate the process and pressure the parties into an agreement. That his decision was met with approval by the parties was seen by Senator Mitchell as a testament to the level of trust that had accumulated between him and the parties (Crocker, Hampson & Aall, 1999: 445). In essence, then, the following conditions led to successful peace negotiations in Northern Ireland: Mitchell’s access to formidable resources, most of which was in the form of positive leverage; Mitchell’s characteristic patience, persistence, and fairness; the nonpartisan position of Mitchell and Clinton and the new precedent they brought to the talks; the parties’ recognition of each other’s strengths and of the possibility of a precipice; and the coherent post-treaty strategy. With these qualities, Mitchell was able to approach the conflict’s implicit prisoners’ dilemma scenario from a position of strength. Given the long history of the conflict and the depth of the division, movement to a cooperative solution presented a substantial challenge. Through leverage and careful persuasion, Mitchell was able to show that the benefits of collaborating were superior to the cost of maintaining the status quo. Mitchell’s standing as a mediator has stood the test of time. In 2009, President Obama appointed him to be the Middle East envoy. Responding to this choice, the Belfast Telegraph declared “There can hardly be a person on the planet who is better qualified for this complex and sensitive role.” The journalist continues by dubbing Mitchell “one of the heroes of the Northern Ireland peace process,” lauding his many qualities as a mediator. Importantly, however, the journalist remarks on the uniqueness of 249 the Northern Ireland conflict, how each mediation scenario, as Mitchell himself recognized, has its own dimensions, root causes, motivations, and cooperation thresholds: [I]t would be a mistake to assume that the Northern Ireland template could simply be used for the Middle East. Both conflicts have deep-seated dimensions of religion and nationality, but the problems are not the same. One of the factors which made peace possible in Northern Ireland was the war-weariness of the main participants, and a realization that neither side could win. . . Sadly, in the Middle East, the protagonists have not yet reached the point of recognizing the primacy of politics rather than violence . . . That said, George Mitchell has the wisdom and experience to move the warring parties cautiously in the right direction.(“Mitchell Will be a Perfect Man for the Job,” 2009: 16) Aside from offering due praise, the journalist elucidates the inherent challenges of mediation. In spite of the notable usefulness of frameworks such as Moore’s 12 steps and the prisoners’ dilemma, no mediation scenario is the same. What a comparison of the case studies of Northern Ireland and Rwanda has manifestly indicated is that disputants themselves must recognize the exigency of third party mediation, must see the “ripeness” of the situation for themselves, and must participate with intention. Mediators such as Mitchell demonstrate the impressive work that can be done, and how careful and conscientious attention to a conflict can culminate in a mutually satisfying resolution— but not until disputants acknowledge that violence is no longer effective. In Northern Ireland, “war weariness” finally drove the parties to recognize that “winning” would not occur through violence, but that a modified version of winning could be effectuated through political strategies of compromise and cooperation. Disputants in the Middle East do not seem ready to replace violent tactics with political dialogue. Those interested in international conflict management will surely be watching Mitchell’s work there very closely. 250 For all the uniqueness of the Northern Ireland situation, what lessons can be gleaned to serve future mediation efforts? In what ways can it contribute to a model for future conflict mediation? Mitchell’s effort to include all parties (including citizens themselves, through the referendum format) in the development of the country’s constitutional roadmap created a broad feeling of support for peaceful and democratic means. By even including marginalized entities, Mitchell preempted problems with spoilers and cultivated a climate of inclusiveness. Furthermore, that he arranged a power- sharing deal involving both the legislative and executive branches, contributed to accountability, checks and balances, and proportional representation. Because of his careful attention to “citizenship concerns relating to the equality of social, economic and cultural rights of all ethnic communities” (English, 2003: 317), the process had the support of the people and stirred in the region a general sense of support and willingness to move past the conflict. Unfortunately, on August 15, 1998, four months after the agreement was made, Republican extremists opposed to the peace process detonated a bomb in the quiet Northern Ireland market town of Omagh. The blast killed 28 people and injured over 200. It was the single worst attack in almost 30 years of violence in Northern Ireland (English, 2003: 318). Encouragingly, however, Sinn Fein chairperson, Mitchel McLaughlin observed, “All of us must change. All of us, Irish Republicans, members of the British and Irish governments, unionists and nationalists, have a duty to ensure that such terrible actions never again occur” (English, 2003: 318). Though the Agreement had not eradicated violence completely, attitudes toward violence were undergoing a change. 251 For the Westminster government, the Omagh attack was, in the words of Gerry Fitt, “the worst outbreak of violence I can ever recall in Northern Ireland” (English, 2003: 318). Instead of cutting the peace process short, however, the attack galvanized the resolve of those working for peace in both communities. Resolution in the Northern Ireland case can be deemed a success according to the four requisites of success. First, after the 1998 bombing, the level of violence decreased dramatically. Second, mechanisms for coexistence were created to help ensure that the agreement succeeded in implementation. Third, the terms of the agreement were recognized and held long enough to establish and legitimize a standard. Fourth, the perceptions expressed by the parties and mediator(s) were in conformity that the agreement was a success and that a new precedent had been established. This impressive conclusion was further legitimized when SDLP leader John Hume and UUP leader David Trimble were awarded the Nobel Peace Prize in 1998 for their bold and courageous work in the negotiation process. By most accounts the Northern Ireland peace process was a success, culminating as it did with the highly regarded Good Friday Agreement. The European Peace Programme described it as “the most successful peace process” and a “blue-print for countries grappling with internal conflicts.” (“Kenya: Lessons From Power Pacts in Other Countries,” 2008). Thus in the introduction to this dissertation--and for purposes of discussion-- I designated Northern Ireland as an exemplar of successful mediation. But the region has not been without its problems since the Agreement was reached, nor can 252 there ever be certainty about the aftermath of any mediation. (As such, any assessment of mediation’s success will be time-bound and impermanent). As recently as October 2009, British Prime Minister Gordon Brown and US Secretary of State Hillary Clinton traveled to the region to assist with the British Government’s devolution, which involves the complicated process of giving over the power of policing and justice to Northern Ireland. What the results of these efforts will be is yet to be determined, but interested parties are holding their breath, hoping that paramilitary violence does not resume. Northern Ireland and, to a much more dramatic degree, Rwanda indicate the importance of post-negotiation support after the so-called “heavy-lifting” has taken place. For the benefit of all mediation, this subject certainly warrants further study. In spite of praise for the “textbook” quality of the Rwandan mediation, the outcome was anything but successful. As such, the prisoners’ dilemma formulation as it applies to Rwanda is extremely complex. Even as the Tanzanian mediator and the RPF received indications that the MRND party was interested in cooperating, it was, in fact, temporizing. In such an inversion of PD’s purposes, then, stalemate covertly served as a stalling technique, specifically affording the MRND and its extremist allies an opportunity to work behind closed doors, take stock of resources, and galvanize forces. Continued play provided the opportunity to systematize and reinforce the very methods the MRND claimed to be repudiating. Substantial leverage was applied in the Rwanda negotiations; perhaps, as I have asserted, too much. 253 Just a cursory comparison between Northern Ireland and Rwanda reveals several differences in the nature of the leverage: where the US applied positive leverage in the former case, in the latter it utilized negative forms of persuasion. Some scholars refer to this difference as a “carrots-or-sticks” hypothesis, which considers the comparative effectiveness of a mediator engaging disputants through a cooperative or conflictual approach. If mediation’s success is determined by its culmination with a signed agreement, then certainly the Arusha Accords represented a triumph of conflict management. But such a characterization, as I hope this dissertation has made clear, is inaccurate at best and irresponsible at worst. First, levels of violence continued to escalate after the Arusha Accords were signed. The magnitude of the genocide that was committed after the signing of the Arusha Accords was so horrific that attempting to revive the peace process became impossible. Second, while mechanisms for coexistence were put in place, weak foundations and lack of credibility made them useless and irrelevant in supporting the peace process. Third, the terms of the agreement were held for such a short period of time that a standard could not effectively be established and legitimized. Fourth, a new precedent was not successfully established and the parties decided to continue to pursue violent tactics. In brief, the following conditions lead to the failure of the peace accords: incoherent treaty terms; failure to meet the needs of both parties equitably; ineffective communication; dearth of intelligence and insight; lack of deliberation about the impact 254 of mediation; insurmountable animosity and distrust between parties; low priority status for the mediator(s); and, negligence by the international community. In the face of all of these flaws, Habyarimana was perfectly willing to go along with the negotiations because he likely had no intention of following through. He thus “stayed at the table,” as the language of PD goes, without ever showing his cards. Still, the inherent risk in holding out through feigned cooperation was inciting extremist anger. At what point he abandoned the cooperative goal or whether he ever was committed to it is unknown. Equally elusive is the degree to which excessive RPF concessions paired with RPF military might became a tipping point toward genocide. Although at the time of the signing of the Arusha Accords many considered it to be the best peace agreement in Africa since Lancaster House, some subsequent commentary condemned the Accords as having, in fact, contributed to the deadly violence in 1994. As previously discussed, compelling theoretical analysis by Kuperman faults the Accords for not neutralizing the extremists, also arguing that excessive international pressure must be factored into the equation (1996: 221). Other critics assert that the Accords broke a fundamental tenet of conflict resolution by failing to give the extremists a stake in the new government (Jones, 1995b: 18). Those who argue this point explain that the extremists were constrained first by exclusion from the government and then by losing control over the military. Hard-liners were left with the stark choice of loss of power or violent opposition (Jones, 1995b: 9). Arguably, extremists might not have perpetuated their plans had they been included in the government--however, circumstantial evidence for this position does exist. 255 While the extremists clearly did not embrace the Arusha process, at least initially, they did not regard it as irrelevant. During the BBTG negotiations, they argued ardently for the inclusion of CDR (Jones, 1995b: 9). At some point during the military discussions, however, they abandoned interest in the process. At opposing sides of Habyarimana, the internal political opposition and the RPF military forces zeroed in with significant pressure. Notably, several observer groups counseled restraint on the part of the RPF, pointing out that the RPF’s military strength, combined with the GoR’s deteriorating economic situation, placed considerable independent pressure on the GoR. For its part, the OAU had the political will to be significantly involved in the implementation of the Accords; therefore the disjuncture between the OAU and the United Nations represents a crucial missed opportunity. The OAU continuously lobbied for involvement and its troops stayed in Rwanda after the fighting began. In the future, cooperative relationships between the United Nations and regional organizations may represent a way of overcoming the perennial lack of political will to undertake remote peacekeeping operations demonstrated by the international community. Such arrangements might be criticized as mercenary but, as with the OAU, these organizations are often interested in addressing their own regional problems and actually can do that with some technical and financial assistance. In the aftermath of genocide, the swath of blame has been wide and deep. As I mentioned earlier, the practical recommendations from US participants were primarily a response to the low priority that the talks represented for the United States. Among the few US government officials to pronounce upon the imminence of crisis in the Rwandan 256 situation, United States Assistant Secretary of State Herman Cohen would later speak out about the shameful negligence of the Clinton Administration in the face of mounting disaster. As he grimly assessed the US response, "We provided zero support for the rebels, zero arms, zero logistics.” Adding further criticism, as he explains, "We did not immediately call these crimes by their rightful name: genocide" (“Ex-U.S. Official, in Paris,” 1998: A10). Though not quantifiable, the degree to which US inattention fueled grim economic circumstances and signaled to the perpetrators that violence would be overlooked by the international community remains unknown. Good intelligence is important to any mediation. Observers argue that those who mediated and implemented the Arusha Accords needed better information. In hindsight the warning signs appear clear, but for many participants at the time the magnitude of the impending crisis was not obvious. Participants and members of the RPF remember dismissing RTLMC precisely because it was so literal and extreme (Adelman & Suhrke, 1996: 35). For some, the January massacres were deciphered as a negotiating tactic. Still others believed that some deaths were the unavoidable result of the transition. Part of the problem may also have arisen because of the dramatically different atmospheres that existed in Arusha and Kigali. For example, Adelman and Suhrke explain that when the Accords were signed, no celebration occurred in the Rwandan capital. Instead the city was dominated by a state of fear, especially acute among the Tutsi population (1996: 187). Tragically, what almost no one anticipated before early 1994, including the human rights community, were the brutally merciless lengths to which the extremists 257 were willing to go in their frantic bid to maintain power. There are strong arguments that a robust response by the international community could have stopped the killings before they spread throughout the country. Instead, the United Nations, after having insisted on unilateral control, simply withdrew. Although the UN Secretary-General Boutros-Ghali was perhaps trying to spread the blame, his fury about the world’s inaction was shared by many: “We are all to be held accountable for this failure, all of us, the great powers, African countries, the NGO’s, the international community. It is a genocide . . . I have failed . . . It is a scandal” (Prunier, 1995: 227). Moreover, as head of the late 1993 UN peacekeeping mission in Rwanda, Canadian General Dallaire also asserts that, “ultimately, led by the United States, France and the United Kingdom, the world body aided and abetted genocide in Rwanda. No amount of its cash and aid will ever wash its hands clean of Rwandan blood” (Dallaire, 2004: 323). In 90 days, as many as a million people died in what President Clinton called the most rapid “slaughter in this blood-filled century we are about to leave. It was a tragedy for which the United States and other members of the international community must take blame” (Charny, 1999: 13). The Arusha Accords stand as a testament to the strength of and the implicit danger of third-party intervention. They reveal that even the most careful resolution is not complete until fully implemented. They also underscore the deep accountability of third parties to maintain their commitment once having accepted the burden of involvement. Especially for small countries, the international community has the ability 258 and power to dramatically alter the course of events. International powers must remain fully aware that partial efforts are likely to be worse than no efforts at all. As of November 2009, France was still seeking to repair its deeply damaged relationship to Rwanda. Following its military and economic support of Hutu forces and allegations of complicity in the genocide, France has had its share of diplomatic work to do. Though a French parliamentary investigation in 1998 found that "errors of judgment ...were made" but rejected the claim that French peacekeepers helped the Hutu militants (Marquand, 2008: 25), relations have been tenuous since 1994 when the RPF seized power in Kigali and forced out the French-supported Hutu regime. On November 30, 2009, President Sarkozy’s administration released a statement emphasizing the country’s support of peace initiatives in the Great Lakes region, vowing to “contribute actively in reinforcing actions of the international community against people responsible for the genocide in Rwanda and armed groups who are destabilizing eastern Congo” (“France to Support Great Lakes Region,” 2009). How can future research build upon the findings in this study? In analyzing the case of Northern Ireland, one of the prominent factors was the personality of George Mitchell. During the course of conducting research, there was no shortage of praise for his charismatic personality, his impassioned dedication to the process, and his personal interest in the well-being of the people of Northern Ireland. Conversely, in discussions of the Rwandan mediation, Mpungwe’s personality goes virtually undiscussed; studies of the Tanzanian negotiation have nothing negative or positive to say about him. A study of the personalities and characteristics of the party 259 leaders participating in the mediation could add to our understanding of mediation processes. Do certain personality traits contribute to a quicker more smooth resolution? What qualities does a leader need in order to ensure that a conflict gets resolved quickly and effectively? Leadership personalities in well-known mediation events could be productively compared; in addition to George Mitchell, figures such as Bill Clinton, Henry Kissinger, Jimmy Carter, Oscar Arias Sánchez, Chester Crocker, Lakhdar Brahimi, Lazaro Sumbelywo, Martin Griffiths, Erik Solheim, Vidar Helgesen, Alvaro de Soto would be useful case studies. How does mediator personality affect the mediation itself? Is a patient, soft-spoken, polite mediator more effective than one who is more assertive and demonstrably charismatic? What particular qualities do effective mediators share? In the words of one scholar, “the words most often used to describe successful mediators” are vision, drive, perseverance, discipline, courage, integrity, strategic thinking, endurance, management skills, intelligence, knowledge, political awareness, ability to communicate, confidence, willingness to take risks, commitment, force of character, respect, fairness, expertise, humility, good listening skills, sound judgment, patience, backbone, sensitivity, and maybe a touch of humor as well. (Jacobsen, 2003: 234) Focusing more specifically on the distinct personalities of prominent mediators could provide new insight into conflict resolution, especially if undertaken through transdisciplinary research that embraces the fields of psychology, political science, history, sociology, and communications. Applying the fundamentals of a prisoners’ dilemma dynamic to current conflicts may serve to move disputants out of a deadlocked situations. Whether it be the 260 Palestinian-Israeli conflict, the Georgian-Russian conflict, or even the issue of gangs in the United States, it is possible that the Prisoner’s Dilemma may serve as a constructive means of modifying disputants’ conflict thresholds. By providing insight into the relative benefits of cooperating versus holding out the mediator(s) can create an appropriate strategy to best advance compromise. Though no panacea, the Prisoner’s Dilemma can give mediators advantages to increase the chances of focusing on a cooperative goal and thus achieving compromise and resolution. For its part, Moore’s methodology provides a valuable framework for breaking down a conflict situation and pinpointing where mediation failed. As I hope to have demonstrated, Moore’s steps prove useful to the researcher or analyst organizing material from extended negotiations. In other words, his 12 steps can serve as a guide to mediations or a heuristic tool for examining mediations that have already taken place. The two cases used in this study illuminate the steps that are more likely to lead to successful mediation processes and how they could be studied and applied to pending conflicts. Even as I write this dissertation, President Barack Obama was awarded the Nobel Peace Prize, an honor that has traditionally gone to peace mediators. A controversial designation, the committee’s selection of Obama to be a peace laureate has largely been justified by the steps he has made and the potential it believes he holds for global peace. As Thorbjoern Jagland, chairman of the Nobel Committee, explained. "In the past year Obama has been a key person for important initiatives in the U.N. for nuclear 261 disarmament and to set a completely new agenda for the Muslim world and East-West relations" (“Obama Wins Nobel,” 2009). Global relations have undergone dramatic changes in the post-Cold War world. As a result, so too have global conflicts. The prevalence of terrorism and so-called war on terror has induced new realizations about the nature both of international disputes and of international conflict mediation. As the cases of Rwanda and Northern Ireland have dramatically indicated, the parties’ desire to participate in mediation is foundational. At a time in which many leaders and participants to disputes remain underground, covert, or anonymous, who will mediators bring to the table? Who will undertake such high stakes mediation? What new qualities will a mediator need in these times of borderless global conflict? What will international conflict mediation look like in a world of rogue players and nationless, ideology-driven disputants? How, quite literally will nation states talk to their enemies? What can models based on local and regional disputes offer to the global scenario? The 2008 book Talking to Terrorists: The Myths, Misconceptions, and Misapplication of the Northern Ireland Peace Process by John Bew and Martyn Frampton is part of important recent work seeking to address this conundrum. Speaking of diplomatic engagement with Hamas, the authors caution “in some instances, the willingness of a state to negotiate might encourage the terrorists to believe that their opposition are [sic.] ready to concede—even when this is not the case” (Coyle, 2008: 6). At the same time, no less than Gerry Adams himself, who has traveled to the Middle East to discuss peace brokering, insists “While no two conflicts are identical, 262 there are key conflict-resolution principles which can be applied in any situation.” There is important research yet to be undertaken about twenty-first century mediation and the technology, framework, and objectives it will involve. Granting Obama the Nobel for his fresh approach to old disputes indicates the continued importance of mediation. We know that mediation can change global relations, what we are yet to see is how global relations will change mediation. 263 BIBLIOGRAPHY Adams, Gerry. A Farther Shore: Ireland’s Long Road to Peace. United Kingdom: Random House, 2005. Adelman, Howard, and Astri Suhrke, “Study 2: Early Warning and Conflict Management.” The International Response to Conflict and Genocide: Lessons from the Rwanda Experience. Joint Evaluation of Emergency Assistance to Rwanda, 1996. ---. The Path of Genocide: The Rwanda Crisis from Uganda to Zaire. New Jersey: Transaction Publishers, 2000. Akenson, Donald. The United States and Ireland. Cambridge: Harvard University Press, 1973. Allison, Graham T. and Philip Zelikow. Essence of Decision: Explaining the Cuban Missile Crisis. New York: Longman Press, 1999. Alonso, Rogelio. The IRA and Armed Struggle. London: Routledge Press, 2007. Arrow, Kenneth et al., eds. Barriers to Conflict Resolution. New York: W.W. Norton and Company, 1995. Augsberger, David. Conflict Mediation Across Cultures: Pathways and Patterns. UK: John Knox Press, 1992. Aumann, Robert J., and Sergiu Hart, eds. Handbook of Game Theory with Economic Applications, 3 vols. Amsterdam: North-Holland, 1992-2002. Baird, Douglas, Robert H. Gertner, and Randal C. Picker. Game Theory and the Law. Cambridge, MA: Harvard University Press, 1994. Barnett, Michael. Eyewitness to a Genocide: The United Nations and Rwanda. New York: Cornell University Press, 2003. Bass, Gary Anthony. Freedoms Battle: The Origins of Humanitarian Intervention. New York: Alfred Knopf, 2008. Beardsley, Kyle. “Politics by Other Means than War: Understanding International Mediation.” Dissertation; University of California, San Diego, 2006. 264 Beardsley, Kyle, David Quinn, Bidisha Biswas, and Jonathan Wilkenfeld, “Mediation Style and Crisis Outcomes.” The Journal of Conflict Resolution 50.1 (2006): 58- 86. Becker, Carl L. “What Are Historical Facts?” The Western Political Quarterly 8.3 (1995): 327-340. Bennett, Peter. “Modeling Decisions in International Relations: Game Theory and Beyond.” Mershon International Studies Review 39.1 (1995): 19-52. Bercovitch, Jacob. Studies in International Mediation. London: Macmillan Press, 2003. Bercovitch, Jacob, J., Theodore Anagnoson, and Donnette L. Wille. “Some Conceptual Issues and Empirical Trends in the Study of Successful Mediation in International Relations.” Journal of Peace Research 28(1): 7-17. Bercovitch, Jacob, and Jeffrey Rubin. Mediation in International Relations: Multiple Approaches to Conflict Management. London: Macmillan Press, 1992. Bew, Paul, Patterson, Henry, and Paul Teague. Between War and Peace: the Political Future of Northern Ireland. London: Lawrence & Wishart, 1997. Bew, Paul and Gordon Gillespie. The Northern Ireland Peace Process: 1993-1996. London: Serif, 1996. Bingham, Gail. Resolving Environmental Disputes: A Decade of Experience. Washington, D.C.: The Conservation Foundation, 1986. Binmore, Ken G. Game Theory and The Social Contract, Cambridge, MA: MIT Press, 1994. Blake, Robert and J. Mouton. Solving Costly Organizational Conflicts. San Francisco: Jossey-Bass, 1965. Bloomfield, David. Political Dialogue in Northern Ireland: The Brooke Initiative, 1989- 92. London: Macmillan Press, 1998. ---. Peacemaking Strategies in Northern Ireland: Building Complimentarity in Conflict Management Theory. London: Macmillan Press, 1997. ---. Developing Dialogue in Northern Ireland: The Mayhew Talks. New York: Palgrave, 2001. 265 Bloomfield, Kenneth. A Tragedy of Errors: The Government and Misgovernment of Northern Ireland. Liverpool: Liverpool University Press, 2007. Bolduc, D. “Implications for Restorative Justice: Mediator Styles and Mediator Behavior,” paper presented at the North American Conference on Peacemaking and Conflict Resolution (NCPCR), 1989. Bowcott, Owen. “Sinn Fein ‘will take risks’ for Brooke Talks.” The Guardian. Feb. 4, 1991. Boyce, David.G. The Irish Question and British Politics, 1868-1996. London: Macmillan Press, 1996. Boyle, Kevin, and Tom Hadden. Northern Ireland: The Choice. London: Penguin, 1994. ---. “The Peace Process in Northern Ireland.” International Affairs 71.2 (1995): 269-283. Brams, Steven J. Superior Beings: If They Exist, How Would We Know? Game Theoretic Implications of Omniscience, Omnipotence, Immortality, and Incomprehensibility. New York: Springer-Verlag, 1983. Brams, Steven, and Jeffrey Togman. “Cooperation through Threats: The Northern Ireland Case.” Political Science and Politics 31.1 (1998): 32-39. Branch, Taylor. The Clinton Tapes: Wrestling History with the President. New York: Simon and Schuster, 2009. Breadun, Deaglan de, “North Parties Gear Up for Bitter Assembly Elections.” The Irish Times 25 May 1998. Burkhalter, Holly. “The Question of Genocide: The Clinton Administration and Rwanda” World Policy Journal,11,4 Winter (1994): 44. Burton, John, and Frank Dukes. Conflict: Practices in Management, Settlement and Resolution. New York: St. Martin’s Press, 1990. CAIN: Conflict Archive on the Internet. 18 October 2009. http://cain.ulst.ac.uk/index.html Callaghan, David. Unwinnable Wars: American Power and Ethnic Conflict. New York: Hill and Wang, 1997. Caplan, Gerald. “Yankees Go Home.” The Globe and Mail (Canada). 21 June 2001: A15. 266 Carlton, Charles. Bigotry and Blood: Documents on the Ulster Troubles. Chicago: Nelson Hall, 1977. “Cautious Optimism in Ireland” The Independent 11 Jan. 1990: 28. Charny, Israel. Encyclopedia of Genocide. United States: ABC-CLIO press, 1999. Chaudhuri, Ananish. Experiments in Economics: Playing Fair With Money. Routledge, 2009. Clapham, Christopher. “Rwanda: The Perils of Peacemaking.” Journal of Peace Research 35.2 (1998): 193-210. “Clinton Apologies.” Times (London). 24 July 2005: 28. “Clinton’s Hope for the IRA” 9 January 1993: 1. Clinton, Bill. My Life. United States: Knopf Publishers, 2004. ---. Giving: How Each of Us Can Change the World. United States: Knopf Publishers, 2007. Cohen, Jared. One Hundred Days of Silence: America and the Rwanda Genocide. Lanham: Rowman and Littlefield Publishers, 2007. Colmon, Andrew, and Clare Wilson, “Antisocial Personality Disorder: An Evolutionary Game Theory Analysis. Legal and Criminological Psychology 2 (1997): 23. Coogan, T. Pat. The IRA. New York: Palgrave Press, 2002. Cook, Susan. Genocide in Cambodia and Rwanda: New Perspectives. New Jersey: Transaction Publishers, 2006. Coughlan, Anthony. Fooled Again? The Anglo-Irish Agreement and After. Ireland: Mercier Press, 1986. Coyle, Colin. “A Tale of Two Peace Processes,” The Sunday Times (London). 17 August 2008: 6. Cox, Michael. “Bringing in the ‘International’: The IRA Cease-fire and the end of the Cold War.” International Affairs 73.4 (1997): 671-693. 267 Cox, Michael, Adrian Guelke, and Fiona Stephen, eds. A Farewell to Arms? From “Long War” to Long Peace in Northern Ireland. Manchester: Manchester University Press, 2000. Crocker, Chester, Fen O. Hampson, and Pamela Aall, eds., Herding Cats: Multiparty Mediation in a Complex World. Washington, D.C.: United Institute of Peace Press, 1999. ---. eds. Leashing the Dogs of War: Conflict Management in a Divided World. Washington, D.C., United States Institute of Peace Press, 2007. Dallaire, Romeo. Shake Hands with the Devil: The Failure of Humanity in Rwanda. Cambridge: Da Capo Press: 2004. “Declaration on Rwandese Refugees Problem," Dar-es-Salaam 19 Feb. 1991. Dekmejian, R. Hrair. Spectrum of Terror. Washington D.C.: Congressional Quarterly Press, 2007. Dekmejian, R. Hrair, and Gordon Doron. “Changes Patterns of Equilibria in the Arab- Israeli Conflict.” Journal of Conflict Management and Peace Science 5.1 (1980): 41-54. Destexhe, Alain. Rwanda and Genocide in the Twentieth Century. London: Pluto Press, 1995. Dingley, James. Combating Terrorism in Northern Ireland. New York: Routledge Press, 2008. Dixit, Avinash. The Art of Strategy: A Game Theorists Guide to Success in Business and Life. New York: W.W. Norton & Co, 2008. Dobble, Peter. “Clinton’s Mission to Meddle in Ulster” Mail on Sunday (London) 14 Feb. 1993. Dowdney, Mark. “Fury as Adams Gets Visa; Unionists Slam Clinton; Gerry Adams Granted New Visitors Visa for the United States.” Daily Mirror 2 March 1996: 1. Dugatkin, Lee Alan, and Hudson Kern Reeve, eds. Game Theory and Animal Behavior. New York: Oxford University Press, 1998. 268 “East and Horn of Africa in Brief; Rwandan President Meets Visiting French Foreign Ministry Official.” Radio Rwanda. BBC Summary of World Broadcasts. 18 July 1991. Elliott, Marianne. The Long Road to Peace in Northern Ireland: Peace Lectures from the Institute of Irish Studies at Liverpool University. Liverpool: Liverpool University Press, 2002. English, Richard. Armed Struggle: The History of the IRA. Oxford: Oxford University Press, 2003. Eriksson, John, et al. “Synthesis Report.” in The International Response to Conflict and Genocide: Lessons from the Rwanda Experience. Joint Evaluation of Emergency Assistance to Rwanda, 1996. “Ex-U.S. Official, in Paris, Again Assails Washington's Failure to Intervene in Rwanda in '94.” The New York Times. 8 July 1998: A10. Fisher, Ronald. The Genetical Theory of Natural Selection. Oxford: Clarenden University Press, 1930. Fisher, Ronald, and Loraleigh Keashly. “The Potential Complementarity of Mediation and Consultation within a Contingency Model of Third Party Intervention.” Journal of Peace Research 28, no. 1 (1991): 29-42. Fisher, Roger, and William Ury. Getting to Yes: Negotiating Agreement Without Giving In. United States: Penguin Books, 1983. Fletcher, Martin and Arthur Leathley “US Visa for Adams Angers No. 10,” Times (London)—31 Jan. 1994: 1. Ford, Richard. “Mass ‘Loyalist’ Rally for Belfast/Opposition to Anglo Irish Agreement Demonstration Planned.” The Times (London) 23 Nov. 1985: 1. Foster, R.F, Modern Ireland: 1600-1972. New York: Penguin, 1988. “France to Support Great Lakes Region” BBC Worldwide Monitoring 30 Nov. 2009. “Frontline: The IRA and Sinn Fein.” PBS Television Broadcast. 21 Oct. 1997. Freedland, Jonathan. “Comment & Analysis: Ten Steps to Peace: Why Some Negotiations Fail While Others, However Feebly, Stay on Track Depends on Following the Rules.” The Guardian (London), 8 August, 2001: 13. 269 Freeman, Alan. “Clinton Snubs Hard-liners; On a Stop in Belfast, President Raises Ire by Refusing to Meet Protestants Opposed to the Peace Process.” The Globe and Mail (Canada) 14 Dec. 2000: A16. Furley, Oliver and R. May, Ending Africa’s Wars: Progressing to Peace. Great Britain: Antony Rowe Ltd., 2006. Gauthier, David. Morals by Agreement. Oxford: Clarenden Press, 1969. Gettleman, Jeffrey. “After 15 Years, Hints of Peace In Burundi.” The New York Times. 16 June 2008: A7. Ghemawat, Pankaj. Games Businesses Play: Cases and Models. Cambridge: MIT Press, 1997. Gintis, Herbert. The Bounds of Reason: Game Theory and the Unification of the Behavioral Sciences. New York: Princeton University Press, 2009. Goldberg, Stephen, Frank E. A. Sander, Nancy H. Rogers, and Sarah R. Cole. Dispute Resolution. 5th ed. Maryland: Aspen Law and Business, 2003. ---. Dispute Resolution: Negotiation, Mediation, and Other Processes. Maryland: Aspen Publishers, 1995. “The Good Friday Agreement,” Signed in Belfast, Northern Ireland on April 10, 1998 by British and Irish governments. Gosling, Paul. “NI Sectarian Divisions ‘Cost €1.5 bn Each Year.” Public Finance 31 August 2007: 10. Gourevitch, Philip. We Wish to Inform You That Tomorrow We Will Be Killed With Our Families: Stories From Rwanda. New York: Farrar, Straus and Giroux, 1998. ---. “Letter from Rwanda: After the Genocide.”: When a people murders up to a million fellow-countrymen, what does it mean to survive?” The New Yorker. 18 Dec. 1995: 78-95. Gowda, M. “Teaching the Prisoner’s Dilemma.” Journal of Policy Analysis and Management 15.4 (1996): 646-653. Grädel, Erich. Automata, Logics, and infinite Games: A Guide to Current Research. Berlin: Springer, 2002. 270 Graham, Williams. “Unionists Bicker as Mitchell Takes Chair.” The Irish Times. 13 June 1996. Greenberg, Melanie. “Consent in International Mediation.” Dispute Resolution Magazine 14.2 (2008): 20-22. Greenberg, Melanie C., John H. Barton and Margaret E. McGuinness, eds. Words over War: Mediation and Arbitration to Prevent Deadly Conflict. Lanham: Rowman and Littlefield, 2000. Greif, Avner. Institutions and the Path to the Modern Economy. New York: Cambridge University Press, 2006. Gribbin, Robert. In the Aftermath of Genocide: The US Role in Rwanda. Nebraska: Universe, 2005. Grunfeld, Fred. The Failure to Prevent Genocide in Rwanda: The Role of Bystanders. Boston: Martinus Nijhoff, 2007. Guelke, Adrian. “The United States, Irish American and the Northern Ireland Peace Process.” International Affairs 27.3 (1996): 521-536. Harkness, David. Ireland in the Twentieth Century. New York: St. Martin’s Press, 1996. Heazle, Michael and Iyanatul Islam, eds. Beyond the Iraq War: the Promises, Pitfalls and Perils of External Interventionism. Boston: Edward Edgar Publishing Limited, 2006. Heims, Steve J., John Von Neumann, and Norbert Wiener. From Mathematics to the Technologies of Life and Death. Cambridge, MA: MIT, 1980. Hennessey, Thomas. The Northern Ireland Peace Process: Ending the Troubles? New York: Palgrave, 2001. Hintjens, Helen. “Explaining the 1994 Genocide in Rwanda.” The Journal of Modern African Studies 37.2 (1999): 241-286. Holland, Mary. “The Debt the North Owes Clinton.” The Irish Times 9 Nov. 2000: 18. History Wiz. < http://www.historywiz.com/rwanda.htm Irish Republican Army. “Irish Republican Army August 31, 1994 Ceasefire Statement.” 271 CAIN: Conflict Archive on the Internet. 18 October 2009. http://cain.ulst.ac.uk/index.html Jacobsen, Diane DeMell,The Difference Makers: The Role of the Mediator in Peace Negotiations, Dissertation Washington University, 2003. Jandt, Fred and Paul B. Pederson. Constructive Conflict Management. London: Sage Publications, 1996. Jones, Bruce, “Roots, Resolution, and Reaction: Civil War, the Peace Process and Genocide in Rwanda.” (paper presented at the University of Toronto conference on Civil Wars in Africa) May 1995. (a) ---. “The Arusha Peace Process: Report for Study II of the Joint Evaluation of Emergency Assistance to Rwanda,” Dec. 18, 1995. (b) ---. Peacemaking in Rwanda: The Dynamics of Failure. Colorado: Lynne Rienner, 2001. Kaufman, Sandra and G. Duncan. “A Formal Framework for Mediator Mechanisms and Motivations.” The Journal of Conflict Resolution. 36.4 (1992): 688-708. Keane, Fergal. “The Blood-Stained Truth Behind Clinton’s Fine Words in Blackpool.” The Independent (London). 5 October 2002: 18. Kelley, Kevin. The Longest War: Northern Ireland and the IRA. London: Zed Books, 1983. Kelly, Michael. “The Visionaries of the Irish Agreement.” Washington Post. 15 April 1998: A19. Keltner, John. “Mediation: Toward a Civilized System of Dispute Resolution,” ERIC Clearinghouse on Reading and Communication Skills, and Speech Communication Association, 1987. “Kenya: Lessons From Power Pacts in Other Countries,” The Nation. 13 April 2008. Khadiagala, Gilbert M. “Intervention in Internal Conflict: The Case of Rwanda.” In Military Intervention: Cases in Context for the Twenty-First Century, ed. William J. Lahneman, 67-88. Lanham: Rowman and Littlefield, 2004. Khosla, Deepa. “Third Party Intervention in Ethnic Conflicts: A Force for Peace or Spiraling Violence,” University of Maryland, 2004. 272 Kiernan, Ben. Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur. New Haven: Yale University Press, 2007. Kinloch, Graham and Raj Mohan, eds. Genocide: Approaches, Case Studies, and Responses. New York: Algora Publishing, 2005. Kleiboer, Marieke. The Multiple Realities of International Mediation. United States: Lynne Rienner Publisher, 1998. ---. “Understanding Success and Failure of International Mediation.” The Journal of Conflict Resolution 40.2 (1996): 360-389. Klerans, John, and John Ryan, “IRA Poised for New Truce as Election Lifts Peace Hopes; Election on Sinn Fein MPs Bring Hopes of Cease-fire.” Sunday Mirror 4 May 1997. Kolb, Deborah. The Mediators. Massachusetts: MIT Press, 1983. Kroslak, Daniela. The Role of France in the Rwandan Genocide. London: Hurst & Co., 2007. Kuperman, Alan. “Rwanda in Retrospect.” Foreign Affairs 79.1 (2000): 94-118. ---. “The Other Lesson of Rwanda: Mediators Sometimes Do More Damage Than Good.” School of Advanced International Studies (SAIS) Review: A Journal of International Affairs 16 (Winter –Spring 1996): 221-240. Kydd, Andrew. “Whose Side Are you On? Bias, Credibility, and Mediation.” American Journal of Political Science. 47.4 Oct. 2003: 597-611. Lemarchand, Rene. Burundi: Ethnic Conflict and Genocide. Massachusetts: Cambridge University Press, 1994. ---. “Review: A History of Genocide in Rwanda.” The Journal of African History 43.2 (2002): 307-311. Levy, Marc. “Mediation of Prisoners’ Dilemma Conflicts and the Importance of the Cooperation Threshold: The Case of Namibia.” The Journal of Conflict Resolution 29.4 (1985): 581-603. Licklider, Roy. “The Consequences of Negotiated Settlements in Civil Wars, 1945- 1993.” The American Political Science Review 89.3 (1995): 681-690. 273 Linden, Ian. “The Churches and Genocide: Lessons from Rwanda,” Svensk Missionstidskrift 83, March 1995.MacDonagh, Oliver. Ireland: The Union and its Aftermath. London: George Allen and Unwin Publishers, 1977. Mamdani, Mahmood. When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda. New Jersey: Princeton University Press, 2001. Marchak, Patricia. No Easy Fix: Global Responses to International War and Crimes Against Humanity. Montreal: McGill-Queen’s University Press, 2008. Marquand, Robert. “The Point Man For Peace in Ireland,” Christian Science Monitor. 9 April 1998: 1. ---. “France Rejects Rwanda’s Genocide Accusations.” Christian Science Monitor. 7 August 2008: 25. Markey, Patrick. “The Clinton Visit”: 11. Maundi, Mohammed. Getting In: Mediators Entry into the Settlement of African Conflicts. United States: United States Institute of Peace Press, 2006. Mayer, Bernard. Beyond Neutrality: Confronting the Crisis in Conflict Resolution United States: Jossey-Bass Press, 2004. Mays, Terry. Historical Dictionary of Multinational Peacekeeping. Maryland: The Scarecrow Press, 2003. McGoldrick, Debbie. “Adams Praises Clinton for Lifting Visa Ban.” The Irish Times 1 Feb. 1994: 6. McKittrick, David. “Irish Thank Clinton For His Part in the Peace.” The Independent (London) 24 May 2001: 12. ---. The Nervous Peace. Belfast: The Blackstaff Press, 1996. Melvern, Linda. A People Betrayed: The Role of the West in the Rwandan Genocide. London: Zed Books, 2000. Menkel-Meadow, Lela Porter Love, and Andrea Kupfer Schneider. Mediation: Practice, Policy and Ethics. United States: Aspen Publishers, 2004. Milgrom, Paul. “Axelrod’s: The Evolution of Cooperation.” The RAND Journal of Economics 15.2 (1984): 305-309. 274 Mitchell, George. Making Peace. California: University of California Press, 2001. “Mitchell Will Be a Perfect Man for the Job” Belfast Telegraph. 23 Jan. 2009: 16. Moghalu, Kingsely Chiedu. Rwanda’s Genocide: The Politics of Global Justice. New York: Palgrave Macmillan, 2005. Moloney, Ed. A Secret History of the IRA. New York: W.W. Norton & Company, 2003. Montgomery, Lane. Never Again, Never Again. . .: Genocide: Armenia, The Holocaust, Cambodia, Rwanda, Bosnia and Herzegovina, Darfur. New York: Ruder Finn Press: 2007. Moore. Christopher. The Mediation Process: Practical Strategies for Resolving Conflict. San Francisco: Jossey-Bass, 1987. Moore, Christopher. The Mediation Process: Practical Strategies for Resolving Conflict, 2nd ed., San Francisco: Jossey-Bass Publishers, 1996. Moriarty, Gerry. “RUC Chief is Optimistic About IRA Cease-fire 60% Chance of it Lasting Until Easter, He Believes.” The Irish Times, 12 December 1994: 8. Mugabowineza, Octave. “Rwanda, What Went Wrong? Revisiting the 1994 Genocide.” The University of Kansas, 2005. Mthembu-Salter, “Mediation and Genocide in Rwanda and Burundi’s Peace Agreement Without Peace,” Track Two 11.5/6 (October 2002). “Nature of the Mediation.” October 2002, Vol.11 no.5&6. CCR Home. 18 November 2009. <http://ccrweb.ccr.uct.ac.za/archive/two/11_56/rwanda_chapter4.html>. “Nightmare for Rwanda.” The Scotsman. 19 July 1994. “Northern Ireland’s Impasse.” The Toronto Star, 13 April 1999: 1. Neuheiser, Jorg and Stefan Wolff. Peace at Last: The Impact of the Good Friday Agreement. New York: Berghahn Books: 2004. Ngowi, Rodrique. “Rwandans Flock to Vote on Constitution: Registration Tops 95%: Limits on Power Designed to Avoid Ethnic Tensions.” National Post. 27 May 2003: A16. 275 Norris, Alexander. “Rubbing Elbows with Evil; Montreal Man Heads Delegation Probing Report of Rwandan Genocide.” Montreal Gazette. 28 Feb. 1993: A3. O’Ballance, Edgar. Terror in Ireland: the Heritage of Hate. Novato: Presidio Press, 1981. ---. Terror in Ireland. United States: Presidio Press, 1981. “Obama Wins Nobel, Vows ‘Call to Action.’” NPR. 9 Oct. 2009. < http://www.npr.org/templates/story/story.php?storyId=113649365> O’Brien, Brendan. The Long War: The IRA and Sinn Fein, 1985 to Today, Ireland: Syracuse University Press: 1993. O’Brien, Marie, and Conor C. O’Brien. A Concise History of Ireland. 3 rd ed. New York: Thames & Hudson, 1981. Odom, Thomas. Journey into Darkness: Genocide in Rwanda. College Station: Texas A&M University Press, 2005. O’Grady, Joseph P. “Forcing the Question of Northern Ireland: The Brooke-Mayhew Talks, 1990-1992.” New Hibernia Review 5.4 (2001): 73-92. O’Kane, Eamon. “Re-Evaluating the Anglo-Irish Agreement: Central or Incidental to the Northern Ireland Peace Process.” International Politics 44.6 (2207): 711-732. O’Reagan, Michael. “Adams Urges Government to Take NI Initiative.” The Irish Times, 22 February 1993: 8. “The Other Lesson of Rwanda: Mediators Sometimes Do More Damage Than Good,” SAIS review 16.1 (1996) 221-240. “Peace: It’s A People Thing.” Belfast News Letter (Northern Ireland) 13 December 2000: 1. Peroff, Kathleen, and Christopher Hewitt. “Rioting in Northern Ireland: The Effects of Different Policies.” The Journal of Conflict Resolution 24.4 (1980): 593-612. Pillar, PR. Negotiating Peace: War Termination as a Bargaining Process. Princeton, NJ: Princeton University Press, 1983. “Politicians Split Over Clinton Trip Timing.” Belfast News Letter 15 April 1998: 8. 276 “President Clinton Discusses Northern Ireland Peace Process.” CNN International Asia Tonight. Karuna Shinsho. CNN International. 12 Dec. 2000. Pruitt, Dean, and Kenneth Kressel. “The Mediation of Social Conflict: An Introduction.” Journal of Social Issues 41.2 (1985): 1-10. Prunier, Gerard. The Rwanda Crisis: History of a Genocide. New York: Columbia University Press, 1995. Rahim, M. Afzalur and Albert Blum, eds. Global Perspectives on Organizational Conflict. London: Praeger Press, 1994. Raiffa, Howard. The Art and Science of Negotiation. Massachusetts: Belknap Press, 1985. Ranelagh, John O. A Short History of Ireland. Cambridge: Cambridge University Press, 1995. Rapoport, Anatol. “Prisoner’s Dilemma: A Study in Conflict and Cooperation.” The Philosophical Quarterly 20.80 (1970): 292-293. Report of the International Body on Arms Decommissioning, Jan. 22, 1996. <http://www.nio.gov.uk/iicd_report_22jan96.pdf> Reynolds, Albert and John Major, “Joint Declaration of Irish Taoiseach,” Dec. 15, 1993. Reyntjens, Filip. The War of the Great Lakes. Massachusetts: Cambridge University Press, 2002. Richmond, Oliver. “Devious Objectives and the Disputants’ View of International Mediation: A Theoretical Framework.” Journal of Peace Research 35:6 (1998): 707-722. Rinaldo, Rachel. “Can the Gacaca Courts Deliver Justice?” Global Policy Forum. 8 April 2004. < http://www.globalpolicy.org/component/content/article/163/29062.html> Ross, Lee D. and Andrew Ward. Naive Realism in everyday life: Implications for social conflict and misunderstanding. In Brown, T., Reed, E., and Turiel, E. (Eds.), Values and Knowledge . Hillsdale, NJ: Lawrence Erlbaum, 1996. “RPF Statement Names New President, Sets Out New Political Arrangements.” Radio Rwanda, BBC Summary of World Broadcasts. 18 July 1994. 277 Ruane, Joseph, and Jennifer Todd, The Dynamics of Conflict in Northern Ireland: Power Conflict and Emancipation. Cambridge: Cambridge University Press, 1996. Rubinstein, Ariel. “Comments on the Interpretation of Game Theory.” Econometrics 59.4 (1991): 909-24. “Rwanda Country Report Accuses France of Role in Killings” Africa News. 11 Aug. 2008. “Rwanda Details of Cease-Fire Agreement Signed with Rebels” Radio Rwanda, BBC Summary of World Broadcasts. 1 April 1991. “Rwanda Herman Cohen on Talks with the Rebels; National Conferences in Africa,” Radio Rwanda, BBC Summary of World Broadcasts. 11 May 1992. “Rwanda Justice Files 1999-2001.” Foundation Hirondelle.10 Dec. 2001. <http://www.hirondelle.org/hirondelle.nsf/0/9117de1eb28f7a4241256beb006b4bc 7?OpenDocument> “Rwanda Observer Force to Start Work; New RPF Attack Claimed” Radio Rwanda. BBC Summary of World Broadcasts. “Rwanda Observer Group to be Deployed in Combat Areas; Threat of Rebel Attack." Radio Rwanda, BBC Summary of World Broadcasts. 20 May 1991. “Rwanda Peace Agreement Postponed ‘Indefinitely.’” AFP report; BBC Summary of World Broadcasts. 26 June 1993. “Rwanda: Plans Under Way to Draft Historic New Constitution.” Radio Rwanda, BBC Worldwide Monitoring. 21 April 2001. “Rwandan President’s Address to the Nation on Refugee and Rebel Problems.” Radio Rwanda, Kigali, BBC Summary of World Broadcasts. 16 Feb. 1991 “Rwanda Prosecutor General Studies Mucyo Report.” The New Times. 17 Aug. 2008. “Rwanda: Ugandan Forces Accused of Deployment in Northern Rwanda in Support of RPF.” Kenya News Agency. 25 Feb. 1993. Sandole, Dennis, and Hugo van der Merwe, eds. Conflict Resolution Theory and Practice: Integration and Application. New York: Manchester University Press, 1993. 278 ---. Conflict Resolution Theory and Practice: Integration and Practice. New York: Manchester University Press, 1993. Savun, Burcu. “Information, Bias, and Mediation Success: Evaluating the effectiveness of International Conflicts,” Rice University, 2006. "Selective Invite to Respectability." The Washington Times 24 August 1997: B5 Sellstrom, Tor and Lennart Wohlgemuth, “Study 1: Historical Perspectives: Some Explanatory Factors,” in International Response to Conflict and Genocide: Lessons from the Rwanda Experience. Joint Evaluation of Emergency Assistance to Rwanda, 1996. Shapiro, Martin, Courts: A Comparative and Political Analysis. Chicago: University of Chicago Press, 1986. Sharlet, Jeffrey. “Fierce Debate Divides Scholars of the 1994 Rwandan Genocide” The Chronicle of Higher Education. 47 3 August 2001: A16-A19. Shropshire, Rob. “Clinton’s Pretend Apology.” The Globe and Mail (Canada) 26 June 2001: A12. Simon, Alexander H. The Sciences of the Artificial. Massachusetts: MIT Press, 1996. “Sinn Fein Criticises Downing Street Declaration.” Deutsche Presse-Agentur 24 July 1994: 1. Sitkowski, Andrzej. UN Peacekeeping: Myth and Reality. Connecticut: Praeger Security International, 2006. Smith, Maynard John. Evolution and the Theory of Games. Cambridge University Press, 1982. Sorbo, M. Gunnar, and Peter Vale, Out of Conflict: From War to Peace in Africa. Sweden: Motala Grafiska, 1997. Stettenheim, Joel. “The Arusha Accords and the Failure of International Intervention in Rwanda.” <www.wilsoncenter.org/subsites/ccpdc/pubs/words/8.pdf> Stevenson, Jonathan. “Peace in Northern Ireland: Why Now?” Foreign Policy 112 (1998): 41-54. 279 Straus, Scott. The Order of Genocide: Race, Power, and War in Rwanda. Ithaca: Cornell University Press, 2006. Totten, Samuel, Paul Bartrop and Steven Jacobs. Dictionary of Genocide: Volume 2. Santa Barbara: Greenwood Publishers, 2007. Townshend, Charles. Consensus in Ireland: Approaches and Recessions. Oxford: Clarendon Press, 1988. Tran, Mark P. “Democrats Pledge Belfast Peace Envoy,” The Guardian 7 April 1992: 6 Tupman, W.A. “Where Has All the Money Gone? The IRA as a Profit-Making Concern.” Journal of Money Laundering Control 1.4 (1998): 303-311. Tynan, Maol Muire, “SF Claims Policy Shift ‘Ambushed’ Dublin.” The Irish Times:12 Ury, William, Jeanne M. Brett, and Stephen B. Goldberg. Getting Disputes Resolved: Designing Systems to Cut the Cost of Conflict. San Francisco: Jossey-Bass, 1988. USC Spectrum Lecture Series, 14 October 1998. <www.usc.edu/student-affairs/spectrum/events.html?event=101498b> Wall, James. “Mediation: An Analysis-Review, and Proposed Research.” Journal of Conflict Resolution 25.1 (1985): 147-80. Wallenstein, P. Understanding Conflict Resolution. London: Sage Publications, 2007. Wallice, Andrew. Silent Accomplice: The Untold Story of Frances Role in Rwandan Genocide. London: I.B. Tauris, 2007. Watt, Nicholas, “Election Plan Cools Early Welcome for Peace Blueprint.” Times (London) 25 Jan. 1996: Watt, Nicholas, Nicholas Wood, Andrew Pierce, and Adam Fresco. “Bomb Shatters IRA Cease-fire.” Times (London). 10 Feb. 1996. Webb, Keith. "Third Party Intervention and the Ending of Wars." Paradigms 9.2 (1995): 1995. Weiss, Ruth. Peace in Their Time: War and Peace in Ireland and South Africa. London: I.B. Taurus, 2001. Wikipedia. 1 November 2009. < http://en.wikipedia.org/wiki/Northern Ireland> 280 Zartman, William. Ripe for Resolution: Conflict and Intervention in Africa. New York: Oxford University Press, 1991. ---.. "The Internationalization of Communal Strife: Temptation and Opportunities Triangulation." in Manus Midlarsky's The Internationalization of Communal Strife, New York: Routledge:27-42, 1992. ---. Negotiating an End to Civil Wars. Washington DC: Brookings Institute, 1995. ---. Negotiation and Conflict Management: Essays on Theory and Practice. UK: Routledge Press, 2008. Zartman, William, and Jeffrey Rubin. Power and Negotiation. United States: University of Michigan Press, 2000. Zartman, William, and Saadia Touval. “International Mediation: Conflict Resolution and Power Politics.” Journal of Social Issues 41.2 (1985): 27-45.

Abstract (if available)

Abstract The nature of global conflict has transformed dramatically in the past few decades thanks to changes in international relations. Many attempts at successful conflict mediation have failed due to the dramatic scope of the situation

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Crimes of honor: an international human rights perspective on violence against women in South Asia

PDF

The elephant and the mouse that roared: the prospects of international policy and local authority in the case of the convention on international species (CITES)

PDF

Augusto Cesar Sandino's political thought and its impact on the Sandinista National Liberation Front

PDF

An alternative model of transition in the Middle East: bounded consociotionalism

PDF

Assessing the effectiveness of transnational activism: an analysis of the anti-whaling and anti-sealing campaigns

PDF

Regulatory burden: a systems analysis of weapons acquisition in the U.S. Department of Defense

PDF

Policing accountability: an empirical investigation of state-sponsored police reform in Riverside, California

PDF

Photography in movement: the right to photograph, ethics in imaging, and the art of making social change

PDF

Compliance and compromise: the jurisprudence of gender pay equity

PDF

Greening law: a socio-legal analysis of environmental human rights in India

PDF

Greater than the sum of her parts: A multi -axis analysis of Black women and political representation

PDF

The impact of 9/11 on the judicial treatment of Middle Eastern asylum applicants in the U.S. courts

PDF

The little Skingin that could: an autobiographical, affirmative look at Native American off-reservation boarding schools between 1970 and 1980

PDF

Petey and Chato: the pitbull's transition from mainstream to marginalized masculinity

PDF

Attacking atheists: doing one's duty to God and country in 21st century America

PDF

Argentina's economic crisis in 2001 and the International Monetary Fund

PDF

Cautionary tales, United States immigration and the Fauziya Kassindja case: Toward a theory of cultural romanticism

PDF

Environmental ABC's -- acting on behalf of children: rethinking representation, risk and responsibility in US environmental policy implementation

PDF

An argument for the criminal hoax

PDF

Transformative occupations: life experiences of performers with disabilities in film and television

Asset Metadata

Creator Gavrieli, Yamit (author)

Core Title Conflict mediation: a comparative study of the effects that mediators have on dyadic conflicts

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Politics

Publication Date 02/16/2010

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag conflict mediation,dyadic conflicts,mediator,OAI-PMH Harvest

Place Name Great Britain (countries), Northern Ireland (countries), Rwanda (countries)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Dekmejian, Richard (committee chair), Burke, Catherine G. (committee member), Renteln, Alison Dundes (committee member)

Creator Email yamit.gavrieli@gmail.com,ygavriel@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2852

Unique identifier UC1174024

Identifier etd-GAVRIELI-2919 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-300994 (legacy record id),usctheses-m2852 (legacy record id)

Legacy Identifier etd-GAVRIELI-2919.pdf

Dmrecord 300994

Document Type Dissertation

Rights Gavrieli, Yamit

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu