Growing Democracy in Africa : Elections, Accountable Governance, and Political Economy [1 ed.] 9781443888448, 9781443885478

Citation preview

Growing Democracy in Africa

The Cornell Institute for African Development/ Cambridge Scholars Publishing Editorial Board Muna Ndulo

Series Editor Professor of Law; Elizabeth and Arthur Reich Director, Leo and Arvilla Berger International Legal Studies Program; Director, Institute for African Development, Cornell University

Christopher Barrett

Stephen B. and Janice G. Ashley Professor of Applied Economics and Management, International Professor of Agriculture, Charles H. Dyson School of Applied Economics and Management, Cornell University

Sandra E. Greene

Professor of History, Cornell University

Margaret Grieco

Professor of Transport and Society, Napier University

David R. Lee

International Professor, Charles H. Dyson School of Applied Economics and Management, Cornell University

Alice Pell

Professor of Animal Science, Cornell University

Rebecca Stoltzfus

Professor of Nutritional Science, Cornell University

Erik Thorbecke

H.E. Babcock Professor of Economics and Food Economics, Emeritus; Graduate School Professor, Cornell University

Nicolas van de Walle

Maxwell M. Upson Professor of Government, Cornell University

Growing Democracy in Africa: Elections, Accountable Governance, and Political Economy Edited by

Muna Ndulo and Mamoudou Gazibo

Growing Democracy in Africa: Elections, Accountable Governance, and Political Economy Edited by Muna Ndulo and Mamoudou Gazibo This book first published 2016 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2016 by Muna Ndulo, Mamoudou Gazibo and contributors All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-8547-9 ISBN (13): 978-1-4438-8547-8

TABLE OF CONTENTS

List of Tables and Figures ........................................................................ viii Acknowledgements .................................................................................... ix Contributors ................................................................................................. x Acronyms .................................................................................................. xv Introduction ................................................................................................. 1 Muna Ndulo and Mamoudou Gazibo Part 1: Institutions and Concepts of Governance Chapter One ............................................................................................... 10 Revisiting the Study of Governance Göran Hyden Chapter Two .............................................................................................. 28 Democratisation in Africa: Achievements and Agenda Mamoudou Gazibo Chapter Three ............................................................................................ 47 The Expansion of Judicial Power in Africa and Democratic Consolidation: Opportunities, Challenges and Future Prospects Charles M. Fombad Part 2: Constitution-Making, Elections, and Conflict Settlement Chapter Four .............................................................................................. 86 Constitution-Making in Anglophone Africa: We the People? Coel Kirkby and Christina Murray Chapter Five ............................................................................................ 114 Measuring the Persuasive Effects of Electoral Campaigns in Africa Jeffrey Conroy-Krutz

vi

Table of Contents

Chapter Six .............................................................................................. 132 Subnational Elections and Accountability: A Study of Political Decentralization and Democratic Governance Rachel Beatty Riedl and J. Tyler Dickovick Part 3: Local Governance and Citizenship Chapter Seven.......................................................................................... 162 Slippery Citizenship: Nationalism, Democracy and the State in Africa Jennifer Riggan Chapter Eight ........................................................................................... 184 Federal Developments and Accountable Government Structures in East Africa Jan Amilcar Schmidt Chapter Nine............................................................................................ 199 Hereditary Rule in Democratic Africa: Reconciling Citizens and Chiefs Kate Baldwin Chapter Ten ............................................................................................. 219 Critical Reflections on Social Accountability and Local Government in Ghana Cyril K. Daddieh Part 4: Political Economy and Corruption Chapter Eleven ........................................................................................ 246 “Emerging” Africa: Long-term Perspectives on Growth and Democracy Antoinette Handley Chapter Twelve ....................................................................................... 271 A Legal Framework for Combating Corruption: Case Study from Zambia Muna Ndulo Concluding Thoughts Chapter Thirteen ...................................................................................... 304 Two-and-a-Half Cheers for Democracy in Africa Nicolas van de Walle

Growing Democracy in Africa

vii

Bibliography ............................................................................................ 313 Index ........................................................................................................ 356

LIST OF TABLES AND FIGURES

Table 1.1. Shifts in the use of the governance concept .............................. 15 Table 1.2. Operationalizing the democratic governance agenda ............... 26 Table 5.1: Aggregated Stated Presidential Preferences ........................... 124 Table 5.2: Percentage of Prior Period Supporters Retained..................... 129 Table 6.1: Subnational Elections, Party Systems, and Accountability .... 142 Table 6.2: Decentralization and Histories of Political Instability ............ 151 Table 9.1. Complaints Against Chiefs by Embeddedness ....................... 217 Figure 1.1. The theoretical origins of the governance concept .................. 13 Figure 1.2. Anchoring-points for select key actors in the governance field ...................................................................................................... 14 Figure 1.3. A framework for analysing democratic governance................ 25 Figure 2.1: Different Models of EMB ....................................................... 32 Figure 9.1. Map of Chiefdoms in Zambia ............................................... 205 Figure 9.2. “Representativeness” of Chief by Selection Method ............ 211 Figure 9.3. “Representativeness” of Chiefs by “Embeddedness” ............ 215

ACKNOWLEDGEMENTS

This book is the result of a symposium titled Elections, Accountability, and Democratic Governance in Africa held at Cornell University April 20–21, 2012. We are indebted to the organizers and sponsors of the symposium for bringing together the scholars who have contributed to this volume. Organizers include the Institute for African Development at Cornell University, with collaboration from the Department of Political Science, University of Montreal; additional sponsorship came from Cornell University units including the Atkinson Center for a Sustainable Future, the Berger International Legal Studies Program, the Department of Government, the Institute for Social Sciences, the Polson Institute for Global Development, and the Mario Einaudi Center for International Studies. The symposium and this book would not have been possible without the help of several people. In particular, we would like to thank the staff from the Institute for African Development: Jackie Sayegh, Program Manager, who did an excellent job in organizing the conference, and Evangeline Ray, Publications Manager, who worked on the compilation of the book. Without them this book would not have been possible. Evangeline Ray deserves special thanks as she has been instrumental in bringing the book project to fruition. She worked tirelessly to liaise with the authors and to painstakingly and diligently edit the manuscript for publication. She remained unfailingly committed to the project throughout its preparation for publication. In addition, we wish to thank student assistants Mihret Tamrat and Alyssa Findley for their extensive help with technical aspects of editing. Thanks also to Cambridge Scholars Publishing/Cornell Institute for African Development Book Series for their assistance and for publishing the work.

CONTRIBUTORS

KATE BALDWIN is Assistant Professor of Political Science and a faculty fellow at the Institution for Social and Policy Studies, Yale University. Her current research projects examine how community-level institutions interact with the national state to affect development, democracy, and conflict, with a regional focus on sub-Saharan Africa. Previously she was assistant professor of political science at the University of Florida. She has held fellowships at the Center for the Study of Democratic Politics, Princeton University, and the Alexander Hamilton Center for the Study of Political Economy, New York University. JEFFREY CONROY-KRUTZ is Assistant Professor of Political Science at Michigan State University. He received his PhD in Political Science from Columbia University. His research focuses on political decision-making, particularly surrounding elections, in sub-Saharan Africa. He is currently working on a book project on mass media, access to political information, and electoral decision-making in Africa, and is also conducting research on comparative electoral campaign behaviour in Africa. He has conducted fieldwork for various research projects in Senegal and Uganda. CYRIL K. DADDIEH is Assistant Professor of Political Science and Director of the Black World Studies Program at Miami University of Ohio. He holds a PhD from Dalhousie University. His teaching and research interests are in contemporary African politics, democratisation and elections in Africa, and international relations and foreign policies of African states. He was the winner of the Hampton Grant for International Research in 2008 and a MacArthur Foundation Grant for research on contemporary African international relations in 1990–1991. J. TYLER DICKOVICK is an Associate Professor of Politics at Washington and Lee University. He conducts research on African and Latin American politics, with a focus on decentralization, federalism, and local governance. His current research projects include a manuscript on decentralization and local governance in Mali and an edited volume that compares processes of decentralization in ten countries in sub-Saharan Africa. He has a PhD in Public Affairs from Princeton University.

Growing Democracy in Africa

xi

CHARLES M. FOMBAD is Professor and Head of the Department of Public Law at the University of Pretoria. He holds an LL.B from Yaounde University and an LL.M and PhD from the University of London. He has been a Senior Lecturer at the University of Yaounde, a professor at the University of Botswana, and Professor Honorarium, University of South Africa. He has been a visiting lecturer at the Universities of Buea and Dschang in Cameroon. His research interests include constitutional and international law, comparative law, and legal history. He is a member of the South African Association of Legal Historians, the Organization for Social Science Research in Eastern and Southern Africa, the African Network for Constitutional Lawyers, and the International Association of Constitutional Law. He is a member of the editorial boards of several academic journals. MAMOUDOU GAZIBO is a Professor of Political Science at the University of Montreal (Canada). He holds a PhD in comparative politics from the University Montesquieu, Bordeaux, France. His research relates to democratisation and institution-building in Africa, China-Africa relations, and theoretical/methodological issues in comparative politics. He is the author, co-author, or editor of several books. He has been a consultant for international organizations including the African Union and the International Organization of Francophone States. He chaired the constitution drafting committee in Niger in 2010 and served as senior adviser to the Nigerien Prime Minister from 2010–11. ANTOINETTE HANDLEY is Assistant Professor in the Department of Political Science at the University of Toronto. She holds an MPhil in International Relations from the University of Oxford and a PhD in Political Science from Princeton. Her research interests include policymaking and economic reform in developing countries and the political economies of Africa (especially Southern and South Africa) and Latin America. Recently, she has focused on business as a political actor, examining business interactions with the state at a macro-level. From 1995–98 she served as Director of Studies at the South African Institute of International Affairs in Johannesburg, South Africa.

GÖRAN HYDEN is a Distinguished Professor Emeritus in the Department of Political Science at the University of Florida. He holds degrees from Oxford and University of California and a PhD from the University of Lund. He was Professor of Political Science at the University of Dar es Salaam, Senior Lecturer in Government at the University of Nairobi, and

xii

Contributors

Lecturer in Political Science at Makerere University, Uganda. He has been President of the African Studies Association, Chairman of the board of the Dag Hammarskjold Foundation, and an Associate of the Nordic African Institute. In addition, he has been a consultant to international organizations including: UBICEF, UNDP, the World Bank, the African Development Bank, Economic Commission for Africa, the Swedish International Development Cooperation Agency (SIDA), the Norwegian Agency for Development (NORAD), and the Danish Development Agency (DANIDA). He has published widely on Africa.

COEL KIRKBY is a McKenzie Fellow at the University of Melbourne Law School. Previously he was an Endeavour Fellow at the University of New South Wales. He was awarded his PhD by the Faculty of Law, University of Cambridge, in 2013. He has co-edited a book and published several articles and chapters in leading journals. He has also assisted democratic transitions as a constitutional law advisor in Fiji and the Democratic Republic of Congo. CHRISTINA MURRAY is Professor of Constitutional and Human Rights Law at the University of Capetown, where she was director of UCT’s Law, Race, and Gender Research Unit from 1992 to 2004. In 2009–10 she served on the Kenyan Committee of Experts appointed to draft the new Kenyan Constitution. In 1994–96 she served on a panel of seven experts advising the South African Constitutional Assembly in drafting South Africa’s Constitution. Since then she has advised a number of South African government departments on the implementation of a new system of multi-level government. She has also done constitution work in Southern Sudan, Nepal, Zimbabwe and Pakistan. MUNA NDULO is Professor of Law; Elizabeth and Arthur Reich Director, Leo and Arvilla Berger International Studies Program; and Director of the Institute for African Development at Cornell University. He is also Honorary Professor of Law at the University of Cape Town, South Africa; and Extraordinary Professor of Law, Free State University. He holds an LLB from the University of Zambia, LLM Harvard University and D. Phil, Oxford University. He is former Professor of Law and Dean of the Law School at the University of Zambia. He served as Legal Officer for the United Nations Commission on International Trade Law (UNCITRAL) and as Senior Political Advisor to the Secretary General’s Special Representative to South Africa. He was Senior Legal Advisor on UN missions to East Timor, Kosovo, and Afghanistan. He has consulted on

Growing Democracy in Africa

xiii

constitution-making in Kenya, Zimbabwe, and Somalia; he has also been a consultant for international organizations including the African Development Bank, where he helped to establish the African Legal Support Facility; the Economic Commission for Africa; the International Foundation for Electoral Systems, and the International Institute for Democracy and Electoral Assistance. He is founder of the Southern African Institute for Public Policy and Research, and he is a member of the Advisory Committee, Human Rights Watch (Africa). RACHEL RIEDL is an Assistant Professor of Political Science at Northwestern University. She holds a PhD from Princeton University. At Northwestern, she is also affiliated with the Program in ComparativeHistorical Social Science (CHSS) and the Program of African Studies. In addition she is a Faculty Associate at the Buffett Center for International and Comparative Studies and a Faculty Affiliate at the Kaplan Institute for the Humanities. JENNIFER RIGGAN is an Associate Professor of International Studies in the Department of Historical and Political Studies at Arcadia University. She is a political anthropologist whose ethnographic research addresses a variety of issues including nationalism, citizenship, state formation, militarism, development, and education in Africa. She has published on the changing relationship between citizenship and nationalism; the decoupling of the nation and the state; and the relationship between militarization, education, and development. Her research has been funded by a Fulbright research fellowship, a Social Science Research Council International Dissertation Field Research Fellowship, and a Spencer/National Academy of Education Postdoctoral Fellowship. JAN AMILCAR SCHMIDT is a Research Fellow in the Africa Project at the Max Planck Institute for Comparative Public Law and International Law. He holds a Masters in European Law from the Universities of Hanover and Strathclyde/Glasgow. He is a member of the German bar. From 2001–06 he was a student assistant and research fellow at the Institute for Federal Studies at the University of Hannover. At the Max Planck Institute, he is involved in capacity-building projects and workshops for Sudanese lawyers and the Independent Federal Constitutional Commission of Somalia. He is also a constitutional advisor for the UNDP Somalia/UNPOS Joint Constitution Unit on the National Constitutional Process in Somalia.

xiv

Contributors

NICOLAS VAN DE WALLE is the Maxwell M. Upson Professor of Government at Cornell University. He holds a PhD from Princeton University. He has published widely on democratisation issues as well as on the politics of economic reform and on effectiveness of foreign aid, with special focus on Africa. In addition, van de Walle has worked extensively as a consultant for a variety of international and multilateral organizations, including the World Bank, USAID, and UNDP. From 2004 to 2008 he was director of Cornell’s Einaudi Center for International Studies and Associate Dean for International Studies, College of Arts and Sciences.

ACRONYMS

AfDB AMA B-BBE B-BBEE CALPERS CalPERS CDD-Ghana CHRAJ CKRC CSO CVC CWSA DA DA DACF DCEs DEC DP DPP ECA ECA EMB EPRDF EPWP ERISA ERISA FDC FGDs FRELIMO FSDEA GBA GBC GEPF GEPF GGP HCR

African Development Bank Accra Metropolitan Authority Broad Based Black Economic Empowerment Broad-Based Black Economic Empowerment California Public Employees Retirement System California Public Employees Retirement System Ghana Center for Democratic Development Commission on Human Rights and Administrative Justice Constitution of Kenya Review Commission Civil society organizations Citizens Vetting Committee Community Water and Sanitation Agency District assembly Democratic Alliance District Assemblies Common Fund District Chief Executives District Election Committee Democratic Party Director of Public Prosecutions Excess Crude Account (Nigeria) Economic Commission for Africa Electoral management body Ethiopian People's Revolutionary Democratic Front Expanded Public Works Program Employee Retirement and Income Security Act (U.S.) Employee Retirement and Income Security Act Forum for Democratic Change focus group discussions Mozambique Liberation Front Sovereign Fund of Angola (Fundo Soberano de Angola) Ghana Bar Association Ghana Broadcasting Corporation Government Employees Pension Fund (South Africa) Government Employees Pension Fund Ghana Gas Project High Council of the Republic

xvi

ICAC ICT IDEA IGEPE IGEPE IMF INC MCEs MLGRD MPs NAPRMGC NCP NDC NDPP NMC NNPC NPP NRM NSIA NSIA NSSF NSSF NUGS ODI ODM OECD OPIC PDC PDP PE PETS PFMA PIC PIC PNDC PPP PRCC RAIDS

Acronyms

Independent Commission against Corruption- in Australia information and communications technology Institute for Democracy and Electoral Assistance Institute for the Management of State Holdings (Mozambique) Institute for the Management of State Holdings International Monetary Fund Interim National Constitution of 2005 Municipal Chief Executives Ministry of Local Government and Rural Development Members of Parliament National African Peer Review Mechanism-Governing Council National Congress Party National Democratic Congress National Director of Public Prosecutions National Media Commission Nigerian National Petroleum Corporation New Patriotic Party National Resistance Movement Nigerian Sovereign Investment Authority Nigerian Sovereign Investment Authority National Social Security Fund National Social Security Fund (Kenya) National Union of Ghanaian Students Overseas Development Institute Orange Democratic Movement Organization for Economic Cooperation and Development Overseas Private Investment Corporation People’s Defence Committee People’s Development Party Private Equity Public Expenditure Tracking Systems Public Finance Management Act Public Investment Corporation (South Africa) Public Investment Corporation (South Africa) Provisional National Defence Council (Ghana) People’s Progressive Party Public Relations and Complaints Committee Resource-based Industrialization and Development Strategy

Growing Democracy in Africa

RBA SAc SADC SFO SITET SLCAC SMDP SNG SOE SONIP SPLM/A SSNIT SWF TCSS TUC UFA UNCAC UNIP UPC WDC

xvii

Retirement Benefits Authority Social Accountability Southern African Development Community Serious Fraud Office Special Investigations Team on Economy and Trade Senate Legal and Constitutional Affairs Committee single-member district plurality Subnational government State-owned enterprise Sonangol Imobiliaria e Propriedades Sudanese People Liberation Movement/ Social Security and National Insurance Trust Sovereign Wealth Fund Transitional Constitution of the Republic of South Sudan Trade Union Congress Uganda Federal Alliance UN Convention against Corruption United National Independence Party Uganda People’s Congress Worker’s Defence Committee

INTRODUCTION MUNA NDULO AND MAMOUDOU GAZIBO

What is the state of governance in sub-Saharan Africa, and is it possible to identify best practices and approaches to establishing political systems that promote accountability, transparency, peace, and civic space to all? These are the questions addressed in this book. This focus is not arbitrary but was chosen for at least three broad reasons. First, as of 2015, a quarter-century has passed since the onset of democratisation in Africa and since democratisation was imposed in the nineties by popular mobilizations demanding better governance; it is now time to critically examine the record thus far in order to understand what has been achieved and what still needs to be done. Second, although the concept of governance is now considered central in political science, our understanding of what it encompasses and how it can be operationalized in the academic study of Africa is still weak. Third, the vast majority of studies on governance in Africa consist either of think-tank indicators (e.g. the World Governance Index, Ibrahim Index of African Governance); managerial perspectives centred on economic management; or strictly political studies on democratisation. This book does not reject the studies mentioned above. However, it seeks a new, integrated, focused approach to governance. Such an approach requires revisiting the concept of governance itself; emphasizing certain specific and decisive components; and giving an integrated view of the most critical governance issues in Africa.

Revisiting Governance: the Need for a New Study The concept of governance has become one of the most often used in the social sciences (Levi-Faur 2014). Its success is partly due to its transversal character, as it is used in a variety of subfields such as public administration, international relations, and comparative politics (Kjaer 2014). With regard to Africa, the concept is often considered as synonymous to “good” governance. This is not surprising given the context in which its application to Africa has blossomed. In fact, even though the

Introduction

2

theoretical roots for the concept are profound and the issues it is applied to in Africa diverse (see chapter 1), the concept of governance became truly central following the 1989 World Bank publication entitled Sub-Saharan Africa: from Crisis to Sustainable Growth. This publication appeared in a time when state economic bankruptcy and political deadlock and conflict were prevalent in most African countries. Africa was considered to be in a “crisis of governance”—a concept referring to economic mismanagement and lack of politically responsive institutions capable of guaranteeing liberty and sound economic policies. Good governance was thus seen as the solution to Africa’s political and economic crisis. We will not discuss the concept of governance further here as the first chapter of the volume (Göran Hyden, “Revisiting the Study of Governance”) is devoted entirely to an in-depth discussion of the complexities and controversies surrounding it. Hyden reviews the theoretical origins of the concept and examines how its uses have changed over time, observing that the term “governance” has become a catch-all notion referring to various reform projects in public administration, macro-economic management, and policies. He specifically considers the challenges of studying governance in the African context, concluding with a proposal for how to study democratic governance—the focus of this book—and an explanation of why such an approach makes sense in the current context. The contributors to this book share two characteristics. On a conceptual level, whether they study democratisation, elections, decentralization, corruption, or economic issues, they all approach their studies from a democratic governance framework, which means they explore either the representative or monitoring aspect of governance in Africa. As Hyden notes, “Governance has two dimensions, one representative (voice) and a second monitory (accountability). The issue is how these two dimensions can be further disaggregated and operationalized. One possibility is to do this with reference to political functions: articulation, aggregation, and so on. Another is to distinguish between different arenas, e.g. civil society, political society, and so forth. Data availability becomes a critical factor in determining how to proceed.” The second thing the contributions share is their methodological approach. No matter the specific issue they study, they do their best to focus their work on the following objectives: x x

To capture the current state of knowledge and discourse and review the lessons learned as well as the problems and challenges; To examine the progress made in the past quarter century since democratisation began on the continent;

Growing Democracy in Africa x

3

To provide, when necessary, policy recommendations that could strengthen democratic institutions and good governance in Africa.

It is also worth mentioning that, rather than taking governance as a unified concept, we have opted for a “disaggregation and prioritization approach.” This means two things. First, we accept the fact that governance, if considered in a democratic framework, can be used to understand voice and accountability issues in a variety of arenas. Second, because we obviously cannot study all these arenas, we have chosen to prioritize those we consider, following the most salient and critical literature on democratic governance. Using this approach, we have identified four arenas for consideration here: institutions and the rule of law; constitution-making, elections, and political conflict settlement; distribution of power and citizenship; and political economy and corruption.

Governance Arenas and Critical Issues in Africa Each contribution in the volume offers particular insights in one of the four arenas, which also constitute the four parts of the book. Together, the four arenas offer a broad—though yet not complete—view of the challenges to and prospects for good governance in contemporary Africa. The contributors all point out the challenges of accountability, participation, and transparency and offer (overtly or not) possible solutions to the problems identified.

Institutions and the rule of law There is consensus that governance is at the heart of Africa’s development challenges and that the institutional aspects of governance are crucial (World Bank 1989; Ndulo 2006). This is at the heart of all the democracy promotion initiatives that have taken place in Africa since 1989. The past three decades have witnessed considerable flux in systems of governance at the political level. There is clearly a realization that institutions are central to the efficiency and effectiveness of public service delivery, poverty reduction, optimum resource exploitation and utilization, public management, transparency and accountability, peace and security, and respect for the rule of law. Thus, consolidating democracy in terms of the building of democratic institutions and the capacity of the state to manage the political and economic processes of society for development remains a major challenge for many African countries. Following the democratic wave that swept the continent a quarter-century ago, it had been hoped that greater

4

Introduction

political freedoms and strong institutions would be crafted and in turn lead to more government accountability. The question is: has this happened? The first part of the book provides conceptual and macro level assessments on the matter. Given the richness and multifaceted nature of the concept of governance, Göran Hyden’s chapter provides an invaluable resource as it traces the state of the art and some general and flexible conceptual guidelines for the rest of the book. Because governance is clearly—and perhaps firstly—about institutions and the rule of Law, Part I provides a general overview of what has been achieved in Africa in the first quarter-century since democratisation began, with particular emphasis on specific institutions such as electoral management bodies (Gazibo, chapter 2) and the judiciary (Fombad, chapter 3). As many contributors— including van De Walle in the conclusion—advise, it is useful to make a clear distinction between the open-ended process of democratisation attempts and the relatively stable institutions of democratic forms of government. Many of the contributors, while focusing on different specific issues, share the notion that the state of democracy in Africa is in a grey zone. The major question is whether the evidence in the book raises hope for democratic consolidation in Africa. The answer is particularly important today as Africa is in the midst of an economic and demographic transformation. Democracy and the rule of law remain the most important preconditions for the institutionalization of good governance and sustainable development practices.

Constitution-making, elections, and political conflict settlement As part of part of the effort to consolidate democracy, promote good governance, and improve the state’s delivery of public goods, a number of African countries have been involved in some type of political change in recent years. Constitutional change is one aspect of this effort; a good constitution, one that takes into account all stakeholders, forms the framework for good governance. Many African constitutions have undergone radical changes in recent decades, with every country except Botswana implementing some fundamental constitutional change since the 1990s. Christina Murray and Coel Kirby (chapter 4) give practical perspectives on the implications of constitution-making procedures, explore emerging norms, and examine the rise of participatory constitution-making in Anglophone Africa over the past two decades. A large majority of African countries now conduct regular elections at national, regional, and local levels, enabling populations to choose their political leaders and ensuring legitimacy of elected government officials.

Growing Democracy in Africa

5

However, in most countries the progress remains fragile and in others there have been setbacks (Lynch and Crawford 2011). In its 2011 Annual State of the Continent report, the African Progress Panel observed— pointing to the crises in Libya and Ivory Coast as example—that nearly two-thirds of African countries have seen some deterioration in political participation, human rights, and the rule of law. The report notes a worrying trend—that of leaders clinging to power. Six of the nine presidential elections held in 2010, for example, were won by incumbents, some of whom have been in power for well over two decades. Holding elections is a massive enterprise anywhere in the world, let alone in a developing country with limited resources. Although holding elections is not in and of itself a sign of democratic consolidation, it is a defining characteristic of democracy and thus an integral part of democratisation and the democratic consolidation process. In post-conflict societies, elections are supposed to provide a means of jump-starting a new, post-conflict political order for stimulating the development of democratic policies, for choosing representatives, for forming governments, and for conferring legitimacy upon the new political order. But at the same time, one must not ignore the potential conflicts provoked by elections in Africa, as the post-electoral violence and civil war in Kenya (2007) and Côte d’Ivoire (2011) remind us. The chapters in Part 2 revisit these issues. Because elections are about replacing armed struggle for power by electoral struggle for citizens’ votes, reflecting on the possibility to influence citizens’ preferences through electoral campaigns (Conroy-Krutz, chapter 5) appears as important as analysing the peace-enforcing power of elections in postconflict countries. The same could be said about elections at the subnational level, because they may create vested interests among large numbers of households for whom elections become an important mechanism of governance that citizens can claim (Riedel and Dickovick, chapter 6). There is a need to assess how free and fair electoral processes are in Africa. However, it is not just the conduct and outcome of elections that must be assessed, but the extent to which consolidation is grounded in an institutional development that ensures democracy’s viability and sustainability. In Africa, questions linger on how best to ensure free and fair elections and on whether elections lead to democratic consolidation—and if they do, in what situations this is likely to occur (for two different perspectives, see Adejumobi 2000 and Lindberg 2006a).

6

Introduction

Distribution of power and citizenship Governance, and particularly democratic governance, is clearly related to citizenship. The notion of citizenship refers to something more than being a mere member of a community; it means also the capacity for such a member to evaluate and take part in the decision-making processes of the polity. Historically, in countries formerly ruled by kings and monarchs, true citizenship has evolved only when these countries have witnessed the development of public arenas that transformed the subjects into full citizens. This process, which can be said to have begun in England in the early thirteenth century, evolved through two complementary processes: democratisation and the people’s right to choose and remove their leaders on one hand, and territorial arrangements to accommodate subgroups sharing the same polity on the other. In Africa, achieving these two processes remains a formidable challenge. African countries are, in fact, very young. They face all kinds of issues such as nation-building, territorial integration, and coordination between traditional and European-style political structures. These issues were generally shadowed by authoritarian regimes during the first three decades following Africa’s wave of independence, but were revived during the new constitutionalism era of the nineties. Beginning around 1990, the new constitutions established institutional arrangements such as human rights commissions and the office of Public Protector to promote accountability, transparency, and service delivery. They also encouraged devolution to bring power and decision-making closer to local communities. Most African countries show signs of becoming increasingly adept at dealing with potential conflicts through some kind of territorial devolution, whether through nationalism and social category construction (Riggan, chapter 7), federal arrangements (Schmidt, chapter 8), rethinking the role of traditional rulers (Baldwin, chapter 9), or subnational decentralized units (Daddieh, chapter 10). The chapters in Part 3 analyse progress toward democratic governance by putting the focus on the citizenship issue. The effectiveness of institutions and policies in holding governments accountable—and indeed how the accountability institutions themselves are to be held accountable—remains a greatly contested issue.

Political economy and corruption There is also agreement that accountable governance is central to all issues relating to the efficiency and effectiveness of governments. Good

Growing Democracy in Africa

7

governance promotes public service delivery and the eradication of poverty, thus fostering the capacity of citizens to participate in public affairs. On the contrary, bad governance, particularly corruption, impacts negatively on the continent’s economic development. This arena is particularly crucial since Africa has been experiencing unprecedented economic development since 2000 (Handley, chapter 11) thanks to several factors such as high commodity prices and Africa’s relations with emerging powers such as China. With an average economic growth rate since 2000 of about 5 percent, the continent has become the new frontier for investments (AfDB 2013). However, this growth has not necessarily translated into sound policies, as corruption and mismanagement—which undermine the capacity of institutions charged with the promotion of good governance, human rights, and the rule of law—often come with it. Corruption is at the heart of governance issues. Yet it is difficult to propose a “one size fits all” definition of corruption. As Wedel (2012, 454) has stated, “the meanings of corruption and the practices the word evoke change over time and place, as do the social and political uses of anticorruption campaigns in the hands of leaders, organizations, and political regimes. Scholarly definitions and approaches to the topic are also subject to the influences of time, place, and politics.” Using a micro-level approach, one may equate corruption to acts of bribery, a phenomenon that, according to Hameed et al. (2014), exacts a global cost of at least one trillion dollars each year. When it is pervasive in society—particularly in the political system—corruption may also be defined at a systemic level. These two approaches have been applied particularly to the study of governance in Africa. Perhaps the concept of neopatrimonialism captures best these two intertwined levels of corruption and how the forms they take vary from one context to another (Bach and Gazibo 2012). According to Médard (1991), the main characteristic of neopatrimonialism is the absence of distinction between private and public affairs. This refers to a wide range of practices, such as the privatization of public resources by state officials, patronage, bribes taken by customs or police officers, collusion between political and economic milieus, and so on (Johnston 2006). The importance of studying (anti) corruption as an essential aspect of any research on good governance (Ndulo, chapter 12) in an era of democratisation is thus obvious. Democratisation is supposedly a good antidote to corruption, while corruption (and bad governance) erode democracy. As we stated above, these arenas are interconnected. For example, many of the new constitutions go beyond the traditional concepts of focusing on organs of government, separation of powers, and the protection of political and civil rights to include a role in service provision. There is a

8

Introduction

growing realization that the state has a vital role to play in provision of basic services, either by direct action or by properly regulating the way services are provided via the private sector, especially in the areas of health, education, infrastructure, and basic necessities like water and electricity. This realization is reflected in the growing number of national constitutions that include provisions for social and economic rights as foundational, and provisions for social goods, the realization of social justice, and preservation of the environment as fundamental objectives. Also, institutional infrastructure is key if the economic growth Africa is experiencing is to be profitable to all citizens. Institutions cannot survive in an environment of poor governance, especially in a corrupt and conflicted state (Gazibo 2012). This book will have achieved its purpose if the discussions contained herein and the various challenges, achievements, and lessons outlined contribute to research, inform teaching, and lead to a greater understanding of the challenges of democratic consolidation in Africa and the relationship between democratisation and development. In addition, our purpose is to share theoretical models and best practices for engaging in the issues around democratisation and economic development in Africa.

PART 1: INSTITUTIONS AND CONCEPTS OF GOVERNANCE

CHAPTER ONE REVISITING THE STUDY OF GOVERNANCE GÖRAN HYDEN

Governance is at the heart of Africa’s challenges because it has implications for the continent’s social and economic development as well as its political stability. But what is governance and how is it best defined and operationalized? Governance as a concept is itself a challenge. When it was first introduced some twenty years ago, it was an attempt by social science scholars to analyse and understand significant changes that were taking place: the rise of global policy issues like environmental protection and human rights, the emergence of a democratic transition, and the demand for greater public accountability in public service delivery. The concept of governance was intended to provide additional analytical leverage in an increasingly complex and challenging policy environment, but as an analytical tool it has fallen short of expectations. Looking back over the past two decades, the problem is that governance has become a catch-all notion referring to all kinds of reform projects—in public administration, macro-economic management, and politics. “Governance” has become a programmatic rather than an analytical tool. With its ideological tie to liberal democracy, it has mostly been used prescriptively. Not surprisingly, the term has been used more by managers and practitioners than by academics. So, should the concept be declared useless for scholarly purposes, or can it be salvaged for future research? This chapter argues the latter: that the time has come to resuscitate interest in the concept of governance for analysing and understanding political processes. More specifically, there is reason to link democracy and governance together in ways that broaden the use of the former and narrow the use of the latter. For political scientists, the concept of democratic governance offers a tool for analysis of politics in a post-transition period and sharpens the focus on what governance is really all about: the politics of rules.

Revisiting the Study of Governance

11

We begin with a brief review of the theoretical origins of the governance concept and how its uses have changed over the past twenty years. We then continue with a discussion of the challenges to studying governance in an African context and conclude with a proposal for how democratic governance might be studied and why such an approach makes sense in the current context.

Theoretical Origins of the Governance Concept “Governance” may have been around as a concept for a long time, but it was essentially dormant until the 1970s, when the term began to be used in public administration literature and subsequently in the works of international relations and comparative politics students. Its sudden rise did not come out of nowhere. A convergence of factors helps explain why scholars in various fields adopted governance as a term to analyse their respective subjects. The growth of inter-jurisdictional policy issues encouraged public administration theorists to look for a concept that allowed them to think beyond conventional terms in their field. International relations scholars took on the term as a means of analysing the emergence of global issues such as the environment and human rights following the end of the Cold War. Students of comparative politics, finally, began to use “governance” during the wave of democratisation that began in the wake of the fall of Communism. Kjaer (2004) and Bevir (2010) have provided a thorough overview of the history of the concept, but a little more elaboration may still be justified here, especially to include the experience outside of academe. Beginning with public administration, the first use of the term “governance” is attributed to Harlan Cleveland in the early 1970s (Frederickson 2005). Writing a guide for future managers, Cleveland (1972) called into question the vertical arrangements of conventional public administration and argued for more attention to horizontal relations between organizations. He wrote that organizations individually needed to become better at using their own, in-house expertise while at the same time, as a group, becoming more collegial and consultative. His assumption was that people wanted less of government and more of governance. The concept, therefore, appears to be the outgrowth of two simultaneous trends in the field of public administration: the blurring of the relationship between what is public and private and a similar tendency to dim the relationship between policy and politics, on the one hand, and implementation and administration, on the other. Governance became gradually associated with the New Public Management School and its prescriptions for reforming public ad-

12

Chapter One

ministration by contracting out responsibilities to non-state actors. A review of the uses of governance in the public administration field includes some advocates who hold the radical belief that it is possible to have governance without government, as discussed by Pierre and Peters (2000). Many scholars, especially in Europe, however, approached the subject of governance in less prescriptive terms and merely acknowledged and described a new decision-making reality characterized by interorganizational dependencies and thus also by the need for multi-level and multi-organizational responses to societal problems (Kooiman 1993). Much of that perspective has made its way into the thinking and use of the governance concept in the international development community. As Frederickson (2005, 293) notes, the problem with the governance concept in public administration is that it lacks a theory; he suggests that scholars should look to international relations studies, where regime theory constitutes the basis for understanding the term. Regimes are deliberately constructed orders at the regional or global level aimed at reducing the risk of unilateral action by powerful nations. They establish shared expectations about desired behaviour. Governance, then, is the exercise of establishing and sustaining such regimes. An example would be the efforts in the 1990s to institutionalize an international human rights regime drawing on the work prepared for the 1993 International Human Rights Conference in Vienna. Thus, as some influential scholars in the international relations field have argued, governance involves managing principles, norms, and decision-making procedures that facilitate the maintenance of an international order (Krasner 1983; Keohane 2001). Much of the literature and practice of global governance incorporates these insights. Rosenau (1992, 13) makes the point that global governance applies to any system of rules that has transnational implications. Governance as related to systems of rule is found also in comparative politics literature, but here it is typically confined to individual countries and associated with regime transition issues spurred by democratisation (Hyden and Bratton 1992). Governance has spun off two distinct directions of research in the field. The first has been inspired by the work of rational choice theorists like Douglass North (1990) and Elinor Ostrom (1990), for whom the main objective has been to devise institutional solutions to issues such as bureaucratic red tape and top-down and centralized problem-solving. The second has been driven by the theoretical insights of Joseph Schumpeter (1942) and Robert Dahl (1971) on building democratic regimes. As can be gathered from the overview, albeit compressed, the intellectual heritage on which governance discourse rests is varied and complex.

Revisiting the Study of Governance

13

It is possible, however, to confine the discourse to two main parameters: 1) effectiveness, and 2) legitimacy. Influences from public administration in particular come closest to the effectiveness dimension, while those from international relations and comparative politics are more closely related to legitimacy. The former tend to encourage a managerial, technocratic approach to governance, while the latter give rise to a focus on the political aspects of governance. The focus on effectiveness tends to treat governance as an instrument to get things done with better results and is manifest in the donor interest in making aid delivery more effective. The focus on legitimacy encourages thinking about how things are done and leads to concerns about respect for the rule of law and how the state interacts with citizens. In the international donor community, this distinction is the difference between a results-based and a rights-based approach to development. The discussion above is graphically summarized in the figure below:

Figure 1.1. The theoretical origins of the governance concept

The continuum between effectiveness and legitimacy defines much of the space within which governance has been put into practice. It covers administrative and economic as well as political aspects of the concept. The various organizations that work in the governance field have anchored their programs somewhere along this line. Thus, as is indicated in the next figure—corporations—the World Bank and other development banks have typically built programs around such concerns as reducing transaction costs, fighting corruption, and enhancing transparency and public accountability in government agencies. At the other end of the spectrum are faithbased and secular organizations, whose objective is to promote social jus-

14

Chapter One

tice. In between these ends, bilateral and multilateral agencies have anchored their governance programs at various points.

Figure 1.2. Anchoring-points for select key actors in the governance field

In developing its more applied approach to governance, the international development community borrowed insights from its mixed intellectual heritage. Even though the emphasis has been on creating specific program interventions and assessing progress toward what is being defined as “good governance,” practice has not been completely unhinged from theory. The two interact even if this is not always in evidence. Practitioners have accepted that governance is about the use of power in the management of a country’s public affairs. This is evident in the early definitions provided, e.g., by the World Bank (1992) and the UNDP (1997). By the end of the 1990s, the donor community had become the most influential stakeholder in the governance field. Agencies had had enough time to fine-tune their instruments, and they were not hesitant to use their prescriptive devices as conditions for further aid. This donor appropriation of the concept was associated with optimism and an emphasis on quick fixes and results. Not surprisingly, given its high expectations, this euphoric period eventually came under criticism in the early 2000s. Donor agencies’ application of the governance concept had taken politics out of the picture. The main criticism in recent years has come from citizen activists and social movements, who have argued for a more transformative approach to governance. Their approach has focused on empowering people to participate in the political process by teaching citizens how to hold public officials to account for their words and deeds. A new discourse on governance has emerged centered on human rights and social development issues. This discourse tends to be driven by social movements and activist groups around the world. It has been especially significant in countries like India, Brazil, and Mexico. The various initiatives are gathered under the “social watch” rubric, some of which—like the participatory budget initiative in Porto Alegre, Brazil; the Social Audit in India; and the Citizen Tribunal in Kenya—have attracted world-wide attention

Revisiting the Study of Governance

15

As this introductory review indicates, over the past twenty years the governance field has been occupied by three different constituencies— academics, donors, and activists. As it turns out, each of these has had a chance to occupy centre stage. Each constituency has been hegemonic at some point in time, but its ability to hold the stage has been challenged by the others with their interest in furthering a different agenda. This process or trend is summarized in the following table.

Agenda setters Main approach Position adopted Principal aim

1990s

2000s

2010s

Academics

Donors

Analytical

Programmatic

Citizen Activists Empowering

Detached

Managerial

Engaged

Generating knowledge

Reforming institutions

Checking power use

Table 1.1. Shifts in the use of the governance concept The current situation in the governance field may be characterized as one of tension between a rights-based and a results-based approach. Donors lean toward the latter, even though some, like the Nordics, also have their governance programs anchored on the side of human rights. Because institutionalizing universal human rights and building democracy take time in societies where these rest on a weak foundation, the expectation of results within set timelines easily leads to disappointments. Donors have no real control of the political process that enhances human rights. At the same time, they can take comfort in the rise of civic activism, since much of that is aimed at promoting and protecting human rights, not only the civil and political but also the social and economic. Even in African countries where civil society is generally quite weak, there has been maturation toward greater strategic intervention on key issues such as land tenure, gender equality, and constitutional reform. Much of this activism has been financially supported by donors—and this still continues—but some of the strongest citizen organizations are enhancing their independence by diversifying their support base.

16

Chapter One

Challenges to Studying Governance in Africa While governance has been established as a global concern, it has become a particularly big issue in African countries. There are many reasons for this. Donor definitions of the concept, with its normative emphasis on “good” governance, have typically been perceptively quite far removed from local legacies of rule. Thus, the distance African countries have been asked to travel in order to meet donor expectations has been quite long. In many African countries, there has also been a sense that national sovereignty has been called into question by the intensity with which the international donor community has called for governance reforms. Furthermore, little attention has been paid to indigenous governance practices and their ability to serve as foundations for improvement. Local practices have typically been dismissed as outdated or problematic. The result is that the study of governance in Africa is characterized by a series of unresolved issues. There is a need to review what might be done to make governance practice better conform not only to international standards but also to local demands and capabilities. The following issues are of special interest and relevance: 1) the tension between formal and informal institutions; 2) the weight attached to the role of governance in development; 3) the types of comparison to pursue; and 4) the assessment process for democracy in a governance context.

Formal and informal institutions This is a perennial issue in the study of African politics. Because the colonial powers introduced institutions that were initially foreign to Africans, there has always been a suspicion toward “foreign” values and models. In some countries that adopted a socialist approach to development after independence, this suspicion served as a rallying call for a transformation based on African values. Some nationalists succeeded in managing this issue better than others—Julius Nyerere of Tanzania having been the most creative in his promotion of a modified model of African communalism, ujamaa. The story of how this model was implemented is associated more with pain and sorrow than with reward and happiness. The final outcome was its abandonment and the return to imposition of formal institutional models based on the historical experience of already developed countries in the West. Ever since structural adjustment in the 1980s, this has been the story of governance in Africa. The African response has been understandably sceptical if not outright hostile. For a long time Africans perceived governance reforms as being

Revisiting the Study of Governance

17

imposed. They were not part of a dialogue but were spoken to in terms of what their governments needed to do to become acceptable to Western donors. Governance conditionalities featured prominently in the late 1990s and early 2000s. Attitudes on both sides have since mellowed. Donors have toned down their rhetoric and have adopted a language referring to recipient governments as “partners” with the right to greater control over the foreign funds pumped into their treasuries. African leaders, for their part, with some notable exceptions such as Robert Mugabe in Zimbabwe, have abandoned much of their scepticism and are now more ready than before to take note of the pros and cons of adhering to global standards. A series of high-level meetings under the auspices of the Organization for Economic Cooperation and Development (OECD) in the past ten years has paved the way for this more collaborative relationship. This change in attitude notwithstanding, there is still some distance, in some cases quite significant, between constitutional and legal rules, on the one hand, and moral and social practice, on the other. In some contexts, as Helmke and Levitsky (2006) note in their study of informal institutions and democracy in Latin America, formal and informal institutions can cohabit a particular field without necessarily contradicting each other. A case in point is African elections, where the formal rules for how elections are conducted are generally accepted by the various contenders, but behaviour remains largely informal, e.g. in the form of competitive clientelism. The latter practice may encourage corrupt and fraudulent behaviour but is typically not deemed to affect the verdict, whether or not the elections were legitimate (Lindberg 2003; Basedau, Erdmann, and Mehler 2007). What is happening in the field of elections is evidence that political actors are learning and ready to change their behaviour in accordance with formal rules, but the degree of conformity varies from field to field. For instance, the independence of the judiciary and other institutions meant to check the behaviour of public officials is still undermined in most countries by executive interventions or client bribery. Even here, there is progress to report, e.g. from Kenya where the judiciary has actively taken steps to rid itself of corrupt judges and enhance its independence in accordance with the new 2010 constitution. Obtaining compliance with formal rules, especially if they have their origin in a foreign model of governance, however, remains a challenge and is taking time. This should be no surprise, but among foreign observers, donors, and analysts there is a tendency to assume that such changes in behaviour can be institutionalized in a short period of time. A relevant example is the pressure that some donors have placed on African governments to adopt international “model” legislation, e.g. on gender or child

18

Chapter One

rights, without first assessing the feasibility of such a move. For instance, the rules guiding how the government bureaucracy and the judiciary operate render such “express” legislation of little value and in some instances cause political backlash. Partly as a result of disillusion in some donor circles with the outcome of reforming governance, and partly because of the new partnership arrangements, there has been a greater willingness among the various constituents to explore the extent to which reforms can be based on existing indigenous institutions rather than those imported from outside. This has been the subject of a major research project based at the Overseas Development Institute in London—the African Power and Politics Project—in which the objective has been to examine how far neo-patrimonialist and clientelist practices are developmental (Kelsall and Booth 2010). Findings from countries like Ethiopia and Rwanda are that such practices may be effective for development provided that they are centralized. This conclusion reflects the point that Evans (2004) makes—that “institutional monocropping” may limit the options countries face in pursuing development. The idea that institutional models imported to Africa from the outside have their own limits when it comes to implementation has also received support in a more unlikely context—the World Bank—where a group of analysts have argued for adoption of a problem-driven approach to governance (Fritz, Kaiser and Levy 2009).

Role of governance in development This growing interest in the African context in which governance reforms are attempted has also raised the question of how important governance reform is in promoting economic and social development. The research focus on transition to democracy that has dominated political science research in Africa for the past two decades has had a distinct institutional orientation. Researchers have started from the assumption that human agency, policy, and institutional design will make a difference. As suggested above, some progress has been made in institutionalizing new behaviour and practice, but much less attention has been paid to how significant governance reforms are for national development. This issue has gained extra attention with the entrance of middle-income countries like China on the African development stage. China has achieved an impressive economic and social transformation in the past three decades without any notable change toward democratic governance. Western observers have not failed to consider the possible effects of the Chinese presence and have expressed fear that the gains made in “good” governance practice in

Revisiting the Study of Governance

19

African countries may be lost. This fear is not wholly unfounded, but from an African viewpoint it smacks of condescendence because it assumes that African actors lack autonomy or integrity. It is in this context that the issue has arisen about what to expect of governance. Can a set of Western norms that have increasingly become part of an international governance regime be fully introduced and implemented in other regions of the world? If not, how good does governance have to be to be acceptable in an international perspective? This set of issues was first raised by Merilee Grindle (2004; 2007) in her analysis of governance reforms in Latin America, where, at least in a U.S. perspective, these efforts were not seen to go far enough. Grindle’s conclusion from her own studies was that it is unrealistic to expect perfect or full compliance with these international norms, at least within the time frame that Western donors and governments tend to adopt. She suggested that as a result many governments will settle for “good enough” governance— and outsiders should accept this. The pursuit of best practice is sometimes misguided because it fails to generate preferred outcomes. Instead, there is a reason to consider “good fit,” i.e. measures that are suitable given the existing political and social context. Some analysts with a foot in the Western policy world are wary about this acceptance of what they consider a pragmatic, if not relativist, approach to governance. It is difficult, however, to escape the question of whether these analysts have not exaggerated the role that governance can effectively play in social and economic development. Governance measures, whether imported or local, have only touched the surface of things. As indicated above, elections and other reforms have not had a transformative effect on society. African countries are changing, not so much because of governance reforms as because of neo-liberal economic reforms and their effects. For instance, one of the more significant changes that has taken place in the past two decades is the rise of a more diversified middle class. Its membership is no longer drawn only from the state bureaucracy but also increasingly from the business and professional worlds. This raises the question of whether this emerging bourgeoisie, as Moore (1966) found in his comparative study of the path to modern democracy in the West, can champion the cause of democratic governance in Africa—or are the conditions for such a scenario not yet in place? The idea that Africans might not be ready for democracy is taboo—it was the reason colonialists gave for refusing independence in the 1950s, and the notion has never been repeated since—but it is hard to avoid the question of what the relationship between democracy and development really is. The research community is divided on the issue and continues to

20

Chapter One

argue whether good governance (or democracy) really is a cause or an effect of development (e.g. Lewis-Beck and Burkhart 1994; Przeworski et al. 2000; Bueno de Mesquita and Downs 2005). The argument by Lipset (1959) that democracy requires a level of modernization before it can be institutionalized is not wholly irrelevant in the contemporary assessment of what governance can do. Good governance is not always a straightforward causal variable. Time has come to reassess the relation between governance and development in Africa, drawing on aggregate studies like those listed above.

Types of comparisons The institutionalist focus that has characterized the study of governance and democratisation in Africa in the past two decades has brought about major changes in the way political scientists go about their business. During the 1960s and 1970s no comparativist in political science would have dared to analyse politics without a structural explanation. Today, the situation is the opposite. Nobody studies anything but institutions. The principal gain from this reorientation has been the rise of crosscountry comparisons. Elections are one arena where important contributions have been made through cross-country comparisons (Bratton and van de Walle 1997; Lindberg 2006b). Other places where cross-country comparisons have been attempted include party systems (e.g. Randall and Svasand 2002; Salih 2003; Kuenzi and Lambright 2005; LeBas 2011) and legislatures (Barkan 2009). Especially important for the emergence of cross-country comparisons in the study of African politics has been the establishment of the Afrobarometer some ten years ago. It collects data on politics and related subjects in twenty African countries and has now conducted as many as four rounds of surveys in some countries. This has allowed the growth of a new database on political behaviour that Africa largely lacked earlier. The existence of this public opinion poll has allowed Africanists to produce quantitative studies that are complementary to similar studies from other regions of the world. Africanists now have a stronger base in regional datasets to compete for publication in top-rated journals in the profession. There has been somewhat of a euphoria over this, and graduate students have been directed to do their basic research based on the data provided by the Afrobarometer. While the barometer has increasingly become a test laboratory for these students and senior scholars relying on its data for analysis, few questions have been raised about the limits of relying primarily on behavioural and institutional data. Little scrutiny, if any, has been devoted to the

Revisiting the Study of Governance

21

reliability of the data. What is more, few questions are being raised about what kind of conclusions one should draw from such studies. Opinion surveys only scratch the surface, and it is not always clear that respondents provide truthful answers. There has been a tendency to blindly believe in the “science” associated with such polls at the expense of critical scrutiny of its findings and how they were arrived at. The lure of cross-country comparisons among Africanists at this point in time is understandable, but there is also reason to embark on comparisons that use a historical or longitudinal perspective. There is nowadays enough data on various dimensions of governance to allow this to take place using quantitative data. Historical comparisons, however, can be done using qualitative methods of analysis. There are good reasons for giving more emphasis to studies that capture change over time within a given country. This approach can be combined with cross-country comparisons as, for instance, students of Latin American politics often do (e.g. Collier and Collier 1991). This approach, which may be described as a version of historical institutionalism, has the advantage of acknowledging the role of structures and how they create “critical junctures” at which governance and policy change in significant ways. It demonstrates progress over time, even if progress may not be linear or non-violent. The study of African politics would benefit from such an approach. It would bring together the often scattered research findings that we have from specialized studies—e.g. elections, political parties, or civil society—to capture the essence of what has happened in terms of democratic governance. Is there a discernible trend? If so, in which direction? How are changes in governance reflected in state inclusiveness, responsiveness, and capacity? Is there a positive correlation between more democratic governance and better state performance in terms of these three indicators? These are research questions that deserve as much attention as those focusing on individual choice and behaviour. Latin Americanists have been able to produce studies that aggregate data analysis into a “big picture” perspective. Africanists have yet to produce such studies.

Democracy in a governance context There has been a lot of confusion regarding the relationship between governance and democracy. The adoption of the notion of “good governance,” derived from norms of liberal democracy, has led to the assumption that good governance and democracy are the same thing. The two concepts, however, are different. Governance refers to how rules are institutionalized, regardless of whether they are democratic or not. In this re-

22

Chapter One

spect, governance is connected to the concept of regime—a mega concept when it comes to studying politics. If governance is understood as the politics of rules, democracy is one step down on the conceptual ladder; it is one type of regime. The notion of “democratic” governance, therefore, is a form of governance that is infused with democratic values. In a governance perspective, it can be argued that most political science research in Africa in recent years has focused on the transition to forms of democratic governance. African research has been part of a global “transitology,” i.e. the study of political transitions from military or oneparty rule. Although transitions in Africa are by no means complete, the study of the phenomenon has run into a blind alley and has largely run its course. Although transition studies have brought the study of African politics into the mainstream of comparative politics,1 these studies produced fewer and fewer insights that are of interest for further research or for policy action. The studies as a class have suffered from two shortcomings. Some scholars, such as Lindberg (2006; 2009), have tended to be overly optimistic in their conclusions about what is happening with these transitions. Although elections are important political events, their contribution to democratisation is not quite as strong as he and some other institutionalists have tended to argue. The other tendency has been to end up in a conceptual quagmire and resign to the notion that everything out there is a “hybrid.” This word has become the most common label of a broad range of countries where the analyst has run aground in terms of identifying whether a particular government is democratic or authoritarian. Levitsky and Way (2010) have tried to make sense of this tendency and add the notion of “competitive authoritarianism” to provide an analytical edge— though the edge is not particularly sharp. Hybridity also remains popular in the grey literature and among media analysts. According to the Economist Intelligence Unit, for instance, at least half a dozen African countries, as diverse as Kenya and Niger, are discussed as having hybrid regimes. One can argue that much of this transition literature has focused on strengthening the representative side of politics. Emphasis has been laid on elections and related institutions such as political parties and parliaments. There has been a tendency to assess democracy through a rather narrow governance lens, e.g. elections. An example would be the inclination to deduce from findings that a reduction in the rate of government turnovers in elections implies a decline in democracy. Not only is democracy more than that, but the transition literature has reduced the concept to what happens as a result of elections only. 1

Transitology was pronounced “over” in 2002 by Thomas Carothers (2002).

Revisiting the Study of Governance

23

Democratic governance is more than the Westminster model. This has become increasingly apparent as attention moves away from transition to current practice. With this change in research orientation has come a growing interest in studying the monitory aspects of democracy. This does not mean that the representative aspects are being ignored, but with demands for more direct citizen participation in the policy process, monitoring government policy implementation has become increasingly significant. Keane calls “monitory democracy” a post-Westminster form in which power-monitoring and power-controlling devices have begun to extend sideways and downwards through the whole political order (Keane 2009, xxvii). The concept goes beyond standard accountability mechanisms, which are part of the original model of representative democracy. Wikileaks may well be the best-known example of this new trend, but monitory governance initiatives are very much part of the political scene in many countries of the South, including those in Africa. Examples of these extra-parliamentary power-monitoring institutions include public integrity commissions, judicial activism, local courts, workplace tribunals, consensus conferences, parliaments for minorities, public interest litigation, citizens’ juries, citizens’ assemblies, independent public inquiries, think-tanks, experts’ reports, participatory budgeting, vigils, “blogging,” and other novel forms of media scrutiny. Many of these are new governance institutions that did not exist—and certainly did not have such a prominent role—in earlier periods when representative democracy was the prevalent model.

Researching Democratic Governance “Democratic governance” is what citizens and their governments do in order to make the rules of the political game acceptable and legitimate in the eyes of as many stakeholders as possible, i.e. by redefining the relations between state and society, government, and citizens in order to make the authorities more inclusive, responsive, and capable in their pursuit of development. It is about changing the nature of the public realm and the rules that make it up. Although the nature of the public realm has been an issue in the study of African politics ever since Ekeh (1975) identified the lack of respect for civic institutions as a major hurdle to African development, this question has been largely ignored in the study of democratisation. To be sure, research has identified communalism and clientelism as manifestations of political behaviour, but we have no studies that have gone so far as to analyse their effects on state institutions. The study of regime transition has

24

Chapter One

been pursued in isolation from state performance. With the decline of the transition literature and the growing interest in current political practice, the time has come to bring regime and state together in research on governance. Much of the literature to date has assumed a positive correlation between “good” governance, on the one hand, and accelerated development or enhanced security, on the other. The assumption has simply been that the more democratic the regime, as measured comparatively with other regions of the world, the more tangible results will be on the ground in African countries. This approach has overlooked the important role that the state as an intermediary plays in determining policy outcomes. The state has been the subject of attention largely through capacity-building and institutional reforms, again based on external models. Little, if any, research has been done on how changes in governance affect the state in terms of inclusiveness, responsiveness, or capacity. There is reason, therefore, to bring the relation between regime and state into focus. Bringing regime and state together in designing future research permits the growth of a new democratic governance agenda. This agenda implies a broader definition and understanding of democracy than has been used in research to date, with an accompanying narrower definition and use of the term “governance.” When we accept that “governance” refers to norms and rules (or institutions) and how they are constituted through social and political interaction between state and non-state stakeholders, the term takes on a specific meaning. This makes the concept useful not only for practitioners but for scholars and analysts as well. When governance is defined as the politics of rules, the concept refers not to the implementation of specific policies but to the normative framework—the regime—within which policies are formed and executed. Governance has both a representative and a monitory dimension. To make government possible, citizens choose their own representatives. To make governance real, the same citizens need to be alert, demanding the right to know what their representatives and government-appointed officials are doing. There is general agreement in the study of African politics that the most serious challenge to development on the continent is the lack of respect for the rules that constitute a civic public realm. This issue has become increasingly salient since the Arab Spring and the growth of demand for constitutional and legal reforms in African countries. The political battles have shifted from being merely about free and fair elections to holding public officials accountable. Corruption and other discretionary uses of power are being increasingly challenged in a growing number of countries

Revisiting the Study of Governance

25

in a manner that suggests the real challenge is getting the public realm to become more civic. Research on democracy and governance has yet to adapt to this shift in political priorities on the continent. Accountability joins voice as a key concept on this democratic governance agenda. In graphic form, the new agenda would look like this:

Figure 1.3. A framework for analysing democratic governance

This framework assumes a process approach to governance based on a simple input-output model. Each political system operates within its own regime, some more democratic than others. Democratic governance, therefore, is the infusion of norms and values that carry an international recognition as being democratic and are seen as enhancing the legitimacy of how the system operates and how goods and services are delivered. It is important to emphasize that the system refers to politics and not just policy, because governance is the framework within which policy is formulated and implemented. Furthermore, governance is not the same as government. Governance refers to the normative environment within which government operates. Governance has two dimensions, one representative (voice) and a second monitory (accountability).The issue is how these two dimensions can be further disaggregated and operationalized. One possibility is to do this with reference to political functions: articulation, aggregation, and so on.

26

Chapter One

Another is to distinguish between different arenas, e.g. civil society, political society, and so forth. Data availability becomes a critical factor in determining how to proceed. The agenda proposed here combines functions and arena approaches, as illustrated below: Representative dimension (voice) Sub-category Data source Citizen Civil Society policy input Index Electoral Election data; participation Afrobarometer Legislative Studies of policy input legislatures

Monitory dimension (accountability) Sub-category Data source Independence Global Integrity of judiciary Index Political Open Budget accountability Survey Social Studies of extraaccountability parliamentary checks

Table 1.2. Operationalizing the democratic governance agenda This is not the place for producing a full research design. The table above is only indicative of what can be done to move the political science research agenda for Africa forward. The composite measure may be tweaked in various ways, especially with regard to relevant data sources. In some instances existing sources have their own relevant indicators that can be selected for a study of democratic governance. In other cases, new indicators have to be developed. There is also merit in using qualitative sources, e.g. official reports or media accounts, to establish measures for activities that so far have been omitted in the study of democratisation in Africa. If necessary, expert panels may be used to do the assessment. Each of the three sub-categories under each dimension may be weighted equally, or some form of weighting standard could be adopted. In the end, however, it should be possible to arrive at a scale, say, of one to ten (or one to five) for each sub-category and thus an aggregate score for each country. Once a first study has been conducted and a baseline established, it would be possible to assess changes over time. This is an approach that differs significantly from the World Governance Indicators (WGI), which have been so influential in shaping the governance discourse to date. The problem with the WGI was a scientific “overreach.” Its architects believed that governance—and by extension politics at large—could be approached and assessed like economic variables. Its principal dimensions were a concoction of factors of programmatic interest to the World Bank itself. For instance, despite their obvious

Revisiting the Study of Governance

27

differences in real politics, voice and accountability were treated as one in order to fit into the Bank’s model of governance assessment. The index itself was an aggregate of surveys, panel-based assessments, and expert surveys that blended a complex set of variables into a mix that in the end was without identity and transparency. Finally, the architects strived for exactness in measuring governance that bordered on the ridiculous: minute statistical difference became the basis for often far-reaching policy conclusions. In all fairness, the authors did report their statistical margins of error, but they might as well have acknowledged from the outset that it is impossible to measure governance in such precise numbers. The agenda proposed here should avoid making the same mistake. The scientific method has its own limits when it comes to studying politics, and it is not precision so much as relevance that counts (or at least should) in the study of politics. Political scientists have a responsibility to other interested parties to convey the limits of their enterprise, not merely in terms of statistical margins of error and such technicalities but also by acknowledging the fact that the scientific method does not work in social and political environment as it does in a laboratory setting.

Conclusions Although academic theorizing in political science laid the foundation for the use of the governance concept in the past two decades, the concept has played a rather obscure role in academic research. Research has instead focused more precisely on issues of democratic transition using a “minimalist” approach to interpreting political trends. Studies of democratic transition have run their course and their explanatory value has been largely lost. The time has come to interpret democracy in its proper context, as a form of regime that has to prove its legitimacy in terms of what it delivers to citizens. The proposal for resuscitating studies of democratic governance is an attempt to move the research agenda in a relevant direction. It should make sense to scholars following political developments in Africa. In addition, this proposal speaks to the need, increasingly recognized in practitioner circles, that the concept of governance needs to be pinned down in a way that addresses the key issues of building a civic public realm where corruption is effectively fought, human rights respected, and public services delivered in a timely and effective fashion.

CHAPTER TWO DEMOCRATISATION IN AFRICA: ACHIEVEMENTS AND AGENDA MAMOUDOU GAZIBO

Introduction If democratic governance is defined as “what citizens and their governments do in order to make the rules of the political game acceptable and legitimate in the eyes of as many stakeholders as possible” (see Hyden’s chapter in this book), then democracy did not became a serious possibility in Africa until the early nineties. It is widely known that at the end of the eighties, only four African countries (Gambia, Senegal, Mauritius, and Botswana) had pluralistic regimes. The remaining sub-Saharan countries were ruled either by a military junta, a one-party regime, or a personal dictatorship (Chazan et al., 1999). This began to change at the end of 1989. On the one hand, with the end of the Cold War, African authoritarian regimes were no longer able to secure support from either the West or the Eastern bloc. In coordination with various international institutions, western democratic countries such as the U.S., France, and Great Britain began to call for African countries to open their political systems while imposing economic restrictions on those who were reluctant to do so and promising rewards to those who complied. On the other hand, this changing international environment encouraged citizens to initiate massive demonstrations, in which they demanded sound economic policies and accountable governments. Within a few years, from 1989 to 1994, virtually all sub-Saharan African nations undertook some kind of political reform. As Bratton and van de Walle (1997, 3) pointed out, “while not unfolding uniformly and to the same extent everywhere, these movements and institutional rearrangements were evident to some degree in almost all African countries.” A quarter-century has passed since the beginning of the democratisation process in Africa. This chapter seeks to evaluate what has been

Democratisation in Africa: Achievements and Agenda

29

achieved during that time and to explore the agenda ahead. My argument is threefold. First, contrary to the holistic and pessimistic view of democratisation, I identify some crucial areas to point out that significant progress has been made. In particular, this has been reflected in the development of institutions and of the administration of elections, in the greater frequency of presidential turnovers, in the growing acceptance of political and social pluralism, and even in the increased adherence to the rule of law. Second, I argue that the quality of democracy in Africa has been slowly improving due to two complementary and new types of pressures: one arising from continental and regional institutions and interventions, and the other from a growth in civic engagement. Both can be said to point toward the attainment of a more balanced relation between state and society. Third, however, I call for a nuanced view. In effect, the achievements not only vary greatly from one country to another, but important challenges remain since the governance (especially its deeper aspects) of most African regimes is still best captured by models that need to be considered “hybrid” rather than strictly democratic. Moreover, when it comes to substantive issues, such as regime responsiveness, economic performance, the redistribution of wealth, and anticorruption policies, results have been mixed at best. In the following sections, I explore the three avenues described above, i.e. the progress achieved, the fostering effect of regional institutions and civic engagements, and finally, the agenda ahead. Rather than concentrating on a specific case study, I draw examples from different countries.

Democratisation after a Quarter-Century: What Has Been Achieved? Two decades after the beginning of the democratisation process in Africa, it is clear that significant progress has been made. In particular, this has been reflected in the development of institutions and of the administration of elections, in the greater frequency of presidential turnovers, and even in the increased adherence to the rule of law. In this section, I discuss these in detail.

30

Chapter Two

The institutionalization of electoral cycles and the administration of elections There has been evidence of improved democratisation in Africa since 1990 in the growth and fortification of the continent’s institutions. During the colonial era and in the subsequent period of authoritarian regimes that dominated the continent until the 90s, institutions were strategically designed to suppress and wield power over populations (Chazan et al. 1999). Democratisation, however, has provided a new foundation for institutional development that contrasts sharply with the models that prevailed throughout most of the twentieth century (Bratton and van de Walle 1997; Lindberg 2009). Institutional analyses state that, once created, an institution “has a formidable capacity for its own reproduction across time” (Young 1994, 2) and tends to buttress its power and persevere over time. Such “dynamics of self-reinforcing or positive feedback processes in a political system” mean that “once established, patterns of political mobilization, the institutional ‘rules of the game,’ and even citizens’ basic ways of thinking about the political world will often generate self-reinforcing dynamics” (Pierson 2004, 10; Pierson 1994; Hall and Taylor 1996; Steinmo 1998). Institutions in charge of the administration of elections are very good examples of this pattern. Although elections do not guarantee democracy, particularly in Africa where most of them are marred by fraud, they are one of its crucial elements. Sub-Saharan African regimes have recorded substantial success in the development of electoral administrations in the past two decades. There is a widely held assumption that there is a correlation between the status of an electoral commission and the quality of the democratic process, which has led to continued and intensive focus in this area. An electoral management body (EMB) is defined as “an organization or body which has the sole purpose of, and is legally responsible for, managing some or all of the elements that are essential for the conduct of elections and of direct democracy instruments.… If these essential elements are allocated to various bodies, then all bodies which have a share of these essential elements can be considered as EMBs.” Essential elements include determining voter eligibility; receiving and validating nominations for parties and candidates; and polling, counting, and tabulating votes cast. Additional functions may include voter registration, boundary delimitation, voter education and information, media monitoring, and electoral dispute resolution (International IDEA 2006).

Democratisation in Africa: Achievements and Agenda

31

EMBs have emerged as the critical institution at the heart of the democratisation process in many African countries. They bear the responsibility of organizing the largest peacetime logistics undertaken in any country, often under severe conditions. Additionally, they often bear the burden of building trust and confidence in the electoral process and teaching all stakeholders the rules and norms of democratic procedures. Almost all forty-seven sub-Saharan states inherited some form of electoral system and practice from their colonial administrations. In the former French colonies in particular, EMBs were largely composed of Interior Ministries responsible for the oversight and administration of elections. Under post-colonial authoritarian regimes, there was a severe lack of confidence in state-organized elections due to several factors: a lack of structural neutrality; electoral fraud; and misuse of state resources for electoral purposes associated with government-organized elections. This is in stark contrast to many advanced democracies, including most of Western Europe and the United States, where public confidence in election management authorities has been established without the creation of an autonomous EMB. The crisis of confidence in sub-Saharan Africa has resulted in the introduction of independent EMBs as the dominant model in the region (Gazibo 2006). Independent bodies are perceived as more objective, open to oversight, and publicly accountable. To date, most sub-Saharan states have adopted some form of independent or mixed arrangement for their electoral management. EMBs may be independent, governmental, or mixed (International IDEA 2006); however, their legitimacy hinges on the degree of autonomy they enjoy. Legal autonomy is determined by three criteria (Gazibo 2006): organic separation from government (Mozaffar 2002; Diarra 2000); functional autonomy (Diarra 2000); and the extent of juridical control over the electoral process (Mozaffar 2002; Diarra 2000). Electoral bodies vary in Africa according to three criteria: x x x

Is the process unilateral (one actor prevailed) or multilateral (compromise between actors)? Who prevailed (incumbents or outsiders)? In the case of a negotiated process, did one camp dominate (Gazibo 2006, 621)?

Different models of EMBs lead to different degrees of institutional autonomy, as demonstrated in empirical studies across the continent (Gazibo 2006). These models include the consensual; the incumbent-led interactive; the outsider-led interactive; the imposed; and the managed (see Figure 2.1.).

Chapter Two

32

The popular move toward independent EMBs in Africa arose in direct response to the institutional credibility crisis; however, the link between this model and the outcomes of democratisation is still unclear. Many hybrid states have independent EMBs. This suggests that the importance of the EMB for democratisation may be trumped by outside variables. These variables could well be captured by the fact that institutions (EMBs) empower political and social actors—who mobilize to ask for free elections even when the EMB is not independent—or by institutional selfreinforcing processes. This is why in 1996 and 1999 in Niger, electors mobilized to protect the commission against the aggressions organized by the president’s militants. It is also why the 2000 elections in Senegal, organized by an EMB forged by the incumbent president Abdou Diouf, led to his defeat and a peaceful turnover. Peaceful transitions of power are indeed another phenomenon that symbolizes the advance of political governance on the continent.

Incumbents Unilateral

Incumbents Multilateral

Managed Togo Burkina, prior to 2000 Niger, 1996

Incumbent-led Interactive Ghana, 1992 Senegal, 1996 Cameroon, 2000 Consensual Niger in 2004; 2010 Benin, 2001 Ghana, 2000

Imposed Benin, 1995 Niger, 1994

Outsider-led Interactive Niger except in 1996 Benin since 1995

Unilateral Outsiders

Multilateral Outsiders

Figure 2.1: Different Models of EMB Source : Gazibo 2006, 621

The frequency of presidential turnover From the 1960s to the 1990s, the majority of African states came to independence with (purportedly) democratic institutions, democratic consti-

Democratisation in Africa: Achievements and Agenda

33

tutions, and multi-party elections; yet the bulk of these countries had low to non-existent presidential turnover, consequently falling under authoritarian rule. One-party or military rule became normalized in many states, with nationalist parties or military strongmen forming the government and destroying the opposition. In a study on leadership in Africa, Arthur Golsdmith (2000, 80) pointed out that while “Africa as a region has experienced 180 political transitions since 1960, ...until the 1990s, only eight established rulers ever retired from the top office in Africa. It is not surprising, therefore, that nearly one African leader in ten died while in office over the past forty years. The lack of retirements and the large number of natural deaths are both signs of entrenched leadership.” Contrary to the situation described by Goldsmith (2000), Bratton and van de Walle (1997, 197) found that from 1989 to 1994, in fourteen African countries, the incumbent was ousted from office peacefully during the first (founding) elections. In 1991 alone, Benin, Cape Verde, Sao Tome, and Zambia all held multi-party presidential elections during which, for the first time, a longstanding incumbent president was defeated in a free and fair contest. Throughout the 1990s, political scholars framed Africa as going through a second liberation marked by increased political reform, multiparty elections, and transitions of presidential power (Barkan 2002, 72). As reported by Bratton and van de Walle (1997), from 1992 to 1994 nine other incumbents were defeated in a regular election. Such peaceful alternations were until then unprecedented. Since then, a growing number of countries—Benin, Cape Verde, Ghana, Mauritius, Sao Tome and Principe, Senegal and Zambia—have each experienced several total turnovers (i.e., presidents coming from different political parties). Others—Botswana, Mozambique, Namibia, Comoro, South Africa and Tanzania—have experienced genuine turnovers even though the successive presidents were from the same party. Because democratic consolidation is often measured by executive turnover (Huntington 1991), the higher incidence of presidential turnover in sub-Saharan Africa during the 1990s symbolizes advances made toward democratic governance, in which presidential elections are not merely a formality to reaffirm the incumbent’s power (Barkan 2002, 73). By the turn of the twenty-first century, the only sub-Saharan African countries that had not held a multi-party election at the presidential level were the two monarchical states of Lesotho and Swaziland, plus the critically failed state of Somalia.

34

Chapter Two

However, it is crucial to note that significant variation exists among African countries and to recognize that presidential turnover in one nation does not necessarily mirror that of other countries (Conteh-Morgan 1997, 112–113). Despite progress, executive turnover has been troubled by undemocratic practices, often engaged in by incumbents who employ patronage and coercion in order to maintain power. Turnover often involves higher risk for incumbents, who (in Burundi, Congo, Rwanda, Sudan and Zimbabwe, for example) could not only lose political office but also be threatened by risks to personal security and loss of legal immunity (Levitsky and Way 2010). Despite progress, there are three categories of sub-Saharan African countries in addition to those that have experienced genuine alternation. The first category comprises seven countries whose current leader (as of 2015) has been in place continually since before 1990. These countries are Angola, Cameroon, Chad, Congo, Sudan, Uganda, and Zimbabwe. In the second category (Eritrea, Ethiopia—until the natural death of Premier Zenawi in 2012, Gambia, and Rwanda), the leader who ousted the longstanding incumbent in the early 1990s was still clinging to power in 2015. Finally, in three other countries (J. Kabila’s DRC, A. Bongo’s Gabon, and F. Gnassingbe’s ), a new phenomenon of monarchical devolution of power appeared—a passing of power from a deceased president to his son.

The growing acceptance of the rule of law Adherence to the rule of law is one of the most critical indicators of democratic governance. The strengthening of the rule of law marks an advanced stage in the transitional process of many countries, after regime change and other basic political and economic openings (Carothers 1998). It implies proper use of state power and authority, which ideally are grounded in respect for human rights and high standards of governance, with mechanisms in place to limit arbitrary or authoritative use of power. According to the UN Economic Commission for Africa (ECA), there is evidence that the African nations adhering most closely to the rule of law have less civil unrest, fewer large-scale conflicts, and generally less social disorder than those who fail to acknowledge it. This overall trend is also confirmed by the Mo Ibrahim Foundation’s annual reports (Ibrahim Index of African Governance 2014; see also previous reports). In this regard, the independent watchdog organization Freedom House defines three broad categories of countries (Gazibo 2010). First, those considered “free” by Freedom House—Namibia, Botswana, South Africa,

Democratisation in Africa: Achievements and Agenda

35

Mauritius, Ghana, and others—have demonstrated increased adherence to the rule of law, which parallels respect for human rights. These rights are cementing norms that raise the accountability of the state, assure equality in representation, and increase transparency of institutions. In the second category (those considered “partly free,” such as Nigeria and Uganda ), adherence to rule of law norms is lower. Not surprisingly, these states embody less democratisation and more human rights violations. Finally, in virtually all third category countries, those classified as “not free” by Freedom House (e.g., Sudan, DRC, Equatorial Guinea, and Zimbabwe), institutions and norms may exist, but are hardly effective. One of the greatest barriers to upholding the rule of law, despite constitutional provisions that explicitly support it, are legal mechanisms that limit it in certain situations, for example in matters of national security. These limits leads to unfortunate abuses of power. If the rule of law is defined as “a system in which the laws are public knowledge, are clear in meaning, and apply equally to everyone” (Carothers 1998), all African states have shortcomings. However, despite these limitations, there are several examples that confirm a growing relevance and acceptance of the rule of law in Africa. Benin’s Constitutional Court is one of the best examples. In Benin (classified as free), the validity of the Constitution of 1990 has never been questioned. The Beninese Constitution is comparatively clear in delimiting the competences of the Constitutional Court, particularly with regard to constitutional review and conflict arbitration between constitutional powers. The Court mainly rules “on the constitutionality of laws and regulatory measures susceptible to infringe on fundamental Human Rights and public liberties.” In those cases, the Court issues its ruling within a timeframe of eight days (article 117 and 120-2). For this purpose, the possibilities of submitting cases to the Court are extensive; any citizen may bring a suit by asking the Court for either an abstract review of a law or constitutional act or a concrete review of a controversy that s/he is engaged in (article 122). Between 1993 and 1999, the Constitutional Court of Benin issued more than four hundred rulings, mostly responding to complaints brought to the Court by individual citizens.1 Among them, there are several cases of concrete review where individuals filed a suit against regulatory measures that were argued to be contrary to the Constitution and in violation of individual human rights. But there are also cases of abstract review where individuals attacked laws or contested the constitutionality of certain acts that did not directly affect them. Prior to the es1

See verdicts of the Constitutional Court, 1991–1998.

36

Chapter Two

tablishment of the Court, the High Council of the Republic (HCR) ruled on those complaints. The nomination of members of the Constitutional Court itself figured prominently among the complaints brought to the HCR for invalidation. In November 2014, the Court confirmed its independence and commitment to safeguarding the legal order when it ruled that neither the current president, Yayi Boni, nor any other Beninese can serve more than two terms in office.

New Directions in Democracy and Governance Promotion in Africa The improving quality of democracy in Africa is due to two complementary pressures: continental and regional regulations and actions, and a general, though unequal, civic engagement. Both point toward the attainment of a more balanced relation between state and society. I discuss these in detail in this section.

Regional and continental regulations and interventions Africa has struggled to gain greater continental and regional cohesion since the 1960s. This struggle led to the creation of the Organization of African Unity in 1963 and its replacement by the African Union in 2000. In parallel, many regional, politically or economically oriented organizations were put into place (Mbaku 1997, 405; Gazibo 2010). Continental and regional integration organizations were not conceived initially to play a role in democratisation and political governance issues in Africa (see Bach in Gazibo and Thiriot 2010). Most of them were created in a period when the majority of African countries were ruled by military or one-party authoritarian regimes. Governance really became an important component of continental and regional institutions in the 2000s as the African continent collectively embraced changes that used regional regulations to guide national systems that were as yet unaccountable. Since then, the collective institutions and mechanisms devoted to good governance and democracy promotion have multiplied. Some of them have emerged in recent years: the Constitutive Act of the African Union, adopted in 2000 at the Lome Summit (in Togo) and enforced in 2001; the African Charter on Democracy, Elections and Governance, enforced in 2007; the ECOWAS2 Protocol on Democracy and Good Governance, 2

Economic Community of West African States

Democratisation in Africa: Achievements and Agenda

37

adopted in 2001; and the African Peer Review Mechanism (APRM) of the New Partnership for Africa’s development (NEPAD), adopted in 2001. The growth of norms, regulations and institutions on a supranational level is both a reflection of and a promoter of greater democratisation, as it provides overarching mechanisms for supporting national governments or keeping them in check when they go outside accepted norms of behaviour. The Constitutive Act, the African Charter, and the ECOWAS protocol all reaffirm Africa’s commitment to enhancing democratic governance and banishing unconstitutional transfers of power. As stated in Article 14 of the Charter, for example, “State Parties shall take legislative and regulatory measures to ensure that those who attempt to remove an elected government through unconstitutional means are dealt with in accordance with the law.” The ECOWAS protocol states in Article 1 that “every accession to power must be made through free, fair, and transparent elections” and calls for “zero tolerance for power obtained or maintained by unconstitutional means.” Contrary to the pessimistic view that institutions are mere facades in Africa, it is now clear that these norms and institutions not only exist, but are also effective. Military coups have not disappeared, but it is not a coincidence that they have drastically diminished in Africa. When they do occur, sanctions and pressures are applied, as in Togo (2005), Niger (2010), Mali (2012), and Guinea Bissau (2012). Regional regulations have produced interesting effects that can be illustrated in numerous situations. The first is the principle of coordination and devolution. This principle is illustrated in the way democracy issues are treated by the different levels of international institutions. Mali, where the military overthrew a democratically elected president in 2012, and Côte d’Ivoire, where the defeated president refused to quit in 2011, illustrate that regional institutions are leading actors in the democracy promotion strategy, and their decisions are systematically followed by the African Union, the European Union, and the United Nations. Second, the self-reinforcing process is illustrated in the case of the African Peer Review Mechanism (APRM). NEPAD anticipated the putting in place of an “African Peer Review Mechanism,” a system of collective controls by which a country voluntarily submits to an outside evaluation to verify that proper political and economic practices are respected. The mechanism relies on the following institutions: x

a committee made up of heads of state from participating governments—the highest level of decision making in the APRM;

Chapter Two

38 x

x x

a panel of distinguished figures who supervise the evaluation process to assure its integrity, examine the evaluation reports, and make recommendations to the committee of heads of state; the Secretariat, which provides technical and administrative support as well as coordinating the activities of the APRM; and the evaluation team, which is given the task of visiting countries and writing (and editing) progress reports, prepared in the context of the action program for each country.

Adhesion to this mechanism is voluntary. Ghana first agreed to submit to this evaluation in 2003. The goal of the evaluation was to show potential investors that the country is transparent and offers a favourable business climate. As of January 2015, eighteen countries were evaluated, and nine have signed the Memorandum of Understanding (MOU). This regional and continental management of governance goes beyond the single issue of elections or transfer of power. Pursuit of sustained peace, a precondition to viable democracy, is sometimes made possible only through regional and continental regulations, particularly in nations such as those in the Horn of Africa or West Africa where national governments and infrastructures are too weak to manage extreme and ongoing conflicts. Many African nations would not enjoy internal peace in the absence of external regulations. It is only thanks to ECOWAS that countries such as Liberia, Sierra Leone, and Côte d’Ivoire were able to end raging civil wars. Also, according to Mbaku (1995), regional regulation can reduce corruption, one of the greatest barriers to African democratisation and development. The creation and maintenance of pluralist and participatory governments across the continent have resulted in benefits from regulation on a broader level. Undoubtedly, regional regulation cannot replace the preeminence of national regulation or local initiatives that are resourcespecific. Lack of political will on a national level can be addressed and compensated for by regional intervention in only some cases, for example in weak states such as Mali, Liberia, or Togo. But it is not clear whether regional regulations and initiatives would be effective if democracy were threatened in bigger and more powerful states such as Nigeria or South Africa.

Growth in civic engagement Civic engagement is integral to democratisation for several reasons, the most prominent of which is its ability to boost public accountability (Ihonvbere 2000, 9). Accountability is causally linked to the improvement

Democratisation in Africa: Achievements and Agenda

39

of political officials’ performance; it is an antidote to corruption that needs to be increasingly present on the African continent in order for democracy to take root and be sustained. Civic engagement fills in gaps left by institutional weaknesses. African nations that are among the most fragile have a wide range of infrastructural fractures that civic engagement can fortify. In essence, while regional and continental regulatory bodies can offer support for fragile governments from above, civic engagement offers invaluable support from below (Gazibo 2008). Historically, increased civic engagement has paralleled democratisation in nations around the world, and accountability and civic engagement are inextricably bound to one another. The World Bank asserts that sociopolitical accountability is entirely reliant upon civic engagement, through which the people are able to participate in exacting governmental accountability (Sarker and Hassan 2010, 383). While many African nations are still burdened by weakness in this area, the continent is experiencing an overall progression toward more civic participation and organized civil society. Specific mechanisms for civic engagement, present to varying degrees in Africa, are public expenditure tracking, citizen monitoring of service delivery, participatory policymaking and budgeting, the boosting of awareness regarding legal rights, and citizen involvement in public commissions and advisory boards (Sarker and Hassan 2010, 384). African civic engagement is concerned with broad themes of poverty reduction, decentralization, and improvements in service delivery, as well as civic and electoral education. It is important not to forget that in many African countries, democratisation was fuelled in the early 1990s by popular movements aimed at asking for more accountable governments. Since then, depending on the country, initiatives have become vast and varied. For example, in some countries like Benin, Burkina Faso, Niger, Senegal, and Zambia, civic engagement has become a watchdog mechanism and integral to the shaping of democratisation itself. Senegal and Burkina Faso offer the most recent examples of how crucial civic engagement is if democracy is to be preserved or established. In Senegal, civic (largely youth) organizations coalesced in 2012 in the y’en a mare (it is enough) movement and, along with opposition parties, mobilized successfully against President Abdoulaye Wade’s re-election for a third term. In Burkina Faso, thanks to the diffusion of the Senegalese experience, civil society mobilized in the Balai Citoyen (“Citizens’ Broom,” a movement created by two popular musicians) in order to “sweep out” corruption and oppose the abolition of the presidential term limit. More than the political parties, it was the Balai Citoyen that was the leading force of the 2014 revolution that ousted

40

Chapter Two

Burkina Faso’s twenty-seven-year strongman Blaise Compaoré in just three days. Benin is another case illustrating the growing democratisation of African societies and the effect of civil engagement. As the Afrobarometer surveys indicate, public opinion across Africa reflects adamant support for democracy over any other type of government (Bratton and Mattes 2001b). The Beninese in particular are remarkably confident in the durability of democracy in their country. The public opinion survey conducted by Afrobarometer in 2008 (fourth round for Benin) suggests that democracy has become “the only game in town” in Benin: 91 percent of the Beninese feel they are completely free to vote for the candidate they prefer; most people know what democracy is; 83 percent prefer it to other types of regimes; and 74 percent disagree or strongly disagree with the idea that a military regime should rule their country. The fifth (2011) and sixth (2014) surveys have confirmed these trends, although a slight decrease has been observed in some areas such as the confidence in public institutions. Ideally, civic engagement ensures multiple dimensions of accountability that will eradicate the potential for corruption. The diversity of the African continent warrants the reflection of similar diversity in civic engagement, with various groups and individuals acting to ensure peace and stability along with greater democratisation (Porter et al. 2010, 32). Civic engagement ensures that the African people are jointly engaged in the development and direction of their own governments. Many democracies on the continent are still transitional or mired in hybrid practices, and individuals that forge relationships with their political and social leaders to make decisions regarding education, health care, law enforcement, and elections are effectively shaping their own democracy, co-creating a people-centred government (Bradley 2005, 542). Overall, civic engagement ensures that the African citizens are not mere recipients of change but active participants in cultivating democracy (Bradley 2005, 543). In turn, enhanced local ownership may foster commitment to and confidence in government and its ideals of equality, freedom, and universal support for human rights. African democracies, especially because of their relative youth, need civic engagement in order to validate democratic institutions and ensure that changes made genuinely support the well-being of the people. Transparency and accountability are both hallmarks of true democracies, and the emergence of civic engagement groups has thus far paralleled growth in both of these areas in Africa (Bradley 2005, 543; Wekwete and Sesay 2001, 63).

Democratisation in Africa: Achievements and Agenda

41

The Agenda Ahead The achievements presented above vary greatly from one country to another and from one region to another, and even if we evaluate democracy from a strictly procedural perspective, important challenges remain since the governance of most African regimes is still best captured by models that need to be considered “hybrid” rather than strictly democratic. Moreover, when it comes to substantive issues such as regime responsiveness, economic performance, the redistribution of wealth, and anticorruption policies, results have been mixed at best.

Escaping the institutionalization of hybrid regimes The quality of democracy is not sealed by passing the hurdle of liberalization, organizing founding elections, or even alternating heads of state. If one looks beyond purely procedural criteria, it is easy to see that the democratisation process in Africa is quite paradoxical, even in those African countries that are considered “free.” By nature, and independent of the African cases, a democratisation process is rarely linear and free of difficulties. As G. O’Donnell and P. Schmitter (1986, 3) have succinctly noted, a democratic transition means the passage from an authoritarian regime to “something else.” In order to classify post-electoral regimes (e.g. in Latin America), scholars proposed concepts such as electoral democracies, delegative democracies, quasi-democracies, ambiguous democracies, and hybrid regimes (Collier and Levitsky 1996; Diamond 2002). This is why it is not satisfactory to simply distinguish between consolidated democracies and all other types of government. Rather, democracy needs to be conceived as a system of interdependent partial regimes: the electoral regime, political rights, civil liberties, separation of powers and correlative horizontal responsibility, and finally, the effective power to govern (Merkel 2004). These then need to be examined for flaws. African regimes are hybrid in the sense that almost all of them exhibit both democratic and authoritarian elements, but they do not all suffer from the same kind of dysfunctional regime. Whereas the electoral regime allows clear distinction between electoral democracies and non-democracies, the other regimes show a greater degree of hybridism in most countries. Merkel (2004) identifies three situations of hybridism, distinguished by the type of dysfunction that applies in Africa: domain democracies, exclusive democracies, and delegative democracies. A number of African countries may, at best, be labelled “domain democracies,” as the elected officials do not possess the effective power to

42

Chapter Two

govern. In Africa, two types of guardianship limit the power of elected representatives to govern, thereby also limiting the exercise of democracy. The first is related to the armed forces. Since the military is one of the few highly organized groups in many African countries, and since many African Presidents (M. Ould Abel Aziz in Mauritania, P. Kagamé in Rwanda, D. Sassou Ngesso in Congo, J. Kabila in RDC, I. Deby in Chad, etc.) are former military officers or guerrilla fighters, the military’s hold on power is significant. This diminishes the credibility of the representative system. The second type of guardianship is related to international financial institutions. Since the mid-1980s, these institutions have been the main decision-makers in economic policy for the majority of African countries. The result is a growing standardization of economic and social policies in the form of structural adjustment programmes, governance programmes, and more recently, strategy papers for poverty reduction and growth. In addition, many African countries are not characterized by genuine democratic governance but are what Merkel calls “exclusive democracies.” In some of these countries, rulers do not hesitate to arrest opponents to the regime, to harass voters, and to grossly rig the electoral process, notably by resorting to recurrent constitutional modifications in order to suppress limits to the presidential terms of office. In other countries, incumbents employ more subtle techniques of fraud: they resort to multiple registration of pro-regime voters, a division of constituencies favourable to incumbents, or the misuse of government funds to partisan ends. It is therefore not surprising that the election results of those countries are always contested by the losers—a clear signal of the defectiveness of the electoral regime. Only those African countries that are classified as “free” have an electoral regime that approximates the ideal of political competition and popular participation. Finally, the majority of African countries display characteristics unique to delegative democracies in which, once in place, elected representatives behave as if they had a blank check and feel no responsibility to their electors. Furthermore, these countries are characterized by a hypertrophy of executive power that also controls legislative and judicial power. This situation prevents the mechanisms of horizontal control and checks and balances. In numerous African countries, the executive branch transforms legislative power into a simple rubber stamp chamber and judicial authority into a means of eliminating opponents. In addition to these defaults, which exist to varying degrees in practically all African countries, there is a common characteristic limiting the scope of democracy in Africa: that is, the persistence of neo-patrimonial practices. Even in countries that have succeeded relatively well in democ-

Democratisation in Africa: Achievements and Agenda

43

ratisation, corruption and clientelism persist (Gazibo and Bach 2012). Some authors argue that, as African countries democratize, clientelism transforms into mass clientelism (different from corruption), and neopatrimonialism will recede (van de Walle 2012). However, in the case of exacerbated neo-patrimonialism, instead of permitting a peaceful resolution of the struggle for power, democratisation transforms itself into a new factor responsible for some of the conflicts that it is supposed to prevent because, in such cases, democracy is inhibited from evolving important aspects such as economic and social rights.

Substantive issues When it comes to the more substantial issues of economic rights, transparency, and accountability, the road toward sound democracies in Africa is hard. For example, Benin is a robust democracy from a formal procedural stand, yet the substance of that democracy is weak. This situation is confirmed by the 2008 Afrobarometer survey, which shows that 62 percent of the Beninese consider the economic situation in their country to be bad or very bad, and 60 percent think the same of their living conditions. Seventy-five percent believe that their living conditions are at least shared by their compatriots, and 64 percent think that the economic policies of the government have been harmful to most of the population and have benefited only a minority of persons. This perception did not improve in the fifth (2011) and sixth (2014) surveys. The 2008 survey also demonstrates that 90 percent of respondents have never contacted a member of parliament and 95 percent have never contacted a member of the Executive because they do not believe in the accountability of officials’ responsiveness. The same can be said about the perception of corruption in the country. This is not unique to Benin. According to Harsch (2006, 16), “A series of public opinion surveys conducted in 2002–03 in fifteen African countries by African and U.S. researchers found that a large majority of people believe conditions are worsening. When asked by the researchers of the Afrobarometer project whether their governments’ economic policies were hurting most people and benefiting only a few, 61 percent of the more than 23 thousand people surveyed agreed. In only three of the fifteen countries (Lesotho, Mozambique, and Namibia) did less than half the respondents think so.” Harsch also warned that Mr. Branko Milanovic, a researcher at the World Bank, reported that in the 1960s the “Gini coefficient”—a common measure of income inequality—was on average ten points higher in Africa than in the rest of the world combined. That difference dimin-

44

Chapter Two

ished in the 1980s, but then in the 1990s “the gap opened up again.” In post-Apartheid South Africa, for example, development has focused on economic growth in various ways (FDI, institutional ministries, etc.), but not in the most important way; as a result, the gaps between the needy and the wealthy have widened (Ndedi 2011). Thus, any analysis of democracy and development in Africa over the past twenty years must account for a serious caveat. Distribution of the benefits of growth (Africa’s GDP has grown by about 5 percent annually since the 2000s) and development have not benefited Africans equally; indeed, many of the neediest populations have seen few of the benefits of development and democracy over the past quarter-century. In addition, corruption has become one of the most serious threats to the introduction and entrenching of democratic governance in Africa, at all levels of state administration. The links between corruption, underdevelopment, and political governance have been established (Kututwa 2005), making the issue of corruption a challenge for progress in multiple areas. In Africa, natural resource wealth has enmeshed many elites and international business actors in corrupt dealings. International aid has been diverted for inappropriate purposes by local state, militia, and nongovernmental actors, just as the aid agencies themselves have not been immune to internal corruption. Civil servants are severely underpaid and often resort to bribes, a practice that has become especially prone to abuse among the army and police. At best, institutions designed to deal with the issue are under-resourced. At worst, they are implicit in the corruption (Transparency International n.d.). Significant strides have been made in this area in the latest wave of democratisation. Most African states have adopted some form of anticorruption legislation reflecting regional and international standards, particularly the African Union Convention on Corruption. Nonetheless, substantial work remains to be done in the areas of reform and accountability of political leadership. Moreover, since Aristotle (Politics), it has been widely recognized that economic development and wealth help democracy endure: they create a large enough middle class to serve as the social basis of democracy and lead to the emergence of new political values (such as enhanced sense of individuality, personal autonomy, and the value of personal freedom and choice) that support democratic institutions and practices (Lipset 1959; Pei 1999; Przeworski 2000). Governance in Africa will improve only if more progress is made on these critical substantive issues.

Democratisation in Africa: Achievements and Agenda

45

Conclusion At the beginning of the 1990s, the scope and speed of the democratic development phenomenon in Africa led many authors to believe that these transformations were going to result in an effortless success. A quartercentury later, however, a number of African countries find themselves between banks of the river, and it is impossible to tell, at least in the short term, whether or not they will move on toward democracy (Herbst 2001). Jean-Germain Gros was convinced long ago (1998) that in no African country, including South Africa, is democracy irreversible. There are some facts that support this cautious assessment. First, many of the countries that seemed to be on the right track, such as Niger and Mali, have recently experienced some serious setbacks. In Mali, the 2012 coup put an end to twenty years of a democratic regime that many observers thought consolidated, and in Niger, another coup overthrew former president Tandja, who refused to quit after his constitutional second term. Even though the two countries have held democratic presidential elections since then, the coups have shown that democracy should not be taken for granted. Second, in several other countries, incumbents have eroded the gains made in the 90s via constitutional reforms aimed at helping them maintain their grip on power. They typically suppress the two-term limit and adopt a first-past-the-post electoral law. Some have succeeded (in Chad, Cameroon, Algeria, and Togo) and others have (as of 2015) firmly decided to try (Rwanda, Burundi, Congo). Third, there are countries where the mere pluralism (Swaziland, Eritrea) or consensus on rules of the political game (Egypt, Libya, Sudan, and South Sudan) is hard to foresee. Certainly these situations show the ambivalent character of democratisation and the difficulty of making it the founding block from which the crisis of the state and governance can end, and from which the continent can be stabilized and its renaissance ignited. Nevertheless, it is important to highlight the progress that has been made during the past twenty-five years. First, it is important to note that, in Africa, those countries that have advanced toward democracy have done so more quickly than countries in other regions. Some of these countries, such as South Africa, Ghana, or Benin, have become a sample of what Africa can accomplish politically despite its dramatic political background. These examples can become the motors of a political revival in Africa if they manage to spread their current norms and values. Second, Africa today is very different because the regional and international contexts have changed and are much more favourable to the

46

Chapter Two

achievement of democratic gains. The reframing of regional organizations’ agendas toward a greater promotion of governance in terms of laws, procedures, and actions, makes it more and more difficult for authoritarian cliques to maintain the kind of domination that prevailed until the 1990s. For example, after the military coups in Mali (2012) and Central African Republic (2013), it was pressure from the regional organizations that forced the military to step down. Despite many weaknesses, the ECOWAS protocol and the African Union’s non-tolerance for non-constitutional change have transformed democracy into a regional priority and produced some effects. Third, institutional diversification and autonomy have increased in many African countries. This is important because institutionalization is a key component of sound democracy and non-arbitrary politics. The way electoral commissions and Constitutional courts manage political competition and juridical conflicts in many countries, and the compliance of political actors, all signal that progress has been made and needs to be further encouraged. Finally, and perhaps more importantly, African societies have changed and have become generally more democratic than their governments. Even in countries labelled “not free” by Freedom House, one can witness a growing engagement on the part of ordinary citizens and civil society organizations in demanding legitimate and responsive government. All these factors are signs of a rebalancing of relations between society and the state in Africa.

CHAPTER THREE THE EXPANSION OF JUDICIAL POWER IN AFRICA AND DEMOCRATIC CONSOLIDATION: OPPORTUNITIES, CHALLENGES AND FUTURE PROSPECTS CHARLES MANGA FOMBAD

Introduction The global tidal wave of democratisation that reached African shores in the 1990s opened up new vistas for constitutionalism on the continent generally and has had a tremendous impact on the role of the judiciary. Because of its potential to supervise and control the exercise of power by the other two arms of government, the judiciary plays a fairly unique role in any society governed by the rule of law. In fact, any genuine and fully functional democracy must be backed by an efficient and independent judiciary capable of acting as a bulwark against abuse of power, authoritarianism, and arbitrariness. Prior to the 1990s, African judicial institutions were weak and substantially compromised because of their close association with dictatorial regimes in power. As a result, they could hardly play the important role of guardians of the constitution, protectors of human rights, and impartial enforcers of the rule of law. The constitutional reforms of the past two decades have tried to introduce fully functional and independent judicial institutions (Tate and Vallinder 1995). Effective independent judicial institutions are not only necessary for the existence and sustenance of constitutionalism and democratic governance but are also vital for its consolidation. This is particularly true at a critical time like the present, when democratic transition is increasingly threatened by ominous signs of a reversal. This chapter seeks to address three main issues: first, whether African judicial institutions are well positioned to play an effective role in promoting constitutional justice and democratic consolidation; second, how they

48

Chapter Three

have performed since the reforms that expanded their jurisdiction; and third, what the prospects for the future are. The first part of the chapter will examine the attempts to establish independent judiciaries and the expansion of constitutional justice through judicial review. The second part will consider the opportunities that have arisen for the judiciary to influence the course of democratic transition. The third will discuss the challenges that African judicial institutions face and assess the prospects for the future. The chapter will show that on the whole, African judicial institutions have had a hard time contributing in an effective manner to sustaining the democratisation process. This is expected to be more difficult in the future because of increasing political interference in various ways in the operation of judicial institutions. As a result of the critical role these judicial institutions play in democratic development, we will, in the light of recent experiences, suggest some of the ways in which their performances can be enhanced. In the final analysis, the reality today is that the threats to Africa’s fragile democracies are so strong that unless the courts are willing to boldly and assertively play their role as guardians and protectors of the constitution and constitutionalism, forces bent on undermining the rule of law and democracy might move the continent back into the dark turbulent days of authoritarianism.

The Growth of Judicial Power in Africa Reform of judicial institutions in Africa came about as part of the general process of constitutional renewal that started in the 1990s and has continued in the most recent constitutions, such as the 2010 constitutions of Angola and Kenya. It is likely that other constitutions which are currently under review such as the constitutions of Tanzania, Tunisia, and Zambia or even the first constitution of the youngest African country, the Republic of South Sudan, will follow this trend. It has been marked by two important developments; first the move to make judicial institutions more independent and second, the expansion of judicial review. These two developments will now be examined.

The establishment of independent judiciaries It is necessary to preface this discussion by pointing out that the concept of judicial independence is a relative—not an absolute—concept. As such, it does not refer to a single kind of relationship or something that a judicial system “has” or “does not have,” but rather what it may have more or less of (Asian Development Bank 2001). Although there is no general

Expansion of Judicial Power in Africa and Democratic Consolidation

49

agreement amongst scholars on exactly what is meant by judicial independence (Larkins 1996), most of them do agree that an independent judiciary should have at least four characteristics: first, that it is impartial; second, that its decisions are accepted by all; third, that it is free from undue influence; and fourth, that it is capable of rendering justice on all issues of substantial legal and constitutional importance (611). From this perspective, an independent judiciary can be defined as one that is free to render justice on all issues of substantial legal and constitutional importance fairly, impartially, in accordance with the law, and without threat, fear of reprisal, intimidation, or any other undue influence or consideration. Various governmental, non-governmental, regional, and sub-regional organizations have produced declarations or statements putting forward what they consider as the basic elements of an independent judiciary (Autheman 2004). The documents containing these declarations are not binding but nevertheless provide evidence of a high level of support for what may be regarded as certain universally agreed upon core elements of judicial independence. On the basis of these diverse declarations and statements, the core elements of judicial independence that seem to be internationally recognized can be summarized as follows: x x x x x x

institutional arrangements for judicial autonomy; financial arrangements for judicial autonomy; arrangements for the security of the judicial office; adequate remuneration of judicial officers; transparent mechanism for judicial appointments; and judicial accountability.1

Two studies, one general (Fombad 2007)2 and the other focused on southern Africa (Van de Vijver 2006), have discussed the prospects for judicial independence under post-1990 African constitutions. These studies, especially the 2007 study, show that although the new and revised constitutions went to great lengths to provide for independent judiciaries, there were significant differences in the approach adopted by Anglophone and Francophone countries. From these studies, it is clear that there is considerable variation in the degree and scope for judicial independence, with

1

These are fully discussed in Dung 2003. For a full discussion of all these elements of an independent judiciary in Africa, see Fombad 2007, 233–257. This section draws largely from this study.

2

50

Chapter Three

the prospects for this being better in Anglophone countries than in Francophone and Lusophone countries. At independence, two main Western approaches to judicial independence were received and adopted in Africa: 1) the British Westminster model, modified with some ideas from the American system (hereinafter referred to as the Common law model); and 2) and the French Fifth Republic model (herein after referred to as Civil law model). The former has been adopted by Anglophone countries, while the latter has been adopted by Francophone countries, with variants of this based on the Portuguese constitution adopted in Lusophone countries. It is necessary, albeit briefly, to say something about the British and French models, both archetypical of the Common Law and Civil Law approaches respectively, that have been widely copied on the continent.

The Expansion of the Scope for Constitutional Justice One of the strongest indications of Africa’s commitment to constitutionalism in the past two decades has been the attempts in many countries, especially Francophone and Lusophone, to provide for a system of judicial review in one form or another. Before the 1990s, constitutional justice was a vain hope because violations of constitutions with impunity were the norm rather than the exception. Generally speaking, two distinct models of constitutional review separated by deep theoretical differences have dominated the world, viz, the decentralised or what is often referred to as the American model, and the centralised or Austrian model. There are numerous variations within these two models. These two main models of constitutional review were received in Africa, but with time, they have taken on several peculiarities worth noting. From an analysis of the current situation in Africa, it is possible to say that three main models of constitutional adjudication operate. Two of these, namely the centralised model and the decentralised model were received during the colonial period. The third, the mixed model, combines elements of both the centralised and decentralised system. It should also be noted that some constitutions do not provide any mechanism for constitutional adjudication.3 3

An example of this is the Libyan constitution of 1969, where the only provisions dealing with the judiciary are fairly obscure. For example, article 28 states that “judges shall be independent” and that “in the exercise of their functions, they shall be free from any authority except that of the law and their conscience”; while article 30 is content to state that “everyone has the right to resort to the courts in accordance with the law.” It should, however, be noted that since the overthrow

Expansion of Judicial Power in Africa and Democratic Consolidation

51

The decentralised, concrete, dispersed, or diffused model is often referred to as the American model because it is usually traced to the famous judgment of Chief Justice Marshall in the 1803 case of Marbury v Madison (U.S. Supreme Court 1803). A good number of African countries, almost all of which are Anglophone,4 have in quite diverse ways adopted the American model. In these countries, constitutional matters are handled by the ordinary courts, although certain cases, such as disputes dealing with the interpretation and application of the bill of rights, may be dealt with only by the superior courts. The second model is the centralised or concentrated model, sometimes referred to as the Austrian model. This was widely adopted in Europe, and the French have developed a particular form of it—the Constitutional Council system—which has been replicated very widely in Francophone and, to a certain extent, Lusophone and Hispanophone Africa. A good number of these countries, such as Equatorial Guinea,5 Cameroon, 6 DR Congo7 and Senegal8 have retained the Constitutional Council model in its classic 1958 French Fifth Republic form, with all its defects and weaknesses. Others, such as Angola,9 Benin,10 Cape Verde,11 Egypt,12 Eritrea,13 Gabon,14 Mali,15 Mozambique,16 Republic of Congo,17 Sudan,18 and Tuniand death of the former Libyan strong man, Colonel Muammar Gadhafi in 2011, a constituent assembly was elected in July 2012 with a mandate to draft a new constitution. 4 See sections 18 and 95 of the constitution of Botswana, sections 22 and 119 of the constitution of Lesotho, sections 11 and 108 of the constitution of Malawi, sections 83–84 of the constitution of Mauritius, articles 25 and 80 of the constitution of Namibia, section 251(1)(9) of the constitution of Nigeria, section 24 of the constitution of Zimbabwe, and sections 127 and 131 of the constitution of Gambia. 5 See Items 94–97 of the constitution of 1991. 6 See articles 46–52 of the constitution of 1996. Although the law introducing the constitution gives the impression that it is merely an amendment to the 1972 constitution, the declared intention until this constitution was promulgated into law had been to make a new constitution; and in fact the number of articles doubled. 7 See articles 157–169 of the constitution of 2005. 8 See articles 74–75 and 89–90 constitution of 2001. 9 See articles 134–135 of the constitution of 1992. 10 See articles 114–124 of the constitution of 1990. 11 See articles 229–237 of the constitution of 1992. 12 See articles 174–178 of the constitution of Egypt 1992. 13 See articles 28–29 and 49 of the constitution of 2001. 14 See articles 83–93 of the constitution of 1991. 15 See articles 85–91 of the constitution of 1991. 16 See articles 180–184 of the constitution of 1990.

52

Chapter Three

sia,19 only retained it after making some slight, or in the case of Benin, substantial modifications to the original design.20 The third category consists of a group of states where the constitution provides for a mixed or hybrid system of constitutional adjudication. Thus, Ethiopia’s system is a mix of judicial and non-judicial mechanisms;21 in Ghana certain matters may be handled by the High Court and others by the Supreme Court;22 South Africa has a mix of centralised and decentralised systems;23 in Tanzania, most constitutional adjudication is handled by the High Court, but certain specified matters are handled by a Special Constitutional Court;24 and in Zambia, concrete review is handled by the High Court.25 One can talk today of the expansion of constitutional justice in Africa mainly because of the developments in some Francophone, Lusophone, and Hispanophone African countries. Two are of particular importance: first, the improvement in the process of judicial appointments, and second, changes related to the powers given to judicial bodies. Under the system of constitutional review practised in the Anglophone countries, the qualifications and manner of appointment of judges is rigorous. In the countries that follow the French model, the appointment system is fundamentally flawed because the appointments are made exclusively by politicians and those appointed need not necessarily be judges. Many revised Francophone constitutions have now changed this.26 Although appointments are

17

See articles 138–151 of the constitution of 1992. See article 105 of the constitution of 1998. 19 See articles 73–75 of the constitution of 1959. 20 The most substantial changes were introduced by the Benin constitution of 1990. 21 See articles 82–84 of the constitution of 1994. 22 See articles 33 and 130 of the constitution of 1992. 23 See sections 38 and 167 of the constitution of 1996. 24 See articles 30 and 125–128 of the constitution. 25 See articles 28 and 28 of the constitution of 1991. 26 For example, article 91 of the Malian constitution states: “The Constitutional Court shall comprise of nine members who hold the title of Counsellors with periods of office of seven years, renewable once. The nine members of the Constitutional Council are appointed in the following manner: x three named by the President of the Republic of which two must be jurists; x three named by the President of the National Assembly of whom two must be jurists; x three Magistrates designated by the Supreme Council of Magistracy. 18

Expansion of Judicial Power in Africa and Democratic Consolidation

53

still made by politicians, their scope for manoeuvre is limited because the constitutions try to ensure—by laying down criteria specifying certain minimum qualifications that appointees must possess—that the choice is not based solely on political considerations. In most cases, it is now clearly spelled out that the appointees must be jurists.27 The second and equally significant change relates to the powers given to these bodies. Unlike in the past where they were restricted merely to dealing with prepromulgation abstract review of legislation, the constitutions now allow for both abstract and concrete review. An example of an expanded scope of constitutional review is provided for under the new Angolan constitution of 2010, which gives its Constitutional Court the power to review the validity of laws and acts of the state, public administration and local authorities,28 abstract review,29 concrete review30 and the novel remedy of declaration of unconstitutionality by omission.31 Many of the constitutions have also given these courts jurisdiction to entertain electoral disputes. Even though many of the locus standi rules limit access to these courts only to certain politicians, who would normally have no reason to challenge any legislation,32 this is mitigated in some circumstances by the right given to citizens to raise the defence of unconstitutionality of any legislation in the course of any legal proceedings.33 Nevertheless, the fact that these are specialised courts—isolated in the capital city, detached from the The Counsellors are chosen from Professors of law, lawyers and magistrates having at least fifteen years of practice, in addition to qualified personalities who have served the state honourably.” 27 Also see article 180(2) of the Angolan constitution, article 159 Central African Republic constitution and article 89 of the constitution of Gabon. But see article 158, which states that at least two-thirds of the members of the Constitutional Council must be jurists from among the judiciary, the bar, and academia. 28 See articles 226 of the constitution. 29 See articles 228–229 ibid. 30 See articles 230–231 ibid. 31 See article 232 ibid. However, the value of this is considerably diminished by article 230(2), which limits access to the court only to a restricted group. 32 Those who could seize the council were restricted to the President of the Republic, the Presidents of the National Assembly, the President of the Senate, and a certain minimum number of parliamentarians. This would often be the very people who were so closely related to the legislation before parliament that they would have little interest in challenging it. 33 See for example, article 166 of the constitution of the Central African Republic, article 86 of the constitution of Chad, article 162 of the constitution of DR Congo and article 2(a) of the constitution of Equatorial Guinea. x

54

Chapter Three

other courts and closely associated in very many respects with the political class—limits their ability to act as an objective and independent institution for promoting constitutionalism, the rule of law and good governance. This also reinforces the view of some cynics who argue that the French Constitutional Council has worked reasonably well by default, not design.34 The question that arises from this is whether the expansion of the scope for constitutional justice has actually seen an effective improvement in the quality of constitutional justice, particularly with respect to the standards of democratic governance. Have the courts taken the opportunities that have been presented to them to influence the pace and nature of the political and constitutional transformation process on the continent?

Opportunities to Consolidate Democracy and Promote Constitutional Justice If the African judiciary could be considered to have been one of the victims of the long years of totalitarianism that enveloped the continent soon after independence, the post-1990 constitutional reforms have provided it with an opportunity to redeem itself. It is clear from the preceding discussion that the new framework is far from perfect. Before considering what needs to be done to improve the standards and the performance of these institutions, it is necessary to pause here and see how they have performed when confronted with issues of democracy and constitutional justice. This brief overview is designed to illustrate the important role that independent and effective judiciaries can play in entrenching constitutionalism, good governance, respect for human rights and democracy. Three particular aspects of democratic consolidation and constitutional justice will be discussed, viz, the settlement of electoral disputes, the curbing of executive lawlessness, and the promotion of socio-economic rights.

34

It is also worth noting that, from its creation in 1958, the French Constitutional Council did almost nothing until 1974, when Parliament modified the Constitution to change the rules for bringing matters before it. As a result of the change, any faction or party able to get sixty members of either chamber of Parliament could request a review of proposed legislation. The rate of referral of matters increased dramatically as minority coalitions for the first time discovered that they could block or delay new initiatives by referring legislation to the Council (Provine 1996).

Expansion of Judicial Power in Africa and Democratic Consolidation

55

Dealing with Electoral Disputes Rightly or wrongly, elections have become the most visible signs of Africa’s transition from one-party dictatorship to multi-party democracy.35 These elections have also become the greatest threat to Africa’s faltering democratic transition. The experiences over the past two decades amply show how elections can in many instances promote democratisation and in others provide an anchor for autocratisation. The latter has been true in far too many instances in Africa.36 Frequent and blatant electoral malpractices by the dominant parties that have progressively replaced the one-party systems have transformed elections into a potential source of fragmentation and conflict. In countries such as Kenya and Zimbabwe, post-electoral conflicts have resulted in violence, deaths and destruction of property. Both incumbents and opposition parties are usually not ready to concede defeat, and very often incumbents will stop at nothing in their attempts to entrench themselves or their parties in office in perpetuity. These conflicts are a manifestation of growing disillusionment by voters with the ballot box as the best means to express their choice of who will govern them. Generally, the judiciary in constitutional democracies plays a very important role in securing the integrity of elections as the main channel of peaceful and democratic change (Gloppen 2006). Many African constitutions empower the courts—especially constitutional courts—to play a role in the electoral process. At least three patterns can be discerned. The first appears in the constitutions of Anglophone states. These usually contain provisions that reserve any disputes concerning the violation of the bill of rights, which regulates elections inter alia, to special courts—usually the High Court.37 In some of these countries however, special election tribu35 Although it is necessary to point out that while an election constitutes a key ingredient of democracy, it is not in or of itself synonymous with democracy. 36 As Lindberg (2009, 86) puts it, elections make democratisation more likely if they serve to make repression expensive and counterproductive, spurring the opposition to unify and mobilize; democratisation is also more likely where a policy of tolerance to the opposition seems to legitimize rulers but in fact triggers defections of state actors and creates self-fulfilling expectations about the continuation of competitive politics. On the other hand, elections make autocratisation more likely if they serve to make repression of opposition leaders cheap and easy, or even unnecessary; and if they make it possible for the regime to control tolerance of the opposition, to split the opposition, and to use elections as a vehicle for patronage; or if elections simply make tolerance too costly for the incumbents. 37 See article 33(1), constitution of Ghana, section 22(1) constitution of Lesotho, section 46(2) constitution of Malawi, and section 84(3) constitution of Kenya. This

56

Chapter Three

nals are provided for to deal with elections.38 The second pattern, which is common in those countries that have adopted the French Constitutional Council model, is that the latter is given the power to review and proclaim election results. For example, article 152 of the constitution of Burkina Faso states that the Constitutional Council “shall monitor the legality, transparency and impartiality of referenda and presidential and legislative elections, and shall be judge of electoral litigations. It shall announce the final results of presidential, legislative and local elections.”39 A third pattern is that of constitutional court having the power only to proclaim the results. An example of this is item 94 (c) of the constitution of Equatorial Guinea, which simply states as one of its functions, “proclaiming the final results of presidential, municipal and legislative elections and referenda.”40 However, even under those constitutions where no specific provision addresses the modalities for settling election disputes, legislation may refer such matters to the ordinary courts or set up special election tribunals. From an analysis of the constitutional framework in Africa, judicial intervention may occur in three ways. First, intervention may be used to resolve disputes over the basic rules to ensure that they conform to the constitution and provide an even playing field. Secondly, intervention may ensure that the electoral rules have been complied with by all the parties and have led to a free and fair outcome. This is probably the most important role of the court. Thirdly, the court may intervene merely to formally declare the results of the elections. This is the most controversial role because the court is reduced to a channel for legitimising election results regardless of any irregularities that may have occurred. It is rare that courts are restricted to playing only this role but as noted above, that is the role that the constitution of Equatorial Guinea has conferred on its Constitutional Council. However, most African constitutions empower the also appears in the constitutions of non-Anglophone African states, for example in section 83(2) of the constitution of Mauritius, and in section 28 of the constitution of Eritrea. A similar right, but differently worded, appears in article 105(b)(c) of the constitution of Sudan, which states that the court shall decide upon claims by aggrieved persons for the protection of liberties and rights which are guaranteed by the constitution. 38 See in the case of Nigeria, sections 239 and 285 of the constitution. 39 See also article 48(1) of the constitution of Cameroon; article 238 of the constitution of Cape Verde; articles 143–145 of the constitution of DR Congo; and article 103 of the constitution of Mali. 40 The obscure language of article 163 of the constitution of Algeria could be read to have this effect too.

Expansion of Judicial Power in Africa and Democratic Consolidation

57

courts to play the first two roles. But much as judicial intervention in most instances provides the courts with an opportunity to ensure that the outcome of the elections reflects the will of the people and that the elections were free and fair, the judicial processes could also be abused and do enormous damage to the course of democracy. Confidence by Africans in the ability of courts to settle election disputes in an honest and fair manner is not very high. In a Gallup survey of 19 African countries in 2006, only 33 percent of those interviewed said that they had confidence in the honesty of elections in their countries while 45 percent said they have confidence in the judicial systems and courts (Tortora 2007). Although the confidence in elections had risen to 41 percent in a 2011 survey, distrust of courts remains the highest amongst national institutions (Rheault and Tortora 2011). In the 2007 controversial elections in Kenya, the main opposition party—the Orange Democratic Movement (ODM) whose leader, Raila Odinga, was believed by most international election observers to have won the presidential elections— refused to take its grievances to what they claimed were “Kibaki’s courts” (Abuya 2010). The ODM argued that the courts were instruments of the state that could not objectively adjudicate any petition involving a sitting president (152–153). The disputed election results led to the violence in which more than fifteen hundred people lost their lives. As will shortly be shown, the pre-2010 Kenyan judiciary has an unenviable record for corruption, incompetence and deference to the executive.41 A number of recent studies have tried to review the performance of African courts in dealing with election disputes. These studies have covered cases from courts in Francophone and Anglophone courts and from countries at different stages in the transition to democracy.42 While it would be inaccurate to draw any definitive conclusions from data analysing election cases in only a few countries, these provide a useful indication of what the courts appear to be doing. A brief review of the outcome of some election cases will be instructive. We will start with presidential elections, which are usually the most hotly contested. In the first, Awolowo v Shagari and FEDECO,43 the elec41 Nigerian presidential candidate Awolowo refused to go to the courts for a second time and Ugandan presidential candidate Besigye did the same after two previous failed attempts, both of which are discussed later in this chapter. 42 The main studies looked at are Ugochukwu and Ononiwu 2000; Fombad 2011a; Abuya 2010, 122–164; Van de Vijver 2006; Oyebode 2006; Gloppen 2007; and Gloppen 2004. 43 1979. NSCC 87.

58

Chapter Three

tion law specified that the winner of the presidential election must record the highest number of popular votes cast and also score 25 percent or more of the votes cast in at least two-thirds of the country’s nineteen states. Shagari won the highest popular votes but scored 25 percent of total votes cast in only twelve states. The Nigerian Supreme Court had to decide what two-thirds of nineteen states meant and held that scoring 25 percent in twelve states was a “substantial compliance” with the law. And for good measure, the court warned that its decision was not to be taken as setting a precedent. This approach was nevertheless followed in Buhari v Obasanjo.44 In this case, although the Supreme Court found that there had been numerous irregularities in the way Obasanjo had been re-elected, it concluded that these did not matter because the elections were conducted in substantial compliance with the principles of the Electoral Act. Similar conclusions were reached in two Ugandan cases: Besigye v Museveni45 in 2001 and Besigye v The EC and Museveni46 in 2007. In both cases, the Supreme Court was very critical of the way the elections had been conducted, particularly the use of security forces that were alleged to have committed acts of intimidation, violence, and partisan harassment and the massive disenfranchisement of voters by deleting their names from voter registers and the partial conduct of electoral officials. However, the court by a majority held that the non-compliance with the provisions and principles of the electoral law did not affect the results of the presidential election in a substantial manner. Once again, the Ugandan courts, very much like the Nigerian courts, were happy to note irregularities but concluded that these were not sufficient to have affected the final results. In Cameroon’s controversial presidential elections of 1992, which most international observers believe was won by the leader of the opposition party, the Cameroon Supreme Court drew attention to numerous irregularities but pointed out that its role was merely to proclaim the results presented to it (NDI 1993). Given such outcomes, it is no surprise that after another presidential loss, Awolowo refused to go to the courts again in 1981 and Besigye also refused to do so after his 2011 loss to Museveni. On the other hand, the performance of most African courts with respect to other electoral disputes, especially those challenging the outcomes of parliamentary elections, appears generally to be less predictable. The analysis of some commentators, such as the account of Siri Gloppen on the elections in Uganda in 2006 (Gloppen 2007) show that in some cases, 44

2005. 13 Nigerian Weekly Law Reports (NWLR) (Pt. 941) 1. 2001. The Supreme Court of Uganda (UGSC) 2(21April 2001). 46 2007. UGSC24, 31 January 2007. 45

Expansion of Judicial Power in Africa and Democratic Consolidation

59

the judges have displayed objectivity, independence, and assertiveness and have not hesitated to rule against the candidates of the ruling party. In other cases, the judges have been conservative and cautious and deferred to the candidates of the ruling party. Nevertheless, it is often uncertain what position the courts will take in any particular case, regardless of its merits. A recent Nigerian Supreme Court decision effectively sacked five state governors and ordered elections to be held immediately (Odemwingie et al. 2012). The five Governors’ 2007 elections were challenged in court, and in 2008 the results were nullified. The Governors contested these results, won the fresh elections that were called, and swore new oaths of office. When elections came due in 2011, the five Governors obtained a High Court order declaring that their four years’ tenure as Governors began when they took the second oath of office in 2008, not upon first oath in 2007. The Court of Appeals agreed, but the Supreme Court held that the terms of office were not personal to them, and that even if they took office in the middle of the gubernatorial term, they still had to step down when the term was over. The Supreme Court pointed out the absurdity of holding to the contrary. Such a ruling would mean that not just the first oath, but also all acts of the Governor after the first elections, would have to be declared invalid and unlawful. Some of the decisions show that even under very harsh and repressive conditions, some judges have been brave and firmly defended the rule of law. An example of this is the Zimbabwean case of Movement for Democratic Change and Mushonga v Chinamasa NO,47 where the applicants (who included one of the unsuccessful candidates in the June 2000 elections) sought a Supreme Court order setting aside the Election Act (Modification) (No. 3) Notice, which President Robert Mugabe had promulgated for the purpose of validating corrupt and illegal practices during the elections. The Notice itself purported to validate the election of candidates regardless of any irregularities committed in the process and declared that such elections shall not be rendered void. The applicants argued that this Notice infringed their constitutional right to the protection of the law and the right to a fair hearing (Hatchard, Ndulo, and Slinn 2004). The Supreme Court noted that the right to full and unimpeded access to the courts is of cardinal importance for the adjudication of justiciable disputes, for such access ensures “a mechanism by which … disputes are resolved in a peaceful, regulated and institutionalised manner” (Hatchard, Ndulo, and Slinn 2004). The court held that the existing civil rights of the applicants was to participate in a free and fair election and that they were legitimately entitled to expect that the results 47

2000. 3 Legal Resource Center (LRC) 673.

60

Chapter Three

in every constituency would be free and fair and properly representative of the will of the voters. Therefore, they had a civil right to challenge the results of an election that they alleged was tainted by corrupt and illegal practices. The Notice had effectively denied them such access. Accordingly, the Notice was declared null and void as being contrary to the constitution. This case illustrates the extremes to which incumbents are prepared to go in order to hang on to power even after losing elections. Empirical research has been carried out to find out whether there are any discernible patterns in judicial attitudes toward dealing with election disputes in African countries. A study based on the comparative analysis of election judgments in Malawi and Zambia to ascertain the pattern of judicial behaviour suggests that judges tend to read signals from the political environment when making their decisions (Von Doepp 2006). According to the author of this study, judicial assertiveness in these disputes depends on the extent of power concentration or fragmentation amongst key political actors. Where power is concentrated with a low likelihood of political change, the study found, judges engage in strategic self-restraint, and the likelihood of anti-government decisions diminishes as has been the case in Zambia. On the other hand, where the system is fragmented and there is high likelihood of change, the judges will defect strategically to distance themselves from current officeholders, as has been the case in Malawi. These findings suggest that unless the position of the incumbent is very weak, judges in Africa will rarely rule against the incumbent. Some of the Zimbabwean and Ugandan election cases support research findings indicating that high party competition induces courts to make decisions less deferent to the incumbent. The converse—lack of political competition—leaves the judiciary susceptible to pressure from the incumbent (Ginsburg 2003). While general evidence suggests that the courts have been able to resist the considerable political pressure they come under when faced with such disputes, many of these cases are still handled in a less than satisfactory manner. Nevertheless, it must be recognised that courts cannot do more than interpret and apply the law. Sometimes what lays the foundation for election fraud and hence the disputes that arise are bad election laws crafted by parliaments firmly under control of dominant parties, which are skewed against the opposition parties or against free and fair elections generally. But besides the challenges of dealing with election disputes, courts also face a tough time trying to ensure that the executive operates within the bounds of its powers.

Expansion of Judicial Power in Africa and Democratic Consolidation

61

Curbing Executive Lawlessness: Lessons from Benin and South Africa Executive lawlessness has been the bane of Africa’s politics. Before the 1990s in Anglophone Africa, the weak position of the judiciary, compounded by the excessive powers conferred on presidents, enabled the latter to arbitrarily interfere with the judicial processes and get away with violations of the constitution. In Francophone Africa, the issue of constitutional review was put beyond the reach of judges. The competent bodies, structured along the lines of the French Constitutional Council, were impotent. With the expansion of the reach of constitutional justice today, the question is whether the standards of justice have improved. Using possibly the best example of modern African constitutional courts, the Benin Constitutional Court modelled on the French Constitutional Council and the South African Constitutional Court (which although a hybrid is modelled on the American approach), it is possible to have some indication of how the courts are using the enhanced powers they now have to promote constitutional justice.

Examples from Benin The Constitutional Court of Benin, established under the 1990 constitution, has been the most active constitutional court modelled on the French system. It had, by the end of 2011, decided more than a thousand cases, far more than any other Francophone constitutional court in Africa.48 A few examples will be briefly discussed here to illustrate the role that the court is playing today in promoting democracy, good governance, and constitutionalism. In Decision DCC 33-94 of 24 November 1994, the Court of Benin declared as a violation of the constitutional right to freedom of association, a provision in Decree No 94 of 26 January 1994, which prohibited members of the Constitutional Court from becoming members of a political party and requiring those who were already members of political parties to resign. It held that neither the executive nor the legislature could suppress or limit arbitrarily a right that had been conferred by the constitution. 48

See in general, Médé 2012. The decisions of the Court are available at: http://confinder.richmond.edu/admin/docs/Benin1990English.pdf. Accessed 9 February 2015. For recent commentaries on this, see Adjolohoun 2011; Kante 2009; Balde nd.; Diarra 2001; and Rotman 2004, 280–314.

62

Chapter Three

In a number of instances, the Court has exercised its powers of prepromulgation review of laws and regulations to prevent legislation contrary to the constitution being enacted. An example of this is Decision DCC 34-94 of 23 December 1994 on the creation of an autonomous national electoral commission. The legislature, which was dominated by the opposition parties, was suspicious of the ability of the Ministry of Interior to handle elections in an independent and objective manner and therefore decided to enact a law creating an autonomous national electoral commission. The President of the Republic, arguing that the legislature had no power to make such a law and had in fact intruded into the executive’s law-making domain, requested the Constitutional Court to declare the proposed law unconstitutional for violating the principle of separation of powers. After carefully examining the constitutional provisions allocating law-making powers to both the executive and the legislature, the Court concluded that the proposed law did not violate the constitution. Decision DCC 05-69 of 26 July 2005 involved Yayi Boni, a Beninese who was President of the West African Development Bank, based in Lome, who wanted to run as a presidential candidate in the 2006 election. To frustrate his ambitions, Parliament decided to enact a new law introducing a residential requirement for presidential candidates, thus automatically excluding Yayi Boni. A number of his supporters petitioned the Constitutional Court. The Court held that the provision in the new law adding a requirement to those clearly specified in article 44 of the constitution was unconstitutional. Decision DCC 06-74 of 8 July 2006 involved an attempt by parliamentarians to extend their four-year mandate by one year. By an overwhelming majority of 71 votes out of 83, the constitutional amendment was approved, but a number of private individuals petitioned the Constitutional Court to have the amendment declared unconstitutional. The Court drew attention to the fact that article 80 of the constitution gave the parliamentarians a four-year mandate with a right to stand for re-election. It then noted that although the constitution allowed for amendments, such amendments had to 1) reflect the desire of the people of Benin to develop a multi-party state subject to the rule of law and 2) take into account the ideals and constitutional values that inspired the 1990 constitution. On this basis the Court concluded that an amendment prolonging the term of parliamentarians was not consistent with the desire and spirit of the constitution to promote a fair and equal opportunity for all to take part regularly in multi-party elections and hence was unconstitutional. The Court has also intervened in disputes involving the electoral process. In Decision EL 95-012 of 24 March 1999, it annulled a ministerial

Expansion of Judicial Power in Africa and Democratic Consolidation

63

order that tried to interfere with electoral functions that had been conferred on the national electoral commission. In Decision EL P-96-006 of 27 January, it received a complaint from the electoral commission that, contrary to article 37 of the constitution, the Government had refused to comply with its duty to make electoral envelopes available to the commission. The Constitutional Court issued an order compelling the Government to comply with this obligation. The Court had to intervene twice with respect to presidential elections. In 2005, the widely unpopular incumbent tried to adopt manoeuvres that were clearly designed to postpone the elections scheduled for March 2006. No money was made available to enable the electoral commission to prepare for the polls, so the latter petitioned the Court. On 17 March 2005, the Constitutional Court ordered the Ministry of Finance to make available within twenty-four hours all the money the commission needed to prepare for presidential polls. This was not done, and on 1 December 2005 the Court, after reiterating that its decisions were not subject to appeal, ordered the Ministry of Finance to immediately obey its earlier decision. In April, 1996, the Court issued a decision dealing with the format for taking the presidential oath of office. Kerekou, who had lost the presidential poll held under the 1990 elections had just won the April 1996 polls and was sworn in on 4 April 1996. The mandatory oath as laid down in article 53 of the constitution included the words, “Before God and our ancestors…” Having become an evangelical Christian during the brief period of his absence from power, Kerekou duly took the oath of office but carefully refused to repeat the part referring to ancestors. The next day, two concerned citizens approached the Constitutional Court seeking a declaration that the oath taken was not in conformity with the constitution and hence invalid. The Court agreed, and the President had to retake the oath in strict compliance with article 53. A number of weaknesses have limited the effectiveness of the court.49 Beginning rather timidly, from 1993 to 2002 the Court made only declaratory judgments. As a result, no action was taken against several ministers and other top officials whom the Court found guilty of several offences, and some were even promoted in spite of the court’s decisions. This changed after 2002 when the Court was given the power to award damages, although it was still not able to compel compliance with its judgments. The Court has also been criticised for lack of consistency in its judgments.50 This may be attributed more to the absence of a strict tradition of 49 50

These are discussed in Adjolohoun 2011, 121–122, 124–129, 135–136. See, ibid.

64

Chapter Three

stare decicis in the civil law system than anything else. Even if these criticisms are valid, they do not diminish the revolutionary changes brought about by the Beninese Constitutional Court. For the first time, individuals have access to constitutional justice. There is an irony here; it was only in 2008 that the French constitution was amended to make this possible in France (Hunter-Henin 2011). Another innovation is the power of concrete review, which enables abusive executive orders, decisions, and rules to be reviewed for conformity with the constitution. For a system in which the executive has extensive law-making powers that were previously not subject to review, this has been a tremendous expansion in the rule of law. South Africa provides an interesting contrast.

The South African Experience While concrete constitutional reviews by a court accessible to private citizens are a novelty to the Beninese constitutional system, such reviews have become the norm in post-apartheid South Africa. The South African Constitutional Court has adopted a very progressive, activist approach in most of its decisions.51 We will limit ourselves here to four important recent decisions where the court intervened strongly to curb executive excesses. Not surprisingly, the government has not taken these judgments well. For example, in November 2011, shortly after one of these decisions, the South African cabinet met and decided to “assess” decisions of the court as part of a broader transformation of the judicial system. Many viewed this move as an attempt to intimidate the court. As background to the first case, in December 2007 the ruling African National Congress (ANC) swept aside the leadership of then President Thabo Mbeki in a historic conference. The Congress resolved that South Africa should have a single police force and that the special corruptionfighting unit, the Directorate of Special Operations, popularly known as the Scorpions, should be disbanded. In the wake of this decision, there was widespread media speculation that the ANC had undertaken a vendetta against the Scorpions because they had investigated many senior officials in the party, including the newly elected party leader, Jacob Zuma. When the bill disbanding the Scorpions and establishing a new unit popularly known as the Hawks became law, Glenister, who had unsuccessfully challenged the cabinet decision to disband the Scorpions, brought an action 51 See generally, Scott and Alston 2000; Davis 2004, 47; and perhaps the most comprehensive account on this is, Corder 2007.

Expansion of Judicial Power in Africa and Democratic Consolidation

65

challenging the constitutionality of this new legislation. By a majority of 5 to 4, the Court declared that the amended Chapter 6A of the South African Police Service Act introducing the Scorpions was inconsistent with the constitution and invalid to the extent that it failed to provide for an adequate degree of independence for the corruption-fighting unit that it sought to establish. Although there was no specific provision in the constitution specifying that the unit be independent, the majority held that the constitutional obligation to set up an independent unit could be inferred from the duty imposed by section 7(2) of the constitution to “respect, protect, and fulfil” the rights in the bill of rights. In this indirect manner, the court declared legislation unconstitutional for violating international law, which became relevant not only because it was binding in South Africa but also because these international laws reflected the ethos of constitutionalism. In other words, the constitutional duty to create a corruption-fighting unit was not discharged by creating a unit under the control of the politicians whom it was required to investigate. The second case involved the appointment of a new Chief Justice to replace the incumbent, whose term was about to expire. The incumbent’s initial appointment had been controversial because the President had overlooked the Deputy Chief Justice, who customarily would have been chosen. Newspapers speculated this was due to the Deputy’s independent line of thinking, which did not go down well with the ANC. In order to prevent the Deputy Chief Justice taking over from the retiring Chief Justice, the President decided to extend the latter’s tenure by five years. In Justice Alliance of South Africa v. President of the Republic of South Africa,52 a number of civil society organizations challenged the constitutional validity of the law on which the President acted, claiming it violated the constitution in that Parliament had made an unlawful delegation of its powers to the President. This was an awkward case in which the judges had to determine the fate of their boss and colleague, the Chief Justice. In a carefully reasoned judgment, the court pointed out that the section of the law on which the President had acted was unconstitutional. A request for an order suspending the declaration of invalidity was also firmly rejected by the court, pointing out that this would be an indirect manner of rendering “valid an extension that [was] invalid.”53 The third case, Crawford-Browne v President of the Republic of South Africa,54 was brought by Crawford-Browne, who had waged a long cam52

2011. ZACC 23. Ibid., 49. 54 Case No: CCT 103/10 (applicant’s submissions). 53

66

Chapter Three

paign against the Government’s 1999 decision to spend about 70 billion Rand on the acquisition of arms. The corruption-fighting unit, the Hawks, had established evidence of massive corruption involving senior members of the Government and the ruling party in the arms deal. There were even allegations of President Zuma’s involvement, but corruption charges against him were inexplicably dropped by the National Director of Public Prosecutions (NDPP). In this case, Crawford-Browne took President Zuma to the Constitutional Court in a bid to compel him to establish a commission of inquiry into the arms deal. He argued that because of the litany of corrupt activities surrounding the procurement of the arms, the refusal and/or failure of the President to appoint a commission of inquiry was reviewable on the grounds that it was illegal and irrational in all the circumstances to fail or refuse to so appoint. It was further argued that under the rule of law foundational to South Africa’s democracy, the President was obliged to ensure accountability, responsiveness and openness. Conduct inconsistent with these foundational values, they argued, was invalid. The case was heard in May 2011 and postponement granted until 20 September 2011 for further papers to be filed. At the eleventh hour, on 16 September 2011, Zuma agreed to appoint the commission. Although the merits of the arguments of the applicant were never considered by the court, it was very likely that the President’s inaction would have been declared illegal and unconstitutional. The precedent for this position had been set in the earlier case of Albutt v The Centre for the Study of Violence and Reconciliation and Others.55 In this case, the court had ruled that any refusal to exercise public powers or exercise public law responsibilities is subject to the rationality review. The constitutional principle emerging from this is that presidential powers must be exercised in accordance with the rule of law, which entails legality, rationality, and fairness. In the fourth case, this time decided by the Supreme Court of Appeal, Democratic Alliance v President of the Republic of South Africa,56 the main opposition party, the Democratic Alliance (DA), challenged the President’s decision to appoint Menzi Simelane as the acting National Director of Public Prosecutions (NDPP). It alleged that the President had ignored overwhelming evidence from different sources proving that Simelane was not a “fit and proper person” as required by the relevant Act. The Democratic Alliance asked the court to set aside the decision on the grounds that it was unlawful, irrational, arbitrary, biased, based on an ulterior motive, and 55

2010 (3) SA 293 (CC). 2011. Supreme Court of Appeal of South Africa (ZASCA) 241 (1 December 2011). 56

Expansion of Judicial Power in Africa and Democratic Consolidation

67

inconsistent with the constitution. The DA not only pointed to Simelane’s limited experience but also argued that the appointment was politically motivated because it was made on the same day that the acting NDPP had dropped corruption charges against Zuma, and it appeared that the purpose was to install someone who would ensure that the charges were never brought again. In an elaborate and well-articulated judgment, the Supreme Court of Appeal agreed with the applicant that the appointment of Simelane was irrational and illegal because it was based on a decision process vitiated by material errors of facts and law. The DA has since won a case in which it wanted the NDP to make public the reasons for the withdrawal of the charges against President Zuma.57 The Benin and South African Constitutional Courts illustrate the contrasting approaches to controlling abuse of powers and promoting constitutionalism in Africa. Although one of the major weaknesses of post-1990 constitutions is that they still confer exorbitant powers on the executives or allow them to arrogate such powers to themselves, the judicial institutions are in a good position to check some of these abuses in the manner the South African Constitutional Court has been doing (Fombad 2012a).

Promoting Socio-economic Rights There can be no constitutional justice that does not entail social and economic justice. The post-1990 constitutional rights revolution is in part failing because the number of angry, unemployed, hungry, and frustrated people who demonstrated on the streets in the 1990s has increased, and their predicaments have grown worse.58 One in two people in sub-Saharan Africa survive on less than one dollar a day;59 33 percent suffer from mal57 At the time, the NPA said it was “neither possible nor desirable” to continue with the prosecution citing an “abuse of process.” The then acting director of the NPPA, Mpshe, pointed to telephone recordings between former NPA head Bulelani Ngcuka and former head of the Directorate of Special Operations Leonard McCarthy, discussing the timing of re-charging Zuma. As such, the case against Zuma was deemed to be politically motivated. “It's not so much the prosecution itself but the legal process that is tainted,” Mpshe said at the time. However, the DA is arguing that the decision was unconstitutional and invalid and that charges should be reinstated. See, Bauer 2012. 58 See “Facts on Poverty in Africa,” http://borgenproject.org/10-quick-facts-aboutpoverty-in-africa/ Accessed 9 February 2015. 59 Most of the economic statistics show regular economic growth in many African countries, but this is all job-less growth, which hits the most vibrant group in the economies, the youth, hardest.

68

Chapter Three

nutrition; fewer than 50 percent have access to hospitals or doctors; and life expectancy is only forty-one years. Poverty, deprivation, and marginalisation pose the primary threats to African democracy and constitutionalism. Democracy will have little meaning if it cannot deliver bread on the table and other positive outcomes for ordinary people.60 In fact, poverty and unemployment are ranked as two of the three most serious problems that Africa faces today (ECA 2005 and 2009). As the Arab spring has shown, the hungry are not going to continue to starve or endure their hardship quietly. Many post-1990 constitutions did not address the issue of socio-economic rights, or where they did, referred to rights in obscure and programmatic language that made rights non-justiciable.61 While no constitutional design can eradicate poverty and unemployment simply by entrenching socio-economic rights, a good constitution can at least provide a solid basis for addressing this problem. In this regard, the South African courts—especially the Constitutional Court—have taken the lead not only on the continent but in the world in showing how economics might be enforced by the courts. These courts have exploded the myth that constitutionally entrenching socio-economic rights may impose an unreasonable burden on the state. They have clearly demonstrated that the right to housing, health care, food, water, and social security in the constitution does not impose any unreasonable burden other than to require the state to “take reasonable legislative and other measures within its available resources to achieve the progressive realisation of these rights.”62 The literature on this is enormous, and the recent Angolan and Kenyan constitutions have gone even further in recognising a broader range of socio-economic rights (Fombad 2011a). The main lesson from the South African cases is that the constitutional recognition of socio60 Former President Nelson Mandela aptly captured the expectations of all African rulers when he said with respect to South Africa: “When we took on the project to transform society one of our rallying cries was “freedom from want.” Our goal was to banish hunger, illiteracy, and homelessness and ensure that everyone had access to food, education, and housing. We saw freedom as inseparable from human dignity and equality….We are fully aware that our freedom and our rights will only gain their meaning as we succeed together in overcoming the divisions and inequalities of our past and in improving the lives of all, especially the poor.” Cited in, UNDP 2002. 61 Fewer than half of the African constitutions recognise the different types of socio-economic rights. See generally, Heyns and Kaguongo 2006, 673–717; Fombad 2011a, 33–64 and Fombad 2011b, 1007–08. 62 For some of the vast literature on the numerous cases decided by the South African courts, see Mbazira 2009; Mubangizi 2007; Liebenberg 2010; and Mia 2005.

Expansion of Judicial Power in Africa and Democratic Consolidation

69

economic rights will not on its own necessarily improve the conditions of the poor and marginalised unless the courts are ready to adopt a more progressive and imaginative approach to dealing with such disputes. Looking forward, it would seem that the law, especially constitutional law, must always be interpreted and applied in a manner that promotes constitutional as well as socio-economic justice. Whether or not the courts in Africa can achieve this will depend on how they overcome the challenges they face and take advantage of the opportunities they have.

Challenges Faced by Judicial Institutions Democratic progress and constitutionalism as a whole depend on the effective operation of each of the three organs of government within their constitutionally defined domains. The basic constitutional framework is therefore critical. We have seen that, unlike in the recent past, the constitutional framework in many African countries now enhances the chances of judicial independence, which is critical to a fair, expeditious, objective, and just resolution of disputes. Because of this, African courts have recently played a role, however limited, in promoting constitutional democracy and justice. However, their ability to influence change is inherently limited because their success depends entirely on the cooperation of the other two branches of government. They remain, in the words of Hamilton in the Federalist 78 (Rosenberg 1991), the least dangerous branch, and vulnerable to manipulation. Besides, court decisions are not selfenforcing. Hence, in spite of a better protective framework today, the future role of African judicial institutions in promoting democracy and constitutionalism will depend on how they can overcome the present challenges and regain the confidence of the public. We shall first consider some of these challenges and in the light of these, see what the prospects for the future are. The African political scene has become more volatile, with elections that often carry the risk of violence due to the increasing tendency of incumbents to hang on to power by all means. As defenders and protectors of democracy and the constitution, judicial institutions have become the battleground for many political fights. Their ability to protect Africa’s fledgling constitutional democracy has been hamstrung by a number of obstacles. The main obstacles, examined below, are the increasing politicisation of the judiciary, judicial corruption, lack of resources, and judicial conservatism.

70

Chapter Three

Politicisation of the Judiciary Political interference with the judiciary and the threat it poses to the ability of the judiciary to operate independently, objectively, and fairly is probably the most insidious problem that has survived the post-1990 constitutional reforms and remains a potent obstacle to arresting signs of democratic reversal. The starting point is with the appointment of judges. Where a judge is appointed not on merit but rather on the basis of perceived or actual loyalty to the government, that judge will feel beholden to the executive and decide matters not on the basis of law but rather in a manner that will favour those who appointed him or her. In fact, many commentators today worry about the growing trend toward executivemindedness or a predisposition to favour the government (Rosenberg 1991). This is particularly so where the constitution places the appointment of judges entirely in the hands of politicians and there is no specific requirement that those appointed should be legal experts, as is the case with some constitutional councils. Where courts are dominated by executive-minded judges, the quality of decisions will certainly be low. After the controversial Ivorian presidential elections of 2010, it was no surprise that the Constitutional Council decided to ignore the results from the election commission, which showed that Mr. Ouattara had won, instead declaring incumbent Laurent Gbagbo the winner (Rosenberg 1991). Other instances where judges have tended to display excessive deference to the executive have occurred in countries where expatriate or contract judges, who tend to favour the government in order to have their contracts renewed, are employed.63 Prior to 1990, judges who dared to rule against the government were often physically threatened or penalised through demotions or transfers to remote parts of the country.64 During this new constitutional dispensation, governments still take very badly to adverse rulings, especially on sensitive political issues such as elections. In Malawi, after a number of antigovernment rulings in 2001, Parliament decided to impeach three of the judges involved, and the impeachment decisions were only revoked after concerted national and international pressure culminating in the Danish 63

See the examples of Botswana, Kenya and Tanzania, in Van De Vijver 2006, 30, 46, and 197 respectively. 64 This was particularly easy in Francophone Africa, where the Supreme Council of Magistracy, chaired by the President with the Minister of Justice acting as cochair, meet at least once each year to decide on issues of discipline, promotions and transfers.

Expansion of Judicial Power in Africa and Democratic Consolidation

71

government suspending aid to the country in 2002 (Van de Vijver 2006, 88–90). In Swaziland, after a series of court judgments against the government and the King, the Prime Minister issued a statement announcing that the government would no longer recognise the judgments of the Appeal Court. In response to this, the judges resigned and only returned a year later when the statement was withdrawn and an apology was issued (Van de Vijver 2006, 174–181). The tactics commonly used by governments today is to instigate sometimes very hostile verbal attacks or criticisms of the judiciary using either government-owned newspapers or NGOs funded by it. Examples of such practices have been reported in Namibia, Zambia, and Zimbabwe (Van de Vijver 2006, 103–104; Widner 2001). Sometimes, even high government officials have been involved in these attacks. For example, in Uganda, President Museveni accused the judiciary of “unprofessional bias against the Movement government (the ruling party) and corruption” (Van de Vijver 2006, 218). It is, however, the ongoing political attacks on judicial independence and integrity in South Africa that have left many constitutional experts, who had hoped that its constitutionalism would set a high standard for the rest of the continent to follow, worried. Two factors make the recent developments in South Africa a dangerous precedent that could have damaging repercussions for judicial independence in the rest of Africa. First, like almost 80 percent of the countries on the continent, South Africa has come under the grip and spell of a dominant party, the ANC, who, like all the other dominant parties on the continent, operate with the philosophy and instincts of the pre-1990 single parties that allowed little or no space for dissenting views (Fombad 2013).65 These dominant parties are now in a powerful position to change the new or revised constitutions and reverse the gains in democracy and constitutionalism that have been made so far. The dangers posed by the dominant party phenomenon are, as we noted earlier, aggravated by the excessive powers conferred on African presidents. The South African government has now used its strong position to 65

In this study, a dominant party is defined as one holding more than two-thirds of the seats in parliament. The study examined a representative sample of twenty countries in Africa and noted that 80 percent of them are controlled by dominant parties. Given that most African constitutions require only a two-thirds majority vote to amend the constitution, that requirement on its own is not a major obstacle to amending the constitution in many countries. But even if the level of difficulty were raised to require a three-fourths majority, the study showed that nine out of twenty, that is, 45 percent, of the countries examined would still have no difficulty in reaching this threshold.

72

Chapter Three

try to pack the judiciary with judges sympathetic to its views. Second, the Government has decided to call for a review of the judiciary. The basis of this review, a discussion document on what it refers to as transformation of the judicial system, is rather vague and does not provide any clue as to the real objectives of the exercise. Significantly, the government initially only mentioned a review of the decisions and powers of the Constitutional Court. A few weeks later, after the Supreme Court of Appeal ruled that the NPA should produce the records of its 2009 decision to suspend criminal charges against President Zuma, the Government decided to expand the review to include the Supreme Court of Appeal (South African Press Association 2012). Many commentators see this exercise as a sinister attempt to intimidate the judiciary because of the numerous cases where it had ruled against the government (Legalbrief Today 2012). There is thus emerging on the continent as a whole, an increasing trend toward political interference, which only adds to the numerous challenges that the judiciary faces, one of the most serious being corruption.

Judicial Corruption Judicial independence requires freedom not only from political manipulation but also from litigation by the socially powerful. Judges all over the continent are reasonably well paid and receive salaries and pensions well above that of the average civil servant. Yet, after the police, the judiciary is generally considered to be the most corrupt institution in Africa.66 The case of Kenya is well documented. In 2003, the Kenyan judiciary was described as “pathologically sycophantic,” “grossly incompetent,” and “shamelessly corrupt” (Van de Vijver 2006, 50). A Ringera Committee Report of 2003 that investigated issues of integrity and corruption in the Kenyan judiciary contained a “price list” showing the cost of bribing a judge. The report also indicated that corrupt magistrates expected 10 to 30 percent of any civil awards they ordered (Van de Vijver 2006, 51). Corruption, especially on the scale reported in Kenya, destroys confidence in the judicial system and distracts judges from their sacred responsibility to do justice and see that justice is done. While corruption is something that the judiciary as an institution and the individual judges can do something about, there are numerous other problems afflicting the judiciary which 66 See for example, Mugwe 2011, Ochieng 2003, Emeka 2011, and Sibanda 2010. Besides these, the two ECA Reports on African Governance, ECA 2005 and ECA 2009, referred to earlier have indicated that corruption is one of the three most serious problems in Africa.

Expansion of Judicial Power in Africa and Democratic Consolidation

73

are to some extent beyond their control. One of these is the problem of lack of resources.

Lack of Resources All the studies on the judiciary in Africa in the past two decades have drawn attention to the problem of lack of resources (Van de Vijver 2006; Gloppen, Gargerella, and Skaar 2004, 17–31; Gender Justice Conference 2008). The judiciary cannot operate either independently or efficiently without adequate resources. In some instances, the lack of resources may be a deliberate political ploy to put pressure on the judiciary. For example, in some countries, judges have complained that their budgets have been cut because of judgments given against the government (Van de Vijver 2006; Gloppen, Gargerella, and Skaar 2004, 17–31; Gender Justice Conference 2008). It is particularly easy to use judicial appropriations to exert undue pressure in certain systems, such as those obtaining in Francophone and Lusophone countries, where the judicial budget is not protected by being part of the consolidated fund. Complaints about lack of resources for the judiciary are widespread in Africa, but the problem is particularly severe in certain countries. For example, there is an acute shortage of trained staff in countries such as Botswana, Swaziland, Lesotho, and Namibia, where governments continue to rely on expatriate staff appointed on short-term contracts. Even countries such as South Africa, Nigeria, and Cameroon still have staffing problems. These are exacerbated in certain situations, for example after elections, when judges are required to leave their normal responsibilities to deal with election petitions. The effect of staff shortages is a huge backlog of cases. Other problems that constrain the performance of the judicial institutions on the continent are poor infrastructure, lack of adequate funds to cover running expenses, insufficient stationery and office equipment, and limited working facilities such as court rooms. In many countries, judges don’t have access to adequate legal resources such as books, case reports, statute books, and gazettes. This is particularly so in the rural areas where there are no libraries, or where such resources have been neglected over the years; and the situation is often exacerbated by lack of computer skills or lack of motivation on the part of users to learn the appropriate skills. In some cases, there is no access or limited access to the internet, with technical problems such as slow and erratic networks, unreliable power supplies, and poor user support and maintenance systems being common. The development of legal capacity on the continent has been retarded or even reversed in some countries, such as Cameroon, by the poor state of

74

Chapter Three

court records, which can make it very difficult for the superior courts to deal with appeals. Perhaps the most damaging condition has been the absence of a reliable and regular system of law reporting in some countries such as Cameroon or the fact that such reporting is erratic in others such as Malawi and Botswana. Without an efficient system for keeping records or reporting court cases, the operation of the legal system, especially where it is based on the stare decicis, cannot function properly. These problems are aggravated by certain trends in the judicial attitude and approach to adjudication.

Judicial Conservatism There isn’t much evidence in African judges’ approaches to constitutional interpretation to indicate that they are alive to the new progressive constitutional rights spirit of the 1990s. There is a certain narrow conservatism that often ignores the new context in which constitutional disputes must be interpreted: the rejection of dictatorship in all its forms and sensitivity to the issues of human rights. Many judges are still too imbued with the past to recognise the imperatives of change necessitated by a fast globalising world where geographical as well as legal boundaries are breaking down regularly. For example, commentators on the performance of the Benin Constitutional Court have bemoaned the inconsistency of its decisions (Adjolohoun 2011, 121–124). It is not difficult to figure out the reason for this inconsistency. Judgments are delivered in the terse and abstract civilian style inherited from the French, which fails to bring out the rationale for the decision. Nor do these judges try to see what is happening in neighbouring countries faced with similar issues. The future of constitutionalism in Africa depends on how judges will, where possible, overcome these limitations.

Prospects for the Future It is now clear that the prospects for judicial institutions in Africa to act objectively and independently in a manner that will promote Africa’s faltering attempts to entrench constitutional democracy and promote constitutional justice are not as good as they appeared after the hectic constitutional changes of the1990s. As the fight by African leaders to cling to power intensifies, there will be more pressure on the judiciary to decide cases, especially politically sensitive cases, with deference to the executive. The prospects for the future depend on the judiciary becoming more assertive and willing to reflect the contemporary desires and aspirations of

Expansion of Judicial Power in Africa and Democratic Consolidation

75

the people and the progressive spirit of the post-1990 constitutional reforms. For the judiciary in Africa to play an effective role in reversing the ominous signs of authoritarian resurgence, there are at least four measures needed. First, the judiciary must be willing to act as agents of change and the last defence against authoritarianism. Second, they will need to join the global judicial dialogue that strives to promote respect for the rule of law and good governance. Third, they will need to use their privileged position to eliminate some of the anomalous remnants of customary practices that still deprive certain sections of society from enjoying the full benefits of modern constitutionalism. And fourth, the training of judges will need to improve considerably.

The Judiciary Acting as Agents of Change and Last Defence against Authoritarianism67 In the face of Africa’s overbearing executives and the constitutional weaknesses of the legislature, a passive judiciary, compounded by the phenomenon of one-party domination, will certainly take Africa back to the dark days of one-party dictatorships and destroy any hopes for constitutionalism and democracy. Judicial independence in the postindependence period was compromised not only by the intolerance of executives but also by the “enthusiastic abdication of judicial responsibilities by the judges” (Odinkalu 1996, 136–137). Constitutionalism and democracy in Africa can only develop and grow with judges who are liberal, progressive, and activist, or who have “bold spirits” and are not the “timorous souls”68 of passive judges of the past. It is contended that constitutional justice can only be realised with a judiciary that is ready to use its powers to negate the continuous authoritarian impulses of elected politicians.69 This, it is argued, can be accomplished in at least four ways. First, judges must as an imperative adopt a more principled and rights-sensitive approach to interpreting constitutions in a manner that takes into account the modern revulsion against dictatorship and the radical political, economic, and social changes of our times. It is manifestly clear that the revised or new constitutions were designed not only to eliminate dictatorship and promote democracy and good governance, but also to promote a new human rights culture that is sensitive to 67

This aspect of the paper is discussed in Fombad 2011b, 1066–1089. Lord Denning in Candler v Crane, Christmas & Co. 1951. 1 AllER 426. 69 This is elaborated upon in Fombad 2011b, 1066–1089. 68

76

Chapter Three

hunger, poverty, unemployment, ignorance, illiteracy, disease, and other social ills that have inflicted so much hardship on a majority of Africa’s population. Attaining these goals requires a judiciary that is willing to reflect the new spirit of constitutionalism when interpreting these constitutions. Second and closely related to the preceding point is the idea that judges should go beyond their traditional role as interpreters of the constitution and strive to do justice by giving effect to contemporary social conditions and values. This idea of progressive judicialism, or what has been referred to in many quarters as judicial activism, has been quite successful in promoting constitutionalism in South Africa and India, and it is argued that the same can be true elsewhere.70 Third, another approach is to adopt broad interpretive techniques when reading constitutions. This, it can be argued, inheres from the very nature of a constitution itself. Far from being a document that contains “timeworn adages or hollow shibboleths,”71 or that resembles a “lifeless museum piece,”72 a constitution must be regarded as a living document that is designed to serve present and future generations as well as embody and reflect the fears, hopes, aspirations, and desires of the people.73 Finally, perhaps the most significant way in which African courts could strengthen good governance and constitutionalism on the continent is to ensure that governments do not breach their international commitments. It is a notorious fact that African governments have no difficulty acceding to international treaties and conventions, but are never in a hurry to domesticate them. While courts cannot compel countries to domesticate the agreements they have joined, they can at least compel their countries not to act in breach of these agreements. The foundation for this approach was laid down in the famous Botswana case. In dealing with the position where a treaty had been signed but had not been domesticated, Amissah JP in Attorney-General v Dow cited with approval the following passage from the judge a quo in the same case: I bear in mind that signing the Convention [the OAU Convention] does not give it the power of law in Botswana but the effect of the adherence by Botswana to the Convention must show that a construction of the section 70

See Ibid. for the discussion of some Indian and South African cases and some of the relevant literature. 71 Earl Warren C.J. in Trop v Dulles 356 U.S. 86(1958) at p. 103. 72 Aguda J.A., in Attorney-General v Dow 1992. B.L.R. 119 at p. 166. 73 Some argue that the constitution is for the living, and the present generation should no longer be ruled by the dead hands of their ancestors. See McConnell 1998, 1127–28.

Expansion of Judicial Power in Africa and Democratic Consolidation

77

which does not do violence to the language but is consistent with and in harmony with the Convention must be preferable to a “narrow construction” which results in a finding that section 15 of the Constitution permits discrimination on the basis of sex.74

In taking this action, the Court basically followed the well-established presumption in statutory interpretation that courts will strive to interpret legislation in such a manner that it will not conflict with international law. The judge went further to explain thus: Botswana is a member of the community of civilised States which has undertaken to abide by certain standards of conduct, and, unless it is impossible to do otherwise, it would be wrong for its courts to interpret its legislation in a manner which conflicts with the international obligations Botswana has undertaken. This principle is used as an aid to construction as is quite permissible under section 24 of the Interpretation Act.75

This approach also appears to have been adopted by the South African Constitutional Court in the Glenister76 case referred to above. Hence, it can be said that judges have an inherent duty to continuously breathe life into the constitution and ensure that the government respects its commitments.

Joining the Global Judicial Dialogue In order to sustain the momentum toward greater constitutional justice in Africa, judges need to abandon the conservative inward-looking culture that was characteristic of the old judiciary and see themselves as members of a global legal community where knowledge and ideas are exchanged across jurisdictions. So far, there has been little evidence of an intraAfrican dialogue either in terms of cross-fertilisation, borrowing or migration of constitutional ideas from one African constitution to another, or through the use of jurisprudence from other African jurisdictions. For example, the South African Constitutional Court during its early years relied extensively on foreign jurisprudence. Hardly any of this was from other African jurisdictions (Udombana 2005; Lollini 2007). Both in constitution-making and in constitutional interpretation, there is still a tendency for African constitutional designers and judges to rely upon and copy ex74

1992. BLR 119 at p 154. Ibid. 76 Supra. 75

78

Chapter Three

tensively or even exclusively from the former colonial power’s legal system. This negative, self-defeating attitude has changed considerably in Anglophone Africa, where constitutional experts look beyond England for inspiration and guidance. In Francophone Africa, constitutional draftsmen have continued to seek inspiration from and rely almost slavishly upon the Gaullist Fifth Republic model, which some still consider to be the most reliable and unassailable constitutional model.77 Three critical issues need to be noted here. First, the progressive nature of the 1996 South African constitution and the high quality of judgments delivered by the South African Constitutional Court, which in certain areas such as socio-economic rights has laid down principles that are copied the world over, suggests that we do not need to continue to look outside the continent for constitutional models. The South African constitution of 1996 not only incorporated the best elements of Western constitutional systems but also adapted these elements to its historical, social, and cultural context to the extent that it is now a model that even some Western countries emulate. Second, the universality of certain constitutional law principles and standards is no longer in doubt. For example, a court can hardly deal with a human rights dispute today without being invited by counsel to consider one foreign authority or another. More often than not, the reliance on these foreign authorities has been rather eclectic, with scant or at most superficial reference to the techniques of comparative law. Even if a constitution does not explicitly or implicitly authorise the judge to refer to or invoke international or comparative law, it can be argued that in the globalised world of today, it would be self-defeating and a dereliction of duty for a judge to completely ignore legal developments, and their actual or potential implications on national law, in the rest of the world (Udombana 2005). The fundamental values that underpin most constitutions, such as equality, dignity of the human being, non-discrimination, freedom of speech, and other rights are now nearly universal. Third, because of the internationalisation of many constitutional law principles, judges must resist the intoxicating notion that they know it all and be prepared to keep up with the latest developments in law (Fombad 2012b). They must adopt an open mind and be willing to research and consider developments abroad, particularly in other legal systems. Fourth, the importance—and embarrassing absence of—an intraAfrican dialogue cannot be underscored enough. The commonality of experience shared by African countries, from colonialism and its aftermath 77

See further Fombad 2008.

Expansion of Judicial Power in Africa and Democratic Consolidation

79

to the challenges of dealing with ethnic, cultural, and religious diversity, means that they can and should learn from one other’s experiences. This does not imply blindly copying what obtains in other jurisdictions but rather drawing inspiration from their dynamic and rich jurisprudence. Legal systems continue to innovate, experiment, and adopt new solutions, from which others can learn, to the problems that arise every day all over the continent. However, the enormous possibilities for enriching national constitutional law through the process of judicial dialogue are unlikely to be felt in Francophone and Lusophone Africa. Following the civil law tradition, the courts in these countries rely very little on precedents, and their judgments contain little factual and legal analysis. Besides referring to the applicable laws, these judgments hardly ever articulate the factors or policies that might have influenced a particular conclusion. This is the main reason for inconsistencies noted earlier in decisions taken by the Benin Constitutional Court. As one commentator has rightly pointed out, there is no reason for Francophone judges to stick to an inherited French style of formulating and rendering judgments when this is detrimental to ensuring predictability, consistency, and fairness in judicial decisions (Adjolohoun 2011, 121– 122; Rotman 2004, 310; and Codjovi 2002). The challenges of access to legal information, especially to hard copies of law reports and textbooks, have been noted. Internet access is expanding, and this opens up new possibilities for promoting access to information, especially from African literature. Today, the latest and most important decisions of African national constitutional courts and regional courts are just a mouse-click away.78 As a result of the rapid advances in information and communication technology in the past two decades, not only judgments but also an extraordinary amount of constitutional material from different jurisdictions is now available. The benefits of this development will be completely lost if national legal experts and judges are too unbending to see the benefits of learning from what is happening elsewhere on the continent. But an important way of overcoming this mental barrier could be through cross-national networking of judges from differ78 The Southern African Legal Information Institute (SAFLII) now provides free online access to the latest decisions from the superior courts in Angola, Botswana, Lesotho, Kenya, Madagascar, Malawi, Mauritius, Mozambique, Namibia, Seychelles, South Africa, Swaziland, Tanzania, Uganda, Zambia and Zimbabwe. It also provides access to the decisions of the following regional courts of justice: COMESA, the East African Court of Appeal, the East African Court of Justice, and the SADC Tribunal.

80

Chapter Three

ent legal jurisdictions. Such personal contacts between judges will enable them to share their experiences and see how judges occupying similar positions elsewhere often confront the same kinds of problems. Although there are in existence many fora where judges from different countries meet, with the exception of the recently established Judges’ Working Group in the African Network of Constitutional Lawyers (ANCL),79 these have often been along regional lines and provided little opportunity for judges from the different African legal systems to intermingle.80 As the numerous regional and continental tribunals in Africa become more active, one would expect the process of internationalisation through crossjurisprudential fertilisation to be intensified. This will be particularly so with national constitutional courts whose decisions are subject to review by these tribunals; such courts will have to take into account the tribunals’ jurisprudence in order to avoid the embarrassment of decisions being reversed on appeal.

Judicial Training Unlike in the civil law tradition of Francophone Africa, where judges undergo formal two-year training, there is no formal training for judges in Anglophone countries. Too many people in the common law world appear to have taken too seriously Lord Devlin’s view that judicial training might be used as a propaganda tool by the executive and thus intrude on the independence of judges (Devlin 1979). This attitude is now changing, and in many common law jurisdictions, there are courses organised to train judges, especially on their first appointment to the bench. Arguably, the main cause of the judicial conservatism alluded to earlier is the relatively poor quality of training available to judges, which creates ample opportunity for their political manipulation. This is reinforced in many cases by questionable systems of promotion, where politics rather than competence is the primary consideration. To enable judges to adequately deal with the contemporary issues of constitutional justice and 79

This mixed judicial forum consisting of constitutional court judges from Anglophone, Francophone, Lusophone and Hispanophone Africa was formally set up during the ANCL Annual conference that took place in Rabat from 2–5 February 2011. 80 Some of these are the East African Magistrates and Judges’ Association, the East African Community Forum of Chief Justice, and the Southern African Chief Justices Forum (SACJF). The only exception is the latter, which includes the Chief Justices from Angola and Mozambique.

Expansion of Judicial Power in Africa and Democratic Consolidation

81

good governance, judicial training should no longer be taken for granted. At least two changes are imperative. First, judges need to be trained in a structured and systematic manner, whether formally as in Francophone countries or informally as in Anglophone countries. Judicial training must be complemented by regular, continuing judicial education programmes carefully designed to expose judges to new ways of thinking, recent developments, new experiences, and the opportunity to share their experiences. Second, in the globalised judicial world of today, the judiciary needs to be familiarised with the techniques and pitfalls of comparative legal research. Once it is accepted that consideration of foreign law is not only unavoidable but absolutely necessary, the next question is how to assess the relevance and applicability of various foreign laws. Too often, judges have simply accepted and adopted legal principles from other common law jurisdictions without heeding Watson’s warning that in law, as in medicine, transplants carry the risk of rejection (Watson 1974). Judges therefore need to be formally trained in comparative law generally, and in the techniques for using and applying foreign and international law in particular. In borrowing from foreign sources there are skills and techniques that pertain and need to be learned. Unless the judiciary is ready to learn how to deal with current developments both nationally and internationally, judicial decisions will retain a rigid, positivist orientation that is bad for both human rights protection and social justice.81 Lord McCluskey in the opening paragraph of his book, Law, Justice, and Democracy, highlights some of the challenges when he says: If I were to be asked what temptations any new judge is exposed to, I should have to admit that they include arrogance, self-esteem and impatience. That answer must alarm all who know that the principal qualities a judge must possess are humility, modesty and tolerance. But just think of the facts. He has been elevated to a position in which he wields a royal authority. The apparatus of state lies ready to enforce his orders. The visible symbols of his office, the way he dresses, the place in which he sits, the manner in which he is addressed, the respect which he is accorded, all are designed to buttress that authority, to intimidate those who might wish to challenge or evade it (McCluskey 1987).

81

See ibid at p. 4, where the author points out that “comparative law has been proving extremely useful in the countries of Central and Eastern Europe in reconstructing their legislation and legal order.”

82

Chapter Three

Judges must therefore resist the intoxicating notion that they know it all and be prepared to keep up with the latest developments in the law.

Conclusion For probably too long, it has been easy and convenient to blame the failures in good governance, human rights protection, democracy, and constitutionalism on colonialism or neo-colonialism (to which one must add apartheid) and African political leaders. The post-independence African constitutions, by design or by default, perpetuated an authoritarian, repressive, and harsh system of governance in which the executive in various ways dominated the legislature and judiciary. The judiciary were reduced to executive accomplices. The 1990 winds of change appeared to have restored the judiciary to its proper role as defender and protector of constitutionalism and the resurgent moves toward multi-party democracy. However, after more than two decades of transition to democracy and constitutionalism, it is clear that the prospects for the future do not look as bright as they initially appeared. There have been opportunities for the judiciary to play a more active, meaningful, and constructive role in promoting the course of democracy and constitutionalism. The African judicial institutions are more independent today than they have ever been, and their powers to review and sanction constitutional violations have also expanded. It was inevitable that the increase in judicial independence and power would involve the judiciary in political battles and controversy with the executive. In dealing with election disputes and abuses of power, the African courts have generally not been as assertive and principled as one would have expected. Apart from South Africa, the courts on the continent have not acted vigorously and imaginatively to promote socio-economic rights. Although judicial boldness and assertiveness on its own will not put an end to election abuses, executive lawlessness, or poverty and deprivation, it will certainly put pressure on election management bodies and governments to improve their performances. Free and fair elections lie at the heart of African democracy. Every election that ends in controversy and dubious results chips away at people’s faith in democracy and constitutionalism—and the faith and confidence of the people are crucial to maintaining the momentum for change. The ability of African judicial institutions to keep the flames of democracy and constitutionalism burning when these are threatened by the enemies of transition are hamstrung by numerous challenges. Some, such as frequent political interference and resource limitations, are external challenges, while others such as judicial corruption and judicial conserva-

Expansion of Judicial Power in Africa and Democratic Consolidation

83

tism are internal, of the judiciary’s own making. The future prospects for democracy and constitutionalism on the continent to a large extent depend on how the judiciary plays it role. It has been argued here that judges must act imaginatively as agents of change, and as the last defence against any authoritarian reversal. To do this, African judges need to be open and ready to share ideas, to learn from each other, and to be actively involved in the global judicial dialogue rather than continuing to blindly follow the inherited colonial stereotypes. To a large extent, building a judiciary with such capacities depends on training. It has been argued further that more serious effort needs to be made to provide a well-structured system of training (either formal or informal) that takes into account the emerging global environment and the benefits of intra-African legal dialogue and cross-regional fertilisation of ideas. The critical role of African judicial institutions in pursuing constitutional justice and protecting the democratic gains of the past two decades assumes an unprecedented importance and urgency because of the many threats that we face today. The analysis of election judgments clearly shows that where elections are quite competitive, the pressure on judges to be deferent to incumbents is considerably reduced. Yet today we are moving ever closer to the non-competitive elections of the pre-1990s, with dominant parties bent on hanging onto power and sit-tight presidents wielding excessive power and using their vantage positions to criticise and intimidate judges. On the other hand, there is a restive public losing faith in elections and in the ability of judicial institutions to defend them against the “robbery of their votes,” as some describe controversial election results. However, the situation is more hopeful than hopeless. African judicial institutions have “no influence over either the sword or the purse,”82 but present constitutions place them in better positions than they have ever been to check the abuse of power and entrench a culture of democracy and constitutionalism. To succeed they must simply be as bold and assertive as the South African Constitutional Court judges have been.

82

Quote from Alexander Hamilton, Federalist Paper 78, in Rosenberg 1991.

PART 2: CONSTITUTION-MAKING, ELECTIONS AND CONFLICT SETTLEMENT

CHAPTER FOUR CONSTITUTION-MAKING IN ANGLOPHONE AFRICA: WE THE PEOPLE? COEL KIRKBY AND CHRISTINA MURRAY

Modern constitutions are frequently amended or replaced, certainly much more often than the venerable American constitution has been. It seems, however, that the pace of change is even higher in Africa. Every Anglophone country in Africa, with the single exception of Botswana, has implemented some form of fundamental constitutional change since 1990.1 There have been twenty-three attempts to create new constitutions; Zimbabwe’s attempt is the most recent, while Ghana, South Sudan, Tanzania and Zambia all have constitutional review processes underway. There have been some successful constitution-making processes, such as those in Namibia, South Africa, and Uganda in the 1990s, and more recently in Kenya (2010) and Zimbabwe (2013). There have also been outright failures, such as Kenya’s 2005 referendum, and apparent successes, such as Swaziland’s a year later, that merely entrenched undemocratic government. What all these cases of constitution-making have in common, however, is a formal commitment to popular participation in the process. In 1990, African states agreed to a charter that declared: “popular participation is the fundamental right of the people to fully and effectively participate in the determination of the decisions which affect their lives at

1

Many countries have had more than one constitution-making process. To be more concise, we usually use the date on which a new or revised constitution was adopted. In cases where processes failed, we usually use the date the process started. On occasion, when the sense of the text demands it, we deviate from this practice and so, for example, when discussing the referendum at the end of the process that started in Kenya in 2000, we use the date of the referendum (2005).

Constitution-Making in Anglophone Africa: We the People?

87

all levels and at all times.”2 In the practice of constitution-making, this means finding ways to make the process democratic, transparent and accountable to the people (Ndulo 2003). If a constitution is to address the basic distribution of state power, then the people must be actively involved to express their desires and restrain political elites from highjacking the process (or being perceived to do so) (Hatchard 2001, 215). All stakeholders must have a say in the process in order to feel it is owned by them and not imposed by others. The notion of “stakeholders” has grown more inclusive in recent years to cover previously excluded groups such as women and particular minorities. For the constitution-making process to truly represent the popular will, some argue that a referendum is required to confirm its legitimacy (Ndulo 2003, 362). In this participatory paradigm, Hart (2003, 3) argued that the old “elite-made constitution ... will lack the crucial cultural element of legitimacy.” Today it is inconceivable that a government would attempt to draft a new constitution without at least a nominal commitment to a process in which the public is consulted. This chapter examines the rise of participatory constitution-making in Anglo-African countries in the past two decades. It first outlines the process by which the “independence” constitutions were adopted during the decolonizing period and examines early experiments in consultation and referenda. Next, recent arguments for and against participation are sketched against their local contexts. The chapter then focuses on constitution-making in contemporary times, looking at the three stages of 1) initiating constitutional review, 2) drafting and 3) ratifying a new constitution. Some of the many difficult problems that confront demands for “peopledriven” constitution-making processes are indicated, but the primary goal is descriptive: to set out the major developments relating to public involvement in constitution-making in Anglophone Africa and thus to pave the way for further work in the area.

From Imposition to Participation in Constitution-Making In the rush to decolonize, the British government and nationalist leaders negotiated independence constitutions modeled on a Westminster-style parliamentary system. The only independence constitution that has survived in Britain’s former African colonies is that of Botswana. The others, and most of their successors, succumbed to the contradictions of post2 African Charter for Popular Participation in Development and Transformation, 1990, E/ECA/CM.16/11, art. 10; see also Hart 2010.

88

Chapter Four

colonial forms of rule, the shared legacy of independent states in Anglophone Africa (Mamdani 1996; Ghai 1991). Many compelling reasons have been given for the problems that beset these new African states in the years following independence. A new reason has recently has been added: the process by which the new polities were established did not allow for the participation of the people. Usually the Colonial Office in Westminster, in its role of imposing constitutions during the rush to decolonize, is identified as the villain of post-colonial Africa. But African nationalist leaders (at least those identified as “moderates” by the Colonial Office) were generally content to negotiate and consent to these constitutions on behalf of the people they claimed to represent (Hyam 2006). The citizens of these newly independent states had little if any involvement. Agreements were negotiated one way or another with political leaders of the new states, but usually this was done away from home, at a considerable distance from the people these leaders were to govern, in a process controlled by British officials, and in a way that conveyed a strong sense that the authority of the outgoing power was to be transferred not to the people of the newly independent state but to the handful of approved nationalist leaders. The idea of public participation had little purchase with either British authorities or national elites. It is true that there were some instances of popular referenda held by British authorities, and later there were experiments with public consultation in new constitutions promulgated by one-party or military governments. But the general trend was to restrict the creation and amendment of constitutions to a handful of recognized leaders and their parties. The “independence” constitutions all followed a similar process and structure, beginning with Ghana in 1957. Westminster passed an “independence” Act declaring each territory a dominion within the British Commonwealth (in later constitutions, states were instead declared republics).3 The constitutions were created by orders-in-council associated with each independence Act. Their amendment was to be by a regular parliamentary process with a few additional requirements such as a two-thirds majority; sometimes, some kind of delay in the legislative process; and, in a few cases, a referendum for changes to particular provisions. In the aftermath of independence, several of the new parliaments formed themselves into constituent assemblies to re-create their states as republics. The Ghanaian Parliament set the trend, constituting itself as a constituent assembly—not provided for in the 1957 Constitution—to declare the state a 3 For example, Constitution of Botswana, 1966, art. 1; Constitution of the Republic of the Gambia, 1970, art. 1.

Constitution-Making in Anglophone Africa: We the People?

89

republic under a new constitution in 1960 (Schwelb 1960). Over the next decade, other states such as Nigeria (1963), Malawi (1966), and Sierra Leone (1971) recreated themselves as republics by adopting new presidential constitutions through acts passed by Parliament. The self-proclaimed republics, with strong executive presidents that rapidly replaced the parliamentary constitutions, paradoxically reproduced the idea that constitutional change is the preserve of a small elite. These “Constituent Assemblies” were merely Parliament acting under another name. The 1963 Nigerian Constitution, for instance, “was decided upon entirely by the prime minister and the regional premiers who, meeting for just one day, agreed among themselves that a republican constitution should reproduce the 1960 imperial constitution with such amendments as would conform it to a republican status” (Nwabueze 1975, 6, cited in Ogowewo 2000, 139). In cases where something more elaborate was considered necessary, a government might establish a commission, but only those proposals agreed upon by the executive would be presented to Parliament, whose decision would be final. This was a problem mainly because a single party or military government, unmediated by meaningful separation of powers or opposition parties, controlled the entire process. One apparent exception was the 1960 Ghanaian process, in which the public was asked to approve a suggested draft constitution in a referendum before the Constituent Assembly met to discuss it. The constitution that emerged also asserted that “Power to repeal or alter this Constitution is reserved to the people.”4 But, as Schwelb (1960) points out, combined with a highly authoritarian system of government with power massively concentrated in the president, this approach effectively gave the President more control over the constitution. It was the President who had the power to call (or refuse to call) a referendum, and rather than a super-majority in the legislature, any proposed amendment put to the people required only a simple majority of votes cast for adoption. More recently, the adoption of the Nigerian constitution of 1979 may have been considered a departure from the trend (Nwabueze 1982). The military government established a Constitution Drafting Committee to write a draft constitution and a Constituent Assembly to discuss the draft. Despite appearances of widespread participation, however, the entire drafting process was tightly controlled by the military, which certified all candidates to the Constituent Assembly and made seventeen unilateral amendments to the Constitution before its promulgation in 1979 (Koehn 1989). 4

Constitution of Ghana 1960, art 3.

90

Chapter Four

Before the third wave of democratisation, the constitution-making processes in Nigeria and elsewhere remained the preserve of a miniscule elite aided by a relatively small group of men chosen by and responsible to these same elites. This changed after the collapse of communist states in Eastern Europe, which inspired civil society movements from Gambia to Zambia to recapture state power monopolized by incumbent governments (Joseph 1997). The opening up of political space was followed by two great waves of constitution-making, with experiments in intensified public participation during the 1990s and the following decade. Most Anglophone constitutions adopted in the past two decades have included a significant shift towards including the public in each stage of the process of constitutional change—agenda setting, the deliberating and constitutionwriting stage, and ratification. The ethos of constitutional reform itself was challenged by new arguments for public participation that rejected the earlier view that an entire constitution, or fundamental elements of it, could be replaced by the same process used for less significant amendments.5

Starting a Process: the Agenda The first step in constitution-making is setting the agenda. In the past little attention was paid to this stage since constitutional review was usually under the control of a dominant or single party, which would use the existing constitutional amendment process to initiate changes. Since parliaments were populated by a single, (initially) very popular nationalist party, it could be argued that these legislatures were the appropriate bodies to initiate constitutional change. In practice, however, dominant parties had virtually unlimited power to manipulate the constitutional framework to consolidate their own hold on power—and most readily did so, even as their popularity declined. In cases where the military had abrogated the existing constitution, those who had come to power in the process often created their own constitution by decree, as happened with Nigeria’s 1979 Constitution. Since 1990, however, there have been increasingly successful attempts by opposition parties and civil society to wrest these decisions from the exclusive control of incumbent governments and secure a participatory constitution-making process. These initiatives are based on a wellfounded fear of executive manipulation of the constitution-making process; people have lost faith in parliaments, politicians, and political par5

Njoya and others v Attorney-General and others [2004] LLR 4788 (HCK), p. 17.

Constitution-Making in Anglophone Africa: We the People?

91

ties. (Possible exceptions to this rule include Namibia and South Africa, where national liberation parties have recently come to power.) The first new demand of civil society is often to entrench the constitution-making process in a dedicated Act of Parliament or (perhaps preferably) in the Constitution itself. Laws may specify the institutions that will be involved in the process, timelines for achievement of goals, and requirements for public engagement. There may also be provisions—as in South Africa in 1993, and Kenya in 2001 and 2008—for a basic framework of values or principles that the new or revised constitution must enshrine.6 This demand is partly explained by the long-standing practice of establishing constitutional commissions under an Inquiries Act, a standard law inherited from Britain throughout Anglophone Africa. Under such laws, the executive (usually the Prime Minister or President) has complete discretion to establish a commission, determine its membership and terms of reference, and decide on which aspects of its findings to adopt and which to reject—or even to do nothing at all (Asare and Prempeh 2010; Simutanyi 2011). Under these processes, citizens are typically consulted but have no formal role in the process, and the government is under no legal obligation to take their views into account. From the perspective of the public, then, setting out the process in legislation specially designed for the purpose of governing constitutional reviews is seen as necessary for several reasons. First, by committing the government to a clear agenda, a proper legal framework makes the process more secure and reflects the need of governments to convince citizens of the seriousness of their intentions. In the absence of an agenda with reasonable timelines, the credibility of a process is likely to be limited. Second, through being stipulated by law, the constitution-making process is immediately more open. The law lays out the basic structure of the process: who is involved, what their authority is, when they will report, and, ideally, what opportunities the public has to engage. If a party to the process violates the agreed-upon agenda, others can challenge the violation in the courts and through the media. Third, the public typically demands that the law governing the constitution-making process must provide for inclusiveness and set out clear roles for citizens. A law can do this by responding to three of the most common demands: 1) a representative body, independent of Parliament, to draft the constitution or to discuss and approve a

6

Constitution of the Republic of South Africa, no. 200 of 1993, sch. 4; Constitution of Kenya Review Act, no. 13 of 1997, art. 2C; Constitution of Kenya Review Act, no. 9 of 2008, section 6.

92

Chapter Four

draft (or both); 2) meaningful public consultation; and 3) a final form of popular public endorsement like a referendum. Although civil societies in different countries share the ambition of establishing “people-driven” constitution-making processes, they have had quite different degrees of success in influencing the agenda. In the past decade, Tanzania, Zambia, and Kenya have exhibited surprisingly different outcomes, even though all three governments enacted specific legislation setting out processes for how different groups could come to agree on constitutional change. In each country, these laws started as an attempt by the executive to retain control. But both opposition parties and civil society fiercely contested them on the grounds that the lack of consultation was itself a symptom of their governments’ unacceptably unilateral approach. In response, each law was amended to some extent to secure a role for civil society and the general public in the process. In Tanzania, opposition parties and civil society rejected a 2011 bill for a constitutional review process prepared by the government on the grounds that its mandate was too limited, the process it established was not inclusive enough, and it did not include sufficient mechanisms for public participation. After a short stalemate the government compromised by passing an Act that met some of these demands.7 In 2003, Zambia started a process with a Statutory Instrument that established a Commission.8 Unlike in Tanzania, the government did not revise the law or otherwise respond much to civil society criticism either at this point or four years later, when it passed a law to establish a National Conference.9 Several key religious groups initially boycotted the process, which collapsed when Parliament rejected a proposed constitution in 2011. Apparently not learning from the mistakes of his predecessors, President Sata established a new commission (called a Technical Committee) at the end of 2011, again without consulting civil society in a meaningful way on either its membership or terms of reference (Chanda 2011). While the Committee has a mandate to engage the public, there is still no formal legal road map for the completion of the process, and as the publication of the promised revised draft is delayed again and again, civil society is once more demanding a formal commitment from government on completion of the process.

7

Constitutional Review Act, no. 8 of 2011 [Tanzania]. Statutory Instrument No. 40 of 2003 issued under the Inquiries Act, Laws, vol. 4, cap. 41. 9 National Constitutional Conference Act, no. 19 of 2007 [Zambia]. 8

Constitution-Making in Anglophone Africa: We the People?

93

Kenya offers a more complex case of a protracted contest over the agenda of constitutional change. In the five years following the 1992 elections, massive public campaigns pressed for constitutional reform. The history of civil society’s struggle for constitutional reform during this period is a complicated tale of shifting coalitions and many conferences that finally led to the 1998 Constitution of Kenya Review Act.10 The Act established the Constitution of Kenya Review Commission (CKRC) to be chaired by Yash Pal Ghai, a distinguished Kenyan constitutional lawyer. But the process set out in the Act, essentially devised by politicians, ensured that the government retained control of constitutional reform. Not surprisingly, it was contentious. At this point Yash Ghai intervened dramatically by refusing to be sworn in. The arrangement that was finally agreed to included 1) an expanded Commission including members drawn from civil society; 2) a set of egalitarian principles that emphasised the need for an inclusive process; and 3) the adoption of a draft constitution by a broadly representative National Constitutional Conference. Overall, the sustained demands of civil society over a long period led to significant adjustments to the executive-controlled agenda originally envisaged by President Moi. Perhaps the ultimate manifestation of public participation in the preliminary stages is a referendum on whether or not to proceed after an agenda and process have been agreed on. These referenda have been used in South Africa, Malawi, Uganda, and elsewhere to initiate (or halt) constitution-making processes. In 1992, South Africa’s white citizens were asked whether they supported the government’s ongoing negotiations with the ANC and other parties (Strauss 1993). In 1993 Malawian voters overwhelmingly approved a return to multi-party politics that resulted in a provisional constitution that in turn called for a National Constitutional Conference to create a permanent constitution.11 Formally, Uganda’s 2000 referendum on the retention of the “Movement” system of politics was similar, asking Ugandans whether constitutional change should be introduced. However, the referendum took place in oppressive political conditions, and just five years later, in a maneuver designed to retain power, Museveni secured constitutional amendments that made substantial changes to the constitutional arrangements (Tripp 2010, 170ff).

10 11

No. 6 of 1998. Constitution of the Republic of Malawi, no. 20 of 1994, art. 212.

94

Chapter Four

Writing a Constitution: the Forums There is no ideal model for an inclusive constitution-writing process. The actual processes since 1990 have varied greatly both in the kind of bodies involved and in their degree of, and dependence on, public participation to inform the substance of new constitutions. The drafting processes have ranged from single-forum processes where Parliament or a constituent assembly adopted a new constitution or amendments to an old one, to more complicated, multi-stage processes involving commissions and broadly representative conferences as well as Parliament. The drafters themselves were some combination of government-appointed experts and elected or nominated representatives. In addition, as we discuss in Part 4, the past twenty-five years have seen increasingly ambitious programmes for public participation during the process. Debate about who should make constitutions is as old as the idea of constitutions itself. Jon Elster (1995–96, 395) has recently restated the argument that the process should not be defined by regular politicians but by a specially elected body that drafts a constitution to bind a nation. This is an increasingly influential idea adopted, adapted, and advocated by civil society and opposition parties (at least until they come to power). Some form of constituent assembly is also likely to be more representative and less captive to the short-term interests of MPs. But as the Kenyan process suggests, the buy-in of political elites may be of equal or greater importance to securing a relatively disinterested assembly. Four types of questions arise when determining which forum should be responsible for drafting a constitution. First, on the assumption that the process should be initiated by the preparation of a draft, who should prepare it, and how should they be selected and appointed? Second, if some form of commission will prepare a draft, to whom should that body report? Third, what body or bodies should deliberate on and amend the commission draft? Lastly, who or what body should be the final decision-maker (a question we address in Part 5)? For advocates of public participation, a satisfactory answer to these questions must secure an inclusive process with adequate representation of marginalized groups and must include some way for the public to intervene meaningfully in each stage. Civil society’s most common demand since 1990 concerning constitution-writing bodies has been for them to be properly representative and, ideally, for new constitutions to be adopted not by Parliament (which is usually controlled by the executive) but by a constituent assembly. Nonetheless, in most Anglophone African countries, unless the constitutional reform is managed by a military government, Parliament has retained a

Constitution-Making in Anglophone Africa: We the People?

95

say (effectively a veto right). The political and formal/legal reasons for this persistent parliamentary control are interlinked. Since 1990, constitutional reforms in Africa have always taken place without a constitutional rupture where there is a period of discontinuity with a legally recognizable authority. Leaving important theoretical questions aside, in practice this has made it hard to persuade the governing elite to hand over authority to restructure the state. The incumbent governments’ appeal to constitutionalism as a shield against radical change is well illustrated in the Mung’omba Commission Report of the Zambian Constitution Review Commission (2005). The Zambian government argued, as have many other incumbent governments, that it would be “a betrayal of confidence of the electorate for Parliament to abdicate its authority to legislate” (706). The Commission countered by drawing on civil society arguments for participatory democracy: “recognising the value of democracy and reaffirming the supremacy of the people in governance and constitutionalism, the Commission [was] persuaded that the new Constitution should be adopted through a Constituent Assembly and a national Referendum” (808). In this case, the government compromised by establishing a National Constitutional Conference to debate the draft proposed by the Commission, yet insisted that, once “adopted” by the Conference, the Constitution should be “enacted” by Parliament. Nonetheless, although Parliament remains central, drafting processes have been more open and not infrequently have included bodies intended to better represent the citizenry. At least four different approaches emerge, three of which depend on commissions to do the preparatory work: 1) a drafting commission (the most traditional process) does the preparatory work, followed by deliberation and approval in Parliament, as in Kenya in 2009 and in Ghana at present; 2) a drafting commission followed first by a representative assembly and then by approval in Parliament, as in Kenya in 2000 and in Zambia (2003); 3) a drafting commission followed by a constituent assembly with the power to adopt the constitution or endorse it to present to the people in a referendum, as in Uganda 1995 (where the assembly was directly elected) and in Tanzania at present (where the assembly is to be composed of politicians and representatives of civil society); and 4) a single, elected representative body that is or becomes Parliament, as in Namibia (1993), South Africa (1996), and Zimbabwe (2013).

96

Chapter Four

Commissions The most common approach to drafting a new constitution is to use a commission for the preparatory work. The two main areas of contention in relation to commissions are their composition and the weight attached to their recommendations. A commission could be representative (like a legislature), including sitting members of parliament or specially elected members; it could be more impartial (something like a judiciary), composed of people with expertise, respected for their independence; or it might include a combination of people with “technical” expertise and people drawn from civil society to bring a deeper understanding of the social context within which the constitution must operate. Whatever the composition of a commission, however, the drafting remains in the hands of a relatively small group of individuals. The question is not whether the public can participate directly in the commission’s work, but whether the appointed members are seen as legitimate by the public, including minority and marginalized groups, and whether the commission will pay due attention to the views of the public. Despite acceptance of the need for public engagement in constitutionmaking processes, the old tendency of the government of the day to appoint a commission with very limited consultation persists. Constitutionmaking processes in Ugandan (1995), Zimbabwe (1999), Zambia (2003 and 2011), Malawi (2004) and Ghana (2010),12 among others, all proceeded with government-selected commissions. Such an executive-controlled approach worked in Uganda, where constitution-making took place under severely controlled conditions after a civil war. It may yet work in Ghana as there seems to be limited concern about the composition of the commission. It may also be possible for an executive-appointed commission to build popular legitimacy through its method of work and proposals, as appears to have been the case with the Mung’omba Commission in Zambia. However, these bodies now usually include some members drawn from civil society, and as we have noted, in some cases—most notably Kenya and Tanzania—civil society has been successful in demands that a commission should be more representative. The question of representation in small bodies is always difficult, but more consultative processes have been more successful in producing bodies that are broadly acceptable. In Tanzania, for example, resistance by opposition politicians and civil society to the first Constitutional Review Bill of March, 2011, which gave the 12

Constitution Review Commission of Inquiry Instrument, no. 64, 11 January 2010 issued under art. 278(1)(a) of the Constitution of 1992.

Constitution-Making in Anglophone Africa: We the People?

97

President broad discretion in appointing the constitutional commission, led to an amendment that required the President to invite civil society to submit proposals for commission members.13 Another approach was used to establish the Kenyan Committee of Experts in 2009—it was not the executive but Parliament that appointed the Committee in the highly charged political atmosphere created by the post-electoral violence of 2008. A multiparty parliamentary committee conducted a transparent process and invited applications, interviewed applicants, and selected the group carefully to include different sectors of Kenyan society. In at least one case— Zimbabwe, 1999—a government-appointed commission was a spectacular failure in part because the government failed to respond to public demands concerning the process (Hatchard 2001). The second issue of concern is the authority of the draft prepared by a commission. Is it to be a document that the government of the day can choose to use or discard as it wishes, or is it to be more influential in the constitution-making process? Sometimes, the government can freely ignore a draft produced by a commission. In Malawi, for instance, President Bingu wa Mutharika asked the Malawi Law Commission, a statutory body, to review the constitution. After two years and (apparently) a great deal of public consultation—including two national conferences—the Commission released a report (2007) with a draft constitution that successive governments have since ignored. For civil society, alert to the vested interests of an incumbent government, it is usually considered important for a commission’s draft to have more weight. The pressure is therefore for the draft prepared by a commission to be submitted directly to another, properly representative deliberative body for endorsement, as was done for example in Kenya (2000) and Zambia (2003). In a number of important examples, the draft produced by a commission has had special status in the next stage of preparation. For example, provisions in the Ugandan Commission’s draft could be revised or rejected only with the support of twothirds of the members of the Constituent Assembly.14 Kenya’s Parliament needed a 65 percent majority to reject any provision in the proposed constitution presented by the Committee of Experts, with agreement of key political leaders needed on contentious issues.15 Of course, the implications of such an arrangement are directly linked to the architecture of the entire process. In Uganda, the arrangement provided security for the government, preventing the larger body of the Assembly from abandoning 13

Constitutional Review Act, no. 8 of 2011. Constituent Assembly Statute, no. 6 of 1993, s. 17(3). 15 Constitution of Kenya, 1963, art. 47A(2)(b). 14

98

Chapter Four

proposals of the government-appointed Commission. In Kenya, by contrast, this arrangement balanced two things: the need to have a constitution that the political elite across Kenya’s considerable political divide would support, and the need to reassure the public that their interests as captured by the Committee would be respected. (Murray 2013).16

Conferences and Assemblies In the past, as noted above, governments in Anglophone Africa have assumed that a constitution drafted by a commission would be debated and adopted by Parliament—whether or not it was put to the people in a referendum. While preparation by a commission and endorsement by Parliament remains the dominant trend, civil society has demanded and, in some important cases, secured processes in which much deliberation is conducted in other bodies as well. These constitution-making bodies are usually composed of both elected and appointed representatives with a mandate to deliberate and sometimes approve the draft constitution produced by a commission. They have various names in Anglophone Africa. “Constituent Assemblies” tend to have the additional power to approve the final draft—in other words, excluding Parliament from the process, as in Uganda (1995) and Tanzania (2011).17 “National Conferences,” by contrast, generally have not had endorsing power. This was the case in Kenya in 2000 and in Zambia (2003). Although the name of the body does not always correlate to its final approval powers, we cautiously adopt this general distinction to stress two distinct models of non-parliamentary bodies. It is paradoxical, perhaps, that the trend in Anglophone Africa towards large representative bodies especially established to deliberate on a proposed constitution has its origins in the Nigerian military’s attempt to return to civilian rule in the 1970s. In 1975, Murtula Mohammed’s military government appointed forty-nine members to a Constitution Drafting Committee. The members included two representatives from each of the twelve states, eighteen academics, and some lawyers and businessmen— 16

The Kenya process was particularly involved. The constitution-making bodies involved were limited in their mandate and could not interfere with matters settled in the earlier (2000) process. 17 Various countries including Ghana and Tanzania converted their parliaments into Constituent Assemblies to amend or replace their post-independence constitutions to become republics. But those Constituent Assemblies were simply parliament operating under a different name and quite dissimilar to the ones we discuss here.

Constitution-Making in Anglophone Africa: We the People?

99

but no representatives from the military (Koehn 1989, 407). The Committee submitted its report and a draft constitution to the military in September, 1976. Around the same time, each state established local authority councils with three-quarters elected (and non-partisan) members, who would in turn elect most members of a proposed Constituent Assembly. For the first time in Anglophone Africa, a constitution-making process had specially elected representatives (as opposed to ordinary parliamentarians) whose sole mandate was to draft a constitution. The military government retained control of the process by certifying 341 eligible candidates for the Constituent Assembly. The local councillors then sat in electoral colleges to elect 203 of those candidates, with an additional twenty members appointed by the government. The Assembly held sessions over the course of almost a year and then approved the draft constitution without making substantial changes. However, the military reserved for itself the full right to revise and promulgate the new constitution, and did so: seventeen unilateral amendments were made to the draft before it was promulgated in 1978. The Nigerian Constituent Assembly, elected from pre-selected candidates, could not be described as an exercise in democratic public participation, yet it did endorse some form of representation in a special body mandated to draft and approve a constitution (subject to the revision and promulgation by the military government). Uganda’s Constituent Assembly, established by statute in 1988,18 had more popular origins. In the Commission’s extensive public hearings, people consistently opposed adopting a new constitution through Parliament, preferring a Constituent Assembly (Brandt et al. 2001, 349). The government responded by establishing a Constituent Assembly composed of 214 members elected directly by Ugandans from existing electoral districts, as well as 74 members representing women, the military, political parties, and other interest groups (Odoki 2005, 265).19 This experiment extended the Nigerian model by adding special interest groups to the people’s elected representatives. Again, conditions in Uganda meant that the process was not entirely open, but while the elections were formally on a “no-party” basis, in practice voters knew which candidates represented which parties (Oloka-Onyango 1995, 175). So despite non-party claims, the Constituent Assembly elections were the first de facto multiparty elections of the new Ugandan political order. This effectively infused a measure of representation into a constitution-making process. 18 19

Uganda Constitutional Commission Statute, no. 5 of 1988. Constituent Assembly Statute, No. 6 of 1993.

100

Chapter Four

Kenya’s National Constitutional Conference was twice as large and even more elaborate than the Ugandan forum. It had over six hundred participants, including all 210 members of Parliament; three delegates from each district (totaling 213); 42 political party representatives; 125 representatives of religious groups, women, youth, the disabled, trade unions, and various NGOs; and the members of the Commission that had prepared the draft (Cottrell and Ghai 2004, 7). The government was clearly outnumbered in the conference, and a two-thirds majority was required to amend any provision of the Commission’s draft, which meant that decisions had to be reached by agreement transcending party, ethnic, and sectoral lines. The constitution-making process, as Bannon (2007, 1833) comments, “was designed to avoid partisan capture and to reflect the will of the populace.” Yet it was Parliament, not the Conference, that had the power to endorse the constitution for submission to the referendum. The government used that opportunity to amend substantially the Conference’s final proposed constitution. The Kenyan model has been influential. Both Zambia and Tanzania have drawn upon it in designing their conferences. However, establishing a representative body is not easy—and not always the intention. For instance, when the Zambian government, in partial capitulation to the demands of civil society, established the National Constitutional Conference in 2007, it drew in a wide range of actors nominated by their respective institutions.20 But the Conference was controversial throughout its existence, in part because some sectors of civil society considered it heavily biased towards the government. Constituent assemblies, that is, bodies that not only debate but also approve the constitution or endorse it for referendum, are rare: Namibia (1990) and Uganda (1995) are the only two examples in Anglophone Africa thus far.21 But Tanzania’s new process will have a constituent assembly rather than a Parliament-approved constitution. The assembly is to be composed of all members of the national (357) and Zanzibar (82) legislatures, plus 160 members of civil society groups.22 The governing Act promises broad representation within the civil society group (with attention paid to the proper representation of Zanzibar), but the President can appoint the civil society members by proclamation with no further consul-

20

National Constitutional Conference Act, no. 17 of 2007, s. 4. As we note above, earlier Constituent Assemblies were simply parliaments acting under a different name. 22 Constitutional Review Act, Laws, ch. 83, s. 22. 21

Constitution-Making in Anglophone Africa: We the People?

101

tation. The final approval of the Constituent Assembly draft will be by referendum.

Parliaments as constitution-making bodies As noted above, Parliaments usually play a central role in constitutional amendment in Anglophone Africa. But, in recent times, if fundamental constitutional change is envisaged, usually another body (or other bodies) prepares the changes. South Africa and Zimbabwe are exceptions: in the tradition of the American and French experiments two centuries earlier, although in rather different contexts, both had their legislative bodies act, in effect, as constituent assemblies. In South Africa, the constitutiondrafting process was immediately preceded by an election for a Parliament/Constitutional Assembly. No formal attempt was made to provide representation for particular groups, although the system of proportional representation allowed the largest party, the African National Congress, to ensure that a significant number of women were included in the Assembly. Fair elections, a widely supported negotiation process, and credible political parties were the basis for legitimate constitution-making and alleviated the need to design a body with special representation for different sectoral groups. Moreover, as in the case of the Namibia, the legitimacy of the process was increased because the African National Congress (the main liberation movement) dominated the elections but did not capture enough votes to write the constitution without the support of other parties (Ebrahim 1998; Cottrell 1991, 56–57). Although it too was a process run through Parliament, the recently concluded process in Zimbabwe was rather different. The constitutional crisis was precipitated by a failure to hold free and fair elections, which then led to a new constitution-making process as part of the broader “Global Political Agreement” (GPA) between the two major political parties— the Zimbabwe African National Union-Patriotic Front (ZANU-PF) and two factions of Movement for Democratic Change (MDC). Article 6 of the GPA set out a process by which the Parliament elected in the contested election would create a new constitution. The constitution was drafted by a Select Committee of Parliament (COPAC), debated in Parliament, agreed upon in a national referendum, and then finally approved by Parliament. Although the GPA demanded public engagement through a large All-Stakeholders Meetings and public outreach (discussed in the next section of the chapter) and, in a novel development, through the establishment of thematic committees under COPAC (composed of members of parliament and civil society), there could have been no question

102

Chapter Four

that the goal was a political deal. Every aspect of the process was agreed upon in carefully calculated negotiations among the three key parties, and while the involvement of civil society drew the approval of international observers and appeased civil society to some extent, for each side it was also a means of mustering broader support.

Participating in a Process: Public Inputs Although civil society has had mixed success in securing representative constitution-making bodies, it has had success with ensuring that the public and civil society groups can engage with the process. Every constitution-making process now includes some form of relatively expansive public engagement to educate people about the process as well as canvas their views. In many cases these processes have been astonishingly ambitious and received a massive public response: the Ugandan Commission gathered 15,693 memoranda from the various resistance councils and 25,547 in total; Kenya’s Committee of Experts (2009) received nearly forty thousand submissions during the thirty-day period public comment period; Ghana’s Constitution Review Commission (2010) received over seventy thousand submissions over the course of its work. The mix of approaches used in these processes reflects the great creativity of the many groups involved. In addition to formal channels such as the media and public meetings, constitutional messages have been conveyed in theatre, poetry, art and cartoons and, more recently, through text messages, Facebook, and Twitter. Strategies of inclusivity have been devised to give special groups opportunities to participate, and many programs have sought to engage minorities, rural and urban poor, and others who have been historically excluded from the public sphere. These processes have usually generated energetic public debate, which in countries emerging from oppressive rule has often been the first opportunity for citizens to discuss constitutional and political options freely (Ebrahim 1998, 244; Jeffries and Thomas 1993). In addition, sometimes the civic education that accompanies the process has had lasting effects on the citizens themselves (Moehler 2008). The role of constitution-making bodies in running public participation programs and collecting the views of citizens has varied. Some processes have relied heavily on elected local bodies, which act as intermediaries between a commission and the public by transmitting peoples’ views to a national constitution-making body while also educating citizens about the decisions made and the implications of those decisions. This model is exemplified by the processes leading up to the Ugandan Constitution of

Constitution-Making in Anglophone Africa: We the People?

103

1995, which in turn resembled the Nigerian process for its 1979 Constitution. More often, public participation is managed by the body mandated to draft the constitution—usually a commission but sometimes also a deliberative assembly, and, as in South Africa and Zimbabwe’s most recent process, an elected constitution-making body. However, a number of the programs run by constitution-making bodies have also drawn on local representative bodies to encourage a more “bottom-up” approach. In Uganda, the National Resistance Movement (NRM) structured the participation process on the five-level local government system (Moehler 2008; Cullimore 1994). In the post-conflict years of the late 1980s, these elected councils had a great degree of popular legitimacy, especially at the lower village and parish levels. The Constitutional Commission first visited every parish in the country to gather people’s views, which were used to develop “Guidelines on Constitutional Issues” to structure later consultations. From 1989 to 1992, the Commission held eighty-six district seminars and educational forums in all 870 sub-counties and then returned to each sub-county to collect over twenty-five thousand oral and written memoranda (Furley and Katalikawe 1997). The key players in the public meetings were the parish council executives, who gathered opinions and prepared reports for district resistance councils (Oloka-Onyango 1995). The danger was that NRM cadres who controlled the councils would vet the public’s views, a problem similar to that faced by Zimbabwe’s COPAC consultations, and there were some complaints that views were silenced. But local government councillors had strong ties to their communities, and at least in NRM-supporting areas, played an important role in marshalling citizens to contribute. The processes in South Africa (1996), Kenya (2000), Ghana (2011), and Zambia (1993, 2003, and 2011) exemplify the more common approach in which the constitution-making body itself manages public participation. In all these cases except South Africa, the regulating law demanded public consultation. South Africa’s distinctive process was based on explicit elite negotiations within an elected assembly with public participation as a secondary concern. The main political parties made the key decisions in relatively closed committees composed of a few negotiators with clear mandates from their respective parties. In the first stage leading up to the 1994 elections, there was no process of public participation at all, and interest groups had to rely on their political allies in the negotiations to represent their views. After the elections, however, the process for drafting the “final” constitution grew to include an extensive public participation program, for which South Africa became famous. It involved an extensive media campaign, a strong educational element, public meetings

104

Chapter Four

specifically designed to reach marginalised and other disadvantaged people, and a process for soliciting submissions. The process included over twenty-six public meetings involving over two hundred Constitutional Assembly members and 20,549 people, and 717 organizations participated in related workshops (Ebrahim 1998, 244). The overall goal was to ensure that all South Africans had a sense of ownership in the constitution (Murray 2001, 817–18, Skjelten 2006). Public participation has been a central and more considered element in more recent processes. For Kenya, the Constitution of Kenya Review Commission (CKRC) started with a huge public education campaign (using both extraordinarily energetic NGOs and its own resources). It established constitutional forums in each of 210 electoral constituencies led by locally elected leaders and supported by a Commission-appointed coordinator (Cottrell and Ghai 2004, 6–7). These forums facilitated the wellattended meetings held when the Commissioners later visited each constituency. The meetings themselves were inclusive of many groups marginalized in the past—women, ethnic minorities, and the disabled. In addition, the Commission received over 37,000 written submissions. Engagement with the public continued through the National Conference, where civil society was well represented and generous access was given to observers (Brandt et al, 340). Overall, the process under the CKRC set high expectations for public engagement. Thus, although the statute under which the 2009 Committee of Experts acted required only limited public participation, the Committee judged that its legitimacy depended in good part on an adequate process of public consultation. The huge number of submissions it received (over sixty thousand total) attests to the importance Kenyan citizens placed on consultation.23 Another element in the public consultation process, for both Ghana’s Constitution Review Commission and Zambia’s new Technical Committee on Drafting the Zambian Constitution (TCDZC), was “national conferences.” Neither of these was a conference in the Francophone (or Kenyan) sense of a deliberative body with decision-making power. Instead, they were forums at which, in the case of Ghana, submissions and, in the case of Zambia, the TCDZC’s draft constitution were discussed. In Ghana in early 2011, at least two thousand people gathered at a five-day National Constitution Review Conference aired live on national television. The Conference was the climax to the Commission’s extensive consultations, where submissions the Commission had received were debated. In their 23 Final Report of the Committee of Experts on Constitutional Review, 2010, pp. 30, 41, 81.

Constitution-Making in Anglophone Africa: We the People?

105

final report several months later, the commissioners stated that their recommendations were largely based on the outcomes of the Conference.24 Zambia’s National Convention of April, 2013 likewise discussed submissions to the Commission. In this case, however, submissions had been filtered through both local and provincial meetings. In addition to soliciting written submissions electronically and by mail or by delivery to its regional offices, the TCDZC had appointed district facilitators and held eighty district consultative forums to discuss its first draft constitution. Matters raised at these forums were then deliberated upon at ten Provincial Constitution Conventions and finally at the National Convention.25 Zimbabwe’s two All Stakeholder Conferences were intended to play similar roles—the first would provide a sense of the direction constitution-making should take and the second, when the proposed constitution was on the table, would be a way of testing its acceptability. In these ways, citizens, usually with the generous support of foreign donor agencies, have been drawn into constitution-making processes in a trend that is unlikely to be easily reversed. But there is also plenty of evidence for sceptics of broad public consultation. The Zambian processes from 1993–97 and 2003–05 included extensive public participation, but failed to produce a new constitution. The protracted Kenyan process discussed above finally succeeded with the development of an agenda that brought politicians back in and created powerful incentives for them to work together. Public consultation played a valuable, but not central, role in reaching a successful conclusion. In addition, in Anglophone Africa as elsewhere, participation opportunities are often designed to elicit the unmediated views of citizens, but in fact submissions are seldom politically neutral—nor should they be expected to be in the politically charged environment of most constitution-making. Thus, as Moehler comments in the context of Uganda, public participation can be divisive; if a process of public consultation becomes extremely politicised, its legitimacy may be undermined. The difficulties of managing the use (and sometimes manipulation) of consultation processes by incumbent governments and political parties (and even by international donors and civil society) to advance partisan interests are significant. Zimbabwe’s most recent process illustrates starkly the challenges and dangers of public consultation in a highly polarized and unstable political environment. As noted above, under the Global Political Agreement, the 24

Report of the Constitution Review Commission, 2011, passim. For example, see the National Constitutional Conference website . 25

106

Chapter Four

drafting process was to be framed by two All Stakeholders Conferences and informed by public hearings and consultations. The first All Stakeholders meeting, in July, 2009 was composed of four thousand MPs, civil society representatives, veterans, and other groups and ended in chaos and violence (Sokwanele 2012, 22). The public consultations involved meetings in each of the 1,857 wards. ZANU-PF launched Operation Vhara Muromo (“Shut Your Mouth”) in an orchestrated attempt to control these ward meetings. In Harare, for example, meetings were disrupted by interruptions and coached participants. There were several logistical and technical problems and the process of collating all the reports into a central database was criticized for loss of data, tampering, and missing diaspora reports. Also, the COPAC co-chair, an MDC-T member, was held by the police for three weeks on dubious charges, reinforcing a growing suspicion that the aggregate results of the participatory process were either biased or unrepresentative. Results were analysed by multiparty thematic committees, but there were major methodology disputes (with significant political ramifications) between ZANU-PF and MDC. Presumably confident that submissions were skewed in its favor by supporters coached in Vhara Muromo principles, ZANU-PF insisted that submissions should be treated as a poll on certain central issues. The second (violence-free) All Stakeholders meeting, held in October, 2012, led to general agreement with a draft. A list of thirty outstanding (and contentious) issues was left to the coalition cabinet to resolve, and critics have pointed out that almost every issue was resolved in favor of ZANU-PF.

Ratification: Approving a Constitution The last step before a constitution is put into effect is some kind of formal approval. The general forms of approval include adoption by the legislature or a constituent assembly, or approval in a referendum after endorsement by another body. Independence constitutions, as discussed above, were creations of the British government. Their approval was legally, if not politically, at the sole discretion of the British government. The basic structure for future amendments was similar in all these constitutions and usually required no more than approval by Parliament (usually but not always with a two-thirds majority). In eight of the fourteen Anglophone countries, Parliament had the final say on all constitutional amendments— even if the amendments purported to replace the entire constitution. Only six countries required a referendum, and then only to amend select, especially sensitive provisions.

Constitution-Making in Anglophone Africa: We the People?

107

In this context, it is not surprising that civil society has insisted that the public should have the final say on amendments that affect fundamental constitutional arrangements. Part of the turn to public participation after 1990, this call for referenda has made it more difficult for a process of fundamental constitutional reform to end in a simple legislative amendment (or unilateral decree by a military government). This is reflected both in the amendment provisions in new (or recently reviewed) constitutions and in recent practice. The constitutions of eleven of the fifteen Anglophone countries (including Namibia) now require a referendum to ratify changes to some aspects of their constitution.26 The exceptions are South Africa, Nigeria (which moved away from the referendum requirement included in its independence constitution), Tanzania (where the new draft constitution proposes a referendum for amendment of certain elements of the constitution), and Sierra Leone (which requires an election before fundamental constitutional changes can be implemented). In many countries, moreover, constitutional provisions for proposing amendments have been strengthened to include more opportunity for public debate. Since 1990 there have been two waves of referenda adoption in nine countries. The first wave, in the 1990s, was focused on the question of whether to return to multi-party democracy after a long period of oneparty or military rule in Sierra Leone (1991), Ghana (1992), Malawi (1993), Sudan (1998) and Uganda (2000, 2005). Voters were asked to approve a constitution prepared under the direction of the existing singleparty or military government. (In some cases, as discussed above, voters were asked to approve the principle of multi-partyism in advance and so prepare the way for a constitution-making process.) Sudan also held a referendum in 1998 in which nearly all voters apparently approved a new constitution. The second-wave referenda were to approve comprehensive new constitutions in Seychelles (1992, 1993), Gambia (1994), Zimbabwe (2000, 2013), and Kenya (2005, 2010). The Seychelles transition from a one-party state was less smooth (Hatchard, Ndulo, and Slinn 2004, 41). The constitution-making process required the approval of 60 percent of voters on a referendum for passage. Dissatisfied with the process, opposition groups campaigned for a “No” vote, and only 53.7 percent of voters approved the constitution. After this rebuff the Constitution Commission resumed its work and redrafted the constitution to strengthen multiparty provisions, which easily passed a second referendum with 73.6 percent in favor. 26 If the proposals in the draft Tanzanian constitution are adopted, it will be 12 out of 15.

108

Chapter Four

The failed first Seychelles referendum foreshadowed two later referenda—in Zimbabwe in 2000 and in Kenya in 2005—where voters rejected a constitution that was driven, or highjacked, by the executive. In Zimbabwe, President Mugabe promised a national referendum before any final constitution would be placed before Parliament. This referendum promise was misleading for two reasons. First, legally it was a mere consultative exercise since the government was not bound by its result. Second, the President could (and did) put forward the constitution he wished instead of the draft recommended by the Constitutional Commission. Not only did this undermine the legitimacy of the participation process, it also allowed the executive to ignore all the provisions in the Commission’s draft derived from popular participation of ordinary citizens (Hatchard 2001, 214). The referendum result was a clear defeat for the President as over half the paltry 26 percent of registered voters who turned out rejected his proposed constitution. As early as 1997 in Kenya, opposition leaders rejected the idea of Parliament being the only legitimate body to approve a new constitution (Stacey 2011, 595–96). And although the constitutional requirement that Parliament approve any amendments was retained in the process initiated in 2000, the public understanding was that Parliament would accept whatever constitution was adopted by the National Conference (Cottrell and Ghai 2004). But the change in government after the 2002 elections and a breakdown in consensus-building in the Conference changed the political environment completely, and the government amended the Conference draft substantially before presenting it to Parliament. It was the government draft, duly endorsed by Parliament, that was finally put to the people—and rejected—in the referendum. The high-profile and high-stakes referendum failure in Kenya led to major changes both in the legal framework for and attitudes towards popular ratification in the follow-up process. Before the reform process was revived in 2009, the Constitution itself was amended to require a referendum to approve a new constitution (the previous referendum had been added to the process by a High Court decision).27 In addition, as we describe above, politicians were brought back into the process and the role of Parliament as endorsing body was circumscribed so that it could not easily overturn political agreements.28 Thus the Kenyan constitution submitted to

27

Constitution of Kenya (Amendment) Act, no. 3 of 2008; Njoya and others v Attorney-General and others [2004] LLR 4788 (HCK), p. 17. 28 Constitution of Kenya, 1969 (as amended), art. 47A.

Constitution-Making in Anglophone Africa: We the People?

109

the people in the 2010 referendum was supported by a substantial section of the political elite and, as a result, approved by 68 percent of the voters. In Zimbabwe in 2013, Parliament still had the final say, but the Global Political Agreement formally required a referendum, and it was the constitution that was approved in the referendum that was presented to and adopted by Parliament. As in Kenya, the proposed constitution was agreed upon by the major opposing sides, and they undertook to support it in referendum campaigns. As in Kenya, with the major political parties campaigning for it, the proposed constitution was approved (by 94 percent). For many, a referendum is considered the most legitimate or even the only acceptable conclusion to a constitution-writing process. However, referenda were not held in three of the most successful cases of constitutional reform in Anglophone Africa: Namibia (1991), South Africa (1996), and Uganda (1995). In Uganda, as discussed above, the Constituent Assembly was established by popular demand. While this Assembly was the final decision-making body, there were two important restrictions on its power. First, the process stipulated that, if a two-thirds majority approval could not be reached on any part of the constitution, the matter was to be put to the people in a referendum. Secondly, the President had the discretion to call a referendum before promulgating the constitution. However, neither of these provisions was invoked (Brandt et al. 2001, 349). There is little doubt that had a referendum been called, the Constitution would have been approved. In Namibia, the directly elected Constituent Assembly had full control of the process—from drafting to adoption. Here, in contrast to most of the other post-1990 processes, citizens had confidence in their political representatives (Wiechers 2010, 97–98). The fact that it was political parties and not civil society groups that were most active in developing constitutional demands in the period leading up to the elections for the Constituent Assembly contributed to this. Accordingly, popular ratification was not an issue. Like Namibia, South Africa’s truly representative political parties and popular leaders gave the elected Parliament/Constitutional Assembly legitimacy and there was no call for a referendum. But in South Africa, a mechanism was required to secure the elite deal created prior to the 1994 elections. Rather than a referendum, which is an essentially majoritarian tool, the South Africa process concluded, uniquely, with certification of

110

Chapter Four

the final constitution against a negotiated set of principles set out in the Interim Constitution by a newly created Constitutional Court.29

Conclusion Constitution-makers in Anglo-African countries can no longer escape calls to include the public in the process of starting, writing, and approving new constitutions. This appears to be a welcome change for several reasons. First, participation promises to infuse new constitutions with a legitimacy not accorded their colonial predecessors. If a constitution is a form of “social contract,” as often portrayed by civil society, then it is more legitimate when more citizens participate in its making. (The reverse claim, that constitutions lack legitimacy if there is little or no public engagement in their making, is common in postmortem critiques of African independence constitutions.) The new constitution-makers in Kenya, Tanzania, Uganda and elsewhere also placed great store in the pedigreed notion of the people constituting the state, which justified calls for the people to have a greater role in legitimizing a new constitutional order (Cottrell and Ghai 2004, 11; Widner 2001, 98–99; Odoki 2005, 155–76). Thus the legitimacy of the state was grounded in the free will of the people, who also played a decisive role in creating their own governing law. The expansion of participatory forums (discussed in the fourth section of the chapter) may also promote the formation of individuals as citizen with an interest in fostering and protecting a national culture of constitutionalism and the rule of law. Participation is a means to secure the consent of citizens and to educate them in their civic duties to uphold the new constitution. Moehler (2008) has suggested that the public participation process in Uganda created more skeptical citizens; as a result, Ugandans became “distrustful democrats” who were both more demanding of their government and, perhaps, deferential to the constitution itself. Likewise, Jesse Majome, as spokesperson for Zimbabwe’s COPAC has said, “I do not think that any process at all by which Zimbabweans actually sit down under a tree or in a hall and stand up to say their views about how they want Zimbabwe to be run is a wasted process. That process itself is actually another means to national healing.”30 As the protracted Kenyan process 29

In re Certification of the Constitution of the Republic of South Africa, 1996, (1996) 4 S.A.L.R. 744; In re Certification of the amended text of the Constitution of the Republic of South Africa, 1996, (1997) 2 S.A.L.R. 97. 30 “COPAC Process a Circus: Madhuku” Zimbabwe Independent 15 July 2010. http://www.theindependent.co.zw/2010/07/15/copac-process-a-circus-madhuku/

Constitution-Making in Anglophone Africa: We the People?

111

shows, an obviously inclusive process of public participation may also contribute to building national unity (even when people disagree) by strengthening a long-persecuted civil society. The process opens up political space for civil society and promotes its growth by strenuous demands. A participatory process may also increase the chance that a new constitution will be implemented and accepted by participants, as when civil society managed to force incumbent governments to set down an agenda by law. Public participation may have two other important roles in the particular context of African states. First, public participation, and the open process it signals, may enrich the text of the constitution itself by enhancing the common good with mechanisms to protect rights and hold the government accountable (though analysis of this potential role is beyond the scope of this chapter). Brandt et al. (2001) point out that a process of public consultation may expand the political agenda to include issues that incumbent politicians may wish to avoid. Sometimes public consultation has prevented self-interested politicians from holding the process captive by restricting it to the old method of simple parliamentary amendment. Second, the depth and breadth of inclusion deserves special attention. Elkins et al. (2009, 208) found that inclusive processes corresponded to more successful constitutions—provided constitution-making did not devolve into endless negotiations. Groups excluded in the past based on colonial categories of race, “tribe,” or gender have an opportunity to participate in making a new constitution. Women in particular have been recognized as major players in several processes. In Uganda, for example, women and women’s groups played a key mobilizing and educational role in the decade-long process. If a constitution’s legitimacy is a function of popular input, then it is necessary to ensure that women have a proportional influence on the process. Some authors have also argued that women in general emphasize collectivism and dialogue in politics (Hart 2003, 11). Thus women may enrich the constitution, as would other groups formerly excluded on the basis of race or “tribe.” In countries where government was synonymous with a certain group, like the white minority in apartheid South Africa, including oppressed minorities (or even majorities) is critical to let every citizen participate in the process. Despite all this evidence in favor of participation, there are grounds for skepticism. To what extent should public participation override other considerations when drafting a new constitution? Some countries have managed relatively well without public participation in the constitutionmaking process—Germany, Japan, and arguably Botswana. In South Afri-

112

Chapter Four

ca, public participation was undertaken only as an ad hoc afterthought, after the constitution-making process was well underway. While it is still too soon to evaluate the success of constitutions written since 1990 in relation to their degree of public participation, there are further concerns. Participation will not necessarily enhance the common good since it cannot completely insulate itself from the self-interest of participants (Blount 2011, 40). The virtue of civil society is often contrasted to the vice of politicians. Public participation is then presented as a way of uncovering national interests and needs untainted by partisan interests. But civil society and the general public are never totally free from partisan politics. They may be willing or unwitting instruments of political interests. Nor is civil society monolithic. The term describes a collection of groups with complementary and contradictory interests. Factions within Kenyan civil society, for instance, held conflicting notions of social justice relating to the question of access to land. This had dangerous undertones in Kenya, with a colonial legacy of “tribalized” access to land. There is a related danger that governments will manipulate the public to legitimize undesirable constitutional changes. In the recent Zimbabwean process, for instance, ZANU-PF successfully influenced public submissions through a carefully orchestrated, nation-wide plan. All these questions reflect both the optimism and disappointment of those who see civil society engagement and public participation as a form of democratic politics that might dislodge or at least unsettle the self-interested politics of incumbent (and aspirant) political elites. They also point to the difficult relationship between civil society activism in Africa and decision-making by political elites and, perhaps, to the limitations of participatory democracy in the big power struggles that serious constitutional reform inevitably involves. A final danger of a highly participatory process is that it may marginalize elites. Horowitz and Cottrell and Ghai, among others, suggest that too much participation with too many actors is more likely to lead to an unworkable constitutional system: rather than enriching the design and securing elite buy-in, over-participation may weaken the resulting constitution. There is the related problem that agreement is more difficult with more actors, and, particularly where compromise is necessary, an open process may make it difficult, even impossible, to reach an agreement. This risks marginalizing or excluding politicians, which might have the perverse outcome of undermining the hard compromises necessary to achieve a large consensus on delicate questions. Too much openness might lead to deadlock and, in worse case scenarios, political violence. Ghai and Galli (2006) add the warning that referenda, by their very yes-orno nature, are especially divisive. This leaves a final observation that for

Constitution-Making in Anglophone Africa: We the People?

113

all the potential good of a participating public, some of the most successful constitution-making processes have carefully circumscribed public participation to allow political actors the space—and sometimes the secrecy—to compromise.

CHAPTER FIVE MEASURING THE PERSUASIVE EFFECTS OF ELECTORAL CAMPAIGNS IN AFRICA JEFFREY CONROY-KRUTZ

Since 1990, multi-party elections have been held in forty-six of subSaharan Africa’s1 forty-nine countries.2 These elections include over 160 in which citizens have directly elected a president and 190 for national legislatures. Elections are thus hardly rare occurrences in Africa. Despite this, relatively little systematic research has been conducted about the campaigns that precede them. In the weeks and months leading up to balloting, candidates and parties organize and advertise, often fervently, in order to maximize their vote totals. Candidates speak, distribute goods, and provide entertainment at rallies and marches; organize the faithful for local get-out-the-vote efforts; and craft exhortations for various media, ranging from radio and television spots, to billboards and posters, to clothing and calendars. Democratic competition is hardly fathomable without such pre-election campaigning, yet political scientists know relatively little about the particular functions that such activities serve in Africa. Conventional wisdom holds that, since electoral behaviour in Africa is driven primarily by ethnic considerations and clientelism, campaigns are mainly designed to mobilize latent support. The existing literature has at least three significant shortcomings, however. First, it largely fails to measure systematically the extent to which campaign activities in Africa actually mobilize political participation by communities and individuals. Second, the notion that electoral campaigns in Africa are largely mobilizational in nature, rather than persuasive, is based on relatively little empirical evidence. In other words, to what extent do citizens’ electoral preferences shift during an electoral campaign? Third, the possibility that campaigns might have informational functions—i.e., improving citizens’ 1 2

Hereafter in this chapter referred to as “Africa.” The exceptions are Eritrea, Somalia, and Swaziland.

Measuring the Persuasive Effects of Electoral Campaigns in Africa

115

knowledge about candidates, political issues, and democratic competition more generally—remains unexplored in relation to Africa. This chapter focuses specifically on the second question. In order to measure possible persuasive effects, the author conducted the first-ever panel survey on campaign effects in an African country, with 675 individuals interviewed at three points during the campaign process surrounding Uganda’s February/March 2011 elections. The overall project is, to the author’s knowledge, the first to measure the various types of possible campaign effects in an African context, and as such it contributes to filling a lacuna in our understanding of electoral competition in developing democracies. The primary goal is to identify and measure possible mobilizational, persuasive, and informational effects of campaigns in Africa. To the extent that these campaigns encourage political participation, critical engagement, and information sharing, they are crucial to democratic development. With respect to persuasion, a substantial literature on campaign effects in established democracies—primarily utilizing data from the United States, but also from Canada, Germany, and the United Kingdom in the developed world, and Brazil and Mexico in the developing world— debates the extent to which campaigns actually matter, as far as determining election outcomes. A “minimal effects” thesis has emerged from many of these studies, using both panel and rolling cross-sectional survey data, suggesting that, if campaigns do have much of an effect, it is primarily mobilizational, rather than persuasive. The project studies individual-level campaign effects in a single country—Uganda—through a three-wave panel study on presidential, parliamentary, and local elections, which took place in February and March, 2011. The first wave of the survey was conducted at the outset of the campaign season (late November/early December 2010), with 1,072 Ugandans in nine different parliamentary constituencies, covering all major regions of the country. Of these, 721 individuals were successfully interviewed a second time, near the end of the campaign (late January/early February 2011), and a third set of interviews was conducted with 675 of these participants after all presidential, parliamentary, and local balloting had been completed (late March/early April 2011). A number of findings are worth highlighting at the outset. First, a substantial proportion (40 percent) of respondents reported having voted differently in 2011 than they had in 2006, while a significant minority (34 percent) changed their stated presidential preference at least once over the course of the panel study. Second, the electoral success of the incumbent, President Yoweri Museveni, who increased his vote share between 2006

116

Chapter Five

and 2011 by 9.1 points, seems to have stemmed from his ability to hold a significant majority of his base from 2006, while steadily adding former opposition voters and undecideds to his sizable coalition. The chapter proceeds in four parts. The first section briefly reviews the literature on campaign effects and presents expectations regarding the magnitude of persuasive campaign effects in developing democracies such as Uganda. The second section presents the case of Uganda and describes the data-collection process. The third section presents results of these analyses, while the fourth section addresses conclusions.

Persuasive Campaign Effects in Africa? While study of election campaigning in the developing world is a relatively new field, campaigns have long been a favoured topic of political scientists focusing on the United States. Many studies have concluded that campaign effects were minimal, at least beyond mobilization.3 The classic Columbia studies, which followed the 1940 and 1948 presidential campaigns, found that few of the citizens under examination were persuaded to vote for one candidate over another because of campaign communications or events; rather, the main effects of the campaign were to minimize defections amongst weak partisans and encourage their turnout (Berelson et al. 1954; Lazarsfeld et al. 1948). Subsequent work cited the longitudinal stability of partisan attachments as contributing to the low probability that most voters could be persuaded by campaign communications (Campbell et al. 1960; Converse 1962), while even later analyses of panel data echoed the Columbia findings, suggesting that campaigns primarily mobilized partisans, or at best “activated” latent predispositions (Finkel 1993). A number of studies have found at least marginal persuasive impacts, however (Herr 2002; Holbrook 1994, 1996; Huber and Arceneaux 2007; Johnston et al. 2004; Petrocik 1996; Shaw 1999b; Shaw and Roberts 2000). Even though Africa is home to more new electoral democracies than any other region of the world, campaigns there have not received significant attention in the political science literature, at least in terms of their potential persuasive effects. Myriad works have covered elections in specific cases in significant depth; these include, inter alia, studies of Ghana (Aubynn 2001; Nugent 1999), Kenya (Throup and Hornsby 1998), Liberia 3

For treatments of the “minimal effects” model, see Abramowitz 1996; Bartels 1992, 1993; Brady et al. 2006; Campbell and Mann 1996; Finkel 1993; Gelman and King 1993; Goldstein and Ridout 2004; Iyengar and Simon 2000; Katz and Lazarsfeld 1955; Klapper 1960; Markus 1992; Rosenstone 1983.

Measuring the Persuasive Effects of Electoral Campaigns in Africa

117

(Harris 1999, 2006), Nigeria (Kew 2004; Miles 1988), Rwanda (Meierhenrich 2006), Senegal (Beck and Conroy-Krutz 2011), Sierra Leone (Kandeh 1998), South Africa (Reynolds 1994), Togo (Seely 2006), and Zambia (Burnell 2002).4 The focus of such works is often on inter- and intra-party (read, elite) politics in the run-up to elections. Missing from most of the literature on African election campaigns has been empirical data on the effects of those campaign activities on individual voters’ behaviours and attitudes, and those that have examined such issues have tended to focus on mobilization (or demobilization), whether in terms of party activities (Bratton 1999; Bratton et al. 2005: 296–99; Kuenzi and Lambright 2007), vote buying (Bratton 2008; Kramon 2009; Vicente 2008), or violence (Bratton 2008; Collier and Vicente 2008). Studies of persuasion in Africa have been much more limited. Most notable here is Wantchekon’s (2003) field experiment during the 2001 presidential election in Benin, which tested voters’ responses to clientelistic (i.e., stressing specific resources that would be distributed to the voters’ home areas) and programmatic (i.e., stressing more national policy concerns, such as fighting corruption and protecting the environment) platforms, which candidates varied in collaboration with the researcher. Wantchekon found that, generally speaking, voters were more responsive to clientelistic platforms. Other studies have identified potentially persuadable voters in Africa, but they have not focused on the role that campaigns might play in individuals’ decision-making processes. For example, in a 2003 survey, Lindberg and Morrison (2005) find that, while most (82 percent) Ghanaians reported stability in their party of support in the 1996 and 2000 parliamentary elections, nearly one-fifth (18 percent) were “swing voters” (i.e., voted for different parties in 1996 and 2000). Given that the National Democratic Congress lost the presidency and its parliamentary majority in 2000, this relatively small segment of the electorate seems to have had great political significance. However, the design of the Lindberg-Morrison survey makes it impossible to determine whether these swing voters were persuaded to switch by campaigns, or whether their attitudes shifted at an earlier point in the intervening four years. Similarly, it is possible that many of the “core” (i.e., non-swing) voters in the sample had no party loyalties in the run-up to earlier election, and that their loyalties happened to be won by identical parties during campaigns in 1996 and 4

William Miles’ (1988) treatment of the 1983 Nigerian elections is perhaps the most in-depth monograph on electoral campaigning in Africa, but like the other works referenced here, its conclusions are limited by the lack of data on campaign effects at the level of individuals.

118

Chapter Five

2000. Young (2009) also finds a significant number of “swing” voters amongst respondents to Afrobarometer surveys in seven countries between 2005 and 2008, but the data do not allow him to draw any conclusions about the role of campaign activities in shifting these citizens’ support.5 The limited attention to possible persuasive campaign effects in Africa likely stems from a consensus that two factors—ethnic identity and clientelistic networks—largely determine Africans’ electoral decision making (Lemarchand 1972; Lindberg and Morrison 2008). On the first count, ethnic considerations are undeniably salient in African electoral politics, with high proportions of citizens voting for candidates from their own ethnic group, or for parties long associated with support for said group (Cohen 1983; Ottaway 1999; Posner 2005; Rothchild and Foley 1988).6 This ethnicization of African politics is most apparent when inter-group violence erupts, as in recent cases such as those in Burundi, the Democratic Republic of the Congo, Kenya, Liberia, Nigeria, Rwanda, Sierra Leone, and Sudan, but ethnic considerations are also central to electoral politics in more stable states, such as Ghana, Malawi, South Africa, and Zambia. Next, distributional considerations are paramount in African politics, given that most citizens are poor and the state controls a large proportion of resources, vis-à-vis typically weak private sectors. Clientelism—in which politically powerful patrons distribute resources through networks in return for the political support (e.g., votes) of more vulnerable clients— is a central feature of African politics (Bratton and van de Walle 1997; Chabal and Daloz 1999; Clapham 1982; Hydén 1980; Lindberg 2003; Schatzberg 2001; van de Walle 2007; Wantchekon 2003). Ethnic and clientelistic logics quite often overlap, with distributional networks mapped according to ethnic loyalties (Fearon 1999). Given these foci on relatively stable features—ethnic coalitions and clientelistic networks—electoral campaigns have not been priorities in political science research on Africa. If voters’ primary considerations are ethnic and clientelistic, then candidates’ and parties’ tasks before elections should be relatively simple: 1) reminding voters, if necessary, of ethnic and clientelistic loyalties; and 2) motivating them to turn out on those bases. In other words, campaigns would be mainly about mobilization; edu5

Studies of the 2000 and 2006 Mexican presidential elections found that large numbers of citizens were persuadable and that a significant number of voters shifted their opinions in response to campaign events, thus affecting outcomes in close races (Domínguez and Lawson 2004; Domínguez et al. 2009; Greene 2011). 6 Some go so far as to characterize elections in such democracies as “ethnic” (Horowitz 2000, 326) or “racial” censuses (Ferree 2011).

Measuring the Persuasive Effects of Electoral Campaigns in Africa

119

cating voters about platforms, positions, and performance and persuading them on those bases would not, therefore, be priorities. There is literature arguing that factors other than ethnic attachments and clientelistic networks, such as incumbent performance, are important in African citizens’ electoral decision-making processes (Bratton and Kimenyi 2008; Kimenyi and Gutiérrez-Romero 2008; Lindberg and Morrison 2008; Miles 1988; Norris and Mattes 2003; Piombo 2005; Posner and Simon 2002; Young 2009), but even this literature does not address campaign effects directly. While much of our knowledge about electoral decision-making and competition in Africa might suggest that campaigns should have little persuasive impact in most countries, we simply have lacked the data to draw any firm conclusions about such campaign effects. Rolling cross-sectional or panel survey data are necessary to measure public or individual-level shifts, respectively, in electoral preferences. To the author’s knowledge, before this project neither type of data had ever been collected during a campaign period in any sub-Saharan African country. A significant contribution of this project, then, is that it allows us to measure campaign effects of various types, including the persuasive variety, in Africa.

Research Methodology Case selection Uganda is a promising focus case for this research project for a number of reasons. Certainly, significant structural and institutional biases exist in the political playing field, with President Yoweri Museveni and his supporters in the National Resistance Movement (NRM) holding strong advantages (Conroy-Krutz and Logan 2012; Izama and Wilkerson 2011; Mwenda 2007; Rubongoya 2007; Tripp 2010, 158–176). And while Museveni and his supporters have prevailed in presidential and parliamentary balloting since elections were re-established in 1996, having first taken power through an armed struggle that ended in 1986, results have differed enough across cycles to suggest some volatility in the electorate’s preferences. This suggests that campaigns might be of significant importance in the country. For example, in the 2006 election, President Museveni’s vote share (59.3 percent) was ten points lower than in 2001 (69.3 percent). Next, as discussed previously, standard treatments of African electoral politics have tended to cite the overriding importance of two factors— ethnicity and clientelist networks. Following this line of thought, researchers have largely ignored the potential functions of electoral campaigns apart from mobilizing these ethnic and clientelistic loyalties. In terms of

120

Chapter Five

the salience of these considerations in electoral politics, Uganda certainly fits the conventional wisdom about Africa. Schisms between Ugandan political parties have long-standing ethnic and regional roots (Apter 1997; Karugire 1980; Kasfir 1976); this ethnicization of political competition has remained despite President Museveni’s putative attempts to foster “individual-merit politics” under a so-called “no-party democracy” (1986– 2005) (Carbone 2008). Ethnic tensions exploded in September 2009, when at least twenty-seven people were killed in the Central Region following a dispute between the kabaka (king) of the Baganda ethnic group and the administration of President Museveni, who is himself a Munyankole. Second, electoral competitors in Uganda have long relied on goods distribution to build support and motivate political participation, particularly given the absence of strong party organizations (Carbone 2008). During campaigns, competitors have been known to distribute relatively lowcost items such as bags of sugar, T-shirts, fuel, and mobile phone cards. At its outset, many expected that the 2011 election would be amongst Uganda’s most competitive (Conroy-Krutz and Logan 2012). Museveni had experienced two electoral cycles of downward trajectory, and recent troubles with the Baganda leadership suggested that members of that group, which forms the largest ethnic constituency in the country, might be ready to end their longstanding support for the incumbent. In addition, many in the opposition believed they were fielding a strong slate of candidates—seven, in total—many of whom might be able to defeat the incumbent were they to face him in a second-round runoff. Dr. Kizza Besigye, a fellow Munyankole of Museveni and the president’s former personal doctor, was standing for the third time as an opposition candidate. While Museveni had seen his vote totals decline in recent elections, Besigye had seen his numbers rise, from 27.7 percent in 2001 to 37.4 percent in 2006. Members of Besigye’s Forum for Democratic Change (FDC) hoped that he could emulate the success of other perennial opposition figures in Africa, such as Abdoulaye Wade, who eventually won the Senegalese presidency (albeit on his fifth attempt). Other major opposition candidates also looked initially promising. Norbert Mao, an incumbent chairperson from the politically important Gulu District in the North, fashioned himself as a “Ugandan Obama,” based on his mixed parentage (an Acholi father and a Munyankole mother). The nomination of Mao was an overt attempt by the leadership of the Democratic Party (DP), which favours increased autonomy for the traditional leadership of the Buganda Kingdom, to merge its Baganda base (which it expected to expand given Museveni’s recent clashes with the kabaka) with the smooth-talking lawyer’s own base in the North. The se-

Measuring the Persuasive Effects of Electoral Campaigns in Africa

121

lection of a non-Muganda as DP flag bearer generated significant controversy amongst the monarchists, spurring a protest movement known as Suubi and yielding the independent campaign of a DP defector, Samuel Lubega. Olara Otunnu, a Langi from the North, was nominated as the candidate of the Uganda People’s Congress (UPC), the party of late dictator Milton Obote (1966–71, 1980–85). Given Obote’s violent expulsion of the current kabaka’s father from the country in 1966 and his administration’s brutal campaign against Baganda in the Luweero Triangle in the 1980s, a UPC affiliation is still poisonous to many in the country’s populous Central Region, and the Party’s hopes of winning significant votes outside of its Northern base are extremely limited. However, given Otunnu’s status as a former Ugandan ambassador to the United Nations and a noted human rights activist, UPC stalwarts hoped that Otunnu would at least generate interest from better-educated urbanites. Other opposition candidates included Jaberi Bidandi Ssali of the People’s Progressive Party (PPP); Abed Bwanika of the People’s Development Party (PDP); and Beti Kamya, a sitting Member of Parliament from Kampala, of the Uganda Federal Alliance (UFA). Ultimately, Museveni claimed an easy re-election victory, garnering over two-thirds (68.4 percent) of the vote. Besigye came in a distant second, with 26.0 percent, while none of the other six candidates garnered even 2 percent.7 Perhaps even more surprising than Museveni’s popular vote share was the geographic scope of his win: the president won 105 of the country’s 112 districts, and he even won a majority of votes in the North (60.6 percent), increasing his 2006 share by some 24.2 points in a region that had long been an opposition stronghold. Buganda also disappointed the opposition, as Museveni actually ended up increasing his vote share to 60.7 percent, from 57.4 percent in 2006. Although a number of observers (DEMGroup 2011a; EU-EOM 2011a, 2011b; COG 2011) and analysts (Izama and Wilkerson 2011) concluded that the incumbent’s significant resource advantages likely determined his victory,8 observer groups also noted the balloting and counting processes as lacking serious problems (AU Observer Mission 2011; DEMGroup 2011b, EACCOMES-IGAD 2011).

7

Mao won 1.9 percent, Otunnu 1.6 percent, Bwanika 0.7 percent, Kamya 0.7 percent, Bidandi 0.4 percent, and Lubega 0.4 percent. 8 For a counterpoint to this conclusion, see Conroy-Krutz and Logan 2012.

122

Chapter Five

Panel design and data In order to collect data on individual-level changes in attitudes across the campaign period, we conducted a three-wave panel study in Uganda between November, 2010 and April, 2011. Panels have figured prominently in studies of campaign effects in countries such as the United States (Bartels 1993; Berelson et al. 1954; Finkel 1993; Hillygus and Jackman 2003; Just et al. 1996; Lazarsfeld et al. 1948; Markus 1982; Norris et al. 1999; Patterson 1980; Patterson and McClure 1976), the United Kingdom (Sanders et al. 2005), Germany (Finkel and Schrott 1995), Canada (Gidengil et al. 2009; Johnston et al. 1992), Brazil (Baker et al. 2006), and Mexico (Domínguez and Lawson 2004; Domínguez et al. 2009; Greene 2011). Panel studies feature at least two advantages over alternative longitudinal designs such as rolling cross-sectional samples. First, panel data enable the tracking of individual trajectories, and they are, of course, necessary for researchers to observe individual responses to campaign stimuli directly (Bartels 2006, 142–47; Wlezien and Erikson 2001, 432). Second, apparent instability in the opinion of mass publics is often an artifact of measurement error in cross-sectional designs (Achen 1975, 1983; Converse 1964; Wlezien and Erikson 2001). However, panel designs with N > 2 waves facilitate adjustment for such measurement errors (Wiley and Wiley 1970), since repeated measures allow for checks on the reliability of observations (Bartels 1993, 269–70). The three-wave panel survey was scheduled to coincide with campaigning for the February/March presidential, parliamentary, and local elections in Uganda.9 One thousand seventy-two individuals from nine parliamentary constituencies10 were interviewed via face-to-face interac9

Besides citizenship, the only eligibility criteria were a) being at least eighteen years old (voting age in Uganda), and b) ability to understand and answer questions in one of the ten survey languages: Acholi, Ateso, English, Langi, Luganda, Lusamia, Lusoga, Runyankole, Runyoro-Rutoro, and Rwamba. 10 Constituencies were selected through a stratified random sample from the full list of 238 constituencies in order to maximize the probability that adequate samples from Uganda’s major ethnic groups (Baganda, Banyankole, Basoga, Acholi, Langi, and Iteso) were included. One additional constituency from urban Kampala District and two additional ethnically diverse constituencies were also randomly selected. The selected constituencies included Bugweri (Iganga District, predominantly Basoga), Gulu Municipality (Gulu, Acholi), Igara West (Bushenyi, Banyankole), Katikamu South (Luweero, Baganda), Maruzi (Apac, Langi), Ntoroko (Ntoroko, diverse), Rubaga North (Kampala), Samia-Bugwe North (Busia, diverse), and Soroti Municipality (Soroti, Iteso). Each of the four major regions of

Measuring the Persuasive Effects of Electoral Campaigns in Africa

123

tions in late November/early December, 2010, near the outset of the official campaign period.11 Of these, 721 individuals (67 percent of the original sample) were successfully re-interviewed in late January/early February, 2011, in the days and weeks preceding the 18 February balloting for president and Parliament. A post-election wave was completed in late March/early April, 2011,12 in which 675 individuals (63 percent of the original sample) were successfully re-interviewed. Given the limited geographical scope of the fieldwork, the sample is not nationally representative. Fieldwork was conducted by Synovate Uganda.

Findings Aggregate Stability in Candidate Preferences? Aggregate levels of support for each of the presidential candidates, at different points in the study, are reported in Table 5.1.13 Of those who participated in the study, just over half (55 percent) reported that they voted for Museveni in 2006. Only 8 percent said that they had voted for Besigye, while other opposition candidates were the reported recipients of only 1

the country is represented here: North (Gulu Municipality, Maruzi, Soroti Municipality), East (Bugweri, Samia-Bugwe North), Central (Katikamu South, Rubaga North), and West (Igara West, Ntoroko). 11 Within selected constituencies, Enumeration Areas (EAs) were selected at random, with an EA’s probability of selection directly related to its population. EAs are delineated by the Ugandan Bureau of Statistics (UBoS). Within selected EAs, households were selected via a random-walk pattern, and individuals within households were selected at random using a kish grid. 12 Uganda is divided into a number of administrative units, including (from largest to smallest) districts, counties, sub-counties, parishes, and villages. Direct elections are held for chairpersons and council members at the district and sub-county levels. In addition, voters in urban municipalities elect mayors and council members. District elections were held on 23 February, municipality elections on 2 March, and sub-county elections on 7 March. Municipal elections in Kampala were abruptly canceled due to widespread reports of irregularities; they were rescheduled for 14 March. Interviews for the final wave of the survey began approximately two weeks after the Kampala elections. 13 Again, the sample is not nationally representative; therefore, we should not expect that respondents’ stated preferences will match those of actual election results, nor those of nationally representative surveys that were conducted prior to the election.

Chapter Five

124

percent of respondents’ votes.14 Finally, 33 percent of the respondents said that they did not vote, for various reasons, in 2006.15 Wave 1

Wave 2

Wave 3

(Early campaign, Nov.– Dec. 2010)

(Late campaign, Jan.–Feb. 2011)

(Reported vote, March– April 2011)

55

63

70

66

8

6

11

12