Civil War in African States: The Search for Security 9781935049982

Citation preview

CIVIL WAR IN AFRICAN STATES

CIVIL WAR IN AFRICAN STATES The Search for Security

Ian S. Spears

Published in the United States of America in 2010 by FirstForumPress A division of Lynne Rienner Publishers, Inc. 1800 30th Street, Boulder, Colorado 80301 www.firstforumpress.com and in the United Kingdom by FirstForumPress A division of Lynne Rienner Publishers, Inc. 3 Henrietta Street, Covent Garden, London WC2E 8LU © 2010 by FirstForumPress. All rights reserved Library of Congress Cataloging-in-Publication Data A Cataloging-in-Publication record for this book is available from the Library of Congress. ISBN: 978-1-935049-20-3 British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. This book was produced from digital files prepared by the author using the FirstForumComposer. Printed and bound in the United States of America The paper used in this publication meets the requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48-1992. 5 4 3 2 1

For Sarah, Jordan, Jack, and Nicholas

Contents

Acknowledgments

ix

1

The Search for Security in Africa

1

2

The Wars of Ethiopia and Eritrea

41

3

Unification and Quasi-Secession in Somalia

115

4

The Angolan Civil War

179

5

The Search Continues

241

Bibliography Index

253 273

Acknowledgments

As others before me have stated, a project such as this involves the contributions of many individuals and organizations. I am very fortunate to have had so many people willing to help in both big and small ways. It is with pleasure that I am finally able to recognize and thank them for their kind assistance, guidance, and support. None of them are, of course, responsible for the arguments or the inevitable flaws that remain. In Angola, I would like to thank Allan and Julia Cain and the Development Workshop (Luanda) for their hospitality and assistance. I would also like to thank Dr. Chris Jennings, the late Dr. George Wahl, Daniel dos Santos, and Eunice Inácio. On and in Eritrea and Ethiopia, I would like to thank Taisier Ali, Brother Ghebres, Sister Thomas, Bruce McKinnon, Ambassador Mike Murray, Lawrence Liu, Josh Lincoln, Jim Borton, Joseph Stern, the Institute for Security Studies (ISS) in Addis Ababa, Dr. Jakkie Cilliers, Kenneth Mpyisi, Yemi Tadesse, Dr. Solomon Inquai, Muhamed Abdulsalem, and Daniel Dissassa. Particular thanks to my friends Berouk Mesfin at ISS, Mr. Teshome Gabre-Mariam Bokan, and Mr. George Jacoby. Thanks as well to John Young, Kjetil Tronvoll, and Alem Abbay for taking the time to read, provide materials, and answer questions on Ethiopia and Eritrea. In Somalia, I would like to thank Dominik Helling and the amazingly insightful Mohammed Hassan Ibrahim. The Academy for Peace and Development in Hargeisa Somaliland provided a very friendly environment for me to work. Thanks also to Ulf Terlinden, Yassin Kahen “Yankie Kilo” Booh, Dr. Mohamed Fadal, Ambassador John Hirsch, Dr. Ali Sheikh Ibrahim, Edna Adan, and Matt Bryden. In South Africa, I would like to thank Deane-Peter Baker, the Baker family, and the School of Philosophy and Ethics at the University of KwaZulu-Natal in Pietermaritzburg for providing such a hospitable environment for the completion of this project. Others who provided careful readings, criticism, and friendship included Elisabeth King, Jay Oelbaum, and Ed Dosman. I would like to say a very special thank-you to my good friends Paul Kingston and

x

Acknowledgments

Robert O. Matthews. Paul travelled with me, read the manuscript, and, in several long and challenging discussions in Addis Ababa and eastern Ethiopia, helped me shape the framework. In reading every word, Bob Matthews, a much-valued friend and mentor, provided essential guidance and friendly criticism. Thanks also to Michael Murphy, Richard Iton, Sid Noel, Jim Kirkwood, Howard Lentner, Bruce Unger, the late Donald Rothchild, Robert Astroff, Rex Brynen, Mark Brawley, and Richard Sandbrook. For a welcome diversion, I would also like to thank my extraordinary crews and friends at Branksome Hall. For many years of comradeship and friendship, I would like to acknowledge David Johnson, Kevin Wright, Julia and Tony Tremain, John Heder, Bob Blunt, and Scott Fleming. I would like to thank my many friends and colleagues at the University of Guelph. These include Ken Woodside, Julie Simmons, Byron Sheldrick, David MacDonald, Dennis Baker, Candace Johnson, Theresa Lee, Adam Sneyd, Janice Hicks, Judith McKenzie, Janine Clark, Melissa Gabler, Jordi Díez, Brian Woodrow, O.P. Divedi, Fred Eidlin, Cathie Hosker, Geraldine McCauley, Shelagh Daly, Debbie Bowie, Carol Dauda, and Renee Tavascia. Special thanks to Craig Johnson, Troy Riddell, and Tim Mau for their friendship since my arrival at the University of Guelph. Special thanks also go to John Sutcliffe, Martha Lee, and Tom Najem at the University of Windsor. Several extraordinary students at both of these universities were instrumental in challenging me and helping me shape and express my thoughts. In particular, I would like to thank Meghan O’Keefe, Fraser Pennie, Max Kaploun, Lilly Whitham, Caitlin Vito, Erin Schneider, Philip Martin, Allison Neufeld, Tressa Allan, and Nick Sowsun. Others who provided assistance included Kym West, Joanne Lebert, and Tess Cecil-Cockwell. Special thanks as well to my good friends Michael Szonyi, Francine McKenzie, Luin Goldring, Peter Vandergeest, Sarah Powell, and Stephen Smith for their many years of friendship and sage advice on this project and more serious matters. For their long hours of editorial work, I would like to offer a very special thank-you to Dorothy “Dot” Graham and Patricia Kennedy. Thanks to the University of Toronto library, the University of Guelph library, the Toronto Public Library, and the Canadian Department of Foreign Affairs and International Trade. I would also like to thank Lynne Rienner and Claire Vlcek for their patience, guidance, and assistance. As part of this process, my gratitude also goes to several referees of the manuscript for their challenging criticisms and thoughtful advice.

Acknowledgments

xi

My family has been a source of endless support: Martha Spears, the Leger family, the Spears-Moshiri clan, the Atkinson family, and the Peterson family. In particular, I would like to thank my parents, Ellen and John Spears, and Betty Anne and Hugh Anson-Cartwright, for their love and guidance. Thanks as well to Adam and Jill Hermant and Judith and Martin Farnsworth. Thanks as well are due to Maria Santos. In the end, my deepest gratitude goes to my wife, Sarah, and my three beautiful children, Jordan, Jack, and Nicholas. They have been a constant source of unconditional love, support, and patience. And surely, by now, they must know that “to the furthest stars and back a million times” is a drop in the ocean in terms of my love for them.

1 The Search for Security in Africa

When the risks of competition exceed the risks of cooperation, [disputants] should direct their self-help efforts towards achieving cooperation. Charles L. Glaser The hunger for a final crushing victory overshadows any spirit of sectarian compromise. David Brooks The desire for separation always springs from the recognition that a certain socio-economic and cultural community is badly governed by the state to which it belongs. Joseph Tubiana How do disputants in civil wars—rebel movements, ethnic groups, state leaders—find security in Africa’s anarchic situations? Why do some rebel movements pursue a secessionist agenda while others seek to overthrow the existing government? Under what circumstances will insurgents agree to share power? Why do some insurgent movements change their strategies midcourse? The answers to these questions can provide insight into which approaches can best address the continent’s most violent conflicts and create sustainable peace. This volume evolved as a consequence of several articles I wrote that questioned power-sharing as a viable form of conflict resolution in African states. Two issues emerged from those articles. The first is the question of alternatives to power-sharing: if power-sharing cannot be achieved, what other options exist? Second, since there are occasions when disputants do opt for power-sharing agreements—even if those agreements are less common or durable—what explains this willingness to share power? For that matter, what explains why disputants choose to accept or reject any given approach to peace and security? By learning

2

Civil War in African States

how the disputants themselves see conflicts—identifying the alternative strategies that they (as opposed to foreign peacemakers) consider in pursuit of their security and explaining the circumstances in which they will opt for these strategies—I seek to offer an important perspective that has, to my mind, received insufficient attention thus far. Beyond this objective, I hope that the discussion in this book accomplishes three tasks. First, my intent is to challenge prevailing assumptions about the possibilities for conflict resolution in African states. Here I draw heavily from the international relations paradigm known as “realism.”1 From my perspective, it is not useful to think about what could or should be. Political behaviors must be seen as givens, as lamentable as they may be. People tend to conduct their affairs on the basis of interests—and virtually all political behavior in conflict situations is directed towards ensuring the primary interests of security and survival. Consequently, conflict resolution requires no expectation that people’s behavior can be changed in meaningful ways. Rather, it involves recognizing existing forces for what they are and managing them by channeling them in constructive directions. I am, admittedly, challenging the view that conflicts can be easily remedied. My preference in this work, however, is to help readers become aware of the ways in which effective political action in civil wars will always be encumbered or advanced by conflicting political interests. To be sure, this approach is pessimistic. Scholars who adopt a perspective of realpolitik do not like the world that they describe. Nevertheless, if this approach is too bleak for the taste of some people, or if it fails to consider adequately the possibilities for peace and reconciliation, it does provide a framework for understanding the persistence of many African conflicts and the apparent intransigence of Africa’s disputants. More importantly, however, this approach is not so pessimistic if one sees political action as being driven less by a political actor’s inherently and unchangingly evil nature and more by that actor’s concern for his or her own security. A second task, then, is to promote a better understanding of violent conflict by challenging the view prevalent in foreign-policy circles that conflicts are essentially contests between good and evil. In a 1995 Foreign Affairs article reflecting on the crisis in Yugoslavia, Charles Boyd emphasized the need to understand what interests and what insecurities drive conflicts.2 Boyd’s argument was that all groups have legitimate interests and fears—one being the fear of becoming a minority in another state. It is the act of demonizing disputants, he claimed, that creates demons.

The Search for Security in Africa

3

Given the brutalities of recent civil wars in Congo, Rwanda, Sudan, and Sierra Leone, this perspective may be difficult for some readers to accept. In the discussion that follows, I do not mean to overlook moral failure. Indeed, it is hard not to see evil in such behavior. From my perspective, however, it is not useful to see conflict in only Manichean terms. The tendency towards violence has more to do with the nature of the African state and the insecurities it creates than the good or evil nature of a given actor or the pathological predispositions of a given society or culture. Moreover, as I endeavor to show, in these circumstances, even saints feel compelled to do evil things if such action enhances their chances for survival. Alternatively, villains can behave in ways similar to saints and embrace peace if aggression does not advance their interests but peace does. Again, the primary concern of the principal actors is their security and well-being. That is why the focus of any approach to conflict resolution must begin with the recognition of security as the driving motivation. Indeed, a third task of this study is to demonstrate that only when these concerns about security are met is it realistic to think in terms of meaningful conflict resolution. This requires outsiders to develop an appreciation of the perspectives of the actors on the ground and acknowledge that what appears to be a rational solution from a collective perspective does not necessarily meet, and often conflicts with, the individual security needs of the disputants themselves. In short, in order to understand how intervention can be most effective, it is necessary, in Barry Posen’s words, “to think about the strategy of the other side.”3 To this end and to the extent that it was possible, I have tried to account for and incorporate the interests and perspectives of the disputants as they defined them. The research presented here is based on statements from the main players as represented in interviews, in published media sources, in documents, and in narratives provided by journalists who are sympathetic to a particular disputant’s cause. I am aware, of course, of the manner in which tactics can be a factor in any given statement of strategic objectives; that is, the actors may have an interest in skewing the truth. To the extent that it was possible, however, like the historian Barbara Tuchman, I sought to avoid making my own judgments on the reasons for people’s actions.4 For every assertion I have made, I have endeavored to provide documentary support. Since some of the events in question took place before I was involved in this study, I have looked to sources produced at the time for this supporting evidence. There is, as I discovered, no single narrative for any of these conflicts, and the narratives themselves can be overwhelming in their

4

Civil War in African States

complexity. Interpretations are frequently aligned with ethnic or clan interests and, accordingly, renditions of history are often politically charged. They are also subject to self-censorship or otherwise engineered for political purposes. Given the sanctions against violence, political groups emphasize the cooperative and inclusive aspects of their struggles in their accounts of events so that they will be looked upon more favorably by the international community. Finding documents or other supporting evidence that account for all aspects of a group’s search for security—from ugly episodes of violence to more agreeable instances of nonviolence—was not an easy task. In Ethiopia, for example, the democratic character of the new EPRDF regime was (and remains) particularly controversial. There, opposition parties pulled out of early elections, claiming fraud and intimidation on the part of a governing party that was bound to have won in any case. In Somalia, events were complicated by the fact that leaders sometimes fought on behalf of several groups or militias or because several opposing leaders fought under the same banner (often against each other).5 Similar processes were at work in Angola—much to the frustration of students of the conflict there. Even at the time of Angola’s independence, John Marcum observed that “the foreign intervention and factional fighting that ensued in 1975 proved so chaotic and opportunistic that its exact sequence may remain forever arguable.”6 Later, Jonas Savimbi’s biographer, Fred Bridgland, wrote: “I have striven to ensure that [my] book is factually accurate. But the trouble with Angola is that every fact is in dispute . . . the facts are so contentious.”7 Needless to say, the narratives provided here are open-ended; they are not the last word on such complex conflicts. While I have made every effort to provide reasonably comprehensive narratives of these conflicts, I do not seek to introduce extensive new facts about any particular case (more detailed descriptions of these conflicts are cited in the endnotes). Instead, my purpose here is to introduce a form of analysis that links theory and description in more useful ways than are allowed for by facts alone, and to provide insights about realistic opportunities to prevent, limit, and end violence. In the remainder of this chapter, I introduce key elements of the African security predicament, and provide a discussion of the strategies of integration/power-sharing, domination/conquest, and separation/ secession and the factors which give rise to them. Chapters 2, 3, and 4 examine the protracted wars in Ethiopia, Somalia, and Angola—situations that demonstrate various combinations of these strategies. In the concluding chapter I consider the prospects for lasting conflict resolution in African states.

The Search for Security in Africa

5

My selection of African case studies, and, indeed, my selection of these case studies in particular, speaks in part to my interest in and familiarity with African politics and my longer-term familiarity with these countries. I believe, as William Zartman has observed, that our purpose should not merely be “to learn about Africa—an exercise of current interest to a small audience—but to learn from Africa—a project of much wider importance.”8 Indeed, during the research and writing of this book, it became increasingly clear to me that the challenges facing these countries, and the conclusions I reached regarding conflict, are also relevant in other conflict zones, both within and without Africa—including Iraq, Sri Lanka, the former Yugoslavia, the Democratic Republic of Congo, Mozambique and South Africa. I will leave it to others who have more expertise in these areas to determine if any insights provided here have application elsewhere. Africa and the Politics of Survival

In his observations of the continent’s political elite, the former American Secretary of State Henry Kissinger observed that African leaders have “survived and prevailed by learning to be finely attuned to the nuances of the power relationships on at least three levels: vis-à-vis the erstwhile colonial power, the American-Soviet competition, and the struggles for pre-eminence within their own movements. They had to be, and were, realists.” Kissinger also suggested that the demands facing Africa's political leaders were more intense than in other continents. African leaders, he concluded, “had no illusions about the grammar of staying in power; politics, in their view, was not a profession for weaklings.”9 More recent analysis suggests that, while much has changed in global and African political life, Kissinger’s assessment of the domestic power struggle endures. Such a conclusion helps explain why African states have so often been arenas for major armed conflict.10 For some, the so-called “third wave” of post-Cold War democratic reforms have “produced few tangible changes in the rules of the political game.” Frustrated with the weakness of African political institutions, with their own continuing inability to unseat incumbent governments, and with the ongoing unwillingness on the part of the international community to risk destabilizing fragile polities by criticizing electoral processes, opposition groups continue to contemplate violence as their most viable option.11 In Africa, in spite of the fact that anti-colonial movements were often united in their political objectives, the states that independence created were rarely coherent expressions of these same movements. Nor

6

Civil War in African States

did African states emanate as expressions of single existing ethnic groups. Instead, owing to the arbitrary nature of colonial partition, most were much more complex, multicultural, multilingual, and often religiously diverse entities. Efforts were made to construct political institutions which could manage this diversity, but these often broke down. To correct this failure, political authority was established and maintained through clientelist ties, the domination of a single ethnic group, or both. As states rarely have a monopoly on force, inevitably, opposition groups emerged to challenge their authority.12 In some cases (particularly since the end of the Cold War), political institutions and authority have been so compromised that political life has degenerated into chaos and violence. The challenges presented by this kind of semi-anarchic or anarchic situation are relevant anywhere they exist, but particularly in Africa given the frequency of state collapse since 1990. “The phenomenon of [state collapse] is historic and worldwide,” according to William Zartman, “but nowhere are there more examples than in contemporary Africa.”13 Not surprisingly, the occurrence of state collapse is intimately connected to civil war. In the post-Cold War era, the Stockholm International Peace Research Institute (SIPRI) repeatedly observed that “Africa is the most conflict ridden region of the world.” A “root cause” of these wars, SIPRI argued, could “be found in the weakness of many of its states.”14 Similarly, observers with the World Bank have asked “Why are there so many civil wars in Africa?” They conclude that “the relatively high incidence of civil war in Africa is due ... to the high levels of poverty, heavy dependence on resource-based primary exports and, especially, to failed political institutions.”15 In other cases, political authority has remained essentially intact but is violently contested by groups who are as powerful, or nearly as powerful, as those who formally control state power. We shall see this in my discussion of Angola where independence arrived with no fewer than three viable anticolonial movements, each of which had a realistic chance of acquiring power in the capital Luanda. While the ruling MPLA has since been recognized as the country’s legitimate government, UNITA rebels maintained control of large portions of southern and central Angola until its military defeat in 2002.16 The problem of weak or contested states is compounded by the fact that even the continent’s most tragic events are not seen as warranting the kind of global attention that is necessary to offset the lack of authority in its vast territories. Many African states became, in Margaret Anstee’s term, “orphans of the Cold War.”17 In cases where the international community has been willing to commit substantial

The Search for Security in Africa

7

resources and troops in an effort to limit these wars, interventions have been geographically limited in scope or have set a higher priority on impartiality than on the restoration of order.18 Moreover, they tended to come after the processes of state collapse or contestation were well under way, or to have been hampered by limited, uncertain, or ambiguous mandates regarding the use of force. Belligerents are also well aware of the international community’s unwillingness to tolerate casualties among foreign peacekeepers in regions of marginal importance and know that a few gruesome acts against peacekeepers will lead the international community to withdraw or stay away.19 The manner in which ethnic groups and rebel movements cope in these uncertain conditions is the central focus of this book. Most immediately, insecurity leads individuals to find strength in numbers and to retreat into a clan or ethnic group which then becomes the bases on which the conflict may be fought. To be sure, most insurgencies and governments comprise individuals from a variety of different backgrounds and perspectives. Furthermore, a member of an elite inner circle may not care about ethnicity or clan until he or she is expelled from or targeted by the regime. Identity groups can then become a means to advance or defend an individual’s cause. An impending conflict may also force individuals of mixed heritage to make choices regarding their identity and, more specifically, which identity will best ensure their survival.20 In this sense, Jack Snyder and Robert Jervis argue, ethnic diversity may not so much cause conflict as conflict causes or leads to a more acute awareness of ethnic identity.21 When individuals do not make their ethnic identity explicit, their adversaries may act on the assumption that they have. During conflicts, people of an ethnic group are often “essentialized” or “corporatized” by their adversaries. Of the principal actors in the conflict in Sri Lanka, for example, Suthaharan Nadarajah and Dhananjayan Sriskandarajah write that “while Tamils and Sinhalese were politically complex communities, they came to be referred to as monolithic wholes.” Individuals of a given ethnic group may be associated with the violent activities of the insurgency irrespective of whether or not they initially participated in or even supported those activities.22 The Marxist government in Ethiopia, and more recently the Islamic government in Sudan, did not distinguish ordinary citizens from rebels who ostensibly fought on their behalf, but rather—with terribly violent results—assumed the former supported the latter. The indiscriminate nature of “draining the pond”—killing or forcibly removing civilians who are assumed to provide support for rebels—has the effect of treating both civilians and rebels as one and thereby turning innocent bystanders into rebel supporters.23 The

8

Civil War in African States

assumption is, of course, self-fulfilling, since such atrocities convince civilians that the government is the enemy and that only the rebels are willing to fight on their behalf. Scholars and journalists have long questioned whether the behavior that has been seen as characteristic of civil wars in Africa and elsewhere is in any way rational. John Garnett writes, for example, that “It may be going too far to describe run-of-the-mill interstate wars as rational and civilized, but there is a grain of sense in the thought. Ethnic wars are quite different. They are not about the pursuit of interests as normally understood. They are about malevolence and they are unrestrained by any legal or moral rules.”24 Stephen Lewis, the former UN envoy on AIDS in Africa, has also referred to rebel movements such as the Lord’s Resistance Army in Uganda as a “lunatic rebel group,” its leader, Joseph Kony, a “madman,” and the Sudan’s leader, Omar Hassan al-Bashir, as “evil incarnate.”25 Certainly much of the journalistic analysis of African leaders questions the rationality of political decisions or actions when, as in the most notorious examples of Sierra Leone and Liberia, rebels were intoxicated or engaged in cruel or seemingly nihilistic behaviors such as chopping off their victims’ hands or dressing in wigs and women’s clothing. Others, however, see logic to the behavior of rulers and insurgents alike. As Danny Hoffman has demonstrated, rebels will undertake the most heinous crimes if they believe that it will result in a pay-off from the international community which helps secure their future.26 As for Africa’s leaders, they too must pay careful attention to cues in their environment and the actions they undertake; those who are merely reckless do not survive. “My experience with Mobutu,” Henry Kissinger writes of the long-reigning president of Zaire, “had been that, however grotesque his public conduct, he was a sharp analyst of the requirements of his own survival.”27 This is why attention to the individual strategies of actors within a state is critical. In his essay on Third World security, Brian L. Job urges scholars and practitioners to consider the perspectives of those on the ground in a way that allows them to understand their actions:28 States (more appropriately, regimes) are preoccupied with the short term; their security and their physical survival are dependent on the strategies they pursue for the moment. Consequently, it is rational for regimes to adopt policies that utilize scarce resources for military equipment and manpower, to perceive as threatening opposition movements demanding greater public debate, and to regard as dangerous communal movements and promote alternative identifications and loyalties. . . . Gaining enhanced security for

The Search for Security in Africa

9

themselves, albeit at the expense . . . of engaging in repression of their own peoples, is an acceptable bargain for many Third World state elites.29

Job concludes that an appreciation of the motivations of disputants does not require condoning these practices. “It is likely, however, to reveal that officeholders in Third World states are not irrational, insofar as their short-term, even long-term, interests are structured by the environment in which they find themselves.”30 For our purposes, and as discussions of the strategies below will reveal, the value of rationality and various theoretical devices which depend on rational decision-making can also be found in the way that they highlight differences in individual and collective gains. The outside observer pursues a misplaced logic that sees a collective gain to a peace settlement and an end to war. Outsiders assume that belligerents will accept and follow the same collective logic that they do. In the prisoner’s dilemma, for example, players would be collectively better off if they cooperated with each other and stonewalled their interrogator.31 But this is not the individual logic of the disputants themselves. From their perspective, the possibilities for cooperation are present but heavily circumscribed because the incentive to satisfy their personal and immediate need for security outweighs this collective gain. Thus, what is collectively rational to foreign mediators and interventionists is different from the perspective of each disputant’s selfinterest. Ethnic groups, insurgencies, and ruling elites think in terms of survival plans or strategies, choosing those which, in the view of the leadership or its people, are the best means to assure survival in hostile or insecure environments.32 This book considers three general types of strategies. These include (1) integration strategies or approaches that involve cooperation, accommodation, or the sharing of power among disputants within a single state; (2) domination strategies or approaches in which one group assumes a dominant or hegemonic position relative to others or which involve the conquest, neutralization or elimination of adversaries; and, finally, (3) separation strategies or approaches that involve secession and the erection of formal state barriers between disputants. These can be discussed each in turn.

10

Civil War in African States

Survival Strategies in African States Strategies of Integration, Cooperation, and Power-Sharing

Post-conflict environments can allow varying arrangements and degrees of cooperation and integration. These can range from highly integrative approaches such as power-sharing, which require significant amounts of cooperation, to competitive multiparty elections and federalism, which require considerably less.33 In each case, however, former belligerents are opting for non-violent ways to manage their differences. Power-sharing is one answer to Africa’s security predicament, and it is one that has been frequently advocated by scholars and practitioners alike. By giving all—or the most significant—parties a slice of power, inclusive agreements lower the political stakes in conflicts and provide an equitable solution to the question of “who rules?”34 For the disputants themselves, power-sharing is also an attractive option, because it solves the enduring problem that minorities face in divided societies where voting patterns reflect ethnic lines; that is, where they are doomed to exclusion by the fact that they can never acquire sufficient votes to win office.35 From the perspective of the international community, powersharing is also appealing because it does not require that decisions be made on the legitimacy of each disputant’s motives.36 Instead, it merely assumes that conflicts arise from parties being denied their legitimate rights to representation and autonomy.37 Indeed, exclusion from power is frequently cited as the principal reason for taking up arms in Africa and elsewhere. If it is true that political actors are compelled to act aggressively only because their exclusion from power leaves them with no other option, then it is difficult to imagine solutions to violent conflict other than power-sharing. In fact, inclusive coalitions have long been a fundamental feature of the African political landscape. Regimes are often dependent on the careful construction of clientelist networks that incorporate a sufficient number of representatives from different ethnic groups and regions in their respective governments.38 Requisite for power-sharing to function is that adversaries actually want such a system, have an interest in its success, and be willing to cooperate with other ethnic elites. As Arend Lijphart has observed in his discussion of so-called “consociational” power-sharing, such arrangements require political elites to “make deliberate efforts to counteract the immobilizing and unstabilizing effects of cultural fragmentation.”39 In this sense, it could be said that power-sharing is something which must be believed if it is to be seen.

The Search for Security in Africa

11

But getting combatants to sign on to peace agreements can be difficult in the aftermath of war. Having just engaged in violent acts against each other or fought a civil war, disputants almost never want to be together, let alone cooperate or share power on equal terms. The misery of war does not lead disputants to think that violence was the wrong decision; it crystalizes ethnic sentiments and ethnic hatred. If groups cannot arrive at an inclusive peace agreement before conflict erupts, they are often in no better position to do so after conflict has begun.40 As one observer has remarked in the case of Rwanda in the aftermath of the 1994 genocide, having Tutsis share power with Hutu extremists was akin to the Jews agreeing to share power with the Nazis or the Armenians cooperating with the Turks.41 As desirable as it might be for outsiders, such arrangements may be too unpalatable for belligerent parties to accept. Consequently, given the psychological barriers that must be overcome, if power-sharing requires an intrinsic desire to come together, it appears to be a scenario that for most disputants is heavily circumscribed. Beyond the psychological problems, there are other barriers to power-sharing or other negotiated forms of settlement. Most African states are structured with weak or fledgling institutions—a problem that, again, is even more acute in the aftermath of a civil war. Political settlements may be forged but, in the absence of a powerful guarantor or substantial authority being allocated to the opposition, the newly created institutions are rarely capable of managing internal disputes and political competition over the long run. In these contexts, disputants would rather have security than representation. That is why, during Zimbabwe’s transition to majority rule in 1980, Robert Mugabe stated that he would rather have control over the army than representation in parliament.42 Similarly, twenty-eight years later, Zimbabwean opposition leader Morgan Tsvangirai argued that he would not enter a power-sharing agreement with the regime which did not give him the authority to rule effectively, stating, “It’s better not to have a deal than to have a bad deal.”43 To be sure, African leaders are adept at forging inclusive coalitions amongst key ethnic groups. After all, most African regimes are built on complex patronage networks which include representation from different clans and ethnic groups. But owing to the difficulties in forging such agreements across the divisions caused by conflict, such alliances and coalitions are often built as a means of defending or projecting power rather than as a means of conflict resolution or altruism.44 In other words, and as my discussion of post-1991 Ethiopia will demonstrate, such coalitions may be regarded unfavorably and rejected by at least

12

Civil War in African States

some local disputants because they are seen more for serving the purpose of domination rather than integration and reconciliation. Nonetheless, while the possibilities for consensus are limited, there are opportunities for more limited forms of cooperation and/or inclusion. In fact, the idea that one would accommodate another or give way on a particular issue often suits one’s survival needs. Reflecting on Chamberlain’s deal-making with Hitler, for example, Isaac Chotiner writes, “appeasement was not about weakness or pacifism or an unwillingness to confront danger; rather it was a cold-hearted realist strategy that saw negotiations with Hitler as the best way to ensure the survival of the British Empire.”45 The key motivation remains selfinterest: disputants will agree to cooperative arrangements when it is expedient to do so and when the perceived risks associated with a peaceful outcome are less than those associated with new or renewed conflict.46 In this way, genuine moments for cooperation or integration, limited or infrequent as they may be, are not based on trust so much as they are on the coincidence of common interest and self-preservation. When Siad Barre in Somalia and Mengistu Haile Mariam in Ethiopia came to terms with each other in 1988 after years of acrimony and war, their efforts were motivated less by an affection for each other than by a shared need to rid themselves of the insurgencies that each was supporting in the other’s state. Domestically, a regime’s final days are also often notable for the near-unconditional offers of peace and inclusion to the most prominent and threatening insurgent movements. More cynically, even when they are not in their final death throes, governments have an interest in encouraging defection from the opposing ranks and will offer willing adversaries protection. Inclusion can then be a means of dividing an adversary or isolating a particularly hardline faction from its moderate base. In Somalia, for example, President Siad Barre bought off and effectively neutralized the rebels associated with the Somali Salvation Democratic Front so that he could continue his oppression of other, more dangerous, opponents to his regime.47 In short, even if cooperative sentiments are not shared by all of one’s adversaries, an inclusive approach can be motivated by the same self-interested logic as the strategies of domination and separation that are discussed below. For rebel movements, a public demonstration of willingness to accept a power-sharing agreement (or even to participate in a process which might lead to one) offers other valuable advantages, most notably international recognition for rogue or otherwise weak insurgent movements or ethnic minorities.48 For the weakest parties or those facing military defeat, power-sharing can be a means to buy time or

The Search for Security in Africa

13

stave off elimination. Moreover, since power-sharing is an approach which, by definition, involves the acceptance of all or most warring parties, it can entail “a major adjustment in the domestic balance of power . . . in favour of insurgents at the expense of state leaders.”49 Unfortunately, while the international community may condemn violence, the effect can be to reward its use by insurgent groups when power-sharing is advocated as a form of conflict resolution. These efforts may be counter-productive because the result is often a proliferation of smaller rebellions and factions who utilize ever greater levels of violence as a means to gain recognition and access to state resources. For smaller movements in particular, any slice of power is sufficient reward after violent conflict; indeed, rebellions are often undertaken with no greater goal in mind.50 Strategies of Domination and Conquest

The state is the institution that communal groups look to and seek to control in order to dominate others. In Africa, ethnic groups and insurgencies have acted on the assumption that capturing the capital city entitled them to rule over the entire country. In most cases, upon independence from colonial rule, political authority was given to the movement that controlled the capital, regardless of whether it had sufficient means to exercise state power over all corners of the territory. Nonetheless, the new leaders deemed it a central function of their regime to extend state authority by eliminating or neutralizing rivals who lived within their territorial boundaries or who threatened to dismember the country territorially. Consequently, to shore up their questionable power, state leaders bought into the idea of “unambiguous sovereignty” from the colonial rulers and held that, in juridical terms, a given territory could be ruled by one, and only one, power. Others who challenged this authority were illegitimate.51 Indeed, groups who threaten a regime’s exclusivist view of the state are perceived as jeopardizing its very existence and risk subjecting themselves to whatever coercive measures the regime believes are required to maintain order. Moreover, regimes may use threats by others to destroy the integrity of a state in order to position themselves as the defender of the nation and justify a strategy of domination and conquest. As Siad Barre’s notorious military commander argued in justification of his government’s unrestrained use of force in the northwest region of Somalia during the late 1980s, “I was defending a country from a guerrilla movement that was backed by the Ethiopian government. I had

14

Civil War in African States

obligations to protect the territorial integrity of Somalia and I was defending my borders.”52 Unambiguous sovereignty is less of a problem in Western societies because power and authority are distributed and deeply entrenched among an array of complex and diverse institutions. In the developing world, however, power at the center is not complemented by the same diffuse array of institutions, a fact which leads states to zero-sum politics and intensifies the stakes over which politics is fought. When the state vests power in a single authority, that power typically becomes the instrument of one societal group—often at the expense of all others. As a result, the state becomes a battleground for a territory’s communal groups. Violent struggles for power among different ethnic groups can be particularly intense because any subsequent compromise requires at least one of the parties to forego its version of the national narrative, one that may have been built on histories of exploitation and domination by another group. Rarely can a single state sustain two such national narratives. The stronger a group’s belief that violence can be justified for the purpose of state-building, the greater their drive for political domination. In this sense, the civil wars which are endemic in much of Africa and elsewhere in the developing world represent ongoing processes of state-building that are not unlike those which took place in the developed world in centuries past. State-building requires leaders to rationalize power and manage dissent. To be sure, and as my discussion has already described, many dissenters can be co-opted with various incentives. But others will remain outside the political process, holding out for benefits that no realistic offer can satisfy. At this point, Edward Luttwak argues, political leaders face a new choice: use whatever force is necessary to crush the extremists and consolidate power, even if it leads to civil war, or risk having to live indefinitely in a divided state. The choice is a false one according to Luttwak: “Better a brief civil war leading to peace and an independent state than an impossible coexistence with armed extremists that would endlessly prolong the suffering.”53 It is not that communal groups cannot tolerate or accommodate opposition; rather, the state cannot tolerate groups or entities whose existence undermines the state’s ability to claim itself as sovereign. It is one thing to share power with groups or individuals who have arrived at the conclusion that killing for political purposes is no longer acceptable, or who live in institutionalized states where there are established mechanisms for managing conflict. It is another thing to share power with an adversary whose very presence threatens one’s existence. For

The Search for Security in Africa

15

the state, like its challengers, the issue is very much an existential one. Dominating one’s adversary is more often regarded as the safer option. Barbara F. Walter concurs that, in the absence of an external guarantor, belligerents would prefer no agreement to an agreement that increases their vulnerability. Civil wars, she observes, rarely end in negotiated settlements of any kind. Instead, most internal wars end “with the extermination, expulsion, or capitulation of the losing side.” Her data shows that “groups fighting civil wars almost always chose to fight to the finish unless an outside power stepped in to guarantee a peace agreement.”54 Such a finding is no surprise to those who argue that international anarchy compels state actors to fight wars in the hope that they can win rather than to seek negotiated settlements that place them in intolerably vulnerable situations. As John Mearsheimer argues, “Given this fear—which can never be wholly eliminated—states recognize that the more powerful they are relative to their rivals, the better their chances for survival. Indeed, the best guarantee of survival is to be a hegemon, because no other [belligerent] can seriously threaten such a mighty power.”55 More succinctly, Henry Kissinger writes, “in a war, it is not enough to endure—it is essential to prevail.”56 For disputants who hold the balance of power, there may be little reason to share power except in the form of mere token gestures intended to satisfy the demands of the international community. Inducing disputants to forego strategies of domination has proven to be difficult both inside and outside Africa. In Sri Lanka, until it launched a final military assault on the Tamil Tigers in 2009, the government in Colombo concluded that its efforts to reach a negotiated settlement were being exploited by the rebels, thus prolonging the war. Crushing the rebels, it said, was not only justified; it was humane.57 In Iraq, American policymakers urged a devolution of power from the center to the country’s regions, a shift envisioned by the new Iraqi constitution. As one New York Times columnist pointed out, however, “Everybody out of power sympathized with their [the American policymakers’] diagnosis but everybody in power rejected it.” He added, “There is a winner-take-all mentality which is not conducive to compromise.”58 The preference for domination can be rooted in a sense of victimhood or historical injustice perpetrated by others. “Through force of arms,” writes another observer, Iraq’s Shiites “intend to dominate the country entirely, taking what they believe was stripped of them when their revered leader Hussein was murdered in the desert of seventh-century Mesopotamia.”59 In Africa, this same logic has repeatedly played itself out. Robert Jervis and Jack Snyder note that “the Tutsi minority in Burundi . . .

16

Civil War in African States

counted for its survival on maintaining a dictatorship and a monopoly over military power. To them, democratization and proportional representation in the military ranks and officer corps, measures pressed on them by the international donors, were indistinguishable from a death sentence, since they felt that maintaining their control was the only way to guarantee against victimization by the Hutu majority they had brutalized.”60 Regarding neighboring Rwanda, Bruce Jones observes that “the victims [of the 1994 genocide] owe their deaths to those members of the akazu who chose to massacre them rather than relinquish their grip on power.”61 In these ruthlessly competitive systems, the most successful actors adopted the worst, and least accommodating, practices in order to survive.62 Aside from assuring one’s physical survival, power opens the door for other opportunities. Not only is one more secure in an existential sense but domination allows one to control other benefits that come from holding high public office in centralized states. In a land of poverty, it is these perquisites which can make political office worth fighting for. As one journalist wrote regarding Angola, “The misuse of oil revenues robs Angolans of more than health care and education. The opportunity to steal millions gives the country’s leaders a reason never to risk this privilege by democratizing.”63 In other words, the opportunities offered by corruption are themselves incentives against power-sharing, democratization, or even decentralization. In sum, there is a perceived finality to conquest which is attractive to communal groups. As the absolute sovereign, it can subdue a rebellion, preserve order, ensure a group’s survival, and acquire control over scarce resources. Even some Western analysts agree that, despite the obvious contradiction, warfare that results in a decisive conclusion can be an effective form of conflict resolution. As Edward Luttwak states, “although war is a great evil, it does have a great virtue: it can resolve political conflicts and lead to peace.”64 Indeed, frustrated with the unwillingness of the Revolutionary United Front rebels in Sierra Leone to honor their commitments to peace, one prominent international advocacy group advocated that “those in the RUF who refuse to demobilise should be defeated militarily.”65 In most cases however—particularly where rebels represent a genuine and significant political constituency—a strategy of domination does not come without costs or risks. Achieving a satisfying “final, crushing victory” may be an example of human desires producing an outcome that is ultimately contrary to one’s own interests.66 In pursuing its own interests, a communal group’s authority is not accepted by those whom it seeks to dominate; ethnic minorities do not want to live in

The Search for Security in Africa

17

states they perceive as hostile, and ethnic majorities resent being ruled by powerful minorities. In both cases, a strategy of domination without any form of political inclusion is, in the longer term, self-defeating insofar as it provokes an endless cycle of revolt and repression. This is a principal limitation of a coercive strategy, and it provides insight into how violence can beget further violence. Strategies of Separation and Secession

The third and final strategy used by disputants in a conflict is separation; that is, they pursue formal independence as a sovereign state. Statehood, the highest level of contemporary political organization, has long been regarded by nationalist groups as a desirable means to satisfying nationalist sentiments. As the black activist Martin R. Delany remarked, “The claims of no people . . . are respected by any nation until they are presented in a national capacity.” But independence is not merely a romantic ideal. In conflict situations, it also serves as a survival strategy insofar as it reduces or eliminates inter-ethnic competition. For some groups, legal recognition is essential because independent states are entitled to acquire the means to defend themselves from their adversaries.67 Equally important, independence shifts the burden of a disputant’s security from one’s partner in a peace agreement to the legal conventions of the larger international system. Given that the international community values respect for state borders and for the stability thereof, and given the rarity of interstate wars compared to civil wars, the formal achievement of statehood represents a viable means to ensuring one’s survival. Indeed, the international conventions of contemporary world politics have proven to be an essential means of survival for even the weakest states in Africa and elsewhere. “Legal recognition,” write Robert Jackson and Carl Rosberg, “has been far more important than material aid in their emergence and survival to date.”68 Inherent in a secessionist strategy is the assumption that one group will ultimately control the political entity that is to be established. While a group may accept that they are not in a position to control the political center, they remain fearful that they will not be secure unless they are sovereign. Sovereignty is the most desirable part of any secessionist strategy. War-weary groups who believe that they have been victimized by others say they will accept nothing less than a state of their own. This they can in turn control, despite the obvious hypocrisy inherent in the fact that such unilateral action may come at the expense of someone else’s security. It is in this sense that secession is not an end in itself but

18

Civil War in African States

a means for a group to gain or regain control of its destiny. It is a strategy which results in domination. One may debate, of course, whether partition is in fact a viable means of conflict resolution.69 If it is not, why, and to what extent, do insurgencies pursue this strategy? Some scholars, such as Donald Horowitz, have argued that secession does not reduce violence or minority oppression once successor states are established because, with the birth of each new secessionist state, comes a new, possibly fearful, minority within it.70 Others who dismiss partition as a form of conflict resolution claim that, in the short term, secession does not in fact lead to peace or to more security for the insurgents. Barbara Walter, for example, argues that regimes are reluctant to allow secessionist movements to achieve their goals and that they resist such efforts because failing to do so sends a message to other would-be secessionists that the regime is incapable of maintaining its territorial integrity. “Once it becomes clear that governments can no longer defend their own sovereign territory,” she argues, “they become targets for any domestic or international foe.”71 Consequently, governments will strenuously resist secessionist movements—an ominous tendency that features strongly in the calculations of would-be secessionist movements. Because of this threat, even if in the end sovereignty leads to a peaceful and more secure outcome, the risks and pain involved may preclude it as a viable strategy. The same processes which give African states their security—that is, international conventions on the sanctity of state borders —also help to make the prospects of secession in Africa extremely unlikely. In theory, as Pierre Englebert and Rebecca Hummel have pointed out, African states should be prone to secessionism. The states which constitute the continent are relatively new and they are usually governed by weak and exclusionary regimes that seek, however ineffectively, to extend their authority over large territories. Moreover, their populations are diverse, a fact which might incline them to seek separate paths of autonomy. Insurgents also often sit astride resources that could sustain them both in their military campaigns and as independent states. When considered in proportion to the continent’s high incidence of violent conflict, however, the number of secessionist conflicts is in fact quite low.72 The claim of Englebert and Hummel is not that the strategy is itself unappealing; rather, they argue that even if insurgencies have secessionist ambitions, most of them (and especially the smaller ones) can be tamed with promises of access to state resources from the central government. Secessionist movements, in short, can be bought off. Undoubtedly, an inhibiting factor to successful secessionism is also the

The Search for Security in Africa

19

widely held convention—represented most clearly in Articles 3 and 4 of the African Union’s Constitutive Act—concerning respect for the sovereignty and integrity of African states.73 While secession is comparatively rare, the motivations behind it merit examination. First among these is that even when secession is not desired, the mere threat to disrupt a country’s territorial integrity can be used tactically to bring about either material rewards or the inclusion of previously marginalized groups in a political process.74 There are, nonetheless, more security-specific reasons for secession. In his insightful essay on the consequences of state formation in the developing world, for example, Mohammed Ayoob notes two key questions that capture the dilemmas facing local actors and explain why secession, as opposed to autonomy, remains an attractive option for some.75 First, even if a peace agreement is achieved, what guarantees are there to ensure that rebellious groups will put down their weapons and reconcile themselves to an autonomous or semi-autonomous existence which is dependent on the good faith of the central government? Second, what guarantees exist to ensure that the central government will in fact abide by its commitment to respect minority rights and regional autonomy over the longer term? From the perspective of local actors, mechanisms such as federalism and power-sharing—both integrative solutions—present considerable risk insofar as, ultimately, there is little to stop a regime from reneging on its pledge to respect another group’s autonomy. Consider, for example, the annulment of the 1972 Addis Ababa peace agreement between Khartoum and the Sudanese People’s Liberation Army (SPLA). Discussing that case, Taisier Ali and Robert O. Matthews write, “The future of the region of the Southern Sudan relied entirely on the whim of one man, who could, as he had fashioned the agreement, just as easily break it. And that is precisely what President Nimeiri began to do as early as 1977, five years after the accord was signed.”76 As I shall demonstrate in chapter 2, this sequence of events also took place with respect to the 1952 Federal Act between Eritrea and Ethiopia. In both cases, the international community failed to punish governments who disregarded minority rights even though provisions for autonomy were guaranteed in formal agreements. Secession prevents the security of a group being so easily compromised. If there is no external guarantor, separation is a convincing strategy because the remedy that emerges from insecure environments allows—even requires—that each party take measures to defend itself. Sovereignty and the consequent entitlement to arm oneself may offer the

20

Civil War in African States

only form of protection, even if it means violence will occur in the short term. Factors That Can Influence Strategy

In this book I do not suggest that one of these strategies is or should be preferable. Political actors will, according to their needs, choose the option (or in some cases options) that best serves their search for security. How can we account for or explain these different choices? One can identify four sets of factors, operating in complex ways, which can incline or disincline disputants to one or more of the above strategies: (1) the balance of power among the disputants; (2) history, memory, and precedent; (3) internal attributes; and (4) global and regional factors. Balance of Power

Amidst the precariousness of a civil war, the disputants are ultimately concerned with their own survival. They must therefore pay close attention to the distribution of power or the balance of coercion among them. The most obvious demonstration of the influence power has on the selection of strategy in civil conflicts is the receptiveness a weak party shows towards integrative strategies. Weak parties are at risk of being eliminated or dominated by their stronger adversaries and, in order to save themselves, will seek to have their goals accommodated by making offers to cooperate with the dominant power. In Chad and the Democratic Republic of the Congo, for example, dozens of insurgencies have operated without any realistic chance of gaining power in the capital city. In these cases, the smallest insurgent groups may seek little more than to gain materially from the havoc they cause. Other groups, such as the Tigray People’s Liberation Front in Ethiopia, found that, as their power increased, other opportunities emerged and they saw the possibility of themselves becoming dominant. Attention to power balances means not simply assessing who has more guns today (the static situation), but also calculating how those power balances are likely to change in the foreseeable future. The end of the Cold War and the decline of superpower patronage, for example, allowed rebel movements to be much bolder in their demands and goals since, in many cases, they knew that further military aid would not be forthcoming to those in power. Additionally, those whose power is declining may see power-sharing as a strategy that allows them to buy time to regroup and rebuild supplies before making a bid to dominate

The Search for Security in Africa

21

their adversaries. In short, rarely is there a losing or weakened disputant who would reject a power-sharing agreement, particularly if it means they would otherwise be shut out of the political process; equally rare is the rising disputant who will consider a power-sharing agreement for anything more than the short term. Power is also likely to bear on secessionist conflicts. Stronger parties have more options available to them and can, for example, consider either secession or domination as options. Conversely, as Englebert and Hummel write, leaders of weaker “culturally distinct, oppressed, or otherwise polarized groups or regions may well initially prefer to go their own way but find it hard to pursue sustainable separatist strategies in Africa’s commodity dependent and sovereigntyconstrained environment.” They add that, since international recognition is so hard to come by, marginalized and excluded groups derive “greater benefits from joining ‘national unity’ governments than from continuing their original struggle.”77 How then does power manifest itself in Africa’s civil wars? Obviously the number and capability of a group’s weapons is fundamental to this calculation. During the Cold War, determined efforts by the superpowers—and the vast military patronage they provided—gave recipient regimes such power advantages over rebels that their defeat at the hands of insurgent movements only rarely happened.78 The end of the Cold War reduced the number of state-tostate arms transfers, but this has been compensated for by the proliferation of the gray (commercial) and black (illegal) market smallarms transfers that have flowed to both regimes and insurgents.79 Civil wars in Africa are, to be sure, often fought with relatively unsophisticated weapons but resource-weak rebel movements can look to other assets to develop their power. The cohesiveness of a disputant is a key variable—although this is a more difficult concept for political scientists to quantify. Barry Posen argues that, since the French Revolution, national identity has been an important determining factor in the relative strength of armies. Because individuals with a strong sense of national identity are more willing to cooperate with each other, they have the advantage over those with a weaker sense of identity. Posen concludes that the collectivity’s sense of “groupness” in ethnic, religious, cultural, or linguistic terms provides them with an important dimension in offensive military power.80 Other scholars concur that, when other forms of hard power are not available or are in decline, state and insurgent leaders invoke ethnic identity to increase their support.81 While one might question the extent to which solidarity can be manipulated by elites, it is compelling to think that power can manifest

22

Civil War in African States

itself not only in terms of weaponry but also in terms of the solidarity of a group and its ability to mobilize itself to advance a collective interest. History, Memory and Precedent

Decisions on strategy are not always made on the basis of objective assessments of the local balance of power. They are also a product of social situations.82 Indeed, a number of scholars have noted the role of history in shaping insurgents’ strategic objectives and their evaluations of potential rivals.83 Insurgent groups may use history as raw material to define and shape both their identity and the type of strategies they adopt. A separate colonial existence, for example, may lead to the establishment of dedicated governing structures that distinguish land from other regions or jurisdictions and form the basis of a future independent state—as has been the case in East Timor in Indonesia, and in Eritrea and Somaliland in the Horn of Africa. The international community tends not to encourage secession, although, as Englebert and Hummel observe, it does consider decolonization as an acceptable form of self-determination.84 Secession may also be a consequence of the historical oppression of a communal group; oppression may not only create a sense of solidarity among people but may induce them to seek an exit as a means of preventing similar treatment in future. Or, insurgents may use a long history as a unified state to justify resorting to a strategy of coercive domination in order to maintain territorial integrity. Communal groups that regard themselves as the historic rulers of the country—such as the Shoa Amhara in Ethiopia—may never consider secession, as to do so would contradict their view of themselves as the rightful guardians of the state’s legacy. Communal groups or states are also inclined to manufacture, embellish, or manipulate historical events for security purposes. As we will see in our discussion of Ethiopia, for example, their rulers have often claimed that Ethiopia has encompassed the entire Horn of Africa and territory stretching into Arabia.85 We can consider the role of memory in light of the fact that all renditions of history are selective and partial. Memory works in at least two ways when it comes to civil conflict. First, memories serve as a mobilizing force for action. What may appear objectively to be unprovoked acts of aggression or domination may be rooted in memories of previous acts of aggression or victimization.86 Elites can manipulate historical events to shape identity and bring coherence to otherwise disparate groups. Political leaders can also hark back to past oppressions to galvanize otherwise unmotivated people and distinguish themselves from others. In Ethiopia in the 1970s

The Search for Security in Africa

23

and 1980s, for example, Tigrayan rebels referred to themselves as Woyane in reference to a rebellion that was crushed by the British in 1943. Second, history and memory—particularly of recent events—teaches lessons, highlights previous security breaches, and serves as a guide for future action. In Angola, as we shall see, the ruling MPLA learned not to repeat the mistake of allowing itself to be disarmed and made vulnerable after its adversary reneged on a pledge to abide by results of elections held in 1992. In Ethiopia, Emperor Haile Selassie insisted on maintaining control of Eritrea because experience had shown it to be a gateway for invaders. In some cases, history and memory can provide decision-makers with the only means to assess the intentions of other communal groups. This becomes particularly important in circumstances when the state is weak and the government is unable to protect all groups. Each one is then left to make its own determinations of how the others can be expected to behave. Frequently, these decisions are made on the basis of a “worst-case analysis” whereby every other group is regarded as the enemy.87 Although memory does not point clearly to a particular strategy choice, one can make predictions. A minority group’s memory of oppression by another group, or its recollection of broken agreements with an adversary, for example, could be expected to lead to a strategy of either secession or domination, as the trust needed for a cooperative arrangement has been broken. Conversely, if memories are more pleasant or if there is no previous experience of close cooperation with a rival (and thus no memory to draw upon), there may be greater opportunities for inclusion and cooperation. Internal Attributes

Strategic models of action are traditionally outward-looking insofar as strategy is determined by external factors. The environment or “structure” (the distribution of power, for example) is assumed to impose itself on the actors. But other scholars have argued that such an approach is unsatisfactory as it does not explain why, for example, apparently similar situations can degenerate into violent conflict in some regions but not in others, or why some political actors are more consistently peaceful than others.88 These scholars supplement purely strategic situations with factors which are internal to the group. Memory, just discussed, is one such internal factor. Others could include, but are not restricted to: (i) the presence of scarce or valuable resources; (ii) geography; and (iii) demography. While these latter three factors are not internal to the actors themselves in the same literal sense

24

Civil War in African States

that memory and history are, each of them can, nonetheless, discretely shape the interests and objectives of a disputant. In some cases, they are critical in influencing the power relations among the disputants. Paul Collier’s thesis on the significance of greed as a mobilizing factor in violent conflict tells us something about the choice of strategy that a given communal group will most likely pursue. Secessionist strategies, he says, might best be seen as the “rage of the rich.”89 Rebel groups will be able to exploit critical resources and the export revenues those resources produce if they know that they can take those resources with them when they secede from the rest of the state. Regions that are currently wealthier than the rest of the territory may also choose to secede to release themselves from the obligation of having to support poorer regions or to pay for post-war reconstruction.90 Secessionist strategies, then, are most likely to be found among wealthier and more prosperous regions or among those people who can readily get access to valuable natural resources. It is therefore no surprise that rebels in the mineral-rich Katanga region of Congo/Zaire, the oil-producing regions of Biafra in Nigeria and Cabinda in Angola, and the relatively prosperous province of Eritrea have all pursued secessionist strategies.91 Even if their political support is concentrated in a given area, rebel movements whose base is not so favored with abundant resources are less likely to pursue a secessionist strategy. They might, however, seek integration or domination often as a means of gaining access to these same resources. While the UNITA rebels did establish a state-like entity referred to as “Savimbiland” in the southeastern corner of Angola, it did so mainly for the purpose of fighting the war and to demonstrate to outsiders its capacity to govern in the capital should that opportunity arise. The principal explanation for the rebel’s desire to take power in Angola’s capital, Luanda, writes Tony Hodges, is that Angola’s substantial oil wealth, and the conflict’s central prize, is not located in the Ovimbundu areas which constituted Savimbi’s power base, but is concentrated mainly in reserves off the northwest coast of Angola between Luanda and Cabinda.92 Groups that want to secede, however, are not always richer than their adversaries and, according to Donald Horowitz, it is backward groups in backward regions who are more inclined to have secessionist ambitions.93 Regions may choose to secede precisely because of their own weakness or poverty. Secession allows them to acquire means to defend themselves that may previously have been denied to them as long as they were part of a larger entity. Secession may also, in effect, be a means to surrender themselves to a larger and more powerful authority that can protect small and otherwise unviable nations. Secessionists in

The Search for Security in Africa

25

Kosovo, for example, have looked to the continent-wide European Union to shield its fledgling independence. In this sense, Kosovo, an economically dubious entity, sought separation (and integration into an international organization) as the best means for defending its interests.94 Geography can also be expected to play a role in the choice of strategy. Landlocked regions or provinces may have a more difficult time pursuing a secessionist strategy and might instead seek domination or integration. On the other hand, geographically distinct regions, such as the exclave of Cabinda, East Pakistan, or the Aceh region of Indonesia, are separated from the rest of the country by a portion of land belonging to another country and/or by a body of water. With geographically distinct regions, then, not only is the territorial basis of a new state easily and clearly identifiable, but central governments have more difficulty exerting their authority over such regions. Central governments wanting to avoid the dismemberment of their state may closely observe any secessionist tendencies in these regions and seek to suppress them forcefully before a dangerous precedent is set. Kongo Central, the western-most province of the Democratic Republic of the Congo, for example, provides that country’s only (albeit limited) access to the coast—making the government in Kinshasa particularly intolerant of potentially precedent-setting secessionist tendencies anywhere in the country.95 Finally, scholars of nationalist and ethnic conflicts have observed the way in which demographic factors—including the proportion and distribution of identity groups within a territory—affect the types of survival strategies that communal groups pursue and the likelihood that central governments will resist. Ted Gurr writes that “the claims and political strategies of ethnopolitical groups vary according to their type and circumstances. National peoples usually seek exit, a goal that often leads to separatist wars and state repression.” Minority groups, by contrast, “want access [to existing state resources], a goal usually pursued by conventional political action and protest campaigns.”96 Stephen Van Evera argues that central governments are more likely to tolerate a secessionist movement if it leaves behind a more homogenous rump state. Permitting secession in that case would set a less-damaging precedent than in a more complex multiethnic state where the departure of one group or region could lead to a dangerous succession of secessionist movements which might ultimately dismantle that state.97 Demography and secession can also overlap in significant ways to further complicate decisions regarding strategy. In the Sudan, the Islamic government in Khartoum indicated its willingness to tolerate a

26

Civil War in African States

referendum on the secession of southern Christian African ethnic groups. It has shown less willingness to tolerate any sort of insurgent movement in western Darfur, a region of Islamic African ethnic groups whose secession, were it to be pursued, could have more damaging consequences for the rest of the country. As Posen has argued, demography can also influence the power that a communal group can wield and the likelihood that it will be able to successfully execute a given strategy.98 Cross-cutting ethnic, clan or regional differences can have a profound impact on the ability of an insurgency to reach consensus on a given strategy and can provide opportunities for central governments to undermine their cause by exploiting differences. Alternatively, the heterogeneity of the rump state’s army can also impede a government’s ability to suppress a more unified secessionist effort.99 Global and Regional Factors

Finally, exogenous conditions can influence the occurrence of both violence and strategy selection among communal groups and governments.100 Since governments tend to respond to rewards and sanctions, external actors can exert leverage more successfully on them than on insurgents who may see themselves as accountable only to the people they claim to represent.101 The most obvious indication of the way in which external factors influence local actors was the post-Cold War transition and the manner in which global factors affected the calculations of clients in disputeprone states. During the Cold War, client regimes or insurgent groups in each of the case studies here could rely on external support which afforded them the independence to avoid cooperative or accommodative arrangements with their adversaries. Also, regional spoilers such as apartheid-era South Africa proved to be effective in challenging regimes it did not like by supporting incipient rebel movements who were willing to engage in violence.102 In the post-Cold War era, by contrast, a consensus has emerged on the need to find integrative and democratic solutions to conflicts. Moreover, there has been a tendency to frown upon solutions which involve either separation or domination. Knowing the international community’s preference for inclusive approaches to conflict resolution, smaller insurgent movements have proliferated and carried out violent acts in order to draw attention to themselves and gain entry into the political system, a place that neither their population size nor the sophistication of their political agendas would normally warrant.103 In this sense, insurgent factions arguably wag the dog and

The Search for Security in Africa

27

exploit the international community’s proclivity towards inclusiveness as a form of conflict resolution. External forces can also alleviate insecurities that accompany civil wars and make cooperative agreements possible—provided such interveners are willing to stay long enough for incipient institutions to take root. South Africa proved to be critical in helping to forge a peace agreement in Burundi in June 2003, under the most trying circumstances.104 External pressures or threats can also provide incentives for elites to cooperate—although an examination of postcolonial African history demonstrates that there are few external threats on states that are sufficiently strong to overcome the deepest internal cleavages.105 It is not difficult to see how global factors can have an impact on other strategies as well. Jack Snyder, for example, has observed that globalization “can make separatism look attractive,” as small regions can claim and profit from lucrative natural resources.106 In the end, however, secessionist strategies are still largely—if not entirely—dependent on the willingness of states in the international community to recognize them. As Donald Horowitz observed, and as our discussion of “Somaliland” will show, while the emergence of a secessionist movement is conditioned by domestic politics, a secessionist movement’s ultimate success is conditioned by international politics.107 Particularly since the breakup of Yugoslavia, scholars and practitioners have discouraged the disintegration of other ethnically complex states. “I hope we do not see the creation of any more nationstates,” observed Britain’s Foreign Secretary Douglas Hurd.108 In a similarly discouraging statement, Chester Crocker, the American Assistant Secretary of State for Africa, wrote that the international community “should think twice about calling for the breakup of more states.”109 In their founding documents, Africa’s continental institutions, the Organization of African Unity and the African Union, have also been explicit about the need to avoid alterations of colonial borders.110 In its communiqués, the African Union continues to iterate its “attachment to the unity, territorial integrity, and sovereignty” of African states.111 While some scholars have wondered what the impact of these prohibitions has been in terms of the choices local actors make, the fact that virtually all African states are, in theory, vulnerable to disintegration among their diverse populations means that the provision is effectively self-reinforcing.112 Mark Zacher has provided compelling evidence that the international norm against border change is very strong and continues to get stronger, particularly in Africa.113

28

Civil War in African States

Despite the fact that both the United States and the Soviet Union facilitated or subsequently endorsed the takeover of state power by clients who espoused their preferred ideologies—Mobutu in Zaire (USA), Mengistu in Ethiopia (USSR), and the MPLA in Angola (USSR) being notable examples—the international community now warns against the violent take-over and domination of state power unless the incumbent regime is particularly odious. The African Union also continues to “reaffirm its total rejection of any attempt to seize power by unconstitutional means,” as stipulated in its Constitutive Act and other relevant documents.114 Notably, the international community tends not to object as strongly or to punish the use of violence as a means of gaining a portion of power when the result is an inclusive or cooperative arrangement. Qualifications on the Use of Strategy

In light of the three strategies outlined above and the four contributing factors, this chapter’s final section will offer four more general qualifications to this discussion. First, the strategies should be regarded as ideal types which do not necessarily manifest themselves in mutually exclusive terms. Instead, as our three case studies will demonstrate, disputants may pursue various combinations of these strategies in order to meet their security needs. For example, disputants may work cooperatively with one adversary as a means of dominating another more dangerous one. Or they may support cooperation in the short term in order to better position themselves for a strategy of domination. These strategies are also highly situational, and their use may change over time in the same way that a quarterback alters his/her plan of action depending on the conditions on the field. While there is a certain consistency to the strategies chosen, some factors (the balance of power) may be more transient than others (history, demography). The fluidity of African politics means that flexibility in strategic planning is also unavoidable if security is to be sustained over the long term. Failure to read the political situation correctly will almost certainly spell doom for a disputant. Inevitably, strategies must reflect the unique conditions facing each of the principal actors in a conflict. This is not, of course, unique to African politics. Second, to say that a given disputant seeks its security by pursuing a given strategy does not mean that it succeeds in securing its survival. Rationality does not presume foresight. Successive Ethiopian governments pursued a strategy of domination even though they could not prevail over their adversaries. Somaliland has pursued a strategy of

The Search for Security in Africa

29

secession since 1991 even though it remains unrecognized by the international community. And the MPLA in Angola agreed to a powersharing formula in the mid-1990s even though the agreement was never consummated and was eventually abandoned. Again, the pursuit of a given strategy reflects that disputant’s perspective on how its security is most likely to be achieved. Third, the argument here suggests that there is consensus in the adoption of a particular strategy. In reality, strategy is the product of numerous factors which interact in complex and sometimes countervailing ways. The embrace of a given approach to security may also obscure an underlying lack of consensus within the insurgent movement itself. Some individuals, for example, may agree to a strategy of domination in order to win a war or overthrow their principal adversaries, but anticipate a more inclusive or cooperative approach once that task has been completed. Once victory has in fact been achieved, these individuals are often disillusioned by the strategy that secured them in power and their own regime’s inability to become more open—an aspiration that they may have believed was at the heart of their struggle. In the case of Rwanda, for example, Filip Reyntjens writes that “When a new government took office on 19 July 1994, the RPF reaffirmed its commitment to the terms and the spirit of the Arusha Accord and the logic of power-sharing it contained. . . . However, a number of amendments made unilaterally by the RPF . . . profoundly modified the political regime agreed in Arusha. They . . . imposed the dominance of the RPF in government.”115 The resulting split in the RPF has since led to the departure of a number of prominent individuals within the movement. On the other hand, the adoption of particular strategies can also have the effect of making corporate an otherwise fractious population. A conviction that a given strategy is essential to the group’s survival generates solidarity among citizens and leaders alike. There can be pressure to conform and, by contrast, risk in dissent if such dissent suggests sympathy with the enemy. With respect to both the Eritrean and Somaliland referenda, for example, a not entirely-secret voting procedure meant that there was social pressure to vote en masse for independence—and in both cases support for secession was close to 100 percent. In this way, certain strategies make the assumption of a corporate identity self-fulfilling. Finally, the idea that disputants pursue various strategies in order to achieve security must also be considered in light of another body of scholarship that purports to show that rebellions are started and wars sustained exclusively through economic incentives.116 Much effort has

30

Civil War in African States

been expended portraying these as rival approaches to understanding and explaining conflict. A more fruitful way forward is to see these approaches as compatible and even interrelated. As Jeffrey Herbst has argued, all conflicts involve varying degrees of both economic and political incentive or “greed and grievance.”117 In the African context, security is not just about survival but about having access to the resources that ensure survival. Groups seek access to the state, either as part of a power-sharing arrangement or as a monopoly, in order to acquire a slice of this resource pie. Groups will also seek to split off from other areas in order to acquire exclusive control of these resources. In other words, the so-called greed approach is not incompatible with the framework proposed above and in fact infuses each of the three strategies. Moreover, while the prospect of untold riches may motivate some leaders and affect how they respond to other incentives, it is less clear that enriching themselves is their exclusive preoccupation. Jonas Savimbi in Angola, for example, appears to have been more interested in power than its trappings. He had a political constituency of his own, and his lifestyle was hardly one of opulence. It appears that resources he gained were a means to sustain his war effort by making it possible to acquire weaponry and soldiers rather than a means to enrich the leadership.118 Indeed, conflicts that have been portrayed solely in terms of the accumulation of wealth often, in the end, can be seen to have had political objectives. While the Angolan government undoubtedly benefitted from the state of emergency coincident with Savimbi’s insurrection, it ultimately sought and achieved a military victory. Continued insurrection was not necessary for the government to profit materially. Even in Sierra Leone, a conflict that has been described as being concerned with nothing more than “diamonds, diamonds, diamonds,” there were political objectives.119 There, the rebel’s rejection of a lucrative peace deal when they launched a high-risk military offensive to capture Freetown suggested that the RUF’s political ambitions were at least as important as its economic ones.120 Conclusion

Disputants in conflicts have long argued that those outsiders who seek to end violent conflicts fail to appreciate the security predicaments and the security threats associated with a peace agreement. South Vietnamese leader Nguyen Van Thieu highlighted the differences in these perspectives to Henry Kissinger, his American ally in the negotiations in 1972, stating, “You are a giant, Dr. Kissinger. So you can probably

The Search for Security in Africa

31

afford the luxury of being easy in this agreement. I cannot. A bad agreement means nothing to you. . . . For us, . . . it is a question of life and death.”121 Indeed, Thieu claimed that he would rather “fight alone” than sign an unacceptable agreement. More recently, Georgia’s president urged European leaders to pay closer attention to his state’s security predicament vis-à-vis Russia. “You should understand,” Mikheil Saakashvili said in mocking reply to European leaders who urged him to make concessions, “that the crocodile is hungry. Well, from the point of view of someone who wants to keep his own leg, that’s hard to accept.”122 If outsiders want peace, peacemakers must first see things from the viewpoint of the adversaries themselves so that they can appreciate the specific security constraints that inform the perspectives of each actor. Moreover, they must make agreements compatible with local security concerns. In the post-Cold War era, conflict management has unfortunately entailed a near-universal resort to power-sharing. Powersharing is a viable option for conflict resolution but, from the perspective of belligerents, it is only one among three—and is a highly limited one at that. Success in conflict management, as the following case studies will show, is contingent on understanding the link between security strategies of the disputants, the factors which give rise to a particular strategy, and the mediator’s solution. In short, peacemakers must understand and appreciate when a given solution is consonant with the disputants’ assessments of their security needs and when it is not. 1. Key international relations texts in the realist canon include Kenneth N. Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979); Hans J. Morgenthau, Politics Among Nations (New York: Alfred A. Knopf, 1973); and John J. Mearsheimer, The Tragedy of Great Power Politics (New York: W. W. Norton, 2001). For a domestic politics version of realism, see Barry R. Posen, “The Security Dilemma and Ethnic Conflict,” Survival, vol. 35, no. 1, (Spring 1993), pp. 27-47. 2. See Charles G. Boyd, “Making Peace with the Guilty: The Truth About Bosnia,” Foreign Affairs, vol. 74, no. 5 (September-October 1995), pp. 22-38. 3. Barry Posen, “The War for Kosovo: Serbia’s Political-Military Strategy,” International Security, vol. 24 no. 4 (Spring 2000), p. 40. 4. Barbara Tuchman wrote that she wanted to avoid speculative statements along the lines of “As he watched the coastline of France disappear, Napoleon must have thought back ....” See “Author’s Note,” The Guns of August (New York: Dell, 1962), p. 8. 5. Both General Mohamed Farah Aidid and Ali Mahdi, for example, the two most central actors in the battle for Mogadishu following the fall of Siad Barre, claimed to represent the United Somali Congress (USC). Ali Mahdi himself belonged to several organizations and militias. Mohamed Farah Aidid and Ali Mahdi were also, however, both members of the Hawiye clan family.

32

Civil War in African States

6. John Marcum, cited in Fred Bridgland, Jonas Savimbi: A Key to Africa (Edinburgh: Mainstream, 1986), p. 451. 7. Bridgland, Jonas Savimbi: A Key to Africa, p. 10. 8. I. William Zartman, “Introduction: Posing the Problem of State Collapse,” in I. William Zartman ed. Collapsed States: The Disintegration and Restoration of Legitimate Authority (Boulder: Lynne Rienner, 1995), p. 2. 9. Henry Kissinger, Years of Renewal (New York: Simon and Schuster, 1999), pp. 904 and 907. 10. From 1999 to 2008, according to the Stockholm International Peace Research Institute (SIPRI), 13 major armed conflicts were recorded for Africa, the highest regional total. See Lotta Harbom and Peter Wallensteen, “Patterns of Major Armed Conflicts, 1999-2008” SIPRI Yearbook 2009 (Oxford: Oxford University Press, 2009), p. 71. In fact, since the end of the Cold War Africa has had more conflicts than any other region. See Lotta Harbom and Peter Wallensteen, “Patterns of Major Armed Conflicts, 1990-2001” SIPRI Yearbook 2002 (Oxford: Oxford University Press, 2002) p.65, 11. Dennis M. Tull and Andreas Mehler, “The Hidden Cost of PowerSharing: Reproducing Insurgent Violence in Africa, African Affairs, vol. 104, issue 416 (2005), pp. 383, 384. 12. African political systems have long been characterized in terms of their weakness. For appropriate descriptions, see Michael Bratton, “Beyond the State: Civil Society and Associational Life in Africa,” World Politics, vol. 41, no. 3 (1989), p 410. Patrick Chabal describes the African state as “politically feeble” and a “vacuous shell.” Patrick Chabal, “Violence, Power and Rationality: A Political Analysis of Conflict in Contemporary Africa,” in Patrick Chabal, Ulf Engel, and Anna-Maria Gentili eds., Is Violence Inevitable in Africa? (Leiden: Brill, 2005), p. 6. 13. I. William Zartman, “Introduction: Posing the Problem of State Collapse,” Collapsed States: The Disintegration and Restoration of Legitimate Authority , ed. I. William Zartman (Boulder: Lynne Rienner, 1995), p. 1. 14. Stockholm International Peace Research Institute (SIPRI), Yearbook 1999: Armaments, Disarmaments and International Security, 20, 24-25. Yearbook 2000, p. 17. 15. Ibrahim Elbadawi and Nicholas Sambanis, “Why Are There So Many Civil Wars in Africa? Understanding and Preventing Violent Conflict,” Journal of African Economies, vol. 9, no. 3 (2000), p. 245. Italics added. 16. In 1989, Herman Cohen, the Assistant Secretary of State for Africa, stated that “Sure, the MPLA is the government, it’s in the UN, it’s been recognized by the OAU (by one vote by the way). But we really see Angola as two governments at the present time. We look at it in terms of what exists on the ground. There are two governments there and the big question for making peace is how to merge those two.” Herman J. Cohen, “Forging a Bipartisan Policy,” Interview by Margaret A. Novicki, Africa Report, September-October 1989, pp. 10-13. 17. Margaret Joan Anstee, Orphan of the Cold War: The Inside Story of the Collapse of the Angolan Peace Process, 1992-3 (New York: St. Martins’ Press, 1996). 18. Peacekeeping countries are rarely able to provide the numbers of troops necessary to maintain order in a large area. New York City has approximately

The Search for Security in Africa

33

37,000 police officers; the UN mission to the Democratic Republic of the Congo, which was the most expensive in the world, consisted of fewer than 19,000 peacekeepers. See Somini Sengupta, “Warring Militias in Congo Test U.N. Enforcement Role,” New York Times, April 11, 2004, p. 8. 19. In an indictment of the international community, Ugandan President Yoweri Museveni observed that Western nations do not have the stomach for bloody African wars, wondering aloud “Which U.N. troops will stay in these mountains for six months? They will just run away like they did in Rwanda. European soldiers go only to areas where there is no death.” Cited in Donald G. McNeil, “Bombing Won in Kosovo, Africa Is a Tougher Case,” New York Times, July 25, 1999, sec. 4, p. 16. 20. An example of this phenomenon is the case of Pauline Nyiramasuhuko, a Tutsi official responsible for women’s affairs in Rwanda, who adopted a Hutu identity at the time of the Rwandan genocide and participated in the killing spree against Tutsi. See Peter Landesman, “A Woman’s Work,” New York Times Magazine, September 15, 2002. 21. Snyder and Jervis, “Civil War and the Security Dilemma,” in Civil Wars, Insecurity and Intervention, Barbara F. Walter and Jack Snyder eds. (New York: Columbia University Press, 1999), p. 22. 22. See Suthaharan Nadarajah and Dhananjayan Sriskandarajah, “Liberation Struggle or Terrorism? The Politics of Naming the LTTE,” Third World Quarterly, vol. 26, no. 1, (2005), pp 90-92. Newbury also notes the “development and propagation of a corporate view of ethnicity” whereby individuals are seen in terms of their group membership rather than in terms of the characteristics as individuals. See Newbury, “Ethnicity and the Politics of History in Rwanda” Africa Today, vol. 45, no. 1 (1998), pp. 7-27. 23. See Gérard Prunier, Darfur: The Ambiguous Genocide (Ithaca: Cornell University Press, 2005), p. 103. 24. See John Garnett, “The Causes of War and the Conditions for Peace” in John Baylis, James Wirtz, Eliot Cohen, Colin S. Gray, eds., Strategy in the Contemporary World: An Introduction to Strategic Studies (Oxford: Oxford University Press, 2002), p. 82. 25. Stephen Lewis, “I Too Accuse Him of Crimes Against Humanity,” Globe and Mail, July 28, 2008, p. A13. For a contrasting view of the LRA, see James Bevan, “The Myth of Madness: Cold Rationality and ‘Resource’ Plunder by the Lord’s Resistance Army,” Civil Wars, vol. 9, no. 4, (2007), pp. 343-358. 26. Danny Hoffman, “The Civilian Target in Sierra Leone and Liberia: Political Power, Military Strategy, and Humanitarian Intervention,” African Affairs, vol. 103, issue 411 (2004), pp. 211-226. 27. Kissinger, Years of Renewal, p. 805. 28. Brian Job, “The Insecurity Dilemma: National, Regime and State Securities in the Third World,” The Insecurity Dilemma: National Security of Third World States, ed. Brian Job (Boulder: Lynne Rienner, 1992), p. 29. 29. Job, “Insecurity Dilemma,” pp. 27-28. 30. Another body of literature which relies heavily on the assumption of rationality is the so-called greed-based approach which looks to economic explanations of material gain as a motivation for violence. See, for example, David Keen, The Economic Functions of Violence Adelphi Paper 320 (New York: Oxford University Press, 1998).

34

Civil War in African States

31. In the game-theory scenario of the prisoner’s dilemma, two suspects in a crime are being held in separate locations by a police officer. The officer presents a choice to the captives: confess or do not confess. The officer states that he has insufficient evidence to convict either suspect of the major crime and can only charge them with minor crimes if they both refuse to confess. But the officer also points out that if one of the prisoners confesses, the confessor will go free and get a large reward, while the other prisoner will be hanged. If both suspects confess on the same day, they will each receive long prison sentences. The scenario demonstrates how the incentives to betray one’s partner-in-crime are greater than those that flow from cooperation. 32. A few scholars consider the strategies of different parties to a conflict. In some cases, the disputants themselves spell out their strategies. See, for example, Yoweri Museveni, “The Progress of the People’s War,” in Yoweri Museveni: Selected Articles of the Uganda Resistance War (Kampala: NRM Publications 1985). See also Alemseged Abbay, Identity Jilted or Reimagining Identity (Red Sea Press, 1998); Assis Malaquias, “Diamonds are a Guerrilla’s Best Friend: the Impact of Illicit Wealth on Insurgency Strategy,” Third World Quarterly, vol. 22, no. 3 (2001), pp. 311-325; Osita Afoaku, “Congo’s Rebels: Their Origins, Motivations, and Strategies,” and Kevin C. Dunn, “A Survival Guide to Kinshasa: Lessons of the Father Passed Down to the Son,” both in John F. Clark, The African Stakes of the Congo War (New York: Palgrave Macmillan, 2002), pp. 109-128; Barry R. Posen, “The War for Kosovo: Serbia’s Political-Military Strategy,” International Security, vol. 24, no. 4 (Spring 2000), pp. 39-84. 33. Donald Rothchild and Philip G. Roeder, “Dilemmas of State-Building in Divided Societies,” in Philip G. Roeder and Donald Rothchild (eds.) Sustainable Peace: Power and Democracy After Civil Wars (Ithaca: Cornell University Press, 2005), p. 8. See also, Arend Lijphart, Democracy in Plural Socieites: A Comparative Exploration (New Haven: Yale University Press, 1977), chapter 2. 34. Richard K. Betts, “The Delusion of Impartial Intervention,” Foreign Affairs, vol. 73, no. 6 (November-December 1994), p. 21. 35. Arthur Lewis, Politics in West Africa (London: Oxford University Press 1965), p. 72. 36. Indeed, there can be ambivalence associated with including certain individuals in a power-sharing government. Observers questioned the inclusion of Sierra Leonean rebel Foday Sankoh in a July 1999 power-sharing agreement signed in Lomé Togo. The United States, writes Allister Sparks, “supported a deal that brought Sierra Leone’s psychopathic Foday Sankoh, leader of the [Revolutionary United] Front, into a ‘government of national unity’ and gave him control of the country’s mineral resources, even as Sankoh’s men were drugging child soldiers and chopping off the hands and feet of ordinary citizens.” From Sparks’s perspective, it was self-evident that this was an objectionable approach to conflict resolution though, for many at the time, it was greeted as a success. See “A View of Rome from the Provinces,” The Wilson Quarterly, vol. xxv, no. 2 (Spring 2001), p. 49. 37. Power-sharing arrangements can have varying degrees of cooperation and autonomy. Matthew Hoddie and Caroline Hartzell, for example, refer to federalism as “territorial power sharing.” See their chapter in Sustainable

The Search for Security in Africa

35

Peace: Power and Democracy After Civil Wars, Ed. Philip G. Roeder and Donald Rothchild (Ithaca: Cornell University Press, 2005). 38. See Donald Rothchild and Michael W. Foley, “African States and the Politics of Inclusive Coalitions,” The Precarious Balance: State and Society in Africa, eds. Donald Rothchild and Naomi Chazen (Boulder: Westview 1988); Robert H. Jackson and Carl G. Rosberg, Personal Rule in Black Africa: Prince, Autocrat, Prophet, Tyrant (Berkeley: University of California Press, 1982). 39. Arend Lijphart, “Consociational Democracy,” World Politics, vol. 21, no. 2 (1969), p. 212; italics in original. 40. Betts, “The Delusion of Impartial Intervention,” p. 23. 41. Chaim Kaufmann, “‘See No Evil: Why America Doesn’t Stop Genocide,’ Review of: Samantha Powers, ‘A Problem from Hell’: America and the Age of Genocide” (New York: Basic Books 2002), in Foreign Affairs, 81, no. 4 (2002), p. 145. 42. Cited in Barbara F. Walter, ‘The Critical Barrier to Civil War Settlement,’ International Organization, vol. 51, no. 3, (1997), p. 351. 43. See Celia W. Dugger, “Zimbabwe Opposition Leader Insists Any Deal Give Him Real Power, New York Times, August 17, 2008, Sec. 1, p. 6. 44. Lustick, “Stability in Deeply Divided Societies,” pp. 335-336. 45. Isaac Chotiner, “Appeasement’s Taint Is All in Hindsight,” New York Times, May 25, 2008. 46. See Charles L. Glaser, “Realists as Optimists: Cooperation as SelfHelp,” International Security, vol. 19, no. 3 (Winter 1994-95), pp. 50-90, especially, p. 60. 47. Jama Mohamed Ghalib, The Cost of Dictatorship: The Somali Experience (New York: Lillian Barber Press, 1995), p. 185. 48. See, for example, Steven Erlanger, “Hamas’s Radical New Strategy: Taking Part in Elections,” New York Times, January 2, 2005. 49. Tull and Mehler, “The Hidden Cost of Power-Sharing,” p. 388. 50. See Tull and Mehler, “Hidden Cost of Power-sharing.” Similarly, Englebert and Hummel write that “In Africa, political violence usually provides the means of fighting for (re)insertion into the system by marginalized and excluded groups. It does not represent attempts to challenge, reform, revolutionize, or break away from the state.” See “Let’s Stick Together: Understanding Africa’s Secessionist Deficit,” African Affairs, vol. 104, issue 416 (2005), p. 417. 51. See Jeffrey Herbst, States and Power in Africa: Comparative Lessons in Authority and Control (Princeton: Princeton University Press, 2000), p. 66. 52. IRIN Interview with Major General Muhammad Gani, Integrated Regional Information Network, July, 2000. 53. Edward Luttwak, “Steering into Civil War,” Globe and Mail, February 14, 2005, p. A17. According to research conducted by Daniel Kahneman and Jonathan Renshon, leaders disabuse themselves of the assumption that violence does not solve anything. Being a “hawk,” they say, is more often regarded as the preferred option among policymakers. Their work suggests that decision-makers are psychologically biased in favor of conflict rather than concession. See “Why Hawks Win,” Foreign Policy (January/February 2007), pp. 34-38. 54. Barbara F. Walter, “The Critical Barrier to Civil War Settlement,” International Organization, vol. 51, no. 3 (Summer 1997), p. 335.

36

Civil War in African States

55. John J. Mearsheimer, The Tragedy of Great Power Politics (New York: Norton, 2001), p. 3. 56. Kissinger, Does America Need a Foreign Policy: Toward a Diplomacy for the Twenty-First Century (London: Free Press, 2002), p. 290. 57. See Lydia Polgreen, “Justifying A Costly War In Sri Lanka,” New York Times, July 19, 2009, p. 6. 58. Ironically, perhaps, the report argued that since it was so difficult to create institutions that did not favor one sect, the end result was more likely partition rather than domination. David Brooks, “A Million Little Pieces,” New York Times, June 5, 2007. 59. Edward Wong, “Iraq’s Curse: A Thirst for a Final Crushing Victory,” New York Times, June 3, 2007. 60. Snyder and Jervis, “Civil War and the Security Dilemma,” pp. 29-30. 61. The term “akazu” means Little House and refers to the extremist members associated with the Hutu regime. See Bruce D. Jones, “Civil War, the Peace Process, and Genocide in Rwanda,” in Taisier M. Ali and Robert O. Matthews eds., Civil Wars in Africa: Roots and Resolution (Montreal and Kingston: McGill-Queen’s, 1999), pp. 60, 78. 62. Francis Fukuyama, “Women and the Evolution of World Politics,” Foreign Affairs, vol. 77, no. 5 (September/October, 1998), p. 36. 63. “Angola’s Elusive Oil Riches,” New York Times, (editorial) June 15, 2004, p. A22. 64. Luttwak, “Give War a Chance,” Foreign Affairs, vol. 78, no. 4 (July/August 1999), p. 36. 65. See International Crisis Group, “Sierra Leone: Time for a New Military and Political Strategy,” April 11, 2001, p. iii. 66. Edward Wong, “Iraq’s Curse: A Thirst for a Final Crushing Victory,” New York Times, June 3, 2007, sec. 4, p. 3. 67. In Somaliland, as we shall see, the opposite was also true. The government of Ibrahim Egal acquired national armed forces because he thought it would enhance the likelihood that Somaliland would receive international recognition. 68. Robert Jackson, and Carl G. Rosberg, “Sovereignty and Underdevelopment: Juridical Statehood in the African Crisis,” Journal of Modern African Studies, vol. 24, no. 1 (1986), p. 2. 69. See for example, Nicholas Sambanis, “Partition as a Solution to Ethnic War: An Empirical Critique of the Theoretical Literature,” World Politics, 52, no. 4 (2000), pp. 437-483; Radha Kumar, “The Troubled History of Partition,” Foreign Affairs, 76, no. 1, (1997), pp. 22-34; I. William Zartman, “Posing the Problem of State Collapse,” in Collapsed States; Paul Collier, “Economic Causes of Civil Conflict and Their Implications for Policy,” and Mohammed Ayoob, “State Making, State Breaking, and State Failure,” both in Turbulent Peace: The Challenges of Managing International Conflict, eds. Chester A. Crocker, Fen Osler Hampson, and Pamela Aall (Washington: United States Institute of Peace 2001). 70. Donald Horowitz ‘The Cracked Foundations of the Right to Secede,’ Journal of Democracy, vol. 14, no. 2 (2003), pp. 5-17. 71. Walter, “Civil War Settlement,” p. 355.

The Search for Security in Africa

37

72. Englebert and Hummel, “Let’s Stick Together,” pp. 400 and 402. While they are somewhat vague on the number, the authors cite Eritrea’s secessionist war against Ethiopia; Katanga and South Kasai in Congo; Biafra in Nigeria; Casamance in Senegal; southern Sudan; Cabinda in Angola; as well as the activities of liberation movements in Mali, Niger and the Comoros. See their chart on p. 401. 73. Article 3 (b) says that the “Objective of the Union shall be to: ... defend the sovereignty, territorial integrity and independence of its Member states.” Article 4 (b) says that the “Union shall function in accordance with the following principles: ... Respect for borders on achievement of independence.” 74. See for example, Laurence Piper, “Nationalism Without a Nation: the Rise and Fall of Zulu Nationalism in South Africa’s Transition to Democracy, 1975-99,” Nations and Nationalism, vol. 8, no. 1 (2002), pp. 73-94. The call for secession was most explicitly made by Zulu King Goodwill Zwelithini. While it was publicly denounced by Inkatha Freedom Party leader Mangosuthu Buthelezi, his nationalist rhetoric was seen as an attempt to draw attention to his movement and allow him a more prominent role in the post-apartheid political process. See also the discussion of various Somaliland clan factions in post-Siad Barre Somalia in chapter 3. 75. Ayoob, “State Making, State Breaking, and State Failure,” p. 138. 76. Taisier M. Ali and Robert O. Matthews, “Conclusion: The Long and Difficult Road to Peace,” in Taisier M. Ali and Robert O. Matthews eds., Durable Peace: Challenges for Peacebuilding in Africa (Toronto: University of Toronto Press, 2004), p. 399. 77. Englebert and Hummel, “Let’s Stick Together,” p. 417. Donald Horowitz also concludes that the infrequency of successful secessions can be attributable not to the legitimacy of contemporary borders or the efficacy of conflict resolution but to the inherent weakness of many secessionist movements relative to the strength and determination of the central governments which seek to defeat them. Horowitz, Ethnic Groups in Conflict (Berkeley, University of California Press, 1985), p. 265. 78. Steven R. David, Third World Coups d'Etat and International Security (Baltimore: Johns Hopkins University, 1987), pp. 79-84. 79. As Herbert Howe has observed, the supplies of weapons are difficult to monitor but have likely involved transfers of several million weapons. See Herbert Howe, Ambiguous Order, chapter 3. 80. Posen, “Security Dilemma and Ethnic Conflict,” Survival, 35, no. 1 (1993), p. 30. 81. Herbert Howe, Ambiguous Order: Military Forces in African States (Boulder: Lynne Rienner, 2001), p. 88. 82. Snyder and Jervis, “Civil War and the Security Dilemma,” p. 26. 83. See Robert D. Kaplan, “History’s Cauldron” Atlantic Monthly, June 1991, 93-104; “A Reader’s Guide to the Balkans,” New York Times Book Review, April 18 1993, pp 1, 30-32; and Balkan Ghosts: A Journey Through History. For criticisms of this view, see Noel Malcolm, “Seeing Ghosts,” National Interest (Summer 1993), pp. 83-88; and V. P. Gagnon, “Ethnic Nationalism and International Conflict: The Case of Serbia,” International Security, vol. 19, no. 3 (Winter 1994/95), pp. 133-134. 84. Englebert and Hummel, “Let’s Stick Together,” p. 418.

38

Civil War in African States

85. Margery Perham writes that, at the time of writing, Ethiopian emperor Haile Selassie claimed “the whole of the modern Eritrea and also Italian Somalia as ‘lost provinces’ . . . . The claim is based . . . upon some rather indefinite references to early history and migrations, almost every sentence of which cries out for comment or correction.” See The Government of Ethiopia (London: Faber and Faber, 1969), pp. 480-481. 86. Posen, “The Security Dilemma and Ethnic,” 30-31; Snyder and Jervis, “Civil War and the Security Dilemma,” p. 26; Stephen Van Evera, “Hypotheses on Nationalism and War,” International Security, vol. 18, no. 4 (Spring 1994), pp. 23 and 25. 87. Posen, “Security Dilemma and Ethnic Conflict,” p. 31. 88. On this point, see Snyder and Jervis, “Civil War and the Security Dilemma.” They argue that a perceptual element is crucial to a complete explanation of the security dilemma. See also John J. Mearsheimer, “Reckless States and Realism,” International Relations, vol. 23, no. 2 (2009), p. 247. 89. Paul Collier, “Economic Causes of Civil Conflict” in Crocker, Hampson, Aall, Turbulent Peace, p. 152. 90. Notably, Horowitz downplays the significance of wealth as an economic motivator for secession. “Economic loss or gain plays the smallest role,” he says, “ethnic anxiety the largest.” Donald Horowitz, Ethnic Groups in Conflict, p. 259 and pp. 249-250. 91. Collier cites Eritrea as a region with twice the per capita income of the rest of Ethiopia. Collier, “Economic Causes of Civil Conflict,” p. 152. 92. Tony Hodges, Angola: Anatomy of an Oil State (Oxford: James Currey, 2004), p. 26. While Angola’s diamond wealth is distributed in a number of regions in Angola, it was generally beyond the territory encompassed by “Savimbiland.” 93. Donald Horowitz, Ethnic Groups in Conflict, p. 236. 94. See Graham Bowley, “Declaring Something a Lot Like Independence,” New York Times, March 2, 2008, sec. 4, p. 1. 95. This tendency is noted by Henry Kissinger, Years of Renewal, p. 944. 96. Ted Robert Gurr, “Minorities and Nationalists: Managing Ethnopolitical Conflict in the New Century,” in Turbulent Peace: The Challenges of Managing International Conflict, Chester A. Crocker, Fen Osler Hampson, and Pamela Aall, eds. (Washington: United States Institute of Peace, 2001), p. 164. 97. Van Evera, “Hypotheses on Nationalism and War,” p. 17. 98. Posen, “The Security Dilemma and Ethnic Conflict,” p. 30. 99. Horowitz, Ethnic Groups in Conflict, p. 267. 100. Tull and Mehler argue, for example, that while Africa’s incidence of violence—the highest in the world—“should not be attributed solely to outside actors, it remains nonetheless true that the foreign policies of Western countries continue to have an appreciable impact on the political processes on the continent.” See “The Hidden Cost of Power-Sharing,” p. 386. 101. That is, at least until they have a realistic chance of achieving power in government. Some scholarship has pointed to the fact that rebel movements will undertake the most heinous acts of cruelty in order to draw in the international community and benefit from the vast resources they provide in their efforts to

The Search for Security in Africa

39

alleviate suffering. See for example, Danny Hoffman, “The Civilian Target in Sierra Leone and Liberia”. 102. South Africa provided various forms of support against regimes in Congo, Mozambique, and Angola. 103. This is the general argument of Tull and Mehler, in “Hidden Cost of Power-Sharing.” 104. See, for example, René Lemarchand, “Consociationalism and PowerSharing in Africa: Rwanda, Burundi, and the Democratic Republic of the Congo,” African Affairs, vol. 106, no. 422, p. 11, footnote 22. 105. Lijphart argues that in all of the consociational democracies “the cartel of elites was either initiated or greatly strengthened during times of international crisis, especially the First and Second World Wars.” See Arend Lijphart, “Consociational Democracy,” World Politics, vol. 21, no. 2 (January 1969), p. 217. 106. Jack Snyder, “One World, Rival Theories,” Foreign Policy (November/December 2004), p. 58. 107. Horowitz, Ethnic Groups in Conflict, p. 240. 108. Douglas Hurd before the Royal Institute of International Affairs, London, June 1993. Cited in Baylis and Smith, Globalization of World Politics, 1st edition, p. 367. 109. Chester Crocker, “Intervention: Toward Best Practices and a Holistic View, in Turbulent Peace, Chester A. Crocker, Fen Osler Hampson, and Pamela Aall, eds. (Washington: United States Institute of Peace, 2001), p. 238. More recently, in 2006 Senator Joseph Biden and Leslie Gelb proposed a plan to decentralize Iraq by “giving each ethno-religious group ... room to run its own affairs.” Referred to by others as the “Biden plan” or “the partition” (because it cited Bosnia as an example), President George W. Bush claimed that such a plan would be “like pouring oil on fire” and was, accordingly, “not even a starter.” Biden was forced to clarify his views and later denied that his intention was partition. See Joseph R. Biden Jr. And Leslie H. Gelb, “Unity Through Autonomy in Iraq,” New York Times, May 1, 2006 and Peter Wallsten and Paul Richter, “Bush Dismisses the Idea of Partitioning Iraq,” Los Angeles Times, August 16, 2006, p. A12. 110. Mark Zacher notes that the “1963 OAU Charter contains a strong article in support of territorial integrity (Article 3), but a much more specific statement was adopted by the Assembly of Heads of State and Government in 1964 after both Morocco and Somalia had launched wars of territorial revision against neighboring states. All member states except Morocco and Somalia approved a resolution calling on members ‘to respect the borders on the achievement of national independence.’” See Mark W. Zacher, “The Territorial Integrity Norm: International Boundaries and the Use of Force,” International Organization, vol. 55, no. 2 (Spring 2001), p. 222. 111. See, for example, the Communique of the 139th Meeting of the Peace and Security Council of the African Union, 29 June 2008, Sharm el Sheikh, Egypt. 112. Englebert and Hummel write: “It can plausibly be argued ... that the OAU set rules of territorial integrity that were more stringent than elsewhere. Yet, the continent-wide nature of these rules fails to account for the few actual instances of African separatism. In addition, the incapacity of weak African

40

Civil War in African States

states to enforce them suggests that they may not per se be an impediment to separatist action.” See “Let’s Stick Together,” 412, footnote, p. 32. 113. See Zacher, “The Territorial Integrity Norm,” pp. 215-250. 114. Communique, 138th Meeting of the Peace and Security Council (of the AU), June 29, 2008, Sharm el-Sheikh, Egypt. 115. By 1995, the RPF had already lost key members of their organization—the prime minister and the minister of the interior being two—who subsequently made allegations about the government’s abuses of power and its violation of human rights. See Filip Reyntjens, “Rwanda, Ten Years On: From Genocide to Dictatorship,” African Affairs, vol. 103, issue. 411, p. 178. 116. The most prominent discussion of this issue is presented in Mats Berdal and David M. Malone eds. Greed and Grievance: Economic Agendas in Civil Wars (Boulder: Lynne Rienner, 2000). 117. Jeffrey Herbst, “Economic Incentives, Natural Resources and Conflict in Africa,” Journal of African Economies, vol. 9, no. 3 (2000), pp. 270-294. 118. Jonas Savimbi’s successor, Isaias Samakuva, emphasized how Savimbi “would punish people who take everything for themselves and not look after the inferiors.” According to Samakuva, “He would not contemplate his generals living nicely and his soldiers starving.” Interview with the author, July 14, 2004. 119. Barbara Crossette, “Singling Out Sierra Leone, U.N. Council Sets Gem Ban,” New York Times, July 6, 2000. 120. See Charles Cater, “The Political Economy of Conflict and UN Intervention: Rethinking the Critical Cases of Africa” in The Political Economy of Armed Conflict: Beyond Greed and Grievance, Karen Ballentine & Jake Sherman eds. (Boulder, CO: Lynne Rienner, 2003), p. 29. Regarding the involvement of Liberia’s President Charles Taylor in the same conflict, a report published by the International Crisis Group also noted that “President Taylor is not just interested in money and diamonds. As one senior Liberian commentator put it, ‘he’s in Sierra Leone not for the money but for his political agenda.’” International Crisis Group, Sierra Leone: Time for a New Political and Military Strategy, April 11, 2001, p. 13. 121. Cited in John G. Stoessinger, Henry Kissinger: The Anguish of Power (New York: W.W. Norton, 1976), p. 68. 122. Cited in James Traub, “Taunting the Bear,” New York Times, August 10, 2008, Sec. 4, p. 1.

2 The Wars of Ethiopia and Eritrea

We think the Derg is too weak to be a partner in serious negotiations. . . . The Derg has to be eliminated. We cannot compromise with it. We have to go ahead with our fight to liberate the country. TPLF leader Meles Zenawi We are not fighting for a change of regime but for independence. EPLF spokesman Andemikael Kahsai. Everything to the war front. Ethiopian President Mengistu Haile Mariam In his memoir of the Cold War-post-Cold War transition in Ethiopia, the American Assistant Secretary of State, Herman Cohen, wrote about the “many players, [and] many agendas” that influenced the Ethiopian scene—an appropriate descriptor given that Ethiopia did have many players who employed a complex mix of strategies in their respective searches for security.1 I will focus largely on three of these players: the Ethiopian government, led by Emperor Haile Selassie and, subsequently, by President Mengistu Haile Mariam, and two rebel movements, the Eritrean People’s Liberation Front (EPLF) and the Tigrayan People’s Liberation Front (TPLF). The Ethiopian government pursued a strategy of domination throughout, though with varying degrees of success and, particularly in its last days, accompanied by limited gestures of accommodation. Both the EPLF and the TPLF also sought to achieve local dominance at the beginning of their struggles; however they subsequently pursued contrary strategies of, respectively, secession and political domination. A principal question of this chapter is why, despite the fact that both rebel movements faced the same threat in the form of a coercive regime,

42

Civil War in African States

each elected a different strategy. This strategic divergence is even more remarkable given that the insurgencies operated out of contiguous northern provinces which were populated in part by ethnically similar people. What circumstances led to the EPLF’s single-minded embrace of political independence from Ethiopia? Equally, why did the TPLF, in neighboring Tigray province, reject secession and opt instead for political domination? To what extent did their respective strategies succeed in realizing their respective security objectives? An understanding of the circumstances in both of these provinces will reveal that the strategies adopted were not random or incidental outcomes. Rather they were consequences of the respective historical origins of the principal disputants, as well as demographic and balance of power factors. Significantly, the analysis will reveal how these strategies, while ultimately securing the survival of the two movements, were in conflict with each other and with a post-Cold War environment that sought to reconcile democratic rule and political stability. It is also no small irony that, in the end, the single largest threat to these one-time allies came from each other. The Origins of Eritrea’s Secessionism

Ethiopia’s claim to Eritrea was based largely on the existence of the long-standing empire of Axum, which was centered in what is now Tigray province and extended north to the coast of the Red Sea, encompassing what is now Eritrea. The empire was eventually brought down in 970 A.D.2 through a combination of foreign encroachment and internal instability. Over the next centuries both the seaboard and the interior were subject to foreign control and erratic and fragmented rule. More consistent authority was achieved in the interior in the middle of the nineteenth century when Tewodros II, one of a line of Ethiopian emperors claiming to be descendent of Solomon, King of Israel, consolidated sufficient power to constitute a viable state that would become the basis for modern Ethiopia.3 The Red Sea coast developed strategic and commercial importance with the opening of the Suez Canal in 1869. Britain had occupied Egypt in 1872 as a means of ensuring that control of the Nile was not subject to the uncertainties of Egyptian politics. But Britain was also eager to keep France from gaining control of the Nile. While it was prepared to stand watch of the White Nile, it needed a means to protect the Blue Nile and its Ethiopian source and tributaries. Provided the newly united state of Italy was willing to serve the British interest, Britain willingly colluded with Rome in its efforts to acquire a colony in Ethiopia. For its part,

The Wars of Ethiopia and Eritrea

43

Ethiopian policy encouraged limited European penetration in exchange for armaments and, it was hoped, European assistance in times of trouble. Eventually, in 1889, Italy signed the Treaty of Wich’al_ (Uccialli) with Menelik II, King of Showa, who ceded what is now Eritrea in the expectation that Italy would in turn support his bid for the emperorship of Ethiopia. Subsequently, Italy sought to expand its presence southward beyond that demarcated by the treaty but was prevented from doing so when it was defeated at Adwa in 1896 by an army under Menelik. While the Italians retrenched northward to Eritrea, Menelik continued to expand the Ethiopian empire south into regions inhabited by Oromo and Somali peoples. Today a visitor to Eritrea’s capital, Asmara, can still see the impact of Italian colonialism: its architecture and its substantial road, port, and railway infrastructures. Eritrea also acquired a significant European population, growing from 4,188 Italians in 1931 to nearly 75,000 in 1939.4 Mussolini envisioned Asmara as a showpiece for a new Roman empire whereby the entire Horn of Africa would come under Italian authority. Eritrea would later serve as a staging ground for Italy’s second—and this time successful—invasion of Ethiopia in 1935. Abandoned by its European allies, and its Emperor, Haile Selassie, deposed and living in England, Ethiopia’s fate was left in the League of Nations’ pledge of collective security. More interested in securing Italy’s compliance against Hitler’s Germany, however, the League’s great powers leveled sanctions against Italy but refused to dislodge the invaders. Not until Italian forces were driven out by a joint BritishEthiopian force in 1941 and Ethiopia’s administration overtaken by the British, was Rome’s claim on Eritrea withdrawn. Emperor Haile Selassie returned to his throne convinced that Ethiopian control of Eritrea was central to his country’s security.5 To be sure, the creation of a new political territory in Eritrea, temporary as it had been, had important consequences for the region’s future insofar as it formalized juridical unity in a demographically diverse territory. The result was an Eritrean nationalism that was contradicted by demographics and Ethiopian sentiment. Not only had Italian colonization brought nine ethnic groups together, but it also integrated Christians in the region’s highlands with Muslims from its lowlands. The border also divided people who, heretofore, had a common language and cultural links: speakers of Tigrinya inhabited both sides of the Ethiopian-Eritrean border. Ethiopians felt that Eritrea had no basis for independence or common identity other than the fact that it had been colonized by the

44

Civil War in African States

Italians, a rogue chapter in a longer history of Ethiopian-Eritrean unity.6 While the Italian defeat at Adwa in 1896 had created its Ethiopian military heroes and ensured the country’s independence from European colonialism, it was a pyrrhic victory insofar as it led to new and unexpected threats to Ethiopia’s integrity. “[I]f Italy had won the battle and occupied all of Ethiopia,” wrote Dawit Wolde Giorgis, a one-time minister in the regime of Mengistu Haile Mariam, “there would be no Eritrean problem today. The entire region, including Italian Somaliland, though conquered, would have been united and after decolonization might have emerged as Greater Ethiopia.”7 Others saw only a tenuous connection between Eritrea and Ethiopia, claiming that, though the Axumite Kingdom preceded the Ethiopian Empire, it did not give birth to it and that, since then, Ethiopia had only briefly controlled Eritrean territory in the years from 1880 to 1885. Dan Connell, a sympathetic journalist who chronicled the modern Eritrean struggle for liberation from Ethiopia, wrote that “To assert an identity between the ancient Axumite empire and the modern Ethiopian state is . . . a bit like trying to argue an Italian claim to modern France based on the conquest of Gaul by the long defunct Roman Empire.”8 Italian colonialism is also regarded by Eritreans more positively than by Ethiopians insofar that it generated a development trajectory more advanced than that of their Ethiopian counterparts. “We look at colonialism more favorably than other peoples in Africa and Asia,” observed Yemane Ghebreab, the political affairs officer for Eritrea’s current ruling party.9 For Eritreans, their nation was not one of the nine ethnic groups that inhabit Eritrean territory; rather it was an Eritrean collectivity whose origins could be found in the common experience of Italian colonialism. While colonial borders are often given as a source of Africa’s contemporary troubles, Eritreans embrace them because they are the country’s raison d’etre. These diverging historical narratives informed perspectives on Eritrea’s future status when this question was taken up by the United Nations at the end of World War II. In 1949, when a Four Power Commission failed to agree on what should be done with the former Italian colony, the UN established another Commission for Eritrea which was asked to make recommendations in light of the following criteria: the interests of the parties within Eritrea, the capacity of the people to rule themselves, the interests of peace and security, and the rights and claims of Ethiopia. The Commission’s delegations—representing Burma, Guatemala, Norway, Pakistan, and the Union of South Africa—took different messages from what they

The Wars of Ethiopia and Eritrea

45

heard during these consultations and, not surprisingly, arrived at different recommendations regarding Eritrea’s future. The Unionist Party, the most prominent group that was heard from in favor of integration with Ethiopia, informed the commission that, as well as historical, geographical, and economic links between the two regions, integration was the course that most Eritreans wanted. Other pro-integration groups and parties maintained that a separate Eritrea would not actually be independent because it was too poor to rule itself and was therefore vulnerable to infiltration and control by Italy. As far as the pro-unionist movement was concerned, the commission recorded, real independence for Eritrea “could be obtained only in union with Ethiopia.”10 The Independence Bloc and its allies, by contrast, claimed that its views were supported by the majority of Eritreans, that Eritreans were indeed capable of ruling themselves, that Eritrea had never been part of the modern Ethiopian empire, and that Eritrea’s Italian community was not to be feared. On the contrary, one of the delegations recorded, Italians were Eritrea’s economic lifeblood, and their exodus following union with Ethiopia would “spell economic disaster.” According to this view, Eritrea already had “higher social, economic and administrative standards” largely as a result of its distinctive colonial history, and “annexation to or federation with Ethiopia would be a retrograde step.”11 The commission was also sensitive to the interests of Ethiopia itself. Haile Selassie was worried about the prospect of an independent and perhaps vulnerable Eritrea becoming a staging ground for a future assault on Ethiopia. Eritrea had little inherent value or logic of its own—it was too fragmented politically, economically, and ethnically—except as a base from which to attack its southern neighbor.12 In the words of Kennedy Trevaskis, who served as a British colonial officer in Eritrea at the time, “experience had taught [Ethiopia] that, unless she had the key to her own front door, she would never be secure.”13 Consequently, Haile Selassie believed that Ethiopia’s independence would best be maintained if Eritrea were under his control. In this view, the emperor also had support from members of the commission who, concerned with Eritrean irredentism, wrote that “the movement for union with Ethiopia has many of the characteristics of a popular movement, and it is more than likely that outright frustration of these wishes would make the position of internal security in Eritrea untenable.”14 Given that the debate had become so polarized, any decision was bound to provoke dissension. The commission’s delegations from Guatemala and Pakistan concluded that no foreign threat existed, that

46

Civil War in African States

Ethiopia’s control of Eritrea “would not save Ethiopia or safeguard it against an aggression,” and, in any event, that an international guarantee by the United Nations offered Ethiopia sufficient protection. In what now appears to have been its most prophetic statement, the two delegations wrote that “annexation . . . against the will of a large portion of the Eritrean population, would create constant internal friction, giving rise inter alia to police measures of repression . . . which could jeopardize the internal tranquility of Ethiopia.”15 The final decision in 1950 was a carefully crafted compromise which sought to satisfy all interested parties but ultimately reflected the majority opinion in favor of association with Ethiopia. The Federal Act between Ethiopia and Eritrea was a contractual agreement whereby Eritrea would be an autonomous unit with its own constitution and assembly but would be federated with Ethiopia under the Ethiopian Crown. Anze Matienzo, the transitional commissioner for Eritrea who had been appointed by the United Nations, stated that such a pact could not rely on a “mere document . . . to ensure life and continuity for the institutions thus created.” Ethiopia and Eritrea, he said, would have to “learn to live side by side, each respecting the proper sphere of activity and jurisdiction of the other.” The UN commissioner maintained that there were indeed convincing grounds for trust. “My conversations with His Majesty the Emperor of Ethiopia,” the commissioner stated, “have convinced me that such good faith exists and that it is the Emperor’s sincere desire that the Federal Act should be implemented in accordance with both the letter and the spirit of the resolution.” He concluded that, among the Eritreans, he was also “convinced . . . that there is a genuine readiness for full cooperation with the federal authorities and a real respect for the unity of the Federation under the sovereignty of the Emperor.”16 Despite the commissioner’s faith in the respect each party in the federation would show for each other, the next ten years saw the gradual dismantling of the arrangement—undermined by partisans of both union and independence. When Haile Selassie finally dissolved the federation on November 16, 1962, and turned Eritrea into Ethiopia’s fourteenth province, he claimed to be doing so on behalf of the Eritrean people whom, he said, had “repeatedly requested us to abolish the federal system and re-establish a unitary form of administration.”17 While the emperor claimed to have done everything possible to ensure that Eritrea “prospered under our rule,” he in fact never reconciled himself to the federation and worked to undermine its autonomy. He believed that a centralized state—including a national army and a pan-Ethiopian identity that prominently featured the Amharic language and

The Wars of Ethiopia and Eritrea

47

culture—was the most effective counter to narrower nationalist forces within. Haile Selassie was particularly fearful of the threat an autonomous Eritrea posed to a multi-ethnic Ethiopia. The emperor’s predecessors had been preoccupied with the expansion of the Ethiopian empire. Now the emperor linked the survival of his regime to maintaining his country’s territorial integrity. “The history of Ethiopia proves that its boundaries in the past stretched over the Horn of Africa and reached out far into Arabia,” he told a visiting delegation. “It has shrunk because of the designs and the attacks on our country by our enemies and because of the apathy of our people.” It was urgent, the emperor concluded, that “we fight for the whole of it rather than agree to lose it bit by bit.”18 Consequently, Ethiopian security interests and the emperor’s absolutism dominated at the expense of Eritrean autonomy and security. The Eritrean flag was banned in 1958 and the National Assembly was steadily emasculated, though it did have sufficient strength to request that the federation with Ethiopia be terminated in an engineered vote on November 14, 1962. Eventually, Ethiopia established Amharic as the official language of Eritrea, replacing Tigrinya and Arabic, both of which had been guaranteed as official languages in the Eritrean constitution. As John Markakis has written, “The imperial regime never considered making any concession to Eritrean demands, even to ostensibly apolitical ones.”19 These actions did not go unchallenged within Eritrea. Those who believed that the United Nations was legally obligated to safeguard Eritrean autonomy protested Ethiopian transgressions of the federation agreement to the UN Secretary General.20 But these remonstrations went unheeded outside of Ethiopia in large part because of the emperor’s diplomatic lobbying and his exploitation of Washington’s interest in the Kagnew communications facility near Asmara.21 The Ethiopian position was that the international community was not in a position to do more than reprimand the regime. Technically, the abrogation of the federation was a violation only of the spirit of the UN resolution that had created the federation, since, with the implementation of the Federal Act, the UN had divested itself of responsibility for the protection of Eritrean autonomy. The status of the federation subsequently became an internal matter dealt with exclusively between the Eritreans and the Ethiopians. Once a majority of Eritrean legislators had “voted” to dissolve the arrangement, and once this dissolution had been accepted by Ethiopia, the UN had no recourse.22 This failure of the federation to protect Eritrean autonomy would serve as the basis for Eritrean skepticism about future arrangements short of independence.

48

Civil War in African States

The Demise of the Haile Selassie Regime

The pro-independence movement at the time of dissolution was largely controlled by Eritrea’s Muslim population, who had lost the protections they had been afforded under the Eritrean constitution.23 As the Eritrean Liberation Front (ELF)—formed by Eritrean exiles in Cairo in 1960—framed their campaign largely in pan-Muslim terms, Eritrea’s influential Christian highlanders were not initially involved. But Ethiopia’s often clumsy handling of the Eritrean question played into the hands of secessionist elements and undermined support among many of those initially in favor of federation. Eventually many disillusioned and disenfranchised Christians joined the ELF. As an advisor to the emperor, John Spencer observed that “The campaign for an independent Eritrea ... would certainly not have persisted with any effect had it not been substantially aided by over-zealousness in asserting federal powers by the central government at Addis Ababa, abuses of authority and of personal position and by errors of sheer stupidity on the part of federal officers.”24 Indeed, grievances were not directed exclusively at the central government in Addis Ababa. Pro-unionists in Asmara undermined their own cause when they harassed and even arrested their political opponents within Eritrea. Even the emperor’s efforts to control his allies in the unionist leadership backfired. Tedla Bairu, the former unionist leader and head of the Eritrean government under the federation, whom the emperor had exiled to an ambassadorship in Stockholm for his excesses in promoting the unionist cause, moved himself to Damascus where he defected to the independence movement.25 The incipient Eritrean movement was not, however, without problems of its own. Despite its stated intention to “smash the Ethiopian economic and military resources which strengthen the occupation of Eritrea,” the movement was in no position to overthrow the emperor or assert its independence.26 During the 1960s and early 1970s, the ELF struggled with internal bickering, insufficient resources, and a narrow political base made narrower by the perception that it was advancing the interests of Arab states. It managed to mount only a few highjackings and assassinations of government officials—activities that were designed mostly to attract international attention to its cause but which did not reveal a strong and unified organization.27 Within Ethiopia, the emperor’s regime was facing pressures of its own. The highly centralized system of imperial rule was viewed as an anachronism by growing numbers of urban youth who were increasingly radicalized, versed in Marxism, and saw revolution as a solution to the

The Wars of Ethiopia and Eritrea

49

challenges of inequality facing Ethiopia. The Ethiopian government was incapable of responding to these protests and demands for reform from their people, both because of the emperor and because of the system itself. The aging emperor’s now-apparent senility was a central obstacle to meaningful reform, as was his insistence on having the final word in all matters of state. In the absence of political parties or a responsible cabinet, the government was unable to advance the institutional means by which the allegiance of ordinary Ethiopians could be won or through which their frustrations could be channeled. By default, coercion became the principal means of social control. The most damaging evidence of the regime’s ineffectiveness and insensitivity to its own people, however, came in 1973 when graphic images from Jonathan Dimbleby’s film “The Unknown Famine” were publicly juxtaposed with the opulent lifestyle of the emperor and his family in Addis Ababa. When the emperor’s regime was overthrown in 1974, those who were involved did not just resent these conditions as much as they took advantage of them. Soldiers’ grievances over their living conditions took the form of a mutiny nourished by the disenchantment felt across Ethiopia. By the time the military made its move, much of the regime’s authority had been compromised and the emperor’s overthrow was executed with relative ease. The famine had provided evidence of the regime’s incompetence while the students—who were insufficiently organized and united to forge a front of their own—provided popular support for the coup leaders. While news of the emperor’s overthrow was greeted with relief and joy, the military’s own limitations—in particular, its dependence on violence—were soon revealed. The decision to unseat the emperor and install a new military leadership was not, according to one participant, planned in advance. “We continued to meet and discuss change, but no one in the circle I worked with ever mentioned the overthrow of Haile Selassie,” wrote Dawit Wolde Giorgis, a early member of the Derg. “At the time we feared that his influence was so powerful, his myth so strong, that trying to bring him down would lead to bloodshed.”28 Indeed, the decision to remove and eventually kill the emperor and much of his family reflected the paranoia felt by the new leaders about the counter-revolutionary potential of the old regime. “Probably no other issue is more crucial to the survival of the council than what to do with the emperor,” wrote one Western journalist in Addis Ababa at the time.29 Convinced that he would somehow regain his crown were he allowed to live, the coup leaders first proclaimed their “unswerving loyalty” to the emperor, and then moved him out of the palace before announcing that he had been found dead in his bed.30

50

Civil War in African States

The Derg in Power

The incremental acquisition of power by the new rulers—calling themselves the Provisional Military Administrative Council (PMAC) or Derg (Amharic for “Council”)—initially aroused cautious optimism from Eritrea on several grounds. The ELF proclaimed that it “welcomed the statements and intimations of the military despite the vagueness of their intentions towards the Eritrean cause and the implementation of the Front’s demands.”31 The view among student activists, Eritreans, and indeed the regime’s new leaders was that the old feudal regime—particularly the Showan Amhara elite who led it—had alone provoked ethno-nationalist sentiments and that the regime was the sole source of the secessionist movement’s anger. The imperial regime had been, proclaimed the ELF, “first and last . . . responsible for the harsh circumstances under which the people have lived.”32 From the Derg’s perspective, the removal of the feudal regime would arrest the exploitation of peripheral regions and thus induce Eritreans and other nationalist groups to, as the ELF put it, “work together in building a new Ethiopia where the concept of self-determination was a reality.”33 There were other reasons for the ELF’s cautious optimism.34 Despite its name, the new regime’s program of Ethiopia Tikdem (“Ethiopia above all”) initially held out the promise of regional autonomy for the country’s nationalist groups. To be sure, the program stated that “the people inhabiting a particular region will be responsible for ensuring that affairs concerning that region are executed in manners that do not affect national unity and progress.” But it also said that “local matters will be run at the local level.”35 Moreover, regional autonomy appeared to be safeguarded by an absence of the same Abyssinian chauvinism that characterized the imperial regime and, more importantly, by the fact that one of Eritrea’s own sons, General Aman Michael Andom, was soon appointed chairman of the PMAC.36 However, the elevation of the popular and familiar Eritrean, General Aman, to the chairmanship in September 1974 over the other contenders, Colonel Mengistu Haile Mariam and Major Atnafu Abate, obscured not only an underlying power-struggle within the new leadership but also a difference over how to manage the issue of Eritrea. According to one member of the council, General Aman was willing to negotiate directly with the ELF, preferring “a peaceful settlement with the rebels, while Mengistu wanted to crush them militarily.”37 Unable to prevail against this emerging group of hardline elements within the Derg, Aman resigned. Both Aman’s Eritrean sympathies (seen by some as in conflict with Ethiopia tikdem) and his subsequent association with

The Wars of Ethiopia and Eritrea

51

anti-Derg elements were regarded as acts of treason. Government troops were dispatched to Aman’s home on November 23 where the General was killed in an ensuing gun battle. It is unlikely that there was unanimous support for these actions but Aman was regarded by many as an outsider and thus a potential threat to the incipient regime’s survival.38 That evening, in events that were referred to as “Bloody Saturday,” another fifty-nine former government officials and dissidents within the military were summarily executed. Subsequent reports claimed that Mengistu Haile Mariam—the man who would eventually become Ethiopia’s head of state, chairman of the Derg, and the revolution’s “solitary leader”—was one of four council members who stood behind the firing squad to ensure that the executions were carried out.39 In terms of the Eritrean liberation movement, the death of Aman ended any hope that the new government would recognize Eritrea’s political aspirations. Having failed to realize their political aspirations through peaceful accommodation with either the colonial powers, the imperial regime, or the revolutionary Derg, the Eritreans now believed that they had “no other recourse than to take their destiny into their own hands.”40 For Mengistu, the removal of Aman was an important step in the consolidation of power. As an officer in the Ethiopian military, Mengistu had developed a reputation as a trouble-maker and a fighter who did not hesitate to hit first and hard.41 Stories emerged of Mengistu’s brutality towards his own colleagues.42 To the extent that these stories instilled fear and concentrated power in his hands, Mengistu cultivated these sentiments. “Although I condemned Mengistu's methods, I still believed that there was no alternative to his leadership of the Revolution,” wrote Dawit Wolde Giorgis. “Mengistu rose to power despite the fact that everyone knew he was ruthless and cruel.”43 According to David Korn, who served as the American ambassador to Ethiopia from 1982 to 1985, Mengistu “was one of the few who genuinely seemed to know what they [the members of the Derg] wanted. Where others were uncertain or timorous over the role that the group should seek for itself, Mengistu had no hesitation in proclaiming that it should seize power.”44 For those within the Derg, Mengistu had other desirable attributes, not the least of which was his eloquence in articulating the Derg’s demands. Dawit Wolde Giorgis thought it important that Mengistu “had come from a poor family background and was not an Amhara, the dominant ethnic group under Haile Selassie. . . . As a person and as a junior officer he represented our rejection of past values. . . . Many of us [in the Derg] felt he would

52

Civil War in African States

embody the spirit of the Revolution and symbolize the change we wanted to bring about.”45 If all revolutions require violence and terror as a means to break free from old ways, then Mengistu’s approach was acceptable—even necessary—for some. But Ethiopia had also enjoyed a comparative sense of stateness in terms of its monarchy and its political institutions—a parliament and a constitution—which could, in theory, have evolved to channel and respond in limited ways to political grievances. The new regime, however, actively sought to undermine even these institutions. “[I]t seems clear that [Mengistu] knew from the outset that he wanted to destroy not only the monarchy but the entire structure of government and society,” wrote David Korn. “He was not out to make modest changes in course. . . . [T]he Derg seems not to have given a moment’s thought to the possibility of strengthening or preserving the parliament, even as a façade.” Korn concludes that “Mengistu wanted, and got, a clean sweep of all institutions.”46 Revolutionary ambition had thus created its own anarchy, and consequently the military—with its vastly superior organization and military power—was well positioned to move into the political vacuum with little or no opposition. But instead of bringing order to the country, the Derg appeared to be the source and beneficiary of disorder. To many, the Derg undertook the same methods as the emperor except that now there was force to mediate the ruler’s power. Aregawi Berhe, a founding member of the Tigrayan People’s Liberation Front, states that the Derg distinguished itself from the emperor’s rule by the excesses that came from being a military dictatorship. “This military regime was not only repressive,” he writes, but also “exceptionally brutal in its use of sheer military force and police terror against all national and ethnic groups that dared raise the democratic question of self-determination or autonomy—even those that conformed their presence within the framework of the then Ethiopian state.”47 If the events before November 23, 1974 had opened up the possibility of nonviolent revolutionary change and a peaceful resolution of the Eritrean conflict, the emergence of hardline elements within the Derg—and their willingness to use shockingly violent means—then changed everything. A feudal and anachronistic imperial rule had been exchanged for one directed towards, at least initially, both Ethiopian nationalism and modernization—although in this case modernization referred to a brutal effort to homogenize the state through the elimination of sub-nationalist movements. The results were as predictable as they were unintended. For Eritreans, the killing of General Aman was equivalent to a declaration of war.48 As Paul Henze

The Wars of Ethiopia and Eritrea

53

wrote, “Mengistu’s uncompromising approach not only hardened the Eritrean insurgent’s will to resist . . . it drove the Eritrean populace into a more negative position toward Ethiopia than had ever occurred in imperial times.” As for the government, he adds, “all major actors concerned with Eritrea [had now] succumbed to the fatal illusion that they could impose their will by force.”49 “The Military Council members approached the Eritrean problem no differently than the Emperor had,” wrote Dawit Wolde Giorgis. “In its first two years the Military Council was absolutely chauvinistic toward Eritrea. The members, as military men, were ruled by a stubborn pride that would not allow them to negotiate with the secessionists while they had the upper hand over Ethiopian troops in the field.”50 Prominent individuals from within the new Ethiopian leadership or the army who strayed from advocating this coercive strategy of domination or who acknowledged that the war against Eritrean insurgents was unwinnable were arrested. The same fate awaited those who allegedly threatened the regime itself. In the post-revolution power struggles, General Aman Michael Andom, Major Sisay Habte, Derg vice-chairman Lt. Col. Atnafu Abate, as well as Aman’s successor as head-of-state, Teferi Bante, were subsequently executed as antirevolutionaries.51 Indeed, the insecurity within the Derg itself created an environment in which only the most ruthless among them survived. Mengistu’s association with the organization of a disastrous “peasant march” into Eritrea in 1976, and the rise of others advocating alternative solutions to the Eritrean problem, might have spelled Mengistu’s demise were it not for the decisive action he routinely took against his rivals.52 The most violent of these was a 1977 palace shootout that left eight senior anti-Mengistu members dead and Mengistu, as it were, the last man standing. When the bloody purging was over, Mengistu justified his actions, stating that he wanted “to have for breakfast those who wanted to have him for lunch.”53 It was only at this time that a more coherent—while still destructive—approach to Eritrea emerged. The Derg’s Move to the Soviet Union

Witnesses to the Bloody Saturday executions claimed that in the moments before his death, Kebede Gebre, the former defence minister under Haile Selassie, called out to his executioners, “You are spilling innocent blood and it will come back to haunt you.”54 Indeed, the violence had serious repercussions for an incipient regime now bent on a strategy of coercive domination. Not only did it provoke condemnation from the United Nations General Assembly, but it further unnerved

54

Civil War in African States

Ethiopia-watchers in Washington. After thirty years of military patronage, Washington announced in 1974 that it was halting new economic and military assistance “pending a better understanding of the situation.”55 This threat to cut military assistance was genuine given that, by the early 1970s, the strategic significance to Washington of the Kagnew communications station had declined as a result of satellite technology and the establishment of another communications station on Diego Garcia in the Indian Ocean. That Mengistu successfully navigated himself and his stillprecarious regime through such uncertain times was a testament to his ability to survive. But in order to survive, Mengistu had to find a new external patron that was more compatible with his ideological orientation and that was willing to supply sufficient weaponry to keep internal and regional adversaries at bay. This latter threat was real: neighboring Somalia saw the political turmoil in Addis Ababa as an opportunity to make good on its plans to claim Somali-inhabited regions of eastern Ethiopia. For some observers, Ethiopia had no alternative but to appeal to Moscow given the regime’s increasingly strained relations with its American patron, and the sense of siege generated by separatist insurgents in the north and an irredentist Somalia in the east. According to John Spencer, “She had nowhere else to turn; it was preferable to being partitioned between Somalia and an independent Eritrea.”56 A differing view is that Mengistu’s intentions vis-á-vis his adversaries were, from the start, offensive rather than defensive. His actions were a consequence, not of weakness arising from an American reluctance to supply weaponry, but of Mengistu’s prior decision to pursue a strategy of domination against internal enemies.57 David Korn claims that the latter interpretation is closer to the truth. As a revolutionary, he says, Mengistu saw himself as “no friend of the United States” and that he was “chief among those in the Derg who wanted the Americans out of Ethiopia and the Soviets in.”58 Given the scale of weaponry required, “Soviet arms were what Mengistu and many others in the Derg wanted from the very beginning. . . . The decision to go for a military solution in Eritrea made this absolutely essential.”59 The Derg managed, however, to maintain sufficiently cordial relations with Washington until this new and more lucrative relationship with Moscow could be consummated. While Washington had been troubled by the events of the 1974 revolution, it was also reluctant to end arms shipments fearing that doing so “would only strengthen the hands of radical elements among the military and further frustrate the moderates.”60 Washington resumed, and for a time even increased, its military supplies to Addis Ababa. A fleet of American-made F-5E

The Wars of Ethiopia and Eritrea

55

fighter aircraft, delivered in April 1976, proved to be crucial in slowing the advance of invading Somali troops one year later.61 But American officials claimed that Washington did not believe that the Derg’s war with the Eritreans was winnable and that they should not continue servicing even these aircraft in the face of human-rights violations and the increasingly frequent public denunciations of the United States emanating from Addis Ababa.62 Endearing his regime to Moscow required Mengistu to shift from a nationalist program to an espousal of a Marxist agenda—one that was more radical than that followed by the rebels in the north to whom Moscow had previously provided limited support. In the first few months of its rule, the policies of the Derg had been ill-defined and without a specific reference to socialist doctrine.63 Following the lead of the country’s radicalized students and after the triumph of socialist elements within the Derg, the new regime proclaimed that its seizure of power had been in the name of socialist revolution, and that the end of capitalist exploitation and the application of a “scientific solution” would render Eritrean secessionism irrelevant.64 Marxist ideology did not, of course, mollify Eritrean demands but it did serve other purposes.65 According to Dawit Wolde Giorgis, “Mengistu and his clique saw the advantages of a totalitarian system. They used Marxist ideology . . . to justify the establishment of a political system that would, in the name of democratic centralism, give a few people absolute power.”66 Indeed, Marxism served the regime’s desire to hold on to power quite effectively. Steven R. David writes that Marxism-Leninism legitimized the Derg and undercut some of its opposition from radicalized students. However, “in no instance did Mengistu put forth Communist programs that did not directly contribute to his hold on power. More tellingly,” writes David, “when ideological concerns clashed with his concerns for survival, ideology always lost out.”67 A full-blown Soviet commitment to the regime in Addis Ababa did not in fact come until October-November 1977.68 Following the establishment of Mengistu’s revolutionary credentials and his emergence as the undisputed leader of the Derg in February, Ethiopia had come under siege from Somali troops who had invaded the Ogaden desert in July. Having failed to bridge the political differences between the Somali capital Mogadishu and Addis Ababa, Moscow aligned itself exclusively with the latter. The vast supply of military hardware (estimated at $1 billion) and the provision of several hundred military advisors and 11,000 Cuban troops allowed Ethiopian forces to reverse Somali gains and push the invaders back across the Somali-Ethiopian border in February and March 1978.69 The July invasion itself, writes

56

Civil War in African States

Dawit Wolde Giorgis, “was a tragic moment for the Ethiopian people ... but for Mengistu it became a blessing in disguise.” A power struggle among various Marxist urban civilian groups—and Mengistu’s insistence on violence to contain it—had already generated public resentment toward the regime. With the Somali attack on the Ogaden, Mengistu was able to “focus attention away from his brutal actions, toward a national danger that would galvanize the patriotic feelings of all Ethiopians.”70 Moreover, Moscow could no longer afford to be associated with Somalia, an avowedly irredentist state which it had previously armed. The invasion was the catalyst that opened the floodgates to Soviet military assistance to Ethiopia that lasted until the early 1990s. Mengistu saw his own achievement of power as signaling a change in the nature of the Ethiopian revolution and the means by which the insurrections would be fought. “Our revolution has been transformed from a defensive to an offensive one since 3 February 1977,” he said, referring to the date he executed his own coup within the Derg and emerged as its strongman. “In the future our revolution will go further on the offensive.”71 According to the American Ambassador David Korn, Mengistu’s “readiness to act quickly, to seize the moment, and his lack of scruple at shedding blood became characteristic” of his rule.72 In the bitter power struggles that marked the early years of the Derg, the man who would ultimately establish the strategy for managing insurgencies in the north was one who also triumphed by dispatching his personal rivals with force. Ethiopia’s relationship with the Soviet Union allowed his regime to spurn a cooperative strategy in favor of a coercive one where he sought to dominate his adversaries.

The EPLF’s Secessionist Strategy and Consolidation of Power

Quashing the rebellion in Eritrea was regarded as the most critical test of the Derg’s anti-insurgency strategy because it was believed that once the rebels were crushed, other emerging movements who were dependent on the Eritreans for training and support would also collapse. Moreover, the Derg believed that failure in Eritrea would have catastrophic consequences for Ethiopian unity. In a country built upon the forceful incorporation of nationalities, a successful Eritrean insurgency—or even a negotiated secession—was expected by hardline elements within the Derg to be the first domino, culminating in Ethiopia’s disintegration.73 The Derg maintained that these insurgencies were a consequence of external meddling. The “reactionary Arab ruling class,” which,

The Wars of Ethiopia and Eritrea

57

Mengistu alleged, was behind the Eritrean insurgency and which promised to enslave the people of Ethiopia, was just another in a long line of invaders, following “Italian fascists” and “British imperialists,” who sought to dismember the country.74 Yet the Ethiopian people seemed less convinced. Until the Soviet Union had committed itself fully to the regime, the Derg was unable to achieve the military capacity it believed it needed to stave off disintegration, despite its efforts to raise a peasant militia, tens of thousands strong, in 1976. “The Ethiopians are fighting because they are paid to,” one Eritrean commander, Saleh Naud Adem, told a Western journalist. “My men are fighting for a cause. Until there is only one of us left, we will fight.”75 By the early 1970s, however, the Eritrean insurgency was breaking apart. Its problems were the result of the ELF’s inability to manage its own ethnic and religious diversity, a decentralized organizational structure, and a Cairo-based leadership which had trouble enforcing its own decisions. The ELF-Revolutionary Council (ELF-RC)—the selfdescribed “parent organization”—was led by Idris Mohammed Adam and, after 1975, by Ahmed Nasser.76 In 1970, a splinter faction calling itself the People’s Liberation Forces (ELF-PLF) left the main body under the leadership of Osman Saleh Sabbe. Subsequently this group divided into two, one led by Sabbe and the other—which would eventually form the Eritrean People’s Liberation Front (EPLF)—by Ramadan Mohamed Nur and Isaias Afeworki.77 While some factions indicated a willingness to negotiate with the Derg at one time or another, a political existence for Eritrea that was independent of Ethiopia was the stated objective for all three. For Sabbe, the leader who most consistently and publicly portrayed the Eritrean struggle in extreme life-or-death terms, the Derg was engaged in what he described as “genocidal operations” in a war of “liquidation” against the Eritrean people.78 Political independence was necessary if this was to be stopped. According to Sabbe, “If we were to achieve independence by force, the United Nations would accept a fait accompli. . . . Only military victory could induce both [the UN] and the OAU to recognize our existence.” Sabbe was unsympathetic to compromise arrangements that would, for example, preserve Ethiopian access to the sea, noting that “Sixteen African countries have no direct access to the sea. Before it annexed Eritrea, the Ethiopian empire had no opening to the sea. We demand complete independence and no other formula would be acceptable to us.”79 But a secessionist strategy was a hard sell to an international community that preferred to maintain political boundaries as they were, and to African leaders who were conscious of ethnic diversity within

58

Civil War in African States

their own borders.80 Even socialist countries who previously had been sympathetic to Eritrea as an “oppressed nation,” were troubled by the prospect of a movement fighting a secessionist war against an explicitly socialist state.81 To improve the chances that a military victory, were it to be achieved, would be recognized by other African states, Eritrean leaders utilized the language of anti-colonialism rather than of secession or separatism.82 Presenting the Eritrean problem in this way reduced the prominence of border change and increased the relative importance of liberation and survival.83 “Our goal is simply to liberate our land as soon as possible,” noted Tesfai Woldemichael, the ELF secretary-general, in the months before Soviet military aid vastly expanded the power of the Derg. “We see in our military strategy that this is now possible. . . . We are in the final stage just before total victory.”84 The Eritrean movements viewed any rhetorical gestures of cooperation and goodwill from the Derg as disingenuous and likely to jeopardize the Eritrean cause. “Experience has shown,” read a statement of the EPLF Central Committee, “that whenever the Derg raised the slogan concerning a ‘peaceful solution,’ it was a fraudulent tactic designed to deprive the Eritrean resistance of its international support and gain [the Derg] time to consolidate itself militarily.”85 The Derg’s dominant approach to the Eritrean rebellion did not succeed, however, and, on the contrary, was counter-productive. Eritrea had few of the characteristics that usually underpin claims for political independence. What had distinguished it from the rest of Ethiopia was Italian colonialism. But colonialism also made Eritrea little different from other former colonies in Africa insofar as it drew an arbitrary border around a diverse collection of ethnic groups, languages, and faiths.86 Haile Selassie’s use of selective coercion had kept the Eritrean insurgency divided—and therefore weakened—because it was premised on the belief that the rebellion was sponsored by Arab states acting against a largely Christian empire. While continuing to proclaim that Arab states were behind the insurgency, the Derg’s use of coercion was more generalized and helped create a unifying Eritrean identity.87 “Haile Selassie treated the movement as a Moslem conspiracy, directed from the outside,” noted one Eritrean official. “This made it possible for us to get aid from Arab states but divided our people along religious lines. The Derg treats all Eritreans, Christians and Muslims, as members of the front to be liquidated. This has united us.”88 The government’s preoccupation with the crisis in the Ogaden had indeed allowed the three factions of the rebellion to achieve a string of military victories that made the prospect of political independence appear closer than ever. But Eritrea’s leaders were wary of getting ahead

The Wars of Ethiopia and Eritrea

59

of themselves. “Unity is a basic issue because without it there will be no solution,” observed Ahmed Nasser of the ELF-Revolutionary Council. “Our general policy calls for the need to unify all the revolutionary factions.”89 Unity was necessary both to avoid the kind of postindependence factionalism that was consuming Angola and, in the shorter term, to achieve military victory on the battlefield. “I believe that we would find it impossible to agree on the formation of a government as long as we cannot manage to insure unity among ourselves,” noted Osman Sabbe. “However, if we were able to achieve unity, the Eritrean cities which are now surrounded by our forces would immediately be taken . . . and a national government would be formed on Eritrean soil.”90 But while unity was desirable in theory, several elements made it elusive in fact. The patronage of the Arab states gave them influence over the direction of the insurgent movement, which was resented by Christian fighters in the field.91 Moreover, within the ELF there had long been suspicion of Christian highlanders because of their historic association with the unionist movement and because of concerns about possible Ethiopian infiltration.92 The Muslim/Christian divide was further complicated by differences over ideology and tactics. The EPLF sought a Maoist-style “national democratic revolution” requiring the intensive politicization of peasants and workers and the achievement of a radical political and economic restructuring of society. The ELF-RC, though more centrist, also accepted a non-capitalist development strategy as dictated by Moscow-oriented Marxists. For Osman Sabbe and the ELF-PLF, however, debates over ideology were irrelevant and likely to undermine the more important task of defeating the Derg. “Isaias’s Eritrean People’s Liberation Front, like Ahmed Nasser’s ELF, refuses to unite [with us] because they think in Marxist terms,” argued Osman Sabbe. “For our part, we consider ideology a secondary matter and the liberation struggle the top priority.”93 Osman Sabbe’s real concern, however, was that the Marxist rhetoric of the other two movements was undermining financial support from conservative Arab states, support that was central to sustaining his movement. More generally, complaints emerged that Eritrea’s national interests were being subordinated to the personal ambitions of the leaders or their patrons. One EPLF spokesman argued that the factionalization of the Eritrean insurgency was “in no way a result of any supposed desire on the part of the EPLF to achieve unity primarily on an ideological basis but, on the contrary, stem[s] from Osman Sabbe’s repeated maneuvers aimed at undemocratically establishing himself, under the pretext of unity, as the only leader of the Eritrean people.”94

60

Civil War in African States

Among themselves, the Eritrean liberation fronts confronted the same dilemma as other parties to a conflict; that is, whether immediate efforts should be directed towards integration, domination or, by default, continued separation of their respective insurgencies. All parties insisted on the desirability of the former but in practice could only achieve the latter two. In the early 1970s, the ELF-RC attacked the ELF-PLF, claiming that the “Eritrean field cannot bear more than one organization and one leadership.”95 But the ensuing civil war between the factions resolved little; moreover, it was costly, killing more Eritreans in four years than in the previous thirteen years of battle with the Ethiopian army.96 While the ELF-RC and EPLF explored the possibilities of integrating their forces during the 1970s, they continued to threaten Osman Sabbe’s much diminished ELF-PLF. In April 1978, the newly partnered leadership of Ahmed Nasser and Mohamed Nur urged Sabbe to bring his remaining guerrillas into the union within one month or face “total annihilation.”97 Sabbe’s unwillingness proved fatal: when he refused, his fighting forces were finally defeated at the hands of ELF-RC in OctoberNovember 1978.98 Efforts at formally merging the two remaining liberation movements in Eritrea, the EPLF and the ELF-RC, were part of another protracted process that ultimately never succeeded. Diplomatic efforts were undertaken in October 1977, April 1978, and again in December 1978. But accusations emerged that the ELF-RC planned to delay the implementation of any agreement in the expectation that the EPLF would soon be weakened by its continuing confrontation with the Derg, leaving the ELF-RC in command of the struggle.99 Instead, owing to a well-organized “strategic withdrawal” during the government offensives of 1978-79, the EPLF gained strength relative to its Eritrean rival, and the ELF-RC, once the most formidable nationalist army in Africa, went into decline as its fighters began defecting to both the government and the EPLF.100 Persistent division of the Eritrean insurgency provided opportunities for both the Soviet Union and the Derg to make inroads. The ELF-RC, whose survival against the rival EPLF now depended on holding separate negotiations with the Derg, actively welcomed Moscow’s mediation assistance.101 The ELF-RC’s plan was to stave off elimination both by making peace with the Derg and by seeking to forge a new relationship with the ELF-PLF which, despite a near absence of fighters, continued to receive diplomatic and financial assistance from Arab states.102 For its part, the EPLF could not afford to have the ELF-RC negotiate a separate deal with the Derg which would not only recognize

The Wars of Ethiopia and Eritrea

61

the latter as the principal voice of Eritrea but would also allow it to turn against the EPLF.103 It became necessary, then, for the EPLF to defeat the ELF-RC once and for all before a betrayal could be consummated. In a joint operation with the TPLF, the EPLF undertook an offensive in August 1980 which eventually pushed remaining ELF-RC fighters into neighboring Sudan.104 By mid-1981, the ELF-RC was a spent force and the EPLF emerged triumphant in its effort to dominate regional rivals. While both the Derg and Moscow had hoped that an agreement with the ELF-RC would allow it to neutralize a one-time formidable opponent and score propaganda points, they now faced in the EPLF a united and battle-disciplined insurgency. Nonetheless, the Derg persisted in its coercive approach to the Eritrean rebellion. In 1982, after its sixth and largest major military offensive, named Operation Red Star, the Derg claimed to have “effectively smashed” the insurgency in Eritrea. On January 18, 1982, the government stated that “the apparent existence of scattered pockets of insurrection can be of no consolation to those who have worked, directly or indirectly, for Eritrea’s estrangement from the motherland.”105 While the sheer weight of the military offensives did achieve periodic gains for the Derg, the government also experienced significant set-backs when it failed to take several key Eritrean towns. Eritrean rebels claimed that morale among the Derg’s troops was in decline, while morale in areas secured by EPLF remained high. Periodic visits to EPLF-controlled territory by foreign journalists confirmed this to be true.106 As the EPLF advanced on the battlefield, it prepared the next step in its drive for political independence: it sought to achieve certain facts on the ground. Earlier, the EPLF had criticized its rivals, for the “ridiculous belief” that, without concrete development or credible plans, “the shooting of a few rounds of ammunition would draw the attention of the UN which would then ‘grant’ Eritrea its independence.”107 The EPLF, however, created the structures and institutions that made Eritrea a virtual state within a state, albeit an unrecognized one. The emergence of the EPLF as the sole voice for independence, coupled with the coercive tactics of the Ethiopian “other,” created a longed-for sense of unity in Eritrea.108 The EPLF also sought to establish a single coherent national myth, a process that involved an intensive inculcation of its cadres as well as, in some cases, the redrafting of its own history.109 In order to increase its independence from foreign suppliers, the EPLF took over foreign-owned plantations and operated them as state farms. Under the slogan “Destroy the enemy with his own guns and bullets,” the EPLF captured virtually all of its military equipment from government

62

Civil War in African States

troops and serviced them in its own underground workshops.110 Finally, it trained physicians and paramedics, operated hospitals in the field, and supplied its own pharmaceuticals to treat its wounded. Self reliance and self-discipline were the watchwords of the struggle. In sum, the EPLF’s struggle was one that sought domination before separation. The former was essential before the latter could be achieved. An effective war effort and a viable state in future required it to remove uncooperative rivals within Eritrea’s borders. And those weakened rivals could not be allowed to be seduced by overtures from the Derg. The TPLF’s Ambivalent Strategy

Axum, in present day Tigray, had been the ancient capital and center of power in Ethiopia. At the height of its power, Axum’s authority extended throughout much of northern Ethiopia and to the Arabian coastline across the Red Sea. Though most of Ethiopia’s kings were Amhara, the Tigrayans remained powerful contenders for power and, on occasion, produced one of Ethiopia’s rulers. Yohannes IV, who ruled as emperor from 1872 until his death in 1889, was Ethiopia’s last Tigrayan emperor. At the time of his death, political power was already shifting southward from Axum to the Showan region and, within it, to Addis Ababa.111 Yohannes’s successor, Menelik II, and Ethiopia’s political and military elite at that time were more interested in expanding to the morefertile agricultural lands to the south and southwest than to the agriculturally marginal lands in the north. Even prior to Yohannes’s death, Menelik II was negotiating treaties of friendship and commerce with Italy that would provide him with arms and ammunition for further expansion.112 Seven weeks after he assumed the throne, Menelik II negotiated the Treaty of Wich’al_, an agreement that not only consolidated his rule as King of Kings but also effectively divided Tigrayans across two jurisdictions (Ethiopia and the then-Italian colony of Eritrea) and further isolated Tigray from the sea. One may speculate as to whether the new frontier with Eritrea was meant to weaken a rival ethnic group or merely to secure a northern border with Italy while Menelik expanded Ethiopia’s territory to the south.113 In any event, the most significant consequence of Italian colonialism, as far as Tigray was concerned, was its much diminished position vis-a-vis the rest of Ethiopia. Others concur with the view that the persistent neglect of Tigray since then has been purposeful insofar as it deprived the region of its former greatness. Negash and Tronvoll write, “earlier governments did not appear to lament the languishing state of development in Tigray because of a historical fear that a strong

The Wars of Ethiopia and Eritrea

63

Tigray would contend for political power.”114 Indeed, Tigray’s lack of infrastructure and development pushed Tigrayans to migrate to other regions of the country in search of seasonal labor.115 Tigrayans also revolted against their dire economic circumstances in 1942, but were forced into submission by Ethiopian ground forces and British air support—an act of repression which is remembered today by Tigrayans as a “mission” by the central government “to punish the Tigrayan people so that they could never rise up again.”116 A contemporary challenge to Addis Ababa in the form of an organized armed movement from Tigray, the TPLF, did not come until 1975, after Haile Selassie had been overthrown and replaced by the Derg. This suggests that the rise of the Tigrayan insurgency and the expansion of the war in Ethiopia, coming as they did so soon after the emperor’s demise, were related to disappointment and disillusionment with the Derg itself. From the Tigrayan perspective, the Derg, despite its Marxist rhetoric and proclaimed attention to subnational grievances, did not represent a meaningful evolution from the Showan-Amhara ruling group that had dominated Ethiopian politics for so long. Tigrayans, like Eritreans, were not won over by a hastily assembled group of military leaders who appeared only to have appropriated the radical Marxist language of university students for themselves and offered brutality in response to manifestations of national opposition. Specific policies of the Derg—such as severe restrictions on the ability of Tigrayans to work elsewhere and send remittances home—were regarded as damaging to Tigray and generated considerable resentment.117 In addition to specific and real grievances, there was also an element of opportunism behind the creation of a Tigrayan movement. The perception was that the Derg was preoccupied with other things: its own internal power struggles, a deteriorating situation in Eritrea, and increasing tensions in the Somaliinhabited eastern region of Ethiopia.118 Despite the sentiment that “conditions in Tigray were by far the worst” and that “Tigrayans, together with their affinitive neighbours the Eritreans, were the first targets of the new repression,”119 the Tigrayan view that they were being singled out from other non-Amhara regions was unfounded. Historically, as John Young writes, “It is true that lack of state investment in Tigray limited development, but there is little evidence that Tigray suffered disproportionately from other parts of nonShoan Ethiopia in this respect.” Indeed, the emergence of postrevolutionary Ethiopia coincided with a proliferation of liberation movements in various regions of the country, not just Tigray. Although not all movements were organized in this way, ethnic or “national” identity provided a useful means for a number of them to mobilize

64

Civil War in African States

against the Derg. While class emancipation remained the stated objective for radical Tigrayan students, it was because the center was dominated by Showans or their lackeys that the TPLF believed the political struggle had to be undertaken on the basis of specifically ethnonational liberation.120 Tigrayan identity was itself a source of security against the state and provided what was believed to be “the most effective and shortest way to uproot the oppressive system.”121 As a precedent that legitimized its resistance, the Tigrayan leadership pointed to the 1942 peasant uprising, known as Woyane (revolt), and the repression that followed.122 What this meant in terms of its security strategy, however, was less clear—or at least more nuanced. The initial message that the TPLF sent to the outside world was that its goals were, like the EPLF’s, secessionist in nature.123 Its most significant early document on this issue, the Manifesto of the TPLF, declared that the first task of the TPLF would be “the establishment of an independent democratic republic of Tigray.”124 But subsequent public pronouncements on secession were made conditional on other factors—most notably, on who controlled power at Ethiopia’s political center. An “independent and democratic state of Tigray” was the only acceptable outcome according to Ghidey Zeratsion, a TPLF official speaking in 1981, “if the existing oppressive conditions” in Ethiopia continued or were made worse. On the other hand, if a “democratic political atmosphere” was established in Ethiopia, Tigray could become a “voluntarily integrated nation with relations based on equality, democracy, and mutual advantage.” The choice, according to Ghidey, was up to the people of Tigray.125 The attractiveness of political independence was also mitigated by the fact that Tigrinya-speakers lived in both Ethiopia and Eritrea. A secessionist agenda—particularly, one based on an historic Tigrayan nation which might have included Tigrayans in Eritrea (as opposed to the existing, smaller and less viable administrative territory)—would inevitably have brought it into confrontation with the EPLF, a much more formidable military force than the TPLF at that time.126 The response of many other groups—Oromos, Afars, Sidamas, and Ogadenis, alike—to the continued domination by a regime that appeared little different from its predecessors, was to seek escape by making secession part of their own political rhetoric.127 This response was exploited by the TPLF: in order to widen its base of support, the TPLF extended its platform of a right of self-determination to other groups as well.128 The strategy of the TPLF (soon to be the most powerful liberation front after the EPLF), was based on the view that, if the nature of the political center could be changed and previously-marginalized

The Wars of Ethiopia and Eritrea

65

groups given cultural freedom and regional autonomy, a unified state was still possible.129 (in this sense, Tigray’s former greatness was not, in theory, a motivating factor. This was because the end result would not be a Tigrayan-dominated state but a democratic one, where power could be shared and all nationalities would be equally entitled to determine their own future).130 On the other hand, if there was any doubt about whether a future state would respect this autonomy, then ethnic groups—starting with Tigrayans but including all others as well—would have the right to secede. In this sense—and despite contemporary efforts to downplay its significance—separation remained an important element in the TPLF strategy. A right to self-determination not only attracted other disaffected groups to the TPLF cause but it provided an essential escape hatch in the event that a future Ethiopian state was not to its liking. For all it would accomplish in subsequent years, the beginnings of the Tigrayan insurgency were humble and sometimes violent. In its early days, most of the urban students who made up the TPLF were unaware of the logistics of fighting a guerilla war. When the insurrection in Tigray first started, its ranks were so sparse that its fighters were kept on the move to give the impression to both the Derg and the Tigrayan peasants that the front was larger than it actually was.131 At that time many TPLF fighters held romantic visions of battle and underestimated the rigors and hardships the struggle required. John Young alludes to one guerilla leader who claimed that early TPLF cadres had a “‘Che Guevara complex’ and saw themselves as heroes who wanted to fight and win the war quickly.”132 Consequently, its early failures against the Derg led to a significant decline in its numbers and a sense that the movement was on the verge of collapse. Within the TPLF there were also repeated and sometimes violent purges of dissidents and critics. The high rate of battle fatalities coupled with defections resulting from these internal political squabbles and difficult living conditions meant that the number of fighters was more than halved from 1,200 to 450 between 1976 and 1978.133 Those individuals who remained were well practiced in the politics of survival. The TPLF was also poorly equipped for battle, a problem which further discouraged new recruits. A TPLF delegation that toured Arab states in November 1979 acknowledged that it had few weapons, but plaintively claimed that “the human factor remains the most important.”134 Some TPLF cadres wondered why the leadership could not secure arms from either the EPLF (which initially supported the EPRP) or the ELF (which was supporting another group, the Ethiopian Democratic Union). There was no expectation that it would receive

66

Civil War in African States

patronage from abroad. The American Assistant Secretary of State for African Affairs, Herman Cohen, claimed that, as with the Eritreans, secessionist and Marxist aspects of the Tigrayan strategy disqualified the movement from receiving military assistance.135 To compensate for the lack of weapons, TPLF guerillas resorted to carrying sticks covered in plastic to give the impression that they were carrying firearms.136 The Derg dismissed the incipient movement as “common bandits obsessed with narrow nationalism,” who would never be able to mount a serious challenge to a well-equipped military force.137 The TPLF’s Strategy of Domination in Tigray

The ability to forge relationships with other emerging movements, both within and without Tigray, was difficult for the TPLF. As in Eritrea, the field within Tigray was regarded as too crowded for more than one rebel movement. For much of the late 1970s and early 1980s, the TPLF sought to make itself the dominant and exclusive movement in Tigray. The most prominent among the Tigrayan-based groups were the Tigrayan Liberation Front (TLF), the Ethiopian Democratic Union (EDU), and the Ethiopian People’s Revolutionary Party (EPRP). But as we shall see below, in its broader struggle against the Derg, the TPLF sought to eliminate some groups while including others and, when that did not work, it sought the creation of compliant alternative movements. The first of the confrontations in the struggle for control of the Tigrayan rebellion emerged between the TLF and the TPLF in the mid1970s. The TLF advocated the unconditional independence of Tigray from Ethiopia.138 While efforts were initiated to avoid the same fratricidal violence occurring among the Eritrean fronts, minor skirmishes continued between the Tigrayan organizations. An agreement was publicized of the merging of the two movements in November 1975 but the subsequent killings and arrests of TLF leaders by TPLF officials suggested to some that the agreement was not (or ever intended to be) honored.139 A more serious threat emerged from the EDU, an anti-Derg party which was operating out of western Tigray. Led by Tigray’s former governor, Ras Mengesha Seyoum, the grandson of the Tigrayan emperor Yohannes IV, the EDU disavowed association with both the political left and right, and claimed to have representation from all regions of Ethiopia.140 The EDU appealed to the Tigrayan peasantry for support, but was unsympathetic to secessionist aspirations on the part of both Eritrean and Tigrayan insurgencies. However, despite the liberal antimonarchist message it sought to project, the EDU was unable to

The Wars of Ethiopia and Eritrea

67

disassociate itself from the Ethiopian aristocracy. “Ras Mengesha’s movement was based on defending the interests of the prominent landowners,” observed TPLF leader Meles Zenawi. “We didn’t have much in common. His movement declined and ours grew because we had the confidence of the people.”141 The EDU was, however, larger in number and better armed than the TPLF, and prevailed in the first of several military encounters. But in their last major engagement in November 1979, the TPLF reduced the EDU to a rump, a condition from which it never recovered.142 The Ethiopian People’s Revolutionary Party (EPRP) was the third challenger of the TPLF from within Tigray province. Owing in part to earlier efforts to assassinate Mengistu, the EPRP was the Derg’s main civilian Marxist opponent. As a result, it was the central target of the Derg’s “Red Terror” campaign of 1977-78.143 While its urban base in Addis Ababa disintegrated, the EPRP concentrated the forces and activities of its armed wing, the EPRA, in Tigray province. Initially, the EPRP and TPLF had regarded themselves as a “necessity for the survival” of the other.144 But fundamental and irreconcilable differences also emerged between the two movements which soon led to violence. EPRP officials maintained that TPLF cadres were essentially “narrow nationalists” and “high school dropouts,” lacking in “sophistication, clarity and articulation of ideas.”145 While the EPRP professed to be representative of all Ethiopians, the TPLF regarded it as an “Amhara” organization because its political ambition did not promise an outcome different from that of the imperial regime. Indeed, these accusations were indicative of differences over Ethiopia’s central “contradiction” and how opposition should be mobilized. The EPRP focused on class struggle: working-class interests were the priority but the movement was ethnically inclusive and diverse. The TPLF, on the other hand, believed that nationality, particularly that of the peasantry, was the central basis on which revolution could be mobilized. Tigrayans were in fact prominent in the EPRP leadership, and party members boasted that there were more Tigrayan members in its ranks than in the entire TPLF.146 The EPRP also acknowledged that, in principle, “Ethiopia is an Empire state in which national oppression is a reality” and that, therefore, “oppressed nations and nationalities have the full right to self-determination up to and including secession.” Finally, not unlike the TPLF, the EPRP remained convinced that a social revolution, which it would lead and which would bring about fundamental change in society and in the Ethiopian state structure, would address national contradictions and keep secessionism at bay.147 What was answered differently by each, however, was the question of whether a specifically

68

Civil War in African States

national (i.e. Tigrayan) or multinational movement could carry out this task. From the TPLF’s perspective, the national orientation of the movement, rather than the mere fact of Tigrayan representation, was critical for any group operating out of Tigray. Since the TPLF had concluded that Tigray’s misfortune and weakness were linked to the domination by the Amhara and, more recently, by the Derg, the primary vehicle for liberation from this tyranny could not be yet another organization accused of Amhara chauvinism. The only way for the TPLF to achieve their objective—a self-determining Tigray or an Ethiopia in which national identity could be protected—was to mobilize on the basis of nationally-based movements. Consequently, while “multinational organisations should be able to freely move, agitate and organise people all over Ethiopia,” the TPLF insisted that it be recognized as the sole representative of the Tigrayan people. In a key document, the TPLF argued that the struggle of Tigrayans would be advanced most if those opposed to the Derg were operating throughout Ethiopia: “instead of concentrating their efforts in areas where a strong democratic and national organisation exists, the multinational movements should operate in areas which have no vanguard to lead the people, since their task would not be different from that of their democratic national movements.”148 The subtext was that there was no room for two movements in Tigray province. There were no fewer than nine meetings between October 1975 and February 1978 in an effort to reconcile these differences or to establish a modicum of “peaceful coexistence” between the two movements.149 Mutual trust could not be found, due in part to an initial hubris on the part of the EPRP. A founding member of the EPRP writes that “many felt that the political line of the EPRP was correct and that it was an invincible and infallible political force.” Besides, the TPLF was regarded as a “small force that would be limited to the Tigrai [sic] area alone.”150 But the TPLF had acquired military experience in its on-going fight with the EDU and, furthermore, the EPLF was shifting its support from its forces to those of the TPLF. The TPLF had already accepted the EPLF’s position that the Eritrean question was a colonial one that could only be rectified through independence. The EPRP, however, rejected the EPLF’s stand, believing that such a position would make Eritrean independence inevitable and irrevocable.151 Consequently, the TPLF had moral if not military support from the EPLF when it finally forced the EPRP out of Tigray in March of 1978.152 When the end came, the TPLF stated that it was not a “question of who should leave whose area,” but which movement “lacked the support of the people.” The EPRP, it

The Wars of Ethiopia and Eritrea

69

claimed, had been “weakened by its own internal contradiction.”153 Kiflu Tadesse, a founding member of the EPRP, wrote in his memoir of the event that “the defeat of the EPRA [the military wing of the EPRP] meant that the only formidable military opposition to the military regime in Tigrai was the TPLF. . . . The victory of the TPLF over the EPRA was one of the first steps that brought the TPLF to power in 1991.”154 TPLF/EPLF Conflict, Cooperation and Coexistence

The emergence of the EPLF and the TPLF as the most significant insurgent forces in their respective regions—and in Ethiopia more generally—created new opportunities and choices for both movements. There was also a natural kinship and merging of interests between the EPLF and TPLF which neither had with other movements. For the ambition of the EPLF to be realized it was essential that Eritrea have a partner in Addis Ababa that was both sympathetic to the idea of an independent Eritrea and militarily significant enough to help achieve it. Moreover, representative as it was of a relatively small region and minority population, the TPLF also benefitted from a tactical alliance with one of Africa’s most powerful rebel movements. But the essence of each movement was different, as were the grounds on which they planned to pursue their strategies. Their ability to forge an alliance against the Derg was conditional on reconciling these differences. Eritrea’s brand of nationalism was necessarily complex given that, as a creation of colonialism, it was, like other African states, made up of different ethnic and religious groups. From the perspective of the EPLF, no single ethnic or religious identity could be privileged—as was the case in more homogenous Tigray—without undermining the Eritrean independence project. If independence was to be achieved, the new state would have to be viable as a coherent unit inclusive of all of Eritrea’s ethnic and religious identities. Moreover, since independence for African states had previously been achieved on the basis of their status as former colonies, independence for Eritrea could only—according to the EPLF—be achieved as a result of having been a former colony itself. Since a right to national self-determination up to and including secession was at the core of the TPLF program for Ethiopia, however, it was natural that the TPLF also believed that any one of Eritrea’s nine nationalities had such a right. In practice, this meant that one movement could not maintain strict adherence to its principles without undermining the legitimacy of the other. If the EPLF claimed that only former colonies could secede (the only feasible option for the EPLF), the ability of the TPLF to mobilize

70

Civil War in African States

on ethnic grounds was jeopardized (since Tigray was not and never had been a colony). On the other hand, if the TPLF claimed that liberation movements must mobilize on the basis of nationality (the only option for the TPLF), then Eritrea would be fragmented (since there is no one Eritrean nationality). Worse, if national groups were democratically entitled to secede, an independent Eritrea was theoretically vulnerable to future secessionist movements within its own territory. “The EPLF has a much more difficult situation than we do,” Meles Zenawi would later observe. “Many of our differences result from that, and we have understanding and sympathy for their position. . . . [T]he closer the EPLF comes to taking power in Eritrea the more dangerous this issue becomes.”155 These differences became public in 1985, when the two movements published statements explaining their respective positions on these issues.156 From that time until 1988, the two movements did not engage in cooperative military operations against the Derg. However, their grievances did not result in violence and, in a series of documents published in the midst of this feud, the TPLF claimed that it remained in “genuine solidarity with the peoples of Eritrea and their just claim for self-determination.” Its criticisms, the TPLF argued, were of a different nature from those expressed by the “enemies of both the Eritrean and Ethiopian revolutions.”157 In other words, the TPLF chose not to challenge the EPLF militarily—something it would have been hardpressed to do—but instead chose to seek cooperation in the future. “It is the position of the TPLF,” one document stated, “that the co-operation of the EPLF and TPLF, which has been of benefit to both organisations in the past, must and should continue despite our differences [and] . . . is not conditional upon full agreement on all basic issues and policies.”158 For the EPLF as well, having eliminated rivals within Eritrea, there was little need to challenge other movements beyond its territory if ultimately Eritrea’s independence goals had been accepted by the TPLF. In short, the EPLF and TPLF could at least co-exist. “The EPLF was very clear on what its goals were,” noted Yemane Ghebreab, a senior Eritrean official. “It was an independent Eritrea. And therefore it was not in any way in competition for political dominance in Ethiopia.”159 Indeed, it was the combination of the growth of TPLF strength, the acceptance of the coexistence of their respective agendas, and, of course, their common objective of defeating the Derg, that allowed cooperative relations to resume between the two movements.160 When relations were restored, their incompatibilities were set aside (though not forgotten) and a common front was re-established on the basis that “the struggles of the Eritrean people and the peoples of

The Wars of Ethiopia and Eritrea

71

Ethiopia are directed against a common enemy.”161 Nonetheless, cooperation was strictly on a military level, and it was clear that the realization of their common political and military objectives would entail further difficulties. The Weakening Derg

Mengistu’s ability to keep his enemies at bay sustained his conviction that violence was his best option. After all, Mengistu had acquired and consolidated his power through violence, executed all real or imagined political rivals within the regime, wiped out virtually all urban civilian opposition, repelled the 1977-78 Somali invasion of the Ogaden and, for a time, dominated opposition in the northern provinces through massive ground offensives. In the culture of killing that followed the revolution itself, Mengistu brought about what the Ethiopian historian Bahru Zewde refers to as “political emasculation” of the population. “A cowed population,” he writes, “stripped of its arms, reduced to seeing its loved ones lying dead on the streets, forbidden to mourn and in extreme cases even forced to celebrate, was ready to accept anything.” The aim, as another scholar put it, was to “crush the living by showing the cost of dissent.”162 Against the insurrections in the north, reliance on a military solution yielded results that were both pathetic and brutal. In one instance, an eye-witness report related how one or two Ethiopian MiG aircraft would launch daily attacks on a TPLF-held airfield until a pilot at last succeeded in destroying the one nominal asset on the ground that was ostensibly “held” by the rebels: an Ethiopian Airlines DC-3.163 In contrast, the June 1988 bombing campaign in Tigray was of “unprecedented severity,” and included an attack on the town of Hausien that was described by Africa Watch as the “most savage on record.”164 The apparent purpose of the attack was to starve out and force the resettlement of large portions of Tigray’s peasant population so that the TPLF could not harness its support. The actual outcome was less favorable to the regime: not only did such violence broaden the disaffection towards the regime—a disaffection which began during the Red Terror—but it drove Ethiopians in increasing numbers toward the rebel fronts. Equally damaging was the fact that the Ethiopian army began to be undermined by defections from within its own ranks. Indeed, within this onslaught of violence, were also the seeds of the Derg’s demise. The Derg’s strategy to defeat insurgent movements required a massive investment in Ethiopian military capability that was made possible

72

Civil War in African States

through the generous provision of advisors and hardware from the Soviet Union. But it also involved a vast and rapid expansion of Ethiopia’s armed forces—from the 40,000-strong and comparatively disciplined British-trained imperial army under Emperor Haile Selassie to the several hundred thousand troops under the Derg. “The good statesmanship of the Emperor together with the admirable diplomatic skill of his appointees . . . enabled Ethiopia to safeguard its territory and maintain its unity with an army of comparatively smaller size,” writes Eyayu Lulseged.165 But the shift to the Soviet bloc, the commitment to socialist dogma, and the eventual resort to the forced recruitment of often-under aged soldiers—measures which were undertaken to achieve the troop numbers the Derg believed were necessary for such massive interventions—in fact reduced the army’s fighting capacity. According to Lulseged, most army “training centres were like concentration camps because of their poor facilities. The trainees were ill-fed. There was not sufficient drinking water. Medicine was in dire shortage. Trainees had no extra uniform to change.”166 Such rapid expansion also meant that military preparation was increasingly superficial. While imperial troops had been subjected to an extended and intensive training program, an increasingly desperate Derg short-circuited this process. When government troops came under increasing pressure from insurgents in 1986, training at its five military training centers was abandoned so that recruits could be put into action more rapidly.167 By contrast, while the two principal rebel movements dealt forcefully with other contending rebel movements within each of their territories, their greatest asset may have been their comparative discipline and their ability to peacefully induce enemy troops to abandon the Derg army and to incorporate them into their own forces.168 PoWs held by the TPLF were given the options of joining the struggle against the Derg, living in the liberated areas, working abroad (usually the Sudan) on behalf of the struggle against the Derg or, finally, returning home unharmed.169 The rebels thus ensured that many PoWs would not return to fight on behalf of the government or, if they did, they would disseminate the insurgents’ benevolent political objectives. Africa Watch observed the generous treatment extended to PoWs, stating that, while “all the fronts have their dissenters who are fiercely critical of certain [aspects] of their actions and policies ... the treatment of civilians and prisoners of war has been good, even exemplary.”170 In fact, as the government’s fortunes declined, the insurgents were increasingly unable to absorb surrendering troops and thus turned them away. According to TPLF leader Meles Zenawi, “we sometimes capture the same men two or three times. We cannot hold them when we capture them because we

The Wars of Ethiopia and Eritrea

73

have no place and we do not want to feed them. So we release them and they try to go back to their homes. But they get caught and put back into the army. So they defect to us again.”171 When government troops surrendered, they provided large quantities of war materiel to the insurgency. Since there were virtually no external suppliers willing to provide the Ethiopian liberation movements with arms, the capture of weapons was the central means by which this weakness was overcome. The Ethiopian prime minister, Fikre Sellassie Wogderes, is said to have wondered aloud if there was anything other than aircraft that the army had not inadvertently “handed over” to the rebels, while Soviet officials complained that the rebels made better use of Soviet equipment they had captured from Derg forces than did the Ethiopian army itself.172 In his account, Dawit Wolde Giorgis stated that, Today the Eritrean and Tigrayan movements do not have to go looking for armaments. They are supplied by the Ethiopian army. Ethiopian troops are so heavily armed, so immobile, so conventional in unconventional warfare, that whenever units retreat they leave behind enormous amounts of weaponry and materiel — tanks, heavy artillery, guns, munitions—to be seized by the rebels. The secessionist movements have become the most sophisticated and well armed guerilla movements on the continent. These are realities.173

While Mengistu would subsequently blame his regime’s decline on reductions in Soviet arms supplies (and indeed few if any regimes had collapsed while under determined Soviet patronage), it was difficult to imagine how the Derg could survive, in light of its inferior discipline, without direct Soviet involvement.174 With the demise of the Cold War, Mengistu no longer confidently couched his rhetoric in Marxist terms and instead returned to the theme of Ethiopian nationalism. On May 31, 1988, following military defeats in Eritrea and Tigray, Mengistu stated his preference for peace but blamed the EPLF and TPLF for the ongoing violence. Mengistu warned, “the bandits of Ethiopia and Tigray stated unequivocally at their last congresses that they will destroy or be destroyed. The question is: Should they be destroyed, or should we be destroyed?” Despite having long referred to the insurgencies as inconsequential “shifta” (bandits), Mengistu now urged Ethiopians not to underestimate the threat the insurgency posed. “Even though the problem takes the form of disruption in one region of our country,” he stated, “the tendency is a very dangerous one because it could jeopardize the sovereignty, historical unity, and independence of the people of Ethiopia.”175 This

74

Civil War in African States

message continued to resonate among urban Ethiopian elites. Many of them hated Mengistu for bringing the country to ruin but feared the prospect of a narrow ethnically-based insurgency even more, and thus regarded the president as the only man who could keep the country united.176 The Victory of the TPLF in Addis Ababa and the EPLF in Asmara

The TPLF’s creation of ethnically-based groups, which then allied themselves with the TPLF, was a way of overcoming one of the principal problems identified by conflict-resolution specialists: that the strength and breadth of insurgent movements in highly fractionalized states is limited by their inability to extend their cause outside their own narrowly defined ethnic group.177 Following its third congress in March 1989, the TPLF announced the creation of the Ethiopian Peoples’ Revolutionary Democratic Front (EPRDF), an organization that it would lead and which included a number of smaller organizations or People’s Democratic Organizations (PDOs). These PDOs had been established by the Tigrayans and were made up in part by captured government officers and soldiers. Since they had been liberated by the TPLF, it was no surprise that they would agree with the TPLF’s view that political mobilization should be on the basis of national identity and accept the appointment of the TPLF to represent Ethiopia’s peoples. Tamrat Layne, the leader of the Ethiopian People’s Democratic Movement, an EPRDF affiliate, defended his new movement: “The Dergue says that the EPDM is a puppet organization of the TPLF, but we cooperate with the TPLF because of our principles. We are an independent organization with its own programs and policies, our own method and leadership and our own nature, which is multinational.”178 Claiming to represent Ethiopia’s large Oromo population, another “PDO”—the O r o m o P e o p l e ’ s Democratic Organization (OPDO)—emerged as a partner of the TPLF despite the existence of the Oromo Liberation Front (OLF), a less coherent and less TPLF-friendly Oromo-based organization. The OPDO accepted the Tigrayan’s deliberately equivocal approach to self-determination, allowing it to ride on TPLF coat-tails when the TPLF secured power in Addis Ababa in 1991. “There is no problem about the viability of Oromia as a separate state, because its economic potential is very great,” noted Kuma Demeksa, the OPDO chairman. “The OPDO’s choice,” he continued, “is a democratic Ethiopia. Unity is preferred . . . because a united and democratic Ethiopia is in the interests of the Oromos, as long as equality is achieved.”179 Through the creation of the EPRDF, the TPLF

The Wars of Ethiopia and Eritrea

75

mobilized Ethiopia’s diverse population and provided a model for the post-Derg organization of the Ethiopian state. Moreover, in John Young’s words, the EPRDF “provided a means by which the TPLF, representing Tigrayans who constitute about 4% of the population, could maintain a dominant position in the country’s political life.”180 If this approach to TPLF hegemony was integrative, the organizations which it chose to work with (indeed which it created) were, in fact, mostly kindred rather than independent political organizations.181 At the end of March 1989, Ethiopian troops were defeated near the Tigrayan town of Enda Selassie, and government ground forces were withdrawn, leaving the province in the hands of the TPLF.182 The victory meant that the Derg no longer had land access to the north of Ethiopia and could only continue its military campaign in the two provinces of Tigray and Eritrea by supplying its few remaining garrison towns by air transport. While negotiations for a peace settlement were initiated as early as 1989, these proved to be fruitless since both sides refused to make concessions. Mengistu, according to his foreign minister, Tesfaye Dinka, continued to believe he “could not give away what his predecessor Haile Selassie had seized”; nor could he preside over the dismantling of the country by “sell-outs.”183 The guerrilla forces, emboldened by their progress against the government army, felt no need to make concessions which would compromise their political ambitions. In spite of the Cold War’s end and the loss of the two northern provinces to rebel control, there was little evidence that a change in Mengistu’s behavior was imminent. Mengistu continued to search the globe for arms suppliers to take the place of the Soviet Union, and his armed forces did not refrain from launching air attacks against rebel targets. Nor did Mengistu shrink from his old ways when it came to controlling internal enemies—in May 1989 he ordered the execution of twelve of his generals who had been involved in a failed coup attempt.184 Nonetheless, the TPLF had not yet decided to take the struggle south to Addis Ababa. A campaign to the capital would involve further risks, and invite additional attacks from the regime—a likely outcome based on the fact that government assaults were continuing. Rebel leaders were not prepared to move into the capital until they had secured cooperation with other anti-Derg organizations. “We are absolutely confident that we could move down . . . into Addis Ababa,” noted one TPLF statement in June 1990. “But we do not want to move on Addis until we have agreement with other opposition movements for a provisional movement.”185 But other factors that pertained more

76

Civil War in African States

specifically to TPLF interests were also at issue. As hated as the regime was, during a southern offensive in August 1989, Tigrayan fighters had encountered resistance to their movement among civilians outside their own province and it was not clear that their presence would be welcomed outside Tigray. Moreover, some Tigrayans questioned the need for their fighters to be sacrificed for the liberation of non-Tigrayan territory.186 Was it better to consolidate power in Tigray and declare an independent state or did the Derg’s belligerence require the TPLF to subordinate Tigrayan nationalism and take control of the capital? According to John Young, the TPLF engaged in a year-long debate during which military activity was halted. One of the conclusions reached was that government air attacks on Tigray would continue as long as the Derg remained in place in the capital. A second conclusion was that Tigray’s association with the Ethiopian Orthodox Church should not be severed; Tigray’s future was to remain within Ethiopia. “The outcome of the debate,” according to Young, “convinced Tigrayans that their peace and security could be assured only with the Derg’s elimination, after which there were significant developments on the political and military fronts.”187 The question of how to reconcile Tigrayan interests with that of the broader Ethiopian state would come later. The conflict would indeed be resolved through military victory. Sporadic efforts at negotiations continued, although it was recognized that a political settlement was now unreachable, since the TPLF had openly committed itself to a strategy of domination.188 “We think that the Derg is too weak to be a partner in serious negotiations,” noted Meles Zenawi. “The same is true, I think, in respect to the EPLF. The Derg has to be eliminated. We cannot compromise with it. We have to go ahead with our fight to liberate the country.”189 The government also remained committed to a military solution, however misguided. According to American Assistant Secretary of State Herman Cohen, “When he [Mengistu] invited us to become involved, we saw our role as helping to achieve a negotiated peace between the government and its internal enemies. As we later discovered, that was not Mengistu’s agenda. What he wanted was to find a substitute for Soviet military support. . . . Until the very end in May 1991, Mengistu’s overriding aim was to win military victories in his unwinnable internal wars.”190 The EPLF also believed military victory was necessary to their goal, which was to generate sufficient leverage for independence to be recognized by other African states and by the broader international community. “The [government’s] bottom line, with which we agreed,” noted Herman Cohen, “was the maintenance of Ethiopia’s territorial

The Wars of Ethiopia and Eritrea

77

status quo.” But Cohen added that other realities would make moot a negotiated solution and a return to another confederal arrangement. “We also had to prepare for the contingency of an EPLF military victory. . . . We might be faced with a declaration of independence by an EPLF in full control of Eritrean territory.”191 In other words, the reluctance of Western states to fully embrace the possibility of secession had made military victory that much more important for the EPLF. Cohen even observed that the EPLF was relieved when the Derg did not offer a new proposal containing substantive concessions during negotiations. That could have made independence less necessary.192 Still, there was a need for an organized transition. When the Siad Barre regime in neighboring Somalia collapsed in January 1991, Mogadishu had descended into anarchy. Four months later, the imperative was to avoid the same outcome in Ethiopia.193 The disciplined nature that the TPLF had demonstrated was an encouraging indication that a “soft landing” could be achieved in Addis Ababa, and Cohen gave his blessing to an EPRDF move into the capital when the Derg finally collapsed.194 On May 21, 1991, as a peace conference was getting underway in London, Mengistu fled the capital bound for Zimbabwe, and the Ethiopian vice president, Tesfaye Gebre Kidan, told the American chargé d’affaires in Addis Ababa, Robert Houdek, “It is all over. The army has fallen apart. I have instructed commanders to welcome in the Woyane.”195 The TPLF/EPRDF Shares Power: A Strategy of Domination Realized

The Derg had survived for so long—sixteen years—in large part because it had acquired and mobilized, albeit with decreasing effectiveness, overwhelming force against its adversaries. The dangerous early days—when political rivalries threatened to tear it apart and Ethiopians had wagered on its imminent collapse—emboldened Mengistu to use force to solve Ethiopia’s most pressing security problems. But the Derg’s military dominance also stimulated disaffection among groups who, nonetheless, would themselves prosper under uncertainty and come to master the instruments of violence.196 Now, having survived their own perilous birthing periods, and successfully executed their political strategies at such great cost, the question was whether the rebel-movements-turned-governments themselves could offer anything other than force to manage the inevitable challenges they would face. Would their respective political visions of domination and separation in a post-Derg Ethiopia and Eritrea

78

Civil War in African States

translate into a better and more peaceful existence for Tigrayans and Eritreans alike? Would insurgents who had acquired power by forcing others out now be able to let their political rivals in? Or would they, like their predecessors, ultimately succumb to the use of coercion to maintain their own political dominance? In order to defeat the Derg, both the EPLF and EPRDF had had to reconcile conflicts within their respective territories and eliminate their rivals. Now, in order to consolidate and protect their achievements and avoid backsliding, they needed to be single-minded in governance, to avoid political processes with uncertain outcomes, and to exercise selfdiscipline and control. It shared power because, according to one EPRDF statement, “it would have been ridiculous for one organization, operating in only a part of the country, to have formed a government.”197 Yet it only shared that power with those who were willing to accept its political vision. The power that it could now wield—forged over years of struggle with the Derg and its own regional rivals—meant that, even if it was less brutal than Mengistu, the TPLF/EPRDF could also forcefully dispense with those whom it regarded as uncooperative. However, the EPRDF also exercised its considerable power with restraint, reflecting a shrewder appreciation of its minority status than the dire predictions of the outgoing Derg would have suggested. It convened a five-day National Conference on Peace and Democracy for July 1, 1991 to draw up a Transitional Charter and create a new Transitional Government of Ethiopia (TGE). The proliferation of ethnically-based parties at the time of the July conference appeared to confirm the TPLF’s view that ethnic identities were indeed the basis around which Ethiopia’s people preferred to organize, in spite of the fact that it was precisely this form of political organization that the EPRDF encouraged for conference attendees.198 Recognizing Ethiopia’s constituent ethnic components and ensuring the right for people to live and express themselves according to their own culture and language would mitigate the most serious grievances. According to the new president, Meles Zenawi, “The key cause of the war all over the country was the issue of nationalities. . . . People were fighting for the right to use their language, to use their culture, to administer. So without guaranteeing these rights [it] was not possible to stop the war, or prevent another one from coming up.”199 But even this ostensibly open political process generated its own dissenters who believed that the creation of PDOs and other EPRDFinspired groups was a deliberate means of undermining anyone who disagreed with the new governing party. According to OLF deputy chairman Leenco Lata, “We agreed not to involve paper organizations

The Wars of Ethiopia and Eritrea

79

[but] all the organizations with a history of some sort, whether you like that history or not.”200 In the days leading up to the conference, however, EPRDF activists sought not only to identify the established “forces who stand up for peace and democracy,”201 but also those individuals who could represent Ethiopia’s ethnic groups, whether or not they had been organized previously. As the number of political parties mushroomed, the OLF was increasingly concerned that its own access to political power would be diluted by hand-picked or EPRDF-created organizations.202 “We had a big showdown,” recalled Leenco. “It came down to whether there will be a conference in which we will participate or not. But just to give the process the benefit of the doubt, with all our complaints, we decided it is better to participate.”203 Twenty-seven parties and organizations participated in the July conference. Within a year, however, most of the non-EPRDF-affiliated parties had withdrawn from the government and did not contest local and district elections, held in 1992. The OLF, which had joined the interim government but refused to join the EPRDF, insisted on maintaining its own army, which clashed with EPRDF troops throughout the year. Despite the withdrawal of the OLF and other parties four days prior to the voting, the EPRDF went ahead with elections in June. As expected, EPRDF candidates dominated the process: of the 1,147 regional assembly members who were elected, 1,108 of them were EPRDF or EPRDF-affiliated candidates. While Ethiopia’s Oromo and Amhara populations were still, in theory, represented by the EPRDF-affiliated parties—by, respectively, the OPDO and the ANDM, both of which were successful by default—neither party was homegrown and it was not clear that they enjoyed significant voter support. In July 1992, the Transitional Government closed OLF headquarters and its forces attacked OLF troops in southern regions of the country, eventually defeating, disarming, and interning most OLF fighters. When it was over, Terrence Lyons wrote: “this brief conflict demonstrated that behind the EPRDF’s political strategy of ethnic affiliates and regional elections was an experienced and battle-hardened military that could act decisively when necessary.”204 Soon after the elections were completed, a constitutional commission was established and preparations were made for a 547member constituent assembly to debate and ultimately approve the new Ethiopian Constitution. The constitution was an amended manifestation of the TPLF platform and, while there was open and sometimes vigorous debate among those elected to the assembly, the document was ratified in December 1994. Article 39 of the constitution claimed that

80

Civil War in African States

“Every nation, nationality and people in Ethiopia has an unconditional right to self-determination, including the right to secession.” While Eritrea was not a ‘nation,’ as envisioned by the TPLF, it had two years in which to hold a referendum (on sovereignty). Pending that resolution, the earlier July 1991 conference had affirmed Eritrea’s right to administer itself.205 The constitution also re-established Ethiopia as a federal system, consisting of nine ethnically-based states.206 In this way, the EPRDF hoped both to deflect local disaffection with the central government and to direct political competition towards Ethiopia’s ethnic homelands. But what the regime called the recognition of peoples’ right to self-determination was, for critics, a Trojan horse that advanced TPLF domination. “For the Tigrayans who control the state, this is a masterstroke,” noted Makau wa Matua, project director of the Human Rights Program at Harvard University. “It decimates their opponents in Addis Ababa and leaves the center exclusively to them.”207 Even the “loss” of Eritrea was seen by some commentators as part of the TPLF’s survival strategy: in this case survival meant acquiescing rather than fighting. While EPRDF leaders claimed it was merely honoring its pledge to the EPLF to respect the wishes of the Eritrean people through a democratic referendum, an independent Eritrea absolved the regime of Ethiopia’s most draining burden. Referring to the EPLF’s formidable military power, Meles Zenawi acknowledged that “even the military force of the Derg failed to stop it,” and he cautioned that refusing Eritrean independence now would lead to “needless fighting” that his country could not afford.208 Moreover, given the extensive damage that Eritrea had suffered in its struggle with the Derg, an independent Eritrea allowed a still-fragile Ethiopia to avoid costly reparations while the EPRDF consolidated its rule. “If Eritrea were to remain part of Ethiopia,” wrote Paul Henze, “much of the costs of rehabilitation and reconstruction would have to come from Ethiopia’s budget in competition with the needs of all other Ethiopian regions. Eritrean independence relieves Ethiopia of this burden.”209 The EPRDF’s next challenge was to establish an electoral system that met its long-standing pledge to create a democratic state and that also preserved its hard-won control of the levers of power. Despite the satisfaction associated with the Derg’s defeat, Ethiopia’s political climate was not conducive to democratic rule given that the participants in the electoral process consisted of a disciplined and imperious ruling party and a fractious and equally self-righteous opposition. As one election observer wrote, describing the zero-sum environment of the 1994 elections: “There is a total and unyielding criticism of each other, a sense of exclusiveness which just does not allow for anything but

The Wars of Ethiopia and Eritrea

81

disruption and destruction to be expected from other groups. Consequently political positions become self-assertive and absolute: every new demasking of the adversary, every new suspicion instantly adds credibility to one’s own positions.”210 Following the 1994 elections for the Constituent Assembly, elections were held in 1995 (regional and parliamentary), 1996 (local), and 2000 (national)—all without substantial participation from opposition parties and, consequently, each resulted in the continued dominance of the EPRDF or EPRDF-affiliated parties. Critics claimed that the ruling party had broken its word and had not provided sufficient political space for opposition to thrive. Most damningly, they pointed to instances where opposition members had been incarcerated and journalists intimidated. “I will judge this government on the basis of what they have promised to do,” noted Beyene Petros, a prominent opposition leader who refused to participate in the 1995 elections. “They promised unlimited political freedom and respect for human rights, and we will evaluate the government on that offer.” He added that Haile Selassie and Mengistu Haile Mariam “were brutal in their performances but they had not promised democracy. Today the methods are much more subtle, but you still cannot oppose the government—just like as under Mengistu.”211 But the new ruling party was not one to compromise. The success of the TPLF’s long struggle against the brutal reign of the Derg had made it not only the principal force in Ethiopian politics but also one less inclined to share the spoils of power. The TPLF attributed its success over the Derg to its iron discipline and single-mindedness. In the postDerg years, this approach re-emerged as “revolutionary democracy,” in which the EPRDF was the vanguard. Concessions to those who did not share its vision merely diluted the approach that had both brought the Derg down and brought the TPLF to the forefront of Ethiopian politics.212 “From its inception,” write Ken Menkhaus and John Prendergast, “the TPLF undertook mass mobilization based on its singular political vision in the context of constant siege and consistent internal and external competition. In this atmosphere—which lasted more than fifteen years—there was little room for compromise. Victory went to the strongest, the best organized.” When the TPLF did prevail, the implications for Ethiopia’s post-Derg politics were obvious, according to Menkhaus and Prendergast: “It is no surprise then, as we fast forward to the present, that when people do not play by the TPLF rule book, they are ‘expelled’ from the game one way or another. . . . In the main, they do not accommodate different visions. There is little space given because it is felt there is none deserved.”213

82

Civil War in African States

The EPRDF was not merely a political party distinct from the Ethiopian state. Its victory and the control of the executive (including the army, the security forces, and the economy) was intended to establish a government which was favorably and permanently disposed to securing Tigrayan interests; it was not a negotiated pact meant to establish a system in which all parties could participate equally. No wonder that, during the run-up to elections, the EPRDF’s presence was regarded by outsiders and Ethiopians alike as not only dominant but also ubiquitous. In a 1995 report commissioned by the Swedish International Development Agency, Eva Poluha wrote that people in the Peasant Associations she studied told her that “there was only one party, the EPRDF, presenting itself for elections so they . . . never had to make any choice. In their views the government party never intended to let anyone else come forward because it was the EPRDF’s victory that had caused the fall of the Derg and made the elections possible.”214 Equally, having brought revolutionary democracy to Ethiopia, the EPRDF could not allow the weakness of its incipient party system to undermine its political ambitions. The TPLF was forced to intervene in regions where political parties were faltering and threatened to collapse, a tendency that increased the perception among some observers that the Tigrayans were dominant throughout Ethiopia and were not committed to ethnic federalism.215 The EPRDF was, however, sensitive to the competing challenges it faced. As a liberation front it had developed its own democratic credibility as a genuine defender of victimized people; however, in eliminating real or imagined threats, it risked undermining its own democratic credentials. It claimed that all it required was that its political rivals participate and follow the rules that had been openly debated in the Constituent Assembly. The EPRDF’s control of the voting process was never as absolute as that of the Derg or as extensive as its critics claimed. While coercion was employed and election monitors have repeatedly stated that ballot boxes were stuffed, the EPRDF’s electoral strength was equally a consequence of the opposition’s fractious nature. Terrence Lyons regarded the Ethiopian opposition as “a collection of diverse groups, often disorganized, sometimes irresponsible, poorly led, . . . split by numerous internal differences,” and seeking “to divert the EPRDF plans for a controlled transition.” The imbalance between the EPRDF and the opposition, he wrote, “encouraged many parties—particularly those unprepared or ambivalent about participation—to seek to discredit rather than strengthen the electoral process.” He concluded, “By remaining out of

The Wars of Ethiopia and Eritrea

83

the elections . . . the leaders of the opposition played a central part in the consolidation of the EPRDF’s control, and it is difficult to see how they will find a way to engage the newly elected Federal Government constructively during the next five years.”216 Given that the EPRDF won only 484 of 547 seats (89%) of the 1994 elections for the Constituent Assembly, with the remainder going to independent candidates, questions were raised as to whether the opposition had missed an opportunity to challenge EPRDF domination. Frustrated with the opposition’s refusal to participate, Andreas Ashete of the watchdog InterAfrica Group wondered whether the opposition had failed its constituents. “The regional elections were terribly flawed,” he claimed, “but the government has made a serious effort to correct the mishaps and will try harder again next time. Free and impartial elections cannot be left to the government alone. Join the election board! Open offices! That is how elections get to be fair.”217 Even one of the government’s most vocal critics and the chairman of the Ethiopian Human Rights Council (EHRCO), Mesfin Wolde Mariam, noted the failure of opposition groups to take advantage of the extraordinary situation in which Ethiopia now found itself, saying, “There are risks in a country like Ethiopia where everything is new. I do not know if they [Ethiopia’s opposition groups] have really done enough to promote their cause, to force the EPRDF to recognize them.”218 Throughout the 1990s, the EPRDF remained dominant in Ethiopia. But critics noted that the EPRDF’s dominance was hedged with a claim of Tigray’s right to self-determination, up to and including secession. “We felt that those decisions were essentially designed as a fall-back position for the TPLF in case the TPLF and the Tigrayans failed to dominate Ethiopia and failed to have their way inside Ethiopia, in case other ethnic groups and nationalities in Ethiopia became dominant,” observed Yemane Ghebreab, a senior cadre in Eritrea’s ruling party. “Then the TPLF wanted a fall-back position in terms of an independent Tigray. The ethnic politics and the constitution . . . were not arrived at with the overall interests of Ethiopia in mind, with the belief that this was the best way to maintain Ethiopian unity. It was essentially a form of self-preservation. That was the agenda.”219 In the mean-time, however, the constitution became a document that the government could point to in order to sustain and legitimize its rule, notwithstanding the fact that it was a reflection of the EPRDF’s vision even before the Derg was overthrown.

84

Civil War in African States

The EPLF’s Independence Strategy Realized

The EPLF’s post-Derg plan for independence was not conditional on whether or not a new political arrangement could be worked out with Addis Ababa. “We are not fighting for a change of regime but for independence,” observed EPLF spokesman Andemikael Kahsai.220 Upon hearing news of the Derg’s fall several days later, he added that the EPLF was not tempted by the prospect of sharing power in Addis Ababa with its EPRDF allies, stating “We have not been fighting to have ministerial posts in a future government of Ethiopia.”221 But the EPLF also recognized that having its military victory converted into formal recognition as an independent state required the blessing of the international community. The value of this formality was well appreciated by the EPLF: once states are established and recognized, they tend to survive.222 The EPLF indicated its willingness to put off a declaration of independence for three years, at which point Ethiopia’s political situation would have stabilized and Eritrea could hold an internationally supervised referendum. When the referendum did come, in April 1993, Isaias Afewerki made clear that it was nothing less than a decision on whether Eritreans wanted survival or death. The outcome was never in doubt. As the soonto-be president remarked, “The commission [established by the EPLF to conduct the referendum] says that there was 99.8 percent approval in the referendum. This result was not unexpected inasmuch as the issue was not a political competition but an issue of the existence or nonexistence of a people.”223 Given that there were only two options on the ballot—independence or continued existence as an Ethiopian province—it was clear that statehood was regarded as the only means to permanently assure Eritrean security.224 A third option—a federal solution that harkened back to the compromise proposed by the UN in the early 1950s and which would have been dependent on the goodwill of Ethiopia to uphold—was regarded as dangerously deceptive and therefore not to be considered, let alone repeated. Once the results were known, Isaias hailed the liberation of Eritrea as a “historical victory for the Eritrean people that ended 50 years of colonial rule that started after World War II.” Indeed, if it was not already clear to the world at large, Isaias recalled the event which made the EPLF’s separation strategy inevitable. The 1952 Federal Act, which established Eritrea’s autonomy status within Ethiopia, had proven to be merely a parchment barrier, and neither it nor the United Nations had been a sufficient guarantee of Eritrean sovereignty. At a speech before the UN General Assembly, Isaias scolded the organization, stating that

The Wars of Ethiopia and Eritrea

85

“not once in 41 years did Eritrea, scene of the longest war in Africa, and victim of some of the grossest violations of human rights, feature in the agenda of the United Nations.” Worse, he said, a failure to recognize Eritrea’s plight, “gave a free hand to aggressors, thereby prolonging our suffering and increasing the sacrifices we had to pay.”225 Implicit in the president’s comments was the view that as a member of the community of independent states, Eritrea’s fate was finally and rightfully in its own hands. Independence was declared on May 24, 1993. The government was led by Isaias Afewerki, and the legislative body—the National Assembly—included seventy-five elected representatives and seventyfive members of the People’s Front for Democracy and Justice (PFDJ)—the name the EPLF subsequently adopted at its February 1994 Congress. There was, however, uncertainty regarding the type of political system the PFDJ was planning to adopt. Initially, Isaias’s vision for a liberal democratic future for Eritrea was unequivocal: “Our policy on human rights and democracy is unshakeable. We are working to give priority to the establishment of democratic institutions. . . . On this basis we will work hard to establish laws and regulations which guarantee the formation of a government which will be based on multipartyism and be elected by the people in a democratic manner for the first time in the history of Eritrea.”226 Within days, however, the commitment to a multiparty system was qualified. Isaias claimed that, while the PFDJ was in favor of democratic rule, such rule would not manifest itself in the familiar Western model but as one that “suits our society and its traditions.” He said that, even though democracy would not be practiced through political parties, pluralism could still be achieved by way of Eritrea’s democratic institutions, its constitution, and its laws. To counter critics who suspected that this non-traditional approach was merely another demonstration of the one-party state found elsewhere in Africa, Afewerki responded that he was indeed obligated “first and foremost to avoid the fatal mistakes committed by liberation movements in most Third World states when they assumed power and monopolized it for many decades without opening a single window for the opposition.” A failure to do so, observed Isaias, was to commit the leadership “to their own death” and their country to the “bottom of a deep political, economic, and social abyss.”227 Nonetheless, both the high voter turnout during the referendum and the remarkably high level of support for independence was held up as legitimizing the EPLF/PFDJ’s dominance of independent Eritrea.228 As for foreign policy, Isaias stated that his primary focus would be his country’s relations with Ethiopia. Based on the experience of two

86

Civil War in African States

years of de facto independence within Ethiopia and, prior to that, several years of collaboration with the TPLF/EPRDF, Isaias claimed that the future would be one of prosperity and promise. Also evident in the postDerg rhetoric was not just the assertion that Eritrea could now control its own fate but also the suggestion that an independent existence meant peace with its neighbors: “We do not want war or problems,” he stated at a 1993 news conference. “We are exhausted by starvation and our economy is ruined. We want neighborly relations founded on basic concepts and principles that primarily boost and open prospects for future cooperation.”229 The year after the independence declaration, the nonpartisan democracy watchdog Freedom House weighed in on Eritrea’s progress toward democracy. The principal democratic event in post-Derg Eritrea was the referendum—the outcome of which was of course congruent with the interests of the PFDJ and which had been achieved thanks in part to both the Derg’s brutality and the PFDJ’s leadership. But even with the Derg gone, the citizens of Eritrea were still not able to change their government through democratic means. Freedom House declared in its 1993/94 assessment that the EPLF/PFDJ, “claiming to represent the interests of the entire population, has monopolized the political life of the country starting from the village councils to the interim legislature, the National Assembly.”230 In subsequent years, Freedom House stated that Eritrea’s political system was either “not free” or only “partly free.” Indeed, in these subsequent assessments, Freedom House became increasingly critical of the PFDJ’s approach to democratic governance. It acknowledged that the PFDJ had ratified a new constitution in 1997 that made provisions for a multiparty government but added that, “in reality, Eritrea has yet to institutionalize a democratic political system” and, more specifically, that “no timetable has ever been drafted and adopted concerning implementing legislation and statutes regarding political parties and elections.” The PFDJ’s dominance continued because the other parties that the constitution had authorized existed only as diffuse and weak factions in voluntary exile, and also because the judiciary had yet to take positions that were at variance with the government’s own positions. The remnant of the ELF, which the EPLF had marginalized during the 1980s, was given no political space and was forced to take refuge in neighboring Sudan and Ethiopia.231 The government was also increasingly sensitive to encroachments by foreign non-governmental organizations, claiming that they often operated “as if no government existed” or as “substitutes for the government.” In Eritrea, Isaias argued, “everybody recognizes that aid

The Wars of Ethiopia and Eritrea

87

of that kind is a form of counter-attack against the state.” Indeed, the government also rejected loans on the basis that they inevitably led to aid dependency, a condition the EPLF had long sought to avoid.232 Yet it was their unapologetic approach to governance which, for a time, won both the EPLF/PFDJ and the EPRDF plaudits from some observers in the international community. In 1998, in an article in the influential journal Foreign Affairs, Dan Connell and Frank Smyth praised the leaders of Eritrea and Ethiopia, claiming that they represented a “new bloc” of African leaders. This new generation of assertive Africans, they argued, had become disciplined in drawn-out struggles with Cold War African despots. Consequently, while not necessarily democrats, they were “pragmatists,” who appeared to have evolved from their violent past and who were achieving significant growth for their countries through free markets, and who insisted that “corruption not class difference” was “the greatest threat to national development.”233 In a remarkable reversal of contemporary attitudes towards governance, the writers overlooked human rights and even sanctioned a kind of non-coercive authoritarianism, praising Isaias’s determined efforts to subvert ethnic fragmentation. The EPLF succeeded in the short term because it successfully exploited the fact that it had extricated Eritrea from Ethiopian control and thus could ride on residual goodwill. Unlike the TPLF who, on the cusp of the Derg’s defeat, quickly expanded its demographic reach by making limited accommodations with willing ethnic representatives, the EPLF had long decided that in order to avoid fatal divisions it had to assume the mantle of the sole representative of Eritrea. To be sure, the regime of Isaias Afewerki demonstrated its probity and proved itself to be administratively competent, even in the absence of national elections. But it also endeavored to convince Eritrea’s diverse groups that domestic consensus and unity needed to be upheld at all costs and thus suppressed the development of opposition parties. In the euphoria of post-independence, most Eritreans were willing to give the government the benefit of the doubt in order to avoid jeopardizing Eritrea’s integrity and hard-won independence. As many within the PFDJ recognized, the ruling party had tied its survival to the continuation of a united and peaceful state: the erosion of one would almost certainly undermine the survival of the other. The Inevitable Clash? The 1998-2000 Ethiopian-Eritrean War

Underlying Eritrea’s demand for independence was an implicit acceptance—based in its own experience—of Lord Palmerston’s dictum

88

Civil War in African States

that there are no eternal allies but only eternal interests. The attraction of independence was that it allowed Eritrea to legitimately seek the coercive means to ensure their survival and to wield its status as a global citizen to generate international attention—important hedges in the event that former allies become enemies. Eritrean leaders acknowledged as much when they dismissed the fact that it was their ally, the EPRDF, which now formed the government in Addis Ababa, and refused to disarm. “We have the same fear, we still have to survive,” observed Sephat Ephrem, the EPLF Chief of Staff. “International law has not been kind to Eritrea. We have paid enough sacrifice. We have to protect our victory by still being vigilant.”234 Ironically, it was Eritrea that was increasingly engaged in disputes with its neighbors—a number of which had been supportive during its struggle against the Derg. These conflicts included a 1995-96 dispute with Yemen over the Hanish Islands in the Red Sea, an undeclared war with the Sudan after Khartoum and Asmara began funding each other’s opposition groups, and a 2008 military stand-off with Djibouti.235 While the Eritrean government claimed that its neighbors were the source of these conflicts, observers speculated that these disturbances were a natural consequence of Asmara seeking to assert its newfound sovereignty. Nonetheless, these events were manipulated to generate a sense of victimization among Eritrean citizens. This not only contributed to an appreciation for the benefits of independence, but also sustained the view that for all its authoritarian inclinations, it was the PFDJ that was the first and last defender of the Eritrean people.236 The most extreme and tragic example of this pre-emptive tendency was the exceptionally violent and ostensibly sudden conflict that erupted between Ethiopia and Eritrea in May 1998 over the border town of Badme. The suspicion that Tigray might forcefully re-embrace its Tigrayan identity had long been a rhetorical concern of independent Eritrea. Some Eritreans claimed that Tigrayan irredentists sought a return to a “Greater Tigray,” which would lead to efforts to reclaim Tigrinya-speaking territories in Eritrea.237 Indeed, several other Eritrean ethnic groups—specifically the Afar, Saho, and Kunama—lived on both sides of the Eritrean-Ethiopian border. It was Asmara’s concern that these groups might also take their example from the Tigrayans. The EPRDF’s approach to Ethiopian unity, which encouraged ethnic groups to celebrate their ethnic roots, was a direct threat to the PFDJ’s nationbuilding process, which effectively required ethnic groups to forego their identity. Eritrean leaders also remained suspicious that Ethiopian nationalists in Addis Ababa had plans to undermine Eritrean

The Wars of Ethiopia and Eritrea

89

independence. Some members of the former Amhara ruling elite made no secret of their desire to bring what they referred to as “Ethiopia’s Red Sea territories” back into the Ethiopian fold, nor did they hide their animosity to the Tigrayan-led government for giving those territories away.238 The specific reasons for this conflict were unclear. There had been tensions between the two governments caused by a number of outstanding disputes between Ethiopia and independent Eritrea—a disagreement over a common currency, Ethiopia’s use of the Eritrean port at Asab, and, not surprisingly, uncertainty over where the border lay. But these issues seemed incommensurate with the level of violence that ensued. After two years of war and efforts by international mediators to negotiate an end to the hostilities, Ethiopian troops pushed the Eritrean army back from the disputed area and threatened to continue north to Asmara. “It was simple,” observed an Ethiopian spokeswoman. The Eritrean army, she said, had been “smashed, kicked, destroyed, devastated, humiliated and sent packing and running.”239 Under international pressure, however, Prime Minister Meles Zenawi halted the offensive in June 2000 and a peace agreement was signed in Algiers on December 12, 2000.240 A decision on the status of the border was referred to an international commission. The Eritrean-Ethiopian Border Commission (EEBC) awarded significant portions of territory along the border to Ethiopia but ultimately assigned the flashpoint town of Badme to the Eritreans in March 2003. Both parties, however, denied that Badme was in fact the source of the conflict and virtually all international observers have since struggled to explain the disputants’ motives or why the dispute escalated so rapidly to full-fledged war.241 “Badme is not bigger than Asmara. Badme is not more important than Assab by any stretch of the imagination. It is some godforsaken village. So it’s not about territory,” insisted Meles Zenawi in an interview.242 Indeed, each party sought only to portray the other as the more powerful aggressor and to characterize themselves as vulnerable and peaceful defenders of their own people, not as the ruthless and disciplined insurgents-cum-rulers who won the struggle against the Derg.243 “Isaias was convinced that it would be a cake walk,” argued Ethiopia’s State Minister for Foreign Affairs, Tekeda Alemu. “The whole rationale in starting the war lay in this: he thought that it would be very easy to rock the boat here in Ethiopia and that the EPRDF would succumb and make compromises or be overthrown. Therefore the survival of the government was under threat without any doubt.” In what became a recurring justification for the EPRDF’s effort to maintain domestic control and garner international

90

Civil War in African States

sympathy, Ethiopian officials linked their survival to that of the state and to the possibility of a much more chaotic situation should the government fall. “It was not only a question of the government not surviving,” noted Tekeda. “With countries such as Ethiopia—where you have weak political institutions and the fusion of government and state—when a government falters, the whole state falters.”244 Eritrean statements during the war declared that Ethiopia’s “territorial ambitions and other larger objectives” had revealed themselves. In fact, Eritrea believed it had good reason to fear that its survival was at stake—though this fear was not related to Ethiopia’s appetite for more territory. Rather, Eritrea knew that Ethiopia considered them a threat. There was a perception among Ethiopian foreign policy officials that Eritrea was arrogant and its overzealous meddling in the affairs of other states was a threat to the region as a whole.245 For some within the TPLF, Eritrea was a state comparable with Israel: territorially compact but militarily powerful. The question of how to respond to Eritrea’s challenge was a source of tension within the TPLF. One view was that to allow Eritrea’s meddling to go unchallenged would only increase its sense of invincibility and its appetite for regional hegemony. Gebre Tsadkan, Ethiopia’s Chief of Staff at the time of the war, was convinced that leaving Isaias in place was unacceptable and that Eritrea “had to be cut down to size.”246 For him, the war was not so much an accident as an opportunity. Tsadkan recommended that having won the border war, the Ethiopian army should continue on to Asmara and replace the Eritrean government with one that was more amenable to Ethiopian interests. Meles Zenawi’s view, by contrast, was that the conflict did not concern threats to Ethiopian territory, that Eritrean aggression had been checked, and that carrying the war to Asmara involved unnecessary risk and would almost certainly jeopardize Ethiopia’s much-valued relations with Washington.247 Following a sequence of events that split the ruling TPLF/EPRDF and at one time included a plot involving Tsadkan to remove the prime minister, Meles won the day and expelled the dissident “clique.”248 Importantly, despite concerns that the longevity of the ruling party was in doubt, local observers contended that the TPLF was not about to let its grip on power slip away. The TPLF was clear that its own unity was central to its survival and that, if internal discipline was not maintained, it would be destroyed by one of Ethiopia’s other morenumerous ethnic groups, particularly the Oromo. As a result, criticisms of the leadership were always “to the point of improving their way of governance” and not to the point of regime change. If Tsadkan had been

The Wars of Ethiopia and Eritrea

91

an Oromo, he might have pressed harder, noted one Ethiopian journalist. The objective, however, was always to maintain TPLF supremacy, not for the TPLF to commit suicide.249 The 1998-2000 border war demonstrated how even the EPLF/PFDJ and TPLF/EPRDF could become engaged in a violent clash for supremacy. Their previous relationship of collaboration and subsequent nonviolent separation proved, in the end, not to be an exception to the rule that Ethiopia’s insurgent groups eventually come to their own struggle for regional domination. States, like insurgent groups, can cooperate when their interests converge, coexist when their interests do not conflict, and wage war when they do. What had changed between the groups was that they no longer had a common enemy in the Derg, they waged their struggle as states, and the political arena in which they competed had become a larger regional one. The lesson for Eritrea, however, did not concern the viability of secession as a means of avoiding conflict with Ethiopia. Clearly secession had not led to conflict avoidance. Rather, the lesson was that allies are not as permanent as the principal interest of security. After having overcome Eritrea’s defenses, there was nothing to stop Ethiopian forces from taking Asmara; what prevented them from continuing the offensive was the fortuitous outcome of a power struggle at the highest levels of the Ethiopian state and, perhaps, a concern about the political fallout from Ethiopia’s invasion of a recognized independent state.250 When it could not win the war militarily, Eritrea was able to rely on this status as a shield and as a means to contain Ethiopia. Had this been a civil war, some Eritreans wondered, would there have been anything to stop an Ethiopian offensive save for a brief statement of moral condemnation from the world community? That Eritrea was right to seek independence had, once and for all, been confirmed. Contemporary Ethiopia and Eritrea

Since the Ethiopian-Eritrean war, both governments have struggled with public hostility and division within their respective regimes. While the EPRDF might have benefitted from the conflict insofar as it strengthened its credentials as a defender of Ethiopian integrity, the results were very much the opposite. Ethiopians who were still angry with the EPRDF for having given up Eritrea, now blamed the regime for the outbreak of the war, for not having taken back the port of Assab, and for not winning Badme in the subsequent negotiations to resolve the conflict. While the EPRDF remained politically dominant, the war failed to create a national consensus behind the regime—in part because of the

92

Civil War in African States

March 2001 split in the leadership of the TPLF. For the time being, the disparity in power between the government and opposition groups limited the latter’s options to those formal political processes that the EPRDF was still obligated to uphold. In the elections of 2005—in which the EPRDF won 327 seats while a bloc of opposition parties won 170 seats—the opposition groups and some election monitors accused the government of electoral fraud and intimidation.251 When the outcome of the voting, which the EPRDF had expected to win, threatened to erode its monopoly of power, the government shut the process down and denied the opposition the important gains it had made. Ethiopia’s Information Minister, Bereket Simon, suggested that the violence which followed in the streets of Addis Ababa was among the “different nationalities of Ethiopia” and, therefore, was not directed specifically at the government. The alternative to the regime’s crackdown, he said, “might have made the Rwandan genocide look like child’s play.”252 The security force’s resort to violence and the statement of its minister were again indications of a regime that equated its survival with the domination of the political process. In a memorandum written after the elections, the government wrote that “Some within the opposition . . . had made it clear that their objective in participating in the elections was not to win seats, but rather to use the opportunity for mobilizing people to change the constitutional order.”253 Since the 2005 elections, the Ethiopian government has come under continuous criticism for its authoritarian ways.254 In Eritrea, there have also been public divisions within the ruling PFDJ since the war with Ethiopia. In September 2001, the government detained a group of eleven former government ministers and accused them of being “traitors” who had sold out the country to Ethiopia during the war.255 The regime clamped down on Eritrea’s independent and government-owned media; by Amnesty International’s account, it has one of the largest number of detained journalists of any country in the world, and possibly also the largest in relation to the country’s population.256 Despite Eritrea’s efforts to maintain national unity domestically, opposition groups have emerged from the Eritrean diaspora and have staged demonstrations in Brussels, San Francisco, and London. These groups—one of which initially called itself the Eritrean Peoples’ Liberation Front-Democratic Party (EPLF-DP) and another, the Eritrean National Alliance (ENA), which emerged from the remnants of the ELF—have continued to demand greater decentralization, democratization, and language rights for Eritrea’s minority groups.257 A third group—a coalition of Eritrean opposition groups formed in 2008 in Ethiopia under the name Eritrean Democratic Alliance—threatened to

The Wars of Ethiopia and Eritrea

93

“kill the spirit of the regime which strongly believes . . . there are no[t] any Eritrean opposition groups.”258 While the Eritrean diaspora was initially supportive of the regime—and continued to provide it with substantial remittances in the years following independence—many expatriates no longer or reluctantly comply with a 2 percent development tax the government requires for them to have access to its embassies abroad.259 This decline in remittances has had dire consequences for Eritrea’s economic health. The government has also sought to maintain tight control over information, by effectively banning opposition parties and criticism of the regime and paying little heed to rights such as “freedom of conscience, expression of freedom, movement, assembly and organization” (Article 19) that were ostensibly protected by the 1997 constitution. According to Amnesty International, “thousands of prisoners of conscience and political prisoners remained in detention after years in prison.”260 Eritrea has also continued its efforts to support dissent beyond its borders. It hosts or supports armed opposition groups fighting in Ethiopia (specifically the Oromo Liberation Front and the Ogaden National Liberation Front) as well as in the Sudan.261 Most recently—and much to the chagrin of the United States—it supported the Union of Islamic Courts in neighboring Somalia—a regime whose interests ran counter to those of both Ethiopia and the United States. Eritrea’s defiance on issues of foreign policy, human rights and democratization continued even in the face of threats by Washington to place it on a list of states that sponsor terrorism. Eritrea’s Information Minister, Ali Abdu, warned the American Assistant Secretary of State, Jendayi E. Frazer, that if she wanted to see “Eritrea kneel down in front of her,” she needed to learn an Eritrean fact: “Eritreans kneel on only two occasions: When they pray and when they shoot.”262 Conclusions

Two of many players on the Ethiopian political scene, both the EPLF and the TPLF confronted the same enemy in their quest to achieve security. However, they initially pursued different strategies, each informed by historical, demographic, and balance of power factors. How do we account for these differences? The regional balance of power not only informed the EPLF’s strategy but it also served as a hedge against efforts by Ethiopia to reclaim Eritrea. In fact, since the resources the Eritrean insurgents could mobilize at the beginning of its insurrection were modest relative to those of the regime, it was necessary to create this favorable balance. It

94

Civil War in African States

did this initially by securing limited support from Arab states and, subsequently and more profitably, by acquiring the vast stocks of weapons abandoned by the Derg Army troops. Later it would need to have in place the internal coherence and fundamental structures of statehood which would make this balance of power permanent. A unified leadership, self-reliance, and a national consciousness were necessary to win political independence—an iron-clad guarantee that Eritrea’s autonomy could not be taken away. The primary determinants of the Eritrean secessionist strategy, however, were historical: its experience as an Italian colony and the perceived betrayal of the federal agreement by Addis Ababa coupled with the international community’s unwillingness to intervene to protect the arrangement it had itself engineered. Integration had been tried and had failed. But if historical factors account for the EPLF’s drive for independence, other factors—most notably, its ethnic and religious diversity—were a challenge to the achievement of statehood and indeed nearly destroyed the earliest manifestations of the independence movement. These difficulties, however, did not lead the movement to the conclusion that Eritrea’s independence should not be sought. Nor did Soviet assistance to the Derg or the rebel movement’s meager beginnings deter the EPLF from pursuing a secessionist agenda. While Moscow’s patronage allowed the Derg to keep the liberation movements at bay and, on occasion, pressured some factions to consider separate deals with the Derg, the EPLF leadership never doubted that the war could ultimately be won. Indeed, the much-too rapid expansion of the Derg’s military and its conviction that only a military solution was satisfactory generated cracks in army discipline which both of the principal liberation movements were able to exploit. In Ethiopia’s case, it is also tempting to look to historical factors—particularly Tigray’s place as mid-nineteenth century rulers of the Ethiopian empire—to explain the TPLF’s efforts to control the state. To be sure, Tigrayans continued to identify themselves with Ethiopia as they had since antiquity. But becoming rulers in Addis Ababa was unrelated to Tigray’s honorable if brief rule except insofar as it legitimized the Tigrinya-speaking minority as part of Ethiopia’s ruling elite. The contemporary idea of Woyane was about making Tigray safe from economic and cultural oppression that arose as a consequence of Amhara (and then Derg) domination centered in Addis Ababa. Controlling power in the capital—even if in conjunction with other ethnic representatives—afforded the TPLF the best protection for the small region it inhabited within Ethiopia. Secessionism, on the other hand, was ultimately ruled out except insofar as, in its milder form of

The Wars of Ethiopia and Eritrea

95

self-determination, it could be used to induce other ethnic groups to its cause and thus facilitate a more favorable balance of power against the Derg. While Tigray’s established administrative boundaries were compact and contained a comparatively homogenous population of Tigrinya-speakers, its landlocked location and poverty made it an unlikely candidate for successful statehood. If these factors led the EPLF and TPLF to their respective strategies of separation and domination, why have they not now been able to embrace the integrative approach their public statements also espoused? What factors have militated against a more democratic approach? The new governments in both states began their tenures with the promise of a truly democratic and cooperative future. But as the TPLF dissident Aregawi Berhe wrote, “the EPRDF leaders took no time to prove that in practice they were no different from their predecessors, the military dictators.”263 Similarly, in Eritrea, while secession has provided the means for its leadership to defend itself from Ethiopia, it has not protected citizens from the EPLF/PFDJ’s own authoritarian ways. Lamentably, as in much of Africa, political events in Ethiopia and Eritrea have been cyclical. While both regimes have established political stability and continuity, reports by human rights groups raise questions about how much inclusiveness and democratic reform has been gained for the Ethiopian and Eritrean people. Observers and participants continue to be surprised by the fact that the pattern continues and that they are unable to change this trend. When they are out of power insurgents emphasize the importance of democratic rule and social justice, but once they are in power they are unable or unwilling to create states that ensure these ideals. They resort in varying degrees to the same coercive proclivities of their predecessors and thus provoke new opposition groups who believe that armed struggle is their only option with which to challenge the regime. There is the often-stated view that rebel movements do not make very good or very democratic governments and that the best thing that a rebel movement can do upon assuming power is to hand authority over to an independent commission, which can then, through an inclusive forum, develop a truly equal and just political system. In the years after the Derg’s fall, those who were left out of the political system (or who refused to participate) have repeatedly criticized the EPRDF and EPLF for failing to accomplish this seemingly reasonable goal. But rebel movements have a difficult time sharing power with others, let alone ceding authority to an independent commission that will pay no heed to the sacrifices it has made. In Ethiopia, too much Tigrayan blood had been expended to begin apportioning significant amounts of power to

96

Civil War in African States

other parties whom the TPLF felt were jumping on the bandwagon late in the day. It was also clear that, having avoided cutting their teeth in the Darwinesque struggle that marked the Derg years, other groups did not have the organizational ability and capacity of those who did endure that struggle. In this sense, it was the process of power consolidation itself, rather than a specific historical event, which precluded inclusion and democratic rule. In fact, in the case of the TPLF and the EPLF, they believed that the very organizational principles and discipline that allowed them to survive and prosper as rebel movements could and should be generalized to run their respective countries. Their experiences have conditioned or reinforced their strategies of domination. But while there is truth to the view that the failure to democratize and find justice lies in part with the principal actors, the discussion provided in this chapter suggests that the problem is with the broader political environment which encourages local actors to see their security in terms of self-help. One need not believe that insurgent movements such as the TPLF and EPLF were disingenuous in their demands for democratization. Rather, particularly after the fall of Haile Selassie, Ethiopia was so ruthlessly anarchic, and the requirements for keeping themselves in contention remained so demanding, that it was unfortunate, though perhaps predictable, that those who emerged from these struggles could not live up to their own claims for benevolent rule. As rulers of territorially static and ethnically diverse states, leaders inevitably have difficulty accommodating all voices—particularly if not all of those voices are themselves inclined to compromise. But neither can they allow those who are disenchanted with their rule to secede for fear that the entire state may dissolve on their watch. Nor can they allow dissidents to undermine the control that sustains their rule. In the end, and despite their rhetoric, democracy is an unattractive option if, as a minority, it jeopardizes their hegemony or if, as victors in an epic struggle, they believe they have a greater entitlement to rule. The selective use of coercion becomes a key instrument in maintaining their rule, even though this approach inevitably invites armed struggle from those who themselves believe they have no other option for finding a voice. The tendency in African states is not ineluctably or exclusively towards violence. More accurately, the political system in which these actors operate forces disputants to secure their own interests first by thinking strategically. Unless they can be used as a means to advance their own interests, the compromises and cooperation that the international community regards as necessary virtues in any democratic

The Wars of Ethiopia and Eritrea

97

system, are more often seen as vices by the rebel movements because they can be exploited by their adversaries. To be sure, the TPLF and the EPLF cooperated in order to defeat the Derg: the EPLF would not likely have received recognition if it had not agreed to work with an entity willing to accept Eritrean secession. For its part, the TPLF provided the means for ethnic representatives from non-Tigrayan areas to administer themselves; additionally small parties who were willing to embrace the EPRDF program were also given a share of the political spoils. But none of this was meant to diminish the control that the EPLF and TPLF had won for themselves when the Derg was defeated in May 1991. On the contrary, these cooperative measures were taken to advance EPLF and TPLF power. Why would any movement undertake the risk of challenging a regime if, in the end, they could not control their own security or had to share it with any or all of the many players who demanded representation? Prior to their defeat of the Derg, the struggles of both the TPLF and the EPLF ultimately involved dominating or eliminating their local rivals as much as accommodating them. A failure to eat was to be eaten. For the EPLF this involved the emasculation of the ELF-RC and the ELF-PLF; for the TPLF, this involved the elimination of the TLF, the EDU, and the EPRP. Survival in many African states requires not simply coercion but vigilance. It means understanding when one stands to benefit from compromise and cooperation and when one needs to remove challenges forcefully. Notably, for some the early days of the Derg heralded a new more-promising future before its revolution turned sour and brutal. The Derg’s subsequent success with military force, and its access to weapons, led it to believe that it could kill its way to victory. But its dependence on coercion led it only so far. Mengistu’s reckless unwillingness to seriously explore non-violent approaches was not only ineffective but counter-productive in that it was exploited by the regime’s adversaries and ultimately led to the Derg’s downfall. The Derg continued to believe that military force was necessary to defeat its adversaries even as it failed to manage the expansion of its own armed forces. By contrast, while the two principal rebel movements dealt coercively with other contenders for power within their own territories, their greatest asset may have been their ability to induce enemy troops to abandon the Derg army, and to organize and incorporate them into their own ethnically-based allies. If it is true that the environment—the expansive and weakly institutionalized units that make u p the African state system—significantly affects the likelihood of conflict and the behavior of disputants, then observers and policy-makers should expect to see the

98

Civil War in African States

authoritarian ways of the EPRDF and the PFDJ challenged by insurgents who will also claim to seek a just and democratic system. Meaningful conflict resolution requires us to reconsider how the states in which these conflicts persist are organized so that cooperation rather than insecurity can be advanced. Otherwise, even the victors in conflicts such as Ethiopia’s are, like those they replace, ultimately doomed to repeat their mistakes, for they too are driven to rely on force to suppress the many players that emerge in their extended territories. 1. Herman J. Cohen, Intervening in Africa: Superpower Peacemaking in a Troubled Continent (Basingstoke: Macmillan, and New York: St. Martin’s Press, 2000), p. 19. 2. There are various dates offered for Axum’s demise. Edmond Keller cites 970 AD. See Borders, Nationalism and the African State, p. 88; Bahru Zewde cites the mid-7th century as marking the beginning of Axum’s decline. See A History of Modern Ethiopia (Oxford: James Currey, 2001), p. 8. 3. Ethiopians claim a three thousand year history based on an alleged link of their kings to Solomon in the tenth century BC. While the evidence for this claim is admittedly thin, legend has it that the Queen of Sheba and King Solomon produced the first descendent in the Solomonic line. Ethiopian emperors have regarded themselves as “King of Kings of Ethiopia” since the thirteenth century when a dynasty claiming to restore the Solomonic line came to power. See Bahru Zewde, A History of Modern Ethiopia, 1855-1991 (Oxford: James Currey, 2002), pp. 1 and 7. 4. See Tekeste Negash, Italian Colonialism in Eritrea, 1882-1941: Policies, Praxis and Impact (Uppsala: Uppsala University, 1987), pp. 51-53. 5. That Haile Selassie was keenly aware of Eritrea’s importance as a gateway to Ethiopia for foreign invaders is noted by Kennedy Trevaskis: “Within less than eighty years it [Eritrea] had served as a base for no less than four foreign invasions: Lord Napier’s successful expedition of 1868, the illfated Egyptian attempt of 1875, the disastrous Italian invasion of 1896, and the Fascist conquest of Ethiopia in 1935. Alien rule in Eritrea,” he concludes, “would always constitute a potential threat to Ethiopian independence.” See G.K.N. Trevaskis, Eritrea: A Colony in Transition: 1941-52 (London: Oxford University Press, 1960), p. 58. 6. For a discussion of the way in which Italian policy distinguished Eritreans in the Italian colony from Ethiopians, see Tekeste Negash and Kjetil Tronvoll, Brothers at War: Making Sense of the Eritrean-Ethiopian War (Oxford and Athens: James Currey and Ohio University Press, 2000), pp. 5-9. 7. Dawit Wolde Giorgis, Red Tears: War, Famine and Revolution in Ethiopia (Trenton: Red Sea Press, 1989), p. 75. Haile Selassie, in contrast, was critical of the “mistake” made by Menelik II of not recovering Eritrea after his victory at Adwa. See John Spencer, Ethiopia at Bay: A Personal Account of the Haile Selassie Years (Algonac: Reference Publications, 1984), p. 253. For the Marxist Derg, which ruled Ethiopia from 1974 to 1991, the battle at Adwa was regarded as a great victory for, in the language of the day, “the broad masses of Ethiopia.” See “Mengistu Scores [sic] Imperialists in Anniversary Speech,

The Wars of Ethiopia and Eritrea

99

FBIS, vol. VIII, no. 42, 3 March 1977, p. B1; and “Banti Warns Against Complacency Over Adowa Victory,” FBIS vol. VIII, 3 March 1976. 8. Dan Connell, “The Birth of the Eritrean Nation,” Horn of Africa, vol. 3, no. 1 (January-March, 1980), p. 15. 9. Cited in Robert D. Kaplan, Surrender or Starve: Travels in Ethiopia, Sudan, Somalia and Eritrea (New York: Vintage, 2003), p. 202. Italy, by contrast, did not see Eritrea as having an economic value independent of Ethiopia. It was, according to Christopher Clapham, “merely a jumping off point for the control of the entire Ethiopian plateau.” See Transformation and Continuity in Revolutionary Ethiopia (Cambridge: Cambridge University Press, 1988), p. 210. For a discussion of Eritrean sentiment towards Italy during its colonization, see Tekeste Negash, Italian Colonialism in Eritrea, 1882-1941: Policies, Praxis and Impact (Uppsala: Upsala University, 1987), chapter 6. 10. “Report of the United Nations Commission for Eritrea,” (A/1285, 8 June 1950) in The United Nations and the Independence of Eritrea (Blue Book Series), Document 4, paragraph 113, p. 72. 11. “Report of the United Nations Commission for Eritrea,” (A/1285, 8 June 1950) in The United Nations and the Independence of Eritrea (Blue Book Series), Document 4, paragraph 118, p. 72. 12. Christopher Clapham, Continuity and Transformation, p. 210. 13. Trevaskis, Eritrea: A Colony in Transition, p. 81. 14. “Report of the United Nations Commission for Eritrea,” (A/1285, 8 June 1950) in The United Nations and the Independence of Eritrea (Blue Book Series), Document 4, paragraph 153, p. 78. 15. “Report of the United Nations Commission for Eritrea,” (A/1285, 8 June 1950) in The United Nations and the Independence of Eritrea (Blue Book Series), Document 4, pp. 87-88. 16. “Final Report of the United Nations Commissioner in Eritrea” in The United Nations and the Independence of Eritrea (A/2188, 17 October 1952), (Blue Book Series), Document 9, p. 133. 17. The Eritrean legislature made two previous calls for union in May 1955 and again in 1957. When union was finally achieved, Haile Selassie added that “Federation contained inherent dangers of creating misunderstanding among the people and furthermore created a duplication of administrative apparatus.” In his view, violent disturbances and disgruntled Eritreans were a consequence of “alien influences” which had infiltrated Eritrea. See African Recorder, December 3-16, 1962, p. 345. 18. Cited in “Report on the Visit to Addis Ababa,” Somali Public Record Office, Reference CO830/26 XC17470. 19. John Markakis, National and Class Conflict in the Horn of Africa (Cambridge: Cambridge University Press, 1987), p. 121. 20. Ruth Iyob, Eritrean Struggle for Independence: Domination, Resistance, Nationalism, 1941-1993 (Cambridge: Cambridge University Press, 1995), p. 90-91. 21. Marcus, A History of Ethiopia (Berkeley: University of California Press, 1994), p. 158. 22. Ethiopia had resisted any effort to guarantee Eritrea’s status by the UN believing that to do so would, in effect, subject both Eritrea and Ethiopia to a “disguised form of UN trusteeship.” See John H. Spencer, Ethiopia at Bay, pp.

100

Civil War in African States

236, 318-319, and 319, footnote 1. For a dissenting view, see Roy Pateman, “The People the World Forgot: The United Nations and Eritrea,” Horn of Africa, vol. XII and XIV, nos 3 and 4, 1 and 2 (1991), p. 28. See also Iyob, The Eritrean Struggle for Independence, p. 85 23. Markakis, National and Class Conflict in the Horn of Africa, p. 106. 24. Spencer, Ethiopia at Bay, p. 303. 25. Spencer, Ethiopia at Bay, p. 304. 26. Africa Contemporary Record, 1969-1970, p. B112. 27. A previous organization, the Eritrean Liberation Movement (ELM), believed that Eritrea’s liberation could be achieved by way of a coup d’etat against the emperor, after which Eritrea’s liberation would be declared. The ELM was eventually crushed by the ELF in 1965. See Iyob, The Eritrean Struggle for Independence, chapter 6. 28. Dawit Wolde Giorgis, Red Tears, p. 9. 29. David B. Ottaway, “Ethiopia’s Government A Shambles,” Washington Post (December 1, 1974) p. A1. 30. David and Marina Ottaway, Ethiopia: Empire in Revolution (New York: Africana Publishing Company, 1978), p. 5. 31. “Eritrean Liberation Front Statement,” Africa Contemporary Record, 1974-75, pp. C77-78. 32. “Eritrean Liberation Front Statement,” Africa Contemporary Record, 1974-75, pp. C77-78. 33. Giorgis, Red Tears, p. 87. 34. Markakis, National and Class Conflict, p. 137, notes that the ELF was more optimistic than the ELF-PLF. 35. “‘Ethiopia Tikdem’ Explained” published in The Ethiopian Herald, 1 November 1974, and in Africa Contemporary Record, 1974-75, p. C78-C81. 36. Noted in Markakis, National and Class Conflict, p. 245. 37. Giorgis, Red Tears, p. 21. 38. Ottaway and Ottaway, Ethiopia: Empire in Revolution, p. 61, 134. 39. Ottaway, “Ethiopia’s Government A Shambles,” Washington Post (December 1, 1974) p. A1. 40. Ruth Iyob, The Eritrean Struggle for Independence, p. 3. 41. Bereket Habte Selassie, “Political Leadership in Crisis: The Ethiopian Case,” Horn of Africa, vol. 3, no. 1 (1980), p. 8. 42. See, for example, David Korn, Ethiopia, the United States and the Soviet Union (London: Croom Helm, 1986), p. 106, 108. 43. Dawit Wolde Giorgis, Red Tears, p. 30. 44. Korn, Ethiopia, the United States and the Soviet Union, p. 110. 45. Dawit Wolde Giorgis, Red Tears, p. 17. 46. David Korn, Ethiopia, The United States and the Soviet Union, p. 111. 47. Aregawi Berhe, “Ethiopia: Success Story or State of Chaos?” in Ronaldo Munck and Purnaka L. de Silva eds., Postmodern Insurgencies: Political Violence, Identity Formation and Peacemaking in Comparative Perspective (Houndsmill and New York: Macmillan and St. Martin, 2000), p. 99. 48. Korn, Ethiopia, The United States and the Soviet Union, p. 11. 49. Paul B. Henze, Eritrea’s War: Confrontation, International Response, Outcome, Prospects (Addis Ababa: Shama Books, 2001), p. 256.

The Wars of Ethiopia and Eritrea

101

50. Giorgis, Red Tears, p. 87. 51. See Bereket Habte Selassie, “Political Leadership in Crisis: The Ethiopian Case,” Horn of Africa, vol. 3, no. 1 (1980), pp. 8 and 9. 52. Ottaway and Ottaway, Ethiopia: Empire in Revolution, p. 139. 53. Cited in Paulos Milkias, “The Great Purge and Ideological Paradox in Contemporary Ethiopian Politics,” Horn of Africa, Horn of Africa, vol. 19, nos. 1-4 (2001), p. 14. 54. David B. Ottaway, “Ethiopia’s Government A Shambles,” Washington Post, December 1, 1974 (p. A1) 55. David B. Ottaway, “Ethiopia’s Government A Shambles,” Washington Post, December 1, 1974 (p. A1) 56. Spencer, Ethiopia at Bay, p. xiv. 57. See for example, Donald Petterson, “Ethiopia Abandoned? An American Perspective,” International Affairs, vol. 62, no. 4 (Autumn 1986), pp. 627-645, especially 636. 58. Korn, Ethiopia, the United States and the Soviet Union, p. 109. 59. Korn, Ethiopia, the United States and the Soviet Union, p. 17. According to Korn, the United States would never have agreed to provide the quantity of weaponry Mengistu demanded. “If they wanted a truly big army,” he states, “they could get the wherewithal to equip it only from Moscow.” 60. State Department “issue paper” of August 29, 1974, cited in Petterson, “Ethiopia Abandoned?” p. 630. This position was followed as late as August 1976 when William E. Schaufele Jr. Assistant Secretary of State for Africa reiterated that “our present policy toward Ethiopia will not only contribute to the stability of this second most populous country in black Africa, but also assist black African states in maintaining the principle of territorial integrity.” See Petterson, p. 635. 61. Petterson, “Ethiopia Abandoned?” p. 635; Korn, Ethiopia, the United States and the Soviet Union, p. 19. According to Korn (pp. 40-41), Soviet aircraft did not arrive until October 1977. 62. See Korn, Ethiopia, the United States and the Soviet Union, p. 14, and Africa Contemporary Record, 1977/1978, p. B246. 63. Ottaway and Ottaway, Ethiopia: Empire in Revolution, p. 7. 64. The obverse of this, however, was that any nationalist (i.e. Eritrean or Tigrayan) expression was regarded as counter-revolutionary. See John Young, Peasant Revolution in Ethiopia: The Tigray People’s Liberation Front, 19751991 (Cambridge: Cambridge University Press, 1997), p. 95. Initially, Mengistu claimed, “progressives” had supported secessionism in Eritrea because it would “weaken the feudal bourgeois regime”. The fact that they did not join the revolutionary movement indicated that Eritrean leaders were in fact “brokers of imperialism and reactionary Arab ruling classes.” See “Mengistu Addresses Nation on Eritrean Issue,” Foreign Broadcasting Information Service (hereafter FBIS), vol. VIII, June 12, 1978, p. B2, and “President Mengistu Addresses May Day Rally,” FBIS, vol. VIII, May 2, 1977, p. B3. The reference to the “scientific solution” can be found in “Seminar on Eritrea Opens in Addis Ababa,” FBIS June 19, 1978, p. B3. 65. In a 1978 speech, Mengistu acknowledged that the Eritrean insurgency did have progressive elements. See “Mengistu Addresses Nation on Eritrean Issue,” FBIS, June 12, 1978, p. B2. Ethiopia’s foreign minister, Col. Dr. Feleke

102

Civil War in African States

Gedle-Giorgis, claimed the opposite. See, Ian Goddard’s Horn of Africa (AprilJune 1978) interview with Ethiopia’s foreign minister Col. Dr. Feleke GedleGiorgis. 66. Giorgis, Red Tears, p. 45. 67. Steven R. David, Choosing Sides: Alignment and Realignment in the Third World (Baltimore: Johns Hopkins University, 1991), p. 117. 68. Africa Contemporary Record, 1977/1978, p. B228. 69. Africa Contemporary Record, 1977/1978, pp. B228-229. 70. Giorgis, Red Tears, p. 34. 71. FBIS, VIII, 7 April 1977, p. B1. 72. Korn, Ethiopia, the United States and the Soviet Union, p. 112. 73. John Young, Peasant Revolution in Ethiopia, p. 118. 74. “President Mengistu Addresses May Day Rally,” FBIS, vol. VIII, May 2, 1977, p. B1. 75. John Darnton, “Eritrean Rebel Army Set for Decisive Test,” New York Times, Monday July 11, 1977, p. A1. 76. Markakis, National and Class Conflict, p. 130. The change in leadership, at the ELF’s second national congress in May 1975, is described by Markakis on page 139. 77. This basic account comes from ELF Chairman Ahmad Nasir in FBIS, March 23, 1977, p. B3. For a different version of these events, see Connell, Taking on the Superpowers, p. 287. 78. “ELF Spokesman Accuses Authorities of Genocide,” FBIS, March 2, 1976, p. B1. 79. “Sabi Views Status of Eritrean Liberation Struggle,” FBIS, March 18, 1977, p. B3. 80. The United States preferred to maintain the territorial status quo of African states even though it meant passing up an opportunity to support an insurrection against a Soviet backed state. Assistant Secretary of State for African Affairs, Herman Cohen writes that “territorial integrity was . . . the only point of agreement between American and Ethiopian governments. Cohen, Intervening in Africa, p. 22. 81. Michael and Trish Johnson, “Eritrea: The National Question and the Logic of Protracted Struggle,” African Affairs, vol. 80, no. 318 (1981), p. 184. 82. See letters by EPLF spokesman Hagos Ghebrehiwet in “Eritrea’s Freedom Fighters,” Wall Street Journal, April 25, 1986, and “Eritreans Want Independence,” Wall Street Journal, January 28, 1987. 83. See, for example, Mengistu’s comments in “Mengistu Addresses Nation on Eritrean Issue,” FBIS, VIII, June 12, 1978, p. B2. 84. Cited in John Darnton, “Eritrean Rebel Army Set for Decisive Test,” New York Times, July 11, 1977, p. A1. 85. Africa Contemporary Record, 1980/81, p. B172. 86. Eritrea’s nine ethnic groups include the Afar (4%), Bilen (2%), Hedareb (2%), Saho (3%), Kunama (3%), Nara (2%), Tigre (35%),Tigrinya (48%), and Rashaida (1%). The population is also divided among Christians and Muslims. 87. Michael and Trish Johnson, “Eritrea: The National Question,” p. 185. 88. Cited in John Darnton, “Eritrean Rebel Army Set for Decisive Test,” New York Times, 11 July 1977, p. A1.

The Wars of Ethiopia and Eritrea

103

89. Ahmed Nasser, in “Al-Ahram Interviews ELF Chairman Nasir,” FBIS, March 23, 1977, pp. B2/B4. 90. “Sabi Views Status of Eritrean Liberation Struggle,” FBIS, March 18, 1977, p. B1/B2. 91. See, for example, the EPLF document “Eritrea: Intervention and SelfReliance,” published in Review of African Political Economy, No. 10, September-December 1977, p. 107. 92. See Pool, From Guerrillas to Government, p. 57, and Markakis, National and Class Conflict, chapter 5, especially pp. 120-122. 93. Sabbe in 1978 interview with Le Monde, published in FBIS June 16, 1978, p. B3. 94. Nafi Kurdi, published in “Le Monde Interviews EPLF Spokesman on Issues,” FBIS, April 8, 1977, p. B3. 95. Cited in Pool, From Guerrillas to Government, p. 70. See also Markakis, National and Class Conflict, p. 134. The first phase of the battle lasted from the spring of 1972 until its climax in February 1973. 96. René Lefort, Ethiopia: An Heretical Revolution? (Oxford: Zed Press, 1981), p. 229. 97. “Two Main Eritrean Independence Groups to Merge,” New York Times, April 29, 1978, p. 3; See also, Africa Contemporary Record, 1978/79, pp. B222224, B231. 98. Connell, Taking on the Superpowers, pp. 288, 305, 496. 99. Africa Contemporary Record, 1980/81, p. B174. 100. Johnson, “Eritrea: The National Questions and the Logic of Protracted Struggle,” p. 192; Connell, Taking on the Superpowers, p. 490, p. 493. 101. On this issue, see Dan Connell, Taking on the Superpowers, p. 495 and 463-464, Africa Contemporary Record, 1980-81, p. B173; and Markakis, National and Class Conflict, p. 248. 102. Connell, Taking on the Superpowers, p. 495, p. 496. 103. The best evidence that this was the perception of EPLF fighters can be found in Dan Connell, “Civil War Threatens in Eritrea,” Boston Globe, September 1, 1980 and reproduced in Dan Connell, Taking on the Superpowers, p. 463-4. 104. Connnell Taking on the Superpowers, p. 530. After the joint operation the two movements confronted the issue of borders between Ethiopia and Eritrea. Allegedly, the two movements agreed at that time to leave the borders as they were but pledged to address the issue once they had toppled the Derg and established formal authority in Addis Ababa and Asmara. Another opportunity to address this issue was in Khartoum in 1988. But the issue was not addressed at that time nor again after the Derg’s fall in 1991. The missed opportunities would provide the basis for the 1998-2000 war between Ethiopia and Eritrea. See Martin Plaut and Patrick Gilkes, “Conflict in the Horn: Why Eritrea and Ethiopia are at War,” Chatham House Briefing Paper, New Series No. 1, March 31, 1999. 105. Africa Contemporary Record, 1981-82, pp. B155-156. 106. See, for example, Andrew Buckoke, “Tanks Back Bandit Warfare in Eritrean Struggle,” The Times, November 22, 1988 p. 13.

104

Civil War in African States

107. See Eritrean People’s Liberation Front “Self-Reliance . . . Only Way to Victory,” published in Review of African Political Economy, No. 10, September-December 1977, pp. 108-114. 108. The generalized and brutal means by which the war was being fought by the Derg meant that there were few Eritreans who were not uprooted, injured or affected in some way. Out of a population of approximately 2.3 million, an estimated 200,000 were killed, 65,000 of whom which were combatants. An estimated 90,000 children were orphaned and 70,000 were injured or suffered permanent disability. See Eric Garcetti and Janet Gruber, “The Post-war Nation: Rethinking the Triple Transition in Eritrea,” Chapter 12 in Michael Pugh ed., Regeneration of War-Torn Societies (New York: St. Martin’s, 2000), p. 221 and 224. 109. See, for example, Alemseged Abbay, “The Trans-Mareb Past in the Present,” Journal of Modern African Studies, vol. 35, no. 2 (1997), pp. 321-334. Notable in this respect is the difference of perspective on the distinct identities of peoples on both sides of the Ethiopian-Eritrean border. Those south of the border tended to see Italian colonialism as having divided an otherwise coherent Tigrinya-speaking people. Eritreans, by contrast, emphasized the distinctness of the peoples on different sides of the border and the claim that the Imperial Emperor Yohannes (who ruled from 1872-1889), a Tigrayan, was a foreigner and conqueror. 110. Quoted in “Self-Reliance . . . Only Way to Victory,” published in Review of African Political Economy, No. 10, September-December 1977, p. 110. 111. The new city of Addis Ababa did not become Ethiopia’s capital until about 1892, three years after Menelik II had been crowned emperor. See Paul B. Henze, Layers of Time (New York: Palgrave, 2000), p. 154, 162 and Bahru Zewde, A History of Modern Ethiopia, 1855-1991 (Oxford: James Currey, 2002), p. 69. 112. Edmond Keller Revolutionary Ethiopia (Bloomington: Indian University Press, 1988), pp. 27-36. 113. See Berhe, “Ethiopia: Success Story or State of Chaos?” pp. 97-98. 114. Tekeste Negash and Kjetil Tronvoll, Brothers at War: Making Sense of the Eritrean-Ethiopian War (Oxford and Athens: James Currey and Ohio University Press, 2000), p. 96. 115. Berhe, “Ethiopia: Success Story or State of Chaos?” p. 99; Markakis, National and Class Conflict, p. 251. 116. Berhe, “Ethiopia: Success Story or State of Chaos?,” p. 99. 117. See Paul B. Henze, “The Tigre People’s Liberation Front: Conversations with Meles Zenawi,” mimeo (April 1990). 118. John Young, Peasant Revolution in Ethiopia, p. 87. 119. Berhe in “The Origins of the Tigray People’s Liberation Front,” African Affairs, vol. 103, issue 413 (2004), p. 576, and Berhe, “Ethiopia: Success Story or State of Chaos?” p. 99. 120. John Young, Peasant Revolution in Ethiopia, p. 83. 121. Berhe, “The Origins of the Tigray People’s Liberation Front,” p. 583. 122. See comments by Meles Zenawi in Paul Henze, “The Tigre People’s Liberation Front: Conversations with Meles Zenawi,” mimeo (April 1990), p. 7.

The Wars of Ethiopia and Eritrea

105

123. See Africa Contemporary Record, 1979-80, p. B200; Sarah Vaughan, “Historical Perspectives on the Ethio-Eritrean Conflict,” paper presented to the conference convened by the Law Forum for Peace, Addis Ababa (9 December 1998), p. 4. Trevaskis also notes the rumors of Tigray’s secessionist aspirations came as early as the Italian defeat in 1941. See, G. K. N. Trevaskis, Eritrea: A Colony in Transition: 1941-52 (London: Oxford University Press, 1960), p. 63. 124. Manifesto of the TPLF, cited in Aregawi Berhe, “Ethiopia: Success Story or State of Chaos,” note 6, p. 123. Former members who have since broken with the TPLF have emphasized that the ultra-nationalists who drafted the Manifesto document had secessionist aspirations but that these clashed with the rank-and-file who immediately insisted that it be dropped. Berhe claims that “This was the stand of the present EPRDF leaders [the successor organization to the TPLF] and was a point of difference with the author [ Berhe] who consistently fought for a democratic unity of Ethiopia.” See also, Berhe, “The Origins of the TPLF,” African Affairs, p. 591. 125. Africa Contemporary Record, 1980-81, p. B175. In subsequent pronouncements the TPLF leadership denied having secessionist aspirations. See, for example, Paul Henze, “The Tigre People’s Liberation Front: Conversations with Meles Zenawi,” mimeo (April 1990), p. 9. 126. Christopher Clapham, Continuity and Transformation in Revolutionary Ethiopia, p. 211. 127. Berhe, “Ethiopia: Success Story or State of Chaos?” p.99. 128. See Martin Plaut and Patrick Gilkes, “Conflict in the Horn: Why Eritrea and Ethiopia are at War.” 129. Jenny Hammond, Fire From the Ashes: A Chronicle of the Revolution in Tigray, Ethiopia, 1975-1991 (Lawrenceville: Red Sea Press, 1999), p. 175, and Young, Peasant Revolution in Ethiopia, p. 100. 130. In fact, according to Berhe, the aim of Tigrayan university students in the mid-1970s was explicitly “not the restoration of Tigrayan hegemony over the whole of Ethiopia as some politicians have presented it.” Aregawi Berhe, “The Origins of the Tigrayan People’s Liberation Front,” p. 576. Other scholarship, however, notes that Tigrayan leaders have always been under suspicion of seeking to restore the “glory of the Tigrean [sic] royal family.” See Marina and David Ottaway, Ethiopia: Empire in Revolution, p. 86. 131. John Young, Peasant Revolution in Ethiopia, pp. 96-97 . 132. John Young, Peasant Revolution in Ethiopia, p. 96. 133. John Young, Peasant Revolution in Ethiopia, p. 115. 134. Africa Contemporary Record, 1979-1980, p. B201. 135. “Whenever conservatives asked why we were not helping the insurgents against the hated Mengistu regime,” noted Cohen, “we replied: Those rebels are Marxist-Leninist-Albanians.” Cohen, Intervening in Africa, p. 22. See also the comments of US Senator Orrin Hatch, “Keep Ethiopia Part of the Reagan Doctrine,” The Wall Street Journal, April 4, 1986. It was noted in jest that when the TPLF captured towns from the Derg they would take down the portraits of Marx, Lenin, and Engels and replace them with even larger ones. See Peter Biles, “Profile: Ethiopian Leader Meles Zenawi” BBC News Online, August 10, 2005.

106

Civil War in African States

136. John Young, Peasant Revolution in Ethiopia, p. 97. At this time the EPLF was providing weapons to the EPRP, rather than the TPLF, because the EPRP was regarded as being more supportive of Eritrean independence. 137. Africa Contemporary Record, 1980-81, p. B175. 138. This is a contested claim. David and Marina Ottaway write that “Despite its name, the propaganda of the [TLF] did not stress Tigrean independence but simply the replacement of the ‘military dictatorship’ in Addis Ababa with a liberal democratic state.” See Ethiopia: Empire in Revolution, p. 86. Berhe claims that the organization which eventually formed the TLF advocated the “outright independence of Tigray.” See “The Origins of the TPLF,” p. 578. 139. Kiflu Tadesse, The Generation: The History of the Ethiopian People’s Revolutionary Party, Part II—Ethiopia Transformation and Conflict (Lanham: University Press of America, 1998), p. 390; Young, Peasant Revolution in Ethiopia, p. 113. 140. Africa Contemporary Record, 1976-1977, p. B193. 141. Henze, “The Tigre People’s Liberation Front: Conversations with Meles Zenawi,” p. 14. 142. Paulos Milkias claims that the demise of the EDU owed as much to an earlier and devastating confrontation with the Derg army as to damage inflicted by the TPLF. See Paulos Milkias, “The Great Purge and Ideological Paradox in Contemporary Ethiopian Politics,” Horn of Africa, vol. 19, nos 1-4 (2001), p. 4. 143. John Markakis, National and Class Conflict, p. 243. For a more detailed discussion of the Red Terror see, Bahru Zewde, “The Red Terror in Ethiopia: Context and Consequences,” in Kjetil Tronvoll, Charles Schaefer, Girmahcew Aneme Alemu eds., The Ethiopian Red Terror trials: Transitional Justice Challenged (Oxford: James Currey Athens: Ohio University Press, 2009). 144. Tadesse, The Generation, Part II, p. 390. 145. Tadesse, The Generation, Part II, p. 392. 146. Tadesse, The Generation, Part II, p. 391; Markakis, National and Class Conflict, p. 254. 147. Markakis, National and Class Conflict, pp. 244, 255. 148. Tigray People’s Liberation Front (TPLF), On Our Differences with the EPLF (People’s Voice: 1986). 149. Tadesse, The Generation, Part II, pp. 390-398. 150. Tadesse, The Generation, Part II, p. 394. 151. Tadesse, The Generation, Part II, p. 395. 152. TPLF officials denied that they received military support from the EPLF. 153. TPLF, On Our Differences with the EPLF, p. 7. 154. Kiflu Tadesse, The Generation, Part II, p. 407. Italics in original. 155. Henze, “The Tigre People’s Liberation Front: Conversations with Meles Zenawi,” p. 9. 156. The documents that explain the positions of the two parties are found in “The EPLF and Its Relations with Democratic Movements in Ethiopia,” and “Statement by the Central Committee of the Tigray People’s Liberation Front (TPLF) on the Occasion of the 24th Anniversary of the Armed Struggle in

The Wars of Ethiopia and Eritrea

107

Eritrea,” both reprinted in Review of African Political Economy, No. 35 (May 1986), pp. 85-94; and TPLF, On Our Differences with the EPLF. 157. TPLF, On Our Differences with the EPLF. 158. TPLF, On Our Differences with the EPLF. Ethiopian officials who were privy to this conflict have stated privately that the TPLF’s willingness to accept Eritrean secession fitted entirely within their democratic principles at the time. As a democratic organization, they were in no position to deny Eritrea’s right to secession from Ethiopia. The TPLF could overlook differences in their respective approaches to the insurgency but resented the EPLF’s attitude which implied that the Eritreans were the senior partner in their relations and could therefore dictate what was in the interests of the TPLF. Gebre Tsadkan Gebretensae, interview with the author, July 31, 2006. 159. Yemane Ghebreab, interview with the author, July 31, 2002. 160. There have been a number of explanations for the mending of relations between the TPLF and EPLF. According to one source, the TPLF deliberately launched attacks against the Derg in order to draw the EPLF’s attention to its growing military strength—particularly since the EPLF had previously believed that nationalist movements were too narrowly based to be useful in the Eritrean independence struggle. See Paulo Milkias “The Great Purge and Ideological Paradox in Contemporary Ethiopian Politics, Horn of Africa, p. 7, and John Young, Peasant Revolution in Ethiopia, p. 157. Sarah Vaughan argues that the TPLF was concerned that the EPLF would forge a deal with the government that would leave the TPLF alone to bear the brunt of the government’s military power. See Sarah Vaughan, “Historical Perspectives on the Ethio-Eritrean Conflict,” p. 7. 161. EPLF-TPLF Joint Statement, April 24, 1988. 162. The latter quote is a citation from Babile Tola. Both can be found in Bahru Zewde, “The Red Terror in Ethiopia” (forthcoming) (p. 49 of mimeo). 163. See John Young, “Ethiopians Destroy Their Own Plane,” New African, July 1988, p. 17. 164. Africa Watch, Evil Days: 30 Years of War and Famine in Ethiopia (New York: Human Rights Watch 1991), pp. 256-264. The attack was especially notable given that Hausien was not in rebel-held territory and apparently had no military significance. 165. Eyayu Lulseged, “Some Institutional Problems of the Ethiopian Army: 1974-1991,” in Harold G. Marcus ed. New Trends in Ethiopian Studies: Papers of the 12th International Conference of Ethiopian Studies, Volume 1: Humanities and Human Resources (Lawrenceville: Red Sea Press, 1994), p. 664. 166. Lulseged, “Some Institutional Problems of the Ethiopian Army,” pp. 668-669. 167. Lulseged, “Some Institutional Problems of the Ethiopian Army,” pp. 668-669. 168. According to one estimate, between 1988 and 1990, the EPLF and TPLF captured 55,000 prisoners of war. See Jane Perlez, “Prisoner No. 14279: Forlorn Pawn in Ethiopia's Long and Ruinous Civil War,” New York Times, February 19, 1990, p. A6. 169. Africa Watch, Evil Days, p. 311; Tekeste Melake, “The Battle of Shire (February 1989): A Turning Point in the Protracted War in Ethiopia,” in New

108

Civil War in African States

Trends in Ethiopian Studies: papers of the 12th International Conference of Ethiopian Studies, Volume 1, Harold G. Marcus ed. (Lawrenceville: Red Sea Press, 1994), p. 965; Young, Peasant Revolution in Ethiopia, p. 128. 170. Africa Watch, Evil Days, p. iii. 171. Henze, “The Tigre People’s Liberation Front: Conversations with Meles Zenawi.” 172. See Eyayu Lulseged, “Some Institutional Problems of the Ethiopian Army: 1974-1991,” in Harold G. Marcus ed., New Trends in Ethiopian Studies: papers of the 12th International Conference of Ethiopian Studies Volume 1(Lawrenceville: Red Sea Press, 1994), pp. 669-670; “Ethiopia: US Peace Talks Go Ahead,” Africa Research Bulletin, vol 26, no. 8 (September 15, 1989), p. 9385A; and Jane Perlez, “On the Ethiopian Front, Rebel Confidence Rises,” New York Times, February 14, 1990, p. A15. 173. Dawit Wolde Giorgis, Red Tears, p. 111. See also the comments by Meles Zenawi in Henze, “The Tigre People’s Liberation Front: Conversations with Meles Zenawi,” p. 14. 174. Steven R. David, Third World Coups d'Etat and International Security (Baltimore: Johns Hopkins University, 1987), p. 79. Mengistu blamed Gorbachev for the collapse of the Derg in an interview with a South African daily, The Star, and subsequently reported in “Mengistu Defends ‘Red Terror,’ BBC News Online, December 28, 1999. 175. Africa Contemporary Record, 1988, p. B281. 176. Harold G. Marcus, A History of Ethiopia, p. 217. 177. See Paul Collier, “Doing Well Out of War: An Economic Perspective,” in Mats Berdal and David M. Malone eds. Greed and Grievance: Economic Agendas in Civil Wars (Boulder: Lynne Rienner, 2000), p. 100. 178. Hammond, Fire from the Ashes, pp. 397-98. 179. Hammond, Fire From the Ashes, p. 409. 180. John Young, “Governance in Post-Cold War Ethiopia,” Paper Presented to the Africa Society Conference, Edmonton, Alberta, February 26-28 1999, p. 3. 181. Bahru Zewde, “The ‘Red Terror’ in Ethiopia” (mimeo, p. 51). 182. Henze, Layers of Time, p. 313. 183. Cited in Cohen, Intervening in Africa, p. 30. 184. Mengistu denied having ever killed anyone during his rule. In a 1990 interview, Mengistu claimed that the accusation was “absurd.” He added, “In the first place, it is not in my nature to kill even an insect or a small living thing, let alone a human being. If anyone perished during the implanting of the revolution, it was certainly not on my orders.” See Carson Black, “Mengistu Courts West with Pledge of Democracy," The Times, June 28, 1990, p. 9. 185. Africa Contemporary Record, 1990-92, p. B 298. 186. Young, Peasant Revolution in Ethiopia, 1997, p. 166. 187. Young, Peasant Revolution in Ethiopia, p. 167. 188. Young, Peasant Revolution in Ethiopia, p. 167. 189. Henze, “The Tigre People’s Liberation Front: Conversations with Meles Zenawi,” p. 16. 190. Cohen, Intervening in Africa, p. 20. 191. Cohen, Intervening in Africa, p. 39. 192. Cohen, Intervening in Africa, p. 43.

The Wars of Ethiopia and Eritrea

109

193. Observers had long worried about such a possibility. See, for example, Lionel Cliffe, “Dramatic Shifts in the Military Balance in the Horn: The 1984 Eritrean Offensive,” Review of African Political Economy, vol. 11, no. 30 (September 1984), p. 96. 194. Cohen, Intervening in Africa, p. 49. See also Neil A. Lewis, “U.S. Plan to Be ‘Midwife’ To a New Rule in Ethiopia,” New York Times, May 26, 1991, sec. 1, p. 10. 195. Africa Contemporary Record, 1990-92, p. B304. 196. According to Meles Zenawi, the Derg “kept the TPLF healthy and alert.” See Harold G. Marcus, “Breakfast Meeting with Prime Minister Meles,” October 20, 1995. Available at: www.hartford-hwp.com/archives/33/008.html. 197. Cited in Sarah Vaughan, “The Addis Ababa Transitional Conference of July 1991: Its Origins, History, and Significance,” Occasional Papers No. 51, Centre of African Studies, Edinburgh University, 1994, p. 31. 198. See Vaughan, “The Addis Ababa Transitional Conference of July 1991,” p. 57. 199. Cited in Vaughan, “The Addis Ababa Transitional Conference of July 1991,” p. 56. 200. Leenco Lata, interview with the author, December 14, 1996. 201. Speech by Meles Zenawi at the opening of the July Conference, July 1, 1991 (mimeo). 202. By January 1992, the US Embassy counted more than 63 political parties. See Paul B. Henze, “A Political Success Story,” Journal of Democracy, Vol. 9, No. 4 (October 1998), p. 46. 203. Leenco Lata interview. 204. Terrence Lyons, “Closing the Transition: The May 1995 Elections in Ethiopia,” Journal of Modern African Studies, vol. 34, no. 1 (1996), p. 128. 205. The US Assistant Secretary of State for African, Herman Cohen, noted that OLF leader Yohannes Lata Wagayo had argued in favor of referendum on the independence of an Oromia state. Cohen discouraged him from pursuing that option. See Cohen, Intervening in Africa, p. 51. 206. Article 47 indicated the states of the Federal Republic. These included: 1) Tigay, 2) Afar, 3) Amhara, 4) Oromia, 5) Somali, 6) Benishangul/Gumaz, 7) Southern Nations, Nationalities and Peoples, 8) Gambela Peoples, 9) Harari Peoples. See The Constitution of the Federal Democratic Republic of Ethiopia. 207. Makau wa Mutua, “The Regionalization Controversy,” Africa Report, September-October 1993, p. 31. 208. “Eritrea Votes to Become Africa’s Newest Nation,” Africa Report, May-June 1993, p. 8. 209. Paul Henze, Eritrea’s War: Confrontation, International Response, Outcome, Prospects (Addis Ababa: Shama Books, 2001), p. 218. This also helps explain why, after two successive regimes that had refused to make concessions on Eritrea, the EPRDF was now willing to endorse Eritrean independence. Ironically, then, while both Haile Selassie and Mengistu Haile Mariam believed Eritrean independence would bring down their respective regimes, the TPLF regarded it as allowing it to consolidate its rule. 210. Siegfried Pausewang, The 1994 Election and Democracy in Ethiopia, cited in Kjetil Tronvoll and Øyvind Aadland, The Process of Democratization in

110

Civil War in African States

Ethiopia—An Expression of Popular Participation or Political Resistance?, Human Rights Report: No 5. (Norwegian Institute of Human Rights, August 1995), p. 7. 211. Cited in Kjetil Tronvoll and Øyvind Aadland, The Process of Democratization in Ethiopia, p. 28. 212. For a brief discussion of revolutionary democracy, see J. Abbink, “Discomfiture of Democracy? The 2005 Election Crisis in Ethiopia and Its Aftermath,” African Affairs, vol. 105, no. 419 (2006), pp. 195. 213. Mark Duffield and John Prendergast, “Liberation Politics and External Engagement in Ethiopia and Eritrea,” Horn of Africa Discussion Paper Series (April 1995), pp. 10-11. 214. Eva Poluha, The 1995 Ethiopian Elections Viewed from the Grassroots, A Report to SIDA. Stockholm: SIDA (1995), p. 15. 215. Medhane Tadesse and John Young, “TPLF: Reform or Decline,” Review of African Political Economy, no. 97 (2003), p. 398. 216. Lyons, “Closing the Transition: the May 1995 Elections in Ethiopia,” p. 139. 217. “EPRDF Landslide,” Africa Research Bulletin, vol. 31, no. 7 (August 21,1994), p. 11509C. 218. Mesfin Wolde Mariam interview with the author, May 25, 1996. 219. Yemane Ghebreab, interview with the author, July 31, 2002. 220. “Initial Rebel Reaction to Mengistu Flight Noted,” FBIS, May 21, 1991, p. 7. 221. Cited both on “EPLF on Transitional Government, Independence,” FBIS, May 29, 1991, p. 8, and “Three Rebel Movements View Future Developments,” FBIS, May 30, 1991, p. 3. The EPLF did not participate in the July 1991 conference except as an observer. 222. Negash and Tronvoll in Brothers at War, p. 92, and discussed at length by Robert Jackson in Quasi-States: Sovereignty, International Relations and the Third World (Cambridge: Cambridge University, 1990). 223. “Eritrean Leader Gives Address on Referendum Results,” FBIS, April 28, 1993, p. 4. 224. Symbolism was associated with the voting procedures in the referendum. The choice of color for the ballots was seen in life-or-death terms: blue, denoting courage, for independence, and red, denoting death, for maintaining Eritrea as a province of Ethiopia. Several sources noted that these colors—ostensibly required because of low levels of literacy—could be used as a means of intimidating those who would otherwise have voted to remain in Eritrea. One election monitor I spoke with personally ridiculed these objections on the grounds that the outcome was so clearly a foregone conclusion that a truly neutral ballot was unnecessary. For a detailed discussion of the symbolism associated with the voting process, see Kjetil Tronvoll, “The Eritrean Referendum: Peasant Voices,” Eritrean Studies Review, vol. 1, no. 1 (Spring 1996). 225. Statement dated September 30, 1993 by the President of Eritrea at the 48th session of the United Nations General Assembly, Document 40, The United Nations and the Independence of Eritrea (Blue Book Series), p. 251. 226. “President Speaks at Independence Celebration,” FBIS Daily Report, May 25, 1993, p. 6.

The Wars of Ethiopia and Eritrea

111

227. “Afewerki on Constitution, Separation of Powers,” FBIS, May 26, 1993, p. 6. 228. Tronvoll, “The Eritrean Referendum: Peasant Voices,” p. 59. 229. “Mubarak-Afewerki New Conference Reported,” FBIS, May 26, 1993, p. 5. 230. Freedom in the World—1993-1994 (New York: Freedom House), p. 248. 231. Amnesty International, “You Have No Right to Ask. . .” (May 2004), p. 4. 232. African Contemporary Record, 1996-98, p. B374. 233. Dan Connell and Frank Smyth, “Africa’s New Bloc,” Foreign Affairs (March/April 1998), p. 82. 234. Cited in Jane Perlez, “Pupil of G.I.'s Does Well, as a Rebel,” New York Times, June 27, 1991, p. A4. 235. See Martin Plaut, “Yemen and Eritrea: Friends Once More?” Review of African Political Economy, vol. 25, no. 78 (1998), pp. 659-661. On the tensions with Djibouti, see Jeffrey Gettleman, “A Red Sea Conflict’s Buffer Zone Is Some Rocks, and a Few Inches,” New York Times, May 25, 2008, p. 6. 236. Freedom in the World, 1999-2000, p. 184. 237. The evidence for this was a new map that was being produced for distribution in Ethiopian primary schools and which incorporated into Tigray several areas of dispute between the two countries. See Plaut and Gilkes, “Conflict in the Horn,” p. 5. 238. This reference is made in the “A Darling of the West Turns Nasty,” Economist, November 12,-18th 2005, p. 52. 239. Cited in Ian Fisher, “Eritrea Gives Up Vital Town to Ethiopia; Peace Hopes Rise,” New York Times, May 26, 2000, p. A3. 240. Meles Zenawi became the Ethiopian Prime Minister in 1995. 241. Sarah Vaughan writes, “A considerable part even of what we know has emerged along the Ethio-Eritrean border . . . remains—quite frankly—inexplicable.” See Sarah Vaughan, “Historical Perspectives on the Ethio-Eritrean Conflict,” p. 2. See also Tekeste Negash and Kjetil Tronvoll, Brothers at War: Making Sense of the Eritrean-Ethiopian War; J Abbink, “Briefing: The Eritrean-Ethiopian Border Dispute,” African Affairs, 97 (1998), pp. 551-565; Dan Connell, “Against More Odds: The Second Siege of Eritrea,” Eritrean Studies Review, Vol. 3, No. 2 (1999); and Ruth Iyob, “The EthiopianEritrean Conflict: Diasporic Versus Hegemonic States in the Horn of Africa, 1991-2000, Journal of Modern African Studies, vol. 38, no. 4 (2000), pp. 659682. 242. Meles added that “According to the latest rendition of the Boundary Commission, Badme would be 800 metres inside Eritrea. What’s 800 metres in a country as big as Ethiopia? What’s 800 metres compared to what we willingly and happily gave up as Eritrea? It’s nothing.” See “Interview with Prime Minister Meles Zenawi,” October 29, 2003, Integrated Regional Information Network: www.irinnews.org/report.aspx?reportid=46954. 243. President Isaias has repeatedly referred to those who fought and died in Eritrea’s war with Ethiopia as “martyrs.” Estimates are that as many as 100,000 Eritreans and Ethiopians were killed during the two year war. “Peace at Last,” African Confidential, vol. 41, no. 24, December 8, 2000, p. 8.

112

Civil War in African States

244. Alemu Tekeda, interview with the author, July 1, 2004. 245. As Meles remarked, “In five years, four of Eritrea’s five neighbors had been assaulted, attacked. The only neighbor which so far has not been assaulted is Saudi Arabia. It is anybody’s guess when and if this is going to happen.” Meles’s statement to the OAU on December 17, 1998 cited in Negash and Tronvoll, Brothers at War, pp. 65. In fact, Eritrea’s relationship with Riyadh had not always been peaceful either. In 1993, the EPLF accused Saudi Arabia of having a “hostile policy towards Eritrea.” During his 1993 New Year’s message, Issayas stated that Saudi Arabia should end its efforts to “sew chaos in Eritrea and in the Horn of Africa by using fundamentalism as a political tool.” See "Eritrea-Saudi Arabia: Tension Rises" Africa Research Bulletin, vol.30, no.2 (February 1-28, 1993). 246. Tsadkan, interview with the author, July 31, 2006. On this discussion, see also “Ethiopia: Arrests Starting to Thin Army Ranks,” Indian Ocean Newsletter, June 16,2001 and “Ethiopia: New Army Chief Named,” IRIN-Horn of Africa Weekly Round-up, No. 39: May 26, to June 1, 2001 . 247. Interview with TPLF official (July 2008). Notable in this respect was a US$660 million post-war reconstruction credit from the World Bank, which was conditional on Meles signing a peace deal with Eritrea. “October Evolution,” Africa Confidential, 42, no. 21 (October 26, 2001). 248. See “Arrests Starting to Thin Army Ranks,” Indian Ocean Newsletter, 16 June 2001. 249. Anonymous, interview with the author, July 20, 2008. 250. See Milkias, “The Great Purge,”, p. 21. 251. In its final report, the EU Election Observers Mission stated that the 2005 elections “fell short of international principles for genuine democratic elections.” See Final Report of the European Union Election Observation Mission (2005). The Carter Center, by contrast, stated that “the majority of the constituency results based on the May 15 polling and tabulation are credible and reflect competitive elections.” See Final Statement on the Carter Center Observation of the Ethiopia 2005 National Elections (September 15, 2005). 252. Martin Plaut, “High Stakes in Ethiopia Stand-Off,” BBC News—Online, 23 June 2005. 253. Memorandum on the Conduct of the European Union Election Observation Mission (EU-EOM) in Ethiopia, mimeo provided to the author. 254. See, for example, Lovise Aalen and Kjetil Tronvoll, “The End of Democracy? Curtailing Political and Civil Rights in Ethiopia,” Review of African Political Economy, vol. 36, no. 120 (2009), pp. 193-207. 255. Those held included former senior members of the government with close previous ties to Isaias, including Vice-President Mahmoud Ahmed Sheriffo, former Foreign Minister Haile Woldetensae, and former EPLF intelligence chief Petros Solomon. The government justified the arrests by claiming that these individuals intended to overthrow the government. See Amnesty International, “You Have No Right to Ask—Government Resists Scrutiny on Human Rights,” (May 2004), p. 7, and Africa Contemporary Record, 2001/02, p. 414. 256. Amnesty International, “You Have No Right to Ask,” p. 8. In 2009, for the third year in a row, the media watchdog Reporters Without Borders ranked Eritrea last (175rd place behind Turkmenistan and North Korea) on its

The Wars of Ethiopia and Eritrea

113

World Press Freedom Ranking. “President Afeworki and his small clan of paranoid nationalists,” it said in its 2008 ranking, “continue to run Africa’s youngest country like a vast open prison.” Ethiopia ranked 140th. 257. The EPLF-DP is now called the Eritrean Democratic Party (EDP). Amnesty International, “You Have No Right to Ask. . .” (May 2004), p. 3. 258. “Eritrean Opposition Groups Agree to Form Full Alliance,” Sudan Tribune, January 23, 2008. 259. Amnesty International, “You Have No Right to Ask,” p. 3. 260. “Eritrea,” Amnesty International Report, 2009. For an extensive report on the conditions inside Eritrea, see Kjetil Tronvoll, The Lasting Struggle for Freedom in Eritrea: Human Rights and Political Development, 1991-2009 (Oslo Center for Peace and Human Rights, 2009). 261. Africa Contemporary Record, 2001-02 (p. B423) states that the SPLA is one of the opposition groups operating from Eritrean territory. The other Sudanese group is the National Democratic Alliance (NDA) a coalition of rebel groups. 262. See Jeffrey Gettleman, “Eritreans Deny American Accusations of Terrorist Ties,” New York Times, September 18, 2007, p. A17. 263. Aregawi Berhe, “Ethiopia: Success Story or State of Chaos?,” p. 104.

3 Unification and Quasi-Secession in Somalia

And, mind you, not I, nor you, nor any other Somali can deflect the course towards our avowed ultimate goal—the reunification of all Somalis. Somali Prime Minister and subsequent President of the Republic of Somaliland Ibrahim Egal It has been demonstrated to us that, unless the Qurmis [Somalilander] and its supporters are subjected to a campaign of obliteration, there will come a time when they will rise again. Maj-Gen. Mohamed Saeed Hirsi (Morgan) The question of unity with the south is impossible. ... We are in the north. ... and our concerns concern the north, and we do not wish to be party to any old or new errors. The people of the north are what concern us. Somaliland Republic President Abdirahman Tour In late January 1991, the long-term Somali dictator Siad Barre was seen fleeing in a tank from Mogadishu, Somalia’s capital city. Two years later, on May 18, 1993, as the security situation in southern Somalia continued to deteriorate, delegates to a conference in the northern city of Burao declared the independence of Somaliland. Their precedent was Somaliland’s previous status as a colonial Protectorate of Britain and its five-day existence as an independent state prior to its union with Italian Somalia in 1960. The crisis and final collapse of the Somali state in 1991 and the reassertion of Somaliland’s sovereignty raise an obvious question: Why did Somalia, a state that sought to unify all Somalis and whose statehood

116

Civil War in African States

had been actively pursued by Somali politicians in both the north and the south, come to be seen by northerners as no longer desirable, and even to be regarded as threatening? This chapter will address this issue in four parts. Part one will provide a brief introduction to the region and to Somali political life. Part two will provide an explanation of why inhabitants of British Somaliland chose to join Somalis in the south in the formation of the independent Somali Republic—that is, why they chose a strategy of integration. Part three will account for the reversal of this sentiment among northerners and, specifically, what led them to want to abrogate their union with the south and pursue a strategy of separation from Somalia. Finally, part four will consider the experience of the “Republic of Somaliland,” an entity that is ostensibly autonomous from the rest of Somalia, but whose independence has yet to be formally recognized by the international community. Origins: People, Clan, and Geography

The Somali people make up a comparatively homogenous ethnic group who practice one religion (Sunni Islam), speak one language (Somali) and, in many cases, practice a pastoral means of subsistence. Yet despite the relative uniformity of faith, language, and livelihood there is diversity within the Somali clan system. Clan structure is highly stratified. Somalis may give their highest level of allegiance to one of five major clan families distributed throughout the Horn: the Isaaq (the predominant clan family in Somaliland), the Hawiye (Somalia’s most numerous clan), the Darod, the Rahanwein, and the Dir. Within each of these clan-families, lineage systems branch out into a complex array of clans and subclans. At various times, Western and Somali observers alike have commented on the precarious nature of pastoral life in the Horn of Africa, the way it has persisted, and the manner in which clan and family have provided security, order, and predictability in a demanding physical environment. Equally, however, observers note the way in which clans have been the basis for organizing violent confrontation within the modern state.1 This feature of social and political life is one factor which distinguishes Somalis from other African societies and states, and is also one of the bases on which Somali society itself has been divided. Importantly, the highest level of allegiance—the clan family—is not necessarily the most significant or stable unit of analysis. The identification of individuals with a more complex lineage below the clan family clarifies the fact that much of the violent conflict that Somalia and Somaliland have experienced—such as

Unification and Quasi-Secession in Somalia

117

the 1991 struggle for Mogadishu, and Somaliland’s 1992-96 civil war—has been within rather than between clan families. Surprisingly, it was not the harsh physical environment of the Horn of Africa—a relatively arid land with low annual rainfall and only two rivers—that independence leaders regarded as the greatest difficulty of Somali existence. “Our misfortunes do not stem from the unproductiveness of the soil, nor from a lack of mineral wealth,” observed Somalia’s first prime minister, Abdirashid Ali Sharmarkay. “These limitations on our material well-being were accepted and compensated for by our forefathers.”2 Nor was the clan system itself seen as a liability—despite the negative light in which it was frequently cast by outsiders. On the contrary, clan reflected a mechanism for coping with the environment in which they lived. In the pre-colonial Horn of Africa, no individual was without the “insurance company” of clan;3 but neither were individuals able to gain access to—let alone accumulate—sufficient resources to allow a concentration of political power. Male Somali society, then, tended to remain comparatively egalitarian rather than hierarchical, diffusing responsibility for individual well-being throughout the clans. “The advantage of this democratic totality among pastoralists,” observed the former Somali president Adan Abdulla Osman in 1962, “is that life is free from political tyranny, and, adapted to more modern conditions, their social system promotes a form of local government with which every male person can identify himself directly, and which secures for him an acceptable standard of human rights and justice.”4 The bane of Somali existence was, however, the manner in which these structures could be exploited and undermined by various external and internal forces. “Nothing is more dangerous than to influence in any way the ... balance of power between these tribes,” wrote John Hanning Speke, a 19th century European explorer of the northeast Somali plateau. “The imperfect attempt of strangers to interfere would be turned to the worst account by the designing adventurer and the turbulent spirit who expects to rise by means of anarchy.”5 Britain learned first hand of the risks of trying to manipulate clan organization to its benefit when its efforts to suppress uprisings in its territories (by distributing arms to friendly clans) met with failure. Contemporary observers of Somali political and social life agree that efforts to circumvent or tamper with the clan system offered more peril than promise for Somalis and interveners alike. As John Drysdale, the former Western advisor to the government of Somalia, has written, “Somali history has demonstrated that serious disorders have often been traced, not to any malfunction of

118

Civil War in African States

the Somali system of authority, but to the unimaginative application of alien systems of government which have inadvertently undermined it.”6 Indeed, for all its benefits, the Somali President also acknowledged how the social system to which he belonged could be manipulated “because the lack of any centralised political direction opened the door to eager partitioners during the scramble for Africa in the nineteenth century.”7 The colonial division of the Somali people into no fewer than five states—Italian Somalia, British Somaliland, French Jibuti (Djibouti), northern Kenya, and eastern Ethiopia—was regarded as particularly cruel. Drawing new borders around people not only politicized existing groups and created opportunities for clan and perhaps regional dominance within Somalia, but, as we shall see, also interfered with the ability of Somalis to get their livelihood. Not surprisingly then, decolonization stimulated a desire to remove colonial borders and perhaps defuse political tensions by way of a broader and, in some way, more natural, if idealistic, state of Somali unity. If the colonial powers had divided Somalis, then the independence era promised to provide Somalia’s leadership with opportunities to realize this ambition. But insofar as it allowed for the concentration of power and the political dominance of one clan or region over another, statehood also created new opportunities for abuse within Somalia. While there were earnest efforts during the early years of independence to be inclusive and to achieve unity through peaceful means, these years—1960 to 1969—were also marked by political uncertainty. This was relieved in some measure by the coup led by Mohamed Siad Barre in October 1969. But his coercive, although unsuccessful, approach to reclaiming Somalia’s “lost territories” and, in the wake of this failure, his ruthless effort to pursue a strategy of domination over challengers to his leadership led to crisis and ultimately to the collapse of the Somali state. From this, only the Republic of Somaliland emerged as a coherent political entity. There is no shortage of irony in the fact that an effort to unify a purportedly homogenous ethnic group instead led to Africa’s most extreme example of collapse and fracture. European Penetration and Conflict

While much is made of the divisions that colonial partition brought upon Somalis—and indeed on much of Africa—the motivation for unification of the northern and southern colonies can only partly be traced to a general desire to act in the name of Somali nationalism. To be sure, Somali politicians portrayed themselves as strong nationalists, and even

Unification and Quasi-Secession in Somalia

119

today Somalilanders acknowledge that the prospect of all five Somaliinhabited territories being together had been a critical motivation in joining the south in 1960.8 But at least with regard to the 1960 merging of British Somaliland and Italian Somalia, their motivations were more prosaic, an indirect consequence of the incompatibility between the mobility of nomadic Somali life and the permanence of the international boundaries that emerged with statehood. As we shall see, the principal grievance of the northern clans was that the political borders that colonialism had created across their seasonal migration routes forced them to enter, not Italian Somaliland, but rather Ethiopia. The first significant European intrusion into the Horn of Africa was made by the British at the time of the scramble for Africa at the end of the nineteenth century. European merchants had previously traveled to the Horn but none had established a permanent presence in the region. Great Britain’s interest was not in Somaliland per se but in India and the coaling station across the Gulf of Aden at Yemen. With the opening of the Suez Canal in 1869, the strategic importance of these coastlines increased. Earlier in the nineteenth century, Britain had concluded treaties with several Somali clans in order to secure shipwrecked British merchants from looting. Then a series of treaties were reached with local clans between 1884 and 1886 that provided migrating herdsmen with the “protection of Her Majesty” in order that they could gain access to seasonal inland grazing areas free from attacks by bandits. In return for the “independence” afforded by this protection, the Somali clans pledged not to make any new agreements with foreign nations or powers “except with the knowledge and sanction of Her Majesty’s Government.”9 Other European and non-European states were establishing their presence in the region as well. In the south, along the Benadir Coast, Italy was building its East African empire and, further west, France created a foothold at Jibuti (Djibouti). Aware of the increasing European interest in the region, Ethiopia also sought to assert its presence on the Horn. Various Ethiopian kings had long claimed to have conquered these territories, but their hold on them was seldom sustained or welldefined. The Ethiopian emperor Menelik was motivated by this vision of a broader Ethiopian empire and by the need to stand its ground against European intrusion. Formally, the British had established only a protectorate in 1887 rather than a colony, since, unlike other in regions in Africa, its interest in Somaliland was strategic and there was no need for a significant European settler community. Nor was there any need to pursue development initiatives among the Somali clans. Despite its skeletal

120

Civil War in African States

presence, however, maintaining the protectorate proved burdensome for the British. Chief among these burdens was the challenge presented by the anti-colonial Dervish movement of Sayid Mohamed Abdille Hassan, a Darod Somali and poet referred to pejoratively by the British as the “Mad Mullah.” In 1899, the Sayid temporarily occupied the town of Burao and began harassing the Habr Yunis and Habr Toljala—clans the British were pledged to protect.10 He also called on all Somalis to join him in a holy war against the colonial infidels. While agreements were concluded with the Sayid in an effort to curb his activities, on-going harassment of protectorate clans eventually forced the British to withdraw to the coast and to provide arms to Isaaq clans who would, it was hoped, keep the Sayid’s forces at bay. But this decision produced only inter-clan violence, which did little to counter the Sayid, and so the British chose to undertake their own armed adventure. This resulted in a crushing British defeat in 1913. Much to Britain’s embarrassment, colonial authorities could neither subdue the insurgency nor endure the political fallout that would come from withdrawal from the Protectorate. The Sayid’s forces were not defeated until 1920, after the conclusion of the First World War, when a major air bombardment was launched. The attack dispersed his fighters, but ultimately failed to kill the man himself. The Sayid’s defiance against foreign rule and his repeated ability to withstand British efforts to contain him elevated him to the status of a hero among some clans and nationalists.11 Most notably, the Sayid would serve as a mythical hero for Somalia’s late-twentieth century dictator Siad Barre, who also saw himself as creating a Somali state free of foreign influences. But the Sayid’s legacy was qualified in part because it was clan based. The Sayid lamented clan divisions among Somalis. Yet he was also frustrated by the resistance he encountered from some northern clans, particularly among the Isaaq—some of whom were enlisted by the British to defeat him and whom he dismissed in his poetry as being no better than the colonizers themselves. Not surprisingly, despite his one-time residence in the north as well as modern efforts to revere him as an early nationalist, the majority of the Isaaq did not support the Sayid or subscribe to his legacy of struggle.12 The interior limits of British protection remained undefined. Theoretically, the protection the British had pledged to the Somali clans extended over all the territory they inhabited, including their migration routes.13 But while migrating clans were obviously familiar with the grazing territories, no precise efforts were undertaken to document the limits of the territories they inhabited on their southern seasonal

Unification and Quasi-Secession in Somalia

121

migrations.14 Since these clans migrated into territory claimed by Ethiopia, clashes between the administering authorities were inevitable. However, the territory south of the British Protectorate was not, strictly speaking, Ethiopian. Italy had sought to establish a protectorate of its own in Ethiopia, and the first treaty concluded by Britain defining the border was with Rome in 1894. The result of Italy’s defeat at the Battle of Adwa in 1896, however, was that Ethiopia would now be recognized as an independent state with which foreign governments would need to deal directly. When Ethiopia rejected the Anglo-Italian Treaty of 1894 and insisted on a new agreement, Britain, eager to improve its standing with Ethiopia, was willing to comply. In a new agreement, negotiated in 1897, the British conceded to the Ethiopians territory they had negotiated earlier with Rome—an area to the south of the Protectorate known as the Haud and another area to the northwest of the Haud which was subsequently referred to as the “Reserved Area.”15 In making these agreements, however, the contracting parties were effectively drawing a line across the seasonal migration path used by Somali nomads. In this sense, the new boundary did not just separate a people who hitherto had had no such divisions, it threatened their way of life and thus, because it was so precarious, their survival. While the British negotiator, James Rennell Rodd, claimed to have avoided any formal recognition of Ethiopian sovereignty over these lands (the control of which by Ethiopian authorities was, in any event, doubtful), the very fact that the agreement explicitly delimited a frontier implied that territory on the other side of the border was Ethiopian.16 In order to address the herdsmen’s need for grazing lands, an annex was added to the 1897 Anglo-Ethiopian agreement which specified that “the tribes on either side of the frontier were free to cross that frontier for the purpose of grazing.”17 Predictably, objections were raised by the Somali clans. Those whose migration now required them to traverse an international boundary had voluntarily entered into the agreements of 1884 and 1886 seeking the preservation of their independence by way of British protection. Whether they had actually transferred their territory to the British was less clear. What was clear was that the Somalis had not been consulted by either party, and that the British had yielded territory to the Ethiopians in the 1897 agreement that was arguably not theirs to give away.18 In this sense, the British had betrayed the Somalis they had pledged to protect. From the British perspective, their Somali critics were unjustifiably unsympathetic to the fact that the British had been negotiating with the Ethiopians from a position of weakness and had, arguably, made the best

122

Civil War in African States

of a bad situation. Given that the whole Protectorate had become a liability worthy of only a minimal British presence and the fact that Ethiopia had its eyes on the entire region of the Horn, the 1897 agreement had possibly saved most Somalis from Ethiopia’s expansionist designs.19 As the British minister of state would later argue, “at that time we had not the forces in that area to resist an attack had the Ethiopians decided to press it.”20 Had the British not at least agreed to delimit the Ethiopian frontier and, instead, acquiesced in giving Menelik everything he desired, Ethiopia could conceivably have been considerably larger than it is and Somalis would be merely another ethnic group in the Ethiopian empire. In this case, despite the arbitrary and imperfect nature of African borders, the European scramble for African colonies had the perhaps inadvertent effect of protecting some, though obviously not all, Somalis from Ethiopian rule. It would not be the last time that the Ethiopian-Somali border would be regarded by one party or another as offering a life-saving refuge. This view, however, provided little solace for those Somalis whose survival still required them to migrate to grazing areas on the other side of the boundary. While most continued with their seasonal migrations beyond the Protectorate (and, in fact, continued to receive limited protection from the British), once they crossed the frontier they became formally subject to Ethiopian authorities, who on occasion sent military expeditions on tax-collecting missions. More worrying were appeals by Ogadeni Somalis to the Ethiopian government to close the border to Protectorate Somalis in order to reduce competition for the grazing lands in Ethiopia. Moreover, while the border offered security for Protectorate Somalis, they were occasionally subject to attacks from the south. Since the 1897 treaty allowed clans from both sides to move freely across the border, southern bandits guilty of harassment in the north could also find sanctuary back across the border, knowing that their actions would go unpunished by Ethiopian authorities. While the assurances offered by the Ethiopian emperor to respect migrating Somalis were in theory as good as those of the British, his failure to ensure Somali well-being reflected poorly on the British and the pledge they had struck in the 1897 agreements.21 Indeed, the difficulty Britain had encountered with the Sayid had been only the start of the challenges it faced in the Protectorate as a whole.22 After the Sayid’s death in 1920, there was optimism that at least some form of limited development could proceed in the Protectorate. But the unwillingness of the British to finance major development projects without local taxation, an almost complete absence of tradable resources or economic opportunity, a society that

Unification and Quasi-Secession in Somalia

123

was regarded by British officials as unreceptive to colonial manipulation, and an inhospitable environment meant that few serious development initiatives were undertaken within Somaliland’s colonial borders.23 Yet for reasons of honor, philanthropy, and strategic interest, the British public would not allow the abandonment of the Protectorate. Consequently, for most of the remainder of the colonial era, British Somaliland was run strictly on a “care-and-maintenance” basis, an approach that, if it did nothing else, “left the bulk of the Somali people untouched and uninfluenced by the colonial power.”24 In 1941, the defeat of Italian forces in Africa by Britain meant that Ethiopia and Italian Somalia came under British Military Administration (BMA), an event which in turn led to the restoration of the Ethiopian emperor, Haile Selassie, to his throne. Provisional agreements struck with the emperor in 1942 and 1944 allowed continued British administration of the Ogaden and the Reserved Areas in order to compensate for the still-fragile state of the Ethiopian government and to allow Britain to continue prosecuting its war against Vichy France in Jibuti. But the agreements also acknowledged that the presence of the BMA in these areas did not prejudice Ethiopian sovereignty over its territory.25 If these affirmations of Ethiopian sovereignty on the part of Britain appeared to confirm Britain’s view that the territory was Ethiopia’s, the Ethiopians remained skeptical. Upset with what it already perceived as Britain’s overbearing attitude toward it, Ethiopia feared that any gestures that continued to acknowledge a British presence or interest in this territory—whether in the form of British officials or Somali subjects of the British Protectorate—could subsequently be used against it or allow future British claims on the territory.26 Indeed, these suspicions received further support after World War II, when some British officials envisioned an ambitious proposal that would unite all Somalis of the Horn. In 1946 the British foreign secretary, Ernest Bevin, proposed the amalgamation of the three largest Somali-inhabited colonies under a single trusteeship. “In all innocence,” the minister stated, “we proposed British Somaliland, Italian Somaliland, and the adjacent part of Ethiopia, if Ethiopia agreed, should be lumped together as a trust territory, so that the nomads should lead their frugal existence with the least possible hindrances to interfere with their meager existence.” The proposal appeared to be motivated by the conviction that a return to pre-colonial normality could be achieved if Somalis were brought under a common authority and if those needing to graze their flocks and herds were not compelled to subject themselves to different jurisdictions. “All I want to do in this case is to give those poor nomads a chance to live,” Bevin

124

Civil War in African States

declared. “[T]o have these constant bothers on the frontiers when one can organize the thing decently—well, after all, it is to nobody’s interest to stop the poor people and cattle there getting a decent living.”27 Others thought differently. Russian and American officials were skeptical that any sort of British-inspired initiative was not merely British empirebuilding. Ethiopia was also unmoved. Ethiopian critics thought it was naive to expect that Ethiopia would cede territory for the sake of creating a “Greater Somalia” (particularly given that Britain had already repeatedly recognized these territories as Ethiopian) and resented the implication that such an initiative was the price Ethiopia would have to pay for restoring the emperor to his throne.28 In fact, the emperor now wanted Britain out of his territory altogether. Irritated by the ongoing British presence in the Reserved Area and the Ogaden, and by Britain’s apparent reluctance to unequivocally acknowledge what was believed to be theirs, Ethiopia formally asked that Britain return territories under BMA to Ethiopian administration. “After the war they kept on handing it back to us in bits and pieces until . . . we insisted on taking back what was left in their hands,” the emperor later remarked. “I cannot recount ... the trouble to which we have gone in regaining control of that part of our land.”29 Having already effectively acknowledged Ethiopia’s sovereignty over the contested areas in 1942 and 1944, however, Britain was now resigned to reaffirming that the 1897 agreement was in fact binding. Various efforts were made to make alterations that would have benefitted the Somalis, while still respecting Ethiopia’s rights—including extending the existing arrangement or leasing the Haud. But in the end Britain had no leverage to exact such changes and the negotiations proved fruitless. The 1954 agreement (published in 1955) allowed clans on both sides to continue crossing the border to graze their flocks and herds and even allowed for a British Liaison Officer to remain in the Haud and the Reserved Area. But, once again, it affirmed the “full and exclusive sovereignty of Ethiopia over the territories.”30 Independence and Unity for the Somalis

Somalis were, of course, indignant. For those Somalis hoping to remove the boundary between the Protectorate and the grazing lands in the south, the agreement proved to be the final insult: if any doubt remained as to the force and effect of the 1897 treaty, these had now been put to rest. One Somali who later served in the government of Siad Barre wrote that the handing over of the Haud to Ethiopian administration was “not

Unification and Quasi-Secession in Somalia

125

only an unmitigated disaster, but a betrayal. As a people we overwhelmingly lost faith in the protection treaties with the British. During the six years after this transfer and prior to independence, the very word protectorate became anathema to most Somalis.”31 Not surprisingly, the movement for independence and unity gained considerable momentum as a result. Somalis in the south had long been motivated by nationalist ideals of a unified Somalia, which accommodated clan and regional differences. “We Somalis are one in every way,” declared Haji Mohamed Hussein, the president of the Somali Youth League (SYL), the most prominent political party in the south in 1948. “We are the same racially and geographically, we have the same culture, we have the same language and the same religion. There is no future for us except as part of a Greater Somalia.” Somali politicians had, like sympathetic British officials, envisioned the elimination of colonial borders as a means to return to some measure of normality. “We want [unity] and the Somalis of the other territories also want it,” Haji Mohamed Hussein proclaimed. “By this union only can we have the opportunity to give full expression to our national spirit and work out our destiny as a nation of normal human beings.”32 Now, eight years later, to advance the cause of unity, Somali nationalists seized on the indignities meted out by their colonizers and neighbors. In February 1956, a prominent Somali advocate, Michael Mariano, broadcast over Somali radio his view that, counter-intuitively, the “great calamity” of partition could prove to be Somalia’s “greatest blessing.”33 From his perspective, the loss of the Haud and the Reserved Areas to Ethiopia was to be exploited insofar as it invigorated a Somali national consciousness and desire for unity. Despite its apparent hostility to the interests of migrating Somali herdsmen, Ethiopia too claimed that it had an interest in uniting the Somali people—provided that this “lost territory” was, as with Eritrea, kept within the confines of the Ethiopian empire. In January 1960, Emperor Haile Selassie told the New York Times that the international community should be aware that Italian, British and French Somaliland had all previously belonged to Ethiopia. It should also be remembered, he said, that Mussolini had gathered troops and materiel in Italian Somalia before he attacked Ethiopia. As for the prospect of ceding territory to Somalia, he said that the ethnic Somalis who inhabited the Ogaden region of Ethiopia “have made it clear they do not wish to join in a Greater Somalia. But they would welcome other Somalis joining them [as part of Ethiopia].”34 Four months later, the emperor repeated his preference for Somalia’s inclusion in Ethiopia. But he also noted the pressures on him

126

Civil War in African States

to close the border that were coming from Ethiopia’s ethnic Somali population. In May 1960, a delegation of Somali officials from the soonto-be independent Somali state had arrived in Addis Ababa requesting that His Majesty “help us by transferring the area in question to us when we become independent.” The emperor’s response was to present the delegation with a petition allegedly written by Ogadeni Somalis which urged the emperor to stop the terrorism and persecution that they had suffered at the hands of northern Somali herdsmen.35 “Taking advantage of the terms of the treaty, the Somalis of the British Protectorate have enjoyed water and grazing privileges within Ethiopian territory,” the petition observed. “However, these British subjects when entering Ethiopian territory, rather than feeling gratitude from your Majesty’s generosity for the rights they enjoy, have on several occasions killed, plundered and outraged the interests of Your loyal people.” Saying that they had “forewarned . . . of the harm that might result” when the most recent agreement with Britain was signed, the petition urged His Imperial Majesty “to ensure us our safety from alien rule by terminating the 1954 Treaty as soon as possible and by enforcing the prohibition of future grazing rights to tribes in the British Somaliland Protectorate.”36 In other words, not only did the petition call on the emperor not to transfer the land to the future independent state of Somalia but to prevent Somalis from crossing the frontier altogether.37 “You are not our enemies,” the Emperor told the visiting delegation in his denial of their request, “... but we wish to indicate how sacred we regard our land.”38 Though one might have had doubts about the authenticity of the petition and the sentiments of its authors, the Ethiopian regime did indeed have support among Somali-inhabitants of the Ethiopian Ogaden.39 From the perspective of Ethiopia’s ethnic Somalis, the Protectorate clans—notably the Isaaq, the Dolbahunte, the Mijertain and the Marehan—were seeking to strengthen and extend their rights to grazing lands and wells deep inside Ethiopian territory. Pressing the regime to enforce a ban was a means for Ogadeni Somalis to defend themselves against these incursions from the Protectorate. For the Ogadeni—some of whom had been physically displaced by these northern intruders—their principal worry was that any cession of land, even if it was in the name of a “Greater Somalia,” would come at their expense.40 While compromise between the clans of the Ethiopian Ogaden and the Protectorate was possible in theory, any measure that served to accommodate the interests of one party was perceived to directly jeopardize the interests of the other. The concerns of Ethiopia’s ethnic Somalis served the emperor’s interests as well. While the prospect of another large minority within its

Unification and Quasi-Secession in Somalia

127

borders would certainly have created its own problems, the Ethiopian government’s priority was to put its security against outside threats over the internal threat of rebellious or feuding Somali clans.41 The emperor thus declared that a united and independent Somalia was merely a front for British interests and a “revival of the old Bevin plan.” To reassure Somalis that he had their interests at heart, the emperor observed that, as the only country in Africa that had never been colonized, only Ethiopia could reliably safeguard Somali interests. It was not, the emperor claimed, that the Protectorate borders served to secure Somali nomads from Ethiopian expansion, as the British alleged; rather, it was the Ethiopians who would protect the independence of Somalis from the British. “When Syed Mohamed Abdulla Hassan, in fighting for the independence of Somalis, was defeated,” the Emperor told the Somali delegation, “he took refuge in this country.” The Ethiopian foreign minister added, “It is not the intention on the part of Ethiopia to stop your tribes from grazing beyond your boundary, but they should do so as other people in Ethiopia. This can be taken as the considered official view of the Ethiopian Government.”42 For many Somalis, however, true independence would not be achieved with pledges by the emperor of Ethiopia to respect their wishes and sovereignty. Nor did the Somali delegates believe the emperor’s claims that all ethnic Somalis would prefer an existence within Ethiopia to one within a united Somalia. And in spite of his allusions to the 1930s, when Ethiopia had been abandoned by the League of Nations and invaded by Italy, Somali delegates also did not accept that the emperor truly distinguished the desire for Somali unity from Ethiopia’s need for a buffer territory from outside attack. In short, they thought that Somali unity within Ethiopia would serve nothing more than Ethiopian security interests. The struggle for the Haud was critical insofar as its loss would throw the construction of any sort of independent Somali state into question. As the Times of London wrote in 1956, if the Ethiopians succeeded in acquiring the Haud, “it will be only a matter of time before Somaliland comes under their control; for British Somaliland cannot survive without the Haud, and without British Somaliland there can be no hope of a federated and independent greater Somalia.”43 Now the immediate problem for the Somalis was no longer the British, who were now receptive to the idea of self-government, but the Ethiopians, whom they saw as deliberately working against Somali interests. “I have interpreted the Emperor’s attitude as a deliberate scheme to worsen the situation, and goad the new Republic into a pugnacious attitude by forcing them into a position of helplessness while their subjects are

128

Civil War in African States

being maltreated across the boundary,” wrote the Somali delegation after their visit to the emperor. The Somali delegation’s visit to Addis Ababa also revealed increasing Somali frustration with diplomacy as a means of advancing Somalia’s foreign policy interests. “I found out during my short stay in Addis Ababa that the Ethiopians are inclined to misinterpret any timidity or over-caution as subservience,” wrote the mission’s scribe. “I feel that they should be made to realise from the beginning that here is a political machine with which they will have to reckon. This does not necessarily mean that we should take an aggressive line, but a little determination and go-getting, as it were, would I think soon straighten the Ethiopian attitude.”44 In fact, as Somaliland’s independence approached, the inability to make progress on gaining access to the Haud led Somali nationalists to advance the cause of Somali unity where it could be achieved—namely with Italian Somalia. But while Somalis may have felt a certain cohesiveness in their common identity, there were other political, physical, and developmental barriers that worked against the realization of a sentimental nationalism. To be sure, many northern politicians espoused nationalist ideals, but the demands of ordinary Protectorate Somalis tended to be limited to the Haud region only, and to be annexationist rather than nationalist.45 In the years prior to independence, Somalis had no written script and had little in the way of infrastructure with which to bring Somalis in distant parts of the expansive territories together. A wedge of Ethiopian territory separated many Somalis in the north and south, and the roads, air links, and telecommunications and postal service of the two colonies remained rudimentary. Perhaps most significantly, the two Somalias—British Somaliland and Italian Somalia—were coming from separate colonial traditions with different colonial languages. Accounting for Somaliland’s Integrationist Strategy

The frustration over the so-called “Haud fiasco” did, however, have a unifying effect. Nationalists became convinced that unencumbered access to these territories could be reacquired only if Somalis were united and, therefore, in a stronger position to bargain with their neighbors. “The hard truth is probably that they have not yet formed any clear feelings of nationality,” wrote one Ethiopian critic soon after the union was consummated in 1960. “Their primary concern is for water and grazing land, and it is hard to expect them to possess national loyalties or to recognise frontiers that restrict their movement in search

Unification and Quasi-Secession in Somalia

129

of the elements of their lives.”46 But in retrospect Somalis too would see the events of 1954 as the watershed that led to union with the south in 1960, insofar as it instilled a belief that only an independent and unified Somalia could provide the clout to reverse this injustice. “[H]ad it not been for the handing over of those territories, so vital to the transhumantic life pattern of herdsmen and herds alike,” wrote Jama Mohamed Ghalib, a Isaaq minister in Siad Barre’s regime, “there would hardly have been the emotional desire for an immediate union with the South, nor even the demand for immediate independence.”47 In this sense, integration with the south was driven by a personal affront to Somali dignity inspired by the 1954 betrayal. For the most part, however, ordinary Somalis of the Protectorate had an immediate and vital interest only in the lands of the Haud and the Reserved Areas, and were less committed to the ideal of a “Greater Somalia”—which would have encompassed the larger Ogaden desert—except as part of a more remote ideal. As a result of the frequent and sometimes violent encounters with the Ogadenis, the possibility of competing with them within the same state for the foreseeable future was most unattractive. The desire to act quickly and make progress on this process of union, however, compromised their ability to reach sustainable terms with the south. A constitutional conference was convened to address the multitude of issues that needed to be reconciled prior to union. British Somaliland’s independence came on June 26, 1960 (at which time it received recognition from thirty-five states); it would be followed, five days later, by Italian Somalia.48 On that day, July 1, 1960, the north was also united with the south to form the Somali Republic. As things turned out it was perhaps ironic that the north pressed hardest for union. In the south, some legislators hesitated and asked that the union be delayed for several months to allow outstanding unity issues to be addressed. Those in the north, however, feared that any postponement would be to their disadvantage in a subsequent federation and insisted that the union move ahead. In the haste, northern interests were not adequately safeguarded. “At the time,” writes Jama Mohamed Ghalib, “I noted with alarm that the people of Somaliland had forced the union upon the South so precipitously that they alone had to pay the price by accepting a southern constitution, southern flag, southern capital and a southern Head of State—who also appointed a southern Prime Minister.”49 When the first national government was formed, southerners were assigned the key ministries of foreign affairs, the interior, finance and commerce and served as commandants of the armed forces. Now known as the “Northern Regions,” Somaliland received 33 seats in the new 123-seat

130

Civil War in African States

national assembly. Somaliland’s prime minister was given the comparatively junior post of Minister of Education.50 In the end, the final preparations for political union had been so hastily convened and independent political consciousness was so undeveloped that there was no opportunity to foresee the incompatibility of interests that was bound to emerge between the two former colonies. Ironically, as Saadia Touval has written, had more extensive negotiations been held in advance of the union, they would likely have led to the crystalization of opposing interests and thus prevented unification altogether.51 Touval found the union all the more remarkable insofar as it was seen as an apparently conscious and deliberate decision on the part of northern leaders to reduce their own political standing and power. “It was undoubtedly obvious both to the [Isaaq] in the North and to the Hawiye in the South,” he writes, “that their relative weight and political influence were bound to decline if the two territories were merged into a unitary state . . . Assessed in light of normal considerations of political interest—namely, that groups normally strive to preserve or increase their influence—the decision to form a unitary state therefore seems surprising.”52 One might wonder what accounted for this apparent action of selfsacrifice. Touval’s explanation points to the political leaders’ “commitment to nationalist ideals.”53 But clearly other factors were also at work: first, the British had reneged on their promises of protection; second, Somali land was seen as having been sold out from beneath them, and diplomatic efforts to get their land back had proved to be ineffective; and, finally, Ethiopia continued to be an obstacle not only to the seemingly reasonable nationalist demands of Somali leaders but to the very survival of northern Somalia’s nomadic population. There was a sense that, if Somalis did not build their state or at least hold their own, they would be consumed by others, most notably Ethiopia. The conviction emerged that only unity and independence would provide Somalis with the strength to reacquire Somali-inhabited territories. If imperialism had divided Somalis, then independence would be the means to achieve unity. Merging the north with the south was going to be the first step on the way to this unity and reassertion of Somali interests. Given the much-diminished authority with which northerners were left, however, they perhaps should have exercised greater caution. While it was true that there was a genuine desire for unification among virtually all Somali elites, there was also much that was impulsive about this integration strategy, and little time had been spent considering the perils of such a hasty union. There was as yet no unpleasant experience

Unification and Quasi-Secession in Somalia

131

with the south to give anyone pause on the merits of a separate existence and, from a nationalist perspective, much that was seductive about the idea of unification. This sentiment would of course change in time. As the human-rights watchdog Africa Watch would later observe in their account of the atrocities inflicted by the Siad Barre regime against the north in the 1980s, “The failure to negotiate any preconditions and to foresee potential conflict of interest [at the time of independence and union] would in time come to haunt the north.”54 Civilian Rule and Somalia’s Undoing

It may be derisory to consider the “undoing” of Somalia so soon after union and independence. But in the longer term, the creation of a larger Somalia would not in fact serve the interests of peace and prosperity as many Somalis had expected. For some years after independence, efforts continued to be directed towards uniting Somalis and building Somalia. “Not I, nor you, nor any other Somali,” stated Somalia’s prime minister, Mohamed Haji Ibrahim Egal, “can deflect the course towards our avowed ultimate goal—the reunification of all Somalis.”55 The irony was that Somalia’s imperfect assets—most notably a comparatively free-flowing, if disorderly, democracy and an often-times pragmatic leadership that demonstrated a grudging willingness to make peace with its neighbors—were increasingly regarded by Somalis as liabilities that needed to be addressed. But the remedies would prove even more costly than the ailments. The quick fixes that initially were undertaken to make Somalia stronger, less corrupt, and more united—a bloodless military coup in 1969 and the subsequent authoritarian regime, and an increasingly assertive and uncompromising foreign policy which culminated in the 1977 invasion of Ethiopia’s Ogaden region—obscured and were sometimes merely compensation for what was in fact the ongoing weakening of the state. Unhappiness in the north was in many respects a consequence of the same processes and problems experienced by all Somalis. But without meaningful or effective political institutions through which to channel their grievances and the military regime’s proclivity to look to coercion in an effort to subdue what it perceived to be a challenge to its authority, the union between the two former colonies came to be regarded by Somalilanders as a catastrophe that needed to be reversed. From afar, Somalia under civilian rule appeared to respect democratic principles, and, while its survival hinged precariously on the ability of its clan representatives to negotiate compromises on various clan demands, for nine years Somalis clung to its parliamentary system.

132

Civil War in African States

The sometimes-chaotic politics did, however, give rise to a stability which, even in the midst of ongoing border conflicts with Ethiopia, allowed for two elections (1964 and 1969) and much-admired freedom of expression. Complementing Somalia’s democratic progress, Zambian president Kenneth Kaunda remarked, “I have known no other country in Africa where former Prime Ministers or Presidents are not only put in a position to continue to be influential by expressing freely their views either in favour of the current government or the opposition, but also who have, in effect, no cause to fear for their lives as they continue to make this contribution.”56 In time, a “triumvirate” of politicians divided the country and control of its political candidates into three spheres of influence, on the understanding that Mohamed Haji Ibrahim Egal, an Isaaq from the north, would preside over the former British Protectorate, while the south would be governed by Somalia’s president, Abdurashid Ali Sharmarke, and the minister of the interior, Yassin Nur Hassan.57 But the union that had coincided with Somalia’s independence vastly increased the scale of the political arena, much to the detriment of the democratic process. Northerners and their problems were now far away from the center of decision-making in Mogadishu, and the traditional elders, who had previously managed local issues, were replaced, in the name of democratic convention, by elected representatives who did not benefit from the same degree of oversight that local politics had allowed.58 While the Italian model of parliamentary rule was followed to accommodate Somalia’s clan system, in reality political representation proved to be different to democratic participation. The complicated system of proportional representation exacerbated clan divisions, and the many weak and small political parties were less a platform for aggregating interests than a means for clans to get illegitimate access to the resources of the state. “While western analysts focused on the process of competitive elections as proof that democracy was ‘alive and well’ in Somalia,” wrote one scholar, “in fact the opposite was true. The multiple parties fissured the very foundation of the state; clan parties fractured into sub-clan parties, then extended-family parties, and finally into one-man parties.”59 Indeed, from the first election in 1960 to Somalia’s tumultuous last election in 1969, the number of political parties ballooned from 5 to 62, while the number of candidates contesting the 123 seats was over 1,000. Democratic failings were coincident with an inability to make progress on Somali-unity. If agreements were reached with neighboring countries which did not ultimately secure Somalia’s “lost territories,” Somali politicians risked being seen as betraying the cause. But efforts to secure these lands also ran up against the general consensus in

Unification and Quasi-Secession in Somalia

133

Africa—as manifested in the charter of the newly created Organization of African Unity (OAU)—that Africa’s colonial borders needed to be left in place. Indeed, Somali leaders had long pledged that progress on unification could be achieved through peaceful means. And, following a clash along the Ethiopian-Somali border, some politicians maintained that the Somali economy, struggling as it was, could not sustain the military expenditures that already consumed 30 percent of the government’s budget.60 But, increasingly, there was impatience with diplomatic efforts. In 1963 Britain ignored the findings of a commission it had established to ascertain the wishes of the inhabitants of the Northern Frontier District and, to the chagrin of Somalis who lived there, incorporated it into the independent state of Kenya.61 The subsequent electoral defeat of “unionists” in French Somaliland in 1966 and the Arab defeat in the Six Day War left Somalia with few allies. These realities led the newly appointed Prime Minister Egal to consider a more moderate policy which, to the relief of Somalia’s neighbors, downplayed Somalia’s territorial ambitions. But while Egal himself and his “new look” approach to normalizing relations won praise in foreign capitals, the 1967 “Arusha Memorandum of Understanding” he forged in order to bring about peaceful relations with Kenya was subject to criticism at home. There, the returning prime minister was met with protests, for abandoning the goal of a “Greater Somalia,” and opposition MPs accused him of “selling out” and of “compromising with imperialism.”62 Military Rule

By the time the military stepped in on October 21 1969, Somali politics had descended into near anarchy, with nepotism, corruption, and occasional violence as the most notable and distasteful aspects of the most recent elections. Indeed, as one minister in the subsequent military regime stated, there was “no shortage of areas to be capitalized on as far as the misdeeds of the old regime were concerned.”63 A failed democracy made the transition to one-party rule easy to justify. Referring to the deposed prime minister as the “bloodsucker of the people,” the coup leaders claimed that Somalia’s corrupt politicians “had endangered the very existence of the nation.”64 By contrast, the newly created “Revolutionary Council,” under the leadership of General Mohamed Siad Barre claimed that, domestically, its plan was to “speedily . . . wipe out the bad and mischievous activities in many sectors of public life in the Civil Service, and to instil unity among the people and eradicate tribalism.” In its first statement on foreign

134

Civil War in African States

relations, the Council also pledged to defend “the people’s right of selfdetermination” and to provide “active support and assistance to the liberation movements whose peoples and countries are under illegal occupation.”65 The order that the military brought to Somali politics cast into sharp relief the disorder that had discredited and deadlocked the previous parliamentary system.66 The response among ordinary Somalis was positive. Jama Mohamed Ghalib writes, “In the initial stages, Siad Barre’s activities were impartial and the general feeling was one of euphoria. For those who had not previously known Siad Barre, he became a savior, while most of those who had known him tended to give him a chance and support his programs of nation building.”67 The new regime was fortified by young, educated technocrats, who, while sensitive to clan representation, were allegedly selected on the basis of merit rather than clan. The coup delivered many of the assets that observers of military rule had long promised: a disciplined and vigorous government that did not need to concern itself with the corrupt politics of clan and which, in turn, could act upon issues that had long eluded the previous civilian government. “One of them,” writes Ghalib, “was the writing of the Somali script. The other was Somalia’s membership of the Arab League. Neither could have been realized without a dictatorial regime, for both lacked any common parliamentary consensus during the previous civilian governments.”68 On the first anniversary of the coup, the regime declared Somalia would be governed by “scientific socialism,” a doctrine that was also, conveniently, above the politics of clan. Then, in 1971, Siad Barre banned tribalism outright, claiming it was associated with nepotism and corruption. As it increasingly espoused a socialist program of action, the new regime was praised by foreign intellectuals for the changes it promised to bring. This was not merely a revolution against the old European model, wrote Basil Davidson in a 1975 article, but a “revolution against all those aspects of Somali structure and custom, developed through the pre-colonial centuries and persisting through the colonial and ‘neocolonial’ phases, that had long become handicaps . . . on progress toward egalitarian modernization.” He added that, after October 1969, the Somalis had taken “charge of their own history once again, and resumed the task of making it.”69 For Davidson and others, the revolution was more about democratization than domination. Since in subsequent years the revolution proclaimed advances in the rights of women, launched a massive literacy campaign amongst its nomadic population, and exchanged corruption for “mass participation in public affairs,” the

Unification and Quasi-Secession in Somalia

135

pragmatic and orderly rule of generals and soldiers was no further from democracy than anything civilian rule had afforded. The bloodless nature of the coup itself suggested that the regime’s writ was welcomed in all parts of the country, including the north, and therefore would not rely on coercive means. Beneath the façade of order and stability, however, was a sterner core. The president’s colleagues within the Supreme Revolutionary Council would come to speak of Siad Barre’s “insatiable’ drive for power and dominance.” In the years after the coup, his rivals—including those who had participated in the coup but who came to represent possible threats to his rule or who questioned his commitment to democratic renewal—were publicly executed.70 Siad Barre’s increasingly autocratic ways coincided with efforts to deify him as the “Father of the Nation.” Jama Mohamed Ghalib would later state that Siad’s view of those around him was “ever Machiavellian.”71 Somalia’s Failed Invasion of the Ogaden

From the outset, Somalia’s neighbors were apprehensive about Siad Barre’s coup in Mogadishu. Tensions, which re-emerged in June 1969 during one of the Ethiopian government’s periodic efforts to enforce a tax-collecting scheme among Somali nomads, were in no way assuaged by the fall of Somalia’s civilian government. Ethiopia and Kenya feared that, having just deposed the Somali politician who had in the face of considerable public pressure espoused a policy of moderation with regard to its “lost territories,” the new regime would now seek to distinguish itself from its predecessor by adopting a more aggressive approach to the Somali question. Indeed a Somali government spokesman’s assertion that “the Revolution will exert every effort for unification of the Somali people” was rendered only slightly less ominous by the qualification that it would do so in a “responsible manner.”72 What allowed the Siad Barre regime to contemplate acting on its foreign policy ambitions was the increasingly close relationship it had forged with the Soviet Union. Somalia had received steadily increasing amounts of Soviet military aid during the 1960s, but this level of assistance was dramatically expanded following Siad Barre’s 1969 coup and his embrace of socialism. Siad’s total military manpower ranked Somalia (a country of three million people) among the top five of Africa’s fifty-three states.73 With its Soviet-supplied equipment — including two-hundred tanks, three-hundred armored cars and personnel carriers, two-hundred heavy guns and SAM missiles, sixty MiG fighters

136

Civil War in African States

and a squadron of Ilyushin light bombers — Somalia could claim to have the finest military force in sub-Saharan black Africa.74 Soviet officials disregarded the regime’s assistance to the Western Somali Liberation Front (WSLF), an increasingly active secessionist movement comprised of the same clan as the president’s mother and operating in Ethiopia’s Ogaden desert. The introduction of Somalia’s own regular troops into the Ogaden in June 1977 in support of the WSLF marked a change from the deference and diplomatic niceties extended to the Ethiopian emperor by the Somali delegation in 1960. Emboldened by its newly acquired force capabilities, the regime was not willing to seek the sort of trade-offs with its neighbors, a diplomatic tactic that might have allowed it to acquire sympathetic allies. Nor did it seek to rationalize its actions by looking to the one issue that might have focused international attention on a compromise agreement: the fact that the grazing lands in the north of the Ogaden (the Haud) were essential to the survival of its northern people.75 Instead it claimed that its grievance was Ethiopian colonialism and Somali self-determination, an issue which, if taken to its logical conclusion, would have undermined virtually all African states. Before the UN General Assembly, the Ethiopian foreign minister Feleke GedleGiorgis seized upon his country’s role as a victim of Somali aggression and played directly to the principal security concern of all other African countries. “If Somalia’s attempt to annex eastern Ethiopia is based on the assumption that there is a Somali minority within Ethiopia,” he said, “then such reasoning . . . is apt to create a dangerous precedent exposing Africa, and perhaps even the rest of the world, to the forces of violence and instability. The fact remains that there is not a single African country which does not have distinct ethnic groups straddling its borders.”76 Not surprisingly, no African state came to Somalia’s defense. Indeed, it is not clear what prompted Siad to take such a catastrophic gamble, other than the hope of taking advantage of the perceived ongoing political uncertainty in neighboring Ethiopia—a result of the overthrow of Emperor Haile Selassie. Implicit in the comments of one of the president’s officials, Jama Mohamed Ghalib, were doubts about whether the Ogadeni Somalis in Ethiopia really intended to be united with Somalia. At least among the ethnic Somalis living in “capitalist Kenya,” he said, there was no longer a desire to be part of “socialist Somalia.” This sentiment was underscored by a 1977 referendum in which Djiboutians had also voted to go their own way rather than unite with Somalia.77 Other observers suggested that an invasion was not part of Siad Barre’s plan, but rather the plan of his

Unification and Quasi-Secession in Somalia

137

overzealous generals, forcing the president to follow rather than lead his troops to battle.78 Still others speculated that Siad Barre was motivated less by a desire to liberate the Ogaden than by a need for Ogadenis to increase his power base and strengthen his rule.79 In any case, Somalia’s headlong and ill-conceived rush into the Ogaden was doomed. Moscow could not afford to be associated with a state that had broken the sacred African trust not to invade another territory and, on November 13, 1977, began delivering significant quantities of military hardware to Addis Ababa. For his part, Siad Barre scorned Moscow’s adulterous behavior and announced that all Russian advisors would be expelled. Soviet arms had given Somalia enough power to overwhelm Ethiopian defenses during the early stages of the invasion, yet Somalia remained dependent enough that, in 1978, Somali forces were routed once the Soviet Union switched sides. The political capital that the president gained among his own people for such a bold and independent foreign policy initiative was soon undermined by the broader fall-out from the offensive’s failure. Indeed, the humiliating blunder in the Ogaden ended any hope for a Somali state uniting ethnic Somalis on the Horn of Africa. The defeat would set in motion events and processes which would ultimately lead to the collapse of the Siad Barre regime. Disillusionment in the North: Second Thoughts on an Integrationist Strategy

Indications that Somalilanders were not satisfied with the union of the two Somalias had manifested themselves prior to Siad Barre’s coup and the war in the Ogaden. In June 1961 Somalis had the opportunity to vote to approve Somalia’s new national constitution but, unlike the southerners, many northerners either boycotted the referendum or voted against it by a significant margin.80 Six months later, in December, eighteen northern army officers, nearly all of them Isaaq, took their southern superiors hostage and declared a military takeover of the north. The original objectives of the disgruntled officers were related to the post-independence process of integrating military units into the Somali National Army. But their actions—limited to the north as they were—implied to some that the officers were bent on secession. While the effort did not succeed, the results of the referendum suggested to these officers that they had broader support from the dissatisfied north. The former British Somaliland did settle into a more amicable relationship with the south in the years after independence—aided in part by the balance Prime Minister Egal brought to the north-south relationship. However, observers saw these two events as the beginning

138

Civil War in African States

of northern dissent towards union with the south.81 As with all Somalis, northerners had found the allegations of elite corruption and vote rigging sufficiently distasteful that even they acquiesced to the inevitability of Siad Barre’s 1969 coup against the increasingly unpopular Egal government.82 The military regime’s decisiveness, initially valued by many Somalis, insofar as it reversed the country’s decline into political anarchy, was in part a consequence of the extensive security apparatus created in the years after the coup d’etat. The organizations that comprised this apparatus not only facilitated political order but they also allowed Siad Barre and, more importantly, his clan, to build and secure a power base from which he could exercise control and domination over other clans. The components of this security apparatus included the notoriously ruthless National Security Service (NSS) led by Siad Barre’s son-in-law (and aided by specialists from the German Democratic Republic), various Regional Security Committees, the Guulwadayaal Militia or so-called “victory pioneers,” and the dreaded military intelligence unit or Hangaash. Less than a year after the coup, Siad Barre’s had his first Vice President arrested and, two years later, three other senior Ministers or Revolutionary Council Ministers were arrested and executed by firing squad. While there were various rationalizations offered for the executions, many interpreted them as warnings to the clans to which the latter three belonged (Abgaal/Hawiye, Habr Yunis/Isaaq, and Majerteen/Darod) not to challenge the rule of the nowundisputed leader Siad Barre. Later, in the aftermath of Somalia’s 1978 retreat from the Ogaden, eighty commissioned and noncommissioned officers were also executed by the regime because of, in the regime’s words, “their opposition of the way the war was handled.” Not surprisingly, others interpreted the executions as a warning to members of the armed forces to desist from any further resistance against the regime.83 Although the regime’s overthrow did not come for several years after, for many, the war with Ethiopia was a turning point. “By the end of the Somali-Ethiopian war of 1977-78,” writes a founding member of the northern insurgent group that would work to unseat Siad Barre, “it became evident to all who could think clearly that the continued existence of this regime undermined the future existence of the nation itself.”84 Despite the regime’s efforts to intimidate potential opposition, the first move against it was made as early as April 1978. After a failed coup attempt, the regime’s anger was directed against the Majerteen clan of the Mudug region, from which the coup leaders had come. In response, those who eluded capture formed the Somali Salvation Front

Unification and Quasi-Secession in Somalia

139

(SSF), which soon after launched a guerrilla campaign against the regime in central Somalia.85 The Hardening of Northern Sentiment Toward the South

While the Isaaq did not hold Siad entirely responsible for the debacle in the Ogaden, the result did not enhance their president’s standing among them. Northern officers claimed that Siad Barre had been too greedy. Some believed that the Ethiopians had been prepared in September 1977 to sign a peace agreement that would have secured for Somalia the Ogaden up to the town of Jijiga and allowed for the recovery of the allimportant grazing lands of the Haud. Instead, hubris led the regime to advance on Harar, a town which would have secured valuable qat and coffee-growing areas, but which, as Somali troops soon found out, the Ethiopians (and their Russian patron) would not concede without a determined fight.86 The experience in the Ogaden would have other consequences for the northern Somalis’ goals in the future. Beyond the misguidedness and incompetence of the president himself, many Isaaq were once again reminded that the broader international community had not acquiesced to Somali demands for the incorporation of other Somali-inhabited lands. Unlike Eritrea, Somalilanders do not disregard their previous desire to join with the south. But the north-south union had always been part of a greater vision—the unification of all Somali-inhabited lands—and when the possibility of that vision disappeared, so too did the Somalilander’s need to unite with the south.87 Even during the war, northern officers felt that they had been undermined by a southern-based military hierarchy and had been manipulated by southern interests.88 This experience was a preliminary step in pushing northern Somalis toward a secessionist strategy. But for northern elites, another lesson of the Ogaden was that, because secession too involved a border change, it was a goal unlikely to be realized and, consequently, one that should be set aside. The shorter-term problem for the northern Isaaq, however, was not merely the failure in the Ogaden but the manner in which it set in motion processes that reflected and generated further resentment towards the south. The regime’s military defeat led to an influx of hundreds of thousands of Ethiopian Ogadenis into the north in 1978 and 1979. While they ostensibly came as refugees—most of whom no longer felt welcome in Ethiopia—it was soon clear that the Somali government regarded them as settlers, who could displace northern Isaaqs. Since the refugees were without a home and were, by-and-large, supporters of

140

Civil War in African States

Siad Barre, they and the militias they eventually formed became instruments of the regime. The arrival of the refugees from the Ogaden and the more favorable treatment they received created hostility among the Isaaq, not only towards the Ogadeni clan, whom they regarded as taking their land, but also towards the regime that had encouraged them to come. In the meantime, the presence of several hundred thousand ethnic Somali refugees from Ethiopia led to an equally vast influx of humanitarian assistance. While their arrival presented enormous challenges for the Somali state, these refugees and the deluge of foreign aid workers who came as a consequence were also exploited by the regime to help prolong its rule and sustain its soldiers.89 An internal report published by the World Bank suggested that between 1981 and 1988, between one-quarter and one-half of the government’s expenditures were “unidentified,” suggesting that they had been embezzled by government officials. Most of the government’s revenues came from foreign sources, and the government did not hesitate to ask donors—particularly the United States—for more money. By the mid1980s, seventy-two percent of total government spending was financed from abroad.90 Most of the revenue that could be accounted for was directed towards financing the Somali military: “For all intents and purposes,” the report concluded, “Somali ordinary budgets in recent years have been financing only defense and general [security] services.”91 In fact, for many the regime’s efforts to manipulate refugee movements was directly related to the opportunities for corruption that resulted. “In my opinion,” observed one expatriate official associated with the World Bank, “the oppression in Barre-time . . . had one single purpose: to control aid. . . . Since the country had virtually no productive basis, the only way to keep the apparatus going was to plunder aid.” By manipulating the flow of assistance, the regime was able to sustain a vast clientelist network. “Those Somalis who were outside of the cleptocracy [sic] saw with great envy and anger the open racketeering and plundering of people and the aid machine,” wrote the official. “Many of them were utterly amazed about the ease with which aid people handed out money and goods to those who kept the country at ransom. . . . Only the monopoly on power and aid of Siad Barre and his cleptocracy kept them out.”92 Among the northern Somalis, the bitterness generated by the regime’s corruption was surpassed only by resentment at the apparent favoritism shown to the Ogadeni at the expense of the Isaaq. As one foreign worker cited by the human rights group Africa Watch commented, “the fact that the government was

Unification and Quasi-Secession in Somalia

141

using the refugees both to get aid and to weaken the Isaaks (sic) politically was a major problem.”93 The regime’s policy in the north was undoubtedly driven by the opportunities to engage in corruption that were provided by the refugee crisis. But the identity of those most affected, the Isaaq, was attributable to existing tensions between the Isaaq and an emerging MarahanOgadeni-Dulbahante (MOD) clan alliance that was at the center of the regime’s power. Despite the official ban on “tribalism,” clan had remained central to Somali political life. For the president, there were doubts about the sustainability of the regime’s military superiority, and Siad Barre had covertly looked to these ties of loyalty to fortify his rule. For ordinary Somalis, however, the absence of all representative institutions and the transformation of state institutions into instruments of oppression meant that clan was the only safe retreat and the only means with which to organize and mobilize opposition to the regime.94 Despite his own reliance on clan support, Siad Barre liked to portray himself in the tradition of Sayid Mohamed Abdille Hassan (the “Mad Mullah”) with respect to his emphasis on Somali nationalism, a rejection of colonialism, and his resentment over how the latter had exacerbated clan divisions.95 He maintained that the Sayid and his broader Dervish rebellion had been defeated by British colonial power with the active support of the Isaaq—a claim that endeared neither the British nor the Isaaq to the president. And, like the Sayid, the president was convinced that imperialist forces and clanism were undermining revolutionary progress. The revolution, he said, was “like a bus from which everyone gets off somewhere; those with personal interests get on or off en route, while true, patriotic Somalis and martyrs remain on board at all times.” Those opposed to the regime, he said, “were all along intent on the nation-wrecking activities in complicity with the colonialists.”96 In this respect, Siad Barre was convinced that the Isaaq, like the Majertein, were “bent on his downfall.” It soon became evident, noted Jama Mohamed Ghalib, one of the president’s Isaaq ministers, that Siad Barre “harbored a special hatred for the Isaaq and soon his ultimate goal became clear—to reduce them to the level of impoverished nonentities.”97 For the President, Sayid Mohamed Abdille Hassan was a nationalist to be emulated; for the Isaaq he was the subject of ridicule and scorn.98 The Isaaq now regarded their colonial experience under the British as comparatively benign and, increasingly, viewed their current circumstances and Marehan domination as colonial oppression. During the civilian era, three clan families—the Majertein, the Hawiye and the Isaaq—had provided both of Somalia’s presidents and all three of its

142

Civil War in African States

prime ministers. To be sure, all three clan families continued to be represented in the military regime, but they were now relegated to an outer circle beyond the MOD coalition or were involved in alliances with their politically less powerful subclans. Disenchantment with their status within the military regime, rising first among the Majertein and then among the Isaaq, was compounded by the policy of harassment and intimidation carried out by the regime’s WSLF surrogates. The movement of Ethiopian-Somali refugees into the north at the regime’s encouragement was, according to Gerard Prunier (who visited the region in the 1980s), interpreted as exposing the same divisions that had, during the Dervish rebellion, led to violence between Isaaqs and the Ogadeni, of which the Sayid had been a member. “And to think that we wanted [the British] out” was a common refrain among the Isaaq. “Look what we got later. That was real colonialism, not what we knew under the British.”99 The fight against the president and his MOD alliance was increasingly viewed by the Isaaq as the real anti-colonial fight, and in fact it was regarded as considerably more dangerous. The emergence of the Somali National Movement (SNM) was a response, not to demands for secessionism in the north, but to a need to defend the Isaaq of the former Protectorate from the regime and, more immediately, from the Ogadeni, who were now threatening to take over Isaaq-inhabited territory.100 “The Isaaq people had urgently to be organized,” wrote Jama Mohamed Ghalib. “That started in late 1978, with a sifting of ourselves . . . to decide who should be involved, and who should be dropped.”101 The movement was the product of two largely Isaaq expatriate groups—one in London and the other in Saudi Arabia—joining forces. At least among the London-based Somalis, there was an ongoing tension between, on the one hand, those who wanted to create a broad-based and inclusive organization that opposed the regime and, on the other, those who believed that immediate needs favored a more narrow clan-based movement that represented those most threatened by the regime’s actions in Somalia. Critics of the former argued that the insurgency would be incapacitated by internal divisions and, specifically, by the fact that the Somali expatriate community also consisted of members of the president’s own clan who sought to temper criticism of the regime. Yet a narrowly based movement would be denigrated by the regime as being “tribalist,” a damaging accusation given that it might limit the SNM’s ability to solicit allies. Since the SNM was regionally based and far from the capital, allies would be essential if the war was to be won.102 In the end, the movement was formally established in 1981 by a group composed primarily, though not exclusively, of Isaaq exiles.

Unification and Quasi-Secession in Somalia

143

Despite secessionist elements within the movement, the SNM claimed that its immediate objective was only the overthrow of the Siad Barre regime and the establishment of a democratic government in a radically decentralized Somalia.103 In early January 1983, the SNM launched its first offensive against the Somali government when it attacked Mandera prison and released a number of northern dissidents. Eventually the SNM moved its headquarters from London to eastern Ethiopia, where it was able to secure military assistance from the regime of Mengistu Haile Mariam. The move was a controversial one for obvious reasons, and there was no shortage of irony associated with the fact that Africa’s arbitrary borders were at once being cursed and exploited by both the region’s regimes and its insurgent movements. Ethiopia, the supposed enemy of all Somalis—including those in the north who earlier had so strenuously resisted the transfer of the Haud to the emperor and had joined with the south in order to reclaim access to Ethiopia’s grazing lands—was now providing resources for the SNM to carry out its attacks against the man who championed the fight to return all Somali-inhabited lands to Somalia. But the SNM’s relations with Ethiopia also disabused northerners of their earlier belief that crossborder access to critical grazing areas could be achieved only indirectly, through union with the south. Clearly this need could now better be met directly by cooperating with Ethiopia.104 Other dramatic events involving relations between the Isaaq and the regime were taking place in the northern capital of Hargeisa. In 1981, a group of young Isaaq professionals who were returning from Mogadishu were dismayed by the city’s state of disrepair and started a series of selfhelp initiatives and programs meant to improve conditions. But as their arrival coincided with the circulation of Somali Uncensored, an antigovernment journal published by the SNM, the government took offense and regarded the group as subversives who were intent on bringing the regime down by exposing its failures in the north. “Officials resented our criticism of the government’s policy of neglect as well as the independence of our initiative,” one member told Africa Watch. “Everything must depend on the government and this is what we were challenging. That for them was unacceptable and dangerous.”105 The heavy-handed treatment that the members of the so-called “Hargeisa Group” received at the hands of government officials in the days leading to their trial in February 1982, as well as rumors that a conviction would result in death sentences for the accused, led to rioting and, on one occasion, to the military shooting blindly into the crowd. The attack, Africa Watch stated, was “one of the most important events that triggered the politically explosive situation in the northern region.”106

144

Civil War in African States

Following its fourth Congress, held in Ethiopia in 1984, the SNM launched a series of attacks on government military posts in the vicinity of the three northern cities of Hargeisa, Berbera, and Burao. However, these efforts drained resources and neither threatened the regime’s hold on power nor led significant numbers of northerners to support the SNM.107 Moreover, for each attack the government responded with its own incursions into SNM-held territory and by instituting harsh security measures throughout northern Somalia. The regime pursued a number of strategies to relieve pressure on the military, with varying degrees of success. In the early 1980s, the president had taken advantage of Majerteen disenchantment with the leadership of the SSF (now referred to as the Democratic Front for the Salvation of Somalia) and began to distribute patronage to Majerteen clans to win back their support. Siad Barre also looked to his own remaining Majerteen ministers to convince their clansmen to reject the DFSS.108 The most startling initiative, however, came in April 1988, when Siad Barre and Ethiopian president Mengistu Haile Mariam pledged to stop providing military assistance to insurgent groups based in each other’s countries, leaving them to their fate. With support for the WSLF no longer forthcoming, the agreement ostensibly allowed the besieged Ethiopian regime to focus attention on their insurgents in Tigray and Eritrea. But the settlement ultimately proved to be counterproductive for Siad Barre. While Somali authorities rounded up DFSS leaders and their weapons, SNM leaders moved their forces from their Ethiopian base camps back to northern Somalia. From there, beginning in May 1988, they launched a major offensive against the cities of Burao and Hargeisa.109 Anticipating their return, General Morgan, Siad Barre’s commander in the 26th Sector North West, threatened reprisals against anyone who provided support for the SNM. In fact, the military’s policy in the north—as evidenced by the notorious January 1987 “letter of death” report to the president from General Morgan—was already well-established. “Unless Qurmis [meaning “the rotten,” referring to the SNM] and its supporters are subjected to a campaign of obliteration,” the report stated, “there will come a time when they will raise their heads again.”110 The document expressed the need to render “uninhabitable the territory between the army and the enemy, which can be done by destroying the water tanks and the villages lying across the territory used by them for infiltration.” The government made good on its threats. Between May 1988 and March 1989, in attacks that were interpreted by some as genocidal, Somali troops and aircraft systematically destroyed the largest northern cities, killed an estimated fifty thousand, and forced up to six hundred

Unification and Quasi-Secession in Somalia

145

thousand people to flee to Ethiopia.111 After the government’s offensive, Mogadishu International Service radio announced that the Armed Forces of the Somali Democratic Republic “had completely wiped out the terrorist gangs” and that the situation was “completely back to normal.”112 The government’s offensive in the north had in fact been a disaster for the SNM. However, the Ethiopian-Somali agreement that preceded it and the savagery of the attack itself had the effect of further alienating both ordinary citizens and powerful allies, thereby widening the size and diversity of the regime’s opposition. Both Siad Barre’s sacking of the Ogadeni defense minister at this time and his rapprochement with Ethiopia were regarded as betrayals among Ogadeni soldiers. They mutinied and formed another insurgent movement, the Somali Patriotic Movement (SPM), kicking one of the legs out of the MOD alliance in the process.113 In the north, each manifestation of government violence hardened the resolve of those Isaaq civilians who remained, and pushed them towards the SNM.114 The plight of refugees who left for Ethiopia captured the attention of the Western media and governments, leading the latter to reconsider financial and military aid that was being provided, albeit in smaller amounts, to the regime.115 As a fighting movement, the SNM had both assets and liabilities. The movement was beset with clan divisions, internal violence, and irregular leadership—weaknesses that would undermine its strength, prolong its struggle against the regime, and continue to haunt it in subsequent years. Even sympathetic observers noted the undisciplined and counter-strategic nature of the movement. At night, Gerard Prunier would later write of his travels with its fighters, “there were no sentries, although fighting was in progress less than ten miles away. When I got up at two o’clock in the morning, snores could be heard all over the place. Five men with grenades could have wiped out the whole detachment in three minutes. What is more serious is that this carelessness extends to tactical and even strategic decisions.” Still, the movement was regarded as a genuine manifestation of a frustrated Somali society. “In a way, the SNM does not exist,” wrote Prunier. “It is simply the Issaq people up in arms.” He added, “This is both a tremendous strength (popular support for the SNM in all areas I went through, seemed completely spontaneous and without reservations), and a tremendous weakness because the Movement seems so immersed into the population as to be quite incapable of ‘leading’ it in the modern political sense.” Despite its apparent lack of discipline, however, foreign aid workers saw in it a remarkably democratic movement in which fighters were

146

Civil War in African States

given a chance to speak, and a highly motivated force where fighters were literally willing to engage their enemy head-on.116 Since it was a movement that was closely connected to its constituents, the Isaaq responded enthusiastically and in ever larger numbers to its call to arms. “When I plan an offensive, men start flocking in,” noted one SNM commander. “They come from everywhere, carrying their guns and 5060 bullets. I start by planning an attack with 600 men, and soon I have 3,000. And there is no way of telling them not to go. . . . Then, after we get back to base, they all disappear and go back to the bush.”117 Although the SNM did document its political intentions, the movement was criticized for a lack of a coherent and unifying strategy—which was not surprising given the conflicting pressures facing the movement. Accompanying its efforts to crush the insurgency, the government announced that the objective of the SNM was to “separate northern Somalia from the south.” Indeed, it was tempting to see the SNM as a secessionist movement and, according to Prunier’s first-hand assessment, “the Ethiopian authorities . . . deal with the SNM as if it were the legal government of Northern Somalia.”118 In fact, however, the SNM’s intentions were not so clear. Publicly, the SNM called only for the overthrow of Siad Barre’s tyrannical government and the recreation of a peaceful democratic state. Early indications of its political intentions focused on the Somali state as a whole and avoided reference to either the north or to a Greater Somalia.119 SNM documents also called for “a new political system built on Somali cultural values of cooperation rather than coercion,” whereby “no man exercised political power over another except according to established law and custom.”120 If the SNM was pursuing a secessionist strategy, the movement was reluctant to state this publicly. Since Somalia was the only ethnically homogenous state in Africa, and since so much of its pre- and post-independence political life had been directed towards uniting Somalis, it was difficult to imagine anything other than a unified Somali state in future. “But the reality is,” observed Gerard Prunier, “that everybody seems keenly aware of two contradictory but complementary elements of truth: one, that the transclan Somali state is a failure. Second, that clan secession has no future.”121 Their experience in a “trans-clan state has been tried,” Prunier wrote, and it could not be made to work without one clan dominating all others.122 Yet, equally, an independent political existence did not appear to be feasible. The comments of SNM leaders at this time reveal sensitivity to both the realities of global politics and to the organization of power in the contemporary Somali state. Several members of the SNM leadership had

Unification and Quasi-Secession in Somalia

147

served as diplomats and thus knew of the African community’s distaste for border change. As desirable as secession might be, the SNM knew there was little chance of going back on its pre-independence decision to merge with the south.123 The failed effort to unite Somalis and the war in the Ogaden taught them the hard lesson that Africa’s borders (and the international conventions that helped sustain them) were unlikely to change even if only to undo the union of the only two Somali-inhabited territories that had joined the Somali Republic. Attempting to secede would invite condemnation from the international community and distract the movement from the more immediate objective, which was to end the regime’s repressive rule in the north. Equally important as this external rationale was an internal logic: had the SNM prematurely announced a wish for secession, then other rebel movements—particularly those in the south—might have betrayed them. In such a scenario, Siad Barre might have succeeded in convincing other clans not to turn against the regime and instead to turn against the Isaaq, whom he could portray as undermining the achievement of a Greater Somalia. Since the SNM could not plausibly overthrow Siad’s regime without the help of other clans, they could not afford to alienate potential allies.124 Yet there were also problems in assuming that Somalia’s challenges were limited only to Siad Barre and his clan, or that his removal would end Somaliland’s conflict with the south. In the post-Siad era, there would be no guarantee that a resurgent south would not, once again, attempt to dominate the north. The ideal of a Somali state that had united all Somalis, and that had earlier been the focus of so much domestic attention and unquestioning nationalist loyalty, was, in the end, still and inevitably subject to the domination of an exclusive circle of clan allegiances. “They believe,” wrote Gerard Prunier of the SNM, “that the officially ‘neutral’ state 1960 model will in fact always be monopolized by one clan or by a group of clans and that the others will suffer.”125 For these reasons, a movement south, as the Tigrayans were in the process of doing in neighboring Ethiopia, was also out of the question. Even if given the chance, and despite its benevolent stated intentions, the SNM feared that its efforts to control the south would be to its own detriment and that of other clans.126 From the perspective of other larger, more powerful and more centrally located Somali clans, such as the Hawiye (who were now forming their own insurgent factions), the prospect of the northern Isaaq single-handedly ruling Somalia was implausible.127 And even if it were possible, unlike the Tigrayans, the

148

Civil War in African States

Isaaq had no tradition of ruling the Somali state. Clan domination of Somalia as it was currently constructed was not an option. For the time being, then, the strategy of the SNM was neither explicitly secession nor domination. But neither was any sort of accommodation with the current regime considered an option. Having alienated its allies and patrons, the regime was less capable than ever of keeping its current and newly-emerging domestic adversaries at bay, much less dominating them. Predictably, in September 1989, as insurgent forces closed in, political reforms were introduced, including a pledge to return to a multi-party system. In September 1990, a new constitution was drafted and free elections were scheduled for February 1991. But the measures were to no avail. Even efforts by the regime and members of the international community to engage the opposition groups in dialogue were rejected by the insurgents on the premise that any changes were unlikely to alter what was, in effect, a ruling family’s hold on power. Subsequently, as Siad Barre’s grip continued to weaken, the regime’s cooperative initiatives and pledges of greater inclusion were regarded as irrelevant, given that the president’s authority was now limited largely to the capital city.128 Those in the international community who were advocating a peaceful approach were also regarded with suspicion, because they came from Western countries who had long continued to supply the regime with weapons. Until Siad Barre’s defeat, the international community was considered to be an unreliable ally.129 For the immediate term, the SNM would, in cooperation with other recently-created insurgent groups, only contribute to the overthrow of the regime, and the issue of what would come after would be negotiated among them.130 In August 1990, these insurgent movements—the SNM, the Hawiye-based United Somali Congress (USC), and the southern-based Somali Patriotic Movement (SPM)—agreed in Ethiopia to form a united front against the Somali regime. In early December, the USC, led by Mohamed Farah Aidid (Siad’s former ambassador to India), led his troops to Mogadishu, where they engaged what remained of the regime in a two-month battle culminating in Siad Barre’s hurried departure from Mogadishu on January 26, 1991. The Decision to Secede and the Experience of Independent Somaliland

If the insurgency in the north was in part “born out of Siad Barre’s misrule,” as is often proclaimed, it still needs to be explained why

Unification and Quasi-Secession in Somalia

149

secession was the option Somalilanders chose once his rule had ended.131 The fact that Somaliland had, since the original merger in 1960, been dissatisfied with the benefits that unity had provided was well established. Somalilanders were also deeply embittered towards the south for its brutality during the war, and there was little reassurance from the United Somali Congress in Mogadishu that the future would be different. A final contributing factor was that, being distant from the capital, the SNM was not in a position to manipulate the fast-changing situation in post-Siad Mogadishu. In the south the political situation had yet to be settled and, in fact, civil strife among various armed clan-based factions was leading to what James R. Kunder, a senior American diplomat, referred to as the “single worst humanitarian crisis in the world.”132 While an increasing number of largely clan-based rebel movements had formed in the final months of Siad Barre’s rule, they were united only by their common willingness to remove the military regime. In the wake of the regime’s collapse, most factions only paid lip service to the idea of a provisional broad-based government that would lead to general elections and multiparty democracy. Any inclination to cooperate was undermined by the preemptive efforts of other groups—most notably by the Mogadishu faction of the USC led by Ali Mahdi Mohamed—to control the political process by establishing a government of their own. “Politics being what they are — some people might have ambitions,” observed Hassan Ali Mireh, a representative of the DFSS (now the SSDF). “The military might want to continue ruling the country. Other opposition groups, including our own, might want to run the country alone.”133 Another Somali faction leader, Abdullahi Daib, stated: “If one rebel group claims they are taking power, that’s going to be a problem. Each person wants the cake — that’s what I’m worried about.”134 Events in the aftermath of Siad Barre’s fall suggested that northern interests would again be a sideshow to the real struggle for power in the capital. Seeing themselves as having “broken the back” of the Siad Barre regime and, consequently, as having prepared the ground for what was only the final coup de grace by the USC, its one-time ally, the SNM expected to be treated as the senior political movement.135 “The people in Mogadishu, who claim to have set up a government, really have done a very stupid thing politically,” observed Omar Issa, a member of the SNM’s executive. “They formed a so-called government without any mandate from any other group in Somalia.”136 The SNM had expected to bring forth its own agenda, which would allow it, in conjunction with other interested parties, to establish a decentralized Somalia—albeit one which, significantly, stopped short of independence for the north. Their

150

Civil War in African States

vision of Somalia, however, did not prevail. The fact that, in the end, northerners had not been directly involved in Siad Barre’s removal and had no forces in Mogadishu meant that, despite the agreement reached in August 1990 with the USC and the SPM, the SNM was no longer regarded as one of the key players who needed to be satisfied or accommodated in any post-Siad peace process. That a faction from one of its partners would form a government without including or even consulting the SNM was unacceptable and confirmed fears that the north would again be dominated by southern interests.137 “As for the decision to separate, it is the outcome of the old practices related to Siad Barre and the new practices related to Ali Mahdi and his supporters,” Abdirahman Tour would later observe. “How can we agree to a government most of whose members joined Siad Barre in destroying us, and whose other members have unknown intentions toward us?”138 The belief that northerners now had to protect their own interests was encouraged by other events and external forces. Somaliland’s western border bisects the land of the Issa clan of Djibouti’s ruling family. In the aftermath of Siad Barre’s collapse, another rebel movement, the United Somali Front (USF), with the support of the Djiboutian government and former Somali government forces, seized control of Issa-inhabited territory on the Somaliland side of the border. While the insurgent USF was subsequently defeated by the SNM, the latter soon realized the need to establish state structures of its own. The declaration of independence, when it came, was an affirmation that northern interests would not be dominated by others.139 Somaliland’s Secessionist Strategy Made Public: The Declaration of Independence

But as late as February 1991, and despite its control of the north, the SNM had yet to formally embrace the secessionist project. If the preindependence desire to unite with the south was driven by clan elites, the push for Somaliland’s post-Siad declaration of independence came from below and against the better judgement of the SNM leadership—including its soon-to-be-named president, Abdirahman Ahmed Ali “Tour,” and his successor, Mohamed Haji Ibrahim Egal. It also came from a relatively broad base of Isaaq and even non-Isaaq clans.140 In this case, the leadership’s reservations about Somaliland’s domestic problems and the slim likelihood that a declaration of independence would be recognized by the international community were no match for the urgency with which Somalilanders called for immediate independence.141 On May 18, 1991, elders gathering at the

Unification and Quasi-Secession in Somalia

151

“Grand Conference of the Northern Peoples” in the town of Burao passed a resolution revoking the 1960 Act of Union that had merged the two colonies thirty-one years earlier and declaring the independent “Republic of Somaliland.” An interim government of the SNM was announced, elections were promised to be held in two years, and Tour, the SNM chairman, was named Somaliland’s first president. These events were celebrated by Somalilanders because they gave hope that the brutal persecution they had endured under Siad Barre’s rule would not be repeated. The declaration was also a means of distancing the north from the ongoing factional fighting in the south and provided the clearest indication to southerners that the north had no territorial claims on their land. Finally, the endorsement that the declaration received from the people of Somaliland was expected to symbolize the stability and unity of the region which would then induce the international community to provide aid.142 For some, however, while the sentiments for independence were sincere and understandable, the hastiness of the declaration was regarded as a strategic error. The consequence of not consulting with the v a r i o u s f a c t i o n s in Mogadishu—their o w n disarray notwithstanding—meant that Somaliland’s declaration went unrecognized by the international community.143 Others argued, prophetically as it turned out, that the declaration would not serve the interests of Isaaq unity. As a clan competing in Siad Barre’s Somalia, the Isaaq were united, but as an independent state, the Isaaq were more inclined to compete among themselves, to the detriment of their security.144 In either case, the benefits of sovereign statehood that were expected to flow from the declaration—the ability to receive foreign aid, the ability to defend itself from external threats (most notably a desire of the south to reassert their authority), and the ability to preserve the unity of the northern people—were not immediately forthcoming. On the other hand, the independence declaration created its own reality, and in time made its revocation virtually impossible. The struggle was no longer framed, as it was earlier, in terms of the overthrow of Siad Barre, but in terms of the achievement of independence. Somaliland’s culminating liberation had been won at great cost, and any thought of a return to union with the south was to obviate the sacrifices that had been made by its people. Moreover, proponents of secession did not see the declaration as impulsive nor as the result of manipulation by a few politicians. “Some people confuse the declaration of the Republic of Somaliland by the SNM in Burao on May 18, 1991 with the fact of separation itself,” wrote Ibrahim Megag Samatar:

152

Civil War in African States

Separation was a political reality long before that. It is a consequence of an historical process whose two protagonists were the cruel persecution by the regime and the stubborn resistance of the persecuted. It is the culmination of the victory of that lonely struggle by the SNM for an extended period. Siyad Barre himself effectively sanctioned the separation and put the last nail on the coffin of the union by his bombardment of the cities of the north and the mass murder of their citizens which led to the fleeing of terrorized civilians into Ethiopia.145

While the Isaaq were the predominant clan family in Somaliland, not all of its clans were in agreement over how power would be shared; nor had they necessarily been convinced of the benefits of the declaration or the manner in which it was achieved. If independence were to be truly and not just nominally realized, managing these divisions, as well as those of other clans outside the Isaaq clan family, would be the principal challenge facing the republic’s leadership. As the evidence will show, the new state’s leadership faced the same decisions on issues of cooperation, coercion, and secession that were found among other regimes. Through dialogue among Somaliland’s influential elders, the leadership sought to accommodate those who wanted to join the political process; yet it would need the judicious use of force when dissenting militias threatened to spoil the integrity of the new country. Indeed, the major issue was the knowledge that, if its broader secessionist objectives were to be realized, it could not afford to mismanage dissenters within Somaliland or allow a descent into southern-style factionalism. Nor could Somaliland’s own independence declaration—and the border change that it would entail—involve a serious disruption of regional affairs. There were, after all, good reasons why the international community viewed any secessionist effort with either casual interest or, by contrast, with trepidation. Given that it took place amidst a flurry of other declarations of power in southern Somalia and a call for autonomy in neighboring Puntland, Somaliland’s declaration did not feature either in the post-Cold War’s list of strategically significant events or even as something to be taken seriously. Herman Cohen, the American Secretary of State for African Affairs, noted that “We did not recognize the new entity [of Somaliland] because we had no confidence in any declarations in Somalia.”146 It was also dismissed because the infrequent occasions on which secessionists had made serious claims to alter Africa’s borders—among them a prolonged crisis in the Congo following Katanga’s 1960 secessionist declaration, and Biafra’s ruinous and failed

Unification and Quasi-Secession in Somalia

153

effort to secede from Nigeria in 1967—did not inspire confidence that it could be done, let alone peacefully.147 The broad and unifying support for the independence declaration among Somalilanders was, however, an asset for the north, particularly in light of the ongoing upheaval in the south. In December 1992, mobilized by the violence and starvation that was occurring in and around Mogadishu, the international community became directly involved in a humanitarian effort. Called the Unified Task Force (UNITAF), the mission was intended to push past Mogadishu’s “warlords” and deliver food aid to those most at risk. But the intervention soon evolved into a more ambitious plan to disarm the factions themselves and even to rebuild the Somali state. Under the UN Operation in Somalia (UNOSOM), a parallel UN mission that had been previously launched, the effort proved too dangerous. It came to ruin when eighteen US Army Rangers, who had been pursuing one of the faction leaders, Mohamed Farah Aidid, were killed in a Mogadishu firefight in October 1993. Following the lead of the United States, UNOSOM’s main sponsor, the UN withdrew its forces and the resurrection of the Somali state—if anyone thought it could still be raised—was left to others. In fact, southern Somalia was resistant to peace-and-reconciliation. In the ten years after the regime’s collapse, more than a dozen peace initiatives were launched, none of which was able to produce a sustainable and inclusive agreement. In his September 16,1999 Report to the United Nations on the Situation in Somalia, UN Secretary General Kofi Annan noted the local and regional consequences that flowed from an ongoing “lack of a functioning central government.” Somalia, he said, is now “being seen as a ‘black hole’ where the absence of law and order is attracting criminals and subversives.”148 The UN and the broader international community welcomed the efforts of the Djiboutian president, Ismael Guelleh, to fill Somalia’s political vacuum in 2000. In the Djiboutian city of Arta, a provisional constitution was created, along with a Transitional National Government (TNG). But the process resulted in the election to a transitional presidency of Abdiqassim Salad Hassan, a minister from the muchdespised Siad Barre years, and in a government that had no mandate to govern Somalis living in Somalia much less Somaliland. Somalilanders also could not help but notice, with justifiable anger, that members of the international community appeared prepared to formally recognize a process, and the government it produced, despite the fact that the latter did not exercise authority over any territory. Meanwhile, an existing and effective government in Hargeisa, albeit one which had declared

154

Civil War in African States

independence, continued to be ignored.149 More problematic still was that processes meant to encourage power-sharing and inclusion in the south were all for naught, since the endless bargaining sessions did little to promote reconciliation even among the southern clans. As one observer of the ongoing Somali peace process wrote, For a decade, national reconciliation in Somalia has been reduced to a crass exercise in pie sharing, in which the sole objective of negotiations is to get Somali representatives to agree to proportional allocation (by clan) of various seats in the government. Critical issues of reconciliation—such as the return of stolen property and compensation—are not even discussed, much less resolved. . . . This inevitably causes rapid inflation in the number of seats in parliament and proposed ministries—designed solely to buy off and satisfy as many clans as possible—leading to a bloated and utterly unsustainable administration.150

It became the view among increasingly impatient members of the international community that, while the armed factions did want to participate in these reconciliation processes, they did so in the expectation of indulging themselves with donor-provided resources of state which were expected to be forthcoming once the new government was formally recognized. However, despite the international cynicism that was apparent whenever a new reconciliation effort was initiated during the 1990s, progress towards the creation of a legitimate or recognized government in Mogadishu was a source of concern in Hargeisa. Since no southern group had demonstrated flexibility on the Somaliland issue in the past, it was unlikely that any new national government in Mogadishu would accept Somaliland’s claim to independence in the future. For the international community to unconditionally recognize a new centralized government in Somalia was also to endorse its claim to the north, an action which would, one United Nations report warned, “make a conflict with the Somaliland administration all but inevitable.”151 Reasserting Somaliland’s Secessionist Strategy

Beginning in 1997, rainwater began to uncover mass grave sites in and around the destruction of Hargeisa which served to remind foreigners and Somalis alike of the previous regime’s ruthlessness and the dangers of efforts to reunite with the south. Somaliland officials spoke in apocalyptic terms. The Vice-Speaker of the Somaliland parliament, Abdulqadir Haji Ismail Jirdeh, warned that, “The TNG has been

Unification and Quasi-Secession in Somalia

155

encouraged to claim sovereignty over other groups, territories and entities which it doesn’t control and which it doesn’t represent. ... They will try to re-arm themselves and try to reconquer by war. We will resist that.” He added, “The immediate plan [of the international community] is to help the Somali people in their crisis. The intentions are good, but the road to hell is paved with good intentions. I have no doubt in my mind, this will lead to more war.”152 The progress made at Arta notwithstanding, there was little chance in the immediate future that a government based in Mogadishu was about to acquire the means to extend its authority northward. On the other hand, while the situation in the north never descended to the factional strife found in the south, the government in Hargeisa during the 1990s had also remained fragile as it struggled with its own lack of resources and internal strife. The authority of Somaliland’s new president, Abdirahman Ahmed Ali Tour, did not convincingly extend throughout Somaliland, and the SNM, the movement that had liberated Somaliland from Siad Barre’s rule, was itself increasingly divided and would soon dissolve entirely. In January 1992, these divisions led to violence between clan-based militias that supported Abdirahman Ahmed Ali Tour, and who regarded themselves as the “national army,” and another faction of the SNM, known as ‘Alan ‘As (Red Flag). After nine months of inconclusive fighting, elders brought the two sides together in the town of Sheikh, where a ceasefire was reached, a national council of elders (based on a traditional council referred to as a guurti), was established, and plans were made for another and more ambitious “national” conference to be held in the town of Borama in January 1993.153 The Borama conference—a five month inclusive gathering of all clans which was overseen by Somaliland’s elders—continues to be regarded by Somalilanders and pundits alike as a defining event in the Republic’s political development and a reaffirmation of its decision in 1991. The Borama conference would also signal the end of the Isaaqdominated SNM, which had had control over all earlier proceedings, and its replacement with a broader multi-clan coalition known as a beel system of government which relied on consensus decision-making.154 Even the choice of Borama, a town in western Somaliland populated largely by the Gadabursi clan, was significant. The Gadabursi, a nonIsaaq clan, had earlier allied itself with Siad Barre’s regime, but had worked to restore relations with the Issaq in 1991. At the conclusion of the conference in May 1993, the Gadabursi hosts were rewarded with the post of Somaliland’s vice-presidency. The conference delegates also produced a peace-and-security accord, a new National Charter reflective

156

Civil War in African States

of the SNM’s vision for a post-Siad Barre administration, and a new government under the leadership of Mohamed Haji Ibrahim Egal, who had been Somalia’s last prime minister before the 1969 coup that brought Siad Barre to power. Berbera, Somaliland’s port, was an important revenue-producing strategic asset, being within the president’s clan territory. Egal’s presidency now provided his government with financial resources that had been unavailable to his predecessor, Abdirahman Ahmed Ali Tour. Somaliland’s problems, however, were not over. The appointment of Egal as prime minister was criticized in some quarters because, in 1991, Egal had opposed Somaliland’s independence declaration and had subsequently exiled himself to the United Arab Emirates. His return to Somaliland did not remove suspicion that he still preferred a united Somalia to an independent Somaliland.155 More seriously, Egal’s rise to power was opposed by the Habr Yunis (who had just lost the presidency) and the Iidegale clans, both subclans of the Isaaq clan family and known collectively as Garhajis. They claimed that the process which propelled Egal, a Habr Awal, into power had been unfair, and that his choice of cabinet ministers included prominent opponents of the Tour administration, as well as members of the ‘Alan ‘As faction that had violently challenged the Tour government in 1992. In July 1993, the Habr Yunis withdrew their cooperation from the government and, one year later, declared Somaliland’s government illegitimate. A group of Garhajis, led by Tour, announced their preference for a renewed federation with Somalia and restored their relationship with the southern faction leader of the USC, General Mohamed Farah Aidid. The relationship had initially been formed when the SNM and USC had agreed to join forces to defeat Siad Barre. Civil War in Somaliland

The standoff between the Egal government and the Garhajis reflected the long-standing clan divisions within the SNM that predated Siad Barre’s fall. It was also the result of the manipulation of General Aidid and the United Nations, both of whom opposed Somaliland’s breakaway declaration and sought to slow its movement to independence.156 Tensions came to a head when Somaliland’s other strategic asset, Hargeisa’s revenue-producing airport, came under the control of an Iidegale militia in November 1994. Fighting broke out first in Hargeisa and then in Burao, where Egal’s government authority was challenged by the Habr Yunis. In the midst of violence, disagreements emerged about whether these rogue clans were more interested in subverting

Unification and Quasi-Secession in Somalia

157

Somaliland’s nascent independence project, or in gaining access to Somaliland’s political process and using their links to southern clans for leverage. The government maintained that the principal disagreement was between those in favor of, and those opposed to, Somaliland’s independence—a characterization that was furthered by Aidid’s notdisinterested material support for the Garhajis. The Garhajis themselves claimed that the conflict revolved around questions of appropriate clan representation in government, Egal’s association with ‘Alan ‘As, and the predominance of Habr Awal business interests—and thus that it had little to do with Somaliland’s independence except insofar as these issues further empowered the president and his clan.157 For many northern Somalis, Somaliland’s own civil war between Egal’s government and the “federalist” forces (i.e. those portrayed as wanting to reunite with southern Somalia), which lasted from 1992 to 1996, was as traumatic as the events in the last years of Siad Barre, not only because of the upheaval resulting from human displacement and death but because it threatened to undermine Somaliland’s ambitions for independence. If the larger project of political independence was not to be a casualty of Somaliland’s civil war, it was essential that the remaining federalist forces be either integrated or neutralized. After government forces had moved to take control of Hargeisa’s airport, opposition forces dug in at Burao, looking on this as their last stand. Neither side was able to win a decisive military victory, however, and this forced the government to seek a peaceful solution. Since both the stalemate and Somaliland’s precarious political status ruled out domination by coercive methods, Egal adroitly looked to other means to outmaneuver his opposition and achieve the same ends. The Habr Yunis and the Habr Jelo (another Isaaq clan who, unlike the Habr Yunis and Habr Awal, had yet to have one of their own in the president’s office) began their own initiative to make peace and would eventually form a single political block to challenge Egal.158 Before the alliance could be consummated, however, Egal advanced the date of a proposed reconciliation conference in Hargeisa and threatened to proceed without them. Egal then abandoned his alliance with ‘Alan ‘As and appealed to the opposition for support. Inducements of up to $30,000 were rumored to be on offer to those willing to break ranks and come to Hargeisa, while those who refused were isolated from the political process. Many opposition members accepted the bribes or conceded to political expediency. By keeping the political center on his own turf, Egal prevailed and consolidated his authority without further jeopardizing Somaliland’s independence ambitions.

158

Civil War in African States

Egal’s Victory

The conflict would be formally concluded in Hargeisa between October 1996 and February 1997. Egal’s position had no doubt been aided by the death in August 1996 of General Aidid, the patron of the putative federalist forces in Somaliland. But Egal had also acquired several advantages that helped to end the military stalemate that had been reached in 1995.159 He was, for example, able to exploit both the legitimacy that the Borama conference had conferred on his rule as well as the port revenues flowing to his government to strengthen his position. Most importantly, however, he was able to take advantage of the deeply-held independence sentiment in Somaliland at the time. By framing the debate as being about Somaliland’s status as an independent state, and by placing himself firmly in the independence camp, he defeated his opponents politically without resorting to undue amounts of force.160 Egal’s new dominant position in Somaliland helped to serve another purpose: to re-engage the international community with the hope that it would recognize Hargeisa’s claims to statehood. This proved to be a formidable task. Since the United Nations had committed substantial resources to resurrect a central authority in Mogadishu, it did not welcome Hargeisa’s continued insistence that it was taking an independent course. To no avail did proponents of Somaliland’s independence highlight the fact that the larger and vastly more expensive peace processes in the south—such as the 1993 UNsponsored $1.3 million Addis Ababa conference—failed to produce a lasting agreement. Conferences in the north, starting with the Borama conference in 1993 and finishing with the conference in Hargeisa in 1996-97, had been run entirely by Somalis and at little expense to the international community, and yet continued to make progress in reconciling differences among Somaliland’s people.161 The growing view among Somalilanders and local advocacy groups that the UN sought to undermine Somaliland had started when a leaked August 1993 diplomatic memo suggested that UNOSOM’s intention was to encourage some regions and political movements in Somaliland to look to Somalia rather than Hargeisa for political leadership.162 Subsequent UN documents also suggested that President Egal’s commitment to independence had “softened.” Specifically, they hinted that he was still sympathetic to the cause of a united Somali state and was only seeking a “special status” for the territory in order to “access international financial institutions” or because he was under pressure from “hardline elements in ‘Somaliland’ [who were] adamantly opposed

Unification and Quasi-Secession in Somalia

159

to any dialogue with the rest of Somalia.”163 Some within the UN hoped that, as an alternative to independence, Somaliland would serve as one of several “building blocks,” an approach which would first entail various regions of the country rebuilding themselves, as Somaliland had done, and then coming together as a united entity.164 After meetings with various regional leaders, the UN Secretary General claimed that President Egal’s intention was to “assist the clans in the ‘South’ in reaching peace and subsequently enter into negotiations on an appropriate formula for uniting the country.” Rest assured, the UN Secretary General wrote after consulting his Deputy Assistant Minister for Bilateral African Relations, “Mr. Egal had a role to play in reuniting the country.”165 If Egal was, indeed, still ambivalent about the independence project, he continued to maintain that its success would require Somaliland to demonstrate to the international community that it was committed to a peaceful and democratic process and that its separate status was producing meaningful economic and developmental benefits for its citizenry. Various regional organizations were, of course, also insistent on a peaceful approach to conflict management in Somalia. Meeting in November 2000, member states of the Intergovernmental Authority on Development (IGAD) had included in their final communiqué a demand that the peace process in Somalia “continue and be completed through dialogue and not by resorting to force.”166 While the statement was directed at the south, it had obvious relevance to events in the north as well—not least of which was that a peaceful north would be secure from a future armed effort by the Mogadishu to reclaim Somaliland. Multiparty Elections in Somaliland

But a peaceful approach to politics was now something that President Egal would point to in order to advance Somaliland’s claim to an independent status. In 1999, President Egal linked Somaliland’s independence to the achievement of multiparty democracy, arguing that failing to do so would undermine its prospects for international recognition.167 Consequently, the government of Somaliland followed through on plans to conduct a constitutional referendum, as well as plans to hold competitive local and presidential elections. As Bradbury, Abokor and Yusuf note, the district council and presidential elections in particular “were perceived by many Somalilanders as a test of their ability to govern themselves and the credibility of their state.”168 The elections were also politically risky, since they marked a movement away from the consensus-oriented approach of the post-Borama years

160

Civil War in African States

towards a Western brand of competitive elections that would, inevitably, produce both winners and losers. A multiparty electoral system, however, was not adopted just for the sake of the international community. Others maintained that, while it was inclusive, the existing beel system solidified clan politics by reducing democratic rule to a process in which representatives were judged only in terms of what they delivered to their own clan. While the b e e l system had served Somalilanders well in the early years of reconstruction, what was required now—both for the sake of Somaliland’s unity and its recognition from abroad—was a meritocracy, in which candidates would be forced to compete for votes through impersonal institutions.169 On May 31, 2001 the first of these three democratic processes took place when Somaliland conducted the referendum on its newly drafted constitution. Among the constitution’s first articles was a statement reaffirming Somaliland’s existence as a sovereign and independent country, a statement which, thanks to a propaganda campaign in Hargeisa, became the central issue in the referendum. Another unrelated event, however, added urgency to the referendum’s outcome: the arrival in Mogadishu in October 2000 of the Arta-generated TNG, a government that publicly aspired to national recognition and control of all Somalia’s territory. This position, accordingly, put the Somaliland administration on the defensive. As one source observed, “Whatever President Egal’s profound convictions may have been with regard to the political future of Somalia, the intention of the TNG to extend its control over Somaliland led him and his administration to adopt a militant secession attitude.”170 While the outcome of the referendum was for many a foregone conclusion, other sources have argued that Egal himself was unprepared for such a massive endorsement of independence. From this perspective, it was the extent of support shown for the independence clause during the constitutional referendum that pushed Egal to finally and convincingly embrace secession. “Whatever he may have believed previously,” one of his ministers observed, “from 31 May [2001], onwards he was a Somalilander.”171 The Death of Egal: A Test for Somaliland Independence

A more critical test for Somaliland’s independence, however, was not Egal’s personal political conversion but his death, on May 3, 2002, during surgery in South Africa. Egal’s death was met by widespread demonstrations of nationalist sentiment. But the response to his death also served as an important indication of the sustainability of Somaliland’s constitutional and governmental process. As required by

Unification and Quasi-Secession in Somalia

161

the constitution, the leaders managing the political transition sought to avoid a political vacuum and looked to Egal’s vice-president, Dahir Rayale Kahin, a Gadabursi (a non-Isaaq clan) to assume the presidency. Democratic elections were to follow at the local level on December 15, 2002, and later, in 2003, Somaliland conducted its presidential elections. “On 14 April, 2003,” the International Crisis Group reported, “the people of Somaliland enjoyed an experience all too rare in the Horn of Africa: an election without a predetermined outcome.”172 The results announced on April 19 suggested that the election had indeed been closely fought—which was celebrated by Somalilanders as an indication of an open election process. The UDUB Party, founded by Egal and led by his successor Dahir Rayale, won over Kulmiye, its nearest rival party, by a mere eighty votes. While the outcome was met with protests from opposition members, the leader of Kulmiye stated that he did not intend Somaliland to follow the path of Mogadishu. The results were contested at Somaliland’s Supreme Court, but Kulmiye eventually conceded victory to UDUB. On May 16, incumbent president, Dahir Rayale, was officially sworn in as Somaliland’s first popularly elected president.173 Since its independence declaration, Somaliland has continued its public campaign to gain international recognition. In a 2002 document published by Somaliland’s Ministry of Foreign Affairs, the government argued that it had all the attributes normally associated with statehood, including a territory defined by international treaties, a stable government, a permanent population, and the capacity to enter into relations with other states.174 More significantly, the government maintained that its status as an independent state was not a violation of the OAU/AU’s rules on the sanctity of African colonial/state borders; nor would recognition allow other regions to make similar claims to statehood. Because Somaliland had been a British Protectorate which, brevity notwithstanding, existed as an independent state prior to its union with the south, recognition would involve not a border change but the acknowledgment of Somaliland’s previous colonial borders. And since independence would be on the basis of colonial borders, no other community could make a similar claim to independence unless they could point to a prior colonial existence. In this sense, the government argued, “Somaliland’s independent status . . . represents the dissolution of a voluntary union between sovereign states, not an act of secession.”175 Finally, the government pointed to its own internationally supervised referendum on independence, which was conducted in an open and free manner and which overwhelmingly endorsed the independence option.

162

Civil War in African States

Conclusions

At the time of independence, the British Protectorate of Somaliland sought integration with the southern Somalis of Italian Somalia. Thirtyone years later, Somalilanders reversed course and declared independence from the rest of Somalia. Thus, Somalilanders pursued a strategy of integration followed by a strategy of separation. As the preceding discussion makes clear, the most important factors that gave rise to the strategy of integration can be identified in terms of demographics, resources, and history/memory/precedent. Ethnic Somalis were spread throughout the Horn of Africa, and colonial borders had divided the Somali people among no fewer than five states. Given these colonially-created divisions, the appeal of a single state for all Somalis is not surprising. But the attraction of a unified Somalia was also driven by a more commonplace need: namely, the ability to use grazing lands to the south in territory that had been, in the eyes of many Somalis, wrongly ceded by the British to the Ethiopian emperor. The combination of a nationalist ideal, an economic imperative, and a sense of injustice was a powerful motivation for unification. The desire for integration, however, was as much a product of what Somilanders did not know as what they did. To be sure, many Somalis were attracted to the idea of unifying as a common people. Until the union of north and south, however, there was no history of a Somali state, and, therefore, no experience of living under a single or common authority. In short, there was much that was appealing about a strategy of integration and no compelling reason why it should not occur. But life in Somalia’s semi-arid land (i.e. one necessitating much of the population to follow water) made the creation of such an entity unlikely. The introduction of a modern state—one which came to be sustained in large part by superpower patronage—was bound to elicit resentment and resistance. When the state became sufficiently authoritarian and coercive, resistance was manifested first among Somalia’s Isaaq clan in the form of the Somali National Movement. Before its declaration of independence in 1991, the strategy of the SNM was, at least publicly, merely the removal of the Siad Barre regime from power. That its conversion to a secessionist strategy was belated reflected both an internal logic that an explicitly secessionist strategy would alienate potential allies before the struggle was won, and the SNM’s awareness that there was well-grounded conservatism in international politics on the issue of state recognition. In other words, factors associated with both its own ability to project power and global conservatism were significant checks on any secessionist aspirations. To

Unification and Quasi-Secession in Somalia

163

be sure, the desire for dissolution was rooted most obviously in the increasingly violent manner in which the central government in Mogadishu sought to suppress opposition in the north. But this chapter has also suggested that the initial enthusiasm for union was in part a function of a more prosaic need to access grazing lands. In this sense, an important setback for Somalia’s unity (the cession of the Haud to Ethiopia) occurred even before the most significant—and only—step forward had been taken (the union itself between the north and the south). Other events—dissatisfaction with civilian government, disillusionment with the promise of the union of all Somali peoples after the 1977-78 Ogaden War, the regime’s reaction in 1982 to the so-called Hargeisa Group, and the relentless aerial attacks by Mogadishu’s military forces in 1988—continued to harden the resentment of northerners against the regime. Arguably, none of these historical factors demanded a specifically secessionist strategy, and northern security needs could have been addressed by Siad Barre’s overthrow. But Somalilanders who had borne the brunt of the government’s attacks saw nothing in the behavior of southern factions in the aftermath of the regime’s collapse to suggest that the relationship with the south was about to improve. As Matt Bryden has written, the Somali Democratic Republic in the last years of Siad Barre was best described as a “‘toxic state’—one that became so noxious to its population that it has been purposefully and comprehensively dismantled by its own people.” He adds, “Rather than counting on better leadership in future, Somalis opted to deny their leadership the means to perpetuate past patterns of abuse by establishing smaller, weaker, more accountable political arrangements.”176 In this sense, a secessionist strategy was deeply rooted in the memory of attacks by the central government and served as a means of preventing any resurgent political authority in Mogadishu from oppressing the north again. Today, reminders of the abuses suffered by northerners at the hands of Siad Barre’s regime are apparent to anyone who visits Somaliland’s main cities.177 Somaliland’s geographical isolation from the capital, and its previous existence as a British Protectorate encompassing the predominant Isaaq clan family, also helped speed the movement from a strategy which sought only to remove Siad Barre from power to an independence declaration. Given the manner in which Somalilanders have cultivated this history of oppression and woven it into their own collective memory, the prospect of a return to a united Somali is unlikely for the foreseeable future.

164

Civil War in African States

But how well have Somalilanders been served by a secessionist strategy? Has secessionism served Somalilanders’s security and development interests, even in the absence of formal recognition?178 Ten years after its declaration of independence, the UNDP described several indicators that life was significantly better in the north than in the south. “In comparison to southern Somalia,” the UNDP stated, “food security in the north has generally been better over the past decade. Better physical security and a greater reliance on pastoral production have been the main reasons.” The report added, “The construction boom in the northern cities has also provided much needed employment for displaced people and returning refugees and has triggered a significant flow of migrant labour, mainly from the south but also former pastoralists and non-Somali migrant workers from Ethiopia.”179 For many, Somaliland is, de facto, an independent country, regardless of the failure of the international community to recognize it. Somaliland’s inability to gain recognition is even regarded as a blessing in disguise—particularly for its minority clans. The would-be state’s need to demonstrate its credentials to the international community has advanced the democratic process, increased the need for political inclusiveness, and ensured that Somaliland has avoided the political excesses of the Horn of Africa’s other now-independent state, Eritrea. When resources are tested and each new challenge overcome, Somaliland’s political institutions are strengthened, regardless of whether recognition is granted.180 In this sense, recognition would in fact be a retrograde step, since the Somaliland state would rely less on the strength of its indigenously constructed institutions and more on the basis of an exogenously-imposed juridical statehood. But Somaliland has also become hostage to the very preconditions it believes are necessary for its security: peace and recognition.181 A successful secessionist strategy still required Egal to achieve a position of dominance within the new state—preferably through peaceful means. To be sure, peace is a desirable outcome much welcomed by Somaliland’s residents. However, as a result of its perennial need to appear to be the tranquil and harmonious success story in the Horn of Africa, political dissent, legitimate or otherwise, is criticized as damaging to peace and, therefore, likely to undermine Somaliland’s prospects for international recognition. The means of suppression are, in this sense, more subtle but no less real. Dahir Rayale Kahin’s presidency outlived the mandate granted by the electorate in the 2003 elections and, while new elections were scheduled for 2010, some Somalilanders worried that they have been unable to make demands on their government that citizens can in more mature democratic states.182

Unification and Quasi-Secession in Somalia

165

This qualification notwithstanding, any realistic and peaceful dispensation for Somalis must now either acknowledge that a single unified Somali state is not possible or, at the very least, allow Somalilanders to keep out of reach of the coercive means of a resurrected Mogadishu. While there are reasons for doubting the commitment of all Somalis in the north to a separate existence, few outside observers question that an overwhelming number of them prefer the relative peace that de facto independence has brought. More importantly, any effort by the south to retake the northern territory—an unlikely scenario given the persistent disarray that has characterized Mogadishu and the comparative unity that has characterized Hargeisa—would almost certainly involve a return to violence. During this era of post-Siad independence, Somaliland has proven itself to be increasingly capable of managing its own problems. By contrast, repeated failures to establish a stable and sustainable government in Mogadishu suggest that it is no longer possible. While Somaliland has not avoided violence within its own declared borders, its desire to be recognized by the broader international community has necessitated the use of peaceful means to find solutions to the problems it faces. Each year of its de facto independence represents another year in which a new generation of Somalilanders know of no other existence beyond life as a separate political entity, thus making a retreat from its quest for independence less and less likely. 1. It is in this sense that one keen observer of Somali politics, Matt Bryden, remarks that in the early 1990s “clanism had destroyed Somalia but saved the Somalis” referring to the manner in which clan had at once divided the state and yet provided a means for Somalis to maintain personal security. Interview, July 14, 2001, Hargeisa Somaliland. 2. Cited in John Drysdale, The Somali Dispute (New York: Praeger 1964), p. 8. 3. John Drysdale, Stoics Without Pillows: A Way Forward for the Somalilands (London: Haan, 2000), pp. 65, 140. 4. Cited in Drysdale, The Somali Dispute, p. 22. 5. In Richard F. Burton, First Footsteps in East Africa; or, an Exploration of Harar (London: Longman, Brown, Green, and Longmans, 1856), p. 502. 6. Drysdale, The Somali Dispute, p. 21. 7. Cited in Drysdale, The Somali Dispute, pp. 22-23. 8. By “Somalilanders” I am referring to the inhabitants of the former British Protectorate or British Somaliland. 9. Cited in D.J. Latham Brown, “The Ethiopia-Somaliland Frontier Dispute,” International and Comparative Law Quarterly, vol. 5, no. 2 (1956), pp. 246-247.

166

Civil War in African States

10. Saadia Touval, Somali Nationalism: International Politics and the Drive for Unity in the Horn of Africa (Cambridge: Harvard University Press, 1963), p. 52. 11. Some have argued that the Sayid’s portrayal as a nationalist is misguided since his appeal was rooted in Islam and he was just as inclined to fight against other clans as against the colonial powers. Gilkes, “Two Wasted Years: The Republic of Somaliland, 1991-1993,” (mimeo), p. 4. The Sayid’s support came from the Dhulbahante clan (his maternal lineage) with whom the British had no treaty of protection, and the Wasengeli; both the Dhulbahante and Warsengeli are from the Darod clan family. 12. Correspondence with Mohamed Ibrahim in Hargeisa, email April 1, 2008. Accounts of Sayid’s life can be found in D.J. Jardine, The Mad Mullah of Somaliland (London: Herbert Jenkins, 1923); Robert L. Hess, “The ‘Mad Mullah’ and Northern Somalia, Journal of African History, vol. 3 (1964), pp. 415-433; Patrick Kakwenzire, “Resistance, Revenue and Development in Northern Somalia, 1905-1939, International Journal of African Historical Studies, vol. 19, no. 4 (1986), pp. 659-677; and E.R. Thurton, “The Impact of Mohammed Abdille Hassan in the East Africa Protectorate,” Journal of African History, vol. X, no. 4 (1969), pp. 641-657. 13. Touval, Somali Nationalism, p. 155. 14. Latham Brown, “The Ethiopia-Somaliland Frontier Dispute,” p. 248. 15. For a brief description of these territories, see Latham Brown, “The Ethiopia-Somaliland Frontier Dispute,” p. 252. 16. Drysdale notes that Rennel of Rodd negotiated the agreement such that it involved “Abyssinian recognition of our Protectorate without in any way admitting recognition on our part of a cession to Ethiopia.” See The Somali Dispute, p. 29. For a discussion of how the Anglo-Ethiopian Agreement superceded the British guarantees, see Latham Brown, “The EthiopiaSomaliland Frontier Dispute,” p. 254. 17. Cited in Latham Brown, “The Ethiopia-Somaliland Frontier Dispute,” p. 250 In practice, however, between 1897 and 1935 (when Italy’s invasion of Ethiopia put the 1897 treaties in question), British officers “followed and administered the British-protected tribes on their annual migrations over the border.” In this sense, migrating Somalis were “effectively protected, so far as circumstances allowed.” Latham Brown, p. 252. 18. Latham Brown, “The Ethiopia-Somaliland Frontier Dispute,” p. 255. 19. Prior to the arrival of Lord Rennell of Rodd, the British representative, the French had negotiated a similar agreement with Ethiopia. Menelik tried to convince Lord Rennell of Rodd that the boundary should be only 100 km from the northern coast—which was what he had negotiated with France for their French colony at Djibouti. Rodd would not agree to this. See Drysdale, The Somali Dispute, p. 28. 20. Cited in D.J. Latham Brown, “The Ethiopia-Somaliland Frontier Dispute,” p. 249, footnote 14. 21. Latham Brown, “The Ethiopia-Somaliland Frontier Dispute,” p. 253. 22. H.B. Kittermaster wrote in 1928 that “The future prospects of the Protectorate are not discouraging, but it has a very bad reputation to live down because of the activities of the Mad Mullah, and it requires careful nursing.” See

Unification and Quasi-Secession in Somalia

167

H.B. Kittermaster, “British Somaliland,” Journal of the Royal African Society, vol. 27, no. 108 (1928), p. 330. 23. For a pessimistic and sometimes unflattering account of some of these efforts and experiences by a former British colonial official, see Harold B. Kittermaster, “The Development of the Somalis,” Journal of the Royal African Society, vol. 31, no. 124 (1932), pp. 234-244. 24. Patrick Kakwenzire, “Resistance, Revenue and Development in Northern Somalia,” p. 677. 25. Latham Brown, “The Ethiopia-Somaliland Frontier Dispute,” pp. 256257. The Ethiopian interpretation of these agreements is more critical. In this view, the agreements were forced upon Ethiopia and, particularly in the case of the 1942 Agreement, “almost every article underlined its dependency and the preponderant role of Britain.” See Bahru Zewde, A History of Modern Ethiopia, 1855-1991 (Oxford: James Currey, 2002), pp. 179-180. 26. Drysdale, The Somali Dispute, p. 80. 27. Cited in both Touval, Somali Nationalism, pp. 79-80, and Drysdale, The Somali Dispute, p. 67. 28. Mesfin Wolde Mariam, “The Background of the Ethio-Somalian Boundary Dispute,” Journal of Modern African Studies, vol. 2, no. 2 (1964), pp. 208-209. 29. “Report on the Visit to Addis Ababa,” Somali Public Record Office, Reference CO830/26 XC17470. p. 7. 30. Cited in Latham Brown, “The Ethiopia-Somaliland Frontier Dispute,” p. 259. 31. Jama Mohamed Ghalib, The Cost of Dictatorship: The Somali Experience (New York: Lilian Barber Press, 1995), p. 28-29. 32. Touval, Somali Nationalism, p. 95. 33. Touval, Somali Nationalism, p. 103. 34. Jay Walz, “Haile Selassie Calls on Big Four To Share Results of Their Talks,” New York Times, January 31, 1960, p. 1. 35. The petition was published in the Ethiopian Herald on 21 May 1960. 36. “Visit to Addis Ababa.” 37. The Ethiopian government initially announced that the right of clans to have access to grazing lands would be annulled after the Protectorate had become independent. Immediately prior to independence, however, Ethiopia claimed that the grazing rights would be recognized, provided the 1897 frontier was recognized by Somalis. See I.M. Lewis, A Modern History of the Somali: Nation and State in the Horn of Africa, 4th edition (Oxford: James Currey, 2002), p. 325 endnote, 22. 38. “Visit to Addis Ababa,” p. 8 39. The Somali delegation maintained that the Ogadeni elders who purportedly signed the petition were mere “quislings,” who were being disingenuous in their claims. “I think for the sake of the record I should say,” wrote the mission scribe, that “they are former leaders of the Ogaden who have succumbed to the Ethiopian intimidations and were therefore disowned by their own people. None of these eight [signatories] can dare to come back to Ogadenia amongst their people.” See “Visit to Addis Ababa,” p. 3. 40. Touval, Somali Nationalism, pp. 134-135.

168

Civil War in African States

41. The Emperor likely downplayed the risks of including Somalis within his multi-ethnic territory because of his confidence that they could, to some extent, be assimilated to Ethiopian culture and interests. Given his conviction that Ethiopians and Somalis were of the same race and drank “from the same river,” he did not discount the possibility that education and the incentive of employment with the Ethiopian state would induce cooperation from the Somali community in Ethiopia. See Touval, Somali Nationalism, pp. 142-146. See also “Visit to Addis Ababa,” p. 2 and Drysdale The Somali Dispute, p. 83. 42. “Visit to Addis Ababa,” p. 5. 43. “Harsh Ethiopian Methods of Subduing Somalis,” The Times, October 25, 1956, p. 8. 44. “Visit to Addis Ababa,” pp. 3 and 4. 45. Touval, Somali Nationalism, pp. 105-107. 46. Mariam, “The Background of the Ethio-Somalian Boundary Dispute,” p. 211. 47. Jama Mohamed Ghalib, The Cost of Dictatorship, p. 28-29. 48. Upon its independence, the U.S. Secretary of State, Christian Herter, sent a congratulatory message to Somaliland. Also on the day of Somaliland’s independence, the United Kingdom also signed several bilateral agreements in Hargeisa. See David H. Shinn, “The Horn of Africa: How Does Somaliland Fit?” Paper Presented at the Discussion Seminar Introducing Somaliland in Umea, Sweden, March 8, 2003, p. 1. 49. Jama Mohamed Ghalib, The Cost of Dictatorship, p. 33. 50. Academy for Peace and Development, “A Self Portrait of Somaliland: Rebuilding from the Ruins” (Hargeisa: Somaliland Centre for Peace and Development, 1999), p. 14; International Crisis Group, “Somaliland: Democratisation and its Discontents,” Africa Report No. 66 (Nairobi: 28 July 2003), p. 4. 51. Touval, Somali Nationalism, p. 111. 52. Touval, Somali Nationalism, p. 118. 53. Touval, Somali Nationalism, p. 118. 54. Africa Watch, Somalia: A Government at War with its Own People—Testimonies About the Killings and the Conflict in the North (New York: Africa Watch, January 1990) p. 14. 55. Africa Contemporary Record, 1968-69, p. 201. 56. Africa Contemporary Record, 1968-69, p. 200. 57. Ghalib, The Cost of Dictatorship, p. 67. 58. Davidson, Stoics Without Pillows, pp. 78-79. 59. Gary Payton, “The Somali Coup of 1969: The Case for Soviet Complicity,” Journal of Modern African Studies, vol. 18, no. 3 (1980), p. 501. The view that these parties were related less to ideological cleavages and more related to personal and clan interests was reflected by the fact that, immediately after the 1969 elections, representatives of 121 of the 124 seats crossed the floor to join or rejoin the most prominent party, the Somali Youth League. See “Somalia: After the Elections,” Africa Confidential, No. 11, May 23 1969, pp. 3-4. 60. Al Castagno, “Somalia Goes Military,” Africa Report, February 1970, p. 26.

Unification and Quasi-Secession in Somalia

169

61. The NFD Commission’s Report is discussed in John Drysdale, The Somali Dispute, chapter 11, 62. Raymond Thurston, “Détente in the Horn,” Africa Report, February 1969, pp. 8-9; Africa Recorder, November 20-December 3, 1969, p. 2391. 63. Ghalib, The Cost of Dictatorship, p. 120. 64. The pretext of the coup was the apparently unrelated assassination of President Abdirashid Ali Sharmarke on October 15, 1969. Castagno, “Somalia Goes Military,” p. 25; “Somali Army and Police Seize Power,” New York Times, October 22, 1969, p. 3. 65. Africa Contemporary Record, 1969-70, p. B175. 66. Gary D. Payton argues that part of this disorder was attributable to the machinations of the Soviet Union itself, who had sought to destabilize the civilian government in Mogadishu so that it could build a new relationship with a client state in a strategically important region. See “The Somali Coup of 1969: The Case for Soviet Complicity,” Journal of Modern African Studies, vol. 18, no. 3 (1980), pp. 493-508. 67. Ghalib, The Cost of Dictatorship, p. 120. 68. Ghalib, The Cost of Dictatorship, p. 141. The regime legally banned public acknowledgment of clan and demanded that individuals be referred to as “comrade.” 69. Basil Davidson, “Somalia: Towards Socialism,” Race and Class vol. xvii, no. 1 (1975), p. 29. 70. See Ahmed I. Samatar, Socialist Somalia: Rhetoric and Reality (London: Zed, 1988), pp. 87, 109-110. Those senior-most officials arrested or executed within the first three years of the military regime included the former police commander, General Jama Ali Qorshel, the SRC Vice President Mohamed Ainanshe, and another SRC member and Defence Minister, Salah Gavere Kedie. See Africa Contemporary Record, 1971-72, pp. B185-186. 71. Jama Mohamed Ghalib, Cost of Dictatorship, p. 132. 72. Africa Recorder, November 20-December 3, 1969, p. 2392. 73. The four African states with larger numbers of soldiers included Nigeria, Ethiopia, Zaire and Uganda. See Ioan M. Lewis, Blood and Bone: The Call of Kinship in Somali Society (Red Sea Press: Lawrenceville, 1994), p. 174, footnote 42. On July 11, 1974, a Treaty of Friendship and Cooperation was signed between the Soviet Union and Somalia. 74. Lewis, Blood and Bone, p. 174, footnote 42. 75. Laitin claims grazing lands did offer room for negotiations and possible compromise because of their role in the survival of Somalia’s nomadic population and because they were less important to the Ethiopian economy. See, David Laitin, “The War in the Ogaden: Implications for Siyaad’s Rôle in Somali History,” Journal of Modern African Studies, vol. 17, no. 1 (1979), p. 115. 76. United Nations, Official Records of the General Assembly (Volume 1), Thirty-Second Session Plenary Meetings, 27th meeting, October 10, 1977, p. 511, paragraphs 133-134. 77. Ghalib, The Cost of Dictatorship, p. 112. Patrick Gilkes contends that this tendency has become particularly strong since 1991. The two liberation movements, the Ogadeni National Liberation Front (ONLF) and the WSLF now

170

Civil War in African States

argue in favor of a Ogadeni state rather than a Somali one. See “Two Wasted Years,” p. 4. 78. David Laitin, “The War in the Ogaden,” pp. 112. There were other suspicions that Siad Barre was a follower rather than a leader—that, for example, if he had not led the coup himself, “the younger army and policy officers would have gone ahead without him.” See “Somalia: After the Coup,” Africa Confidential, No. 21, October 24, 1969, p. 7. Others have maintained that Siad Barre demonstrated such ambitions much earlier. See Samatar, Socialist Somalia, p. 115, endnote 9, and Jama Mohamed Ghalib, The Cost of Dictatorship, pp. 67-68. 79. Interview with the author, Hargeisa, July 28, 2008. See I.M. Lewis, “The Ogaden and the Fragility of Somali Segmentary Nationalism,” African Affairs, vol. 88, no. 353 (1989), p. 574. Ogadeni was the clan of Siad’s mother’s brother and would form a critical alliance that would sustain the president’s rule particularly in later years. 80. The significance of the constitutional referendum is contested. Some have argued that the rejection of the constitution was a consequence of a campaign by northern politicians unhappy with the distribution of political portfolios among their colleagues. See Abdi Ismail Samatar and Ahmed I Samatar, “The International Crisis Group Report on ‘Somaliland’ in Horn of Africa, vol. XXIV, (2006), p. 23. 81. On the significance of 1961failed Somali coup see Ghalib, The Cost of Dictatorship, pp. 123-125; and I.M. Lewis, A Modern History of the Somali, 4th edition (Oxford: James Currey, 2002), pp. 173-178. 82. With regard to the 1969 coup, Africa Confidential wrote that “things were so peaceful in the northern town of Hargeisa . . . that [unlike in the south] there was not even a curfew in force.” See “Somalia After the Coup,” Africa Confidential, No. 21, October 24, 1969, p. 8. The only reported unrest following the coup was in Majertenia in the north-east and in the Upper and Lower Juba in the south. “Somalia Military Policies,” Africa Confidential, vol. 11, no. 16, August 7, 1970, p. 8. Gilkes, by contrast, claims that the coup did not alleviate northern sense of neglect. See “Two Wasted Years,” p. 5. 83. Abdisalam M. Issa-Salwe, The Collapse of the Somali State: The Impact of the Colonial Legacy (London: Haan, 1994), p. 65. 84. Ibrahim Megag Samatar, “Light at the End of the Tunnel: Some Reflections on the Struggle of the Somali National Movement” in Hussein M. Adam and Richard Ford eds. Mending Rips in the Sky: Options for Somali Communities in the 21st Century (Lawrenceville: Red Sea Press, 1997), p. 24. 85. The SSF later to merge with other groups to become the Democratic Front for the Salvation of Somalia (DFSS) and subsequently the Somali Salvation Democratic Front (SSDF). Ghalib suggests that the coup was not motivated solely by the war. The regime, he says, was already engaged in a program of harassment, surveillance, arrests and restrictions on their business against the Mijertein, a clan that was economically and, particularly during the civilian era, politically influential. See Ghalib, The Cost of Dictatorship, p. 128. 86. See Gilkes, “Two Wasted Years,” p. 5. See also Laitin, “The War in the Ogaden,” p. 115. 87. According to one prominent Somalilander, “the dream was that every Somali had to seek to bring the five parts together politically. And our union

Unification and Quasi-Secession in Somalia

171

with the south was the first step in that direction. It was not a desired union per se. It was a means to an end. If you take away, why should the means be pushed together again. The international community has taken away the end.” Abdulqadir Haji Ismail Jirdeh, interview with the author, July 13, 2001. See also Ibrahim Megag Samatar, “Light at the End of the Tunnel,” pp. 46-47. 88. Academy for Peace and Development, “A Self Portrait of Somaliland,” p. 16. 89. Bradbury, Somaliland (CIIR Country Report 1997), p. 7. The windfall of humanitarian assistance reaped by the regime is estimated to have been $120 million per year. According to Ghalib, the regime also engaged in a practice of doubling the estimated refugee population so that half of the food allocation could be diverted towards the army. See Cost of Dictatorship, p. 149. According to Peter Bridges, the US ambassador in Mogadishu from 1984-1988, Siad Barre claimed that the country was underpopulated. “We need, more young men for soldiers,” Siad Barre told the Ambassador, “to defend ourselves against Ethiopia.” Bridges notes the Somali government’s estimation of its population at 9.4 million; the World Bank said that Somalia’s population was only 6 million. See Peter Bridges, “Safirka: Envoy to Somalia,” in Michigan Quarterly Review (Winter 1998), p. 7. 90. World Bank, Somalia: Crisis in Public Expenditure Management, Report No. 8727-SO (March 8 1991), p. 11. According to the US ambassador Peter Bridges, “there was a rapacious side to the Somalis. Siad Barre’s insistence that Somalia needed still more American aid was echoed by every minister that I called on, and many of their requests were clearly not justified.” See Bridges, “Safirka: Envoy to Somalia,” p. 9. 91. World Bank, Somalia: Crisis in Public Expenditure Management, p. 20. 92. Private correspondence from official. 93. Africa Watch, Somalia: A Government at War with its Own People, p. 34. 94. I.M. Samatar, “Light at the End of the Tunnel,” p. 40. 95. Ghalib, The Cost of Dictatorship, p. 160. For a discussion of the linkages between Siad Barre and Sayid Mohamed Abdulle Hassan, see, David Laitin, “The War in the Ogaden,”pp. 95-115. 96. “Siad Barre Addresses Mogadishu Rally,” 19 June 1982, FBIS: Middle East and Africa, vol. 10, no. 119 (June 21, 1982), p. R1. 97. Ghalib, The Cost of Dictatorship, p. 160. 98. Gerard Prunier writes, “In a way, it was back to 1890, and I never heard so much abuse heaped on poor old Muhammad Abdulle Hassan as during that trip. See Gerard Prunier, “A Candid View of the Somali National Movement,” Horn of Africa, vol. xiii, no. 3 and 4, vol. xiv, nos 1 and 2 (1990-1991), p. 116. 99. Prunier, “A Candid View of the Somali National Movement,” p. 119. Another traveler to the north, Ioan Lewis, wrote in 1989 that during his most recent visit in 1985, the north “began to look and feel like a colony under a foreign military tyranny.” See I.M. Lewis, “The Ogaden and the Fragility of Somali Segmentary Nationalism,” p. 576. 100. Africa Watch, Somalia: A Government at War with Its Own People, p. 196. According to Patrick Gilkes, there was spirited discussion on the issue of secession at this time, but this was ultimately rejected. Instead the SNM sought

172

Civil War in African States

to portray itself as northern rather than specifically Isaaq. See Gilkes, “Two Wasted Years,” p. 6. 101. Ghalib, The Cost of Dictatorship, p. 162. 102. Lewis, Blood and Bone, 192-193. 103. For an early statement of the SNM, see Lewis, Blood and Bone, pp. 198-199. 104. Bradbury, Somaliland, p. 19. By contrast, John Drysdale notes that, by this time, “the problem of the grazing lands all but faded into colonial history.” See Stoics Without Pillows, p. 74. Ghalib also writes that “had Siad Barre succeeded in absorbing the Ogaden into his realm during the 1977-78 war over that territory, the Isaaq people would have enjoyed no neighbouring base or sanctuary from which to launch their liberation struggle.” See The Cost of Dictatorship, pp. 183-184. 105. Africa Watch, Somalia: A Government at War with its Own People, p. 39. 106. Africa Watch, Somalia: A Government at War with its Own People, p. 37. John Drysdale also regarded these events—specifically February 24, the date of the government’s clampdown—as a “turning point in the Somalilander’s calendar of struggle against Barre.” See John Drysdale, Whatever Happened to Somalia?: A Tale of Tragic Blunders (London: Haan, 1994), p. 136. 107. Gilkes, “Two Wasted Years,” p. 7. 108. “Somalia: Post Reshuffle,” Africa Confidential, vol. 23, no. 6 (March 17 1982). 109. The attacks served to demonstrate that the Ethiopian-Somali agreement had not doomed the movement. “The occupation of Hargeisa,” noted an SNM representative, “confirms that the revolt against Barre is anything but over, despite the accord the head of state signed with Ethiopia ... to end external support to each others’ respective opposition groups.” FBIS—Daily Report (Sub-Saharan Africa), June 6, 1988, p. 5. 110. The report was leaked to the public and published by Richard Greenfield as “Somalia’s Letter of Death,” New African, July 1987, pp. 14-16. A translation of the report was also published in its entirety online at www.civicwebs.com/cwvlib/africa/somalia/1987/letter_of_death.htm. 111. Mark Bradbury, Somaliland, pp. 11 and 20. General Morgan is reported to have asked the following rhetorical question: “If the SNM attacks us, do you think that we will leave this town of yours as it is?” Africa Watch, A Government at War, p. 144. 112. “Siad Barre Says Northern Rebels ‘Wiped Out,’” FBIS-Daily Report (Africa) June 20, 1988, p. 1. 113. Bradbury, Somaliland, p. 11; Abdisalam M. Issa-Salwe, The Collapse of the Somali State, p. 73. 114. Bradbury contends that this happened in 1988: “Up to this point,” he says, “the SNM had not been widely supported by civilians. These brutal attacks finally united the Issaq [sic] behind the SNM for an all-out war against the regime.” Somaliland, p. 11. Africa Watch contends that this process was already well underway and began “in earnest” in 1984 after the SNM stepped up military activity in Somalia’s urban areas. See Africa Watch, A Government at War, p. 63.

Unification and Quasi-Secession in Somalia

173

115. In addition to the Africa Watch report, A Government at War with Its Own People, two key documents can be consulted: United States General Accounting Office (GAO) Somalia: Observations Regarding the Northern Conflict and Resulting Conditions (Washington: May 1989), and Robert Gersony, Why Somalis Flee: Synthesis of Accounts of Conflict Experience in Northern Somalia by Somali Refugees, Displaced Persons and Others (Bureau for Refugee Programs, Department of State: Washington, 1989). 116. This assessment comes from a presentation and further discussion with Immanuel Deisser of Medicins Sans Frontiers at the 8th Congress of the Somali Studies International Society, Hargeisa, Somaliland, July 11, 2001. He argued that the high frequency—eight out of ten occasions—in which MSF would treat specifically frontal wounds suggested a highly motivated group of fighters. 117. Prunier, “A Candid View of the Somali National Movement” p. 113. 118. Prunier, “Candid View of the Somali National Movement,” p. 110. 119. Ioan M. Lewis, Blood and Bone, pp. 198-199. Article 6 of the SNM’s Constitution adopted at the 5th Congress, 1987, reads, ‘The Somali National Movement is committed to the total liberation of the Somali Republic and shall oppose any division of the country into regions or mini-states that are prejudicial to the unity of the country.” Cited in Ghalib, The Cost of Dictatorship, p. 217. 120. Cited in Ahmed I. Samatar, Socialist Somalia: Rhetoric and Reality, p. 142. 121. Prunier, “A Candid View of the Somali National Movement,” p. 117. 122. Prunier, “A Candid View of the Somali National Movement,” p. 118. 123. Gilkes, “Two Wasted Years,” p. 9. 124. Mohamed Yuusuf Duale, interview with the author, July 28, 2008. 125. Prunier, “A Candid View of the Somali National Movement,” p. 118. 126. There was, for example, little inclination for SNM units to move south, as the TPLF was currently doing in Ethiopia. Prunier writes: “It was often pointed out to me that the SNM units in the Central Area were not composed of Issaq. And that, in fact, it was a necessity for them not to be. ‘Who knows,’ I was often told rather realistically, we might behave as badly down there as THEY have up here.’ See, “A Candid View of the Somali National Movement.” pp. 118-119. See also Ioan Lewis, Blood and Bone, p. 200. 127. M.A. Jesow, Interview with the author, June 20, 1997. 128. Even at the time of Siad Barre’s fall in January 1991, eight of the president’s sons and daughters held key positions within the administration. See Bradbury, Somaliland, p. 13. In his final days, the president was referred to mockingly as the “Mayor of Mogadishu” by the opposition fronts and officials of his former patron, the United States. See Herman J. Cohen, Intervening in Africa: Superpower Peacemaking in a Troubled Continent (Basingstoke: Macmillan, and New York: St. Martin’s Press, 2000), p. 203. Siad Barre asked for peace negotiations and offered to abide by them “whatever the outcome.” The rebels refused, claiming they were willing to meet with the other opposition groups but not the President. In a statement, they said, “Let him leave or resign.” Jane Perlez, “U.S. and Italy Evacuating Foreigners in Somalia,” New York Times, January 6, 1991, p. A3. 129. Ghalib, The Cost of Dictatorship, p. 196.

174

Civil War in African States

130. Prunier notes that “contribute” was the key word for the SNM: “they never think they can do it alone.” See Prunier, “A Candid View of the Somali National Movement,” p. 118. 131. Ghalib, The Cost of Dictatorship, pp. 171-172. 132. Cited in Jane Perlez, “U.S. Says Airlifts Fail Somali Needy,” New York Times, July 31, 1992, p. A9. 133. Robert M. Press, “Somali Rebels Seek Peace After Ousting Ruler,” Christian Science Monitor, January 30, 1991, p. 3. 134. Robert M. Press, “Somali Rebels Seek Peace After Ousting Ruler.” 135. Ghalib, The Cost of Dictatorship, p. 183. The SNM’s role in overthrowing the regime is, not surprisingly, overstated by the SNM leadership. Compare with Patrick Gilkes, “Two Wasted Years.” 136. Cited in Koert Lindijer, “Why the SNM Spurns USC Overtures,” New African, May 1991, p. 12. 137. See “Reaction to USC Presidential Appointment Noted: Senior SNM Member Comments,” FBIS-Daily Report (Sub-Saharan Africa), January 31, 1991, pp. 8-9. 138. “Somaliland’s President Tur Interviewed,” FBIS-Daily Report (SubSaharan Africa), December 7, 1992, p. 12. 139. Gilkes, “Two Wasted Years,” p. 8. 140. Gilkes, “Two Wasted Years,” pp. 8 and 9. 141. According to Drysdale, at the time, the SNM’s reservations about a declaration of independence were rooted in the belief that “the country’s economy was not in a robust enough condition to withstand international diplomatic rejection.” See Whatever Happened to Somalia, p. 139-140. 142. Bradbury, Somaliland, p. 18. 143. Not surprisingly, proponents of secession maintained that there was no single entity in Mogadishu with which to negotiate Somaliland’s secession. See Ibrahim Megag Samatar, “Light at the End of the Tunnel,” pp. 47-48. 144. Ghalib, The Cost of Dictatorship, p. 219. 145. Ibrahim Megag Samatar, “Light at the End of the Tunnel,” p. 46. The current government of Somaliland now argues that the country’s roots go back even further. “Our country,” one government statement proclaims, “is not an entity which was born after the disintegration of The Siad Barre dictatorship. ... [I]n ancient pre historic times a country existed in Somaliland which had its own entity and its own geographical contours.” See “Official Government Statement” Republic of Somaliland, April 23, 1997, p. 1. 146. Cohen, Intervening in Africa, p. 205. 147. Anonymous, “Government Recognition in Somalia and Regional Political Stability in the Horn of Africa” Journal of Modern African Studies, vol. 40, no. 2 ( 2002), pp. 250-251. 148. Report to the United Nations on the Situation in Somalia, 16 September 1999, p. 13 of 17. 149. Anonymous, “Government Recognition in Somalia and Regional Political Stability in the Horn of Africa,” p. 253-254. The United Nations, African Union (AU), Arab League, and the Organization of Islamic Conference extended membership to the TNG. Matt Bryden, “State Within a Failed State: Somaliland and the Challenge of International Recognition,” in Paul Kingston

Unification and Quasi-Secession in Somalia

175

and Ian S. Spears eds. States Within States: Incipient Political Entities in the Post-Cold War Era (New York: Palgrave, 2004), p. 173. 150. Menkhaus, “Somalia: In the Crosshairs of the War on Terrorism,” Current History (May, 2002), p. 212. The directionless nature of some of the Somali negotiating processes was perhaps best demonstrated earlier when, after weeks of fruitless negotiations in a Kenyan hotel, the UN finally threatened to stop paying the participants’ hotel bills. An agreement was quickly signed, but, like many others in the interminable peace process in the south, it did not last. 151. Matt Bryden, “Somaliland and Peace in the Horn of Africa: A Situation Report and Analysis,” United Nations Development Programme and UN-Emergencies Unit for Ethiopia (November 1995), p. 9. 152. Abdulqadir Haji Ismail Jirdeh, interview with the author, July 13, 2001. Egal had initially opposed the creation of a Somaliland army, citing insufficient resources to allow it. Later Egal changed his mind, arguing that a national army would increase the chances of international recognition of his government. Africa Contemporary Record, 1992-94, vol. 24, p. B392. 153. In the next several years Somaliland elders would not only facilitate the organization of these conferences but would also accept many of Somaliland’s administrative and security functions. Mark Bradbury, Adan Yusuf Abokor and Haroom Ahmed Yusuf, “Somaliland: Choosing Politics Over Violence,” Review of African Political Economy, no. 97 (2003), p. 459. 154. Beel refers to clan or community. See Bradbury, Abokor, Yusuf, “Somaliland: Choosing Politics Over Violence,” p. 460. 155. In 1991, Egal had taken part in peace talks in Djibouti meant to reestablish a national Somali government in Mogadishu. International Crisis Group, “Somaliland: Democratisation and Its Discontents,” Africa Report #66 (July 28, 2003), p. 10. 156. Bradbury, Somaliland, pp. 25, 26. 157. Bradbury, Somaliland, p. 26. Some have argued that the former representation was a political ploy on the part of Egal to discredit opposition to his government and that the latter interpretation was the correct one. Writing in the midst of the conflict, Matt Bryden states that “the federalist agenda was quickly proven hollow and collapsed, leaving both the government and opposition in the position of defending Somaliland independence.” He concludes that, “It is therefore somewhat confusing that the two groups should still be fighting at all. ...” See Matt Bryden, “Somaliland and Peace in the Horn of Africa: A Situation Report and Analysis,” UNDP and Emergencies Unity for Ethiopia (November 1995), p. 2. 158. Bryden, “Somaliland at the Crossroads,” p. 7. 159. Bradbury, Becoming Somaliland, p. 121. 160. On this point, see Matt Bryden, “Somaliland and Peace in the Horn of Africa: A Situation Report and Analysis,” United Nations Emergencies Unit for Ethiopia (November 1995), p. 2. 161. Drysdale, Whatever Happened to Somalia, p. 143. 162. The leak led Egal to ask UNOSOM to leave Somaliland. UNOSOM did not leave, although it did stop its work. Drysdale (1994), pp. 146-147; Bradbury, Somaliland, p. 30. 163. Report of the Secretary-General of the United Nations on the Situation in Somalia, 16 September 1999, pp. 3-4.

176

Civil War in African States

164. See Matt Bryden, “New Hope for Somalia? The Building Block Approach,” Review of African Political Economy, vol. 26, no. 79 (March 1999), pp. 134-140; “Somalia: Are ‘Building Blocks’ the Solution?, Integrated Regional Information Network (IRIN), July 19, 1999. Aside from Somaliland, the “Puntland” to the east was seen as another potential “building block.” 165. Report of the Secretary-General of the United Nations on the Situation in Somalia, September 16, 1999, p. 4. 166. Proceedings of the Eighth Ordinary IGAD Summit of Heads of State and Government, November 23, 2000, Khartoum, Republic of Sudan, Annex III. 167. Bradbury, Abokor and Yusuf, “Somaliland: Choosing Politics Over Violence,” p. 463. 168. Bradbury, Abokor and Yusuf, “Somaliland: Choosing Politics Over Violence,” p. 459. 169. Interview, Hargeisa, July 2008. 170. Anonymous, “Government Recognition in Somalia and Regional Political Stability in the Horn of Africa,” p. 262. 171. Cited in International Crisis Group, “Somaliland: Democratisation and Its Discontents,” p. 12. 172. International Crisis Group, “Somaliland: Democratisation and its Discontents,” p. 1. 173. Dahir Rayale Kahin is a Gadabursi, a member of the Dir clan family. Some Somalis pointed to the election of a non-Isaaq as an indication of the government’s inclusiveness. Others said that, as a non-Isaaq, he was attractive because in order to maintain his credentials in an Isaaq-majority state, he would be even less likely to make a deal with the south. Interviews conducted by the author, Hargeisa, July 2008. 174. Somaliland claims that its borders are in fact “better established in international law than those of Somalia, whose border with Ethiopia remains an undefined and undemarcated provisional boundary.” See, Ministry of Foreign Affairs, Briefing Paper: The Case for Somaliland’s Recognition as an Independent State (Hargeisa: August 2002). 175. Ministry of Foreign Affairs, Briefing Paper: The Case for Somaliland’s Recognition as an Independent State. 176. “Reviving the Somali Peace Process: Perspectives and Prospects in the Post-Arta Period.” Paper presented to the 8th Congress of the Somali Studies International Association, Hargeisa, July 2001, p. 2. 177. The most notable of these was an actual Soviet-built MiG-15, erected in the center of Hargeisa and surrounded by paintings of atrocities committed by the Siad Barre regime. In 1997 mass graves were discovered, which some Somalilanders have since sought to have preserved as a tangible reminder of atrocities committed by southerners against Somalilanders. 178. Some practitioners have advised against the recognition of southern Sudan, for example, as a means of promoting viable state structures because, once statehood has been recognized, the incentive to continue the process of democratic reform is reduced and the incentive to assert one’s authority increased. Kenn Crossley, “Why Not to State-Build New Sudan,” States Within States: Incipient Political Entities in the Post Cold War Era (Palgrave, 2004), p. 148.

Unification and Quasi-Secession in Somalia

177

179. United Nations Development Program, Human Development Report: Somalia (New York: 2001), pp. 68 and 70. 180. Mark Bradbury notes that a lack of recognition has forced the government of Somaliland “to develop its own capacity.” See Bradbury, Somaliland, p. 20. 181. Interview, Hargeisa, July 2008. 182. Somaliland’s presidential elections were postponed in 2007 and again in 2008. The are now expected to take place before April 2009.

4 The Angolan Civil War

We are reaching the end of a long journey to democracy. That is what we wanted. I will not go back to fight in another bush war. UNITA leader Jonas Savimbi We are not going to lay down arms. There is a saying, ‘if you want peace, keep the powder dry.’ UNITA leader Jonas Savimbi The radical posture by Mr. Savimbi permits me to conclude that political compromise was never part of his agenda. Angolan President José Eduardo dos Santos This chapter focuses on one of Africa’s longest and most destructive conflicts, the Angolan civil war, from its beginnings in 1961 to its conclusion in 2002. In what follows, I examine two sets of questions. First, what were the circumstances that pushed two parties—who had until the end of the Cold War resisted efforts to come together—to agree in 1994 to one of Africa’s most promising examples of equitable powersharing? More specifically, why was the ruling MPLA—the movement that had most vociferously resisted calls for integration—willing to share power in 1994? Second, given the significance of such an agreement, what factors undermined this apparently mutual willingness to adopt integrative strategies and, in the end, doomed them both to more bloodshed? In particular, why was UNITA’s leader—an individual who repeatedly demanded that he be included in Angola’s governing structure—ultimately unwilling to come to the Angolan capital of Luanda and take up his position in a power-sharing government?

180

Civil War in African States

From independence onward, Angola’s liberation movements idealized domination as their preferred strategy. But, at least for UNITA, an unfavorable balance of power precluded the realization of such as strategy. The MPLA’s stranglehold on Luanda combined with its backstopping from Moscow and Havana meant that, until the end of the Cold War, the most UNITA could hope for was negotiations leading to multiparty elections. Beginning in 1994, however, both parties publicly accepted the logic of a power-sharing approach to conflict resolution. But an integrative strategy was meant to serve other needs. In the case of the MPLA, not only did power-sharing satisfy external pressures for a peaceful accommodation, but it proved to be an important means for the government to divide its opposition and isolate UNITA’s leader Jonas Savimbi from the rest of the movement. For UNITA, a renewed commitment to cooperation was a lifeline that allowed it to buy more time to rearm itself. In the end, neither leadership made a genuine or unqualified commitment to power-sharing. When the agreement had served its purpose or proved to be unworkable, both parties returned to war in a final effort to defeat the other. This chapter will be divided into sections. The first section will consider the emergence of Angola’s principal independence movements, along with the forces that prevented them from arriving at political accommodation. Subsequent sections will examine the turn of events that led both parties to accept an integrative approach to Angolan politics and those forces that undermined these efforts. Consideration will also be given to the ruling MPLA’s decision to finally neutralize UNITA as a military threat. Decolonization in Angola and the Emergence of the MPLA and UNITA

Most African states entered the post-colonial era with one liberation movement that claimed, rightly or wrongly, to speak on behalf of that colony’s people. Angola had three such movements, each based in varying degrees on class, region, and ethnicity. Having three movements could have been a blessing—they represented what some Angolans saw as “the beginning of democracy.” For others, a divided anti-colonial movement was a curse, insofar as differences between them generated an intense and ultimately violent rivalry. One of the first of the three movements to challenge Portuguese rule during the 1960s, the FNLA, was led by Holden Roberto and drew its support largely from northern Angola’s Bakongo population. Its rival, the MPLA, established in 1956, was led by Agostinho Neto and drew its

The Angolan Civil War

181

support from the Luanda region and its environs. While the MPLA was a party composed largely of Mbundu and urban intellectuals, it was more ethnically diverse than the other two movements. Both its heterogeneity and its substantial following in Luanda would serve the MPLA well during Angola’s transition to independence when control of the capital would prove to be decisive.1 The third movement, UNITA, emerged in 1966 under the leadership of Jonas Savimbi. While it represented elements of an exploited rural working class, the rationale behind UNITA’s creation was also largely ethnic: UNITA’s population base was the Ovimbundu of the highlands in central and southern Angola. Since the Ovimbundu were the colony’s single largest ethnic group, its supporters regarded a liberation movement as an essential counter-balance to the movements representing the other two major ethnic groups. “In this country these [ethnic] sentiments exist,” observed UNITA’s Isaías Samakuva. “The MPLA might be bad but they are from my area. UNITA might be bad but they are from my area. Politics in this country are still based on these factors.”2 Portuguese colonial rule persisted largely because it was successful in fragmenting the nationalist movement. In the end, Angola’s independence was achieved not as a direct result of the individual or collective efforts of the nationalist organizations, but indirectly as a consequence of a coup within Portugal itself. The fall of the fascist Estado Novo regime in Lisbon generated considerable unease, given that the legacy of colonial rule was not a coherent class of Angolan educated elites but a fratricidal struggle among contending movements. If there had ever been unity among the anti-colonial movements, this ended with the sudden prospect of a winner-takes-all power-grab in the capital Luanda. The respective parties did, however, make public pledges to find an amicable negotiated transition to an independent Angola. Talks began in Alvor, Portugal, in January 1975, and an agreement was reached to proceed with Angola’s independence on November 11, 1975. Until then Angola was to be ruled by a coalition transitional government comprised of all three nationalist movements. The armed forces of the three movements would also be integrated into a single army, and general elections for a constituent assembly would be held in October 1975. Having spent the years between 1961 and 1974 terrorizing the liberation movements and fomenting suspicion among them, the departing Portuguese were now hard-pressed to find common cause among them. Even less propitiously, the Lisbon coup had been inspired by a desire to end Portugal’s chronic colonial wars, and its leadership was no longer inclined to commit further blood and treasure to ensuring order in a state

182

Civil War in African States

in which it had no interest. Because of this and despite their pledges to respect the Alvor Agreement, the three movements sought to position themselves to take the capital. They also solicited support from foreign governments that might allow them to either defeat their adversaries or block them from securing power. Aware of the escalating tension among the leaders of the three Angolan nationalist movements, one Portuguese official bleakly wondered “who among them would be alive a year from now.”3 In the case of UNITA, however, there was reason to believe its calls for a peaceful outcome were genuine, if still self-serving. Even prior to the Lisbon coup, the MPLA had secured financial and logistical support from the Soviet Union, and the FNLA was receiving support from Zaïre and China. UNITA, by contrast, could field fewer guerrillas than either of the two other movements and was the only movement without a foreign patron.4 However, given its links to the most significant ethnic group, the Ovimbundu, UNITA could claim the high-road of peacemaker with the expectation that it would prevail in free and fair competitive elections. The most charismatic of the three leaders, Savimbi was increasingly confident that his travels across Angola were generating support for his movement. But it was increasingly clear that a democratic solution was unlikely and that UNITA would become irrelevant if they did not respond in kind to the military build-up of their adversaries. Its ascension to power blocked by the other two factions, UNITA was unable to remain above the fray when its offices and several of its cadres were attacked by MPLA forces. UNITA, however, even after it entered the war, still claimed that it was the party of peace and blamed its adversaries for the violence. “The entry of UNITA into the war,” it declared, “is entirely the fault of the MPLA which has systematically violated all the agreements which have been made, and has attacked the forces of UNITA in a cowardly way.”5 Maintaining that the MPLA could never control such an expansive country, UNITA also prophesied that a unilateral declaration of Angolan independence by the MPLA would entrench Angola in a protracted civil war.6 Believing that the military balance of power favored them, both the FNLA and the MPLA rejected any proposal for a negotiated solution. Daniel Chipenda, a senior FNLA cadre (though previously a defeated rival for the MPLA leadership) stated, “We do not think that new negotiations are possible any longer; we are going to Luanda, not to negotiate, but to lead.”7 Holden Roberto, the FNLA’s leader added, “We have signed a number of agreements, all of which have been violated by the MPLA. Now we will no longer be tricked. Now we will go

The Angolan Civil War

183

forward.”8 The MPLA was equally intransigent: “We are 100% enemies and we can never come to an agreement. Our fight must go on until FNLA is defeated as the American imperialists were in Vietnam.”9 The MPLA refused any further offers to negotiate a settlement, claiming that they would only be presented with a fait accompli, which would almost certainly be less than what they had already achieved through military force. As Independence Day approached, all parties knew how important military control of the capital was to winning international recognition. Each party solicited outside assistance so that it could either prevail over the others or deny victory to their adversaries. Interested foreign powers included the Soviet Union and Cuba, both of whom intervened on the side of the MPLA with military and economic aid, troops and advisors. The United States and South Africa intervened to support the FNLA and later UNITA. The outside powers saw the conflict in their own global terms. The U.S. Assistant Secretary of State for African Affairs, Nathaniel Davis, later wrote that Secretary of State Kissinger, in his refusal “to abandon the competition against our great adversary, would freely acknowledge, I believe, that he saw Angola as part of the U.S.Soviet relationship, and not as an African problem.”10 Kissinger himself states in his memoirs that the issue “was not the intrinsic importance of Angola but the implications for Soviet foreign policy and long-term East-West relations.”11 Superpower involvement, then, was not undertaken with the intention of resolving differences in Angola but, on the contrary, had the effect of amplifying an otherwise local conflict. The involvement of outside powers put one more obstacle in the way of a negotiated solution and, once these links were forged, the political orientation of the respective movements crystalized so that power-sharing—if it ever was seriously considered by the parties—became among the remotest of possibilities. John Stockwell, the CIA’s station chief in Angola, remarked that, in September 1975, UNITA leader Jonas Savimbi caused “embarrassment” in Washington when he made overtures to the MPLA for a negotiated solution.12 Similarly, Washington rebuffed any overtures the MPLA made to it. According to Stockwell, when a delegation came to Washington in October to “plead the MPLA’s potential friendliness towards the United States,” they were met only by a “low-level State Department officer who reported perfunctorily to the working group.”13 The introduction of South African troops to bolster non-MPLA forces in late October 1975 prompted a massive shipment of Soviet arms and the direct intervention of Cuban troops in order to rescue the MPLA,

184

Civil War in African States

whose control of Luanda was increasingly threatened. When November 11 arrived, there were two declarations of independence. From the capital, Luanda, Agostinho Neto was sworn in as president of the People’s Democratic Republic of Angola and, in Lisbon, UNITA announced that a government would be established in Huambo (Nova Lisboa). Since the Portuguese government had studiously avoided handing power over to any one faction, the fate of Angola’s liberation movements was decided on the battlefield and in the world’s foreign ministries. No state recognized the government in Huambo, and few were willing to recognize the MPLA government in Luanda. The tide turned in favor of the MPLA when it was revealed that troops from apartheid South Africa had intervened in support of UNITA and the FNLA. Pretoria’s intervention, then, inadvertently saved the MPLA insofar as it prompted neighboring governments to rethink which side of history they wanted to be on.14 In late November, Nigeria recognized the Luanda government, an action which was soon followed by Tanzania. In both cases, foreign governments justified their actions by citing the South African role in Angola.15 Believing that the resort to Cuban troops was now justified in the face of South African aggression, and that the MPLA no longer had any obligation to reconcile with the other movements, other African states followed suit by offering diplomatic recognition to the MPLA government in Luanda. The transition to independence, and Angola’s new balance of power, would influence the subsequent strategies of the surviving movements. From UNITA’s standpoint, the MPLA’s military fait accompli had blocked elections or a negotiated solution, either of which would have given them access to political power. 16 Now the MPLA remained in control of the capital only at the pleasure of Soviet and Cuban power—foreign intervention UNITA claimed it had wanted to avoid from the beginning.17 Consistently, UNITA’s subsequent rhetoric emphasized the MPLA’s undeserved domination of political power in Luanda and the need to expel the 20,000 to 30,000 Cuban troops now estimated to be in Angola. But in justifying its war against the MPLA, UNITA vacillated between two extremes. Publicly, as the weakest of the three parties, UNITA was the most willing of the three to share power. UNITA’s representative in Washington, Jeremias Chitunda, claimed that war had been forced upon UNITA and that their efforts were directed only towards pressuring the government to allow it to contest elections that had been denied at the time of independence. “No government can effectively function in Angola unless it fully includes UNITA as a major national party,” he argued. “Notwithstanding its devastating effect, war is being pursued because it has been imposed upon Angola. Peace in

The Angolan Civil War

185

Angola requires a political solution based on the formation of a government of national unity rather than on a foreign-fanned attempt at eliminating the grass-roots party, UNITA.”18 Jonas Savimbi, by contrast, claimed in a less-guarded moment that “Any stability in southern Africa, or any peaceful solution of the problems facing Africans in the area, will only materialize if the bogus M.P.L.A. Government in Luanda is liquidated immediately.” He added that, “So long as the Western countries flirt and compromise with the Luanda regime, this will only help to consolidate its position.”19 For its part, the MPLA saw itself as the only legitimate movement in Angola—a standpoint that justified its exclusion of other movements. “We, the MPLA, hold that the only liberation movement in Angola is the MPLA,” stated Agostinho Neto. “And for this reason we do not have to sit again at the negotiating table with UNITA and FNLA.”20 The MPLA viewed their adversaries a s illegitimate political movements—mere fronts for the evil outside interests of the United States and South Africa, whose military assistance was the only reason UNITA and the FNLA survived. “We do not regard UNITA or FNLA as liberation movements. We regard them as invading forces, at the service of imperialism, with whom we no longer have anything to talk about.”21 This had an obvious impact on MPLA strategy: negotiations between UNITA and the government were said to be “an historic impossibility,” and UNITA was unequivocally regarded as an illegitimate force. For the MPLA, like Savimbi, domination meant elimination: “The People’s Republic of Angola sees as a priority and a vital and inalienable task the expulsion from our country of the army made up of [South African] and Zairian troops, Portuguese fascists, Angolan puppets [i.e. UNITA and the FNLA] and mercenaries who represent the combined force of imperialist aggression in our country.”22 Other events in early 1976 reaffirmed that the balance of power was shifting in favor of the MPLA. The fall of several key cities, including Silva Porto—the headquarters of UNITA during the independence struggle—forced UNITA to announce the establishment of a new headquarters at Serpa Pinto, a plan which also had to be abandoned as Cuban-backed MPLA forces rolled south. Additionally, South African troops had decided to withdraw before being confronted with an expected January joint offensive by Cuban troops and the MPLA’s armed wing, the People’s Armed Forces for the Liberation of Angola (FAPLA). Even the United States appeared in full retreat after the U.S. Congress passed the Clark Amendment which ended American covert military assistance to the Angolan rebels. As reported at the time, “With convincing victories on military and political fronts achieved by the

186

Civil War in African States

middle of February, it had become abundantly clear that the . . . (MPLA) had emerged victorious,” and the Organization of African Unity (OAU) recognized the MPLA “as the sole government of Angola.”23 The defeat of UNITA appeared to have finally been achieved once and for all. In fact, statements of total victory proved to be premature. As President Neto admitted on April 5,1976, “the war is not entirely over.”24 While the FNLA had been fatally weakened, UNITA was transforming itself from a semi-conventional force into a clandestine guerrilla movement. UNITA’s retreat south had ended in 1979 when the rebel leader established a new base at Jamba in the remote south-easterly corner of the country. From there, Savimbi would continue his war against the government. Since South Africa’s chief foreign policy concern was the prevention of new Marxist regimes on or near its borders, Pretoria continued to provide substantial military and financial support to insurgent movements such as UNITA.25 While it was principally a base of operations for UNITA, Jamba also came to resemble a mini-state, which UNITA officials referred to as “The Freeland of Angola,” providing schools and health services, and housing UNITA’s putative government. Savimbi’s state-within-a-state, under the constant protection of the South African military in neighboring South West Africa, served as a way-station where UNITA could rebuild its forces for the day when it would be able to challenge government authority in the rest of the country. Jamba was not a secessionist enclave—the barren and inhospitable territory contained neither sufficient quantities of Angola’s mineral wealth nor enough UNITA supporters to be a viable homeland.26 Instead, the territory served a number of other functions, not least of which was as a public relations center. Foreign journalists, politicians, and supporters were flown into Jamba and given the opportunity to report back on the movement’s capacity to provide capable leadership for the three million people who resided in the territory. Just as importantly, however, Jamba was a symbolic provocation to the government in Luanda. As one journalist with the New York Times observed, “Lodged like a bone in the throat, [Jamba] offered a permanent challenge to Luanda’s authority, to its ability to implant policies that might ordinarily have improved the lives of Angola’s people. It denied the very title that MPLA had won for itself as the Government of the People’s Republic of Angola.”27

The Angolan Civil War

187

UNITA Emboldened, the MPLA Recalcitrant: The Cold War Stand-Off

While having relatively modest beginnings at the time of Angola’s independence, UNITA’s strength in numbers of personnel and amount of equipment increased during the 1970s and 1980s, as a result of expanded South African protection and support and, after 1985, renewed military patronage from the United States. In 1985, even before the repeal of the Clark Amendment restrictions on American assistance to the rebels, UNITA officials were buoyant about the movement’s military achievements. “We have made some impressive breakthroughs, some irreversible gains, and militarily it is only a matter of time” before the MPLA would be forced to negotiate, boasted Jeremias Chitunda, UNITA’s foreign secretary. “We have a totally winning situation today,” he continued. “If you asked me a year ago, I would not have been as optimistic. ... [But now] we are everywhere, north, south, east and west.”28 At this time, while its base of operations remained in Jamba, UNITA claimed to be dominant in more than one-third of the country and to have a significant presence in two-thirds. Not only did this presence allow UNITA to challenge government control, but it allowed the rebels to exploit Angola’s lucrative diamond fields.29 Nonetheless, UNITA’s military strength remained inferior to that of the MPLA. The American Assistant Secretary of State for African Affairs, Chester Crocker, remarked that an outright UNITA victory “would become possible only if the SADF [South African Defence Force] intervened massively and the Soviets and the Cubans walked away—an unlikely scenario.”30 So too did more sober assessments of UNITA’s military capability by UNITA officials tend to play down the possibility of total military victory. According to a senior UNITA official, Abel Chivukuvuku, “UNITA’s perspective was always that the war would have to end in negotiations, because a guerrilla movement rarely goes to [the point where it can] take power by force. Its [purpose] is to force the established government to negotiate a peaceful settlement.”31 UNITA members claimed that UNITA had other key assets which could be brought to bear on the conflict. One former UNITA guerrilla fighter noted that although there was no question that the MPLA government was capable of mobilizing more weapons and more personnel, UNITA’s relative weakness in these respects was compensated for by its political sense of purpose: “Their political ideals, their determination, their expectations—they had hope for the future.” UNITA fighters, he said, were frequently reminded that “Angola is

188

Civil War in African States

being led by Cubans and Russians, so we have to change the situation. We are fighting until we have elections in Angola. If we have elections we will win.”32 Savimbi himself conceded that there was “no prospect of a military victory” for UNITA.33 Instead, Savimbi claimed that his objectives were, first, like the United States, the removal of Cuban troops from Angola and, second, the establishment of a multiparty system for Angola. UNITA linked itself with other liberation movements aligned with the West, including the Contras in Nicaragua and the Mujahadin in Afghanistan—both of whom received American weapons supplies under the Reagan Doctrine.34 During a 1986 visit to Washington under the guidance of a Virginia public relations firm, Savimbi encouraged the view among conservatives that Angola was another test of the resolution of the United States to stand up to communism. “If you lose Angola to communism,” Savimbi warned his host at the end of his tour, “you lose southern Africa. By losing southern Africa, you lose access to strategic minerals that are critical to your economy. Also, you lose the strategic [trade] lines of the South Atlantic and Indian Ocean.” According to Savimbi, all that was required to accomplish UNITA’s objectives was a small demonstration of assistance from the Reagan administration: “With a little support,” he noted, “we can really bring the MPLA to the negotiating table.”35 While the Reagan administration did provide UNITA with military aid, the State Department refuted accusations that it was fueling the war in Angola, claiming instead that U.S.-provided assistance was insignificant compared to the $1 billion in military hardware recently provided by the Soviets. “It wasn’t us that made this into an East-West issue,” one administration official offered. “The Soviets have poured about $8.2 billion into Angola over the past 13 years. . . . Soviet aid dwarfs our aid to UNITA by 80 or 90 to 1 [and] they aren’t volunteering any cuts.”36 While U.S. aid to UNITA never matched the quantity of weaponry Angola acquired from the Soviet Union, aid did increase beyond the $15 million that was originally intended, albeit without full disclosure of the amount. “All we have said is that we are providing appropriate assistance–without spelling out how, where or how much,” Bea Russell, a State Department spokeswoman told the New York Times. “We are trying to provide some balance to the vast amount of assistance coming from the Soviet Union to the other side.”37 This increasing American and South African assistance put pressure on the Angolan government to reach a settlement. Moreover, UNITA was able to operate freely in the sparsely populated southern provinces of Angola, and the sheer vastness of the territory, along with the

The Angolan Civil War

189

certainty of South African reprisals should FAPLA encroach on Jamba, precluded a conclusive elimination of UNITA. The on-going attacks launched by UNITA guerrillas similarly prevented the MPLA from sending Cuban troops home. That Cuban troops provided essential security was not in doubt. The Angolan government claimed that, from 1981 to 1985, South Africa had conducted 168 bombing raids, 234 airborne troop landings, 90 strafing incidents, and four naval landings.38 In a letter to the United Nations, Angolan President José Eduardo dos Santos (who succeeded Neto in 1979), wrote that with the ongoing military threat from South Africa, a complete Cuban withdrawal from Angola would be “suicide” for his country.39 South African troops had also been found in Angolan territory after Pretoria had publicly announced its withdrawal from southern Angola in March 1984, and it was no secret that frequent South African incursions continued.40 “We cannot sign our political suicide,” one Angolan official said concerning Cuban withdrawal. “How can we trust the South Africans? They are a few hours away from Angola, while Cuba is at the other side of the world.”41 Both Cuban and Angolan officials resented what they regarded as a double standard on the part of the Americans, who recognized UNITA’s security interests but not those of the MPLA government in Luanda. Cuba’s deputy foreign minister remarked, We are not the source of the problem. The Cubans did not create the problem. We didn’t invade Angola, nor Namibia, we haven’t imposed apartheid on anybody in the area. We are not to blame for anything. There is no international resolution asking the Cubans to withdraw. And on the contrary, there are more than one expressions of recognition of the role Cuba has played there—the Non-Aligned Summit conference, the UN, and so forth.42

Savimbi’s erstwhile-sympathetic biographer, Fred Bridgland, argued that the obvious solution was to find accommodation for everyone, allowing for power-sharing, peace, and the repatriation of foreign soldiers. Savimbi had offered the MPLA a five year coalition government with UNITA, to be followed up by free multi-party elections. The deal was contingent on the departure of Cuban troops. Without that, wrote Bridgland, there would be no guarantee of safety for UNITA. It looked like a cycle of conflicting interests that would be impossible to break.43 Indeed, it nearly was. Angola’s president, José Eduardo dos Santos, had long held that negotiations were out of the question because it would mean that, through military aid, outside interests could force their

190

Civil War in African States

client-insurgent into a power-sharing deal. Any foreign state, he argued, could organize a destabilizing insurgency and then insist that the weakened government accommodate the insurgents.44 More critically, from the MPLA perspective, UNITA still lacked credibility because it was relying at this time entirely upon South African power. The MPLA claimed that UNITA’s military capacity was so closely intertwined with South Africa that, once that aid ended, UNITA would collapse as a military force. “Without Pretoria’s help, Savimbi couldn’t last a month,” argued Angola’s ambassador to the United Nations, Elísio de Figueiredo. “Savimbi is South Africa.”45 Similarly, Major Agostinho Nelumba, of the army’s general staff, stated, “We view the South Africans as our principal enemy. . . . If it were not for the South Africans, we would have resolved the [rebel] problem already.”46 Indeed there was considerable public frustration on the part of Angolan officials, who claimed that UNITA would have been wiped out during one of FAPLA’s major southern offensives in 1985 had it not been for the timely intervention of South African forces.47 By default, then, the MPLA pursued a strategy of domination. South African power delegitimized UNITA and thus ruled out any possibility of adopting a cooperative strategy which accommodated the rebels. “We do not have any interest in forming a government with the puppets,” argued Angola’s Vice-Minister of Foreign Relations, Venancio de Moura, in a 1988 interview: “That would mean sharing our power in Angola with the apartheid regime—which wouldn’t be desirable for our people. We fought a long liberation struggle to win our independence as Africans, Angolans, non-aligned, and for the political option we chose, Marxism-Leninism. The possibility of forming such a government is nonexistent.”48 For his part, Savimbi remained unapologetic that circumstances forced him to appeal to Pretoria for military assistance and claimed that he was less dependent on South Africa than the MPLA was on its Soviet and Cuban patrons.49 UNITA would undoubtedly have preferred more support from Washington, but, from 1976 until 1985, had been prevented from receiving it by the Clark Amendment. “Guns don’t drop from the skies,” Savimbi said in 1981. “I have to get them where I can.”50 The End of the Cold War: The Promise of Cooperation and Integration

In light of its inability to challenge the Angolan government militarily, Washington concentrated on the removal of Cuban troops from Angolan

The Angolan Civil War

191

territory. The diplomacy of “constructive engagement,” led by Assistant Secretary of State Crocker, linked two issues: the Cuban presence in Angola and Namibian independence.51 It sought to place Washington in the role of an “honest broker” in order to bring about an agreement that committed Angola to the withdrawal of Cuban troops in exchange for a promise from South Africa to grant Namibian independence under UN Resolution 435. Constructive engagement had the potential of providing simultaneous benefits to all the parties. “Constructive engagement was, I think, a very clever way of giving everybody something,” noted Marcos Samondo, a senior UNITA’s cadre.52 “The independence of Namibia,” noted President dos Santos, “can unchain solutions for all other problems in our region—it would create better conditions of security in the southern border of Angola which could be followed by the progressive reduction of Cuban troops.”53 With both parties agreeing on the possibility of a deal, what was required was an event that would move the protagonists to reconsider their present positions and options. The battles at Mavinga and Cuito Cuanavale in 1987 and 1988 proved to be important catalysts in this regard. For years South Africa had waged a mostly low-level war in southern Angola in defence of UNITA and against SWAPO guerrillas seeking the independence of South African-controlled South West Africa (Namibia).54 Few dispute the view that at Mavinga, as elsewhere, the South African forces had the upper hand. This changed dramatically at a subsequent battle in early 1988 at Cuito Cuanavale. As one former SADF officer recalled, “We never really expected them to fight back. In fact, it was really more like sport. Then there was Cuito Cuanavale. That was a shock!”55 An infusion of Cuban troops led to a battle that proved to be painful enough to speed up the on-going negotiations that were part of Crocker’s constructive engagement. On December 22, 1988, the New York Accords, consisting of a bilateral agreement between Angola and Cuba, and a tripartite agreement involving Cuba, South Africa and the government of Angola, were signed. The agreements included the following provisions: the withdrawal of South African troops from Namibia and the independence of Namibia under Security Council Resolution 435, a timetable for a UN-monitored Cuban withdrawal from Angola, and a pledge by foreign parties to respect the principle of noninterference in the internal affairs of the states of southwestern Africa. As Chester Crocker would later note, the agreement worked because it addressed a number of critical security concerns simultaneously: “UNITA would lose its regional military ally at the moment when the MPLA was losing the Cubans. The ANC would lose its last remaining

192

Civil War in African States

sanctuary within reach of South Africa just when SWAPO would become free to return home and seek power through democratic elections. The South Africans, rapidly losing credible enemies, would put Namibia on the path to independence and begin shedding their interventionist military role in neighboring countries.”56 The Accords were not without peril. While there were signs of reform in South Africa, the end of the apartheid regime was not yet a reality. Moreover, despite assurances in the form of American guarantees to keep South Africa in check, the New York Accords did not explicitly involve UNITA. Nonetheless, while the Accords did involve considerable risk for the MPLA, at least three factors made the agreement attractive. First, Angola’s increasingly dire economic predicament almost certainly was a factor. Coffee production was one-twentieth of what it had been at independence. The profitability of diamond mining was continually jeopardized by UNITA attacks on the key diamond areas. Angola’s Benguela railway had been sabotaged so many times that it was, for all intents and purposes, shut down. The extensive use of antipersonnel mines had left Angola with only two percent of its land under cultivation. Even Angola’s oil industry, which for the most part had been insulated from UNITA and South African attacks, was suffering from uncertainty in the price of oil during the mid-1980s. Most notably, however, in 1986 the ongoing need to fund Angola’s armed forces consumed an estimated 75 percent of Angola’s budget. In addition to putting thousands of troops in the field, Angola was purchasing billions of dollars in Soviet weaponry.57 A deal which gave Namibia its independence was in effect a trade-off, whereby Cuban troops were exchanged for a credible security cordon and the prospect of Western investment and recognition. A second factor in Angola’s willingness to sign an agreement came from pressure from the Soviet Union. Financial troubles in Angola constrained its ability to pay for continuing arms, and the Soviet Union was reluctant to provide support without payment. The reception President dos Santos received in Moscow in 1988 was reportedly cooler than a similar trip made in 1986.58 Increasingly there was an expectation that Luanda would finance its own war effort and that it had failed to meet its financial obligations. “We are counting our money everywhere,” observed a Soviet journalist in Luanda. “We are not making any profit here, political or economic. They were supposed to pay for the arms. Then they sent a letter to the Soviet bank saying that Angola was a poor country, a developing country, and they couldn’t pay.”59 The implication was that the Soviets had welcomed Angola as a

The Angolan Civil War

193

client because they were capable of paying. When this ability was constrained, Luanda would be treated like any other unreliable client and jettisoned. A final factor in Angola’s decision was the accepted view that, if they agreed to send Cuban troops home and get South African troops out of southern Angola (the latter, they hoped, guaranteed by Namibian independence), UNITA would be rendered powerless. “Let’s assume that South Africa fully abides by the agreements and puts an end to its help to UNITA,” said Cuba’s Deputy Minister for Foreign Affairs, Ricardo Alarcón, in an interview just prior to the signing of the agreement. “Then UNITA would not be able to use Namibia as a rearguard or a starting point for their attacks on Angola, and that certainly would put UNITA in its normal perspective. This big giant invented or fabricated in some parts of the Western media would be reduced to its real proportions.”60 For all the changes that were taking place, these events did not bring about a fundamental change in strategy. The MPLA was still embarking on a strategy of domination. From its perspective, any new regional or global realignment only offered renewed opportunities for the government to vanquish UNITA once and for all. Indeed, there was a genuine perception among Angolan officials that, with this agreement, the balance of power in Angola would shift decisively towards the government. Thanks in part to Cuban and Soviet training, FAPLA itself was seen as a much stronger military force and certainly one that was superior to what Luanda believed was the soon-to-be neutralized UNITA.61 The MPLA was willing to make a deal to send Cuban troops home—even if it was not yet prepared to consider a negotiated solution with UNITA—on the basis that it would still have the upper hand militarily, economically, and politically. The government’s principal strategy thus remained domination but now, it was hoped, the struggle was against a much-diminished adversary. From Washington’s perspective, the New York Accords were expected to help the parties to the Angolan conflict reach an agreement that benefitted both of them. “This is the end of two wars and we hope a third,” stated Assistant Secretary of State Crocker at the signing ceremony in New York. He added, “We think it can also lay the basis for Angolans to at long last recognize that for their internal question they have no real option except for a political solution, for political dialogue and reconciliation.”62 But Savimbi was annoyed at having been excluded from the talks and the possibility that an independent Namibia would cut UNITA supply lines to Jamba. He warned that an agreement signed without him

194

Civil War in African States

would not stop him from continuing the war.63 To be sure, while official South African assistance to UNITA had ended with the New York Accords, both the rebels and the MPLA continued to receive military assistance from their superpower patrons. Savimbi’s concern was that Washington was negotiating a deal that would not only fail to give UNITA a role in a proposed government of national reconciliation, but would also use the promise of an end in aid to UNITA as an incentive to get the Angolan government to sign the regional agreement.64 As if to remind the world that UNITA’s struggle continued as before, Savimbi claimed that he had the means to be the spoiler in any settlement on southern Africa. “I will have the means. . . . and I know how to buy guns from all over the world,” Savimbi stated. “I have some dollars with me already which didn’t come from South Africa or America.”65 If peace—and not some other superpower objective—really was the goal, Savimbi believed he would have to press Washington into maintaining its commitment to UNITA until national reconciliation had been achieved.66 On January 26, 1989, the Angolan government publicly acknowledged for the first time that it had recently engaged in limited discussions with UNITA officials. It also made an informal call for a ceasefire, suggesting that a more cooperative strategy was considered an option in at least some quarters of the government.67 But clashes between the two movements had continued well into 1989, and it became increasingly clear that, even without its South African patron, UNITA’s power was not as marginal as the MPLA had hoped. “To say UNITA’s impact is negligible is ridiculous—when you can barely drive out of Luanda,” noted one diplomat.68 While few believed that UNITA had any chance of overthrowing the government, it was equally clear that the MPLA could not suppress the UNITA challenge entirely. “The real problem is not one of numbers, but rather the type of war,” observed one relief worker, referring to UNITA’s guerrilla warfare tactics. “[The Army] can’t destroy UNITA altogether, because they are everywhere and nowhere all at once.”69 Indeed, UNITA remained a force to be reckoned with, and, not surprisingly, the MPLA felt a sense of betrayal and frustration, claiming to have been led to believe that U.S. aid to UNITA would end with the signing of the New York Accords.70 In January 1989, President Bush sent Savimbi a letter informing that him that the United States would continue to provide “all appropriate and effective assistance to UNITA” until a political settlement could be reached.71 Nonetheless, there were accusations that there had been a deliberate campaign to give the impression that American military supplies to UNITA had ended. The

The Angolan Civil War

195

November 1989 crash of a CIA plane that was delivering military supplies to UNITA proved embarrassing to Washington, given that it occurred only days before a meeting between U.S. President George Bush and Soviet President Mikhail Gorbachev off Malta where the former had planned to complain about on-going Soviet weapons shipments to various clients.72 In December 1989, when the MPLA launched yet another offensive against Mavinga, the United States rushed in an emergency infusion of military aid, including advanced Stinger missiles.73 Howard Wolpe, chairman of the House Subcommittee on Africa, also accused South Africa in June 1989 of having “extended a golden handshake of up to three years of military supplies to UNITA” before it formally ended aid on April 1, 1989.74 There were, however, continuing pressures on the superpowers to end their support for their Angolan clients and to facilitate a negotiated solution to the Angolan conflict. “Angola was a conflict that for the better part of two decades had been elevated by geostrategists into a pillar of Cold War rivalry,” wrote U.S. Secretary of State James Baker. “While this confrontation was understandable in an earlier time, by 1989 neither of the superpowers had overriding reasons for being embroiled in this conflict. It was time to move on to more critical issues—and the Angolan civil war seemed ripe for resolution.”75 Indeed, with fewer ideological justifications for continued violence and with both sides increasingly ready to acknowledge that the war could not be won militarily (but might be won via the ballot box), an agreement between UNITA and the MPLA now seemed to be a distinct possibility. A Negotiated Settlement

The first effort to achieve this end was a failure. In May 1989, President dos Santos held a summit of African heads of state in which he presented his own plan for peace with UNITA. The substance of the agreement was an unpublished offer to induce Savimbi to “temporary and voluntary retirement,” while UNITA cadres would be integrated into the “institutions of the People’s Republic of Angola.” Mobutu Sese Seko, President of (then) Zaïre and a strong supporter of UNITA, was asked to present the plan to Savimbi. According to Assistant Secretary of State Herman Cohen, Mobutu now saw power-sharing—albeit one that personally excluded Savimbi—as “the key to peace in Angola.” Negotiations, he said, “would consist of putting all the top governmental and military jobs on the table and sharing them between the MPLA and UNITA.”76

196

Civil War in African States

There was, however, no mention of Savimbi’s demand for elections or an interim government. While Savimbi had publicly signed the agreement—in front of African heads of state at a subsequent summit at Mobutu’s retreat in Gbadolite, Zaïre—it soon became unclear whether Savimbi was aware of, or fully grasped, the nature of its provisions. Savimbi had dismissed earlier proposals that included UNITA but excluded him: “Sometimes people try to divide UNITA and Savimbi. The Angolan government may be so afraid of me, but let me stop dreaming. When you are saying UNITA you say Savimbi. Talk about UNITA, I am included.”77 Speaking of the Gbadolite proposal, Savimbi became emphatic in his denials that he had accepted such an agreement. “Why should UNITA, which has never been stronger militarily and politically,” Savimbi asked, “surrender to the Luanda Government in negotiations what the Government could never obtain on the battlefield? UNITA has not fought 22 years, first against the Portuguese colonial rule and then against the Soviets and Cuban-backed regime, only to throw up its hands in surrender.”78 Savimbi then gave an indication of the only kind of political bargain that he would now find acceptable: “UNITA is not afraid of elections. Only through elections will the Angolan people get a government that they see as legitimate, one that will inspire them to work together to reconstruct the nation.”79 Herein lay the core of the problem: neither party was willing to accept a political outcome that was not commensurate with their respective aspirations to dominate the other. “Mr. Savimbi is no democratic freedom fighter,” wrote Flora Lewis in the New York Times, “but he is an effective leader with a genuine constituency. He is too formidable for any idea of power-sharing to be taken seriously.” For his part, President dos Santos put it another way: “In Africa we have oneparty states. We think this system suits us in this phase. There can only be one party.”80 According to the MPLA’s rhetoric, an obstacle for power-sharing was not the difference—ideological or otherwise—between the movements as much as the personal problem of the two avowed enemies. While the Angolan government was willing to conduct limited, secretive negotiations with UNITA, it was not yet prepared to consider power-sharing with Savimbi himself. The integration of UNITA and the MPLA might be negotiable; the merging of its respective leaderships was not. Angola’s Foreign Minister, Pedro de Castro Van-Dunem ‘Loy,’ stated: “As Savimbi is the spiritual father of the hideous crimes against our people, at this moment he cannot be encompassed within the policy of clemency. Maybe perhaps after some years, but we cannot find any

The Angolan Civil War

197

justification before our people to explain why we would include Savimbi at this point.”81 Nonetheless, the government was beginning to make concessions. In March 1990, President dos Santos announced that explicit references to Marxism-Leninism would be dropped from official documents and that non-MPLA members could run for political office. He also accepted the “possible evolution of our political system into two parties.”82 Finally, in early April 1990, in Cape Verde, the MPLA publicly announced its plans for “exploratory talks” with UNITA. The first of several rounds of negotiations between UNITA and MPLA representatives began in late April 1990 in Portugal. By November 1990, debate continued over UNITA’s demand for recognition by the MPLA.83 Then, during a December 1990 meeting between Soviet Foreign Minister Eduard Shevardnadze and James Baker in Houston, the two superpowers agreed to apply simultaneous pressure on their respective allies and present them with a settlement plan.84 A peace plan drawn up in part by the American Assistant Secretary of State, Herman Cohen, and his Soviet counterpart, Yuri Yukalov, on December 11 was then presented to representatives of each movement at a meeting in Washington two days later. With the Washington plan forming the basis of an agreement that linked a ceasefire to an election date, the most problematic issue became deciding on the timing between the two.85 Following another round of negotiations in April 1991, the Estoril Accords were initialed, providing both a ceasefire and a settlement to the civil war. On May 31, 1991, President dos Santos and Jonas Savimbi formally signed the Angola Peace Accords (formally called the Bicesse Accords). Included in the accords were the following provisions: (1) An agreement on a ceasefire, the demobilization of FAPLA and UNITA’s armed wing, the Armed Forces for the Liberation of Angola (FALA), and the creation of a new integrated army; (2) Free and fair multi-party elections to be held in September 1992 under UN supervision. A president and a national assembly were to be elected, the latter of which would be elected according to a system of proportional representation; (3) A “triple-zero” option, whereby each of the United States and the Soviet Union pledged not to provide further military aid, and the superpowers would publicly urge other countries (most notably South Africa) to follow suit; and (4) The formation of a joint MPLA-UNITA commission to monitor the ceasefire under UN supervision. Critical to bringing the parties into an agreement was the threat that their relative military power would be undermined by the superpowers if they were found to be the main obstacle to an agreement. As Assistant

198

Civil War in African States

Secretary of State Cohen later acknowledged, superpower patronage generated confidence and invulnerability among the protagonists, which precluded them from making concessions.86 Only an end to patronage could unlock the Angolan conflict. Following the meeting between Baker and Shevardnadze, the MPLA was reminded by Washington that U.S. support for UNITA would not end unless a settlement was reached. But UNITA was also told by Shevardnadze that its American support would end if it was viewed as the only obstacle to a negotiated settlement.87 In fact, however, elections now offered each movement an appealing alternative to war. As Linda Heywood has argued, the signing of the Bicesse Accords was a personal victory for Savimbi: although still powerless, he could claim to have been vindicated for fifteen years of guerrilla attacks and conventional warfare against the government.88 While there had been earlier indications that UNITA was willing to accept a power-sharing deal, now Savimbi reaffirmed his view that the only way forward was an election which did not necessarily involve power-sharing concessions: Some might argue that it is unrealistic to expect the outcome of the Angolan negotiations to be democratic elections, that some form of integration is all that can be expected. This brand of elitism seems to assume that the Angolan people are incapable of democracy, their long struggle to achieve it notwithstanding. It fails, moreover, to explain why the world community supports free elections in Poland and Namibia—our southern neighbor—but would settle for integration in Angola.”89

Indeed, elections offered Savimbi much more, not least the opportunity to win total power. A winner-take-all election meant that, in the event of their victory at the ballot box, no substantive concessions would be required. UNITA had long held that its population base was sufficiently broad to assure electoral victory. UNITA’s confidence in this was now bolstered by the fact that Marxist regimes were collapsing in Africa and elsewhere.90 Given the government’s own economic mismanagement, it seemed likely that the same would happen in Angola. In the face of increasing lack of interest from the United States, elections offered UNITA a strategy which would finally, legitimately, offer them a realistic chance at power—an opportunity which, despite their military gains, appeared not to be achievable through military means. “UNITA, largely Savimbi, has been very clear,” observed a diplomat in Luanda. “Their goal is to take power basically. If one accepts that simple premise everything that they have done makes sense.

The Angolan Civil War

199

They joined the Bicesse process because they really did believe they would win the elections. If they hadn’t believed that they wouldn’t have opted for democracy. Democracy was a strategy not a policy of UNITA at that time.”91 The Angolan government also had reason to celebrate the agreement at Bicesse. To be sure, the MPLA could no longer dismiss UNITA now that the rebel movement had been recognized in a negotiated settlement. However, what had been up for negotiation was only the electoral rules, not Savimbi’s guaranteed allocation of power. Power-sharing had long been unacceptable for the reasons stated by dos Santos—that Savimbi could not personally be allowed to ride into power on the coattails of American and South African power. For the incumbent, elections held obvious risks, and the MPLA certainly had more to lose than UNITA. Yet, the end of the Cold War and the internal stalemate that continued meant that, like UNITA, winner-take-all elections were the only option now for the MPLA. While the MPLA fully expected that they would win, they also found competitive elections acceptable because UNITA would be judged on its own merit and, the MPLA thought, would finally be revealed for what it was. Since the time of independence from Portugal, the MPLA had believed that Angolans saw it as the only legitimate liberation front. An electoral triumph would mean, not cooperation with UNITA, but that the rest of the world would have no choice but to recognize its rightful role as the dominant party of Angola. For both parties it was the reverse of Clausewitz’s famous dictum: politics had become a continuation of war by other means. Unlike the New York Accords, however, Bicesse did not conclusively satisfy the security needs of both parties. Instead of resolving Angola’s conflict, Bicesse put it in abeyance. In the New York Accords, by contrast, each party had different and complementary objectives which could be met concomitantly provided they were matched by quid pro quo in another area by their respective adversary. An agreement to conduct winner-take-all elections, on the other hand, provided a means to avoid or temporarily delay dealing with the central objectives of two movements that, by their very nature, could not be met concurrently. At the end of the negotiations process at Bicesse, then, there were still two formidable military powers contending for a political outcome in which, as it stood, there could be only one winner. Power-sharing might have mitigated this tension, but it would have required the leaderships of both the MPLA and UNITA to accept less than their optimal outcome—something neither was prepared to commit to in a formal peace agreement. Nor did they feel the need to do so. Even in a situation in which a movement could conceivably lose, there

200

Civil War in African States

were few incentives to forming a power-sharing deal when they had the option of absolute political power. Elections as Part of a Strategy of Domination

The first UNITA officials arrived in Luanda in June 1991 and held a pro-UNITA rally, which attracted an estimated ten thousand supporters. UNITA leader Jonas Savimbi was not among the officials. He stated that he was waiting for an invitation from President dos Santos to come to Luanda: “I have asked President Eduardo dos Santos for a house ... and the day that they tell me it is ready, I will go to stay.”92 While the Bicesse Accords made no mention of power-sharing or a transitional government of national unity, such ideas did receive limited attention among some of Angola’s political leaders who were questioning the wisdom of a winner-take-all electoral outcome. Uncertain of the electoral outcome, President dos Santos gave vague assurances on Portuguese television that a deal could be considered, and his advisor, Lopo do Nascimento, remarked that a coalition government might even be preferable: “I think it is possible,” he told the Christian Science Monitor. “I think it is desirable. I think a government of national unity would be appropriate to cure the wounds of war.”93 Savimbi, however, convinced that he would win any election race, spoke less about power-sharing, but promised to respect the decision of the people in the event of a loss: “If I lose the election, this is my country and I am an ordinary citizen,” declared Jonas Savimbi. “No one will push me to go back to the bush anymore.”94 Indeed, UNITA officials were brimming with confidence that the election was theirs, and, two months before voting day, were busy considering the logistical formalities of the anticipated transition to UNITA rule.95 But events in UNITA’s recent history came back to haunt Savimbi and to undermine its image as a democratic movement. In February 1992, two senior UNITA cadres defected from UNITA, bringing with them revelations of Savimbi’s authoritarian nature and, in particular, his role in the murder of two high-ranking UNITA officials—Tito Chingunji, UNITA’s former representative in Washington, and Wilson dos Santos, Savimbi’s brother-in-law.96 Worse, it was alleged that Savimbi had engaged in the practice of witch burning, apparently as a means of eliminating these and other perceived rivals to his leadership. UNITA officials claimed that Savimbi had had no knowledge of the deaths, but nonetheless these reports prompted the U.S. Secretary of State to call for a “full explanation” of UNITA’s alleged human rights abuses.97

The Angolan Civil War

201

On the campaign trail, UNITA’s fortunes were also souring. UNITA officials began to worry privately that the campaign would not be able to attract the support it would need to win.98 Publicly, hostile political rhetoric between the MPLA and UNITA caused uneasiness. One diplomat warned in June 1992 that both sides “were finding it extremely hard to make the transition from being bitter antagonists to accepting the notion that they can peacefully coexist. . . . They still tend to respond to every challenge with threats of violence.”99 While strained relations were to be expected, given the years of violent conflict between the MPLA and UNITA, the tension was no doubt a result of increasing uncertainty over the election’s outcome. While dos Santos did enjoy personal popularity as the leader who had secured peace for Angola at Bicesse, his government was held in contempt for its corruption and incompetence. To improve its fortunes, the MPLA hired a Brazilian public relations firm to direct and soften its political campaign. By contrast, Savimbi’s own rhetoric was increasingly belligerent. “If I lose then the elections were rigged,” Savimbi allegedly told a British television crew. “I will send my men back to the bush to fight again. We will not accept defeat.”100 As well, Savimbi’s use of anti-white and antimestiço language, and his pledge to purge state-sector employees who had previously supported the MPLA, led many Angolans who were dependent on the government to worry that support for UNITA could be costly.101 The UNITA-led electoral sweep predicted by many had unexpectedly turned into a contest in which the outcome was too close to call. More disconcerting than the apparent shift in voter support away from UNITA towards the MPLA was a growing and contrary shift in the balance of power between their respective armed forces. Reports emerged that UNITA was not complying with its demobilization obligations. On September 9, Savimbi privately informed Assistant Secretary of State Cohen that, while he had urged his commanders to respect the electoral outcome, they were telling him that “an election was not necessary to win power.”102 The government, by contrast, was having trouble keeping its demobilized troops from deserting the assembly areas. By the end of September, 65 percent of MPLA and only 26 percent of UNITA troops had been demobilized.103 Fearing the consequences of this discrepancy between military power and electoral support, Cohen admitted that it was not hard to wish for a UNITA victory.104 Indeed, the central concern was not just who would win the elections but whether the two leaders could now even find a way to coexist during the rebuilding of a democratic Angola. In the month prior

202

Civil War in African States

to the election, there was renewed pressure put on the two leaders from the international community to form a government of national reconciliation or coalition government. The MPLA and UNITA, however, resisted the latter and paid only lip service to the former, claiming that their preference was to win the election and then invite representatives of various political organizations to participate in the new administration. Margaret Anstee, the UN Special Representative to Angola, remarked that the kind of partnership which a coalition would require was not likely, given the two leaders’ purported ideological differences, their radically dissimilar personalities, and the “deep personal animosity” that characterized their relations. “The gulf of personal mistrust between the two was so vast as to be probably unbridgeable,” she stated. “I hardly ever attended a meeting with either that did not contain some comment of suspicion or disdain—even contempt—for his rival.”105 Peace was, however, maintained during the pre-election period in spite of the animosity and bellicose rhetoric between the two movements. Elections still provided the most direct and least costly means by which one side or the other could legitimately gain complete political power and neither UNITA nor the MPLA was prepared to jeopardize this opportunity by undertaking overtly violent actions. Anstee states that “the ceasefire held—even if sometimes by a hairline—simply because, when push came to shove, neither side wanted to imperil the elections, which each expected to win (at least in the latter stages).” The elections would be, noted Anstee, the “critical watershed,” and cooperation, to the extent that it occurred, was to advance an electoral process which both sides expected to win.106 When the elections came on September 29 and 30, 1992, the initial reports from both sides were encouraging.107 However, by October 3, before the vote-counting was complete, but after early reports put the MPLA and dos Santos in the lead, Western sources were hearing accusations from UNITA that the election had been rigged. A statement made by Savimbi on UNITA’s radio station, Radio Vorgan, accused the MPLA of stealing and “misappropriating the ballot boxes.” He said that the “MPLA is not winning and the MPLA cannot win,” and claimed instead that “UNITA is running ahead in the legislatives and presidentials, as a just and noble recognition of those who fought for the liberation of the country.” More ominously, he issued the vague threat that “all this will make UNITA take a position which could deeply disturb the situation in this country.”108 In an interview with Portuguese radio, a senior UNITA official, Elias Salupeto Pena, was more to the point: “The situation is so grave,” he stated, “that we cannot imagine the

The Angolan Civil War

203

publication of such fraudulent results because this will mean immediate war.”109 With United Nations officials and other observers refusing to acknowledge that any sort of systematic vote-rigging had occurred and, given UNITA’s failure to substantiate its own claims, these threats were widely viewed as a reflection of UNITA’s frustration with its apparent loss in the elections. As the post-ballot situation intensified, some election observers speculated that President dos Santos had received more than 50 percent of the vote—a dangerous outcome if it meant Savimbi would take up arms again. “Presidentially, in fact, I think dos Santos actually won the election,” noted one observer. “Then there was some backpedaling and recounting in a desperate attempt to allow a second round—hoping that would entice [Savimbi] into not walking out.”110 The official results, announced on October 17, put President José Eduardo dos Santos at 49.57 percent of the presidential ballots, and Jonas Savimbi at 40.07 percent. In the legislative voting, the MPLA won 53.74 percent and UNITA won 34.1 percent.111 Since dos Santos officially won less than the 50 percent required for an outright victory, a second, run-off election was required to decide the presidency. Given this requirement, frantic efforts were made to keep Savimbi in the election process and to persuade him to contain his rhetoric. Then, on October 16, Savimbi stated his willingness to accept the “admittedly fraudulent” legislative election results and to participate in a run-off election for the presidency. By the end of October, however—and Savimbi’s pledge not to return to the battlefield aside—Angola slid back into a war more violent than it had seen in recent years. Obviously, Savimbi’s rejection of the results was inconsistent with UNITA’s long-professed demand for elections in Angola and with his promise to respect their outcome.112 But UNITA’s real political aim was always the achievement of power. War, UNITA argued, had been necessary to bring about free and fair elections. And elections, which UNITA believed it would win owing to the broad population base among the Ovimbundu that it claimed to represent, were merely the latest means to advance UNITA’s strategy of domination. Now, in addition to the public humiliation of losing the election and having his political aspirations thwarted, Savimbi had to face the prospect that his seventeen-year war against the MPLA had been in vain. The Return to War: Savimbi’s Shifting Strategy

Given the stakes and his pre-election bellicosity, it is hard to believe that Savimbi did not see war as a possible outcome. However, several factors

204

Civil War in African States

point to the view that the threat of a return to war had more to do with brinkmanship designed to pressure the government into a favorable power-sharing settlement than with a pre-meditated plan to take over the country. Much of UNITA’s leadership was lodged in downtown Luanda during and after the election. If UNITA had expected war, these cadres would have been, and indeed were, vulnerable to a government counterattack. The diaries of Jeremias Chitunda, a senior UNITA cadre killed during the subsequent fighting, expressed surprise that UNITA was being portrayed by the international community “as the ones causing the new civil war.” He states that UNITA’s military operations were only for the purpose of “strengthening our negotiating hand” in the aftermath of the election loss. Chitunda was also aware of the limitations of military pressure and opined that “thinking that one can only negotiate after a Gana II [war] is a bad habit.”113 If, however, war had not been part of Savimbi’s plans, it was because he did not expect an electoral loss—other than the possibility of one engineered by MPLA malfeasance. UNITA’s leadership had interpreted assurances from American and South African officials to mean that they were sure to win.114 Ironically, Savimbi may have put too much trust in the democratic process to deliver on his long-sought ambition and thus failed to recognize its inherent unpredictability. His lashing out in the days following the voting was indicative of his frustration and surprise that UNITA’s—and his own—political aspirations had been, in his opinion, undermined yet again.115 As Assis Malaquias has argued, at independence UNITA maintained that the Ovimbundu’s relegation to subordinate status was accomplished only with the aid of the Russians and Cubans, who had helped place and keep the dominant MPLA in power. Now the Ovimbundu were being kept out of power by a grossly manipulated political process.116 Indeed, statements in the Chitunda diary suggest that the charge of rigged elections was not just a ploy but something that UNITA leaders genuinely believed to be true.117 The election had been close enough that the significance of any electoral irregularities—and some irregularities did occur—was magnified.118 UNITA’s conviction that the voting process had been faulty took on greater significance given the fear that the 1992 elections would legitimize the MPLA strategy of domination once and for all. As one election observer noted, “I think that the perception was that this was the one election. . . . [There was this] mistrust and fear of each other and I don’t think there was a tremendous amount of confidence that the party that wins the elections would not find an excuse to change the constitution to maintain their power.”119 To no avail, Margaret Anstee

The Angolan Civil War

205

downplayed the significance of the elections and urged Savimbi to consider the role of an opposition leader until elections were held again.120 Given the enormity of the task ahead—rebuilding a wardevastated country, reforming a Marxist-oriented economy, and reining in the electorate’s inflated expectations for post-election prosperity—UNITA’s best chance might come in a subsequent round of elections.121 UNITA’s decision-makers, however, preferred a return to a military route to achieve their objectives, rather than waiting for another election. Chitunda records in his diary that there was still “a great distance between the two parties,” and added that “the government puts a priority on stopping the war, while we put priority on stopping the causes of war.”122 Stopping the war meant a new deal. While negotiators from both sides were willing to consider an agreement, UNITA was willing to proceed as if the elections had never happened: “They [the government] want it to reflect the results of the legislative elections, which means they would have a majority,” Chitunda wrote in his diary. “[W]e want a GTRN [Transitional Government of National Reconstruction] administered on a parity basis, working together—at least in the key ministries.” Later an emboldened Chitunda wrote, “Under the present circumstances, our new negotiating position will be rejection pure and simple, of the election process and we will start from zero. . . . But it all depends on the outcome of this fighting.”123 In the meantime, the political balance of power shaped by the MPLA’s electoral victory meant that a negotiated agreement, likely a power-sharing one, was UNITA’s only lifeline. The government’s sense of vulnerability, combined with UNITA’s fear of being shut out of the political process, created an extremely precarious environment. It was not yet clear what the outcome would be. From that point on, each group’s efforts to reduce its own vulnerability was interpreted by the other side as evidence that its opponent was out to destroy it. For the MPLA, the most obvious threat to the peace was Savimbi’s rejection of the election outcome. This was followed by the mobilization of UNITA troops against several Angolan cities, including Huambo, and other strategic locations throughout the country. Sensing its own vulnerability, MPLA officials believed that it could not afford to see UNITA’s actions as anything other than aggression. Savimbi likewise perceived the government’s actions as threatening. The government had distributed weapons to pro-MPLA vigilantes and had organized attacks by the government’s remaining military force, the emergency anti-riot police, or ninjas. In the days after the election,

206

Civil War in African States

Savimbi’s two nephews, General “Ben-Ben,” and his brother and UNITA’s second-in-command, Salupeto Pena, remained in Luanda as part of UNITA’s negotiating team. As tension and violence continued to increase between the two sides, Savimbi voiced fears for their safety, stating that, if they were killed, full-scale civil war would erupt.124 When reports emerged that they, along with Jeremias Chitunda, had been ambushed when trying to escape from Luanda, UNITA accused the government of luring them into a trap and then trying to decapitate the senior leadership. Later Savimbi responded to the reports of widespread killings of the Ovimbundu people in Luanda and elsewhere by referring to them as “genocide.”125 In light of the certain condemnation that UNITA would face from friends and foes alike, UNITA’s return to war might have been a colossal error, “except for the fact that [UNITA] almost won,” noted a diplomat in Luanda. “It was awfully tempting. Through most of 1993, my sense was that they were still winning. They were all over the country.”126 And from Savimbi’s perspective, denying political power to the MPLA was no worse a crime than what UNITA had experienced at the time of Angola’s independence. Indeed, with much of the Angolan armed forces in disarray and with UNITA troops making rapid advances in both rural and urban areas, Savimbi was undoubtedly convinced that, despite his electoral loss at the polls, he could now realize his long desired domination of the country. “Savimbi wants the whole pie,” one Western diplomat told the New York Times. “And in his position, with most of the country effectively under his control, why should he compromise?”127 UNITA’s victory in the key oil town of Soyo in the northwest in January 1993, along with its control of the diamond areas in the northeast, assured that, at least temporarily, the government would be denied the revenues that had financed its earlier military efforts.128 Then, following an eight-week battle, UNITA captured Angola’s secondlargest city, Huambo, on March 6, 1993.129 The battle appeared to be a turning point: “This was a new kind of war,” noted Jorge Valentim, UNITA’s Information Secretary. “We are no longer a guerrilla army. It is conventional now. From now on, the towns and cities are our battlefields. We will defend them to the death.”130 With the victory in Huambo, UNITA’s ability to set the terms of a new bargain where UNITA was in control appeared to have increased exponentially. In a statement on Radio Vorgan broadcast from his Huambo headquarters, Savimbi boasted, “We have now captured many, many tanks here in Huambo. We have anti-aircraft guns, long-range guns and we have ammunition. We can, therefore, continue this war for a long time.”131

The Angolan Civil War

207

The shift in the balance of power was so dramatic and the prospects for MPLA success so grim that foreign observers wondered aloud if UNITA could even take Luanda. “It’s not inconceivable that UNITA leader Jonas Savimbi could conquer the joint,” noted one diplomat. “But he would have to kill just about everyone in Luanda to achieve that.”132 Indeed, the struggle between UNITA and the MPLA in the aftermath of the 1992 elections provides insight into the relationship between the balance of power and the strategies of the two principal parties. While Savimbi initially pursued brinkmanship as part of a strategy to force his way back into an electoral process he had already lost, his early success afforded him the opportunity to consider even more ambitious goals. With the government’s forces in disarray and UNITA on the offensive, his strategy changed from one seeking integration to one seeking domination. But it was not to be. In an interview in early 1993, President dos Santos acknowledged to the New York Times that his troops had not been prepared for the UNITA offensive in late 1992 and had been losing territory ever since. But he also asserted that, as soon as the military structure was reorganized and rearmed, “the military evolution will be favorable.”133 Indeed, while the government had lost the symbolically important city of Huambo, it had not yet been defeated. “If UNITA was going to win the war militarily, their chance was between October and April,” observed a Western military analyst. “They didn’t do it.”134 Moreover, the government had undertaken initiatives which would allow it to regain the upper hand politically and militarily. Several factors altered the balance of power back in favor of the MPLA. The first of these was the formal recognition by the United States of the MPLA as the legitimate government of Angola in May 1993.135 Washington’s belated recognition of the government and its condemnation of UNITA’s actions signaled that it was running out of patience with the party already determined by international opinion to have upset the peace process. Since Washington’s recognition meant that the government now had the right to defend itself, the way was also paved for the eventual removal of an international embargo on arms supplies to Angola, thus allowing the MPLA to begin a massive military rearmament program. On April 23, 1993, the government unilaterally declared that the “triple zero” provision of the Bicesse Accords, which banned military aid to either party, was now dead. Luanda called on “all countries with which the Angolan government has diplomatic relations of cooperation to help it equip its forces with material and logistical means.”136 By August 1993, all three members of the “Troika” that had pledged to observe the triple-zero provision of the Bicesse Accords

208

Civil War in African States

(Russia, Portugal, the United States), as well as other countries (such as the United Kingdom), had stated their intention to lift their bans on military aid to Angola. As the first reports of a new government military offensive against UNITA came in, Luanda was investing heavily in its armed forces, acquiring an estimated $1-2.5 billion in new military hardware. Finally, the government employed a South African private security firm, Executive Outcomes, to train and direct its war effort. The combination of these factors would be crucial to saving the government. The MPLA Assumes a Dominant Position … and Offers to Share Power

The Angolan government learned hard lessons from its experiences in Luanda at the end of 1992. The obvious and unfortunate message of the elections and the subsequent return to war was that it was to no benefit and at considerable risk that they followed their pre-election demobilization orders. As a consequence, Luanda remained determined not to let itself be caught again in such a weakened position. “The Government played largely by the rules, and lost big time,” noted a Western military analyst following the elections and return to war. “You can be assured they won’t make that mistake again.” The government knew that unless UNITA was facing military defeat, they would not come to the table. “If there is military equilibrium, which is what we have at the moment,” observed the Angolan Chief of Staff, General João de Matos, “there will not be a lasting agreement. . . . It is necessary to have a military imbalance to reach an agreement.” Such an outcome was not long in coming. Beginning in Lusaka, Zambia, in November 1993, and under the mediation of the UN’s new Special Representative, Alioune Blondin Beye, several sets of negotiations took place. On October 17, 1994, Beye stated that the UNITA and MPLA negotiators had finally reached an agreement. Fighting, however, continued as both sides maneuvered for military superiority before the two leaderships were to meet on November 20, 1994. Much to the chagrin of UNITA and the international community, which regarded the MPLA’s offensives as “needless” and as violations of the spirit of the peace process, the government recaptured Huambo prior to the formal signing of the Lusaka Agreement.137 A negotiated settlement was now critical for UNITA because it alone could save it from inevitable defeat. While UNITA continued to control portions of Angolan territory, it was clear that, given its isolation within the international community and its recent battlefield losses, only

The Angolan Civil War

209

a negotiated agreement could stop the government steamroller. UNITA also faced the prospect of UN sanctions, which would restrict international travel for its leaders and inhibit its ability to solicit external diplomatic support. Sanctions also threatened to further disrupt the trade in diamonds, an increasingly important source of funds for UNITA. By contrast, an agreement would avoid outright military defeat and allow UNITA to enter the post-Lusaka era with much of its military force still intact. Since it was clear that the international community supported an inclusive power-sharing arrangement, an agreement also provided UNITA with access to the corridors of power in Luanda and, more importantly, could allow it to legitimately tap into Angola’s diamond wealth. While the government had to rein in its own hardliners, who were tempted by the prospect of an outright military victory over UNITA, a negotiated agreement had certain benefits for it as well. Following a previous and unsuccessful round of peace talks in the Ivory Coast during 1993, the UN Security Council had applauded the government’s apparent willingness to search for a negotiated settlement, while castigating UNITA for continuing to pursue a military solution. Although the government’s receptiveness to the Ivory Coast agreement at that time had much to do with its fortunes on the battlefield, President dos Santos was careful not to further offend potential allies, even though, in military terms, it now had the upper hand.138 Having only recently secured recognition from Washington, the government believed that its compliance in the peace process, combined with UNITA’s status as a pariah, would serve it well over the longer term. As Paul Hare, the American special representative to the Angolan peace process, remarked, “Although these developments favored the government, they also made it difficult for Luanda to refuse to enter into new negotiations.”139 Lusaka and the End of Integration

There was little evidence, however, that either side was genuinely satisfied or even secure in an integrative arrangement, except insofar as it served a larger purpose. As impressive as the Lusaka Protocol was, it was still power-sharing with a balance of power that favored the government. The signing ceremony itself was tarnished by the government’s last-minute capture of Huambo and, partly as a consequence of the government’s actions, Savimbi’s refusal to attend the signing ceremony personally.140

210

Civil War in African States

Nonetheless, the Lusaka Protocol sought to avoid many of the pitfalls of the previous agreements. Most significantly, while UNITA would inevitably have a role that was secondary to the MPLA, the substance of the agreement was that the two parties were to be integrated in political, military, and economic terms. At the core of the Protocol, however, was another trade-off: UNITA would gain a share of political power and representation in the Angolan Armed Forces (FAA) in exchange for its own disarmament and transformation into a political party. UNITA would also be required to hand over territory under its control to the central government. These and other issues—Savimbi’s status in the new government, the inclusion of UNITA’s deputies in a Government of Unity and National Reconciliation (GURN), wealthsharing, disarmament, and territorial handover—deserve particular attention, since each of them proved to be contentious and to present its own challenges to implementation. The first of these issues was Savimbi’s post-Lusaka status. The Protocol stated that “as President of the largest opposition party, the President of UNITA shall be guaranteed a special status.”141 While the UN’s Special Representative to Angola, Blondin Beye, had gone to great lengths to spell out various provisions of the agreement in detail, the exact nature of this “special status” was left unresolved. In subsequent meetings between the two leaders over the next three years, a number of options were considered. Savimbi’s continued reluctance to come to Luanda, however, was attributed variously to his concerns about personal insecurity, his indecision about what role he should play, and to his lingering reluctance to give up the military option for a subservient political role under his arch-enemy. At one point, an agreement appeared to have been worked out, whereby Savimbi would accept one of two vice-presidential offices. But subsequent public statements by UNITA put Savimbi’s acceptance in doubt. In a 1996 speech on the thirtieth anniversary of UNITA’s founding, Savimbi himself asked, “Do you really need Savimbi as vice-president? Perhaps I will be more useful telling the truth I tell rather than going around gagged. I would also not like to die from a heart attack just because I was not received by the president’s office. I can be far more useful outside the cabinet than inside it.”142 Savimbi also suggested that he sensed a trap, claiming that his own cadres in Luanda were warning him that he would be ambushed upon arrival.143 MPLA officials claimed not to be surprised at Savimbi’s reluctance to take up the offered post in the capital. One MPLA official acknowledged that the government would not allow Savimbi to use the power-sharing agreement as a Trojan horse. “Of course, you can

The Angolan Civil War

211

understand that in Angola we are not politically naive,” stated Paulo Jorge. “How can we create one vice-president if by some impeachment of our President, Savimbi would automatically become President?” Nor was Savimbi expected to sign on if he were offered anything less powerful than the office of vice or co-president. “I said to my colleagues that I didn’t believe that Savimbi would accept,” noted Jorge, “because, in my opinion, Savimbi would not accept to be subordinate to anybody.”144 UNITA officials insisted, however, that Savimbi had indeed reconciled himself to the prospect of a power-sharing arrangement, “as long as it was not simply to cut ribbons.”145 Extensive negotiations had simply failed to settle on a satisfactory role for the UNITA leader. According to one UNITA official, “from the government’s point of view, the idea was simply, you come here, we give you the vicepresidency, we give you five Mercedes, you’re going to have a cook, you’re going to have a good salary, and just forget about everything else. From that perspective, it was difficult for someone like [Savimbi] to take. It is very hard to simply come out and to say now you are the vice-president but all you do is wake up in the morning, read the papers, and then look at your Mercedes.”146 Other observers claim that Savimbi’s “failure to grasp that [UNITA’s] long-term future lay in the political arena” undermined his own cause. It was, according to one source, another missed opportunity for the UNITA leader, insofar as “he was unable to exploit the discontent swelling in the cities” at this time, once again proving “his inability to adapt to civilian politics.”147 But other factors may have been involved in Savimbi’s decision not to come to Luanda. UNITA’s control of diamond mines in northern and central Angola in addition to his relatively austere lifestyle in the African bush had allowed Savimbi to sustain himself without reliance on government inducements. Unlike Afonso Dhlakama in Mozambique, who had received financial incentives to stay in the political process, Savimbi could not be bribed by cheques. Moreover, Savimbi’s conviction that coming to Luanda involved personal risk, and his belief that he could achieve more if he returned to the battlefield, meant that finding the right inducement would never be an easy task. In short, Savimbi perceived greater security and brighter prospects outside of Luanda. The newsletter Africa Confidential quipped that a political strategy was most attractive “except perhaps to someone who has spent the past 33 years of his life fighting in the Angolan bush in the absolute conviction that he will one day be President of his country.”148 Indeed, when so much pointed to an integrative strategy, Savimbi continued to opt for the military one.

212

Civil War in African States

Not only did Savimbi find UNITA’s proposed status as a junior partner in a power-sharing agreement incommensurate with what had been sacrificed in the war, but he believed that nothing less than parity could ensure UNITA’s survival. UNITA Secretary General Eugenio Manuvakola said that Savimbi told him the best way to ensure UNITA’s survival “is to go back to the war because the peaceful solution will not give life to UNITA.” UNITA, he said, “could die by respecting the agreements.”149 If parity could not be achieved, UNITA officials contended, the choice was either to submit to MPLA domination or return to war and keep the movement alive. A second issue, the inclusion of UNITA deputies, was more successful. Despite Savimbi’s reluctance to come to Luanda, provisions were made in the Protocol for members of UNITA “to participate adequately at all levels and in the various institutions of political, administrative, and economic activity.”150 Here too, however, there were challenges. UNITA had initially demanded a power-sharing arrangement whereby the government and UNITA would operate on the basis of parity and where there would be an equal division of spoils. Other UNITA officials, citing the memory of those killed in Luanda following the 1992 elections, raised concerns about the security risks of having senior UNITA officials in Luanda. “We’ve been lured to Luanda twice before, in 1975 and 1992, and our men were killed both times,” noted one UNITA cadre. “We won’t make that mistake again.”151 But others within UNITA were less adamantly against coming to Luanda. While Savimbi preferred to remain in the bush rather than accept a second vice-presidency, his MPs were being promised all the comforts of urban life in Luanda, an offer that for many compared favorably with the uncertainties of life in the bush. In April 1997, seventy UNITA deputies arrived in Angola’s capital to participate in the newly established GURN. As one diplomat observed soon after, “The GURN is in place and is functioning. The parliament sits and there is free debate. It functions like any good parliament should with a good opposition. That is what provides some hope in all of this.”152 David Wimhurst, the UN’s spokesman in Luanda added, “They have very open and free debates which are televised—and they are calling each other all sorts of names, which is great. They are lancing the boil. A war of words rather than a war of bullets, that’s what we say.”153 Even UNITA’s own ministers welcomed the GURN: “It is very important as far as the country is concerned because it is a real symbol of unity,” observed the new Minister of Mines, Marcus Samondo. “Obviously it starts from the top level and needs to go to the bottom level. But we work as a team.

The Angolan Civil War

213

Once you are here, you are not invited as a Minister from UNITA. You are a Minister of the Angolan people.”154 A third element of the Protocol, the sharing of Angola’s diamond wealth, was formally outside the Lusaka Protocol. Yet it was critical to the Protocol’s success. The U.S. encouraged a “peace for diamonds” deal on the assumption that UNITA would be unlikely to hand over diamond territory under its control, as required by the Lusaka Protocol, unless UNITA benefitted from future diamond revenues.155 Control of the diamond mines had been, according to Savimbi, “a question of survival.” In 1992, he argued, “the Government was supposed to finance our campaign. It did not do so. We therefore had to secure our own means of finance because, one day, there will be lots of new elections, and we must finance the campaign.”156 Until the end of the 1980s, UNITA had relied largely on external patronage to fund its war effort; but as that evaporated they had to secure their financing elsewhere. Even as the peace process that led to the Bicesse Accords had gotten under way, UNITA was infiltrating diamond territories in the north. In the run-up to the election, UNITA had not been required to hand-over diamond territories under its control and, indeed, following the electoral loss, they were further consolidated militarily by UNITA. Impatient with the slow progress of extending central administration to areas under UNITA control, in 1995 Luanda made a surprise statement that it was about to clear out those engaged in illegal mining. The announcement alarmed UNITA and the peace accords suddenly appeared in jeopardy, until the government backed down and sought another approach.157 “There were all sorts of initiatives taken to try and resolve the problem of the diamonds,” noted UN spokesman David Wimhurst. “Luanda recognized that the only way to get UNITA to agree to the return of territorial administration to central authority was to do a deal on the diamonds.” In this way, another exchange of sorts was worked out, whereby UNITA pledged to allow government control to be extended throughout the diamond territories in return for its right to legally exploit the diamond resources for its own benefit. “You legalize it essentially,” observed Wimhurst. “It’s a pragmatic solution to a thorny problem and it has northing to do with the Protocol. But at the same time we recognized that if they can’t resolve the diamond thing, we won’t get much further.”158 The diamond issue, however, proved to be just as nettlesome as the others. Wealth-sharing arrangements ran into difficulties when UNITAperpetrated attacks continued in the diamond regions, forcing the government to withdraw its offer of a percentage of diamond revenues. For its part, UNITA claimed that the territorial concessions granted by

214

Civil War in African States

the government were of little value, since they had no proven diamond reserves.159 By the end of 1997, the “peace for diamonds” proposal had come unraveled. The fourth issue was disarmament. UNITA’s ability to come within a hair’s breadth of taking the country, including Luanda, led the MPLA to vow never to let itself be so unprepared again. UNITA “probably has enough men in arms to cause damage, disruption, and so on, but it can’t win the day,” noted David Wimhurst. “Essentially we are trying to close off all the military options and leave open the political option, the democratic option. And that’s the door through which UNITA has to go.”160 UN officials stated that sanctions against UNITA were not undertaken in an effort to “punish” UNITA, but rather to encourage compliance on their disarmament obligations.161 Militant members of the MPLA continued to emphasize the need for vigilance against a possible UNITA offensive in the future and to ensure UNITA’s compliance in the peace process. “UNITA only understands one language—force, hard line,” noted one such MPLA militant. “When we took these hard positions, UNITA became more flexible.”162 But efforts to control UNITA’s behavior and hardline rhetoric did not alleviate Savimbi’s sense of insecurity. While the international community and the government claimed that they were only pushing the insurgents to meet their obligations under Lusaka, UNITA claimed that the government was preparing to destroy them. “I am facing the most critical situation in UNITA’s history,” Savimbi had told delegates at the March 1996 thirtieth anniversary of UNITA’s founding. “To confine soldiers is not a problem, but disarmament is. . . . You must understand how difficult and critical this is for me. I would prefer to surrender weapons and die afterward; my life will become meaningless.” Even if he were to accept the vice-presidency, Savimbi did not believe that his post had any significance unless the office could guarantee UNITA’s continued capacity to coerce. “Once weapons are surrendered, I will not be suitable for the vice-presidency,” he maintained. “I have not known of any historic leader who disarmed his forces and stayed in power.”163 The ongoing tension, frequently characterized by Angolans at that time as nem guerra, nem paz (neither war nor peace), led UNITA to believe it had little choice but to maintain weapons, hold out, and threaten to go back to the bush. Yet an armed UNITA was, for obvious reasons, intolerable for the government. As Robert Fowler, head of the UN sanctions committee on Angola, stated before the Security Council, UNITA was using “the period of partial implementation of the Lusaka peace accord to replenish their armaments, to procure and store large quantities of petroleum and to otherwise prepare for war.”164 Yet, for at

The Angolan Civil War

215

least one senior American official, until 1998, as precarious as it was, the peace process was never over.165 The U.S. Special Representative, Paul Hare notes in his memoirs, “The real question focused on intentions and capabilities. Was UNITA’s purpose to retain a defensive capability only until the overall political and military situation became clearer? Or did UNITA intend to keep an offensive military capability in order to strike at the government? No one knew the answer to these questions. . . . So much depended on the calculations and decisions of one man: Jonas Savimbi.”166 The final and perhaps most contentious issue (one obviously related to the diamond issue), however, was UNITA’s obligation to hand over to central administration territory under its control. As for “why the agreement broke down,” remarked UNITA’s Jardo Muekalia, “it is something which of course depends on which side you are on.” As far as UNITA was concerned, the main problem was “when it started turning over territory to the government,” observed Muekalia. “The government would come into UNITA area as a conquering army. UNITA raised those aspects to the UN, the Troika, and its basic point was that if the extension of state authority means the end of UNITA, then we are not interested in this agreement.”167 Similarly, David Wimhurst, the UN’s spokesman in Angola, observed: The whole process of giving back territory was fraught. We had situations where there would be a ceremony, the government would come in, they would take over the local administration .... They would put up the Angolan flag, the national police force would come in and everyone would applaud. We’d go home and then UNITA would start to resist. . . . They would provoke incidents, and the police—being brutal and short-witted—would go in and beat up local people, kill them. And UNITA would say, ‘You see, every time we extend the state administration, all they do is come in and harass and hunt us down and beat us to death.’ They provoke it themselves. But the Angolan national police play right into the hands of UNITA.168

An additional complication for UNITA was the perceived partisan response of the international community. World opinion had grown impatient with Savimbi’s reluctance to meet his obligations under the Lusaka Protocol and thus was unsympathetic to UNITA charges that the MPLA was using unnecessary force. The UN and Washington pledged to raise UNITA’s concerns with the MPLA, but only after UNITA had handed back the remaining four towns they held to central administration. According to Muekalia, “The international community said ‘You have to go all the way, because then we cannot trust your

216

Civil War in African States

good faith. And after you have finished we can talk to the government.’ I think there this agreement came to a point where UNITA did stop, and the international community kept adding more sanctions, hoping that, with sanctions, UNITA would move. UNITA did not move and the sanctions continued. And we came to a point where everybody was backed into a corner.”169 Muekalia later added, “The government portrayed it as unwillingness on the part of Savimbi; we saw it as a death sentence.”170 But UNITA’s protestations continued to fall on deaf ears. Even among UNITA’s traditional allies, its excuses for not handing over land under its control, for not disarming, and for the failure of the rest of UNITA’s deputies to come to Luanda, were wearing thin. Some American officials had come to believe that UNITA was merely using the peace process to rearm itself and strike again. “For UNITA,” Herman Cohen wrote in his memoirs, “peace negotiations were in effect another form of warfare, not an attempt to achieve compromise, consensus, and reconciliation. The fact that two parties to a conflict engage in a peace process does not necessarily mean that one or both are serious about peace. We learned that the hard way in Angola. Savimbi was the master of delay and deceit.”171 Power-Sharing to Divide UNITA and Isolate Savimbi

As South African and American patronage waned in the late 1980s, UNITA had migrated beyond its Jamba enclave in south-east Angola into the diamond regions of north-eastern Angola. It was here that UNITA would continue to acquire the diamond wealth that, the government feared, could sustain it throughout the 1990s. This march northward had several implications. First, without an external patron to be accountable to, UNITA had even less need to maintain a democratic façade. On the contrary, as Assis Malaquias has pointed out, UNITA’s strategy had shifted from a Maoist strategy of winning the “hearts and minds” of the local peasantry, to a predatorial one, which sought to sow fear in order that Angola’s diamonds could be readily exploited.172 As diamond territories were depopulated, UNITA’s constituency among the Ovimbundu was correspondingly lessened. Secondly and more importantly, the struggle between the government and UNITA over diamond territory became zero-sum in nature. During the Cold War, UNITA received military support regardless of whether it controlled territory. Now, diamond territory was UNITA’s lifeblood; in order to contain UNITA, the government needed to deny access to this crucial source of funding. On the other hand, for UNITA to give it up was

The Angolan Civil War

217

necessarily to undercut its own means of survival. Yet this was the very action required for compliance with the Lusaka Protocol. From the government’s perspective, UNITA intransigence on the handover of territory to central administration was only the latest indication of Savimbi’s noncompliance and, therefore, noncooperation. UNITA’s rejection of the 1992 elections had shown that its intentions had always been domination and that it could never accept a decision made by Angolans in a free and fair election unless UNITA was the winner.173 UNITA’s failure to disarm clearly suggested that its intentions were not peaceful and, on the contrary, that it was now using the peace process as an opportunity to regain its military strength. “UNITA does not just have a plan for peace. Its leadership has alternate plans,” stated Eugenio Manuvakola upon his defection from UNITA. “There are plans for the peace process but they want to take advantage of this interim period, which they want to make as protracted as possible, to prepare their military force which was downsized and degraded.”174 It is also not clear that the government did in fact seek to induce Savimbi to come and take his rightful place in a governing coalition. While the government claimed to have opened a political space for Savimbi, hardliners continued to see his containment—or preferably his removal—as the only options.175 “When UNITA proclaims its desire for peace, it is trying to sell a vision to world opinion which does not correspond with reality,” said the Angolan Chief of Staff, General de Matos, in a 1995 interview with Le Monde. He was clear about what needed to be done: “Only the absolute military defeat of Savimbi will assure peace.”176 Ironically, however, the existing power-sharing agreement was a means to this end. Inclusion proved to be a way to isolate Savimbi, insofar as it played upon the belief among some UNITA cadres that their war, fought for a legitimate cause, was now being hijacked by an increasingly totalitarian Savimbi. Having achieved UNITA’s stated objective of an election, many cadres were willing to come to Luanda, where, it was promised, they would be “welcomed as brothers” and enjoy a “dignified life.”177 “Multiparty elections were what we had fought for and now [Savimbi] was destroying all that,” said General Geraldo Nunda, a senior UNITA commander, after his defection to the government.178 Nunda also recognized the not-too-subtle shift in the balance of power that was the inevitable result of UNITA’s financial lifeline of diamond resources, the trade in which was increasingly being choked off by international sanctions. The government, by contrast, was able to fund its massive rearmament by auctioning off drilling licences

218

Civil War in African States

for new blocs in Angola’s oil-rich deep waters—a resource which made the government’s pockets considerably deeper than UNITA’s. Recalling a conversation that took place prior to his defection, Nunda claimed that Savimbi told him that he could have “fifty million dollars in a year.” According to Nunda, he responded by telling Savimbi that “the government can have fifty million dollars from oil in a month.” “UNITA can’t win the war against the government,” Nunda told Savimbi. “War is suicide.”179 Savimbi’s efforts to maintain control of his own movement also undermined his cause. His seemingly brutal willingness to eliminate dissidents or perceived challengers of his authority proved selfdefeating. Not only did Savimbi’s actions suggest that he could no longer be a viable partner in a power-sharing arrangement but they instilled fear among his own cadres that even they could face death if they stepped out of line. The opportunity to participate in a government of national reconciliation, by contrast, provided a useful and legitimate escape hatch for those wanting to evade Savimbi’s wrath. Eugenio Manuvakola was one of several high-profile UNITA cadres who defected in the late 1990s.180 Having been under house arrest since February 1995, Manuvakola escaped with his family from the UNITA redoubt of Bailundo and traveled secretly to Luanda. Once he was there, the government used this opportunity to showcase not only Savimbi’s crumbling support among his most senior cadres, but also that Luanda was a secure place for those willing to engage in peace. The MPLA Closes in on UNITA

Luanda’s next step was to engineer a new UNITA, which would accommodate those opposition members who had already formally broken with Savimbi and force the hand of those who had not. On August 31, 1998, one year after Manuvakola’s escape, the government suspended all UNITA members who had yet to renounce their links with Savimbi and removed all perquisites of political office. Several days later, the creation of UNITA-Renovada was announced, under the leadership of Manuvakola, the Hotels and Tourism Minister Jorge Valentim, and three other UNITA dissidents. The party was launched under the guidance of one of the MPLA’s leading figures in the peace process, Higino Carneiro, and the government had already secured the support of the four UNITA ministers and vice-ministers. Hoping to draw in the majority of the remaining UNITA members of parliament, the government publicly recognized the new party and reportedly offered $3 million to each of the dissidents.181 Not all played the government’s

The Angolan Civil War

219

game. Predictably, UNITA-Renovada was accused of being too close to the government and of cheapening the genuine grievances that UNITA members sought to advance. But even those who did not formally join Renovoda maintained their distance from Savimbi and pledged that they would not challenge the leadership of the new party.182 “I have had disputes with Andulo [Savimbi’s home town],” remarked Abel Chivukuvuku, a prominent UNITA parliamentarian who nonetheless refused to join Renovada. “From now on I am a free man, a citizen of peace.”183 The government’s strategy to isolate Savimbi worked on a regional level as well. The government was well aware that Savimbi’s troops in the field were being supplied with military reinforcements from Angola’s neighbors. In mid-January 1998, an Angolan MiG-21 had forced down a DC-4 aircraft, and the crew of nine South Africans was arrested. The captain, Peter Bietzke, claimed to have flown more than three hundred missions into Angola, dropping weapons supplies at Bailundo and Andulo in central Angola. A subsequent report commissioned by the UN in consultation with UNITA dissidents claimed that, throughout the 1990s, in exchange for packets of Angolan diamonds, regional sympathizers in South Africa, Zaire, Togo, Congo (Brazzaville) and Burkina Faso had allowed their countries to serve as sources or transit points for UNITA-bound weaponry.184 According to one of the dissidents interviewed for the report, UNITA often received fifteen to twenty planeloads of supplies each day, in violation of UN sanctions. Determined that it would not suffer the same fate that it had following the 1992 elections, the government undertook its own regional initiatives to contain its adversary. Angola provided military support to Laurent Kabila’s struggle to overthrow Mobutu Sese Seko in neighboring Zaire (later to be renamed the Democratic Republic of the Congo or DRC). Mobutu had long been a staunch supporter of UNITA, and his overthrow in May 1997 meant an end to Kinshasa’s assistance to UNITA. Luanda kept a force of some seven thousand troops in the DRC to sustain the new regime and to protect their long common border, but when Kabila no longer served their interests, Luanda was a likely culprit behind his assassination.185 The Angolan government was also instrumental in the 1997 overthrow of President Pascal Lissouba in Congo Brazzaville, where UNITA had some of their last external bases. Finally, amid allegations that UNITA was funneling weapons in and diamonds out across the Angolan border with Zambia, the Angolan government mounted pressure on the regime of Frederick Chiluba to cooperate in enforcing UN sanctions on UNITA. This careful realignment

220

Civil War in African States

of the regional context had important balance of power benefits for Luanda: at a September 1998 summit of the Southern African Development Community (SADC) in Mauritius, Savimbi was formally branded as “the person solely responsible for the increasing deterioration of the security situation in Angola,” and as a “war criminal” who was “objectively incapable of leading his party onto the road of peace in Angola.”186 The situation had changed, from the Cold War years when Angola was surrounded by countries whose leaders allowed their territory to be used as conduits or bases for UNITA rebels, to one in which virtually all of Angola’s neighbors were led by people installed by or friendly to Luanda. The balance of power had shifted in favor of the MPLA government and Luanda used the opportunity to press its domination of UNITA.187 The Lusaka Protocol itself became a useful template by which UNITA’s behavior could be measured and from which support for its final offensive could be mobilized. By demonstrating UNITA’s failure to live up to its obligations under the Protocol, the government was better placed to persuade the international community that war was the only option. A final set of sanctions went into effect in June 1998 when it became clear that UNITA was not going to comply with either of the agreement’s disarmament or territorial commitments. After four years of UNITA’s obstruction of the Protocol’s implementation, the patience of the government and the international community had worn thin. As one diplomat in Angola observed, “the sanctions are clearly a message to UNITA. But I think that the hidden message here to the government—which isn’t so hidden—is that, if the next stage of the sanctions fail, the international community would have no objection to the government reasserting their authority in whatever way they feel appropriate.”188 Indeed, on June 21, 1998, Information Secretary João Lourenço, the head of the MPLA’s parliamentary caucus, drafted a resolution giving the government carte blanche to restart the war. The MPLA Realizes Its Strategy of Domination

In September 1998, the Troika sent a letter to Savimbi pointing out that UNITA had neither handed over the four remaining towns under its control to the government nor met its demilitarization obligations. The Angolan government, on the other hand, “has fully honoured its undertakings under the Lusaka Protocol.” The letter concluded, “This is UNITA’s last chance to secure a legitimate and constructive place in Angola’s political context. Should it fail to respond with irreversible steps, such as extending the central government’s administration to the

The Angolan Civil War

221

whole of Angola and full demobilisation, we shall be obliged to conclude that it does not want peace.”189 The letter was interpreted as an implicit acknowledgment that, should this last warning fail to elicit the appropriate response, the government could proceed with a military offensive against UNITA. Citing a source in Washington, Lynne Duke of the Washington Post wrote that, “In an earlier time, the government’s military actions would have been roundly condemned in foreign capitals. Now, however, ‘there’s been no international criticism of this because this is a sovereign government . . . [which] has the right to defend its population from armed forces as well as prevent them from taking territory.’”190 At the same time as it was making it harder and more expensive for Savimbi to acquire new weapons, the Luanda government continued to invest heavily in its own arsenal. In June 1998, dos Santos traveled to Russia, reportedly prepared to spend tens of millions of dollars on new military equipment. UNITA officials claimed to be dumbfounded. “Our question was always, if there is a peace agreement—this is a peace agreement—why are they so busy purchasing weapons?” said UNITA’s representative, Jardo Muekalia, in Washington. “If you keep buying, from my perspective, you are not protecting yourself against some threat out there. You are preparing to do something internally.”191 Indeed the government was. Until 1998, the objective of the government’s efforts had ostensibly been to minimize the threat that Savimbi presented and, if possible, to force him to comply with his obligations under the Lusaka Protocol. The result was, of course, the opposite: Savimbi’s own sense of insecurity was increased and the unlikelihood that he would fulfill Lusaka’s terms or come to Luanda grew even greater. The final coup de grâce was the government’s military offensive, launched in December 1998. However, while the government’s rhetoric was bombastic, it too continued to fear the threat that Savimbi posed. “Everyone was afraid of him,” remarked Paulo Jorge, Angola’s former foreign minister. “I used to say that today our problem is to eliminate the cause of this conflict and not the effects of the conflict. The principal cause was Savimbi.”192 Indeed, the conflict between the MPLA and UNITA had become intensely personal; ending the war could only mean eliminating Savimbi himself. In the years prior to the 1998 offensive, UNITA had walked a careful line, preferring to avoid a direct confrontation with the government. But once the fighting was under way, the government suffered battlefield reversals, indicating that UNITA was indeed prepared to do battle. Once again UNITA sought to press its claim that it was only holding out for a fairer negotiated agreement. “We did not start

222

Civil War in African States

this offensive, therefore it is not up to us to stop it,” noted UNITA Secretary-General, Paulo Gato. “If the MPLA stops its offensive, the UNITA leadership might consider the possibility of returning to the negotiating table.”193 But the government pressed on. In a November 1999 interview on Angolan state television, General Joao de Matos claimed that they knew of Savimbi’s whereabouts and would continue to “thrash” UNITA “until we capture him or annihilate him.”194 President dos Santos continued to offer a peaceful resolution of the conflict—provided Savimbi “says when he is going to stop the war and how he intends to conclude the implementation of the Lusaka Protocol and prove it by concrete steps.” Dos Santos added, however, that “the radical posture by Mr. Savimbi permits me to conclude that political compromise was never part of his agenda.”195 As the government offensive proceeded, the Angolan government continued to receive moral support from an international community now sympathetic to its cause. “It is essential that Savimbi is isolated and even removed from UNITA,” said the British Foreign Office Minister, Peter Hain. “He is the cause of the war more than anybody else. . . . Savimbi needs to be taken out of the picture.”196 Savimbi’s end did not come until February 22, 2002.197 With the assistance of American and Brazilian intelligence, Israeli communications specialists, and defectors from UNITA’s own ranks, Savimbi’s movements within the eastern province of Moxico were traced until government troops caught up with him and gunned him down in a shootout. As expected, the death of the rebel leader precipitated the collapse of UNITA as a military movement. The remaining rebel holdouts signed a new peace agreement in Luena two months later, on April 4. The aftermath of the war in Angola has indisputably been a more peaceful country. While there were enormous logistical challenges associated with reining in and accommodating UNITA rebels still in the field, the government for the most part received approval for its management of the post-war transition. For its part, without its leader, UNITA proved to be remarkably cooperative. Its congress, held in June 2003, led to the election of Isaías Samakuva as UNITA’s new leader and the reincorporation of UNITA’s various post-Lusaka factions into the movement. While the government was slow to commit to a specific date for new elections, Angolans voted in parliamentary elections in September 2008. The outcome of the election was never in doubt. Samakuva himself had long predicted the MPLA would win handily. “Is the MPLA going to win?” he asked in a 2004 interview. “I say yes. I don’t see

The Angolan Civil War

223

another party that can do it.” He claimed only that the “fraud will be bigger and more sophisticated.”198 Angola’s electoral commission rejected UNITA’s claims that the MPLA was using its massive state revenues to fund its electoral campaign and line its pockets, saying that they were without foundation. For its part, the MPLA officials claimed that the victory was rightfully theirs. “The people recognize what the MPLA is doing for them,” noted Rui Luis Falco Pinto de Andrade. “We brought the country peace. We stabilized the economy and we are rebuilding the country.”199 Indeed, the government had finally brought stability to Angola through domination of its principal adversary and the killing of Jonas Savimbi. The government’s largest irritation is no longer UNITA but a collection of outspoken and prolific journalists and critics who have continued to demand greater accountability from the government. But even they acknowledge their limitations in a state effectively dominated by one political party. “We are powerless,” observed Fernando de Macedo, president of the Association for Justice, Peace and Democracy. “By political means you can’t change the situation. . . . They have too much at stake to let themselves lose.”200 Conclusions

In the early post-independence years, both the MPLA and UNITA remained profoundly suspicious that each sought to eliminate the other. Each side claimed, as a result, that, in order to protect and sustain themselves, they had to maintain links with outside patrons. UNITA’s rhetoric spoke of a desire to have access to the political system and of a willingness to share power; their actions spoke more of a desire to dominate the MPLA and remove them from power. Unlike UNITA, the MPLA at that time made no pretensions about the possibility of powersharing, since it maintained that UNITA was merely an instrument of South African power and did not have a meaningful constituency of its own. There was no need to be inclusive of it—only a need to eliminate it. With the end of the Cold War came a new need to consider other options to break the Angolan stalemate. It was a new era in African politics and conflict management, and the promise of a new world order led many to believe that long-standing violent conflicts could be brought to an end—as they already had in neighboring Namibia—through democratic elections or power-sharing arrangements. International factors, it seemed, played a role in bringing parties to a peaceful resolution of conflict and, indeed, Angola initially followed this pattern

224

Civil War in African States

as both the MPLA and UNITA seemed willing to pursue nonviolent approaches to politics. It is unlikely, however, that either party saw the Bicesse Accords and the elections that followed as part of an integrative strategy. Both parties expected that the elections would require only a minimum amount of cooperation—and likely only when it came to the demobilization of their adversary. The elections were attractive precisely because, over the longer term, they allowed one party to dominate the political process. Even better, after years of having to defend its bellicosity, participation in an internationally sanctioned election would allow the winner to be recognized as the legitimate authority in Angola. In short, domination, which each party was assured would be theirs, could be achieved with virtually no concessions and would receive the blessing of the international community. It is perhaps surprising, then, that the Lusaka Protocol—an agreement that promised to bring both parties together in a powersharing agreement—would come after the failure of Bicesse, the elections of 1992, and Angola’s return to war. But while both parties did sign the agreement—an apparent shift from a strategy of domination to one of integration—neither party truly committed itself to its consummation. Jonas Savimbi did not sign the agreement himself, sent his subordinate in his place, and he never came to Luanda to assume his position in the Angolan government. Savimbi was clearly not won over by the benefits of inclusiveness, except insofar as the peace provided by the Lusaka Protocol could be exploited as a time-buying means to rebuild his military strength. This, of course, is hardly an approbation of inclusiveness on its own merits. Nor was the MPLA entirely convinced by the merits of inclusiveness, although this determination is more difficult given that their more passive role in this process required them only to make political positions available to compliant UNITA members. Savimbi’s refusal to take up his post relieved the MPLA of having to accommodate the UNITA leader in any way. One of the main sticking points, however, was that if integration was to be meaningful, Savimbi wanted to assume a position with real power. This was an accommodation that the MPLA was clearly not prepared to allow. Rather, the MPLA continued to seek to neutralize UNITA as a military force by using power-sharing to isolate Savimbi from his own movement. Its heavy-handed approach to the return of UNITA-held territory to central administration did nothing to instill confidence within UNITA’s senior-most leadership that the government’s commitment to power-sharing was sincere. If UNITA’s threat as a military force could have been reduced, there is reason to

The Angolan Civil War

225

believe that power-sharing would have become a reality. But at the cost of his military power, even an inclusive agreement was too much for Savimbi to bear. For him, political office was useless without the military capacity or the financial resources to back it up. No longer having the support of a superpower or even a regional patron to defend him, the UNITA leader responded to the increasingly unfavorable balance of power by remaining out of reach. Savimbi preferred the relative safety of the bush to political impotence or potential death in Luanda. Do integrative solutions ever serve the purpose of reconciliation? And how has the MPLA’s strategy of domination served ordinary Angolans? Initially, when asked to give an explanation for the breakdown of the Bicesse peace agreement, the UN Secretary General’s Special Representative, Margaret Anstee, stated that the winner-take-all approach of Bicesse “left no consolation prize for the loser” of the 1992 elections. Power-sharing, she said, would have allowed for a more gentle introduction to the rigors of competitive democracy and would have lessened the political and economic stakes that were at the center of the return to violence.201 This is indeed a compelling argument for this kind of arrangement. But political leaders may not see things this way. As Savimbi himself remarked, there was something condescending about Western leaders proposing integrative solutions, because they imply that peoples in developing countries are incapable of managing the acrimony of democratic politics. Anstee herself was lectured by the Namibian prime minister to the effect that power-sharing was a “second-class” form of democracy which was being imposed on Africa by the international community. “In the UK one party wins and governs, the others lose and don’t and that is the way it should be,” he raged. “That is the way it is here.”202 This is, of course, the fundamental problem: peaceful integrative solutions appear to be the solution, but they are difficult to implement in the aftermath of civil wars and, when they are, they are rarely utilized for the purposes of reconciliation. On the other hand, integrative solutions were acceptable to UNITA’s subalterns. If Savimbi believed that he had no choice but to submit to MPLA domination or continue the war, his subordinates faced a different calculation. For many of them, continuing with Savimbi’s destructive path meant certain death. After years of struggle in the bush, some were willing to engage in political discourse in Luanda. As became increasingly evident, the government was willing to welcome and accommodate UNITA dissidents in Luanda—provided they were

226

Civil War in African States

willing to break publicly with Savimbi. Their ability to challenge the MPLA in any meaningful way remained, however, limited. Angolans remain ambivalent about Savimbi’s strategy and the war in general. A favorable balance of power, combined with the memory of their near death experience following the 1992 elections, led the MPLA to return to a domination strategy which persists to this day, much to the regret of some Angolans. For all of the war’s terrible destruction, many Angolans continue to praise Savimbi for having had the courage to challenge the MPLA monopoly on power. From this perspective, it was not the end of the Cold War which brought about the 1992 elections; it was Savimbi’s willingness to bear the costs of war. In the end, however, despite having brought democracy by forcing elections, Savimbi did not win them and would not accept the outcome. Savimbi’s mistake was not in signing the Bicesse Accords but in failing to abide by the results. In Angola no one disputes that the 2002 killing of Jonas Savimbi brought the war to a welcome conclusion after some twenty-seven years of post-independence struggle. Yet, while many Angolans are relieved to see Savimbi gone, his death leaves behind a vacuum and a profound sense of vulnerability among many Ovimbundu. Savimbi allowed his personal ambition to take precedence over the interests of those on whose behalf he claimed to be fighting. And they now resent their deceased leader for failing to take advantage of opportunities for peaceful engagement. They know that his death leaves them under the dominant authority of a party which, they fear, demonstrates more arrogance than magnanimity. One wonders if in time a new force will emerge to challenge the MPLA and if it will once again be a violent challenge. In this sense, at least in the short term, conflict does not end, and it is not clear that one can say there is conflict resolution—only varying degrees of security and insecurity. 1. Assis Malaquias, “Ethnicity and Conflict in Angola: Prospects for Reconciliation,” in Jakkie Cilliers and Christian Dietrich, Angola’s War Economy: The Role of Oil and Diamonds (Pretoria: Institute for Security Studies, 2000), p. 106. 2. Isaías Samakuva, Interview with the author, July 14, 2004. 3. Henry Giniger, “For Independent Angola, A Great Threat of Strife,” New York Times, January 16, 1975, p. 12. 4. Prior to the Lisbon coup the MPLA had an estimated 6000 troops, the FNLA had 15,000, and UNITA had 1000. Africa Contemporary Record, 19757 6 , p. B424. In contrast, according to John Stockwell, CIA estimates of respective troop strengths at independence were as follows: MPLA 20,000; FNLA, 15,000; UNITA, 4,000. See John Stockwell, In Search of Enemies: A CIA Story (New York: W.W. Norton, 1978), pp. 91 and 155.

The Angolan Civil War

227

5. “Where is the Legality in the Civil War in Angola?” Declaration of the Political Bureau of the Central Committee of Angola, in Africa Contemporary Record, 1975/76, p. C90. 6. “Where is the Legality in the Civil War in Angola?”Africa Contemporary Record, 1975/76, pp. C90-91. 7. Cited in Africa Contemporary Record, 1975-76, p. B427. 8. Cited in Africa Contemporary Record, 1975-76, p. B427. 9. Cited in Africa Contemporary Record, 1975-76, p. B427. 10. Nathaniel Davis, “The Angolan Decision of 1975: A Personal Memoir,” Foreign Affairs, vol. 57, no. 1 (1978), pp. 123-124. 11. Kissinger, Years of Renewal (New York: Simon and Schuster, 1999), p. 810. 12. Stockwell notes that “we wanted no ‘soft’ allies in our war against the MPLA.” See Stockwell, In Search of Enemies, p. 193. 13. Stockwell, In Search of Enemies, p. 193. 14. Thomas Turner, “Angola’s Role in the Congo War,” in John F. Clark ed. The African Stakes of the Congo War,” (New York: Palgrave-Macmillan, 2002), p. 79. 15. Africa Contemporary Record, 1975-76, B430. See also Howard W. French, “From Old Files, A New Story Of U.S. Role in Angola War,” New York Times, March 31, 2002, p. 4. 16. In an “open letter to the free peoples of the world,” Jonas Savimbi claimed that UNITA had the “support of more than two-thirds of the Angolan population.” The MPLA, he said, “came to power in Angola through the Soviet tanks and not through peoples’ choice. . . . The M.P.L.A. Government in Luanda is very much a minority Government that will have to give way no matter how long our struggle will take, no matter how hard our resistance will be.” See “A Refusal to Become ‘Black Russians’ or ‘African Cubans,” New York Times, December 8, 1976, p. A27. 17. In a statement, UNITA claimed that “On 11 November 1965 a minority of white colonials seized power in Zimbabwe against the will of the black majority. Africa condemned this and it is fighting against it. On 11 November 1975, ten years later, if a handful of men who are thirsty for power should legally seize power in Angola Africa must show this same courage in reestablishing political legality and national reconciliation. At the side of legality is the truth. At the movement they are weak but they will end in triumphing.” See Declaration of the Political Bureau of the Central Committee of Unita, Nova Lisboa, 15 October 1975, cited in Africa Contemporary Record, 1975-76, pp. C89-C91. 18. Jeremias K. Chitunda, “Angola: Blueprint for Peace,” (Letter to the editor, dated June 9, 1976) New York Times, June 22, 1976, p. 34. In subsequent years, Chitunda claimed that elections were what UNITA was calling for. “If a free election were held today,” Chitunda observed in 1985,” Jonas Savimbi would win a massive landslide.” Cited in Associated Press, “Guerrilla Chief Predicts Victory Over Angola,” Washington Post, January 2, 1985, p. A24. 19. Jonas Savimbi, “A Refusal to Become ‘Black Russians’ or ‘African Cubans,” New York Times, December 8, 1976, p. A27. 20. “Statement by Dr Agostinho Neto on 1 October 1975,” in Africa Contemporary Record, 1975/76, p. C88.

228

Civil War in African States

21. “Statement by Dr Agostinho Neto on 1 October 1975,” in Africa Contemporary Record, 1975/76, p. C88. 22. “New Angola Move By Cuba Reported,” New York Times, June 4, 1976, p. A7; “Dr Agostinho Neto’s Independence Day Speech,” November 10, 1975 in Africa Contemporary Record, 1975/76, p. C95. 23. “MPLA Emerges Victorious, Recognition by MPLA” Africa Diary, March 11-17, 1976. 24. Africa Contemporary Record, 1976/77, p. B448. 25. By the mid-1980s, South Africa was providing an estimated $200 million in financial support. South African military units also provided assistance for UNITA from bases in northern Namibia (which was under South African control). See Jakkie Potgieter, “‘Taking Aid from the Devil Himself: UNITA Support Structures,” in Jakkie Cilliers and Christian Dietrich, Angola’s War Economy: The Role of Oil and Diamonds (Pretoria: Institute for Security Studies, 2000), p. 260. 26. Colin Legum alleges, however, that in the early 1980s South African strategists did consider the creation of a separate Ovimbundu state which would encompass either south Angola, or northern areas of Namibia and southern Angola. Colin Legum, The Battlefronts of Southern Africa (New York: Africana Publishing Company, 1988), p. 245. 27. Alan Cowell, Killing the Wizards: Wars of Power and Freedom from Zaire to South Africa (New York: Simon & Schuster, 1992), pp. 106-107. See also Fred Bridgland’s account of Jamba in Savimbi: The Key to Africa Edinburgh: Mainstream, 1986), pp. 319-325. 28. “Guerrilla Chief Predicts Victory Over Angola,” Washington Post, January 2, 1985, p. A24. 29. According to Jakkie Potgieter, Savimbi first acknowledged that diamonds were an important source of revenue for UNITA during his November 1979 tour of the United States. See Potgieter, “‘Taking Aid from the Devil Himself: UNITA Support Structures,” p. 260. 30. Chester Crocker, High Noon in Southern Africa: Making Peace in a Rough Neighborhood (New York: WW Norton, 1992), p. 173. 31. Abel Chivukuvuku, representative of the President of UNITA to the President of Angola. Interview with the author, November 24, 1997. 32. Luis Graciano, interview with the author, December 9 1997. 33. Steve Mufson, “Angolan Rebel Leader Courts U.S. Aid,” Wall Street Journal, December 17, 1981, p. 29. 34. As articulated by the American Assistant Secretary of State for Africa, Chester Crocker, the “Reagan Doctrine” sought to ensure that “those who have a positive alternative to the Soviet model” receive American support. The objective was “to help the Soviets learn that it was one thing to seize power using instruments of coercion; it was something altogether different to prop up year after year a collection of feckless and bankrupt regimes that lacked a popular mandate and depended overwhelmingly on Soviet guns to survive.” See Crocker, High Noon in Southern Africa, p. 290. The doctrine was first articulated by President Reagan in a speech to the United Nations on October 24, 1985. For a critical view of the Reagan Doctrine see Michael McFaul, “Rethinking the Reagan Doctrine in Angola,” International Security, vol. 14, no. 3, (1989-1990), pp. 99-135.

The Angolan Civil War

229

35. George D. Moffett III, “US Debates Support for Angolan Rebels vs. Broader African Policy,”Christian Science Monitor, February 11, 1986, p. 6. 36. E. A. Wayne, “African Leaders Want US Mediation in Angola, But No Rebel Aid,” Christian Science Monitor, April 18, 1988, p. 7. 37. Angolan officials, for their part, claimed even at that time that the $15 million figure was a “gross underestimation” of the extent of U.S. assistance to UNITA. “Fifteen million dollars is a joke,” remarked General António França N’dalu, commander of Angola’s army. “That barely supplies a battalion.” See James Brooke, “Angola Says U.S. Uses Zaire Bases to Train Rebels,” New York Times, May 26, 1988, p. A10. 38. James Brooke, “Cuba’s Strange Mission in Angola,”New York Times, February 1, 1987, sec. 6, p. 24. 39. James Brooke, “Cuba’s Strange Mission in Angola,”New York Times, February 1, 1987, sec. 6, p. 24. 40. In February 1984, the Lusaka agreement was signed between Angola and South Africa whereby South Africa agreed to withdraw from southern Angola on the condition that Luanda restrain SWAPO guerrillas from interfering in Namibia. Over the next year, South Africa continued to enter Angolan territory, claiming that Angola had not lived up to its side of the bargain. See Novicki, “Against All Odds,” Africa Report, January-February 1985; Patrick Laurence, “S. Africa Stalls on Namibian Independence,” Christian Science Monitor, April 19, 1985, p. 7; and “Crossing Borders: A Glance at S. Africa’s Record,” Christian Science Monitor, December 23, 1985, p. 9. 41. Cited in “Angola’s Feelings Don’t Get in the Way of Profits,” New York Times, January 12, 1986, sec. 4, p. 3. 42. See the interview by Margaret A. Novicki with Cuba’s Deputy Minister of Foreign Affairs, Ricardo Alarcon, Africa Report, November-December 1988, pp. 39-41. 43. Bridgland, Key to Africa, p. 458. 44. “Angola: The Struggle Continues” in Frontline Southern Africa: Destructive Engagement, Phyllis Johnson and David Martin, eds. (New York: Four Walls Eight Windows, 1988), p. 147. 45. Ted Morello, “Angolan Envoy Warns Against US Aid to Savimbi,” Christian Science Monitor, January 31, 1986, p. 9. 46. Cited in Karl Maier, “Angolan Civil War: Air Power is Key,” Christian Science Monitor, August 19, 1986, p. 9. The view that South Africa gave UNITA its teeth was shared by Western observers as well. “UNITA would be a less formidable fighting force,” wrote Robert I. Rotberg, “if it were not assisted financially, militarily and logistically by South Africa. Indeed, it is easy to argue that UNITA has packed a robust punch only because of the steady transfer of supplies from South Africa and because of the direct assistance of a South African-officered battalion, and the provision of South African air cover, radar surveillance, and shared intelligence.” See Robert I. Rotberg, “Angola Sting,” Christian Science Monitor, April 10, 1986, p. 16. 47. See comments by President José Eduardo dos Santos on the South African intervention in 1985 in his interview with Margaret A. Novicki in Africa Report, January-February, 1986, pp. 4-7.

230

Civil War in African States

48. Africa Report interview with Venancio de Moura, July-August 1988, pp 46-48. 49. See “Two Views of Angola’s Problems,” Christian Science Monitor, July 1, 1988, p. 4. 50. Steve Mufson, “Angolan Rebel Leader Courts U.S. Aid,” The Wall Street Journal, December 17, 1981, p. 29. Savimbi made a similar statement in 1988 during his trip to the United States: “I think that anything that we got from South Africa was out of necessity. Because when one is fighting a war, you have to get support where you can get it. It is not only UNITA that has been getting support. The founding fathers here [in the United States] got support from the monarchies of France and Spain, and they did not approve at all of the systems which were there because they were fighting to have a more democratic system at the time.” See Africa Report’s interview with Jonas Savimbi, in the July-August 1988 issue, p. 49. 51. Although there is no official document outlining the policy of constructive engagement, the approach was elaborated in an article by Chester Crocker in “South Africa: Strategy for Change,” Foreign Affairs, vol. 59, no. 2 (Winter 1980-81), pp. 323-351. The thrust of the Foreign Affairs article is concerned mostly with the politics within South Africa rather than a regionwide policy of conflict resolution in southern Africa. 52. Marcos Samondo, Minister of Geology and Mines in the Government of National Reconciliation. Interview with the author, December 5, 1997. 53. Cited in Andrew Meldrum, “At War with South Africa,” Africa Report, January-February 1987, pp. 29. 54. SWAPO stands for South West Africa People’s Organization. 55. Confidential interview, November 20, 1997. For a discussion of the events and impact of Cuitu Cuanavale, see W. Breytenbach, “Cuito Cuanavale Revisited: Same Outcomes, Different Consequences,” Africa Insight, vol. 27, no. 1 (1997), pp. 54-62; and Edward George, The Cuban Intervention in Angola, 1965-1991: From Che Guevara to Cuito Cuanavale (London: Frank Cass, 2005), chapter 10. 56. Crocker, High Noon in Southern Africa, p. 442. 57. Statistics from James Brooke, “Angola’s Economic Struggles,” New York Times, January 31, 1986, p. D1; Karl Maier, “Angolan Civil War Keeps Potential Economic Powerhouse Poor,” Christian Science Monitor, September 12, 1986, p. 15. 58. W. Raymond Duncan and Carolyn McGiffert Ekedahl, Moscow and the Third World Under Gorbachev (Boulder: Westview, 1990), p. 172. Assistant Secretary of State Chester Crocker, in contrast, claims that he “saw no evidence of Soviet arm-twisting” and states that in 1988 Moscow still had not begun to reduce aid to Cuba. See Crocker, High Noon in Southern Africa, p. 422-23.. 59. James Brooke, “Angola Gives Hints of Willingness to Seek Political Solution to War,” New York Times, January 28, 1989, p. 4. 60. Margaret A. Novicki, “Interview: A Conversation with Ricardo Alarcón,” Africa Report, November-December 1988, pp. 39-41. 61. Jorge interview #1 with the author, November 29, 1997. 62. Cited in E. A. Wayne, “‘Big Prize’ for Southwestern Africa,” Christian Science Monitor, December 22, 1988, p. 1. The two resolved wars were 1)

The Angolan Civil War

231

South Africa and SWAPO, and 2) South Africa and Angola. The remaining unresolved conflict was, of course, UNITA and the MPLA. 63. Anticipating an agreement of which it was not a part, Savimbi put the parties on alert that, “If there is no agreement between UNITA and [Luanda] there will be no cease-fire.” Cited in Lynda Schuster, “Angola’s Rebels Decry US Peace Plan,” Christian Science Monitor, August 22, 1988, p. 1. Four days later, Jeremias Chitunda, UNITA’s vice-president, echoed the sentiments, stating that “a deal without UNITA is no deal at all.” Cited in Lynda Schuster, “For Angola Rebels Business Is Booming,” Christian Science Monitor, August 26, 1988, p. 1. In the end, the New York Accords did not include UNITA in national reconciliation, but neither did they stipulate an end to superpower aid to their respective clients. 64. See Jeane Kirkpatrick, “Deal Making, Soviet Style,” Washington Post, October 3, 1988, p. A11; and David B. Ottaway, “Fate of Angolan Guerrillas Unresolved by Accord,” Washington Post, July 21, 1988, p. A15. On September 19, Savimbi wrote a letter to President Reagan urging him of the need “to make clear to all parties that the United States will not stop providing necessary assistance to UNITA until the Soviets, the Cubans, the Eastern bloc countries . . . cease providing military supplies to the MPLA” and to continue pressing for a Government of National Reconciliation. 65. Cited in Schuster, “Angola’s Rebels Decry US Peace Plan,”Christian Science Monitor, August 22, 1988. 66. On the repercussion of the New York accords for UNITA, Abel Chivukuvuku noted the concerns in UNITA’s leadership at the time: “We had to think about it. Dr Savimbi was very attentive to those discussions. As far as our view at the time was concerned this was a threat to UNITA because we could no longer count on the South Africans and their strong backing as far as logistics. That is when we also started launching several political initiatives along with the effort of the war. ... But it was a problem to think about having to do all the logistics by plane. The Americans at that point were doing the logistics to support us, but it was not comparable to South Africa’s logistics. The American logistics was in quality superior, but in quantity inferior. Stingers and TOWs and anti-tank weaponry and training and so on as far as quality is concerned superior, but as far as quantity is concerned inferior.” Interview with the author, November 24, 1997. 67. The call for a ceasefire was made in an interview and did not represent a formal initiative on the part of the MPLA. UNITA insisted that it would not consider any offer until it became a formal position of the MPLA. See James Brooke, “Angola Gives Hints of Willingness to Seek Political Solution to War,” New York Times, January 28, 1989, p. 4. 68. Collen Lowe Morna, “Peace Slow to Follow Angola Accord,” Christian Science Monitor, February 17, 1989, p. 6. 69. Morna, “Peace Slow to Follow Angola Accord,” Christian Science Monitor, February 17, 1989. 70. In an interview with Africa Report, Pedro de Castro Van-Dunem ‘Loy’, Angola’s new minister of External Relations, stated that, “At a certain stage in the negotiations, the American administration said that when Angola presented a calendar for Cuban withdrawal acceptable to all the parties, then the U.S. would begin a different type of relationship with Angola, recognizing Angola

232

Civil War in African States

diplomatically, ending assistance to UNITA. We attained an acceptable calendar and signed a global accord, but to our surprise the American administration publicly affirms that it will continue support to UNITA and the destabilization of Angola.” Pedro de Castro Van-Dunem ‘Loy,’ “Creating Conditions for Peace,” interview with Margaret Novicki, Africa Report, March-April 1989, p. 24. Similar assurances were reported by other sources. See, for instance, David B. Ottaway, “Angola Offers Removal of All Cuban Troops,” Washington Post, February 14, 1988, p. A32. Representative Howard Wolpe also argued before Congress that, since the Cuban presence in Angola had been the original rationale for aid to UNITA, that assistance could now be terminated. Congressional Record, Intelligence Authorization Act for Fiscal Year, 1990, House of Representatives October 12, 1989, p. H7014. Washington was now arguing that support for UNITA would be sustained until a political agreement was worked out between the government and the rebels. See J. Stephen Morrison, “Mr. Savimbi Goes to Washington,” Africa Report, SeptemberOctober 1988, p. 56. 71. The letter was subsequently leaked to the media. See “President-Elect Bush Makes First Foreign Policy Commitment,” Washington Post, January 12, 1990, p. A1. 72. Robert Pear, “C.I.A. Crash Called Problem in Angola,” New York Times, December 1, 1989, p. A9. “U.S. Shows Its Hand in Angola Plane Crash,” Africa Report, November-December 1989, p. 8. 73. James Baker III, The Politics of Diplomacy (New York: G.P. Putnam and Sons, 1995), p. 599. 74. Howard Wolpe, “More U.S. Aid For Savimbi,” New York Times, June 10, 1989, p. 27; John Battersby, “Namibia Success Raises Hopes for Angola Cease-Fire,” Christian Science Monitor, March 23, 1990, p. 1. Savimbi claimed that his forces had sufficient stockpiled arms to carry on a country-wide war for the two years it would take for Cuban troops to complete their withdrawal. See “U.S. Aid to Savimbi: Bush Pledges to Keep the Money Coming,” Africa Report, March-April 1989, p. 6. 75. James Baker, Politics of Diplomacy, p. 599. 76. Herman J. Cohen, Intervening in Africa: Superpower Peacemaking in a Troubled Continent (Houndsmills, Macmillan: 2000), p. 91. 77. Karl Maier, “The Military Stalemate,” Africa Report, May-June, 1988, p. 35. See also the interview with Jonas Savimbi in Africa Report, July-August 1988, pp. 49-50. 78. Jonas Savimbi, “I Have No Secret Agenda,” New York Times, October 30, 1989, p. 19. In an October 5, 1989 speech to the Heritage Foundation, Savimbi stated that “If this plan had been presented to us, we would have rejected it, because we did not fight for fourteen years against the Soviets, Cubans, and the communist system in order to join them. We have resisted for fourteen years in order to achieve freedom and have elections and democracy in our country.” Jonas Savimbi, “The Coming Winds of Democracy,” (Heritage Foundation, 1989). 79. Jonas Savimbi, “I Have No Secret Agenda,”p. 19. 80. Flora Lewis, “The New Climate, New York Times, January 26, 1989, sec. 4, p. 25.

The Angolan Civil War

233

81. “Pedro de Castro Van-Dunem ‘Loy’: Creating Conditions for Peace,” Interview with Margaret A. Novicki. Africa Report, March-April 1989, pp. 2325. 82. James Brooke, “Angola Sees Gain in Free Namibia,” New York Times, March 21, 1990, p. A12. 83. UNITA had already announced its recognition of the MPLA on May 1, 1990. 84. “These guys can’t make it on their own in negotiations,” Baker reports Shevardnadze as saying. “We need to give them a push. ... Then we need to bring in all the parties and sell it to them.” Baker, The Politics of Diplomacy, p. 599. 85. The MPLA had long insisted on a fifteen-month period between a ceasefire and elections; UNITA argued that it would accept a year as the maximum time. A compromise proposed by the Portuguese mediator, José Duao Barroso (the Portuguese Secretary of State for Foreign Affairs) whereby elections would be held between fifteen and eighteen months after a cease-fire was eventually accepted. 86. Cohen, Intervening in Africa, p. 123. 87. James Baker, Politics of Diplomacy, p. 600; Chivukuvuku, interview with the author, November 24, 1997. 88. Linda Heywood, Contested Power in Angola: 1840s to the Present (Rochester: University of Rochester Press, 2000), p. 215. 89. Jonas Savimbi, “I Have No Secret Agenda,” New York Times, October 30, 1989, p.19. 90. Abel Chivukuvuku noted: “I think ... there was a kind of momentum. By 1991 the communist regimes were falling. We and our friends, the Americans, we were all led to that belief that okay maybe it spreads to Angola and the process continues.” Interview with the author, November 24, 1997. 91. Allan Cain, interview with the author, November 22, 1997. Furthermore, in 1988, UNITA dissidents revealed a Practical Guide to the Cadre which consisted of lessons taught to UNITA cadres at the Command Kapessi Kafundanga Study Center in Jamba. Believed to be authentic, the guide stated that “the idea of reconciliation is part of our tactical position. But it does not form part of our strategy, because that is not possible.” Cited in Congressional Record, Intelligence Authorization Act for Fiscal Year, 1990, House of Representatives October 12, 1989, p. H7023. 92. John Battersby, “Angolans Begin to Disarm Under Successful CeaseFire,” Christian Science Monitor, July 3, 1991, p. 6. 93. John Battersby, “Angolans Begin to Disarm Under Successful CeaseFire.” 94. John Battersby, “Angolans Begin to Disarm Under Successful CeaseFire,”p. 6. 95. In his diaries, Jeremias Chitunda worried about whether Savimbi would take his oath of office on a Constitution of Angola written by the MPLA and a flag chosen by the MPLA. On July 27, his diary advised that UNITA should “prevent the present symbols from being imposed when our elected president is sworn in.” Chitunda did not think it would be tactically appropriate to discuss the matter in public, “so we do not appear to be pretentious.” See “The Diaries of UNITA’s Jeremias Chitunda,” FBIS-AFR-93-018-S.

234

Civil War in African States

96. The defectors claimed that Savimbi was indeed responsible for the deaths of Tito Chingunji and Wilson dos Santos. “I had access to the normal prisoners who are under police guard,” stated Miguel Maria N’Zau Puna in an interview. “But those people were under the guard of the President.” Savimbi, he says “wanted to show me that they were killed by the security people. But as my experience—I’ve been in UNITA so long—I know that ... they can’t go and [make] their own decision to do that.” Interview with the author, March 23, 2005. 97. One of the UNITA defectors, Miguel Maria N’Zau Puna, claimed to be present with Savimbi on the occasion of one of the witch burnings. “The day we went to attend the burning of those people—I was present. I said ‘Mr President, no.’ He said, ‘You came from Cabinda. It is far away. You don’t know the customs of the people of this area. I know them. We must punish them in such a way.’” Interview with the author, March 23, 2005. For discussions on UNITA human rights abuses see Imani Countess, “US Must Hold Savimbi Accountable, Support Angola Peace,” Christian Science Monitor, April 23, 1992, p. 23; Human Rights Watch-Africa, Angola Civilians Devastated by 15 Year War, vol. 3, no. 1 (February 5, 1991); and Africa Watch, “Angola: Violations of the Laws of War by Both Sides.” See also Fred Brigdland’s account, Death in Africa (Cold Type: www.coldtype.net. 2004). 98. In the 14 August entry of his diary, Jeremias Chitunda wrote: “What is it that attracts the youth to MPLA? (...) We need to be more effective in mobilizing the young people (...). Assign articulate personnel to work with youth, someone able to appeal to the young of the asphalt [city].” Chitunda also worried that UNITA’s message was not reaching Angolan intellectuals and the more educated classes. On 20 August, with just over a month before the election, he wrote: “Take a good look at the slogans. ‘Savimbi is our choice’ will not work for the intellectuals – the ones in the cities, for example.” “The Diaries of UNITA’s Jeremias Chitunda,” pp. 1-2. 99. Kenneth B. Noble, “Election Makes Skeptics of Angolans,” New York Times, June 28 1992, p. 8. 100. Cited in Andrew Meldrum, “Hungry to Vote,” Africa Report, November-December 1992, p. 27. Anstee notes a similar phenomenon: “Dr Savimbi’s electioneering caravan was roaming all over the country and his speeches were seen as increasingly aggressive. One phrase of his, frequently quoted by alarmists and opponents alike, was ‘If UNITA does not win, then it means fraud!’” See Margaret Joan Anstee, Orphan of the Cold War: The Inside Story of the Collapse of the Angolan Peace Process, 1992-3 (Houndsmill: Macmillan 1996), p. 151. 101. Human Rights Watch, Angola: Arms Trade and Violations of the Laws of War Since the 1992 Elections” (New York: Human Rights Watch, 1994), p. 11. 102. Herman J. Cohen, Intervening in Africa: Superpower Peacemaking in a Troubled Continent (New York: Palgrave 2000), p. 116. 103. Vladimír Kr_ka, “Peacekeeping in Angola (UNAVEM I and II),” International Peacekeeping, vol. 4, no. 1 (Spring 1997), p. 87. 104. Cohen, Intervening in Africa, p. 116. 105. Anstee, Orphan of the Cold War, p. 147. 106. Anstee, Orphan of the Cold War, p. 78.

The Angolan Civil War

235

107. Anstee, Orphan of the Cold War, pp. 200-201. Both UNITA and the MPLA had representatives throughout the country on September 29 and 30, the two days for voting, in order to verify the vote. National Election Council officials claim that on those days UNITA representatives did not make any claims that vote-rigging had occured. See Testimony of the President of the NEC, Dr Caetano de Sousa, in Angola: White Book About Peace Process, volume 1 (Luanda: Republic de Angola, 1995), pp. 126-127. 108. “Declarations by UNITA Leader Jonas Savimbi,” Radio Vorgan, October 3, 1992, cited in Angola: White Book, p. 198. 109. John Battersby, “Angolan Rebels’ Threat to Bolt Puts Democratic Vote in Jeopardy,” Christian Science Monitor, October 13, 1992, p. 1. 110. Confidential interview, October 4, 1992. The New York Times reported that Angola’s National Electoral Commission had released figures showing dos Santos with about 55% of the vote in the presidential contest and Savimbi with 25%. See Kenneth B. Noble, “Savimbi, Trailing, Hints at New War,” New York Times, October 4, 1992, p. 15. 111. The results of legislative balloting, according to these figures, provided the MPLA with 129 seats in the 223-seat parliament. UNITA won 70 seats. 112. Most notably, Savimbi told Margaret Anstee, “I give you my word of honour that I will not begin a war, even if we lose the election. ... That is a guarantee ... I can tell you that even if some officers on either side might want to start another war, the troops on both sides do not want that.” Anstee, Orphan of the Cold War, p.148. In the election aftermath, both President dos Santos and Savimbi allegedly reassured the South African Foreign Minister Roleof “Pik” Botha that they would not return to war to settle their differences. See “Angolan Rivals Meet To Hash Out Deal On Presidential Vote” Christian Science Monitor, October 19, 1992, p. 6. 113. See “The Diaries of UNITA’s Jeremias Chitunda,” FBIS-AFR-93018-S. 114. Patrick Smith, “Angola: Free and Fair Elections,” Review of African Political Economy, no. 55 (November 1992), p. 101. 115. See Smith, “Angola: Free and Fair Elections,” p. 101. 116. Assis Malaquias, “Ethnicity and Conflict in Angola,” p. 107. 117. In the diary uncovered in the days after the election, Jeremias Chitunda, UNITA vice president, stated the ways in which the elections were fraudulent: “coercion by the AM [anti-riot] police; missing ballot boxes; extra polling places; no accounting or balancing sheets; strange coincidences.” See “The Diaries of UNITA’s Jeremias Chitunda.” 118. Anstee and other observers did not deny the existence of irregularities but claimed they were “mainly due to human error and inexperience.” On the same day the results were announced, Margaret Anstee detailed the observations of the United Nations stating that with “all deficiencies taken into account” the elections could be considered to have been “generally free and fair.” See Angola: White Book, p. 135. The OAU, the European Community (EC) and other international organizations made statements supporting the claim that the elections had been free and fair. 119. Allan Cain, interview with the author, November 22, 1997. Other scholars have pointed out that Angola’s wealth, and its lack of dependence on

236

Civil War in African States

donors who might have been able to exert ongoing political pressure, meant that there were no ways of reassuring UNITA that Angola’s incipient democracy would be maintained in future. See Paul Collier, “Economic Causes of Civil Conflict and Their Implications for Policy,” in Chester A. Crocker, Fen Osler Hampson, and Pamela Aall eds.,Turbulent Peace: The Challenges of Managing International Conflict (Washington: United States Institute of Peace, 2001), p. 159. 120. Savimbi’s “great hero,” Winston Churchill, she told him, “had also suffered unexpected defeat in the elections after the war, a war that he had brought to a victorious conclusion. Yet he had continued to play an important role as the leader of ‘His majesty’s Loyal Opposition’ ... and five years later had been voted back into power.” Orphan of the Cold War, p.217. 121. Anstee, Orphan of the Cold War, p. 217. 122. “The Diaries of UNITA’s Jeremias Chitunda,” p. 5. 123. “The Diaries of UNITA’s Jeremias Chitunda.” Chitunda was killed while trying to escape from Luanda on November 1, 1992. His diary was recovered soon after. UNITA officials did not deny that the diary was authentic. Underline in original. 124. Anstee, Orphan of the Cold War, p. 276. 125. Anstee, Orphan of the Cold War, p. 309. 126. Confidential interview with the author, November 22, 1997. 127. Kenneth B. Noble, “A New Crisis Engulfs Angola As the Rebels Make Big Gains,” New York Times, January 29, 1993, p. 1. 128. While UNITA was able to exploit the diamond areas for its own purposes by sending its own troops to work in the mines, it was not able to generate revenue from Angola’s oil region. When UNITA took over the town of Soyo, the major oil companies withdrew their workers. Thus, while control of diamond towns such as Cafunfu generated revenue for UNITA’s war effort, the sole effect of UNITA’s control of Soyo was to deny these revenues to the government. 129. Earlier, on January 20, 1993, UNITA had announced on BBC radio that its troops had captured both Soyo and Huambo. The statement was premature. In fact, Huambo was not captured until several weeks later, on March 6, 1993. “Angola Rebels Seize Oil Center and Old Base,”New York Times, January 21, 1993, p. A5. 130. John Battersby, “Angolan Rebels Win Concessions After War Gains,” Christian Science Monitor, April 22, 1993, p. 8. 131. Noble, “Angolan Rebels Rebound,” New York Times, April 13, 1993. 132. John Battersby, “Violent ‘Cleansing’ Part of Angola War, Aid Workers Say,” Christian Science Monitor, February 3, 1993, p. 7. 133. Cited in Kenneth B. Noble, “Angola President Says His Forces Are Rearming After Rebel Gains,” New York Times, February 2, 1993, p. A6. 134. Bill Keller, “After Big Gains, Angolan Rebel Offensive is Halted,” New York Times, September 24, 1993, p. A3. 135. To the Angolan government’s frustration, the outgoing Bush administration had claimed that it would not recognize the Angolan government until the election process had been completed—a stipulation that required dos Santos to go through a second ballot, as required under the Bicesse agreement. An exasperated President dos Santos responded by claiming that his hands were

The Angolan Civil War

237

tied: “To maintain a position which does not recognize Angola means to favour UNITA. ... There hasn’t been a second round of the ballot yet because Savimbi decided to go back to war.” He added, “There is no control over the whole territory because Savimbi managed to occupy some of the territory by violating the Bicesse accords.” See John Battersby, “Angolan President Urges US Role to Help End Renewed Civil War,” Christian Science Monitor, January 25, 1993, p. 1. 136. Human Rights Watch: Angola: Arms Trade and Violations of the Laws of War Since the 1992 Elections, p. 36. 137. Government officials claimed that they did not intend to take the city but only to encircle it so that UNITA could not bring heavy weapons into the city once the peace agreement was signed. Bill Keller, “New Path to Peace in Angola: Clamp Rebels and Hold Tight,” New York Times, November 6, 1994, p. 1. 138. The government offensive was regarded by the international community as a violation of the spirit if not the letter of the Lusaka Protocol. According to Paul Hare, “the American administration was incensed because it had thought the Angolan government had made firm commitments not to engage in military operations against UNITA positions, especially after the peace agreement was effectively secured.” See Paul Hare, Angola’s Last Best Chance for Peace: An Insider’s Account of the Peace Process (Washington: United States Institute of Peace, 1998), pp. xiii-xiv. 139. Hare, Angola’s Last Best Chance for Peace, p. 18. 140. UNITA’s lead negotiator and Secretary General, Eugenio Manuvakola, signed on UNITA’s behalf. In a subsequent interview, Manuvakola stated that he had urged Savimbi to go to Lusaka “because it was a very important moment for the party and it was the duty of the President to be there. I said I had had contacts with the Zambian government and I could guarantee there was enough security and that he would not suffer an attack. The response to my request was that I was trying to attract Dr. Savimbi into an ambush.” See “Press Conference by Eugenio Manuvakola, Former UNITA Secretary General Regarding His Escape From UNITA,” ReliefWeb, August 28, 1997. 141. Annex 6 of Lusaka Protocol. 142. FBIS-AFR-96-053 (18 March 1996), p. 10. Savimbi also claimed that he would not accept the role of vice-president unless other opposition parties were accepted into the government. 143. Savimbi suggested that “people tell me to reject it [the vice presidency] because they think it is a trap. I do not know what you want any more.” FBIS-AFR-96-053 (March 18, 1996), p. 10. Other UNITA dissidents have informed the author that UNITA parliamentarians sent letters to Savimbi warning him that coming to Luanda would be at great personal risk. 144. Paulo Jorge, interview with the author, November 29, 1997. 145. Jardo Muekalia, interview with the author, July 7, 1999. 146. Jardo Muekalia, interview with the author, July 7, 1999. The list of amenities to be provided to Savimbi in one offer included a residence, driver, cook, domestics (two), first-class air tickets (two per year), and a monthly salary equal to that of the prime minister. See Hare, Angola’s Last Best Chance for Peace, p. 110-111.

238

Civil War in African States

147. Tony Hodges, Angola: From Afro-Stalinism to Petro-Diamond Capitalism (Oxford: James Currey, 2001), p. 53. 148. “Angola: Savimbi’s Last Stand,” Africa Confidential, vol. 39, no. 5 (March 6, 1998). 149. Eugenio Manuvakola, interview with the author, July 21, 2004. 150. Lusaka Protocol, Annex 6. 151. “Demobilising but Still Divided,” Economist, September 14, 1996. 152. Allan Cain, interview with the author, November 22, 1997. 153. David Wimhurst, interview with the author, February 6, 1998. 154. Marcus Samondo, interview with the author, December 5, 1997. 155. Tony Hodges, Angola: Anatomy of an Oil State, 2nd edition (Oxford and Bloomington: James Currey/Indiana University Press, 2004), pp. 191-192. 156. Angola Peace Monitor, vol. 2, no. 9 (1996), p. 4. 157. Suzanne Daley, “Foes in Angola Still at Odds, Over Diamonds,” New York Times, September 15, 1995, p. 1. 158. David Wimhurst, interview with the author, February 6, 1998. 159. “Angola II: Talking Drums,” Africa Confidential, vol. 39, no. 10 (15 May 1998). 160. David Wimhurst, interview with the author, February 6, 1998. 161. Robert Fowler, Testimony Before the UN Security Council, January 18, 2000. 162. Paulo Jorge, interview with the author, November 29, 1997. 163. Foreign Broadcast Information Service-Daily Report (Sub-Saharan Africa) March 15, 1996. 164. Robert Fowler, Testimony Before the UN Security Council, January 18, 2000. 165. Paul Hare, interview with the author, June 20, 2000. 166. Hare, Angola’s Last Best Chance for Peace, p. 105. 167. Jardo Muekalia, interview with the author, July 7, 1999. Foreign observers backed up UNITA’s claims that the government was heavy-handed. See “Angola on the Way to War,” Globe and Mail, July 27, 1998. 168. David Wimhurst, interview with the author, February 6, 1998. 169. Jardo Muekalia, interview with the author, July 7, 1999. 170. Jardo Muekalia, interview with the author, June 9, 2004. 171. Cohen, Intervening in Africa, p. 121. 172. Malaquias, “Diamonds are a Guerrilla’s Best Friend,” pp. 315-317. 173. Savimbi’s conviction was that, prior to Bicesse, UNITA had had sufficient momentum for military victory, and that continuing the military course would have been a better bet than having to subject himself to the uncertainties of an election. “I would like to make it clear that one of our greatest errors was to sign the Bicesse Accord in 1991,” he stated in 1996. “It was a major error, and I am here to admit that error. UNITA had everything to continue its unstoppable struggle. It was an error and I fully admit it.” “Angola: More on Savimbi’s Speech Marking UNITA’s Anniversary,” FBIS, 18 March 1996, p. 10. 174. “Press Conference by Eugenio Manuvakola, Former UNITA Secretary General Regarding His Escape From UNITA,” ReliefWeb, August 28, 1997. 175. When asked whether Savimbi’s reluctance to come to Luanda was rational, Savimbi’s successor, Isaías Samakuva, claimed that the UNITA leader

The Angolan Civil War

239

had feared for his life. According to Samakuva, “It was part of the strategy to get Dr. Savimbi under the impression that his life was in danger.” When asked if Savimbi’s life really was in danger, Samakuva says, “I have no doubt.” Interview with the author, July 14, 2004. 176. Georges Marion, “Le chef de l’armée angolaise estime que la guerre pourrait reprendre,” Le Monde, 16 February 1995, p. 5. 177. “UNITA Suffers Key Defections,” Angolan Mission Observer, December 1999. 178. Fred Bridgland, “It Took a Rebel to Catch a Rebel,” Globe and Mail, April 3, 2002, p. A11. 179. General Geraldo Nunda, interview with the author, July 19, 2004. 180. See, for example, “UNITA Suffers Key Defections,” The Angolan Mission Observer, December 1999. 181. “Angola: Strings Attached,” Africa Confidential, vol. 39, no. 20 (October 9, 1998), p. 1. 182. All but two of the seven suspended UNITA ministers and viceministers were reinstated. Geology and Mines Minister Marcos Samondo emigrated to the United States while the Vice-Minister for Social Communications, João Evangelista, left the government. 183. Angola Peace Monitor, no. 1, vol. v., (October 7, 1998). Savimbi did his best to undermine these efforts to speak on UNITA’s behalf in Luanda. In April 1998, Savimbi met with the MPs and told them that their project would not work and tried unsuccessfully to replace Chivukuvuku with Celestino Kapapelo, a Savimbi loyalist. See “Angola: Facing Up to Savimbi,” Africa Confidential, vol. 39, no. 23 (November 20, 1998), p. 4. 184. United Nations Sanctions Committee, Report of the Panel of Experts on Violations of Security Council Sanctions Against UNITA, March 10, 2000. 185. See Thomas Turner, “Angola’s Role in the Congo War,” in John F. Clark ed. The African Stakes of the Congo War,” (New York: PalgraveMacmillan, 2002). 186. Angola Peace Monitor, no. 1, vol. v., (October 7 1998). 187. Ian S. Spears, “Angola’s Elusive Peace,” International Journal, vol. 54, no. 4 (Autumn 1999), p. 567. 188. Confidential interview with the author, November 22, 1997. 189. Cited in Angola Peace Monitor, no. 1, vol. v., (October 7 1998). 190. Lynne Duke, “Heavy Fighting in Angola Eclipses Peace Accord,” Washington Post, October 13, 1998, p. A12. 191. Jardo Muekalia, interview with the author, September 17, 1999. 192. Paulo Jorge, interview with the author, June 21, 2004. 193. “Interview — UNITA Sets Conditions for Angola Talks,” Reuters, December 16, 1998. 194. Lara Pawson, “‘End of War in Sight’—Angolan General,” BBC News, November 16, 1999. 195. “Angola’s President Reaffirms Commitment to Democracy; Calls on Savimbi to Take the Path to Peace,” Angola News Flash, May 11, 2001. 196. Buchizya Mseteka, “UK’s Hain Says Angola Rebel Boss Must Be Removed,” Reuters, January 31, 2000.

240

Civil War in African States

197. General Geraldo Nunda, formerly one of UNITA’s senior-most commanders before his defection, led the final offensive, named Kissonde, after an aggressive African ant. 198. Isaías Samakuva, interview with the author, July 14, 2004. UNITA did in fact accept the results of the 2008 elections. “Despite all that has happened,” Isaías Samakuva stated, “the UNITA leadership accepts the results of the elections.” He added that he hoped the MPLA “governs in the interests of all Angolans.” “Angola: Election Free and Fair, Sort of,” Integrated Regional Information Network, September 9, 2008. 199. Stephanie Nolen, “Angola’s Long March to Peaceful Elections,” Globe and Mail (Toronto), September 5, 2008, p. A17. 200. Stephanie Nolen, “With a Booming Economy Benefiting the Elites, Regular Angolans Find Little Solace in Democracy,” Globe and Mail, September 6, 2008, p. A1. 201. Anstee, Orphan of the Cold War, p. 519. 202. Anstee, Orphan of the Cold War, p. 519.

5 The Search Continues

Consociationalism [power-sharing] is designed to minimize conflict in ‘deeply divided’ societies. Yet it would appear that such societies would not adopt consociationalist measures in the first place until levels of hostility had diminished substantially. Vincent Maphai Violence begets violence. Martin Luther King The secessionists are no better than the governments. ... Secessionists, who fight against the principle of territorial integrity of the state as a whole, insist on the territorial integrity of the seceding region. Benyamin Neuberger

In each of the cases considered in this volume, political transitions came with great optimism and promise that a new democratic era was upon them. The same optimism was felt everywhere at the end of the Cold War. In 1988, political commentator Joseph C. Harsch wrote: “Take a moment . . . and you will discover that the difference between now and a year ago is truly remarkable. Peace is by no means universal. There are still horrible examples of unnecessary famine, misery, war, and brutality. But there is immensely less war and danger of war than there was a year ago.”1 Reviewing the world situation, albeit four years later, newly-elected President Bill Clinton saw things differently: “On balance, let’s make one thing clear,” he said. “It is a wonderful thing

242

Civil War in African States

that the Cold War is over. But let’s also admit that the end of the bipolar world has made it possible to peel off a layer of human aggression and made it possible in some parts of the world for people to be starved, brutalized and killed with much greater abandon.”2 Indeed, seven years after the overthrow of Mengistu, Ethiopia and Eritrea returned to a bloody cross-border struggle; Somalia, for its part, remains in a perpetual state of anarchy; and, despite its moments of peace in the early and mid-1990s, Angola’s civil war did not end for another ten years. Two decades after the end of the Cold War none of the recognized countries in the three case studies examined here had experienced a democratic change in government and all were accused in varying degrees of human rights violations. Politics remains a struggle and, while the prevalence of violent conflict has ebbed and flowed in Africa, the search for security continues. In this sense, without fundamental changes in the nature of the African state system, there is an inevitability about conflict that the international community needs to understand. But to what extent can the strategies of disputants in Africa’s civil wars lead to peaceful outcomes or be shaped in more inclusive and/or less destructive ways? More importantly, perhaps, what can mediators and the broader international community do to help African states avoid violent conflict or reduce its frequency and duration? Contemporary conflict resolution theory would suggest three approaches. If anarchy is the source of conflict that necessitates the resort to survival strategies, then one approach is for the international community to support interventions that effectively end anarchy or reduce its harmful and limiting effects. Security, from this perspective, is necessary for trust to be built. Otherwise, disputants in these environments are unable to take chances in forging peace, because mistakes will be fatal. A second approach is to recognize that the international community’s ability to reduce anarchy is limited and that the strategies of local actors have much more bearing on conflict outcomes. Left to their own devices, parties who have a stake in the reconciliation of their own security interests will eventually sort themselves out. In this approach, the world community should either stay out or work with the disputants themselves and accept the consequences of the disputants’ respective strategies. A third, more open-ended and radical approach—one that is not often addressed in the existing literature—involves Africans identifying and abandoning those structural aspects of the contemporary state system which perpetuate conflict. Let us briefly consider each of these options.

The Search Continues

243

Efforts to reduce anarchy should be expected to improve the chances for inclusive or cooperative strategies. In the preceding case studies, integrative solutions such as power-sharing that spanned the primary division of conflict rarely emerged on their own. This is not to say that there is no desire for cooperation. Most actors might prefer cooperation and their public rhetoric often shows a desire for peaceful outcomes. But to the extent that choices are conditioned by the environment, their strategies prioritize security over peace and reconciliation. Moralizing or wishing for peaceful outcomes does not make them happen if such outcomes compromise or jeopardize vital interests. An outside guarantor that can protect both sides in a dispute, on the other hand, may be able to open up a greater array of options beyond separation or domination provided one can be found that both parties can agree on. In that case—assuming they have the desire to do so—local actors can afford to integrate or share power because their security is assured even if they are required to undergo an otherwiserisky disarmament. But what would an end to anarchy look like in Africa? And what problems are likely to emerge? In the 1990s, when a number of states were on the verge of collapse, it became fashionable to advocate some sort of liberal imperialism or trusteeship—an externally imposed form of the domination strategy—as a means of restoring order and encouraging development.3 A number of commentators who advocated such measures have since changed their position, in part because of America’s troubled experiences during interventions in Somalia, Iraq, and Afghanistan.4 Even states that are verging on or in the midst of total collapse are not necessarily amenable to an externally imposed peace, and well-meaning efforts to re-colonize Africa—or elsewhere—would undoubtedly be met with resistance.5 In that case, interveners face the challenge of reconciling the need to crack down in order to enforce peace with the ambition to promote freedom.6 More recent statements and less ambitious approaches to ending anarchy have emphasized “shared sovereignty” or have focused on the intervention of peacekeeping troops in the moments when conflict-prone states need assistance most.7 Proponents of intervention point to transitions in Rhodesia-Zimbabwe, Namibia, Mozambique, Burundi, Liberia and Sierra Leone to show how countries can benefit from the presence of peace-keeping forces. But problems remain, particularly with intervention. The physical size of some of Africa’s largest states—Congo, the Sudan, and Angola, to name three—present obvious challenges for both their governments and the international community. In many cases the national military is too small compared with the

244

Civil War in African States

population and territory it seeks to control, and international assistance is often inadequate for filling the gap of monitoring such large states. In the case of Angola, for example, the mission that was established in February 1995 to monitor the Lusaka Protocol (UNAVEM III), supplied only 7,000 troops—less than half of the 15,000 that were requested—to police a country of over 1.2-million square kilometers.8 Even MONUC, the United Nations’ largest and most expensive peacekeeping mission, has relied on only 20,000 troops to help make peace in the Congo, a country the size of Western Europe. While the forces required to maintain order need not necessarily be large, in the face of a determined insurgency they do need to be effective. However, the international community has often been unable to provide suitably well-trained and disciplined peacekeepers to maintain order in states that are regarded by the guarantor as strategically marginal. Other approaches to reducing anarchy focus on identifying propitious moments to intervene. Practitioners recommend that preventative measures be undertaken in the early stages of a conflict. But determining the optimal moment when a preventative strategy should be implemented is difficult. The warning signs in the early phases of conflicts are often unrecognized or are ignored, and interventions, when they do come, are often too late. In a debate on the possibilities of intervention in Rwanda during the 1994 genocide, for example, Alan J. Kuperman wrote that “the hard truth is that even a large force deployed immediately upon reports of genocide would not have been able to save even half the ultimate victims.”9 Even if the international community is able to intervene in a timely fashion, how long does it need to stay in order to convince would-be disputants that their survival will be assured? Some scholars have recommended that peace-building missions might on average last a total of seven to nine years.10 But time is not often on the side of those who want to deliver peace. Nor is it clear that patience will be rewarded with meaningful reconciliation. When the international community brought Somalia’s southern faction leaders to a Nairobi hotel in March 1994 in the hopes of establishing an integrative settlement, concessions—and these were perfunctory in nature—were only made when the UN refused to pay the hotel bills that had been incurred. In fact, in some cases, internationallyfunded efforts to find or “build” peace—whether pre-emptive or ex-post facto—may be counter-productive. Local initiatives, as the experience in northern Somaliland demonstrates, often demand fewer resources and are therefore less likely to be exploited by the disputants for material gain.11 When it is the parties to a conflict who must cover the expenses

The Search Continues

245

there is a strong disincentive to prolong either the conflict or the peace process unnecessarily.12 In any case, violence, such as that in Rwanda, is also not confined to a single isolated event. While the 1994 genocide was the most tragic example, the post-colonial history of both Rwanda and neighboring Burundi have involved repeated episodes of extreme violence. Similarly, the conflicts in both Ethiopia and the Sudan have persisted through different regimes in their respective capitals. If the structural conditions that give rise to the conflicts transcend the rulers themselves, the idea of an optimal moment for successful intervention is essentially irrelevant. Finally, in their attempt to limit anarchy, foreign powers may support regimes whom they believe have the capacity to maintain authority in the face of increasingly nihilistic insurgents or the rapid degeneration of states into chaos.13 By taking this approach, an authoritarian Eritrea is given priority over an anarchic Liberia or Sierra Leone. But, in the early days of a rebellion, and assuming such a distinction exists, how does the international community distinguish benevolent dictators from predatorial monsters? As was the case in Ethiopia and Rwanda, the groups or organizations that start the conflict (and might therefore be considered nefarious) are also often the ones who are subsequently relied on to deliver and maintain peace. If the international community is forced to rely on authoritarian leaders because viable alternatives are unattractive or do not exist, does one tolerate a Felix Houphouet Boigny in the Ivory Coast, a Mobutu Sese Seko in Zaire, or even a Robert Mugabe in Zimbabwe for the sake of stability, even though they themselves threaten their citizens?14 In short, there are problems and terrible ethical dilemmas involved in efforts to end anarchy in African states. The second general approach to conflict resolution is for the international community either to stay out of conflicts and let the respective strategies of the disputants run their course, or make its approach to conflict resolution compatible with the strategies. As the case studies revealed, even in the absence of a guarantor or some other external means to end anarchy, disputants do still choose non-violent means. But integrative strategies, as with domination and separation strategies, are selected because they satisfy some vital security need. When it came to their adversaries, in none of the cases in this study was it clear that power-sharing or even competitive elections served the purpose of reconciliation. Power-sharing almost always had other motives. In Ethiopia, for example, the Derg sought negotiations with the ELF-RC in order to weaken the Eritrean independence movement not because it wanted peace with it. Subsequently, the TPLF shared power

246

Civil War in African States

only in order to advance its own hegemony rather than to support truly independent ethno-nationalist organizations. In Somalia, northerners forged their own inclusive pacts, but this has been at least partly in response to threats from the south. Notably, Somalilanders have not reconciled with the south because the atrocities committed by Mogadishu hardened northern antipathy to the point where the mere mention of an integrative solution (i.e. a re-unified Somalia) was viewed as tantamount to betraying the Somaliland “national” cause. In Angola, both UNITA and the MPLA looked to the electoral reckoning of 1992 to consolidate their domination. Subsequently, in 1994, the MPLA signed a power-sharing pact only when it came under extreme pressure from the international community, an action which, in fact, helped legitimize its position as the recognized government. Savimbi, for his part, signed on to the Lusaka power-sharing agreement as a means of avoiding outright defeat. No real or lasting reconciliation was achieved between Savimbi and the MPLA leadership. Strategies of domination and secession are also the consequence of the pursuit of self-interest and the absence of other means of security. Separation is often the preferred option because, in the absence of a guarantor, actors know that with statehood comes the right to maintain a separate, independent means of defense. The EPLF, the insurgency that was most—out of this study’s examples—adamantly in favor of secession, now has a capacity to overhaul or “re-manufacture” its own weapons, one of their principal reasons for seeking statehood.15 Groups were also more likely to see separation as a possible alternative to power-sharing or domination if there was some historical basis for independent statehood, as was the case in both Eritrea and Somaliland. Separation may be particularly welcome if it provides one of the disputants the exclusive right to a resource that has previously been denied to it, or which it has had to share. Even in Angola, at least one senior member of UNITA, a movement generally without secessionist aspirations, intimated that UNITA would be interested in regional autonomy if it meant that they could control the country’s diamond regions.16 Domination also has its attractions for groups seeking to preserve their security. But no matter how benevolent a given insurgency’s intentions are, the immediate outcome of a successful strategy of domination is only short-term security and peace, not democracy. The victors are too busy congratulating themselves on their achievements, helping themselves to the perquisites of power, and establishing rules that protect their victory, to provide room for others. Because they often have so few resources, other incipient movements are too small and

The Search Continues

247

disorganized to do more than complain about the lack of political space provided by the victors. Establishing a political landscape that they control and which is free of other threats or challenges inevitably suits the new rulers best—as has been the case in Ethiopia, Somaliland, and Angola. The relief that ordinary citizens attach to such outcomes which bring peace but fall short of democracy is, one assumes, inversely proportional to the misery that they experienced during the war. This second approach to conflict resolution, like the first, offers no panacea. Power-sharing is perhaps the most desirable strategy but, particularly when it involves agreements directly between the principal disputants, its durability is often questionable. On the other hand, there are inevitable and difficult trade-offs that result from the alternatives. Domination in the form of authoritarian rule may go some way to provide order in the short-to-medium term but, in the long term—even aside from ethical questions associated with exclusive rule—it too may invite a violent counter-response from those left out. As one political realist has noted, “success leads to failure”; the excessive accumulation of power by one group often only provokes the opposition of others.17 Nor does secession necessarily mean conflict avoidance—as the recent interstate war between Ethiopia and Eritrea shows—or that the seceding state’s security problems are forever solved. On the contrary, in both Eritrea and Somaliland, the declaration of independence reintroduced the very same dilemmas that had plagued the regimes from which they had broken: how to deal with those within one’s own borders who do not want to be part of the seceding state or who otherwise dissent from the regime. Does one accommodate them, crush them, or allow their secession? The EPLF/PFDJ proved to be proactive—no doubt as a consequence of its own perilous birthing—and has suppressed all other claimants to power. The SNM, by contrast, was too frail to do so, but Somaliland’s inability to receive recognition has meant that dissenters keep quiet so as not to jeopardize Somaliland’s chances for recognition. Neither of these responses has served the broader interest of democracy. The case studies reveal that the strategies insurgents adopt and the rules they impose on others as regimes have their origins in their own particular demographic, historical, or power-related circumstances. Disputants are opportunists who respond to their environment and choose strategies in light of their security needs and their experience with their adversary. As circumstances change or new opportunities emerge, the strategies may also shift. Consequently, there are possibilities for the international community to shape these strategies by providing different incentives and sanctions. The MPLA’s willingness to

248

Civil War in African States

share power with UNITA was a consequence of considerable international pressure. Similarly, in Somaliland, the tendency towards a secessionist strategy was kept in check until 1991 by the awareness among the SNM leadership that recognition would not likely be granted by the international community. But external interveners also need to be aware of their own limitations in shaping these strategies and to consider whether their priorities produce genuine long-term peace and reconciliation. Global and regional influences do have bearing on the choice of strategy, but they are only one group of factors and not necessarily the most important ones. To the extent that strategies are products of a complicated and reasonably stable set of factors, disputants are unlikely to respond well to methods of conflict resolution which are incompatible with their perceived security interests. Successful conflict resolution requires, at the very least, that mediators pay close attention to the strategies of the disputants with which they are dealing. In many conflicts it has been the disputants themselves who have been most effective in bringing seemingly intractable conflicts to an end. In Ethiopia in 1991, Tigrayan and Eritrean rebels defeated the ruling Derg and delivered internal order to their countries despite the fact that, unlike their adversary, neither received significant military patronage from abroad. In Somalia, northern rebels delivered peace to their citizenry after years of repressive rule from Mogadishu and, notably, in the absence of the massive international intervention that characterized the south. Finally, in Angola, the government effectively ended the twenty-seven-year civil war, not when it participated in an internationally mediated peace process in 1991, but when it killed UNITA’s leader Jonas Savimbi in 2002. None of the strategies considered in this book offers a guarantee for peace. They all involve difficult decisions and sometimes compromises on peace, democratization, and security for others. Only one of these strategies—power-sharing—specifically promotes reconciliation among disputants. But as should be clear by now, actors emerging from civil wars are more concerned with—and more successful in promoting—their interests than they are reconciliation. Indeed, there is little doubt among most Tigrayans that they are better off without Mengistu, just as there is little doubt amongst Eritreans that they are better off without Ethiopia. In Somaliland, the development gains that have been achieved by northerners in recent years are also a compelling case for a political existence separate from Mogadishu.18 In Angola, the MPLA can hardly be blamed for losing patience with its mercurial adversary Jonas Savimbi. In none of these cases did power-sharing with

The Search Continues

249

their principal adversary prove to be a realistic option. Nor do the cases discussed here prove, as some political scientists have tried to do, that strategies such as partition entail risks and do not always lead to peaceful outcomes or that citizens prefer democratization over stability and security. Unless reconciliation can be made compatible with security, local actors will always set a higher value on the latter than the former. As the sub-title of this volume suggests, the search is primarily one for security rather than for peace or democracy. Victorious insurgents, however, like the regimes they replace, are themselves ultimately subject to failure, for they too must rely on force to suppress the challenges that will inevitably emerge within their extended territories. Observers continue to be perplexed by the fact that, despite pledges of democratization and inclusiveness, African regimes inevitably behave in self-interested ways and frequently resort to violence as a means of suppressing dissent within their borders. Scholars and politicians explain this violence by arguing that state-making in Africa is “still in its early stages,” suggesting that violent processes associated with state formation have been a necessary or inevitable condition for the subsequent achievement of order, prosperity and justice.19 Given time, the expectation is that these conflicts will run their course. Susan Rice, the former U.S. Assistant Secretary of State for Africa, argued that the “European Renaissance lasted over two centuries” and that “bloody, protracted war—and often plague—dominated at least half that period.” Africa’s troubles, she noted, may be the “‘birth pains’ of a new Africa.”20 The key for her was patience, not change, and recognition that, as was the case in Europe, development and state consolidation is a long, evolutionary process. But the history of state-making in Western Europe was considerably different from that in African states or, for that matter, elsewhere in the former colonized world. European states were not constrained by the same rigid state borders that characterize the contemporary state system; rather they were more fluid reflections of the power held at the center. While European state formation did indeed involve horrendous and prolonged violence, the political entities that emerged were also often indigenous outgrowths of local polities. What if the conditions in the African state system—most notably, this obsession with maintaining current borders—are such that peace and development cannot occur unless structural reforms are undertaken? In other words, what if territorial stasis, reliance on international recognition, and the ongoing discrepancies between borders and political power have themselves become barriers to peace and development? In that case, it is not merely a question of time before African states begin to resemble their Western

250

Civil War in African States

counterparts, as Susan Rice has suggested. If peace and development require the establishment of an effective state and for groups to be free of the endless security dilemmas that exist in so many African countries, then Africans and the international community may have to consider a third approach to conflict resolution. This involves fundamental changes in the nature of the state itself including reforms in the way borders are sustained and sovereignty is recognized. Many of the problems of African development arose because of the expectation that peaceful and coherent states would emerge from a supposedly more humane short-cut to state-building in which governments were inclusive, borders were guaranteed, and violence was condemned. In the end, however, territorial stasis creates and sustains security dilemmas that frequently result in more violence. The solutions to Africa’s troubles are not merely technical and the problems will not be solved by technocrats in the West. For every African conflict reaching its end, there is another one emerging. As conflicts in Angola and Ethiopia were concluding, new conflicts in Rwanda, Sierra Leone, and Liberia were brewing. And yet Africa’s map—the borders and the vast territories over which rulers endeavor to exert their authority—remains essentially unchanged. Much of the violence in Africa appears nihilistic because the states are themselves arbitrary, weak, or without purpose. In previous eras states emerged and governments were formed because they solved a fundamental problem—they offered protection. Contemporary states in Africa do not offer such protection and on the contrary are too often a threat to their own populations. Africa’s violence will finally subside when states, emerging in a variety of forms, serve the interests of Africans rather than the international community’s preference for the political status quo—that is, when states become purposeful. Otherwise Africans will be subject to recurring episodes of violence and insecurity. 1 Joseph C. Harsch, “This Quieter World,” Christian Science Monitor, December 29, 1988, p 10. 2 Cited in Thomas L. Friedman, “Clinton Inherits Conflicts That Don’t Follow Rules,” New York Times, December 13, 1992, sec. 4, p. 3. 3 See, for example, Tim Weiner, “An Evolving Idea for Liberia Envisions U.N. Trusteeship,” New York Times, August 17, 2003, sec. 1, p. 4; David Rieff, “A New Age of Liberal Imperialism?” World Policy Journal (summer 1999); William Pfaff, “A New Colonialism? Europe Must Go Back into Africa,” Foreign Affairs, January-February 1995, pp. 2-5.; Edward Luttwak, “Kofi’s Rule: Humanitarian Intervention and Neocolonialism,” National Interest (Winter 1999/2000), pp. 57-62; “Bringing Back the British,” Africa Confidential, 8 December 2000 (vol. 21, no. 4); Ali A. Mazrui, “Decaying Parts of Africa Need Benign Colonization,” International Herald Tribune, August 4,

The Search Continues

251

1994; Gerald B. Helman and Steven R. Ratner, “Saving Failed States,” Foreign Policy, #89, Winter 1992-93, pp. 3-20; Charles Krauthammer, “Trusteeship for Somalia: An Old—Colonial—Idea Whose Time Has Come Again,” Washington Post (October 9, 1992), p. A27; and Paul Johnson, “Colonialism’s Back—and Not a Moment Too Soon,” New York Times Magazine, April 18, 1993. 4 In at least one case, there was an outright retraction. See, for example, Sebastian Mallaby, “Liberal Imperialism, R.I.P.” Foreign Affairs, June 30, 2004. 5 This has, of course, most recently been noted in non-African contexts. Many Iraqis and Afghanis, for example, remain ambivalent about their current respective occupations. Many want American troops out but fear the repercussions of being left alone. See John F. Burns, “The Road Ahead May Be Even Tougher,” New York Times, March 7, 2004, sec. 4, p. 1. 6 See David E. Sanger, “Iraq Paradox: Cracking Down While Promoting Freedom,” New York Times, October 28, 2003. 7 On the former, see Stephen D. Krasner, “The Case for Shared Sovereignty,” Journal of Democracy, vol. 16, no. 1 (January 2005), pp. 69-83. 8 Gumisai Mutume, “World Should Learn from its Angola Failure,” Daily Mail and Guardian, April 5, 1999. 9 Alan J. Kuperman, “Rwanda in Retrospect,” Foreign Affairs (January/February, 2000), pp. 94-95. 10 Roland Paris, “Peacebuilding and the Limits of Liberal Internationalism,” International Security, vol. 22, no. 2 (Fall, 1997), p. 88. 11 For an extreme example of this tendency, see Danny Hoffman, “The Civilian Target in Sierra Leone and Liberia: Political Power, Military Strategy, and Humanitarian Intervention,” African Affairs, vol. 103, issue 411 (2004), pp. 211-226 12 Anonymous paper (forthcoming), p. 13. 13 Herbert Howe, Ambiguous Order: Military Forces in African States (Boulder: Lynne Rienner, 2001), especially chapter 3. 14 By the year 2000, when his regime began seizing white-owned farms, Robert Mugabe was portrayed, almost universally, as a tyrant. But in the early post independence years he was seen as a relatively magnanimous peacemaker. “Once again I pondered to myself over [Mugabe’s] maturity, reasonableness and sense of fair play” remarked Ian Smith, Rhodesia’s former Prime Minister, after a May 1980 meeting with Mugabe. Cited in Martin Meredith, Our Votes, Our Guns: Robert Mugabe and the Tragedy of Zimbabwe (New York: Public Affairs, 2002), p. 42. 15 Howe, Ambiguous Order, p. 82. 16 UNITA representative Jardo Muekalia. Interview with the author, September 17, 1999. 17 Waltz presents this as a lesson that should be clear to disputants, in his case states. But he adds that those who accumulate such power “have nevertheless been able to persuade themselves that skillful diplomacy and clever strategy would enable them to transcend the normal processes of balance of power politics.” See Kenneth N. Waltz, “War in Neorealist Theory,” in Robert I. Rotberg and Theodore K. Rabb eds., The Origin and Prevention of Major Wars (Cambridge; Cambridge University Press, 1989), 49.

252

Civil War in African States

18 Mark Bradbury and Ken Menkhaus, “Human Development in Somalia, 2001: An Overview” Paper presented to the 8th Congress of the Somali Studies International Association, Hargeisa (July 2001). 19 Samuel Makinda, Book Review of Africa in the New International Order: Rethinking State Sovereignty and Regional Security, Edmond J. Keller and Donald Rothchild eds., in Survival, vol. 39, no. 4 (1997-98), p. 188. 20 Susan E. Rice, Speech before the International Breakfast Club, African Medical and Research Foundation, October 23, 1998, Toronto, Canada.

Bibliography

“A Darling of the West Turns Nasty.” Economist, November 12-18th, 2005. “A Refusal to Become Black Russians or African Cubans.” New York Times, December 8 1976, p. A27. Aadland, Oyvind, and Kjetil Tronvoll. The Process of Democratization in EthiopiaCAn Expression of Popular Participation or Political Resistance, Human Rights Report: No 5. (Norwegian Institute of Human Rights, August 1995). Aalen, Lovise, and Kjetil Tronvoll. “The End of Democracy? Curtailing Political and Civil Rights in Ethiopia.” Review of African Political Economy 36, no. 120 (2009): 193-207. Abbay, Alemseged. Identity Jilted or Reimagining Identity. Red Sea Press, 1998. Abbay, Alemseged. “The Trans-Mareb Past in the Present.” Journal of Modern African Studies 35, no. 2 (1997): 321-334. Abbink, J. “Briefing: The Eritrean-Ethiopian Border Dispute.” African Affairs 97 (1998): 551-565. Abbink, J. “Discomfiture of Democracy? The 2005 Election Crisis in Ethiopia and Its Aftermath.” African Affairs 105, no. 419 (2006): 173-199. Academy for Peace and Development. A Self Portrait of Somaliland: Rebuilding from the Ruins. Hargeisa: Somaliland Centre for Peace and Development, 1999. Afoaku, Osita. “Congo’s Rebels: Their Origins, Motivations, and Strategies.” The African Stakes of the Congo War, edited by John F. Clark, 109-128. New York: Palgrave, Macmillan, 2002. Africa Contemporary Record. Annual Survey and Documents. London: Africa Research Limited. African Recorder, December 3-16, 1962, p. 345. Africa Recorder, November 20-December 3, 1969, p. 2391. Africa Watch. Angola: Violations of the Laws of War by Both Sides. New York: Human Rights Watch, 1989. Africa Watch. Evil Days: 30 Years of War and Famine in Ethiopia. New York: Human Rights Watch, 1991. Africa Watch. Somalia: A Government at War with its own People: Testimonies about the Killings and the Conflict in the North. New York: Human Rights Watch, January 1990. African Union. The Communique of the 139th Meeting of the Peace and Security Council of the African Union, 29 June 2008, Sharm el Sheikh, Egypt.

254

Civil War in African States

Ali, Taisier M., and Robert O. Matthews, eds. Durable Peace: Challenges for Peacebuilding in Africa. Toronto: University of Toronto Press, 2004. Amnesty International. “Eritrea.” Amnesty International Report, 2007. Amnesty International. “Eritrea: ‘You Have No Right to Ask’: Government Resists Scrutiny on Human Rights.” May 2004. “An Interview with TPLF.” Horn of Africa. vol. 4, no. 3 (1981), pp. 32-35. “Angola II: Talking Drums.” Africa Confidential 39, no. 10 (May 15, 1998). “Angola on the Way to War.” Globe and Mail, July 27, 1998. Angola Peace Monitor, vol. v, no. 1 (October 7, 1998). Angola Peace Monitor, vol. 2, no. 9 (1996). “Angola: Facing Up to Savimbi.” Africa Confidential 39, no. 23 (November 20, 1998). “Angola: Strings Attached.” Africa Confidential 39, no. 20 (October 9, 1998). “Angolan Rivals Meet To Hash Out Deal On Presidential Vote.” Christian Science Monitor, October 19, 1992. “Angola’s Elusive Oil Riches.” New York Times, (editorial) June 15, 2004, p. A22. “Angola’s Feelings Don’t Get in the Way of Profits.” New York Times, January 12, 1986, sec. 4, p. 3. “Angola’s President Reaffirms Commitment to Democracy; Calls on Savimbi to Take the Path to Peace.” Angola News Flash, May 11, 2001. “Angola Rebels Seize Oil Center and Old Base.” New York Times, January 21, 1993, p. A5 “Angola: Savimbi’s Last Stand.” Africa Confidential 39, no. 5 (March 6, 1998). Anstee, Margaret Joan. Orphan of the Cold War: The Inside Story of the Collapse of the Angolan Peace Process, 1992-93. New York: St. Martins’ Press, 1996. Anonymous. “Government Recognition in Somalia and Regional Political Stability in the Horn of Africa.” Journal of Modern African Studies 40, no. 2 (2002): 247-272. Associated Press. “Guerrilla Chief Predicts Victory Over Angola.” Washington Post, January 2, 1985, p. A24. Ayoob, Mohammed. “State Making, State Breaking, and State Failure.” Turbulent Peace: The Challenges of Managing International Conflict, edited by Chester A. Crocker, Fen Osler Hampson, and Pamela Aall. Washington: United States Institute of Peace 2001. Baker III, James. The Politics of Diplomacy. New York: G.P. Putnam and Sons, 1995. Battersby, John. “Angolan President Urges US Role to Help End Renewed Civil War.” Christian Science Monitor, January 25, 1993. Battersby, John. “Angolan Rebels’ Threat to Bolt Puts Democratic Vote in Jeopardy.” Christian Science Monitor, October 13, 1992. Battersby, John. “Angolan Rebels Win Concessions After War Gains.” Christian Science Monitor, April 22, 1993. Battersby, John. “Angolans Begin to Disarm Under Successful Cease-Fire.” Christian Science Monitor, July 3 1991. Battersby, John. “Namibia Success Raises Hopes for Angola Cease-Fire.” Christian Science Monitor, March 23, 1990. Battersby, John. “Violent Cleansing: Part of Angola War, Aid Workers Say.” Christian Science Monitor, February 3, 1993.

Bibliography

255

Baylis, John, and Steve Smith. Globalization of World Politics. 1st edition. New York: Oxford University Press, 1997. Bennett, John. “Tigray: Famine and National Resistance.” Review of African Political Economy 10, no.26 (July 1983): 94-102. Berdal, Mats, and David M. Malone, eds. Greed and Grievance: Economic Agendas in Civil Wars. Boulder: Lynne Rienner, 2000. Berhe, Aregawi. “Ethiopia: Success Story or State of Chaos?” Postmodern Insurgencies: Political Violence, Identity Formation and Peacemaking in Comparative Perspective, edited by Ronaldo Munck and Purnaka L. de Silva. Houndsmill and New York: Macmillan and St. Martin’s Press, 2000. Berhe, Aregawi. “The Origins of the Tigray People’s Liberation Front.” African Affairs 103, no. 413 (2004): 569-592. Betts, Richard K. “The Delusion of Impartial Intervention.” Foreign Affairs 73, no. 6 (November-December 1994): 20-33. Bevan, James. “The Myth of Madness: Cold Rationality and ‘Resource’ Plunder by the Lord’s Resistance Army.” Civil Wars 9, no. 4 (2007): 343-358. Biden Jr., Joseph R., and Leslie H. Gelb. “Unity Through Autonomy in Iraq.” New York Times, May 1, 2006. Biles, Peter. “Profile: Ethiopian Leader Meles Zenawi.” BBC News Online, August 10, 2005. Black, Carson. “Mengistu Courts West with Pledge of Democracy." The Times, June 28, 1990, p. 9. Bowley, Graham. “Declaring Something a Lot Like Independence.” New York Times, March 2, 2008, sec. 4, p. 1. Boyd, Charles, G. “Making Peace with the Guilty: The Truth About Bosnia.” Foreign Affairs 74, no. 5 (September-October 1995): 22-38. Bradbury, Mark. Somaliland. London: CIIR, 1997. Bradbury, Mark, and Adan Yusuf Abokor and Haroom Ahmed Yusuf. “Somaliland: Choosing Politics Over Violence.” Review of African Political Economy 30, no. 97 (2003): 455-478. Bradbury, Mark, and Ken Menkhaus. “Human Development in Somalia, 2001: An Overview.” Paper presented to the Eighth Congress of the Somali Studies International Association, Hargeisa, July 2001. Bratton, Michael. “Beyond the State: Civil Society and Associational Life in Africa.” World Politics 41, no. 3 (April 1989): 407-430. Breytenbach, W. “Cuito Cuanavale Revisited: Same Outcomes, Different Consequences.” Africa Insight 27, no. 1 (1997): 54-62. Bridges, Peter. “Safirka: Envoy to Somalia.” Michigan Quarterly Review 37, no.1 (Winter 1998): 1-21. Brigdland, Fred. Death in Africa. Cold Type: www.coldtype.net. 2004. Bridgland, Fred. “It Took a Rebel to Catch a Rebel.” Globe and Mail, April 3, 2002, p. A11. Bridgland, Fred. Jonas Savimbi: A Key to Africa. Edinburgh: Mainstream Publishing, 1986. “Bringing Back the British.” Africa Confidential 21, no. 4 (December 8, 2000). Brooke, James. “Angola Gives Hints of Willingness to Seek Political Solution to War.” New York Times, January 28, 1989, p. 4. Brooke, James. “Angola Says U.S. Uses Zaire Bases to Train Rebels.” New York Times, May 26, 1988, p. A10.

256

Civil War in African States

Brooke, James. “Angola Sees Gain in Free Namibia.” New York Times, March 21, 1990, p. A12. Brooke, James. “Angola’s Economic Struggles.” New York Times, January 31, 1986, p. D1. Brooke, James. “Cuba’s Strange Mission in Angola.” New York Times, February 1, 1987, sec. 6, p. 24. Brooks, David. AA Million Little Pieces,@ New York Times, June 5, 2007. Brown, D.J. Latham. “The Ethiopia-Somaliland Frontier Dispute.” International and Comparative Law Quarterly 5, no. 2 (1956): 245-264. Bryden, Matt. “New Hope for Somalia? The Building Block Approach.” Review of African Political Economy 26, no. 79 (March 1999): 134-140. Bryden, Matt. Somaliland and Peace in the Horn of Africa: A Situation Report and Analysis. United Nations Development Programme and UNEmergencies Unit for Ethiopia. November 1995. Bryden, Matt. “State Within a Failed State: Somaliland and the Challenge of International Recognition.” States Within States: Incipient Political Entities in the Post-Cold War Era. Paul Kingston and Ian S. Spears eds. New York: Palgrave, 2004. Buckoke, Andrew. “Tanks Back Bandit Warfare in Eritrean Struggle.” The Times, November 22, 1988, p. 13. Burns, John F. “The Road Ahead May Be Even Tougher.” New York Times, March 7, 2004, sec. 4, p. 1. Burton, Richard F. First Footsteps in East Africa; or, an Exploration of Harar. London: Longman, Brown, Green, and Longmans, 1856. Castagno, Al. “Somalia Goes Military.” Africa Report, February 1970, p. 26. Cater, Charles. “The Political Economy of Conflict and UN Intervention: Rethinking the Critical Cases of Africa.” In The Political Economy of Armed Conflict: Beyond Greed and Grievance, edited by Karen Ballentine and Jake Sherman. Boulder: Lynne Rienner, 2003. Chabal, Patrick. “Violence, Power and Rationality: A Political Analysis of Conflict in Contemporary Africa.” Is Violence Inevitable in Africa? Patrick Chabal, Ulf Engel, and Anna-Maria Gentili eds. Leiden: Brill, 2005. Chitunda, Jeremias K. “Angola: Blueprint for Peace.” (Letter to the editor, dated June 9, 1976) New York Times, June 22, 1976, p. 34. Chotiner, Isaac. “Appeasement’s Taint Is All in Hindsight.” New York Times, May 25, 2008. Clapham, Christopher. Transformation and Continuity in Revolutionary Ethiopia. Cambridge: Cambridge University Press, 1988. Cliffe, Lionel. “Dramatic Shifts in the Military Balance in the Horn: The 1984 Eritrean Offensive.” Review of African Political Economy 11, no. 30 (September 1984): 93-97. Cohen, Herman J. Intervening in Africa: Superpower Peacemaking in a Troubled Continent. Basingstoke: Macmillan, and New York: St. Martin’s Press, 2000. Collier, Paul. “Doing Well Out of War: An Economic Perspective.” Greed and Grievance: Economic Agendas in Civil Wars, edited by Mats Berdal and David M. Malone. Boulder: Lynne Rienner, 2000.

Bibliography

257

Collier, Paul. “Economic Causes of Civil Conflict and Their Implications for Policy.” Turbulent Peace: The Challenges of Managing International Conflict. Chester A. Crocker, Fen Osler Hampson, and Pamela Aall eds. Washington: United States Institute of Peace, 2001. “Colonial Office Report on the Somaliland Protectorate for the Years 1950 and 1951.” Congressional Record, Intelligence Authorization Act for Fiscal Year, 1990, House of Representatives October 12, 1989. Connell, Dan. “Against More Odds: The Second Siege of Eritrea.” Eritrean Studies Review 3, no. 2 (1999): 191-207. Connell, Dan. “The Birth of the Eritrean Nation.” Horn of Africa 3, no. 1 (January-March, 1980): 14-24. Connell, Dan. “Civil War Threatens in Eritrea.” Boston Globe, September 1, 1980 Connell, Dan. Taking on the Superpowers: Collected Articles on the Eritrean Revolution, 1976-1982. Trenton: Red Sea Press, 2003. Connell, Dan, and Frank Smyth. “Africa’s New Bloc.” Foreign Affairs 77, no. 2 (March/April 1998): 80-94. Constitution of the Federal Democratic Republic of Ethiopia Countess, Imani. “US Must Hold Savimbi Accountable, Support Angola Peace.” Christian Science Monitor, April 23, 1992. Cowell, Alan. Killing the Wizards: Wars of Power and Freedom from Zaire to South Africa. New York: Simon & Schuster, 1992. Crocker, Chester. “Intervention: Toward Best Practices and a Holistic View.” In Turbulent Peace, edited by Chester A. Crocker, Fen Osler Hampson, and Pamela Aall. Washington: United States Institute of Peace, 2001. Crocker, Chester. High Noon in Southern Africa: Making Peace in a Rough Neighborhood. New York: WW Norton, 1992. Crocker, Chester. “South Africa: Strategy for Change.” Foreign Affairs 59, no. 2 (Winter 1980-81): 323-351. Crossette, Barbara. “Singling Out Sierra Leone, U.N. Council Sets Gem Ban.” New York Times, July 6, 2000. “Crossing Borders: A Glance at S. Africa’s Record.” Christian Science Monitor, December 23, 1985. Crossley, Kenn. “Why Not to State-Build New Sudan.” In States Within States: Incipient Political Entities in the Post Cold War Era, edited by Paul Kingston and Ian S. Spears. New York: Palgrave, 2004. Daley, Suzanne. “Foes in Angola Still at Odds, Over Diamonds.” New York Times, September 15, 1995, p. 1. Darnton, John. “Eritrean Rebel Army Set for Decisive Test.” New York Times, Monday July 11, 1977, p. A1. David, Steven R. Choosing Sides: Alignment and Realignment in the Third World. Baltimore: Johns Hopkins University, 1991. David, Steven R. Third World Coups d'Etat and International Security. Baltimore: Johns Hopkins University, 1987. Davidson, Basil. “Somalia: Towards Socialism.” Race and Class 17, no. 1 (1975): 19-37. Davis, Nathaniel. “The Angolan Decision of 1975: A Personal Memoir.” Foreign Affairs 57, no. 1 (Fall 1978): 109-124. “Demobilising but Still Divided.” Economist, September 14, 1996.

258

Civil War in African States

Deisser, Immanuel. Paper presented at the 8th Congress of the Somali Studies International Society, Hargeisa Somaliland. July 11, 2001. “Disarming: Questions for Gerry Adams.” New York Times Magazine, August 14, 2005, p. 13. Drysdale, John. The Somali Dispute. New York: Praeger 1964. Drysdale, John. Stoics Without Pillows: A Way Forward for the Somalilands. London: Haan Publishers, 2000. Drysdale, John. Stoics Without Pillows: A Way Forward for the Somalilands. London: Haan Publishers, 1994. Duffield, Mark, and John Prendergast, “Liberation Politics and External Engagement in Ethiopia and Eritrea.” Horn of Africa Discussion Paper Series (April 1995): 10-11. Dugger, Celia W. “Zimbabwe Opposition Leader Insists Any Deal Give Him Real Power.” New York Times, August 17, 2008, Sec. 1, p. 6. Duke, Lynne. “Heavy Fighting in Angola Eclipses Peace Accord.” Washington Post, October 13, 1998, p. A12. Duncan, W. Raymond, and Carolyn McGiffert Ekedahl. Moscow and the Third World Under Gorbachev. Boulder: Westview, 1990. Dunn, Kevin C. “A Survival Guide to Kinshasa: Lessons of the Father Passed Down to the Son.” In The African Stakes of the Congo War, edited by John F. Clark. New York: Palgrave Macmillan, 2002. Elbadawi, Ibrahim, and Nicholas Sambanis. “Why Are There So Many Civil Wars in Africa? Understanding and Preventing Violent Conflict.” Journal of African Economies 9, no. 3 (2000): 244-269. Englebert, Pierre, and Rebecca Hummel. “Let’s Stick Together: Understanding Africa’s Secessionist Deficit.” African Affairs 104, no. 416 (2005): 399427. “EPLF-TPLF Joint Statement.” April 24, 1988. “EPRDF Landslide.” Africa Research Bulletin 31, no. 7 (August 21,1994). “Eritrea Votes to Become Africa's Newest Nation.” Africa Report, May-June 1993, p. 8. “Eritrean Opposition Groups Agree to Form Full Alliance.” Sudan Tribune, January 23, 2008. Eritrean People’s Liberation Front. “Self-Reliance ... Only Way to Victory.” Published in Review of African Political Economy no. 10 (SeptemberDecember 1977): 105-114. “Eritreans in Rome Report Rebels Still Block Ethiopians in Asmara.” New York Times, May 18, 1978, p. A3. “Eritreans Want Independence.” Wall Street Journal, January 28, 1987. “Eritrea’s Freedom Fighters.” Wall Street Journal, April 25, 1986. “Eritrea-Saudi Arabia: Tension Rises.” Africa Research Bulletin 30, no.2 (February 1-28, 1993). Erlanger, Steven. “Hamas’s Radical New Strategy: Taking Part in Elections.” New York Times, January 2, 2005. “Ethiopia: Arrests Starting to Thin Army Ranks.” Indian Ocean Newsletter, June 16, 2001. “Ethiopia Tikdem: Explained.” The Ethiopian Herald, November 1, 1974. “Ethiopia: US Peace Talks Go Ahead.” Africa Research Bulletin 26, no. 8 (September 15, 1989). Ethiopian Herald. May 21, 1960.

Bibliography

259

FBIS (Federal Broadcast Information Service—Daily Report). VIII, 7 April 1977, p. B1. FBIS. “Afewerki on Constitution, Separation of Powers,” FBIS, May 26, 1993, p. 6. FBIS. AAl-Ahram Interviews ELF Chairman Nasir,@ FBIS, March 23, 1977, pp. B2/B4. FBIS. AAngola: More on Savimbi=s Speech Marking UNITA=s Anniversary,@ FBIS, 18 March 1996, p. 10. FBIS. AAngola: Savimbi Meets Press Before Departing on Tour,@ FBISCAfrica, June 20, 1988, p. 3. FBIS. ABanti Warns Against Complacency Over Adowa Victory,@ FBISAfrica. Vol. VIII, 3 March 1976. FBIS. “Diaries of UNITA=s Jeremias Chitunda,” FBIS-AFR-93-018-S. FBIS. “ELF Spokesman Accuses Authorities of Genocide,” FBIS, 2 March 1976, p. B1. FBIS. “EPLF on Transitional Government, Independence,” FBIS, May 29, 1991, p. 8. FBIS. “Eritrean Leader Gives Address on Referendum Results,” FBIS, April 28, 1993, p. 4. FBIS. “Interview with Le Monde.” FBIS. June 16, 1978, p. B3. FBIS. “Initial Rebel Reaction to Mengistu Flight Noted,” Federal Broadcast Information Service FBISCDaily Report (Sub-Saharan Africa), May 21, 1991, p. 7. FBIS. “Le Monde Interviews EPLF Spokesman on Issues,” FBIS, April 8, 1977, p. B3. FBIS. “Mengistu Addresses Nation on Eritrean Issue,” FBISCDaily Report (Sub-Saharan Africa), June 12, 1978, pp. B1-5. FBIS. “Mengistu Scores [sic] Imperialists in Anniversary Speech,” FBIS, vol. VIII, no. 42, 3 March 1977, p. B1. FBIS. “Mubarak-Afewerki New Conference Reported,” FBIS, 26 May 1993, p. 5. FBIS. “President Speaks at Independence Celebration,” FBIS Daily Report, 25 May 1993, p. 6. FBIS. “President Mengistu Addresses May Day Rally,” FBIS, vol. VIII, 2 May 1977, p. B3. FBIS. “Reaction to USC Presidential Appointment Noted: Senior SNM Member Comments,” FBIS-Daily Report (Sub-Saharan Africa), January 31, 1991, pp. 8-9. FBIS. “Sabi Views Status of Eritrean Liberation Struggle,” FBIS, March 18, 1977, p. B3 FBIS. “Seminar on Eritrea Opens in Addis Ababa,” FBIS 19 June 1978, p. B3. FBIS. “Siad Barre Addresses Mogadishu Rally,” FBIS: Middle East and Africa, vol. 10, no. 119 (21 June 1982), p. R1. FBIS. “Siad Barre Says Northern Rebels >Wiped Out,=” FBIS-Daily Report (Africa) June 20, 1988, p. 1. FBIS. “Somaliland=s President Tur Interviewed,” Federal Broadcast Information Service FBIS-Daily Report (Sub-Saharan Africa), December 7, 1992, pp. 11-12. FBIS. “Three Rebel Movements View Future Developments,” FBIS, May 30, 1991, p. 3.

260

Civil War in African States

Fisher, Ian. “Eritrea Gives Up Vital Town to Ethiopia; Peace Hopes Rise.” New York Times, May 26, 2000, p. A3. Final Statement on the Carter Center Observation of the Ethiopia 2005 National Elections. (September 15, 2005). Final Report of the European Union Election Observation Mission (2005). Foley, Michael W., and Donald Rothchild. “African States and the Politics of Inclusive Coalitions.” In The Precarious Balance: State and Society in Africa. Edited by Donald Rothchild and Naomi Chazen. Boulder: Westview, 1988. Fowler, Robert. Testimony Before the UN Security Council, January 18, 2000. Freedom House. Freedom in the World: The Annual Survey of Political Rights & Civil Liberties, 1993-1994. Washington: Freedom House, 1994. Freedom House. Freedom in the World: The Annual Survey of Political Rights & Civil Liberties, 1999-2000. Washington: Freedom House, 2000. French, Howard W. “From Old Files, a New Story of U.S. Role in Angola War.” New York Times, March 31, 2002, p. 4. Friedman, Thomas L. “Clinton Inherits Conflicts That Don’t Follow Rules.” New York Times, December 13, 1992, sec. 4, p. 3. Fukuyama, Francis. “Women and the Evolution of World Politics.” Foreign Affairs 77, no. 5 (September/October, 1998): 24-40. Gagnon, V. P. “Ethnic Nationalism and International Conflict: The Case of Serbia.” International Security 19, no. 3 (Winter 1994/95): 130-166. Garcetti, Eric, and Janet Gruber. “The Post-war Nation: Rethinking the Triple Transition in Eritrea.”Chapter 12 In Regeneration of War-Torn Societies. Edited by Michael Pugh. New York: St. Martin’s, 2000. Garnett, John. “The Causes of War and the Conditions for Peace.” In Strategy in the Contemporary World: An Introduction to Strategic Studies, edited by John Baylis, James Wirtz, Eliot Cohen, and Colin S. Gray. Oxford: Oxford University Press, 2002. George, Edward. The Cuban Intervention in Angola, 1965-1991: From Che Guevara to Cuito Cuanavale. London: Frank Cass, 2005. Gersony, Robert. Why Somalis Flee: Synthesis of Accounts of Conflict Experience in Northern Somalia by Somali Refugees, Displaced Persons and Others. Washington: Bureau for Refugee Programs: Department of State, 1989. Gettleman, Jeffrey. “Eritreans Deny American Accusations of Terrorist Ties.” New York Times, September 18, 2007, p. A17. Gettleman, Jeffrey. “A Red Sea Conflict’s Buffer Zone Is Some Rocks, and a Few Inches.” New York Times, May 25, 2008, p. 6. Ghalib, Jama Mohamed. The Cost of Dictatorship: The Somali Experience. New York: Lilian Barber Press, 1995. Gilkes, Patrick. Two Wasted Years: The Republic of Somaliland, 1991-1993. Bedfordshire: Save the Children Fund UK, 1993. Gilkes, Patrick, and Martin Plaut. “Conflict in the Horn: Why Eritrea and Ethiopia are at War.” Chatham House Briefing Paper, New Series no. 1, March 1999. Giniger, Henry. “For Independent Angola, A Great Threat of Strife.” New York Times, January 16, 1975, p. 12. Giorgis, Dawit Wolde. Red Tears: War, Famine and Revolution in Ethiopia. Trenton: Red Sea Press, 1989.

Bibliography

261

Glaser, Charles L. “Realists as Optimists: Cooperation as Self-Help.” International Security 19, no. 3 (Winter 1994-95): 50-90. Goddard, Ian. “Interview with Ethiopia’s Foreign Minister Col. Dr. Feleke Gedle-Giorgis.” Horn of Africa (April-June 1978). Greenfield, Richard. “Somalia’s Letter of Death.” New African, July 1987, pp. 14-16. “Guerrilla Chief Predicts Victory Over Angola.” Washington Post, January 2, 1985, p. A24. Gurr, Ted Robert. “Minorities and Nationalists: Managing Ethnopolitical Conflict in the New Century.” In Turbulent Peace: The Challenges of Managing International Conflict, edited by Chester A. Crocker, Fen Osler Hampson, and Pamela Aall. Washington: United States Institute of Peace, 2001. Hammond, Jenny. Fire From the Ashes: A Chronicle of the Revolution in Tigray, Ethiopia, 1975-1991. Lawrenceville: Red Sea Press, 1999. Harbom, Lotta, and Peter Wallensteen. “Patterns of Major Armed Conflicts.” In SIPRI Yearbook 2006. Oxford: Oxford University Press, 2006. Hare, Paul. Angola’s Last Best Chance for Peace: An Insider’s Account of the Peace Process. Washington: United States Institute of Peace, 1998. Harsch, Joseph C. “This Quieter World.” Christian Science Monitor, December 29, 1988. “Harsh Ethiopian Methods of Subduing Somalis.” The Times, October 25, 1956, p. 8. Hartzell, Caroline, and Matthew Hoddie. “Power Sharing in Peace Settlements: Initiating the Transition from Civil War.” In Sustainable Peace: Power and Democracy After Civil Wars, edited by Philip G. Roeder and Donald Rothchild. Ithaca: Cornell University Press, 2005. Hatch, Orrin. “Keep Ethiopia Part of the Reagan Doctrine.” The Wall Street Journal, April 4 1986. Helman, Gerald B. and Steven R. Ratner. “Saving Failed States.” Foreign Policy 89 (Winter 1992-93): 3-20. Henze, Paul B. Eritrea’s War: Confrontation, International Response, Outcome, Prospects. Addis Ababa: Shama Books, 2001. Henze, Paul. B. Ethiopia in Mengistu’s Final Years: Until the Last Bullet. Addis Ababa: Shama Books, 2007. Henze, Paul B. Layers of Time. New York: Palgrave, 2000. Henze, Paul B. “A Political Success Story.” Journal of Democracy 9, no. 4 (October 1998): 40-54. Henze, Paul B. “The Tigre People’s Liberation Front: Conversations with Meles Zenawi.” Mimeo (April 1990). Herbst, Jeffrey. “Economic Incentives, Natural Resources and Conflict in Africa.” Journal of African Economies 9, no. 3 (2000): 270-294. Herbst, Jeffrey. States and Power in Africa: Comparative Lessons in Authority and Control. Princeton: Princeton University Press, 2000. Hess, Robert L. “The ‘Mad Mullah’ and Northern Somalia.” Journal of African History 5, no.3 (1964): 415-433. Heywood, Linda. Contested Power in Angola: 1840s to the Present. Rochester: University of Rochester Press, 2000. Hodges, Tony. Angola: Anatomy of an Oil State. Oxford: James Currey, 2004.

262

Civil War in African States

Hodges, Tony. Angola From Afro-Stalinism to Petro-Diamond Capitalism. Oxford: James Currey, 2001. Hoffman, Danny. “The Civilian Target in Sierra Leone and Liberia: Political Power, Military Strategy, and Humanitarian Intervention.” African Affairs 103, no. 411 (2004): 211-226. Horowitz, Donald. “The Cracked Foundations of the Right to Secede.” Journal of Democracy 14, no. 2 (2003): 5-17. Horowitz, Donald. Ethnic Groups in Conflict. Berkeley: University of California Press, 1985. Howe, Herbert. Ambiguous Order: Military Forces in African States. Boulder: Lynne Rienner, 2001. Human Rights Watch. Angola: Arms Trade and Violations of the Laws of War Since the 1992 Elections. New York: Human Rights Watch, 1994. Human Rights Watch-Africa. Angola Civilians Devastated by 15 Year War. New York: Human Rights Watch, 1991. International Crisis Group. “Somaliland: Democratisation and Its Discontents.” Africa Report No. 66, July 2003. International Crisis Group. “Sierra Leone: Time for a New Political and Military Strategy.” Africa Report No. 28, 2001. International Crisis Group. “Somaliland: Democratisation and its Discontents.” Africa Report No. 66, 2003. “Interview: UNITA Sets Conditions for Angola Talks.” Reuters, December 16 1998. “Interview with Jonas Savimbi.” Africa Report, July-August 1988, p. 49. “Interview with Venancio de Moura.” Africa Report, July-August 1988, pp 4648. IRIN. “Angola: Election Free and Fair, Sort Of.” Integrated Regional Information Network, September 9, 2008. IRIN. “Ethiopia: New Army Chief Named.” IRIN-Horn of Africa Weekly Round-up, May-June 2001. IRIN. “Interview with Prime Minister Meles Zenawi.” Integrated Regional Information Network, October 29, 2003, www.irinnews.org/report. aspx?reportid=46954. IRIN. “IRIN Interview with Major General Muhammad Gani.” Integrated Regional Information Network, July, 2000. IRIN. “Somalia: Are ‘Building Blocks’ the Solution?” Integrated Regional Information Network, July 19, 1999. Issa-Salwe, Abdisalam M. The Collapse of the Somali State: The Impact of the Colonial Legacy. London: Haan Publishers, 1994. Iyob, Ruth. Eritrean Struggle for Independence: Domination, Resistance, Nationalism, 1941-1993. Cambridge: Cambridge University Press, 1995. Iyob, Ruth. “The Ethiopian-Eritrean Conflict: Diasporic Versus Hegemonic States in the Horn of Africa, 1991-2000.” Journal of Modern African Studies 38, no. 4 (2000): 659-682. Jackson, Robert H. Quasi-States: Sovereignty, International Relations and the Third World. Cambridge: Cambridge University, 1990. Jackson, Robert H., and Carl G. Rosberg. Personal Rule in Black Africa: Prince, Autocrat, Prophet, Tyrant. Berkeley: University of California Press, 1982.

Bibliography

263

Jackson, Robert H., and Carl G. Rosberg. “Sovereignty and Underdevelopment: Juridical Statehood in the African Crisis.” Journal of Modern African Studies 24, no. 1 (1986): 1-31. Jardine, D.J. The Mad Mullah of Somaliland. London: Herbert Jenkins, 1923. Jervis, Robert, and Jack Snyder. “Civil War and the Security Dilemma.” Civil Wars, Insecurity, and Intervention. Edited by Barbara F. Walter and Jack Snyder. New York: Columbia University Press, 1999. Job, Brian L. The Insecurity Dilemma: National Security of Third World States. Boulder: Lynne Rienner Publishers, 1992. Johnson, Michael, and Trish Johnson. “Eritrea: The National Question and the Logic of Protracted Struggle.” African Affairs 80, no. 318 (1981): 181-195. Johnson, Paul. “Colonialism’s Back: And Not a Moment Too Soon.” New York Times Magazine, April 18, 1993. Johnson, Phyllis, and David Martin, eds. AAngola: The Struggle Continues@ Frontline Southern Africa: Destructive Engagement. New York: Four Walls Eight Windows, 1988. Jones, Bruce D. “Civil War, the Peace Process, and Genocide in Rwanda.” Civil Wars in Africa: Roots and Resolution. Edited by Taisier M. Ali and Robert O. Matthews. Montreal and Kingston: McGill-Queen’s, 1999. Kahneman, Daniel, and Jonathan Renshon. “Why Hawks Win.” Foreign Policy, January/February 2007, pp. 34-38. Kakwenzire, Patrick. “Resistance, Revenue and Development in Northern Somalia, 1905-1939.” International Journal of African Historical Studies 19, no. 4 (1986): 659-677. Kaplan, Robert D. “History’s Cauldron.” Atlantic Monthly, June 1991, pp. 93104. Kaplan, Robert D. Balkan Ghosts: A Journey Through History. New York: St Martin’s Press, 1993. Kaplan, Robert D. “A Reader’s Guide to the Balkans.” New York Times Book Review, April 18, 1993, pp. 1, 30-32. Kaplan, Robert D. Surrender or Starve: Travels in Ethiopia, Sudan, Somalia and Eritrea New York: Vintage, 2003. Kaufmann, Chaim. “See No Evil: Why America Doesn’t Stop Genocide.” Review of: Samantha Powers, “A Problem from Hell”: America and the Age of Genocide. New York: Basic Books, 2002. In Foreign Affairs, 81, no. 4 (2002): 142-149. Keen, David. “The Economic Functions of Violence.” Adelphi Paper 320. New York: Oxford University Press, 1998. Keller, Bill. “After Big Gains, Angolan Rebel Offensive is Halted.” New York Times, September 24, 1993, p. A3. Keller, Bill. “New Path to Peace in Angola: Clamp Rebels and Hold Tight.” New York Times, November 6, 1994, p. 1. Keller, Edmond. Revolutionary Ethiopia. Bloomington: Indian University Press, 1988. Kirkpatrick, Jeane. “Deal Making, Soviet Style.” Washington Post, October 3, 1988, p. A11. Kissinger, Henry. Does America Need a Foreign Policy: Toward a Diplomacy for the Twenty-First Century. London: Free Press, 2002.

264

Civil War in African States

Kissinger, Henry. Years of Renewal. New York: Simon and Schuster, 1999. Kittermaster, Harold B. “British Somaliland.” Journal of the Royal African Society 27, no. 108 (1928): 329-337. Kittermaster, Harold B. “The Development of the Somalis.” Journal of the Royal African Society 31, no. 124 (1932): 234-244. Korn, David. Ethiopia, the United States and the Soviet Union. London: Croom Helm, 1986. Krasner, Stephen D. “The Case for Shared Sovereignty.” Journal of Democracy 16, no. 1 (January 2005): 69-83. Krauthammer, Charles. “Trusteeship for Somalia: An Old ‘Colonial’ Idea Whose Time has Come Again.” Washington Post, October 9, 1992, p. A27 Kr_ka, Vladimír. “Peacekeeping in Angola (UNAVEM I and II).” International Peacekeeping 4, no. 1 (Spring 1997): 75-97. Kumar, Radha. “The Troubled History of Partition.” Foreign Affairs 76, no. 1 (January/February 1997): 22-34. Kuperman, Alan J. “Rwanda in Retrospect.” Foreign Affairs 79, no.1 (January/February 2000): 94-118. Laitin, David. “The War in the Ogaden: Implications for Siyaad’s Rôle in Somali History.” Journal of Modern African Studies 17, no. 1 (March 1979): 95-115. Laitin, David, and Said S. Samatar. Somalia: Nation in Search of a State. Boulder: Westview, 1987. Landesman, Peter. “A Woman’s Work.” New York Times Magazine, September 15, 2002. Larémont, Ricardo René, ed. Borders, Nationalism and the African State. Boulder: Lynne Rienner Publishers, 2005. Laurence, Patrick. “S. Africa Stalls on Namibian Independence.” Christian Science Monitor, April 19, 1985. Lautenbach, Dale, and Stephen Taylor. “Savimbi Puts Aside Weapons of War.” The Times, May 31, 1991, p. 9. Lefort, René. Ethiopia: An Heretical Revolution? Oxford: Zed Press, 1981. Legum, Colin. The Battlefronts of Southern Africa. New York: Africana Publishing Company, 1988. Lemarchand, René. “Consociationalism and Power-Sharing in Africa: Rwanda, Burundi, and the Democratic Republic of the Congo.” African Affairs 106, no. 422 (2007): 1-20. Lewis, Arthur. Politics in West Africa. London: Oxford University Press 1965. Lewis, Flora. “The New Climate.” New York Times, January 26, 1989, sec. 4, p. 25. Lewis, Ioan M. A Modern History of the Somali: Nation and State in the Horn of Africa, 4th Edition. Oxford: James Currey, 2002. Lewis, Ioan M. Blood and Bone: The Call of Kinship in Somali Society. Lawrenceville: Red Sea Press, 1994. Lewis, Ioan .M. “The Ogaden and the Fragility of Somali Segmentary Nationalism.” African Affairs 88, no. 353 (October 1989): 573-579. Lewis, Neil A. “U.S. Plan to Be ‘Midwife’ To a New Rule in Ethiopia.” New York Times, May 26, 1991, sec. 1, p. 10. Lewis, Stephen. “I Too Accuse Him of Crimes Against Humanity.” Globe and Mail, July 28, 2008, A13.

Bibliography

265

Lijphart, Arend. “Consociational Democracy.” World Politics 21, no. 2 (January 1969): 207-225. Lijphart, Arend. Democracy in Plural Societies: A Comparative Exploration. New Haven: Yale University Press, 1977. Lindijer, Koert. “Why the SNM Spurns USC Overtures.” New African, May 1991, p. 12. Lulseged, Eyayu. “Some Institutional Problems of the Ethiopian Army: 19741991.”New Trends in Ethiopian Studies: Papers of the 12th International Conference of Ethiopian Studies, Volume 1: Humanities and Human Resources, edited by Harold G. Marcus. Lawrenceville: Red Sea Press, 1994. Lustick, Ian. “Stability in Deeply Divided Societies: Consociationalism versus Control.” World Politics 31, no. 3. (April 1979): 325-344. Luttwak, Edward. “Give War a Chance.” Foreign Affairs 78, no. 4 (July/August 1999):36-44. Luttwak, Edward. “Kofi’s Rule: Humanitarian Intervention and Neocolonialism.” National Interest (Winter 1999/2000): 57-62. Luttwak, Edward. “Steering into Civil War.” Globe and Mail, February 14, 2005, A17. Lyons, Terrence. “Closing the Transition: the May 1995 Elections in Ethiopia.” Journal of Modern African Studies 34, no. 1 (March 1996): 121-142. Maier, Karl. “Angolan Civil War: Air Power is Key.” Christian Science Monitor, August 19, 1986. Maier, Karl. “Angolan Civil War Keeps Potential Economic Powerhouse Poor.” Christian Science Monitor, September 12, 1986. Maier, Karl. “The Military Stalemate.” Africa Report, May-June, 1988, p. 35. Makinda, Samuel. ABook Review of Africa in the New International Order: Rethinking State Sovereignty and Regional Security.@ Survival. vol. 39, no. 4 (1997-98): 88. Malaquias, Assis. “Diamonds are a Guerrilla’s Best Friend: the Impact of Illicit Wealth on Insurgency Strategy.” Third World Quarterly 22, no. 3 (June 2001): 311-325 Malaquias, Assis. “Ethnicity and Conflict in Angola: Prospects for Reconciliation.” In Angola’s War Economy: The Role of Oil and Diamonds, edited by Jakkie Cilliers and Christian Dietrich. Pretoria: Institute for Security Studies, 2000. Malcolm, Noel. “Seeing Ghosts.” National Interest, Summer 1993, pp. 83-88. Mallaby, Sebastian. “Liberal Imperialism, R.I.P.” Foreign Affairs, June 30, 2004. Marcus, Harold G. A History of Ethiopia. Berkeley: University of California Press, 1994. Marcus, Harold G. “Breakfast Meeting with Prime Minister Meles.” October 20, 1995. Available at: www.hartford-hwp.com/archives/33/008.html. Mariam, Mesfin Wolde. “The Background of the Ethio-Somalian Boundary Dispute.” Journal of Modern African Studies 2, no. 2 (July 1964): 189-219. Marion, Georges. “Le chef de l’armée angolaise estime que la guerre pourrait reprendre.” Le Monde, February 16, 1995. Markakis John. National and Class Conflict in the Horn of Africa. Cambridge: Cambridge University Press, 1987.

266

Civil War in African States

Mazrui, Ali A. “Decaying Parts of Africa Need Benign Colonization.” International Herald Tribune, August 4, 1994. McFaul, Michael. “Rethinking the Reagan Doctrine in Angola.” International Security 14, no. 3 (Winter 1989-1990): 99-135. McNeil, Donald G. “Bombing Won in Kosovo, Africa Is a Tougher Case.” New York Times, July 25, 1999, sec. 4, p. 16. Mearsheimer, John J. “Reckless States and Realism.” International Relations 23, no. 2 (2009): 241-256. Mearsheimer, John J. The Tragedy of Great Power Politics. New York: Norton, 2001. Mehler, Andreas, and Dennis M. Tull. “The Hidden Cost of Power-Sharing: Reproducing Insurgent Violence in Africa.” African Affairs 104, no. 416 (July 2005): 375-398. Meldrum, Andrew. “At War with South Africa.” Africa Report, JanuaryFebruary 1987, p. 29. Memorandum on the Conduct of the European Union Election Observation Mission (EU-EOM) in Ethiopia. “Mengistu Defends Red Terror.” BBC News Online, December 28, 1999. Menkhaus, Ken. “International Peacebuilding and the Dynamics of Local and Natioal Reconciliation in Somalia.” Learning from Somalia: The Lessons of Armed Humanitarian Intervention, edited by Walter Clarke and Jeffrey Herbst. Boulder: Westview, 1997. Menkhaus, Ken. “Somalia: In the Crosshairs of the War on Terrorism.” Current History 101, no. 655 (May 2002): 210-218. Menkhaus, Ken. “The Crisis in Somalia: A Tragedy in Five Acts.” African Affairs 106, no. 204 (2007):357-390. Melake, Tekeste. “The Battle of Shire (February 1989): A Turning Point in the Protracted War in Ethiopia.” In New Trends in Ethiopian Studies:Papers of the 12th International Conference of Ethiopian Studies Volume 1, edited by Harold G. Marcus. Lawrenceville: Red Sea Press, 1994. Meldrum, Andrew. “Hungry to Vote.” Africa Report, November-December 1992, p. 27. Meredith, Martin. Our Votes, Our Guns: Robert Mugabe and the Tragedy of Zimbabwe. New York: Public Affairs, 2002. Milkias, Paulos. “The Great Purge and Ideological Paradox in Contemporary Ethiopian Politics.” Horn of Africa vol. 19, nos. 1-4 (2001): 1-99. Ministry of Foreign Affairs. Briefing Paper: The Case for Somaliland’s Recognition as an Independent State. Hargeisa: August 2002. “MPLA Emerges Victorious, Recognition by MPLA.” Africa Diary, March 1117, 1976. Maphai, Vincent T. “The New South Africa: A Season for Power-Sharing.” In Democratization in Africa, edited by Larry Diamond and Marc F. Plattner. Baltimore: Johns Hopkins, 1999. Moffett III, George D. “US Debates Support for Angolan Rebels vs. Broader African Policy.” Christian Science Monitor, February 11, 1986. Morello, Ted. “Angolan Envoy Warns Against US Aid to Savimbi.” Christian Science Monitor, January 31, 1986. Morna, Collen Lowe. “Peace Slow to Follow Angola Accord.” Christian Science Monitor, February 17, 1989.

Bibliography

267

Morrison, J. Stephen. “Mr. Savimbi Goes to Washington.” Africa Report, September-October 1988, p. 56. Mseteka, Buchizya. “UK’s Hain Says Angola Rebel Boss Must be Removed.” Reuters, January 31, 2000. Mufson, Steve. “Angolan Rebel Leader Courts U.S. Aid.” Wall Street Journal, December 17, 1981, p. 29. Museveni, Yoweri. “The Progress of the People’s War.” In Yoweri Museveni: Selected Articles of the Uganda Resistance W a r . Kampala: NRM Publications, 1985. Mutua, Makau wa. “The Regionalization Controversy.” Africa Report, September-October 1993, p. 31. Mutume, Gumisai. “World Should Learn from its Angola Failure.” Daily Mail and Guardian, April 5, 1999. Nadarajah, Suthaharan, and Dhananjayan Sriskandarajah. “Liberation Struggle or Terrorism? The Politics of Naming the LTTE.” Third World Quarterly 26, no. 1 (2005): 87-100. Negash, Tekeste, and Kjetil Tronvoll. Brothers at War: Making Sense of the Eritrean-Ethiopian War. Oxford and Athens: James Currey and Ohio University Press, 2000. Negash, Tekeste. Italian Colonialism in Eritrea, 1882-1941: Policies, Praxis and Impact. Uppsala: Uppsala University, 1987. Neuberger, Benyamin. National Self-Determination in Post-Colonial Africa. Boulder: Lynne Rienner, 1986. Newbury, Catharine. “Ethnicity and the Politics of History in Rwanda.” Africa Today 45, no. 1 (January-March1998): 7-24. Noble, Kenneth B. “A New Crisis Engulfs Angola As the Rebels Make Big Gains.” New York Times, January 29, 1993, p. 1. Noble, Kenneth B. “Angola President Says His Forces Are Rearming After Rebel Gains.” New York Times, February 2, 1993, p. A6. Noble, Kenneth B. “Angolan Rebels Rebound.” New York Times, April 13, 1993. Noble, Kenneth B. “Election Makes Skeptics of Angolans.” New York Times, June 28 1992, p. 8. Noble, Kenneth B. “Savimbi, Trailing, Hints at New War.” New York Times, October 4, 1992, p. 15. Nolen, Stephanie. “Angola’s Long March to Peaceful Elections.” Globe and Mail, September 5, 2008, p. A17. Nolen, Stephanie. “With a Booming Economy Benefiting the Elites, Regular Angolans Find Little Solace in Democracy.” Globe and Mail, September 6, 2008, p. A1. Novicki, Margaret A. “Against All Odds.” Africa Report, January-February 1985. Novicki, Margaret A. “Forging a Bipartisan Policy, Interview with Herman J. Cohen.” Africa Report, September-October 1989. Novicki, Margaret A. “Interview: A Conversation with Ricardo Alarcón.” Africa Report, November-December 1988. Novicki, Margaret A. “Creating Conditions for Peace: Interview with Pedro de Castro Van-Dunem Loy.” Africa Report, March-April 1989. Novicki, Margaret A. “Interview with President José Eduardo dos Santos.” Africa Report, January-February, 1986.

268

Civil War in African States

“October Evolution.” Africa Confidential, 42, no. 21, October 26, 2001. “Official Government Statement” Republic of Somaliland, April 23, 1997. Ottaway, David B. “Angola Offers Removal Of All Cuban Troops.” Washington Post, February 14, 1988, p. A32. Ottaway, David B. “Ethiopia’s Government A Shambles.” Washington Post, December 1, 1974, p. A1. Ottaway, David B. “Fate of Angolan Guerrillas Unresolved by Accord.” Washington Post, July 21, 1988, p. A15. Ottaway, David B, and Marina Ottaway. Ethiopia: Empire in Revolution. New York: Africana Publishing Company, 1978. Paris, Roland. “Peacebuilding and the Limits of Liberal Internationalism.” International Security 22, no. 2 (Fall 1997): 54-89. Pateman, Roy. “The People the World Forgot: The United Nations and Eritrea.” Horn of Africa vol. XII and XIV, nos 3 and 4, 1 and 2 (1991). Pawson, Lara. “End of War in Sight: Angolan General.” BBC News, November 16, 1999. Payton, Gary. “The Somali Coup of 1969: The Case for Soviet Complicity.” Journal of Modern African Studies 18, no. 3 (September 1980): 493-508. “Peace at Last.” African Confidential, vol. 41, no. 24, December 8, 2000, p. 8. Pear, Robert. “C.I.A. Crash Called Problem in Angola.” New York Times, December 1, 1989, p. A9. Perham, Margery. The Government of Ethiopia. London: Faber and Faber, 1969. Perlez, Jane. “On the Ethiopian Front, Rebel Confidence Rises.” New York Times, February 14, 1990, p. A15. Perlez, Jane. “Prisoner No. 14279: Forlorn Pawn in Ethiopia's Long and Ruinous Civil War.” New York Times, February 19, 1990, p. A6. Perlez, Jane. “Pupil of G.I.'s Does Well, as a Rebel.” New York Times, June 27, 1991, p. A4. Perlez, Jane. “U.S. and Italy Evacuating Foreigners in Somalia.” New York Times, January 6, 1991, p. A3. Perlez, Jane. “U.S. Says Airlifts Fail Somali Needy.” New York Times, July 31, 1992, p. A9. Petterson, Donald. “Ethiopia Abandoned? An American Perspective.” International Affairs 62, no. 4 (Autumn 1986): 627-645. Pfaff, William. “A New Colonialism? Europe Must Go Back into Africa.” Foreign Affairs 74, no.1 (January-February 1995): 2-6. Piper, Laurence. “Nationalism Without a Nation: The Rise and Fall of Zulu Nationalism in South Africa’s Transition to Democracy, 1975-99.” Nations and Nationalism 8, no. 1 (January 2002), 73-94. Plaut, Martin. “Yemen and Eritrea: Friends Once More?” Review of African Political Economy 25, no. 78 (December 1998): 659-661. Plaut, Martin. “High Stakes in Ethiopia Stand-Off.” BBC News Online, June 23, 2005. Pool, David. From Guerrillas to Government: The Eritrean People’s Liberation Front. Oxford: James Currey, 2001. Polgreen, Lydia. “Justifying A Costly War In Sri Lanka.” New York Times, July 19, 2009, 6. Poluha, Eva. The 1995 Ethiopian Elections Viewed from the Grassroots, A Report to SIDA. Stockholm: SIDA (1995).

Bibliography

269

Posen, Barry R. “The Security Dilemma and Ethnic Conflict.” Survival 35, no.1 (Spring 1993): 24-47. Posen, Barry R. “The War for Kosovo: Serbia’s Political-Military Strategy.” International Security 24, no. 4 (Spring 2000): 39-84. Potgieter, Jakkie. “Taking Aid from the Devil Himself: UNITA Support Structures.” In Angola’s War Economy: The Role of Oil and Diamonds, edited by Jakkie Cilliers and Christian Dietrich. Pretoria: Institute for Security Studies, 2000. Press, Robert M. “Somali Rebels Seek Peace After Ousting Ruler.” Christian Science Monitor, January 30, 1991. “President-Elect Bush Makes First Foreign Policy Commitment.” Washington Post, January 12 1990, p. A1. Press Conference by Eugenio Manuvakola, Former UNITA Secretary General Regarding His Escape From UNITA,@ ReliefWeb, August 28, 1997. “Proceedings of the Eighth Ordinary IGAD Summit of Heads of State and Government.” November 23, 2000, Khartoum, Republic of Sudan. Prunier, Gérard. “A Candid View of the Somali National Movement.: Horn of Africa 13, nos. 3 and 4, 14, nos. 1 and 2 (1990-1991):107-120. Prunier, Gérard. Darfur: The Ambiguous Genocide. Ithaca: Cornell University Press, 2005. Report of the Secretary-General of the United Nations on the Situation in Somalia, 16 September 1999. “Report on the Visit to Addis Ababa,” Somali Public Record Office, Reference CO830/26 XC17470. Reporters Without Borders. World Press Freedom Ranking 2008. Reporters Without Borders: 2008. “Reviving the Somali Peace Process: Perspectives and Prospects in the PostArta Period.” Paper presented to the 8th Congress of the Somali Studies International Association, Hargeisa, July 2001. Reyntjens, Filip. “Rwanda, Ten Years On: From Genocide to Dictatorship.” African Affairs 103, no. 411 (2004): 177-210. Rice, Susan E. “Speech before the International Breakfast Club, African Medical and Research Foundation, October 23, 1998, Toronto, Canada. Richter, Paul, and Peter Wallsten. “Bush Dismisses the Idea of Partitioning Iraq.” Los Angeles Times, August 16, 2006, p. A12. Rieff, David. “A New Age of Liberal Imperialism?” World Policy Journal 16, no.2 (Summer 1999): 1-10. Roeder, Philip G., and Donald Rothchild. “Dilemmas of State-Building in Divided Societies.” In Sustainable Peace: Power and Democracy After Civil Wars, edited by Philip G. Roeder and Donald Rothchild. Ithaca: Cornell University Press, 2005. Rotberg, Robert I. “Angola Sting.” Christian Science Monitor, April 10, 1986. Samatar, Ahmed I. Socialist Somalia: Rhetoric and Reality. London: Zed, 1988. Samatar, Abdi Ismail, and Ahmed I Samatar. “The International Crisis Group Report on >Somaliland= in Horn of Africa,” vol. XXIV, (2006). Samatar, Ibrahim Megag. “Light at the End of the Tunnel: Some Reflections on the Struggle of the Somali National Movement.” In Mending Rips in the Sky: Options for Somali Communities in the 21st Century. Edited by Hussein M. Adam and Richard Ford. Lawrenceville: Red Sea Press, 1997.

270

Civil War in African States

Sambanis, Nicholas. “Partition as a Solution to Ethnic War: An Empirical Critique of the Theoretical Literature.” World Politics 52, no. 4 (July 2000): 437-483. Sanger, David E. “Iraq Paradox: Cracking Down While Promoting Freedom.” New York Times, October 28, 2003. Savimbi, Jonas. AThe Coming Winds of Democracy,@ (Heritage Foundation, 1989). Savimbi, Jonas. “I Have No Secret Agenda.” New York Times, October 30, 1989, p. 19. Schuster, Lynda. “Angola’s Rebels Decry US Peace Plan.” Christian Science Monitor, August 22, 1988. Schuster, Lynda. “For Angola Rebels Business is Booming.” Christian Science Monitor, August 26, 1988. Selassie, Bereket Habte. “Political Leadership in Crisis: The Ethiopian Case.” Horn of Africa vol. 3, no. 1 (1980). Sengupta, Somini. “Warring Militias in Congo Test U.N. Enforcement Role.” New York Times, April 11, 2004, p. 8. Sisk, Timothy D. Power Sharing and International Mediation in Ethnic Conflicts. Washington: United States Institute of Peace, 1996. Shinn, David H. “The Horn of Africa: How Does Somaliland Fit?” Paper Presented at the Discussion Seminar Introducing Somaliland, Umea, Sweden, March 8, 2003. Smith, Patrick. “Angola: Free and Fair Elections.” Review of African Political Economy no. 55 (November 1992): 101-106. Snyder, Jack. “One World, Rival Theories.” Foreign Policy no. 145 (November-December 2004): 52-62. “Somalia: After the Coup.” Africa Confidential, no. 21, October 24, 1969, pp. 78. “Somalia: After the Elections.” Africa Confidential, no. 11, May 23 1969, pp. 34. “Somali Army and Police Seize Power.” New York Times, October 22, 1969, p. 3. “Somalia Military Policies.” Africa Confidential, 11, no. 16, August 7, 1970, p. 8. “Somalia: Post Reshuffle.” Africa Confidential, 23, no. 6, March 17 1982. Sparks, Allister. “A View of Rome from the Provinces.” The Wilson Quarterly, 25, no. 2 (Spring 2001). Spears, Ian S. “Angola’s Elusive Peace.” International Journal 54, no. 4 (Autumn 1999): 562-581. Spencer, John. Ethiopia at Bay: A Personal Account of the Haile Selassie Years. Algonac: Reference Publications, 1984. “Statement by the Central Committee of the Tigray People’s Liberation Front (TPLF) on the Occasion of the 24th Anniversary of the Armed Struggle in Eritrea.” Review of African Political Economy, no. 35 (May 1986): 85-94. Stockholm International Peace Research Institute (SIPRI), Yearbook(s): Armaments, Disarmaments and International Security. Stockwell, John. In Search of Enemies: A CIA Story. New York: W.W. Norton, 1978. Stoessinger, John G. Henry Kissinger: The Anguish of Power. New York: W.W. Norton, 1976.

Bibliography

271

Tadesse, Kiflu. The Generation Part II: Ethiopia Transformation and Conflict: The History of the Ethiopian People’s Revolutionary Party. Lanham: University Press of America, 1998. Tadesse, Medhane, and John Young. “TPLF: Reform or Decline.” Review of African Political Economy 30, no. 97 (September 2003): 389-403. TPLF. “On Our Differences with the EPLF.” People’s Voice, 1986. “The EPLF and its Relations with Democratic Movements in Ethiopia.” Review of African Political Economy no. 35 (May 1986): 85-92. The Reporter (Addis Abba), vol. 1, no. 2 September 9, 1996. Thurston, Raymond. “Détente in the Horn.” Africa Report, February 1969. Thurton, E.R. “The Impact of Mohammed Abdille Hassan in the East Africa Protectorate.” Journal of African History 10, no. 4 (1969): 641-657. Touval, Saadia. Somali Nationalism: International Politics and the Drive for Unity in the Horn of Africa. Cambridge: Harvard University Press, 1963. Traub, James. “Taunting the Bear.” New York Times, August 10, 2008, Sec. 4, p. 1. Trevaskis, G.K.N. Eritrea: A Colony in Transition: 1941-52. London: Oxford University Press, 1960. Tronvoll, Kjetil. “The Eritrean Referendum: Peasant Voices.” Eritrean Studies Review 1, no. 1 (Spring 1996): 23-67. Tuchman, Barbara. The Guns of August. New York: Dell, 1962. Turner, Thomas. “Angola’s role in the Congo War.” In The African Stakes of the Congo War, edited by John F. Clark. New York: Palgrave-Macmillan, 2002. “Two Main Eritrean Independence Groups to Merge.” New York Times, April 29, 1978, p. 3. “Two Views of Angola’s Problems.” Christian Science Monitor, July 1, 1988. UNDP. Human Development Report: Somalia (New York: 2001). “UNITA Suffers Key Defections.” Angolan Mission Observer, December 1999. United Nations. The United Nations and the Independence of Eritrea -- Blue Book Series. New York: United Nations, 1996. United Nations Sanctions Committee. Report of the Panel of Experts on Violations of Security Council Sanctions Against UNITA, March 10, 2000. United Nations. Official Records of the General Assembly (Volume 1), ThirtySecond Session Plenary Meetings, 27th meeting, 10 October 1977. United Nations. Report to the United Nations on the Situation in Somalia, September 16,1999. United States General Accounting Office (GAO) Somalia: Observations Regarding the Northern Conflict and Resulting Conditions (Washington: May 1989). “U.S. Aid to Savimbi: Bush Pledges to Keep the Money Coming.” Africa Report, March-April 1989, p. 6. “U.S. Shows Its Hand in Angola Plane Crash.” Africa Report, NovemberDecember 1989, p. 8. Van Evera, Stephen. “Hypotheses on Nationalism and War.” International Security 18, no. 4 (Spring 1994): 5-39. Vaughan, Sarah. “The Addis Ababa Transitional Conference of July 1991: Its Origins, History, and Significance,” Occasional Papers No. 51, Centre of African Studies, Edinburgh University, 1994.

272

Civil War in African States

Vaughan, Sarah. “Historical Perspectives on the Ethio-Eritrean Conflict.” Paper presented to the conference convened by the Law Forum for Peace, Addis Ababa, December 9, 1998. Walter, Barbara F. “The Critical Barrier to Civil War Settlement.” International Organization 51, no. 3 (1997): 335-364. Waltz, Kenneth N. “War in Neorealist Theory.” In The Origin and Prevention of Major Wars, edited by Robert I. Rotberg and Theodore K. Rabb. Cambridge: Cambridge University Press, 1989. Walz, Jay. “Haile Selassie Calls on Big Four To Share Results of Their Talks.” New York Times, January 31, 1960, p. 1. Wayne, E. A. “African Leaders Want US Mediation in Angola, But No Rebel Aid.” Christian Science Monitor, April 18, 1988. Wayne, E. A. “‘Big Prize’ for Southwestern Africa.” Christian Science Monitor, December 22, 1988. Weiner, Tim. “An Evolving Idea for Liberia Envisions U.N. Trusteeship.” New York Times, August 17, 2003, sec. 1, p. 4. White Book About Peace Process, volume 1. Luanda: Republic de Angola, 1995. Wolpe, Howard. “More U.S. Aid For Savimbi.” New York Times, June 10, 1989, p. 27. Wong, Edward. “Iraq’s Curse: A Thirst for a Final Crushing Victory.” New York Times, June 3, 2007, sec. 4, p. 3. World Bank. Somalia: Crisis in Public Expenditure Management, Report No. 8727-SO (March 8 1991). Young, John. “Ethiopians Destroy Their Own Plane.” New African, July 1988, p. 17. Young, John. “Governance in Post-Cold War Ethiopia.” Paper Presented to the Africa Society Conference, Edmonton, Alberta, February 26-28 1999. Young, John. Peasant Revolution in Ethiopia: The Tigray People’s Liberation Front, 1975-1991. Cambridge: Cambridge University Press, 1997. Zacher, Mark W. “The Territorial Integrity Norm: International Boundaries and the Use of Force.” International Organization 55, no. 2 (Spring 2001): 215250. Zartman, I. William. “Introduction: Posing the Problem of State Collapse.” In Collapsed States: The Disintegration and Restoration of Legitimate Authority, edited by I. William Zartman. Boulder: Lynne Rienner, 1995. Zenawi, Meles. “Speech by Meles Zenawi at the opening of the July Conference.” July 1, 1991 (mimeo). Zewde, Bahru, A History of Modern Ethiopia, 1855-1991. Addis Ababa: Addis Ababa University Press, 2002. Zewde, Bahru. “The Red Terror in Ethiopia: Context and Consequences.” The Ethiopian Red Terror Trials: Transitional Justice Challenged, edited by Kjetil Tronvoll, Charles Schaefer, and Girmahcew Aneme Alemu. Oxford and Athens: James Currey, and Ohio University Press, 2009.

Index Borama conference, 155, 158 Bridgland, Fred, 4, 189 Britain: actions against Tigray; administration of Ethiopia, 43; in Egypt, 42; Suez Canal, 42; in Somalia, 117, 118-124, 127, 141, 142, 162. See also British Somaliland, Ethiopia, Somalia British Somaliland, 119-121, 126128; independence, 129-131 Bush, George, 194, 195 Chitunda, Jeremias, 184-185, 187, 204, 205 Clans: Somalia, 116-118; Darod, 116; Dir, 116; Isaaq, 116, 120, 130, 137, 139, 140, 141, 142, 143, 145, 150, 152; Hawiya, 116, 130, 138, 141, 142; Majerteen, 138, 141, 142; Rahanwein, 116 Clark Amendment, 185, 187, 190 Clientelism, 6 Clinton, Bill, 241 Cohen, Herman, 41, 66, 76-77, 152, 195, 198, 201, 216 Cold War, 73, 179, 180, 187, 190, 195, 199, 226, 242 Colonialism: 249; in Angola, 180181; in Eritrea, 42-44; Eritrean view of, 43-44, 58, 94; Tigrayan view of, 62; in Somalia, 118-124, 125, 141, 142; borders in Somaliland, 161 Communism, in Angola, 180, 183, 187, 188 Conquest, See Domination Constitutive Act of the African Union, 19, 28 Cooperation, 9; international community’s view of, 10-13, 28, 96; role of memory with, 23; TPLF and EPLF, 70; 97, 98, 143, 146; in Angola, 190, 199, 202, 207, 217, 223. See also Integration, Power-Sharing

Addis Ababa, 62, Amhara domination, 94; 1972 Peace Agreement, 19; Selassie in, 49; US supplies, 54; Soviet interest, 55, 137; challenge from TPLF, 63, 69, 74, 75, 77, 84, 94; EPRP, 67; EPRDF, 77, 84, 88; nationalists in, 88; violence in, 92; EPLF, 69, 70, 84, 94; 1993 UN conference, 158 Afeworki, Isaias, 57, 84-87 Africa Watch, 12, 71, 140-141, 143 African Union, 28, 161. See also Organization of African Unity Aidid, General Mohamed Farah, 148, 153, 156, 158 Alan ‘As (Red Flag), 155, 156, 157 Ali, Abdirahman Ahmed “Tour”, 150, 151, 155, 156 Aman, General Michael Andom, 5051 Amhara, 46, 51, 67, 68, 79 Anarchy, 242-245 Angola, 179-240; Angola: coalition government, 200; diamonds, 209, 211, 213, 216, 219; economy, 192; 1992 election, 200-203, 204; 2008 election, 222-223; independence, 183-184; mineral wealth, 186; post-colonial movement, 180; support from South Africa, 183, 184, 186, 189, 190 Angolan Armed Forces (FAA), 210 Anstee, Margaret, 6, 202, 205, 225 Arab League, 134 Asmara, 43, 47, 48, 74, 88-91 Axum, 42, 62, See also Tigray Badme, 88-89 Baker, James, 195, 197, 198 Barre, Siad, See Siad Barre Beye, Blondin, 208, 210 Bicesse Accords, 197, 198, 199, 200, 207, 224

273

274

Civil War in African States

Crocker, Chester, 187, 191 Cuba: in Angola, 183, 184, 188, 189, 191, 193; in Ethiopia, 55 Delany, Martin R., 17 Democracy, 78, 80, 242, 247; in Angola, 180, 199, 201, 225, 226; in Eritrea, 85, 86; in Ethiopia, 8083, 95-96; proportional representation, 132; in Somalia, 131-132, 134, in Somaliland, 159-161 Democratic Front for the Salvation of Somalia (DFSS), 144, 149. See also Somali Salvation Front, Somali Salvation Democratic Front Democratic Republic of Congo (formerly Zaire), 152 Derg: attack on Hausien, 71; collapse of, 77; desire to destroy government, 52; domination strategy in Eritrea, 56; fear of Arab states, 58; move to Soviet Union, 53; negotiations with ELF-RC, 245; rise of, 50-53; weakening of, 71-74 Dervish, 120, 141, 142 Dimbleby, Jonathan, 49 Djibouti, 118, 119, 136, 150, 153 Domination, 9, 13-17, 41, 246; Derg’s strategy of, 50-53, 61; EPLF’s use of, 61-62; TPLF’s strategy of, 66, 76, 77-78, 94; MPLA’s use of, 190, 193, 204, 220; UNITA’s use of, 203, 206207, 216 Dos Santos, Jose Eduardo, 189-190, 195, 196, 199, 200, 201, 203, 207, 221, 222 EDU. See Ethiopian Democratic Union Egal, Mohamed Haji Ibrahim, 116, 131, 132, 133, 137, 150, 156, 157, 159, 160, 161 Elections: in Angola, 180-184, 188, 189, 192, 196-203, 204-208, 212, 213, 217, 219, 222, 224, 226; in Ethiopia, 79, 80-83, 92; in Eritrea, 86-87; in Somalia, 132, 133, 148, 149, 151, 159-164

ELF. See Eritrean Liberation Front Englebert, Pierre, 18, 21 EPLF. See Eritrean People’s Liberation Front EPRDF. See Ethiopian Peoples’ Revolutionary Democratic Front EPRP. See Ethiopian People’s Revolutionary Party Eritrea: Asmara, 43; colonization, 4243, 44; connection between Ethiopia and Axum, 42, 44; Christian population, 48; defiance of international opinion, 93; democracy in, 86, 87, 93; independence, 44-45, 80, 84-87, 91; division of insurgency, 57; Federal Act (1950), 46; Independence Bloc, 45; Italian invasions, 43; Muslim population, 48; secessionist strategy, 69-70, 94; war with Ethiopia, 87-91 Eritrean Liberation Front (ELF), 48; early view of Derg, 50; suspicion of Christians, 59 Eritrean People’s Liberation Front (EPLF), 57; disputes with neighbors, 88; post-Derg 84;. rise to power, 61-62; secession, 6971, 76-77, 246. See also Peoples Front for Democracy and Justice Estoril Accords, 197 Ethiopia: Amharic, 47; assistance to SNM, 143; claim on Somalia, 125; control by British, 43; defeat of, 75-77, democracy in, 79-83; Derg, the rise of, 50; dispute with Somalia, 119, 121-124; Federal Act (1950), 46; Harar, 139; historical connection to Eritrea, 42-43, 45; impediment to Somali independence, 130; invasion by Italy, 43; invasion of Ogaden, 131; Soviet patronage, 53-56; Unionist Party, 45; US response to dissolution of 1950 federation, 47; war with Eritrea, 87-91 Ethiopian Democratic Union (EDU), 66,

Index 275

Ethiopian People’s Revolutionary Party (EPRP), 67; antisecessionism, 67; view on Eritrea, 68; Ethiopian Peoples’ Revolutionary Democratic Front (EPRDF), approach to unity, 88; creation, 74; governance, 77-83, 92; public criticism, 91-92 Ethnicity, importance to Angolan politics, 181 Exogenous conditions, 26-28 FAA. See Angolan Armed Forces Factors influencing strategy, 20; Balance of Power, 20-22; History, Memory and Precedent, 22-23; Internal Attributes, 23 FAPLA. See People’s Armed Forces for the Liberation of Angola Federal Act between Ethiopia and Eritrea (1950), 46Federal Act of 1952 (Eritrea), 84 FNLA. See National Front for the Liberation of Angola France, colony (Djibouti), 119 French Somaliland, 118, 133, 136. See Djibouti Garhajis, 156-157 Ghalib, Jama Mohamed, 129, 134, 135, 136, 141, 142 Ghebreab, Yemane, 44, 83 Government of Unity and National Reconciliation (GURN), 210, 212 Guella, Ismael, 153 GURN. See Government of Unity and National Reconciliation Guulwadayaal Militia, 138 Haber Awal, 155, 156, 157 Haber Jelo, 157 Haber Yunis, 156, 157 Haile Selassi, Emperor, 23; claim on Somalia, 125-126, 127; exile in England, 43; mistrust of independent Eritrea, 45, 46; opulent lifestyle, 49; overthrow of, 49; restoration to throne, 123 Hangaash, 138 Harar, 139 Hargeisa Group, 143, 163

Hargeisa, 143, 144, 153-158, 160, 163, 165 Hassan, Sayid Mohammed Abdille, 120; “Mad Mullah”, 120, 141 Haud, the, ceding to Ethiopia, 121123, 163; struggle for, 124, 127, 139; unifying effect of, 128-129 Hawiye clan, 116, 130, 138, 141, 142 Horowitz, Donald, 18 Huambo, 184, 205, 206, 208, 209 Hummel, Rebecca, 18, 21 IGAD. See Intergovernmental Authority on Development Integration, 4, 9, 10-13, 24, 25, 225, 245; in Angola, 179, 190, 198, 207, 209, 211, 224, 225; in Eritrea, 45, 60, 69-71, 95; in Ethiopia, 45, 74-75; in Somaliland, 116, 128-131, 137, 162. See also Power-Sharing Intergovernmental Authority on Development (IGAD), 159 Iraq, 15 Isaaq, 116, 120, 130, 137, 139, 140, 141, 142, 143, 145, 150, 152. See also Somali National Movement Italian Somalia, 119, 121, 123, 128; independence, 129-131 Italy: Battle at Adwa, 44, 121; Eritrean view of colonialism, 44; Mussolini, 43; in Somalia, 118, 121, 123; Treaty of Wich’ale, 43, 62; Ivory Coast, 245 Jamba, 186, 193 Kahin, Dahir Rayale, 161, 164 Kenya, 133, 135, 136 King, Martin Luther, 241 Kissinger, Henry, 5, 8, 15, 30-31, 183 League of Nations: pledge of security for Ethiopia, 43. See also United Nations Lewis, Stephen, 8 Luanda, 180, 204; power-grab in, 181; control of, 184, 200 Lusaka Protocol, 208, 209, 210, 213, 214, 215, 217, 220-222, 224, 244, 246 Majerteen clan, 138, 141, 142 Malaquias, Assis, 204, 216

276

Civil War in African States

Maphai, Vincent, 241 Marahan-Ogadeni-Dulbahante (MOD), 141, 142, 145 Marxism: EPRP, 67; in Eritrea, 59; Mengistu’s use of, 55, 73; in Tigray, 63; in Angola, 186, 190, 197, 198 Mavinga, 191, 195 Menelik II, 43, 62, 119, 122 Mengistu, Haile Mariam, 12, 28, 41; agreement with Siad Barre, 144; alliance with Soviet Union, 5356, 73; assistance to SNM, 143; attitude to Eritrea, 50; belief in violence 51, 71, 73, 75; deposal of Selassie, 52; Marxism, 55, 73; overthrow, 73, 75-77, 242; Zimbabwean asylum, 77 Mobutu Sese Seko, 8, 28, 195, 219, 245 MOD. See Marahan-OgadeniDulbahante Mogadishu: dispute with Ethiopia 55; 77, 115, 117, 132, 135, 143, 145, 148-155, 158, 159, 160, 161, 163, 165, 246, 247 MPLA. See Popular Movement for the Liberation of Angola Mugabe, Robert, 11, 245 Mussolini: 43; treaties with Menelik II, 62 Namibia, 191, 192, 193, 223, 225, 243 National Front for the Liberation of Angola (FNLA), 180; help from China, Zaire, 182; power-sharing, 182-183, 185; weakening, 186 National Security Service (NSS), 138 National Union for the Total Independence of Angola (UNITA), 6, 24; cooperation as strategy, 180, 187; diamonds, 209, 211, 213, 216, 219; domination as strategy, 203; growth, 182, 186, 187; guerilla attacks, 189, 194; power-sharing, 183-184, 189, 195, 210, 212, 247, 248; rally for, 200; secession, 246 Neto, Angostinho, 180, 184, 185, 186 Neuberger, Benyamin, 241

New York Accords, 191, 192, 194, 199 OAU. See Organization of African Unity Ogaden, 122, 124; Barre’s invasion, 131, 136-137, 139, 163; disputes with northern Somalis, 125-126, 129; refugees, 139, 142; support for Ethiopian regime, 126; benefit from Barre’s regime, 140 OLF. See Oroma Liberation Front OPDO. See Oromo People’s Democratic Organization Organization of African Unity (OAU), 133, 161, 186 Oroma Liberation Front (OLF), 74, 78-79, 93 Oromo People’s Democratic Organization (OPDO), 74 Ovimbundu, 180, 182, 203, 204, 226 People’s Armed Forces for the Liberation of Angola (FAPLA), 185, 189, 193 Peoples Democratic Organizations (PDOs), 74 Peoples Front for Democracy and Justice (PFDJ), Badme dispute, 88, 91; contemporary public opinion, 92; creation, 85; governance, 86-87. See also Eritrean People’s Liberation Front PFDJ. See Peoples Front for Democracy and Justice Popular Movement for the Liberation of Angola (MPLA): 6, 23, 28, 29, control of Luanda, 183-186; cooperative strategy, 194; creation, 180; discussions with UNITA, 194; domination strategy, 190, 193, 204, 214, 216; elections, 246; loss of Huambo, 206, 207; recapture of Huambo, 208; power-sharing, 180,182-183, 195, 208-210, 247, 248; US recognition, 207 Portugal: Alvor Agreement, 181-182, colonial rule in Angola, 181, Lisbon coup, 181; post-colonial involvement, 184

Index 277

Posen, Barry, 3, 21, 26 Power-Sharing, 9, 10-13, 225, 247, 248; answer to security predicament, 10; difficulties in managing, 11; Mobuto, 195; psychological problems with, 11; TPLF’s strategy, 69-71, 74, 246; between UNITA and MPLA, 180, 183, 184, 189, 195, 194, 196, 198, 199-200, 209, 212, 224, 225, 249; See also Integration Realism, 2 Republic of Somaliland, 150. Resources: motivation for secession, 24; Resources: part of security, 30, use of secession in acquiring, 24, in Angola, 206, 209, 211, 213, 216, 219 Roberto, Holden, 180, 182 Rwanda, 3, 11, 16, 244, 245 Sabbe, Osman Saleh, 57, 59 SADC. See Southern African Development Community Samakuva, Isaias, 222 Savimbi, Jonas: 30, 180, 182, 196, 199; austerity, 211, 212; coalition government, 198, 202; death, 222, 226, 248; domination strategy, 203-208, 211, 214; 1992 election, 201-203, 204; human rights abuses, 200; integrative strategy, 211; isolation from UNITA, 216-218; Mobuto’s support, 195, 219; power-sharing, 183-184, 189, 210, 211-212, 248; political campaign, 201; political objectives, 181, 184, 186, 188, 189, 196; post-Lusaka, 210, 211; UNITA-Renovada, 218-219; with Washington, 188, 193-194; witch burning, 200 Scientific socialism, 134 Secession, 9, 17-20, 246; Eritrea, 4247, 94, 247; effect of power balance, 20-22; importance of international recognition, 17-18, 27; means of acquiring resources, 24; strategy for self-defense, 2425, 94; role of demography, 25;

role of geography, 25; role of history (Eritrea), 94; role of history (Ethiopia), 94; SNM’s strategy of, 128-131, 137, 139; of Somaliland, 150-156, 160, 164165; TPLF’s strategy of, 42, 64; Security concerns: 1, 3, 4, 8-11, 1720, 22, 28-30, 242, 243, 245-250; Angola, 189, 191, 192, 199, 211, 226; Ethiopia and Eritrea, 41-47, 64, 76, 77, 84, 91, 93, 96; in Somalia, 115, 116, 122, 127, 136, 138, 144, 163, 164 Seko, Mobuto, Sese, See Mobutu Separation, See Secession Seyoum, Ras Mengesha, 66 Siad Barre, 12, 120, 124, 134, 135, 161-163; aftermath of collapse, 149-150, 151; agreement with Mengistu, 144; attacks against northern Somalia, 144-145; clanism, 141; collapse of regime, 77, 115; 1969 coup, 118, 138; embezzlement, 140; dislike of Isaaq, 141; invasion of Ogaden, 135-137, 139; nationalism, 141; ruthlessness, 138 Sierra Leone, 16 Six Day War, 133 SNM. See Somali National Movement Snyder, Jack, 7, 15, 27 Somali National Movement (SNM), 142-148, 149-150, 155-156, 162, 247, 248 Somali Patriotic Movement, (SPM), 145 Somali Republic: Act of Union (1960), 129, 151 Somali Salvation Democratic Front (SSDF), 149 Somali Salvation Front (SSF), 138139, 144 Somalia: Britain in, 117-124, 133, 141; civil war, 156-157; clan structure, 116-118, 132; collapse of Barre regime, 77; conflict with Ethiopia, 121-124; humanitarian aid to, 140; independence, 124128, 129-131, 150-154, 158;

278

Civil War in African States

integration strategy, 128-131; invasion of Ogaden, 55, 135-137; military rule, 133-135; physical environment, 117; proportional representation, 132; secession strategy, 150-156; Somali Republic, 129, 151; Republic of Somaliland, 151; Soviet patronage, 135; Soviet withdrawal, 137 Somaliland, 29, 244; beel system, 160; British Protectorate, 119, 161; Borama conference, 155; civil war, 156-158; colonial borders, 161; geographical isolation, 163; governance, 154155; independence, 150-154; multi-party elections, 159, 161; Quasi-secession, 150-156, 164, 248 South Africa: in Angola, 183, 184, 186, 189, 190 South African-controlled South West Africa (Namibia) (SWAPO), 191, 192 Southern African Development Community (SADC), 220 Sovereignty, 17, 250; in Eritrea, 80, 88; Ethiopian claims in the Haud, 121-124; Somalian desire for, 127 Soviet Union: in Ethiopia, 53-56, 72, 73; in Somalia, 135-136, 139; in Angola, 182, 183, 188

Soyo, 206 SPM. S e e Somali Patriotic Movement Sri Lanka, 7, 15 SSF. See Somali Salvation Front Strategy, influencing factors: demographic distribution, 2526; geography, 25; greed, 24; history, 22-23; memory, 22-23, precedent, 22; resources, 24 Sudan, 19, 25 Suez Canal, 42, 119

Supreme Revolutionary Council, 133, 135 SWAPO. See South Africancontrolled South West Africa (Namibia) Tewodros II, 42 TGE. See Transitional Government of Ethiopia Tigray, 62-65; division of, 62; Eritrean fear of, 88; view of Derg, 63. See also Tigrayan People’s Liberation Front Tigrayan Liberation Front (TLF), 66 Tigrayan People’s Liberation Front (TPLF), 20; Badme dispute, 9091; beginnings, 63, 65; challenge from EPRP, 67, challenge from EDU, 66-67; challenge by TLF, 66; domination as strategy, 76, 77-78; goal of secession, 64, 69; power-sharing, 246; relationship with EPLF, 68, 69-71; as ruling power, 79-81; treatment of prisoners, 72 TLF. See Tigrayan Liberation Front TNG. See Transitional National Government TPLF. See Tigrayan People’s Liberation Front Transitional Government of Ethiopia (TGE), 78, 79 Transitional National Government (TNG), 160 Tsadkan, Gebre, 90-91 Tsvangirai, Morgan, 11 UDUB. See United People’s Democratic Party UNDP. See United Nations Development Program Union of Islamic Courts, 93 UNITA, See National Union for the Total Independence of Angola UNITAF. See United Task Force United Nations Development Program (UNDP), 164 United Nations Operations in Somalia (UNOSOM), 153 United Nations: Addis Ababa conference, 158; in Angola, 190,

Index 279

191, 209, 214; Commission for Eritrea, 44; 1950 EthiopianEritrean decision, 46; Kofi Annan, 153, 159; MONUC, 244; response to dissolution of 1950 federation, 47; in Somaliland 158-159; UNITAF, 153; UNOSOM, 153 United People’s Democratic Party, 161 United Somali Congress (USC), 148, 156 United Somali Front (USF), 150 United States: in Angola, 183, 185, 191, 207; Clark Amendment, 185, 187, 190. See also Washington United Task Force (UNITAF), 153 UNOSOM. See United Nations Operations in Somalia USC. See United Somali Congress USF. See United Somali Front Walter, Barbara F., 15, 18 Washington: aid to Somalia, 140; interest in Ethiopia, 47, 54-55, 90; response to Mengistu, 54-55, 76; response to TPLF, 66; response to EPLF, 66, 77, response to MPLA and UNITA, 183, 187, 189; assistance to UNITA, 188, 190, 194-195; in Angola, 190, 191, 193, 194, 207, 209, 215 Western Somali Liberation Front (WSLF), 136, 142, 144 Wimhurst, David, 212. 213. 214. 215 World Bank, 140 Woyane, 23, 64 WSLF. See Western Somali Liberation Front Yohannes IV, 62, 66 Zaire: help to Angolan rebels, 182, 219; minerals, 24; US support, 28. See also Democratic Republic of Congo (DRC) Zenawi, Meles, 41, 67, 72-73, 78, 89, 90 Zimbabwe: security interests, 11; asylum for Mengistu, 77

About the Book

How do disputants in Africa's civil wars—rebel movements, ethnic groups, state leaders—find security in the midst of anarchic situations? Why do some rebel movements pursue a secessionist agenda while others seek to overthrow the existing government? Under what circumstances will insurgents agree to share power? Proposing answers to these questions, Ian Spears offers a fresh perspective on the possibilities for ending violent political conflict in Africa. Spears focuses on the security predicaments of the disputants themselves as he closely examines the roots and dynamics of civil wars in Angola, Ethiopia, and Somalia. His original analysis leads to conclusions that challenge prevailing assumptions about the nature both of conflict resolution and of peacebuilding in postconflict societies. Ian S. Spears is associate professor of political science at the University of Guelph. His publications include States Within States: Incipient Political Entities in the Post–Cold War Era (coedited with Paul Kingston).

281