Power Politics in Zimbabwe 9781685850692

Zimbabwe's July 2013 election brought the country's "inclusive" power-sharing interlude to an end an

260 94 2MB

English Pages 281 [290] Year 2014

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Power Politics in Zimbabwe
 9781685850692

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Power Politics in Zimbabwe

Power Politics in Zimbabwe Michael Bratton

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

Paperback edition published in the United States of America in 2016 by Lynne Rienner Publishers, Inc. 1800 30th Street, Boulder, Colorado 80301 www.rienner.com and in the United Kingdom by Lynne Rienner Publishers, Inc. 3 Henrietta Street, Covent Garden, London WC2E 8LU Hardcover edition published in 2014 by Lynne Rienner Publishers, Inc. © 2014 by Lynne Rienner Publishers, Inc. All rights reserved ISBN 978-1-62637-388-4 (pbk. : alk. paper) 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

Contents

vii

Acknowledgments

1 Power Politics

1

2 Political Settlements

13

3 The Colonial Political Settlement

33

Part 1 The Roots of Repression

4 The Independence Political Settlement 5 A Period of Crisis, 2000–2008

6 African Experiences with Power Sharing

Part 2 Power-Sharing Settlements

51 73 97

7 Zimbabwe’s Power-Sharing Settlement, 2008–2013

123

8 Rewriting Zimbabwe’s Constitution

145

Part 3 The Challenges of Reform 9 Improving Electoral Conduct?

10 Approaching Security Sector Reform 11 Tackling Transitional Justice

v

165 189 211

vi

Contents

12 Reflections on Theory and Policy

Part 4 Conclusion List of Acronyms References Index About the Book

233

247 251 271 281

Acknowledgments

I did not choose to write this book. It chose me. As a child in Salisbury, Rhodesia, I was aware only of the blissful delights of an outdoor life in Africa’s most beautiful country. As I grew up, however, I slowly became conscious of a different vista: the surrounding society was sharply split between a privileged white elite and a far larger black African society. I also sensed a looming political confrontation. Without ever making a conscious decision to do so, I have been trying ever since to understand African aspirations for political freedom and human dignity—and why these hopes have been so frequently dashed by self-interested political elites. Sometimes this journey has taken me into remote rural areas, not only in Zimbabwe, but also in Zambia and Malawi, and at other times to the doorstep of human rights organizations in Uganda, think tanks in Ghana, or government ministries in South Africa. My inquiry has involved conversations with politicians, civil servants, farmers, and market traders in more than a dozen countries from Kenya to Mali, as well as countless hours in libraries, lecture halls, computer labs, and seminar rooms on several continents. But now the itinerary has come full circle; that is, back to Zimbabwe. Zimbabwe today is a troubled country. Its fabulous natural potential has been squandered. Its promising institutions have been turned to wicked ends. Its people have been violently abused. Why did a predatory clique of black rulers cause such destruction? Why have rival political elites in Zimbabwe never been able to reach a valid political settlement, up to and including the recent charade of “power sharing”? What, if anything, can be learned about current political outcomes in Zimbabwe from the country’s own history, and from the histories of other countries that have also experimented with governments of national unity? If the pages that follow cast any light on these questions, credit should go mainly to others. For many years I have leaned heavily on my colleague

vii

viii

Acknowledgments

and friend Eldred Masunungure for thoughtful reflections on Zimbabwe’s politics. He will recognize bits and pieces of ideas that we developed while working together on Afrobarometer surveys or writing reports on Zimbabwe for the World Bank, the US Agency for International Development (USAID), and the Development Leadership Program supported by Australian Aid. Early versions of some material in the book were presented at a conference on “Becoming Authoritarian” at Cornell University in 2011 and in revised form at the annual meeting of the American Political Science Association in Chicago in 2013. Other brief passages first appeared in articles in the Journal of Democracy (Chapter 7), Foreign Affairs (Chapter 9), and the Journal of Modern African Studies (Chapter 11). I also learned a great deal from conversations with (in alphabetical order) Joel Barkan, Roy Bennett, Abdenico Bhebhe, Tendai Biti, Valerie Bunce, Nelson Chamisa, Abel Chikomo, Annie Chikwana, Patrick Chinamasa, Heather Chingono, Rindai Chipfunde-Vava, Eddie Cross, Dumiso Dabengwa, Staffan Darnolf, Larry Diamond, Boniface Dulani, Geoff Feltoe, Chad Gandiya, Peter Godwin, Tony Hawkins, Richard Joseph, Richard Klein, Norma Kriger, Adrian Leftwich, Frances Lovemore, Okay Machisi, Lovemore Madhuku, Farai Maguwu, John Makamure, Ian Makone, Simba Makoni, John Makumbe, Ibbo Mandaza, Elton Mangoma, Charles Mangongera, Isaac Maphosa, Sydney Masamvu, Jennifer Lynn McCoy, Gorden Moyo, Sam Moyo, Beatrice Mtetwa, Deprose Muchena, Fidelis Mudimu, Jestina Mukoko, Elphas Mukonoweshuro, Rukudzo Murapa, Edwin Mushoriwa, Douglas Mwonzora, Daniel Ndlela, Irene Petras, Kucaca Phulu, Brian Raftopoulos, Tony Reeler, John Robertson, Mandivamba Rukuni, Martin Rupiya, Lloyd Sachikonye, Otto Saki, Eileen Sawyer, Tim Sheehy, Useni Sibanda, Masipula Sithole, Jameson Timba, Morgan Tsvangirai, Laurence Whitehead, Jenni Williams, Rosaline Zigomo, and Austin Zvomo. Visits to Zimbabwe every year since 1980—including stints for various universities, foundations, and nongovernmental organizations—helped me keep abreast of events. A sabbatical fellowship at the United States Institute of Peace—where I gained new insights from Cecile Aptel, David Backer, Steve Heydemann, Chantal de Jonge Oudraat, and Valerie Rousseaux and benefited from research assistance by Kevin Malone—provided the perfect opportunity to write most of the manuscript. I especially want to thank Adrienne LeBas and Timothy Scarnecchia for providing terrifically incisive comments on a draft of the manuscript. All errors, though, are mine. Anne and I had always hoped to eventually return to live and work in Zimbabwe. But those plans are presently suspended. Until the country’s political leaders can find a genuine political settlement, the lyrics of the Biddle City Band (in which I play the electric bass) will have to suffice:

Acknowledgments

I guess I won’t be going back there. I’ll have to find another home. I miss that red Zambezi sunset, A place where wild illusions roam.

ix

—Michael Bratton

1 Power Politics

In Zimbabwe’s presidential, parliamentary, and local elections of July 31, 2013, the Zimbabwe African National Union–Patriotic Front (ZANU-PF) meted out a crushing defeat to the Movement for Democratic Change (MDC), its erstwhile partner in a shaky power-sharing government. Robert Mugabe won a seventh presidential term with an official 61 percent of the vote against rival Morgan Tsvangirai’s reported 34 percent. ZANUPF seized well over two-thirds of the elected seats in Zimbabwe’s House of Assembly (160 out of 210), a supermajority that would allow the party to change the country’s newly minted constitution at will. Moreover, by effectively terminating an MDC presence in the cabinet and its control of the lower chamber of parliament, the result called into question the very viability of opposition politics in Zimbabwe. In short, the election represented a resounding reassertion of one-party power and defeat for a decade-long attempt to introduce a more inclusive set of rules for the conduct of politics. To be sure, the lopsided vote confirmed that the old ruling party had succeeded in remobilizing a political base in Zimbabwe’s majority rural areas, albeit in good part by intimidation. Using every strategy in the playbook of power politics, ZANU-PF outfoxed, out-organized, and out-muscled a well-meaning but inexperienced popular opposition movement, whose followers were concentrated in the towns. Indeed, Mugabe’s party was so determined to emerge as the winner that it flagrantly manipulated the procedures and institutions of democratic elections, thus inadvertently calling into question the legitimacy of its own apparently overwhelming victory. These striking events threatened to derail the country’s fragile recovery from a dire period of deep economic and political crisis. Once one of Africa’s most bountiful and promising lands, by 2008 Zimbabwe had descended into political violence, economic deprivation, and institutional

1

2

Power Politics in Zimbabwe

decay. A brief interlude of enforced power sharing between 2008 and 2013 helped deliver a degree of social peace and economic stability to the country’s long-suffering population. But despite so thoroughly recapturing the commanding heights of state power in 2013, ZANU-PF appeared to have little to offer other than an unchecked return to economic folly, elite corruption, and the bitter politics of exclusion. An obvious, but all too easy, explanation for misrule in Zimbabwe lays blame at the feet of the country’s only leader during more than 30 years of independence: President Robert Gabriel Mugabe. His biographers have lent merit to this case (Chan 2003; Meredith 2007; Holland 2008). A veteran of a colonial liberation war, Mugabe is a quintessential purveyor of power politics (Blair 2002). His path to the apex of state power—by bullet as well as ballot—shapes the way he has subsequently governed. As Zimbabwe’s top official, Mugabe has concentrated authority in the presidency and thus gained sweeping discretion over political decisions. Displaying his trademark political symbol of a clenched fist, he is usually seen in public surrounded by a phalanx of uniformed security forces. When challenged, he has seldom hesitated to apply his self-proclaimed “degrees in violence” against enemies—real and imagined. Although African “big men” set the tone for a regime of governance, they almost never govern entirely alone. After all, politics is the art of the possible: leaders govern using the endowment of material resources, inherited institutions, and political alliances available to them at any given time. These legacies create as many constraints as opportunities. To survive in office, a leader must, at minimum, serve the vested interests of his immediate coalition of supporters, an elite group that often shares a comradeship born from a common political struggle. Together, these allies seek to maximize the group’s collective advantage, usually to the exclusion of political rivals and ordinary citizens. While the life story of an authoritarian leader may personify the ruling group, its regime of governance—described in this book in terms of power politics—is internalized in the attitudes of dominant elites and institutionalized in how they exercise power. A powerful clique of party and military hardliners may sometimes limit a top leader’s freedom of choice. And an authoritarian regime may persist long after the original “big man” leaves the political stage. However, even the most powerful ruling group can never have everything its own way. Especially at critical junctures in a country’s history, they discover that they must work with others to create new rules for the political game. The nature of the resultant regime is determined in good part by the political deals—this book calls them “political settlements”— reached among the preeminent political actors of the day. Sometimes these actors arise from factions within the ruling group, where they have learned to operate in the treacherous, fratricidal world of autocracy. At other times,

Power Politics

3

an opposition emerges from civil society, usually as a counterreaction to the excesses of an exclusive, narrow ruling coalition. At some exceptional moments, when rival elites attract a popular following, ordinary people may occasionally enter the power struggle as members of mass movements. Complicating the scenario, international actors from neighboring countries and further abroad may also intervene; if they align themselves with challengers rather than supporting the status quo, external interventions tend to further limit the discretion of would-be dictators. In independent Zimbabwe, the country’s ruling group emerged from the crucible of an anticolonial liberation war. Its political strategies and tactics cannot be understood apart from this heritage. Political leaders opted for violence, first as a reaction against the brutalities of settler colonialism, later as a means of retaining their tight grip on the bountiful privileges of state power. In imposing their preferred political settlement on the Zimbabwean population, ZANU-PF created a militarized version of an electoral authoritarian regime (Schedler 2006; Levitsky and Way 2010). In response, the rival MDC sought to use the formal mechanisms of constitutional reform and competitive elections to unseat a group of rulers who had outstayed the population’s welcome. But, as I will show, their strategy of nonviolence and formal legalism repeatedly failed to dismantle the ruling elite’s power structure. Instead, election campaigns became focal points for state-sponsored violence, overt or implied. And international actors played ambiguous roles, sometimes pressuring the country’s leaders to compromise and other times turning a blind eye to political abuses. The disturbing case of Zimbabwe raises general questions for subSaharan Africa, a region that, in the past half century, has suffered an excessive share of the world’s armed conflicts, civil wars, regime changes, and state failures. These devastating events are only the most visible manifestations of a more generic deficiency: the failure of many African states to create a legitimate political order (Zolberg 1966). Far from establishing a consensual right to rule, African political leaders have too often asserted an unregulated desire to capture control of the state. To the extent that narrow political elites have been willing to resort to violence to secure this prize, political life in Africa has been governed by raw power politics. These African experiences raise broad and fundamental questions. How can developing countries avoid civil strife? What are the institutional requirements for strong and legitimate states? Do power-sharing pacts provide answers to recurrent problems of legitimacy? Do these kinds of political settlements foster or undermine the consolidation of democracy? In recent years, analysts have recommended power-sharing settlements as the most appropriate antidote to persistent political conflict (Lijphart 1985, 1999, 2008; Norris 2005; Reynolds 1999, 2002). In Africa, these institutional devices—which seek to marry political rivals in coalition gov-

4

Power Politics in Zimbabwe

ernments of national unity—have been applied to problems of settler decolonization, civil war, and, more recently, protracted electoral crisis. But, one size does not fit all. Power sharing is unlikely to work when entrenched authoritarians lose an election but are unwilling to surrender power. Under these circumstances, the former ruling elite is instead likely to call upon established institutional and cultural legacies to reassert exclusive control of the state. To illustrate these dynamics, consider a snapshot of a recent confrontation between rival political elites in Zimbabwe. The vignette provides a dramatic example of the power politics that undermine Zimbabwe’s persistent search for an inclusive and legitimate—but ultimately elusive—political settlement. Specifically, in this book I offer an explanation of the failure of Zimbabwe’s ill-fated power-sharing settlement of 2008–2013.

A Settlement Under Stress

In October 2010, Zimbabwe’s fragile “unity” government was close to collapse. A deep rift had developed between the main protagonists: President Mugabe and Prime Minister Tsvangirai. Expressing frustration with political and policy deadlock, each called for fresh elections as the only way forward. Yet the prospect of a new round of political campaigning raised the specter of an unwelcome return to state-sanctioned violence. This stalemate threatened to escalate into a full-blown constitutional crisis. The precipitating events were President Mugabe’s decisions to appoint governors, judges, and ambassadors without informing the prime minister or seeking his consent. These unilateral acts were only the most recent violation of the terms of a power-sharing deal struck just two years earlier by Zimbabwe’s main political parties. Mugabe’s high-handed maneuvers seemed to confirm that power sharing will rarely work when one partner dominates decisionmaking and lacks a sincere commitment to cooperation. From the outset, power in the government of national unity (GNU) was unequally shared. ZANU-PF retained control of the core instruments of coercion, notably the ministries of defense and justice as well as the intelligence service. The MDC gained leadership of the ministries of finance, education, and health. So, while the former opposition party took on the demanding responsibility of ensuring socioeconomic recovery (which it calculated would rebound to its electoral advantage), ZANU-PF concentrated on shoring up its apparatus of political control and national security (also in preparation for forthcoming elections). The power-sharing agreement required the president to consult with the prime minister on important decisions, for example, appointing top officials

Power Politics

5

or announcing government policy. Yet Mugabe repeatedly acted as if he retained the untrammeled powers of an absolute ruler by failing to honor, even in the breach, the consultation requirement. He treated Tsvangirai with disdain, refusing him the chairmanship of the cabinet in the president’s absence and condoning the refusal of army and police chiefs to salute the prime minister. The fledgling Prime Minister’s Office and inexperienced MDC-T parliamentary delegation struggled to find reliable means to block such ZANU-PF power plays. But Mugabe’s unilateral political appointments were a step too far. Aside from the lack of consultation, they directly violated an understanding that these posts would be shared proportionally among the main parties and with the prime minister’s consent. In response, and abandoning his customary reserve, Tsvangirai raised the political stakes. He framed the issue as a constitutional dispute that threatened the viability of the coalition government. In a carefully worded public statement, the prime minister confessed that “the events of the past few months have left me sorely disappointed in Mr. Mugabe and in his betrayal of confidence that I and many Zimbabweans have personally invested in him.”1 He called upon the Senate to refuse to seat the governors, the chief justice to void the appointments of illegally appointed judges, and the international community to reject the credentials of unconstitutionally appointed ambassadors. While boycotting a cabinet meeting, Tsvangirai declared, “Neither I, nor the MDC, can stand back any longer and just allow Mr. Mugabe and ZANU-PF to defy the law, to flaunt the Constitution, and to act as if they own this country.”2 Mugabe reacted with anger, denouncing the “foolish and stupid” events unfolding within the government. He declared that the power-sharing experiment was only meant to last for two years and called for an end to the unity government. By this time, the president and prime minister were communicating only through the press; even before Tsvangirai’s boycott they had reportedly gone for many weeks without talking directly to each other. These developments held ominous warnings for Zimbabwe. In the aftermath of a violent election in 2008, none of the contenders was ready to run an effective campaign. Given the uncertainty surrounding a constitutional reform process, it was unclear whether the next elections would be held under old rules or new. Moreover, the official electoral management body was ill prepared to guarantee a free and fair contest. And, most important, military and militia leaders were already circulating in rural areas warning would-be voters that violence would befall anyone who dared to vote for the MDCs, whether the larger faction led by Tsvangirai (MDC-T) or the smaller splinter group headed by Arthur Mutambara. Yet, despite stresses and strains, Zimbabwe’s so-called inclusive government continued to limp along. Neither ZANU-PF nor the MDCs had any

6

Power Politics in Zimbabwe

other potential political partner. Mugabe and his party continued to try to goad Tsvangirai and his supporters into leaving the coalition. Despite provocation, the latter refused to take the bait—or the blame that would go with it. However, given the unworkable dual political executive, a stalemated GNU was unlikely to accomplish much in the way of meaningful political or policy reform.

The Problem of Power Politics

Power is a relational concept. It refers to the relationship between A, a dominant person or group, and B, a compliant individual or collectivity. A is said to have power over B if A is able to get B to do what A wants. If B refuses to comply, then a power struggle ensues, much as occurred between Mugabe and Tsvangirai (and their respective elite coalitions) in the sequence of incidents just described. In any such struggle, political protagonists may employ three essential power resources: coercion, incentives, or persuasion. The domination of A over B may be established by violence, for example, when A uses an arsenal of repression—or the threat thereof—to compel submission. Alternatively, A may reward B for obedience by offering attractive material inducements, whether in cash or in kind. The dominant party thus co-opts the compliant party, often by distributing political offices with attendant perquisites. Finally, beyond force or favor, A can use reasoned arguments to persuade B to acquiesce, for example, by appealing to traditional symbols, popular ideas, or ethical values. It should be clear that these basic power resources lie on a continuum of political legitimacy. This continuum runs from coercion to persuasion; that is, from the least to the most justifiable form of power. All political leaders—whether democratic or authoritarian—employ mixtures of these power resources as instruments of rule; none relies on just one approach alone. But leaders in authoritarian regimes assert political control by using methods that tilt toward the coercive end of the scale, for example, when they make decisions without consultation. By contrast, leaders in democratic regimes rely more heavily on persuasion, for example, trying to convince voters to choose parties based on the policy platforms presented during election campaigns. The strategies employed to legitimize these two main regime types overlap in the middle since each leader must deliver economic and social goods to ordinary citizens in order to substantiate a political right to rule. But the distribution of rewards, usually in the form of public goods, is expected to be broadly based (that is, inclusive) in democratic regimes. In autocratic regimes, leaders can afford to attend to the needs of a narrower ruling group by

Power Politics

7

supplying them with private goods and by backing up exclusionary policies with coercion. In power politics, might makes right. In this form of politics, the main sources of authority are military strength and the selective distribution of economic resources. Little room remains for ethical values or constitutional rules to constrain the unlimited exercise of power. The practice of power politics creates a dog-eat-dog world in which only the fittest (who may also be the fattest) survive. The concept of power politics derives from the realist school of international relations, which views states as locked in blunt competition to achieve self-defined national interests in the absence of an overarching external authority (Wight 1946; Morgenthau 1967; Mearscheimer 2001). In this book, I focus on domestic politics within a state rather than on international politics between states (Mills 1956; Svolik 2012). But the concept of power politics aptly characterizes the domestic political behavior of dominant political elites in authoritarian settings. Leaders in these regimes are motivated by the self-interest of the ruling group, determined to maintain dominance over opponents and rivals, and they resist independent constraints embodied in a rule of law. Instead, autocrats prefer to employ primal means to attain their desired political ends. In political struggles with rivals they rarely hesitate to unleash the full panoply of power politics, including manipulation of the law, economic exclusion, political intimidation, covert operations, and even physical violence. The power politics approach has been criticized for granting undue attention to the biographical details and personal quirks of the towering gladiators who play starring roles in the political arena. But, as stated earlier, my goal is to de-emphasize individual leaders and instead emphasize the persistent elite coalitions and inherited political institutions through which they operate. My conception of power politics does not ignore political agency; rather, it situates groups of actors collectively within political structures. In my view, the ruling coalition gets its way mainly by monopolizing the key repressive institutions at the heart of the state, such as the army, police, and prisons. The prevalence of force in political life imparts “dismal,” even “gruesome,” features to authoritarian rule (Svolik 2012, 13– 15). Because violence is the ultimate arbiter of conflicts, incumbent leaders cannot make commitments that other actors regard as credible. For example, even though rulers may seemingly agree to the negotiated terms of power-sharing settlements, they seldom hesitate to renege later if breaking promises becomes convenient. Because all parties, particularly those with less power, cannot trust rulers to abide by their word, elite settlements rarely hold. Moreover, authoritarian politics are marked by the absence of an independent, sovereign authority that can enforce agreements. Absent external

8

Power Politics in Zimbabwe

sponsorship or a widespread culture of constitutionalism, formal political institutions are worth little more than the paper they are written on. Thus, political outcomes are driven less by democratic constitutions and other legal codes and more by the informal, often arbitrary, practices of power players. Rather than submitting to the rule of law, authoritarian elites typically employ the coercive powers of a rule by law (Ginsberg and Mustafa 2008). This mode of governance abandons due judicial process in favor of the selective prosecution of political rivals, frequently with trumped-up charges of legal obstruction or economic corruption. Authoritarian rulers also employ coercion to accumulate economic resources, often through the confiscation of private property. They design extractive economic institutions in order to remove assets and income from producers in society and to transfer benefits to a political class (Acemoglu and Robinson 2012). Political predation of this sort undermines property rights and perverts incentives for innovation and investment, thus constraining economic growth. Instead, the proceeds of the economy— whether derived from land, mining, or commerce—are distributed to political loyalists, especially those within the leader’s inner circle. In this regard, rulers have an incentive to minimize the size of the governing coalition, the better to maximize the share of spoils accruing to each of its members (Bueno de Mesquita and Smith 2011). For this reason, in the context of power politics, political institutions are not only economically extractive but also politically exclusive. In times of economic crisis, there is even a tendency for the ruling coalition to shrink down to its coercive core. Leaders will always accord special treatment to the armed forces in order to maintain their essential loyalty; even if civil servants sometimes go unpaid, every effort is made to compensate the military. But because autocrats rely heavily on repression, they inadvertently strengthen the hand of the armed forces, who, in turn, are able to claim a share of both economic bounty and political decisions. Although autocracies are built on foundations of political violence and economic extraction, some authoritarian leaders still try to use a comprehensive system of ideas to justify their rule. These ideologies may invoke either ancient traditions or a vision of revolution, but their purpose is to persuade people to support the political regime on a voluntary basis. Ruling parties that ascend to power via armed struggle may even have unusually powerful abilities to foster elite cohesion and regime durability (Levitsky and Way 2012). A violent mode of political transition generates strong political identities, social solidarity, and organizational discipline that serve as supplementary resources of authoritarian rule. The problem with appeals to past revolutionary glory, however, is that they have a limited shelf life, especially for new generations born since the initial struggle. Ideology can all too easily become a smokescreen that barely conceals the vested interests

Power Politics

9

of a narrow ruling clique. Leaders can try to renew the fervor of ideological commitment by conjuring up threats from internal or external “enemies.” But, ultimately, autocrats can build political legitimacy only by delivering economic and social goods, which requires a growing economy. If they cannot perform the basic tasks of economic management, they have little choice but to fall back on violence. The problem of authoritarian rule can be summarized as the domestication of power politics. Can political actors channel open conflict along peaceful, legal, and institutional lines? In other words, can ruling and opposition elites arrive at a comprehensive mutual agreement—an inclusive political settlement—about the basic rules for governing society? The authoritarian purveyors of power politics have distinct advantages in this struggle. To the extent that a partisan minority enjoys a monopoly over the means of coercion, they can effectively deter opponents from mobilizing popular support in the streets of the capital city. One negative consequence of rule by force, however, is that it leaves the opposition with few options but to resort to violence themselves, thus perpetuating a vicious cycle (Nkomo 1984; LeBas 2011). Moreover, elites who capture state power invariably extract and deploy economic resources to consolidate power. Even if authoritarian rulers find it useful to go through the motions of competitive elections, they do so only after ensuring an incumbency advantage over campaign resources, thus tilting the electoral playing field in their own direction and ensuring that they cannot lose. The more exclusive the ruling coalition and the more extensive their control of the private economy, the less any opposition movement is able to mobilize resources to mount a successful challenge (Arriola 2013). For their part, challengers to authoritarian rule can usually offer appealing political messages: an end to state-sponsored violence, guarantees of personal political freedom, and the prospect of broad-based economic opportunity. One might expect that the deeper the pathologies of power politics, the easier it is for rival elites to persuade a mass population that they offer a brighter future. Much depends, however, on whether opposition leaders can convincingly get out their message in the face of official propaganda campaigns and mobilize their supporters in the face of the ruler’s efforts to suppress electoral turnout. Sometimes, widespread fear of political violence and extended experience with economic deprivation discourage citizens from participation in political life. Moreover, alternative leaders may have tainted their own reputations by participating as junior partners in power-sharing arrangements with entrenched dictators. Finally, the potential for a meaningful division of power depends on whether conditions exist for free, fair, and credible elections. Can electoral challengers withstand the organized efforts of entrenched authoritarian rulers to buy or bludgeon their way into retaining power?

10

Power Politics in Zimbabwe

Outline of the Book

This book analyzes the resilience of authoritarian rule in Zimbabwe through the lenses of power politics and elite political settlements. I show how, by capturing the state, leaders have sometimes been able to entrench lasting arrangements for organizing and exercising power. Zimbabwe’s exclusive and defensive political regimes have never effectively incorporated the interests of a wide range of political players, let alone respected the will of a mass electorate. At certain critical junctures, however, key political actors have had little choice but to enter elite pacts to maintain their influence, most recently in the form of a power-sharing agreement signed in 2008. But a lasting political settlement based on an inclusive social compact has always remained elusive. This book places power sharing in Zimbabwe in a comparative perspective, both over time within the country and laterally across other African countries. The empirical focus is on the political contest between a nascent democratic movement and a resilient authoritarian regime, especially during a transitional period of enforced coalition government. More broadly, the book aims to draw theoretical and policy lessons from one particularly problematic power-sharing experiment. The text is organized in four parts. Chapter 2 is conceptual and theoretical. I begin by defining political settlements with reference to an emerging literature on the subject. Applying a power-politics perspective, I propose a framework of three different types of political settlement: power capture, where a dominant elite unilaterally imposes its own rules; power sharing, where contending elites struggle to institutionalize competing sets of rules; and power division, where elites agree on rules for periodically circulating power among contending elites, such as democratic elections under civil liberties and the rule of law. I also argue that the way political elites rule depends in good part on the way they ascended to power. Thus, Zimbabwean politics can best be understood by tracing the trajectory of the dominant political party, whose collective identity was forged during a violent guerrilla war. Once in power, the party readily appropriated the institutional legacy of a strong colonial state. Part 1 compares institutional arrangements across time. A brief historical narrative first traces political settlements in Southern Rhodesia from the earliest colonial contacts, when British adventurers enticed traditional chiefs to strike dubious deals for land and minerals, to the last-ditch effort of white settlers to preserve political power by offering moderate African nationalists a so-called Internal Settlement (Chapter 3). The narrative then turns to independent Zimbabwe, starting with a compromise agreement hammered out at Lancaster House in London in 1979 to grant majority rule in return for guarantees of economic continuity. Also considered are the so-

Power Politics

11

called Unity Accord of 1987, aimed at mending rivalries among the principal liberation movements, and a landmark set of concessions that the ruling party made in late 1997 to veterans of the liberation war. The upshot of each of these political settlements was the maintenance of ZANU-PF’s exclusive grip on state power (Chapter 4). I then provide a detailed account of the escalating political and economic crisis after 2000, which ended only when the international community prodded rival political elites into a power-sharing agreement in 2008 (Chapter 5). Part 2 concentrates on power sharing. I place power-sharing settlements in cross-national context by examining the proliferation of such pacts in contemporary Africa (Chapter 6). The power-sharing approach has been used to bring a peaceful end to settler colonialism—as in South Africa— and to terminate civil wars—as in Sierra Leone. I distinguish the utility of power sharing in these cases to its more questionable application in crises of electoral democracy, as illustrated by the case of Kenya following disputed elections in 2007. I argue that, when entrenched elites retain control over the coercive instruments of the state, they are poised to act as spoilers for any negotiated political settlement. Moreover, while elite pacts help reduce overt political conflicts in the short run, sharing power has negative effects on the consolidation of democracy in the longer run. The internal dynamics of coalition politics in Zimbabwe’s so-called inclusive government are analyzed with special attention to the challenges of establishing a legitimate political order (Chapter 7). Part 3 draws attention to a set of fundamental reforms that together determine whether countries in transition are able to move toward more open forms of government. With reference to Zimbabwe since 2008, I consider constitution making (Chapter 8), election management (Chapter 9), security sector reform (Chapter 10), and transitional justice (Chapter 11). Since little systematic analysis has been written about these issues in Zimbabwe, I trace the trajectory of reform efforts in some detail. Ultimately, progress on political reform depends on the outcome of elite bargaining. Where relevant, I reference reform experiences in other countries. In addition, I assess popular support for political reforms and power sharing in Zimbabwe with a unique set of public opinion data. The book concludes with lessons learned (Chapter 12). Some lessons are abstract, requiring scholars to revise theories about institutional change and power sharing. Other lessons are practical and are relevant to policymakers in the government of Zimbabwe, regional organizations in Africa, and the international donor community. The conclusion reiterates the abiding theme of the book: legitimate political settlements are elusive because autocratic elites who capture power are unwilling to share it. This explanation of authoritarian resilience is vividly illustrated in contemporary Zimbabwe.

12

Power Politics in Zimbabwe

Notes

1. “Statement by the Prime Minister of the Republic of Zimbabwe and President of the MDC, Rt. Hon. Morgan R. Tsvangirai.” October 8, 2010. http://www .zimbabweprimeminister.org. 2. Ibid.

2 Political Settlements

Political settlements provide an essential entry point for understanding the origins of power politics, as well as the potential constraints upon this form of rule. As elite bargains, settlements define the balance of power and frame the rules of the political game. In Zimbabwe, an armed liberation movement captured a strong state that was first designed for the purposes of colonial domination. Once in power, the Zimbabwe African National Union–Patriotic Front (ZANU-PF) relied heavily on core coercive institutions (army, police, and intelligence services) and on the distribution of economic patronage rewards to proven political loyalists. The nature and style of this regime can be traced back to the conditions and causes from which it arose. The civilian-military coalition that came to dominate political life in Zimbabwe favored an exclusive form of political settlement.

Framework

A political settlement is a common understanding, initially among elites, about the organization and exercise of power (DiJohn and Putzel 2009; Southall 2013). Since power—the ability to obtain compliance—is a contested resource, settlements run the gamut from exclusive, benefiting a select group of elites, or inclusive, representing a broader set of interests. Because power determines the distribution of resources, political settlements establish who gets what, when, and how (Lasswell 1936). At the center of power lies the state. Political elites struggle to control the instruments of state—especially the institutions of coercive, extractive, and legal power. As rational practitioners of power politics, rulers would prefer to govern alone by capturing solo control of the state. Often, how-

13

14

Power Politics in Zimbabwe

ever, a given political elite cannot exclude others and instead must enter into arrangements to share or divide power. The broader and more inclusive the ruling coalition, the more likely that nonstate actors can be persuaded to comply with commands from public officials because ordinary people are inclined to regard the state as legitimate. A settlement has both formal and informal aspects. On the formal side, political settlements may be inscribed in the written terms of ceasefire agreements, peace treaties, property laws, coalition pacts, and national constitutions. On the informal side, settlements are validated by less tangible but equally important qualities such as common interests, popular ideas, credible commitments, interpersonal trust, and shared visions of the future. These qualities may be in short supply. Yet, to stick, settlements require congruence between the formal and informal dimensions of any agreement and between institutional rules and the distribution of material benefits. Moreover, the legal agreements that give effect to political settlements must be flexible enough to evolve according to social practices. Ideally, a political settlement represents a society-wide consensus that is codified in a written constitutional and legal code. In practice, however, political elites tend to obey an iron law of oligarchy. 1 They reach deals about power without much popular or legal validation, often reproducing the exclusive and extractive institutions of predecessor regimes. In other words, “The critical element that holds a political settlement together is the alignment of interests within the dominant elite coalition,” although no rulers can indefinitely ignore “the dynamic relationship between elite interests and the broader array of interests in society” (Parks and Cole 2010, viii). Thus every political regime—whether democratic, authoritarian, or hybrid—is based on a political settlement; that is, a set of operating rules that reflects the actual balance of political forces in the populace. In practice, therefore, settlements more often resemble elite pacts than social contracts. Political settlements provide an immediate alternative to the open conflicts engendered by power politics. Settlements arise when political adversaries conclude that peaceful procedures are less costly than the continued use of violence for resolving conflicts of interest. Building on this foundation, political settlements promise to one day deliver permanent institutions with routine capacities to domesticate violence, reduce discord, arbitrate disputes, involve citizens, and provide desired public goods. As such, political settlements bridge the terrain between peacemaking, on the one hand, and state-building, democratization, and socioeconomic development, on the other (Brown and Gravingholt 2010). The acid test for the durability of any state is whether its institutions can win widespread political legitimacy. In other words, do key stakeholders at various levels come to accept that the occupants of state office have a

Political Settlements

15

right to rule? At the outset, elites often agree among themselves to grant a narrow group of rulers a measure of political authority. A political settlement of this sort can prove quite durable as long as elites embed their power in a reliable set of institutions. International actors often prop up the regime by recognizing elite claims to state sovereignty (Jackson and Rosberg 1982; Englebert 2009). But in the long run, the ultimate arbiters of political legitimacy are the adult population in the country at large, who determine whether the prevailing political order is worthy of their consent.

A Wide-Angle Lens

By drawing attention to the actual division of power established by leading actors, the political settlements approach offers the possibility of understanding both the sources of political instability and the challenges of building enduring institutions. As a wide-angle lens, the political settlements approach is consistent with, and casts light on, several existing themes in the comparative study of politics. These include conflict management, state building, and regime transition. Each will be considered briefly in turn. In societies divided by violence, peace agreements are a necessary first step in arriving at a durable political order. One stream of literature on peace agreements suggests that stability following civil wars depends mainly on security guarantees enforced by external third parties (Walter 1999; Walter and Snyder 1999). Third-party enforcement may be important, but it runs the familiar risk of lacking authentic indigenous roots. Thus, there is no substitute for settlements reached by domestic actors, for example, through direct, face-to-face talks between the principals involved in a conflict. Recent research suggests that the comprehensive nature of internally negotiated peace agreements influences their subsequent durability (Hartzell and Hoddie 2003). If the participants agree to allocate power across various realms—economic, military, and territorial—peace agreements are likely to last longer than if they focus on political decisionmaking alone. The inclusion of assurances over multiple dimensions of power enables participants in peace agreements to gain a cumulative sense of security and guarantees against the possible failure of the accord. But power sharing is difficult following civil conflict because military enemies and prospective governmental partners are often one and the same. Trust between political adversaries is hard to establish when each may have recently used lethal violence against the other. As Spears puts it, “In civil war contexts, power sharing is equated with making a deal with the devil Conflict Management

16

Power Politics in Zimbabwe

and thus such deals, even when they are forged, are unlikely to last” (Spears 2002, 127). Other analysts even suggest that power sharing in the name of peace actually encourages insurgent leaders to resort to violence as an effective means of staking a claim to power (Tull and Mehler 2005; Mehler 2009). Moreover, international mediators too often assume that their job is largely done after a signing ceremony wraps up a peace accord. But these symbolic events do not always reliably represent a viable, longterm political settlement among indigenous political elites. In the absence of civil strife, or in its aftermath, the state plays the central role in maintaining political order. But, in practice, many questions remain: Who controls the state? Toward what ends? And do societal groups submit to state domination? Analysts now reject the facile assumption that elites at the pinnacle of politics always prevail in efforts to shape society (Migdal 1988; Scott 1998). State capabilities cannot be taken for granted in the developing world, where political leaders are routinely unable to get ordinary people to comply with their commands. Instead, social control is distributed among “the many organizations in society that vie to make the rules about how people should behave” (Migdal 1988, xx). Faced with entrenched social interests, the local agents of the state often operate far differently than central planners intended. And, paradoxically, if politically insecure rulers fear that rivals will build social power bases within the state apparatus, they will actually work to undermine the strength and independence of such institutions. A view of the state as one organization that must compete for domination among other social forces is compatible with the notion of political settlements. From this perspective, state building resembles less the imposition of a centralized order and more a process of negotiating a series of compromises. In other words, the coalition that occupies state office must bargain for the compliance of other organized social actors—whether military forces, armed militias, opposition parties, business associations, ethnic groups, or traditional authorities. Insightful studies of state building in Africa have described the resultant give-and-take process, notably across ethnic lines, in terms of “hegemonic exchange regimes” (Rothchild 1995) or “the reciprocal assimilation of elites” (Bayart 1993). In this socially embedded mode of state building, the neat boundary between the public and private realms is blurred. On the positive side, the adaptation of state structures to cultural norms and practices helps to build local ownership and a popular sense of authenticity for institutions. On the negative side, political settlements may open the door to collusion among public and private elites to employ state power toward personal or narrow group ends. State Building

Political Settlements

17

Regime Transition

Wherever elite coalitions capture the state, they set about attempting to consolidate power. In developing countries, leaders who were anxious about their hold on office after independence typically constructed a defensive array of political institutions to insulate themselves from social pressures. In some places these reorganizations took the form of military juntas that outlawed elections and legislatures; elsewhere, one-party regimes invented controlled forms of these institutions. Either way, the object was to prevent the unregulated turnover of political power. But those elites who were excluded from the ruling coalition—as well as a growing proportion of citizens—interpreted their inability to evict underperforming leaders as a betrayal of the nationalist social contract. In reaction, and as part of a global wave of regime change in the aftermath of the Cold War, they began to call for democratization. An influential literature on regime transitions is consistent with the political settlements approach. O’Donnell and colleagues argue that all transitions begin with divisions among rival political elites and end when these differences are resolved through negotiated political pacts (O’Donnell, Schmitter, and Whitehead 1986). The survival of the old regime depends on the level of cohesion among the practitioners of power politics. But if splits arise among these actors over the advisability of political reforms, a moderate faction may defect from the hardline core in order to join moderate elements in the opposition.2 In many cases this new alliance is able to wrest concessions from rulers, sometimes leading to competitive elections and a turnover of office holders. According to this model, the critical threat to a democratic transition is a reactionary backlash, notably by security chiefs or business elites. A successful outcome therefore depends on assurances extended to the old guard in a series of behind-the-scenes pacts that guarantee key political, military, and economic interests. In other words, regime transitions rest on negotiated political settlements.

Actors, Interests, and Institutions

Unpacking the concept of political settlements, Parks and Cole propose that “the key elements . . . are powerful actors, operating in pursuit of their interests, leading to the establishment or reshaping of institutions” (Parks and Cole 2010, 6, emphasis added). This approach nicely depicts how the most influential players in the political game seek to mold the institutions of the state to address their own ends. Far from being formal structures fixed over time, state institutions usually evolve informally to reflect the power politics of an underlying, “elite-driven social order” (DFID 2010, 22).

18

Power Politics in Zimbabwe

Actors

By definition, the most powerful actors are elites. These are the leading individuals, groups, and coalitions in society. They possess superior resources, whether of wealth (as with landowners and business people), coercion (e.g., the army, police, prisons, and intelligence services), or the ability to allocate legal rights and social values (e.g., judges, religious leaders, and traditional authorities). In addition, the leaders of political parties use organizational resources to contend for state power and, as such, constitute a distinct class of political elites. In developing countries, a few elite figures commonly cohere into a dominant coalition that rules exclusively for an extended period of time. As a result, some elites are state insiders; others are outsiders. The ruling coalition enjoys control over the top offices in the state hierarchy and the main sources of political patronage. The breadth, or inclusiveness, of the ruling coalition depends on whether political officeholders incorporate business, traditional, and security elites (or even rival political parties) into the inner circles of decisionmaking and resource distribution. Almost inevitably, some elite groups will be left out of the core alliance. Because exclusion threatens their interests, they must choose between two possible responses. One option is to seek co-optation into the official spoils system, for example, by aligning with the interests of the ruling group. The alternate approach is political resistance, such as mobilizing mass opposition in an attempt to open up the system or to capture it. Of special interest among actors in settlements are so-called spoilers; individuals who are able to wield veto power over outcomes (Stedman 1997; Tsebelis 2002). Spoilers gain power by possessing an inordinate share of resources. For example, a small guerrilla army may be able to torpedo a peace agreement by virtue of superior familiarity with a contested upland territory. A group of senior military officers may block a truth commission by insisting that an outgoing government grant a blanket amnesty. Or corrupt business elites may delay a democratic transition or economic reforms in order to protect material gains made under a previous regime. If they opt for renewed conflict over peaceful compromise, the actions of spoilers weigh disproportionately on the outcomes of political settlements. Interests

Political elites strive to gain state office and, once successful, to stay in power. If incumbent elites try to legitimize their rule by delivering economic development, they will have an interest in broadening the ruling coalition to include representatives of other communities. They also will have an incentive to choose policies that make their country’s economy at-

Political Settlements

19

tractive to foreign investors. In these ways, political and economic elites may come to share common cause in expanding the national economic pie and limiting predatory behavior. If ruling elites feel politically insecure, however, they are more likely to sacrifice ties with economic producers in favor of cultivating support within the security forces, and to rely on their control of coercion to retain power. A retreat into force or predation is especially likely if the economy experiences external shocks or prolonged contraction, resulting in the dwindling availability of patronage. The risk under these circumstances is that a shrinking core coalition will serve only its own limited interests. A fundamental tenet of the political settlements framework is that institutions—formal and informal—work distinctively across countries because their political contexts vary (Khan 1995, 2009; Englebert 2000). Institutions arise from particular conflicts and from their negotiated management or resolution. Ruling elites hold the upper hand in these processes, as they likely control the institution-building agenda. The viability of formal institutions depends critically on the balance of power; rules that are attuned to the interests of the most powerful elites are more likely to be enforced. As Khan explains, “Institutions and the distribution of power have to be compatible, because if powerful groups are not getting an acceptable distribution of benefits from an institutional structure, they will strive to change it” (2009, 4). Postcolonial societies offer numerous examples of elite-driven institutional reforms. Rulers may revise national constitutions in self-serving ways, such as restricting rights of political association or by concentrating power in an imperial presidency. And many failures of international development assistance can be attributed to the mismatch between imported institutions and the economic interests of political elites. The same lack of agreement about the rules of the political game lies at the root of fragile and failing states. Institutions

Explaining Settlement Outcomes

Settlement outcomes range from state failure to the establishment of a durable and legitimate order. This book proposes several factors to explain settlement outcomes:

• Inclusion. This is the most important factor. How representative of society are the elites, coalitions, and movements that participate in drawing up the rules of the political game?

20

Power Politics in Zimbabwe

• Scope. Is the settlement a purely political deal, or does it also comprehensively address economic and security interests? • Incentives. Does every party to a settlement have reason to believe that they have more to gain from abiding by institutional rules than engaging in violence? • Balance of power. How are power and other resources distributed among parties? Do a dominant set of actors enjoy resource advantages? Do spoilers enjoy special veto capacities? In particular, who controls the military and other agencies of coercion? • Credibility. Are all parties sincerely pledged to honor the terms of the agreement? Can leaders convince rivals of their trustworthiness? • Ownership. Does agreement arise primarily from indigenous political interests? Or is it driven by external third parties?

It should be clear that viable political settlements hinge on a favorable combination of these features. To repeat, an inclusive social base should counteract the exclusive interests of a narrow elite coalition. Beyond this essential foundation, the most conducive combination for consolidating a legitimate political order is the following: a comprehensive settlement, incentives for cooperation, an internal balance of power, and credible participant promises. While local leaders may prefer to own the settlement, they may have little choice but to turn to external third parties for leverage against a ruling elite intent on perpetuating power politics. The view through the lens of political settlements reveals distinctive patterns of power allocation. The parties to a political settlement may decide to: Types of Political Settlement

• Concentrate power in their own hands, • Exercise power in cooperation with rival groups, or • Divide power by accepting that one set of interests counterbalances another.

In shorthand terms, this book refers to these alternative forms of political settlement as power capture, power sharing, and power division. Each is an alternative means of regulating power politics and is discussed in turn. Some political settlements allow a segment of the political class to seize control of the commanding heights of the state. To the extent that leaders Power Capture

Political Settlements

21

can exercise wide discretion over decisionmaking and sideline rivals from public office, they will do so. But an exclusive and unbalanced form of power allocation risks predictable political pathologies. If arrangements for political accountability are absent, then predation, corruption, and other abuses of power can be expected. As the core instrument of legal authority, the state endows officeholders with the capacity to order others to do what they otherwise would not choose to do. With few means available for upward social mobility in Africa’s weak economies, state office also presents its occupants with opportunities to accumulate prestige and wealth (Callaghy 1984; Herbst 2000). Given such desirable prizes, political winners are tempted to seize as much state power as they can. But the snag of an exclusive, winner-take-all approach to governance is that losers receive nothing. Indeed, the emergence of dominant political elites has the unfortunate effect of consigning unsuccessful contenders to the political wilderness, where their livelihood chances are gravely reduced. When the stakes of political rivalry are so high—sometimes literally a matter of life and death—politicians instinctively try to hold power close rather than seek ways to share it. Concentrations of power at the apex of the state intensify chances for collusion and corruption. When a narrow group of political elites captures state power, they are tempted to use political office as a means to arrange lucrative personal business deals with economic operators (Reno 1995; Bayart et al. 1999). Thus, any broadening of a constricted political settlement tends to favor special interests, whether multinational corporations, indigenous entrepreneurs, or, in the extreme, minerals smugglers and drug cartels. Some analysts suggest that these sorts of organized firms can attain “state capture” on their own behalf (Hellman et al. 2000). But it is readily apparent that power capture can be initiated from any quarter: “Any group or social strata external to the state [can exercise] decisive influence over state institutions and policies for its own interests against the public good” (Pesic 2007, 3). In particular, ruling political parties often systematically appropriate the financial power of the state, employ relatives and cronies in the public service, and bend public policy toward partisan ends. Military elites, cautious about the international approbation that accompanies overt coups d’etat, may quietly infiltrate ruling circles behind a civilian façade. Political settlements based on power capture are inflexible. Once elites develop vested interests in these arrangements, they are loath to surrender. Accustomed to arbitrary authority and moneymaking opportunities, exclusive ruling coalitions often try to extend terms of office and rule indefinitely. Citizens occasionally read these initiatives as violations of an implicit social contract and react with mass protests (Wiseman 1996). Unchecked autocrats may then turn the coercive capabilities of the state against the populace, with state-sponsored violence leading to abuses by the

22

Power Politics in Zimbabwe

army or police. Ironically, these developments tend to deepen the incumbents’ resolve to cling to power, especially if they come to fear prosecution—if not for corruption, then for human rights violations—once they leave office. The horrors of civil violence may prompt interested parties—domestic and international—to call for power sharing. In an effort to end conflict, rival elites sometimes agree to parcel out state offices, for example, by creating a post for a prime minister alongside the presidency or by distributing cabinet posts among political parties on a proportional basis. They may further make commitments to work together on institutional or policy reforms. The general goal of power-sharing settlements is to moderate the raw competition for power, usually in the name of a greater good, such as peace, national unity, or a transition to democracy. Power sharing has long been recommended as a mechanism for governing deeply divided societies, especially those split by the sorts of ethnic and religious cleavages common in Africa. A power-sharing settlement—or “consociational democracy” in classic terminology (Lijphart 1999, 2008)— guarantees all main contenders a stake in the political process. According to Lijphart, four key institutions define power sharing: grand coalitions of all relevant parties, proportional distribution of public offices, spheres of autonomy for cultural groups, and veto powers for minority groups. All these rules challenge the majoritarian principle of winner-take-all by ensuring that prospective political losers have refuge from majority domination, a share in decisionmaking, and avenues to pursue their political interests. Outside Western Europe, experiments with power sharing have rarely displayed all the features of a full-blown consociational model. Instead, these deals have most commonly involved the formation of governing coalitions (grand or otherwise) and the distribution of public offices among participants. As a familiar consequence, the size of executive cabinets has tended to expand, sometimes to an unwieldy scale. Moreover, power-sharing bargains may not sit well in centralized presidential systems as compared with the more conducive context of parliamentary regimes. All told, power sharing in Africa is more appropriate for temporary resolutions of conflicts than for constructing an interlocking set of permanent institutions. Put differently, power sharing works better for peace building than state building. Nevertheless, power-sharing approaches have been applied in Africa to a variety of political situations. The formula was often adopted in the last days of Africa’s decolonization in an effort to convince minority groups of white settlers to accept black majority rule. Since that time, power sharing Power Sharing

Political Settlements

23

in governments of national unity has become a standard policy response to the resolution of postcolonial civil wars. More recently, the concept has even been stretched to embrace situations of political breakdown in Africa’s fledgling democracies, such as following disputed elections. I argue that power sharing is not appropriate under all these conditions, especially the resolution of electoral crises. But because power sharing seems to offer an antidote to deep-seated conflict, international and domestic elites have, with increasing frequency, fallen back on coalition government as a preferred settlement formula. If contenders for power find violence repugnant or costly, they will be inclined to make sincere commitments to compromise or unity. But if the relevant elites see power sharing as merely an interim detour from a larger struggle to capture state power, they will continue to have strategic incentives to exploit cooperative arrangements to their own advantage. In this regard, power sharing will delay—and may even permanently derail—a transition to democracy. Where authority remains concentrated, power-dividing settlements seek to disperse the authority of the state. The division may be horizontal, where legislative or judicial bodies (or agencies of restraint like an independent central bank or a human rights commission) impose limits on a political executive. Or the division may be vertical; for example, where rights-bearing citizens vote in competitive elections or where lower tiers of regional or local government gain control over central government functions and budgets. Participants in such arrangements accept that they can rarely attain all available power and settle for only a slice. But weak parties can obtain a measure of satisfaction because independent power bases are institutionally protected. If it turns out that political adversaries are insincere about surrendering control, remedies are available through the checks and balances provided by countervailing institutions. Roeder and Rothchild tout the advantages of power division: “One limits majorities . . . by expanding individual liberties and rights at the expense of government and by empowering different majorities in independent organs of government” (2005, 15). The logic of power division rests on the emergence of multiple majorities around diverse issues and within different institutions. These plural majorities disperse authority among different governmental branches and strengthen civil society in relation to the state. Such countervailing powers provide defenses against state capture by dominant elites. In short, the institutions of liberal democracy are the essential channels for resolving conflicts, thus enabling peace through democratization. Power Division

24

Power Politics in Zimbabwe

Institution building based on the division of political power works in three main ways. First, a clear line is drawn between the public and private spheres. The power to make certain important decisions, often concentrated in the state, is assigned to individuals, households, and private associations. Because social actors have diverse and crosscutting interests, the dispersion of authority across society actually reduces, rather than increases, the likelihood of domination by a single, permanent majority. Second, power division rests on the strict separation of governmental functions into specialized institutions. Rather than fusing executive and legislative functions—as in a parliamentary system—the powers of government reside in independent bodies, each with its own mandate and resources. The existence of multiple decisionmaking forums increases the likelihood that individuals will find themselves in a majority on at least some policy issues. Finally, to prevent the holders of state office from overreaching, power is pitted against power. In terms of constitutional design, power division therefore implies a presidential system with an even balance of authority between the executive branch and other organs of government. The popular demand for democracy in Africa in the 1990s was driven by a concern that African presidents had accumulated too many powers. This upsurge prompted a wave of constitutional reforms aimed at limiting the arbitrary abuses of one-man rule, for example through presidential term limits. In a startling turnabout, one-party and military regimes became largely extinct (at least formally) and were replaced by a range of fragile democratic and hybrid regimes. The rise of electoral institutions, however, has yet to be fully complemented by the institutionalization of independent legislatures, judiciaries, or local governments, whose development lags behind. And the realization of individual liberties and rights has been partial at best. Instead, African presidents and their elite coalitions have launched rearguard actions to resist the loss of executive powers, rendering power division in Africa an unfinished project.

Institutions: Continuity and Change

The account of political settlements presented so far has emphasized the self-interested actions and strategic interactions of elite political actors. The type of political settlement—whether power capture, power sharing, or power division—has been portrayed as if it were mainly a product of leadership decisions. While the rational calculation and gamesmanship of leaders certainly help determine who prevails, these considerations do not fully account for the development of institutions. Political actors are embedded in structural and cultural contexts that shape motivations, opportunities, and outcomes (Lichbach and Zuckerman 2008). Thus analysis of the search for

Political Settlements

25

a legitimate political settlement, whether in Zimbabwe or anywhere else, must include consideration of institutional legacies and cultural contexts. The field of comparative politics addresses institutional legacies through a school of studies known as historical institutionalism (Moore 1966; North 1990; Thelen 1999; Erdmann et al. 2010). It accounts for the development of political institutions with reference to the causal mechanisms underlying both continuity and change. Several concepts are key. First, “path dependence” refers to the inertia of initial conditions, which tends to reproduce institutions over time and establish patterns of longterm continuity (Mahoney 2000; Pierson 2000). Second, “critical junctures” are disruptive periods of political uncertainty—such as regime transitions—that provide political actors with unusual opportunities to initiate change (Pierson 2004; Capoccia and Keleman 2007). Third, “compliance” refers to the discretion enjoyed by political actors to exploit gaps between rules as formulated and their actual enforcement. Because different actors may interpret the same rules in divergent ways, opportunities sometimes arise for gradual institutional change (Mahoney and Thelen 2010). By tracing the complex interactions among actors, interests, and institutions, “Historical institutionalists . . . go beyond the truism that ‘history matters’; they examine not only what happened but how, when, and why” (Erdmann et al. 2010, 5). As will be shown, evidence is readily available of path dependence in the development of political institutions in Zimbabwe. In the precolonial era, successive Shona and Ndebele kingdoms developed the formal organization and informal practices of chieftainship. Precedents of patronage and violence as instruments of rule can arguably be traced back this far. After colonial conquest, white settlers incorporated indigenous political institutions, at least as far as they understood them, into a so-called native authority system aimed at managing the political life of black Africans. This network was overlaid by the formal organization of a strong state; while settler rule allowed parliamentary representation for whites (through a dominant one-party system) it forcefully restricted political activity by blacks. When an armed liberation movement eventually attained political independence for Zimbabwe in 1980, its leaders found advantage in preserving the apparatus of the colonial state, especially those coercive elements that enabled control over political opponents. Two examples will suffice. From 2000 onward, ZANU-PF revived the Joint Operations Command (JOC), originally conceived by the Rhodesian military as a directorate for prosecuting counterinsurgency and as a shadow cabinet of security chiefs for making strategic policy decisions. And, in 2002, the government promulgated a Public Order and Security Act (POSA) that, like the colonial-era Law and Order Maintenance Act (LOMA) severely restricted democratic freedoms; for example, by requir-

26

Power Politics in Zimbabwe

ing police permission for political gatherings and allowing the detention of political activists. In these ways, institutional legacies from the past were projected forward through time, providing a structural template for the perpetuation of exclusive authoritarian rule. Even though political settlements draw on deep structural roots, they are also subject to change. Indeed, settlements arise and adapt as a consequence of crisis, which is another way of saying at critical junctures. Often leaders only discover incentives to bargain with rivals when a deepening predicament leaves them with no other feasible option. A crisis may manifest in an enduring war that no side has the military capability to end, which is an apt depiction of the crisis prevailing at the tail end of settler rule in Rhodesia. A critical juncture may also take the form of a collapsing economy, as when a currency becomes worthless or basic public services cease to function, a situation brought about in postcolonial Zimbabwe by 30 years of economic mismanagement and institutional neglect. Either way, adversaries often come to recognize the need to reach an accommodation when faced with an irresolvable situation of military or human insecurity. Finally, a durable political order depends on the adaptability of political settlements, which are best conceptualized not as a single event but as a sequence of agreements. Political actors will always exploit ambiguities in institutional rules by trying to implement the settlement to their own advantage. While defenders of the political status quo will promote a narrow interpretation of the terms of the agreement, advocates of change will endeavor to enforce expansive interpretations of compliance. In so doing, each side may cite the law and appeal to the judiciary for a definitive ruling. Either way, through the dynamics of political struggle, the institutions envisaged in political settlements are usually subject to ongoing processes of incremental change. Moreover, the parties to a political settlement are unlikely to “get the institutions right” on the first attempt. Instead, they usually focus on the art of the possible, putting in place a piecemeal agreement that addresses only the most pressing disputes. But institutions designed to end a civil war or resolve a violent electoral dispute may not be well suited to consummating a democratic transition or constructing an effective state. For example, whereas power sharing might be the right approach for ending armed hostilities, power division based on an institutional separation of powers may be more appropriate for ensuring democratic accountability. Moreover, if arrangements to share political power fail to address issues of economic inclusion or the role of the security sector, the interests of key elites may remain unaddressed. Thus, subsequent negotiations are needed; it becomes necessary to revisit the original settlement and revise its terms in what has sometimes been called a “post-settlement settlement” (du Toit 2003; Sisk 2010).

Political Settlements

Elite Political Culture

27

Political culture is equally important. As well as having interests, actors have values, beliefs, and preferences that motivate their involvement in political life (Almond and Verba 1965; Putnam 1971; Diamond 1993; Inglehart and Welzel 2005). Actors, elite as well as mass, justify their political actions in terms of what they consider to be right or just. At the heart of any political value system is a theory of legitimacy that validates why a particular elite has a right to rule. On the one hand, authoritarians tend to claim that, as a vanguard group, they possess some kind of superior virtue; as such, they should serve as guardians of the public interest. On the other hand, democrats usually point to the peak value of political equality, which they claim endows every citizen with a right to participate in choosing leaders and policies (Dahl 1989). Thus, political values prove helpful in explaining why different sorts of elites opt for exclusive or inclusive political settlements. Where does elite political culture come from? And how does it shape political settlements? This study emphasizes the potent effect of common political experiences as the formative factor in the creation of elite cultures. Political leaders develop mutual values by engaging in collective action, often in young adulthood or when mounting resistance to an unpopular political regime. Facing a common adversary such as an entrenched authoritarian government, opposition movements develop a shared worldview and ties of solidarity that potentially persist over time. If opposition movements ever attain state power, they carry with them into office the values that provided group cohesion in the first place. Moreover, rulers are prone to govern according to the same code of conduct—whether by ballot or bullet—that initially brought them to power. In Zimbabwe, the elite coalition that led the country to national independence shared a common political culture that originated from several strains of authoritarianism (Masunungure 1998, 2004; Sachikonye 2011). First, based on precolonial political precedents, the new leaders gravitated toward new forms of chieftaincy and gerontocracy—rule by elders. In searching for authentic institutions, they cultivated the convenient expectations that citizens would rally around a dominant party and that leaders would govern for life. Second, most members of the incoming elite had suffered under the heavy hand of colonial repression, having been detained, jailed, or exiled for daring to organize nationalist resistance. These experiences imprinted in their minds an appreciation of the overwhelming power of the modern state, especially insofar as the law and military force could be used to stifle opposition political activity. Third, and most important, Zimbabwe’s ruling coalition was forged in the crucible of a national liberation war. The country’s guerrilla movements

28

Power Politics in Zimbabwe were not structured democratically. Authoritarian militarism was the chief and common characteristic. . . . The movements paid scant attention to issues of individual and civic rights . . . and both advocated an implacable internal unity. The liberation struggle was fraught with intense intrigues, factionalism, violent purges and assassinations. . . . There was a lot of witch-hunting, intimidation and torture, “enemies” being summarily dealt with. (Masunungure 2004, 150–151)

The liberation war was seminal in several respects. It gave birth to a coalition in which civilian and military elements were in periodic tension over political leadership. Indeed, at key moments in the life of Zimbabwe’s dominant party—including struggles over leadership choices, survival, and succession—the military was more likely to control civilians than vice versa. The tense atmosphere of the liberation struggle encouraged a polarized outlook among leaders in which the political world was divided starkly between a small circle of trusted confidants and a hostile landscape full of implacable “enemies.” Secrecy and loyalty were valued above all. This distrustful worldview was conducive to factionalism, splits, and purges. Indeed, an inward-looking “laager mentality”3—which ironically mirrored the defensive stance of the severely outnumbered white settlers of colonial Rhodesia—was not conducive to political inclusion. In addition, the war also ensured that the ruling elite readily resorted to violence as a standard operating procedure. The logic of power politics suggested that the ends (control of the state) justified the means (political violence), which, in turn, was justified in terms of revolutionary values. Whenever the dominance of Zimbabwe’s new rulers was threatened, they defaulted to coercion—using both formal state forces and informal party militias—as a political trump card. The party inculcated a culture of violence by transmitting to young people an ideology of chimurenga (revolutionary war) as a just form of warfare. By encouraging violence, however, rulers exposed themselves to the risk of prosecution for rights abuses, either in national tribunals or an international criminal court. As such, repressive tactics have usually been followed by leaders’ efforts to co-opt or emasculate institutions dedicated to legal redress. In short, elites with blood on their hands have sought self-preservation by dismantling the rule of law. Finally, the top leaders in independent Zimbabwe used the sacrifices of the guerrilla fighters—“we died for this country”—as the ultimate justification for their own political and economic entitlement. As liberators, they claimed to own Zimbabwe in the fullest sense of the term; namely, that the country belongs to them and not to anyone else. They traced political legitimacy not to universal political liberties and open elections—whose procedures, results, and validity they readily dismiss—but to an armed victory in a liberation war. Thus, the leaders of a vanguard party have won not only a right to rule in perpetuity; they are also entitled to seize the nation’s wealth

Political Settlements

29

as they see fit. On the other side of the coin, leaders deemed not to have partaken in the liberation war, or not to have done so with sufficient vigor, are seen as permanently ineligible to enjoy the political and economic fruits of independence. In fact, by challenging the ruling coalition, they reveal themselves as “sellouts”; that is, agents of imperialism. * * *

The chapters that follow explore the evolution of political settlements in Zimbabwe. I trace the agency of successive political elites within a structure of institutional precedents and a context of cultural parameters. The narrative reveals that political settlements are rarely forged from scratch; instead their design often builds on past institutional and cultural practices. As a result, the internal dynamics of power sharing can differ quite substantially from the letter of the law in formal agreements. Instead, the history of political settlements in Zimbabwe—up to and including the power-sharing agreement of 2008—continuously reverts to power politics. In this regard, the so-called inclusive government of 2009–2013 was inclusive in name only. A façade of power sharing concealed a persistent reality of power capture and blocked the possibility of power division via democratization. A narrow coalition of civilian and military leaders with roots in the country’s liberation struggle continued to turn the instruments of the state toward controlling elections and looting the country’s wealth. Far from sharing power with a democratic opposition, ZANU-PF elites sacrificed democracy, state building, and economic development for their own political survival.

Notes

1. The German sociologist Robert Michels (2001) proposed this law, in which one form of tyranny is replaced with another and successive governments intensify the abuses of their predecessors. 2. Mahoney and Thelen (2010, chap. 1) explain the prospects for institutional change in terms of the relative power balance between “predators,” who seek to use institutions solely in their own interests, and “subversives,” who try to change institutions from within. I take up this distinction in the conclusion. 3. “Laager” is an Afrikaans word in common currency in Southern Africa that refers to a defensive circle of wagons.

Part 1 The Roots of Repression

3 The Colonial Political Settlement

The colonial encounter in the territory of Southern Rhodesia brought together groups of people who failed to comprehend each other. On the one hand, indigenous societies enjoyed systems of chiefly rule based on kinship lineages and ancestral ties to land. On the other hand, colonizers introduced models of government derived from the British Westminster experience that featured territorial boundaries, a unitary state, an elected parliament, written laws, and property rights, especially over land. Thus conditions were hardly conducive for arriving at a shared political settlement. Instead, the scene was set for mutual misunderstanding and institutional incongruence; the political rules of the colonial state failed to match the customs of the colonized. Despite these differences, precolonial and colonial rulers faced the common challenge of establishing political authority. Thus, by dint of circumstance, they tended to develop similar repertoires of governance. Indeed, indigenous chiefs and white settlers alike favored the construction of institutions that distributed resources preferentially to their own followers. Successive rulers relied on the material incentives of patronage, whether by collecting and redistributing tribute in the precolonial era or by allocating land and paid employment as society modernized. In each era, rulers also created ideologies to justify their right to rule, ranging from claims of access to rainmaking spirits or a Christian mission to spread commerce and civilization. The most common factor across historical periods was the use of force. Every form of government relies, to some degree, on specialists in violence to subdue rivals and to bring outlying groups within the orbit of a political community. To maintain a contingent of men under arms—whether a warrior band, an expeditionary police force, or a standing army—rulers must find the means to feed them and to acquire military technology. In Southern

33

34

The Roots of Repression

Rhodesia and Zimbabwe alike, surpluses from agriculture and mining—including iron, gold, and, later, diamonds—have repeatedly provided the wherewithal for enforcing political authority. So, even as colonizer and colonized devised distinctive political institutions, they responded to common challenges within a shared material environment. These historical approaches to governance, in turn, created institutional legacies—especially coercion and patronage—that are echoed in rulers’ decisions and methods today. The main line of cleavage and competition in the colonial period was between a tiny white ruling group and a vastly larger populace of black subjects. Complicating the interaction in Zimbabwe, however, were internal rivalries among elites on both sides. Within African society, political and cultural blocs took on distinctive identities that even an emerging nationalist solidarity could not erase. And, as nationalist sentiments arose, the colonizers became divided between European settlers and the metropolitan government over the direction of the country’s political development. The short century of colonial domination (1896–1979) started and ended with political violence. In between, a colonial political settlement, while relatively durable, never gained full political legitimacy; ultimately, race-based minority rule proved unsustainable. Especially after World War II, the colonial history of Southern Rhodesia became a marathon series of attempts to renegotiate the political settlement and reconcile the divergent interests of the British metropolitan authority, the white settlers, and the increasingly radicalized African nationalist movements. A short-lived experiment in power sharing—the so-called Internal Settlement of 1978—was little more than a last-ditch effort by white settlers to govern through conservative black agents. In the end, independence under majority rule proved possible only when the settlers concluded that they could no longer bear the costs of containing the rising tide of political violence.

Precolonial Foundations

The African occupants of the forests, river valleys, and grasslands of what is now Zimbabwe lived mostly in scattered, small-scale village communities (Mazarire 2009; Ranger 2010). Distinctive cultural groups coexisted in an environment of shifting boundaries, population migrations, cross-village barter, and low-level warfare that featured raids to seize control of cattle, food, and women. Heads of family lineages gained power through marriage alliances, offering sisters and daughters to young men in exchange for labor and security. The institutions of proto-states sometimes arose through the vertical integration, trade expansion, and militarization of these patrimonial networks.

The Colonial Political Settlement

35

Great Zimbabwe, from which the modern country takes its name, was characterized by the distinctive stone structures whose remnants are found today from Botswana to Madagascar. This civilization, which reached its peak in the fourteenth century, probably collapsed as a consequence of natural environmental change. Its peoples dispersed northward across the highveld plateau that forms the spine of present-day Zimbabwe and into the Zambezi lowlands. Mazarire describes the process of state formation among these groups in terms of the expansion of patronage: Lineage heads transformed . . . their son-in-law or vakushwa, normally from unrelated totemic groups, into councilors (makota) and their totemic kinsfolk into machinda, or governors. . . . When this arrangement was replicated more widely it formed a hierarchy growing in scale from village to district to provincial level. . . . In the plateau region, where the control of people became more important than the control of material objects, these vaNyai became . . . visible evidence of their patron’s followings. (2009, 13)

By aggregating patron-client networks, the Mutapa dynasty rose to dominance in the fifteenth century in the northern highlands and Zambezi valley. But this state-like system could not resist the combination of border rebellions and Portuguese military force. Among the terms of a political settlement imposed by the Portuguese, the munhumutapa (king) was required to sign away land rights and mineral concessions and to permit forced labor. Only by the turn of the eighteenth century were successive dynastic leaders able to drive the Portuguese off their mines and estates and to revive the vaNyai system of federated clientelism. One of these successors established the Rozvi dynasty. Located in the southwest portion of present-day Zimbabwe, this line of chiefs became a distinctive ruling elite based on military prowess and claims of magical skills. As a political class operating a centralized state, the Rozvi kings parceled out land to loyal clients in return for tribute, and they extracted labor from vassal groups in return for protection. But the state’s expansion proved to be its undoing, as village leaders on the widening periphery found opportunities to escape the center’s scope of control. By the eighteenth century, even the heart of the Rozvi state itself had begun to dissolve amid political factionalism, court intrigues, and predatory militarism. The resulting political vacuum allowed new groups to come to power by virtue of superior weapons: in this case, the short stabbing spear. Waves of Nguni warriors and refugees, dislodged by the mfecane—the great scattering—south of the Limpopo, arrived in or migrated through the territory in the early nineteenth century. The mfecane was a period of widespread social violence that originated in the consolidation of the Zulu nation under Shaka and spread northward into what was then the Transvaal under Mzilikazi, potentate of the Ndebele peoples. These regimes specialized in

36

The Roots of Repression

killing, property destruction, and forced population movement, all with the objective of stifling political rivals. By the mid-nineteenth century, the immigrant Ndebele under Lobengula (Mzilikazi’s son), who had settled in the area around what is now Bulawayo in southwestern Zimbabwe, effectively subjugated the various peoples that spoke dialects of Shona. The span of the Ndebele state almost matched the scope of the societies it displaced, although it never fully penetrated all of present-day Mashonaland. In so doing, however, Lobengula’s hierarchical military establishment found ready parallels with the preexisting structures of the Rozvi houses. Even though the identity of the ruling elite became explicitly ethnic, Mazarire notes: The Ndebele were careful not to disturb existing non-secular structures, and in some cases invited back the defeated Rozvi as loyal subjects. Thus, even as they pursued their enemies up and down the plateau, this did not stop them deferring to Shona spirit mediums  .  .  .  to whom they paid tribute. Similarly . . . Lobengula solidified his relations with the Mwari cult in order to become an important rainmaker himself. (2009, 33)

For precolonial peoples, incorporation into a centralized state was an intermittent exception rather than a daily rule. Cultural identities remained fluid, especially at the margins. For example, peoples who saw themselves as distinctively “Shona” or “Ndebele”—terms used to signify a coherent linguistic and political identity—did not crystallize until the later years of the nineteenth century and then largely as a product of—and reaction against—waves of black and white immigration from the south, including colonial settlement and administration (Ranger 1985). Finergrained cultural identities among Shona-speakers—such as Zezuru, Manyika, and Karanga—remained politically salient. And, except when chiefs called for tribute or labor or when state predation spilled into the periphery, the daily reality of politics remained local, with village headmen and other elders remaining the preeminent figures in the lives of a rural population.

The Colonial Settlement

The arrival of an expeditionary column mounted by the British South Africa Company (BSAC) in 1890 and the declaration of a British protectorate in the following year marked the onset of a founding political settlement. Though Cecil Rhodes had grand imperial ambitions, his immediate motivation was commercial, since the country was said to hold gold wealth on the same glittering scale as the Witwatersrand in South Africa. As a means of gaining entry, Rhodes’s agents secured concessions granting monopolies over mining

The Colonial Political Settlement

37

rights from Lobengula and other chiefs. But the parties to these agreements had divergent understandings of what was entailed (Hole 1926). The company wrongly assumed that the Ndebele king enjoyed undisputed control over the vast northern territories, including the Mashonaland provinces. And Lobengula surely did not intend to undermine his own sovereignty by ceding to the newcomers extensive uncompensated rights to land, settlement, and administration. The establishment of a European outpost at Fort Salisbury (now Harare) was backed by a mounted infantry unit that later became the British South Africa Police (BSAP). By 1893, following an armed clash with Ndebele warriors near Fort Victoria (now Masvingo), a force composed of company police and settler militiamen drove Lobengula from the territory. Shortly thereafter, the so-called Matabele and Mashona rebellions laid siege to Bulawayo and Salisbury in 1896. Despite energetic resistance, sometimes aroused by Mwari spirit mediums, the settlers used their superior weaponry, now the Maxim gun, to suppress the uprisings (Ranger 1967; Cobbing 1977; Beach 1979; Ndlovu-Gatsheni 2009). From the outset, therefore, the colonial state was forged in conflict and conquest. And the colonial regime—even in its civilian functions, especially policing and native administration—embodied a militaristic ethos. Initially based on a Royal Charter (1889) for commercial exploration, BSAC rule was formalized by the Southern Rhodesia Order in Council (1898), the country’s first constitution. It provided for an administrator recruited from the colonial civil service, a small legislative council that included settler nominees, and a British army officer to command the paramilitary police force.1 The company’s main aim was to generate enough profit to reimburse shareholders for the costs of early investments in railways and other infrastructure. When a gold rush failed to pan out, the company turned to agriculture as an alternative foundation for the colony’s economic development. The administrator and council set about seizing arable land for white occupation, moving Africans into lower-quality native reserves, and putting in place a network of policies to underwrite commercial farming. The BSAC’s effort to harvest returns from its new agricultural strategy was predicated on the cooperation of the settlers. In return, the latter demanded a share of political representation and, eventually, an end to company rule. In reality, the BSAC was anxious to rid itself of its unprofitable venture in Southern Rhodesia. Thus, once a British court ruled in 1918 that the land belonged to the Crown, the company gladly relinquished responsibility for public administration. A 1922 referendum on the country’s constitutional future asked the settler community—then just 35,000 strong—to choose between joining the Union of South Africa or seeking responsible government. They opted to govern themselves.

38

The Roots of Repression

The Southern Rhodesia Constitution Letters Patent of 1923 was a historic political settlement between the British government and a small group of European settlers. The terms for the establishment of a British colony— which shifted the balance of power from external to internal elites—remained in effect without major changes for the next four decades. The new rules of the political game provided the colonists with a free hand to pursue their interests with a minimum of interference from abroad. Formally, the British government retained reserved powers to protect Africans against discrimination under the law, but these were seldom, if ever, used (Palley 1966). Africans themselves were not consulted about the terms of the settlement and lacked even token rights of representation. In this regard, the colonial political settlement was highly exclusive. Once the settlers had captured state power, they used it to build an economy based on privilege and patronage. The Land Apportionment Act of 1930 entrenched an uneven distribution of land that prevented Africans from purchasing property in designated European areas or in towns. In addition, white farmers enjoyed subsidized inputs for certain crops—for example, to launch a tobacco industry—and protected markets that blocked African farmers from selling foodstuffs, including maize. At the same time, taxes were imposed to drive Africans out of smallholder and subsistence agriculture and into working for commercial farmers. Africans were also prohibited from prospecting and mining. Instead, the government provided cheap electricity and tariff protections to encourage white immigrant entrepreneurs to enter minerals production and manufacturing. Through a network of public enterprises—ranging from coal, iron, and steel to the processing and marketing of meats, sugar, and cotton to railway and airline corporations—the state came to dominate the economy. While blacks faced stagnant incomes and declining living standards in overcrowded native reserves, whites became accustomed to full employment and comfortable Western lifestyles. Under the Industrial Conciliation Act of 1934, skilled jobs, along with rights to unionize, were reserved for persons of European origin, as were most educational, health, and other social services. As such, the very emergence of the race-based elite depended on an apparatus of political and legal controls; given the working class origins of many immigrants, Rhodesians were elite only in the sense that they reserved the higher paying jobs for themselves. Especially after World War II, they differentiated themselves from the British—whom, under Labour governments, they regarded as socialist and decadent—with an ideology of “civilized standards” that mixed pioneering individualism with the conservative values of a disappearing imperial age. The political system was split along lines that segregated white citizens from black subjects (Mamdani 1996). The settler community enjoyed political representation in Westminster-style parliamentary institutions, includ-

The Colonial Political Settlement

39

ing cabinet government, an elected legislature, an independent judiciary, and a professional civil service. Given the small size of the settler population, members of the political, economic, and military elites came from similar backgrounds and were mutually acquainted; they easily fell back on old school and sports club ties to conduct the country’s business. Africans, however, remained excluded from political life. Instead, a Native (later Internal) Affairs Department governed the bulk of the population directly through a phalanx of rural district officers with discretionary powers over land allocation, tax collection, population movements, and civil and criminal law (Weinrich 1971). Chiefs and headmen—designated less in terms of their local social standing than on their political pliability—were deputized as salaried agents of the central state. The style of governance was paternalistic: it ignored the citizenship aspirations among an increasingly urbanized African population by assuming that peasants would be content to use chiefs as their channel of representation. In practice, governance in colonial Rhodesia produced distinctive institutional legacies. First, interest groups in commercial agriculture exerted strong political influence that was disproportionate to the number of white farm families, which never exceeded 6,500. Not only was the electoral map biased toward rural voters, but the public institutions that regulated agriculture—from price-setting procedures to marketing boards—were controlled by farmer representatives. In this way, the national farmers union captured part of the state, as did the chamber of mines and confederation of industries in their own sectors. Second, the political party system produced de facto one-party dominance. Although, over time, settler politics generated a multiplicity of political parties with changing names, the underlying pattern was stable. From 1923 to 1962, one party (known successively as the Rhodesia Party, the United Party, the Federal Party, and the United Federal Party) controlled the levers of government; with no alternation of the faction in power, the regime amounted “essentially . . . [to] a one-party system” (Leys 1959, 133). Together, the ruling party, economic interest groups, and government agencies established reinforcing roles to further the economic goals of the settler community.

Toward Mass Nationalism

Organized political resistance was slow to get off the ground. The Southern Rhodesian Bantu Voters Association, formed in the 1920s to respond to the advent of responsible government, involved a minuscule class of educated Africans. The aspirations of this emerging middle stratum of black teachers, lawyers, nurses, and entrepreneurs did not always coincide with those of ordinary people, whom they held at a social distance. Until the late 1950s,

40

The Roots of Repression

African leaders sought inclusion in the prevailing political settlement by calling for participation in the electoral and economic life of the colony. Their petitions for fair governance—that is, the repeal of discriminatory laws—fell on the deaf ears of a disengaged British government that failed to address the fact that “the mass of the population in Southern Rhodesia were politically inert, passive, and virtually powerless” (Gray 1960, 167). To be sure, there was widespread popular resentment about rural land expropriation, hut taxes, and restrictions on urban settlement enforced by the requirement that every African carry an identity document (“pass”). The Native Land Husbandry Act of 1951—which mandated individual land tenure, soil conservation, and de-stocking of cattle—undermined chiefly authority and prompted outbreaks of protest against the native authority system. At the same time, a railway workers strike in Bulawayo in 1945 and a general strike in 1948 over falling wages and poor working conditions signaled that trade unionists would come to play a political role. “This was not yet a nationalist moment, however, because Africans were still thinking in terms of an amelioration of white rule rather than sovereignty and independence” (Mlambo 2009, 98, emphasis in original). Key disputes concerned franchise expansion and national independence. Was either politically feasible? And which would happen first? Leaders of the settler community consistently resisted what they saw as interference from Britain, which they feared would favor African interests; instead, they pushed for dominion status under a settler government along the lines of Australia and Canada. By contrast, for many years African political leaders called for including more people in the country’s qualified voting system, only later expanding their demands to include national independence under majority rule. Conflicts over these constitutive issues led to cosmetic modifications of the colonial political settlement. The creation of the Central African Federation in 1953, which temporarily joined Southern Rhodesia with Northern Rhodesia and Nyasaland, prompted amendments to the Southern Rhodesia Electoral Act that enfranchised several thousand of the wealthiest and besteducated Africans. In a timid step toward majority rule, more token voters were added in 1961 on a segregated B-roll for 15 African seats in parliament. In negotiating the 1960 Constitution, however, the British relinquished virtually all imperial powers in return for a truncated declaration of rights and weak oversight by a multiracial council. These electoral and constitutional reforms in no way vitiated white political control but instead extended southern dominance over the copper wealth of Northern Rhodesia. Settlers in Southern Rhodesia opted for the Central African Federation and accepted the gesture toward power sharing mainly as a bid for dominion status within the Commonwealth. For their part, African nationalists were deeply suspicious of the federalists’ promises of multiracial “partnership”

The Colonial Political Settlement

41

that purported to replace the old system of native administration. This vague settlement, described by one commentator as “a sophisticated cover for what was in fact a hard political bargain” (Bowman, 1973, 18), never promised more than a half-share of power, and then only after endless decades of political apprenticeship. Indeed, in a trenchant analysis, Colin Leys drew “one clear lesson” about the durability of the colonial political settlement: The transfer of effective power to an immigrant community, which was largely dependent on a discriminatory legal apparatus for its very presence in the country, precluded the possibility that that power would later be voluntarily shared with the rest of the population, or that the apparatus would be dismantled. (1959, 294, emphasis added)

While mass opposition to the Central African Federation provoked demands for majority rule and political independence in Northern Rhodesia and Nyasaland, the reaction was more muted in Southern Rhodesia. Only in 1957 did Africans form a first national political party, the Southern Rhodesian African National Congress (later simply the ANC) under the leadership of Joshua Nkomo. While the ANC quickly attracted a mass membership, many in the African elite still pinned their hopes on the illusory promise of “partnership” and continued to take part in white-led organizations such as the United Federal Party (West 2002). But the settler regime revealed its intrinsically coercive character in 1959 when Prime Minister Edgar Whitehead declared a national state of emergency, banned the ANC, seized its assets, and detained 500 of its members. Other parties immediately sprang up to take the ANC’s place—first the National Democratic Party (NDP), then the Zimbabwe African People’s Union (ZAPU)—but each was in turn outlawed and its leaders imprisoned. The upshot of state repression was the emergence of mass nationalism. Based on urban and rural grievances, African political leaders finally made a clear demand for political independence under majority rule.

The White Rebellion

For all intents and purposes, the British government had long conceded the principle of settler political domination in Southern Rhodesia. The metropolitan power declined to intervene in the country’s internal political affairs even as the settlers embarked on a systematic campaign to eradicate the nationalist threat. But British colonial policy toward the rest of Africa began to change in the late 1950s when faced with the Mau Mau emergency in Kenya, the march to political independence in Ghana, and growing nationalist protest in Nyasaland. The 1960 Monckton Commission concluded that

42

The Roots of Repression

hostility between the races in Central Africa was so great that the Central African Federation could be maintained only by force or massive constitutional revisions. Preferring to renegotiate the existing political settlement, the British therefore recommended broadening the franchise in Southern Rhodesia, abolishing discriminatory laws, and instituting self-government for Nyasaland and Northern Rhodesia. In fact, the latter territories soon attained full political independence—as Malawi and Zambia in 1964. These groundbreaking developments provoked a hardening of attitudes in settler circles further south. The government of Southern Rhodesia put in place a battery of repressive legislation, culminating in the Law and Order Maintenance Act (LOMA) of 1960, which prohibited political meetings, banned or censored publications, expelled journalists, allowed warrantless searches, legalized detention without trial, and imposed draconian prison sentences for a range of political “offenses.” In 1958, the ruling United Federal Party had already deposed the prime minister of Southern Rhodesia, Garfield Todd, for supposed liberal sympathies toward African advancement. Federal Prime Minister Roy Welensky berated the British for betraying white interests in Central Africa, accusing them of losing the will to govern. Most importantly, the electorate in Southern Rhodesia moved to the right in 1962 by electing a conservative opposition party, the Rhodesian Front (RF), on an anti-federation, whitesupremacy platform.2 The electoral turnover occurred in a turbulent context: Malawi and Zambia were approaching independence, all nationalist parties were outlawed in Southern Rhodesia, and dispossessed whites were fleeing Katanga in the aftermath of postindependence conflict in the Congo. The RF made the most of this heightened political atmosphere, warning that Britain could no longer be trusted to uphold its end of any political settlement. A parallel radicalization was under way among the African nationalists. When Nkomo indicated a willingness to accept the 1961 Constitution’s 15-seat formula for African representation (versus 50 seats for whites), he was forced to backtrack in a hail of criticism. Later, his illconsidered plan for a government-in-exile prompted a split in the party, with Ndabaningi Sithole forming the breakaway Zimbabwe African National Union (ZANU). Even though most Ndebele speakers stuck with Nkomo, the cleavage in the nationalist movement was not purely ethnic. Nor was it based on divergent political goals, since all leaders espoused anticolonial and pan-African ideals. The difference was more a matter of temperament, having to do with the perceived willingness of leaders to stand up and fight (Sithole 1999; Scarnecchia 2008). The government’s decision in 1964 to detain Nkomo, Sithole, and others, including Mugabe, made it clear that all peaceful avenues of political advancement were closed. Excluded from a process of negotiation, the country’s nationalist leaders therefore faced the unpalatable choice of shelving their political

The Colonial Political Settlement

43

demands or resorting to armed struggle. While ZANU and ZAPU opted for a military strategy, it also became clear that they would fight separately and, perhaps, even against each other. The colonial political settlement between Britain and the white settlers broke down on November 11, 1965, when Prime Minister Ian Smith and 15 cabinet colleagues signed a unilateral declaration of independence (UDI). This illegal rebellion consummated the capture of state power by settler elements in a renamed state—Rhodesia. It was preceded by a flurry of failed negotiations in which London made a bid to insist on five principles—unimpeded progress toward majority rule, guarantees against retrogressive constitutional amendments, improvements in the political status of Africans, progress in ending racial discrimination, and assurances that any new settlement was acceptable to the people as a whole.3 The settler government in Salisbury rejected all these terms and instead prepared to go its own way. In a quest for legitimacy, the RF induced an indaba (conference) of chiefs to express support for the impending UDI and gained near-unanimous support from whites in a 1964 referendum and 1965 election. Opposition came only from the handful of African voters; the head of the Rhodesian army, who opposed a rebellion against the Queen (he was subsequently dismissed); and some tobacco farmers and other business elites who feared economic isolation. The UDI excluded not only the African nationalists, but also the former colonial power. British judges ruled that the rebellion was illegal but the Rhodesian courts granted the government first de facto, and later de jure, recognition. Britain was reduced to a weak, punitive role; having forsworn military intervention, it could exert pressure only by imposing economic sanctions. Along with other Commonwealth countries, the United Kingdom froze Rhodesia’s overseas assets, imposed exchange regulations, and blocked imports of sugar and tobacco. When these measures failed to return the rebel regime to constitutional rule, Britain called upon the United Nations to impose comprehensive mandatory embargoes on arms and oil. But the UDI regime survived in part because fellow white settlers in Portuguese Mozambique and apartheid South Africa colluded in evading the embargoes. Furthermore, sanctions unintentionally stimulated growth and diversification in the country’s small manufacturing sector. Continuing efforts to end the settler rebellion via negotiations were fruitless. Even when Britain compromised, the Rhodesians, fearing a trick, could not bring themselves to accept a new settlement. Instead, the settlers declared a republic in 1970, severing ties with the Crown. A seeming breakthrough in the impasse occurred in 1971, when a new Conservative government in London offered a deal in which a property-qualified franchise would eventually (time unspecified) lead to parity of white and black seats in parliament, with half of the latter elected indirectly by chiefs. The British willingness to sacrifice the principle of progress to majority rule was eased

44

The Roots of Repression

by a requirement that any agreement should be broadly acceptable to the people as a whole. A visiting commission in 1972, headed by a retired high court judge, Lord Pearce, found that Africans—quickly mobilized by an African National Council headed by Bishop Abel Muzorewa and backed by ZANU and ZAPU—roundly rejected the proposed settlement. Even the chiefs were divided, thus exposing the fiction that traditional leaders spoke for all Africans in support of the RF government.

The Independence War

With an inclusive political settlement more elusive than ever, the country sank into a period of violent struggles. The two leading nationalist parties developed their own armed units. Fighting for ZAPU, the Zimbabwe People’s Revolutionary Army (ZIPRA) trained in the Soviet Union and were based in Zambia. ZANU had the Zimbabwe African National Liberation Army (ZANLA), which trained in China and operated from Mozambique. Throughout the 1960s, the two guerrilla armies engaged in sporadic acts of sabotage and isolated firefights with Rhodesian troops. They were plagued by poor logistics and did not coordinate with one another. While ZANU came to favor popular political mobilization, ZAPU stuck with more conventional military tactics (Martin and Johnson 1981). A faction of fighters from both wings tried to unify the guerrilla movements under a Zimbabwe People’s Army (ZIPA) while simultaneously marginalizing the political elites of the two movements. The ZIPA initiative illustrated the pervasive jockeying for supremacy between military and civilian leaders, a feature that later came to characterize postcolonial party politics. The war setting also established the precedent of prohibiting dissent within leadership ranks, with strict disciplinary measures—from expulsion from the party to physical assassination—being leveled at individuals deemed “sellouts” (Mhanda 2011). The war escalated between 1972 and 1979, with atrocities mounting on both sides. The guerrillas attacked isolated farmsteads, ambushed enemy convoys, and destroyed the outlying infrastructure of the settler state, including schools, bridges, and the offices of district administration. The Rhodesian state’s counterinsurgency response ranged from search-anddestroy missions (by mobile Selous Scouts and Fire Force units), through assassination plots against exiled nationalists leaders (like Herbert Chitepo and J. Z. Moyo), to large-scale bombing raids on guerrilla rear bases in Mozambique and Zambia (Moorcraft and McLaughlin 2008). Caught in the middle of the fighting was the rural population (CCJPR 1975, 1978). While rural dwellers who sympathized with the goals of national liberation were keen to provide the guerrilla fighters with food, shel-

The Colonial Political Settlement

45

ter, information, and support services, there is ample evidence that ZANLA and ZAPU used coercion—including summary executions, beatings, torture, and rape—to extract political and material support from unwilling segments of the population (Kriger 1992). Although some young people eagerly served as auxiliaries to the fighters, others had to be press-ganged to abandon their studies and report for military training. During this period ZANU developed its trademark political mobilization techniques featuring all-night indoctrination sessions (pungwes) that castigated not only the Rhodesians but also rival guerrilla movements. Motives were mixed on the other side, too. Some Africans served willingly in the Rhodesian African Rifles (RAR) and the BSAP or as district security assistants attached to the civilian administration. But many more resented being herded into so-called protected villages, subjected to collective punishments, and held under curfew to prevent contact with the guerrillas. Indeed, the brutality of the Rhodesian counterinsurgency campaign in the late stages of the war made a mockery of the settler government’s claims to uphold civilized standards. By the end of the 1970s, the adversaries had arrived at a standoff: the state security forces enjoyed a secure base in the cities, while the guerrillas controlled large swathes of the countryside. Though the momentum was with the insurgents, neither side was positioned to secure a decisive military victory any time soon (Davidow 1984). Rather, ordinary blacks and whites alike were so weary of war that their leaders began to see advantage in a negotiated settlement. For the Rhodesians, international sanctions had finally started to bite, causing debilitating shortages of fuel and spare parts. Faced with an unsupportable war budget, escalating taxes, and military conscription for all able-bodied males, the settlers were losing their comfortable way of life. Meanwhile, the guerrilla armies suffered from extended supply lines that often left the rank and file short of food and equipment. Moreover, personal, ideological, and ethnic struggles among political leaders and military commanders often played out in cruel disciplinary crackdowns. Finally, the frontline states of Tanzania, Zambia, Botswana, and Mozambique, fatigued with divisive squabbles between and within the liberation movements and concerned about the mounting costs of the conflict for their own economies, pushed ZANU and ZAPU to shelve their differences and enter talks with the RF regime. Indeed, external third parties drove the effort to settle the liberation war. US Secretary of State Henry Kissinger reacted to Cuban and Soviet interventions into Angola by reversing US support for white rule in Southern Africa; in 1973 Washington repealed sanctions-busting legislation and pledged to help African states resolve the Rhodesian crisis (Lake 1976). President John Vorster of South Africa came to see the Rhodesians as a dispensable liability in his country’s efforts to reach out economi-

46

The Roots of Repression

cally to black-ruled states. From 1974 onward, he began to work with President Kenneth Kaunda of Zambia to pressure their respective clients in the region to share power. While ZANU’s hand was strengthened by Mozambique’s achievement of independence in 1975, the party viewed with alarm the news of secret meetings between Nkomo and Smith in Lusaka, a flirtation with the Vorster-Kaunda détente that further damaged Nkomo’s reputation. At an all-parties conference in Geneva in 1976, the frontline states pushed ZANU and ZAPU into a “Patriotic Front”—a loose accord of convenience that fell well short of a merger and the United States and Great Britain forced Smith to concede that majority rule was inevitable. 4 Due mainly to discord among the nationalist delegations, the Geneva talks failed. While leaders in exile—Mugabe and Nkomo, whose guerrillas were making gains—were not ready to abandon the armed struggle, the internal leaders—Muzorewa and Sithole, who had no armies of their own— favored the restoration of British rule for a brief interim period leading to independence.

An Internal Settlement

By this time, it was becoming clear that the Rhodesians were losing the war, although the RF government refused doggedly to admit as much in public. Between 1976 and 1979, ZANLA and ZIPRA fielded ever-larger numbers of trained fighters with increasingly sophisticated weapons, effectively controlling rural areas during nighttime hours. As schools, clinics, dip tanks, and grinding mills were closed or destroyed, civil administration collapsed in the countryside. In the towns and commercial farming areas the white settler population suffered steep declines in morale. Emigration rates, especially among young people, climbed. In the face of these challenges, Smith’s last-gasp objective was “a power-sharing arrangement that would leave whites in a dominant role” (Mtisi et al. 2009, 148). He hoped that a black-white coalition government would project an image of racial partnership and prove acceptable to the international community. The Internal Settlement of 1978 also attempted to undercut support for the guerrillas by sharing the top offices of state, formally at least, among Smith, Muzorewa, Sithole, and Jeremiah Chirau, a conservative chief dragooned to supposedly represent the rural population. Not least because the “internal” parties (led by Muzorewa and Sithole) resented the Organization of African Unity’s (OAU’s) recognition of the externally based liberation movements (led by Mugabe and Nkomo), the latter were deliberately excluded. The institutional structure of the Internal Settlement, which created the hybrid state of Zimbabwe-Rhodesia, helped to cement in place a Westmin-

The Colonial Political Settlement

47

ster model of government. It vested sovereign authority in an elected parliament (and thus, effectively, in any political party that could control parliament) rather than in a written constitution (Southall 2013, 65). Thus persisted a legacy of winner-take-all that created incentives for political leaders to engage in power politics at the expense of the rule of law. Under the Internal Settlement, whites retained special representation under a separate voter roll in a House of Assembly (28 out of 100 members) and Senate (10 out of 30 members, with a further 10 seats allocated to chiefs). Each party in parliament also enjoyed proportional representation in the cabinet, headed by a prime minister, initially designed to rotate among the parties that signed the internal agreement. Following a 1979 election, Abel Muzorewa was installed as Zimbabwe’s first black chief executive, though the Rhodesian Front still controlled the civil service, held the loyalty of the armed forces, and had sympathizers in the local business community. As a result, Zimbabwe-Rhodesia failed to win international recognition. Nor did this uneasy attempt at racial partnership end the war, which cost an estimated 30,000 lives. As a failed experiment, the ill-fated Internal Settlement held important lessons for later efforts to use power sharing as an solution to power politics. The commitment of the Rhodesian Front to share power was never genuine or credible, especially in the light of Smith’s hard-line position that Africans would govern “never in a thousand years.” Instead, the settler regime saw power sharing as a useful device to buy time against a nationalist onslaught while still securing political dominance behind the scenes. If anything, the Internal Settlement implicated leaders like Muzorewa and Sithole in a regime that preached partnership but practiced power politics. After all, it was during the short life of Zimbabwe-Rhodesia that these leaders intensified the program of forced resettlement and collective punishment in the countryside and campaigned for election using the Rhodesian security forces and hastily recruited party militias (like Muzorewa’s Pfumo Revanhu). Through these formal policies and informal initiatives, violence became further embedded in Zimbabwe’s political culture. In August 1979, the Commonwealth heads of state and government met in Lusaka, where they resolved that Britain should resume its colonial responsibility for decolonizing Rhodesia. By way of response, newly elected British Prime Minister Margaret Thatcher convened a constitutional conference at Lancaster House, London, in September. Using shrewd and determined diplomacy, British Foreign Secretary Lord Carrington cajoled the two delegations—the Patriotic Front headed by Mugabe and Nkomo, and Zimbabwe-Rhodesia represented by Muzorewa, Smith, and Sithole—to come to terms. Neither side was fully satisfied with the outcome. To arrive at agreement, Presidents Kenneth Kaunda of Zambia and Samora Machel of Mozambique had to threaten the PF with withdrawal of rear-base support.

48

The Roots of Repression

And Smith, who sought to undermine Muzorewa’s leadership at every turn, had to be reined-in by his own delegation. The provisions of the Lancaster House Agreement and their implications for the maneuverability of Zimbabwe’s postcolonial government are analyzed in the next chapter. For the moment, suffice it to note that the agreement represented a landmark in the country’s political development. It signaled the temporary reassertion of British authority over the colony for the purpose of supervising a ceasefire, convening a universal suffrage election, and overseeing the transition to majority rule. As such, the Lancaster House Agreement amounted to a new political settlement that ended the colonial era and launched Zimbabwe on a path of internationally recognized independence.

Colonial Legacies

This chapter has viewed a brief history of colonial Southern Rhodesia through the wide-angle lens of political settlements. As argued earlier, political settlements—defined as elite agreements about the organization and exercise of power—shape processes of state formation, regime change, and conflict management. In this case, the colonial political settlement facilitated the formation of a strong state and diversified economy overseen by a racially based electoral authoritarian regime. The British agreed to take a hands-off attitude toward “responsible self-government” by allowing white settlers to capture power; they would govern not only themselves, but also the colony’s African subjects. The settlers created a parliamentary framework of institutions that effectively imposed discriminatory property and employment laws, protected agriculture and industry, and promoted stateled economic growth. This structure of power proved relatively durable, lasting over 40 years between 1923 and 1965. The staying power of the colonial political settlement can be traced to several factors. The parties to the settlement—the British government and the white settlers—both had incentives to bring an end to BSAC rule: the former to extricate itself from a costly economic commitment, the latter to gain the latitude to write their own rules for governing the country’s resources. The 1923 constitutional formula provided Britain with the convenient fiction that it was ultimately responsible for African interests while vesting the instruments of state in settler hands. Only with time would it become apparent that the British commitments were not credible and, once abandoned, difficult to regain. Moreover, the scope of the settlement was comprehensive insofar as, alongside political authority, whites gained control over the means of production (principally land and labor) and the means of coercion (namely the police force, the courts, and the Department of Internal Affairs). In short, the colonial political settlement was an economic and military pact as well as a political one.

The Colonial Political Settlement

49

The principal weakness of the colonial settlement, however, was that it was politically exclusive. African interests were barely considered. Even token government concessions that seemed to hold out the promise of power sharing—the “partnership” of the Central African Federation, the Internal Settlement—were insincere because settler leaders had no intention of yielding real power. Instead, the colonial regime personified power politics; it rested on an imposed institutional order backed up by coercion. Importantly, no elite political settlement was ever ratified by the population at large. Accordingly, the period of relative political stability between 1923 and 1965 was bracketed by periods of conflict. The colonial regime began and ended in violence. Its roots lay in a phase of occupation and direct administration at the hands of a commercial company and paramilitary police force chartered by the Crown. The conquest of the Shona chieftaincies and the Ndebele kingdom was achieved by force of arms and the theft of resources, resulting in the subordination of indigenous political and economic systems. By expropriating African land and cattle and restricting opportunities for education and employment, settler rule created an array of grievances that were eventually ripe for mobilization by African nationalist leaders. Because settler governments were unwilling to accommodate African demands for inclusion, many opportunities were lost to renegotiate the colonial political settlement by gradually charting a path to majority rule. Instead, political polarization became the order of the day. The biggest risk for any leader was to settle prematurely and thus be labeled a “sellout.” Just as moderate white leaders (like Todd and even Whitehead) were outflanked on the right in the 1950s, moderate black leaders (like Muzorewa, Sithole, and even Nkomo) were superseded on the left during the 1960s and 1970s. These moderates—who were backed by political and economic interests in South Africa, Britain, the United States, and some local business elites— were squeezed out of the settlement process. In the end, the colonial political settlement was abandoned by a right-wing settler faction that embarked on an illegal seizure of sovereignty in a quixotic bid to resist the inevitable erosion of white privilege. The intransigence of the Rhodesian Front, in turn, prompted the radicalization of demands for national liberation and increased the appeal of armed struggle. The war of independence added new layers of violence to a preexisting legacy of state repression. On the surface, the settler regime was modeled on a British system of government that included formal features like civil liberties, due legal process, and an independent press, at least for white citizens. But, as Africans began to demand political rights and especially as the rural insurgencies took hold, successive settler governments used executive decrees or the powers of a dominant ruling party to ram emergency regulations through parliament. The colonial government assembled an elaborate battery of security measures to censor the mass media, spy on society, imprison opponents without trial, and provide legal immunity to its

50

The Roots of Repression

own troops for extra-judicial killings. When Zimbabwe eventually became independent, the incoming government inherited a strong state whose coercive instruments were primed for repressive rule. By closing peaceful avenues of political representation, the settler regime helped to radicalize African nationalism, pushing it in a sharply militant direction. And Western countries drove the nationalists into the embrace of China and the Soviet Union by refusing their pleas to arm the fledgling guerrilla forces. By opting for a political strategy of armed struggle, ZANU and ZAPU made their own fateful commitments to violence, thus embarking on a slippery slope that had profound and lasting consequences. The commencement of a second chimurenga (revolutionary struggle) (the first was in 1896) marked the escalation of ruthless military tactics on a number of fronts: between the Rhodesian security forces and the guerrilla armies over the political allegiance of Zimbabwe’s rural population; between rival ZANLA and ZIPRA armies over contested operational zones within the country, notably in the Midlands province; and within each movement itself, especially between ZANU political leaders in exile and ZANLA fighters on the front lines. The revolutionary violence that subsequently came to underpin ZANUPF’s ideology and methods of rule can be traced directly to these precedents. And the characteristics of elite political culture described in the previous chapter—defensiveness, illegality, cruelty, and impunity—were direct products of the leaders’ turn to violence as a standard operating procedure. The liberation war was the source of the exclusive view of the political world in which ZANU-PF saw itself as the only entity entitled to rule Zimbabwe. This view excluded even PF-ZAPU (as ZAPU was renamed), which ZANU-PF leaders regarded as too eager to make political compromises and to share power in a moderate political settlement.

Notes

1. See “BSA Police History,” which emphasizes the consistently paramilitary character of policing in the history of Rhodesia. http://www.bsap.org. 2. This alternation did not disturb the de facto one-party system, since the RF easily won all subsequent whites-only elections. Note, however, that the civil service and not the political party managed the country’s affairs (Godwin and Hancock 1993, 59). 3. Under pressure from the Commonwealth, British Prime Minister Harold Wilson later added a sixth principle: NIBMAR—no independence before majority rule. 4. According to Martin and Johnson, “The principal obstacle to greater unity within the Patriotic Front was the question of political leadership. ZANU would not accept Nkomo and ZAPU would not accept Mugabe as overall leader” (1981, 306).

4 The Independence Political Settlement

Zimbabwe gained independence under leaders whose political sensibilities were shaped by the harsh experiences of racial discrimination, political detention, and the liberation war. They entered politics in the context of a colonial state with coercive and administrative institutions designed for centralized control. Even though independence was eventually secured by means of a negotiated settlement, ZANU-PF drew from the guerrilla struggle the principal lesson of power politics: that political power is rooted in military might. As Mugabe believed: “Our votes must go together with our guns; after all any vote . . . shall have been the product of the gun. The gun, which provides the votes, should remain its security officer, its guarantor” (1980, 12; see also Meredith 2003). The winning party of the independence election, already tied closely to its own military force, was well positioned to inherit a strong, authoritarian state from its colonial predecessor. Thus, ZANU-PF’s opponents worried about what lay ahead: The new black leader took over a white-constructed dictatorship with the panoply of secret courts, secret hangings, and emergency powers. Would Mugabe now turn those weapons the other way? If he did and the whites fled, what good would peace be to an economically prostrate country? (Moorcraft and McLaughlin 2008, 179; see also White 2003)

The elite pact reached at Lancaster House in London on December 21, 1979, involved an implicit bargain: blacks would ascend to positions of political leadership in a sovereign state while whites would continue to enjoy ownership rights in the economy. Because parts of this bargain were informal, each side had its own understanding of exactly what had been agreed upon and for how long it would remain in force, thus opening the future to

51

52

The Roots of Repression

misperceptions, backtracking, and conflict. Certainly, the independence settlement did not redress stark social and economic inequalities and therefore failed to fully address the aspirations of black Zimbabweans. As one commentator noted: The 1980 constitution . . . was as much about peacemaking as it was about drafting fundamental legislation.  .  .  .  The deeply entrenched protection of white-owned farms was necessary for peace, but has since become an occasion of much trouble. (Benomar 2004, 83)

Moreover, ZANU-PF made full use of its leading position to consolidate state power in its own favor and against the interests of political minorities. As soon as legally permissible, the government amended the independence constitution to strengthen the hand of the political executive and to remove restrictions on land transfers. Thus, the independence settlement was short-lived and renegotiation soon began.

The Lancaster House Agreement

The 1979 Lancaster House Agreement—which codified the terms of the independence settlement—called for an immediate ceasefire, an election within three months, and a new constitution. The resulting 1980 Constitution sought to reconcile the fundamental interests of the protagonists. Africans would receive rights of universal suffrage—one person, one vote—leading to a government elected by majority rule. Whites would retain 20 seats in the 100-member lower house of a bicameral parliament for a transitional period of seven years along with restrictions on constitutional change. The armed forces of the liberation movements would be integrated into a national army, and civil servants from the old regime would remain fully vested in their government pensions, which would be payable by the Zimbabwe government even if retirees were abroad. In short, Africans achieved their main political goal—majority rule—in return for guaranteeing that white economic interests would not be fundamentally disturbed. On the critical issue of land, the Lancaster House Agreement favored the status quo. Contrary to ZANU-PF’s preference for confiscation of commercial farms without compensation, the constitution mandated that land could only change hands on market terms—that is, between willing sellers and willing buyers. Although a new African government would lack the financial resources to buy and develop much land, the British and US governments remained vague about promises of foreign assistance for land purchases. In this way, the independence settlement deferred a fundamental problem that would later come back to haunt the country.

The Independence Political Settlement

53

In a bold move before the final agreement was signed, Foreign Secretary Peter Carrington sent Lord Soames to Salisbury (later Harare) as the new British governor, accompanied only by a small Commonwealth Monitoring Force (CMF). As it happened, the restoration of British authority went largely unchallenged; instead, the governor’s arrival ended the settler rebellion, restored lawful administration, and triggered the lifting of economic sanctions. But remnants of the war lingered: some guerrillas entered official ceasefire assembly points supervised by the CMF, other insurgents refused to leave the bush, and still others buried their arms in secret caches and melted back into the civilian population. As election fever spread, intimidation and violence mounted on all sides, especially in areas controlled by ZANLA. Although the guerrillas enjoyed widespread popular support, Rhodesians convinced themselves that Bishop Abel Muzorewa would emerge victorious. In the event, the election of February 1980 provided Mugabe’s ZANU-PF with 57 seats, Nkomo’s PF-ZAPU with 20, and Muzorewa’s United African National Council (UANC) with just three. Reflecting historical cleavages reinforced by clashes between contending liberation movements during the war, ZANUPF swept the Mashonaland provinces, as did PF-ZAPU in Matabeleland.1 Dubious of the outcome, Nkomo and Muzorewa rejected the election results as a sham. True, ZANLA had instilled fear among voters by warning that the war would continue if ZANU-PF did not win; but there was little reason to question the election outcome, as ZANU-PF was the country’s most popular party at the time. Despite rumors of a possible right-wing coup, the Rhodesian military maintained its formal subservience to civilian authority and reluctantly acknowledged the election result. On April 18, 1980, Zimbabwe became independent, and Robert Mugabe became the country’s first majority-rule prime minister.

Consolidating State Power

In his first major speech to the nation, Mugabe surprised his critics by emulating Jomo Kenyatta in Kenya and announcing a policy of racial reconciliation. His sincerity in this regard has never been ascertained: the departing governor may have played a part in drafting the speech, and Mugabe may simply have been buying time without ever abandoning the goal of indigenizing the economy. But, at least provisionally, ZANU-PF acknowledged the need for a truce with white business and administrative elites. The new prime minister appointed two white cabinet ministers, including a leader of the commercial farmers to the agriculture portfolio, and temporarily retained the sitting heads of the national army, police, air force, and national intelligence agency. After all, members of the ZANU-PF elite had grown to

54

The Roots of Repression

political maturity in prison, combat, or exile and thus lacked the requisite technical and managerial experience to easily assume control of an extensive state apparatus. They knew firsthand from events in Mozambique that a rapid exodus of settlers along with their skills risked institutional and economic collapse. They also feared that Zimbabwe’s economic dependence on the apartheid regime in South Africa made the inherited state and economy vulnerable to external destabilization. Perhaps guided by Kwame Nkrumah’s injunction to “seek ye first the political kingdom,” the ZANU-PF government focused on the consolidation of political power, especially within the coercive organs of the state. Rex Nhongo (later known as Solomon Mujuru) and Emmerson Mnangagwa, loyalists from the liberation war, took up leadership posts in the army and state security ministry, respectively. Both men later emerged as key faction leaders within the ruling party. Because other senior leaders from the front lines of the nationalist struggle—like Leopold Takawira and Josiah Tongogara—had died of natural causes or been killed, the composition of the cabinet tilted toward younger intellectuals, such as Dzingai Mutumbuka, who drew the education portfolio, and Herbert Ushewokunze, initially appointed minister of health. Provincial barons like Eddison Zvogbo (from Masvingo) and Kumbirai Kangai (from Manicaland) were brought on board, not only for their professional expertise but also for regional balance. At the outset, Mugabe even sought to embrace nationalist rivals: he appointed four cabinet ministers from PF-ZAPU and offered the ceremonial presidency of Zimbabwe to Nkomo, who turned it down in favor of a ministerial post. Thus, the leadership of the new government was a coalition of oldguard nationalists, young radicals, battle-hardened guerrillas, and professionals returned from exile. Unlike Nkomo, who rose to the pinnacle of his party in the typical dominant style of an African “big man,” Mugabe at first occupied a less secure position. Historically, ZANU had suffered roiling internal divisions between generations of activists (some free and some in prison), among armed factions within the guerrilla armies, and between fighting forces and political negotiators. Mugabe had emerged in 1977 as a compromise leader who was minimally acceptable to all sides including, critically, the military commanders.2 Although he had secured the acting title of president of the party while still in exile in Maputo, there was a sense, even after independence, that Mugabe was merely first among equals and that the military retained influence over civilians. Acceding to the office of prime minister clearly enhanced his political status; but not until ZANU-PF’s second congress in 1984 was he confirmed substantively as president of the party. On the military front, a Joint High Command was set up to merge three rival armies—ZANLA, ZIPRA, and selected Rhodesian forces—into a unified Zimbabwe National Army. In practice, military integration favored

The Independence Political Settlement

55

ZANLA, whose members became commanders, and punished ZIPRA, many of whose cadres were summarily discharged (Kriger 2001; Todd 2007). From the outset, Mugabe drew defense affairs into the prime minister’s office, adding intelligence and provincial administration to his portfolio by 1985. He also created a Fifth Brigade of Shona-speaking ZANLA excombatants that bypassed the normal command structure and reported directly to the top leader. In the early 1980s, ZIPRA “dissidents” tested the coherence of the ruling coalition with sporadic insurgent activities, which ZANU-PF chose to see as part of a plot against the government, aided and abetted by South African interests. Using this excuse, Mugabe dismissed Nkomo and other PF-ZAPU ministers from the cabinet in 1982 and unleashed the Gukurahundi, a violent pogrom against the rural Ndebele population of Matabeleland and the Midlands. In addition, senior ZIPRA officers, including Lookout Masuku, then deputy commander of the army, and Dumiso Dabengwa, former head of ZIPRA intelligence, were arrested and charged with treason. After an attempt on his life, Nkomo fled the country and remained in exile from 1983 to 1985. Thus, the independence settlement between ZANU-PF and PF-ZAPU dissolved within a few years. ZANU-PF members regarded their former nationalist allies as “enemies of the state,” fit only for destruction.3 Moreover, atrocities committed by the Fifth Brigade in the Gukurahundi campaign led to the permanent alienation of most Ndebele-speakers. 4 The schism lasted until 1987, when an imposed Unity Accord revived the grand coalition established at independence, rehabilitated Nkomo and other senior PF-ZAPU leaders, and restored a semblance of peace in the countryside. In practice, however, the Unity Accord aimed to do what the Gukurahundi had failed to do—conquer the last frontier of resistance to ZANU-PF hegemony by subordinating the Sindebele-speaking region of Matabeleland and incorporating PF-ZAPU structures into the Shonadominated party. In the same year, Constitutional Amendment Act No. 7 created an executive presidency with Mugabe as president and Nkomo as one of two national vice presidents. Given its roots as a rural insurgency, ZANU-PF also moved quickly to penetrate the colonial-era state apparatus in the peasant farming areas, for example by appointing party loyalists as district administrators and replacing the old system of native administration with representative district councils. The party leaders’ hard work paid off, as their candidates were elected to these local government bodies and to a hierarchy of provincial and district planning boards and ward and village development committees in all regions except Matabeleland. Because district councils delivered an expanded range of social services after independence, ZANU-PF used performance in the locality to claim political legitimacy. Indeed, the party made the most of the patronage opportunities presented by a District Devel-

56

The Roots of Repression

opment Fund disbursed to councils by the powerful Ministry of Local Government, Urban, and Rural Development. To assert political control over the state at the center, the ZANU-PF elite moved quickly to Africanize the civil service. At first, the cabinet constituted a thin African veneer atop a largely untransformed state apparatus still manned by unsympathetic white personnel who could not be relied upon to implement reform policies. Over time, however, Africanization was facilitated by a doubling in size of the civil service, the promotion of blacks long held back from advancement, and an influx of skilled returnees from abroad. These institutional developments strengthened the state bureaucracy in relation to the ruling party and helped to maintain commitments to legal and technocratic standards in public management. As a signal of reassurance to the West, Mugabe appointed Bernard Chidzero, a former senior United Nations official noted for his orthodox approach to economic policy, as minister for economic planning and development. Over the next few years, Chidzero gradually expanded his influence by taking over the finance portfolio, moving economic decisions from party to cabinet, and emerging as the chief architect of the country’s economic strategy.

Development or Corruption?

At independence, Zimbabweans could boast that the country possessed a young and well-educated cohort of cabinet ministers. But an ideological gulf remained between the “comrades” in the party—whose organization was modeled on Leninist lines, complete with Presidium, Politburo, and Central Committee—and bureaucrats trained in colonial-era public administration. Chidzero’s “Growth with Equity” (1981) strategy was a compromise tailored to adapt to the structural inheritance of a state-capitalist economy. Despite rhetorical commitments to Marxism-Leninism, ZANU-PF did not attempt radical reforms that would have pushed the economy in an explicitly socialist direction, most obviously with regard to land redistribution. But the core party elite, schooled in Marxist economics and nationalist resentment, could not overlook easily the evidence that the economy remained mainly in foreign and settler hands. While not abandoning the rhetoric of socialist revolution, Zimbabwe’s new leaders adjusted to the economic situation in which they found themselves. For example, the new government depended on white farmers to feed the hundreds of thousands of rural dwellers who had been displaced from smallholdings during the independence war. Although inequality in land distribution had been a central motive of the armed struggle, the new leaders avoided radical land reform in the medium term. Instead of expropriating farms without compensation, the government opted

The Independence Political Settlement

57

to tax the export revenues of the commercial agricultural sector. This outcome was partly a product of legal constraints—the constitution mandated that land could only change hands willingly—and partly a consequence of resource shortages: aid donors were reluctant to foot the bill for land reform. Moreover, the new government’s fledgling Ministry of Lands lacked the technical and organizational expertise to operate an extensive resettlement program. Instead, government leaders chose to invest in improving access to agricultural and social services for the bulk of the population, most of whom were small-scale “communal area” farmers. Policies were revised to raise the prices paid for food crops, extend agricultural credit to smallholders, and build secondary schools and health clinics throughout the country. Although these policies fulfilled promises made by ZANU-PF, the government often adopted plans for service expansion directly from the shortlived Zimbabwe-Rhodesia government of 1979. But some ZANU-PF politicians pressed for land reform, at least rhetorically, as a campaign issue in the 1985 and 1990 elections. Regional party barons attempted to exert influence at the local level by mobilizing landless peasants and war veterans to occupy commercial farms. In the first independence decade, however, the government usually sided with the rule of law by removing squatters, sometimes by force. In attempting to accommodate the demands of organized interest groups, ZANU-PF had to scale down many of its stated developmental goals. Incomes policy is a good example. The party had come to power with the goal of establishing a national minimum wage that would lift wage earners above a set poverty line. A struggle ensued over where to draw that line, with the Employers’ Confederation of Zimbabwe (EMCOZ) representing employers against the Zimbabwe Congress of Trade Unions (ZCTU) representing labor. Ultimately, in 1984, the government enacted legislation that restricted the unions’ right to strike and, in 1986, established a review board that essentially accepted the employers’ argument that wage increases should be contained in favor of increased employment. This defeat for the labor movement not only revealed the unrealistic demands and weak organization of ZCTU, but it deepened a growing rift between industrial workers and the ZANU-PF, a predominantly rural party. Moreover, party leaders were not immune from predatory temptations. The notorious “Willowgate” affair burst into public view in 1989 and involved ministers who bought cars at subsidized prices from a state-run assembly plant and then illegally sold them at much higher prices. It revealed the vulnerability of state-owned enterprises to political interference, as well as a web of personal ties among politicians and businessmen who were profiteering outside of the law. When the media broke the story, Mugabe set

58

The Roots of Repression

up a commission of inquiry that led to the resignation of five cabinet ministers and one provincial governor, including senior officials like Maurice Nyagumbo, then senior minister of state for political affairs, who soon committed suicide. But the president ultimately granted pardons to all those implicated, subsequently returning ministers like Enos Nkala and Frederick Shava to top party positions. Following this high-profile scandal, ordinary Zimbabweans began to question the motivations of the ruling elite and to wonder if the country’s judicial and administrative institutions were up to the task of controlling corruption. Generally, however, the 1980s were a decade of moderate economic growth and rapid social development in Zimbabwe. The country’s gross domestic product expanded at an average 4.5 percent per year between 1980 and 1989, though droughts in 1982–1984 and 1987 caused fluctuations. Small-scale farmers proved an asset to the national economy, quadrupling the output of maize and doubling cotton production. Within five years of independence, nearly all children were in primary school and more than 80 percent of eligible students were moving on to secondary school. The minister of health was able to make the landmark announcement that the infant mortality rate had been cut in half. As for institutional outcomes, ZANU-PF met its main goal of becoming the dominant party in Zimbabwe. Having accumulated considerable political capital from its leading role in the national liberation struggle, it enjoyed leeway for policy missteps, at least in the early years. Outside of Matabeleland, where a culture of political fear was taking root, few people knew the full scale of the Gukurahundi atrocities, leaving the government’s legitimacy largely undamaged. As Figure 4.1 shows, the ruling party secured over 80 percent of the votes cast in the 1990 elections and won 117 out of 120 elected National Assembly seats. But these figures obscure the fact that the practice of power politics had led to a narrower elite coalition than policies of racial reconciliation or economic redistribution might predict. Mugabe began to withdraw the hand of friendship from the white community after Ian Smith’s party (renamed the Conservative Alliance of Zimbabwe) won 15 of the 20 white seats in the 1985 election. Minister of State for National Security Emmerson Mnangagwa denounced the white vote as a deceptive betrayal of the reconciliation policy. In response, ZANU-PF marshaled a cross-party parliamentary coalition to promulgate Constitutional Amendment No. 6 in 1986, which scrapped the white voter roll, thus ending the practice of reserving some seats for whites only. And because of persistent mutual distrust, especially over the government-controlled system for allocating foreign exchange, the black ruling party and white business and farming communities warily held each other at arm’s length.

The Independence Political Settlement

59

Figure 4.1 Elections in Zimbabwe: Performance of ZANU-PF, 1980–2008

Sources: African Elections Database, 1980–2005; EISA Observer Mission Report, 2008. Notes: Data are percentage shares of total votes cast in elections or of total elected seats in the National Assembly, which had 80 elective seats in the 1980s, 120 seats in the 1990s, and 210 seats in 2008. a. 2008 results are for the first round only.

Importantly, the 1987 Unity Accord failed to cement the ruling coalition. Although the top layer of PF-ZAPU leaders was enticed into the ruling group, the residents of Matabeleland and parts of the Midlands did not follow. With residents who remembered the 20,000 victims of the Gukurahundi campaign, the region remained a reservoir of discontent. In the 1985 elections residents voted unanimously for PF-ZAPU, provoking in some places violent retaliation from mobs led by the ZANU-PF women’s and youth leagues. Instead, ZANU-PF’s major achievement during this decade was to consolidate its own political position and to subordinate its main nationalist rival. By 1990 it controlled almost all the seats in the National Assembly. A government that began with a nod toward power sharing had, within seven short years, accomplished a decisive feat of power capture. By establishing party hegemony under a banner of “national unity,” ZANU-PF ensured that political competition would center increasingly on factional rivalries within the ruling party itself. At the same time, the governing coalition created a protected political space in which leaders could pursue their own personal and business interests, even as objective conditions began to turn against the economy.

60

The Roots of Repression

Politicizing the State

With the expiry in April 1990 of the Lancaster House Agreement, the rulers of Zimbabwe were freed from the last legal obstacle to the creation of a de jure one-party state. ZANU-PF leaders made well-worn arguments for abandoning the terms of the independence political settlement by contending that single-party rule was consistent with African traditions, suitable for a “unified” and “classless” African society, a necessary alternative to imported multiparty models, internally democratic, and a prerequisite for coherent development planning. Mugabe was especially effusive on the theme that Britain had no right to teach democracy to Zimbabweans. But party leaders could not prevent debates from breaking out in ZANU-PF’s Politburo and Central Committee, as well as in the independent press and civil society, about the cost to democratic accountability. The single-party idea always enjoyed less elite and popular support than imagined by the inner circle of the ruling party. A 1985 public opinion survey conducted by the University of Zimbabwe showed that some 60 percent of respondents favored one-party rule, whereas the remainder opposed it. By 1990, however, the proportions had reversed, with 60 percent opposed (Moyo 1992). In the end, mounting internal and societal opposition ensured that the position articulated by Canaan Banana, who had served as ceremonial president between 1980 and 1987, carried the day. Specifically, since ZANU-PF had already attained a de facto one-party state via the ballot box, there was no need to legally entrench the arrangement in the constitution. In any event, such a move would have been anachronistic since, at the time, other nearby countries—notably Zambia, Mozambique, and Kenya—were in the process of opening up to multiparty competition. Regardless of the formal rules of the political game, the ruling elite in Zimbabwe had long demonstrated an inability to tolerate the expression of political dissent. In 1990, ZANU-PF marshaled all its efforts, legal and otherwise, to frustrate an emerging opposition movement. Edgar Tekere, a firebrand populist who was once ZANU-PF’s secretary-general and a cabinet minister, had been dismissed from the party in 1988 for blowing the whistle on what he called a “vampire class” of corrupt leaders (Tekere 2007). In 1989, he formed the Zimbabwe Unity Movement (ZUM) on a platform that opposed a one-party state and promised employment, housing, and market reforms. In a move guaranteed to infuriate the national leadership, ZUM formed an electoral coalition with the white-led Conservative Alliance of Zimbabwe. The incumbents resorted to a standard bag of dirty tricks in the 1990 elections—monopolizing the public airwaves, disrupting opposition rallies, and directing official resources into the ruling party campaign. Even though

The Independence Political Settlement

61

ZUM ran a disorganized crusade, ZANU-PF overreacted, falling back instinctively on violence. Just days before the poll, state security agents attempted to assassinate Patrick Kombayi, national organizing secretary of ZUM. Following the vote, some ZUM candidates and supporters went into hiding, fearing retribution by revengeful ZANU-PF mobs. To his great discredit, Mugabe used the power of the presidency to pardon security personnel and others convicted of electoral violence, thus adding to a growing culture of impunity. Once reelected, the ruling elite developed a systematic set of controls to organize its own supporters and weaken would-be opponents. Although a long-standing state of emergency was allowed to lapse in 1990, ZANU-PF retained the colonial Law and Order Maintenance Act, which enabled bans on political meetings and movements. In order to penetrate and politicize society, the president’s office set up its own Ministry of Political Affairs in 1988, housed at party headquarters in Harare and in 1992 expanded the cabinet to add residential ministers in every province. All such positions went to the most reliable party loyalists. When backbench members of parliament tried to exercise their legal authority to question the government budget, the ZANU-PF speaker of the National Assembly condemned their efforts as treasonous. When independent-minded individuals won party primary elections, the central organs of ZANU-PF intervened to overturn the election results. The election to parliament of Margaret Dongo, a former ZANLA combatant who broke with the party in 1995, was an exception that only drew attention to the fact that independent candidacies were usually futile (Makumbe and Compagnon 2000). With regard to party-state relations, the ZANU-PF Politburo and Central Committee gradually usurped the policymaking roles of the cabinet and parliament, thereafter “ensuring that [the party] dominates virtually every political institution in Zimbabwe” (Makumbe 2003). For example, the Politburo established watchdog committees to enforce the supremacy of the party, prompting a public clash between Minister of Home Affairs Moven Mahachi and Minister of the Public Service Eddison Zvogbo over whether the party had the right to supervise the work of civil servants. The slow pace of policy reform and delays in routine administration in Zimbabwe during this period were partly due to such political struggles, leading many competent senior officials, including experienced permanent secretaries, to opt for early retirement. ZANU-PF chose to fill these posts with individuals whose qualifications leaned more toward partisan fealty than managerial skills. These personnel changes led to the politicization of the bureaucracy; henceforth, cabinet ministers would rely on permanent secretaries to dependably execute the party’s bidding; and in the run-up to the 1995 elections, ministers threatened to dismiss civil servants who appeared to support an opposition party.

62

The Roots of Repression

The state is the most valuable prize in African politics because state power can be used to create opportunities for private gain. Mandaza lays out this logic in the Zimbabwean context: The quest for power and wealth expressed itself sometimes in open corruption and nepotism. The long years of colonial domination and deprivation, not to mention imprisonment and the hard days of the struggle, became almost the licence . . . to accumulate quickly; and the state . . . appeared the most viable agency for such accumulation. (Mandaza 1986, 56–57; also see Taylor 2007)

Thus top politicians succumbed to predation, sometimes transforming themselves into wealthy patrons. By 1990, senior party leaders—including the two national vice presidents and the president’s sister—owned commercial farms, thanks to the land reform program. At the local level, ZANU-PF town councilors allocated housing stands to themselves, their families, and their followers. Fraud in Zimbabwe’s parastatal enterprises and corruption in the public service sector reportedly reached “epidemic” proportions (Economist Intelligence Unit 1993). Most brazenly, the corrupt allocation of public resources was institutionalized in Zimbabwe in the Political Parties (Finance) Act of 1992. It provided that any party with at least 15 representatives in parliament—which, in practice, meant only ZANU-PF— would receive government financial support. In this way, the ruling party openly appropriated taxpayer resources to finance the construction of its own formal organization, party headquarters, and informal patron-client networks.

Liberalizing the Economy

The positive performance of the Zimbabwe economy in the short term masked fundamental, long-term problems. The government’s fiscal deficit—over 10 percent of GDP in 1990—was high and rising, and an excess of imports over exports led to ballooning trade deficits. Prospects for economic growth were hampered by a lack of foreign investment and the obsolescence of the country’s productive infrastructure. The job market could not keep up with the more than 100,000 graduates leaving secondary school each year. And, with the emergence of a free South Africa in 1994, Zimbabwe began to face a formidable regional competitor whose productive capacity was the largest and most sophisticated on the continent. The ruling elite thus faced a basic policy choice. Would leaders stick with professed commitments to economic self-reliance and socialist transformation? Or would they reverse direction by choosing to follow other African countries that were embracing globalization and market reforms?

The Independence Political Settlement

63

As it happened, Zimbabwe adopted an Economic Structural Adjustment Program (ESAP) in 1991 with support from the World Bank. The politics underpinning this policy shift involved a partial realignment of the governing coalition: the ruling party moved decisively against urban workers and consumers, for example, by cutting jobs, maize-meal subsidies, and social services. The apparent ease of the policy turnabout suggests that there were few authentic Marxist-Leninists in the party hierarchy. The ZANU-PF leadership was instead most strongly attracted to pragmatic policies that promised to expand the economic pie. In Jonathan Moyo’s words: It was not surprising that the pre-independence socialist rhetoric of ZANUPF dwindled with the tick of time. . . . In 1980, the party was subjected to lofty talk about egalitarianism. Now the same public is being subjected to talk, under the veil of socialism, about long overdue economic liberalization and structural adjustment. (1991, 89)

In opting for ESAP, state leaders forged new alliances with both established and newly formed business associations. The most influential lobby was the Confederation of Zimbabwe Industries (CZI), which cultivated allies with bilateral donors, international financial institutions, and the dwindling technocratic corps of the government. The CZI explicitly recognized the advantage of avoiding direct confrontation with the political authorities and instead engaged in a campaign of persuasion that resulted in a set of policies—convertible currency, open imports, export incentives, foreign exchange retention, remittance of profits by foreign investors—that closely corresponded to its members’ wish list. That is not to imply that the government’s partnership with manufacturing interests led to completely free trade. Rather, economic policy in Zimbabwe in the 1990s exemplified a partial reform syndrome that selectively retained administrative controls (Hellman 1998; van de Walle 2001). Public marketing boards continued to set agricultural producer prices, tariffs replaced quota restrictions on imports, and parastatals were never completely privatized. As such, economic liberalization did not eliminate opportunities for strategically placed politicians to continue to extract economic rents. Nor did adjustment policies necessarily lead to economic success. A vicious drought in 1992 led to economy-wide recession. Inflation in bread prices provoked food riots in Harare in 1993. And the plan to finance ESAP through growth in exports fell flat, forcing the government to turn to bilateral donors, the World Bank, and the International Monetary Fund to finance huge fiscal and current account deficits. Thus, one of the major outcomes of the economic reform program was mounting national debt.

64

The Roots of Repression

Moreover, few economic interest groups were as successful as the CZI at aligning with policymakers. The white-led Zimbabwe National Chamber of Commerce (ZNCC) lost influence to newly established associations of black entrepreneurs such as the Indigenous Business Development Centre (IBDC) and the Affirmative Action Group (AAG). “Indigenization” of productive enterprise based on racial criteria became a guide not only to policy, but also to patronage. At the same time, the Commercial Farmers Union (CFU) found themselves at loggerheads with government, especially on matters of rural security, the merger of farmers’ unions, and land reform. Workers’ unions, led by the ZCTU, were dead set against ESAP, especially the laying off of over 10,000 low-level public service workers.

A Nascent Opposition

Because the labor movement emerged in the 1990s as a building block for an opposition coalition, its rocky relationship with the incumbent party is worth examining. Before independence, African workers were organized in four rival federations—the weakest affiliated to ZANU-PF—that were prohibited from strikes and collective bargaining. After independence, the authorities forced unions to amalgamate under ruling party control beneath the ZCTU umbrella. Party cadres (often former ZANLA combatants) were active in workplaces, seizing leadership of workers’ committees and pressuring employers to hire veterans from the liberation war. The new government passed a Labor Relations Act in 1985 that reproduced existing colonial restrictions on workers’ rights and continued to marginalize labor representatives from wage-setting negotiations between government and the business community. Moreover, the government did not hesitate to deploy the police and army to put down wildcat strikes for wages and benefits in the 1980s or against ESAP austerity measures in the 1990s (Nordlund 1996; LeBas 2011, chap. 3). As a consequence, the ZCTU gradually distanced itself from ZANUPF, especially after the election of Morgan Tsvangirai as the federation’s secretary-general in 1989. Once a foreman in the mining industry and a disaffected former political commissar in ZANU-PF, Tsvangirai emerged as the essential censorious voice of the labor movement. In the 1980s, union leaders had called for worker representation in ZANU-PF structures; by the 1990s, they came out openly in support of an alternate, multiparty political system. In 1989, Tsvangirai condemned the government’s closure of the University of Zimbabwe and asserted solidarity with students protesting corruption; for his pains the government detained him for a month. Then in 1992 when he led union opposition to government legislation that undercut the ZCTU’s financial base (by eliminating automatic union dues deduc-

The Independence Political Settlement

65

tions), the minister of labor warned ZCTU not to behave like an opposition political party or it would risk harsh treatment. Tsvangirai later survived at least three assassination attempts, including in 1997 when assailants burst into his tenth-floor office and tried to throw him out of the window. Under Tsvangirai’s leadership, the ZCTU began to correct some of its organizational weaknesses—including nepotism in job assignments and misappropriation of funds—that had roiled the union federation in earlier years. Moreover, rising inflation and lagging wages offered opportunities to revitalize the labor movement. By 2000, average real incomes were onefifth lower than in 1980, and three-quarters of the population was estimated to be living in poverty. Yet government food subsidies had been eliminated. As the government turned to policies of cost recovery for social services, people found that access to education and health care, after rising dramatically in the 1980s, reversed direction in the 1990s. Symbolizing these setbacks, schoolteachers refused to mark exam papers for want of adequate salaries paid on time. In protest, ZCTU officials organized a national public sector strike in 1996, a general strike in 1997, and mass stay-aways in 1998 (Saunders 2001). Importantly, the unions expanded their critique of the government beyond bread-and-butter issues to broader demands for political accountability. Labor leaders nurtured alliances with other social groups, notably university students and human rights activists, demonstrating a newfound capacity to coordinate confrontation with the government. According to Brian Raftopoulos: The disparity between the de jure rights and freedoms enshrined in the Zimbabwean constitution and the de facto political rules developed by the state . . . provided . . . important openings for contesting ZANU-PF domination. The battles in the courts over the abuse of executive powers and the uneven playing field provided by electoral laws became the focus for action by civil society groups, especially in the 1990s. Such issues were to feature as a central part of the campaign for constitutional reform after 1997. (2004, 7)

In 1997 a voluntary National Constitutional Assembly (NCA) raised popular awareness about the need to replace the 1980 Constitution (now amended multiple times to cement the president’s dominance) with a new national charter (McCandless 2011). The NCA—a broad grassroots alliance of professional, labor, women’s, and religious organizations, including prominently the ZCTU—created an educational campaign on constitutional issues disseminated through a network of provincial meetings. Through the courts, the NCA forced the government to remove a ban on its advertisements in the electronic mass media, enabling the campaign to reach rural dwellers via radio. The government reacted to this perceived setback by threatening to regulate media outlets. At the same time, the NCA initiative

66

The Roots of Repression

prompted the government to form its own Presidential Commission of Inquiry into constitutional issues, a step leaders would rather have avoided. By fostering a national public debate on the way Zimbabwe should be governed, the NCA “used the language and politics of constitutionalism to expand the meaning of development in Zimbabwean politics” (Moyo et al. 2000, 39). By the end of the 1990s, therefore, scattered social interests had begun to crystallize into a nascent opposition coalition (Muzondidya 2009). All but two opposition parties5 boycotted the 1995 elections because of the absence of electoral and constitutional reforms. For its part, the NCA advocated that citizens shun the government’s official constitutional commission. While some people heeded this call, others spoke up against the government’s handpicked body. As an independent civic organization, the NCA claimed (somewhat disingenuously since its members favored a change of government) to have no partisan agenda. Instead, ZCTU took the lead, announcing in 1999 the formation of a political party known as the Movement for Democratic Change (MDC) whose principal objective was “a struggle for jobs, decency, and democracy.” Tsvangirai—by then NCA chair—was elected president of MDC at the inaugural party congress in January 2000. Thus an emergent democratic movement arose to challenge an entrenched ruling party whose mismanagement and corruption had called into question its right to rule. The ZANU-PF leadership had become narrower and uncohesive: few former PF-ZAPU members remained in the cabinet, rifts had begun to emerge among rivals to succeed Mugabe, and parliamentary backbenchers were restive. The party’s loss of political legitimacy was starkly illustrated in the 1996 presidential elections: although Mugabe won over 90 percent of the vote, barely one-third of the registered electorate bothered to turn out on polling day (see Figure 4.1). To offset its loss of political support, ZANU-PF tried to shore up its heartland among the Shona-speaking peasantry, for example, by providing rural voters with drought relief and free seed and fertilizers. In addition, rural and district councils were legally merged, thus transferring tax revenues from commercial to communal farming areas. At the same time, ZANU-PF reversed its relations with traditional chiefs and headmen by restoring some of their lost powers and including them in the party’s patronage network. Formerly the leading source of progressive ideas in Zimbabwe, the party elite thereby transformed ZANU-PF into a force for social and political conservatism. In a significant break with the independence political settlement, the government also repudiated the property rights of white farmers. Previously, the transfer of land from the commercial sector to peasant producers had proceeded legally and deliberately: just 52,000 farm families had been resettled by 1990 against a waiting list of up to half a million (Bratton

The Independence Political Settlement

67

1990; von Blankenberg 1994; Moyo 2000). In 1992, however, a Land Acquisition Act allowed compulsory (not willing) acquisition at “fair” (not market) prices. In 2000, the constitution was amended to permit the state to confiscate designated land and only pay compensation for buildings and other improvements. Land reform, once mainly an intermittent election issue, now became the central pillar in Mugabe’s ideological assault on the MDC. He accused the opposition party of being a neocolonial stalking horse for the economic interests of imperialists and settlers. In an increasingly racist discourse that found appeal within the country and beyond, Mugabe proclaimed himself a pan-African liberator in a life-and-death struggle for the restoration of black Africans’ lost land rights (Raftopoulos 2001; Alexander 2006). By embarking on fast-track land seizures, the ruling party drove the last nail into the coffin of racial reconciliation, a key plank in the independence political settlement. This radical departure required that Mugabe cultivate new allies; he turned to the veterans of the liberation war. As with the Unity Accord of 1987, which bent PF-ZAPU to the will of ZANU-PF, the agreement with the “war vets” can be regarded as an informal, intra-party “settlement within a settlement.” It realigned the composition of the ruling party by incorporating elements that had hitherto been largely excluded. Almost overnight, the war vets became ZANU-PF’s most-valued political resource, wielding what amounted to veto power on strategic policy matters. In so doing, Mugabe allowed ZANU-PF policy to be captured not only by armed militants, but also by the security chiefs of the armed forces, with whom the war veterans shared a political affinity forged by the liberation war. In a serious internal convulsion, the Zimbabwe National Liberation War Veterans’ Association (ZNLWVA) led by Chenjerai “Hitler” Hunzvi challenged the authority of the party in 1997. The veterans bitterly protested their exclusion from state patronage, especially after it came to light that senior members of the ZANU-PF inner circle had received massive handouts from a War Victims’ Compensation Fund. The rank-and-file war vets noisily interrupted the president’s annual speech at Heroes’ Acre and invaded party headquarters, where they held members of the Politburo hostage. Unable to withstand the potential defection of this key constituency, the president promptly conceded to demands for generous gratuities and pensions for 50,000 veterans. These arbitrary and unbudgeted expenditures precipitated a crash in the value of the Zimbabwe dollar, necessitated tax increases on working people, and initiated a downward spiral in the economy as a whole. Wilbert Sadomba, a war veteran turned academic, variously labels Mugabe’s concessions as a “pact,” a “truce,” and an “agreement” and concludes that it signaled “a new political era of war veteran dominance in politics” (Sadomba 2011). According to Eldred Masunungure, the deal

68

The Roots of Repression

between Mugabe and the ZLWNA is of “cardinal significance in the political development of Zimbabwe post-independence” because it “almost completely reconfigured [the country’s] political economy.”6 By caving to the ex-combatants’ demands and elevating war veterans to a strategic role in the ruling coalition, the ZANU-PF elite put a definitive end, not only to the era of ESAP and the influence of technocrats on policymaking in Zimbabwe, but also to the basic provisions of the independence political settlement. Instead, leaders opted to abandon economic rationality in a violent and reckless search for political survival.

An Unraveling Settlement

The political settlement struck at the time of Zimbabwe’s independence began to come undone soon after it was installed. The agreement among black and white elites to trade political power for economic moderation lasted only seven years; that is, until the expiry of reserved clauses in the constitution. In this regard, the independence settlement was less permanent than the preceding colonial settlement, which endured from 1923 to 1965. Why did the postcolonial bargain, which initially won considerable popular and international legitimacy, ultimately prove short-lived? The answer suggested by this chapter is that this particular elite pact placed inconvenient hurdles in the path of ZANU-PF leaders’ quest to consolidate state power and use it for personal gain. To be sure, the independence settlement looked promising at the time. It featured an inclusive group of participants, a convergent set of incentives, and an agreement of comprehensive scope. First, the independence settlement was politically inclusive. The Lancaster House Agreement was the product of an all-parties conference that created conditions for the country’s first universal and credible election. No major political party was prevented from contesting the April 1980 poll, and voter turnout was quite high. Moreover, ZANU-PF at first offered to share a token measure of power with rivals by granting them cabinet seats. Because the party enjoyed a large enough electoral majority to govern alone, this gesture could also be withdrawn at any time—and it soon was. What looked like an inclusive regime on the surface soon turned exclusive: promises of racial reconciliation and the inter-party Unity Accord notwithstanding, ZANU-PF accumulated the lion’s share of political power. But, at least at the outset, the independence agreement seemed to allow an unprecedented degree of engagement in national political affairs to all principal stakeholders. Second, the Lancaster House Agreement provided strong incentives for accommodation. Just as ordinary citizens longed for peace, white Rhode-

The Independence Political Settlement

69

sians and African nationalists both stood to benefit from the end of a debilitating civil war. Since the new constitution delivered independence under majority rule, it removed the guerrillas’ rationale for fighting. At the same time, international recognition of a new government paved the way for the suspension of international sanctions, resumption of trade, and renewed economic growth. All parties could agree, for example, that to meet popular aspirations, it was essential to repair and expand the country’s wardamaged networks of schools, health clinics, and agricultural services. Third, the independence settlement was relatively comprehensive. The scope of the elite pact addressed not only the core political concern, namely, who would lead a majority-rule government. It also encompassed economic and military issues, such as the disposition of property and the immediate status of the security forces. By guaranteeing public sector pensions, the agreement forestalled an exodus of civil servants and helped to maintain effective governance. And, by initially keeping white commercial farmers on the land, the government was able to meet both popular demands for food security and its own tax base in export revenues. By negotiating a ceasefire and gathering guerrillas into assembly points, the independence agreement also paved the way for the eventual integration of rival forces into a national army. And by subsequently providing training programs, British military instructors helped to moderate ZANU-PF’s instinctive proclivity to transform the national security forces into an armed wing of the ruling party. As such, the transition to independence was initially accompanied by the maintenance of professional capabilities in both public and private sectors. In retrospect, however, several unfavorable factors were also at play. The independence political agreement was externally driven, weakened by less than credible leadership commitments, and, most importantly, susceptible to power capture by a dominant party. In this last respect, the independence settlement allowed the ruling elite to develop a repertoire of political controls that drew on the colonial legacy. First, the Lancaster House Agreement lacked an adequate degree of local ownership. The 1980 Constitution was an external initiative led by the British government and backed by South Africa and the frontline states of Zambia and Mozambique. The Zimbabwe-Rhodesia delegation and the two liberation movements would likely never have accepted the terms of the settlement had they not been threatened with the withdrawal of funding, training, and logistical support by their patrons in the region. Only after agreement was reached on the principles of independence and majority rule did the nationalist leaders reluctantly swallow the unwelcome racial provisions retained in the constitution. In the end, the Lancaster House conferees arrived at a least-worst compromise, beloved and fully embraced by no leader within the country. Rather, agreement was reached because nobody

70

The Roots of Repression

wanted to shoulder blame for a breakdown of the conference and because all parties trusted in their own ability to win an election. Because ZANUPF did not feel a strong sense of ownership in the independence settlement, its leaders later had few qualms about amending, abandoning, or otherwise violating its provisions. Second, there was a good deal of uncertainty at the time about the credibility of leaders’ commitments. Whites, primed by their own anticommunist propaganda, panicked at the thought of a self-professed Marxist assuming the post of prime minister. Blacks wondered about the loyalty of the Rhodesian defense forces, including whether the white security chiefs would take orders from a new black premier. Mugabe’s reconciliation speech helped to allay some fears. But the various factions also hedged their bets, whether the ZANLA forces that refused to enter ceasefire assembly points, dissident elements from ZIPRA who hid their weapons in the bush, or disgruntled settlers who continued to speculate about a reactionary military coup. In the end, these threats proved to be overblown; but the new nation nevertheless embarked on independence with deep reservoirs of distrust among former adversaries. And distrust deepened when the government moved to politicize the armed forces and, later, to confiscate commercial farmland. Finally, the political balance of power within the independence government clearly rested with the ZANU-PF majority. At best, Joshua Nkomo— arguably the father of African nationalism in Zimbabwe—never seemed destined for more than a secondary role after independence; at worst, he and his PF-ZAPU followers were marginalized. Unaddressed in the arrangements for a transfer of political power was the festering division between the two liberation movements, which could not be papered over with the fiction that they together constituted a cohesive “patriotic front.” Within two years of independence, a renewed conflict between ZANU-PF and PFZAPU broke out in Matabeleland, with deadly and lasting consequences for the supporters of the minority party. The price for the re-inclusion of sections of the PF-ZAPU leadership in the governing coalition, as stipulated in the Unity Accord of 1987, was nothing less than their utter subordination to ZANU-PF. During the first two decades of independence, ZANU-PF embarked on a systematic campaign to capture the Zimbabwean state. The result was an imbalance of power symbolized by the reemergence of a de facto one-party regime. Unlike the single-party system of the settler colonial era, in which the cabinet and bureaucracy were responsible for policymaking, decisions in the postindependence one-party system became concentrated in the party Politburo. Coupled with ill-advised policy choices and creeping corruption, the heavy-handed dominance of ZANU-PF rule ultimately called into question the party’s mandate. Over time, as the

The Independence Political Settlement

71

economy faltered and living standards dropped, ZANU-PF lost support among economic elites, urban workers, and ethnic minorities in the rural margins. In the final analysis, the ruling party’s quest for supremacy— which catalyzed a counterhegemonic opposition movement—sounded the death knell of the independence settlement.

Notes

1. PF-ZAPU eked out one seat in Mashonaland West province. There were no delimited electoral constituencies in the 1980 elections; a province-based proportional party-list system was used. Regarding power distribution, this electoral system produced a more inclusive result than the constituency-based first-past-the-post system adopted after 1985, which reinforced an exclusive pattern of ZANU-PF dominance. 2. During the liberation war, Mugabe had favored combining military and political roles with political commissars attached to each ZANLA combat unit. Moreover, civilian politicians and military commanders cohabited on the ZANU-PF Politburo (Ranger 1980, 83). 3. At this time, Mugabe likened PF-ZAPU and its leader to “a cobra in the house. . . .  The only way to deal effectively with a snake is to strike and destroy its head” (Nkomo 2001, 2). 4. Translated from the Shona as “the spring rain that washes away the chaff from the last harvest,” Gukurahundi suggests clearing out debris. It involved two overlapping conflicts: a low-intensity counterinsurgency campaign to root out “dissidents” by the regular defense forces; and an ethnic pogrom by the Fifth Brigade against unarmed Ndebele civilians (CCJPZ 1997/2007). 5. The Forum Party and ZANU (Ndonga). 6. Personal communication, August 1, 2013.

5 A Period of Crisis, 2000–2008

The unraveling of Zimbabwe’s independence settlement gave way to a period of open conflict. As ZANU-PF’s dominance came under challenge, its leaders resorted to the more extreme tactics of power politics, including violence. By ignoring the rule of law and the principles of sound economic management, they led the country into a devastating economic and political decline. In the electoral arena, the incumbent rulers lost ground to the opposition, first in a constitutional referendum and later in a series of increasingly ferocious elections. To offset these losses, the ruling elite dismantled the last remnants of property rights and judicial independence, while, at the same time, encouraging militia and security force elements to play enhanced roles in governance. Faced with unemployment, displacement, and hyperinflation, Zimbabwean citizens retreated to the informal economy in search of safety and security. When confronting existential threats, leaders tend to govern by the methods they first used to ascend to power. Because ZANU-PF entered politics via a war of liberation, its leaders intuitively revived the party’s civilian-military coalition from the war years and reverted to familiar techniques of strong-arm rule. Indeed, ZANU-PF never fully transformed itself from a guerrilla movement to a democratic political party. Instead, a selfinflicted national crisis brought back to the fore the ideological and coercive elements in ZANU-PF’s authoritarian temperament. Only this time around, the party had at its disposal the full array of state institutions inherited from colonial rule and adapted to its own defensive and extractive purposes. In clinging to power, the ZANU-PF regime—characterized in this book as a militarized form of electoral authoritarianism—intensified its reliance on ideology, patronage, and violence. Reminding his followers of resistance to settler conquest in the 1890s and the struggle for independence in the 1970s, Mugabe rhetorically urged a third phase of chimurenga (revolution-

73

74

The Roots of Repression

ary war) to complete the emancipation of the country. In this phase, land rights would be restored to the “sons of the soil,” who were defined narrowly in the party’s version of “patriotic history” as those who had actively supported ZANU’s side in the struggle (Ranger 2004; Kriger 2006). Land was not only the ultimate revolutionary symbol, but also a valuable political resource for cementing political loyalty. Resorting to forcible land seizures, the party redistributed commercial farms, not only to land-hungry peasants but also to key members of the ruling elite, including cabinet ministers, senior civil servants, security force commanders, judges, and parliamentarians. Because land and other resources were treated as prizes of patronage, economic production and related social services suffered. Central to the “lost decade” of 2000–2008 was an escalation of political violence; by the peak of the crisis in mid-2008, the party was on a war footing. Abandoning any pretense of political tolerance, Mugabe and his military allies unleashed a no-holds-barred campaign to dispossess white farmers and crush any opposition movement that threatened their permanent hold on power. At the core, the regime came to rest on two interpenetrated organs of authority: the ruling party and the security forces. Political elites intentionally blurred the boundaries between party and state—both at the top, between ZANU-PF structures and the apparatus of national security, and at the bottom, between regular military forces and informal party militias. As a result, the military became politicized and the polity became militarized. For its part, the MDC tried to confront the ZANU-PF’s clenched fist with a more inclusive and peaceful symbol: a raised open palm. The opposition party promoted a liberal discourse about democracy, economic opportunity, and human rights that was more in tune with a global, post–Cold War world. Over time, a majority of Zimbabweans recognized the declining relevance of the politics of national liberation; for example, they rejected propaganda about ZANU-PF war heroes as being a shabby cover for the personal interests of a narrow clique. Importantly, citizens showed a growing willingness to vote against the ruling party in elections, despite intimidation and fear. But, even though the opposition eventually won a working parliamentary majority in 2008, a weak and divided MDC was still unable to take control of the levers of state power.

Descent into Violence

The turning point in the emerging conflict was the 2000 referendum on a new constitution. Prompted by demands from civil society, the government went through the motions of consulting the electorate on constitutional reform, but, in reality, its constitutional commission firmly con-

A Period of Crisis, 2000–2008

75

trolled the process (Dorman 2003; Ndulo 2010). The commission chair forced through a draft that lacked broad support, and the government added substantial “corrections and clarifications,” notably to allow compulsory land acquisition. The NCA, led by nongovernmental organizations (NGOs), charged that the procedure was opaque and insufficiently participatory; moreover, the constitution did not go far enough in reducing presidential authority. In this regard, the referendum represented not only “a plebiscite on ZANU-PF’s rule since 1980” (Raftopoulos 2009, 211), but also the first concerted effort by organized forces in civil society to define a more inclusive political settlement. In the February 12, 2000, referendum, a 54 percent majority voted “No” on the government’s constitutional proposals, thus handing the ruling party its first-ever defeat at the polls. The shocked official response was swift and brutal. Mugabe blamed the loss on the emergent opposition movement, which he portrayed—in increasingly exclusive and racist terms—as a front for Western imperialism (Muzindidya 2007). Henceforth, these “unpatriotic” forces—no longer limited to PF-ZAPU “dissidents” but now including urban professionals, rural teachers, religious leaders, commercial farmers, and farm workers, especially those from neighboring countries—would be treated as “enemies of the state” who could expect no protection under the law. In girding for a crackdown, party leaders gave tacit credence to the war veterans’ critique that corrupt party leaders had allowed the provincial structures of ZANUPF to wither away. Unless remobilized, a weak party organization at the grassroots seemed to portend another defeat for the ruling group in the upcoming June 2000 parliamentary elections. The party, therefore, turned to tried-and-true tactics: it revived an alliance from the liberation war and whipped up grievances over land. War veterans—who had long complained about the slow pace of land reform— and unemployed urban youth were encouraged to seize large farms and estates (Moyo 2001). The targets were white landholders and farm workers, notably those suspected of supporting MDC. In the process, unruly gangs engaged in jambanja (chaos, violence, lawlessness) destroyed crops, livestock, and equipment and forced the farm occupants to flee.1 The post-2000 invasions were more centrally orchestrated than the scattered and spontaneous takeovers of earlier decades. The party ordered government agencies, including the army and intelligence service, to provide finances, transport, and other logistical backup (ZHRNGO Forum 2000). The police were instructed not to intervene except to protect invaders from prosecution for assaults or property crimes. Then, in July 2000, the government announced a Fast Track Land Reform and Resettlement Program in which many landless individuals received plots of land: between 2000 and 2002, some 6 million hectares were confiscated from 4,000

76

The Roots of Repression

white farmers and redistributed to an estimated 127,000 small-farm families and 7,200 black commercial farmers. But the prime properties were allocated to ZANU-PF chefs (privileged party leaders) and allied elites, some of whom accumulated several farms (Moyo 2011; Southall 2013, 256–257). The ruling party strategy for subsequent elections—in June 2000 (parliamentary), March 2002 (presidential), and March 2005 (parliamentary)— built upon previous experience with political violence (Kriger 2005; ICG 2005). The goal was to create “no-go” zones in the countryside that were closed to opposition campaigns. Under the direction of the party hierarchy, local ZANU-PF officials ignored constitutional guarantees of free association and assembly by effectively banning MDC from operating. In a practice derived from the liberation war, people were compelled to attend pungwes (all-night rallies), where they were subjected to political indoctrination and where suspected “sellouts” were publicly humiliated. In a mounting campaign of aggression against MDC candidates and supporters, party militias administered beatings, destroyed property, and even carried out rapes and murders. Table 5.1 shows how acts of state-sponsored violence peaked during the long election season between 2000 and 2002. When drought threatened to kill thousands of rural Zimbabweans in 2002, Didymus Mutasa, a key ZANU-PF insider, seemed unconcerned, declaring, “We don’t want all these extra people . . . we would be better off with only six million [of] our own people who supported the liberation struggle.”2 For its part, the opposition used ZCTU and NCA structures to build a rival network of activists among public service workers—teachers, nurses, and agricultural extension workers—in the ruling party’s rural strongholds and elsewhere. The MDC slogan—chinja maitiro (change your ways)— held special appeal for urban youth, workers, professionals, and the resiTable 5.1 Politically Motivated Human Rights Violations, 2000–2001 Murder Assault Abduction Assault threats Death threats Property offenses Displacement % State-sponsored

February–June 2000 8 569 142 439 567 263 >2,000 94.5

July–December 2000 1 176 9 33 10 74 57 93.2

January–June 2001 2 192 70 199 71 66 274 88.7

July–November 2001 5 548 35 805 73 360 >4,000 92.1

Sources: Eldred Masunungure, “Travails of Opposition Politics in Zimbabwe Since Independence.” In Zimbabwe: The Past Is the Future, edited by David Harold-Barry (Harare: Weaver Press, 2004), 182–183. Based on data in Amani Trust, Analysis of Zimbabwe Presidential Election, March 9th and 10th, 2002 (Harare: Amani Trust, 2002), 9.

A Period of Crisis, 2000–2008

77

dents of disaffected regions—mainly Matabeleland, but also Manicaland. It was a new party with a fresh agenda whose coalition of supporters (including the private media, particularly the Daily News) had few associations with ruling or opposition political parties from the past. In reaction, the ruling party employed government-controlled television and radio stations to restrict coverage of the MDC, except to depict members as neocolonial pawns (Bratton et al. 2006). Despite state-sponsored violence and electoral irregularities, MDC made significant electoral gains in June 2000, winning almost half of the elected seats in parliament, taking 57 to ZANU-PF’s 62. The opposition challenged the official election results in 39 constituencies and won four court cases, thereby forcing the ruling party to rely for a reliable majority in the National Assembly on the 30 members appointed by the president. For the first time, Zimbabwe had a full-fledged opposition party with enough legislative votes to block further constitutional amendments. The MDC was less successful in the March 2002 presidential contest, however, when Mugabe officially claimed 56 percent of the valid vote to Tsvangirai’s 42 percent. The campaign was marred by increased violence and intimidation. The incumbent’s hands were strengthened by the strict application of new legislation: the Public Order and Security Act (POSA) outlawed meetings of five or more people without police permission, and the Access to Information and Protection of Privacy Act (AIPPA) required the registration of journalists and banned foreign correspondents. Together, these instruments amounted to the suspension of constitutional protections, the return of colonial era repressions, and the effective re-imposition of a state of emergency. Moreover, a Citizenship Act served to disenfranchise farm laborers and mineworkers who retained ties to Malawi, Zambia, and Mozambique, as well as the few remaining whites who had retained foreign passports. Tsvangirai contested the 2002 presidential election despite a trumpedup treason charge—allegedly plotting to assassinate Mugabe—hanging over his head. After a drawn-out trial that stalled the opposition politically, the MDC leader was acquitted on all counts. The March 2005 elections were relatively more peaceful. But fewer than half of the fatigued and battered electorate turned out to vote: of the 120 contested seats, ZANU-PF captured 78 to MDC’s 41, with the expanded ruling-party majority a testament to the effects of vindictive prosecutions, hate speech, violence, and fear. However, the MDC was also in disarray, flip-flopping on plans for an election boycott, which confused the electorate. In 2005 the party split over the issue of whether to participate in elections for a reintroduced Senate. The emergence of an MDC splinter group (MDC-M) led by Arthur Mutambara and Welshman Ncube was prompted in part by concerns that Tsvangirai was making unilateral decisions and violating internal party rules to advance his own stalwarts.

78

The Roots of Repression

Abandoning the Rule of Law

In a further bid to reestablish its lost authority, ZANU-PF drastically curtailed the residual independence of state institutions. The main victim was the rule of law. Until then, the judiciary had retained a good measure of professionalism and autonomy, often protecting the rights of civic organizations and trade unions and issuing verdicts against the government in constitutional test cases. But the planned anarchy of 2000 and beyond, whether land invasions, political intimidation, or election tampering—along with an amnesty cynically granted by Mugabe for perpetrators of violence in the 2000 election—all undermined the legal order (Feltoe 2004; Compagnon 2011). In ruling on a case brought by the Commercial Farmers’ Union, the Supreme Court found that the fast-track land reform program had not been carried out in conformity with the constitution or with laws that the government had itself enacted. The chief justice also overturned a presidential decree that banned defeated candidates from challenging election results in court. Rather than complying with the court on these matters, however, senior government officials (including the minister for information in the president’s office, Jonathan Moyo, and the minister of justice, Patrick Chinamasa) condemned the judges. Emboldened by signals from party leaders, war veterans invaded the Supreme Court. Ultimately, in 2001, the chief justice and two High Court judges, who happened to be white, were forced into early retirement under threats of physical harm. To replace Chief Justice Anthony Gubbay, for example, Robert Mugabe appointed Judge Godfrey Chidyausiku, a former ZANU-PF minister and reliable ally. When others on the Supreme Court refused to resign, the government expanded the bench by promoting sympathetic jurists from the High Court. Interference in the provision of justice was ongoing. The party routinely assigned politically sensitive cases to judges deemed ideologically reliable. In the lower reaches of the judicial system, prosecutors and magistrates came under intense political pressure, up to and including attacks on their families and property. The erosion of public and judicial service salaries due to inflation and budget shortages led to vacancies and delays. Understaffed, many district courts in Zimbabwe operated with a single magistrate. Under these circumstances, it was remarkable that the justice system maintained even a semblance of capacity, if only for routine civil and criminal matters. But when it came to sensitive cases with political overtones, the Zimbabwean justice system “ceased to be independent and impartial; the legal culture [had] been subverted for political ends” (Irwin 2004, 1; see also Gubbay 2009). In short, the ruling elite dropped any pretense of constitutionalism. While not hesitating to rule by law—for example, by concocting legal

A Period of Crisis, 2000–2008

79

charges against political opponents—they openly scoffed at a rule of law as represented by the independent authority of the courts. The 2005 Constitutional Amendment Act No. 17 effectively vested ownership of acquired land in the state and ended court jurisdiction over any challenge to acquisition. The fast-track land reform program was justified on political grounds as a return of land to its rightful owners, implying that all commercial farmers were illegal occupants, even those who had bought farms after independence, sometimes from the government itself. Invoking a political rationale addressed at a historical injustice, ZANU-PF leaders instead enquired rhetorically, “Where was the rule of law when the land was being taken away from black people?” Beyond the judiciary, political purges also occurred at the grassroots, where teachers, civil servants, and local government employees who had worked for MDC during the election season were forced out of their jobs (McGregor 2002). Violence was also systematically organized in the ZANU-PF strongholds in the Mashonaland provinces, presumably to discourage former party loyalists from crossing over to the opposition (LeBas 2011). Land committees led by war veterans or party or security officers sidelined local government structures by taking over the distribution of occupied farmlands and drought relief. ZANU-PF dismissed the elected MDC mayors of Harare and Chitungwiza, replacing them with appointed city managers. Previously independent bodies like the University of Zimbabwe and the Zimbabwe Republic Police were politicized by amendments to their governing legislation or the promotion of ZANU-PF sympathizers to command and management positions. The party even penetrated civil society by targeting church leaders like Catholic Bishop Pius Ncube, who was placed under surveillance by the Central Intelligence Organization (CIO), and Lutheran Bishop Ambrose Moyo, who, having taken a public stance against political violence, felt compelled to flee the country. Recognizing that working people had crossed over to the opposition, the ruling party also tried to reassert control over the labor movement. Senior ZANU-PF officials created an ad hoc labor committee and tried to breathe life into a dormant splinter organization known as the Zimbabwe Federation of Trade Unions (ZFTU). In addition, they prompted war veterans to infiltrate the ZCTU membership, run for vacant posts in the ZCTU leadership, and get involved in labor disputes. Indeed, the party relied increasingly on informal militias. In return for personal immunity for having embezzled war veterans’ funds, Hunzvi delivered the ZNLWVA to ZANU-PF where it became an official auxiliary force. With the emergence of Joseph Chinotimba as their leader, the war veterans adapted tactics from the rural land invasions to raid urban businesses, occupy factories, and extort payments, all purportedly in the

80

The Roots of Repression

name of defending workers’ rights. Another ally was Border Gezi, governor of Mashonaland Central province, who transformed the National Youth Service Program into a school for paramilitary hit squads known colloquially as the “green bombers,” due to the color of their uniforms. While it remains unclear who gasoline-bombed the printing press of the Daily News, once Zimbabwe’s most popular newspaper, the attack in 2001 followed one day after ZANU-PF Minister of Information Jonathan Moyo threatened to silence the paper for posing a security threat to the nation. The breakdown in the rule of law led to international isolation. A Commonwealth ministerial action group sought to engage the ZANU-PF leadership in a commitment to restore due process in land reform. After a Commonwealth observer group found irregularities in the 2002 presidential elections, Zimbabwe was suspended from the Commonwealth. In 2003, defiantly announcing that his regime would not be bound by international legal or human rights norms, Mugabe pulled the country out of the organization. Other international actors also stepped up pressure. In 2001, the United States promulgated the Zimbabwe Democracy and Recovery Act (ZDERA). This legislation promised to restructure the country’s debt and to support an equitable land reform program but only after an end to politically motivated violence, the completion of free and fair elections, and the subordination of the armed services to an elected civilian authority. In the meantime, donors suspended all but the most urgent humanitarian aid, namely, items to combat drought or HIV/AIDS. In addition, because Zimbabwe had defaulted on international loans in 1999, it became ineligible for development assistance from the World Bank and International Monetary Fund. In 2003, Western governments imposed targeted sanctions on about 200 individual party, military, and business leaders, later extending the same measures to their selected family members and to state-owned enterprises. For persons identified with rights violations, overseas assets were frozen and prohibitions put in place against financial transactions and international travel. In addition, the United States and the European Union prohibited the sale of arms to the Zimbabwe government. During the mounting crisis, the labor movement, civil society, and MDC continued to receive modest amounts of international assistance. The ZANU-PF publicity machine used these transfers to bolster its claim that the opposition lacked authentic nationalist roots and was merely a surrogate for Western interests. As relations deteriorated with the West, the government’s foreign policy increasingly began to “look East” by cultivating ties with partners in China, Malaysia, Libya, and Iran, countries that were unconcerned about human rights, good governance, and the rule of law.

A Period of Crisis, 2000–2008

Militarization of the State

81

During the 2000s, Zimbabwe’s electoral authoritarian regime hardened considerably. To understand this trend, it is helpful to recall how military commanders often challenged civilian leaders during the armed struggle. The militarization of politics and the politicization of the military actually started in the 1970s. These trends ebbed during the 1980s and 1990s, when a British Military Advisory and Training Team (BMATT) introduced a professionalization program for the armed forces and ZANU-PF faced no credible threat. However, when a serious challenge emerged in the form of MDC, the military leadership—who still thought of themselves as ZANLA cadres in Zimbabwe Defense Force uniforms—reasserted an undying commitment to the party of liberation. This development coincided with—and probably also caused—the departure of BMATT in 2000 and, with it, a force for moderation.3 Since the independence war, military commanders had always sat on the party’s central policymaking bodies. But they remained largely behind the scenes until ZANU-PF began to lose seats in parliamentary and local government elections, at which time senior officers were seconded into strategic political posts formerly occupied by reliable civilians. A quick count in early 2008 turned up 44 current or retired military officers serving as cabinet ministers (4), permanent secretaries or directors (7), ambassadors (4), members of parliament (7), and managers or board members of parastatal corporations (22) (Zimbabwe Institute 2008). Among the penetrated ministries were foreign affairs, energy, transport, and justice, where Attorney General Sobusa Gula-Ndebele was a former director of military intelligence. Military leaders also appeared at the helm of strategic economic institutions, including the National Oil Company of Zimbabwe (NOCZIM), the National Railways of Zimbabwe (NRZ), the Commercial Bank of Zimbabwe (CBZ), and the Grain Marketing Board (GMB). In response to a perceived national security threat posed by a rising opposition movement, the party restructured the top levels of the state apparatus. A Joint Operations Command (JOC) of security agencies became the supreme, but unofficial, decisionmaking body. Representing a small cabal of top political and military leaders, the JOC sidelined the civilian cabinet and even rivaled the party Politburo. Meeting weekly and attended by the commanders of the defense forces, army, police, air force, intelligence service, and prisons, as well as the ministers of defense and state security, the JOC reported directly to the president in his capacity as commander-inchief of the armed forces. The governor of the Reserve Bank, an ex officio member, was charged with providing JOC with revenue on demand, including supplemental financing for the security forces. Operating as a private fiefdom, the JOC defined its remit as any policy issue deemed to impinge

82

The Roots of Repression

on national security, broadly defined. It reportedly had nine task forces with responsibilities across all economic sectors, including the management of foreign exchange and monetary policy. The political rise of the security elite imbued the management of the party-state with military-style “operations.” Without prior warning and with little advance planning, the regime would suddenly announce an emergency policy to be implemented by the army, police, or armed auxiliaries. The civilian cabinet, line ministries, parliament, and local authorities were seldom consulted. The arbitrary mode of governance that originated in the fast-track land reform program became standard operating procedure. Examples of such JOC decisions include Operation Murambatsvina (Clean out the Trash) in 2005, which was designed to stifle the informal economy by ransacking business properties and punish the MDC’s urban followers. This campaign—in which some 700,000 urban Zimbabweans were rendered homeless or lost their livelihood, and up to 2 million were indirectly affected—was condemned by the United Nations and other international agencies as a gross violation of human rights (Tibaijuka 2005; Potts 2006; Bratton and Masunungure 2007; Vambe 2008). It was followed by Operation Garikayi (Live Well), a housing scheme directed by army commanders and ostensibly meant for Murambatsvina victims but that ended up benefiting members of the security forces and their extended families. Also in 2005, the army used Operation Taguta (Eat Well) to seize agricultural properties and equipment, effectively sidelining producers on commercial and public estates. They also took control of food distribution, forcing newly resettled farmers to plant maize at the expense of other crops and requiring sales to the state marketing board at below-market prices, measures that proved deeply unpopular in rural areas. The JOC and ruling party—whose roles and top personnel were deeply fused—relied upon the Central Intelligence Organization to provide surveillance of the population and on the Zimbabwe Republic Police (ZRP) to crack down on unauthorized political activity. War veterans and youth militias associated with both the ruling party and a national service scheme were also added to this lethal mix (Scarnecchia 2006). As these elements were incorporated into the security forces via recruitment to the army, prisons service, and police, the distinction between regular military and paramilitary functions became blurred. Because shadowy militias acted as proxies for the party-state, it was often unclear exactly who was ordering intimidation, abductions, and torture and who was carrying out these orders. And even when abuses in police or intelligence services were documented, perpetrators were rarely charged and invariably escaped penalty. Thus, human rights violators among the political and security elites came to expect that they would be protected by a culture of impunity.

A Period of Crisis, 2000–2008

83

Ordinary Zimbabweans were traumatized by such developments. Public opinion research showed that more than four out of five citizens were afraid to speak openly about politics.4 Political fear, in turn, led people to adopt a risk-averse approach to public life: citizens became less willing to join strikes, participate in protests, or otherwise engage in organized resistance (Masunungure 2006). After leading a successful stay-away campaign in March 2003, the MDC’s calls for mass action increasingly fell on deaf ears, and party debates about opposition strategy—confrontation or negotiation—led to divisions and, eventually, a split within the party.5

A Collapsing Economy

As the decade began, the government’s management of the economy was already unsteady: unrestrained borrowing and spending had produced mountainous debt, along with galloping inflation and exorbitant interest rates. Beginning in 1998, the president deployed the Zimbabwe National Army to prop up the fragile regime of Laurent Kabila in the Democratic Republic of Congo, an ill-advised, costly move partly motivated by a search for new sources of national income. Instead, the deeply unpopular Congo adventure ended up monopolizing scarce supplies of fuel and foreign exchange and squandering public money at the rate of US$1 million per day. Meanwhile, health and education spending were sacrificed to the ballooning military budget. Looking for scapegoats, the president blamed the business sector for profiteering and inflation, accusing it of being “in sympathy with white landed interests.” Desperate for foreign exchange, the government formulated unworkable on-again, off-again rules requiring exporters to surrender export earnings in convertible currencies at official rates. This hostile business climate led to the failure or contraction of many firms, thereby swelling the ranks of the unemployed, conservatively estimated by 2003 at 50 percent of the workforce. Over time, ZANU-PF’s economic policies became increasingly erratic and delusional. An alphabet soup of emergency recovery programs— ZIMPREST, MERP, NERP, and NECF—was introduced and abandoned in quick succession. Like King Canute trying to halt the incoming waves, Mugabe used the president’s decree powers in a vain effort to repeal the basic laws of market economics. Faced with sky-high inflation at the end of 2006, for example, the government announced Operation Sunrise, which required citizens to turn in their old currency in order to receive devalued banknotes marked in higher denominations. Operation Reduce Prices in June 2007 mandated that retailers cut the amounts charged for basic commodities by 50 percent. Teams of inspectors recruited from the army, po-

84

The Roots of Repression

lice, and intelligence services oversaw the clampdown. Not surprisingly, stocks of basic commodities like maize-meal, bread, sugar, and cooking oil soon disappeared from shop shelves as producers found themselves unable to supply goods at prices below the cost of production. Instead, parallel markets thrived, where goods were sold at even higher prices and often only in foreign currency. Despite these policy failures, few within the political elite “dared to suggest a course of action at odds with the preferences of the president” (Kanyenze 2004, 139; see also Makina 2010). As a result, the last of the remaining technocrats in the inner circle resigned in frustration, including Minister of Industry and International Trade Nkosana Moyo and Minister of Finance and Economic Development Simba Makoni. In 2004 Mugabe appointed Gideon Gono as governor of the Reserve Bank of Zimbabwe (RBZ), whose remit was to generate the resources for the party’s core political and military projects, now including the salaries of the regular and auxiliary armed forces. Gono did so by pumping up the money supply and assuming quasi-fiscal functions previously held by the government’s economic and spending ministries. The RBZ not only required commercial banks to place half their assets in reserve accounts, it also raided the foreign exchange reserves of private companies, pension funds, and NGOs. Side-stepping not only the Ministry of Finance and line ministries, but also parliamentary oversight, Gono acted as a de facto prime minister: he used decree powers to announce major economic policies and fiscal expenditures. By 2007, the RBZ’s reckless policy of printing money led to six-digit hyperinflation that rendered the Zimbabwe dollar virtually worthless (see Table 5.2). The period 2000–2008 ended in a full-blown economic crisis. At a time when real per capita incomes were beginning to rise in the rest of Africa, Zimbabwe was battered by the world’s lowest growth rate and highest inflation. The economy contracted in every year between 1998 and 2008, Table 5.2 Key Economic Indicators in Zimbabwe, 2000–2008 2000

GDP growth (annual %) −6.8 GDP per capital (PPP US$) 3,358 Fiscal deficit (% of GDP) 18.6 Inflation (annual %) 56

2001

2002

3,144

3,068

112

199

−2.7 7.0

2003

2004

2005

2006

2007

2,775

2,737

2,606

2,533

2,346

599

133

586

1,281

−4.4

−10.4

2.7

0.2

−3.8 7.6

−5.3 6.1

−4.8 3.1

2008

−7.4

−14.5

n/a

32.1

2,005

100,000 231,000,000

Sources: International Monetary Fund, “Zimbabwe: Staff Report for the 2010 Article IV Consultation” (Washington, DC: IMF, April 2010); United Nations Development Programme, Comprehensive Economic Recovery in Zimbabwe (Harare: UNDP/Zimbabwe, 2008).

A Period of Crisis, 2000–2008

85

shrinking by half over the whole period. The last official report on inflation in mid-2008 pegged the rate at 231 million percent, although private economists estimated far higher levels. All key production sectors—agriculture, industry, and manufacturing—operated at a fraction of former capacity. Consumers faced extreme shortages of staple foods, motor fuels, foreign currency, and even the worthless local banknotes. Electricity blackouts became a serious and persistent problem. Unemployment surpassed 80 percent. By 2009, up to half the population was dependent on international food aid. Prisoners were dying of starvation and disease in the country’s overcrowded jails (Alexander 2009). Social services crumpled: schools closed countrywide and a cholera epidemic killed 4,000 people (TI 2007; PHR 2009). Adding to this ruin was the specter of AIDS. At the peak of the economic crisis, the national HIV prevalence rate exceeded 30 percent, making Zimbabwe one of Africa’s hardest-hit countries. In urban areas, the level of infection rate was estimated to be around 40 percent and perhaps as high as 80 percent in the army. With funeral attendance a cultural tradition, an estimated 2,000 deaths per week further dragged down economic productivity (Bollinger et al. 1999). A decade previously, Zimbabwe’s health care system had been among the best in Africa. By the mid-2000s, severe shortages of basic drugs, medical equipment, and trained personnel had pushed hospitals and clinics close to collapse. Between 1985 and 2003, infant mortality rates jumped by 15 percent, and average life expectancy plummeted from 61 years to just 42 (World Bank 2009). The economic crisis left the population in despair. The daily life of ordinary Zimbabweans became a difficult ordeal. In a national survey in 2004, more than half of all adults (54 percent) described their standard of living as “bad” or “very bad” (Afrobarometer 2005). Asked to rank themselves on a spectrum ranging from “poor” to “rich,” Zimbabweans said they were worse off than Africans in 15 other countries and better off than just Malawians. When they compared their personal circumstances to those of their parents ten years previously, Zimbabweans perceived greater declines than did residents of any other African country surveyed, including Malawi. This darkening public mood reflected limited access to basic human needs. Take food, for example. As food production slumped, hunger grew. By 2004 only one out of five adult Zimbabweans (18 percent) could report that they and their families had “never” gone hungry during the previous year. As a consequence, perhaps a quarter of the country’s 12 million people took the exit option by migrating to neighboring countries and overseas (UNDP 2010; Maphosa 2010). Their remittances helped to sustain relatives at home, often becoming the single largest source of income, food, and new clothing for households in Zimbabwe’s southern districts.6 However, the brain drain cut into the country’s potential pool of qualified leaders, thus

86

The Roots of Repression

adding to a professional culture of mediocrity and making eventual economic and institutional recovery all the more difficult. Even as many ordinary Zimbabweans sank into poverty, a small, politically connected elite reaped benefits from arbitrage and corruption. Key ruling party loyalists—an ever-shrinking coalition—had vested interests in maintaining the status quo, which included preferential access to land, trading and import licenses, urban housing, petroleum products, and many other commodities in short supply. The most valuable perk was foreign currency speculation: purchased at favorable official rates, hard currencies could be transformed into small fortunes in local dollars when sold on the black market. In short, ZANU-PF leaders quite literally treated politics as a business. Through M&S Syndicate, a holding company, the party purchased interests in real estate properties, motor vehicle sales, the import and distribution of industrial machinery, water pumps, steel, building materials, and mining. These companies enjoyed sweetheart deals to supply government departments with essential goods and services. Even more importantly, ZANU-PF barons (including faction leaders Emmerson Mnangagwa and Solomon Mujuru) built business empires and used those resources to advance personal political careers rather than corporate party interests. Beyond the party, the extension of the official patronage network undermined the integrity of key governmental institutions. The judiciary became corrupted as judges, their salaries decimated by inflation, were wooed with gifts of commercial farms, sport utility vehicles, and HDTV sets. The Zimbabwe Electricity Supply Authority administered a Rural Electrification Fund to extend the national electricity grid to rural areas, especially chiefs’ homesteads, further cementing their political fealty. Agricultural inputs and maize intended for food relief were sold by party functionaries or were awarded to card-carrying acolytes of ZANU-PF, while these supplies were withheld from persons suspected of opposition sympathies. The economic crisis had devastating effects on the quality of governance. Government operations broke down because poorly paid public sector workers lacked motivation or were absent from duty while seeking livelihood in informal markets or lining up to withdraw cash or buy gasoline. By expanding the number of official jobs, ZANU-PF not only sought to purchase the loyalty of its own followers, but it also tempted MDC supporters to buy into a corrupt system. As ordinary citizens realized that they could not survive without breaking currency or trading laws, corruption became a moral crisis that infected all quarters of society.

The Disputed 2008 Elections

The more the government lost control of the economy—which, in turn, reduced the amount of available patronage—the more it tightened control po-

A Period of Crisis, 2000–2008

87

litically. In late 2006, the police beat and arrested leaders of ZCTU and Women of Zimbabwe Arise (WOZA) who were protesting for economic rights and constitutional reform. Teachers and junior doctors who went on strike in early 2007 were also detained. When the labor movement called for another national stay-away in April 2007, the state imposed a ban on political rallies and demonstrations. In a landmark event in March 2007, the police prevented the Save Zimbabwe Campaign (a coalition of church and civic groups) from convening a “prayer meeting” in Harare. Leaders of the opposition, including Tsvangirai, were attacked: one person was killed, 50 were hospitalized, and nearly 200 were arrested. Media coverage led to an international outcry, not only from the United Kingdom, the European Union, and the United States, but also from the African Commission on Human and Peoples’ Rights. Inside the country, the Catholic Bishops issued a pastoral letter asserting that “black Zimbabweans today fight for the same basic rights they fought for during the liberation struggle.” Mugabe’s reaction to this criticism was to announce: “If they [protest] again, we will bash them again” (OSI 2007, 9; see also HRW 2007). In an upsurge of state-sponsored violence, the frequency of arrests, assaults, and reports of torture increased in 2007 compared with previous years. The principal targets were individuals holding leadership positions in MDC, which the regime tried to brand as “terrorists.” Especially disturbing was the appearance of new tactics, including home invasion and kidnapping, in which victims were beaten in front of their families or abducted and taken to secret sites. Lawyers who sought to locate and defend victims were also targeted for violence. The crackdown and growing regional unease about the broader national crisis prompted the Southern Africa Development Community (SADC), the region’s official intergovernmental body, to convene an extraordinary summit meeting in Dar es Salaam, Tanzania. Even as SADC expressed continued solidarity with the ZANU-PF regime, the summit’s final communiqué instructed South African President Thabo Mbeki to facilitate dialogue between the government and opposition in Zimbabwe. Whereas elements within the MDC appeared committed to the mediation process, ZANU-PF displayed much more reluctance. Mugabe delayed sending envoys to the talks in Pretoria and repeatedly refused to enter into face-to-face deliberations with MDC leaders. Representatives of civil society were denied a seat at the table. Critics also rejected Mbeki’s strategy of “quiet diplomacy,” which essentially amounted to an unwillingness to publicly criticize Mugabe’s policies, as inadequate to secure meaningful concessions. Ultimately, however, a measure of progress was made. In September 2007, a small group of senior leaders from the main ruling and opposition parties met behind closed doors at a resort on Lake Kariba to negotiate a draft constitution. The Kariba Draft, as it became known, sought to recon-

88

The Roots of Repression

cile the existing charter with elements of the rejected 2000 proposal, while leaving executive powers broadly untouched. In December, the government announced piecemeal legislative reforms to POSA (to allow political rallies so long as police identified no threat of violence) and the Broadcasting Services Act (to guarantee balanced coverage of election campaigns and to selectively allow licensing of journalists and broadcasters). No sooner had agreement been reached, however, than Mugabe unilaterally declared a timetable for elections in March 2008 without addressing the agreed-upon precondition of comprehensive constitutional reform. Moreover, it soon became apparent that ZANU-PF did not intend to abide by the new laws: the police continued to block or harass opposition gatherings and the government-controlled media continued to praise the ruling party and castigate the opposition (MMPZ 2009). And ZANU-PF—intent on bolstering its main social base among the peasantry—turned traditional leaders into appendages of the ruling party. Pampered with development services and consumer goods, the chiefs were expected to act as the party’s eyes, ears, and whips in their local districts and to deliver the rural vote. The political crisis came to a head with general elections in March and June 2008. At first it looked as if the opposition had made the breakthrough that it had long sought. MDC-T won more seats (99) than ZANUPF (97) in the House of Assembly and also secured control over most municipal councils. In the House of Assembly, the balance of power was held by the MDC-M splinter group headed by Arthur Mutambara, which won 10 seats. After suspicious delays with the presidential election results, however, the Zimbabwe Electoral Commission (ZEC) announced that no candidate—neither Tsvangirai with 48 percent of valid votes nor Mugabe with 43 percent—had achieved the absolute majority necessary to be declared winner in the first round.7 A constitutionally mandated runoff election was, therefore, scheduled for June 27, 2008. According to reliable accounts, Zimbabwe’s top military leaders quietly seized political control after the first round of the election. Mugabe reportedly informed the security chiefs that he had lost the presidential vote and was considering surrendering power.8 But the commander of the Zimbabwe Defense Forces (ZDF), Constantine Chiwenga—backed by police chief Augustine Chihuri, air force head Perrance Shiri, CIO director general Happyton Bonyongwe, and director of prisons Paradzai Zimondi—allegedly vetoed this proposal. Perhaps fearing exposure to prosecution for rights abuses, they insisted that Mugabe contest and win a runoff election. As the International Crisis Group observed: The security establishment—police, intelligence and army—has always perceived itself as a praetorian guard for the country’s “nationalist revolution” but it has progressively become a bastion of the ZANU-PF architecture of vi-

A Period of Crisis, 2000–2008 olence, eroding professional neutrality and making it a threat to public security and democracy. (ICG 2006, 8)

89

These developments represented an evolution of the governing coalition. Due to the requirements of armed struggle, ZANU-PF politicians and guerrilla commanders had always enjoyed a mutually beneficial alliance. At independence, the integration of rival armies into a professional military establishment forced ZANU-PF’s civilian and military wings to part company temporarily, though Mugabe made sure to install political and ethnic loyalists in top security positions. ZANU-PF leaders apparently decided that the crisis conditions in the economy and electoral setbacks in 2000 and 2008 meant it was time for a new liberation war. As such, with the party’s military wing now in the lead, they resorted to the familiar tactics of armed struggle by forcing a largely rural electorate to vote their way. On the brink of losing the presidency, ZANU-PF launched a terror campaign that revived memories of the Gukurahundi bloodbath of the 1980s. The JOC security chiefs (“with the party chairing”9) were in charge. The country was divided into 10 provincial command centers staffed by 200 serving army officers, who were dispatched mainly to rural areas to mobilize militias and supervise the election campaign. To finance the terror, the JOC ordered the Reserve Bank to print money to fund pay hikes for war veterans, allowances for youthful fighters, and payoffs to chiefs and headmen. These combined forces were ordered to invade farms, burn down houses, and incite other forms of violence, all to force the population into voting for ZANU-PF. In order to conceal responsibility, auxiliaries were sometimes issued police or army uniforms, while undercover, plainclothes, or masked intelligence operatives carried out kidnappings and murders. Persons suspected of supporting MDC were beaten in public or snatched away to torture centers often located at rural schools. The police obligingly turned a blind eye to any politically motivated acts of abuse (Godwin 2010). The rulers’ strategy for the runoff—code-named Operation Mavhotera Papi (How Did You Vote?)—was electoral cleansing. The objective was to kill MDC officials and polling agents, displace qualified electoral officials like schoolteachers, and punish known MDC supporters, thus eliminating the opposition as an electoral threat. The targets of intimidation were not so much the solid MDC strongholds in the cities and the southwest, but the politically contested areas in the country’s middle belt and northeast, where first-round voters had swung away from ZANU-PF and toward MDC. The object of electoral cleansing was to create “no-go zones” where the ZANUPF monopoly could be enforced at the local level through the direct and demonstration effects of violence. In a further sanction, the party ordered food relief withheld from opposition sympathizers and commercial food

90

The Roots of Repression

supplies distributed only to shops operated by ZANU-PF supporters or personnel. A villager from Shamva, Mashonaland Central Province, whose homestead had been burned, succinctly summarized the strategy: “Only ZANU-PF people will vote. There are no opposition supporters. It will be a big advantage for them.”10 In Harare, top MDC leaders were forced into hiding as a result of police raids on their party’s national headquarters. Threats against the opposition leadership were rendered shockingly real when the wife of the newly elected MDC mayor of Harare was abducted and executed. Tsvangirai embarked on an international tour of African and European capitals to draw attention to the crackdown and to encourage international isolation of the Mugabe government. Perhaps missing an opportunity to mobilize his supporters in a protest campaign, he returned to Zimbabwe only shortly before the runoff. But, citing the unacceptable cost of the violence, he ultimately withdrew his candidacy and took refuge in the Netherlands Embassy. Like Burma’s defensive military junta at the time, the Zimbabwe government kept foreign relief agencies and international journalists out of the hinterlands. Despite the government’s best efforts, however, news of the regime’s brutality and the dire costs of its policy to use food as a political weapon began to trickle out. Unlike in the 1980s, when the Matabeleland massacres were hidden from public view, the 2008 crackdown was carefully documented and publicized by local and international civic bodies (HRW 2008; SPT 2008; ICG 2008a). Up to 200 MDC officials and supporters were killed, thousands injured in politically motivated beatings, and perhaps 200,000 people displaced from their homes. The raw power politics of ZANU-PF undoubtedly succeeded in raising the cost of supporting the opposition. But in so doing, the ruling party revealed its own deeply antidemocratic tendencies. By blocking free and fair electoral processes and outcomes, the militarized authoritarian regime showcased its own bloody credentials for all the world to see. On June 27, 2008, Mugabe claimed a hollow victory with 85 percent of the vote from a brutalized and shrunken electorate But few in the West regarded him as the legitimate president of Zimbabwe. For the first time, other African leaders began to call collectively for a transitional arrangement, including perhaps a power-sharing settlement. The United Nations Security Council considered a resolution to impose an arms embargo and targeted sanctions, but the measure was vetoed by two permanent members—Russia and China—and opposed among others by Libya, Vietnam, and South Africa. By joining the international pro-Mugabe bloc, Pretoria indicated the deep ambivalence in its foreign policy toward meaningful political change in Zimbabwe. Back inside the country, Tsvangirai claimed that “the country has witnessed a de facto coup d’etat and is now effectively run by a military junta.”11

A Period of Crisis, 2000–2008

91

More accurately, the resultant regime was a hybrid civilian-military coalition. The visible leadership continued to be drawn from party ranks: Mugabe retained his position as national president and Emmerson Mnangagwa, a senior Politburo member and a contender for presidential succession, chaired the JOC.12 In part to deflect international condemnation, the Zimbabwean generals seemingly preferred to present a civilian political façade. But their apparent dominance in the tense 2008 period between the first and second rounds of the presidential election pointed to the mutual interdependence of military and civilian elements in Zimbabwe’s governing elite. And it raised the question, at least during moments of political crisis, as to which leaders—civilian or military—were in ultimate control.

A Critical Juncture

In sum, by mid-2008, Zimbabwe’s economic crisis became linked with a crisis of power politics. An autocrat, having manipulated the system to put himself back into power after almost (and probably actually) losing an election, faced a rapidly collapsing economy. Such was the leadership legacy of three decades of ZANU-PF rule. As a result, Zimbabwe’s ruling and opposition elites were profoundly polarized; they could not agree on who had the right to rule. On one hand, the dominant party persisted with its claim that it should govern forever by virtue of its contribution to liberating the country from settler colonialism. The 2008 election confirmed that ZANU-PF had no intention of recognizing an opposition victory in the presidential poll or surrendering control of a government that they regarded as theirs alone. Statements by ZANU-PF barons continued to reflect a persistent sense of political entitlement. With reference to the presidential runoff, a Politburo member warned that the party did not intend to abide by the ballot box: “We’re giving the people of Zimbabwe another chance to mend their ways, to vote properly . . . this is their last chance.”13 Likewise, Grace Mugabe (the president’s wife) publicly stated that “even if people vote for MDC, Morgan Tsvangirai will never step foot inside State House.”14 On the other hand, only the electorate could freely grant popular political legitimacy. The opposition claimed that state-sponsored violence and the manipulation of election procedures had invalidated ZANU-PF’s bid for authority. At the same time, the MDC leader bore responsibility for not making the most of opportunities offered by the slow-motion theft of the 2008 elections. A split opposition meant that MDC-T and MDC-M, whose combined support could have been enough for an outright victory, entered the contest at a disadvantage.15 Moreover, Tsvangirai’s decision to spend extended periods outside the country to seek international backing left his

92

The Roots of Repression

party leaderless at key moments during the runoff campaign. Finally, the decision of an otherwise courageous leader to withdraw from the election was read by some supporters as a loss of nerve that cleared the way for a Mugabe sweep (Chan 2008). As a result, MDC-T found itself unprepared to mobilize its supporters into the streets at the crucial interval between the news that Mugabe had lost the first round of the presidential election and before the security chiefs could organize a reign of terror. Following the disputed 2008 elections, the political preferences of the rival parties hardened into non-negotiable positions. There was little appetite for dialogue or movement toward a middle ground on either side. And yet, common sense led ordinary Zimbabweans to believe that a negotiated political settlement between rival coalitions was a prerequisite for a return to normalcy in the country. Some 70 percent of adults nationwide had long agreed that “problems in this country can only be solved if the MDC and ZANU-PF sit down and talk with one another” (Bratton et al. 2006, 105). In this regard, the peak of the crisis was a critical juncture. A collapsing economy and disputed elections served as catalysts to break, at least temporarily, the fever brought on by unrestrained power politics. By mid2008, ZANU-PF had proved that it could enforce its political dominance over a cowed population by deploying military might. But Mugabe’s victory in the presidential runoff was pyrrhic: formally speaking, he “won” the uncontested election; but, in forcing himself upon the electorate, he lost a leader’s most valuable informal resource: popular legitimacy. Furthermore, increasingly isolated from the West and even from former African allies, the party could no longer rely on legitimacy granted from the international community. Unable to ignore growing pressures from internal and external sources, the ruling party would have to reach a new political settlement with adversaries in the democratic opposition.

Notes

1. For an argument interpreting land occupation as a more orderly process see Chaumba et al. (2003) and Scoones et al. (2010). Scoones and colleagues argue that some resettled farmers subsequently made gains in agricultural productivity and living standards. The most balanced account is Hanlon 2013. 2. Quoted in Scarrnechia 2009, 16. 3. I am indebted to Eldred Masunungure for the insights in this paragraph. 4. In an Afrobarometer survey in October 2005, 88 percent of adult Zimbabweans said that they “always” or “often” had to “be careful of what they say about politics.” Zimbabwe: Summary of Results. http://wwwww.afrobarometer.org. 5. At issue was whether the MDC was itself justified in resorting to reactive violence (LeBas 2011, 201–210).

A Period of Crisis, 2000–2008

93

6. The International Fund for Agricultural Development estimated that official and informal remittances could be between US$360 million and US$490 million annually. The Zimbabwean (December 24, 2010). http://www.thezimbabwean.co.uk. 7. The Zimbabwe Election Support Network conducted a parallel vote tabulation (PVT) based on a random sample of polling stations. It suggested that, subject to a margin of sampling error, Tsvangirai had achieved an absolute majority on the first round. But it appears that ZANU-PF’s electoral magicians made the most of the confidence interval around the PVT estimate to concoct a defensible “final” figure for Tsvangirai that fell just short of the required majority. 8. The most authoritative rendition of the events leading to the JOC takeover is Craig Timberg’s “Inside Mugabe’s Violent Crackdown.” Washington Post (July 5, 2008). See also Celia Dugger, “Slow Motion Coup.” New York Times (April 26, 2008) and Allister Sparks, “Zimbabwe’s Military Feels the Heat.” Cape Times (April 30, 2008). 9. “Zimbabwe Campaign: Secret Documents.” BBC (June 12, 2008). 10. BBC, May 19, 2008 11. “Zimbabwe Opposition Leader Says Country Run by Military.” Voice of America (April 10, 2008). 12. A hardliner with an intelligence background, Mnangagwa was minister for state security during the Matabeleland massacres of the 1980s. He was brought in to head Mugabe’s presidential runoff campaign reportedly because of his reputation for strong-arm tactics. He was later appointed minister of defense. 13. “Violence in Zimbabwe Disrupts Schools and Aid.” New York Times (May 9, 2008). 14. “Zimbabwe’s Mugabe Will Never Step Aside for Rival: Wife.” Agence France Press (May 29, 2008). 15. While Mutambara did not contest the presidency, he endorsed Simba Makoni’s third-party run. MDC-T lost parliamentary seats to MDC-M, especially in the southwest provinces.

Part 2 Power-Sharing Settlements

6 African Experiences with Power Sharing

After the debacle of the 2008 election, political elites in Zimbabwe turned to power sharing as a means to address the devastating costs of unbridled power politics. Before discussing Zimbabwe’s five-year experiment with coalition government (2008–2013), it is necessary to place power sharing in a continental context. The Zimbabwe case unfolded against a wave of power-sharing agreements in Africa and raises general questions about political settlements in Africa and beyond. This chapter therefore asks: Are authoritarian political elites, including party and military spoilers within their own ranks, able to distort power-sharing settlements? Does a short-term quest to bring an end to political conflict undermine longer-term institutional development? Or are there cases where power sharing fosters both a durable peace and democratization? And what are the conditions that explain any differences in outcomes?

The Proliferation of Power Sharing

The proliferation of power sharing in Africa allows for comparative examination of these questions. Over the last several decades, various forms of coalition government have been used to address deep-seated problems of political disorder. Of course, a predilection for power sharing is hardly new among policy analysts. Arthur Lewis was early to argue that coalition governments are more appropriate than winner-take-all elections for ethnically divided African societies (1965, chap. 3). Heeding this insight, some African presidents sought to ameliorate conflict by broadening the base of political regimes, for example, by balancing ethnic or sub-regional appointments to senior state offices. The goal was to manage discord by ensuring

97

98

Power-Sharing Settlements

that various organized communities felt represented at the political center and enjoyed a share of the fruits of patronage. According to a similar logic, national and international elites have converged in recent years on power sharing as a preferred remedy for a wide range of political crises (Spears 2002; Roeder and Rothchild 2005; Lijphart 2008). Negotiators have applied this mechanism to at least three situations. The first was the transition from settler regimes, where outgoing white minorities received special political representation as an interim step toward genuine popular rule. Apart from Zimbabwe’s independence settlement, the clearest example in Africa is the interim government of national unity in South Africa from 1994 to 1996. In this case, power sharing helped pave the way toward a multiracial democracy based on principles of political equality and majority rule. Second, comprehensive peace agreements to end civil wars have routinely provided for the formation of unity governments. In these cases, former adversaries have negotiated a share of executive power as a way to gain a foothold within the state, thus lessening and domesticating the stakes of political struggle. Examples abound in Africa, including in Rwanda (1993), Sierra Leone (1999), Democratic Republic of Congo (2002), Liberia (2003), Burundi (2004), and Sudan (2005) (Gates and Strom 2009). The range of outcomes from civil war settlements is wide: from a genocidal intensification of conflict (as in Rwanda), through the creation of a shaky and temporary peace (DRC), to the halting passage toward a more democratic regime (Liberia and Sierra Leone). Third, power sharing seemingly offers a solution to disputed election outcomes, notably where practitioners of power politics refused to cede their grip on the presidency. By the end of the first decade of the current century, poor-quality elections were increasingly frequent in places like Togo (2005), Madagascar, Nigeria, and Kenya (2007), Djibouti (2008), as well as Ethiopia, Rwanda, Sudan, and Côte d’Ivoire (2010). As contending parties clashed over the legitimacy of official voting results, policymakers, especially external third parties, looked around for solutions that would restore order and legitimacy. Much as reformers had once recommended free and fair elections as a promising therapy for civil war and bad governance, they alighted on power sharing as a panacea for postelectoral conflict. But, as stated earlier, one size does not fit all. This chapter argues that power sharing has worked better for easing transitions from settler colonial rule and ending long-standing civil wars than for resolving election disputes. Among the critical differences across these situations are the nature of the state and the relations of elites to state power. When ending settler rule, governing elites have inherited a strong state whose institutions were largely intact. Power passed formally to an African

African Experiences with Power Sharing

99

majority, and any token political representation for non-African groups was time-limited by constitutional sunset clauses. To be sure, settlers retained a measure of political influence through the state bureaucracy, but this power advantage was soon diluted as Africans were promoted into public sector posts. In time, settlers retreated into the economy, where they still enjoyed privileged positions, leaving African political elites to take over and adapt political institutions, usually by entrenching one-party dominance. Except for the brief interlude of power sharing at the beginning of the independence period, no political elite was able to capture the institutions of a strong state since the incoming black leadership had yet to consolidate its own power. In this chapter, the case of South Africa’s Interim Constitution of 1993 will illustrate the features of power sharing at the close of settler rule. By contrast, civil wars undermine state capacity. The parties to powersharing agreements under these conditions inherit state institutions that are extremely fragile, perhaps even close to collapse. No party—whether an embattled group of incumbent leaders or factionalized groups of rebels— enjoys a firm grip on power. Because resources are limited and because all parties are organizationally weak, it is again difficult for any one group to capture the state and turn it against rivals. Because no one is able to dominate, everyone has an incentive to cooperate, and power sharing at the end of civil wars often allows a reprieve from violence. Much less certain, however, is the resolution of the underlying causes of the conflict. Because former enemies have few common interests and often harbor malevolent intentions, written agreements are inherently unstable or dependent on third-party enforcement. In this chapter, the case of the 1999 Lomé Peace Accord in Sierra Leone will represent the features of peace agreements at the end of civil wars. As a formula for addressing disputed elections, power sharing occurs under distinct institutional conditions. In these cases, state power is neither strong and intact (for example, as a product of settler rule) nor fragmented and fragile (for example, as a result of debilitating civil war). Rather, the overall capacity of the state is intermediate, being strong enough to manage a disputed election but not legitimate enough to win widespread public or international approval. In this chapter, the case of Kenya in 2008–2103 (a time period that coincided with power sharing in Zimbabwe) will epitomize political settlements between rival political elites in the aftermath of disputed elections. The general lesson is as follows: power sharing does not work well when an intransigent rump of incumbent leaders controls the intact parts of a moderately strong state. The very fact that such leaders seek to continue ruling even after losing elections indicates that they are primed to disregard the agreed-upon rules of the political game. Autocrats who lose elections may be pushed by international patrons to sign negotiated accords, but they

100

Power-Sharing Settlements

cannot be compelled to implement them. In Kenya, leaders who were responsible for electoral violence colluded to use the residual power of the state to avoid accountability. Because rulers had little respect for the rule of law, moderates discovered that their efforts to introduce the constitutive political reforms necessary for legitimate political settlements were blocked.

Dismantling Settler Rule: The Case of South Africa

Because South Africans devised an indigenous process of direct negotiations, they possessed an unusually strong sense of ownership of the resulting political settlement. Contenders for power engaged in informal contacts and talks before and during a formal series of negotiations. Civil society actors similarly launched parallel dialogues within local social institutions. Critically, the powerful institutions of state security, which remained intact throughout the transition, eventually pulled back from sponsoring violence and pledged loyalty to a new civilian authority. Although international sanctions helped to weaken the apartheid regime, outside mediation in negotiations was unnecessary because domestic actors were able to resolve differences largely on their own. In early 1993, the governing National Party (NP), headed by South African President F. W. de Klerk, and the African National Congress (ANC), under the leadership of Nelson Mandela, held a string of bilateral, closed-door meetings to hammer out a transition to democracy. In order to move peaceably toward a common political destination—majority rule—the main rivals agreed to a number of interim steps. Principal among these was the creation of a transitional government of national unity (GNU). Inaugurated following the first multiracial elections in 1994, the GNU was designed to last for five years; but it effectively dissolved in October 1996 when de Klerk pulled the National Party out of the coalition. But the brief powersharing interlude served its purpose as a bridge between minority settler rule and majoritarian, nonracial democracy. Given South Africa’s violent political past, a peaceful transition was unexpected (Friedman and Atkinson 1994; Sparks 1995; Maphai 1996). At the time, observers across the political spectrum feared that apartheid was destined to provoke a racial bloodbath. The countless hardships unleashed by apartheid—declared a “crime against humanity” by the United Nations—fueled mass protests against pass laws at Sharpeville in 1960, student riots against Afrikaans-language instruction in Soweto in 1976, and a united front of community-level resistance in the 1980s. With nationalist Settlement

African Experiences with Power Sharing

101

organizations banned and a state of emergency in effect, the ANC and other parties were forced to operate underground or from abroad. As the state resorted to ever more brutal tactics against political activists—executions, assassinations, disappearances, torture—the nationalist movements stepped up their own campaigns of mass disruption and insurgent violence. By the late 1980s the country’s future looked bleak. The South African economy was under stress from international sanctions, ungovernable urban townships, and the rising costs of defending borders. Recognizing the risks of escalating conflict, a few far-sighted members of the Afrikaner intelligentsia and English-speaking business elites reached out to ANC leaders in exile. In a sequence of private discussions in Harare, Lusaka, Dakar, London, and Geneva, “Both sides came to realize that they had a strong love of South Africa in common” (Bouckaert 2000, 243). In a watershed event in February 1990, de Klerk announced Mandela’s release and the un-banning of nationalist parties. Talks began immediately. The ANC called for an end to emergency laws, the release of political prisoners, a moratorium on political trials, and the departure of security forces from the townships. Once granted, these concessions brought other issues onto the table, such as the return of exiles and the suspension of the armed struggle. The two principals then reaffirmed their mutual commitment to negotiations and invited other parties to join the process.1 The next three years see-sawed between negotiation and violence. The major political parties participated in a Convention for a Democratic South Africa (CODESA) that produced the outlines of an elite consensus on nonracial democracy and the reintegration into the state of the nominally independent ethnic homelands. Other issues—such as property ownership, immunity for political crimes, and the distribution of powers to lower tiers of government—proved much more contentious. Indeed, the transition period (1990–1994) was marred by frequent incidents of violence, especially in KwaZulu-Natal and the East Rand, where supporters of ANC and Inkatha (a Zulu cultural and political movement) battled for grassroots supremacy. Despite the signing of a national peace accord in 1991, CODESA ended in acrimony when ANC walked out in protest against attacks on its supporters by a shadowy, state-sponsored “third force.”2 The outlines of a compromise settlement were proposed by Joe Slovo, lead strategist for the South African Communist Party (SACP). Recognizing that neither side would surrender and seeking to preclude a counterrevolutionary threat from the state bureaucracy and security forces, he proposed power sharing as a means to break the logjam (Sisk 1995). The plan envisaged a two-stage solution beginning with a grand coalition formed on the basis of an all-races election. The critical feature was a “sunset clause” that would limit the coalition’s life to five years, during which time a constituent assembly and a constitutional court would approve a new national

102

Power-Sharing Settlements

charter based on universal principles. Thereafter, the powers of the South African state would pass—via a process of power division—to the winners of elections. This agreement was set down in a significant “Record of Understanding” in September 1992. On this basis, a broadly inclusive Multiparty Negotiating Forum was convened at Kempton Park in April 1993 and attended by a full range of minority parties. By this stage, potential spoilers could no longer afford to boycott the talks for fear that a growing bilateral convergence between the ANC and NP would leave them behind. Even so, Chief Gatsha Buthelezi formed an alliance with the white right and the homeland leaders in order to insist on political and cultural autonomy for ethnic minorities. A succession of violent incidents—the assassination of Chris Hani, chief of staff of the ANC’s armed wing, an attack on the negotiating chamber by a posse of white extremists, and escalating interparty clashes in KwaZulu-Natal— seemed to signal further breakdowns in talks. But the threat of the alternative—an all-out civil war—only made the main conferees more determined than ever. In November 1993, the negotiating forum inaugurated a Transitional Executive Council to oversee preparations for a democratic election. The Interim Constitution of 1993 implemented the power-sharing settlement.3 Its terms required an Independent Electoral Commission (IEC) to conduct an all-races election according to a list system of proportional representation. The results would determine the make up of a transitional government: the GNU executive would feature a 27-member cabinet with posts proportionally assigned to all parties that won at least 5 percent of the vote. The legislature would double as a Constituent Assembly, whose main function was to draft a new constitution. Nine administrative provinces would be permitted to adopt their own charters as long as these were consistent with agreed-upon principles, including the rights of both property owners and trade unionists. Critically, the comprehensive settlement provided for a reformed South African Police Force and a unified South African National Defense Force (SANDF). The election of April 1994 began badly but ended well (Johnson and Schlemmer 1996; Bratton 2008). Armed clashes between supporters of the ANC and the Inkatha Freedom Party (IFP) led to the declaration of a state of emergency in KwaZulu-Natal province. As a potential spoiler, Buthelezi appeared willing to risk a bloody showdown in pursuit of personal political ambitions and ethnic group interests. Eventually realizing that the transition would occur with or without him, he announced one week before the elections that the IFP would take part. In the event, the 1994 election in South Africa was a celebration, both solemn and jubilant, of a momentous historImplementation

African Experiences with Power Sharing

103

ical shift. The election brought an end to 350 years of settler colonial rule and, with it, the perverse idea that the right to self-government was the preserve of one race but not others. To be sure, South Africa’s founding election did not run entirely smoothly: voters had to wait in long lines to cast ballots, and the result in KwaZulu-Natal owed as much to an elite bargain as a valid vote. Nevertheless, high voter turnout and peaceful polls enabled the IEC to declare the election “substantially free and fair.” The ANC won a solid majority with 63 percent of the vote, followed by the National Party and the IFP with 20 and 11 percent, respectively. As part of the political deal, the IFP vote percentage was raised just enough to allow it to become the majority party in the Zulu political heartland (Sisk and Stefes 2005, 304). Formally, the three main parties formed a government of national unity with Mandela as president, de Klerk as a deputy president, and Buthelezi as minister of home affairs. In a signal of economic continuity, the NP’s Chris Stals stayed on as finance minister. The ANC’s Joe Modise took charge of the Ministry of Defense, where his first task was to integrate guerrilla and homeland armies into the new SANDF. Analysts disagree on the military’s role during this period, with some seeing it as an independent force committed to derailing the transition and others regarding it as being subject to civilian control and concerned mainly with securing its own future.4 The army repeatedly stepped in to restore order during the transition and, by saluting Mandela at the presidential inauguration, the service chiefs publicly indicated that a reactionary military coup was not in the offing. The parties in the GNU coalition, however, soon discovered the limits of power sharing. Frustrated by the contradiction of simultaneously being a part of the government and the country’s principal opposition party, the NP voluntarily withdrew in May 1996, two years ahead of schedule. Power sharing—in the form of high political office for party leaders—nonetheless remained a positive inducement for the IFP to prefer normal politics to political violence. Following elections in 1999, all parties effectively accepted that the ANC would control the main instruments of state. The GNU’s main achievement was constitutional reform. Several aspects of the constitution-making process framed future political prospects (Sisk and Stefes 2005). The process was participatory, with many ordinary South Africans partaking in public education programs and submitting suggestions. The ANC and NP led the reform process by continuing to seek consensus whenever sticking points arose, for example, over the death penalty or the appointment of judges. Finally, the Constitutional Court required revisions to strengthen provisions for the rights of individuals, the autonomy of provincial governments, and public oversight by watchdog agencies. With these steps complete, President Mandela signed a new constitution into law on December 10, 1996.

104

Power-Sharing Settlements

In contrast to the interim bargain, however, the final settlement featured a division of power rather than power sharing.5 The South African Constitution of 1996 posed no requirement for coalition building, grand or otherwise. Instead, the party with the largest share of seats in the National Assembly would form a government. After 1994, the ANC consistently won about two-thirds of the popular vote and was free to govern on its own. By subordinating the legislature to the party leadership, it adopted some of the features of de facto one-party rule (Giliomee and Sinkins 1999). Moreover, the constitution provided political minorities with no guaranteed power of veto. Instead, individuals and groups had to rely for protection on institutional checks and balances built into a system of provincial administration and judicial independence. The Constitutional Court proved a stout defender of the rule of law: careful to frame its judgments narrowly, the court tilted against the executive branch and in favor of social justice. On paper, South Africa’s constitution promised other antidotes to executive power in the form of independent agencies of accountability. A Human Rights Commission was charged to redress violations and educate the citizenry; a public protector was mandated to receive citizen complaints about lax or unfair administration; and an auditor-general provided oversight on the government’s use of taxpayer money. In practice, however, the operations of all public watchdog agencies were limited by budgetary constraints, and none functioned comprehensively at local levels, where most abuses occurred. As a result, the country experienced growing elite corruption, persistent economic inequality, and an epidemic of criminal and social violence. Furthermore, while South Africa’s vaunted Truth and Reconciliation Commission cast light on atrocities committed in the name of apartheid, it left many victims wondering whether truth without justice would necessarily lead to reconciliation (Gibson 2004). In the initial phase of the transition from apartheid in South Africa, the majority’s willingness to share power helped to alleviate minority fears. For supporters of the NP and the IFP, power sharing provided a measure of reassurance by delaying the day when, under universal suffrage, they would be outnumbered in elections. For their part, supporters of the ANC were frustrated by the need for power sharing and skeptical of its consequences. But they were willing to make a temporary concession if it helped provide stability in the short run without permanently inhibiting progress toward majority-rule democracy. Despite ongoing violence, numerous factors contributed to an unexpectedly smooth transition. Because neither the NP government nor the Interpretation

African Experiences with Power Sharing

105

ANC-led resistance could impose its will by military means nor govern the country alone, a temporary interlude of power sharing seemed like the only viable option. By the end of the 1980s, the isolation of South Africa from the West had weakened the economy and, as communism collapsed, the ANC was losing sources of support from Eastern-bloc countries. While encouraged by international actors to engage in dialogue, the South Africans nonetheless devised an indigenous set of procedures and institutions. They set their own pace that required only minimal and intermittent interventions from outside (Lyman 2002). The transition also benefited greatly from the inspired leadership—meaning common vision and political skills—of Mandela and de Klerk. The final settlement emerged incrementally. The power-sharing agreement of 1993 represented only one phase in a sequence of elite pacts that unfolded between Mandela’s release in 1990 and the adoption of the 1996 Constitution. Step-by-step negotiations provided the parties with numerous opportunities to build confidence and develop trust; in particular, a gradual approach allowed time for minorities to test the credibility of the majority’s commitment to respect individual and group rights. The settlement process also featured a mix of formal and informal consultations. Notably, the ANC and NP would sometimes retreat into relaxed bosberaad (bush meetings) to discuss the problematic sticking points before arriving at a consensus position to submit to broader negotiating forums. As the same time, the practices of compromise and moderation used in the political negotiations diffused into other parts of society (Marks 2000). In efforts to quell interparty violence, the ANC and IFP established liaison committees at various levels, and NGOs formed community peace committees that sometimes even included the police. Churches, trade unions, and women’s groups were also active in building bridges across partisan divides and demanded to be consulted about the transition. In short, informal institutions created multiple arenas of bargaining that helped to overcome a legacy of violence and laid a foundation for democratization.

Ending Civil Wars: The Case of Sierra Leone

In contrast to South Africa, the settlement in Sierra Leone required intervention from international mediators backed by military force, not only to cultivate an agreement but also to ensure implementation. External enforcement was necessary because state institutions were exceptionally fragile; the civilian government lacked a loyal and reliable army and was unable to protect civilians. But potential spoilers from a rebel insurgency, who never fully renounced violence, were eventually squeezed out of a share of state power when they were defeated in democratic elections.

106

Power-Sharing Settlements

Settlement

In July 1999, President Ahmad Tejan Kabbah of the governing Sierra Leone People’s Party (SLPP) signed a peace agreement with Corporal Foday Sankoh, leader of the rebel Revolutionary United Front (RUF) to bring to a close a decade-long civil war in Sierra Leone (Lord 2000; Hirsch 2001; Keen 2005). The Lomé Peace Accord, as the settlement was known, provided for a ceasefire and a coalition government that would share executive power. Former combatants benefited from an amnesty granted to those who had committed atrocities during an unusually pitiless conflict.6 The civil war began in 1991 when a small force of RUF guerrillas, armed in Libya and backed by Charles Taylor’s National Patriotic Front of Liberia (NPFL), penetrated the diamond-rich regions of southeastern Sierra Leone. Their aim was to overthrow the government of President Joseph Momoh, whose All People’s Congress (APC) had ruled Sierra Leone for 23 years. The RUF was a populist, millenarian movement that appealed to dispossessed elements in society but soon turned to mass abduction of recruits—including children—and the looting of natural resources (Richards 1996). The national army, weakened by internal dissent and military coups, ultimately fragmented; renegade soldiers abandoned their posts to enter the diamond trade and threw their lots with the rebels. In 1992, Momoh was ejected from office by a group of returnees from the war front, who proceeded to form a National Provisional Revolutionary Council (NPRC). Under domestic and international pressure, they accepted a 1996 election that brought to power President Kabbah and the SLPP. Since the army was unable to contain the insurgency, the government relied on Executive Outcomes, a firm of South African mercenaries and an ethnic militia known as the Civil Defense Forces (CDF). But by 1996, the RUF effectively controlled large parts of the countryside. A tentative ceasefire and peace agreement was signed later that year in Abidjan. But Sankoh reneged on the agreement, plunging the country back into war (Bangura 1997). Another military takeover followed, led by a group calling itself the Armed Forces Revolutionary Council (AFRC) under Johnny Paul Koroma, an army major awaiting trial for a previous attempted intervention (Zack-Williams 1999). The new junta shocked Sierra Leoneans and the international community by inviting the RUF to join the government. Sankoh was awarded the deputy presidency, and other RUF members received cabinet posts. In reaction, the Economic Community of West African States (ECOWAS) deployed a monitoring group of mostly Nigerian troops (the Economic Community of West African States Monitoring Group, ECOMOG), which marched in to restore the Kabbah administration.

African Experiences with Power Sharing

107

But the conflict continued, reaching cruel heights in 1999, when the RUF-AFRC alliance mounted Operation No Living Thing against the capital, Freetown. The six-week campaign left much of the city devastated and the population traumatized. The rebels’ tactics featured extreme violence against civilian populations, including looting and burning of property, gang rape, sexual slavery, forced enlistment and labor, and collective starvation. It was during this period that the RUF’s signature mode of terrorism—the mutilation of innocents by amputating arms, legs, and other body parts—reached a peak of international condemnation. While the guerrillas achieved a military advantage in the countryside, ECOMOG was able to hold Freetown. Since neither side was able to bring closure to the war on its own, negotiations became necessary. As conditions for ending hostilities, the RUF demanded a share of executive power, the ejection of foreign troops, and guarantees of amnesty for war crimes. In the Lomé negotiations, Kabbah had an incentive to settle quickly once it became clear that Nigerian forces were likely to soon withdraw. For his part, Sankoh searched for a way to overcome the RUF’s pariah status. In six months of preparatory discussions, both the principals and their respective delegations met regularly. And while the government was initially dead set against power sharing, Kabbah finally bowed to international pressure to accept many of the RUF’s demands, incorporating them into the July 1999 agreement. The Lomé Peace Accord provided for a ceasefire that would be overseen by a neutral peacekeeping force jointly constituted by a United Nations contingent (UNAMSIL) and ECOMOG. To enable the demobilization and disarmament of fighting forces and their reintegration into civil society, the agreement envisaged the creation of a new national Sierra Leone Army (SLA). It also promised the Commission for the Consolidation of Peace, whose role was to “ensure reconciliation and the welfare of all parties to the conflict, especially victims of war.”7 Importantly, but controversially, the agreement granted “an absolute and free pardon” to Sankoh and all rebel combatants in relation to their actions during the war and sanitized the RUF by endorsing its transformation into a political party.8 In terms of governmental power sharing, the RUF received four ministerial and four deputy ministerial positions in an expanded coalition cabinet—less than the parity it desired. Formally, Sankoh was appointed vice president—with the explicit provision that he answer only to the president—and chairman of a commission on strategic resources. While the commission’s stated purpose was to reestablish government control over the country’s gold and diamond mines, many wondered how Sankoh would informally exercise its new powers. Observers judged that, along with amnesty, de facto access to flows of mineral revenues was the decisive consideration inducing the RUF to enter the agreement (Francis 2000).

108

Power-Sharing Settlements

Implementation

The agreement did not last long. Disputes soon arose over the familiar problem of allocating state offices. The RUF complained that it had not received senior cabinet positions and that, along with the AFRC, they had been sidelined from high-level decisionmaking. At lower levels, the disarmament of combatants and the formation of an integrated national army fell short of targets and well behind schedule. Elements within the RUF, especially Sankoh’s second-in-command, Sam Bockarie, therefore tried to keep their options open by refusing to disarm or enter demobilization camps. For their part, members of the AFRC protested the exclusion of their loyalists from incorporation into the SLA. Clashes with peacekeepers escalated out of control in May 2000 when the RUF took hostage several hundred newly arrived UNAMSIL soldiers. A grassroots counterdemonstration demanding their release turned fatal when RUF soldiers opened fire, killing civilians. Recognizing the blunder, Sankoh tried to escape, but he was captured and jailed. With these events, Sierra Leone’s experiment with power sharing effectively dissolved less than one year after its inception. Despite this disappointing short-term outcome, the brief interlude of power sharing set the stage for more positive political developments down the road. The first was the attainment of relative peace. Once Sankoh exited the scene, many young RUF fighters lost heart and began to look for alternative means of livelihood. The rebel organizations disintegrated into factional infighting and suffered military setbacks, both against the Guinean army on the northern borders and a growing UNAMSIL presence in Freetown. The resource base of the insurgency also began to dry up in the face of international sanctions against Liberia and a global campaign against the trade in “conflict diamonds.” Once a full contingent of over 20,000 UN troops arrived—then comprising the largest peacekeeping force in the world—the rebels were forced to retreat. Sierra Leone continued to remain vulnerable to ongoing destabilization by the Charles Taylor regime in Liberia. But by January 2002 the internal disarmament and demobilization exercise was complete with 70,000 weapons collected and 50,000 former combatants returned to civilian life. President Kabbah was able to declare the war over. A second positive development was the rapid convocation of elections. A tight timetable left precious little time for the RUF to transform itself from rebel movement to a competitive electoral party. In the May 2002 polls, its political wing received less than 2 percent of the popular vote, garnered no parliamentary seats, and soon disappeared as a political force. President Kabbah was re-elected in a landslide, securing over 70 percent of the vote and a sweeping majority of SLPP seats in the legislature. The elec-

African Experiences with Power Sharing

109

toral process was far from perfect, however; the campaign and count were marred by fraud charges; the returns revealed an ethnic divide between Kabbah’s Mende south and the opposition’s Temne north; and the security forces held back full support for the new order (ICG 2002; Kandeh 2003). Nonetheless, the 2002 election marked a step toward democracy ahead of the 2007 national elections, which ushered in a new president and a turnover of ruling parties. Third, the international community assisted in setting up a transitional justice mechanism to investigate political crimes committed during the civil war. A Special Court for Sierra Leone issued indictments in March 2003 against top RUF, AFRC, and CDF commanders—and later against Charles Taylor of Liberia. In this regard, the Sierra Leone case demonstrated that guarantees of amnesty were not sacrosanct; they could be annulled if rebels failed to live up to commitments previously made in peace agreements (ICG 2003). External ownership of the transitional justice process, however, undermined domestic efforts to set up a Truth and Reconciliation Commission, whose inception was delayed repeatedly and whose budget was always inadequate (Keen 2005, 303). Moreover, Sierra Leoneans had little appetite for truth telling about the horrors of the war, seemingly preferring to forgive and forget (Shaw 2005). Thus, the TRC’s final report in 2004 had only a modest impact on the prevailing climate of legal impunity. What explains this sequence of events? Because of the extreme fragility of the state in a country overrun by civil war, international actors loomed large. The Lomé Peace Accord was reached several years after the government lost control of its own security sector; instead, President Kabbah relied on international forces and local auxiliaries to hang on to power. From this vulnerable position, he came under irresistible pressure from neighbors—especially Nigeria, but also Ghana and Guinea—to open a dialogue with the rebels. Meanwhile Britain began to explicitly tie development aid to negotiations. US Special Envoy Jesse Jackson played a key role in instigating the ceasefire that enabled the start of substantive talks, and US officials reportedly drafted language for parts of the accord.9 Indeed, international mediators and donors were responsible for bringing the RUF and AFRC in from the cold by extending recognition to them as legitimate negotiating partners. This rapprochement occurred against the expressed preferences of most government representatives, including the president, and in the face of evidence that the insurgents lacked support among the population (Kovacs 2008). Because external third parties were unable to quell all violence, they share partial responsibility for the rapid breakdown of the peace agreement. Interpretation

110

Power-Sharing Settlements

The UNAMSIL peacekeeping effort was weak and halting. The UN Security Council did not authorize a multinational force to monitor adherence to the ceasefire until four months after Lomé, and all of the promised troops did not arrive until several months later. The lightly armed peacekeepers were a motley crew, spottily deployed within the country, and quick to surrender when faced with stiff resistance. The absorption of an ECOMOG contingent into UNAMSIL did little to boost the reputation of the peacekeepers, since some Nigerian soldiers had been implicated in trading diamonds and arms. With the credibility of the UN force in question, and given a renewed RUF drive against Freetown, the UK decided to send in troops under its own command in May 2000. Motivated by an escalating set of concerns—to evacuate British citizens, to strengthen UNAMSIL, and to engage in security sector reform—the British intervention shifted the balance of power decisively in favor of the incumbent government. As for domestic factors, President Kabbah made the most of his limited control over the remnants of a weak state. Perhaps because he commanded no formal army, he and his negotiators insisted on adherence to the constitution, a factor that not only tilted the power balance in the favor of the existing government, but also helped to win national and international legitimacy. Within the confines of a reluctantly entered power-sharing settlement, Kabbah was still able to restrict his opponents to a minority role and outmaneuver their extra-legal strategies for claiming power. In this endeavor, the government was aided by internal splits within the RUF and AFRC. As a legacy of a decade-long civil war, Sierra Leone inherited a fragile and fragmented state. The long-term prospects of the postwar political settlement therefore hinged on the rehabilitation of the state’s monopoly of violence under a rule of law. But security sector reform is a demanding agenda that often takes years to complete. For example, the pro-government civil defense militias were excluded from the demobilization process; they retained their own arms, much to the chagrin of the SLA security chiefs. Weapons found their way across international borders and into private caches, for example, as part of the illicit diamond trade. Finally, demobilized combatants, as well as families affected by the war, found it difficult to secure sustainable livelihoods. At best, only about half the young male fighters found jobs or self-employment (Fithen and Richards 2005, 132). As a result, groups with grievances continued to have access to weapons, “opening the door to them becoming politicized and potential agents of future violence” (Ginifer 2006, 808). Above all, the Lomé Accord was controversial because it rewarded rebels with a share of power in a coalition government. By being willing to do business with craven thugs, the international community and the government elites signaled that foul atrocities could yield political dividends. In

African Experiences with Power Sharing

111

one analyst’s condemnation, “The granting of amnesty and government posts to the rebels made a mockery of the basic principles of human rights and democracy” (Bangura 2000). In the end, however, potential spoilers were unable to upset the political settlement and, by overplaying their hand, lost any temporary advantage. Thus, although the Lomé Accord apparently “overcompensated the RUF on paper . . . the government was able to undercompensate [them] during the implementation of the agreement” (Binningsbo and Dupuy 2009 17; see also Jarstad 2008). Thus, the SLPP used power sharing to end a civil war, marginalize rivals, and create the conditions for a return to elected government.

Surviving Disputed Elections: The Case of Kenya

Power sharing in Kenya occurred under much different circumstances than in either South Africa or Sierra Leone. The trigger for elites to negotiate a political settlement came not from authoritarian breakdown or civil war but from an eruption of violence following a democratic election. To address this crisis, elite negotiation in Kenya was neither homegrown (as in South Africa) nor imposed from abroad (as in Sierra Leone). Instead, the resolution of postelection violence required a delicate mix of internal and external agency. The internal parties were unable to resolve their differences on their own, yet external third parties lacked the authority and leverage to impose a solution. In Kenya, the state had not collapsed, as in Sierra Leone, but the principles of constitutional rule, civilian control of the military, professionalism in the security sector, and judicial independence were less well-established than in South Africa. In February 2008, President Mwai Kibaki of the Party of National Unity (PNU) and opposition rival Raila Odinga of the Orange Democratic Movement (ODM) arrived at a political settlement aimed at resolving a crisis triggered by the 2007 presidential elections. The principals noted that widespread postelection violence had “brought to the surface deep-seated and long-standing divisions . . . [which] if left unaddressed . . . threaten the very existence of Kenya as a unified country.”10 Intense domestic bargaining and skilled international engagement were required to craft a power-sharing compact. Even so, the resulting government was a shotgun marriage in which reluctant partners had little choice but to cohabit in a nominal coalition. Nevertheless, they cooperated to foster constitutional and electoral reforms and colluded to resist efforts to prosecute the agents of violence. The roots of Kenya’s political crisis ran deep. Under founding father Settlement

112

Power-Sharing Settlements

Jomo Kenyatta, but especially under his successor, Daniel Arap Moi, the powers of the presidency had expanded, while the autonomy of legislative, judicial, and civil service institutions eroded. Both presidents aggressively promoted co-ethnic supporters—Kikuyu and Kalenjin respectively—into leadership positions in the public and armed services (Bienen 1974; Widner 1992; Throup and Hornsby 1998). Facing few constitutional constraints, political elites also indulged in spectacular acts of corruption involving land, banks, and the import-export trade (Wrong 2009). Those Kenyans without access to government patronage, especially the Luo, interpreted their group’s exclusion in terms of communal injustice. Pushed by international donors and civil society to accept multiparty competition in 1991, President Moi—a consummate power politician — turned to scare tactics to stay in office. In the 1992 elections, the ruling party deployed Kalenjin “warriors” to kill political opponents—the Kikuyi, Luo, and Luhya—or chase them from their homes in the Rift Valley (Barkan 1993; Barkan and Ng’ethe 1998). The same pattern of ethnic violence was repeated for the 1997 elections, this time spreading to culturally mixed areas along the Indian Ocean coast. Moi survived both contests with a narrow plurality of votes against a hopelessly divided opposition. By mobilizing auxiliary militia forces to secure re-election, however, he chipped away at the authority of the security services and undermined the state’s monopoly of legitimate force (Mueller 2011). The 2002 election seemed to promise a respite from disunity and conflict. Opposition parties combined for the first time in a National Rainbow Coalition (NARC), an electoral alliance that propelled Mwai Kibaki into the presidency with broad, cross-ethnic support (Anderson 2003; Ndegwa 2003). But once in power, Kibaki—a Kikuyu—reneged on promises to govern even-handedly, for example, by failing to install Raila Odinga—the leading Luo politician—as prime minister. As one result, Kibaki’s further effort to entrench presidential prerogatives in a constitutional referendum went down to defeat in 2005. A polarized electorate and an ugly campaign of ethnic stereotyping set the scene for the disputed presidential election of December 2007. Without consulting other parties, Kibaki dismissed the Electoral Commission of Kenya (ECK) and appointed his own loyalists. Public opinion polls before the election showed Odinga ahead in a tight presidential race (Bratton and Kimenyi 2008; Gibson and Long 2009). But ECK announced that Kibaki had won by a solid margin of votes, a surprising result that was probably due to irregularities from several quarters; both political parties—ODM and PNU—put pressure on election officials to manipulate the count in their own strongholds. In addition, officials at the ECK tally center allegedly inflated Kibaki’s overall vote share, especially in the Central province (ICG 2008b). Since the courts were politically compromised, there was no independent institution to arbitrate charges of a rigged election. Instead, with

African Experiences with Power Sharing

113

unseemly haste, Kibaki went on national television to be sworn in as president for a second term; he then retreated to a military base after authorizing the internal security minister to suspend all live media broadcasts. Violence immediately broke out. Angry throngs set up roadblocks, beat up persons who spoke the “wrong” ethnic vernacular, and set fire to houses, shops, and cars. Some of the violence was spontaneous, but a significant portion was planned, organized, and financed by various groups. At first, Kalenjins and Luos were the main perpetrators; but Kikuyu victims quickly regrouped and retaliated. The Kibera slum saw the worst of the violence in Nairobi, but clashes also occurred in outlying towns like Kisumu, Mombasa, and Eldoret, the scene of a shocking incident in which a church burned with dozens trapped inside. Police and paramilitary units either stood by or joined in with indiscriminate shootings. The multiethnic Rift Valley was most seriously affected, where popular frustration with the election outcome was intensified by historical grievances over land and jobs. In all, about 1,500 people were killed and half a million displaced, creating a lasting humanitarian crisis and losses to the economy estimated at several billion US dollars. The violence, the worst since independence, shattered Kenya’s international reputation as one of Africa’s most stable democracies. In response, the African Union urgently convened a mediation team of eminent Africans and led by former UN Secretary-General Kofi Annan. Four former African presidents visited Nairobi, as did representatives of the United Nations, African Union, European Union, and United States. After numerous false starts and a month of intransigence from the principals—especially the Kikuyu hardliners who surrounded the president—Annan cajoled Kibaki and Odinga into a political compromise (KAF 2008). The parties conveniently acknowledged that it was impractical to recount votes or re-run the election. Instead, they agreed to form an Independent Review Commission (IREC) to investigate the bungled presidential vote and to recommend improvements to the electoral process. Furthermore, recognizing that “a political settlement is necessary to promote reconciliation and unity,” they also committed to reform Kenya’s “governance structure” (KNDR 2008a, 4; KNDR 2008b). Accordingly, the two main political parties entered a government of national unity, supposedly as equal partners. The president (an established post retained by Kibaki) would appoint a prime minister (a new post assumed by Odinga, as leader of the largest party in the National Assembly). The president would head the armed forces and the prime minister would supervise government ministries. Moreover, the partners committed themselves to mutual consultations and cooperative governance. They agreed that the coalition would last only until the next regularly scheduled election; but they left unresolved the question of what would happen if the agreement fell apart (Horowitz 2008).

114

Power-Sharing Settlements

In order to address fundamental societal problems, the parties promised an enormously comprehensive reform program: they undertook to investigate the violence, to provide humanitarian relief, to write a new constitution, to address land and other economic grievances including youth unemployment, and to undertake institutional reforms to electoral administration, the police, and the judiciary. The very ambition of the promises seemed to guarantee that some would never be fulfilled. Kenya’s grand coalition accumulated a mixed record. On the one hand, the power-sharing partners began to recognize historical grievances, for example, by introducing in 2009 a sweeping land policy that, if ever executed, would clarify Kenya’s muddled land ownership records and assign secure rights of land tenure. And, in a defining event in 2010, parliament passed a new, publicly approved national constitution that provided a framework for dividing and distributing power. On the other hand, the coalition government bogged down in controversies (KNDR 2009). It prematurely shut down camps for persons displaced by the violence, forcing them to return home without guarantees of livelihood or security. Parliament passed a new media law that threatened to muzzle critical news reports. And top politicians continued to be implicated in dubious deals, including a scheme to sell food reserves for profit at a time of national food shortages. Ordinary Kenyans began to wonder whether the parties to the coalition would serve as a check on each other’s behavior or conspire to cover up each other’s misdeeds (Cheeseman and Tendi 2010). Moreover, since leaders took only token initiatives to reconcile communal grievances in the Rift Valley and the urban slums, the momentary peace that followed the political settlement seemed tenuous. Critically, partisan bickering over the division of official state positions almost derailed the GNU before it got rolling, leading to a perception of two governments in one. Kofi Annan had to rush back to Kenya to oversee the structure of the GNU: the PNU retained the ministries responsible for internal security and finance, whereas ODM took the portfolios of public service, local government, and agriculture. One drawback was the costly expansion of the cabinet to 42 ministries, each with ministers, deputies, and staff. The PNU violated the spirit of power sharing by using the powers of incumbency to unilaterally block ODM appointees from taking up positions in the civil and security services. Kibaki insisted that, as president, he enjoyed the prerogative to make appointments without consulting the prime minister. For his part, Odinga tried unsuccessfully to suspend cabinet members over mounting corruption charges. Implementation

African Experiences with Power Sharing

115

Thus, as the crisis of early 2008 receded from view, political elites faced diminishing incentives to work together and often lapsed into mutual recriminations. Before long, they began to position themselves for the next election. As part of the overall settlement, the IREC, under the chairmanship of South African Justice Johann Kriegler, recommended new electoral rules.11 The commission, which included political party appointees, stated that it could not determine who had won the disputed 2007 presidential contest because the integrity of the count was “irretrievably polluted.” Hence, it called for the reorganization of the electoral management body and a package of institutional reforms, including a new electronic voter registration system. In early tests, a new body, later named the Independent Electoral and Boundaries Commission (IEBC), proved effective at administering byelections and a nationwide constitutional referendum. In a parallel reform, a Commission of Inquiry into the Post-Election Violence (CIPEV), chaired by Appeals Court Justice Philip Waki and including foreign commissioners, probed the root causes of the conflict (Republic of Kenya 2008). The CIPEV report noted that concentrated presidential power and winner-take-all elections created incentives for ethnic solidarities and extreme political tactics. The commissioners found that state security agencies—notably, the national Kenya Police, the paramilitary General Services Unit, and the provincial Administration Police— failed to anticipate, prepare, or coordinate to prevent violence. Individual members of these agencies were themselves responsible for acts against civilians that amounted to violations of human rights. Importantly, political patrons systematically planned and centrally directed some postelection attacks and then retreated behind a veil of impunity. The Waki Commission therefore recommended a merger of police agencies and a special tribunal to prosecute the masterminds of the violence. As a safeguard, the commissioners prepared a sealed envelope containing a list of alleged senior perpetrators. When the cabinet and the parliament each repeatedly failed to launch prosecutions within Kenya, the envelope was passed to the International Criminal Court (ICC) in The Hague. In December 2010, the ICC’s chief prosecutor released the names of six persons: three sitting cabinet ministers (including Uhuru Kenyatta and William Ruto, later the successful presidential ticket in 2013 elections), Kenya’s top civil servant, a former police chief, and a radio broadcaster. Public opinion polls at the time indicated that popular majorities saw the ICC as the preferred venue for trials and that vigorous legal prosecution was warranted. Since the court leveled indictments at individuals from both sides of the political spectrum, PNU and ODM found common cause in colluding to resist the ICC. In retaliation for the ruling, the government sought to withdraw Kenya from its obligations under interna-

116

Power-Sharing Settlements

tional law and campaigned for political support from the African Union and UN Security Council. In a signal achievement, however, the coalition government consummated 20 years of debate about constitutional reform. Rather than engage in a time-consuming process of public hearings, as tried previously, the government commissioned an expert committee of African jurists and legal scholars to prepare a draft, which was approved by parliament in April 2010. The two principals in the coalition government found unity of purpose in together pressing for the adoption of the draft constitution, including by barnstorming the country together. The document was ratified by 68 percent of the public in a referendum in August 2010 and promulgated into law later that month. Perhaps the proceedings were peaceful because the referendum was never a close contest and because control of the presidency and government were not immediately at stake. On paper, the constitution was transformative; it shifted the basis of Kenya’s elite political settlement from power sharing to power division. The charter divided power along both horizontal and vertical lines. First, the new law abandoned the ambiguous role of prime minister by vesting coherent authority in a president. An indissoluble National Assembly, elected for a fixed term and with extensive powers to confirm official appointments would oversee government performance and, if necessary, impeach the chief executive. Second, the constitution created 47 elected county governments with devolved powers and revenues distributed from the political center. This quasi-federal arrangement aimed to ensure equity in the allocation of national resources and to break up political blocs organized by ethnic groups. In praising the constitution—the first indigenous charter that Kenyans had ever given themselves—analysts ventured that “it raise[d] the prospects for peace and stability in East Africa’s anchor state . . . [and] return[ed] Kenya to a path of democratization and economic growth” (Barkan and Mutua 2010; see also Kramon and Posner 2011). This prognosis proved overly sanguine. To be sure, elections in March 2013 were relatively peaceful, if not entirely free of pre-poll violence. But only two-thirds of eligible adults were registered to vote and the electronic system for tabulating votes broke down, which required the IEBC to revert to a manual count (Barkan 2013). The official result—which gave Deputy Prime Minister Uhuru Kenyatta a narrow first-round victory in the presidential poll (with 50.7 percent)—seemed to fly in the face of pre-election surveys and a postelection exit poll. Kenyans were again left wondering about the competence and integrity of electoral institutions and whether the constitution could really guarantee an electoral climate unbound by fear. International observers could not bring themselves to pronounce the proceedings free and fair, but only “credible.” In an eerie premonition of conditions that would unfold in Zimbabwe’s election later the same year, observers

African Experiences with Power Sharing

117

concluded that “while peace is obviously preferable to violence, it does not necessarily indicate a fully democratic process” (Long et al. 2013). Interpretation

The vitality of the power-sharing settlement in Kenya depended on whether political elites would apply the terms of the constitution as intended by the drafters. In the past, the rule of law was often circumvented, making the prospects uncertain at best. The parliament was faced with the daunting task of passing many pieces of enabling legislation to put the charter into effect. Some sitting legislators were not sincere supporters of the new order, and candidates for office sought to politicize the steps for implementing constitutive rules as elections approached. As such, Kenya’s political settlement remained precariously balanced between a constitutional vision of democratic checks and balances and a reality of an alternately fractious and collusive political oligarchy. Moreover, the Kenyan case illustrates the role of state power in securing and shaping political settlements. By retaining the executive presidency, Kibaki was able to gain the upper hand in the coalition and, more often than not, to reduce Odinga to a junior partner. As president for five years before negotiations began, Kibaki had accumulated control over the state machinery, thanks to senior appointments to the judiciary, police, intelligence services, administration, and army. Thus, even though the reliability of the army’s rank and file remained untested, the loyalty of the top military brass to the civilian government was never in serious doubt. To be sure, civilian leaders were probably induced to find a negotiated settlement by the knowledge that the service chiefs would not indefinitely tolerate chaos. But instead of deploying the army, the government relied upon the Kenyan police services to maintain civil order. Once tested, these units were found wanting; they lacked professional standards, suffered internal divisions along ethnic and political lines, and tended to use disproportionate force. All told, the dismal performance of the police drew attention to the need for security sector reform as an institutional guarantee of the new constitutional order. At the same time, the self-serving behavior of politicians further weakened the state by turning public institutions to private ends. Not content with already being highly paid, Kenyan members of parliament voted themselves increased salaries and allowances. The “informalization” of the state continued apace as unofficial groups took it upon themselves to provide security services (Branch and Cheeseman 2008; Mueller 2011). Over time and at the margins, state agencies conceded the official monopoly of force to an array of auxiliaries, including vigilante groups, crime gangs, and ethnic militia (ICTJ 2010). Increasingly, Kenyan politicians turned to these in-

118

Power-Sharing Settlements

formal private armies to protect themselves from attacks by opponents, mobilize electoral support, or protest losses in elections. In sum, Kenya’s power-sharing settlement of 2008 had a wholesome short-term effect: it put a momentary halt to overt political violence. In addition, it broke a persistent deadlock over the content of a new constitution, thus creating a framework for a lasting political settlement. In the medium term, however, power sharing allowed a president who may well have lost an election (Kibaki) to retain power and provide a breathing space to organize support for the election of a co-ethnic successor (Kenyatta, Jr.). The 2013 election in Kenya also signaled that African voters, rightly or wrongly, are prone to close ranks around leaders perceived as targets of Western criticism and sanctions (Mueller 2014). For the long term, the GNU power-sharing interlude failed to address the deep structural domination long enjoyed by an extractive Kikuyu elite. Moreover, the implementation of new institutional rules, and their acceptance by all political players remained an open question. By colluding to avoid the prosecution of perpetrators of violence or to seriously address underlying structural inequalities in society, Kenya’s political elites failed to guarantee that crises would never recur.

Similarities and Differences

This chapter concludes with reflections on the similarities and differences in the power-sharing experiences of South Africa, Sierra Leone, and Kenya. The purpose is to draw attention to dissimilar goals—ending settler rule, emerging from civil war, surmounting a disputed election—that cast light on the power-sharing experiment in Zimbabwe described in the next chapter. In retrospect, power sharing was an appropriate mechanism for conflict management in South Africa and Sierra Leone. This form of interim political settlement addressed escalating violence, whether from street clashes in Soweto or a rural insurgency encircling Freetown. Because political adversaries saw advantages in preserving rather than destroying state institutions, they had incentives to work together and to resolve differences by talking instead of fighting. By contrast, in Kenya and Zimbabwe the trigger of political violence was an election in which none of the parties was willing to concede that they had lost a popular vote. Under these conditions, the winners of the election were always bound to regard coalition government as an unwelcome second prize; it was hard to reconcile themselves to cohabitation with opponents whom they believed they had beaten at the polls. For former incumbents, who would have preferred to govern alone, power sharing at least offered a way to avoid complete surrender. They saw entry into a coalition govern-

African Experiences with Power Sharing

119

ment as an opportunity to regroup politically and as a platform for mounting a bid to recapture the state, a strategy that bore fruit in Kenya. The question arises as to whether power sharing was an appropriate response to the crises brought about by the practice of power politics in Kenya and Zimbabwe. Much hinges on the credibility of the elections that preceded power sharing in these two countries. If the results of the presidential contest of December 2007 in Kenya were too close to call—as Kofi Annan and Judge Kriegler argued, and as the political principals eventually seemed to acknowledge, even if only for purposes of convenience—then it was impossible to definitively designate election winners and losers. Thus, the balance of incentives for adhering to a cooperative agreement was positive in Kenya, especially since both sides were implicated in postelection violence. In Zimbabwe, however, the first round of the 2008 presidential election was almost certainly rigged, and the second round was blatantly stolen through a systematic campaign of state-sponsored repression. By insisting on a power-sharing compromise under these circumstances, the mediators openly violated the democratic principle that the losers of elections should step down from office. And, by rewarding the perpetrators of onesided, state-sponsored violence with a share of power, they created incentives for actors to continue to resort to the same tactics in the future. On the basis of evidence from the cases considered here, power sharing can lead to positive outcomes. In South Africa, a short interval of coalition government paved the way to a political settlement that has since proven peaceful, legitimate, and, so far, durable. In Sierra Leone, a barely longer interlude of power sharing allowed the country to exit from an especially destructive civil war and to introduce—if not yet to consolidate—a set of democratic political rules. In Kenya, the experiment in power sharing produced within the first two years a set of constitutive political reforms but also ended in an election that confirmed the authority of a dominant power elite. The outcome in Zimbabwe seemed destined to resemble that of Kenya more closely than that of South Africa or Sierra Leone. As the next section will document, rival political elites in Zimbabwe remained poles apart in debates over how to revise the constitution, regularize election procedures, and reform the security and justice sectors. And, as confirmed by the outcome of the 2013 election, power sharing in Zimbabwe did not yield a legitimate political settlement. The cases in this chapter provide fresh insight into the relative roles of domestic political actors and external third parties in power-sharing agreements. The success of South Africa’s transition to democracy rested on the ability of opposing camps to devise a homegrown process of negotiation and to recognize a shared (or at least compatible) set of national interests. By contrast, the next chapter will show that power sharing was problematic in Zimbabwe because the leaders of the old regime never made credible commitments to cede any meaningful measure of political

120

Power-Sharing Settlements

control. As a result, rival political elites could not arrive at a shared vision of the country’s future that could serve as the basis for an internally driven settlement. Instead, external agents provided the impetus behind coalition governments in Zimbabwe, and, for that matter, in Kenya and Sierra Leone. In Kenya, a high-profile, African Union–sponsored mediation mission was critical in getting Kibaki and Odinga to shake hands in public, designate negotiating teams, develop an agenda for talks, and, ultimately, to make mutual concessions. The diplomatic initiatives of Kofi Annan were critical in getting to “yes.” And Annan stayed involved in the postagreement phase, for example, referring the cases of violent offenders to the ICC when the government failed to set up a special tribunal to try them within Kenya. In Sierra Leone, the entire transition from civil war to electoral democracy was the product of international pressure. A regional peacekeeping force enabled a ceasefire and international diplomacy ensured that rebels were included in the Lomé Accords. In this case, domestic actors—notably President Kabbah, Foday Sankoh, and Johnny Paul Koroma—never developed high levels of mutual trust. Rather, the progress in moving toward open elections depended on the willingness of international actors to follow through with implementation of the agreement, up to and including a British military intervention to prop up the shaky coalition government. Compared with these cases, the balance between weak external pressures and distrustful relations between domestic elites was the worst of all combinations in Zimbabwe. The international community—within Africa and beyond—passed the task of guaranteeing the implementation of power sharing in Zimbabwe to SADC. Admittedly, South African mediation was critical in producing a power-sharing agreement, but, in practice, the permissive orientation of successive South African presidents toward Zimbabwe was critical; “quiet diplomacy” was at best soft on—and at worst biased toward—ZANU-PF. Thus, external third parties are helpful in promoting elite settlements only to the extent that they remain sincerely involved in guaranteeing the implementation of the agreements they sponsor. Most importantly, the extent to which governments of national unity can foster lasting political settlements is affected by the nature of the state and the relationship of political elites to state institutions. The fate of power-sharing experiments depends on whether political leaders control the functioning institutions of a strong state and how they wield these resources. The key distinction is whether these leaders agree to govern according to the rule of law or whether they resort to extra-constitutional tactics, including political violence. The cases considered here constitute a descending scale of constitutional rule. In South Africa, leaders voluntarily subordinated a strong state to a constitutional transition. In Sierra Leone, a failed state was pulled back from the brink by a combination of external

African Experiences with Power Sharing

121

support and new set of electoral rules. In Kenya, political elites managed to agree on a formal program of constitutional reform but held in reserve informal resources of violence for use in the event that the law did not work in their favor. The next chapter considers these issues in the context of Zimbabwe. Would power sharing work there? Or would the country’s latest effort at an elite political settlement prove dysfunctional? Would the old regime accept the series of constitutional, electoral, security, and justice reforms favored by its new partners? Or, most critically, would ZANU-PF continue to wield the instruments of power politics?

Notes

1. At this stage, the NP sought to permanently entrench power sharing in a rotating presidency. 2. The “third force” combined elements of Inkatha, the South African Police, and the military intelligence service (Sparks 1995, 153–178). 3. Constitution of the Government of South Africa Act, No. 200 of 1993. http://www.info.gov.za/documents/constitution/93cons.htm. 4. Contrast the position of Sparks (1995) with Gutteridge (1994). These debates are summarized in Kriger and Bond (1995). 5. Constitution of the Republic of South Africa, 2006. http://www.info.gov.za /documents/constitution/1996/index.htm. 6. Peace Agreement Between the Government of Sierra Leone and the Revolutionary United Front of Sierra Leone, July 7, 1999. US Institute of Peace, Peace Agreements Digital Collection. http://www.usip.org/files/file/resources/collections /peace_agreements/sierra_leone_07071999.pdf 7. Ibid. Peace Agreement, Article VI, 1. 8. Ibid. Articles IX, 1 and 2. As a guarantor of the accord, however, the United Nations, at the express instructions of the Security Council, declined to overlook acts of genocide, crimes against humanity, and war crimes 9. Ismail Rashid, “The Lomé Peace Negotiations,” in Lord, 2000. See also Keen 2005, 251, quoting Ryan Lizza, “Where Angels Fear to Tread.” New Republic (July 13, 2000). 10. “Acting Together for Kenya: Agreement on the Principles of Partnership of the Coalition Government.” (Nairobi: February 28, 2008): 1. 11. Kreigler had chaired the Independent Electoral Commission for South Africa’s 1994 founding election.

7 Zimbabwe’s Power-Sharing Settlement, 2008–2013

At the peak of the political and economic crisis of 2008, a bankrupt ZANU-PF government was in retreat. A disintegrating economy, a worthless currency, and a disputed election had severely compromised the party’s capacity to govern. The state apparatus had become quite fragile, as evidenced by the hollowness of formerly strong public institutions and the wholesale breakdown of social services.1 The party’s residual competence to deploy a centralized security establishment, to confiscate property, and to control elections, however, suggested that Zimbabwe had not yet become a completely failed state. The entry of incumbent and opposition elites into a power-sharing pact nonetheless marked a setback for the old regime. A formal agreement to apportion political authority between competing parties signaled the loosening of ZANU-PF’s grip, even if the nature of the successor regime was highly uncertain. The ruling elite discovered that they had lost so much political legitimacy—not only within the country but, critically, among allies in the region and continent—that they could no longer justify their right to rule. At the same time, opposition leaders reconfirmed what they had long known: that they could win a popular mandate from the electorate but still be blocked from assuming state power. The pressures to arrive at a power-sharing settlement were both external and internal. Zimbabwe’s international allies, including many in the African Union, could no longer ignore the fact that Mugabe’s “victory” in the 2008 presidential election did not reflect the genuine will of the Zimbabwean people. As a result, President Mbeki reconvened the SADC mediation team in yet another effort to extract a negotiated solution. By this time, crisis conditions inside the country had provided the necessary incentives for leaders to come to terms. Not only did neighboring states wish to avoid instability in a member state, which would add to outward popu-

123

124

Power-Sharing Settlements

lation flows, but also rival elites within Zimbabwe faced the prospect of deepening political schisms and impending economic collapse. For their part, the long-suffering public was desperate to return to a semblance of normal life. The new settlement was no leader’s first choice. Both Mugabe and Tsvangirai entered into it reluctantly and under duress. The ZANU-PF leader likely acceded to power sharing as an interim measure to avoid having to step down from office and as a means of buying time to figure out a fresh survival strategy. For their part, the MDC leaders probably concluded that, rather than remaining perpetual political outsiders, they were better advised to try to change the system from within. From an opposition perspective, half a loaf seemed better than none.

The Global Political Agreement

At a stiff ceremony at the Rainbow Towers in Harare on July 21, 2008, the leaders of ZANU-PF, MDC-Tsvangirai, and MDC-Mutambara signed a Memorandum of Understanding. The MOU, brokered by President Mbeki of South Africa, provided for talks about “establishing the framework of an inclusive government” (Raath 2008). People cheered as the leaders left after the signing; they welcomed the fact that political rivals had shaken hands for the first time in a decade. However, Mugabe asserted that he would not serve in a government that he did not lead, and opposition leaders still refused to recognize his victory in the presidential runoff poll or to serve under him as president. These seemingly irreconcilable positions were hashed out in an on-again, off-again series of negotiations in Pretoria that centered on the distribution of offices in a transitional coalition government. Among other critical sticking points, the parties could not agree on the disposition of the Ministry of Home Affairs, which controls the police. The impasse finally broke when the MDC-T reluctantly bowed under pressure to Mbeki’s compromise proposal that the leadership of Home Affairs be shared by co-ministers, one from each of the two main parties. Even so, the principals to the agreement had a hard time convincing members of their own inner circles that a temporary coalition government was the best deal they could get. On the ZANU-PF side, the security chiefs on the JOC, including the top defense force and police commanders, Constantine Chiwenga and Augustine Chihuri, remained dead set against granting authority to anyone without liberation war credentials. On the MDC side, the party’s secretary general, Tendai Biti, and treasurer, Roy Bennett, held out until the eleventh hour against cohabitation with ZANU-PF, their perennial nemesis.

Zimbabwe’s Power-Sharing Settlement, 2008–2013

125

Yet, on September 15, 2008, ZANU-PF and the two wings of the MDC entered into a self-styled Global Political Agreement (GPA).2 The political protagonists—Mugabe, who remained in office as president; Tsvangirai, who assumed a newly created post of prime minister; and Mutambara as deputy prime minister—ostensibly accepted a deal to divide state power and to govern cooperatively. The advent of power sharing in Zimbabwe was both a landmark breakthrough and a flawed compromise. The terms of the GPA envisaged a balanced and cooperative apportionment of executive authority. The key to the power-sharing deal was Article 20: Acknowledging that we have an obligation to establish a framework of working together in an Inclusive Government  .  .  .  the Parties hereby agree that . . . the executive authority of the Inclusive Government shall vest in, and be shared among the President, the Prime Minister and the Cabinet, as provided for in the Constitution and legislation. (Agreement 2008)

More helpfully, specific provisions stated that the president would chair the cabinet and that the prime minister would lead a Council of Ministers and serve as deputy chair of the cabinet. It was also clear that the president alone would head a new National Security Council and that the prime minister would be the leader of government business in the parliament. But, the locus of ultimate control over the policy process was left unresolved. Intentionally or not, the framers of Zimbabwe’s power-sharing agreement blurred the guidelines about how executive authority would actually be exercised (Matyszak and Reeler 2011). Moreover, the agreement was insufficiently specific about time. Each side regarded the coalition as transitional, not least because they were uncomfortable entering into a long-term cohabitation with political rivals whom they deeply distrusted. But the GPA document was silent on the question of exactly when the power-sharing interlude would end. The completion of a transition had to be inferred from a timetable laid out for the promulgation of a new constitution, which implied that a fresh election could be organized some 18 months to two years after the swearing-in of a GNU (Veritas 2010a). But this timetable was conditional on the constitutional reform sequence staying on schedule. There was no mistaking the intent of the GPA drafters that constitutional reform was an inescapable prerequisite for either ending or renewing the power-sharing arrangement: “This Agreement and the relationship agreed to hereunder will be reviewed at the conclusion of the constitution-making process” (Agreement 2008). Certainly, no new election—or any return to power division—could be legally held before a new constitution was in place. The haziness of the settlement set up rival leaders for future conflicts, especially over senior political appointments. The GPA specified that the

126

Power-Sharing Settlements

president and prime minister would “agree on the allocation of Ministries.” The stated goals were to share executive positions in rough parity between ZANU-PF and the combined MDCs and to make senior appointments only after consultations between the president and prime minister. The GPA made no mention of recourse if either party felt aggrieved. Instead, in a bid to activate power sharing, the agreement simply reiterated the informal aspiration that all parties would “act in a manner that seeks to promote cohesion both inside and outside government” (Agreement 2008). Beyond the distribution of executive power, the terms of the GPA addressed some two dozen issues held to be important by some or all of the principals. Among the more significant goals were the following: • to restore economic growth; • to provide for a new constitution and to reestablish the rule of law; • to guarantee free political activity; • to depoliticize key state institutions including traditional leadership, the youth service, and the security forces; • to establish a National Security Council and National Economic Council; and • to reassert Zimbabwe’s sovereignty while calling for the removal of restrictive measures on international aid.

The substance of the agreement revealed the priority positions of each political camp: ZANU-PF insisted on the recognition of the existing distribution of land and on the removal of targeted sanctions on its leaders; the MDC parties held out for open political activity, including a free mass media and the restoration of a legal order. Each side compromised by symbolically acknowledging the non-negotiable demands of its opponents. The fact that the new political settlement was based on compromise had both advantages and disadvantages. While the elite deal promised welcome relief from political violence and the cultivation of an atmosphere perhaps conducive to political reforms, it was unstable from the outset because the main protagonists had little in common. The GPA was less a reflection of a heartfelt, deep-seated meeting of the minds than a series of shallow, papered-over disagreements. In short, because the parties lacked the requisite level of mutual trust to reliably work out conflicts when these would inevitably arise, the GPA seemed like a recipe for institutional deadlock. There was also reason to question whether all parties were sincere, especially ZANU-PF. Mugabe had a long history of reneging on written agreements: in 2001 he broke a promise to the Commonwealth to desist from legal and violent measures to seize land; in late 2007 he abrogated an understanding in SADC-mediated talks that constitutional reform would precede elections; and, during the final stages of negotiations in August

Zimbabwe’s Power-Sharing Settlement, 2008–2013

127

2008, he violated a stipulation that parliament would not be reconvened without MDC concurrence. Thus, much depended on whether SADC, as guarantor of the GPA, would punish leaders for breaking commitments taken in bad faith. Finally, there was a risk that the MDCs, by compromising on a secondbest solution, would lose sight of the ultimate goal of a democratic transition. Struggles over the actual distribution of executive power—and the perquisites that accompany it—could easily prove distracting. MDC leaders would have to repeatedly remind themselves that the privileges and rewards of state office were not ends themselves. Rather, the main purpose of the settlement was to advance the political process toward a free and fair election that would divide power and in which the winners would form the next government.3

A Balance Sheet

A transitional government of national unity (GNU)—described in the GPA as the “Inclusive Government”—was sworn in on February 11, 2009. As soon as the government took office, rival elites began contesting the delicate equilibrium of power sharing. On one hand, ZANU-PF repeatedly claimed the unilateral powers of the presidency to make key executive decisions without prior consultation. On the other hand, the MDC formations attempted to resist the fate of PF-ZAPU in the Unity Accord of 1987—namely, absorption into a governing coalition as a powerless junior partner.4 This tense political standoff inhibited the construction of effective or lasting institutions. Instead, all GNU partners became preoccupied with seeking short-term partisan advantage. While leadership consensus was sometimes obtained on narrow technical matters—for example, the mechanics of reopening schools or repairing water treatment systems—larger issues of political reform or socioeconomic development were usually disputed and blocked. Most importantly, the balance of power between rival coalitions tilted in favor of entrenched ZANU-PF hardliners, including the security force commanders and the permanent secretaries at the head of civilian ministries, and against the minority of determined reformers within the MDCs, who lacked experience at governing. To all appearances, ZANU-PF’s strategy was to frustrate the MDCs’ leaders to the point that they would withdraw from the unity pledge, thus ensuring the failure of the GNU without ZANU-PF incurring blame. Tsvangirai showed great forbearance in the face of these provocations, but at the risk of appearing reactive, rudderless, and overly accommodating. After almost four years of joint governance, the GNU had met the terms of only a handful of GPA provisions. A donor-funded analysis esti-

128

Power-Sharing Settlements

mated that, by December 2012, just four out of 40 GPA articles had been fully implemented, some 21 articles had been partially implemented, but 15 “outstanding” articles had not been implemented at all (see Table 7.1).5 Even under a generous interpretation of these data, barely half of the power-sharing provisions had been put into effect. A selected balance sheet of political transition in Zimbabwe reveals more setbacks than gains, a selection of which is listed below (Idasa 2009– 2013; see also Raftopoulos 2013b). Economic Gains

Hyperinflation was rapidly tamed. In a key policy shift in January 2009, the outgoing ZANU-PF government bowed to market realities by abandoning the Zimbabwe dollar and adopting a basket of foreign currencies as legal tender. This drastic policy change put an immediate end to rampant profiteering in black market foreign exchange and contained price increases to the low single digits through 2013.

A measure of economic recovery ensued. Mineral exports rose, becoming the largest source of foreign exchange earnings. Consumer goods reappeared on supermarket shelves, though South African and Chinese products displaced many consumables previously produced locally. Schools and hospitals began to reopen, boosted by priority aid programs to supplement the salaries of medical staff and provide textbooks for primary schools. In time, government revenues more than doubled to an estimated 36 percent of GDP in 2012, allowing the further restoration of basic public services. Economic Setbacks

Economic stabilization fell short of full recovery. The use of factory capacity rose from a low of 10 percent in 2008 to more than 40 percent by the end of 2010, but many industrial and manufacturing enterprises did not revive. As a result, unemployment remained stubbornly high, perhaps 80 percent or more. Meanwhile, due to drought and disruption, food production continued to fall short of the country’s needs; one-third of children under the age of five remained chronically malnourished and stunted (GRZ 2010). Commercial farmers suffered a fresh wave of land invasions. Officers from the Zimbabwe National Army and Zimbabwe Republic Police

129 Table 7.1 Implementation of the Global Political Agreement, February 2009– December 2012 Article

1 2 3a 3b 4 5a 5b 5c 6a 6b 6c 7a 7b 7c 8 9 10 11 12 13a 13b 14 15 16 17 18a 18b 19 20a 20b 20c 20d 20e 20f 21 22a 22b 22c 23 24

Total

Full Implementation

Establish power-sharing government X Work together Economic stability and growth Establish National Economic Council Lift sanctions and aid restrictions Accept existing land distribution Land audit Britain pays land compensation Set up constitutional review commission X Consult public on constitution Prepare draft constitution Correct regional, ethnic, gender imbalances Establish organ for national healing Repatriate skills from diaspora Share respect for national symbols Reaffirm national sovereignty Enable free political activity Reestablish rule of law Enable free assembly and association Separate party and state Train security forces in human rights Reaffirm neutrality of traditional leaders De-politicize youth training De-politicize food relief programs Develop legislative reform agenda Renounce and prevent violence Review political court cases Enable free expression and media President and PM share executive authority Cabinet decides policy President and PM jointly allocate ministries Establish National Security Council Council of Ministers oversees implementation Cabinet limited to 31 ministers, 15 deputies No by-elections for 12 months X JOMIC monitors GPA Security forces observe public order laws SADC/AU guarantee GPA Joint annual review of GPA Enactment of GPA by constitutional amendment X 4

Partial Implementation X X

X X X

X X X

X X X X X

X X X X X X X

X

21

No Implementation

X X

X X X X

X X X X X

X X X

X

15

130

Power-Sharing Settlements

claimed ownership of prime properties and used enlisted men to enforce eviction orders. Eventually, only 300 white commercial farmers remained on the land, of which half faced threats of dispossession. The total estimate of farm workers and family members displaced by land seizures rose above one million (CFU 2010; IDMC 2010).

ZANU-PF pressed for further asset nationalization. In February 2010, and without first advising the prime minister or cabinet, Minister of Youth Development, Indigenization, and Empowerment Savior Kasukuwere, announced that a pending Indigenization and Economic Empowerment Act would be put into effect. The law required that investors cede a controlling share of 51 percent of ownership of all major commercial enterprises to indigenous Zimbabweans, defined as black Africans. The MDCs joined economists and the business community in condemning the initiative as yet another attempt by predatory elites to loot economic assets.

Diamond revenues changed the national balance of power. Beginning in October 2008—after the GPA was signed but before the GNU was installed, and as the value of Zimbabwe’s currency imploded—the army launched a series of raids to secure control of the Marange diamond fields in Eastern Zimbabwe. According to eyewitness accounts, some 200 miners were killed, including those gunned down by helicopter gunships and bulldozed into mass graves (HRW 2009). Operation Hakudzokwi (“you will not return”) appeared to have two goals: “to ensure control of the diamond deposits for the ZANU-PF elite and to reward the army for its loyalty” (Global Witness 2010, 6). The bulk of the proceeds accrued to predatory elites, allegedly including members of the president’s immediate family, a national vice president, top security officials, and several cabinet ministers. As a result, the coalition government realized only a tiny fraction of potential diamond revenues (GRZ 2012). Political Gains

The level of violence declined from a peak in mid-2008. There was a significant decrease in the incidence of reported political abuse cases from a total of 1,320 nationwide in December 2008 to 319 in the same month in 2011 (ZPP 2012).

Parliament established a measure of political independence. The MDC-T was able to elect its own candidate, Lovemore Moyo, as speaker of the National Assembly in August 2008 and again in March 2011.6 But the speaker still had to work with a clerk of the house who was attached to the

Zimbabwe’s Power-Sharing Settlement, 2008–2013

131

old regime and experienced at the obstructive arts of parliamentary procedure. Thus, the parliament never succeeded in repealing repressive legislation (like POSA and AIPPA) and was hamstrung by the coalition arrangement in government; all controversial matters were negotiated by the GPA principals or within the executive branch. Political Setbacks

Political persecution persisted. The attorney general authorized the arrest of two dozen MDC legislators; one apparent goal was to sentence enough of them to jail terms that the MDCs’ parliamentary majority would be imperiled.7 Harassment of students, journalists, and activists continued; for example, a leading human rights advocate, Jestina Mukoko, was abducted by state security agents, tortured at an undisclosed location, and detained without trial. The public service remained politicized. Because political appointees remained on the public payroll, reformers found it difficult to restore a performance ethic to public administration (World Bank 2010). For example, the MDC-T minister of public service found his reform initiatives challenged by a Public Service Commission, whose chair had been in office since independence. A report on a payroll audit to review the scope of public employment and to eliminate ghost workers was repeatedly embargoed from release.

Security sector reform was blocked. In accordance with the GPA, a National Security Council (NSC) with multiparty civilian representation was intended to replace the Joint Operations Command. In practice, ZANU-PF refused to dismantle the JOC and six months passed before the NSC held a pro forma introductory meeting; but the NSC transacted no serious business and later met only sporadically. For example, the security chiefs obstructed the prime minister’s requests to monitor police posts and prisons and to visit the army-controlled diamond fields. In reality, Tsvangirai had little choice but to tacitly resign himself to playing second fiddle to Mugabe with regard to military matters.

These setbacks revealed a fundamental problem with the transitional government: power was not genuinely shared. ZANU-PF and MDC-T exercised power separately within exclusive and competing zones of authority. Thanks to intransigence during power-sharing talks, ZANU-PF had retained exclusive control over the coercive instruments of state. The party’s ministers were in charge of the security, intelligence, and judicial services,

132

Power-Sharing Settlements

as well as the politically strategic agencies responsible for land and agriculture, including provincial and local administration. For its part, MDC had an edge—though hardly complete ascendancy—in the representative institutions of the state, namely the House of Assembly and the chambers of local government councils. In addition, as nominal heads of economic and social ministries, MDC ministers served as gatekeepers of international aid. But, because no party was dominant, a divided coalition of political elites was unable to push through major pieces of reform legislation. Instead, democratization and development would largely be stalled as long as Zimbabwe remained one country with two governments. Moreover, a patronage culture endured. The GPA called for a large, sixperson executive (a president and prime minister, each with two deputies) and a bloated cabinet of 31 ministers and 15 deputy ministers. Yet the accord was violated at birth when ZANU-PF and the two MDCs colluded to appoint 41 ministers and 19 deputies, the largest and most expensive cabinet in Zimbabwe’s history. The expansion of official posts to accommodate political allies suggested that both sides were willing to sacrifice the careful management of scarce public resources in order to distribute political spoils. And some MDC cadres may well have regarded a government position as an opportunity to gain access to assets and rents previously enjoyed by ZANU-PF, as reflected in the demands of new MPs for state-of-the-art vehicles and other official benefits.

A Fragile Social Coalition

The limited impact of power sharing on institutional development was due as much to the weaknesses of would-be reformers as to the old regime’s strengths. The organizations that emerged under the MDC banner struggled to make the transition from an opposition movement to a disciplined and effective partner in government. When Tsvangirai took up the newly created position of prime minister he did so with few organizational resources. Not only did the prime minister possess ill-defined authority, but he also lacked basic facilities like office space, furniture, communications links, and a budget. The young staff of the Prime Minister’s Office (PMO), including eager returnees from the diaspora, were well-educated but inexperienced. They faced an entrenched and reluctant civil service whose officers—from permanent secretaries to gate guards—did their best to hamstring the newcomers. As a result, the prime minister—nominally the head of government business—failed to develop a coherent policy agenda. The strategy documents of the MDC-T party and the GNU administration, including a medium-term plan repeatedly rejected by cabinet and donors, were vague

Zimbabwe’s Power-Sharing Settlement, 2008–2013

133

on action steps. MDC ministers, who had never served previously in government, too often made undisciplined and contradictory policy statements without party approval. Most important, the instruments for policy formulation, legislation, and coordination remained under centralized control in the president’s office. In any event, the PMO was never intended to be an implementing agency; its role was to propose policy, not to execute it. That much became clear when the prime minister’s 2009 drought mitigation scheme to distribute agricultural inputs to small farmers was hijacked by ZANU-PF structures at the local level. As one PMO interlocutor said: “We are in office, but not in power.”8 Under these circumstances, some civic leaders, journalists, and diplomats—and even reformers within the MDC-T—charged that Tsvangirai was too accommodating to Mugabe. As a senior MDC official put it: “MT always backs down; RM never does so.”9 Intermittently throughout the GNU period, the PM made sincere efforts to cooperate by declining to publicly portray the president as the key source of governmental gridlock, but rather as part of the solution. He claimed a functioning working relationship with the man who previously had him jailed, beaten, and prosecuted. Yet critics wondered why the main MDC leader was not more forceful, for example, by making use of his party’s parliamentary strength to introduce political reforms or by speaking out more vigorously against abuses by security forces. As for building social alliances, the MDCs made modest progress in restoring the government’s relations with the business community. The MDCs’ neoliberal economic program and good governance discourse resonated with existing and prospective investors. The MDC-T appointed Eddie Cross—a former head of the Confederation of Zimbabwe Industries—as its secretary for economic affairs. The business community welcomed the dollarization policy and MDC-T Finance Minister Tendai Biti’s insistence that only over his “dead body” would he allow a premature resuscitation of the discredited Zimbabwean currency. Of course, ZANU-PF ministers often countermanded the MDC’s economic policy initiatives. For example, the PMO was blindsided when the president’s office announced “indigenization and empowerment” regulations that would require the state to have 51 percent ownership of all major economic enterprises. Business leaders worried, however, that politicians—MDC leaders included—did not fully realize how badly the economy was damaged nor how long it would take to recover. They were discouraged by the GNU’s failure to convene a National Economic Council (NEC), which left the country without an institutionalized forum for government, business, and labor to resolve policy differences. To restore stability and confidence, private sector interests wanted a longer period of political transition than that envisaged in the GPA; they argued that elections should be deferred in order for the government to implement business-friendly policies and a stronger recovery.

134

Power-Sharing Settlements

Relations between civil society and the MDC also evolved. Tensions first emerged within the original opposition–civil society alliance when, in 2007, SADC excluded civic actors from political negotiations over the country’s future, which the latter correctly saw as an exclusive process of elite pact-making. Thereafter, civil society developed independent positions for the 2008 election campaign, holding its own “peoples’ conferences” to call for economic and constitutional reform. There were also differences of opinion among NGOs about the wisdom of MDC-T joining a coalition that required the party to share power with ZANU-PF. The director of the NCA, still Zimbabwe’s leading NGO on constitutional affairs, charged the MDC with “capitulation” and the trade union confederation in South Africa opined that, in the power-sharing coalition, “the loser has become the winner and the winner the loser” (Bracking and Cliffe 2009). After the GNU was sworn in, civic associations took it upon themselves to monitor the implementation of the GPA: for example, civic organizations were more active at tracking the constitutional reform process and incidents of political violence than the government’s official Joint Monitoring and Implementation Committee (JOMIC). And other civic coalitions—like the Zimbabwe Election Support Network (ZESN) and the Election Resource Center (ERC)—took the lead in preparing the ground for free and fair elections by raising concerns about voter registration, press freedom, election monitoring, and campaign violence. As with the government itself, governance problems affected voluntary organizations in Zimbabwe, which themselves operated with little or no oversight. Reflecting a persistent elite-oriented political culture, a founder’s syndrome was prevalent, whereby the first leader of the organization ran it like a private fiefdom, sometimes for life. While the rest of the economy shrank after 2000, a civil society industry boomed, courtesy of donor funding. But Zimbabwe’s civil society organizations (CSOs) suffered technical deficits, in part because some of their best talent was drawn into the transitional government. Notable exceptions were the professional operations of the Zimbabwe Lawyers for Human Rights, which prosecuted cases of human rights abuses, the Counseling Services Unit, which provided medical and other assistance to the victims of political violence, and ZESN, which in 2008 ran the first independent parallel vote tabulation in the country’s history. Otherwise, the social coalition between the MDC and civil society was much more tenuous by 2013 than it had been in 2000.

An Evolving Public Mood

Despite the halting performance of the GNU, the citizens of Zimbabwe saw the power-sharing settlement as a welcome alternative to political conflict.

Zimbabwe’s Power-Sharing Settlement, 2008–2013

135

A nationally representative public attitude survey conducted in May 2009 indicated that two-thirds of all adults (66 percent, increasing to 72 percent by October 2010) agreed that “creating a coalition government was the best way to resolve the recent post-election crisis.” Only one-quarter (26 percent) thought that “coalition government [was] ineffective; leaders should have found another way” (Afrobarometer 2009). As such, ordinary people apparently lent much more legitimacy to the awkward GNU arrangement than did political elites. Yet, in other respects, Zimbabweans also viewed the GNU with eyes wide open. A plurality of citizens regarded power sharing as an elite bargain that fell short of their preferred method of choosing governments. More than four out of ten saw power sharing as “a second-best solution, to be used only when elections fail.” The rest of the electorate was divided, with one-quarter seeing power sharing as “a good alternative to elections, which rarely work well” and another quarter as “a bad alternative that should never replace competitive elections.” Moreover, citizens harbored skepticism about commitments to cooperation made by political elites. Fewer than half saw their leaders’ undertakings as “genuine” (46 percent), with one-third (32 percent) regarding these as insincere and 23 percent saying “don’t know.” A September 2009 survey revealed that more citizens saw MDC-T than ZANU-PF as “fully committed” to “making the Inclusive Government succeed” (42 versus 15 percent) (MPOI 2009). Realistically, they came to recognize that political power in the GNU rested more with the president (67 percent) than with the prime minister (6 percent). Finally, the general public was evenly split on the expected fate of the GNU: in September 2009, 47 percent expected it to survive, and 44 expected it to collapse; one year later, in October 2010, a slim majority (53 percent) expected the GNU to collapse. These increasingly skeptical attitudes toward the coalition government were reflected in trends in electoral support. At first, as shown in Table 7.2, MDC-T’s share of the intended vote (“if an election were held tomorrow”) rose from a low of 25 percent in 2007 to a high of 57 percent shortly after the inauguration of the GNU. Over a similar period, ZANU-PF’s share as the preferred party of eligible adult voters declined from a high of 41 percent in 2006 to a low of 10 percent in 2009. As the life of the GNU unfolded, however, the trends began to reverse direction. By October 2010, ZANU-PF had apparently regained some lost ground (to 18 percent) and support for MDC-T had dipped (to 36 percent). These trends continued through July 2012, at which time the two main parties were neck and neck, each attracting about one-third of the overtly expressed preferences of the electorate. Partisans interpreted these data to their own advantage. Proponents of ZANU-PF were tempted to see early evidence of their party’s resurgence,

136

Power-Sharing Settlements Responses to the question, “If a presidential election were held tomorrow, which party’s candidate would you vote for?”

Table 7.2 Voting Intentions by Political Party, Zimbabwe, 2006–2012 (percentage) October 2006

ZANU-PF 41 MDC-T 26 3 MDC-M/Na Other n/a Don’t know n/a Refused to answer 31 I will not vote n/a

December 2007 19 25 1 7 n/a 43 4

January 2008 21 30 1 5 1 41 2

March 2008 20 28