Ethnicity and Democracy in Africa 0852558600, 9780852558607

Citation preview

Ethnicity & Democracy in Africa

Ethnicity & Democracy in Africa Edited by Bruce Berman, Dickson Eyoh & Will Kymlicka

JAMES CURREY OXFORD

OHIO UNIVERSITY PRESS ATHENS

Electronic edition published in 2017 Ohio University Press www.ohioswallow.com James Currey Boydell & Brewer Ltd PO Box 9, Woodbridge Suffolk IP12 3DF (GB) www.jamescurrey.com Boydell & Brewer Inc. 668 Mt Hope Avenue Rochester, NY 14620-2731 (US) www.boydellandbrewer.com All rights reserved. No part of this book may be reproduced in any form, or by electronic or mechanical means, including information storage and retrieval systems, without permission in writing from the publishers, except by a reviewer who may quote brief passages in a review. © James Currey Ltd, 2004 First published 2004 ISBN ISBN ISBN ISBN

0-85255-861-9 0-85255-860-0 0-8214-1569-7 0-8214-1570-0

(James Currey cloth) (James Currey paper) (Ohio University Press cloth) (Ohio University Press paper)

British Library Cataloguing in Publication Data Ethnicity & democracy in Africa 1. Ethnicity - Political aspects - Africa I. Berman, Bruce (Bruce J.) II. Eyoh, Dickson, 1954III. Kymlicka, Will 323.1'6 Library of Congress Cataloging-in-Publication Data available on request

ISBN 978-1-78204-792-6 (James Currey eISBN) ISBN 978-0-8214-4267-8 (Ohio University Press eISBN)

Contents Contributors Preface Acknowledgements

ix xiii xv

I Ethnicity & Democracy in Historical & Comparative Perspective 1 1 BRUCE BERMAN, DICKSON EYOH & WILL KYMLICKA

Introduction Ethnicity & the Politics of Democratic Nation-Building in Africa

1

2

PETER EKEH

Individuals' Basic Security Needs & the Limits of Democratization in Africa

22

BRUCE BERMAN

Ethnicity, Bureaucracy & Democracy: The Politics of Trust

38

WILL KYMLICKA

Nation-Building & Minority Rights: Comparing Africa & the West

54

V

II The Dynamics of Ethnic Development in Africa

73

5

JOHN LONSDALE

Moral & Political Argument in Kenya

73

6

DICKSON EYOH

Contesting Local Citizenship: Liberalization & the Politics of Diflference in Cameroon

96

7

CHERYL HENDRICKS

The Burdens of the Past & the Challenges of the Present: Coloured Identity & the Rainbow Nation

113

8

JACQUELINE S. SOLWAY

Reaching the Limits of Universal Citizenship: 'Minority' Struggles in Botswana

129

9

TOYIN FALOLA

Ethnicity & Nigerian Politics: The Past in the Yoruba Present

vi

148

Ill Ethnicity & the Politics of Democratization

167

10 E. S. ATIENO

ODHIAMBO

Hegemonic Enterprises & Instrumentalities of Survival: Ethnicity and Democracy in Kenya

167

11 SHULA MARKS

'The dog that did not bark, or why Natal did not take off5: Ethnicity & Democratization in South Africa KwaZulu Natal

183

12 GITHU

MUIGAI

Jomo Kenyatta & the Rise of the Ethno-Nationalist State in Kenya

200

13 MAMADOU

DIOUF

Between Ethnic Memories & Colonial History in Senegal: The MFDC & the Struggle for Independence in Casamance

218

14 BOGUMIL JEWSIEWICKI & LEONARD N ' S A N D A BULELI

Ethnicities as 'First Nations' of the Congolese Nation-State: Some Preliminary Observations 240

15 A. RAUFU

MUSTAPHA

Ethnicity & the Politics of Democratization in Nigeria

257

vii

IV Ethnicity & Institutional Design in Africa

277

16

RICHARD SIMEON & CHRISTINA MURRAY

Multi-Level Governance in South Africa

277

17

JOHN BOYE EJOBOWAH

Liberal Multiculturalism & the Problems of Institutional Instability

301

18

BRUCE BERMAN, DICKSON EYOH & WILL KYMLICKA

Conclusion African Ethnic Politics & the Paradoxes of Democratic Development

317

Index

325

viii

Contributors BRUCE J. BERMAN

is Professor of Political Studies at Queen's University in Kingston, Ontario. He is the author of Control and Crisis in Colonial Kenya (1990) and, with John Lonsdale, of Unhappy Valley: Conflict in Kenya and Africa (1992). His recent publications include 'Ethnicity, Patronage and the African State: the politics of uncivil nationalism,' African Affairs, 1998; 'African States, Bureaucratic Culture and Computer Fixes' (co-authored with W.J. Tettey), Public Administration and Development, 2001; and Critical Perspectives on Politics and Socio-Economic Development in Ghana (co-edited with W.J. Tettey and Korbla Puplampu), 2003. He was president of the Canadian Association of African Studies in 199091 and is vice-president and president-elect of the African Studies Association. LEONARD N'SANDA BULELI

received his doctorate from Laval University for a political history of Maniema in the 20th century. He was director of the Institut superieur pedagogique de Kindu from 1993 to 1999, and Professor of History in the Institut superieur pedagogique de Bukavu from 1983 to 1994. He recently published 'Le Maniema, de la guerre de l'AFDL a la guerre du RCD', Politique africaine, 2001; and his book, La bataille de Kindu was published in the collection Cahiers africaines in 2003. MAMADOU DIOUF

received his Ph.D. from the Universite Paris I Sorbonne. He is Professor of History and African American and African Studies and Director of the African Initiative Studies program at the University of Michigan, Ann Arbor. His books include Le Kajoor au 19eme Siecle: Pouvoir Ceddo et Conquete Coloniale (1990), Histoire du Senegal (2001) and (with D. Cruise O'Brien et M. C. Diop) La Construction de VEtat au Senegal (2002). JOHN BOYE EJOBOWAH

is an assistant professor in the Global Studies Program at Wilfrid Laurier University, Waterloo. He specializes in making normative analysis of practical claims to equality and of constitutional and institutional arrangements for accommodating such claims. His book, Competing Claims to Recognition in the Nigerian Public Sphere, was published in December 2001 by Lexington Books (Lanham, M D and Oxford). PETER EKEH

received his Ph.D. in sociology from the University of California, Berkeley in 1970 and is currently Professor of Sociology and former chair of African American Studies at the State University of New York at Buffalo. He was formerly head of the Department of Political Science at the University of Ibadan, Nigeria, and chair of Ibadan University Press. He was a Visiting Fellow of the Woodrow Wilson Center in Washington in 1988-9. His publications include Social Exchange Theory (1974) and he co-edited with Eghosa Osaghae, Federal Character and Federalism in Nigeria (1989). ix

Contributors

X

DICKSON EYOH

is Associate Professor of Political Science and African Studies and past Director of African Studies at the University of Toronto. He has published extensively on the state and ethnic politics in Africa. His recent publications include The Encyclopedia of Twentieth Century African History (Routledge, 2003), co-edited with Paul Zeleza. TOYIN FALOLA

is the Frances Higginbothom Nalle Centennial Professor in History at the University of Texas at Austin. He is the author of numerous books, including Key Events in African History: A Reference Guide, Nationalism and African Intellectuals; co-editor of the Journal of African Economic History; and editor of Rochester Studies in African History and the Diaspora, and the series Culture and Customs of Africa (Greenwood Press). He has received various awards and honors, including the Jean Holloway Award for Teaching Excellence, the Texas Exes Teaching Award, the Chancellor Council Outstanding Teaching Award for Teaching Excellence, and the Ibn Khaldun Distinguished Award for Research Excellence. His students and colleagues have presented him with a Festschrift edited by Adebayo Oyebade, The Transformation of Nigeria: Essays in Honor of Toyin Falola. CHERYL HENDRICKS

earned a Ph.D. in Government and International Studies from the University of South Carolina, and is currently Lecturer in Political Studies, University of the Western Cape, South Africa. She was a post-doctoral fellow in 2001-02 at the Frederick Douglass Institute of African and African-American Studies at the University of Rochester. BOGUMIL JEWSIEWICKI

Professor of History at Laval University, Quebec, holds the Canada Research Chair in the comparative history of memory. He is a research fellow at the Centre d'etudes africaines at CNRS/EHESS in Paris. His most recent book is Mami wata. La peinture urbaine au Congo (Paris, Gallimard, 2003). He also edited with Richard Banks Politique africaine no. 84, 2001 on the war in the Congo. WILL KYMLICKA

is the author of five books published by Oxford University Press: Liberalism, Community, and Culture (1989), Contemporary Political Philosophy (1990; second edition 2002), Multicultural Citizenship (1995), Finding Our Way: Rethinking Ethnocultural Relations in Canada (1998), and Politics in the Vernacular: Nationalism, Multiculturalism, Citizenship (2001). He is also the editor of The Rights of Minority Cultures (1995), and co-editor of Ethnicity and Group Rights (1997), Citizenship in Diverse Societies (2000), Can Liberal Pluralism Be Exported? (2001), and Language Rights and Political Theory (2003). He is a Professor of Philosophy at Queen's University, and a Visiting Professor in the Nationalism Studies program at the Central European University in Budapest. JOHN LONSDALE

is Fellow of Trinity College, Cambridge and Professor in African History. His

Contributors

xi

most recent publication is Mau Mau and Nationhood: Arms, Authority and Narration (2003), edited with Atieno Odhiambo. He is currently working on the documentation of Kenya's decolonization for the British Documents on the End of Empire series, and, with Bruce Berman, on the joint intellectual biography of Jomo Kenyatta and Louis Leakey in their rival constructions of the Kikuyu. He was president of the African Studies Association of the United Kingdom in 2000-01. SHULA MARKS

is Emeritus Professor of Southern African History of the School of Oriental and African Studies, University of London and Distinguished Research Fellow of the School of Advanced Study in the University of London. A former Director of the Institute of Commonwealth Studies, London, she is a Fellow of the British Academy and holds honorary degrees from the Universities of Cape Town and Natal. She has lectured and written widely on South African history. GITHU MUIGAI

obtained his legal education from the University of Nairobi and his doctorate from the Columbia University School of Law. He is currently senior lecturer in the department of public law of the Faculty of Law at the University of Nairobi. He has published extensively on politics and law in Kenya, and is currently serving as a commissioner of the Constitution of Kenya Review Commission. CHRISTINA MURRAY

is Professor of Constitutional and Human Rights Law at the University of Cape Town. Between 1994 and 1996 she served on a panel of seven experts advising the South African Constitutional Assembly in drafting South Africa's 'final' Constitution, and has advised a number of government departments on the implementation of the new system of multi-level government. She teaches and writes on the law of contract, human rights law (particularly relating to gender equality and African customary law), international law, and constitutional law and is also the director of the Law, Race and Gender Research Unit at UCT. ABDUL RAUFU MUSTAPHA

completed a doctorate on rural Nigerian politics at St Peter's College, Oxford. He is currently a University Lecturer in African Politics at Queen Elizabeth House, and Kirk-Greene Fellow at St Antony's College, Oxford. His key areas of research are in the politics of rural Africa, identity politics and ethnicity, and the politics of democratization in Nigeria. Recent publications include Agrarian Production, Public Policy & the State in the Kano Region, 1900-2000 (Drylands Research, Crewkerne, 2000); 'Civil Rights and Pro-democracy Groups in and Outside Nigeria,' in Amuwo, Bach & Lebeau, eds, Nigeria during the Abacha Years (1993-1998), (FRA/CEAN, Ibadan, 2001); and c Coping With Diversity: The Nigerian State in Historical Perspective,' in A. Samatar and A.I. Samatar, eds, The African State: Reconsiderations (Heinemann, Portsmouth N H , 2002). E. S. ATIENO ODHIAMBO

is a Professor of History at Rice University in Houston. His latest publications include Mau Mau and Nationhood (co-edited with John Lonsdale) (2003), and The Burdens of History and Leadership in Africa: The Essays of Bethwell Allan Ogot

Contributors

xii

(co-edited with Toyin Falola) (2002). He is currently reading the archives of Tom Mboya. RICHARD SIMEON

is Professor of Political Science and Law at the University of Toronto. He is a close observer of the constitutional process in South Africa, including teaching and research on decentralization and multi-level governance. He has held several visiting appointments, including at Harvard and the University of Cape Town, and is a member of the advisory council of the 'Club de Madrid'. His recent research has focused on comparative federalism, decentralization and institutional design in ethnically divided societies. His major publications include Federal Provincial Diplomacy (1972), Redesigning the State: the Politics of Constitutional Change (1983, with Keith Banting), State, Society and the Development of Canadian Federalism (1990, with Ian Robinson), and Political Science and Federalism: Seven Decades of Scholarly Engagement (2002). JACQUELINE SOLWAY

is a social anthropologist in the International Development Studies Programme at Trent University, Canada. She has conducted research in Botswana for over two decades on rural political economy, social differentiation, and development and more recently on the formation and politicization of local identities. In conjunction with local scholars and activists, she is engaged in on-going research that examines multiculturalism and the state in Botswana. She has also conducted research in southern Egypt on natural resource management. Her recent articles appear in Journal of Southern African Studies, Ethnos, Anthropological Quarterly, Interventions, and she guest-edited an issue of Anthropologica.

Preface This book has grown out of a concern that ethnic politics was a primary factor shaping the success or failure of the renewed efforts at democratic development and political reform that emerged in Africa in the 1980s. The recognition of ethnic differences and the amelioration of ethnic conflicts through both institutional reform and policy initiatives appeared to be crucial to the reconstruction of African states and the establishment of stable and enduring democratic processes. At the same time, it was equally apparent that the ethnic communities of Africa were not atavistic survivals of a pre-modern world, but dynamic social creations of the colonial and post-colonial eras in which the state played an important and often determining role in the definition and development of ethnic communities and identities. The intimate embrace of ethnicity and the state raised a host of both empirical and normative questions regarding the effectiveness and legitimacy of varying institutional means for accommodating or overcoming ethnic diversity; the relationship between nation-building and assimilation or multi-cultural recognition; how to deal with ethno-regional differentiation due to uneven socio-economic development; and the proper relationship between individual and collective rights in liberal democratic theory and practice. There were no obvious answers to either the practical or the normative concerns about what both worked and should work in the diverse sociocultural and political contexts of African states. To address these concerns we decided to assemble a group of scholars who approached the issues from a diversity of disciplinary perspectives including history, law, sociology, anthropology and philosophy as well as political science. We were particularly interested in bringing political theorists focused on the normative issues of rights and democratic principles, and political scientists specializing in the comparative analysis of institutional systems like federalism, together with Africanists concerned with the empirical analysis of the African experience of ethnic politics and democratization. The result was three days of fruitful discussion that led, if not to definitive answers, then to new ways of addressing the questions and approaching the formulation of tentative answers in different contexts. One of the editors, Will Kymlicka, was unable to attend the conference due to scheduling conflicts, although his paper was presented at the conference and formed part of the discussions, as was the paper by Githu Muigai, who was prevented at the last minute from attending by the government of Kenya. Bogumil Jewsiewicki and Leonard Buleli, who were also unable to attend the conference, submitted their contribution to the project. Events since the conference have reinforced our sense of the importance of the issues dealt with in this volume. Ethnic conflict continues to be the major source of violence ripping apart African states, and the peaceful accommodation of ethnic differences remains key to successful democratic development. The following chapters address two major issues. First, that the development of ethnic communities and identities and patterns of ethnic competition and conflict are the result of the contingent and often idiosyncratic interaction of indigenous cultures and institutions with the intrusive external political, economic and cultural forces of Western modernity. Second, that democratic development in multiethnic societies in Africa depends upon the contingent interactions and adaptaxiii

xiv

Preface

tions of both indigenous and exogenous institutions and cultural elements. Successful democracies in Africa will probably neither look like, nor function as facsimiles of, familiar forms of Western liberal democracy, but rather produce distinctive African variants as the fundamental issues are argued out and negotiated in each state. Theory, both normative and empirical, must deal with understanding the bases of such complexity.

Acknowledgements The organization of the conference in March 2000 and subsequent development of this book could not have been possible without the help and support of numerous institutions and individuals. The Centre for the Study of Democracy at Queen's University, Ontario encouraged us to develop this project and provided constant organizational support. We would like to thank, in particular, George Perlin, Professor of Political Studies and the founding Director of the centre, Michael Martin, executive officer, and Andrew Parkin for their belief in the importance of the project and support through the often convoluted labyrinth of organization and funding. The resources for the conference and a grant towards the publication of the proceedings, and distribution of a substantial number of copies in Africa, came from a grant from the Ford Foundation, and we would like to record our gratitude for its generosity and support. We would also like to thank Michael Levin of the University of Toronto, Daniel Weinstock of the University of Montreal, and Shaheen Mozaffar of Bridgewater State University for their active participation and generous contributions to the conference sessions, as well as Chweya Ludeki, Jonathan Sears and Jacqueline Klopp, graduate students at Queen's and McGill Universities, for their enthusiastic participation and assistance. The Canadian Broadcasting Corporation turned the themes of the conference into a two-hour program involving most of the conference participants, that was broadcast on the CBC Newsworld Network in May 2000. We would particularly like to thank Tony Burman, director of CBC News, Ann Petrie, who served as the on-air moderator, and their colleagues for their professionalism and ability to turn the complex material of the conference into accessible and engaging programming. They set a high standard for the integration of television and academic research in public service broadcasting. We would also like to thank James Currey and Douglas Johnson of James Currey Publishers for taking on this substantial volume, and to them and their anonymous readers for comments and suggestions that helped turn the original manuscript into a much better book. Finally, we thank our spouses for their patience and forbearance over the past four years for our frequent preoccupation with turning our initial ideas into a material reality. BRUCE BERMAN, DICKSON EYOH & WILL KYMLICKA

Kingston & Toronto

XV

I

Ethnicity & Democracy in Historical & Comparative Perspective

i

BRUCE BERMAN, D I C K S O N EYOH & W I L L KYMLICKA

Introduction Ethnicity & the Politics of Democratic Nation-Building in Africa

T

HE flow of commentary on political responses to the multifaceted crisis of development in which African countries have been enveloped in the past three decades betrays the propensity of Africanist social sciences to alternate, seemingly without much effort, between moments of exaggerated optimism and despair about Africa's development prospects. 1 Competing explanations of the causes of the crisis agreed that the authoritarian post-colonial state was the primary culprit. The groundswell of popular opposition to authoritarian rule in the late 1980s and early 1990s was, for many, a welcome sign of the re-animation of the agency of Africans to design for themselves more promising futures - futures that would be based on liberal politics and market economies. This euphoria did not last long as successive electoral cycles reaffirmed the resilience of clientelism and patronage as the dominant practice of African politics. 2 Civil society, whose supposed resurgence was much vaunted, turned out to be riven by communal divisions, particularly of ethnicity and religion. Civic associations reflecting such cleavages have had scant positive effect on party formation and electoral competition, and often demonstrate little interest in 1 Colin Leys, The Rise and Fall of Development Theory. Oxford: James Currey; Bloomington, IN: Indiana University Press, 1996, pp. 107-8. 2 We have tried to avoid using the term 'neo-patrimoniaP in describing contemporary African political systems, although it is widely used by political scientists. This seems to us a misuse of the original Weberian concept, which referred to an explicit and formal, albeit highly personalized, system of administration in what Weber classified as traditional 'oriental despotisms' (succinctly summarized by Reinhard Bendix in Max Weber: an Intellectual Portrait, New York: Anchor Books, 1962, pp. 334-59). Although the behavior of African office holders is often very similar to that of officials in a patrimonial system, the difference is that in contemporary Africa the formal institutions of state are those of the legal-rational authority of the modern national state. A fundamental part of the political crisis of Africa is the weakness of such institutions, as they are continually undermined by the pervasive informal practices of clientelism and patronage.

1

2

Bruce Berman, Dickson Eyoh & Will Kymlicka

promoting liberal democracy. Everywhere the politics of identity and ethnicity appears resurgent. 3 Accounts of the travails of current experiments in multiparty politics turn on unavoidably normatively laden definitions of democracy and the criteria used to gauge its progress. To oversimplify, two, but by no means exclusive, perspectives are prevalent in the African literature.4 There is the mainstream perspective which favors a minimalist (procedural) definition and sees periodic changes in governments through multiparty elections as the hallmark of democracy. For this perspective, unrelieved economic adversity, the paucity of middle and independent capitalist classes, cultural fragmentation, etc. jointly explain the resilience of c neopatrimonial' politics. Ranged against it are critics of liberal (minimalist) democracy, many of whom are advocates of 'popular democracy'. For them, the politics of clientelism endures because the main purpose of the elite-driven multiparty politics advocated chiefly by the United States is to widen the circulation and recruitment of elites and legitimate neo-liberal 'reforms', and not the transformation of existing inequalities in the distribution of economic and political power.5 We cannot in this context evaluate the important conceptual and normative differences that separate these and other perspectives in the worldwide and Africa-focused debates on current democratic experiments. We can suggest, however, that the two broad perspectives share a view that the social pluralism of African societies (a phenomenon for which ethnicity has come to serve as an all too convenient shorthand) is the taproot of clientelistic and patronage politics, and by extension, a leading, if not the primary, obstacle to democratic nation-building. They also incline toward an elite-centered perspective, based on the presumption that African politics is saturated with a mercenary ethos, that regards ethnic politics as 'the shadow theater of accumulation'. 6 The material preoccupations and personalistic nature of patronage networks that are the conduit of ethnic politics continue to eliminate the relevance of formal institutions, ideology and policy differences in the organization of the wider civic political arena. The contributors to this collection are preoccupied by obvious questions which this commonsense about multiparty politics and ethnicity in Africa elicits. Why is ethnicity a political problem? How is the problem manifested? And which institutional models offer the best prospect of ameliorating the challenges that ethnicity poses to democratic nation-building? The interdisciplinary perspectives offered in the papers in this collection differ from the dominant perspectives in contemporary African political analysis in a number of key respects. First, they are all attentive to and built upon the growing body of more than two decades of research by historians and anthropologists that has demolished the view of African 'tribes' as atavistic survivals of primordial stages 3

Patrick Chabal and Jean-Pascal Daloz, Africa Works: Disorder as Political Instrument. Oxford: James Currey; Bloomington, IN: Indiana University Press for the International African Institute, 1999, pp. 17-30; Richard Sandbrook, Closing the Circle: Democratization and Development in Africa. Toronto: Between the Lines; London and New York: Zed Press, 2000, ch. 2; Robert Fatton, 'Africa in the age of democratization: The civic limitations of civil society,' African Studies Review, 38, 2, 1995; and Michael Bratton and Nicolas van de Walle, Democratic Experiments in Africa: Regime Transition in Comparative Perspective. Cambridge: Cambridge University Press, 1997. 4 Sandbrook, Closing the Circle, pp. 4-6; Dickson Eyoh, 'African Perspectives on Democracy and the Dilemmas of Post-Colonial Intellectuals,' Africa Today, 45, 3-4, 1998, pp. 281-306. 5 William I. Robinson, Promoting Polyarchy: Globalization, U.S. Intervention and Hegemony. Cambridge: Cambridge University Press, 1996. 6 Jean-Francois Bayart, The State in Africa: The Politics of the Belly. London: Longmans, 1993.

Ethnicity & the Politics of Democratic Nation-Building

3

of social development and of African ethnic politics as simply a cynical instrument of elite manipulation. Instead, African ethnicities are viewed here as complex and protean expressions of the often distinctive African experiences of modernity, grounded in the changing material realities of state and market, and the confrontations of class, gender and generation. From this perspective, the incorporation and reinterpretation of pre-colonial elements of culture and community and their instrumental invocation and manipulation are contextually located within the internal and external dimensions of communal politics. Thus, the essays which follow, especially those that analyze African country cases, are unambiguous about the key role played by elites in the politicization of ethnicity and the inherently authoritarian and exploitative character of clientelistic politics and patronage networks. However, as a needed corrective to the elitecentered thrust of much contemporary analysis, they suggest that fuller appreciation of the dynamics of ethnic politics and the challenges it poses to democratic nation-building must remain alert to two related pitfalls: (i) the inclination to treat both ethnic communities and elites as homogeneous and static; and (ii) the tendency to view the politicizations of ethnicity as phenomena manufactured by corrupt elites and consumed by more or less gullible masses. 7 Instead, we attempt to build from the premise that ethnic pluralism is and will remain a fundamental characteristic of African modernity that must be recognized and incorporated within any project of democratic nation-building. For this reason, the historians and anthropologists who contributed to this volume have been joined in a dialogue with political theorists concerned with the development of democracy in multi-cultural societies and the institutional means for its realization. In the following sections of this chapter we shall briefly outline, first, the historical and cultural origins of modern African ethnic communities; second, the patterns of politicized ethnicity in contemporary politics, their relationship to existing states and market economies, and the challenge they pose to democratic development; and, finally, the concepts and institutional options available for creative adaptation in the development of multi-ethnic democratic nation-states in Africa.

The Construction of Ethnic Communities and Identities in Africa As noted earlier, all of the contributions to this volume are based on the premise that African ethnicities are not atavistic, primordial survivals of archaic primitive cultures, but rather modern products of the African encounter with capitalism and the nation-state in the colonial and post-colonial eras. Contemporary ethnic communities and identities in Africa did not and will not fade away with the inevitable advance of global modernity, but rather represent critical aspects of the particular African experience of modernity itself. They are the outcomes of continuous and continuing processes of social construction emanating from the encounters of indigenous societies with the political economy and culture of the 7

Jan Pieterse, 'Varieties of Ethnic Politics and Ethnicity Discourse,' in Edmund Wilmsen and Patrick McAllister, eds, The Politics of Difference: Ethnic Premises in a World of Power. Chicago: University of Chicago Press, 1996; Dickson Eyoh, 'Conflicting Narratives of Anglophone Protest and the Politics of Identity in Cameroon,' Journal of Contemporary African Studies, 16, 2, 1998, pp. 250-52.

4

Bruce Berman, Dickson Eyoh & Will Kymlicka

West, as well as the deliberate manipulations of diverse political actors. 8 These processes are both historically specific and contingent on the outcome of internal and external struggles defining the membership and boundaries of ethnic communities, and their relations with the other communities with whom they share the same state. As several of the later chapters make clear, ethnicities have been constructed from diverse indigenous and foreign cultural materials and continue to be defined and redefined up to the present (see particularly the chapters by Diouf, Eyoh, Hendricks, Solway, and Falola). Any approach to democratic development in Africa that does not recognize the diversity and dynamism of ethnic communities in any foreseeable future cannot succeed. The development of ethnicity in Africa for more than a century has been marked by a dialectic of expansion and differentiation.9 Contemporary ethnicities are both much larger in social scale and population, and more sharply demarcated from other such groups, than the smaller and more fluid communities of the pre-colonial past. 10 At the same time, African ethnic groups are not univocal, and the content of culture and custom as well as the boundaries of communities remain matters of frequent conflict and negotiation. As the chapters by Falola and Mustapha illustrate, most large ethnic communities in Africa continue also to contain local sub-groups and identities whose relations are often problematic. The social forces shaping ethnic development and identity have been fundamentally material, and ethnic politics has focused on defining the terms of access both to traditional assets of land and labour and the material resources of modernity in both the state and the market. It has become customary to distinguish the internal and the external aspects of this process. The internal dimension, concerned with relations inside the group, has been termed 'moral ethnicity', a contested process of defining cultural identity, communal membership and leadership. It is important to emphasize that the attachment many Africans have to their ethnic group and ethnic identity is not simply an atavistic or irrational attachment to kith and kin, or to blood and soil. It is rather tied up with a complex web of social obligations that define people's rights and responsibilities, and that protect people when they are most vulnerable and alone (for example when traveling, ill or dying). This indeed is the point of calling it 'moral ethnicity' - membership in an ethnic group entails subordinating one's behavior to certain moral imperatives when dealing with other group members. It is equally important, however, not to romanticize moral ethnicity. Relations within the group are not egalitarian or harmonious. Indeed, as Lonsdale notes, moral ethnicity has primarily been a culture of personal accountability, with little concern for the poor and no articulation of a concept of universal, equal citizenship. He and other contributors note the tendency of African ethnic cultures to a conservative authoritarianism (see also the papers by Berman, Marks, and Falola). Competing elites and subgroups are continually contesting the 8 J.D.Y. Peel, 'The Cultural Work of Yoruba Ethnogenesis,' in Elizabeth Tonkin et al., eds, History and Ethnicity. London: Routledge, 1989; Thomas Hyland Eriksen, Ethnicity and Nationalism: Anthropological Perspectives. London: Pluto Press, 1993. 9 Shaheen Mozaffar, 'The Institutional Logic of Ethnic Politics: a Prolegomenon,' in Harvey Glickman, ed., Ethnic Conflict and Democratization in Africa. Atlanta, GA: African Studies Association Press, 1995, pp. 60-61. 10 The seminal statement is Aidan Southall, 'The Illusion of Tribe,' Journal of Asian and African Studies, v, 1, 1970. See also Leroy Vail, ed., The Creation of Tribalism in Southern Africa. London: James Currey, 1989; Carola Lenz, 'Tribalism and Ethnicity in Africa: a review of four decades of Anglophone research,' Cahiers des sciences humaines, 31, 2, 1995.

Ethnicity & the Politics of Democratic Nation-Building

5

meanings of group membership, and seeking to renegotiate their assigned responsibilities. Nonetheless, within this conservative hierarchical system, there are relations of trust and solidarity: people can rely on others in the group to fulfil their responsibilities, minimal and inegalitarian as they often are. By contrast, the external dimension, concerned with relations between an ethnic group and the state, or between two or more ethnic groups, is essentially amoral. This process, often called 'political tribalism', describes the competitive confrontation of 'ethnic contenders' for the material resources of modernity through control of the state apparatus (see the papers by Lonsdale and Kymlicka). Here success is defined as maximizing the power and resources available to one's own group, whatever the consequences for other groups or for the functioning of the state as a whole. The interaction of moral ethnicity and political tribalism describes a complex process of ethnic definition and identity, of who belongs to what community and what access to material resources such membership makes possible. At the heart of ethnic politics is the use of historical and cultural resources of past and present in a struggle for control of the future and definition of the terms of social change (see the chapters by Diouf, Solway, Hendricks, Eyoh, and Marks). Modern African ethnicities were shaped by a particular relationship with the institutions of the colonial state. Colonial bureaucracies played a key role in the construction of'tribal' identities out of earlier kinship groups and political units, building upon indigenous power relations of clientage between big men and their supporters and dependants to forge terms of collaboration facilitating the typical pattern of indirect rule. 11 A subordinate apparatus of thousands of chiefs and village headmen, whether incorporated indigenous positions of authority or new colonial creations, exercised a 'decentralized despotism' of local control within ethnically defined administrative units. 12 These cadres provided the colonial state with its knowledge of the distinctly patriarchal and authoritarian versions of indigenous culture and custom it sought to sustain as the basis of law and order in the countryside. 13 The linkages of the chiefs and headmen and other local agents to the colonial state provided conduits of patronage resources of modernity as well as authoritative control over local land and labor, and made ethnic membership rather than any broader concept of citizenship the basis for rights and property. As Ekeh discusses in his chapter, for ordinary individuals contact with the colonial state always contained an element of danger and uncertainty, requiring powerful patrons for protection and intermediation, and making ethnicity the essential community of trust and security in opposition to the alien and amoral state and competing ethnic communities. The cultural content of ethnic construction was powerfully promoted, if largely unintentionally, by Christian missions and mission education that created standardized print versions of 'tribal' languages from related vernacular dialects, created a literate intelligentsia, and, with translations of the bible, provided them with potent literary resources for the imagining of ethnic history and 11 Bruce Berman, 'Ethnicity, Patronage and the African State: The Politics of Uncivil Nationalism,' African Affairs, 97, 388, 1998; Peter Ekeh, 'Social Anthropology and Two Contrasting Uses of Tribalism in Africa,' Comparative Studies in Society and History, 32, 4, 1990; Catherine Boone, 'States and Ruling Classes in Post-Colonial Africa,' in Joel Migdal et al, eds, State Power and Social Forces. Cambridge: Cambridge University Press, 1994. See also the chapter by Muigai in this volume. 12 Mahmood Mamdani, Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton, NJ: Princeton University Press; London: James Currey, 1996. 13 Sally Falk Moore, Social Facts and Fabrications. Cambridge: Cambridge University Press, 1986; Martin Chanock, Law, Custom and Social Order. Cambridge: Cambridge University Press, 1985.

6

Bruce Berman, Dickson Eyoh & Will Kymlicka

culture. With these tools, ethnic intelligentsias, strikingly similar socially to the petty-bourgeois intelligentsias that created the ethnic nationalisms of Europe, 14 were able to combine and reinterpret indigenous and borrowed cultural elements into 'imagined communities' larger in scale and more culturally coherent than any that had previously existed. 15 On a more limited, but nonetheless politically significant scale, the development of Western anthropology based on firsthand fieldwork in Africa, created integrated, logically coherent and scientifically authoritative versions of the ostensibly homogeneous cultures and institutions of several important ethnic communities. Anthropology, too, provided compelling cultural resources for African intellectuals to create influential accounts of their peoples that were also resources of political mobilization. 16 Ethnic development in colonial Africa had distinct, although linked, urban and rural contexts. The diverse, polyglot cities built on internal migration from various regions of a colony provided a cockpit for encountering the ethnic 'other' and conceptualizing the ethnic 'self for both the educated and literate intelligentsia and the illiterate laborer. Through cultural and mutual aid societies and a developing vernacular press, the 'authentic' values, language and communal interests of the rural ethnic 'home' could be articulated and debated. At the same time, the urban encounter with other groups doing exactly the same thing crystallized the competition and confrontation of political tribalism. 17 In the rural homelands, ethnic development involved the collaborating chiefs and headmen, the local intelligentsia and petty bourgeoisie, Christian converts and traditionalists, women and youth in a politics of moral ethnicity focused on contested issues of custom, the moral claims of leadership, and the relations of genders and generations. Urban and rural contexts were linked by numerous personal journeys between them, including the final passage home for burial. 18 The material substance of ethnic politics derived from the impact of colonial capitalism on indigenous processes of class formation and socio-regional differentiation. Conflicts over class formation were imbricated with ethnicity both within and between communities (see the chapters by Berman and Odhiambo). Changes in the social relations of production mattered, creating new disputes within kin groups and making ethnicity into an arena of conflict over the moral and material alienations of class formation that threatened established relations of indigenous moral economies. The differential access of men and women to the 14

Benedict Anderson, Imagined Communities. London: Verso, rev. edn, 1991. John Lonsdale, 'The Moral Economy of Mau Mau,' in Bruce Berman and John Lonsdale, Unhappy Valley: Conflict in Kenya and Africa. London: James Currey, 1992; Terence Ranger, 'The Invention of Tradition in Colonial Africa,' in Eric Hobsbawm and Terence Ranger, eds, The Invention of Tradition. Cambridge: Cambridge University Press, 1983; Lenz, 'Tribalism and Ethnicity'; Peel, 'Cultural Work', and Falola's chapter in this volume. 16 Jomo Kenyatta, Facing Mount Kenya. London: Seeker and Warburg, 1938; Bruce Berman, 'Ethnography as Politics; Politics as Ethnography: Kenyatta, Malinowski and the Making of Facing Mount Kenya,' Canadian Journal of African Studies, 30, 3, 1996; Kofi Busia, The Position of the Chief in the Modern Political System of Ashanti. Oxford: Clarendon Press, 1951. 17 Berman, 'Ehnicity, Patronage and the African State,' pp. 323-9; Lenz, 'Tribalism and Ethnicity,' pp. 308-10; Eriksen, Ethnicity and Nationalism, pp. 10-69 passim. 18 Crawford Young, 'Nationalism, Ethnicity and Class in Africa: a retrospective,' Cahiers d'Etudes africaines, 103, xxvi-3, 1986, pp. 445-6; Carola Lenz, 'Home, Death and Leadership: Discourses of an educated elite from north-western Ghana,' Social Anthropology, 2, 2, 1994; Leroy Vail, 'Ethnicity in Southern African History,' in Vail, ed., The Creation of Tribalism, pp. 7-11; Terence Ranger, 'The Invention of Tradition Revisited: the case of Colonial Africa,' in Terence Ranger and O. Vaughan, eds, The Legitimacy of the State in Twentieth Century Africa. London: Macmillan, 1993, pp. 20-7, 46-8; Sara Berry, No Condition is Permanent. Madison, WI: University of Wisconsin Press, 1993, pp. 32-9. 15

Ethnicity & the Politics of Democratic Nation-Building

7

proceeds of cash crop production and trade, and of youth to the wages of labor created, in particular, new axes of confrontation between genders and generations as male elders sought to sustain their power and control over women and young men (see chapters by Marks, Diouf, and Lonsdale). The constant debate over rights and obligations, relations of inequality and reciprocity shaped the character of political leadership and made moral ethnicity into a multi-layered dialogue between leaders and followers (see chapters by Lonsdale and Odhiambo). Within developing ethnic communities class conflict was thus expressed in confrontations over an implicit moral contract of leadership between the wealthy and powerful and their poorer clients and dependants; those who claimed leadership through their wealth and power and those who accepted such claims and expected to be protected and rewarded for doing so were bound in a community of trust and reciprocity. However, whose claims to leadership would be recognized was a matter of conflict between chiefs and headmen, educated teachers and clergymen, small businessmen and traders, and prospering cash-crop farmers. Similarly, whose claims to the patronage and support of leaders would be recognized and rewarded was a matter of conflict among the rest of the community, with contingent and revisable outcomes in particular groups. The uneven spread of colonial economic and infrastructure development between cash-crop and labor reserve regions, mining zones and centers of urban commerce and industry introduced significant regional differentiation in access to cash-crop production, trade, education, wage labor and state employment amongst different ethnic communities to produce sharper edges in the confrontation and competition of political tribalism. The significant economic growth and rapid amplification of the 'development' programs of the state in the late colonial period after World War Two greatly increased the resources available in both state and market, and the stakes of ethnic competition for them. Sustaining positions of leadership in ethnic communities increasingly rested on effectively claiming a share of the 'national cake' providing resources for both collective projects of 'development' in the community and individual benefits of patronage. At the same time, the growing conflicts of political tribalism allowed leaders to manipulate appeals to ethnic solidarity that could override the internal conflicts of moral ethnicity and obscure the development of class cleavages and confrontations (see chapters by Marks, Lonsdale, and Odhiambo). Understanding the complex interactions between indigenous societies and the intrusive forces of Western modernity in the colonial state and capitalism that have produced the internal and external contests of ethnic formation in Africa leads to another important point: the plurality of trajectories of change and the varying experiences of modernity that have resulted. Against the pervasive teleological assumptions of Western development theories, whether in modernization theory, neo-Marxism or neo-liberalism, of a unilinear path to a singular modernity, we are called to remember Malinowski's warning in the 1930s that the encounter of European and African cultures would produce something strikingly different from either, rather than a replication of European modernity or a preservation of African 'tradition'. 19 The continued production of ethnic difference suggests the need to explain complexity and singularity in the African experience of change and the plurality of potential outcomes of economic and political development. The connection of ethnic development with the growth of the institutions of 19 Bronislaw Malinowski, 'Preface,' in Methods of Study of Culture Contact in Africa. London: International African Institute, Memorandum XV, 1938.

8

Bruce Berman, Dickson Eyoh & Will Kymlicka

the modern state under colonialism, in which individuals were linked to the predominantly bureaucratic institutions of the state through pervasive patron-client networks, meant that in Africa people related as subjects and clients, rather than citizens, to an authoritarian and paternalistic state. It was a state, moreover, that, as Peter Ekeh notes in his chapter, was typically subject to the pressure of foreign interests to exploit the indigenous population and provided little in the way of protection or security. At the same time, throughout the colonial and most of the post-colonial eras the state also was the greatest source of wealth and power, and became the central focus of attempts to accumulate both. Ordinary people sought patrons for access to resources of the state as well as protection from its abuses; and such patrons were primarily available within ethnic communities in which they could claim membership. The critical consequence of this pattern of state-society linkage was the generation of a politics of opportunistic materialism, the 'politics of the belly', that made the maintenance of patron-client networks and the conditions of successful leadership increasingly dependent on the distribution of material benefits. 20 Moreover, the growing materialization of patron-client relations sharpened the internal conflicts of moral ethnicity over the obligations and reciprocities of leadership, and reduced the external confrontations of political tribalism to an amoral free-for-all for control of state resources. Finally, the contradictions between the processes of ethnic construction and patronage politics and the anti-colonial nationalist movements of the late colonial period and post-colonial projects of nation-building are becoming increasingly clear. Nationalist thought, with visions of a united post-colonial nation, stumbled on the question of ethnic diversity. The most typical position was a vigorous rejection of a supposedly archaic and atavistic tribalism, as in Samora Machel's often quoted dictum that 'for the nation to live, the tribe must die.' 21 At the same time, nationalist movements were frequently identified as under the domination of particular ethnic communities which sought ascendancy in the new state (see chapters by Falola, Mustapha, Lonsdale, and Odhiambo), and this could provoke the competitive political mobilization of other ethnic communities that feared marginalization. Even before independence, nationalist movements were frequently rent by internal confrontations of political tribalism and held together by tenuous coalitions of ethnic leaders based on promised divisions of the resources of the state. Regardless of their visions of the nation, in the rough-and-tumble competition of political tribalism within nationalist movements, leaders relied on an ethnic base of support and their links with its patronage networks. Such contradictions were rarely resolved. As Githu Muigai notes in his chapter, Jomo Kenyatta remained torn between being the President of the Republic of Kenya and paramount chief of the Kikuyu. Post-colonial attempts at nation-building were overlaid on top of ethnically defined patronage politics, which rapidly reproduced itself within national institutions of states and parties. Given this capture of national institutions by ethnically defined groups and parties, it is perhaps surprising that nation-building efforts have had any success at all. Yet many Africans do have at least an incipient sense of attachment to national institutions, and a desire to see them work for the benefit of all 20 Bayart, The State in Africa; Rene Lemarchand, 'The state, the parallel economy, and the changing structure of patronage systems,' in Donald Rothchild and Naomi Chazan, eds, The Precarious Balance: State and Society in Africa. Boulder, CO: Westview Press, 1988. 21 Samora Machel quoted in Mamdani, Citizen and Subject, p. 135.

Ethnicity & the Politics of Democratic Nation-Building

9

citizens, so long as this does not threaten the basic survival, cultural identity or economic well-being of their own group. Since the process of ethnic construction continues into the present, the issue of how sub-national ethnic communities and identities can co-exist with the development of universal citizenship, national identity and strong national democratic institutions remains the enduring dilemma of African politics.

'Democratization' and Ethnic Conflict in Contemporary Africa There is little doubt that the wave of 'democratization' in Africa since the 1990s has seen an increase rather than decrease in the visibility of ethnic politics and conflict. The country case studies presented here concur with the widespread opinion that the return to multiparty electoral competition has led to an intensification of the 'politics of primary patriotism'. 22 Multiparty politics reshapes the contexts of struggle among elites seeking to defend or challenge the distribution of state power and resources. It obligates both incumbent elites, long accustomed to rule without popular mandate, and opposition elites, to compete openly for the support of ordinary citizens. In effect, by prompting rearrangements of power relations at all societal levels, multiparty politics opens spaces for the 'venting' of long-entrenched elite and communal cleavages. The fragmentation of broad, urban-based opposition movements into parties with core ethnic constituencies and the consequent regionalization of political competition are the most obvious expressions of this trend. It is a trend that is strikingly reminiscent of politics at the terminal phase of colonialism when elite attention shifted from the defeat of colonialism to struggle for control of the resources of successor states. While the form, intensity, and ramifications vary in accordance with the ethno-regional make-up of societies, the fount of ethnic competition remains the weaving of communal cleavages into the fabric of state power and uneven regional processes of economic transformation from which these cleavages derive their material content. The informal clientelistic networks that dominate politics have involved hierarchical patterns of incorporation and exclusion of ethno-regional elites and communities within the state system of power. This has ensured that both elite and popular evaluations of the relationship between the distribution of state power and material opportunities is framed in terms of class and communal advantage or disadvantage. The military-managed centralization of political and economic power after the civil war (1970) in Nigeria, for example, that was made possible by the emergence of petroleum as the pre-eminent source of public income, resulted in the central (federal) state eclipsing regional governments as the main theater of accumulation and class formation. The continually reconstituted alliances amongst the three dominant ethno-regional elite blocs (North, East and West) and elites of minority communities for control of the central state has been regulated by the determination of the 'northern' elites (civilian and military) not to relinquish control of the central state (see chapters by Mustapha and Falola). The undiminished salience of this fundamental future of Nigerian state organization and politics since independence is evident in the pattern of electoral competition for national office between and within regions in the most recent 22

Peter Geschiere and Josef Gugler, 'Introduction,' Africa, 68, 3, 1998, pp. 309-19.

10

Bruce Berman, Dickson Eyoh & Will Kymlicka

democratic experiment. It also incubates the post-electoral sectarian violence especially in the northern states, where elites who have come to view the current regime as inimical to 'northern interests' have been threatening the imposition of Sharia as the juridical foundation of public authority (Mustapha). In Kenya, the consolidation of the Moi regime called for a remake of the Kenyatta regime's multi-ethnic elite alliance in a way that seriously diminished the economic and political power of the Kikuyu faction. The regime has responded to the challenges of multiparty politics by a combination of stateorchestrated violence and an alliance between elites from Luo, Kalenjin and other minority ethnic communities who find common purpose in precluding the political resurgence of the Kikuyu elite (see chapters by Odhiambo and Muigiai). The inner sanctum of Biya's regime in Cameroon has been composed of elites from his Beti and co-ethnic groups. Here, too, the regime has succeeded in deflating mass-based opposition to its incumbency by a mixture of state violence, manipulation of administrative rules and pressures on elites who are wedded to state patronage to become political leaders of their communities (Eyoh). South Africa is held, with some justification, as an exception in the modern African political trajectory. Its present constitutional arrangements seem designed, and have been lauded, for succeeding in dissuading ethnic political mobilization (Simeon and Marks). Yet the post-apartheid state is not without traces of the above characteristics of African post-colonial states. The apartheid state was built on a tripartite race-based hierarchy of citizenship. Despite its impeccable commitment to an equal and universal citizenship, the ruling African National Congress (ANC) is overwhelmingly supported by the African (black) majority, while the major opposition parties are backed mainly by nonblack (White, Indian and Colored) minority ethnicities (Hendricks, Simeon and Murray). Another powerful propellant of 'politics of primary patriotism' is competition for power between competing elites from the same ethnic communities. This is hardly surprising as elites of ethnic communities are never homogeneous; the symbols, grievances, and expectations that are marshaled by elites to foster ethnic political consciousness are typically contested by other elites from the same community (see chapters by Lonsdale, Mustapha, Falola, Hendricks, Marks, and Odhiambo). Multiparty political competition accentuates the ingroup competition for leadership characteristic of the process of moral ethnicity by making control over local and regional populations all the more imperative for political success. Although instigated by urban-based social groups, current processes of political liberalization have lent a new or renewed significance to rural society, which, with the exception of heavily urbanized South Africa amongst the cases, is demographically predominant in African states. Across African societies, then, the vast number of political constituencies outside the 'cosmopolitan' cities and towns continue to be differentiated by markers of cultural difference, ethnic and/or religious. This reality, understandably, encourages and rewards elite manipulation of kinship ideologies and communal identities in the quest for local and regional leadership. Two recent and increasingly common forms of such manipulation contribute significantly to inter- and intra-communal political competition and conflict. First, there is the invocation of distinctions between the 'natives/indigenous' and 'stranger/migrant' groups to assert the rights of communities to be represented by elite 'sons of the soil'. Second, there is the resort to ever narrower definitions

Ethnicity & the Politics of Democratic Nation-Building

11

of kinship boundaries to found claims for leadership within culturally-related groups. The varied uses of autochthony, as opposed to residence, as the core principle for determining local/regional leadership buttress conceptions of politics as primarily a struggle for supremacy between ethnic communities or kinship groups. The political efficacy of the manipulation of localized kinship ideologies rests on the ability of elites to repress internal dissent over their conceptions of ethnic and kinship boundaries. It feeds the increasingly violent politically motivated confrontations in local society that often pit one ethnic minority or segments of the same ethnic group against another (Mustapha, Falola, Marks, and Eyoh). The internal as well as external dynamics of political ethnicity challenge the often implicit assumption of elite-centered perspectives that ordinary citizens are unwitting victims of a form of political competition organized by and for the benefit of corrupt elites. The prevailing social, cultural, economic and political factors outlined earlier that have led to the historical development of African ethnic communities and their particular relationships to the state continue to predispose ordinary citizens to privilege kinship and communal affinities as a premise for political participation. On the one hand, electoral competition has accentuated conflict within ethnic communities over elite claims to leadership and classbased confrontations over the moral obligations and reciprocities of rich and poor. The example of Kenya stands out strongly (Lonsdale, Muigai, Odhaimbo). On the other, political liberalization also has opened up space for the articulation of inter-ethnic regional grievances that were often repressed by authoritarian regimes. These grievances stem from the hierarchical incorporation of ethnoregional communities within the state system of power and the attendant intercommunal inequalities of access to resources in modernity. The moral validity and proposed redresses of regional grievances are matters for debate. The examples of Nigeria (Mustapha, Falola, Ejobowah,) Cameroon (Eyoh) and Senegal (Diouf) suggest, however, that the 'feelings' of collective disadvantage that impel regionalist movements are shared by cross-sections of elites and commoners of concerned communities. Neo-liberal programs of reform that sanctify freemarket principles of efficiency in the allocation of public investments promise to exacerbate regional economic, social and political disparities. However constructed, transformed and instrumentalized politically, ethnicity is always or nearly always metaphoric kinship 23 . For the vast majority of contemporary Africans, the metaphorical kinship of ethnicity remains crucial to securing basic security, and similar to the 'horizontal kinship' of nationalism of peoples all over the world, 24 to their conceptions of selfhood and social belonging. It is, thus, the durability of kinship as the most fundamental unit of social trust (Ekeh and Berman) that ultimately grounds the vitality of ethnicity as the idiom of political identity and competition in post-colonial Africa. The historical experiences and repertoires of cultural practices that structure and differentiate kin-based ethnic communities embed normative references for judging claims to communal belonging and the exercise of political authority within. To recall Lonsdale's persuasive argument, moral ethnicity gets perverted into political tribalism when ethnic groups collide in competition for resources in stateordered arenas, and the measure of effectiveness of political representation is the 23 Thomas Hyland Eriksen, 'A Non-Ethnic State in Africa? A Live World Approach to the Imaging of Communities/ in Paris Yeros, ed., Ethnicity and Nationalism in Africa: Constructivist Reflections and Contemporary Politics. Basingstoke: Macmillan, 1999. 24 Anderson, Imagined Communities.

12

Bruce Berman, Dickson Eyoh & Will Kymlicka

ability of elites to promote the interests of their primary community through state institutions dominated by ethnic patrons and their clienteles 25 (Berman and Ekeh). Subordinate groups, even if they are ultimately the losers in the flow of the resources of modernity through patronage networks, are equally adept at the deployment of kinship and ethnic ties for political purposes for a simple reason: it enables evaluation of the legitimacy of elite political leadership through a shared, if constantly contested, moral vocabulary. And few modern elites can escape without personal conflict the moral obligations of their primary group membership, which is also central to their individual and social identities (Ekeh, Odhiambo, Lonsdale). The failure of post-colonial states to service competently and without bias the most elementary material and security needs of their citizens has compounded reliance on kinship and ethnic networks by individuals and groups seeking ways to cope with unrelenting economic hardship and for sanctuary from elite-orchestrated political violence (Ekeh, Berman). This has prompted calls not only for decentralization of state administrations, but also for new approaches to nationmaking that recognize and allow for the political expression of the social pluralism of African societies. The arguments behind such advocacy are fashioned differently; their common premise is that citizenship in ethno-cultural communities retains its far greater relevance to the political behavior of the majority of Africans than the abstractions of (universal) national citizenship. What are needed to enhance the accountability of political systems are institutions that validate indigenous precepts of political community and authority and permit representation of both individual and communal interests. 26 Several of the essays in this volume review the normative premises and problems of institutional models of how best to reconcile the competing demands for individual and communal representation in democratic nation-building processes (Berman, Kymlicka, Ejobowah, and Simeon). We discuss these models in the next section. However, it is worth noting that some African countries have already experimented with new approaches to nation-building that give greater space to indigenous identities and greater representation of traditional authorities. Some countries, for example, have attempted to decentralize power to sub-national political units that are defined along ethnic or cultural lines. By itself, however, this has not resolved problems of political ethnicity or political accountability. Reorganization of states in this fashion risks encouraging solidification of ethnic boundaries and cementing competition between ethno-regional oligarchies as the basis of national politics (see Eyoh). Historical and contemporary migrations have ensured that localities and regions in African nation-states that are ethnically homogenous are increasingly rare. As Mamdani has forcefully argued, colonial use of cultural groups' membership, rather than residence, as the exclusive criterion of citizenship in rural society imposed as one of the most compelling tasks of post-colonial nationbuilding resolution of the question 'when does the stranger become a citizen?' 27 . 25

John Lonsdale, 'Moral Ethnicity and Political Tribalism,' in Preben Kaarsholm and Jan Hultin, eds, Inventions and Boundaries: Historical and anthropological approaches to the study of ethnicity and nationalism. Roskilde, Denmark: Institute for Development Studies, University of Roskilde, 1994. 26 Patrick Chabal, Power in Africa. New York: St. Martin's Press, 1992. 27 Mahmood Mamdani, When does a Settler become a Native?: Reflections on the Colonial Roots of Citizenship in Equatorial Africa and South Africa. Inaugural Lecture as A. C. Jordan Chair of African Studies, Cape Town: University of Cape Town, 1998.

Ethnicity & the Politics of Democratic Nation-Building

13

Unless this question is answered in a fair and democratic way, the use of cultural criteria, in effect the principle of autochthony, to delineate citizenship rights within sub-national units risks further encouragement, if not the formal institutionalization, of differential and unequal local citizenship. Other countries have experimented with the use of traditional political institutions as the basis of local democracy. Whatever the merits (and they are considerable) of such proposals, it remains the case that these institutions are almost by definition hierarchical and authoritarian. Shaped by and addressed to the needs of primary local groups, they tend to exclude minorities and migrant groups from political participation (Solway on Botswana). They are also inherently patriarchal and deny women the right to equal participation in local politics. Not surprisingly, the interests of so-called traditional African women were least well served by attempts to constitutionalize chiefly authority in postapartheid South Africa that conflict with the gender equality enshrined in the constitution (Marks). These examples show that greater formal accommodation of social pluralism (through such things as ethnic federalism or traditional chiefly authority) is not necessarily or inherently beneficial to democracy. If we are to find a way of reconciling ethnicity and democracy, we need to find ways of encouraging pluralism without compromising norms of freedom and equality. That is the challenge we discuss in the next section.

Strategies of Reconciliation and Democratization Despite the media stereotype that Africa is uniquely afflicted by ethnic and tribal conflicts, the fact is that many other countries around the world have faced comparable problems. Very few countries are united by common descent, language, religion and culture. Perhaps only Iceland, Portugal and the Koreas could plausibly be described in this way. In every other country, the sense that there is one united people inhabiting one connected country has had to be constructed. Developing a sense of common citizenship, and of loyalty to a common state, amongst people who differ in their identities, cultures and religions is a difficult task, but by no means an impossible one. Many multi-ethnic states have proved to be remarkably successful in it, and there is no basis for the pessimistic assumption that multi-ethnic states are inherently unviable or unstable, or incapable of democratization. It is often said that states in Africa are 'artificial', and that state borders were drawn without much attention to the historic identities and cultures of the people who live within them. But that was also originally true of many borders in Europe, Asia or the Americas, which were the result of conquest, imperial treaties, or dynastic marriages. Very few arose from the democratic will of the people themselves. If these borders now seem natural or reasonable, it is because states have succeeded in giving citizens a reason to identify with the state, and to see it as 'their own'. Indeed, a survey of the countries of the world should teach us, not that multi-ethnic states are unviable, but that states have many tools for constructing common loyalties amongst a diverse population. In this section, we discuss five (potentially overlapping) approaches that have been adopted by democratic states to reconcile ethnic diversity and common citizenship, and that have been adopted or recommended in the African context:

14 -

Bruce Berman, Dickson Eyoh & Will Kymlicka

a 'neutral' or 'difference-blind' state Jacobin republicanism (nation-building from above) civil society (nation-building from below) federalism/decentralization consociationalism

All of these are intended to create freedom and equality for the members of all ethnic groups within a democratic regime. There are, of course, other alternatives which are premised on ethnic hegemony rather than equality. These typically take the form of what Ian Lustick calls the 'control' model of regulating diversity: one ethnic group not only controls the state and its major institutions (including the bureaucracy, police and army), but also dominates the other groups, containing them within a position of political marginalization and socioeconomic disadvantage. 28 This has perhaps been the dominant model in most of the world historically. In most cases, such 'control' regimes are undemocratic. But forms of ethnic hegemony can exist even within nominally democratic regimes: they then take the form of what Sammy Smooha calls 'ethnic democracy'. 29 In such systems, there is universal franchise, but the state is nonetheless defined as the property of the dominant group, and a variety of techniques are used to ensure that all the important decision-making positions in the state are reserved for members of that group. Oft-cited examples include Israel, Latvia and Estonia, Northern Ireland (under Home Rule), and Malaysia. Regimes of control and ethnic democracy can be quite stable, and vary in their level of oppression. A stable and minimally oppressive ethnic democracy may be the best that can reasonably be expected in some circumstances. But these regimes are obviously flawed from the point of view of justice and democracy. So our focus in this section is on proposals - such as the five listed earlier which at least aspire to a more inclusive democratic regime, in which the state would be seen as equally belonging to, and serving the needs of, all its citizens. In thinking about the relevance of these five models in Africa, we must not lose sight of the distinctive conditions and challenges facing African countries today. While ethnic diversity and 'artificial' borders are not, by themselves, unique to Africa, there are other ways in which the African context is distinctive. To oversimplify, we can say that, in the West, the process of accommodating ethnic diversity has taken place within states that have reasonably well-functioning market economies and democratic political systems. The challenge, in short, was to pluralize already existing and functioning liberal-democratic economic and political systems. Elsewhere in the world, however, the challenge of ethnic conflict is magnified by the fact that it is occurring simultaneously with other radical transformations of the state and the economy. Latin American states, for example, are undergoing a double transformation: they have to deal with problems of ethnic conflict (particularly relating to indigenous peoples) at the same time as they are moving away from systems of military dictatorship. Eastern European states are also facing a triple transformation: they are having to deal with ethnic conflict (particularly from linguistic and national minorities) at the same time as they are both moving away from a system of one-party Communist dictatorships, and shifting from a centrally 28 Ian Lustick, 'Stability in Deeply Divided Societies: Consociationalism versus Control', World Politics, 31, 1979. 29 Sammy Smooha and Theodore Hanf, 'The Diverse Modes of Conflict-Regulation in Deeply Divided Societies', International Journal of Comparative Sociology, 33, 1, 1992, pp. 26-47.

Ethnicity & the Politics of Democratic Nation-Building

15

planned economy to a market economy. It is not surprising, therefore, that issues of ethnic conflict have proved more dangerous and destabilizing in Eastern Europe than in either Western Europe or Latin America. Fears about ethnic conflict are exacerbated when people are already suffering from broader forms of economic, political and social insecurity. But even in Latin America and Eastern Europe, functioning states and national economies existed to be democratized and liberalized. In many African countries, by contrast, state construction is still incipient and incomplete. There is an urgent need to build state capacity, so that states can effectively secure public order, uphold the rule of law, and meet the basic needs of citizens. 30 We could say, then, that African states face a quadruple transformation: they must negotiate ethnic diversity at the same time as they are building state capacity, democratizing political systems and liberalizing economic institutions. Given these factors, Western models of democratic pluralism cannot simply be exported into the very different circumstances of Africa today. Some pessimists argue that these difficult circumstances make the very idea of liberaldemocratic pluralism impossible in Africa. The best we can expect, they argue, is a less oppressive form of authoritarian control. And indeed, as we shall see, the track record when these proposals have been implemented in Africa is not good. Yet this sort of pessimism is premature, and potentially misdirected. After all, the track record for regimes of authoritarian control in Africa is also not good. If some of these proposals seem idealistic, one could respond that it is even more naive to suppose that current systems of hegemonic control will work in the long term. All around the world today people demand a voice in the way their lives are governed, and demand a level of security and respect from the state. Older models of authoritarian control can sometimes gain the passive acquiescence of their subjects, but successful states today require the active allegiance and participation of citizens. Moreover, it is important not to hold African states to standards that Western countries themselves do not meet. It may be difficult to imagine eliminating ethnic bloc-voting, or ethnic favoritism in patronage, in many parts of Africa. But these phenomena exist in the West as well. Political life in many Western democracies remains divided along ethnic or linguistic lines. The goal should not be to somehow purify politics of all forms of partiality or favoritism, which is impossible, but rather to establish political institutions and conventions which make all citizens feel secure and respected by the state, whatever their ethnicity, language or religion. The goal, in short, is not Utopian harmony or altruism amongst all ethno-cultural groups, but simply learning to manage ethno-cultural diversity and ethno-cultural conflict in a constructive rather than destructive way. Given the quadruple challenge facing many African countries, there can be no magic formulas or simple solutions. No matter what approach African countries adopt, they will face difficulties not faced in other parts of the world, and will have to come up with their own variations, adapted to local conditions and customs. With these provisos in mind, let us now turn to the five models: (i) The Difference-Blind State: When asked how states should respond to ethnocultural diversity, many people, particularly those trained in the liberal tradition, will want to say that the state should simply ignore these differences. The state 30

On the inability (and/or unwillingness) of African states to meet the basic security needs of their citizens, see the provocative chapter by Peter Ekeh in this volume.

16

Bruce Berman, Dickson Eyoh & Will Kymlicka

should allow people to develop and express their cultural practices and identities in private - in the home, church or private associations - so long as they respect the rights of others to do the same. But the state should neither promote nor discourage cultural affiliations and practices. Ethno-cultural diversity should simply be privatized, and the state should be blind to the private cultural choices of individuals. This is a popular model amongst many liberals because it extends to the case of ethno-cultural differences a model which has proved very successful in the case of religious differences. During the Wars of Religion in Western Europe, Catholics and Protestants fought over a century of civil wars to decide which religion should be the official state-sponsored religion. Both sides agreed that there must be an official religion, and that believers in any other religion should be subject to discrimination and persecution. They simply disagreed about which religion it should be. However, when it became clear that neither side had the military capacity to defeat the other, a compromise slowly emerged, which involved the separation of church and state. People would be free to attend whatever church they wanted in private, but the state itself would have no official religion, and would be indifferent to the religious beliefs of its citizens. This model of a separation of church and state has proved to be surprisingly successful in the West. The division between Catholics and Protestants, which produced untold and seemingly unending violence, has now become almost entirely pacified and depoliticized in most Western countries. The state has become more stable by privatizing religion, and the religious groups themselves have thrived and prospered without state support or sanction. Given this apparent historic success with religion, many people naturally want to apply the same model to ethno-cultural differences. However, there are two obvious limitations to this model. The first is that it requires considerable selfrestraint on the part of dominant groups who control the state, and hence who have the power to adopt state policies supporting their culture. This selfrestraint was only acquired by Catholics and Protestants in the West after over a century of bloodshed. It would be naive to suppose that dominant groups will not always be tempted to use their control over state resources to promote their identities and practices. More importantly, this difference-blind strategy is in fact impossible. The fact is that the state cannot avoid implicitly or explicitly supporting some cultures over others. Most obviously, the state must make decisions about the language of public administration, public health care, schools, public media, road signs and so on. Any group which manages to get its language adopted as a state language in this way can gain enormous benefits, while other groups will face pressure to assimilate to this state-sponsored language group. Many African countries have tried to avoid the danger of linguistic favoritism by simply adopting the colonial language as the state language. But this does not solve the problem of language policy at the more local level. As most of their populations are illiterate in official languages, public institutions at more local levels tend to operate in the local vernacular(s). Moreover, there are many other areas where states cannot avoid giving public recognition and support to some ethno-cultural groups. For example, the state makes decisions about which holidays to recognize, which authors to teach in schools, and which heroes or events to celebrate when naming streets, towns and topographic features. These decisions almost invariably involve giving public recognition and support to certain ethno-cultural groups over others (usually the majority or dominant group, of

Ethnicity & the Politics of Democratic Nation-Building

17

course). For these reasons, talk about 'difference-blindness' often simply obscures the fact that states are inevitably involved in making decisions which recognize and benefit some ethno-cultural groups while ignoring or disadvantaging others. (ii) Jacobin republicanism: A second strategy accepts the premise that the state is unavoidably involved in promoting a particular language, culture and identity, but tries to turn this into a virtue rather than a vice. The goal, on this view, is for the state to deliberately support and diffuse a common language and culture which will be defined as the 'national' language and culture, to which all citizens should assimilate. While this language and culture may historically have originated in a particular ethnic group, the state should redefine it as a 'universal' language and culture, belonging equally to all citizens. The goal, in other words, is not to privatize culture, but rather to de-ethnicize it, so that a particular language and culture becomes the official and public language and culture, and is redefined as the joint possession of all citizens, not of any particular ethnic group. This is, of course, the French model of citizenship, in which all citizens are expected to assimilate to a particular national language, republican political heritage, and secular culture. It has had considerable success in producing a unified and cohesive political community within France, into which many groups have assimilated. The sharing of a common language and national culture has helped strengthen democratic trust and solidarity across ethnic, religious and regional lines within France. And yet this strategy also has serious limitations. In fact, it probably only succeeded in France because massive coercion (for example, forbidding people to speak or publish in a minority language) was used in the nineteenth century to assimilate groups like the Basques and the Bretons. The same was true about the coercion needed to suppress minority groups in other Western states that now take for granted a common national language and culture. It is not clear that this strategy can work today, or that the international community would tolerate the level of coercion needed to make it work. David Laitin provides a nice example of how our views regarding state coercion have changed over the centuries: It is said that in Spain during the Inquisition gypsies who were found guilty of speaking their own language had their tongues cut out. With policies of this sort, it is not difficult to understand why it was possible, a few centuries later, to legislate Castilian as the sole official language. But when Emperor Haile Selassie of Ethiopia pressed for policies promoting Amharic, infinitely more benign than those of the Inquisition, speakers of Tigray, Oromo, and Somali claimed that their groups were being oppressed, and the international community was outraged. Nation-building policies available to monarchs in the early modern period are not available to leaders of new states today.31 To be sure, many African countries have tried to pursue this sort of top-down Jacobin nation-building strategy, particularly in Francophone Africa. But as the papers by Diouf (on Senegal) and Eyoh (on Cameroon) show, this model is bitterly resisted by some minority groups. 32 What is presented by the state as an 31 David Laitin, Language Repertoires and State Construction in Africa. Cambridge: Cambridge University Press, 1992, p. xi. 32 See also the chapters by Falola (on Nigeria), Hendricks (on South Africa), Muigai (on Kenya) and Solway (on Botswana) for other examples of how minorities have felt excluded from the 'unified national culture' promoted by their respective states.

18

Bruce Berman, Dickson Eyoh & Will Kymlicka

effort to promote a 'national' or 'universal' language and culture is perceived by minorities as an effort by the dominant group to privilege its particular language and culture. Where minorities cannot see themselves properly represented and respected in this 'national' culture, or have not played an active role in defining it, they typically respond by challenging these nation-building strategies, and in the extreme case may even challenge the very legitimacy of the state's authority over them. Moreover, African nations are composed for the most part of collections of ethnic minorities. Most lack ethnic majorities who may be tempted by their numerical preponderance to try to impose their language and culture as the 'national' language and culture. The limitations of this centralized, top-down nation-building strategy are a common theme in virtually every academic analysis of the post-colonial African experience. 33 As Jibrin Ibrahim notes, 'For a very long period, African institutions - the schools, the media, state organs and sometimes even religious institutions - have been propagating the virtues of national unity and the necessity of developing the national state.' 34 Yet the level of identification with the state remains very low, the strategy has simply not worked, and in many cases has backfired, by fuelling fear and resentment amongst groups who feel excluded. Indeed, most analysts would agree with John Markakis that this approach has been a 'disastrous failure'. 35 (iii) Civil society. A third strategy seeks to avoid the flaws of 'top-down nationbuilding' by arguing instead for 'nation-building from below'. The idea here is that a common national language, culture and identity will emerge, not as a result of imposition from a centralized and authoritarian state, but rather as a result of the mixing of peoples in the institutions of civil society, such as churches, trade unions, newspapers, environmental groups, women's groups, and so on. Out of these everyday and non-threatening interactions in civil society, inter-ethnic trust will develop, as will a new pan-ethnic vernacular and identity. In this way, nation-building will occur as a result of gradual evolution and consensus-building in civil society, not state imposition. This is obviously an attractive model, and Thomas Eriksen argues that it is in fact occurring in some African countries. He claims that Mauritius, for example, is in the process of developing a common set of supra-ethnic national myths and symbols which is invested with meaning and relevance by the bulk of the population, and has a 'high level of cultural integration, which makes a national public sphere possible', based on a Creole version of the colonial French language. 36 This, he argues, is the result of compromise and consensus from below, rather than (or in addition to) nation-building from above. 37 But, as Eriksen notes, Mauritius is unique in its island isolation, high education levels, and comparative lack of a gap between elites and masses. In most countries, the reality is that the associations of civil society are themselves already 33 'At the core of the problem is the nation-building strategy adopted in post-colonial Africa. So far, most African countries have pursued the strategy of imposing the nation and the polity rather than trying to generate a democratic consensus' (Lidija Basta and Jibrin Ibrahim, Federalism and Decentralisation in Africa: The Multicultural Challenge. Fribourg, Switzerland: Institute of Federalism, 1999, p. 4. 34 Ibid, p. 25. 35 John Markakis, 'Nationalism and Ethnicity in the Horn of Africa' in Yeros, Ethnicity and Nationalism in Africa, p. 70. 36 Thomas Hylland Eriksen, 'A Non-ethnic State for Africa?', in ibid., pp. 52-3. 37 The model of inclusive nation-building adopted by the ANC in South Africa, discussed in the chapter by Hendricks, has some affinities with this approach.

Ethnicity & the Politics of Democratic Nation-Building

19

defined and divided on ethnic lines. Proponents of this civil society strategy tend to describe the state as the home of vice and civil society as the home of virtue. But as Berman notes in his chapter, 'hopes that the development of civil society would be a force for democratization are particularly unrealistic, since with distressing frequency, the rhizomes of ethnic factionalism and patronclient politics reproduce themselves within these parties and associations, rendering them, like so much of the apparatus of state, into ideological and institutional facades covering the reality of business as usual'. Insofar as this is true, focusing on civil society simply relocates the problem, rather than providing a means of resolving it. (iv) Multination federalism: Given the limitations of both top-down and bottom-up nation-building in deeply divided societies, one obvious response is to give up the goal of forming a unified nation-state. If there is no feasible route to developing a cohesive sense of national identity, or a common sense of loyalty to the nation-state, why not abandon the very idea of a nation-state, and accept that the state is 'multinational'? Such a multination state can be seen as a federation or partnership of various groups, each of which will retain its distinctiveness and its right to autonomy or self-government. This multination state can take two forms. Where groups are more or less territorially concentrated, it is likely to take the form of federalism. In a multination federal system, the country is divided into several sub-units whose borders are drawn in such a way that each of the various groups will form a local majority in one or more of the sub-units. By de facto controlling a sub-unit, even if they are a minority in the country as a whole, each group is able to feel a sense of security, and can use the levers of sub-state power to protect and promote its identity and culture. Such a model of multination federalism has been successfully adopted in several Western democracies, including Canada, Switzerland, Belgium and Spain. Quasi-federal regimes have also been adopted in the United Kingdom, Italy and Finland to create autonomous sub-units in which national minorities form a local majority. And we see similar developments in India, Russia and Malaysia. There is a long history of promoting similar forms of federalism in Africa. To date, however, it has been relatively unsuccessful. Many African federations have failed (the Mali Federation); others exist primarily on paper (Ethiopia). Some critics have concluded that federalism is an 'abject failure' in Africa. But as Ladipo Adamolekun notes, unitary states in Africa do not have a superior record in terms of democracy, human rights, peace or economic development, and so 'the failures of federalism in Africa are not peculiar to federalism; they are part of the general failure of democratic governance on the continent'. 38 The merits of federalism are discussed in several of our chapters, and elicited a great variety of views. Some authors were sceptical. Berman, for example, says that federalism often simply devolves power to levels where problems of patronage and political tribalism are even greater. Eyoh notes that federalism might exacerbate the problem of the exclusion of internal migrants. If a sub-unit is seen as 'belonging' to a particular group, then 'sons of the soil' preference will be given or will agitate for preference over mere 'citizens' from elsewhere in the state. In effect, the problem of ethnic hegemony is solved at the central level by creating a series of ethnic hegemonies at the sub-state level. Mustapha goes so 38 Ladipo Adamolekun and John Kincaid, 'The Federal Solution: Assessment and Prognosis for Nigeria and Africa', Publius: The Journal of Federalism, 21, 1991, p. 174.

20

Bruce Berman, Dickson Eyoh & Will Kymlicka

far as to say that ethnic federalism is 'ahistorical and repugnant'. Other authors, however, are more optimistic. Ejobowah argues that only a federal system can work in Nigeria, and Simeon claims that the adoption of federalism has helped diffuse what many commentators expected to be a violent conflict in KwaZuluNatal. Kymlicka gives more general arguments for the appropriateness of federalism in multination states. (v) Consociationalism: In countries where ethnic groups are not territorially concentrated, federalism is obviously not a solution to the issue of ethnicity. In these cases, the idea of a multination state may instead be implemented through some form of consociationalism. In a consociational regime, the state may be unitary and centralized, but there are guarantees that all ethnic groups will share power at the central level. This may be achieved through rules regarding the representation of ethnic groups in the legislature, in cabinet, and in the civil service. Electoral systems can play an important role in encouraging or requiring powersharing in the central legislature. 39 For example, systems of proportional representation can ensure that political parties representing minority communities gain representation. Alternatively, electoral rules can oblige national parties to seek support across ethnic lines by requiring that they win a certain level of support in all regions in order to gain power. Consociationalism may even involve some form of veto right, so that all of the major groups in the country must agree on a policy, particularly if it involves constitutional changes or if it affects the basic interests of the groups. Like federalism, this model has been successfully adopted in some Western states, such as the Netherlands, Austria and Belgium. And it too has been promoted in Africa, with only limited success. The most obvious attempt to implement it, in Rwanda and Burundi, failed completely, but it remains a topic of debate in other African countries, including Liberia and Angola. Proponents argue that it may help to provide a sense of security amongst the members of the various groups, and help them develop some sense of identification with and loyalty to the state. It also eliminates the fear of secession or irredentism which is often raised in federal systems, since groups are not given control over territory. On the other hand, critics see it as rigidifying ethnic categories, and indeed as perpetuating and exacerbating the very problem of political tribalism it was meant to solve. It seems to reduce politics to a question of how to divide up the state's offices and resources amongst the various groups. Both federalism and consociationalism can help to provide a sense of security to groups, but a long-term solution requires not only security within groups, but also the development of trust and solidarity across groups. It should be clear by now that none of these five models offers any magical solutions. Nor should they be seen as mutually exclusive. Each offers insights, and states can combine them in new and experimental ways. Indeed, it is surely the case that African countries will need to be both courageous and experimental if they are to tackle effectively the challenges of ethnicity and democracy. Lastly, we must add a word on the global system and the prospects of democratic nation-building projects in Africa. Although this is not an explicit focus of 39

Some advocates of electoral engineering may view their approach as an alternative to Lijphartian consociationalism. We believe, however, that it belongs here, since for our purposes the key feature of consociationalism (as opposed to federalism) is power-sharing across ethnic lines at the central level, and all of the different models of electoral engineering aim at this result. We are not implying that electoral engineering is applicable only in explicitly consociational regimes, and believe, instead, that it is possible to combine elements from the different models.

Ethnicity & the Politics of Democratic Nation-Building

21

any of the chapters, the contributors take for granted that the new (post-Cold War) global order poses significant constraints on the choice of nation-building policies that are available to African states. In the past three decades, while Africa has been mired in unrelenting economic adversity, economic globalization has been proceeding at what appears to be an inexorable pace. Growing marginalization is the only fitting description of Africa's fate in a rapidly globalizing world economy. The region's share of world trade and foreign investment has been declining, and the dependence of its states and economies on Overseas Development Assistance is intensifying.40 The new global order is hostile to state-dominated strategies of development and dictatorial rule (top-down approaches to nation-building). The reigning ideology of development advocates market-driven approaches, and considers democracy to be an integral precursor of development, rather than a by-product. African intellectuals and policy-makers endorse the broad tenets of the new development orthodoxy. But, due to the diminution of sovereign authority in economic policy caused by high aid dependence, the responses of African states to the challenges that globalization poses to economic development and democratic nation-building are, to an extent that is unparalleled amongst developing country regions, under the sway of external actors. The weighty influence of external actors is most evident in the ubiquitous neo-liberal structural adjustment programs which have become synonymous with development strategy in Africa over the past three decades. In sentiment and practice, neo-liberal adjustment programs presume a single universal model of economic organization and development. In their insistence on the sanctity of free-market principles of efficiency in the allocation of public investments, they dissuade the conscious use of economic development policy as an instrument of statecraft and nation-building, including the management of ethnic diversity. The risks they pose to the exacerbation of economic, social and political disparities that feed ethno-cultural political competition and conflict cannot be underestimated. 40

For an elaboration of the challenges of the new global order and development orthodoxy, see Dickson Eyoh and Richard Sandbrook, 'Pragmatic Neo-Liberalism and Just Development in Africa,' in A. Kohli, Chung-In Moon, and G. Sorenson, eds, States and Markets: Just Development in the 21st Century. Tokyo: United Nations University Press, 2001.

2

PETER EKEH

Individuals' Basic Security Needs & the Limits of Democracy in Africa

T

HE dominant constructs of African political sociology were formulated in the 1960s. It was then fashionable to forge supraindividual categories as the essential foundation of society, thanks to the formidable presence of Parsonian sociology in that era. It was an age of theory-building in which individuals and their needs counted for little. Despite George Homans' (1964) exasperated plea for enhancing the individual's place in social theories, urging social scientists to choose the path of 'Bringing Men Back I n / the theoretical orientations that influenced African studies in the 1960s were essentially collectivistic. They led to the recognition of such ideational ensembles as state, society, ethnic groups and tribes. They were far more reluctant to recognize the potency of unique individuals and their needs in the formation of these entities. The neglect of the unique individual and his (or her) needs was not confined to academic practitioners who chose their theoretical models from Western social sciences. On the contrary, the neglect of the unique individual and his needs has been historic in Africa. The slave trade was based on spite for the worthiness of the unique individual. The most important state, which arose and functioned in the African history of the nineteenth century was the Fulani Sokoto Caliphate. It sprang from a revolution that was organized against Hausa kings who were accused of allowing too much freedom in the lives of their citizens (see Ekeh, 1997). The succeeding regimes of European colonialism, which occupied much of African history of the twentieth century, allowed very little room for the needs of the individual. In the post-colonial era, African states have done nothing to overcome the hostile relations between states and individuals in previous eras of African history. In a sense, therefore, the academic belittlement of the individual in African studies is a refraction from African history of the eighteenth, nineteenth, and twentieth centuries. As Africans march (or are dragged) into the twenty-first century with the rest of humanity, there is a severe dissonance between the imagery of African political sociology, suggesting strong relationships between the state and elements of society, and the precariousness of the state and, indeed, society in African countries. African political sociology continues to be modeled on the exigencies of Western historical realities which gave birth to a scholarship that celebrates a heathy relationship between states and individuals. I do not deny that it is proper to model Africa's political sociology on Western European experiences in the relationships between the state and individuals. Indeed, I urge it. However, it is not the current instances of Western European experiences with relationships between states and individuals that should count in the construction of African political sociology. On the contrary, I contend that pre-modern European political sociology and modern African political experiences are conjoined at a very 22

Basic Security Needs & the Limits of Democratization

23

elementary level of individuals' security needs and their significance for the evolution of state and society. At stake in such considerations is the methodology of comparison between African states and societies, on the one hand, and Western European states and societies, on the other. In much of African studies, the European state has been invoked as an exemplar from whose properties the structures and processes of African states may be understood. But the instances and patterns of the European state so invoked have varied enormously. Many have taken the mature European state, of the nineteenth and twentieth centuries, as the proper model for studying African political sociology. For instance, Afro-Marxists spent much intellectual capital in the 1980s in their efforts to study the behaviors of nonexisting 'ruling classes' and 'national bourgeoisie' in African countries in false imitation of these dominant groups in European and Western societies. This mode of comparison is based on the notion of the European state as an epiphenomenon with developed political parties, bureaucracies, and electoral practices. In such a comparison, the mature European state is taken as a finished social formation. Employing the alternative method of examining the historical processes through which European states and societies evolved into maturity would appear to provide a more productive method of comparison in studying African social formations. This mode of comparison will reveal both similarities and differences in the relations between the state and individuals in the differing paths of evolution of European and African states and societies. In addition, this latter form of comparison will incline us towards more elementary conditions of political sociology rather than higher-order conceptions, such as the ruling political classes that Afro-Marxists emphasized in their studies. It is a form of comparison that is liable to lead us to the lessons in Thomas Hobbes' discussions of the brute security needs of the unique individual, rather than to the superimposition of a paradigm of a mature Western state on African political sociology. A major conclusion from Hobbes' political philosophy is that individuals' basic security needs have an intrinsic power that compels the direction of the evolution of state and society. The European state evolved into a healthy maturity, differentiating itself from other elements of society, because it was able to attend to individuals' basic security needs. In the course of its evolution, a strong bond has emerged between states and individuals in Western societies. In Africa, the modern state has not undertaken to attend to the business of providing for individuals' basic security needs. Rather, and following colonial practices, this awesome responsibility has been allocated to ethnicity and other forms of kinship. This historic pattern of the state's neglect of individuals' basic security needs has emboldened the sentimental ties and bonds of trust between individuals and their ethnic groups and other forms of kinship, while the state is badly starved of individuals' loyalties.

The Hobbesian Complex Although it has almost exclusively been considered an aspect of Western history and philosophy, Thomas Hobbes' Leviathan poses a universal problem, concerning the individual's existence, which requires variable solutions. This universal problem of a degraded human condition and its variable solutions are meaningful to different political cultures, inside as well outside Western civilization. At its

24

Peter Ekeh

base, Hobbes' theory demonstrates the evil consequences for the individual's existence that flow from social disorders in the relationships between the elements of state and society. The theory weighs the problems created for the unique individual's existence by a condition of disorder in which he/she is robbed of the state's organizational prowess or society's normative protection. But it also offers opportunities for solutions, which are partial to particular societies and cultures, to such a universal problem. In the proximate case of Revolutionary England that supplied the context of Hobbes' theory, the acute problem of disorder arose from the notional absence of the state, threatening the individual's existence with anarchy in a painful life that was destined to be 'solitary, poor, nasty, brutish, and short'. The absence of the state and its organized governmental institutions yields to a state of nature embodying a war of everyone against everyone else in which 'the notions of right, justice and injustice have no place'. Hobbes' preferred solution was to summon the organizational potentials of the state in aid of restoring the order which individuals help to create by contractually giving up some of their native freedoms. Such a package of a generic universal problem, concerning individuals' insecurities that are generated by the political circumstances of the state, and its variable solutions, constitutes what may be termed a Hobbesian complex. The Hobbesian complex is manifest in different shapes and forms, depending on the configuration of the relationships between state, society, and the individual. The Hobbesian complex is concerned with the security needs of the individual. It is concerned specifically with ways in which the absence, ineffectiveness, or inadequacies of the state may endanger the individual's basic security needs. It may be assumed that whenever the individual's basic security needs are threatened the imperatives of the Hobbesian complex are invoked. Stated in such terms, the Hobbesian complex may be considered to be crossculturally pervasive. It should certainly be useful for considering problems which confront the individual's existence in corners of mankind where individuals have experienced dangers that render their livelihoods vulnerable. Although Hobbes' description of the putative state of nature was most immediately patterned on the civil strife of his era, the conditions he portrayed have been approximated much more closely in other regions of the world than in the England of his day. Where else was the individual's life nastier than in slave trade Africa? Nor is the misery of the individual that is the subject matter of Hobbes' theory solely a problem of the historic past. On the contrary, events in several regions of Africa in recent times recall Hobbes' concern with individuals' security needs with painful freshness. And yet Hobbes' theory has not been employed in the study of Africa with the same ardor with which, say, Karl Marx's theory of society has been invoked in studying African conditions. Two reasons may be given for such a failure. First, the unique individual has not had the prominence in African studies that other constructs of Western social science and history have been privileged to enjoy. Such terms as society, state, and culture have been used extensively in studying Africa. But as a construct of social science and history the unique individual is largely absent from African studies. This neglect dates back to colonial social anthropology. Indeed, Radcliflfe-Brown saw the differences between the simpler societies that social anthropology studied in Africa and the more complex Western societies that were the concerns of political philosophers in terms of the place of the individual in them:

Basic Security Needs & the Limits of Democratization

25

In the simplest societies, there is little more than the very important differentiation on the basis of sex and age . . . As we pass from the simpler to the more complex societies [in the West] we find increasing differentiation of individual from individual and usually some more or less definite division of the community into classes. (RadcliffeBrown, 1940: xxi) Put differently, colonial social anthropology subsumed individuals under kinship and assumed that there was an absence of unique individualism. Individuals had no unique needs outside the kinship realm. They were created by their kinship groups and their existence outside them was deemed to be meaningless. Having disputed the relevance of political philosophy's methodology of inquiry for examining Africa's political circumstances (see Fortes and Evans-Pritchard, 1940: 4-5), social anthropologists studying colonial Africa did not ask any existential questions pertaining to the individual's autonomy in African societies. Remarkably, the unique individual is still very much suppressed in modern African studies, especially by 'progressive' Marxist writers who continue to hide the unique individual within the 'masses', just as social anthropologists squeezed him into kinship categories. The second reason why Hobbes' theoretical perspectives have not been invoked for analyzing non-Western, particularly African, circumstances derives from the character of Hobbesian theory. It has hardly grown beyond its mythical presentation by Thomas Hobbes. Unlike Marxist theory, it has not been related to historical epochs which would endow it with a dynamism of variability in its employment in other cultures. Yet, even within the West, Hobbes' theory can be aligned with different historical experiences. For an illustrative example, individuals' security needs in England in the age of Oliver Cromwell's iron-fisted authoritarian rule were vastly different from the misfortunes of Irish peasants who were persecuted by social forces that they barely understood and who were unprotected by any state organization. This chapter explores problems confronting the individual's existence in Africa by employing elements of the Hobbesian complex - emphasizing the problem of the individual's insecurities and the solutions to it in the context of the individual's relationships to state and society. It does so, first, by isolating the unique individual as an autonomous player in the social formations that have dominated Africa's existence in the last several centuries and, second, by relating the existence of the individual to Africa's historical epochs of the slave trade, colonialism, and the post-colonial state. Because the formulation of African political sociology has usually employed categories developed from Western scholarship of state and society, it is important that we begin this exercise by supplying an intellectual background from the Western European encounter between the state and individuals.

Individuals5 Security Needs as the Motor of Western History In many ways, Western Europe's political history provides a useful method for studying African political sociology by way of a sharp contrast. The intellectual justification for studying the destiny of the individual in Africa from the point of view of Hobbes' theory must be sought in the wealth of individualism in Western European history and its contrasting thinness in Africa. By serving as its foil, Western history's wealth of individualism provides us with a mode of examining the individual's smallness in Africa's history. Such a contrast also helps to point up the individual's precarious relationship with the state in much of African history.

26

Peter Ekeh

Individuals' desires and needs are the driving engine of Hobbes' theory. Indeed, Hobbes' brand of social theory points up a major strand of Western European history that contrasts sharply with the African genre of history. It may be stated as follows: individuals' needs and desires have been the motor driving the historical forces of Western civilization. This is certainly the case with such major epochs of Western history as the rise of feudalism and the formation of the nation-state. I shall choose an example for illustrating my contention that individuals' needs and desires have driven Western history from a point and level of that history that will lend themselves to a worthwhile comparison with African historical experiences. The transition from kinship to feudalism in portions of Western European history was largely compelled by individuals' security needs. Before feudalism, individuals in Western Europe entrusted their protection from violence to their kinsfolk. The failure of kinship, in the face of expanded dangers and violence, to provide individuals with their security needs, led to the social formation of the feudal order. It bears hearing from three authorities on feudalism who have examined this issue from various perspectives. Relying on themes of ancient Germanic social structures, Otto Hintze indicated the historical response to the declining efficacy of kinship as a protector of individuals, as follows: Vassalage went back to the time when the family as a protective group was disintegrating . . . The peasant... in these troubled times needed someone else's protection. With the collapse of clan institutions - which had given their members the necessary protection, and support, had avenged injuries, and also provided security to the public authorities for their members' good behavior and the performance of public duties - a need arose for new communal groupings on the one hand, and for the protection of a powerful lord on the other. The village community did not guarantee that protection. Many free peasants thus gave themselves, by an act of commendation, to the protection and command of a powerful lord, who often took their land and became their landlord. (Hintze, 1929: 27) Analyzing ancient Irish legal data, Henry Sumner Maine came to about the same conclusions on the rise of feudalism from the ruins of kinship: Commendation went on all over Western Europe with singular universality of operation and uniformity of result. . . Yet there is considerable mystery about men's motives for resorting to so onerous a proceeding . . . Perhaps the most precise assertion . . . as to the reasons of so large a part of the world for voluntarily placing themselves in a condition of personal subordination is, that they must have been connected with the system of civil and criminal responsibility which prevailed in those times. Families . . . were responsible for the offences and even for the civil liabilities of their members; but corporate responsibility must have been replaced . . . by the responsibility of a single lord . . . the general disorder of the world had much to do with the growth of the new institutions; a little society compactly united under a feudal lord was greatly stronger for defence or attack than any body of kinsmen. (Maine, 1888: 155-6) Similar conclusions on the rise of feudalism were put forward by the French economic historian, Marc Bloch, in his study of feudalism: Yet to the individual, threatened by the numerous dangers bred by an atmosphere of violence, the kinship group did not seem to offer adequate protection . . . That is why men were obliged to seek or accept other ties . . . The tie of kinship was one of the essential elements of feudal society; its relative weakness explains why there was feudalism at all. (Bloch, 1940: 142)

Basic Security Needs & the Limits of Democratization

27

What unites all three writers - what Otto Hintze, Henry Sumner Maine, and Marc Bloch all took for granted as a commonplace of medieval European history - was an emphasis on the security needs of the individual as the motivator of historical change. Implicit in all three interpretations of medieval European history is the supposition that social institutions respond to individuals' changing desires and security needs in society. Thus, all of these authorities saw the origins of feudalism in the security needs of individuals which kinship could no longer provide and which required a new order of protection. By extension of their reasoning, it may be claimed that feudalism and the feudal state rose and died in response to individuals' changing needs and desires, yielding room to the modern nationstate that is better equipped to serve the security needs of modern man. Individuals' basic security needs pose threshold problems whose resolutions should precede attention to higher-order political problems. Modern democracy is an historical outcome issuing from processes that helped to resolve elementary problems posed by individuals' security needs. Indeed, democracy is an historical outcome in the development of Western societies that followed from the fullness of individualism. The state's full commitment to providing the security needs of its citizens is at the base of the foundation of democracy in Western societies.1 However, in those instances where the underlying relations between individuals and the state are less positive than in Western history, democracy is liable to be a lot more problematic. Indeed, democracy is troubled in Africa because the relationship between individuals and states has been haunted in the African history of the last four centuries, spanning the eras of the slave trade and European colonialism.

Kinship and the State in African History Modern African studies have been dominated by paradigms borrowed from Western European historical experiences and their uses in the social sciences (see Oyovbaire, 1983). One such dominant paradigm was the central assumption of modernization theory of the 1960s, namely, that kinship organizations would wither away from African societies, and from those of other underdeveloped countries, as modern nation-states took root in their political cultures - in effect, repeating the paths of Western European history. This formulation attained a critical level of theorizing in political anthropology. Two leading theoretical anthropologists made such an assumption the platform of their theories of the state. Morton Fried's evolutionary theory of the state postulates that the state emerges when it is able to capture kinship: 'A state is better viewed as the complex of institutions by means of which the power of the society is organized on a basis superior to kinship' (Fried, 1967: 229). Richard Fox (1971: 129) contends, with a Marxist finality, that 'Kinship withers away as society passes from primitive to the complex. Familial etiquette gives way to class relationship. Lineage and clan are by-passed in favor of country, province, and state.' Fox 1 Erich Fromm (1940: 270-1) puts this point as follows: 'The future of democracy depends on the realization of the individualism that has been the ideological aim of modern thought since the Renaissance . . . The victory of freedom is possible only if democracy develops into a society in which the individual, his growth and his happiness, is the aim and purpose of culture, in which life does not need any justification in success or anything else, and in which the individual is not subordinated or manipulated by any power outside of himself, be it the state or the economic machine . . . These aims could not be fully realized in any previous period of modern history; they had to remain largely ideological aims, because the material basis for the development of genuine individualism was lacking. Capitalism has created this premise.'

28

Peter Ekeh

(1971: 131) presses this point further with the following: 'The incongruity between the principles of kin-based societies and those of state societies is a valid theoretical distinction which highlights the direction of social change.' He insists that in the course of such social change, 'Kinship is slowly extinguished . . . As the evolution of state society proceeds, kinship becomes a shabby remnant of its former self, a relic of tribal times' (Fox, 1971: 131-2). There is no doubt, of course, that such views are patterned from the corpus of Western history. The growth of kinship and the development of the state were inversely related in European history. Kinship disappeared from much of Western Europe with the emergence of a new feudal order. In other areas, such as Scandinavia and the Celtic districts of the British Isles, where feudalism did not arise to displace it (see Bloch, 1940: 142; also compare Phillpotts, 1913: 254-7), kinship eventually yielded to the forces of the modern nation-state. The evidence from African history runs counter to such a Western European pattern in the historical relationship between state and kinship. In contrast, kinship has grown in stature with the development of the state in Africa. As the state seeks dominance in the public affairs of African societies, kinship has not shied away nor yielded place to the state's institutions. In several instances, kinship has flourished alongside the state; in others it has survived attacks from state institutions and has even grown in opposition to the state. More remarkably, kinship has insinuated itself into the public domain and seeks to control the conduct of the state. If there is one firm contrast between Western history and the African historical experience, it must be sought, on the one hand, in the atrophy of kinship in the face of developing state institutions in Western European history and, on the other, in the sharply different circumstances of Africa in which kinship has increasingly become dominant alongside the enlarged operations of the state. This appears to be the clear pattern since about the sixteenth century and well up to the present stages of African history in which kinship's largest manifestation, namely ethnicity, has grown triumphant. The Expansion of Kinship Under the Epochs of Slave Trade and Colonialism The growth of kinship in African history may be traced to the epochs of the slave trade and colonialism in Africa. The Arab Slave Trade (c.950-1850) and the European Slave Trade (1450-1850) from Africa implicated indigenous African states in various ways. Aside from a handful of states that declined to participate in it (thus, see Ryder, 1969: 198, Davidson, 1971: 65, and Ekeh, 1990: 676-7), most powerful states in Africa were converted into participation in the evil trade. In addition, a few states emerged into existence during and from the slave trade, with the primary purpose of servicing the needs of European traders for whom they acted as middlemen (see Dike, 1956; Forde, 1956; Ikime 1969; Jones, 1963). Kinship expanded in these areas in two ways. First, it flourished and seemed to have been strengthened inside states that participated in the slave trade, such as Dahomey (see Manning, 1982), Oyo (see Law, 1977), the Fulani Empire (see Lugard, 1912-19: 51), and Ashanti (see Wilks, 1975). Indigenous African states' involvement in the slave trade strengthened kinship in another direction. Most of the depredations of the slave trade were raids against ethnic groups that had no significant state organizations and that bordered on powerful states which participated in the slave trade. These stateless and segmentary lineage societies became the main focus of colonial social anthropology because their elaborate kinship systems provided a style of politics that was not typical of state

Basic Security Needs & the Limits of Democratization

29

societies. Kinship permeated their entire order (see Fortes, 1945, 1949; EvansPritchard, 1951; and Bohannan and Bohannan, 1953). The victims of slave raids in different parts of Africa all enjoyed great intensity of kinship in their social organization. They range from the Igbo in southeastern Nigeria (see Isichei, 1976) to the most remarkable instance of an area of modern central Nigeria that responded to Fulani slave raids by organizing myriad kingroups, resulting in a multiplicity of languages in a small restricted area (see Isichei, 1982; Morrison, 1982). Similarly, European conquests and colonialism strengthened kinship in Africa. On the ground in Africa, colonialism made use of the same institutions as the slave trade. Whereas the slave traders and their national sponsors befriended Africa's indigenous states as their chief mechanism for gaining control of the captives from raids compelled by the slave trade, European imperialists subjugated Africa's indigenous states during the last two decades of the nineteenth century, turning them to new uses in the governance of their colonies. The resulting colonial state in Africa was an imported bald imitation of the Western state. It was supplemented in its operations by subjugated indigenous African states. Colonialism also used the resources of kinship which had become exceedingly well entrenched in the centuries of the slave trade. The major European colonial powers funded social anthropological researches whose main purpose was to discover the principles of kinship organization in Africa. The chief goal of the European imperialists in these efforts was to discover Africa's folkways that would allow their new subjects to develop in their customary paths of kinship. Colonialism thus deliberately encouraged kinship whose institutions would enable Africans to ply their wonted existence, causing little disruptions in the colonies. Colonialism entailed administration on the cheap, encouraging any low-cost method of governance, which kinship and its ideologies readily supplied. Colonialism's fraternization with kinship helped its expansion into a major social formation during the colonial era. Under the regime of the slave trade, kinship was compact, usually bonding immediate kinsfolk in village settings. Under colonialism, its range and ideology expanded into ethnic groups, many of which used the new technologies of literacy to knit together fragments of kinship groupings and languages into composite ethnic groups (see Ekeh, 1983; Abernathy, 1969: 110). Ironically, the difficult years of European colonialism saw the crystallization of large ethnic groups that expanded the meaning of kinship from little bodies of kinsfolk, villagers, or small communities of agnates, to vast numbers that the ethnic groups now claimed and served. Thus, under the slave trade and colonialism, since about the sixteenth century in African history, kinship has expanded along with the activities of the state. In the post-colonial era, since 1960, the strength of kinship in Africa's public affairs has not diminished. In a number of instances, kinship forces have threatened the very survival of state institutions. This then raises the question: Why is kinship so persistent in Africa? Why is it that, contrary to its history in Western Europe, kinship in Africa has expanded along with the state? The answer to such a question lies in the investigation of two related features of African political sociology. In order to understand the kinship phenomenon in Africa, we need to explore the relationship between the state and the individual, on the one hand, and that between kinship and the individual, on the other. It is because the individual has had very difficult relationships with the state in African history that he has created a kinship system that provides him with the

30

Peter Ekeh

sort of protection which the state has failed to give him in the two eras of the slave trade and colonialism and now in the post-colonial era.

The State Versus the Individual in African History In 1591, the most complex and expansive state system in African history died a sudden death. It was in that fateful year that Songhai, bearing the burden and responsibilities of a fruitful history of a well co-ordinated state enterprise that spanned close to a millennium, came to a tragic end in the hands of Moroccan Arab invaders. As a state organization Songhai inherited the principles of government that the pagan state of Ghana innovated in the sixth century, CE, and that was carried forward by Mali, Ghana's successor state. Ghana, Mali, and Songhai exploited their opportunities near the edges of the Sahara to build a state system that was heavily reliant on international trade across the vast Sahara. 2 But all three of them also emphasized internal stability and justice in their public affairs. There was little doubt that these three states organized a system of public order that relied on the goodwill of their citizens. These were states that were clearly on the side of their citizens, not inclined to sacrifice the welfare of their peoples for the comfort of friendship with foreign potentates (see Fage, 1964; Levtzion, 1973.) From one view of its history, Songhai was sacked because it banked too heavily on the internal legitimacy of respect and goodwill from its internal citizens. After Songhai's fall at the end of the sixteenth century, the autonomy of African history began to unravel. After Songhai, the constant choice confronting African states was that between serving their people or joining in the international exploitation of ordinary Africans in a partnership with foreign interests. This struggle came to define the character of African states and their relationship with their own citizens and neighborhoods. It was a dilemma whose resolution was weighted against the interests of ordinary individuals. A review of African history will show that after Songhai the bulk of the major indigenous states formed alliances with Arab and European interests to exploit ordinary Africans. In one sense, it was a matter of survival for these states. The trend for African states was to form these alliances with foreign adventurers who supplied them with the machinery for warfare, thus limiting their reliance on their own citizens. African states that resisted this trend and insisted on their autonomy paid a heavy price for their assertion of independence. In the centuries and decades of the slave trade and colonialism, the character of the typical African state was changed from the autonomy and independence that Songhai craved in vain. The slave trade and European colonialism were international enterprises that needed the exploitation of ordinary Africans for their successes. But to achieve its goals, each of them also needed partnership with indigenous African states. In this development, the indigenous African state joined an international organization that persecuted and victimized ordinary Africans. Where there was an absence of co-operating indigenous states, the new European imperialism installed its own state structures. While King 2

Levtzion (1973: 163): 'Among the virtues of the Sudanese [as the Arabs labelled these states], Ibn Battuta [the itinerant Arab scholar] counted the care they took in protecting the trade routes. On his way from Walata to the capital of Mali, he travelled with only one guide, because it was not necessary to go in caravan. In these conditions trade flourished and foreigners settled and conducted business in the Sudan.'

Basic Security Needs & the Limits of Democratization

31

Leopold IPs 'Congo Free State' attained a level of violence that was difficult to match, German colonialism in Tanganyika and South West Africa was not far behind. British and French imperialism relied on the intimidation of their colonial 'subjects' by way of implements of violence in their colonies. 3 From the foregoing thoughts, it is fair to conclude that the African state featured prominently in the operations of the slave trade and colonialism in the past four centuries of African history. In these centuries, foreign interests and states teamed up with indigenous African states and state functionaries to carry out the mandates of the slave trade and colonialism in Africa. In these arrangements, the participating states and their governmental institutions failed to provide ordinary Africans with the type of protection for which the feudal order and the succeeding nation-state were celebrated in European history. On the contrary, the state has been the source of pain and persecution for the ordinary individual in African history. Remarkably, this trend has not been reversed in the post-colonial era of military dictatorships and personal rule (see Davidson, 1992; Ekeh 1994, 1997). This leads to the following question: If the individual has not been protected by the state in African history, and has on the contrary been victimized by it, how has he survived in these centuries? In order to gain a meaningful answer to this question, we must now examine the historic relationships between kinship and the individual in Africa. '

Individuals' Security Needs and Kinship in African History and Society Kinship was the Hobbesian response to the untold suffering and insecurities that the alliance of foreign interests and the African state system inflicted on ordinary individuals in the centuries of the slave trade and colonialism. Kinship provided whatever little protection there was to offer, to individuals victimized by the slave trade. During European colonial rule, kinship institutions rose to the assistance of individuals in their efforts to cope with the circumstances and difficulties of colonialism. Kinship has mushroomed into an elaborate network of relationships in which the individual has, nevertheless, a clear identity and from which he expects his basic security needs to be met. Kinship in post-colonial Africa has grown so big that its modern structures and functions deserve to be delineated. Moreover, the nature of the relationships between individuals and their kinship complexes has a sociological form that bears some examination. 3 Here is a picture of this aspect of imperialism as painted by Hannah Arendt (1951: 185): 'Race was [South Africa's] Boers' answer to the overwhelming monstrosity of Africa - a whole continent populated and overpopulated by savages . . . This answer resulted in the most terrible massacres in recent history, the Boers' extermination of Hottentot tribes, the wild murdering by Carl Peters in German Southeast Africa, the decimation of the peaceful Congo population - from 20 to 40 million reduced to 8 million people; and finally, perhaps worst of all , it resulted in the triumphant introduction of such means of pacification into ordinary, respectable foreign policies . . .' Arendt's use of 'savages' to characterize subjugated Africans compels one to invoke Mark Twain's (1897) sarcasm: 'In many countries we have chained the savage and starved him to death . . . in many countries we have burned the savage at the stake . . . we have hunted the savage and his children and their mother with dogs and guns . . . in many countries we have taken the savage's land from him, and made him our slave, and lashed him every day, and broken his pride and made death his only friend, and overworked him till he dropped in his tracks . . . There are many humorous things in the world; among them the white man's notion that he is less savage than other savages.'

32

Peter Ekeh

Levels of Kinship Organizations and their Functions in Post-colonial Africa Colonial social anthropologists sought to demarcate the outlines of kinship in Africa as they met it at the beginning of the twentieth century. That was the kinship left behind by the ravages of the slave trade. Such a compact notion of kinship, consisting of an ancestral founder and his descendants, real or imagined, has been challenged by its own growth during the colonial and postcolonial eras. Behaving as the prime protector of ordinary Africans, kinship responded to the exigencies of the slave trade, colonialism, and the dictatorship of the post-colonial state - all of which have historically victimized ordinary individuals. In doing so, it has become multi-layered in its meanings. Today, we must reconstruct its meanings from the behaviors and self-definitions of ordinary Africans whose creation kinship ultimately is. A close examination of the everyday language and behaviors of ordinary people would suggest that the old concept of kinship subsists in modified ways in modern Africa. In addition, the notion of kinship has been expanded by new structures of relationships carrying out the traditional functions of kinship in situations far removed from village settings and including far more persons than were usual at the beginning of the twentieth century. Kinship norms and behaviors have been carried from their ancestral homelands into cities by immigrants who have turned them to new uses. Along with this expansion of kinship is a change in its intuitive definition. More and more, kinship has come to be phrased as a final cause phenomenon, not as a practice or behavior that is enacted for its own sake. Even so, its norms and moral codes of behavior have followed its routes of expansion. Such modernization of kinship is not a subject that has attracted any body of scholarship so far. I suggest the following levels of kinship organization in post-colonial Africa. Primary Kinship. I use the term primary kinship to denote that portion of the kinship domain that social anthropologists discovered from their studies of African social structures in the first half of the twentieth century. Starting from an ancestral point of origin, it consisted of a corporate body of persons who shared a mandated pattern of behaviors in matters of inheritance, land ownership, rules of marriage, funeral rites, and their common security problems. As British social anthropology so usefully revealed, this unit of kinship was paramount in stateless societies (see Middleton and Tait, 1970), whereas its authority was less in societies ruled by chiefly hierarchies. During the era of the slave trade, such a primary kinship group was most important for the survival of the individual, particularly in stateless societies. It was less able to function completely during colonialism, with changes that came with literacy, expansion of communications and migration of kinship members into cosmopolitan areas far from the ancestral homeland. In the post-colonial era, primary kinship's functions have dwindled in several respects, particularly in the sphere of the security of its members. The dangers with which individuals are confronted in modern Africa are larger than any safety net that such core primary kinship can provide. But such lapses in its functions have not been bridged by any efforts from the modern African state. Instead, kinship has grown beyond its primary variant in order to cope with more demanding functions that it must now fulfill. Primary kinship continues to be important because it maintains vital ties to units of the expanded kinship system functioning outside its domain. Secondary Kinship. Primary kinship rarely functioned in isolation, even in the heyday of pristine kinship in the nineteenth century in stateless societies. Indeed, segmentary lineage theory was based on the capacity of kinship systems to meet

Basic Security Needs & the Limits of Democratization

33

new situations of danger by invoking the assistance of wider fragments of kinship including persons more distantly related to endangered individuals. In the postcolonial era, with the failure of the state to be helpful to individuals, those units not traditionally called upon in the first instance by individuals in the hierarchy of expectations have taken on greater significance. These secondary cells of kinship, beyond the primary ones, have become important for individuals in such areas of their lives as funeral and nuptial rites, which have become enormous in modern times. They also play key roles in protecting individuals from harassment, for instance from agents of the state. However, secondary kinship will have little role in matters related to land ownership and inheritance which continue to be the property of primary kinship. Extended Kinship. Secondary kinship is physically practised in the neighborhood of primary kinship. However, the dramatic expansion in kinship practices in Africa has arisen from their employment by migrants to cities. Neither primary nor secondary kinship can be of immediate relevance for migrants far away from home. But kinship practices are enacted in these polyglot cities by persons who claim common kinship because they come from an area that has a cultural distinction. They seek to perform the functions of kinship, sometimes fulfilling them on behalf of the primary kinship groups to whom such kinship activities are reported for final sanction. Thus, for instance, nuptial ceremonies carried out away from home by extended kinship cells are sanctioned by enabling ceremonies at the sites of primary kinship in the ancestral homelands. Extended kinship's greatest value inheres in the measures of security it provides to its members. These are in many instances the only meaningful protection from the dangers that confront immigrants to Africa's new cities. They include most prominently the notorious harassments from the police, including arrests on false accusations. These could be deadly (see, for example, Lawyers Committee for Human Rights, 1992) and frequently require intervention from cells of extended kinship. To give examples that will illuminate, a wise Nigerian immigrant will seek out these kinship cells upon arrival in a new urban setting. They are also most effective in the case of deaths of persons from their kinship areas. They not only arrange the repatriation of the remains of the deceased to the ancestral homelands, they also shield the grieving relatives of the deceased from hunger and other dangers. Expressions of extended kinship are more formal than those of primary or secondary kinship. Cell managers often organize their affairs in periodic meetings, but these are closed to outsiders and are open only to those who can claim to be kinsmen in some fashion. They certainly use the language of kinship. As an organizational pattern, extended kinship will link together several cells, using their resources according to the degrees of the problems confronting their members. Sociologists have included in their prolific studies of voluntary associations in Africa categories of what I here label as extended kinship. In what must be regarded as a substantial mimicry of Western European history, sociologists of urban life in Africa invest so-called voluntary associations with a touch of Tocquevillian voluntarism, touting them as free associations with characteristics of European medieval guilds. In fact, however, many of these have closed membership; they are clearly not open to those whom their members regard as nonkinsmen. 4 4

Typically, the urban migrant would join many associations, some of which would be of a professional nature with membership open to persons from different ethnic groups. But kinship-based organizations have restricted membership. In many essential respects they mimic primary kinship's

34

Peter Ekeh

Ethnic Kinship. A newer variant of kinship has grown from the turmoil of civil wars and other forms of national crises that have beset a number of African countries. This concerns the virtual conversion of a whole ethnic group into a mega-kinship system. During the Nigerian civil war, especially towards its end, when the secessionist Igbos were hard-pressed, they behaved as one kinship system. For a group that was notorious for confrontations among its fractions, this rearrangement of their forces into one putative kingroup was remarkable. The absence of a state system pushed the Igbos into a new quest for survival under the aegis of an imagined kinship system. The same notions of an ethnicwide kinship have been discernible in Rwanda and Somalia following the collapse of the state in these countries. But these examples also indicate the limits of kinship's efficacy. In all instances of extreme violence, kinship has proved to be a limited protector of its own. Clearly, kinship's role in offering protection to violated Africans cannot be fully equated to the roles of feudalism and the nation-state in providing protection for individuals in European history. Nor must it be imagined that kinship plays its roles because it is efficient. One only needs to watch the affairs of kin gatherings for a short while to be aware of complaints about their inefficiencies. Kinship continues to perform its roles because individuals still turn to it in the face of the state's greater inefficiencies, indifference, or outright victimization of the individual. Elements of Individuals3 Basic Security Needs The most prominent elementary security need in the centuries of the slave trade and colonialism was protection from violence. Whereas the Western European medieval feudal state sought to protect individuals from violence emanating from external sources (Vikings from the north, Hungarians from the east, and Arabs from the south), in Africa the state has been the predominant source of violence from which the individual seeks protection. Protection from state violence continues to dominate contemporary African affairs, from Rwanda to Nigeria. A second basic security need is the socialization and raising of children. Historically, most African states have only shown interest in those who have survived the travails of growing up in difficult circumstances permeated by violence. A third basic security need is the social welfare of the individual and his family. Fourthly and lastly, a basic security need about which most Africans are concerned is responsibility for funeral and burial rites of individuals and care for those they leave behind. All these basic security needs are the responsibilities of kingroups in the typical African state. From colonial times, they are deemed to be the sphere of responsibility of kingroups and their ethnic extensions. Protection from violence continues to dominate in the considerations of ordinary individuals. The Sociological Nature of Individuals' Relationships to Kinship A clear feature of modern Africa is that ordinary men and women still regard the state as a foreign construction managed by those who bear no allegiance to (4 contd) ways of doing things. Modernization theory's attempts to suggest large departures from rural folkways on the part of urban voluntary associations have not been borne out by subsequent developments in African history, especially in the wake of national crises. Such claims as the following from Kenneth Little should be regarded as an exaggeration as far as the organization of many urban kinship associations goes: 'owing to the rapidity of technological and other changes, fresh groupings and institutions . . . tend to develop on their own. Their raison d'etre is alien to traditional culture and their personnel is recruited according to entirely different principles from kinship and other factors which people brought up tribally customarily use as criteria' (Little, 1965: 163).

Basic Security Needs & the Limits of Democratization

35

them. But it is a totally different matter with their kinship institutions. They assert their ownership of these with vehemence and they freely argue with authority about their rules of operation. In the world of kinship, a person of little worth in the civil sphere of state matters becomes a being with passion, asserting his rights and duties with dignity. There appear to be rules governing the relationships that individuals maintain with their kinship organizations. We detail two of these as follows. Implicit Contracts Between Individuals and Their Kinsfolk. Actually, the ordinary African affirms, in his everyday conversations, the notion that he is tied to his kinship group by the bonds of a contract. Those involved in kinship practices make demands of one another in terms of what they owe to the group and to one another, not necessarily in terms of what they may volunteer to contribute. An example will illuminate this point. Imagine a young man traveling with his sick father away from home somewhere in Nigeria. Imagine that his father's condition worsens and that he dies in a strange place for which he is not prepared. The young man's first instinct is to search for 'relatives,' from the same kinship network of ethnic groups. On finding them, he expects their help and he will justly feel betrayed if he does not receive their assistance in solving his crisis. But he will not feel compelled to approach any agencies of the state, from which he will normally not receive any help. If he did approach them, it would be to plead that they assist him in locating someone from his kingroup. There are thus rights and duties attached to kinship, which are largely absent from the individual's relationship to the state in much of modern Africa. Attached to kinship behaviors, as reflected in the everyday language of kin usages, is the notion that there is an implicit contract whose terms compel participation in kinship affairs. There are rewards for participation and sanctions for breach of these implicit contracts of kinship. Who enforces them? The invisible hand of kinship does, very much like the mysterious invisible hand in Adam Smith's economy. There is gentle suasion that bonds the well-to-do to the illiterate poor in kinship behaviors in modern Africa. In this kinship domain, morality counts - unlike the state realm which is distrusted by the ordinary man and woman who see it as seeded with amoral arrogance. But if the miserable poor will abide by such implicit contracts of kinship, what compels the educated and the well-to-do to keep to their terms? They do so because kinship is powerful in Africa. It is difficult to be a good man outside one's kinship network if one is condemned by the forces of kinship to which one belongs. Rich men and women, important professors and professionals, have found their stock of goodwill exhausted because they are perceived as unfaithful to the terms of these implicit contracts of kinship. Kinship is strong enough to attend to certain of the needs of outstanding persons as well as those of the poor and the miserable. Indirect Reciprocity as a Method of Kinship Interactions Part of the strength of kinship's invisible hand lies in its operational principle of dispensing benefits. Members of a kinship network rarely expect to receive back any benefits directly from those kinsmen they favor. But they expect benefits to flow back to them from their kinship domain some time in the future. This payback may be in a form and kind that are different from the medium of the original favor. A generous wealthy kinsman may receive his reward in the amount of honor he acquires from the entire kinship dominion, maybe by way of immense funeral rites on his death! Such indirect reciprocity enriches the

Peter Ekeh

36

solidarity of kinship because it is based on the conception of a pool of moral goods that is replenished from time to time. 5

Conclusion: Democracy as a Relationship Between the State and Individuals Ultimately, democracy is the political expression of a relationship between the individual, as a citizen, and the state. A citizen is a political participant who has a major stake in the political process that governs the state. He (or she) imagines himself (or herself) as belonging to a community of fellow-citizens who own the state. Where there is an absence of such citizenry, democracy is imperilled. Where the state is at odds with individuals who people the state, democracy is non-existent. Wherever the state and its functionaries are perceived as persecutors of individuals in their domain as outsiders, the democratic process is inverted. It is the absence of such fundamentals that has besmirched the prospects of democracy in Africa. Ethnicity, which stands for what I have more broadly labeled as kinship in this chapter, has assumed the role of the state in several instances because individuals do not trust the state. Indeed, historically ethnicity has protected the individual from acts of violence and persecution by agents of the state. The bonds of mistrust between states and individuals in Africa are replaced with bonds of moral sentiments binding individuals who share a common ethnicity. Democracy must be understood as a historical formation of Western history that awaited the maturation of the individual in his interaction with the state. In order to nurse democracy in Africa, the state must be trained to treat individuals as citizens who own the state. Then the individuals will gradually learn to trust the state. The individual will gradually be weaned from his bondage to ethnicity when he can rely on the state to supply him with basic security needs and when the state ceases to be the source of threats to his existence. Until these developments occur, democracy will continue to be in jeopardy in Africa. 5 For a conception of indirect reciprocity, or generalized exchange, see Ekeh (1974: 52-6; 208-10) and Levi-Strauss (1949: 245-54).

References Abernathy, David B. 1969. The Political Dilemma of Popular Education. An African Case. Stanford, CA: Stanford University Press. Arendt, Hannah. 1951 The Origins of Totalitarianism. New York: Harcourt, Brace & World. New Edition, 1966. Bloch, Marc. [1940] 1961. Feudal Society. Chicago: University of Chicago Press. Bohannan, Laura and Paul Bohannan. 1953. The Tiv of Central Nigeria. London: International African Institute. Davidson, Basil. 1971. 'Slaves or Captives? Some Notes on Fantasy and Facts', in Nathan Huggins, Martin Kilson, and Daniel M. Fox, eds, Key Issues in the Afro-American Experience. New York: Harcourt Brace and Janovich. Davidson, Basil. 1992. The Black Man's Burden. Africa and the Curse of the Nation-State. Oxford: James Currey; New York: Times Books. Dike, K. Onwuka. 1956. Trade and Politics in the Niger Delta 1830-1885. An Introduction to the Economic and Political History of Nigeria. Oxford: Clarendon Press. Ekeh, Peter P. 1974. Social Exchange Theory. The Two Traditions. Cambridge, MA: Harvard University Press. Ekeh, Peter P. 1983. Colonialism and Social Structure. An Inaugural Lecture Delivered at the University of Ibadan. Ibadan: Ibadan University Press.

Basic Security Needs & the Limits of Democratization

37

Ekeh, Peter P. 1990. 'Social Anthropology and Two Contrasting Uses of Tribalism in Africa', Comparative Studies in Society and History, 30, 4 (October): 660-700. Ekeh, Peter P. 1994. 'Historical and Cross-Cultural Contexts of Civil Society in Africa', in USAID, Civil Society, Democracy and Development in Africa. Washington, DC: USAID. Ekeh, Peter P. 1997. 'The Concept of Second Liberation and the Prospects of Democracy in Africa: A Nigerian Context', in Paul A. Beckett and Crawford Young, eds, Dilemmas of Democracy in Nigeria. Rochester, NY: University of Rochester Press. Evans-Pritchard, E. E. 1951. Kinship and Marriage among the Nuer. Oxford: Clarendon Press. Fage, J. D. 1964. 'Some Thoughts on State Formation on the Western Sudan Before the Seventeenth Century', in Jeffrey Butler, ed., Boston University Papers in African History, vol. 1. Boston, MA: Boston University Press, pp. 19-34. Forde, Daryll, ed. 1956. Efik Traders of Old Calabar. Containing the Diary of Antera Duke, an Efik Slave-Trading Chief of the eighteenth century. London: International African Institute. Fortes, Meyer. 1945. The Dynamics of Clanship Among the Tallensi. London: International African Institute. Fortes, Meyer. 1949. The Web of Kinship Among the Tallensi. Oxford: Oxford University Press. Fortes, M. and E. E. Evans-Pritchard, eds, 1940. African Political Systems. London: Oxford University Press. Fox, Richard G. 1971. Kin, Clan, Raja and Rule: State Hinterland Relations in Preindustrial India. Berkeley, CA: University of California Press. Fried, Morton H. 1967. The Evolution of Political Society. New York: Random House. Fromm, Erich 1941. Escape From Freedom. New York: Holt, Rinehart & Winston. Hintze, Otto. 1929[1968] 'The Nature of Feudalism', in Frederick L. Cheyette, ed., Lordship and Community in Medieval Europe. Selected Readings. New York: Holt, Rinehart & Winston. Hobbes, Thomas. 1651 [1945]. Leviathan; or, The Matter. Forms, and Power of a Commonwealth, Ecclesiastical and Civil. Michael Oakeshott, ed. Oxford: Basil Blackwell. Homans, George C. 1964. 'Bringing Men Back In', American Sociological Review, 29: 809-18. Ikime, Obaro. 1969. Niger Delta Rivalry. Itsekiri-Urhobo Relations and the European Presence 1884-1936. New York: Humanities Press. Isichei, Elizabeth. 1976. A History of the Igbo People. New York: St. Martins Press. Isichei, Elizabeth, ed. 1982. Studies in the History of Plateau State, Nigeria. London: Macmillan Press. Jones, G. I. 1963. The Trading States of the Oil Rivers. A Study of Political Development in Eastern Nigeria. London: Oxford University Press. Law, Robin. 1977. The Oyo Empire, cl 600-1836: A West African Imperialism in the Era of the Atlantic Slave Trade. Oxford: Clarendon Press. Lawyers Committee for Human Rights. 1992. The Nigerian Police Force: A Culture of Impunity. Washington, DC: Lawyers Committee for Human Rights. Levi-Strauss, Claude. 1949 [1969]. The Elementary Structures of Kinship. Boston, MA: Beacon Press. Levtzion, Nehemia. 1973. Ancient Ghana and Mali. London: Methuen & Co. Little, Kenneth L. 1965. West African Urbanization: A Study of Voluntary Associations in Social Change. Cambridge: Cambridge University Press. Lugard, Sir F. D. 1912-19. Lugard and the Amalgamation of Nigeria. A Documentary Record. Compiled and Introduced by A. H. M. Kirk-Greene. London: Frank Cass, 1968. Maine, Sumner Henry. 1888. Lectures on the Early History of Institutions. New York: Henry Holt & Company. Manning, Patrick. 1982. Slavery, Colonialism and Economic Growth in Dahomey, 1640-1960, Cambridge and New York: Cambridge University Press. Middleton, John, and David Tait. 1970. Tribes Without Rulers: Studies in African Segmentary Systems. London: Routledge & Paul. Morrison, J. H. 1982. 'Plateau Societies' Resistance to Jihadhist Penetration', in Isichei, ed. Oyovbaire, S. Egite. 1983. 'The Tyranny of Borrowed Paradigms and the Responsibility of Political Science: The Nigerian Experience', in Yolamu Borongo, ed., Political Science in Africa. A Critical Review. London: Zed Press. Phillpotts, Bertha Surtees. 1913 [1974]. Kindred and Clan in the Middle Ages and After. A Study in the Sociology of the Teutonic Races. New York: Octagon Books. Radcliffe-Brown, A. R. 1940. 'Preface', in Fortes and Evans-Pritchard. Ryder, A. F. C. 1969. Benin and the Europeans, 1485-1897. New York: Humanities Press. Twain, Mark. 1897. Following the Equator. Hartford, CT: The American Publishing Co. Wilks, Ivor. 1975. Asante in the Nineteenth Century: the Structure and Evolution of a Political Order. New York: Cambridge University Press

3

BRUCE BERMAN

Ethnicity, Bureaucracy & Democracy: The Politics of Trust Ethnicity and Bureaucracy: Some African Encounters

M

Y first awareness of the relationship between ethnicity and bureaucracy in Africa came while I was a graduate student working at the Institute of Development Studies at the University of Nairobi more than thirty years ago. Preparing for a trip to Tanzania, I was told that I required a re-entry permit for my Kenyan visa. I duly presented myself at the passport and visa department equipped with a letter from the university stating that I would be returning to resume my post at the institute. It was signed by the registrar, Mr Karanja, a common Kikuyu name from central Kenya. I handed my passport and other papers to a clerk at the visa office whose name plate told me he was Mr Ochieng, an equally common Luo name from western Kenya. Glancing at the signature at the bottom of the registrar's letter, he looked up in some annoyance and said 'Karanja, Karanja, Karanja, everybody at that university's a Karanja'. 'Please', I said, 'I don't know Mr Karanja, I was just told I needed his signature on this letter, and I have no idea how many Karanjas work at the university.' Fortunately, he processed my re-entry permit without further comment. A few months later, while working in the national archives, one of the other students with whom I shared the reading room noted that all of the clerks and stack assistants we dealt with were Luhya from western Kenya, just like the chief archivist, Mr Fedha. Moreover, while I am not sure whether the staff were actually kinsmen of his, they were all certainly from the same Maragoli sub-group of the Luhya. When I reflected at the time on the significance of these links between ethnic groups and public institutions, they illustrated for me the sharp awareness of ethnic identity that marked Kenyan society. That particular sections of public service appeared to be dominated by particular ethnic groups struck me as redolent of the complex ethnic arithmetic of politics in the New York City in which I had grown up in the 1940s and '50s, and not terribly surprising in a society of such ethnic diversity. In more recent years, however, I find a deeper and less benign significance in the linkages of ethnicity and bureaucracy in Africa - one that is grounded in the particular connections between ethnic development and the state established during colonial rule and that has had profound consequences on the trajectories of economic and political change. Study of the relationship between ethnicity and democracy has focused primarily on political institutions of representation or self-government as means of accommodating the interests and protecting the rights of diverse cultural communities or minorities. The bureaucratic apparatus of the state has received far less academic or practical attention. This imbalance seems to me a mistake. The state bureaucracy constitutes at all levels the principal arenas of encounter 38

Ethnicity, Bureaucracy & Democracy

39

between ordinary citizens and the institutions of the nation-state. The nature of these encounters shapes the orientations of ordinary people to the realm of politics and the state, as well as the possibility of a civic politics of collective organization in the pursuit of common interests and ideological principles that are regarded as the modal forms of liberal democracy. Western capitalist nationstates experienced a long, painful and still contested process of development of ostensibly politically disinterested, professionally competent and relatively honest civil services - the neutral cadres of executive power of Max Weber's ideal type. This development has been of crucial importance in creating and sustaining the widespread social trust that characterizes both industrial capitalism and the nation-state. In Africa, however, disinterested competence and probity are the last things that ordinary people expect in their interactions with public officials. Instead, the state bureaucracy is a realm of nepotistic appropriation of office, ethnically biased distribution of patronage, extortion of bribes and kick-backs and direct theft of public revenues. Such experiences produce a vicious circle reinforcing reliance on the ethnic solidarity and patron-client networks that dominate bureaucratic processes in post-colonial African states. My central thesis in this chapter is that democratic reforms in Africa cannot succeed, the bonds of ethnic communities cannot significantly relax and a civic politics of broader ties of cooperation cannot develop without a corresponding transformation of the bureaucratic apparatus. To understand why this is necessary and how it might be approached, we should first look at the relationship between bureaucracy, social trust and modernity in the Western experience.

Social Trust and Modernity The young Max Weber, in his early career as an economist, studied the development of stock and commodity markets in Germany and tried to explain why they displayed significant internal variability in their stability and susceptibility to speculative disruption and unethical business practices, and were generally less stable than their British counterparts. While he thought the exchanges were essential to the development of capitalism by promoting the widest extension of trade in time and space, he noted that their ability to perform their function effectively was related to the existence of the strong cultural traditions of a cohesive merchant class that defined a recognized consensus on ethical behavior, such as he found existed in Hamburg but not in Berlin, and to the development of guild-like organizations of market traders, such as existed in Britain but not in Germany, which regulated access to the markets and enforced standards of conduct on their members. He would later explain the development of such norms of ethical conduct in social transactions as the result not only of the beliefs of puritan sects but of the even earlier drive towards autonomy and selfrule in medieval European cities. Through this work and his studies of bureaucracy Weber exposed the extent to which the instrumental rationality of both the state and the market rested on the existence of widespread social trust in the probity and competence of widely scattered, mutually anonymous actors. 1 At the most fundamental level, 'basic' trust is related to the establishment of an individual's ontological security in the reality and predictability of the physical and social world, i.e., 'the confidence that most human beings have in the 1

Reinhard Bendix, Max Weber: an intellectual portrait. New York: Anchor Books, 1962, pp. 14-79, passim.

40

Bruce Berman

continuity of their self-identity and in the constancy of the surrounding social and material environments of action'. 2 As such, it is the foundation of rational human agency and free choice. Beyond this level of basic trust, which is apparently established for most human beings during the earliest stages of life, we can identify three dimensions of trust based on increasing levels of uncertainty and risk in social action - habit, based on internalized, taken-for-granted and largely unvarying behavior; normative trust, based on emotional ties to individuals and community and an assumed shared identity, culture and values; and collaborative trust, focused on the terms of co-operation for rational agency. 3 Based on the degree to which these three dimensions are involved, there are significant differences in the character of social trust in pre-modern and modern societies. In pre-modern gemeinschaft societies, whose disappearance in Europe concerned so many nineteenth-century social theorists, social interactions were dominated by face-to-face relations in which actors were personally known to each other. Social trust was grounded in ties of kinship and community, with clear differentiation between insiders and outsiders, and shaped by the ontological framework of meaning of religious cosmologies and the meaningful routines of social tradition itself.4 In other words, habit and normative trust prevailed and were highly personalized, deriving from the moral character of mutually known individuals whose relationship defined their 'trustworthiness', rather than a generalized trust in abstract social norms and institutions. As Alan Silver notes, 'trust is simultaneously a personal and political quality. Judgments of others - of their capacities, intentions, characters - are central to the conduct of risky enterprises in politics, war and economy'. 5 The paradigms of trust relations were those of kinship and the personal bonds of friendship based on sincerity, loyalty and honor, their converse a mistrust based on fear of betrayal. Non-hostile interaction with anonymous others was virtually impossible, unless formalized bonds of 'friendship' were entered so that interacting parties became 'known' to each other. 6 Unequal relations of power were legitimated by the personal and highly formalized bonds of feudalism and the less formal but equally intimate 'lopsided friendship' of patrons and clients focused on reciprocity, mutual support and loyalty. During the early stages of the development of the modern state in the sixteenth and seventeenth centuries, the persistence of personalized trust relations among political actors produced a pervasive 'concern with faction, intrigue, insinuation, dissimulation, flattery, and betrayal . . . [characteristic of] the complexity of Renaissance politics, the multiple possibilities of betrayal, the instability of expectations'. 7 By the eighteenth century, with the expanded bounds of both national states and markets becoming more clearly defined, trust had become a central preoccupation of liberal theorists, but trust redefined as a calmer, more routinized and universal reliability of behavior in a newly uniform civil society. The act of promising, of self-imposed obligation, especially that contained in contracts, was seen as essential to promoting the 'mutual com2 Anthony Giddens, The Consequences of Modernity. Cambridge: Polity Press, 1990, p. 92 and see also pp. 93-9. 3 Barbara Misztal, Trust in Modern Societies. Cambridge: Polity Press, 1995, pp. 9-11. 4 Giddens, Consequences of Modernity, pp. 100-05. 5 Allan Silver, '"Trust" in social and political theory', in Gerald Suttles and Mayer Zald, The Challenge of Social Control. Norwood, NJ: Ablex, 1985, p. 61. 6 Giddens, Consequences of Modernity, pp. 118-19. 7 Silver, '"Trust"', p. 54.

Ethnicity, Bureaucracy & Democracy

41

merce of good offices', facilitating wider exchanges, and sustaining the very basis of social organization. 8 Such concern reflected and shaped the shift from personal trust bound to the face-to-face linkages of small-scale societies to the impersonal, abstract trust of modernity, based on trust in the honesty and competence of unknown others in complex large-scale institutions reaching beyond the local context. This collaborative trust permits the exercise of an increasingly refined instrumental rationality. Modernity, as Anthony Giddens has pointed out, is based on the disembedding of social relations in time and space from local contexts. This removal of social relations from the immediacy of context and their expansion over unprecedented reaches of time and space linked individuals in complex interactions with hundreds, thousands, and ultimately millions of others whom they would never know or even meet. The ability of such abstract institutions to function depends on the ability of people to have confidence in the probable outcomes of their actions, and this, in turn, rests on their trust in the predictable operation of the institutions through the probity and competence of the anonymous others also acting in it. Furthermore, such trust in abstract institutions has rested increasingly on the operation of expert systems of technical and professional expertise that organize the material and social environment. Such systems of expert knowledge provide security and predictability in daily life and are the condition for the time-space 'distanciation' of modernity, but require faith in their impersonal principles and in the abilities of their practitioners. Trust is thus always partially an act of faith, a pragmatic judgment that the systems will behave as they are supposed to, and must always be understood in relationship to some calculation of risk.9 In capitalist societies, where a rigorous ideological and institutional separation is enforced between the economic and the political, social trust in its abstract collaborative form is essential to the operation of both state and market, 'a pervasive social lubrication' that makes possible the impersonal, universalistic relations of modern abstract institutions. 10 The growth of the price system, especially in the markets for factors of production that define capitalism as a mode of production, and the decline of the personalistic and particularistic hierarchic bonds of medieval society, created universalistic roles of formally equal buyers and sellers that framed the problem of social exchange and the liberal preoccupation with contracts. As Weber understood, trust provided the socio-cultural basis for market exchange, where expanding markets brought increasing business with strangers and a need to rely on their honesty and competence in exchange transactions of mutual and equivalent utility. A hegemonic bourgeois culture provided a framework of business ethics applicable to all transactions that both constrained self-interest in the interest of the integrity of the market itself, and disciplined individual behavior through the threat of the loss of the reputation for trustworthiness. 11 Trust is thus a way to deal with the freedom of others, to predict the behavior of other free agents in a way that facilitates strategic choice and rational agency.12 In the market place this made possible the counterfactual future orientation of capitalist modernity. As markets and capitalist enterprises became larger 8

See the discussion of John Locke, David Hume and William Paley in ibid., pp. 53-6. This paragraph is based on Giddens, Consequences of Modernity, pp. 16-35. 10 Silver, '"Trust"', p. 57. 11 Bendix, Max Weber, pp. 24-30, 36-7, 50-5. 12 John Dunn, 'Trust and Political Agency', in Diego Gambetta, ed., Trust: Making and Breaking Cooperative Relations. Oxford: Basil Blackwell, 1988, pp. 73-5.

9

42

Bruce Berman

and more complex, trust in the probity of individuals was increasingly superseded by abstract, depersonalized system trust in the market itself and reliance on expert systems and their professional practitioners. This collaborative trust makes possible the rational calculation of risks and benefits and the reduction of uncertainty that is the basis for the large-scale routinization and standardization of production and exchange of industrial capitalism.13 Trust is no less essential to the functioning of the liberal democratic state in capitalist societies, and abstract system trust in state institutions is crucial to sustaining trust within the apparently separate spheres of the market and civil society. Silver notes that social trust always includes 'the idea of risk and danger in the setting of contending interests'. 14 In capitalist society this focuses on the possible failure or breach of relations of contract and exchange, particularly in the crucial labor contract that is always fraught with potential conflict. In such instances the parties cannot resort to self-help to impose a solution to the conflict, since any resort to force destroys the very basis of a supposedly free exchange of equivalent values. The enforcement of contracts and repair of breaches have both logically and historically required the existence of the state as an authoritative, disinterested arbiter capable of enforcing universal rules governing social transactions. This meant in practice the development within the context of the nation-state of a single, universal body of law applying equally to all citizens and administered professionally and impartially by judicial and bureaucratic organizations. Two further features of the nation-state also facilitated the orderliness of exchange relations and the growth of capitalism. First, the effective monopoly of the legitimate use of force by the state created a protected arena of non-violent interaction and safe movement in which the market could develop; and second, the state's constantly expanding surveillance of society and reflexive monitoring of its own actions created a base of social knowledge and expertise essential for the planning and implementation of largescale and long-term economic activities.15 The national state, like the national market, is an abstract and depersonalized institution involving interaction among anonymous citizens and requiring trust not only in the probity and competence of politicians and public officials who can rarely be personally known, but also in the fairness and efficacy of the institutions of the state itself. In the Western nation-state the functioning of liberal democracy is based on the development of a widespread trust, cutting across cleavages of class, region and ethnicity, that political institutions act as disinterested arbiters of clashing interests. The Russian legal theorist Evgeny Pashukanis wrote in the 1920s that the state had 'to detach itself from the ruling class and take on the form of an impersonal apparatus of public power', while the American political scientist, Kalman Silvert, noted in the 1960s that the nation-state rested on its acceptance 'as the ultimate and impersonal arbiter of secular dispute'. 16 This is the sphere of the 'relative autonomy of the state' so important in neo-Marxist state theory. 17 This trust involves an essential public 13

Giddens, Consequences of Modernity, pp. 79-91; Silver, '"Trust"', pp. 60-3. Silver, '"Trust"', p. 62. These are central themes of Anthony Giddens, The Nation-State and Violence, Cambridge: Polity Press, 1985. 16 Evgeny Pashukanis, Law and Marxism: A General Theory. London: Ink Links, 1978 (first published in 1924), p. 139; Kalman Silvert, Expectant Peoples: Nationalism and Development. New York: Random House, 1963. 17 See the discussion in Bruce Berman, Control and Crisis in Colonial Kenya. London: James Currey and Athens, OH: Ohio University Press, 1990, pp. 23-34. 14 15

Ethnicity, Bureaucracy & Democracy

43

belief that the political process can be used to pursue differing visions of social interests and that institutional rules provide for transparency and accountability in the formulation and implementation of public policy. The effective civic equality of citizens, the rights to organize freely and compete for support in institutionally defined arenas of political conflict, and the fair and open administration of access to public goods, provide the basis for the development and maintenance of such trust. The acceptance or rejection of the capitalist liberal democratic state as a disinterested arbiter and implementer of the outcome of fundamental social conflicts marked the cleavage between social democrats and revolutionary socialists in the European experience. Trust, therefore, is the fundamental basis of the legitimacy of political authorities and institutions, and of political agency. Without it risk and unpredictability render collective organization and action towards political goals literally unthinkable. 18 It makes possible and encourages the pursuit of political objectives defined by ideological values and broad collective interests, a politics of principle and policy, rather than a politics of narrow materialism and selfinterest. It also facilitates acceptance of opponents in the political process and delegation of power to authorities to literally exercise a 'public trust'. Thus, we can act politically in the arena of mass politics in the nation-state in pursuit of supra-personal principles because we trust that elections are fair; that opponents will abide by the rules and not sabotage policies with which they do not agree; that unknown bureaucrats are competent and unbiased in the administration of policy; and that politicians and leaders of interest associations, whom we support but seldom know personally, will not give all the jobs and contracts to their friends and relatives, sell out to the opposition and run off with the money. In the Western experience the professionalization of the civil service of the capitalist state was a crucial factor in the establishment of such abstract collaborative trust and the 'civic' politics it makes possible. It was, however, a protracted and difficult process that was shaped by the patterns of European culture, class structure and class conflict. On the one hand, newly enfranchised bourgeoisies pressed for higher levels of probity and competence from state officials. In Britain, after the Reform Act of 1832, the capitalist middle class demanded accountability for the use of its taxes and elimination of aristocratic hangers-on and 'placemen' in the state apparatus in favour of open, competitive recruitment of 'qualified' civil servants. Their conservative and aristocratic opponents, however, articulated a nationalist anti-modernism of 'invented tradition' that dominated the state apparatus. 19 Civil servants became the Hegelian 'universal class' sustaining social order above the fractious struggles and crude self-interest of bourgeois civil society. By the end of the nineteenth century, paternalistic 'red Tories' and social democrats melded with new forms of professional expertise in shaping a career public service of disinterested competence focused on containing and ameliorating the disorder and exploitation of capitalist development. The consolidation of the modern state apparatus coincided with the profound structural transformation, escalating class struggles, and political reforms of the fin de siecle second industrial revolution. The public 18 Murray Hausknecht, 'On Politics and Trust', Dissent, Fall 1992; Dunn, 'Trust and political agency', pp. 83-7. 19 The creation of such 'invented traditions' is explored in the essays in Eric Hobsbawm and Terence Ranger, eds, The Invention of Tradition. Cambridge: Cambridge University Press, 1983.

44

Bruce Berman

service became the literal agent of the state's benevolent trusteeship of society.20 In the twentieth century major periods of crisis in the 1930s and 1960s were closely related to the further extension of social reforms under the control of a 'technocracy' of state elites guided by the 'policy sciences' in the practice of a disinterested social engineering. As one British colonial official put it to me, 'Without bureaucracy there can be no social justice'. Finally, we must note that trust in the political system is, as John Dunn and Barbara Misztal separately note, a complex human project, a goal to aim at, 'an active political accomplishment' requiring constant renewal, not a destination ever finally reached. 21 It is an achievement of the civic nationalism of the eighteenth-century revolutionary tradition and its values of universalism, rationalism, detachment and reflexivity. It expresses a 'reflective politics of self-control and self-limitation'22 focused on all participants playing by the rules of civic institutions that are valued in themselves above the outcomes of specific political struggles. While civic trust is never established permanently in any state, it is also neither created nor destroyed by any single event or social transaction. Rather, an overall 'trust account' is augmented or depleted by the continuous renewal of political decision and administrative implementation. Civic nationalism and its accompanying political trust have existed, however, in continuous tension with the particularism of both ethnic nationalism and the ties of personal trust that have never disappeared from the political process. The cool rationalism of abstract collaborative trust cannot provide the mutuality or intimacy of the cohesive forms of personal or communal trust relations 23 that express the primacy of shared identity and emotional attachment. Abstract trust is undermined by particularistic ties of individuals and factions in politics and the state apparatus defined by ties of ethnicity, religion or region. Civic trust in the bureaucracy, in particular, is compromised by evidence of administrative bias and opaque decision-making, as well as of personal or communal bias in nepotism, special deals and preferential access to public goods. Moreover, abstract trust and civic politics have developed more effectively at national rather than local-level politics, where narrowly personal and communal politics often remain dominant. Finally, in the Western experience the tension between abstract and particularistic trust has been most intense during periods of capitalist transition 24 when earlier standards of political and economic behavior have been challenged and movements of political reform have attempted to redefine and eliminate 'corruption' in the political process and the state apparatus. During the most recent phase of capitalist 'restructuring' in the last quarter of the twentieth century, there has been a noticeable increase in corruption, the visible domination of the state by corporate interests and a corresponding decline of civic trust in Western liberal democratic states, although the reform movements of disinterested expertise have yet to appear. 25 20 The concept of 'trusteeship' and its relationship to development in capitalist societies is explored in the important work by Michael Cowen and Robert Shenton, Doctrines of Development. London: Routledge, 1996. 21 Dunn, 'Trust and political agency', pp. 88-9; Misztal, Trust in Modern Societies, p. 7. 22 Ibid. 23 Giddens, Consequences of Modernity, pp. 114-15; see also Silver, '"Trust"'., pp. 64-5. 24 Misztal, Trust in Modern Societies, pp. 4-5; Marek Korczynski, 'Trust, Power and the Market: A Critical Overview of the Economic Sociology of Inter-Firm Relations', Centre for Corporate Change, University of New South Wales, Sydney, 1996, pp. 18-19. 25 See the essays in Donatella Delia Porta and Yves Meny, eds, Democracy and Corruption in Europe. London: Pinter, 1997.

Ethnicity, Bureaucracy & Democracy

45

Ethnicity, Patronage and Social Trust: State-Society Linkages in Africa If the development of broad and abstract collaborative trust in the institutions of the capitalist market and the nation-state has been a crucial attribute of Western modernity, then its virtual absence is a central characteristic of contemporary Africa. Instead, the colonial legacy of bureaucratic authoritarianism, pervasive patron-client relations and a complex ethnic dialectic of assimilation, fragmentation and competition has produced contained and divided communities of trust based on particularistic and personalistic linkages that undermine the functioning of universalistic institutions, as well as a ruthlessly competitive and materialistic politics devoid of ideology or principle. 26 While these characteristics of state and society in post-colonial Africa have generated diverse local variations around the general patterns, they have also made the development of capitalism and liberal democracy, in particular, difficult, if not impossible. The colonial state in Africa was an authoritarian bureaucratic apparatus, not a school for democracy. However haphazard and ramshackle the reality of colonial state power, the European proconsuls struggled to maintain effective control behind a facade of omnipotence and omniscience. For the subject African population, masses or elites, dealing with the colonial state was always a mixture of opportunity and danger - the opportunity to gain access to the diverse resources at the disposal of the state and its agents, and the danger of running afoul of its often apparently arbitrary and capricious actions and its coercive taxes and punishments. Dealing with so dangerous and unpredictable an entity required the protection and support of a powerful intermediary or protector. The other open, but never officially acknowledged, dimension of the colonial state was the pervasive patron-client networks that linked it through its local African agents to the competitive factions of lineage and clan. Colonialism in Africa rested largely on the institutionalization of 'Big Man-Small Boy' politics in rural society, built on the hierarchies of the 'decentralized despotism' of local chiefs and headmen. 27 These cadres of African collaborators, whether directly appointed by the colonial regime or holding indigenous offices incorporated into the colonial state, were clients of local European administrators who rewarded their loyalty through access to the resources controlled by the state. Such access to the state and its patronage became the key to the accumulation of wealth and was controlled by local African officials in the interests of their kinsmen and extensive clientages. Linked to them to varying degrees in different societies were members of a growing literate intelligentsia occupying other positions in the state apparatus and small groups of wealthy farmers, cattle owners and traders who also played patron-client politics, using their surplus to invest in social networks to build their own clientage and position themselves for access to the wider patronage networks of the state. 28 While the patronage networks of colonialism were often built on the base of the complex relations of clientage and dependence 26 I have analyzed these factors in more detail in 'Ethnicity, Patronage and the African State: the Politics of Uncivil Nationalism', African Affairs 97, 388, July 1998, pp. 305-41. 27 The vivid phrase and conceptualization of local-level colonial politics is Mahmood Mamdani's in Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton, NJ: Princeton University Press; Cape Town: David Philip; Kampala: Fountain; Oxford: James Currey, 1996, ch. 2. 28 Sarah Berry, No Condition is Permanent: the Social Dynamics of Agrarian Change in Sub-Saharan Africa. Madison, WI: University of Wisconsin Press, 1993 passim; and Angelique Haugerud, The Culture of Politics in Modern Kenya. Cambridge: Cambridge University Press, 1995, esp. pp. 180-91.

46

Bruce Berman

that existed in many pre-colonial societies, in the context of a colonial political economy based on money and markets the relationships became increasingly focused on access to short-run material benefits rather than establishment of long-term ties of mutual support. 29 The 'politics of the belly' thus originated in the fundamental institutional structures and relationships of the colonial state. At the same time, the social networks of patrons and clients shaped and were, in turn, shaped by the colonial development of ethnic identities and communities. The internal and external factors of ethnic construction, cultural invention and political negotiation encompassed a process of both an increasingly clear definition and enclosing of ethnic cultures and identities, and a significant expansion of the scale of ethnic communities. The 'dual dynamic of assimilation and differentiation'30 shaped diverse stories of ethnic development in Africa in a continuous process of reformulation, including the creation of new groups and identities and the disappearance of old ones. Colonial administrators, missionaries and anthropologists combined in an invention of tradition through efforts to define clearly bounded tribal societies and identities that would help preserve social stability and facilitate political control. Africans responded in the development of ethnicity through a process of cultural imagining based on real cultural experiences and resources, created and refashioned out of both old and new elements. 31 In particular, the emotive power of the cultural symbols and identities of kinship and 'home' were transferred to larger social collectivities in the context of the development of colonial states and markets. 32 At the same time, ethnicity provided individuals and groups with their most important political resources in the competition for the scarce goods of modernity, as well as for access to local resources of land and labor. Internally, this took the form of moral ethnicity, the discursive and political arena within which ethnic identities emerged out of the renegotiation of the bounds of political community and authority, the social rights and obligations of moral economy and the rights of access to land and property. Externally, it took the dialectically related form of political tribalism, which was a search not for a moral community of rights and obligations, but rather communal solidarity and political organization across the boundaries of communities defined by moral ethnicity, first against the alien power of the colonial state and then, increasingly, against the competing interests of rival emerging ethnicities for access to the state and control of its patronage resources. 33 The particular pattern of state-society linkages that developed in colonial Africa - patron-client networks centered on the local African agents of colonial power and largely contained within emergent ethnic communities - defined a 29

Rene Lemarchand, 'The State, the Parallel Economy, and the Changing Structure of Patronage Systems', in Donald Rothchild and Naomi Chazan, eds, The Precarious Balance: State and Society in Africa. Boulder, CO and London: Lynne Rienner, 1988, pp. 150-1. 30 Shaheen Mozaffar, 'The Institutional Logic of Ethnic Politics: a Prolegomenon', in Harvey Glickman, ed., Ethnic Conflict and Democratization in Africa. Atlanta, GA: African Studies Association Press, 1995, pp. 50-1. 31 Berman, 'Ethnicity, Patronage'; Terence Ranger, 'The Invention of Tradition in Colonial Africa', in Hobsbawm and Ranger, The Invention of Tradition', Ranger, 'The Invention of Tradition Revisited', in Preben Kaarsholm and Jan Hultin, eds, Inventions and Boundaries: Historical and Anthropological Approaches to the Study of Ethnicity and Nationalism. Roskilde, Denmark: Institute for Development Studies, Roskilde University, 1994. 32 Carola Lentz, 'Tribalism and ethnicity in Africa: a review of four decades of anglophone research', Cahiers des sciences humaines, 31, 2, 1995, pp. 322-4. 33 John Lonsdale, 'Moral Ethnicity and Political Tribalism', in Kaarsholm and Hultin, Inventions and Boundaries.

Ethnicity, Bureaucracy & Democracy

47

fragmented plurality of communities of trust, within which individual probity, rights and responsibilities were the concern of an active political process, while between them an amoral competition for access to the material resources of the state became increasingly intense. More than twenty years ago Peter Ekeh captured the dimensions of moral ethnicity and political tribalism with his concept of the two publics in Africa. The primordial public is 'closely identified with primordial groupings, sentiments and activities which nevertheless impinge on the public interest. The primordial public is moral and operates on the same moral imperatives as the private realm.' The other, the civic public, developed in relation to the colonial state and 'has no moral linkages with the private realm . . . The civic public in Africa is amoral and lacks the generalized moral imperatives that operate in the private realm and in the primordial public. ' 3 4 The very different contemporary usage of the terms 'primordial' and 'civic' makes these terms difficult to use in this sense today. However, the essential meaning behind moral ethnicity and political tribalism or of the two publics is clear: social trust is largely contained within ethnic communities and imbedded in the personalistic ties of patronclient networks. Such networks were the fundamental state-society relationship of colonial Africa, a relationship that was opportunistically focused on access to material benefits. There was little basis for the development of impersonal, systemic civic trust in the state as an impersonal arbiter of conflict or an honest and disinterested distributor of public resources. In post-colonial Africa the colonial combination of bureaucratic authoritarianism and clientelism has continued essentially unchanged, especially in the structure of rural control and collaboration between the bureaucratic apparatus and local 'big men'. The nationalist regimes that came to power at independence faced the same problems as the colonial state in exercising effective power in the countryside, and most chose not only to rely upon the existing apparatus, including the prefectural field administration immediately over the local tribal authorities, but also to extend and intensify colonial modes of domination. At the same time, on the day after independence, the often tenuous ethnic and class coalitions of the nationalist movements began to unravel into competing factions struggling for control over the very material rewards of state power. Given the continuity of colonial structures of surplus appropriation and market controls, as well as the influx of external development aid, the state remained the essential focus for the accumulation of wealth in an environment in which 'any official decision affords an opportunity for gain, from a fiscal control to a technical verification, from the signature of a nomination form or a concessionary market to an industrial agreement or an import license. Civil service departments and public enterprises constitute virtually bottomless financial reservoirs for those who manage them, and for the political authorities who head them.' 35 The result has been the extension of ethnically based patron-client networks to the very center of the state apparatus, with their ramifying linkages reaching from cabinet to village to produce what J. F. Bayart graphically describes as the 'rhizome state'. 36 The post-colonial state in Africa has been for a generation a conglomeration of agencies and offices to be captured and manipulated, beneath the facade of the official 'development' ideology, for individual and communal benefit. The arbitrary and authoritarian use of state power to accumulate wealth reflects the 34 Peter Ekeh, 'Colonialism and the Two Publics in Africa: a Theoretical Statement', Comparative Studies in Society and History, 17, 1, 1975. p. 92. 35 Jean-Francois Bayart, The State in Africa: the Politics of the Belly. London: Longman, 1993, p. 78. 36 Ibid., pp. 218-27.

48

Bruce Berman

limited development of both the impersonal exchange relations of the capitalist market and of the state as disinterested arbiter of political conflict. Moreover, not only does ethnic and patronage politics inhibit the development of a coherent national dominant class with a project of social transformation, but also the fragmentation and privatization of state power undermines its ability to act as the agent of such a project of national development. The rapid demise after independence of the thin and hastily constructed facade of liberal democratic institutions reinforced the mutual dependence of patrons and clients on the increasing authoritarianism of the political process in one-party states and military regimes. The destruction of an open political process and effective elimination of civic freedoms of speech, information and organization left the personalistic ties of patron-client networks as the only available mode of access to the state for ordinary people and the appropriation of the patronage resources of office as the only source of support and power for politicians and state officials. The politics of political tribalism and moral ethnicity became linked to the ability of the 'big men' of ethnic communities holding positions in the state to obtain for their districts and regions both a significant share of the large-scale collective benefits of development through projects for roads, schools, dispensaries, etc. as well as the more individual rewards apportioned through discrete personal contacts with their clients. In circumstances of economic and political instability, where even the wealthy face the daily insecurities and uncertainties of life, ethnicity and tradition provide networks of mutual support and trust, defining a political community within which demands are made on both the state and its own leadership. 37 For the ethnic community to be sustained, however, there must be real material, political and symbolic goods to deliver. Elites and masses within them draw on the same issues of moral ethnicity to legitimize or challenge the distribution of wealth and power. Ethnicity has been the fundamental context and idiom of class formation and struggle in Africa, with contested processes of class formation being contained largely within the internal politics of developing ethnic communities. Externally, elites are expected to be the spokesmen of their people, defending their interests in national institutions and getting for them their share of 'the national cake'. Internally, as a dominant class they are expected to meet the moral obligations of reciprocity to both kin and other members of the community. 38 Patronage networks not only distribute material benefits, but also are expected to sustain an intimacy of relations between rulers and ruled, a reproduction of the personal ties of the traditional pre-colonial social order rather than the impersonal relations of modernity. Distribution of patronage by the wealthy and powerful both displays their status and subordinates to them those who accept their largesse. Such relations are cloaked in the mantle of tradition and often politically expressed in elaborate ceremonial. 39 The dialectic of ethnicity and class has had several important consequences reinforcing the process of ethnic affiliation and differentiation. First, internally, 37

Haugerud, Culture of Politics, pp. 33-8, 192-8; Peter Ekeh, 'Social Anthropology and Two Contrasting Uses of Tribalism in Africa', Comparative Studies in Society and History, 32, 4, 1990, pp. 691-3. 38 Carola Lentz, 'Home, Death and Leadership: discourses of an educated elite from north-western Ghana', Social Anthropology, 2, 2, 1994, pp. 153, 163-6; Okwudiba Nnoli, Ethnicity and Democracy in Africa; Intervening Variable. Lagos: Malthouse Press, 1994, pp. 18-19. 39 Achille Mbembe, Tower and Obscenity in the Post-Colonial Period: the Case of Cameroon', in James Manor, ed., Rethinking Third World Politics. London: Longman, 1991, pp. 176-8; Lentz, 'Home, Death and Leadership', pp. 158-60.

Ethnicity, Bureaucracy & Democracy

49

the 'lopsided friendships' of patron-client networks reinforce local and ethnic identities and mute the development of class consciousness and conflict by reproducing ties of reciprocity across class lines. 40 Second, externally, ethnicity inhibits the development of broader, national affiliations of class, especially with regard to the failure of a pan-ethnic dominant class to emerge in most African states. Instead of a national dominant class acting towards a project of national development, we find ethnic class factions claiming internal leadership in the contestation of moral ethnicity and serving externally as the principal agents of political tribalism. The constant analogies and metaphors in political discourse throughout subSaharan Africa referring to politics as 'eating' or 'devouring', the repeated references to getting one's share of the state's resources or 'slice of the national cake', express vividly the personal, materialistic and opportunistic character of African politics, and the relative unimportance, if not irrelevance, of ideology, principle or policy.41 Politics as eating also captures the duality of relations with the state as both opportunity and danger; those who aspire to eat can also be eaten in the amoral food-chain of politics. To survive in such a dangerous world requires both support and protection, which are precisely what patrons and clients are supposed to provide for each other, cemented by ties of kinship and ethnicity, and the reason why wealth is invested in developing and maintaining social networks. But both Big Men and Small Boys are also potential rivals who can turn and devour each other, a tension inherent in their relations and accentuated by preoccupation with the flow of patronage. The materialism of relations makes loyalties shallow. The contradiction of the politics of the belly is that both rich and poor, elites and masses, share the same materialistic and opportunistic view of politics and the state. As Dickson Eyoh notes: because of the sharing of ideological repertoires, popular resistance to political oppression in the (African) post-colony evinces ambiguity and ambivalence as their main trait; a ready condemnation of the excesses of the consumption of the spoils of office by the powerful, but not a rejection of the values which sanction such behavior.42 40

See, for example, Haugerud, Culture of Politics, pp. 45-50. The way the informal system works behind the facade of state institutions is well understood by African publics. This is clear in the succinct summary offered in a column in the Daily Nation in Nairobi by Jaindi Kisero on 26 December, 2000: 41

. . . in Kenya as in many developing countries corruption is not incidental to politics, it is often the very reason for engaging in politics. Our leaders seek political power so that they can get an opportunity to acquire riches and prestige and be in a position to hand out benefits in the form of jobs, contracts and gifts of money to tribesmen and political allies. In many of our local languages, political power and positions in government are analogized as food. Each community demands its 'share of the "national cake'". A popular saying has it that 'a goat eats where it is tethered'. Another variation is: 'You eat where you work.' Our political parties are bereft of platforms and ideology and are no more than institutions whose only reason for existence is the intention to divide the spoils of state among the ethnic communities that support them. These attitudes are captured well by the Yoruba saying that 'an elder who is brought a plate does not lick it clean, or he will not find young ones to run errands for him'. It is these entrenched attitudes that explain why members of the public in this country hardly ever get enraged over corruption . . . Instead of getting annoyed about cases of corruption, we Kenyans get more enraged when fruits of corruption are not shared out according to settled expectations. 42

Dickson Eyoh, 'From Economic Crisis to Political Liberalization: Pitfalls of the New Political Sociology in Africa,' African Studies Review, 39, 3, 1996, p. 65.

50

Bruce Berman

Neither capitalism nor the nation-state actually exist as dominant, hegemonic or fully-formed institutional and cultural systems in sub-Saharan Africa, with the possible exception of South Africa. The pattern of state-society relations, the patronage systems and ethnic communities and identities that have developed over the past century display no inherent tendency to develop fully into those politico-economic forms. Indeed, neither state power nor ruling coalitions have been organized around transformative projects directed towards either capitalism or the nation-state. 43 Such projects actually threaten the established bases for the accumulation of wealth and power and for the patronage politics that sustain elites and ethnic factions. Instead, African states are stalled in a heaving, chaotic pluralism of contradictory institutional and cultural elements in which people attempt to find in ethnic communities and identities, which are themselves internally divided and contested, a degree of support and security. Ethnicity provides some semblance of cultural and moral coherence in an environment of intense, even desperate, competition for resources. What remains missing is the development of a trans-ethnic public arena grounded in universalistic norms and civic trust governing both political and economic transactions. The reliance on the loyalties and personal ties of kinship and ethnicity reflects the pervasive fear in Africa that, if you lack powerful patrons and supporters, some other individual or group will get the goods. African states show all the characteristics of pervasively low-trust societies; even the affective trust linkages of kinship and ethnicity are under increasing strain. In such an environment there can be little, if any, faith in the fairness and competence of state institutions or that opponents will play by the rules of the game in politics or the market. Lack of trust undermines largescale and risky enterprises in both the state and the market. Businessmen cannot count on either the behavior of their collaborators, competitors or customers, or the effective enforcement of contracts or honest business practices by the legal apparatus of the state. Partnerships seldom survive, I was repeatedly told in Ghana, because the partners never fully trust each other, especially those from different ethnic communities. It is equally difficult to mobilize for collective political goals when the unreliable competence and probity of leaders and followers makes citizens constantly wary of being turned into suckers or victims.

The Possibilities of Democratization and Bureaucratic Reform The outcome of the trajectory of development in sub-Saharan Africa over the past century that has produced this situation remains uncertain, but it is not likely, given current circumstances, that it will reproduce the paradigmatic forms of Western modernity, industrial capitalism and liberal democracy. What, then, can be done to make such political and economic developments more likely?44 In particular, what forms of instrumental agency are available to create a professionally capable and politically disinterested state bureaucracy in the face of practices that represent a now deeply ingrained bureaucratic culture in African states that is intimately linked to the institutions and cultures of the surrounding society? WTien I asked a Nigerian civil servant what people would call an 43 Catherine Boone, 'States and Ruling Classes in Post-Colonial Africa,' in Joel Migdal et al. eds, State Power and Social Forces. Cambridge: Cambridge University Press, 1994, pp. 124-9. 44 This assumes that capitalism and liberal democracy are currently the only viable goals of development currently available to African states. It is not a statement of personal preference.

Ethnicity, Bureaucracy & Democracy

51

honest official who did not use his office to directly benefit his family and friends, he answered simply, 'stupid'. In the face of such powerful 'common sense', the expectations that the widely promoted package of economic reforms, development of civil society and democratization through multi-party elections will produce significant social and political renewal in contemporary Africa leading to capitalism and liberal democracy are, I believe, sorely mistaken. First, economic crisis punctuated by sporadic spurts of growth has accentuated both inter-ethnic political competition and intra-ethnic class confrontations. The draconian reforms of the Structural Adjustment Programs imposed by the IMF and World Bank often represented further threats to the maintenance of established structures of political and economic power and to the politics of patronage. Not surprisingly, political elites dependent on state patronage have periodically clashed with the international financial institutions over the terms of adjustment programs, as both the civilian Moi regime in Kenya and the military magnates of Nigeria did in the early 1990s. 45 Second, hopes that the development of civil society would be a force for democratization are particularly unrealistic. Typical conceptions of civil society based on idealizations of Western experience tend to focus on borrowed sociocultural forms - churches, professional organizations, labour unions, universities, etc. - and ignore the dense networks of indigenous institutions that surround and pervade them, i.e. precisely those features of historical experience and the socio-cultural landscape that are expressed in moral ethnicity and political tribalism. This is not to deny the reality of the opposition parties, law societies, churches, trade unions and student organizations that have struggled to establish democratic institutions and free elections, and to defend human rights. These do represent the presence in African societies of many who are committed to the civic institutions and values of Western modernity. However, it is their ability to overcome the resistance of existing structures and practices of wealth and power and shift the historical trajectory of change that is in doubt. With distressing frequency, the rhizomes of ethnic factionalism and patron-client politics reproduce themselves within these parties and associations, rendering them, like so much of the apparatus of state, into ideological and institutional facades covering the reality of business as usual. The heart of the problem is that ethnic patron-client politics is not quite the same thing as corruption, although both involve appropriation of public resources for private purposes. Clientelism, however, is based on a recognized inequality between the persons involved that contains at least some degree of a reciprocal differentiation of authority and loyalty.46 In the context of ethnic communities, it is based on a degree of shared identity, mutual expectations and affective trust within a political and moral economy, however contested and tenuous, that is the outcome of the process of moral ethnicity. How, then, can the intra-ethnic focus of authority and trust relations be muted and the amoral and crudely materialistic free-for-all of political tribalism be replaced by a widerreaching civic politics and abstract collaborative trust focused on the institutions 45

Benno Ndulu and Francis Mwega, 'Economic Adjustment Policies', in Joel Barkan ed., Beyond Capitalism and Sociallism in Kenya and Tanzania. Boulder, CO and London: Lynne Rienner, 1994, pp. 102-17, 125-7; Tom Forrest, Politics and Economic Development in Nigeria. Boulder, CO: Westview Press, updated edition, 1995, pp. 242-8. 46 Donatella Delia Porta and Yves Meny, 'Democracy and Corruption: Towards a Comparative Analysis', in Delia Porta and Meny, Democracy and Corruption in Europe, pp. 173-5.

52

Bruce Berman

of the national state? Free, multi-party elections, the central agency of democratization for most political scientists, 47 are certainly necessary, but not sufficient in themselves to produce such an outcome. The renewal of competitive politics may actually intensify the struggle for state resources and the use of patronage by incumbent parties to buy electoral support, thus perpetuating the politics of cynicism and opportunism. 48 Eliminating bureaucratic corruption and patronage entirely is no more possible in Africa than it has been in Europe or North America. Nevertheless, transforming the state bureaucracy from the instrument and prize of patronage politics into something resembling a disinterested professional cadre is clearly an essential component of democratization and the creation of civic politics and trust. Doing so, however, involves both recognizing the ethnic pluralism of African societies and detaching the state bureaucracy from its linkages to it two potentially contradictory objectives. The issue of bureaucratic recruitment, for example, is crucial. We can see that official positions in the state should be filled through separate, politically isolated and disinterested agencies according to professional criteria of training and experience, rather than remaining the patronage reward of senior officials and the enclaves of specific ethnic groups. But bureaucracies in multi-ethnic states must also be representative of the different communities present; systematic exclusion or under-representation of any group would be a continuing source of conflict and focus of political tribalism. In such circumstances, which reflect the enduring differences of relative wealth and access to education among different ethnic communities, programs of 'affirmative action' may be called for. In themselves, however, such programs only perpetuate the materialistic 'our slice of the national cake' politics. Similarly, programs of bureaucratic decentralization, or federal institutions that give specific communities control over their own piece of the state apparatus, sustain the same orientation at levels in which personalistic patron-client politics is even more likely to thrive. They also do little to promote a civic politics at the national level or deal with the politics of Africa's cosmopolitan multiethnic cities. The professionalization of Western state bureaucracies involved an often tense and contradictory combination of both isolation from the political currents of the wider society and close alliances with powerful professional interests and reform movements within it. In the Western experience a significant degree of social and ideological distance was established between the higher cadres of the state bureaucracy and external social interests through distinctive elite educational paths that were crucial to developing in the former a commitment to an abstract public interest and a state-directed project of national development. 49 At the same time, however, it often produced an authoritarian bureaucratic caste that resisted its subordination to, and ultimate control by, popularly elected governments. This experience was reflected in the desultory efforts of the European powers to train their local successors in the elite secondary schools of colonial Africa and through access to metropolitan and, later, local universities. In Africa, however, within the context of established state-society linkages, this tended to 47

Michael Bratton and Nicolas van de Walle, Democratic Experiments in Africa. Cambridge: Cambridge University Press, 1997, pp. 12-13. 48 As is vividly depicted in David Throup and Charles Hornsby's comprehensive analysis of the 1992 general election in Kenya in Multi-Party Politics in Kenya. Oxford: James Currey, 1998, passim. 49 John Armstrong, The European Administrative Elite. Princeton, NJ: Princeton University Press, 1973.

Ethnicity, Bureaucracy & Democracy

53

provide an overlay of paternalistic authoritarianism on existing patterns of big man-small boy clientage relations, rather than an austere commitment to nation-building. The enduring dilemma of African political reform is that it must involve both weakening the linkages between the state and its agents and the ethnic communities of the wider society as the essential condition for muting the fires of political tribalism and achieving a wider collaborative trust promoting the development of both state and market, as well as providing free and open access to the professional cadres of the state for members of all ethnic communities by eliminating the discriminatory impact of differentials of numbers, wealth and educational access. It is strikingly similar to the American dilemma of 'black or blue5 occasioned by the recruitment since the 1960s of significant numbers of African-Americans into urban police forces and other civil service agencies. Once in these agencies, are such individuals to conduct themselves as representatives of the interests of their community or as disinterested professionals whose primary loyalty is to the ethos and solidarity of the public service? If they pursue the first, they will never gain the trust of their professional colleagues; if the latter, they risk losing the trust of their own community. 50 The contradictions involved probably admit of no single or permanent institutional or ideological solutions. Instead, the equity of ethnic composition and the disinterested professional competence of the state bureaucracy will remain constantly contested features of the process of democratization in Africa, and the necessary focus of continual institutional innovation and reform. 50 I leave aside here the question of whether either orientation necessarily guarantees the disinterested enforcement of law. It is significant, however, that the enormous personal stress to which individuals in such positions are subject is a common topic of both American politics and popular culture.

4

WILL KYMLICKA

Nation-Building & Minority Rights: Comparing Africa & the West

A

RE Western models of nation-building and minority rights relevant to Africa? In this chapter, I shall offer a qualified 'maybe' to that question. I begin by explaining in the following two sections what I take to be the main outlines of a common Western approach to nationbuilding and minority rights. There are many differences amongst the Western nations, but I shall try to show that there have been important areas of convergence in recent decades, which can be seen as defining a distinctively Western approach to the issue. I shall then discuss its possible application to Africa.

Nation-States and Nation-Building States We all know that the term 'nation-state' is misleading. There are some 190 independent states in the world today, but most commentators agree that there are upward of 5000-8000 distinct 'peoples' or 'nations'. Simple arithmetic dictates that most states (at the moment over 90 per cent) will be shared by more than one national group, and often by dozens. And yet, in another sense, the term 'nation-state' is apt, since modern states typically aspire to be nation-states, and have adopted various nation-building programs to achieve greater national integration and homogeneity. They may not be nation-states, but they are certainly nation-building states. This fact is obscured by the myth of 'ethno-cultural neutrality', according to which Western liberal states are 'neutral' with respect to the ethno-cultural identities of their citizens, and indifferent to the ability of ethno-cultural groups to reproduce themselves over time. On this view, liberal states treat culture in the same way as religion - i.e., as something which people should be free to pursue in their private life, but which is not the concern of the state (so long as people respect the rights of others). Just as liberalism precludes the establishment of an official religion, so too there cannot be official cultures which have preferred status over other possible cultural allegiances. In reality, however, liberal-democratic states are far from ethno-culturally neutral. Consider the policies of the United States, which is the allegedly prototypically 'neutral' state. First, it is a legal requirement for children to learn the English language in schools. Secondly, it is a legal requirement for immigrants (under the age of 50) to learn the English language to acquire American citizenship. Thirdly, it is a de facto requirement for employment in or for government that the applicant speak English. Fourthly, decisions about the boundaries of state governments, and the timing of their admission into the federation, were deliberately made to ensure that anglophones would be a majority within each of the fifty states of the American federation. These policies regarding the language of education and government employ54

Nation-Building & Minority Rights

55

ment, the requirements of citizenship, and the drawing of internal boundaries, are not isolated exceptions to some norm of ethno-cultural neutrality. On the contrary, they have shaped the very structure of the American state, and have played a pivotal role in determining which ethnolinguistic groups prosper, and which ones diminish. For example, since governments account for 40 per cent50 per cent of GNP in most countries, decisions about the language of government employment and contracts are a major factor in determining which language groups survive. These measures were adopted with the intention of promoting integration into what I call a 'societal culture'. By a societal culture, I mean a territorially concentrated culture, centered on a shared language which is used in a wide range of societal institutions, in both public and private life (schools, media, law, economy, government, etc.). I call it a societal culture to emphasize that it involves a common language and social institutions, rather than common religious beliefs, family customs or personal lifestyles. Societal cultures within a modern liberal democracy are pluralistic, containing Christians as well as Muslims, Jews and atheists; heterosexuals as well as gays; urban professionals as well as rural farmers; conservatives as well as socialists. Such diversity is the inevitable result of the rights guaranteed to citizens - including freedom of conscience, association, speech, political dissent and privacy - particularly when combined with an ethnically diverse population. The American government has deliberately promoted integration into such a societal culture - that is, it has encouraged citizens to view their life-chances as tied up with participation in common societal institutions that operate in the English language. Obviously, the sense in which English-speaking Americans share a common 'culture' is a very thin one, since it does not preclude differences in religion, personal values, family relationships or lifestyle choices. While thin, it is far from trivial. On the contrary, as I discuss below, attempts to integrate people into such a common societal culture have often been resisted. Some groups have vehemently rejected the idea that they should view their life-chances as tied up with the societal institutions conducted in the majority's language. The United States is not unique in this respect. Virtually all liberal democracies have, at one point or another, attempted to diffuse a single societal culture throughout all of their territory. Decisions regarding official languages, core curriculum in education, and requirements for acquiring citizenship, were all made with the intention of diffusing a particular culture throughout society, and of promoting a particular national identity based on participation in that societal culture. In some countries, these nation-building policies have been strikingly successful. Who could have predicted in 1750 that virtually everyone within the current boundaries of France or Italy would share a common language and sense of nationhood? Other paradigmatic 'nation-states' include England, Germany, and Portugal. There are other countries, as we shall see in the next section, where minorities have successfully resisted attempts to integrate them into the dominant societal culture, and governments have had to back down on their nation-building policies. But the quest to become a nation-state has been a powerful one in most Western democracies. Why has nation-building been so ubiquitous? As Canovan puts it, nationhood is 'the battery' which makes Western states run: the existence of a common national identity motivates citizens to act for common political goals (Canovan, 1996: 80). Modern states need to be able to mobilize citizens in pursuit of a

Will Kymlicka

56

wide range of goals. The 'battery' of nationalism can be used to promote liberal goals (such as social justice, democratization, equality of opportunity, economic development) or illiberal goals (chauvinism, xenophobia, militarism, and unjust conquest). The fact that nationalism can be used for so many functions helps explain why it has been so ubiquitous. Liberal reformers invoke nationhood to mobilize citizens behind projects of social justice (for example, comprehensive health care or public schooling); illiberal authoritarians invoke nationhood to mobilize citizens behind attacks on alleged enemies of the nation, be they foreign countries or internal dissidents. This is why nation-building is just as common in authoritarian regimes as in democracies. Consider Spain under Franco, or Latin America under the military dictators. Authoritarian regimes also need a 'battery' to help achieve public objectives in complex modern societies.

Minority Responses to Nation-Building How does the existence of nation-building affect the claims of minorities? As Charles Taylor notes, the process of nation-building inescapably privileges members of the majority culture: If a modern society has an 'official' language, in the fullest sense of the term, that is, a state-sponsored, -inculcated, and -defined language and culture, in which both economy and state function, then it is obviously an immense advantage to people if this language and culture are theirs. Speakers of other languages are at a distinct disadvantage. (Taylor, 1997: 34) This means that minority cultures have limited options. If all public institutions are being run in another language, minorities face the danger of being marginalized from the major economic, academic and political institutions of the society. Faced with this dilemma, minorities have (to oversimplify) three basic options: (i) they can accept integration into the majority culture, and perhaps attempt to renegotiate the terms of integration; (ii) they can seek the sorts of rights and powers of self-government needed to maintain their own societal culture - i.e., to create their own economic, political and educational institutions in their own language; (iii) they can accept permanent marginalization. We can find ethno-cultural groups in the West which fit each of these categories (and other groups which are caught between them). For example, some immigrant groups choose permanent marginalization. This would seem to be true of the Hutterites in Canada or the Amish in the United States. But the option of accepting marginalization is only likely to be attractive to religious sects whose theology requires them to avoid all contact with the modern world. The Hutterites and Amish are unconcerned about their marginalization from universities or legislatures, since they regard such 'worldly' institutions as corrupt. Virtually all other ethno-cultural minorities, however, seek to participate in the modern world, and to do so they must either integrate into the majority society or seek the self-government needed to sustain their own modern institutions. Faced with this choice, ethno-cultural groups have responded in different ways. I shall discuss four types of ethno-cultural groups in Western democracies: national minorities, immigrants, metics, and racial caste groups. In each case, I shall discuss how they have been affected by majority nation-building, and what

Nation-Building & Minority Rights

57

sorts of claims they have made in response. (a) National minorities: By national minorities, I mean groups which formed complete and functioning societies in their historic homeland prior to being incorporated into a larger state. This incorporation was usually involuntary, as a result of conquest, colonization, or the ceding of territory from one empire to another, but in a few cases was the result of a voluntary federation between two or more national groups. National minorities can be subdivided into two types: 'stateless nations' (for example, Catalans, Basques, Puerto Ricans, Flemish, Scots, Quebecois), and 'indigenous peoples' (for example, Sami, Inuit, Maori, American Indians and Australian Aborigines). However they were incorporated, national minorities have often been the first target of majority nation-building programs, since one of the aims of nationbuilding states is to deny that there is more than one 'nation' within the state. To achieve this aim, nation-building states often seek to (a) pressure national minorities to integrate into the majority's societal culture; and (b) undermine the minority's ability to reproduce its own societal culture (for example, by replacing its traditional educational, political and legal institutions with majority-controlled institutions). As a general rule, national minorities have resisted such nation-building policies, and have fought instead to maintain their traditional institutions, and to reproduce their own societal culture. And this has often required that minorities adopt their own competing nation-building on a sub-state level. Indeed, they often seek to use the same nation-building tools that the majority uses; for example, they seek control over the language and curriculum of schooling in their region of the country, the language of government employment, settlement policies, and the drawing of internal boundaries. In short, national minorities have typically sought to maintain or enhance their political autonomy. At the extreme, this may involve claims to outright secession, but more usually it involves some form of regional autonomy. And they typically mobilize along nationalist lines, using the language of 'nationhood' to describe and justify these demands for self-government. How have liberal democracies responded to such minority nationalist claims? Historically, they have suppressed minority nationalisms, often ruthlessly. In the eighteenth and nineteenth centuries, for example, France banned the use of the Basque and Breton languages in schools or publications, and banned any political associations which aimed to promote minority nationalism; Canada stripped the Quebecois of their French-language rights and institutions, and redrew political boundaries so that Francophones did not form a majority in any province; Canada also made it illegal for Aboriginals to form political associations to promote their national claims; and when the United States conquered the Southwest in 1848, it stripped the long-settled Hispanics of their Spanishlanguage rights and institutions, imposed literacy tests to make it difficult for them to vote, and encouraged massive immigration into the area so that Hispanics would become outnumbered. All of these measures were intended to disempower national minorities, and to eliminate any sense of possessing a distinct national identity. This was justified on the ground that minorities which view themselves as distinct 'nations' would be disloyal, and potentially secessionist. Moreover, it was said that economic development required access to the lands and resources in the minority's territory. And it was often claimed that minorities were backward and uncivi-

58

Will Kymlicka

lized, and that it was in their own interests to be incorporated (even against their will) into more civilized nations. National minorities, therefore, were often the first target of majority nation-building campaigns. 1 But the attitude of liberal democracies has changed. It is increasingly recognized that the suppression of minority nationalism was mistaken. Pressuring national minorities to integrate into the dominant national group has simply not worked. Western states misjudged the durability of minority national identities. Despite centuries of legal discrimination and social prejudice, national minorities have maintained the sense of forming distinct nations, and the desire for national autonomy. As a result, when the state attacks a minority's sense of distinct nationhood, the result is often to promote rather than reduce the threat of disloyalty and secessionist movements (Gurr, 1993). In the experience of Western democracies, the best way to ensure the loyalty of national minorities has been to accept, not attack, their sense of distinct nationality. This trend is present in most Western states with national minorities. Some countries have adopted a federal system which grants national minorities significant language rights and regional autonomy; they include Canada (for the Quebecois), Switzerland (the French and Italians), Spain (Catalonia and the Basque Country), and Belgium (the Flemish and Walloons). Other countries have adopted quasi-federal forms of territorial autonomy for national minorities, including Britain (Scotland and Wales), the United States (Puerto Rico), and Italy (the Germans in South Tyrol). Also, many countries, including Canada, the US, New Zealand and the Scandinavian countries, have accorded self-government rights to indigenous peoples. In all of these countries, the goal of eliminating minority national identities has been abandoned, and it is now accepted that these groups will continue to see themselves as separate and self-governing nations within the larger state into the indefinite future. So earlier attempts to suppress minority nationalism have been abandoned as unworkable. But they have also been rejected as morally indefensible. After all, on what basis can liberal-democratic theory justify the suppression of minority nationalisms while allowing majority nation-building? If the majority can engage in legitimate nation-building, why not national minorities, particularly those which have been involuntarily incorporated into a larger state?2 As a result, an increasing number of Western democracies accept that they are 'multination' states, rather than 'nation-states'. And an increasing number of them recognize that national minorities are best accommodated through some form of federalism, since it allows the creation of regional political units, controlled by the national minority, with substantial powers of self-government. What we see emerging within several Western democracies, therefore, is a form of 'multination federalism' - i.e., a model of the state as a federation of region1 This raises the question captured in the title of Walker Connor's article: are nation-states 'NationBuilding or Nation-Destroying?' (Connor, 1972). They are both: they have sought to build a common nationhood by destroying any pre-existing sense of distinct nationhood on the part of national minorities. 2 Liberal principles set limits on how national groups go about nation-building: they preclude attempts at ethnic cleansing, or stripping people of their citizenship, or the violation of human rights. Liberal principles also require that any national group engaged in a project of nation-building respect the right of other nations within its jurisdiction to protect their own national institutions. For example, the Quebecois are entitled to assert national rights vis-a-vis the rest of Canada, but only if they respect the rights of Aboriginals within Quebec to assert national rights vis-a-vis the rest of Quebec. But these limits apply equally to majority as to minority nation-building: they are not a reason to endorse majority nation-building while attacking minority nationalism.

Nation-Building & Minority Rights

59

ally concentrated peoples or nations, in which boundaries have been drawn, and powers distributed, in such a way as to ensure that each national group is able to maintain itself as a distinct and self-governing societal culture (Kymlicka, 2001). (b) Immigrants: By immigrants, I mean groups formed by the decision of individuals and families to leave their original homeland and emigrate to another society, often leaving their friends and relatives behind. This decision is typically made for economic reasons, although sometimes also in order to move to a freer or more democratic country. But we need to distinguish two categories of immigrants - those who have the right to become citizens, and those who do not. I shall use the term 'immigrant group' only for the former case, and will discuss the latter case, which I call 'metics', below. Immigrants, then, are people who arrive under an immigration policy which gives them the right to become citizens after a relatively short period of time say, 3-5 years - subject only to minimal conditions (for example, learning the official language, and knowing something about the country's history and political institutions). This has been the traditional policy in the three major 'countries of immigration' in the West - the United States, Canada and Australia. Historically, immigrant groups have responded differently to majority nationbuilding from national minorities. The option of engaging in competing nationbuilding has been neither desirable nor feasible for immigrant groups. They are typically too small and too territorially dispersed to hope to recreate their original societal culture from scratch in a new country. Instead, they have traditionally accepted the expectation that they will integrate into the larger societal culture. Few immigrant groups have objected to the requirement that they must learn an official language as a condition of citizenship, or that their children must learn the official language in school. They have accepted the assumption that their life-chances, and the life-chances of their children, will be bound up with participation in mainstream institutions operating in the majority language. What immigrants have tried to do, however, is to renegotiate the terms of integration into mainstream institutions. Recent debates over 'multiculturalism' in immigrant countries are precisely about the terms of integration. Immigrants are demanding a more tolerant approach which would allow them to maintain various aspects of their ethnic heritage even as they integrate into common institutions operating in the majority language. Immigrants insist that they should be free to maintain some of their old customs regarding food, dress, recreation, religion, and to associate with each other to maintain these practices. This should not be seen as unpatriotic or 'unamerican'. Moreover, the institutions of the larger society should be adapted to provide greater recognition of these ethnic identities; for example, schools and other public institutions should accommodate their religious holidays, dress, dietary restrictions, and so on. How have liberal democracies responded to such demands for immigrant multiculturalism? Here again, liberal democracies have historically resisted these demands. Until the 1960s, all three major immigrant countries adopted an 'Anglo-conformity' model of immigration. Immigrants were expected to assimilate to existing cultural norms and, over time, become indistinguishable from native-born citizens in their speech, dress, leisure activities, cuisine, family size, identities, and so on. This strongly assimilationist policy was seen as necessary to ensure that immigrants become loyal and productive members of society. However, beginning in the 1970s, it was increasingly recognized that this assimilationist model is unrealistic and unnecessary. It is unrealistic because no

60

Will Kymlicka

matter how much pressure is applied, immigrants never fully lose their distinctive identities. Immigrants do indeed integrate into common institutions and learn the dominant language, but they often remain visibly, and proudly, distinctive. Assimilationist pressure is also unnecessary, since the evidence shows that immigrants who maintain a strong sense of ethnic pride can nonetheless be loyal and productive citizens. As a result, the assimilationist model has increasingly been rejected in favour of a more multicultural model of integration, which accepts the premise of majority nation-building but revises the terms of integration. Mainstream institutions - schools, workplaces, courts, police forces, and welfare agencies - have been reformed to provide greater recognition of immigrants' ethno-cultural identities, and greater accommodation of their ethno-cultural practices, so that they will feel more at home within these institutions. These changes are not only prudent, they are morally preferable, since the old assimilationist model denies equal respect to immigrants, and turns integration into an oppressive process. Immigrant demands for a more 'multicultural' model of integration are, I think, a fair response to majority nation-building. If liberal democracies are going to pressure immigrants to integrate into common institutions operating in the majority language, then we need to ensure that the terms of integration are fair (Kymlicka, 2001; Spinner, 1994). These first two types of groups - national minorities and immigrants - have both been pressured by the state to integrate into the majority's societal culture. The final two types of group I shall discuss - metics and racial caste groups - are different. Far from being pressured to integrate, they have been prohibited from doing so, and been forcibly kept separate, even if they wanted to integrate. (c) Metics: Some migrants are never given the opportunity to become citizens, either because they entered the country illegally, or because they entered as students or 'guest-workers' but have overstayed their initial visa. When they entered the country, these people were not conceived of as future citizens, or even as long-term residents, and would not have been allowed to come in in the first place if they had been seen as permanent residents and future citizens. However, despite the official rules, they have settled more or less permanently. In principle, many face the threat of deportation if they are detected by the authorities or are convicted of a crime. But they nonetheless form sizeable communities in certain countries, engaging in some form of employment, legal or illegal, marrying and forming families. This is true of Mexicans in California, Turks in Germany, or North Africans in Spain. Borrowing a term from Ancient Greece, Walzer calls these groups 'metics' - long-term residents who are nonetheless excluded from the polis (Walzer, 1983). Since metics face enormous obstacles to integration, they tend to exist in the margins of the larger society. The most basic claim of metics is to regularize their status as permanent residents, and to gain access to citizenship. They want, in effect, to be able to follow the immigrant path to integration into the mainstream society, even though they were not initially admitted as immigrants. How have Western democracies responded to this demand for access to citizenship? Some countries - particularly the traditional immigrant countries - have grudgingly accepted these demands. Guest-workers who overstay their visas are often able to gain permanent residence, and periodic amnesties are offered for illegal immigrants, so that over time they become similar to immigrants in their status. But other countries

Nation-Building & Minority Rights

61

- particularly those which do not think of themselves as immigrant countries have resisted these demands. These countries may have no established infrastructure for integrating immigrants. Also, many metics have either broken the law to enter the country (illegal immigrants), or broken their promise to return to their country of origin (guest-workers), and so are not viewed as worthy of citizenship. Moreover, countries with no tradition of accepting newcomers may be prone to regard all foreigners as potential security threats, or simply as unalterably 'alien'. In these countries, of which Germany, Austria and Switzerland are the best-known examples, the official policy is not to try to integrate metics into the national community, but to get them to leave the country, either through expulsion or voluntary return. In short, the hope was that if metics were denied citizenship, so that they had only a precarious legal status within the country, and if they were told repeatedly that their real home was in their country of origin, and that they were not wanted as members of the society, then they would eventually go home. But it is increasingly recognized that this approach is not viable. Metics who have lived in a country for several years are unlikely to go home, even if they have only a precarious legal status. This is particularly true if they have married and had children in the country. At this point, it is their new country, not their country of origin, which is their 'home'. Indeed, it may be the only home that their children and grandchildren know. Once they have settled, founded a family, and started raising their children, nothing short of expulsion is likely to get metics to return to their country of origin. So a policy based on the hope of voluntary return is unrealistic. Moreover, it endangers the larger society. For the likely result of such a policy is to create a permanently disenfranchised, alienated, and racially, or ethnically, defined underclass. The predictable consequences can involve some mixture of political alienation, criminality, and religious fundamentalism amongst the metics, particularly the second generation, which in turn leads to increased racial tensions, even violence, throughout the society. To avoid this, an increasing number of Western democracies, even non-immigrant countries, are adopting amnesty programs for illegal immigrants, and granting citizenship to guest-workers and their children. In effect, long-settled metics are increasingly allowed to follow the immigrant path to integration. This is not only prudent, but also morally required. For it violates the very idea of a liberal democracy to have groups of long-term residents who have no right to become citizens. A liberal democratic system is a system in which those people who are subject to political authority have a right to participate in determining that authority. To have permanent residents who are subject to the state, but unable to vote, is to create a kind of caste-system which undermines the democratic credentials of the state (Walzer, 1983). (d) African-Americans: One final group worth considering is the Blacks who are descended from the African slaves brought to the United States in the eighteenth and nineteenth centuries. Under slavery, Blacks were not seen as citizens, or even as 'persons', but simply as the property of the slave owner. Although slavery was abolished in the 1860s, and Blacks were granted citizenship, they were still subject to segregation laws which required that they attend separate schools, and serve in separate army units, until the 1950s and 1960s. And while such discriminatory laws have been struck down, Blacks remain subject to informal discrimination in hiring and housing, and disproportionately concen-

62

Will Kymlicka

trated in the lower class and in poor neighborhoods. Like metics, African-Americans were historically excluded from becoming members of the American nation. But unlike metics, the justification for this was not that they were citizens of some other nation to which they should return. Blacks in America can hardly be seen as 'foreigners' or 'aliens', since they have been in the US as long as Whites, and have no foreign citizenship. Instead, they were effectively denationalized, they were denied membership in the American nation, but they were not viewed as belonging to some other nation. African-Americans are unlike most other ethno-cultural groups in the West. They do not fit the voluntary immigrant pattern, not only because they were brought to America involuntarily as slaves, but also because they were prevented (rather than encouraged) from integrating into the institutions of the majority culture. Nor do they fit the national minority pattern, since they do not have a homeland in America or a common historical language. Moreover, before emancipation, they were legally prohibited from trying to recreate their own cultural structure (for example, all forms of black association, apart from churches, were illegal, as was the use of African languages). In light of these unique circumstances, African-Americans have raised a unique set of demands. The civil rights movement in the US in the 1960s was often seen as enabling Blacks to follow the immigrant path of integration, through a more rigorous enforcement of the anti-discrimination laws. Those African-Americans who were sceptical about the possibility of following the immigrant path to integration, however, have pursued the opposite tack of redefining Blacks as a 'nation', and promoting a form of black nationalism. Much of the history of African-American political mobilization can be seen as a struggle between these two competing projects. But neither project is realistic. The legacy of centuries of slavery and segregation has created barriers to integration which immigrants do not face. As a result, despite the legal victories of the civil rights movement, Blacks remain disproportionately at the bottom of the economic ladder, even as more recent nonwhite immigrants have integrated. But the territorial dispersion of Blacks has made the option of national separatism equally unrealistic. Even if they shared a common national identity, which they do not, there is no region of the United States where Blacks form a majority. As a result, it is increasingly recognized that a sui generis approach will have to be worked out for African-Americans, involving a variety of measures. These may include historical compensation for past injustice, special assistance in integration (for example, affirmative action), guaranteed political representation (for example, redrawing electoral boundaries to create black-majority districts), and support for various forms of black self-organization (for example, subsidies for historical black colleges, and for black-focused education). These different demands may seem to pull in different directions, since some promote integration while others seem to reinforce segregation, but each responds to a different part of the complex and contradictory reality in which African-Americans find themselves. The long-term aim is to promote the integration of African-Americans into the American nation, but it is recognized that this is a long-term process that can only work if existing Black communities and institutions are strengthened (Spinner, 1994). This shift in approach reflects both moral and prudential considerations. African-Americans suffered perhaps the greatest injustices of all ethno-cultural groups, and so we have an urgent moral obligation to remedy these injustices. Moreover, as with metics, the result of this ongoing exclusion has been the

Nation-Building & Minority Rights

63

development of a separatist and oppositional subculture in which the very idea of pursuing success in 'white' institutions is viewed by some Blacks with suspicion. The costs of allowing such a subculture to arise are enormous, both for the Blacks themselves, many of whom are condemned to lives of poverty, marginalization, and violence, and for society at large, in terms of the waste of human potential, and the escalation of racial conflict. I have discussed four types of ethno-cultural groups, and tried to show how their demands are best understood in relation to majority nation-building. Each group can be seen as claiming that majority nation-building has imposed certain injustices on them, and as identifying the conditions under which majority nation-building would cease to be unjust. 3 It is important to note that, in all of these cases, minorities are not saying that nation-building programs are impermissible. But they insist that nation-building programs should be subject to limitations. If we try to summarize these demands, we can say that majority nation-building in a liberal democracy is legitimate under the following conditions: (a) no groups of long-term residents are permanently excluded from membership in the nation, such as metics or racial caste groups. Everyone living in the territory must be able to gain citizenship, and become an equal member of the nation if they wish to do so; (b) insofar as immigrant groups and other ethno-cultural minorities are pressured to integrate into the nation, the terms of integration should leave room for the expression of individual and collective differences, both in public and private, and public institutions should be adapted to accommodate the minority's identity and practices. Put another way, if minorities are pressured to integrate into dominant institutions, these institutions must be adapted to accommodate these minorities; (c) national minorities are granted self-government to maintain themselves as distinct societal cultures. These three conditions have rarely been met within Western democracies, but there is growing acceptance of them. The patterns I have been discussing are generalizations, of course, not iron laws. Some metics, immigrant groups and national minorities have not mobilized to demand minority rights, and even when they have, some Western countries (for example, Greece) continue to resist these demands. And even where these demands have been accepted, they often remain controversial. Still, the general trend is clear: Western states today exhibit a complex pattern of nation-building constrained by minority rights. On the one hand, Western states remain 'nation-building' states: all Western states continue to adopt the sorts of nation-building policies I discussed in the American context. On the other hand, these policies are increasingly modified to accommodate the demands of minorities who feel threatened. Minorities have demanded, and increasingly been accorded, various rights which help ensure that nation-building does not exclude metics and racial caste groups, or coercively assimilate immigrants, or undermine the self-government of national minorities. 3 There are groups in the West which do not fit into these four categories (e.g., the Roma). But I would argue that the same basic framework applies to their claims as well: whatever the group, minority rights claims are best understood as a response to perceived threats which arise as a result of majority nation-building.

64

Will Kymlicka

Comparisons with Africa I believe that this emerging pattern of nation-building and minority rights in the West is desirable in principle and successful in practice (Kymlicka, 2001). Can it work in Africa as well? I lack the expertise to answer this question, so I shall simply try to raise a few of the relevant issues. One question is whether the categories I have been using make sense in Africa. Can we speak about 'immigrants', 'indigenous peoples', 'national minorities' and 'metics' in the African context, or do we need a different vocabulary? I shall return to this question below. However, before we examine the nature of minorities in Africa and their claims against the state, we must first examine the nature of the state in Africa, and the claims it makes on minorities, since, as we have seen, the former are often a response to the latter. So the first question we need to ask is whether African states have been nation-building. Many commentators argue that they have not, or at least not in any sense which is comparable to the West. Some argue that African states are less prone to nation-building since fewer are democracies. Democracies must be sensitive to majority preferences, and since majority groups typically want to extend the reach of their language and culture, democracies face populist pressure for nation-building. By contrast, countries which are not democratic can resist majority demands for nation-building, and be more sensitive to minority claims. It may be true that authoritarian regimes can ignore populist pressures more readily than democracies. But it is a mistake, I think, to suppose that nationbuilding policies in the West were adopted in response to populist pressure. Even in Western democracies, nation-building began as an elite-initiated project, which only later became a matter of passion for the masses. Far from being the result of majoritarian preferences, nation-building policies were initially adopted by elites precisely in order to create a cohesive sense of 'nationhood' amongst the masses, who could then be mobilized in pursuit of various public objectives. And, as I have noted, this need to mobilize citizens applies to all modern states; it applies as much to authoritarian regimes as to democracies. But there is a more important reason for thinking that African states are not 'nation-building' in the Western sense. In the West, state nation-building has been a matter of majority nation-building. Even if nation-building policies have been elite-initiated, they nonetheless have involved diffusing the majority (or dominant) group's language and culture throughout the territory of the state. The majority's language and culture are imposed on national minorities, indigenous peoples and immigrants. In Africa, by contrast, most states do not have a majority in this sense. They are composed of several ethno-cultural groups, none of which form more than 25 or 30 per cent of the population. This has enormous consequences both for the groups and for the state. From the point of view of the ethnic groups, few have had the power or aspiration to form 'nation-states' in the Western sense, in which their language and culture would monopolize public space and public institutions. As Markakis puts it, this is not a realistic possibility in countries 'with a legion of ethnic groups of all sizes and manners of spatial distribution, whose identities, upon closer inspection, have a tendency to dissolve into smaller components like clans and lineages. Comparatively few ethnic groups in black Africa are candidates for

Nation-Building & Minority Rights

65

nationhood, and few of them have such aspirations' (Markakis, 1999: 77). And just as ethnic groups cannot aspire to their own nation-state, so too states cannot attempt to integrate citizens by diffusing a particular ethnic group's language and culture throughout the territory of the state. To be sure, most African states are interested in developing a common identity, and developing common public institutions and a common public sphere operating in a common language. But this has not typically involved diffusing the majority's language, history, and identity (since there is no such majority). Rather, it has typically involved diffusing the colonial language as the language of state institutions, and trying to develop pan-ethnic bases for state identification (or 'patriotism', if you like) which would appeal to most or all ethnic groups in the territory of the state. For example, rather than basing national identity on the glorious history of the dominant group (as is true of most Western national identities), African states have tried to develop a common identity by appealing to a common future in which all groups can share (Eriksen, 1999: 49). Eriksen insists that Mauritius, for example, is 'in the process of developing a common set of supra-ethnic, national myths and symbols which is invested with meaning and relevance by the bulk of the population', and has a 'high level of cultural integration, which makes a national public sphere possible', based on a Creole version of the colonial French language (ibid.: 52-3). We can call this 'nation-building', but it does not involve building the state around the language, culture and history of the dominant ethnonational group. As a result, nation-building of this pan-ethnic sort is, in theory, not as much of a threat to minorities. In Western states, 'nation-building' has typically meant 'nation-destroying': national minorities were often the first target of majority nation-building campaigns. But in Africa, by contrast, where there are no majority groups, state nation-building is not majority nation-building, and may not privilege the largest group over smaller groups. It would be paradoxical if it turned out to be Africa rather than the West which was the real home of 'civic' nationalisms, whose official language and symbols really are neutral amongst the various cultures and identities of the ethnic groups living in the territory. And in such a context, the dialectic of nation-building and minority rights would be different. Where nation-building does not privilege a hegemonic majority group, there may be less need for specific rights to protect minorities from the injustices which arise as a result of majority nation-building. 4 4 This sort of nation-building can be seen as a threat to all ethnic groups, large or small, since it privileges the colonial language, and provides little public space or public support for local languages and cultures. Indeed, this sort of nation-building has often been seen as requiring, not only indifference to local ethnic identities, but hostility towards them. State leaders in Africa have often supposed that this pan-ethnic identification can only be achieved by trying to suppress more local ethnic identities. Hence the old FRELIMO slogan 'Kill the tribe to build the nation'. As Okoh notes, this could have been 'the credo of almost all African rules and intellectuals' (Okoh, 1997: 150). This was supposed to apply to all 'tribes', including the largest, not just to the smaller minority tribes, and so did not give rise to inter-ethnic conflicts. But it raises the question why have ethnic groups not insisted on greater public support for their languages? As Mazrui notes, Africans have shown little loyalty to indigenous languages (he claims that only Somalis and Afrikaners have manifested linguistic nationalism). He attributes this to several factors, including the relatively recent development of a written literature in African languages and more specifically, the fact that none of the African languages have sacred literature. He also claims that the weak attachment to indigenous languages reflects the fact that 'the humiliation of black people has been much more on the basis of their race than on the basis of their language. African nationalism is therefore much more inspired by a quest for racial dignity than by a desire to defend African languages' (Mazrui and Mazrui, 1998: 4-6). Also, the pragmatic necessity of multilingualism has a long history in Africa, preceding either colonialism or the nation-state.

66

Will Kymlicka

Ethnic conflicts can still arise in this form of post-colonial pan-ethnic nationbuilding. But they are likely to take the form of struggles for a share of state power at the central level, rather than ethnonationalist struggles for self-government and autonomy at a regional level. This indeed is a common feature in African politics. The problem is that while the state may be neutral (i.e., its language, culture and symbols are not tied to any particular ethnic group), access to the state remains ethnified. Access to state power is tied to networks of clientelism/patronage and to political parties that are predominantly defined along ethnic lines. The state itself may be more or less neutral, but the avenues for accessing state power go through ethnic parties and networks. And this raises the danger that some ethnic groups will have much better access routes to the state, while other ethnic groups are excluded. Where this is the problem, ethnic groups are likely to mobilize as what Gurr calls 'communal contenders', fighting for a share of state power, rather than as 'ethnonationalists', fighting for autonomy. And indeed Gurr says that the distinctive feature of ethnic conflict in Africa is that it primarily involves communal contenders rather than minority nationalism (Gurr, 1993). 5 Where this is the case, the closest analogy in the West would not be any of the models discussed earlier, but rather the sort of consociationalism which was adopted to accommodate religious groups in Belgium, the Netherlands, Austria and (now) Northern Ireland. In these countries, Protestants and Catholics were in effect 'communal contenders', rather than ethnonationalists. Under consociationalism, each religious group was guaranteed proportional shares in political power and in access to public resources, a veto power over certain decisions, and guarantees of a grand coalition government, so that neither group was excluded from state power. Insofar as ethnic conflicts in Africa take the form of conflicts between communal contenders for a share of state power, consociationalism is the most relevant Western model. Consociationalism has been tried in Africa, with mixed results. It failed totally in Rwanda and Burundi, 6 but Lemarchand says that it may work elsewhere (for example, Liberia, Angola), and that the main obstacle is often simply the lack of will to make it work (Lemarchand, 1997: 99). In any event, Western countries provide no clear models other than consociationalism for this sort of case. So this is one important pattern of the relationship between state nationbuilding and minority rights in Africa. Where state nation-building is based on a colonial language and pan-ethnic symbols that are neutral amongst the various ethnic groups, minorities will mobilize as communal contenders to ensure that they are not excluded from a share of state power. Consociational forms of minority rights are one possible outcome of this struggle. But this is not the only sort of ethnic conflict in Africa, in part because it is not the only form of nation-building. In some countries, state nation-building is not neutral amongst ethnic groups, but is shaped by the dominant ethnic group 5

Gurr defines communal contenders as groups which have a share of state power, and which have and cherish their own separate cultural institutions, but which do not seek territorial autonomy or independent statehood. Unlike national minorities and indigenous peoples, they do not see themselves as distinct 'nations' with rights to self-government; unlike immigrants, they do not expect to integrate into the dominant group; and unlike metics or racial caste groups, they share state power, rather than being excluded from it. 6 Some pessimists argue that the only option left is separation (via ethnic cleansing) into a Hutuland and Tutsiland, or the incorporation of Rwanda and Burundi into a larger federation with Tanzania or Uganda, so that they will be 'swallowed' by other larger tribes (Gasana, 1997).

Nation-Building & Minority Rights

67

to reflect its language, history and culture. In these cases, we see something closer to Western-style forms of majority (or dominant-group) nation-building. In Ethiopia under Selassie, for example, nation-building was not based on a colonial language, but involved diffusing the Amharic language and culture throughout the territory of the state. As in the West, this sort of dominant group nation-building entailed minority nation-destroying, as concerted efforts were made to suppress the other languages in the country. Similarly in Sudan, nationbuilding has involved attempts to impose the dominant Arabic language and Islamic culture throughout the territory of the state, at the expense of the languages, cultures and religions of southern Sudan. And in other countries, conflicts have arisen from the fear that the state might engage in dominant-group nation-building, for example, the fear that Nigeria would adopt a conception of nation-building built around the language, religion and culture of the Hausa. 7 In these cases, what we see is something closer to Western-style dominantgroup nation-building. And not surprisingly, minorities often respond with Western-style ethnonationalism. 8 Confronted with Amharic nation-building in Ethiopia, minorities responded with calls for autonomy, federalism and even secession in Eritrea. Confronted with Arabic nation-building in Sudan, minorities demanded autonomy, federalism and even secession in the South. Confronted with the threat of Hausa nation-building, minorities in Nigeria demanded federalism and even secession in the east and south. 9 In all of these cases, the closest analogy in the West is the struggle by national minorities for multination federalism. There are many parallels between minority nationalisms in Africa and those in Western democracies, such as the Quebecois, Flemish or Catalans. In both contexts, we see that groups are seeking some form of regional autonomy; that this mobilization was triggered, or intensified, in response to the threat posed by dominant-group nation-building; that this has generated demands for the adoption of federalism; and that there is the threat of secession if this desire for autonomy is not met. 10 How have African states responded to such minority nationalisms? As in the West, they have generally tried to suppress them, and have often given the same 7

Similarly, some Zulu and Afrikaner leaders argued for federalism in South Africa because of fear that the ANC conception of nation-building would not be neutral, but rather dominated by the Xhosa. However, these leaders have failed to persuade their own members that this is a serious threat, as shown by decreasing support for the NP and IFP, and so most Afrikaners and Zulus are acting more like communal contenders than ethnonationalists. 8 Cameroon is an interesting variant on the theme. Although Cameroon adopted a colonial language (French), this was not a 'neutral' language because Cameroon includes territory which had been administered by Britain, and hence has English as 'its' colonial language. So the imposition of French is seen as (a kind of) dominant-group nation-building, and the anglophone minority has demanded autonomy and federalism (and perhaps secession) as a response to this dominant-group nation-building. Cameroon is the exception which proves the rule about colonial languages being neutral amongst African ethnic groups. 9 Somalia is another example of majority nation-building, but it has virtually no ethnic or linguistic minorities. So majority nation-building has not led to minority nation-destroying, although this ethnolinguistic homogeneity has not prevented violent civil war along factional lines. 10 There are often other reasons, beyond fear of dominant-group nation-building, underlying demands for territorial autonomy. In some cases, a minority's demand is related to the control of natural resources in its homeland, which the central government is exploiting without benefit to the local group. The Ogoni in Nigeria became more enthusiastic about autonomy when it was discovered that they possessed large amounts of oil from which they received little benefit. But this too is familiar from the West: Scottish support for autonomy went up when oil was discovered in the North Sea. In both the West and in Africa, demands for federalism reflect a mixture of economic interests and ethnocultural fears.

68

Will Kymlicka

justifications for doing so - i.e., that national minorities are likely to be disloyal; that the minority is backward and uncivilized and so needs to be brought into the modern world; and that the minority's territory contains land and resources needed for the country's economic development. This mixture of security concerns, paternalism and desire for resources permeates state relations with national minorities all over the world. As in the West, this suppression has had limited success in Africa, and many people are looking for an alternative approach. Western models of multination federalism may be relevant here. Many people think that federalism was (and remains) the only viable solution in Sudan, and, whatever the flaws of federalism in Nigeria, many people think that any other approach would have been worse. 11 However, federalism only works where boundaries can be drawn in such a way that the national minority forms a local majority (i.e., where the ethnic groups have relatively distinct homelands, within which they form a clear majority). In some cases, ethnic groups are simply too inter-mixed to create subunits where the minority form a clear majority. And even where there are minority-dominant regions, there will always be 'internal minorities', i.e., smaller groups located within the minority's self-governing homeland. This means that where federalism is adopted, there arises the question of the rights of internal minorities. While federalism in Nigeria helps to protect the Ibo and Yoruba from Hausa domination, 'it did little to assuage the fears of nonIslamic minorities in the north, non-Yoruba in the west, and non-Ibo in the east' (Lemarchand, 1997: 103). While few groups in Africa seek a country of their own, many groups seek a sub-state of their own, in which their language and culture are dominant. Therefore, what we often see is dominant-group nationbuilding occurring at a sub-state levels which excludes non-members of the locally dominant group from access to state power within that particular subunit of the country. 12 This shows that federalism is never enough to accommodate diversity; it must be supplemented with non-territorialized forms of protection of civil, political, social and cultural rights. This is clear from the Western experience, where territorial autonomy is always supplemented with firm constitutional guarantees of basic rights. The particular mix of territorialized and non-territorial minority rights varies in each Western federation, and one would expect similar variations in Africa. This, then, is a second pattern of nation-building and minority rights in Africa. Where state nation-building is based on the language, culture or religion of the dominant ethnic group, territorially concentrated minorities may mobilize as ethnonationalists to gain some form of autonomy and self-government (which in turn leads to demands by internal minorities for protection of their rights from self-governing minorities). Multination federalism is one possible outcome of these struggles. 13 11

In these cases, federalism is required not simply to accommodate the minority, but also the dominant group's aspirations. Federalism might not be needed in Sudan if the northerners were willing to abandon the promotion of Arabic language and Islamic culture in state institutions, and instead adopt a more pan-ethnic conception of Sudanese national identity grounded in the colonial language and a multireligious culture. But there is no evidence that the northerners are willing to do so, and so there is no option but autonomy (or secession) for the southerners. 12 On the privileging of 'indigeneity' on 'sons of the soil' in the sub-units of the Nigerian federation, see Bach, 1997; Ejobowah, 1998. 13 Which of these two patterns is preferable? It might seem preferable to seek a more pan-ethnic state which is not inherently biased towards the dominant ethnic group. This avoids one source of ethnic

Nation-Building & Minority Rights

69

The two patterns of nation-building and minority rights I have discussed are oversimplified ideal types, and a comprehensive survey would find intermediary cases as well as other distinct forms of nation-building. 14 But enough has been said to draw some general conclusions. The main lesson from my point of view is that we must examine minority rights in the context of, and as a response to, state nation-building. It is the nature of state nation-building which determines what sorts of threats are faced by minorities, and hence what sorts of rights they demand in response. That is equally true, I believe, of Africa and of the West. In the two cases of nation-building I have examined, I have suggested that there is a possible Western model that might be relevant: where nation-building is pan-ethnic, Western-style consociationalism might be appropriate; where nation-building is shaped by a dominant ethnic group, then Western-style multination federalism might be appropriate. But these are just vague models, not concrete proposals. No one would propose transplanting any particular form of consociationalism or federalism from the West to Africa, given their different economic, political, historical and demographic circumstances. A thorough comparison of the West and Africa would have to consider other forms of ethnic conflict. For example, I have not addressed issues of migration, either of economic migrants (for example, from Mozambique to South Africa) or of refugees (for example, from Rwanda into neighbouring states). These large-scale movements of people often generate conflict. Are there any models in the West for these cases? One comparison for the economic migrants in South Africa would be the guest-workers in northern Europe or illegal immigrants in southern Europe. At ( 13 contd) conflict, and diminishes the danger of separatism. But the pan-ethnic model avoids the alienation of minority groups from the state by alienating the masses of all groups. By defining the public sphere in terms of the colonial language, it widens the gulf between the state and the masses of all ethnic groups. The use of colonial languages has helped the horizontal integration of elites across ethnic lines, but has not enabled the vertical integration of government and the masses. The majority-group nation-building model, by contrast, helps link the masses of the dominant group with the state, but at the price of excluding both elite and masses of minority groups (unless supplemented by strong minority rights). Mazrui argues that the optimal solution is the use of African (rather than colonial) lingua francas - eg., Swahili in Kenya or Tanzania, which he argues will enable horizontal integration across ethnic groups while improving vertical integration of government and the people. But the Swahili example may be difficult to generalize in Africa. Most regional lingua francas are not seen as neutral by all the ethnic groups. Swahili was an acceptable lingua franca in Kenya and Tanzania because the ethnic Swahili group (for whom Swahili is their historic mothertongue) is tiny: around 1 per cent of the population, and so other groups did not feel threatened by adopting Swahili as a lingua franca. But other regional lingua francas (e.g., Hausa; Fulfulde) are the mother-tongue of powerful groups, and so the increased use of these lingua francas by the state is seen as a threat by other groups. Even in the case of Swahili, its use is controversial in Uganda, where it is seen by some as the regional language of the north, and hence biased against the peoples of the south (Mazrui and Mazrui, 1998: 131-2). Each form of nation-building, whether based on the colonial language, a regional lingua franca or the dominant group's mother-tongue, has its strengths in terms of either the horizontal or vertical integration of citizens, and its weaknesses in terms of potential exclusions, and each needs some form of minority rights to address these exclusions. 14 Consider Malawi. Banda's nation-building policies included regional favoritism for Chewi-dominated regions and a policy of official nationalism that attempted to give the Chewi language and culture official status in Malawi. This was resisted by the non-Chewi groups in Malawi, a fact which was reflected in the 1994 elections, and the new government has reversed many of these policies. However, Kaspin argues that Banda's nation-building policies were partly successful. While they failed in their effort to subordinate minority identities to an official Chewi-based national identity, they did succeed in focusing the attention of all groups on the national level. Minority groups may have disagreed with the attempt to imbue Malawian identity with Chewi content, but the response of these groups has been to politicize the issue of national identity (i.e., to become communal contenders), not to retreat from the national level into ethnonationalist separatism (Kaspin, 1997).

70

Will Kymlicka

the moment, many African countries seem to be treating these migrants the way European countries used to do: i.e., as metics, who are not granted citizenship or other rights in the hope that they will return home. Will African countries come to accept, as most European countries have, that such migrants are here to stay, and need to be integrated into their new country? 15 We could also compare the case of 'indigenous peoples' in the West and Africa. Some people deny that we can identify a subcategory of ethnic groups in Africa who are 'indigenous'. But if we define indigenous peoples, not as those who first possessed a particular piece of the earth (such that all non-European and non-Arabic groups are equally indigenous in Africa), but rather as a conquered people who 'live mainly in conformity with traditional social, economic and cultural customs that are sharply distinct from those of dominant groups' (Gurr, 1993), then perhaps we can identify some groups which are indigenous in this cultural sense. For example, Gurr argues that the San in Botswana and Namibia, the Kirdi in Cameroons, the Maasai in Kenya and the Tuareg in Niger all qualify as 'indigenous' peoples. It would be interesting to compare their treatment and claims with those of conquered/traditionalist indigenous peoples in the West, such as the Sami or Inuit.

Conclusion In this chapter, I have made some preliminary comparisons between minority rights issues in Africa and the West. I have argued that in both contexts we find a similar dynamic of minority rights being claimed by ethno-cultural groups in reaction to the nation-building policies of the state. To be sure, the nature of the groups and their claims often differs. But I believe that the underlying dialectic of nation-building and minority rights is similar. All parts of the world have been integrated into a system of modern nation-states; most modern states have embarked on programs of nation-building which are often directed at minorities; these nation-building programs are often perceived as a threat by minorities; minorities respond to this threat by demanding certain minority rights which set conditions or limitations on nation-building. This, I believe, is true of Europe, Africa, Asia and the Americas. States today differ from premodern or feudal multi-ethnic empires. In the past, multi-ethnic empires were often content to let minorities alone, so long as they paid their taxes or tributes, obeyed the laws, and co-existed peacefully with other ethnic groups. Today, however, few states around the world are content with this sort of co-existence. They want minorities to exhibit a stronger sense of loyalty to the state, so that they will actively co-operate in the projects of the state, be they militaristic wars, economic modernization, or social justice. And to gain the active support of citizens, states around the world have adopted nation-building programs which aim to turn co-citizens, bound only loosely to each other by certain common laws and taxes, into co-nationals, who share a strong bond in virtue of a common national identity and a common commitment to national projects. I believe that most African countries are nation-building in this sense. As we 15 There is no analogue in the West for the large-scale refugee movements displaced by civil war in East Africa. Western countries accept refugees, but in smaller numbers from countries far away. There is no comparable case of accepting tens or hundreds of thousands of people from a civil war in a neighboring country, sometimes involving ethnic groups which live on both sides of the border. The closest analogies in the European context are in the Balkans and Caucasus.

Nation-Building & Minority Rights

71

have seen, this nation-building takes different forms in different African countries, some of which have little in common with Western forms of nation-building. But each form of nation-building puts its own pressures on minorities, and it is only reasonable to expect that they will respond by seeking to modify these nation-building policies, to ensure that they are not disadvantaged by them. In my view, justice in multi-ethnic countries will always require some balancing of nation-building and minority rights.

References Bach, Daniel. 1997. 'Indigeneity, Ethnicity and Federalism', in Diamond, ed., Transition without End: Nigerian Politics and Civil Society under Babangida. Boulder, CO: Lynne Rienner. Bachler, Gunther, ed. 1997. Federalism Against Ethnicity. Zurich: Verlag Ruegger. Canovan, Margaret. 1996. Nationhood and Political Theory, Cheltenham: Edward Elgar. Connor, Walker. 1972. 'Nation-Building or Nation-Destroying', World Politics 24: 319-55. Ejobowah, John Boye. 1998. 'The Political Public and Difference: The Case of Nigeria', Ph.D. Thesis, Department of Political Science, University of Toronto. Eriksen, Thomas Hyland. 1999. 'A Non-ethnic State for Africa?', in Yeros. Gambari, Ibrahim and Martin Uhomoibhi. 1997. 'Self-Determination and Nation-Building in PostCold War Africa', in Wolfgang Danspeckgruber, ed., Self-Determination and Self-Administration. Boulder, CO: Lynne Reiner. Gasana, James. 1997. 'Factors of Ethnic Conflict in Rwanda and Instruments for a Durable Peace', in Bachler. Gurr, Ted. 1993. Minorities at Risk, Washington, DC: Institute of Peace Press. Kaspin, Deborah. 1997. 'Tribes, Regions and Nationalism in Democratic Malawi', in Ian Shapiro and Will Kymlicka, eds, Ethnicity and Group Rights. New York: New York University Press. Kymlicka, Will. 2001. Politics in the Vernacular: Nationalism, Multiculturalism, Citizenship. Oxford: Oxford University Press. Lemarchard, Rene. 1997. 'Ethnic Conflict Resolution in Contemporary Africa: Four Models in Search of Solutions', in Bachler. Markakis, John. 1999. 'Nationalism and Ethnicity in the Horn of Africa' in Yeros. Mazrui, Ali and Alamin Mazrui. 1998. The Power of Babel: Language and Governance in the African Experience, Oxford: James Currey; Chicago: University of Chicago Press. Okoh, Peter. 1997. 'The Nation-state and Ethnopolitical Conflict in Nigeria' in Bachler. Spinner, Jeff. 1994. The Boundaries of Citizenship, Baltimore, MD: Johns Hopkins University Press. Taylor, Charles. 1997. 'Nationalism and Modernity', in J. McMahan and R. McKim, eds, The Morality of Nationalism. New York: Oxford University Press. Walzer, Michael. 1983. Spheres of Justice. New York: Basic Books. Yeros, Paris, ed. 1999. Ethnicity and Nationalism in Africa. Basingstoke: Macmillan.

II

The Dynamics of Ethnic Development in Africa

5

JOHN LONSDALE

Moral & Political Argument in Kenya In the first two years of the civil war Kumase office holders recklessly squandered their material resources in factional struggle. And, simultaneously, through their excessive demands they liquidated their control over their subjects and society in general. (Tom McCaskie, JAH 25, 1984, p. 180)

Introduction

T

HE nineteenth-century model of the nation-state, steampowered, armour-plated, culturally homogeneous and monoglot, is today said to be in crisis. In the twentieth century, and in the northern world, it was still a ghostly imagination of moral community for which to die, for whose governance to vote and pay one's tax. Today many have seen it as a redundant, potentially spiteful, middleman, caught between opposing forces without and within. From without, global capitalist corporations demand to cut their local, social, costs - those welfare services and workers' rights that many electorates still expect their states to defend. From within, the divisive politics of identity has arisen to articulate resentment of such brute market forces and the inability of nation-states to soften them. National politics can thus seem to be increasingly devoid of a public sphere, an argued sense of the common good - and governments powerless to better the lives of citizens. For a state to protect its people against hard bargains, dressed up as market disciplines, is to risk investors seeking easier profits elsewhere. On the other hand, without effective states nations can soon fall apart, with more local loyalties gripping each other by the throat. Yugoslavia is but one example. Before 11 September 2001 many commentators thought the demise of the nation-state might be something to be desired. Regional confederation might better match market power and social responsibility. Since then, more would say that competent nation-states - not merely well policed but also attentive to This paper has benefited from comments made by participants at the Queen's University conference, by my colleagues Richard Waller, Maia Green, Frangois Grignon, Mottie Tamarkin, Jacqueline Klopp & Tom Wolf, & by two graduate students at Egerton University, Kenya: Babere Kerata Chacha & Kibet Ngetich.

73

74

John Lonsdale

citizens - remain essential supports for an international, rather than a global, borderless, order. Both before and after c 9/l 1' African states have been analyzed more critically than their counterparts elsewhere in the world. Before that date it was agreed that they were in a worse pickle than most. The continent had seen little of the steam-powered, parade-ground, or schoolroom state-building that had created nation-states in the northern world. Colonial rule had been culturally timid compared with the social engineering Europeans had endured themselves. In Europe the railway, rifle drill, and the three Rs had fostered cultural majorities, within which political trust can emerge. Without such historical aid, 'national' politics has tended to be merely a tumult of aggrieved minorities. It was scarcely surprising, therefore, that some African countries failed or collapsed as nation-states - if these be defined by a will and capacity to govern all citizens equally, subject to a blindfold rule of law. Many regimes discriminated between citizens on grounds of ethnicity, race, or creed. In extreme cases they ruled by means of civil war. Some thought this a rational enough response to the seductions of a concession-hunting capitalism with mafioso leanings. 1 September 2001, however, made these failings a source of alarm. Global terror can hide in the ruins of failed states like Afghanistan; it has become globally important that African states should work. Western powers might have done little more than shrug their shoulders at Robert Mugabe's theft of the Zimbabwean election, had the World Trade Center still stood; they had done little enough to avert the Rwanda genocide of 1994. Indeed, Africa's crisis surely owes more to the indifference than the self-interest of the northern world and its business corporations. Africa is locked into world capitalism. With less than 2 per cent of global trade, and with its export prices dropping by 60 per cent in the last twenty years, Africa is also in practice locked out. For lack of bargaining power, Africans pay the costs of market inequality, and enjoy few of its benefits. 2 The popular Western view ignores such market logic and blames Africa's sorry condition on its own failings and on these alone. On this view, its states are drained of revenue by official theft and deprived of political trust by the private sale of supposedly public services. More fundamentally, for lack of critical public opinions, Africa's archaic tribalisms can only engage in fruitless, factional, strife. State kleptocracy - power's private grip on the public purse - and ethnicity are said, indeed, to be natural partners, with ethnic calculation substituting for honour among the salaried thieves. In the Western view these afflictions have local roots. Africa, it is said, was not ready for the independence conceded nearly half a century ago. Africans lacked the common histories that are (erroneously) thought necessary for nationhood; they had no national languages, no sense of the public good. Foredoomed to destroy imported, prefabricated, statehoods by resort to narcissistic cultures of ethnic survival, Africans are now left standing by the wider world. Few legal businesses risk long-term investment in Africa; a quick, clandestine and, if necessary, violent buck will often seem the more prudent choice. Europeans, Americans, and many Asians, by contrast, are learning to enjoy ever more widely networked knowledge and contractual trust between strangers.3 1 C. Clapham, Africa and the International System: The Politics of State Survival. Cambridge, 1996; J.-F. Bayart, S. Ellis and B. Hibou, The Criminalization of the State in Africa. Oxford and Bloomington, IN, 1999; Chabal and J.-P. Daloz, Africa Works: Disorder as Political Instrument. Oxford and Bloomington, IN, 1999. 2 P. Collier, 'Africa and the Study of Economies', in R. H. Bates, V. Y. Mudimbe and Jean O'Barr, eds, Africa and the Disciplines: The Contributions of Research in Africa to the Social Sciences and Humanities. Chicago and London, 1993, pp. 58-82. 3 For trust see Bruce Berman's chapter in this volume.

Moral & Political Argument in Kenya

75

This chapter questions the core premise behind this 'Afropessimism'., the assumption that African citizenries cannot exist. I leave aside the key question of how good governance may be measured and secured in any small country locked into yet out of global trade. No political theory yet exists to test the accountability of international capital in its local relations, much less any legal process to enforce it. 4 One is obliged therefore, with some analytical frustration, to focus only on the thought and practice of Africans whose agency is confined to the lowliest corners of global power 5 - yet whose self-rule is said, without irony, to lack responsibility. The dual premise of this essay is, then, a paradox. It is that, despite their relative lack of power, Kenyans - like other Africans engage in critical and comparative, neither petty nor generally derivative, political thought; and that their imaginations of ethnicity, too often destructive, can nonetheless be among their most fruitful sources of nationally active citizenship. Kenyans hold stern views on public accountability and its moral tests. But their thought and practice are contradictory. They can be ethnically honourable and tribally factional. They can be paternally authoritarian and democratically plural, theologically liberal and fundamentalist. The question arises, whether these diverse politics will prove in retrospect to have been the struggles of a nation-in-formation, debating how to control an arbitrary state. That is my prognosis, hesitant as it must be, impossible to prove or disprove other than by a history still in motion. Since 11 September there is greater virtue to be claimed for the possibility - one cannot yet in Kenya call it a project - of such nationhood. While, in this third millennium, we live in a world exhausted by the nation-building wars of the second, no other known political community has a greater chance of controlling a state and its relations with capital than a militant nation, however multi-ethnic. Nor is any institution better able to foster political community than an intendedly national state. Yugoslavia and Afghanistan remind us of the alternatives, as does the warlordism that flourishes in the Congo, Liberia, Somalia, and elsewhere in Africa.6 To substantiate the thesis that their currently contradictory political culture may be contributing to a nation-in-formation I investigate Kenyan arguments on two intertwined issues. The first is the seemingly but perhaps deceptively divisive matter of their imagined ethnicities, and the second, their potentially unifying desire for accountability in governance. Two important analyses of African politics have cast doubt on the possibility of thus probing popular thought. In 1992 Chabal urged scholars to 'devise ways of analysing low politics' by decoding the languages of Africa's civil society. It was not enough to grasp the thought of formal institutions; one must also listen to the ethnic and other groupings in Africa's 'underbelly'. He did not, however, take up his own challenge, perhaps accepting Bayart's view that it was not worth the effort. The latter had questioned whether one could prove that the different voices that anthropologists discerned within ethnic groups possessed an 'ideo-logic' of their own, one that could oppose the 'only one word' or 'virtual totality' which Africans attributed to power.7 Bayart's analysis of the 'belly' of African power had many subtleties but could not answer his own question. The image of rhizome-like power, 4

C. N. Murphy, 'Global Governance: Poorly Done and Poorly Understood', International Affairs, 76, 4, 2000, pp. 789-803. 5 J. Lonsdale, 'Agency in Tight Corners: Narrative and Initiative in African History', Journal of African Cultural Studies, 13, 1, 2000, pp. 5-16. 6 W. Reno, Warlord Politics and African States. Boulder, CO and London, 1998. 7 P. Chabal, Power in Africa: an essay in political interpretation. New York, 1992, pp. 93-4; J.-F. Bayart, Uetat en Afrique: la politique du ventre. Paris, 1989, pp. 74-5, 69.

76

John Lonsdale

intricately rooted in mutually accommodating elites; the insistence that ethnicity is only one of several living identities; the suggestion that ethnic chauvinism is often the product of political faction - all were illuminating insights. But, like others, 8 Bayart discussed politicized ethnicity too exclusively from an instrumental or top-down angle, at the cost of the view from below. This chapter does not so much turn political science's ethnic telescope upside down as look through it from both ends. My argument is historical, with an eye to long processes. It is based on two political distinctions and the dialectical tension between them. These tensions argue against any analysis that separates power from its 'underbelly', or the thoughts and practices of dominant from 'subaltern' classes. All power relationships are represented and enacted within a polemical arena. 9 Power may control the public address system, but crowds also have a voice, generally a murmur, at times a roar. 10 In order to listen to both forms of speech I distinguish, first, between what I call 'moral ethnicity' and 'political tribalism', and then between high, low, and deep politics. These distinctions require discussion before my narrative of Kenyan political thought can suggest the potentially productive tensions that have developed between them. Moral ethnicity is a process of 'ourselves-ing', rather neglected in Africanist literature. Political tribalism is one of 'othering', often the political analyst's staple fare. Neither constructed facet of regional identity can be termed ethnic nationalism. Few African ethnic groups have demanded sovereignty. Moral ethnicity nonetheless arises out of internal discourses of social responsibility comparable, in all but their lack of demand for a state of their own, to those of European nationalism. 11 In colonial times these polemical processes wrought socio-moral communities out of formerly more fractured, less bounded, societies. The chief stimulus to creative debate was the need to establish anew how people should behave among 'us, ourselves', when previous norms of inequality were shocked by social class formation and religious change, both given definition by literacy. Such vernacular patriotic argument was and is potentially a popular defence against exploitation by patrons or rulers. On the other hand, political tribalism or political ethnicity determines how 'we' behave in relation to 'others' in the arena of a multi-ethnic state. Ethnic rivalry for a share in public resources has been the most reliable protector of ruling elites against the emergence of a nationally critical public. The tension between these two sides of ethnic politics - morally questioning within, at times politically ruthless without - informs my periodization of modern Kenya's political history. The second distinction lies between high, low, and deep politics. High politics is what princes and politicians handle, the bargains that share out the commodity 'power' under rules that deter competitors from overplaying their hand and endangering the state. For unruly power ceases to exist; it slips into the street, where the low politics of survival, if equally unrestrained, can only be a war of all against all. What disciplines both and may even make them fruitful are the deep connections between patronage on high and clientage below. These are bound by contracts of obligation, differently interpreted on either side, and daily renegoti8

L. Vail, ed., The Creation of Tribalism in Southern Africa. London, Berkeley and Los Angeles, 1989. J. Lonsdale, 'Political Accountability in African History, in P. Chabal, ed., Political Domination in Africa: Reflections on the Limits of Power. Cambridge, 1986, pp. 126-57. 10 A. Haugerud, The Culture of Politics in Modern Kenya. Cambridge, 1995. 11 'Comparison' describes the relationship between African ethnic and European nationalist thought more accurately than 'derivation'. African territorial nationalisms were more directly derivative. See the critique of nationalisms in P. Chatterjee, Nationalist Thought and the Colonial World: A Derivative Discourse. Minneapolis and London, 1986. 9

Moral & Political Argument in Kenya

77

ated in a dance that tests the tolerances of the other. The possibilities of fundamental change inherent in this unstable relationship are illustrated in McCaskie's account of crisis in the precolonial West African kingdom of Asante, quoted above. The monarchy's high-political faction fights unwisely laid upon its subjects demands too heavy for the deep conventions of unequal reciprocity between high and low politics to sustain. In such situations ordinary people who cannot normally act in their own cause may find they have no option but to fend for themselves, being unable any longer to trust in the powerful to protect their lives and livelihoods for them. There is nothing new in this observation of how, contingently, high and low politics can interact through their deep relations. It is a staple recourse of European historiography in explaining rapid, or revolutionary, political change. Its insights can well be applied to Africa, as the Asante example shows. When, in Kenya, elites have overplayed their hand, their fault is interpreted from below, in the main, as an excess of greedy political tribalism (or racism, in the colonial past) offensive to idealized moral economies of productive ethnic patronage. In Kenya moral ethnicity predated political tribalism. Ethnicity is a universal cradle of civility. It socializes human inequalities in local ways. It naturalizes unequal societies of disciplined, productive, moral 'straightness' and barren fecklessness.12 Quotidian ethnicity is thus always argued between rich and poor. Its stern norms invite tension within households, between generations, classes, genders, and localities. Normal routines are often resented; unequal reciprocities are questioned. Cultural festivals often bear a double meaning, celebratory and subversive. Saturnalia are not confined to Rome, nor carnival to Romans. 13 And this argued nature of ethnicity is at the core of my view that Kenyans, and Africans more generally, possess the cultural means with which to resist the 'totalizing' efforts of their states. 14 However, this universally lived ethnicity, in all its local forms, has no necessary connection with political mobilization. For ethnicities to become political counters they must not only be negotiated from within, but also contrasted with other ways of living and linked to inequalities of power. They must be looked to for group solidarities beyond those needed to perform the chores of daily life. Such externally alert ethnicity was weakly developed in precolonial Kenya. 'Tribes' could not then compete with each other, whether in 'tribal war', trade or production, since tribes did not then exist, in the sense of self-contained, externally bounded, collective groups. Working social relations were either smaller or more extended than the term tribe suggests. Interdependent ethnicities - or ways of managing and celebrating productive livelihoods - got along with each other. Warfare was as likely within a culture group as between them. Each ethnicity might have its own tongue; its own institutions; and distinct territory, often suited to one way of living rather than to another - farming or herding or hunting. But vernaculars could be closely related; similar institutions, especially those that marked age and generation, were widely diffused or borrowed; ethnic regions could support mixed economies; specialized subsistence goods encouraged exchange as much as war. There were few concentrations of power; each ethnic group was internally divided by locality and clan. People lived an ethnicity rather than rationalized it, having little need to defend its ways against outside oppres12 J. Lonsdale, 'Authority, Gender and Violence: The War within Mau Mau's Fight for Freedom', in Atieno Odhiambo and J. Lonsdale eds, Mau Mau and Nationhood. Oxford, 2003, ch. 3. 13 E. Le Roy Ladurie, Le Carnaval de Romans. Paris, 1979. 14 I thus disagree with the argument in A. Mbembe, 'Provisional Notes on the Postcolony,' Africa, 62, 1992, pp. 3-37.

78

John Lonsdale

sion. It was the culture in which they earned their living, discharged social obligations, resolved disputes, earned reputation and trust as self-realized men or women - hommes oufemmes accompli/es.15 People evinced pride in their subsistence civilization and disdained other ways of taming nature. Cultural competition energized precolonial Africa.16 But prudence set limits to cultural chauvinism. In an uncertain environment different specialist producers - hunters, herders, farmers sought reinsurance among neighbouring others whom they might also see as enemies. Cultures were permeable. Big men took wives from ethnic 'others'. The poor might take to inter-ethnic migration in search of alternative protection. Marriage and trade, clientage at time of famine - such necessities required ethnic frontiers to be porous, facilitating canny cross-frontier co-operation.17 Ethnic consciousness unquestionably became sharper in colonial times. Alien rule, economic change and literacy profoundly affected social relations. Local power became concentrated, unprecedentedly, in the hands of official chiefs and their kin; migrant labour subverted household disciplines and generational relations; women's market production called in question existing gender relations. Religious change challenged the elders' monopoly on access to unseen power; literacy gave a scandalous new assurance to the young, to women, or the poor. Many smelt social decay, and urged the need to renovate relationships. Kenyatta summed up best a litany of ritual amnesia, social disorder, and moral failure. 'Religious rites and hallowed traditions are no longer observed by the whole community. Moral rules are broken with impunity, for in place of unified tribal morality there is now . . . a welter of disturbing influences, rules and sanctions, whose net result is that a Gikuyu does not know what he may or may not, ought or ought not, to do or believe, but which leaves him in no doubt at all about having broken the original morality of his people.' 18 Such disorder can be intimately apprehended only within local, ethnic, arenas - each with an imaginable social history, an easily idealized moral economy of unequal reciprocity. A community that mythifies diverse pasts into a common history, to face up to the present, could be one definition of an ethnic group. Only within story-telling groups can one know the moral alienation that comes from confronting as strutting superiors or sullen subordinates those with whom one shares, or can imagine, a co-operative history of achievement. Class formation may make such associations more consciously national or ethnic than before, as mutual obligations are redefined from above and below. It was to catch this sense of societal renegotiation of what one 'ought or ought not, to do or believe' in relation to kin and neighbours, patrons and clients, that I coined the term 'moral ethnicity'. 19 It does not imply any inherent virtue in ethnicity. Our socialization necessarily makes ethnic, and other, identities for us all. I 15

Y. Droz, Migrations Kikuyu: des Pratiques Sociales a ITmaginaire. Neuchatel and Paris, 1999. I. Kopytoff, ed., The African Frontier: The Reproduction of Traditional African Societies. Bloomington and Indianapolis, 1987. 17 R. D. Waller, 'Ecology, Migration, and Expansion in East Africa', African Affairs, 84, 1985, pp. 347-70; C. H. Ambler, Kenyan Communities in the Age of Imperialism. New Haven, C T and London, 1988. 18 J. Kenyatta, Facing Mount Kenya. London, 1938, p. 251. 19 J. Lonsdale, 'The Moral Economy of Mau Mau', in: B. Berman and J. Lonsdale Unhappy Valley: Conflict in Kenya and Africa. London and Athens, OH, 1992, pp. 265-504; idem, 'Moral Ethnicity and Political Tribalism', in P. Kaarsholm and J. Hultin, eds, Inventions and Boundaries: Historical and Anthropological Approaches to the Study of Ethnicity and Nationalism. Roskilde, Denmark, 1994, pp. 131-50; idem, '"Listen while I Read': the Orality of Christian Literacy in the Young Kenyatta's Making of the Kikuyu', in L. de la Gorgendiere et al. eds, Ethnicity in Africa: Roots, Meanings and Implications. Edinburgh, 1996, pp. 7-53; idem, 'Moral Ethnicity, Ethnic Nationalism and Political 16

Moral & Political Argument in Kenya

79

intended to signal only that an imagined common ethnicity makes an urgent case for people aware of emerging class divisions to devise new agreements to act socially^ or morally, among themselves. Again, there is an argued, negotiated aspect to African political culture that analyses devoted to the hegemony and theatre of the state tend to neglect. Moral ethnicities were in part stimulated by literacy in standardized vernaculars and access to biblical allegory.20 The new technology and imagery of words fostered ethnic consciousness among larger populations than would have previously thought themselves related. Literacy also projected identity outwards, giving small groups their place in world history. Many of Kenya's 'tribal histories' now begin with a migration from misri or 'Egypt' 21 - as the Yoruba of Nigeria look to Mecca, or Ethiopians to Israel. Some Kikuyu believe they must be one of Israel's lost tribes. 22 The politics of ethnicity is not generally a search for refuge, nor obsessed with 'difference'; it is more often a claim for equivalence, for recognition of an equal right to exercise that global reach once exclusively claimed, in the African case, by white colonizers. But literacy also underpinned new inequalities and a new distancing of power. At the heart of Kenya's debates on moral ethnicity there has been a tension between old moralities of household reputation, generated within laborious economies of unequal obligation, and a new, potentially impersonal, politics of literary skill and managerial responsibility. In the 1920s such friction once fired generational struggle.23 In the 1950s it split Mau Mau, as will be seen. In the 1960s it lay beneath the surface of the quarrels within the Kenya African Union (KANU) and then between KANU and its successive opponents, the Kenya African Democratic Union (KADU) and the Kenya People's Union (KPU). 24 At the turn of the twenty-first century it is at the core of conflict between opposition claims to democracy and meritocracy, and the prevailing political system of patrimonial faction and dynastic alliance. The social context of these debates has, however, changed. In earlier times there was silence over the status of the very poor and, therefore, little imagination of an equal citizenry. Today, as in nineteenth-century Asante, there is some, if fleeting, evidence that the contempt shown by patrons for old norms of reciprocity with clients has shattered the latter's acceptance of gross social inequality and opened the way for new perceptions of citizenry. Political tribalism has become the obverse side of moral ethnicity. It governs external relations with other perceived ethnic groups, now visualized as bounded constituencies or 'tribes'. Its seedbed was those twentieth-century changes that militated against the inter-ethnicity of former times - the alien imposition of a state that fostered new regional inequalities of power and its more urgent factionalization; the diversion of much farm production from local exchange to the export (19 contd) Tribalism: the Case of the Kikuyu', in P. Meyns, ed., Staat und Gesellschaft in Afrika: Erosions- und Reformprozesse. Hamburg, 1996, pp. 93-106; B. J. Berman, 'Ethnicity, Patronage and the African State: The Politics of Uncivil Nationalism', African Affairs, 97 1998, pp. 305-41. 20 A. Hastings, The Construction of Nationhood: Ethnicity, Religion and Nationalism. Cambridge, 1997. 21 A. M. M'Imanyara, The Restatement of Bantu Origin and Meru History. Nairobi, 1992. 22 Droz, Migrations Kikuyu, p. 343; A Heyer, 'The Mandala of a Market: a Study of Capitalism and the State in Murang'a District, Kenya.' PhD thesis, London University, 1998, p. 291. 23 Lonsdale, "'Listen while I Read."' 24 For a KANU radical's complaint against men of education, see B. Kaggia, Roots of Freedom, 1921-1963. Nairobi, 1975, p. 182. KANU's greater comfort with bureaucratic arguments, by contrast with KADU, was evident in 1962: see 'Kenya Regional Boundaries Commission: Record of the Oral Representations', Nairobi, mimeograph, 1962, especially the evidence of Christopher Kiprotich and Mwai Kibaki, 14 August 1962, pp. 60, 62-4. Evidence for the KANU-KPU differences in this respect is provided later.

80

John Lonsdale

trade; inter-ethnic competition in labour markets; demographic pressure on the choicest land. Inter-ethnicity remains, however, a precondition for the parallel, informal, economy in which so many Africans, with great ingenuity, not only hide from the greedy eye of the state but also negotiate a subsistence entitlement with 'others'. The main spur to political tribalism has, by contrast, been the search for protective solidarity in face of political uncertainty and inequity within Africa's new states. Clients may gain as much from obedience as patrons from mastery. There have been two phases of political unease in modern, statist, Africa. The colonial period has cast a long shadow over the post-colonial. Africans could not predict European likes and dislikes; in post-colonial times, all the levers of wealth and power, however weakened by market manipulation, have been held by a state whose strength lies in its distributive discretion in response to factional interests. It would be hard to say which has been the more unpredictable, the alien colonial or the indigenous post-colonial regime. A long history of mistrust marks the gulf of opportunity between those the state favours and those it excludes. Mutual suspicion governs political practice. It puts a premium on local electoral solidarity, fearful of dissent within an ethnic constituency. In a patrimonial high politics qualified, as in Kenya, by electoral competition, faction alone cannot determine who shares in the private rental income of the public estate. Ethnic champions must stake their followers' solidarity in the auction room of power for their bids to count. Deep questions of accountability from below are better unheard. Political tribalism thus represents a contradiction of the face-to-face struggle for reputation by patrons in the field of moral ethnicity. It calls for unquestioning reliance on a representative until he falls from favour - and a willingness to accept that the choice of local hero rests as much on arbitrary high-political deals as on conditional trust or pressure from below. This politics has little patience with principled debate. Power no longer springs from a reputation earned by managing a productive household whose fruits are, with strategic hospitality, deployed within a knowing politics of local obligation. Power today, shockingly, is often seen to derive from what would have been, and to some extent still is, regarded as occult power, sorcery. For the state's patronage resources have no visible social origin; they have not accumulated sociable debts; thus unencumbered, they cannot be other than potentially anti-social in effect.25 At the level of high politics, political tribalism is in consequence the politics of bought and brittle loyalties, of factional intrigue, of 'bribe and tribe'. 26 In short, there is a fundamental tension within Kenya's political culture. This brings us to the relations between high and low politics, and back to a truism of European historiography in the explanation of rebellion or revolution. This holds that the low politics of survival has most readily discarded the deep politics of subservience, to ascend to the plane of independent action, at moments of high-political crisis. Crisis forces rulers to break customary reciprocities, to demand of their subjects too much in blood or treasure, for too little and uncertain a reward. The prudential reluctance of political elites to dig too far into their supporters' reserves of loyalty is thrown to the winds when, typically, they 25

R. Austen, 'The Moral Economy of Witchcraft: an Essay in Comparative History', in J. and J. Comaroff eds, Modernity and its Malcontents: Ritual and Power in Postcolonial Africa. Chicago, 1993, pp. 89-110; P. Geschiere, The Modernity of Witchcraft: Politics and the Occult in Postcolonial Africa. Charlottesville and London, 1997. For fears of occult power in modern Kenya see Droz, Migrations Kikuyu, pp. 331-90; Heyer, 'Mandala', ch. 15. 26 D. W. Throup and C. Hornsby, Multi-Party Politics in Kenya. Oxford, Nairobi and Athens, OH 1998. J. D. Barkan, 'Kenya: Lessons from a Flawed Election', in F. M. Hayward, ed., Elections in Independent Africa. Boulder, CO, 1993, pp. 213-38.

Moral & Political Argument in Kenya

81

have fallen out violently between themselves, or face ruinous external war. At such a moment, to take an example from the English revolution, Parliament's New Model Army gave birth to the monstrously radical Diggers and Levellers in the course of its soldiers' debates in 1647, on London's Putney Common.

Argument It is time to address general discussion to the Kenyan point at issue: the tension between the country's deep ethnic cultures and its state's high-political practice. A question hangs over this chapter. It is answerable only by future history. Past history, now to be outlined, will, however, exert an influence. The question is, might Kenya's national high politics so alienate its deep ethnic supports that political crisis could bring fundamental change? Could ethnic electorates pool their local critiques of power, encouraged by the inter-ethnicity that often underwrites survival among the poor, to build a common citizenship against the prejudices inflamed by political tribalism? In the absence of the socio-economic conditions for liberal or social democracy, can the only indigenous sources of change, namely, outraged moral ethnicities, unite? Could Kenya experience a radicalization of its politics such as occurred in England in the 1640s, France in the 1780s, or Asante a century later? Might there come a time when Kenya's 'ghosts without name' take to Nairobi's streets, not simply to loot, as in the recent past, but with the same political effect as others did in Michelet's revolutionary Paris? 27 Such an eventuality may appear unlikely. To contemplate it, however, is to open up the prospect of a dynamic relationship between local moral ethnicities and national accountability. 28 The prevailing approach has little room for this possibility. It grants too perennial a primacy to the divisive, top-down, calculus of political tribalism. 29 It distinguishes too sharply between urban citizenries and rural subjects in allegedly 'bifurcated' ex-colonial states. 30 My enquiry into the possibility of political movement has four narrative episodes. Each shows how divisions within ethnic and national politics have, in the recent past, affected the other. Local and national are not segregated, mutually exclusive, bifurcated realms of power or argument. A century of social change, that has reshaped but not destroyed old inter-ethnicities of survival, and has extended rural ethnicities into the labour markets of town, means that politics cannot be so divided. Nor, in the realm of political thought, can there be 'only one word'. Reason of state may dictate high-political tribalism; the low imperatives of survival and deep critiques of patronage still promote moral ethnicity, not only within groups but also between them. Norms of inter-ethnicity draw on the symbiotic subsistence ethic of a stateless past. They are best remembered by Kenya's women, who regularly cross masculine 'tribal' boundaries in marriage and in trade. 31 People cannot ignore the needs of strangers, 27 B. Fontana, 'Democracy and the French Revolution', in J. Dunn, ed., Democracy, the Unfinished Journey, 508BC to AD 1993. Oxford, 1992, p. 113. 28 J. Klopp, 'Electoral Despotism in Kenya: Land, Patronage, and Resistance in the Multi-Party Context.' PhD thesis, McGill University, 2000. 29 D. W. Throup, 'The Construction and Destruction of the Kenyatta State', in M. G. Schatzberg ed., The Political Economy of Kenya, New York, Westport, C T and London, 1987, pp. 33-74; Throup and Hornsby, Multi-Party Politics. 30 As in M. Mamdani, Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton, NJ, Kampala, Cape Town and London, 1996. 31 Heyer, 'Mandala'; L. S. White, The Comforts of Home: Prostitution in Colonial Nairobi. Chicago and London, 1990; C. C. Robertson, Trouble Showed the Way: Women, Men and Trade in the Nairobi Area

82

John Lonsdale

however much high-political faction encourages them to do so. Political tribalism has sanctioned bloody ethnic cleansings in recent years, a horrendously logical conclusion to the divisive high politics that for a century has sustained a 'weak state', colonial and post-colonial. 32 Moral ethnicity, however, confronts this logic, sustained by knowledge that most ethnic groups have extended diaspora in rural tenancies or urban slums, vulnerable to the vigilantism of historically more native others. The two political idioms have been constantly at odds. On that point Kenya's modern history is clear. Political tribalism has hitherto held the trumps. But in any high-political crisis the rules of the game are broken; anything can happen. In my first episode of modern Kenyan history, in the late 1940s, the Mau Mau movement opposed both the moderate, elite, 'Kenyan' nationalism of the Kenya African Union and the local ethnic conservatisms to which the KAU also gave voice. There was no bifurcation in politics here, ethnic from national, but a dynamic argument between them. Secondly, Mau Mau itself combined both class and ethnic militancy. From the perspective of moral ethnicity (not of vulgar Marxist sociology) there was every reason why the two should unite. Mau Mau constituted the 'deep' repudiation of clientage; many poor people felt already betrayed by their patrons, white and black. But the movement's military wing then consumed itself in its own 'Putney debate', conducted in the forests of central Kenya. Here the parochial politics of reputation confronted the managerial pride of literacy, a perennial problem for moral ethnicity in a peasant economy. This internecine conflict within an ethnic radicalism made it possible, thirdly, for Kenyatta, as President of Kenya, to disarm 'the left' at the national level. He was free to pursue the quasi-feudal politics of a conservative, and capitalist, ethnic accommodation. In his high-political hands a generalized moral ethnicity became the ideology of the state. Until his last, more arbitrary, years, this was the most productive, progressive, period of Kenya's modern history. Finally, the recent disenfranchisement of Kikuyu and Luo, Kenya's two largest ethnic groups, has helped to stir up a strange, unstable, brew of liberal and pentecostal visions. 33 The one has a national, high-political, purchase, the latter more often local. Informed by contradictory theologies, both have opposed Moi's 'Nyayo' model of a tutelary democracy, which supposedly follows in Kenyatta's 'footsteps'. But I end with the question to which this past history of alternating ethnicity and nationality seems to lead: whether a new generalization of critical moral ethnicity, coming now from below in deep resistance to the latest phase of selfish patronage, may not provide a new energy of nation-building, forcing a new high politics on the state. Each of these four eras has been rich in principled controversy; the failures of one have warned the next. There are now many Kenyan memories of conflict between freedom and order. 34 Kenyans are far from silenced by the 'unanism', ( 31 contd) 1890-1990. Bloomington and Indianapolis, 1997; D. Cohen, and E. S. Atieno-Odhiambo, Burying 'SM': The Politics of Knowledge and the Sociology of Power in Africa. Portsmouth, N H and London, 1992; W. W. Otieno, Mau Mau's Daughter: A Life History. Boulder, CO and London, 1998. 32 R. Jackson, 'Violent Internal Conflict and the African State: Towards a Theoretical Framework.' Paper presented to the African Studies Association of the UK biennial conference, Cambridge, 2000. 33 It remains to be seen how far the dynastic dispositions being made for the Moi succession will end the disenfranchisement of Kikuyu and Luo. 34 R. Buijtenhuijs, Mau Mau Twenty Years After: the Myth and the Survivors. The Hague and Paris, 1973; G. Prunier, 'Mythes et histoire: les interpretations du mouvement Mau Mau 1952-1986', Revue frangaise d'outremer, 74, 277, 1987, pp. 409-29; E. S. Atieno Odhiambo, 'The Production of

Moral & Political Argument in Kenya

83

the consensual blanket, with which Africa's rulers try to stuff the mouths of dissenting, democratic, thought. 35 The dynamic relationship between ethnic and national politics has, to the contrary, kept them on their toes. Nonetheless, the peoples of Kenya have only begun to extend to the national level, in any organized way, their ethnic expectations of unequal but reciprocal social contracts. Too often disabused of their trust by patrons more reliant on the state than upon their constituents, too often re-divided by the factionalism of opposition party politics, it is not surprising that Kenyans have only now begun, haltingly, to act as a common citizenry, for the first time since the 1960s. 36 What may look like a 'beginning', however, may simply be another 'episode'. Fed and armed by the state, political tribalism can all too easily bite back. Kenyans are not alone in these contradictions. Nowhere in the world is the dialectic between nationality and ethnicity, citizenship and clientage, likely to make an end to its turbulent history. KA U's Decade: Liberals, Conservatives, and Radicals The failures of the Kenya African Union (KAU) have inspired the classics of Kenyan political historiography. Founded on hopes of post-war colonial reform in 1944, the KAU was banned as an alleged figleaf for terror nine years later. Its president, Jomo Kenyatta, was convicted by a bought judge, on perjured evidence, on a charge of managing the Mau Mau insurgency. 37 Here I want to focus on what existing accounts have tended to neglect, the divisions in KAU's thought on the relations between authority and action. 38 These spelt trouble from the start. They promoted paralysis in face of white settler intransigence and British indecision, opening the way for Mau Mau. The KAU's first leadership was provided by a pan-ethnic cadre of educated liberals. Many were old boys of the leading missionary boarding school, the Alliance High School. Theirs, it has been said, was a nationalism of the school tennis court. 39 Their politics was based on two principles. First, educational privilege required of them a duty to their people. This call resembled more the duty of prodigal son to parent than ( 34 contd) History in Kenya: the Mau Mau Debate', Canadian Journal of African Studies 25, 2, 1991, pp. 300-7; J. Lonsdale, 'The Political Culture of Kenya', in B. Frederiksen et al. eds, Political Culture, Local Government and Local Institutions. Roskilde, Denmark, 1993, pp. 87-107; Haugerud, Culture of Politics; G. Sabar-Friedman, 'The Mau Mau Myth. Kenyan Political Discourse in Search of Democracy,' Cahiers d'etudes africaines 137, 1995, pp. 101-31; M. S. Clough, Mau Mau Memoirs: History, Memory & Politics. Boulder, CO and London, 1998; G. G. Githiga, The Church as the Bulwark against Authoritarianism: Development of Church-State Relations in Kenya with Particular Reference to the Years after Political Independence, 1963-1992. Nairobi, 2002. 35 P. J. Hountondji, African Philosophy: Myth and Reality. London, 1983; M. G. Schatzberg, Political Legitimacy in Middle Africa: Father, Family and Food. Bloomington and Indianapolis, 2001, seems to me to give too little room for the variety of African political thought. 36 Klopp, 'Electoral Despotism'. 37 C. G. Rosberg and J. Nottingham, The Myth of (Mau Mau': Nationalism in Kenya. New York and London 1966; J. Spencer, KAU: The Kenya African Union. London and Boston, MA, 1985; D. W. Throup, Economic and Social Origins of Mau Mau. London, Nairobi and Athens, OH, 1987; B. J. Berman, Control & Crisis in Colonial Kenya: The Dialectic of Domination. London, Nairobi, and Athens, OH, 1990, ch 7; E. S. Atieno Odhiambo, 'The Formative Years, 1945-55,' in B. A. Ogot and W. R. Ochieng, eds, Decolonization and Independence in Kenya, 1940-93. London, 1995, pp. 25-47; J. Lonsdale, 'Kenyatta's Trials: the Breaking and Making of an African Nationalist', in P. Coss, ed., The Moral World of the Law. Cambridge, 2000, pp. 196-239. 38 I here summarise J. Lonsdale, 'KAU's Cultures: Imaginations of Community and Constructions of Leadership in Kenya after the Second World War', Journal of African Cultural Studies, 13, 1, 2000, pp. 107-24. 39 B. E. Kipkorir, 'The Alliance High School and the Origins of the Kenya African Elite, 1926-1962'. PhD thesis, Cambridge University, 1969, p. 244.

84

John Lonsdale

of patron to client, a scripturalization of the canons of moral ethnicity. Was not their school motto 'strong to serve'? Secondly, the Whig imperial history they derived from the classroom had taught that power must ineluctably devolve on to those who by learning and property were best qualified to respect liberty. They accepted the hegemony of the British civilizing mission and its corollary: that the educated must lead by example; they even proposed adult education as a remedy for urban crime. They expected the British conscience, informed by reasoned argument rather than jolted by popular pressure, to promote reform. Their heirs indeed feared the tribalism of uneducated mass action. 40 Scarcely democrats, therefore, neither did the liberals draw strength from moral ethnicity. Their main spokesman was Tom Mbotela, son of freed slaves, translator of Tom Paine, vice-president of KAU. For his denunciation of savage custom, for his support of the British co-optation of respectable elites, if also on suspicion of informing the police, Mau Mau did him to death in 1952. 41 Dispirited tennis-court liberals - whom the South African ANC's youth league would have called 'gentlemen with clean hands' 4 2 - gave place to sterner ethnic conservatives. Of these the Kikuyu Kenyatta and Luo Oginga Odinga were the leaders. They were impatient not only with British gradualism but also with the spineless lack of ambition in their constituents that betrayed the laborious traditions of imagined ethnic history. For these conservatives the civilizing mission could never be British. It had to be African. No nation, they said, in effect, could liberate another. 43 The basis of conservative ethnic politics was moral self-criticism. If one's people would only show more solidarity, more commercial vigour, more educational effort, less male crime or less female vagrancy in town, then the British, in the deep logic of productive patronage, would be forced to give Africans their just deserts. Odinga, a gumbooted elder, was as critical of Luo indiscipline and lack of enterprise as Kenyatta was of Kikuyu. 44 Ethnic conservatives, landed elders, men of straight reputation and patrons of compassion (andu a tha in Gikuyu), were in turn opposed by radical men of violence. This next conflict was confined almost entirely to Kikuyu. Such ethnic singularity was scarcely surprising. Of all Kenya's peoples Kikuyu were the most divided by the opportunities and oppressions of British rule. Conservative elders, Kenyatta among them, rested their reputation on property, proof that they could bear responsibility.45 Accountable to a censorious ancestry buried in their land, the gentry could be trusted not to follow policies that risked more pain than gain. But there was little they could legitimately do; they could not offer to protect the land of others without denying its owners their own ances40 Makerere Kikuyu Embu Meru Students' Association. 'Comment on Corfield.' Makerere University, Kampala, mimeograph, 1960. 41 J. E. Harris, Repatriates and Refugees in a Colonial Society: The Case of Kenya. Washington, D C , 1987, pp. 123-47. 42 A. M. Lembede, 'Congress Youth League Manifesto' (1944), in T. Karis & G. M. Carter, eds, From Protest to Challenge: A Documentary History of African Politics in South Africa 1882-1964, vol. 2. Stanford, CA, 1973, pp. 300-08. 43 Cf., A. M. Lembede, 'Some Basic Principles of African Nationalism' (1945) in Ibid., pp. 314-16; idem, 'Policy of the Congress Youth League' (1946). in Ibid., pp. 317-18. 44 Oginga Odinga, Not yet Uhuru. London, 1967, ch. 5; E. S. Atieno Odhiambo, '"Seek ye First the Economic Kingdom": a History of the Luo Thrift and Trading Corporation 1945-1956'. in B. H. Ogot, ed. Hadith 5: Economic and Social History of East Africa. Nairobi, 1975, pp. 218-56; H. M. Muoria, ed., Guka kwa Njamba iitu Nene. Nairobi, 1946; idem, ed. Kenyatta ni Muigwithania Witu. Nairobi, 1947; idem. I, The Gikuyu and The White Fury. Nairobi, 1994, ch. 4. 45 G. Kershaw, Mau Mau from Below. Oxford, Nairobi and Athens, OH, 1997, ch. 6.

Moral & Political Argument in Kenya

85

tral duty. A concerted ethnic nationalism - let alone a pan-ethnic one - was thwarted by the precautionary principles of moral economy. Kenyatta was clear about that on his return from Britain in 1946: CI have not come to order you about, telling you to do this and that.' 46 Nor had the discourse of moral ethnicity imagined an equal citizenry for Kikuyu - let alone Kenyans. Kenyatta complained that the one mark of citizenship Kikuyu possessed, the age-graded comity of circumcision, made them too proud to obey anybody but themselves.47 But young and land-poor men could not afford to stand on propertied principle. They lacked the resources of land and employment on which to marry and confirm their adulthood. By the tenets of organic, ethnic, political thought they were not entitled to wield responsibility. But they were angry, and most of Nairobi with them. They blamed their poverty on patrons who had turned against them. Kikuyu lineage seniors, the gentry, and white farmers were in this respect alike. In pursuit of agricultural efficiency, the former had expunged junior claims on lineage property, the latter the customary rights of labour tenants to hoe and graze on white-owned land. The poor needed action, to remove the British and reopen access to property. These were the men of Mau Mau. Their movement, many-headed, parochial, secretive, had no single strategic direction. It was the outcome of unresolved conflict within Kikuyu moral discourse, between dynastic rights of property, now abetted by literacy, and generational hopes of moral renewal, to which literacy was also allied. 48 In the course of the struggle those with natural authority, the ethnic conservatives, lost their hold over popular sentiment to their radical opponents. But the contradiction between the clannish authority of property and the need for concerted action was not easily overcome. Lacking in authority, but needing loyalty and secrecy, the radicals were obliged - in a move of questionable Kikuyu legality - to take control of the customary oaths of initiation and anti-witchcraft solidarity that were formerly a gentry monopoly. 49 Mau Mau, then, challenged the disciplines of a moral ethnicity that, in the insurgent view, was itself compromised when patronage, reputed to be generous, closed its dynastic fist around property. But Mau Mau and British counterinsurgency also sharpened political tribalism. The colonial regime had no trouble in recruiting troops from other peoples to reinforce the Kikuyu 'loyalist' Home Guard in the Mau Mau war. Many African leaders accepted the British tribalist argument that Mau Mau wished to eject the colonialists only to impose a Kikuyu ethnocracy thereafter. But some also believed that nocturnal Kikuyu oaths breached ethnic morality and smelt of sorcery: 'we do not know . . . what they may do at night.' 50 One cannot otherwise understand why so many Kenyans argued for a 'regional', or majimbo> rather than unitary state at independence ten years later. 51 Majimbo was designed to defend smaller ethnic groups against domination by Kikuyu agrarian wealth, allied to the proletarian 46

Muoria, I, the Gikuyu and the White Fury, p. 18. Muoria, Kenyatta ni Muigwithania Witu, p. 15. 48 J. Lonsdale, 'The Prayers of Waiyaki: Political Uses of the Kikuyu Past', in D. M. Anderson, D. M., and D. H. Johnson, eds, Revealing Prophets: Prophecy in Eastern African History. London, Nairobi, Kampala, and Athens, OH, 1995, pp. 40-91. 49 Kershaw, Mau Mau from Below, ch. 7. 50 Evidence of William Murgor, Daniel arap Moi, Morris Rajula, in Kenya Regional Boundaries Commission: Oral Representations, pp. 150 (for the quote), 152, 157. 51 B. A. Ogot, 'The Decisive Years, 1956-63', in: B. A. Ogot and W. R. Ochieng', eds, Decolonization and Independence in Kenya, 1940-93. London, Nairobi and Athens, OH, 1995, pp. 48-79. 47

86

John Lonsdale

interests of Luo migrant labour. 52 The regionalists were able to press their case for majimbo because overwhelming British force in the field, backed by 'compelling force' in the detention camps, 53 had meanwhile helped to defeat Mau Mau. Mau Mau's subversion of the KAU, then, had come about because an ethnic conservatism, itself suspicious of liberal nationalism, was felt by many Kikuyu to have been betrayed from within by the mutation of productive ethnic patronage into greedy private property. The ethnic regionalism of majimbo to a large extent represented the return of the conservative strand in KAU. But Mau Mau, too, had been weakened from within. This internal Mau Mau conflict is a critical episode in illuminating Kenya's criss-cross ethno-national dialectic of political accountability. Mau Mau's 'Putney Debate' In the forest war Mau Mau leaders argued out their own legitimacy even as they fought the British. 54 The insurgents devoted almost as much energy to conference as to action against the regime. Mau Mau continued the debates of Kikuyu moral ethnicity, now heated by experience of a war that made old gender norms unworkable and the burdens of leadership intolerable. 55 Sex and gender are problems for any army. Far fewer women than men were actively engaged in the forest war. Mau Mau generals tried to cool the fires of male sexual jealousy first by sexual segregation; then, when that failed, by strictly regulated cohabitation; and a ban on sexual competition between officers and men. They were soon accused of flouting their own rules. Field-Marshal Sir Dedan Kimathi, though himself to enjoy too many women, was betrayed to the British by a man he had sentenced to be flogged for presuming, as a common soldier, to sleep with but one. 56 This conflict over gender was only one aspect of a deeper forest struggle over authority that is not inappropriate to call Mau Mau's 'Putney debate'. Mau Mau fighting units ('gangs') never acknowledged a unified high-political command. They nonetheless split in two over the proper source of deep authority: education or household reputation. Kikuyu, and others, had been disputing their respective merits for at least thirty years. In the forest, the educated cause was centred in the 'Kenya Parliament' that supported Kimathi. Their opponents rallied to his rival, Stanley Mathenge, under the collective name of 'Kenya riigiy. Riigi was the wattle door that fastened a household enclosure, symbol of civilization, against the animal and spiritual dangers of the wild. One cannot imagine a more morally charged name with which to epitomize the ideals of Kikuyu leadership. Proverbially, 'nobody else can close the door of another 52 F. Grignon, 'La Democratisation au Risque du debat? Territoires de la Critique et Imaginaires Politiques au Kenya (1990-1995)', in D-C. Martin, ed., Nouveaux Langages du Politique en Afrique Orientale. Paris and Nairobi, 1998, pp. 29-112. 53 C. M. Elkins, 'Detention and Rehabilitation during the Mau Mau Emergency: The Crisis of Late Colonial Kenya', PhD thesis, Harvard University, 2001. 54 For further discussion, see Lonsdale, 'Authority. Gender and Violence'. 55 Sources: D. L. Barnett, and K. Njama, Mau Mau from Within. New York and London, 1966; W. Itote, 'Mau Mau3 General. Nairobi, 1967; H. K. Wachanga, The Swords ofKirinyaga. Nairobi, 1975; M. wa Kinyatti, ed., Kenya's Freedom Struggle: The Kimathi Papers. London, 1987. Previous discussion: L. S. White, 'Separating the Men from the Boys: Constructions of Gender, Sexuality, and Terrorism in Central Kenya, 1939-1959,' International Journal of African Historical Studies, 23, 1990, pp. 1-25; Lonsdale, 'Moral Economy', pp. 456-7; idem, 'Moral Ethnicity and Political Tribalism', pp. 142-50; Clough, Mau Mau Memoirs, pp. 163-8). 56 I. Henderson, with P. Goodhart, The Hunt for Dedan Kimathi. London, 1958, p. 67. This white evidence is substantiated by the Kikuyu sources cited above, and others.

Moral & Political Argument in Kenya

87

man's hut'. 5 7 Riigi stood, therefore, for the parochial politics of household reputation against the wider impersonality of management. Management was morally insolent; it presumed to close the doors of other men's huts. Against this charge of arrogance, literates defended leadership as a burden of unselfish, creative, responsibility - in the same terms as Presidents Kenyatta (below) and Moi have used since. 58 Without wise patronage, they argued, the energies of subordinates lacked order and direction, and could bear no fruit. Kimathi's men also pleaded that policy and ideology were a sounder base for organization than personality, less vulnerable to mortality. They might just as well have spat into the wind. Riigi men stuck to their view that for the educated to claim overall command was to deprive fighters of the right to test the integrity of their own leaders in the face-to-face cockpit of the every day. On their side Parliament men regarded such politics of reputation with the same disdain as Plato looked on democracy - as a shameless auction of easy popularity that fostered an egalitarianism, an unbridled low politics, fatal to military discipline. According to their riigi critics, Kimathi's supporters, like all educated men, were also cowards, too impressed by white power. This was scarcely fair. There were many educated forest fighters; and they were right to respond that in modern times the pen was mightier than the spear. Riigi men were also readier than Parliament to try to negotiate surrender terms with the British. The riigi attributed supposed literate cowardice to a further failing, a missioninspired alienation from ethnic culture, and contempt for the deep demands of moral ethnicity on personal honour. It was not a riposte that Parliament men thought to use but in fact Kikuyu writers had done much to memorize a culture that oral practice was beginning to forget. Still more seriously, riigi fighters accused literates of exploiting illiterate guerrillas as 'merely stone walls' for the protection of elite ambition. Forest fighters wished to level these, to end the 'slavery' of performing their leaders' domestic chores. Diggers as well as levellers, they were also adamant that no Mau Mau general should become 'their master who would divide unto them their lands', after they had driven out the white landlords. 59 Land must go to the tiller: the squatters who had cultivated the 'white' lands and who formed a major element in riigi ranks. To protect the workers' interests, riigi generals insisted, there could be no single party within Mau Mau such as Parliament desired. Within a single party, critics rarely found others brave enough to support them against the leadership. The argument even reached into the realms of theology. A Parliamentary spokesman defended a hierarchy of command with the axiom 'that there was no equality of persons on this earth'. He was answered with the humanist dictum, unknown to either traditional Kikuyu or imported missionary teaching, that 'man was the master of this earth and he could make changes to suit his desires'. 60 I find it remarkable that such fundamental political debate should be pursued under such stress, not between an 'internal' and 'external' wing as in other liberation movements but between the 'internals' themselves. Here was indige57

G. Barra, 1000 Kikuyu Proverbs. Nyeri, 1939, proverb no. 782. Atieno Odhiambo tells me Luo say the same. 58 D. T. arap Moi, Kenya African Nationalism: Nyayo Philosophy and Principles. London and Basingstoke, 1986, pp. 76-106. 59 Barnett and Njama, Mau Mau from Within, p. 402. 60 Ibid., pp. 397-8, 406-9.

88

John Lonsdale

nous dispute over the virtues and failings of a national or bureaucratic bourgeoisie and its single-party dictatorship - entirely without benefit of Frantz Fanon, years before the appearance of Les damnes de la terre. The Mau Mau forest orators had, however, nothing explicit to say on Kenya's multi-ethnic character. They ignored it. Parliament and riigi both assumed that the sweat of Kikuyu labour tenants in the 'White Highlands' and the blood of Mau Mau entitled Kikuyu to inherit land to which other groups such as Maasai and Kalenjin held the longer historical title. But their debate on accountability, characteristic of moral ethnicity, is surely a necessary, if scarcely sufficient, condition for inter-ethnic relations to be treated as an issue for public debate. The debate would no doubt be hot, centring on the peasant proviso of the African frontier, that ethnic entitlement to land should be based not on equal human rights but on the unequal sweat of collective contributions to agrarian civilization. The Kenyan 'left' has tended to dismiss the possibility of a reasoned ethnic politics altogether, as an outdated mirage of 'false consciousness'. 61 It is true that practising politicians have used political tribalism as a resource to exploit, rather than inter-ethnicity as a basis for negotiation. Hitherto, these practitioners of power have had the best of the argument. But their rivals, the 'radical nationalists', rather misleadingly called the 'left', had lost the day soon after independence, weakened by the same divisions that had hastened the end of Mau Mau. They were in no position to oppose the ethnic conservatism of Kenyatta, released from a martyr's prison in 1961 and anointed to unite the nation. His politics had not altered during his eight-year absence. Under him, he promised, there would be no 'hooligan' government. Most Kenyans found that reassuring; the household disciplines that gave the young freedom, but not licence, would be restored. 62 In any case, the 'African Socialism' of his radical opponents was not very different from his own. More populist than socialist, the Kenya People's Union, founded by the radicals in 1966 and destroyed by Kenyatta three years later, was solicitous of smallholder property and the rights of tribe and clan. Riigi-like in its anger at the greed of big men, the KPU was equally uncomfortable with central management - without which socialism is defenceless against capitalist accumulation. 63 The KPU, like the riigi or Mau Mau in general, owed more to an outraged moral ethnicity than to Marx. And it was to a restored moral ethnicity that Kenyatta looked in building the nation. Kenyatta's Ethnic Feudalism Intellectually attuned to the functional anthropology that for him had hallowed, if not invented, Kikuyu 'custom' between the wars, Kenyatta had long held the view that there was no worse delinquency than 'detribalization'. He blamed the religious pluralism of modernity for breaking up households and an imagined tribal unity. In his ethnographic work, Facing Mount Kenya, a resistance work of cultural nationalism, he had in 1938 urged on the British an educational policy 61 M. wa Kinyatti, 'Mau Mau: The Peak of African Political Organisation in Colonial Kenya', Kenya Historical Review, 5, 2, 1977, pp. 287-310; reprinted, with other essays, in idem, Mau Mau: A Revolution Betrayed. New York and London, 1992, 2000. 62 J. Lonsdale, 'Mau Maus of the Mind: Making Mau Mau and Remaking Kenya', Journal of African History, 31, 1990, pp. 393-421. 63 KPU. Wananchi Declaration: The Programme of the Kenya People's Union. Nairobi, 1967; C. Gertzel, The Politics of Independent Kenya. Nairobi and London, 1970, pp. 84-8; G. Lamb, Peasant Politics. Lewes, 1974, pp. 32-7.

Moral & Political Argument in Kenya

89

that would, instead, maintain 'tribal solidarity in social services', to secure 'the continuity of [a tribe's] distinctive features over against other tribes'. With his belief in 'a tribal organisation of the spirit world', he was convinced that ethnic pride was a precondition of liberation. 'It is the culture which he inherits that gives a man his human dignity as well as his material prosperity. It teaches him his mental and moral values and makes him feel it worth his while to work and fight for liberty.' 64 After sixteen years of voluntary exile in Britain Kenyatta had returned to Kenya in 1946, his views unchanged and, at the age of fifty, unlikely to do so. He became the KAU's most eloquent ethnic conservative. Addressing Kamba supporters, he had said of inter-ethnic co-operation: 'we need to decorate ourselves, . . . we need to respect each other and then speak in one voice'. 65 The KAU, it followed, must be a congress-like umbrella over equal ethnic organizations, not a pan-ethnic vanguard trampling on local effort. All the evidence suggests that he struggled to keep the KAU's executive multi-ethnic in character, failing only in face of the determination of Kikuyu Mau Mau militants to take it over themselves. Kenyatta's multi-culturalism was, it seems, rooted in respect for the human need to exercise agency, 'the spirit of manhood' as he called it, that which closed its own riigi. Against the British charge, in court, that he had mesmerized the masses with his oratory he replied, with an anger impossible to fake, 'you think those people have no intelligence? They would listen to any rubbish if I spoke rubbish?' 66 On his release in 1961 he united ethnicity and class in this language of esteem: 'We should respect other tribes and not think of the Kikuyu as the big men.' 6 7 Everything Kenyatta said in public, as Prime Minister and then as President, echoed the sentiments expressed in the days of 'hopelessness', as he later remembered them, when he was a mere agitator. 68 He became the more true to this younger self as he grew into power and faced publicly, as if in open baraza, an opposition that served to sharpen his wit. 69 Kenyatta's most insistent theme was the need to be free to govern oneself, 'to enjoy the fruits of initiative and self-respect'.70 It gave a dignity greater than wealth. 71 Freedom, as if embodying a straight elder or patron, released people's energies. 72 It allowed one to practise the disciplines of tradition, so different from the idle prostitution of loafing in Nairobi. 73 As if to affirm the riigi's deep critique of managerial insolence, he urged Kenyans to solve their own problems and not come running to him. 74 The unemployed should return to the land; 'it has made us what we are'. 75 Kenya was its people; it was a nation by virtue of their self-reliance, a moral fibre they owed to ethnic tradition. 76 'Our people look straight, not down64 J. Kenyatta, Facing Mount Kenya: The Tribal Life of the Gikuyu. London, 1938, pp. 120, 267, 317; J. Lonsdale, 'Jomo Kenyatta, God, and the Modern World', in J-G. Deutsch, P. Probst and H. Schmidt, eds, African Modernities. Oxford, 2002, pp. 1-56. 65 Muoria, Kenyatta ni Muigwithania Witu, pp. 21-2, emphasis added. 66 M T T , 3 March 1959, p. 1253. 67 J. Kenyatta, Suffering Without Bitterness: the Founding of the Kenya Nation. Nairobi, 1968, p. 153. 6S Ibid., p. 313. 69 cf. Haugerud, Culture of Politics. 70 Kenyatta, Suffering, p. 328. 71 Ibid., pp. 341, 347. 72 Ibid., pp. 285, 247, 294. 73 Ibid., pp. 330, 345, 347. 74 Ibid., p. 346. 75 Ibid., pp. 232-4, 282, 291, 329. 76 Ibid., p. 327.

90

John Lonsdale

wards as in colonial times.' 77 The nation was 'a kind of United Nations in miniature'. 'Positive tribal feeling' fostered enterprise. 'Disruptive and negative tribalism', on the other hand, was the fruit of idleness, tempted by socialism's 'fairyland' economics and the foreign bribery of little leaders, 'pitiful victims of flattery, allied to purchase'. 78 To hand out 'free things' in the form of socialist goods was - as if, now, to echo Kimathi's parliament - 'a cowardly way of trying to win popularity'. 79 It also misjudged the sturdy independence of Kenya's people. 80 Politicians, Kenyatta said, must respect this quality - as he himself had done years back, first in rejecting the 'insolence' of a proffered Comintern patronage in the 1930s, 81 then in defending his followers' intelligence in court. Elites must obey 'a sterner moral integrity'; they must show compassion; and respect a man 'for what he is and how he strives'. 82 It would be difficult to find a more direct definition of the deep contract of patronage. This moral contract underpinned what has been aptly called his 'ethnic feudalism'. 83 Nobody would suggest that President Kenyatta lived up to what he preached. He did not. Kenyans knew that. Many, especially non-Kikuyu, saw him as a dictator who refastened the shackles of colonialism - whose imprint his own ankles carried to the grave. Deep politics was not deceived. But Kenyatta thought words, and these particular words, were needed to woo it. He was trying to create 'the only one word' or 'virtual totality' that is the goal in any struggle for political mastery. It was the voice of the elder, the voice of order that at the same time granted discretion to its juniors' energetic freedom. Kenyatta tried to claim, in the always risky public theatre of power, legitimacy for his rule. He conjured up, from ethnic cultures of laborious self-respect and unequal merit, and in the contingent course of a campaign to destroy 'the left', the common principles of nationhood. Kenya, he nearly said, was a nation precisely because its people were not detribalized. They retained their moral drive to achieve a culturally disciplined liberty, unsapped by idle, cosmopolitan, urban, vagrancy. He also argued that one-party rule was the modern means whereby 'many tribes [could act] in concert', as in the past when they coped with famine by exchange with neighbours. The single party also preserved - a point the riigi would have disputed - a traditional culture of 'constructive opposition from within'. 84 What Kenyatta well knew - he would not otherwise, in his seventies, have spoken so tirelessly - is that there is an inescapable dialectic between the need to claim legitimacy, as he did, and its accountability, tested in political practice. It is an ideological competition no polity can escape. In claiming that the state was the enabling patron of its citizens' energy, Kenyatta was giving a reasoned basis for his control, such as men had long justified, in their households, over wives.85 Any discourse of deep legitimacy sets its own test, however, by which its protagonists cannot avoid being judged. Arap Moi has made the same claim as Kenyatta, that unquestioned leadership - what political science used to call 77

Ibid., p. 295. Ibid., pp. 247, 313-4, 303. Ibid., p. 309. 80 Ibid., pp. 303-4. 81 A. M. Pegushev, 'The Unknown Jomo Kenyatta', Egerton Journal (Njoro, Kenya), 1, 1996, pp. 172-98. 82 Kenyatta, Suffering, p. 261. 83 G. Prunier, 'Jomo Kenyatta et son temps,' in F. Grignon and G. Prunier, eds, Le Kenya contemporain. Paris and Nairobi, 1998, p. 133. 84 Kenyatta, Suffering, pp. 229-30. 85 B. Brinkman, Kikuyu Gender Norms and Narratives. Leiden, 1996. 78

79

Moral & Political Argument in Kenya

91

'developmental dictatorship' - releases the client ambitions of its citizens. 86 But if the high politics of state patronage has failed, inevitably, to live up to its own claims, how has that failure been judged from below? In Kikuyu households many women would formerly have wondered if their bruises were incontestable evidence for their husbands' reason. And as early as 1972 a barefoot heckler, rejecting 'the only one word', shouted back at Kenyatta, in a public rally, 'It is all lies!' 87 NyayOy the Devil, and the Future To sympathize with Kenyatta's heckler one must epitomize in a paragraph political Kenya's last four decades. A number of points are clear. They suggest that Kenyatta's ethnic feudalism, with its unequal contract of vassalage underwritten by a normative discourse of moral ethnicity, in time gave way to a patrimonial absolutism that relied for loyalty on political tribalisms that competed for an abject dependence on the state. Democracy and citizenship were not part of the story, not even rhetorically so, but then democracy is an historically derived culture, not a portable set of political practices. Kenya's political culture is not democratic. It does, however, exhibit many of democracy's characteristics. It is privately grasping, publicly censorious, alert to the merest whisper of shifts in the bargains of power, resentful - perhaps envious - of privilege and its corruptions, but scarcely, as yet, fired and disciplined by the doctrine and practice of citizen equality.88 One must take in at a glance then - as Kenyans have watched it over time the degradation of Kenya's actual political culture, its acute sense of social contract, of reciprocity.89 Both the Kenyatta and Moi presidencies, rather like Kabakas' reigns in precolonial Buganda, began with a legitimizing increase in freedoms that were then suppressed. Evidence is plentiful. To start with, there is the manner in which the 'Harambee' habit of community funding of schools and other such improvements came to be appropriated by local big men. Their ever larger contributory cheques, for sums that soon demeaned the proceeds of collective effort, were drawn on one or another state slush fund; what had once shown 'deep' respect for constituents now proves only a dexterity in allying with the powerful. High politics, over time, has bought out low. A corollary, to take a second example, has been the increasing presidential use of the district administration and police to harass and even destroy awkward customers who refuse to be co-opted or bought. Here the disposal of J. M. Kariuki in 1975 still stands out. Other political murders have been less clearly provoked by political tribalism's fury at rebuke from moral ethnicity. Collusion, thirdly, between ruling party and state officials, barbed with violence in the new multiparty era, has also overturned Kenya's formerly 'upside-down' patronage system. Peasant electorates used to fancy themselves patrons, with politicians their clients. The system is now the right, that is, high, side up. A degree of 'order' has been 86

arap Moi, Kenya African Nationalism, R. Sklar, 'Democracy in Africa', African Studies Review, 26, 1983, pp. 11-24. 87 E. S. Atieno Odhiambo, 'Democracy and the Ideology of Order in Kenya', in M. Schatzberg, ed., The Political Economy of Kenya. Westport, CT, New York and London, 1987, p. 178. 88 Haugerud, The Culture of Politics; Grignon, 'La Democratisation au Risque du debat?' See also two magnificent studies of the 1992 and 1997 elections: Throup and Hornsby, Multi-Party Politics; and M. Rutten, A. Mazrui, and F. Grignon, eds, Out for the Count: The 1997 General Elections and Prospects for Democracy in Kenya. Kampala, 2001. Neither massive work makes more than a fleeting reference to what could be called 'policy'. 89 F. Grignon and H. Maupeu, 'Les Aleas du Contrat Social Kenyan', Politique africaine, 70, 1998, pp. 3-21.

92

John Lonsdale

imposed on the expectant vigour of independence. Party and state have tried to intimidate the arguments of deep politics into parroting 'the only one word'. 90 A further corollary has been the increasingly shameless equation between a district's voting record and its access to development spending. Under arap Moi, the shrinking patronage fund has also been topped up by a suicidal piracy of state resources, to pay off importunate allies or fund a political campaign. Recently there has been an appalling rise in the official theft of private or public land, generally at the expense of the poor, seized in order to gentrify new members of the political elite.91 Finally, violent ethnic cleansing - mainly of Kikuyu farmers from the former 'White Highlands' - has left many dead and a quarter of a million displaced. These are the depths to which an officially sanctioned, youthfully thuggish, political tribalism can descend, protected by a collusive silence in the law. It is the bottom of a downward spiral of competition that, when economic growth gave way to decline in the later 1970s, revealed the tribal bacillus that had lurked within a comparatively benign ethnic feudalism from the start. In the 1960s Kenyatta had urged politicians not to forget compassion but to respect a man 'for what he is and how he strives'. In the 1990s, the severest criticism of an M P who kept his district clientele too narrowly confined was that he was creating, even among ethnic kin, 'a constituency of beggars'. 92 He did not care how they strove. His behaviour repudiated any deep reciprocity between patron and client and, by extension, between state and citizen. The question for twenty-first century Kenya, as for nineteenth-century Asante, is whether, through private greed and dereliction of public duty, Kenya's rulers risk liquidating 'their control over their subjects and society in general', so making way for a popular redefinition of politics. Such radical change may, to repeat, seem unlikely. Hitherto there has always been a supply of aspirants grateful for a share of the shrinking national cake, eager to step into the mendicant shoes of any politician disgraced. Nor have the multi-party oppositions of recent years imagined any politics other than a mirror to the one they oppose. Mooted changes in the constitution have had little purpose other than to open up the high-political game. The question is why this should be so. It is not as if people do not know by whom they are oppressed. It is no longer the case, as perhaps was true of the first years of independence, that the taxation (or exploitation) of peasant farming is 'highly indirect, complex and impersonal', largely via the profits of the crop-marketing boards. 93 The constant passage of reshuffled elites through the revolving doors of parastatal office shows clearly who enjoys a presidential licence to 'harvest other men's shillings', being too idle to grow their own maize. Politicians have been rebuked for such thievery since the 1920s. 94 Few in Kenya now question the source of their misery. It is the regime. Even should their local champion enter government, the mass of the electorate can expect little reward, once they have feasted upon the electoral fund. But is it imaginable, is it desired, to concert a public opinion clamorous 90 Atieno-Odhiambo, 'Democracy and the Ideology of Order,' pp. 184-5; cf. Lonsdale, 'Political Accountability', p. 155. 91 K. Kanyinga, 'The Land Question in Kenya: Struggles, Accumulation & Changing Polities'. PhD thesis, Roskilde University, 1998; Klopp, 'Electoral Despotism'. 92 K. Kanyinga, 'The Changing Development Space in Kenya: Socio-Political Change and Voluntary Development Activities', in P. Gibbon, ed., Markets, Civil Society and Democracy in Kenya. Uppsala, 1995, p. 111. 93 C. Leys, 'Politics in Kenya: The Development of a Peasant Society', British Journal of Political Science, 1, 3, 1971, pp. 313-43. 94 Lonsdale, 'Moral Economy', p. 384.

Moral & Political Argument in Kenya

93

for an alternative deep politics, one of programme rather than personality? With Kenya's history, that is difficult to imagine. The electorate is all too easily divided by differential access to what, like many Africans, they call 'eating'. The regime has relied for control on dividing aspirant elites in a hungry contest for the patronage resources they hope to be able to claim for local development. 95 But Kenya has by now been pauperized by rising population, 96 a weak global market for tropical produce, and the fiscal drain of years of official burglary. The state has been driven to grab more and more private and public assets with which to benefice its supporters. The question is, will this increasingly barefaced political banditry unite the grievance of rejected suitors with deep outrage from a dispossessed public, to produce something more than a reshuffle of the court cards of high-political faction?97 Kenya is certainly abuzz with diverse political languages, voiced by a press that knows how to exploit the conditionalities of the Western 'donor community'. Tennis-court liberals - golf-club liberals nowadays - are more numerous and professionally confident than sixty years ago. The Law Society has a brave record. Some of its members have endured prison and torture in the cause of human rights, some have served as the intellectuals of multi-partyism. But arguments for an imported liberal democracy have thus far been aborted by the indigenous politics of ethnic reputation, its deep criteria still imprinted with the codes of different regional modes of production. Until his death in 1995, the sage Luo patron, Oginga Odinga, 98 was chief among the ethnic conservatives. Yet he could not bring himself to share a platform with the self-made, courageous but, in many eyes, too crafty Kikuyu businessman Kenneth Matiba, symbol nonetheless of how to rise without benefit of government favour. Then there were the watajiri, the Democratic Party's plutocrats, who split the Kikuyu against Matiba, worthy of trust because their wealth protected them from the temptation to pilfer the public purse. Their critics said they had no further need; they had done it earlier, under Kenyatta. Finally, there was the ruling party KANU, widely felt, for all its faults, to be the little people's sole defence against immizeration by the rich. It was also the party of the only order that, however oppressive, could until recently be relied upon to keep the country safe from the horrors of its neighbours, Somalia, the Sudan, Rwanda. Moi, his critics had to admit, seemed a canny herdsman of peacefully client cattle, a Kalenjin pastoralist to his fingertips. Kenya also teems with moral languages, theologies of citizenship on the one hand and of salvation on the other. The Trinitarian theology of God's incarnation in the daily sufferings of his children informs the political critique of the 'mainstream' Protestant and Catholic churches. These preach that Caesar should be rendered only such conditional obedience as is his due. 99 Their cler95

Kanyinga, 'Changing Development Space'. cf., J. Iliffe, Africans: the History of a Continent. Cambridge, 1995, pp. 243-6, 253, 269-70. 97 The organising theme of Klopp, 'Electoral Despotism'. 98 H. Odera Oruka, Oginga Odinga: His Philosophy and Beliefs. Nairobi, 1992. 99 H. Okullu, Church and Politics in East Africa. Nairobi, 1974; idem, Church and State in Nation Building and Human Development. Nairobi, 1984; N C C K (National Christian Council of Kenya), Why You Should Vote. Nairobi, 1991; T. M. Njoya, Human Dignity and National Identity: Essential for Social Ethics, Nairobi, 1987; D. Gitari, Let the Bishop Speak. Nairobi, 1988; idem, In Season and out of Season: Sermons to a Nation. Carlisle, 1996; G. P. Benson, 'Ideological Politics versus Biblical Hermeneutics: Kenya's Protestant Churches and the Nyayo State', in H. Hansen and M. Twaddle, eds, Religion and Politics in East Africa. London, Nairobi, Kampala and Athens, OH, 1995, pp. 177-99; M. Ngunyi, 'Religious Institutions and Political Liberalisation in Kenya', in P. Gibbon, ed., Markets, Civil Society and Democracy in Kenya. Uppsala, 1995, pp. 121-77; G. Sabar-Friedman, '"Politics" and "Power" in the Kenyan Public Discourse and Recent Events: the Church of the 96

94

John Lonsdale

ical critique of patrimonial absolutism often creates a stir, but there is reason to suppose that it may be becoming a minority voice among Kenya's Christians. Like other countries around the world, Kenya has witnessed a rapid growth in Pentecostal churches, thanks largely to a second missionization of Africa, since independence, funded by America's religious right. 100 Pentecostal, dualist, theology and its apocalyptic vision of warfare between good and evil sits only too easily with traditional fears of sorcery. Depending on their favoured or excluded ethnicity, some Pentecostalists see the state as a defence against the Devil, as in Romans 13. Others believe the President sups with Satan, with too short a spoon, and look for salvation to the Second Coming, not politics. 101 Biblical theology, the most universal of Kenya's public languages, is also a Babel, not a joint manifesto of reform. 102 The small size, divided opinion and, often, divisive high-political intrigue of the 'civil society' of lawyers, politicians, and clerics, is reason to argue that any future crisis of government in Kenya might be answered more effectively from below by a revived, and generalized, deep moral ethnicity. There is fragmentary evidence of that possibility. Protests against the state's theft of local assets such as forest reserves, market places, even schools, or its bulldozing of shack settlements to make way for property speculation, have begun to draw strength from each other. Some local activists even talk of 'the nation' being killed by its state. Remarkably, junior local politicians have been known to protest against the clearance of 'stranger' squatters from their area, to disencumber the grant of land to a local big man. Their protests have reconstituted the inter-ethnicity of the past. Of this inter-ethnicity perhaps the most extraordinary example, in the year 2000, was the private, multi-ethnic effort to feed the starving and ordinarily neglected Turkana of Kenya's northern frontier, in a collective effort that rejected all sponsorship by the state. As Kenyatta once did from on high, so today - and for the moment - from below, 'an imagined nation of communities' seems to be taking root in many angry hearts and minds. 103 But, to conclude, there can be no certainty. Democracy in Kenya is only a rulebook, periodically taught by the National Christian Council before elections. Neither liberal nor social democracy can yet be foreseen. Their cultures require economic prosperity and a 'horizontal' sense of trust between equal citizens, not yet found in Kenya. One wonders, indeed, if cultural chauvinism has not increased during the long years in which ethnic groups have looked more to the state than to their neighbours for survival in hard times. Kikuyu showed a startling disdain for the uncircumcized Luo during litigation, in 1987, over the burial of 'SM' Otieno. 104 All the same, Kenya's deepest culture of accountability remains the moral ethnicity of household production. With these contradic( " contd) Province of Kenya', Canadian Journal of African Studies, 29, 1996, pp. 429-54; Githiga, The Church as Bulwark. 100 J. Hearn, 'The Development Implications of American Evangelism in Kenya'. PhD thesis, University of Leeds, 1997. 101 Droz, Migrations Kikuyu, pp. 364-77; J. Lonsdale, 'Kikuyu Christianities: a History of Intimate Diversity', in D. J. Maxwell, ed. with I. Lawrie, The African Roots of African Christianity. Leiden, 2002, pp. 187-9. 102 A recent, excellent, survey of ecclesiastical politics is H. Maupeu, 'The Churches and the Polls', in Rutten et al. Out for the Count, pp. 50-71. 103 Examples from Jackie Klopp, personal communications, and from idem, 'Electoral Despotism', quote from p. 252. 104 J. B. Ojwang and J. N. K. Mugambi, eds, The S. M. Otieno Case. Nairobi, 1989; D. Cohen and E. S. Atieno-Odhiambo, Burying 'SM': The Politics of Knowledge and the Sociology of Power in Africa, Portsmouth, N H and London, 1992; Otieno, Mau Mau's Daughter.

Moral & Political Argument in Kenya

95

tory observations in mind one must end by asking two troubling questions about relations between freedom and the state in Kenya. Is it possible, first, for the growing number of landless poor to be deemed to be honourably 'straight', with a legitimate stake in the future, when they have no hope of bearing responsibility for a propertied household? Might moral ethnicity, in other words, no longer work on behalf of even the deserving poor? Yet secondly, is it possible for even the propertied to exercise self-mastery when one's own ethnic spokesmen are dependent on a state in hock to the Devil of monetary greed? Can personal honour continue to tolerate political tribalism? 105 Moving on to possible answers, could the anger of outcast Nairobi force a new deep politics rather than merely burn itself out in the low politics of looting? Could its landless people be accepted, appeased, and incorporated as full citizens of £a nation of communities', or does that suggestion itself beg two different answers? Fifty years ago slum Nairobi pushed open the closed colonial question of political change. In those days its force and coherence lay in an intimidating ethnicity. Could its opposite, an urban, cosmopolitan, citizenry carry a similar weight nowadays? Or would more reason be found to reform the deep politics of state power on the basis of an inter-ethnic communalism that demanded more respect for the strivings of the poor? And in response to either intimation of citizenship, does not the dividing power of political tribalism still hold the trumps? The votes of the poor, after all, are cheaply bought at election times, even if a low peasant cynicism may defend their independence of spirit. Such questions can only be answered by history. Past history has wound its way between one such answer and another. It will probably continue to do so. Moral ethnicity may still represent Kenya's sternest culture of personal accountability, but political tribalism is well practised in fiddling the public accounts. 105

H. Odera Oruka, 'Why the Future Will Judge Us Harshly', Daily Nation (Nairobi) 11 December 1994.

6

D I C K S O N EYOH

Contesting Local Citizenship: Liberalization & the Politics of Difference in Cameroon

T

territories inherited from colonialism into stable development-sustaining nation-states remains a defining aspiration of African nationalism. For post-independence political rulers and elites, nation-building called for the use of the state to nurture national identity at the expense of particularistic identities. This state-driven strategy of nation-building encouraged an approach to the management of ethnic pluralism that was marked by ambivalence. On the one hand, as political rulers moved to consolidate power shortly after independence they became even more hostile to the formal organization and expression of sectional interests. A common feature of early post-independence revisions of constitutions, justified by the quest for national unity, was the proscription of ethnic-based political parties (Le Vine, 1997). On the other hand, whether governing through civilian regimes or military autocracies, they sanctioned the use of ethnicity as the key ingredient of elite competition for power and resources. African intellectuals, committed disciples of the modernist projects of nation-building and development, were equally, if not more, hostile to ethnic-based political competition. This hostility was stoked by the judgement that the self-serving manipulation of communal sentiments by elites explained the ongoing politicization of ethnicity. It well may be the case, as Crawford Young (1993: 23-4) has suggested, that the stigmatization of ethnicity in 'intellectual discourse and state doctrine' was extreme in the case of Africa. The stigmatization of ethnicity in politics and the resulting limits on attention to institutions that could facilitate more effective management of pluralism within participatory political contexts accorded with the suppositions of the prevailing doctrine of development which equated political modernity with growing national unity. The post-Second World War international order condoned this bias towards national unity. As the decolonization process wound down, the international order placed the normative emphasis on the territorial integrity of sovereign states which were vested with the right to self-determination and was unaccommodating to the claims of minorities within states (see Young, 1998: 1-3; Jackson, 1990: 139-59; andHerbst, 2000: 97-138). Developments in the closing decades of the twentieth century - ethnic conflagrations following the implosion of the Soviet Union and the East European Socialist states, the tragedy of Rwanda, the apparent intensification of identity politics in the context of the 'Third Wave of Democracy', etc. - have combined to undermine the belief that cultural homogeneity was a desirable and attainable attribute of national unity in a world in which multi-ethnic nation-states were preponderant. This has led to a growing argument for the need to dispense with the 'pretense of the ethno-cultural neutrality of states' (Kymlicka in this volume) and voices in support of 'pluriethnic approaches to nation-building that decouple citizenship with nationality' (Eriksen, 1999: 47-8). According to this advoRANSFORMING

96

Liberalization & the Politics of Difference in Cameroon

97

cacy, a key challenge to democratic nation-building in multi-ethnic societies is the design of institutions and policies which recognize and effectively mediate the competing claims of universal and ethno-cultural citizenship. In the case of Africa, moral and pragmatic defence of pluri-ethnic approaches to national building has, to a significant degree, rested on the consensus about the roots of the failure of post-colonial nation-building projects: regime consolidation through excessive centralization of power and neo-patrimonial networks. Besides the usual litany of evils (rampant corruption, economic mismanagement, casual violence of rule, etc.), the ubiquitous model of regime consolidation did not dampen the salience of ethnicity in politics; it simply led to the internalization of ethnic competition within state and party structures (see 'Introduction' in this volume). This prognosis of the roots of Africa's predicament supported decentralization of state power and recognition of ethno-cultural differences as guiding principles for the restructuring of state-society relations along democratic lines. The pervasiveness of ethnoregional competition since the return to multiparty politics in the 1990s has increased the appeal of this approach to state reorganization. Several of the chapters in this volume examine the practical and normative challenges that confront the choice of institutional models and policies for managing ethnic pluralism in the context of democratic nation-building (see especially Ejobowah, Kymlicka and Simeon). As none can claim universal applicability, the appropriateness of institutional models and policies will depend on the particular circumstances of African nations (see Conclusion in this volume). Whatever the case, African nation-states remain incipient and approaches to the management of ethnic pluralism that are predicated on the decentralization of power have to remain attentive to the need to build the capacities of central governments to promote national unity and pursue the collective objectives of the societies over which they preside. It will suffice, then, to note two broad perspectives on decentralization, the management of ethnic pluralism and democratic nation-building in the current African conjuncture. The first is represented by the 'New Public Management (NPM)' approach, which is grounded in the reigning neo-liberal orthodoxy on development that insists on the simultaneous advance of free markets and liberal democracy as the only path to sustainable development. The programs of the multilateral financial institutions which have structured the debate and orientation of policies for the reform of governance institutions in Africa are predicated on the assumptions of NPM. For them, decentralization is largely a matter of divesting states of functions that are supposedly more efficiently executed by sub-national political-administrative units or private bodies. In invoking the populism of civil society discourse, proponents of N P M tend to leave the impression that the downsizing of states somehow automatically translates into the growing empowerment of local societies. They do not seem to be excessively concerned about the implications for nation-unity posed by the downsizing of states. In fact, as has often been remarked, the fixation on economic efficiency as the main criterion for the decentralization of power makes the N P M unreceptive to, for example, redistributive policies which may be necessary to compensate for regional imbalances in resources and opportunities that often feed the alienation of ethnoregional groups and undermine the quest for national unity (see Bangura, 1999; Larbi, 1999). The second perspective, in contrast, places at the centre of concern the need for recognition and representation of ethnic difference as the guide to state

98

Dickson Eyoh

decentralization and democratic nation-building strategies. The arguments favoring this approach to nation-building are differently composed, and the proponents vary in their interest in and preference for the enabling institutions and policies.1 However, it is possible to discern some common assumptions in their arguments. At the core of these assumptions is the belief that, like their predecessors, post-colonial states continue to be divorced from culturally embedded and variegated understandings of the meanings of political community, the principles of representation and the moral anchors of political authority. This is because of what may be termed, for lack of a better word, the 'hybrid' character of post-colonial states and societies that are shaped and continually reshaped by the intersection of local and external social, economic, cultural and political forces. The result is political systems that are ordered by overlapping and often conflicting notions, one ethnically defined, the other national and state-centred. It is the resilience of 'kinship as the most significant local principle of social organization' (Eriksen, 1999: 57) or the most fundamental unit of social trust (Berman in this volume) in the everyday interactions engaged in by people in the pursuit of material and non-material goals that underscores the vitality of ethnic citizenship. It is thus hardly surprising that, with the failure of post-colonial states to address the basic needs of most societal groups, ethnic citizenship would retain or have reinforced its greater import on the organization of everyday life (including engagement in public spheres) than the abstractions of national citizenship. I am generally sympathetic to arguments in support of pluri-ethnic approaches to nation-building, and therefore consider as a major challenge of the current African political conjuncture the search for policies and governance institutions that can effectively mediate the legitimate interests of sub-national collectivities without circumscribing the fundamental democratic rights of all citizens and the capacity of states to pursue collective social aspirations. Ethnicities, as suggested in the Introduction and several of the case studies in this volume, attest to the genuine cultural diversity of African societies. Yet, it is the rare exception that ethnic political mobilization in the post-colonial era has been motivated by the need to defend the cultural autonomy of groups pummeled by the homogenizing impulses of state incorporation and economic modernization (Scarritt, 1993). The main engine of ethnic politics and conflict has been elite-led struggles for power and the resources of modernity over which states exercise overwhelming control. Therefore, however robust the utilitarian and normative arguments for pluri-ethnic approaches to nation-building, any design of institutions and policies to acknowledge the rights to representation and power of sub-national communities is likely to confront two unexceptional but nonetheless significant and related practical problems. The first problem pertains to the demarcation of group boundaries. Ethnic identities are not static and are always subject to contestation from within and without. Morever, the vast majority of African nations are composed of ethnic minorities, and even in instances of relatively high degrees of convergence of ethnicity and space, historical and contemporary migrations have ensured that most local societies are multi-ethnic. Given these circumstances, what are and who decides the criteria for demarcating ethnic group boundaries and citizen1

African proponents include the late Claude Ake (1993, 1994), Maxwell Owusu (1996, 1997) and P. Ekeh (see contribution to this volume). A noticeable sea change in the attitude of African intellectuals is a rejection of centralist models of nation-building and agreement on the imperative of recognition and effective management of cultural pluralism.

Liberalization & the Politics of Difference in Cameroon

99

ship and what happens to local groups that are excluded by the chosen criteria? The second concerns the tension between the institutionalization of groups' rights and democratic participation and accountability. Put starkly, given the reality of the resilience of neo-patrimonial networks that are the bedrock of political power and the wellspring of politicized ethnicity, what is to guarantee that recognition and institutionalization of ethnic group rights do little more than firm local elite monopolies of power and a pattern of state rule based on ethnic elite compacts? The rest of this chapter attempts to illustrate the 'ordinariness' of these two problems in contemporary struggles to refashion state-society relations in Africa. This is done with reference to Cameroon, a country in which the parceling together of demands for democracy, state decentralization, and recognition of the rights of sub-national groups to representation and power characterized, from the onset, the political liberalization process in the 1990s. In Cameroon, as elsewhere, the return to multiparty competition has obliged elites previously accustomed to the exercise of power without popular mandate to seek the support of ordinary citizens, and therefore has made control over local societies all the more imperative for political success. As a consequence, elite-driven demands for the decentralization of power and the regime's response to these demands have been predicated on ideologies of autochthony which insist that indigenous groups be accorded greater rights to representation and power in local society. The exploration proceeds with an overview of how the management of ethnic pluralism within the context of an authoritarian model of nationbuilding set the basis of identity-driven political competition and demands for state decentralization in the 1990s. To further elaborate the 'structural' and 'experiential' roots of the two problems noted, it concludes with an account of the basis and pattern of agitation by the officially English-speaking (Anglophone) communities for increased political autonomy. 2

Nation-Building and the Management of Pluralism Under Ahidjo and Biya Cameroon is a post-colonial state formed out of a union in 1961 of United Nation trust territories that were administered by France and Britain, and were formally colonized by Germans from 1885 to 1916. With a population composed of over 250 ethnic groups (based on indigenous language differences), it is amongst the world's most ethnically heterogeneous societies. Besides effective regulation of such ethnic pluralism, success in nation-building called for the blending of territories with different inheritances of official languages, administrative, legal, educational and political traditions as well as levels of economic development. British-administered Southern Cameroons, renamed West Cameroon at independence and currently the contiguous North West and South West Provinces, was less economically developed. Its more pluralistic and participatory political system contrasted with East (ex-French) Cameroon's centralized system, with a dominant executive. Under the direction of Ahmadou Ahidjo, inaugural Prime Minister of East Cameroon in 1958 and founding President of the officially bilingual post-colonial state from 1961 to 1982, the 2 Much of what follows builds on and extends analysis in Eyoh, 1998b and 1998c, which contain references to the pertinent literature on politics and regime formation in Cameroon.

100

Dickson Eyoh

quest for national unity was used to excuse the elaboration of a highly authoritarian and centralized state. Ahidjo succeeded in rendering East Cameroon a one-party state by I960, dominated by his Cameroon National Union (CNU). West Cameroon's political parties were dissolved in 1966 and the CNU proclaimed the sole legal party. A rump referendum in May 1972 replaced the federal republic with a unitary state and lent constitutional legitimacy to the established reality of state rule that had never pretended interest in regional autonomy. Authoritarian rule positioned the state as the dominant locus of power and the presidency as the central node of a constitutionally sanctioned network of state and party institutions. Seasoned politicians whose careers began in the colonial era remained important as senior managers of party and state structures, but by the late 1960s, the regime had become more reliant on so-called 'politicians by decree', that is, academics and civil administrators who were appointed to senior positions in the government and party by the President. This phenomenon promoted a strong identification with the regime on the part of the mostly stateemployed professional classes from both official language communities. It also eliminated the divide between politics and public administration, as it fixed expectations of elevation to cabinet and other high state-party offices as part of the career ambitions of senior bureaucrats and academics. State power was propped up by a combination of patronage, an extensive secret police apparatus and careful attention to ethnoregional representation in the composition of the top echelons of the state and party bureaucracies. The regime's attention to what became commonly known as 'regional equilibrium' extended to the selection of candidates for professional academies which served as the vehicle for recruitment and socialization into the administrative-cum-political elite. An obsession with national unity and hostility towards the organization of any political interests outside the single party acted as powerful checks against ethnic political mobilization. While members of the political-administrative elite were positioned as representatives of regions or ethnic groups at the summit of power, they understood that they owed their positions to Presidential discretion and were proscribed from cultivating an independent basis of political support. For regional and local political barons, what mattered to remain in power were relations with officials of the territorial administration (Governors and Divisional Officials) and a clean record from the secret police apparatus, rather than effective articulation of the preferences of the constituencies they were supposed to represent. Consequently, since elite political ambitions were better served by undermining competitors from within one's own ethnic group, intra-regional ethnic elite competition overshadowed inter-regional ethnic elite competition. This strategy of managing ethnic pluralism benefited from Cameroon's intense ethnic fragmentation which worked against the formation of powerful ethnoregional elite blocks, as has been the case in Nigeria and to a lesser extent Kenya (see chapters by Mustapha, Falola, Odhiambo, Lonsdale and Muigai). It nonetheless assured the continued salience of communal identities in politics and popular evaluation of the benefits of political power. The constitution of post-colonial regimes as alliances of ethnoregional elites has been recognized as a lead factor in the politicization of ethnic differences, as it entails the differential and hierarchical incorporation of groups within the state system of power. Ahidjo's regime was not immune from this tendency. In popular perception, if not in fact, its inner sanctum was dominated by Ahidjo's Hausa-Fulbe compatriots from the north and certain groups from the south

Liberalization & the Politics of Difference in Cameroon

101

were also seen to be favored. The importance ascribed to regional identities encouraged popular representation of political-administrative regions as culturally homogenous and thereby the misleading attribution of a common ethnicity to related and unrelated regional groups. For example, Fang-Beti groups in the south became Betis, Ewondos or Yaoundes, peoples from the north Hausas, and peoples from the South West and North West Provinces Anglophones. The regime assisted this inclination, especially with respect to the North whose population it projected as culturally Muslim, whereas an agglomeration of ethnic minorities, misleadingly and derogatorily lumped togther as 'Kirdi' (non-believers), accounted for more than 70 per cent of the region's population. By this myth, it preserved the dominance of power in the region by the Muslim HausaFulbe minority that had been institutionalized by the colonial state and made mobility within the state system of power for members of northern minority groups contingent on forced or voluntary conversion to Islam and adoption of Hausa-Fulbe cultural repertoires (Schilder, 1994). The incorporation of traditional rulers into networks of patronage formed a large part of the strategy of control of local society, one which enhanced the political value of more localized ethnic affinities and kinship ideologies. In a process which Bayart (1993) labeled 'reciprocal assimilation of elites', local groups increasingly favored appointment of their modern elites as traditional leaders or according them neo-traditional titles. For commoners, the benefit lay in a new type of traditional leader who could enhance communal access to the resources of modernity and guard against modern elites' straying from moral obligations to their primary communities. For modern elites, traditional leadership roles were 'symbolic capital' which could be deployed to ease access to land and other local resources as well as to advance political ambitions at various levels. The massive reorganization of internal boundaries (regional, divisional, sub-divisional, etc.) to serve political and economic objectives had similar effects. As Mbembe (2000) has suggested for the continent as a whole, the redrawing of these boundaries was supposed to capture groups with shared cultures and languages. In fact, a political project of commoners and ethnic elites in Cameroon was to have their area accorded status as a divisional or sub-divisional administrative unit. Such designation was deemed to confirm state recognition of the primacy of claims to local representation and power and collectively owned resources by groups considered indigenes of the area. In sum, the strategies of state penetration of local society and the management of ethnic pluralism created a matrix for the constant re-imagining and invention of regional and more localized kin-based identities. In their complex intersections, these identities at times reinforced and at other times weakened the political salience of each other. In 1982, Ahidjo resigned and was succeeded by Paul Biya, a 'politician by degree' par excellence, who had served as his Prime Minister since 1975. This change did not bring about substantial alterations in state-society relations. The most conspicuous adjustment was the ethnoregional realignment at the summit of state-party power. An abortive coup d'etat in 1984, in which Ahidjo and 'northern political barons' (to use a local term) were rumored to be implicated, gave Biya the opportunity to steady his grip on power. In the aftermath of the coup, ministers, bureaucrats and politicians who were closely associated with the Ahidjo regime or whose loyalties to the new regime were suspect were cashiered or demoted en masse (many were subsequently rehabilitated) and, in 1985, the CNU was remade into the Cameroon Peoples Democratic Movement

102

Dickson Eyoh

(CPDM), with Biya in firm control. The regime continued to cultivate the support of the Hausa-Fulbe political elite which it considered the greatest threat to its hold on power. To counteract possible challenges from this group, it carved the North into three administrative provinces (Adamawa, North and Far North) and turned to a greater reliance on elites from northern minority communities as part of its regional support base. Members of Fang-Beti groups in the Centre, South and East provinces increasingly dominated the regime's inner core. It increased pressure on civil servants in particular and others most directly dependent on state resources for greater political involvement in their home regions. These subtle rearrangements of state-society linkages would be highly conducive to the identity-based political competition and demands for state decentralization in the 1990s.

Challenging Authoritarian Rule in the 1990s: From Citizen Protest to the Politics of Belonging With buoyant oil revenues and a large surplus inherited from the Ahidjo era, the Biya administration embarked on profligate investments in showcase projects and massive expansion of the state bureaucracy. In the mid-1980s, the economy, which seemed protected from the downward trend of continental economies, began to deteriorate rapidly and feed increased public anger against the regime. Pressured by the multilateral financial institutions, the regime had to abandon many ill-conceived projects, freeze public sector employment, and undertake other unpopular measures. The unfolding crises tasked its ability to maintain patronage networks, and fueled official corruption and anti-regime rage. In the popular imagination of the mid-1980s, the worsening official corruption was due to the capture and privatization of the state by the 'essingan' (a closely knit circle of senior politicians, bureaucrats and academics from Beti and related ethnic groups) and 'les Betis* were collectively the beneficiaries.3 The inclination of senior Beti politicians, academics, and journalists to demonize other ethnoregional groups as enemies of the regime abetted the perception of an 'ethnic monopolization' of state power and resources. In the aftermath of the abortive 1984 coup, for instance, Hausa-Fulbe politicians were lambasted as agents of an 'Islamic menace' in state-controlled information organs. As the economy worsened, Bamilekes, reputed to be the most enterprising group and the dominant force in southern commercial networks, were accused of 'ethnofascism', that is, flexing their economic muscle to undermine the regime's management of the economy. The demonization of Bamilekes grew frenzied in the late 1980s when they were blamed for the sudden collapse of major state-owned banks, which was due in large measure to unsecured loans to the Presidential couple and the political and business interests of Beti-related ethnic groups. Bamileke businessmen who, like others, felt Betis were being heavily favored, had turned to a greater reliance on informal savings networks (tontines) for capital (see Takougang and Krieger, 1998: 101-3; Onomo, 2000). They became, along with the Anglophones whose disenchantment with the post-colonial regime was growing more vocal, 'les enemies dans la maisort - groups dedicated to the dismemberment of the nation. 3

Mbembe (1993) speculates that this ethnic bias in access was due to the interest of Beti elites in creating a Beti economic bourgeoisie comparable to the Bamileke.

Liberalization & the Politics of Difference in Cameroon

103

As across Africa, the marshaling call of the urban-based popular opposition to authoritarianism was the return to multiparty politics and the rule of law. The first multiparty elections since the mid-1960s were held in 1992; in March for parliament and November for the presidency. Without doubt, and this needs to be iterated, the mix of grievances - growing economic destitution, narrowing avenues of social mobility for the educated, rampant corruption, arrogance of ethnic power, etc. - that incubated and sustained popular opposition to the regime cut across socio-economic and communal boundaries. Given the hierarchical representation of ethnoregional groups at the summit of state-party power and the privileging of autochthony in local representation and power, political liberalization opened up space for the articulation of long-entrenched feelings of communal injustices in some instances and heightened presumptions of the need to defend communal advantage in others. These would sustain the unrelenting regionalization and ethnicization of political competition in the 1990s. By the time of the 1992 elections, four of the over fifty registered political parties had emerged with the strongest claims to national constituencies. However, the leadership and core support base of the four parties were identified with particular regions, although, in fact, the party of the political and bureaucratic elite, the ruling CPDM, was perceived as the party of 'les Betis\ The National Union of Democracy and Progress (NUDP), headed by Bouba Bello, derived the core of its leadership from Hausa-Fulbe groups in the northern provinces. The Social Democratic Front (SDF), which remains the largest opposition party, was considered a party of the Anglophones backed by Bamilekes in the West Province and 'allogenes' (ethnic diasporas in the southern provinces). Its leadership was dominated by north-westerners. The stronghold of the Cameroon Democratic Union (CDU) was the Sultanate of Foumban in West Province. In the March 1992 parliamentary elections, the CPDM failed to secure a clear majority but performed strongly in southern (Center, South and East) provinces, and in the North West. Its impressive victory in the North West as well as the NUDP's strong showing in the South West and West provinces was largely due to the SDF's boycott of the elections, because of the regime's refusal to set up an independent electoral commission. The N U D P dominated the North, and split seats with the CPDM in the Far North province, and with the MDR, a regime-sponsored minority party, in Adamawa province. The presidential election of the same year, a three-way contest between Biya, the SDF's Fru Ndi and NUDP's Bouba Bello Maigari, reiterated this trend towards the regionalization of political competition. Biya held sway in the southern provinces; Fru Ndi prevailed in the SDF's provincial bastions (North West and West), and in the South West and Littoral provinces with substantial immigrant communities from these two provinces. The 1997 parliamentary elections affirmed the pre-eminence of region-based political competition and its remarkable consequence. With 110 of 180 seats, the CPDM had reasserted its position as the only truly national party (with a seemingly unassailable claim on the Presidency), and the SDF (with 43 seats) was reduced to a parliamentary opposition party with a primarily regional support base in the North West and West Provinces and amongst immigrant populations from these provinces in southern urban centers. The presidential elections of 1998, which were boycotted by the SDF and gave the victory to Biya, reconfirmed the 'new order' of multiparty politics: the transformation from what appeared as a politics of citizen protest against authoritarian rule at the start of the liberalization process to a 'politics of

104

Dickson Eyoh

belonging' whereby political participation was determined first and foremost by kinship ties and communal affinities. The result, at the close of the 1990s, has been security of power for a regime which is uninterested in democracy and a process of state decentralization that would confer substantive rights and powers to sub-national units. To return to the relevance of the Cameroon experience to the two practical problems noted concerning the articulation and redress of demands for the recognition and political representation of sub-national communities, the pattern of political competition and its consequence have been significantly determined by the intense ethnic fragmentation of Cameroonian society. Set against the background of the history of post-colonial regime consolidation, the ethnic pluralism of regions explains, in structural terms, how in their complex intersections regional identities compete with, and are most often overwhelmed by, more substantive ethnic identities. Whilst the material and symbolic relations that bind urban elites and local populations are highly varied across time and space, the return to multiparty electoral competition lent a new political significance to local society. Because of the preponderance of rural society, the majority of political constituencies outside polyglot urban centers are more or less demarcated by ethnicity and other markers of group difference. This has encouraged and rewarded elite reliance on 'localizing strategies of competition' (Nyamnjoh and Rowland, 1998) that are predicated on appeals to kinship ties or communal membership. The governing understanding of the effectiveness of political representation as reflected in the ability to access resources for individual and communal benefit sustained the preference of 'ordinary' citizens for political representation by 'sons, occasionally daughters, of the soil'. Across the nation, including the domains of opposition, political competition came to be increasingly framed as contests over local citizenship and rights to political representation. These contests have been predicated on the differentiation and hierarchization of local/regional citizenship: between the so-called indigenous peoples and 'strangers' or migrants (regardless of the length of settlement in areas of migration). In the increasingly virulent rhetoric of these contests, the 'native born' have insisted on their inherent right to local/regional political power and representation, based on the specter of 'stranger domination' and the certainty of the incapability of 'strangers' to commit to the development of their 'adopted' homes. The contests penetrated to lower levels of society, often pitting clans and lineages from the same ethnic group against one another. In challenging this overvaluation of autochthony, so-called strangers have invoked residence and universal citizenship as the valid criteria for political participation and representation. Yet, even groups that have been targets of antiimmigrant hostility resisted attempts to vest them with 'second-class citizen' status, while defending autochthony as the basis of power and representation in their ancestral regions and communities. In effect, with this elite and mass conviction about the primacy of the politics of belonging, interest in state decentralization has been motivated mainly by the quest to build or defend local/ regional bases of elite power by constraining the rights to power and participation of non-autochthonous groups. The incumbent regime has accentuated the politics of belonging in order to domesticate the challenges of multiparty competition. Its success in containing the opposition has rested on the vulgar use of localizing strategies of competition to fashion supportive regional elite alliances and to frustrate the emergence of alternative interregional elite alliances. It has invested enormous energies and

Liberalization & the Politics of Difference in Cameroon

105

patronage to retain traditional elites as part of the hegemonic alliance, and obligated the bureaucratic elite and others dependent on state patronage to ongoing involvement in politics in their ethnic communities or regions. Whether by choice or for fear of being labeled a supporter of the opposition, bureaucrats, academics and managers of state corporations became regular fixtures in the CPDM's local campaign teams and in state delegations to their home areas. Judging by the continuous strength of the SDF in both provinces, the manipulation of regional elites has been less successful in the North West and West Provinces. Ethnic communities of both provinces have produced a commercial elite which is less dependent on state patronage, and has been the target of regime-abetted anti-immigrant chauvinism and thus more willing to side with the Social Democratic Front (see Rowlands, 1993). Paradoxically, the existence of a relatively autonomous commercial elite inclined to support the opposition increased pressures on elites whose fortunes were wedded to the regime for more vigorous assertion of regional leadership. Although not restricted to it, in the North West Province especially, the resultant intensification of intra-region elite competition for local and regional leadership has kindled an increasingly violent local politics as elites resort to exploitation of clan-based divisions and boundary disputes. The regime routinely misused legal political-administrative arrangements to fashion and refashion regional-ethnic elite alliances. For instance, in the last municipal elections (1998), the SDF emerged as the overwhelming victor in the major urban centers in the North West, West, Littoral, and South West Provinces. The regime responded to this victory by appointing so-called indigenes as Government Delegates, with control of budgets and patronage resources, to administer urban councils in Limbe, Douala, Bamenda Bafoussam and Kumba. The victorious opposition was left with the largely ceremonial office of Mayor. More significant, and perhaps ominous, was the revised constitution promulgated in 1996, but whose provisions were yet to be fully implemented at the time of writing. The constitution provides for the decentralization of power based on the existing ten provincial administrative units, and stipulates that the interests of autochthonous groups must be respected in the composition of local and regional political institutions. This provision has been commonly interpreted and welcomed by the cross-section of elites as signaling that members of autochthonous communities must head the executives of local and regional governments and constitute the majority in their deliberative organs. A small group of academics have been the only voices warning against the repercussions for national unity of this institutionalization of ethnic preference in the arrangement of local and regional power (see Melone et al. 1997; Ebousi-Boulaga, 1997; Collectif Changer Le Cameroon, 1992, 1994). The diversity of regional histories and cultural ecologies ensures marked variations in the ways the tensions between the universalistic rights of national citizenship and the particularistic rights of sub-national groups to political power and representation have been constructed and played out at different levels. The experience of the Anglophone communities of the North West and South West provinces aptly illustrates the dynamic of these tensions. Anglophone agitation for greater political autonomy was the best organized and most threatening regionalist challenge of the 1990s. The Biya regime's response to demands for state decentralization has, to a significant degree, been informed by a determination to tame this threat. The experience is particularly revealing of the extent

106

Dickson Eyoh

to which the struggle for decentralization can and has been driven by the competitive regionalisms of elites as well as inspiring the constant re-imaging and invention of identities.

Contest of Local Citizenship and the Deconstruction of Anglophone Identity 4 Although Anglophone communities shared in the vilification of the Biya regime's ethnic chauvinism, for them 'Beti hegemony' was simply another expression of the domination of state and economy by Francophones. Anglophone grievances turned on a self-conception of former West Cameroon as a distinct community defined by differences in official language and inherited colonial traditions of education, law, and public administration. In the popular wisdom of Anglophones, the post-colonial nation-state building had been defined by the singular determination of the Francophone elite to erase the cultural and institutional foundations of Anglophone identity. The mainstay of the inequalities in distribution of political and symbolic power and access to state resources between both official language communities was what Cameroonian scholar, Kofele-Kale (1985), referred to as the 'gallicising of public life': the preeminence of French as the official language and of inherited French institutions and bureaucratic practices in all aspects of state administration and public life. In the heavily partisan English language press and Anglophone popular discourse, the gallicizing of the post-colonial state was evidence that reunification had been tantamount to the 'recolonization' of West Cameroon by their Frenchspeaking compatriots. The resulting marginalization of their community was borne out by the exclusion of Anglophones from the most senior positions in state and party, a pattern of distribution of public investment that favored Francophone regions, and discrimination in employment in the public and private corporate sectors. The emergence of oil in the South West Province as the major source of state income in the 1970s fueled convictions that Anglophones were responsible for the bulk of resources which fed, at their expense, an obscene level of corruption. This was in spite of the fact that oil income was in decline by the late 1980s. Anglophone complaints about the gallicizing of the state and public life were not without merit as the four-to-one population ratio warranted the numerical preponderance of the Francophone elite within the state and economy. When combined with intolerance of sentiments that appeared to challenge the authoritarian model of nation-building, the demographic imbalance meant that the Anglophone elite was under greater pressure to adapt to the language and cultural codes of its Francophone counterpart in order to more effectively navigate state-centered networks of power and accumulation. These pressures were pronounced for the older generation with limited competence in French, but they were not easily evaded by the younger generation of the Anglophone elite who were fluent in both official languages. The complaints thus highlighted the significance and contradictions of the official language-based regional division as one of the multiple axes of a state-driven process of elite formation. On the one hand, from the perspective of the elite, the complaints were self-serving. By 4

See Eyoh, 1998c for detailed analysis of how the contest of Anglophone identity is played out through conflicting elite narratives of the post-colonial trajectory.

Liberalization & the Politics of Difference in Cameroon

107

attributing to them a homogenous regional identity, they allowed the Anglophone elite to shift on to their Francophone counterparts responsibility for the pathologies of the post-colonial state and to discount the fact that many of the grievances they sought to particularize were being voiced by other subnational groups. On the other hand, because of the manner in which class and communal divisions had been interwoven to maintain the networks of patronage on which state power was predicated, these complaints signaled a common optic through which Anglophone elites and ordinary citizens had come to view the post-colonial state as a culturally coded and therefore contestable set of institutions and practices. The alienation expressed in this representation of post-colonial experience propelled the Anglophone elite into leadership of the pro-democracy movement as it gathered steam in the early 1990s and to establish the SDF as the first nation-wide opposition party in the face of violent resistance from the regime. Besides the broader demand for a return to multiparty politics, Anglophone elites called for the reorganization of state power through a return to some form of federalism. The regime responded to popular pressures for liberalization and state decentralization with a highly controlled process of constitutional reform. It became apparent, in the initial fora on constitutional reform established between 1991 and 1993, that the regime and the Francophone elite schooled in the centralist tradition of France preferred the retention of the unitary state, but with the devolution of some decision-making authority to regional and local bodies. A range of popular movements of varied size, organizational coherence and durability served as channels of Anglophone agitation for a return to the federal state. Though united in the wish to see the unitary state dismantled, the Anglophone elite soon revealed themselves to be divided along regional lines as to the composition of a re-federalized state. These divisions coalesced around two constitutional positions: return to the pre-1972 federal arrangement (the socalled 'two-state option') and a 'ten-state' federal system based on extant provincial administrative units. While the various movements behind the first option drew their supporters from both Anglophone provinces, the elite champions were mainly from the North West Province and were associated with the SDF. In contrast, the main advocates of the latter option were South-western elites, the majority of whom were supporters of the ruling CPDM. They employed elite organizations with overlapping memberships, notably the Southwest Elite Association (SWELA) and Southwest Chiefs Conference (SWCC) , that had been formed in the wake of liberalization to defend what they construed as the interests of their province. The parliamentary and presidential elections of 1992 affirmed the widespread resentment of the post-colonial regime amongst Anglophones. At the same time, the divergent pattern of voting in the two Anglophone provinces with respect to support for the ruling and opposition parties in these elections revealed, and subsequent elections in the 1990s confirmed, the deep-seated regional cleavages amongst Anglophones. These cleavages framed the contending Anglophone elite positions on state decentralization and animated the quarrel over the meaning and political salience of Anglophone identity. South-western elites shared in the standard Anglophone grievances against the centralized state, but advanced an interpretation of the post-colonial trajectory that challenged assertions of an all-embracing Anglophone identity and positioned North-westerners as co-architects with the Francophone elite of the post-colonial state. The United Nations had offered Southern Cameroons the

108

Dickson Eyoh

choice of independence with Nigeria or reunification with French Cameroon. Politicians from the South West Province led the fight for independence with Nigeria, while the Kameroun National Democrat Party, a party dominated by North-westerners, advanced the successful course of reunification with French Cameroon. For south-western elites, because their province was the bastion of opposition to reunification, north-westerners emerged as the favored Anglophone group in the post-colonial regime and deployed their preponderant presence within the centralized state to the disadvantage of the South West province. As with the overarching Francophone/Anglophone elite cleavage, through this indictment South-western elites sought to shift onto their northwestern counterparts responsibility for a shared nation-building project. Southwestern elites in order to ameliorate what many perceived as the overwhelming dominance of North-westerners of the government of West Cameroon, were amongst the most enthusiastic supporters of the abolition of the federal state in 1972. 5 Morever, their narrative of the post-colonial experience necessarily exaggerated the ethnic and cultural similarities of north-western societies, while downplaying similar divisions amongst south-western communities. This narrative of post-colonial trajectory was effective in deconstructing Anglophone identity and subverting the challenges to the regime represented by its mobilization, because of a set of historical circumstances that have allowed 'ordinary' citizens of autochthonous communities of the South West province to take part in its production and consumption. Principal among these are demography and a historical pattern of cross-region migration which have profoundly altered the ethno-cultural ecology of the South West province. With a population currently estimated at approximately 1.3 million against South West's 800,000, the North West has always been the more populous of the two provinces. In a common pattern of colonial change whereby coastal communities were initial sites of commercial contact with Europe and colonial state penetration, south-westerners led the early ranks of the modern Anglophone elite; by the 1950s the advantage had shifted to the North West. The centrality of education as the main, and still relatively egalitarian, vehicle for socio-economic mobility, favored the numerical preponderance of North-westerners among the Anglophone faction of the post-colonial hegemonic alliance and, through better access to state resources, their prominence in the private business sector. Perhaps dating back to the era of the transatlantic slave trade, but heightening with the movement of labor from the North West to the German-developed plantations in the South West at the turn of the twentieth century, the interregion migration regime has remained unidirectional. By the last decade of colonialism, migrants from the North West had become the leading population group within plantation communities in the South West province. With the intensification of cross-region migration in the post-colonial era, indigenes of the North West and other provinces (notably the West) had by the 1980s outnumbered the 'native' born and were the dominant economic force in urban centers in the South West province. The spread of migrants from the North West has extended to remote rural communities, shifting from an earlier pattern where previously they were itinerant sharecroppers to being now mostly resident farmers. The deconstruction of Anglophone identity was fostered by the normally 5 See Mbile (1999) for revealing recollections of a veteran South-western politician whose career goes back to the nationalist era; and Awason (2000) and Chiabi (1997) for the history of decolonization and reunification.

Liberalization & the Politics of Difference in Cameroon

109

innocuous mutual stereotyping of 'self and 'other' that is characteristic of the co-mingling of groups. Thus, for instance, in the popular folklore of 'indigenous' south-westerners, peoples of the North West were hardworking, frugal, ambitious, aggressive and prone to ethnic solidarity and excessive deference to traditional authority. Peoples from the North-West, who share with the Bamileke a national reputation for economic dynamism, have come to internalize this order of stereotypes as positive markers of their communal identity (Rowlands, 1993). They reciprocate by characterizing south-westerners as unenterprising, with a weak sense of community and an incapacity to unite around common goals. For south-westerners, the history of inter-region migration has led to a grudging admiration of the economic dynamism of north-westerners. Their concern about north-westerners' superior capacity for ethnic political solidarity was grounded in popular beliefs about the significance of traditional institutions and kinship ties in the ordering networks of power and accumulation. Modern elites and diasporic communities of the North West have retained a far greater regard for their traditional institutions of authority and allegiance to the repertoires of cultural practices and kinship structures which sustain the translocal reproduction of networks of communal solidarity. In contrast to societies of the North West that are organized in highly centralized political systems with elaborate hierarchies of titled positions, the traditional political institutions have remained weak in the lineage societies of southern forest regions where chieftaincy is more of a colonial invention. In furtherance of their dismissal of the foundations and political value of Anglophone identity, south-western elites and their leading organizations began to put forward an alternative regional identity that was based on the supposed cultural affinities of coastal peoples and transgressed the official language divide. Stretching the concept of coastal peoples beyond reason, they claimed that autochthonous communities of the South West province shared common cultural bonds and traditions with the Duala and related groups in the Littoral province. The leaders of SWCC and SWELA attempted to give political content to this alternative identity through an alliance with SAWA, a group which represented Duala elite supporters of the CPDM. Formed in the colonial period as a cultural association dedicated to promoting a pan-Duala ethnicity, SAWA was reconstituted in the 1990s. With Bamilekes accounting for over 70 per cent of the population of Duala (the largest city and economic capital) and dominating its commerce, the Duala elite feared that multiparty competition would diminish further its insubstantial political power in the city. Their panic over the dominance of 'allogenes' intensified after the 1996 municipal elections, when none of the Mayors of the six districts won by the SDF was an ethnic Duala (see Mentah, 1997; Sonne, 2000). Against the backdrop of these demographic realities and constructions of difference, the uncharted course of multiparty competition provoked anxieties amongst the south-western elite and commoners about 'stranger' domination of local power and representation. These anxieties were sharpened by the inability of the south-western elite to build a regional party comparable to the SDF. Though its claims as an effective institutional alternative to the incumbent party had been seriously eroded by the end of the 1990s, the SDF was justifiably seen as a party organized and controlled by north-western elites, and backed by the majority of north-westerners nationally. The breadth of support for a return to an officially bilingual two-state federal system amongst northwestern elites raised the prospect of the continued vulnerability of the south-

110

Dickson Eyoh

western elite in representation and power at the national, regional and local levels. The Biya regime played to its advantage the regional cleavages that underpinned these anxieties. Thus, for example, the appointment of Achidi Achu, a veteran politician from the North West province, to the revived and second highest government post of Prime Minister following the parliamentary elections in 1992 and the reduction of the South West Province's representation in cabinet, was widely perceived as the 'punishment' of south-western elites for their failure to deliver constituencies. To south-western elites, the CPDM's victory in the North West Province (the result of the SDF's boycott of the elections) and Achu's appointment revealed the adroitness of north-westerners at playing both sides of the political field: simultaneously fanning opposition and support for the regime when their interests dictated but always to the disadvantage of their province. With the SDF's dominance seemingly unassailable in the North West Province, Achidi Achu was replaced as Prime Minister in 1996 by Peter Musonge, a veteran technocrat from the South West.6 This was the first occasion since independence when a south-westerner had been appointed to the second highest government position. South-western elites aligned with the CPDM premised their campaigns for the 1997 parliamentary and 1998 presidential elections on the communal significance of this appointment. They argued that support for the regime was imperative to maintain this Victory' in a long struggle to bring recognition to the province's disadvantaged political position since independence. The revised constitution of 1996 was likewise heralded and continues to be defended as an endorsement of their argument about the contemporary political irrelevance of Anglophone identity and the imperative of protecting the rights of autochthonous communities to local representation and power.

Conclusion It has not been the intention of this chapter to present Cameroon as paradigmatic of the dynamic of identities in politics and demands for state decentralization in the current African political conjuncture. The main purpose was to draw attention to two major problems - how to decide ethnic community boundaries and the order of local citizenship rights and the potential for creating elite monopolies of power - that are not easily avoided in the design of institutional frameworks and policies that are geared to reconciling the competing claims for representation and power based on shared national citizenship versus citizenship in ethnic communities. Cameroon, not least because of its intense ethnic pluralism and unique colonial history, offers an interesting example of the sources and difficulties of resolving these competing claims in a manner which promotes justice and respect for the democratic rights of all citizens. Here, as across much of Africa, the accentuation of identity-based politics and demands for state decentralization has, as in the past, been driven by elite struggles over power and resources and not the determination of communities to defend their cultural autonomy and traditions. The particular form and virulence derive from the manner in which the consolidation and maintenance of the post-colonial regime created and sustained an environment that encouraged appeals to com6 As of early 2004, Musonge had not been replaced, making him the second longest serving occupant of the position after Biya.

Liberalization & the Politics of Difference in Cameroon

111

peting regional and more localized kinship-based identities in the struggle for power and material resources at all societal levels. No effort was made to adjudicate the moral or empirical foundations of claims of communal advantage and disadvantage in the distribution of power and resources. Nonetheless, it is not difficult to appreciate that the logic of ethnoregional representation in regime consolidation and maintenance, including strategies of local control, would generate strong popular convictions about, and pressures to redress inequalities in, the distribution of power and resources amongst communities through state decentralization. Despite some erosion of its material base, the state in Cameroon, as across Africa, remains the most important vehicle of wealth. It is thus easy to see the intensification of the politics of belonging and demands for state decentralization in the era of liberalization as little more than elite strategies in competition for wealth and power in a new or altered political context. Yet, as highlighted by the contest over Anglophone identity and the positions on state decentralization that flowed from it, the pressure for recognition of the rights of 'native' groups to representation and power is in no small measure animated by varied localized anxieties about collectively owned resources such as land and relations of power within and between groups in local society. In other words, elites, ordinary citizens and the state have all been implicated in the valorization of the politics of belonging. The consequence thus far has been a paradoxical situation in which the power of the monolithic state has been barely disturbed, while fanning a pattern of political competition which guarantees the continued crippling of its capacity to promote national unity and development. It seem obvious, then, based on the chapter's sympathy for pluri-ethnic approaches to nation-building, that the crucial challenge in the search for institutional mechanisms that strive to strike a balance between sub-national groups and universal citizen rights in Cameroon-type situations has to be the rejection of exclusionary conceptions of local citizenship that are authorized by ideologies of autochthony. It is difficult to imagine how this can come about without a profound reorientation of elite attitudes in particular away from the conception of politics as simply a self-interested game for control of state resources to a means for pursuing broader collective objectives.

References Ake, C. 1994. Development and Democracy in Africa. Washington, DC: Brookings Institution. — 1993. 'The Unique Case for African Democracy', International Affairs, 69 (2): 240-4. — 1987. 'The African Context of Human Rights', Africa Today 34 (1-2). Awason, N. 2000. 'The Reunification Question in Cameroon History: Was the Bride an Enthusiastic or Reluctant One?', Africa Today 47 (2): 99-119. Bangura, Yusuf. 1999. New Directions in State Reform: Implications for Civil Society in Africa. UNRISD Discussion Paper No. 113. Geneva: UNRISD. Bayart, J-F. 1993. The State in Africa: The Politics of the Belly. London: Longmans Publishers. — 1985. Uetat au Cameroun. Paris: Presses de la fondation nationale des sciences politiques. Chiabi. E. 1997. The Making of Modern Cameroon: A History of Substate Nationalism and Disparate Union, 1914-1961. Lanham, MD.: University Press of America. Collectif Changer Le Cameroun. 1994. Ethnies et developpement national: actes du colloque de Yaounde. Yaounde: Editions C3 and Friedrich Ebert Stiftung. — 1992: Le cameroun eclarte: anthologie commentee des revendications ethniques. Yaounde: Editions C3 and Friedrich Ebert Stiftung. Ebousi-Boulaga, F. ed. 1997. La democratic a Vepreuve du tribalisme. Yaounde: Friedrich Ebert Stiftung and Gerddes Cameroun. Eriksen, Thomas H. 1999. 'A Non-ethnic State for Africa? A Life-World Approach to the Imaging

112

Dickson Eyoh

of Communities,' in P. Yeros, ed., Ethnicity and Nationalism in Africa: Constructivist Reflections and Contemporary Politics. Basingstoke: Macmillan Press. Eyoh, D. 1998a. 'Through the Prism of a Local Tragedy: Political Liberalization, Regionalism and Elite Struggles for Power in Cameroon', Africa 68 (3): 338-59. — 1998b. 'Community, Citizenship and the Politics of Ethnicity in Post-colonial Africa', in E. Kalipeni, and P. Zeleza, eds, Public Spaces and Public Quarrels. Trenton, NJ.: Africa World Press. — 1998c. 'Conflicting Narratives of Anglophone Protest and the Politics of Identity in Cameroon', Journal of Contemporary African Studies 16 (2): 249-76. Herbst, J. 2000. States and Power in Africa: Comparative Lessons in Authority and Control. Princeton, NJ.: Princeton University Press. Jackson, R. 1990. Quasi-States: Sovereignty, International Relations and the Third World. Cambridge: Cambridge University Press. Kofele-Kale, N. 1985. 'Ethnicity, Regionalism, and Political Power: A Post-Morten of Ahidjo's Cameroon', in M. Schatzberg and I. W. Zartman eds, The Political Economy of Cameroon. New York: Praeger Publishers. Larbi, G. 1999, The New Public Management Approach and Crisis States. UNRISD Discussion Paper No. 112. Geneva: UNRISD. Le Vine, V. 1997. 'The Rise and Fall of Constitutionalism in West Africa', Journal of Modern African Studies 35 (2): 181-206. Mbembe, A. 2000. 'At the Edge of the World: Boundaries, Territoriality and Sovereignty in Africa', Public Culture 12 (1): 259-84. — 1993. 'Crise de legitime, restauration autoritaire et deliquescence d'etat', in P. Geschiere and P. Konings. eds, Pathways to Accumulation in Cameroon. Leiden: Afrika-Studiecentrum; Paris: Karthala. Mbile, N. 2000. Cameroon Political Story: Memories of an Authentic Eye Witness. Limbe, Cameroon: Pressprint. Melone, S. A. She and L. Sindjoun, eds, 1997, La reforme constitutionelle du 18 Janvier 1996: aspects juridisques et politiques. Yaounde: Friedrich Ebert Stiftung. Mentah, T. 1997. 'Constitutionalism, Press and Coverage of SAWA minority Agitation', in Melone et al. Nyamnjoh, F. and M. Rowland. 1998, 'Elite Associations and the Politics of Belonging in Cameroon', Africa, 68 (3): 320-37. Onomo, J. 2000. 'Symetries hegemoniques beti-bamileke et rivalites politiques au Cameroun', in D. Zongong and I. Mouiche, eds, Democratization and Ethnic Rivalries in Cameroon. Ethno-Net Africa Publications, http: www.unesco.org/most/p95cirl.htm. Owusu, M. 1997. 'Domesticating Democracy: Culture, Civil Society and Constitutionalism in Africa', Comparative Studies in Society and History 39 (1): 120-52. — 1996. 'Tradition and Transformation: Democracy and the Politics of Popular Power in Ghana', Journal of Modern African Studies 34 (3): 307-43. Rowlands, M. 1993. 'Accumulation and the Cultural Politics of Identity in the Grassfields', in P. Geschiere and P. Konings, Pathways to Accumulation in Cameroon. Paris: Karthala; Leiden: Afrika-Studiecentrum. Scarritt, J. 1993. 'Communal Conflict and Contention for Power in Africa South of the Sahara,' in T. Gurr, ed., Minorities at Risk: A Global Survey of Ethnopolitical Conflict. Washington, DC: US Institute of Peace Press. Schilder, K. 1994. Quest for Self-Esteem: State, Islam and Mudang Ethnicity in Northern Cameroon. Aldershot: Avebury; Leiden: African Studies Centre Research Series. Takougang, J and M. Krieger, 1998. African State and Society in the 1990s: Cameroon's Political Crossroads. Boulder, CO: Westview Press. Sonne, W. 2000. 'De la dynamique de la qualite d'autochtone dans le processeus de democratisation au Cameroun: le cas de la region de Douala', in D. Zongong and I. Mouiche, eds, Democratization and Ethnic Rivalries in Cameroon. Ethno-Net Africa Publications, http:www.unesco.org/ most/p95cirl.htm. Young, C. 1998: 'Ethnic Diversity and Public Policy: An Overview', in C. Young, ed., Ethnic Diversity and Public Policy. New York: St. Martin's Press for UNRISD. — 1993. 'The Dialectics of Cultural Pluralism', in C. Young ed., The Rising Tide of Cultural Pluralism: The Nation-State at Bay? Madison, WI: University of Wisconsin Press.

7

CHERYL H E N D R I C K S

The Burdens of the Past & Challenges of the Present: Coloured Identity & the 'Rainbow Nation 5

E

identity has been a significant variable for political participation and the organization of state power in Africa. Recent studies have shifted our focus towards the constructed nature of ethnic identities, their multiple functions, and their imbrication with forms of governance in post-colonial states. Mamdani's (1996) emphasis on the 'bifurcated nature of the state', Bayart's (1993) stress on the 'reciprocal assimilation of elites', Lonsdale's (1994) differentiation between 'moral ethnicity' and 'political tribalism', and Ekeh's (1975 ) theory of the 'two publics,' are key texts highlighting both the specificity of African states and the change and continuities in the construction and employment of ethnic identity. South Africa, too, bears the burdens and challenges of identity politics. Previous regimes used the markers of race and ethnicity to institutionalize privilege and forge control over the state. The post-apartheid state seeks to reverse the effects of ethnically based politics and race-based citizenship and distribution of resources through the dual process of creating a unified nation and the implementation of policies for redress. In constructing the new nation and effecting redistribution, the state is faced with the reality of fragmented identities and its corresponding socio-political tensions. The challenges of constituting the 'Rainbow Nation' were evident from its inception, i.e., as seen in the spiral of violence initiated by the dissent of aspiring ethnonationalist Afrikaner and Zulu groupings to the genesis of individually based citizenship, in the identity-based voting patterns of the first national democratic elections, and in the cries of 'reracialization' that accompanied the transformation process. The challenges were also located in the political salience and representations of marginal identities, for example, Coloureds, who began to reassert claims to particularity. These claims gained political expression when the majority of the members of the group voted for the National Party in the 1994 elections, in an attempt to limit the ANC's power and nationalist vision in the Western Cape, and through the formation of political and cultural organizations that emerged to 'protect' Coloured 'interests'. This chapter focuses on Coloured identity, highlighting the processes of identity construction, the tensions around group recognition, and the contexts in which group identities become politicized. It points to the ideological and structural factors that shaped the identity and how expressions of the identity articulate with post-1994 nation-building discourses and practices. The central argument is that, although ethnic identities are socially constructed, they acquire a solidity of effect: they become real for their bearers and have actual consequences, especially in contexts where access to resources is allocated on the basis of one's identity. Although identities are continuously under processes of construction they are not infinitely malleable. Their representations are constrained by historical narratives, ideologies and the socio-economic contexts in which THNIC

113

114

Cheryl Hendricks

they are instituted, ordered and consolidated. Redefining racial/ethnic1 identities to suit differing contexts is a pursuit predominantly of the middle class whose education, social circles, and political consciousness provide the resources with which to existentially reflect and reconstruct their identities and to be in a position where the reconstituted identities have experiential validity (i.e. they can both move out of a group and/or become cosmopolitan beings). Conversely, those who do not have the necessary resources rely on the familiarity and security which belonging to a group affords. Capacious reconstruction of an identity, as in the case of Coloureds where what was envisaged is their transformation into 'sameness' within a homogenized black identity, is a slow process that can only result in changed social, structural and ideological settings in which the old meanings and relational characteristics of all racial/ethnic identities within the given context no longer have resonance. We contend that a social and material basis for Coloured identity emerged and was consolidated through shared lived experiences dating from the period of slavery; legitimating discourses positing them as an intermediate racial group; the institutionalization of their 'difference'; their separation from other groups in most social and residential spaces; and the relative advantages the group acquired under the Coloured Labour Preference Policy (CLPP) in the Western Cape. This is not simply a government-imposed identity, as is often argued. This conception of Coloured identity has become theoretically engraved and is part of the problematic of a meaningful analysis of the identity and an understanding of the driving forces behind its political and social expressions. It is a portrayal permeated by the same limitations of many of the constructivist accounts of African ethnicities in which Africans are denied agency in the making of their identities and/or where the remaking (a process that is constitutive of all identities) is then simultaneously used as a measure of authenticity. The processes of Coloured identity formation are not noticeably different from those of other social groups, that is, they are formed at historical conjunctures in which various forces enable the 'imagining' of a common identity. All racialized and ethnic identities are simultaneously imposed and accepted, celebrated and contested, and they are all constituted by both officiation and the bearers thereof and in the interplay between discourses and contexts. The construction of Coloured identity is therefore no more or less suspect than any of the other South African identities, and it resonates precisely because of the constructions of 'other' groups and the power relations between them. The attempts to render the differential experiences of Coloureds irrelevant to group formation, the negation of their own role in the assertion and framing of their identity, and the need to reconstitute this identity whilst the others remain 'stable' are, however, suspect. We argue that this interpretation is due to the ideology of the liberation movement, one that simultaneously sought to racially dichotomize the struggle and to build non-racialism. In this process the burden fell on the 'intermediate' groups to re-racialize themselves paradoxically (as black) and become nonracial, non-ethnic beings in a profoundly racially and ethnically constructed society. We also assert that the discourse of multi-culturalism used to construct the new nation encourages the kinds of reimagining and assertions of Coloured identity that have resulted in the post-apartheid state. Coloured identity in the 1

Identities like Afrikaner, Coloured, Indian, in the South African context, are both racially and ethnically constituted.

Coloured Identity & the Rainbow Nation

115

post-apartheid era should therefore be analyzed in terms of both its historically constituted 'difference' and its interaction with current nation-building pursuits.

The Processes of Coloured Identity Formation The historicity of Coloured identity can be traced to the dynamics of colonization at the Cape, which, through slavery and enserfment, created a race-based underclass, largely of 'mixed descent'. Through the centuries this group established a distinct social identity.2 Slaves were brought in from Madagascar, Mozambique, the Indonesian archipelago, Bengal, South India and Ceylon, whilst the indigenous Khoisan were dispossessed of their land and livelihoods and subordinately incorporated into the colonial political economy. The sexual exchanges (forced and consensual) between white colonialists, slaves, and the indigenous Khoisan reproduced a context in which, by the early nineteenth century, nearly 70 per cent of the Cape's underclass were of 'mixed descent'. It is this characteristic that became the signifier of Coloured identity in the twentieth century. But there was certainly more to the identity than 'racial hybridity'. The Cape's underclass were subjected to similar processes of exploitation, subjugation, and colonial acculturation. They also developed various clubs, associations and patterns of socializing. It is these shared experiences, common structural positions, and shared spatiality which provided further markers upon which a group identity could be formed. 3 The framing and fixing of Coloured identity occurred via both internal and external processes and in interaction with the dominant discourses and changing structural conditions at the Cape and at the national level. Slavery and indentured labor were abolished by the British government in the 1830s. What became known as the 'Cape Liberal Tradition' was seen as a distinguishing feature between the Cape and the other republics. The Cape government removed all references to race (but not gender) from the statutes, giving black men who qualified the possibility of exercising the rights of citizenship (the vote and social mobility). However, money was never a sufficient criterion for social mobility. The British discourse of civility framed who could be considered as equal. Within this discourse British culture was the yardstick for civility: a discourse centered around respectability, prudence, discipline, education and the acquisition of the English language. This discourse, holding out the possibility of assimilation, coexisted with the then advancing discourse on the innate differences of 'races' refuting the possibility of equality. Coloured identities were in part shaped by the interplay of these discordant discourses. On the one hand, it produced a cultural schism that overlapped with class in the community. Those with the necessary resources who were eager to acquire the status of 'civility' and access to the rights of citizenship deemed to accompany this designation, cultivated and sported the etiquette prerequisite for assimilation. For the majority this project was out of reach and they retained and regenerated the socialization patterns that they had been acculturated into, including the continued use of Afrikaans as a first language. On the other hand, the discourse positioned people of 'mixed-race' as an 'intermediate' racial group innately different from both black and white. The approximation of European codes of behavior therefore did little to exclude the emerging elite's representation as an 'Other'. 2 3

See Hendricks (2000) for a more in-depth analysis of these processes. These concerns are dealt with by Adhikari (1992, 1993,1997).

116

Cheryl Hendricks

In the early twentieth century, leaders in the community, such as Dr Abdullah Abdurahman, 4 still punctuated their speeches with the need for Coloureds (a term which was by then in common use) to act 'civilized' and continued to make appeals to the government for the extension of the franchise on the basis of their 'civility'. Mimicking colonial culture in the hope of assimilation was a trend throughout the colonial world, a trend offset only by the continued 'othering' and oppression of the colonized. This too was the case in South Africa. But here the emergence of a common black identity was circumvented by the tri-racialization of the society, the divergent history of Coloureds at the Cape and the differential oppression of Africans. At the turn of the twentieth century Segregation (stipulated as a formal policy, for it has always been de facto) was ushered in, cementing the new nation-state on the basis of racial exclusion. White hegemony was consolidated through a series of laws disenfranchising Africans, restricting urbanization, racially spatializing urban residence patterns, limiting African employment to the unskilled category, and so forth. In the Cape, the Natives Location Act of 1899 legislated the removal of Africans from District Six (a mixed inner-city suburb). They were relocated to an area on the outskirts of the city. Coloureds, mindful that the 'cleansing'/ 'bleaching' of the city might soon incorporate their removal and anxious that the pending unification of the country would erode their limited rights, began to mobilize collectively. The African Political Organization was formed in 1902 championing Coloured interests. 5 The construction of identity is always relational and tied to the existing power relations. To avoid being subjected to the gamut of discriminatory legislation applicable to Africans, it became prudent for Coloureds to insist on their 'difference'. Race was accentuated as the marker of 'difference', for a racial ideology was the overarching framework in which power relations were determined. Culture in itself was deemed as a product of one's blood rather than one's socialization. The national state, operating with the logic of divide and rule, provided the 'opportunity structures' 6 for Coloureds to maintain a distinct identity. It procreated the identity through the official use of the designation Coloured (the term had been in common use since the mid-nineteenth century) to denote those of 'mixed descent', the creation of institutions specifically for Coloureds and by continuing to place them in an ambiguous position vis a vis both Africans and Whites. For example, though discriminatory legislation was being applied to them, it was not as austere as that against Africans and government representatives also continued to hold out the possibility for Coloureds to be included as full citizens of the state, as discernible in the debates on the 'Coloured Question'. The framing of, and desire to maintain, Coloured 'difference' can be seen in the following quotations from the APO newsletter of 24 May 1909 and speeches by Dr Abdurahman: Everyone is well aware that in South Africa there is a large population of coloured people as opposed to natives . . . They are the product of civilization - in its most repellent manifestations according to some. They are of varying degrees of admixture. Their complexions vary from the black skin of the kaffir to a light tint that hardly discloses any trace of the Negro . . . and their mode of life conforms with the best European model, (cited in Adhikari, 1993a: 99) 4

Leader of the African Political Organization, formed in 1902. The naming of the organization seems to indicate a broader-based movement. But from the campaigns and speeches of the organization its concerns were primarily related to Coloureds. 6 To use Esman's (1994) concept. 5

Coloured Identity & the Rainbow Nation

117

The black races must be allowed to develop in their own way . . . We have a deep interest in the native races of South Africa . . . but it is my duty as president of the APO, on the present occasion, to deal with the rights and duties of the Coloured people of South Africa, as distinguished from native races. (1910 Presidential address reprinted in Van der Ross, 1990) It [is] not the aim of the conference [Non-European conferences] to attempt to break down group consciousness. That would be fatal, (cited in Van der Ross, 1986: 100) From the above we note that Coloured leaders bought into the discourse of separate races and that they did not see their aim as breaking down group consciousness. However, they did not see 'civility' as innately determined by race nor did they advocate that Africans should be excluded from the rights of citizenship. By 1948, when the National Party assumed the mantle of governmental power, Coloured identity was already an embedded identity. Neither the National Party nor previous governments created this group simply by 'governmental decree'. But successive regimes certainly created the contexts in which this identity could be socially constructed. In the era of Apartheid the identity is reproduced and bolstered through the enactment of laws requiring group identification (the Population Registration Act), ensuring residential and social separation (Group Areas Act and other petty apartheid legislation), forcing inbreeding (Prohibition of Mixed Marriages combined with an Immorality Act), through material rewards achieved via the CLPP (the restriction of Africans into the Western Cape, and the provision of differential educational, residential and employment opportunities for Coloureds relative to Africans) and the creation of Coloured political institutions. These factors gave rise to a solidity of effect of Coloured identity, i.e., a social and economic materiality. And through these processes both the state and Coloureds gained an interest in their differentiation. Two key points emerge from the above discussion: (i) that agency in the framing of Coloured identity belongs to both the state and Coloureds themselves and (ii) that this enunciation was possible because of the specificity of the Cape's history and processes of racialization. Though identity brokers can shape identity construction, they cannot will it out of nowhere. Not all those so defined necessarily accept their constructions as the intense debates about the nature of Coloured identity and the methods for political mobilization attest (see Hendricks, 2000; Lewis, 1987; Goldin, 1987). Intragroup tensions in the Coloured community, prior to 1994, can broadly be categorized as between those who sought to mobilize on an ethnic/racial basis and those political organizations which sought a broader alliance with other organizations, and, from the 1970s, those who sought to cast off the identity and those who sought to retain it (this apart from the finer ideological differences). It was therefore at first the political significance of the 'difference' which was in dispute, and later the identity itself. The remaining sections explore the articulation between, first, Coloured identity and anti-apartheid discourses and, second, Coloured identity and post-1994 nation-state-building discourses and practices.

118

Cheryl Hendricks

Re-framing Coloured Identity: 1970s-1994 The Black Consciousness Movement (BCM) The Black Consciousness Movement, formed in 1969, was pivotal in introducing the conceptualization of a broader black identity for the oppressed. The BCM sought to achieve the psychological liberation of blacks through positive affirmation of the aesthetic of blackness and a re-evaluation of their history, the creation of black unity, and the overthrow of Apartheid. For the BCM, black identity was constituted through one's structural position in the South African society, thus enabling the extension of the identification to Coloureds and Indians. The BCM's vision of blackness resonated amongst many Coloured intellectuals. They embraced the identity as one which provided psychological empowerment and an escape from the marginality which Coloured identity conferred, i.e., being situated in the interstice of the dominant white and African identities and being peripheral to a struggle largely couched in binary (black vs white) terms. Henceforth, when forced to use the designation Coloured, they would put it in inverted commas to signify their rejection of it. But the majority of Coloureds did not participate in the political project of the Black Consciousness Movement and did not identify with the identity re-inscription. They continued to identify themselves as Coloured, and their ability to access un/semiskilled and service sector employment and live in moderately better residential areas, in the Western province in particular, was premised on their being Coloured. The centuries of negative stereotyping of Africans, their differing socio-cultural experiences, and a by now internalized racism, inhibited a desire to embrace a black identity. Those Coloureds who embraced a black identity did not necessarily view this as an exclusive identity, nor was the re-inscription necessarily a fixed one. Identification varied in relation to differing political conjunctures, opportunity structures, and social spaces. For example, the Reverend Alan Hendrickse, leader of the now defunct Labour Party of South Africa, at first defined himself as black, but when the tri-cameral parliament was formed (in 1983) and the opportunity was afforded him to mobilize a Coloured constituency and enter the portals of power on this basis, he reclaimed his Coloured status. Also many political activists, as discussed below, re-asserted their Coloured identity in the post-1994 era, several for less opportunistic reasons. The political conjuncture allowed for a diffusion of the strategic homogenizing identity narratives of the apartheid era. One could also traverse identities by being Coloured in some contexts and black in others. For example, at home and in 'Coloured spaces' one could easily become one of the group enjoying the familiarity that accompanied it, whilst in other contexts, particularly the public sphere, one reasserted a broader black identity. 7 Identification is not wholly a personal exercise, for one needs to be accepted as such by others and this too is contextual. Scholars therefore need to take account of the difference between identity in the private and public realms and the distinction between social and political identities. In the 1980s the discourse of non-racialism eclipsed that of black consciousness. The United Democratic Front (UDF), which popularized the discourse, gained widespread support amongst Coloureds. But, as we outline, this appeal was based more on the methods of mobilization than on any fundamental alteration of racial/ethnic consciousness. 7

See Zimitri Erasmus' narrative of negotiating her identity in Erasmus (2001).

Coloured Identity & the Rainbow Nation

119

Non-racialism Non-racialism's discourse has largely centered around rights, positing that an 'individual's citizenship, legal rights, economic entitlement and life chances should not be decided on the basis of "racial ascriptions "'(Marks, 1994: 2). This was an ambiguous discourse, for it simultaneously highlighted the fallaciousness of racial categories yet was forced to use those racialized categories in the framing of the struggle. Proponents of the discourse were able to disaggregate a racialized white category into supporters and oppressors, but they sought to homogenize the oppressed as exclusively Black. Those who did not see themselves as such were said to be suffering from a false consciousness. And, though they denied the existence of a Coloured identity (and note that there has never been a similar disavowal of Indian identity - its 'difference', though noted as constructed, was appreciated), they also played to its idiosyncracies by employing 'Coloured' activists to attract Coloured support for the liberation movement, i.e. a simultaneous cognizance and rejection of the identity. The United Democratic Front (UDF), launched in 1983, succeeded in mobilizing a broad cross-section of the South African populace, and it is the support afforded to this organization that led to the belief that Coloureds (as a group, rather than merely individuals within it) transcended their racial/ethnic identity. Activists also assumed that the support for the U D F would translate into a vote for the ANC. Steven Mufson remarked that the formation of this organization marked an era when 'ethnic politics - "Black politics", "White politics" "Coloured politics", and "Indian polities'" - became simply 'South African politics' (Mufson, 1991: 3). This negation of sub-identities was a widely held presumption. Part of the explanation for the UDF's success in obtaining large-scale Coloured support can be attributed to its organizational structure and method of mobilization. The UDF's membership was afforded on an organizational basis. The front attracted a host of social movements that had mushroomed in the early 1980s: organizations concentrating on improving the material conditions in either the workplace or residential areas in which they were located, for example, rent increases, evictions, unfair dismissals, salary negotiations, student concerns, and so forth. The issues dealt with were immediate localized ones with a direct appeal to a broad majority enduring the hardships of the 1980s economic recession. Though the U D F could boast a 'non-racial' following, on closer inquiry the front was trapped, in similar vein to all other organizations that attempted to break down the racial barriers of their support base, by the confines of Apartheid's racial spatialization. The majority of associations belonging to the U D F were community-based and confined by the Group Areas Act to being largely racially homogenous. Individuals belonged to civic associations within their own communities and it was the leadership of these organizations that had frequent contact, at regional and national levels, beyond their racially defined groups, not rank and file members or supporters. Whilst many in leadership positions identified themselves as black, or just as South Africans, the majority of Coloureds were seldom engrossed in posing existential questions. Their concerns were related to improving their life chances, not the finer ramifications of the concept of non-racialism. Edgar Pieterse, a U D F activist noted: 8 8

Edgar Pieterse (1994) 'The treacherous turn of the Coloured vote in the South', cited in Crommelin (1997: 65).

120

Cheryl Hendricks

In fact, when we pronounced that non-racialism and oppressed solidarity were won during the peak of the U D F era, it was probably more wishful thinking and projection because, from the vantage point of our activist sub-culture, the world, and especially the Cape Flats, did indeed look like a very revolutionary place. We were probably our own biggest converts . . . The fact is that often we did become weird, radical, communist activists, distanced from the sentiment, morality and culture of our communities.

Furthermore, although the U D F popularized non-racialism, it hardly interrogated identity issues. Marx (1992: 16) had commented that 'the U D F opposed the physical aspects of national domination and accordingly was concerned less with changing individual consciousness or identity than with mobilizing material resources and followers across perceived racial divisions'. U D F activists deemphasized identity issues and highlighted common oppression and opposition to apartheid in order to attract support. They then argued teleologically that support against apartheid also implied the acceptance of a common identity and support for an ANC-led government. In the minds of many Coloureds the ANC and U D F were two distinct organizations. The ANC was perceived as an African organization championing black (in its narrow usage) majority rule. And since the majority did not perceive themselves as such, they concluded that it did not represent their interests. Although the policies of the U D F and ANC were similar, the U D F had the advantage of being an internally based movement, generating a familiarity with the organization, and it had a leadership profile which portrayed the kaleidoscope of identities in South Africa. Representation of the ANC, for Coloureds, was largely gleaned from the Afrikaans media, in which the ANC was portrayed as a 'black terrorist/communist organization' and therefore offensive to many Coloured sensibilities. As the 1980s drew to a close the realities of identity fragmentation, within the oppressed category, became evident. By the late 1980s the ANC had become a visible presence in South Africa and the apartheid regime, unable to stem the tide of protest, appeared willing to negotiate itself out of exclusive political power. Politics swiftly shifted from the local to the national level, with a simultaneous feeling of marginality setting in amongst Coloureds. The same ambivalence that characterized their perception of place in the preceding periods was transferred to the post-apartheid era, only this time the fear of being dominated by an African majority loomed large and appeared more threatening than one of domination by whites. This fear stemmed from the rapid demographic change in the Western Cape and their own sense of subjectivity, and it was fueled by the shenanigans of the National Party and the neglect of the ANC. The Western Cape, because of its historical past and the functioning of the Coloured Labor Preference Policy (CLPP), was a predominantly Coloured residential and employment space. However, between 1982 and 1992 Cape Town was one of the fastest growing cities with an annual population increase of 13 per cent (cited in Finnegan, 1994: 34). The city's African population more than tripled, and along with the new demographic make-up came racialized tensions, as groups competed for employment, housing, school and university placements, scholarships, and so forth. The scale of the urban transformation of the Cape seemed to overwhelm many Coloureds previously sheltered by the CLPP, and they began to retreat into a laager mentality. As at the turn of the twentieth century, the assertion of racial/ethnic 'difference' by Coloureds was a consequence: 'us Coloureds' and 'the blacks' became frequent echoes. In the new

Coloured Identity & the Rainbow Nation

121

ostensibly deracinated space racial/ethnic 'differences' gained saliency. The early 1990s also saw the transformation of the N P from an exclusively white organization to a 'multi-racial' one actively courting Coloured and Indian support. Reinvoking the construction of the 'Brown Afrikaner'9 or 'step-child', the NP sought to imprint an affinity between Afrikaners and Coloureds and conversely amplify differences between Coloureds and Africans (a similar tactic to that employed in the era of Segregation). Seizing on the disquiet amongst Coloureds, the N P played on their paranoia of being subordinated by Africans. Whilst the ANC was still debating which Coloured 'leaders' it should use to attract the Coloured vote and whether or not it should address Coloured fears, there was a growing reverberation amongst Coloureds of 'first we were not white enough, now we are not black enough' and a plea to the then president De Klerk 'Papa Save Us'. The tables had turned: oppressors were now viewed as potential saviors, and emancipators as future oppressors. The NP's trope of an extended family who needed to put aside the past and unite to protect themselves in the future held sway over the undifferentiated delivery-oriented policy approach of the ANC, amongst a large proportion of Coloureds. The N P gained approximately 60 per cent of the Coloured vote in the 1994 national elections and maintained control of the provincial government in the Western Cape. Many pollsters dubbed the elections a 'racial census' indicating the correlation between race and party support. However, amongst Coloureds, class mediated voting patterns. It was the working class who felt most vulnerable within the new socio-political environment, and it was primarily they who afforded the N P a large Coloured constituency, whilst the vote of the middle class was split between the three major parties, ANC, NP, and Democratic Party (DP). Identity construction and contextual politics account for the shift in Coloured attitudes. In the 1980s the sense of 'we-ness' was a tenuous one, with oppression as the common denominator. As power relations were being reconfigured, latent 'difference' became heightened, with renewed fears of suppression crystalizing. There was, and largely remains, a visceral lack of 'social trust' (to employ Berman's use of the term in this volume) on the part of Coloureds towards Africans. At conjunctures of rapid social change, such as the 1990s in South Africa, the uncertainties and insecurities elevate awareness of 'difference'. The effect is a retreat to the familiar, an ethno-space which simultaneously provides a protective sphere and needs to be protected.

The Rainbow-Nation Configuration and Coloured Identity Reconstruction This section addresses the ways in which Coloured identity expression articulates with post-apartheid nation-state-building discourses. Central to this inquiry are the tensions between 'difference' and 'unity' and 'race-consciousness' and 'non-racialism'. 10 The metaphor of the 'Rainbow-Nation' was employed by the post-apartheid state to capture its vision of a multicultural society united by common constitutionally entrenched principles (democracy, non-racialism and non-sexism) that would serve as a foundation for building a unified nation (united in their quest for territorial integrity and the emergence of a transcendent South African identity, and in their approaches to structural 9

A paternalistic portrayal used to gain Coloured votes in the era of Segregation. See Halasi (1998) for a more in-depth interpretation of black politics by looking at the tensions between non-racialism and race-consciousness and liberal versus republican citizenship. 10

122

Cheryl Hendricks

transformation). In pursuit of this ideal the government made concessions to Inkatha and the Afrikaner right wing; legislated a 'reveal and forgive' catharsis through the establishment of the Truth and Reconciliation Commission; set up a Pan South African Language Board entrusted to promote all languages, and recognized eleven languages as official; established various Commissions to oversee the transformation of relations in regard to human rights, culture, gender and land; and created the fetishes needed for symbolic representation of the nation (anthem, flag, monuments and so forth). The tensions of the new state are connected to its aspirations of simultaneously proffering a political community based on individual rights and group recognition, building non-racialism through race-based policies designed to correct previous inequities, and constructing a common South African culture through an affirmation of previously publicly marginalized cultures. The contradictions emerging in this process often serve to reinforce cleavages. With the advent of the post-apartheid state a number of Coloured political parties and cultural movements surfaced, seeking to authenticate the identity and find a place for it either within the kaleidoscope of the Rainbow Nation, or through secession from the embryonic nation. The re-framing of the identity can be gleaned through the social memory informing the development of the Kleurling Weerstand Beweging (KWB), the December 1st Movement, and the Cultural Heritage Development Council, and in the commemoration and memorialization of events, spaces and people associated with the group. The linkages to slavery and a Khoisan heritage became central to the re-framing of Coloured identity. However, the reclaiming of these 'lost' histories also accentuated the divisions within the Coloured community. The KWB, formed in February 1995, represented an extreme ethno-nationalist portrayal of Coloured identity. Mervyn Ross, a founding member and chief spokesperson (and a previous ANC activist) stated: We are proud that we are ethnic. And once we are ethnic and being recognized by various other people, we can also go further and say, 'Look, we are ethnic. We have our own language, our own culture, our own land and we want to govern ourselves.' We are not prepared to be governed by the white man anymore - he made a mess of it for 300 years. We are not prepared to be governed by black people, (cited in Caliguire, 1996: 10) The confluence of race and ethnicity in the formation of Coloured identity is evinced in this confusing account of what the identity is and its political implications. The stress on 'we are ethnic' is a double manoeuver intended to legitimate group recognition and seek self-determination (which is internationally legitimated through a discourse of culture rather than race). The reason conveyed for secession is race-based, i.e. not wanting to be governed by blacks. The KWB rested its claims of cultural and land rights on an excavation of a Khoisan heritage. The organization received little support for, though many Coloureds were sympathetic to the sentiments expressed, few wanted to establish their own 'Colouredstan.' However, the re-presentation of Coloured identity from one framed around a discourse of race to one imbued with more cultural associations became a prominent undertaking. The Cultural Heritage Development Council, the Brown Movement, the Griqua National Conference of South Africa, and others, soon emerged as organizations seeking either to uncover their cultural pasts or to claim rights on the basis of this past. Remnants of the Khoi and San communities demanded

Coloured Identity & the Rainbow Nation

123

first nation status (similar to Native Americans) and the right to their ancestral land. The propriety and property of representations of the Khoisan were questioned, for example, their display in museums, and in the debate on the Miscast exhibition as to the validity of a white woman re-presenting Khoisanness. They can also be seen in the demands for the return of Saartjie Baartman's remains from France. The revival of these identities was evident at the widely publicized conference in July 1997 that brought together, from as far afield as Angola, and displayed people of Khoi and San heritages. One of the organizers of the conference noted 'We hope this conference will start a debate in South Africa about the people indigenous to this region and what their rights are as people.' {Cape Times, 14 July 1997). This renaissance of Khoi and San identities is also linked to a global network of indigenous/aboriginal communities seeking restoration and retribution. Crawhall (1998) estimates the number of Khoisan people living in South Africa at about 5000. A contradictory process was, however, under way amongst these organizations, for the Khoisan representatives sought to reclaim their distinct identity (as opposed to being labeled Coloured) and the accompanying rights they deemed their groups entitled to. But organizations like the KWB and the Cultural Heritage Development Council (CHDC) sought to use Khoisanness to bolster cultural claims of a Coloured identity. The CHDC saw its task as: 'to foster unity among historically Coloured people and give them pride in their origin' (Weekly Mail, 25 July 1997). Joe Little, a founding member, bestowed chieftaincy status on himself and selected other chiefs, chieftainesses and headmen (constructionism in its most banal form). They could often be seen parading around Cape Town in 'traditional' dress (replacing their 'dashikies with the loin cloth') in an attempt to make visible their reconstructed identity. The difficulty with this kind of representation is that it essentializes a group whose cultural sensibilities arise from lived experiences rather than genealogy. Not all Coloureds have Khoisan ancestry and neither do those who may have it necessarily want to highlight this aspect of their identity. The negative imagery of the Khoisan still lingers in the Coloured community, making it a challenge for an endearment of this construction. 11 The December 1st Movement was formed in 1996 by a group (many of whom were ANC activists) disillusioned by the ANC's neglect of Coloured concerns in the 1994 election campaign. The naming of the organization reflects their use of slavery as a common marker for Coloured identity: on 1 December 1834, slaves were emancipated. However, to avoid accusations of essentialism, they asserted that: we do not wish to celebrate a position of slavery, nor do we wish to dwell on the pain of slavery . . . or suggest that all Coloured people are the descendants of slaves, [furthermore] That the name claims all our forebears: the slave and the free, the European and the African. It is a past that has been hidden in shame that we are now uncovering with pride. Many in the slave community were skilled artisans, others were political prisoners . . . Their wisdom subverted the order of the day inspiring various forms of resistance.12 Here, too, there was a search into history to capture memories which would simultaneously serve as cultural glue for the group and insert the group into the 1 * The term 'Hottentot' (usually expressed in Afrikaans - 'Hotnot') is still used amongst Coloureds to deride each other. 12 Draft Statement of Intent, 'Freedom Is For Us All', of the December 1st Movement.

124

Cheryl Hendricks

South African structural past and identity mosaic. Reinvoking memories of a slave heritage is, however, also a thorny task. Slavery is a memory filled with pain and shame which Coloureds have largely sought to erase. There has not been in South Africa the same interrogation that places the burden of the history of slavery on the shoulders of the slave masters rather than the enslaved, which has occurred in other previously slave-based societies. It is also a history that distinguished Coloureds from other Africans; hence the reluctance by the liberation movement to lay emphasis on this aspect of colonialism during the struggle (see Ward and Worden, 1998). Recapturing the slave past also elicits religious differences (Muslim/Christian) within the group, for example, as displayed in the 'Sheikh Yusuf Tricentenary Commemorations' held just prior to the April 1994 elections. Sheikh Yusuf, a Muslim leader from Java, exiled to the Cape as a political prisoner in the 1690s, is seen as a 'spiritual liberator of slaves through the propagation of Islam' (ibid: 213). The commemoration marked an assertion of Muslim identity in South Africa and an attempt to posit the identity as a 'Malay diaspora'. 13 The multiplicity of slave identities would therefore bring into dispute what needs to be remembered for what purpose in the present, i.e., a broader Coloured identity or a narrower Muslim/'Malay' identity. Nevertheless, the slave past is being commemorated, as it rightly should be, through museums, tourist attraction 'slave routes,' and so forth. Slavery is something not only Coloureds have to come to terms with but all South Africans. How South Africans should employ the knowledge and meanings of this past in the present and future is, however, what is at issue. For the December 1st Movement slavery, because it produced 'racial hybridity', becomes almost venerated. Its members asserted in their Statement of Intent that Coloured identity is 'uniquely South African' and that Coloureds represented 'a symbol of a rich future in a South Africa where race plays no role'. We argue to the contrary; the identity itself is indicative of how much of a role race has played both in the period of slavery and thereafter. The larger point is that they are trying to provide indigeneity of, and future relevance for, the identity. The movement defined its purpose as seeking recognition of Coloured identity and to transcend their socio-political marginality. Like the KWB, this organization was short-lived, disappearing in 1998. It was subjected to a wide range of criticisms, ranging from simply being dismissed as a group of disgruntled activists who had missed the 'gravy train' (euphemism for those who obtained government posts) to more substantive ones of resorting to ethnic mobilization. The importance of these organizations is that they turned the spotlight on the need to engage with the historicity of Coloured identity and their contemporary anguish. What was taking place at the national level that produced this search for authenticity and the political salience of Coloured identity? Part of the explanation can be located within the conceptualizations of the 'Rainbow Nation', policies promoting diversity and the need for race-based policies for advancement. Prior to the 1990s, those organizations that advanced the exclusive interests of Coloureds were ostracized, and claims of self-determination would have been tantamount to political suicide. In the immediate aftermath of apartheid an environment was created in which minority groups felt both the need to, and more enabled to, assert ethnic claims. 13 As noted by Ward and Worden (1998). Also, only a small proportion of East Asian slaves were from Malaysia.

Coloured Identity & the Rainbow Nation

125

The post-apartheid state, recognizing, and being somewhat conditioned by, the global celebration of diversity, reveled in its multi-cultural nature. Both the interim and the revised constitutions gave prominent protection to cultural rights. Political representation at the local and provincial levels was afforded to 'traditional leaders' who, through the formation of the Council of Traditional Leaders (CONTRALESA), have become an organized political force. Culturally based identities were therefore given protection, representation, and public visibility: culture gained prominence as a signifier of belonging and as a mechanism for validating resource claims. However, Coloured identity, as an internally culturally fragmented and racialized identity, has always been perceived as lacking cultural validity and the very existence of the group, qua group, has been challenged. By reclaiming slave and Khoisan heritages, Coloured identity could be invested with the 'cultural capital' deemed necessary to avail them of the new 'opportunity structures.' Moreover, through the employment of a Khoisan identity they could affirm their indigeneity and equal right to belong to the nation, the perceived need to do this speaking volumes about their sense of insecurity. The Khoi and San cultures have been constitutionally recognized but not their demand for first nation status. But the collapsing of these identities with Coloured identity has not taken root in the broader community. The community(ies) seek no such authentication; for them, as opposed to those segments which recognize the limitations and precariousness of a largely racially differentiated community, Coloured identity is not problematic. It is their differential past and present lived experiences, which produced a sense of 'we-ness', and that they want to have acknowledged, not a retention of slave memories or Khoisan culture. In addition, the lamentations of many Coloureds of 'first we were not white enough, now we are not black enough' have little to do with wanting the designation of either category. They stem more from a perceived position of discrimination and marginalization. Attempting to redress past imbalance, the ANC adopted an Affirmative Action policy and enacted the Employment Equity Bill that would facilitate the upliftment of previously disadvantaged collectivities. Given the legacy of apartheid, this would necessarily be race-based: it was seen as a temporary necessity in order to induce more equitable power arrangements in a non-racial society. The policy seems to have increased racial awareness, with a number of 'non-blacks' feeling that their citizenship rights have been diluted. For Coloureds, their differing identity constructions and internal class inequities impact on the way in which they relate to Affirmative Action as part of a strategy for nation-building. For those who see themselves as Coloured, Affirmative Action is seen as beneficial only for Africans; they are, for the most part, the majority working class who do not have the necessary education or skills to take advantage of the opportunities. For those who do have the skills the affirmation is evidenced in their rapid social mobility, though they too echo a disgruntlement with employers' preference for a bodily encoded blackness quite easily acquired through a camouflaging of racialized criteria via a stipulation of the language required. The above raises questions of how to address the concerns of Coloureds and their socio-political future in the new nation. Coloureds need to come to terms with the political reality, the demographics and the competitive environment of post-apartheid South Africa. By the same token the state needs to take into account the historical factors that have produced and contained the group (much like the other groups), that account for their latent racism, and structure

126

Cheryl Hendricks

their present position. Their lack of constituting a national political threat to the government is not a sufficient reason to ignore their concerns in a country that prides itself on creating an equitable socio-economic and political space. In the 1999 election campaigns, the ANC began to revise its strategies in relation to the group. It reverted to the game of ethnic politics by placing 'Coloured' leaders into prominent positions in the Cape province and courting high-profile Coloured politicians who, in the 1994 election, had helped secure the Coloured vote for the NP. Ebrahim Rasool, who became provincial chairperson, campaigned under the slogan 'A Home for All' and emphasized the issues deemed important for the community, namely, crime, education, pensions, housing, land and unemployment (cited in Lodge, 1999: 76). There was some dividend to this strategy as the ANC's vote in the national elections increased from 4 per cent Coloured support, in 1994, to 7.5 per cent in 1999, whilst major inroads were made in the Western Province provincial elections, though not enough to secure legislative control (see Reynolds, 1999). Though it is symbolically important to have Coloured representation in the governing structures of both the Party and the state, it is more important to seek to create the conditions that reduce the fear and make bogus the racism that underpins some of the political expressions of this identity. The 'moral ethnicity' of Coloureds translates into 'political tribalism' when they feel the need to legitimate their identity or transform themselves to become politically acceptable, or are marginalized by discourses and policies operating on the assumption of a shared identity.

Conclusion This chapter has examined the processes of Coloured identity construction and its interrelation with practices of nation-building/state-society reconstruction. By emphasizing 'unity'/non-racial subjects, the liberation movement obscured the very real historical differences between oppressed groups in South Africa. In the simultaneity of 'unity' and 'difference,' 'non-racialism' and 'race-consciousness', the government's nation-building project highlighted the precariousness of Coloured identity. South Africa's burdens of the past and challenges of the present, although having their own peculiarities, bear a marked resemblance to those of other African countries, i.e., the burdens of institutionalized 'difference', cultural/racial inequities, homogenizing nationalist discourses, and a politicized scholarship which prematurely attempted to dig graves to bury ethnic allegiances. The challenges of the present are to find ways in which to constitute viable political communities that are able to deal with claims for group recognition. There are no easy answers or single prescriptions. The management of these competing claims will, and should be, worked out by those implicated in the process. In attempting to find ways in which to affirm multi-culturalism and promote nation-building we need to understand the processes of identity construction and the conditions that induce politicized ethnicity. This study of Coloured identity indicates that, despite the fluidity and malleability of identities, they acquire a solidity of effect that needs to be engaged, not simply dismissed in the hope that the identity will automatically fade into oblivion. The malleability of social identities is circumscribed by the historical and current socio-political conditions and discourses enunciated to legitimize the contexts. A democratization process that ignores historically constituted 'difference' and that conse-

Coloured Identity & the Rainbow Nation

127

quently leaves some groups feeling peripheral will induce the kinds of identity closures and laager mentalities exhibited by the Coloureds, and equally discernible amongst Indians and Whites. The challenges of finding appropriate context-specific models for managing the tensions between the universal (nation) and the particular (sub-identities), will likely preocupy policy-makers and scholars for the foreseeable future. What is evident, from a century of dealing with these problems, is that unmediated applications of liberalism, with its proclivity towards individualism and assimilation, as well as ethnonationalist discourses and practices are both problematic for pluralistic societies.

References Adhikari, M. 1992. 'The Sons of Ham: Slavery and the Making of Coloured Identity', South African Historical Journal 27:95-112. — 1993a. 'Protest and Accommodation: Ambiguities in the Racial Politics of the APO, 1909-1923', KRONOS 20: 92-106. — 1993. 'Let us live for our children9: The Teachers'League of South Africa, 1913-1940. Cape Town: Buchu Books. — 1997. 'Voice of the Coloured Elite: APO, 1909-1923' in L. Switzer, ed., South Africa's Alternative Press: Voices of Protest and Resistance 1880s-1960s. Cambridge: Cambridge University Press. Bayart, J-F. 1993. The State in Africa: The Politics of the Belly. London: Longmans. Caliguire, D. 1996. 'Voices from the communities' in W. James, D. Caliguire, and K. Cullinan, eds, Now That We Are Free: Coloured communities in a democratic South Africa. Cape Town: Creda Press. Crawhall, N. 1998. 'Still Invisible: San and Khoe in the New South Africa', Southern Africa Report 13 (3). Crommelin, C.A. 1997. 'Brown People in the Grey Area: Ethnic politics and the Coloured communities in the South African Western Cape'. Ph.D. Thesis, Instituut voor Sociale Geografie, Universiteit van Amsterdam. Ekeh, P. 1975. 'Colonialism and the Two Publics in Africa: a Theoretical Statement', Comparative Studies in Society and History', 17 (1). Erasmus, Z. 2001. 'Re-imagining Coloured Identities in Post-Apartheid South Africa' in Z. Erasmus, ed. Coloured By History, Shaped By Place. Cape Town: Kwela Books. Esman, M.J. 1994. Ethnic Politics. Ithaca, NY and London: Cornell University Press. Finnegan, W. 1994. 'The Election Mandela Lost' New York Times, 20 October. Goldin, I. 1987. Making Race, The Politics and Economics of Coloured Identity in South Africa. London: Maskew Miller. Halasi, C.R.D. 1998. 'Citizenship and Populism in the New South Africa' in Africa Today, 45 (3-4). Hendricks, C. 2000. 'We Knew Our Place!' A Study of the Constructions of Coloured Identity in South Africa'. Ph.D. Thesis, University of South Carolina. Lewis, G. 1987. Between the Wire and the Wall. A History of South African Politics. Cape Town: David Philip. Lodge, T. 1999. Consolidating Democracy: South Africa's Second Popular Election. Johannesburg: Witwatersrand University Press. Lonsdale, J. 1994. 'Moral Ethnicity and Political Tribalism' in P. Kaarsholm and J. Hultin, eds, Inventions and Boundaries: Historical and Anthropological Approaches to the Study of Ethnicity and Nationalism. Roskilde, Denmark: Institute for Development Studies, Roskilde University. Mamdani, M. 1996. Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton, NJ.: Princeton University Press; Oxford: James Currey. Marks, S. 1994. 'The Tradition of Non-Racism in South Africa' Paper presented at the History Workshop 'Democracy: Popular Precedents, Practice, Culture'. University of Witwatersrand. 13-15 July. Marx, A. 1992. Lessons of Struggle: South African Internal Opposition, 1960-1990. Cape Town: Oxford University Press. Mufson, S. 1991. 'Introduction: The Roots of Insurrection' in T. Lodge and B. Nasson, eds, All Here, And Now : Black Politics in South Africa in the 1980s. Cape Town: David Philip. Reynolds, A. ed. 1999. Election '99: South Africa From Mandela to Mbeki. New York: St. Martin's Press. Van der Ross, R. ed. 1986. The Rise and Decline of Apartheid: A Study of Social Movements among the Coloured People of South Africa, 1880-1985. Cape Town: Tafelberg.

128

Cheryl Hendricks

Van der Ross, R. ed. 1990. Say It Out Loud - The APO Presidential Addresses and Other Major Speeches. Cape Town: The Western Cape Institute for Historical Research, University of the Western Cape. An in-house publication of the Institute. Ward, K. and Worden, N. 1998. 'Commemorating, Suppressing, and Invoking Cape Slavery' in S. Nuttall and C. Coetzee, eds, Negotiating the Past: The Making of Memory in South Africa. Oxford and New York: Oxford University Press.

8

JACQUELINE S. SOLWAY

Reaching the Limits of Universal Citizenship: 'Minority5 Struggles in Botswana

I

N 1965, on the eve of Botswana's independence, the residents took to the polls to elect their first national government. To train people in this fundamental civic responsibility the Bechuanaland Protectorate administration prepared special materials: two films, 'Mosupi Claims his Vote' and 'Mosupi Casts his Vote', and a poster. Six Land Rovers, four of which were fitted with projectors and described as mobile cinemas, showed the films and the posters were widely distributed (Winstanley, 1965). In March 1965 74 per cent of the registered electorate (representing 58 per cent of the potential total) voted (Botswana Parliament, 1996), launching one of Africa's greatest successes in liberal multiparty democracy. For the purposes of this chapter, what is important to note is that the materials were prepared in a number of languages; versions of the films were in Setswana, English, Afrikaans, Ikalanga, and Otjiherero and the posters were in Setswana, English and Ikalanga. After independence English became the official, and Setswana the national, language. Government affairs, the official media and education were then limited to these languages, a fact that, in spite of recent government agreement in principle to change the existing policy, persists. The exclusion of other languages - now called 'minority' languages and spoken by 'minority' peoples - from national recognition is symbolic of wider exclusionary practices that are increasingly being seen as limiting, if not perverting, Botswana's democracy. This chapter examines the rise of minority struggles in Botswana, a relatively recent but steadily intensifying phenomenon. The topic is of interest because of the increasingly significant role minority/majority discourse plays in the expression and manifestation of political strife on all continents in the postCold War era. In addition, the timeliness and importance of the issue are highlighted by the Botswana case study, not because they are especially typical or emblematic of minority and/or ethnic struggles, but rather, because of their socalled 'exceptionality' (Good, 1992), especially in the African context. In much of Africa, ethnic conflict is attributed to state weakness. Patronage politics, imposed austerity fiscal programmes, corruption, and so on leave states unable to deliver services efficiently and contain internal competition and conflict. Politicized ethnic consciousness is frequently represented as a fall-back posi/ wish to thank Michael Dingake, Moitaly Dikgokgwane, Deborah Durham, Donald Kgathi, Danger Kgogwane, Michael Lambek, Bathlalhefi and Doreen Moeletsi, Serara Selelo-Kupe Mogwe, Gordon Mokgwathi, Patrick Molutsi Lydia Nyati-Ramahobo, Itumelang Sabone, Gao Selolilwe, Gloria Somolekae, and Jeff Tsheboagae for sharing their time and knowledge with me. I am fortunate to have had the opportunity to present an earlier version of this paper at the extraordinarily stimulating Queen's University Conference on Ethnicity and Democracy organized by Berman, Kymlicka, and Eyoh. I thank all participants for the rich and useful discussion. I wish to thank Trent University and SSHRC for providing me with research funding.

129

130

Jacqueline S. Solway

tion, a residual, reinvented, or reinvigorated atavistic force, but powerful nonetheless in providing people with meaning, structure, and the means to obtain material support in the absence of a functioning state. It is there to be activated when more 'modern' edifices of order fail. Botswana defies this stereotype; it not only provides a powerful and meaningful civic identity for its citizens but, to a significant extent, it 'works'. It has a functioning bureaucracy that, within reasonable limits, provides a multitude of services to the vast majority of the population. Thus the growth of minority identification, contemporary resistance to continued assimilation, and the politicization of identities is interesting not for what it tells us about the failing post-colonial state but, rather, what it may tell us about the realities of modernity - on the periphery, but perhaps, also, more generally. Current theories increasingly recognize that, despite its changing form and content, difference endures (see Kymlicka and others, this volume). The question then becomes, how best can society and the state endure difference? How can the nation-state be both re-imagined and internally re-configured/structured to enable enduring difference? This chapter will not answer these questions but may provide some insights into ways of thinking about them. Minority struggles in Botswana must be considered in the context of the nation-building policies pursued by the government since independence and their contradictory consequences. In fact, the very success of Botswana's project of liberal bureaucratic rationalization has exposed its limitations, primarily certain 'illiberal' constitutional components and questionable assumptions regarding the forces of assimilation. In doing so, it has created the basis for its own challenge. These processes will be explored and illustrated below. E t h n i c H e t e r o g e n e i t y : ' M i n o r i t y a n d Majority 5 in B o t s w a n a Botswana is often considered amongst the most ethnically and linguistically homogenous countries in Africa. But, as Gladney (1998: 1) reminds us, 'majorities are made, not born . . . numerically, ethnically, politically, and culturally, societies make and mark their majorities and minorities under specific historical, political, and social circumstances.' Constituting a majority is therefore not an unproblematic process; that it has been so successful in Botswana is striking, especially as many argue that 'minorities' may, if taken together, constitute a majority of the country (Mpho, 1989; Charlton, 1993: 348; Parsons, 1985: 27). l Although all citizens of Botswana share a civic identity, the degree to which those of Tswana background (defined here according to Tswana understanding of descent) compose a majority is a contentious issue. The 1946 census, conducted during the colonial period, was the last to record 'ethnic' identity and the results of this exercise led Schapera (1952) to state that 'at least 97 different stocks are represented' in what was then the Bechuanaland Protectorate. 2 The fact that subsequent censuses have avoided ethnic identity reflects, to a 1 As used in Botswana, 'minority' has multiple meanings: it refers abstractly to relative political, jural and numerical status; and, in practice, it refers to any group in Botswana that is not one of the eight 'tribes' listed in the constitution. Generally, 'minority' groups are former subject peoples; they include all the non-Setswana-speaking groups that at some point were brought, voluntarily or by force, within the political orbit of a Tswana chiefdom and subsequently the modern state of Botswana. 2 Schapera's terms may reify and distort the complexities of local identities in Botswana but the

'Minority' Struggles in Botswana

131

large extent, the political sensitivity of the issue and the desire of the state to perpetuate its vision of ethnic homogeneity. Indeed, the degree to which Setswana is now the predominant language and the wide acceptance of the popular view of Botswana as ethnically homogeneous reflect the success of the politically dominant Tswana in asserting their cultural hegemony (cf. Parsons, 1985: 28). Like the process of establishing official nationalism elsewhere, 'Tswanafication', reminiscent of 'Russification' as emphasized by Anderson (1991), has been an attempt to create and promote a unified nation while sustaining, if not strengthening, existing suzerainty. To a great extent this endeavor has been effective but, as noted earlier, its very effectiveness has, ironically, revealed the policy's shortcomings and provided legitimate political space for its contestation. Botswana's assimilationist policy, its emphasis on individual rights, and the success of its process of bureaucratic rationalization have enabled some members of 'minority' groups to transcend certain boundaries and attain valuable positions in national society. But this process of modernity has not had a singular impact; it entails a dialectic in which both assimilation and difference are produced. It has led to a greater generalization of class distinctions so that conditions of poverty and opportunity are experienced more evenly throughout society.3 All 'ethnic' groups in Botswana are internally differentiated along class lines and certainly not all members experience discrimination in equally debilitating ways. But few citizens of minority background, regardless of achievement or wealth, can entirely escape the stigma of 'minority' status. Thus modernity has not eliminated particularistic difference and has, in many instances, contributed to a crystallization of identity on the part of both 'minority' peoples and the dominant Tswana whose hegemony is being challenged in myriad ways. Critical contradictions in Botswana's official policy towards ethnicity contribute to the fact that difference is simultaneously produced and blurred. There has been a sincere effort to downplay ethnicity's importance. The fact that censuses do not collect this information reflects this position, as does the state's official stance that ethnicity should be irrelevant for obtaining land, employment, social services, political office and so on. Yet, the neutral ethnic stance of the state is undermined by certain policies (to be discussed below) that seem to some to communicate the message that ethnic equality is fine as long as the Tswana remain 'more equal' than others and their language and culture are the ones to which others assimilate. Thus the nation exists as a site of struggle: those who are marginalized desire to reconfigure the basis of their inclusion, while those at the center attempt to maintain their place. In 1985, Parsons wrote 'What is remarkable in Botswana is how much, up till (2 contd) evidence does clearly point to significant diversity. In Botswana, as elsewhere, people maintain multiple and nesting identities and they are often multilingual. As will be discussed later, Tswana 'tribes' have, over the years, absorbed numerous immigrants, some of whom came gradually to be regarded as members of that particular Tswana group. But this did not, in all cases, necessarily mean that they forgot or rejected older or alternative identities. Some people, especially those of stigmatized origins, never came to be accepted as Tswana, but were, for administrative purposes, affiliated to a particular 'tribe' and the chief that presided over it. 3 While wealth and poverty are more evenly distributed today across ethnic lines, there are important exceptions. The Sarwa (San or Bushmen) remain amongst the most impoverished in Botswana, more wealth is concentrated in urban areas, and some rural areas, especially those defined as 'remote', have higher rates of illiteracy, destitution, under-five mortality, etc. than national averages (UNDP; 1998). In addition to the San, other minority groups such as the Birwa, Kgalagadi, Yei, Mbukushu, Subia, etc. occupy the poorest (or 'remote') regions of the country.

132

Jacqueline S. Solway

now, the legitimacy of Tswana-dom has been accepted and even supported by non-Tswana groups . . . That legitimacy is considerable at present, but has been and may be challenged later as social conditions change'(p. 28). Since Parsons wrote this, conditions have, indeed, changed. It was in the 1980s that most Tswana began to experience the benefits that were being generated by the country's mineral wealth. 4 Primary school fees were eliminated in 1980 and by the end of the decade, secondary school fees were as well. New health facilities and schools opened throughout the country and, perhaps most significant, the massive drought relief campaign that began in 1980 and has been in effect most years since has had an impact on virtually all rural citizens (Solway, 1994a). Old age pensions and orphan benefits have been introduced and formal sector employment opportunities (private sector but especially government-civil service) have also grown significantly, particularly for the educated. A new urban-based technocratic elite has emerged. 5 Urbanization has been critical in promoting minority consciousness. At independence, Botswana was overwhelmingly rural; its administrative capital was outside the country in South Africa and there were only a few small towns. What resources there were (schools, civil service centres, hospitals, etc.) were located largely in the district - that is the old 'tribal reserve' - capitals. These are Tswana-dominated areas, the seats of Tswana chiefs, and the spaces in which minorities feel most peripheral, stigmatized, and 'minor' in the jural sense of the term. Urban areas, on the other hand, are far more ethnically neutral places. Since independence, Botswana has experienced one of the highest urbanization rates in the world. 6 Town populations may reflect the ethnic groups predominant in the region in which they are located but only to a limited extent. Urban spaces have proved to be locations where minority status is less debilitating; where the publicly declared stance of ethnic neutrality is most legitimate; and urban life has provided an enabling context for minorities to join and partially 'assimilate' to national life. Thus the urban experience has contributed to the creation of a unifying national identity on the part of Botswana's citizens. However, like the process of urban migration elsewhere, it has also fostered the recognition of difference as migrants encounter others in the heterogeneous space. Botswana's economic growth has occurred in a context of relatively success4

Botswana has been the fastest growing economy in the world for the past three decades (World Bank, 1996: 189). Minerals, largely diamonds, have fueled economic growth. Diamond mining is notoriously capital- and not labour-intensive and while employment opportunities have expanded unemployment remains a serious concern. However, social services, many of which were introduced during the drought years of the 1980s, have dramatically expanded. While one only has to open one's eyes to see that, in spite of the country's wealth, there are severe pockets of both rural and urban poverty, the degree to which the country is becoming more or less unequal is a subject of debate. Hudson and Wright (1997) present figures supporting the view that inequality has slightly contracted since independence, while Good (1993) argues that inequality is deepening. 5 In 1989 Molutsi wrote 'It would appear that the political class is becoming less dependent on cattle holding and more focused on commercial activity' (125). A decade later this is even truer. Most elites still pursue a mixed economic strategy combining commercial enterprises, farming, employment, and/or civil service or political positions. In 1998, I frequently heard the rumour that Mogae (the current president - an economist who has worked for the World Bank) owns no cattle. It may or may not be true but, nevertheless, it does indicate the degree to which elite status has become uncoupled from cattle holdings. It also indicates the degree to which high political achievement is increasingly independent of the rural-based patronage links that were so vital to it in the past. 6 Just before independence in 1963 approximately 1 per cent of the population lived in towns (Campbell, 1998: 263). By 1991, the time of the most recent census, the percentage of town dwellers had grown to 45.7 per cent (Mosha, 1998: 282).

'Minority' Struggles in Botswana

133

ful bureaucratic rationalization and expansion. This has had a range of implications; for the purposes of this chapter, I wish to emphasize two. First, the importance of interethnic patron-client networks has diminished (but not disappeared). This means that members of minorities, especially those of lower status, have increasingly found themselves in a position in which they could directly access the state in order to gain rights to land, education, employment, services, subsidies, the courts, etc., instead of having to rely upon the patronage of an ethnic intermediary. Many members of minority groups took advantage of these opportunities and, as mentioned earlier, one result has been the further disassociation of ethnic and class position. 7 Second, the achievement of academic and formal sector success has had, in many instances, an empowering impact upon 'minority' groups and has led some members to re-evaluate their place in Botswana society (see Solway, 1994b). A consequence of the re-evaluation of minority status has been growing agitation for fuller representation, what Phillips (1995) refers to as 'presence' in the national sphere on the part of many minorities. This is evident in a variety of ways. In the official sphere, subaltern demands to change language policy and amend discriminatory sections of the constitution have increased in the past decade. Cultural organizations are on the rise. They are promoting the interests, languages, and cultures of minority groups and, in the process, are further creating culture and difference. And in everyday attitudes and behavior, 'minorities' are increasingly defying Tswana cultural dominance by resisting continued assimilation, in some instances subverting Tswana dominance (Durham, 1994). For example, many members of several different 'minority' groups that I interviewed in the capital in 1998, told me that they prefer to speak their own language or English rather than Setswana which they defined as 'the language of oppression'. A heightened ethnic consciousness is visible in disputes over language use on the Botswananet discussion group, in intense factionalism between the 'north' and the 'south' in the ruling Botswana Democratic Party (BDP), 8 in the degree to which people now use languages other than Setswana in the urban areas, and in the frequency with which ethnicity and 'the nationalities question' are increasingly subjects of discussion and debate. 9 The degree to which particularistic identity is emerging (not re-emerging) as a potent social and political force is surprising to me as an observer of the country for over two decades, and to many of my local associates. Before proceeding with further consideration of the above-mentioned processes, it needs to be said at the outset that, despite the increasing stridency of'minority' rhetoric and the sophistication of their organizations, the pursuit of 'minority' rights is being conducted within the framework of the law, through conventional political channels, and in the absence of violence. 'Minority' 7 The extent of the association varies; for some groups the association was never strong and for others, such as the Sarwa, it remains strong (Wilmsen and Vossen, 1990). 8 BDP fractiousness has quieted with the appointment of Seretse Khama Ian Khama (son of the first president and non-acting chief of one of the larger Tswana 'tribes') to the vice-presidency in 1998. 9 This is based upon three five-week periods of fieldwork (1994,1998, and 2000) in which I pointedly observed signs of such behavior and discussed the issue with numerous people who, generally, confirmed my perceptions. I compare this to four other periods of fieldwork: twenty months in 1977-9; eight months in 1986; and one month each in 1990 and 1991. The Botswananet is an email (listserv) discussion group composed largely of Botswana students studying abroad and interested fellow travelers such as myself. In late 1995, the list manager shut the net down for some days to allow tempers to cool from the ethnic-based heat that damaged, in his estimation, the intended spirit of the net. At the time the address was .

134

Jacqueline S. Solway

groups do not want to seize the state, take over via a coup; they are, in the Botswana 'tradition', peaceful. 'Minority' groups may seek to alter internal boundaries of the country but not the external ones; they are not irredentist or nationally secessionist. It is not Botswana citizenship that is in question, but the terms of that citizenship. 10 And it is not the Botswana nation that is in question, but, rather, its cultural basis and terms of inclusivity. Many state leaders recognize minority discontent to be a potentially destabilizing, although not revolutionary, force. But it is also an extremely sensitive issue and not one for which widely acceptable resolutions appear obvious. 11

Constitutional Rights and their Limits An examination of 'minority' struggles within Botswana also raises questions regarding the co-existence of group/collective rights and individual rights and how the form and degree to which they are respectively privileged by the state affects its capacity to promote social equality. Post-Cold War conflicts and the rise of identity politics have led to a renewed consideration of the moral basis of collectivities as rights-bearing agents as opposed to the more conventional liberal democratic stance that individuals alone bear rights to be protected by the state (cf. Kymlicka, 1995). To what extent, in the absence of granting some sort of group rights, can the state prevent itself from entirely reflecting the interests (or the culture) of the majority? And how, in granting rights predicated upon the existence of group differences, can the state stem the tide of what some fear would be lifting the lid off a 'Pandora's box', resulting in a possible proliferation of groups demanding possibly unending rights and resources? 12 Botswana's constitution enshrines individual rights premised upon an abstract citizen; there is universal adult franchise and access to resources of all kinds, education, state grants and subsidies, political office and so on is, theoretically, available to all. The only explicit privileging of group rights within the constitution pertains to membership in the House of Chiefs (the upper house of the legislature with advisory powers only). Ex-officio membership is limited to chiefs of the eight Tswana 'tribes' (or merafe - nations [singular - morafe or setshaba]) listed in the constitution, while seven other elected members have sub-chief status in the House. 13 These eight groups or 'tribes' as they are specified in the constitution repre10 To make this point further, there is virtually no Botswana diaspora; it is not a country which people want to leave. To illustrate, Samatar (1997: 707) notes 'As of 1994, only one student out of the hundreds sponsored to study overseas had not returned to the country after completing his training.' 11 President Mogae, for example, when I interviewed him in 1998, claimed to be 'indifferent' regarding specific measures to reduce minority exclusion. 12 See, for example, Fraser (1995) who, while not entirely opposing some form of group rights to remedy injustice, emphasizes the need to frame carefully both the group and the right so as to avoid further stigmatizing the disadvantaged group by making it appear 'inherently deficient and insatiable' (85). 13 The eight 'tribes' as listed in the constitution are the Bakgatla, Bakwena, Bamalete, Bamangwato, Bakgwaketse, Barolong, Batawana, and Batlokwa (all Setswana-speaking); their hereditary paramount chiefs have full status in the House. In addition, sub-chiefs are elected by their own number in the four districts that, in colonial times, were crown or freehold, and not 'tribal', land. These four represent districts, not a 'morafe', or 'tribal' unit and constituency, and they must be elected every five years or upon one's retirement. Three more members are specially elected by the other twelve (Republic of Botswana, 1987a: 45) to make a total of fifteen. A draft White Paper that proposes to amend this portion of the constitution was released in late 2001. See footnote 30 for further elaboration.

'Minority' Struggles in Botswana

135

sent the historical metamorphosis of the merafe ('tribes' - pre-colonial polities) into tribal reserves in the colonial period and, since independence, into districts (Peters, 1994). In the district capital the chief presides over the central kgotla (court - seat of authority and public arena and where many government policies and personnel are introduced). The chief embodies the moral and political authority of the tribe (Comaroff and Comaroff, 1997: 129). District capitals are the political and geographic as well as moral and social centers of the 'tribes' transformed into districts and as such they represent a fusion, at least in appearance, of 'tribal' and 'modern' political presence and authority. Included within tribal boundaries were 'tribespeople proper' as well as foreigners and other subject peoples; many of the latter category are currently termed 'minorities'. 14 Preserving the spatial basis of Tswana hegemony, especially the tribal capital and its kgotla, continues to marginalize minorities. While the House of Chiefs remains an important exception, Botswana's constitution is otherwise a modernist document emphasizing generalized human rights that are, in large measure, respected (Nengwekhulu, 1998). 15 For example, in a recent high-profile legal case, the universalistic basis of the constitution was affirmed. The 1982 citizenship act which stipulated that Batswana women married to expatriates could not pass on their citizenship to their children 16 was challenged and eventually amended after appeal and counter-appeal on the basis of it being discriminatory against women and therefore in violation of the constitution (Molomo, 1998: 203-4). The fact that many senior politicians as well as popular sentiment were largely against it (President Mogae, 1998: personal communication) makes this human rights victory all the more impressive. Apart from the constitution, there are other means by which supposedly group-blind state policy can privilege the majority and institutionalize its cultural hegemony. In Botswana, as elsewhere, the state embodies the symbols and rituals of its numeric and/or political majority (Young, 1982). More specifically, a state that accords individual rights to all citizens 'may appear to be "neutral" between the various national groups. But in fact it can (and often does) systematically privilege the majority nation in certain fundamental ways' (Kymlicka, 1995: 51). The marginalization of minorities can occur as a result of 'the drawing of internal boundaries; the language of schools, courts and government services; the choice of public holidays; and the division of legislative power between central and local governments' (ibid.). And in such ways the disposi14

Some districts, such as the Central District, are very large, heterogeneous, and the 'tribal' group (Ngwato) whose paramount chief presides over the district capital's kgotla is easily outnumbered by other groups in the district. The size and inclusion of 'subjects' reflects the power of Chief Khama and his co-operation with the British when the district boundaries were drawn at the close of the nineteenth century. The district is now divided into four sub-districts but many former subject peoples remain resentful and want further autonomy. For instance, many key people in Bobirwa, a sub-district named for the Birwa people, want a separate district. They claim that not only is it humiliating to remain under the Ngwato (and be treated as second-class citizens with 'odd' dialects when they attend meetings in the district capital) but that it is also inefficient. They would be much better served, they argue, by interacting directly with the central government rather than having to go through the district intermediary - and, as they say, 'get the crumbs from Serowe's (the district capital) table' (personal communication, 2000). 15 Indeed, Botswana is regularly applauded for its record on human rights (see for example, US, 2001 although this report is less laudatory than previous ones). Amnesty International ranked Botswana above Britain on its human rights record (Stedman, 1993: 1). 16 This had perverse consequences in that some women chose to remain unmarried to their nonTswana partners so that their children could become Botswana citizens.

136

Jacqueline S. Solway

tions of the majority tend to saturate the fabric of daily social, political, and cultural life.17 The consequences, in Botswana as in many other countries, are such that the ability of 'minorities' to enjoy the full benefits of citizenship may be compromised. This can be manifest in a range of widely varying exclusionary and discriminatory processes and practices. Some of the most subjugated minorities, such as the Sarwa, experience stark material deprivation. In their case ethnic status is largely homologous to class status (Wilmsen and Vossen, 1990). While the Sarwa have experienced some internal differentiation (Wilmsen, 1989), few, if any, Sarwa have found their way into the national bourgeoisie. This is not so for other former subject (current 'minority') peoples. 18 The marginalization of 'minorities' affects their access to the national public sphere. Minorities are not prohibited from speaking their own language or engaging in their own cultural practices, but the public sphere, while portrayed as neutral, clearly reflects the language, interests, everyday practices and dispositions of the majority. Minority languages and practices are thus relegated to the private sphere, while the public is the preserve of the Tswana. When members of minority groups deliberately speak their languages in a government office, a university setting, or a large shop (as they increasingly do), they are calling attention to the practice. Frequently, this is done playfully to remind others that 'we are not all Tswana here' (cf. Durham, 1994). But it is also done more aggressively in order to make an overt political statement (cf. van Binsbergen, 1994). The above examples entail minorities deliberately asserting their non-Tswana status for political purposes. However, some members of minorities, especially the less educated, may not know the appropriate language, codes or behavioral rules-protocols that would lead them to feel comfortable or entitled to participate in the national public sphere. Many whose first language is not Setswana feel their children are disadvantaged in school, especially in the early primary grades. And, of course, this impacts upon their later success in the job market. Non-Setswana speakers feel disadvantaged in the courts, in dealing with the bureaucracy, in accessing the media, in understanding and obtaining the various services and subsidies available to Botswana citizens, and so on. Thus neutral public space is not, in fact, neutral. The so-called 'normal' or unmarked public cultural space is coterminous with the perceived majority (Tswana), and the perceived minority (non-Tswana) then becomes the marked or exceptional category in need of 'special' dispensation. Cultural or symbolic injustices also result from the institutionalization of the majority's predilections, in Bourdieu's (1977) words, the 'naturalizing of their arbitrariness'. Representation then becomes subject to majoritarian discourse. Minorities frequently suffer negative representation, denigrating images, and in Taylor's terms 'nonrecognition or misrecogniton.' '. . . identity is partly shaped by recognition or its absence, often by razsrecognition of others, and so a person 17 The 1982 citizenship act, before it was amended, was a case in point and aptly illustrates how the majority's culture can be imposed on all. Not only did its prohibition on women married to expatriates devolving their citizenship violate the universalistic basis of national Botswana citizenship, but it also reflected the patrilineal basis of Batswana 'tribal' culture. Not all citizens of Botswana share patrilineal ideology; for example, many Sarwa, Yei, and Herero groups are based upon other principles of descent. 18 For example, there are members of parliament from several of the groups that, according to Schapera's (1984) classification, were 'serfs' in the past (e.g. Kgalagadi, Yei, Tswapong). But there are no Sarwa MPs (Mmegi 4-7-97). As mentioned earlier, the home areas of many former 'serf' peoples remain 'remote' and poverty-stricken. 'Remote', it should be noted here, connotes social and moral as well as geographical distance from Tswana centers.

'Minority' Struggles in Botswana

137

can suffer real damage, real distortion if the people in a society mirror back . . . a confining or demeaning or contemptible picture . . . Nonrecognition or misrecognition can inflict harm, can be a form of oppression' (1994: 25). Such cultural and symbolic injustices can affect the sense of moral worth and entitlement that 'minorities' may embrace (cf. Fraser, 1995). And, indeed, many 'minority' peoples have internalized feelings of inferiority (e.g. Datta and Murray, 1989; Solway, 1994b). The creation of the House of Chiefs, the drawing of district boundaries, the choice of official and national languages did not, in themselves, create the inequality that exists between groups in Botswana; this has a prior history. And, as mentioned earlier, post-independence policies, national prosperity, widely distributed infrastructure, and political openness have resulted in a partial leveling of inter-group inequality (while contributing to increased intra-group inequality). But, at the same time, the policies and practices listed above have set limits on the elimination of inter-group inequality and, it can be argued, they have also contributed to the intensifying of group distinctiveness. It is widely acknowledged in Botswana that eradicating these official exclusionary barriers will not be sufficient to end inequality. But it is also widely accepted that their removal remains a necessary prerequisite for attaining higher levels of equality.

Minority Struggles It is perhaps ironic that at the precise point when Botswana's assimilationist policy began to bear fruit and a new generation of previously excluded people were beginning to join the national bourgeoisie, many of these same people and the groups to which they belong were redefining their identities, coalescing around them, and agitating for more substantial change than they were experiencing. The subjective experience of being a minority, of being different, took on new meaning outside the insulatory nest of kith and kin. Working and studying in town led many minorities to perceive their identity with a new level of intensity. For some the experience was epiphanal. At some moment a particular experience or thought would summarize, integrate or, possibly, spark what might have otherwise been perceived to have been vague and undefined notions or feelings. The integrity of one's ethnic identity, its disparaged past, and what should be the road forward - the road to redress - would be realized in a new and heightened form. Identities are re-imagined, forged, claimed, reified, and hardened in this process; this is true both for the minorities and for the Tswana who are placed in a defensive position. Thus, in characteristic fashion, it is on the social margins and in counterpoint to, but at the same time, in conjunction with, rationalist modernity that we witness the ongoing production of ethnicity (cf. Comaroff and Comaroff, 1997). By the 1980s the antagonism in the voices of many minorities was beginning to be audible. Through a variety of methods minorities are seeking means to alter the basis of their political representation, material entitlements, and cultural recognition. These struggles are motivated by a combination of redistributive and recognition goals (cf. Fraser, 1995) and emphasize, to varying degrees, cultural and/or material transformations. However, most minority strategies are premised on the fact that justice on one front can positively impact on the other, and most desired outcomes entail retribution on both fronts. Apart from the everyday counter-hegemonic acts entailing the use of minority language, refusal to behave in a subservient manner, and assertions of cul-

138

Jacqueline S. Solway

tural pride and difference, there are more organized and directed methods being adopted by minority groups to challenge Tswana hegemony. These include (i) the formation of cultural organizations, (ii) support for the political opposition which has been seen to be more sympathetic as well as more proactive on minority issues than the ruling party, and (iii) direct challenges to the constitution and government policy.

Cultural Organizations There are several cultural societies within the country. Amongst the more well organized and established are SPIL, founded in 1981 and devoted to promoting Ikalanaga language and culture, and Kamanakao, founded in 1995, devoted to revitalizing and preserving Yei culture and language. These have national presence; they are referred to in the media; they have published materials in their languages; they are officially registered with the Registrar of Societies in the Office of the President; and they have websites. 19 Birwa maintain an organization and the Herero youth have organized Otjira tjOmitanda (Durham, 1993). In the mid-1980s Kgalagadi attempted to create a cultural organization; they designed an elaborate constitution but failed to bring the project to fruition. These are the minority organizations of which I am aware (apart from the Sarwa organizations which tend to have greater external involvement and are not being considered here). Some of the organizations tend to reflect the concerns of elite members of a particular group and to emphasize language development and cultural/recognition ends, while for others the focus is on remedying material deprivation, with cultural recognition a component of that overall goal. SPIL (the Society for the Promotion of the Ikalanga Language) (van Binsbergen, 1994) is an example of the former, while some of the Sarwa movements are examples of the latter. With few exceptions the founders and leaders of the non-Sarwa organizations are young, educated, urban-based, and hold valued formal sector employment (cf. Durham, 1993). Several are associated with the University of Botswana (cf. van Binsbergen, 1994: 161-2). Writing in 1994, van Binsbergen noted that SPIL remained an elite organization, but by 1998 I was told that it had gained more grassroots popularity. Regardless of the extent to which SPIL itself, as well as other organizations, have broadened their base of support, their mandate of promoting and preserving language and culture have wide and growing appeal. The critical role of youth in minority organizations highlights the fact that such movements are less about the past than they are about the present. 20 It is not elders seeing their culture slip away, but, rather, young largely successful members of modern Botswana who desire to claim their particularistic identity, not instead of, but in addition to, Botswana citizenship. The cultural organizations may be backward-looking in seeking 'traditions', custom, and languages spoken more in rural areas and less by children than their parents, but as the oftused phrase suggests, these 'Janus faced5 organs of ethnic consciousness are equally present and forward-oriented (cf. Marks, 1989; Vail, 1989; ComarofT, 1987). Their stated purpose is not to recreate the past but to empower their 19 The websites are: and . 20 Even their form, with their elected offices and complicated constitutions, is more consistent with the modern Botswana state. As Durham (1993: 283-4) notes of the Herero organization, it is 'uncompromisingly bureaucratic'.

'Minority' Struggles in Botswana

139

members to move 'forward' by increasing their self-confidence and pride in their own backgrounds. This can be achieved by promoting and preserving, in the face of assimilationist forces, the language and cultural forms produced in the past. 21 Equally important is the perceived need to attach positive associations to minority cultures and languages to replace the negative ones common in the Tswanacentric point of view. Colonial and post-colonial Africa have witnessed the rise of numerous ethnically based organizations and cultural societies. These 'voluntary associations' have been analyzed for the instrumental role they have played in facilitating urbanization amongst 'detribalizing' natives. In turn, the associations' leaders have gained clients, political followers and power in an effort to claim a portion of the state's resources, its 'national cake' (Young, 1982: 80; Geertz, 1973). This somewhat cynical interpretation of ethnicized politics has a great deal of credibility (see, for example, Berman, 1998 and this volume; Ekeh, this volume; Eyoh, 1998 and this volume; Lonsdale, this volume; Young, 1982) but its applicability to the Botswana case is limited. The leaders of the cultural organizations whom I know of are little interested in personal political gain or advancement. Botswana's bourgeoisie is fortunate in having many legitimate opportunities available for individuals to engage in rewarding work. Many leaders of the cultural organizations have valued, professional careers that they have worked hard to achieve. To my knowledge, most have not run for political office. Indeed, I know several cases in which they have declined to run for office when it has been suggested to them. Their involvement in their cultural societies stretches their schedules and consumes a good deal of their non-work time. For most it is a moral commitment, not a road for personal gain. 22 That these societies are allowed to register officially within Botswana attests to the country's tolerance of multiculturalism, and this tolerance (as far as it goes and as long as the expression of cultural difference remains outside the public sphere and does not challenge national institutions) defuses what might otherwise be more aggressive or antagonistic expressions of difference. Denying official status to the organizations could well lead to their radicalization - and, indeed, the organizations maintain explicit apolitical stances. But such stances are difficult to sustain in an increasingly politicized environment. Some local branches of SPIL have urged Kalanga members of government, of which there are many, 23 to openly identify themselves as such and to support change in the country's language education policy. As a result, some ruling party members called for the organization's disbanding (van Binsbergen, 1994: 158; see also Mannathoko 1991). 21

Marks (1989) emphasizes the contradictory forces of such organizations. Heightened ethnic consciousness and glorifying the past can transform into politicized cultural xenophobia and internally repressive policies. Some of these possible dangers correspond to what Kymlicka (1995) refers to as the internal restrictions sub-national groups may attempt to impose, but that could be inhibited by liberal national state policies. 22 Of course, there are exceptions, but surprisingly few. 23 For a complex of reasons including their location on the borders of South Africa and the then Rhodesia and their stigmatized status and material constraints relative to the Tswana, Kalanga people availed themselves of educational opportunities earlier than most of their Tswana neighbours. As a result, they hold a disproportionate number of high-level civil service, education, and political positions. The first post-independence generation of successful Kalanga 'kept quiet,' it is said, enjoyed the fruits of their labor, and willingly subjected their identity to the forces of Tswana hegemony.

140

Jacqueline S. Solway

Party Politics Botswana achieved independence in 1966 without conflict or controversy; the political transition was smooth and did not follow a period of anti-colonial struggle. Nor were there a significant cadre of civil servants, major commercial 24 or settler interests, or a mobilized peasantry or other group that might have emerged from an independence struggle to challenge those who assumed power (Samatar and Oldfield, 1995). The new ruling cadre, many of whom became leaders in the Botswana Democratic Party (BDP), represented a bi-racial elite, several of whom had worked together in the Legislative Council (LEGCO) that had been created by the colonial government to facilitate the transition to independence (Gabasiane and Molokomme, 1987). The BDP does not simply or overtly foster Tswana hegemony. The party leaders would quickly deny such allegations and would point to their ethnic 'blind' policies, and the fact that the party includes many minority members, and they would highlight their attempts to distribute infrastructure widely. They might also add that their promotion of bureaucratic rule and the concomitant reduction of chiefly authority (especially with the establishment of Land Boards that removed chiefly control over land [Comaroff, 1980]) have lessened Tswana control. 25 However, at the same time, the BDP government, as noted earlier, eliminated all languages other than Setswana and English, maintained the geography of Tswana power, and most painfully for minorities, wrote and maintains a constitution that lists only Tswana groups as 'tribes' and has a House of Chiefs whose full membership is restricted to Tswana. The BDP has maintained a majority since independence. In 1994 the main opposition party, the Botswana National Front (BNF), took 13 of the 40 parliamentary seats, over 35 per cent of the popular vote, and posed, for the first time, a serious threat to the BDP. In the subsequent election (1999), the main opposition, now hampered by the fact that it had splintered into two competing parties, took only 7 seats. Any number of factors could bring the opposition back in force by the next election in 2004. Continuing BDP support derives largely from Botswana's record of peace, political openness, prosperity, and positive record of infrastructural development and social services. However, in spite of impressive macroeconomic indicators, poverty remains a serious problem (people say 'we are a rich country of poor people'). The opposition finds its strongest support in urban areas where the government's rural subsidy programs have minimal impact, where unemployment and controls on union activity are felt most strongly, and where the contrast between wealth and poverty and the disassociation between the rich and poor render the subjective experience of poverty acutely painful (cf. Wiseman, 1998: 257). 26 24

Botswana's vast and extremely high quality diamond deposits were discovered after independence; one can only speculate how the transition to independence might have been different if these had been known earlier. At independence, the cattle industry was the dominant enterprise. 25 Many authors point to a diminution of chiefly powers since independence and, at present, chiefs continually complain that they are not taken seriously. See, for example, Comaroff (1980); Koma (1984); Gillett (1973); Jones (1983); Proctor (1968); and especially Sekgoma (1994) who discuss the erosion of chiefly powers. See also 'Hands off Chieftainship and mind your own business - chiefs tell politicians' Botswana Gazette, 16 October 1996, p. 4 ; and 'Chieftainship relegated to a status of lower civil service', Midweek Sun, 28 February 1996. 26 Parts of the Southern district and Kgatleng, both rural Tswana areas, have supported the opposition in several elections. In these instances the chiefs have been in conflict with the government and opposition support reflected these quarrels.

'Minority' Struggles in Botswana

141

The 13 seats it garnered in 1994 reflect, to a great extent, this urban appeal and the fact that rapid urbanization led to a re-apportioning and increase of electoral seats in the towns. But the opposition's explicit support for greater inclusivity and its intention to reconfigure the House of Chiefs has won it growing support amongst minorities. 27 Both the BNF and the Botswana Congress Party (BCP) which broke from the BNF in April 1998, cultivate and capitalize upon their appeal amongst minorities. These parties do not identify with any particular group; they are not ethnic parties nor has there been a proliferation of ethnicbased parties to accompany minority politicization. Rather, their support of minority goals has been a means by which these parties have extended their bases of support into the rural areas. The opposition has taken only one rural, predominantly minority area, parliamentary seat but it has taken many local council seats. Several parliamentary seats have been extremely closely fought, and in the 1999 election the results of two minority constituencies (Ngwaketse west and Letlhakeng - both largely Kgalagadi areas) were subject to a recount. I have written previously about the politicization of minority interests amongst Kgalagadi 28 of western Kweneng, the constituency of Letlhakeng (see Solway, 1994b, 1995). This is an area that has witnessed a dramatic rise in political party activity. For instance, in 1979 the rate of voter turnout in the constituency was 41 per cent of registered voters, representing the second lowest turnout in the country (Republic of Botswana, 1979). In 1994 over 78 per cent of registered voters voted, ranking the constituency turnout fifteenth of forty (Republic of Botswana, 1994). In 1999 voter turnout remained above-average. 29 In 1979 the local parliamentary seat was contested, but bland acceptance of the BDP was uncontested; that was quick to change. Voting behavior reflects the general politicization of life in the region where party politics has become the stuff of everyday conversation and debate, where people gossip about who supports which party, where friendship is influenced by party affiliation, and where village clubs and committees are often associated with one party or another. One of the key local opposition supporters describes his politicization as follows: I first fought for Kgalagadi e m p o w e r m e n t in the late 1970s while a b o a r d i n g student at secondary school [in the district capital]. We experienced intense tribalism there. In the hostels, K w e n a [Tswana] said they could n o t sleep next to a Kgalagadi. We were despised and we b e c a m e rebellious. We visited relatives working in the capital w h o 27 For example, Pamphlet 1 of the BDF endorses the establishment of a House of Representatives that would include leaders of several organs of civil society in addition to the chiefs of all the country's groups. Its will would preside over that of parliament if they clashed. Michael Dingake, formerly of the BNF, now leader of the BCP, and a minority member himself, also endorses minority agendas and the party attempted to print its 1999 election materials in several languages. 28 Kgalagadi are an amalgam of disparate peoples some of whom have resided in the Kalahari for centuries and some who fled into the desert as refugees from wars in the early nineteenth century. They were absorbed into the Tswana (Kwena) political economy as batlhanka (servants), forced to render tribute in the form of desert products such as furs and hides, and forced to herd Kwena cattle. Although official servitude status ended with British rule in 1895, hierarchical relations have been slow to disappear. The area they inhabit has been relatively neglected in terms of infrastructure and Kgalagadi of the region were able to access educational opportunities only quite late by Botswana standards. Despite the fact that Kgalagadi are composed of peoples of diverse origins, their common residence, intermarriage, and similar placement within the encompassing political economy have led to a process of ethnogenesis. 29 Because of the split in the opposition national voter registration was low for the 1999 election and many citizens were confused, disheartened, or angry with the opposition parties. Many had predicted that the BNF was poised to win this election, had it continued on course.

142

Jacqueline S. Solway

were embarrassed when we spoke Shekgalagadi. It made me sad and angry to see their shame. Then I saw the BNF [Botswana National Front - main opposition party] pamphlet and on page 16, Koma [the leader] said he was indebted to the minorities for paying taxes so that he could study and to the Sarwa who herded his cattle while he was at school. Then, from 1979 to 1994, I spent up to 65 per cent of my time campaigning for the BNF in western Kweneng. (personal communication 1998) Opposition supporters in Western Kweneng have other reasons to support the parties than minority rights. But for most supporters voting for the opposition is the edge of a wider goal of local self-determination that they hope will eventually result in the formation of a new district. The new district will have its own capital and its own chief who will be a full member of the House of Chiefs (or whatever may replace it); Shekgalagadi will be spoken in local offices, the courts, and primary schools, and local residents will have greater authority over the dispensation and disposition of resources than they now exercise. District secession and all that it entails is a goal embraced by many minorities.

Constitutional and Policy Challenge Since the late 1980s the number and range of minority challenges to official policy and law have dramatically increased. However, repeated efforts to expand language representation in education and the official media have failed to date. In 1998, for instance, the government reaffirmed its restrictive media policy by denying a radio licence to SPIL for an Ikalanga station, despite its rhetorical stance of media openness. Efforts to amend sections 77-9 of the constitution and thereby alter its list of 'tribes' and reconfigure the House of Chiefs have also failed. On these matters, Botswana, once seen as progressive on the subcontinent for its non-racial and liberal multiparty system, is now viewed by many as lagging behind the 'new democracies' in the region such as Namibia and South Africa that have constitutions and policies granting more rights to diverse groups (Molutsi, 1998, personal communication). But the momentum for change is powerful. Early in 2000 the President announced that a task force would be established to review the House of Chiefs and its constitutional basis. The Balopi Commission, charged with this task, submitted its report in 2001. 3 0 Since 1988, there have been many attempts to alter the basis of the House of Chiefs. Within the house itself there have been motions to upgrade the subchiefs (non-Tswana) to paramount chiefs. In early 1994, a motion to do this was flatly defeated, but after the 1994 election, when the opposition posed, for the first time, a serious threat to the ruling party, the motion, when put forward again, was passed. However, no change has yet taken place. Similarly, in 1988, when a member of the opposition proposed in parliament to neutralize the tribal basis of the constitution, a fellow parliamentarian arrogantly reminded his colleague that, in regard to non-Tswana peoples - 'We defeated them' - (Republic of Botswana, 1988: 475) there was little uncertainty as to who was and who was not included in the Tswana fold. The motion was defeated. It was raised again in 1995 (after the 1994 election) by a member of the ruling party and not treated in the same dismissive manner as seven years earlier. The discourse was much more cordial and conciliatory; the motion passed, but, again, concrete change 30 In December 2001, a draft White Paper was presented to Parliament. It proposes to eliminate mention of specific 'tribes' in the constitution, expand representation in the renamed upper house (the report suggests replacing House of Chiefs with a Setswana title), but some vestiges of the old chiefly hierarchy remain.

'Minority' Struggles in Botswana

143

has yet to occur. The scrutiny of international human rights agencies (a force already in play) could have an impact. But this is a deeply sensitive issue that goes to the very core of Tswana patriarchy, pride, and power. While change now appears inevitable, it will be a painful process for many, and not all constituencies will be satisfied with the compromises they are bound to make.

Kamanakao Kamanakao means remnants in Shiyeyi and is devoted to reclaiming, recovering, and valorizing Yei culture and language. The organization has been very successful in gaining local followers and outside support and is seen by many former subject peoples as a model to emulate. Kamanakao maintains an explicitly apolitical stance but the implications of its actions are anything but apolitical. One of Kamanakao's leaders, an urban-based intellectual, does not mince words when she says How can they mean democracy when it's minority rule over majority? What do they mean about democracy when they keep a tribalist constitution for over thirty years? Development follows the constitution; if you look you'll see that the non-Tswana areas have been neglected and do poorly by most indicators. But resources are in nonTswana minority areas - tourism in the Okavango and diamonds in the Kalahari. (Nyati-Ramahobo, 1998; personal communication) Kamanakao has developed a broad base of grassroots support in its region. According to Nyati-Ramahobo (personal communication, 2000), it is now 'localized' and sustainable with minimal or no intervention from urban-based leaders. Its members have worked carefully to foster local involvement and loyalty. Their meetings are held in different villages, elders are invited to speak about past customs and experiences, others are entreated to participate as well, and all meetings open and close with prayers. Kamanakao has produced a Shiyeyi bible, a collection of folk tales, a calendar, and T-shirts, which are all proudly displayed, reflecting the increased pride in Yei identity. NyatiRamahobo's (1999) observation that Shiyeyi is spoken more frequently in the region than in the past is evidence of Kamanakao's impact; in 1988 it was considered a 'threatened language' (Vossen). Kamanakao members see themselves carrying forward the battle initiated by their ancestors in the 1940s, when several Yei delivered a set of demands to the Tawana chief (predominant Tswana group of the region, although numbering roughly half the Yei, Anderson and Janson, 1997). 31 Kamanakao is writing the script of minority struggles in Botswana. Kamanakao and the Yei whom it represents have engaged in all forms of minority struggle; their cultural organization is the most active in the country, opposition support is strong in the region, and in April 1999 a paramount chief, complete with leopard skin and associated regalia, was installed. In late 1999, they filed a law-suit arguing that sections 77-79 of the constitution violate the constitution's stated rights, and they have demanded that their chief receive official recognition. The case is unprecedented in that the Yei claim that the constitution is internally contradictory - that the sections listing 'tribes' and setting out membership in the House of Chiefs contradict earlier universalistic human 31

The majority of the Wayeiyi (Yei) reside in Botswana's northwest district (especially in the Okavango region) where, it is argued, they constitute a numeric majority. Despite their high numbers, the district is represented in the House of Chiefs by the Tswana chief.

144

Jacqueline S. Solway

rights sections. The hearing of the Yei case was repeatedly postponed by the government, and then, in July 2000, the Yei themselves asked for a postponement to clarify certain constitutional issues. This is a high-profile, high stakes case and all parties are being extremely careful not to undermine their positions by technical faults. The case has been covered in the media and was indirectly referred to several times by the government minister addressing a conference I attended in Botswana in May 2000. And it is probably this case, more than any other single incident, that has led to the recent attempt by the government seriously to address the concerns of minorities by establishing the Balopi Commission. Kamanakao's actions also raise questions regarding the degree to which they represent either a true challenge to, or an affirmation of, Tswana hegemony. To borrow an overused phrase, the Yei are not simply recovering their tradition, but inventing it. The Yei do not have a legacy of political centralization or paramount chiefly rule. The recently installed chief was amongst a number of individuals identified as belonging to more prominent or 'senior' kin groupings, but he was also elected from a short list that was arrived at in a manner similar to an academic job search. One could say the Yei are beating the Tswana at their own game or one could say the Yei's decision to install a paramount chief simply affirms the strength of Tswana hegemony. In trying to valorize their own culture, why did they model it upon that of the Tswana and adopt a practice inconsistent with their past? Does this indicate acceptance of the Tswana political model as the one to adopt, or is the reason more strategic? Perhaps they believed it necessary to challenge the Tswana on their own terms. The degree to which they have been successful, in itself, subverts Tswana hegemony and Tswana views of the capacity and rights of former subjects. Conclusion Botswana's much celebrated political and economic success, captured in the title of Samatar's (1999) recent book, An African Miracle, rests upon a variety of fortuitous factors that have their basis in internal as well as external sources, events, decisions and sets of circumstances (see Solway, 1995). 32 The fact that Botswana was able to fashion significant components of a liberal system without entirely discarding existing structures (and thus not alienating those empowered by them) has enhanced its success. Botswana developed a multiparty democracy that rests upon an expanding bureaucracy, that emphasizes leadership based on elections and merit, and that promotes an assimilationist policy towards internal difference. But, at the same time, it has maintained the spatial basis of 'tribal' power and authority and granted roles for chiefs at both the national and local levels. Botswana was able to create a path that neither 'wiped the slate clean' nor allowed what Mamdani (1996) refers to as 'decentralized despotism' to persist into the post-colonial period. Elected officials and chiefs sit in the national government. While the premises of personhood, rights, and authority upon which these political offices are based appear as incommensurate, they not only coexist but mutually inform one another. Their co-presence has contributed to peace and prosperity in the country. However, the continued capacity of Botswana's seemingly contradictory policies to satisfy or contain minority interests and aspirations appears, to many minority individuals and groups, increas32

Other titles such as 'Enduring Elite Democracy' (Good, 1999), Poverty and Plenty (Nteta et al., 1997), and 'Poverty Amidst Plenty (Gulbrandsen, 1996), suggest that the 'African Miracle' is less stellar than many reports and institutions, such as the World Bank, would have one believe.

'Minority' Struggles in Botswana

145

ingly limited. Botswana is faced with an ambitious challenge. How can a peaceful, prosperous and liberal state create the conditions to render itself more inclusive? Can a meaningful sub-national identity be formulated that does not compromise the state's integrity and, at the same time, promotes equity?

References Anderson, B. 1991. Imagined Communities. London: Verso. Anderson, L-G., and T. Janson, 1997. Languages in Botswana. Gaborone: Longman. Berman, B. 1998. 'Ethnicity, Patronage and the African State: The Politics of Uncivil Nationalism', African Affairs.97: 305-41. Bourdieu, P. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press. Botswana, Republic of. 1979. Report to the Minister of Presidential Affairs and Public Administration on the General Election, 1979. Gaborone: Government Printer. — 1987. Constitution. Gaborone Government Printer. — 1988. Hansard. Gaborone: Government Printer. — 1991. National Development Plan. Gaborone: Government Printer. — 1993. Budget Speech. Gaborone: Government Printer. — 1994. Report to the Minister of Presidential Affairs and Public Administration on the General Election, 1994. Gaborone: Government Printer. — 1995. Budget Speech. Gaborone: Government Printer. — 1996. General Elections in Botswana: The Road to Botswana Parlaiment. Gaborone: Research, Information and Public Relations Division, Botswana Parlaiment. Gaborone: Governement Printer. Campbell, E. K. 1998. 'Population Distribution and Urbanisation', in W. A. Edge and M. H. Lekorwe, eds, Botswana: Politics and Society. Pretoria: J.L. van Schaik. pp. 261-80. Carter, G. M. and E. P. Morgan, eds. 1980. From the Frontline: Speeches of Sir Seretse Khama. London: Rex Collings. Charlton, R. 1993. The Politics of Elections in Botswana', Africa 63: 330-70. Comaroff, J. L. 1980. 'Class and Culture in a Peasant Economy: The Transformation of Land Tenure in Barolong', Journal of African Law 24: 85-116. — 1987. 'Of Totemism and Ethnicity: Consciousness, Practice, and the Signs of Inequality', Ethnos 52: 301-23. ComarofI, J. L. and J. Comaroff. 1997. Of Revelation and Revolution: The Dialectics of Modernity on a South African Frontier, Volume Two. Chicago: University of Chicago Press. Datta, K. and A. Murray. 1989. 'The Rights of Minorities and Subject Peoples in Botswana: A Historical Evaluation', in. J. Holm and P. Molutsi, eds, Democracy in Botswana. Gaborone: Macmillan. pp. 58-74. Durham, D. 1993. 'Images of Culture: Being Herero in a Liberal Democracy'. Unpublished Ph.D. dissertation. University of Chicago. — 1994 'Greetings in another Key: Playing with Language and Identity in Botswana'. Paper presented at the African Studies Association annual meetings, Toronto. Eyoh, D. 1998. 'African Perspectives on Democracy and the Dilemmas of Post-colonial Intellectuals', Africa Today 45(3-4): 281-306. Fraser, Nancy. 1995. 'From Redistribution to Recognition? Dilemmas of Justice in a "PostSocialist" Age', New Left Review 212: 68-93. Gabasiane, O. and A. Molokomme. 1987. 'The Legislative Council' in F. Morton and J. Ramsay, eds, The Birth of Botswana. Gaborone: Longman, pp. 161-71. Geertz, C. 1973. 'The Integrative Revolution: Primordial Sentiments and Civil Politics in the New States', in The Interpretation of Cultures. New York: Basic Books, pp. 255-310. Gillett, S. 1973. The Survival of the Chieftaincy in Botswana', African Affairs 72(287): 179-85. Gladney, D. 1998. Making Majorities. Stanford, CA: Stanford University Press. Good, K. 1992. 'Interpreting the Exceptionality of Botswana', Journal of Modern African Studies, 30(1): 69-95. — 1993. 'At the Ends of the Ladder: Radical Inequalities in Botswana', Journal of Modern African Studies 31 (2): 203-30. — 1999. 'Enduring Elite Democracy in Botswana'. Democratization: 50-66. Gulbrandsen, O. 1996. Poverty in the Midst of Plenty. Bergen: Norse. Jones, D. 1983. 'Traditional Authority and State Administration in Botswana', Journal of Modern African Studies 21(1): 133-39. Hudson, Derek and Matthew Wright. 1997. 'Income Distribution in Botswana: Trends in

146

Jacqueline S. Solway

Inequality since Indepepndence', in D. Nteta, J. Hermans, and P. Jeskova, eds, Poverty and Plenty: The Botswana Experience. Gaborone: Botswana Society, pp. 105-32. Koma, K. 1984. Chieftainship in Crisis. Mahalapye: Botswana National Front: Pamphlet No. 5. — n d. BNF its Character and Tasks. Pamphlet No. 1. The Basic Document of the BNF. Kymlicka, Will. 1995. Multicultural Citizenship. Oxford: Oxford University Press. Kymlicka, Will, ed. 1995. The Rights of Minority Cultures. Oxford: Oxford University Press. Mamdani, M. 1996. Citizen and Subject. Princeton, NJ: Princeton University Press; Oxford: James Currey. Mannathoko, C. E. 1991. 'Kalanga Language and Ethnicity: A Historical Perspective', in Catrien van Waarden, ed., Kalanga: Retrospect and Prospect. Gaborone: Botswana Society, pp. 38-45. Marks, S. 1989. 'Patriotism, Patriarchy, and Purity: Natal and the Politics of Zulu Ethnic Consciousness', in L. Vail, ed. The Creation of Tribalism in Southern Africa. Berkeley: University of California Press; London: James Currey. Molomo, M. G. 1998. 'The Role and Responsibilities of Members of Parliament in Facilitating Good Governance and Democracy, in W. A. Edge and M. H. Lekorwe, eds, Botswana: Politics and Society. Pretoria: J. L. van Schaik. pp. 199-212. Molutsi, P. 1989. 'Whose Interests do Botswana's Politicians Represent?' in J. Holm and P. Molutsi, eds, Democracy in Botswana. Gaborone: Macmillan. pp. 120-32. — 1998. 'Elections and electoral experience in Botswana', in W. A. Edge and M. H. Lekorwe, eds, Botswana: Politics and Society. Pretoria: J. L. van Schaik. pp. 363-77. Mosha, A. C. 1998. 'The Impact of Urbanization on the society, economy, and environment of Botswana', in W. A. Edge and M. H. Lekorwe, eds, Botswana: Politics and Society. Pretoria: J. L. van Schaik. pp. 281-99. Mpho, M. K. 1989. 'Representation of Cultural Minorities in Policy Making', in J. Holm, and P. Molutsi, eds, Democracy in Botswana. Gaborone: Macmillan. pp. 133-38. Murray, Andrew. 1990. Peoples' Rights: The Case of Bayei Separatism. Human & People's Rights Project Monograph No. 9. Institute of Southern African Studies, National University of Lesotho. Nenkgwekhulu, R. 1998. 'Human Rights, Development and the Rule of the Law in post-colonial Botswana', in W. A. Edge and M. H. Lekorwe, eds, Botswana: Politics and Society. Pretoria: J. L. van Schaik. pp. 351-62. Nteta, D., J. Hermans, and P. Jeskova, eds. 1997. Poverty and Plenty: The Botswana Experience. Gaborone: Botswana Society. Nyati-Ramahobo, Lydia. 1997. 'Language in Education ad the Quality of Life in Botswana', in Nteta et al., pp. 251-72. — 1999. 'Oppression in Democracy: The Case of the Wayeyi of Botswana'. Paper presented at the ASA, Philadelphia. Parsons , N. 1985. 'The Evolution of Modern Botswana: Historical Revisions', in L. A. Picard, ed., The Evolution of Modern Botswana. London: Rex Collings, pp. 26-39. Parsons, Neil, Willie Henderson, and Thomas Tlou. 1995. Seretse Khama 1921-1980. Braamfontein: Macmillan. Peters, P. 1994. Dividing the Commons. Charlottesville, VA: University of Virginia Press. Phillips, Anne. 1995. The Politics of Presence. Oxford: Clarendon Press. Proctor, J. H. 1968. 'The House of Chiefs and the Political Development of Botswana', Journal of Modern African Studies 6(1): 59-79. Samatar, A. 1997. 'Leadership and Ethnicity in the Making of African State Models: Botswana and Somalia', Third World Quarterly 18 (4) : 687-707. — 1999. An African Miracle: State and Class Leadership and Colonial Legacy in Botswana Development. Portsmouth, NH: Heinemann. Samatar, A. and S. Oldfield. 1995. 'Class and Effective State Institutions: The Botswana Meatr Commission', Journal of Modern African Studies 33: 6651-68. Schapera, Isaac. 1952. The Ethnic Composition of Tswana Tribes. London: The London School of Economics. — 1970. Tribal Innovators. London: Athlone Press. — 1984. The Tswana. London: Routledge & Kegan Paul. Sekgoma, G. 1994. 'Does Chieftaincy have a Future in the Next Century?' in S. Brothers, J. Hermans, and D. Nteta, eds, Botswana in the 21st Century. Gaborone: The Botswana Society. pp. 403-18. Solway, J. 1994a. 'Drought as a 'Revelatory Crisis': An Exploration of Shifting Entitlements and Hierarchies in the Kalahari, Botswana'. Development and Change 25(3): 471-95. — 1994b. 'From Shame to Pride: Politicized Ethnicity in the Kalahari, Botswana', Canadian Journal of African Studies 28(2): 254-74.

'Minority' Struggles in Botswana

147

— 1995c. Political Participation, Ethnicity, and Multiparty Democracy in Botswana', in Dan O'Meara, ed, The Politics of Change in Southern Africa, Montreal: Canadian Research Consortium on Southern Africa, pp. 241-60. Stedman, J., ed. 1993. Botswana: The Political Economy of Democratic Development. Boulder, CO: Lynne Rienner. Taylor, C. 1994. Multiculturalism. Princeton, NJ: Princeton University Press. UNDP. 1998. Botswna Human Development Report 1997. Gaborone: UNDP. United States, Department of State. 2001. Human Rights Reports for 2000: Botswana. Washington, DC. Vail, L., ed. 1989. The Creation of Tribalism in Southern Africa. London: James Currey; Berkeley: University of California Press. van Binsbergen, William. 1994. 'Minority Language, Ethnicity and the State in Two African Situations', in R. Fardon and G. Furniss, eds, African Languages, Development and the State. London: Routledge. pp. 143-88. van Waarden, Catrien, ed. 1991. Kalanga: Retrospect and Prospect. Gaborone: The Botswana Society. Vossen, R. 1988. Patterns of Language Knowledge and Language Use in Ngamiland in Botswana. Bayreuth: Eckhard Breitinger. Wilmsen, E. 1989. Land Filled with Flies. Chicago: University of Chicago Press. Wilmsen Edwin and Rainer Vossen. 1990. 'Labour, Language and Power in the Construction of Ethnicity in Botswana'. Critique of Anthropology 10: 7-37. Winstanley, G. (Chief Electoral Officer). 1965. 'The Bechuanaland General Election of 1965: Ballot Envelopes and Voting Counters'. Mimeo. Wiseman, J. 1998. 'The Slow Evolution of the Party System in Botswana', Journal of Asian and African Studies 33(3): 241-64. World Bank. 1996. World Development Report 1996. New York: Oxford University Press. Young, Crawford. 1982. 'Patterns of Social Conflict: State, Class, and Ethnicity', Daedalus 111 (2): 71-98. Newspapers Midweek Sun Botswana Daily News Botswana Gazette Mmegi

9

T O YIN FALOLA

Ethnicity & Nigerian Politics: The Past in the Yoruba Present

N

IGERIAN politics since the early 1990s has turned the Yoruba into the principal focus of national discourse. The Yoruba political class has now had its turn to have one of its own as the country's President, after its struggles since the 1950s. The Igbo expressed a similar aspiration, and the assumption that the Yoruba had actually benefited more from the system than they claimed. The northerners' support for a southern candidate was partly because of the political crises that the military had created and the need to use a pro-north candidate to retain control. Ethnic calculations remain deep, very much as before, only slightly altered to give the Yoruba greater visibility. This reverses the long trend that privileged the North and its quest for political domination, the short-lived fear of Igbo domination in the 1960s, and the secession and civil war that followed from 1967 to 1970. The rise of Chief M. K. O. Abiola in the early 1990s, his victory in the 1993 elections and the annulment of these elections by the military regime of General Ibrahim Babangida dominated Nigerian politics for the rest of the decade. The rise of General Abacha to power met with organized opposition among the Yoruba and pro-democracy movements. During Abacha's tenure, various Yoruba groups not only emerged to advocate democracy; some wanted secession or, at the very least, the re-negotiation of the basis of the federation to give power to the different nationalities to shape their own policies, free of a powerful center. The discourse of democracy put ethnicity at its very center. It was no longer how a group of nationalists from different parts of the country would govern and bring progress, but how the representatives of the component units would share power and resources along ethnic lines. The assumption was that the military generals and politicians would always represent ethnic interests, and in this kind of arrangement, the North would always seek domination at the expense of others. If a number of Yoruba politicians had advocated forging alliances with the northerners, many now regarded them as 'dangerous enemies'. Abacha's death in 1998 paved the way for a military disengagement that ended in 1999 with the victory of a Yoruba, Olusegun Obasanjo, as the country's President. Since his coming to power in May 1999, there has been a spate of inter-ethnic and inter-communal rivalries, two of which pit the Yoruba against the northerners. The Sharia is being declared as state law in some northern states, a development that some interpret as an attempt to destabilize the Obasanjo regime. Is there a backlash to the growing influence of the Yoruba in politics? Will the East and North create an anti-Yoruba alliance? Has the emergence of Obasanjo benefited the Yoruba? Should Obasanjo govern as a Yoruba or a Nigerian? These are some of the questions that politicians and analysts pose, revealing that the understanding of politics is still very much conditioned by ethnicity.

148

The Past in the Yoruba Present

149

This chapter explores the historical and contemporary development in the formation of Yoruba identity and how this intersects with Nigerian politics. The premise is that contemporary ethnic politics can best be understood against the background of past events. The massive support that Abiola received was in part because of the previous ordeals of Chief Awolowo from the 1950s to the 1980s, especially his efforts at becoming the country's leader. The lack of support by the Yoruba for Obasanjo during the elections was due to the widely held belief that he did not favor the Yoruba when he was the head of state in the 1970s, that he represents northern interests, that he worked against Awolowo in the 1970s, and that he was lukewarm in his support of Abiola in the 1990s. The current support for him now that he is in power is partly based on the fear that northerners will prevent a Yoruba from succeeding. Having failed to prevent Obasanjo from winning the elections, in preference to their favorite son, Chief Olu Falae, the Yoruba now lend their support to him because they interpret anti-Obasanjo sentiments and campaigns as anti-Yoruba.

The Ethnic Factor The role of the Yoruba in politics has been shaped by five inter-related factors: (i) The manipulation, by Yoruba politicians, of Yoruba and Nigerian history for specific ends presented as collective interests. In other words, political leaders operate as the representatives of an ethnic group. (ii) The promotion of a pan-Yoruba consciousness and cultural exclusiveness built on the myth of Oduduwa, the progenitor of the Yoruba people. In other words, political leaders and representatives subscribe to the historical narratives that regard all Yoruba as the descendants of one ancestor. (iii) The use of ethnocentric traits (for example, common language, boundaries, beliefs and group identity) for inter-ethnic competition in a plural society. The Yoruba politicians act on the assumption that the Yoruba constitute one single nation, and that their interests have to be protected against those of competitors from other ethnic groups. (iv) The claim to early contact with Western education and Christianity which, from the colonial period onwards, became significant criteria in access to jobs, business and politics; and arising from this, (v) A stronger claim to modern civilization than any other group in the federation. Indeed, on the basis of the last point, many members of the ruling class used to argue, during the colonial period and immediately after, that the Yoruba were better qualified than any other group in Nigeria to provide the leadership for the country and serve as the agents of modernization. According to this belief, not only would the Yoruba transform themselves, they would disperse to transform others and their leaders would use their modernization to move Nigeria forward. In 1947, Chief Obafemi Awolowo, the most important Yoruba hero of the twentieth century, justified the claims of the Yoruba to the pre-eminent leadership position partly on the basis of their exposure to superior Western culture, and the varying degree of the infiltration of Western civilization and education into the different ethnic groups. 1 All the claims of the Yoruba to leadership have not gone unchallenged, and 1

Obafemi Awolowo, Path to Nigerian Freedom. London: Faber and Faber, 1947, p. 49.

150

Toyin Falola

certainly not this statement. The leaders of the other groups hardly agreed with this claim. For instance, Chief Nnamdi Azikiwe, equally as ambitious as Awolowo and also regarded as the leader of a rival ethnic group, the Igbo, also articulated the belief in the superiority of his own group. 2 If Awolowo limited himself to Nigeria, Azikiwe went further to regard the Igbo as the leaders of Africa. These men and others made new claims in later years, all leading to a similar conclusion: a particular ethnic group must lead the country for its own sake and that of others. As this belief shaped the nature of political competition, it is important to identify its outcome: (i) Building a pan-Nigerian consciousness or nationalism to sustain the Nigerian nation-state is complicated, since there are competing ethnic nationalisms. Secession and the advocacy of separate developments are strong examples of the clash of nationalisms. The belief of a superior-inferior group makes consensus-building more difficult. (ii) Building strong and viable networks of political associations and massbased political parties is difficult, as the challenges of ethnic loyalty may undermine them. It has always been a convenient game to play off one ethnic group against the other and to disguise class and self-interests as ethnic. (iii) If the representatives of the big ethnic groups are only interested in themselves and their groups, the resolution of divisive issues and the protection of minority interests are difficult, if not impossible. Decisions may reveal self-interest, rather than common interest, thus laying the foundation of a future crisis. (iv) The political elite regards the manipulation of ethnic loyalty as the cheapest and most reliable strategy to acquire and consolidate power. The ethnic identity is manipulated to lay claim to leadership. At the same time, the political elite falls back on this same identity to forestall the underprivileged members of its own group creating alliances with members of other groups, since the 'Others' have been portrayed as rivals, enemies, competitors and people with negative attributes. Would Nigerian politics have been different if poor Hausa, poor Igbo and poor Yoruba would have been united in the politics of change?

The Agency of History As with other Nigerian groups, Yoruba identity is rooted in history. The ethno characteristics were formed during the precolonial period, ethnicity was accentuated during the colonial period, and profound conflicts occurred after the country's independence from the British in 1960. The interactions among the various Yoruba polities during the precolonial period continue to have relevance in modern politics. The Yoruba did not build one political kingdom or empire, but the various sub-units had strong contacts with one another. The factors that promoted the contacts in the past (for peace and war) have been activated in the modern era to build a pan-Yoruba consciousness. While there were various Yoruba groups in the past, each with its own political authority and boundary, in the context of contemporary politics, it is convenient for the political class to occasionally forget this in the agenda of forging a Yoruba 'nation' in a strong enough position to compete with the other 2

Nnamdi Azikiwe, My Odyssey: An Autobiography. London: C. Hurst, 1970, p. 243.

The Past in the Yoruba Present

151

equally large and viable 'nations'. My narrative is intended to underscore two inter-related points: (i) the importance of the states formed by different Yoruba groups, and their relationship to one another; (ii) the attempts by a few states to expand at the expense of others, to extend their territories and 'nationalities'. Boundaries were occasionally adjusted, and people were forced to migrate from one area to another. I want to begin my analysis with the historical past that has been used to provide much of the evidence for Yoruba unity in the political present. I shall isolate the historical evidence that now constitutes the political 'memory' of the Yoruba intelligentsia. It is the 'memory' of this past that enables the contemporary intelligentsia to create a basis for Yoruba regional unity in modern Nigeria and to have a strong foundation for articulating a Yoruba identity. A selective reading of the past does not ensure that present politics will unfold as planned. There is also evidence of warfare and rivalries among the Yoruba, which some may recall when it serves modern politics. Territory, Territorial Sovereignty and Citizenship Two major criteria have been used to create a 'Yoruba map': language and the migration of dynastic leaders from Ile-Ife. Areas where the various dialects of Yoruba are spoken are included in the map of Yorubaland. Language, of course, may indicate that there are other cultural affinities. The second criterion includes all areas whose legends claim that their founders migrated out of IleIfe. Thus in Nigeria, to limit ourselves to one country, the Yoruba occupy most of the southwest. To treat the Yoruba-speaking areas as a unit can be justified only on the basis of cultural and linguistic similarities. These similarities have been further promoted by modern politicians who talk of a common historical experience, and the emergence of a standard Yoruba language. Yorubaland was never a single socio-political unit. But what geographical labels did the precolonial people invent for their territories, and by what name(s) did they refer to themselves? In addition, were they conscious of such labels and names and willing to defend them in the face of threats, both internal and external, to destroy the body of ideas or meanings which the labels and names represented? Were the Yoruba conscious of their ethnicity and identity? The answers to these questions reveal the emphasis on groups (such as the Ekiti, Ijebu and Egba) rather than to a single 'Yoruba nation'. Groups had their territories while citizenship was defined in relation to the membership of a state within a group. Land, an aspect of territorial sovereignty, was communal, and no pan-Yoruba authority ever emerged to control all land. Territorial sovereignty and citizenship were not defined in a pan-Yoruba framework since there were different autonomous groups and city-states. These states did not even evolve a loose political federation. The ruling classes in the different states forged relations with one another, but certainly not one in which one of them assumed an overwhelming control over all the others. It was the British who made the Alaafin of Oyo superior to many kings before the 1930s, and it was the modern political party of the Yoruba, the Action Group, that made the Ooni of Ife superior to the others from the 1950s. 3 Every oba was sovereign in his own domain and he, 3

On the changing nature of Yoruba chieftaincy, see for instance, O. Vaughan, Traditional Power in Modern Nigeria: Nigerian Chiefs and Politics. Rochester, NY: University of Rochester Press, 2000.

152

Toyin Falola

together with his chiefs and lineage representatives, constituted the leading members of the political class. The exception was when imperial control had been imposed, and the territory became a vassal to another kingdom, as in the case of the Old Oyo empire that was able to dominate a number of other groups. Several sovereign city-states belonging to different groups dotted Yorubaland. Each consisted of a big city surrounded by outlying villages of varying sizes. The city was like a metropolis, with a network of adjoining farmlands, hamlets and villages. Some of these city-states were large, like Oyo, the metropolis of a huge empire and the largest in the eighteenth century, and Ibadan the largest in the nineteenth century. The majority were medium-sized, like Owu, which was destroyed in 1826, Ondo, Ile-Ife and Ilesa. Others include the Egba states before the nineteenth century, the Ekiti states and a host of others among the Iyagba, Owe, Oworo and Dumu in the northeast, and Ijaye and Ikale in the southeast. The claim to different autonomy and sovereignty depended partly on the nature and pattern of state formation among the Yoruba. Sources do not mention the formation of one Yoruba nation but rather of several states and groups. Some in fact describe pre-Oduduwa communities, that is, in reference to autochthonous communities that existed before the imposition of dynasties with a connection with Ile-Ife and/or before a process of political centralization. State formation in Yorubaland is associated in myths with the activities of Oduduwa and his 'children'. These myths do not mention the unity among these 'children' to establish a single nation, but rather a host of nations. One inference that could be drawn from the myths is that political centralization did take place in various states at different historical periods. This was, however, a process which involved many people over a long period of time. Those who established dynasties maintained control over large territories. Many names were given to these states. Some of these names derived from the features of the environment such as hills, rivers, soil and vegetation, and others from the personal experiences of the dynastic founders or even of the entire migration of people. Expansion took place until boundaries were established with another citystate belonging to the same group or a different one. The commitment to, and the struggles over, boundaries again indicate the place of different groups and of a state-oriented territorial sovereignty in Yorubaland. A state had its territory, from which the citizens, who in turn belonged to recognized lineages, had their own share of land. A ruler held control of the territory and was recognized by other rulers as having the power and sovereignty; his power was limited to this territory. There were traditions to legitimize claims and the limits of expansion. 4 As the Owa of Ilesa put it in 1882, '[T]he boundaries of each state were well defined and regarded as sacred, so that nothing could induce any one to intrude into the territory of another.' 5 Hills, footpaths, streams, and other physical features were used as demarcations. Diplomacy, treaties and rituals were also used to strengthen the agreements on boundaries. These boundaries separated areas of jurisdiction and interest. When these were threatened, disputes arose between states. 6 Sovereignty had relevance at the micro-level of the city-state and group. So, too, did citizenship. The people used their towns or groups to identify them4

See, for instance, the National Archives, Ibadan (NAT), Nigeria, Ekiti Division, 1/1/518, AkureIkere boundary dispute, 1853, pp. 44-9. 5 British Parliamentary Papers, Vol. 63, Nigeria, p. 68. 6 See, for instance, O. Adejuyigbe, Boundary Problems in Western Nigeria: A Geographical Analysis. IleIfe: University of Ife Press, 1975.

The Past in the Yoruba Present

153

selves. For instance, a person from Owo saw himself as an Omo Owo or shortly as Owo (Owo's citizen) and would not define himself in a pan-Yoruba context. Strong group tendencies gave rise to, and probably promoted, parochial tendencies and cultural variations. The widespread definition of being a Yoruba, instead of an Omo Owo, began to gain currency only from the nineteenth century. The mode of production recognized the central role of households. Production mechanisms depended on local initiatives and more on local raw materials. 7 The point here is not to suggest that there were no economic interactions, but that the mode of production could function, as a system, at the level of the community connected to a regional network. The same was true of the political system. Each state was sovereign and constituted a recognizable territorial unit. Its government was also sovereign, and its rulers were the symbols of the state. They exercised judicial, executive and legislative powers. The political system of the states rested on a combination of the lineages, various associations, and the paraphernalia of a central authority. At the lineage level, the Baale (or olori ebi), that is, the lineage head, was in charge. He exercised a penal authority limited to fines and minor impositions, and chastisement. Associations included age-grades and trade guilds, both of which exercised moral authority over their members. The institutions of the central state's authority were dominated by the oba and the chiefs, both acting in a council. 8 The institutions of government were replicated in many areas, but the pattern of authority showed variations. Oyo represented the model of an empire, with a powerful monarch. The Egba formed a loose confederation; the Ife a centralized provincial administration; and other states such as the Ekiti, Ijebu, Ondo, Owo and Awori lacked large central political organizations. New changes were introduced during the nineteenth century. 9 The military became pre-eminent in politics, primarily because of the insecurity of the age. The military had new ideas of government, arising from their crucial strategic functions and substantial wealth. Because they owed their prominence to exploits in war, they tended to have little regard for monarchical institutions. New ideas were put into practice in Ibadan where a military aristocracy was established; in Ijaye, which had a military dictatorship; and in Oke-Odan and Abeokuta where a military federation was practised. 10

Centrifugal Tendencies The various Yoruba states maintained close relations with one another. The dominant ruling class in each state also had strong interests in promoting centrifugal tendencies, primarily to widen the power base. In the first place, the myth of Oduduwa and of a common ancestry was popularized, partly to strengthen the link between the ruling class and to prevent, as much as possible, the emergence of a counter-hegemonic force which could destroy the dynasties 7 Toyin Falola, The Political Economy of a Pre-colonial African State: Ibadan, 1830-1900. Ile-Ife: University of Ife Press, 1984. 8 Toyin Falola and Dare Oguntomisin, The Military in Nineteenth Century Yoruba Politics. Ile-Ife: University of Ife Press, 1984, chap. 1. 9 Toyin Falola, Tower Drift in the Political System of Southwestern Nigeria in the 19th century', ODU: A Journal of West African Studies, 21, January/July, 1981, pp. 109-27. 10 Falola and Oguntomisin, The Military in Nineteenth Century.

154

Toyin Falola

and royal families. The myth also reinforced the ownership and control of territories as well as of land. The affinal relationship referred to below served as a further justificatory ideology; so also did the exchanges of gifts, and diplomatic practices and exchanges. Secondly, inter-group relations widened the network of markets and trade, thus increasing the avenues for profit by the enterprising members of the ruling class as well as the peasants. Thirdly, and more importantly, wars were fought to subjugate neighboring and distant polities. This enabled the accumulation of extensive booty in goods and men, as well as considerable access to gifts and tribute. These massive gains have been clearly demonstrated in the case of Ibadan, 11 as well as in the activities of the leading war heroes of the nineteenth century. 12 Factors of inter-group relations were diversified: trade, migrations, diplomacy and war. Migrations within the Yoruba region were common, brought about by such reasons as the flights from wars and political upheavals, and the search for new and better abodes. The nineteenth century witnessed massive migrations which had the effect of altering the map and the demographic compositions of the region. 13 Such migrations allowed for borrowing and the spread of culture. The diffusion of artifacts and aspects of social institutions represents major evidence of the impact of inter-group relations and their integrative role in the Yoruba region. For instance, the Egungun-Oyo (Oyo masquerade) spread from the north to the south and northeast during the nineteenth century. Similarly, the Orisa-oko (the cult of farming and fertility) and Sango (the deity of lightning and thunder) spread from the Oyo to the other Yoruba groups. Other institutions emanated from other areas and spread to the Oyo; an example of these was the Epa-type masquerade headpieces. The ruling dynasties in the various states forged relationships with one another by promoting 'brotherhood relations'. The cordial relations among them were sometimes explained in affinal relationship. Several of the royal families (members of the ruling class) claimed Oduduwa as their ancestor. These 'children of Oduduwa' constituted the dynasties in a good number of kingdoms. The number of these 'children' has been variously put at between seven and twenty-one. 14 Inter-group relations among the kingdoms were fostered by this myth of common origin. This was occasionally demonstrated in coronation rituals when some of the kingdoms asked Ile-Ife for the staff of office for their new rulers. 15 Some claimed secondary relations with the sons of Oduduwa, that is, the founders of their kingdoms were children of the sons of Oduduwa. For instance, the Akure and Osogbo dynasties claimed blood relationship with the Owa of Ilesa (a 'son' of Oduduwa), while Iwo claimed a connection with the son of a female Ooni of Ife. The rulers who claimed to have been born of the same mother had common rituals to celebrate her. There were cases when the ruling dynasties in contiguous communities claimed a common descent from the same mother. In situations like this, members of the ruling class regarded themselves as siblings and forged deep relationships. This 'blood' relationship was used to justify other important unifying steps, such as the exchange of gifts, custody of 11

Falola, The Political Economy. Toyin Falola and Dare Oguntomisin, Yoruba Warlords of the Nineteenth Century. Trenton, NJ: Africa World Press, forthcoming. 13 Toyin Falola and Dare Oguntomisin, 'Refugees in Yorubaland in the Nineteenth Century', Asian and African Studies, XXII, 1 and 2, pp. 67-79. 14 Samuel Johnson, The History of the Yorubas. Lagos: CMS, 1921, p. 11. 15 See, for instance, J. O. Olubokun, Itan Uyin. Yaba: Forward Press, 1952, p. 5; and NAI, Oyo Prof 2/2/1372, Yoruba Crowns, Rights and Privileges to wear crowns by certain chiefs. 12

The Past in the Yoruba Present

155

one another's princes for effective training, exchange of visits by chiefs, and trade relations. Affinal relationship was also used to prevent conflicts among the members of the ruling class in the same state. Two or more ruling houses existed in most Yoruba towns. Marriage ties could cement relationships among them. The use of marriage, however, cut across states. For political strategy and to widen the network of relationship, princesses were betrothed to princes and chiefs in other states. Examples of this strategy are many, and well narrated in various traditions. The rulers of such states treated one another as 'cousins' and their people as 'kinsmen'. When the products of such marriages grew to acquire political power, inter-group relations were further strengthened because of their dual loyalty to two states. The conjugal and affinal ties between rulers impacted upon how their subjects related to one another. The people followed their examples, that is, of exchanging sons and daughters. The notions of security, safety and 'kinsmen' were implied in such exchanges; in other words, they were exchanging with people in states where it was safe to do so. The network of social and blood relationship was, therefore, wide. Though the several wars of the nineteenth century destroyed a good number of these affinal relationships, they were still remembered and also occasionally served to prevent large-scale hostilities and total destruction of communities. Ibadan, which built the most effective military machine of the century, occasionally considered this factor. It did not attack most Oyo-Yoruba states because their people were regarded as kinsmen. Even when Ijaye - the rival Oyo-Yoruba state - was attacked in 1860, the attack was undertaken after prolonged deliberations. 16 Concrete diplomacy was used to back 'brotherhood relations'. The Yoruba, like most other African people, were conversant with the art and practice of diplomacy. 17 There were several conventions: diplomatic agents were recognized; the status and power of different diplomats were known; political agents were exchanged; the inviolability of the person of the diplomatic agent was recognized; diplomatic communications were generally immune from interdiction; and symbols and signs were used as the secret language of diplomacy. War was yet another factor which brought communities together. Successful imperial wars put separate states and sub-groups under a common political umbrella. The Oyo were able to achieve this in the seventeenth and eighteenth centuries when they established an empire whose territorial extent was very wide. 18 Most of the vassal colonies enjoyed autonomy, but Oyo had to ratify the appointment of their Oba or Baale. In addition, they went to Oyo to obtain their titles and the Alaafin could also depose them. They appeared periodically at Oyo where they re-affirmed their loyalty. The orders emanating from Oyo had to be obeyed, and Oyo's political agents were resident in the colonies to monitor local developments. Among the other factors making for cohesion were the blood and social relationships among the members of the ruling class in the various cities and vil16

Johnson, History of the Yorubas, p. 335. See, for instance, Robert S. Smith, Warfare and Diplomacy in Pre-Colonial West Africa. London: Methuen, 1976; second edition London: James Currey 1989; and J. F. Ade Ajayi and R. S. Smith, Yoruba Warfare in the Nineteenth Century. Cambridge: Cambridge University Press, 1964. 18 J. A. Atanda, The New Oyo Empire: Indirect Rule and Change in Western Nigeria, 1894—1934. London: Longman, 1973, p. 13. 17

156

Toyin Falola

lages; Oyo's military might and ability to subjugate and coerce the colonies; and the domination of commerce, which enhanced the economic power of the metropolis at the expense of the colonies. Finally, many areas were linked by roads, which served as arteries of trade and communications. If the preceding evidence points to opportunities for solidarity among the Yoruba, there is also evidence of alliances forged with others within Nigeria. The groups in the peripheries as well as in the border zones inevitably had to interact with their neighbors. All the Yoruba sub-groups had contacts with other nonYoruba in the West African region. Both forms of interaction revolved around diplomacy, trade, cultural ties and wars. They were also well developed and encouraged the penetration of ideas, goods and people. They were promoted by ecological differences, the necessity of mutual inter-dependence, the benefits from a wider commercial and economic network, and imperial desires. Only a few examples can be cited. Oyo's relations with its northern neighbors, notably the Nupe and Bariba, brought a great deal of exchanges and occasional conflicts. The economic integration of Yorubaland into the Hausa commercial network was very deep. 19 Hausa traders brought goods of diverse origins to Yorubaland in exchange for kola nuts and other products. The Benin expansion into the area of modern Ondo and Lagos State from the sixteenth to the nineteenth centuries was equally significant. There is more evidence to show that the Yoruba did not limit interactions to fellow Yoruba groups. Long-established relations have continued to this day, in spite of new boundaries, new administrative centers, and ethnic rivalries.

Manufacturing Yoruba: An Idea and A Consciousness If the idea of a united Yoruba can be contested, there remain many factors which the modern intelligentsia have used to forge the consciousness of a Yoruba nation. In spite of variations, the Yoruba can refer to cultural similarities and social formations with communal boundaries. There are myths and events that can generate ethnocentric pride and construct an identity. The most vigorous attempt to create the 'Yoruba nation' began during the nineteenth century. It was during that century that the name 'Yoruba' was used to refer to all the Yoruba sub-groups. Names such as Aku, Nagun, Anago, Olukumi, and Yoruba had been used by neighbors and European visitors to describe various Yoruba groups. 20 Yoruba became the most popular name. Its popular usage began in Sierra Leone, when the missionaries were interested in studying African languages, including that of Oyo. By the 1880s, Yoruba had become a common usage among the educated elite. More important than the name was Yoruba consciousness, which developed among the liberated slaves in Sierra Leone and which they imported into Nigeria from 1838 onward. 21 From among the returnees and the products of missionary education introduced in the second half of the nineteenth century emerged an educated elite. This elite was interested in key positions in the church and 19 See for instance, Paul Lovejoy, 'Interregional Monetary Flows in the Precolonial Trade of Nigeria', Journal of African History, XV, 1974, pp. 536-85; and Mahdi Adamu, The Hausa Factor in West African History. Zaria: Ahmadu Bello University Press, 1978. 20 See, for instance, S. W. Koelle, Polyglotta Africana or a Comparative Vocabulary of nearly 300 words and phrases in more than one hundred district African Languages. London: CMS, 1963 reprint; original, 1853 p. 5. 21 For information on the emigrants, see J. H. Kopytoff, A Preface to Modern Nigeria: The 'Sierra Leonians* in Yoruba, 1830-1890. Madison, WI: University of Wisconsin Press, 1965.

The Past in the Yoruba Present

157

civil service, in dominating commerce and also in sharing power with the 'traditional' elite. 22 They perceived Europeans as opponents who stood in their way of achieving their ambitions. One way to overcome the threat posed by Europeans was to promote ethnocentric values, partly to strengthen intra-class unity and provide ideological rationalization for some of their enterprising moves to indigenize the church and set up businesses of their own. Several ethnocentric values assumed great importance. The myth of a common origin was popularized to create a pan-Yoruba identity and destroy the group loyalties which the nineteenth century wars had effectively consolidated. Language, too, became another factor, while they also began to emphasize the similarities in Yoruba culture. Yoruba acquired a written form, one of the earliest in the country with standard alphabets. Although it was the Oyo-Yoruba dialect that was chosen, other groups accepted it, and it became a school subject as well. The standardization enabled the language to unite all those who could read and write. Whether one was Ijebu or Egba, the standard language was available for dialogue. From 1875 till the end of the century, some Yoruba elite promoted Ethiopianism, that is, 'African nationalism expressed through the medium of the church'. 23 Ethiopianism and other aspects of missionary activities and reactions to Europeans engendered cultural nationalism and ethnocentric values. During this period, the Lagos press became virulent; the educated faction advocated several reforms and changes, including the establishment of an African university; and respected Yoruba 'scholars' emerged. From among the latter, a flourishing Yoruba historiography was born, culminating in the writing of Samuel Johnson's work. Patriotism to the Yoruba underlined most of these writings. In the last words of his monumental work, Samuel Johnson spoke for the elite of the nineteenth century who had accepted a common Yoruba identity and hoped for progress in the twentieth century. 24 Johnson's desire for unity was against the background of the century-old civil war among the Yoruba and the need to enjoy peace and economic boom. Johnson's successors also desired unity, but in the context of ethnic competition in a colonial and post-colonial setting.

Beyond the Ethno During the twentieth century, ethnocentric values gave way to ethnic rivalries, as the former became transformed in the service of the latter. There is one decisive reason for this big leap: British rule. The colonial period introduced farreaching changes which have, in many ways, laid the foundation of many future problems and challenges. 25 Colonialism certainly accentuated group differences, encouraged the formation of classes which benefited from promoting group differences and hostilities, and introduced measures which made it difficult to achieve political integration and rapid economic development. 26 22 See, for instance, E. A. Ayandele, The Educated Elite in the Nigerian Society. Ibadan: Ibadan University Press, 1974. 23 E. A. Ayandele, The Missionary Impact on Modern Nigeria, 1842-1914. London: Longman, 1966, p. 177. 24 Johnson, The History, p. 642. Abiodun is capitalized in the original. He was the Alaafin in the late eighteenth century, at a time of great prosperity for the Oyo Empire. 25 For some of these changes, see, for instance, Toyin Falola, ed., Britain and Nigeria: Exploitation or Development? London: Zed, 1987. 26 Okwudiba Nnoli, Ethnic Politics in Nigeria. Enugu: Fourth Dimension, 1978.

158

Toyin Falola

The Yoruba elite began to articulate and manipulate a common consciousness of the Yoruba identity in relation to other ethnic groups. This engendered competition with others, and was expressed in inter-ethnic conflicts and inter-ethnic discriminations in job opportunities and access to national resources. The development of ethnicity owed much to the colonial policies and programs which created new class structures and relations. The dominant class in the colonial era, which subsequently inherited power in I960, benefited from the manipulation of ethnicity for private ends. It was not difficult for them to do so because the British did so much to promote divisions along ethnic lines. Okwudiba Nnoli has studied some of these measures and has shown how ethnicity was deliberately politicized for private ends. 27 A genuinely nationalist, pan-Nigerian political party had not emerged by 1960, when the country obtained its independence. Rather, what the country had during the crucial era of decolonization were parties organized along regional and ethnic lines. In the case of the Yoruba, a far-reaching political expression of ethnic solidarity was adopted in 1945 with the formation of a cultural-cum-political society known as the Egbe Omo Oduduwa (the descendants of the children of Oduduwa) by Chief Awolowo.28 One of the aims behind it was to manipulate the idea of Yoruba identity and consciousness in order to compete with other Nigerian groups. It held its inaugural meeting in Lagos on 28 November 1947, and subsequently couched its aims in an ethnic context. These included, inter alia: the fostering of the spirit of co-operation, unity and brotherhood among the descendants of Oduduwa; the co-ordination of educational and cultural programs among the Yoruba; the discouragement of 'intra-tribal prejudice among descendants of Oduduwa who for linguistic differences failed to recognize themselves as branches of the same stock'; and the importance of striving for the preservation of a 'traditional monarchical form of government of western Nigeria so as to fit properly in any future political set up for the government of Nigeria'. 29 From 1945 onwards, the Egbe designed a host of strategies to pursue their aims and compete with other ethnic groups, notably the Igbo (who also had the Ibo State Union) and the northern intelligentsia who, in 1948, also established a cultural-cum-political union known as the Jam'yyar Mutanen Arewa (The Association of Peoples of the North). Certain aspects of the Egbe's strategy were built on ethnocentric values. It created the Oduduwa National Day, an annual event fixed for 5 June during which young people engaged in athletic competitions and thanksgiving services were held in churches and mosques. 30 The Obas and chiefs were mobilized, and many of them identified with the aims and activities of the Egbe. As a political strategy, the Egbe justified the need for chiefs in any new political arrangement. It was the Egbe Omo Oduduwa that later constituted the core of the Action Group, a political party with a substantial Yoruba base. 31

27

Ibid. S. O. Arifalo, 'Ethnic Political Consciousness in Nigeria, 1947-1951', Geneve-Afrique, XXXTV, 1, 1986, pp. 7-34. 29 The Constitution of the Egbe Omo Oduduwa. Ijebu-Ode, 1948. 30 See O. A. Sobande, 'Path to Unity and Culture: Being a Comprehensive Survey of Principles and Practices of Egbe Omo Oduduwa' Lagos, 1952, mimeo. 31 For details on the AG and the other political parties of this period, see Richard L. Sklar, Nigerian Political Parties: Power in an Emergent African Nation. Princeton, NJ: Princeton University Press, 1963. 28

The Past in the Yoruba Present

159

Chief Obafemi Awolowo and the Limits of Yoruba Ethnicity Chief Awolowo dominated Yoruba politics from the 1950s till his death in the 1980s. He was successful in two major ways: the consolidation of ethnicity and the 'modernization' of the Yoruba. He took the essential elements of Yoruba history that brought them together during the precolonial era and refined them into a political and cultural ideology. All Yoruba, now children of one ancestor, Oduduwa, should turn a large family into a political party under him. It was as if Oduduwa passed the baton to him to continue the great work of unification. This was the ultimate ethnic game, and Awolowo had to confront Ahmadu Bello and Nnamdi Azikiwe who were also playing the same game. In most of the 'struggles' - conflicts for power and wealth - the dominant political leaders and their followers obscure the pursuit of selfish interests by championing the cause of ethnic groups. How politicians play this game of combining self- and ethnic interests has been explained by B. J. Dudley. 32 Awolowo was both brilliant and astute. He sought the means to create a solid constituency behind him, and then to articulate a bigger role for himself and his ethnic group in the larger national front. He was a modernizer - his second success - and arguably the country's most successful leader in his ability to intellectualize an agenda and execute it. His premiership of the Western region, the Yorubaland homeland constituted into a regional unit in a federal system by the British, from 1952 to 1959 is now regarded by his admirers as the 'golden age' of the Yoruba in their modern history. A literature, whose significance will become clearer in the following section of this chapter, summarizes Awolowo's successes in a superlative manner. 33 For the Yoruba who wanted Awolowo as the country's President, he would successfully modernize the country and further empower the Yoruba. Awolowo lost the election of 1959 to become Prime Minister. He suffered during the First Republic, as he was arrested for treason and sentenced to a ten-year prison term. The AG split so badly that it lost control of the region to a splinter party and a state of emergency was declared in 1962. The common interpretation of his ordeal as persecution at the hands of his northern political opponents turned Awolowo into a god among his people. His stature actually rose in the 1960s, even more than when he was in power. Following the second coup of 1966, he was released from prison, and became a federal minister the following year. His role as Finance Minister during the Civil War alienated him from the Igbo leadership who expected that a Yoruba alliance would have dealt a final blow to northern domination. When Awolowo re-entered politics in the 1970s, he repeated his success and his failure. As before, he built a formidable political machine and party that had the Yoruba behind him. Again, as before, he lost to a northerner and in the last election of his career in 1983, he lost again. 34 The Yoruba never expected Awolowo to lose the elections. He did so well in the 1979 elections, and the narrow victory of his NPN opponent had to be sustained by a controversial legal decision by the Supreme Court. Obasanjo was then in power, as a military head of state, and he was believed to be pro-NPN. 32

B. J. Dudley, Instability and Political Order: Politics and Crisis in Nigeria. Ibadan: Ibadan University Press, 1973, p. 164. 33 Egbe Omo Yoruba, Family Handbook and National Blueprint. Washington, DC: Egbe Omo Oduduwa, 1996, pp. 4-5. 34 His career is one of the best documented in Nigerian politics. See, for instance, Toyin Falola et al., Chief Obafemi Awolowo: The End of an Era? Ile-Ife: Obafemi Awolowo University Press, 1988.

160

Toyin Falola

Abiola was a leading member of the NPN and a critic of Awolowo. Both Obasanjo and Abiola were later to pay the price for what was regarded as a betrayal of Awolowo, although both were also to be forgiven much later. Awolowo's failure reveals the limits of the ethnic game at two levels. The first, and the most important, is that the political actors in the other regions were also astute in consolidating their power bases and using ethnicity for the formation of political parties and in the elections. In the east and north, Awolowo was portrayed as an outsider. He was able to do well in the Middle Belt and Niger Delta minority areas. Azikiwe and the Igbo leadership successfully ensured that he would not succeed among the Igbo. In the North, the pioneer politicians successfully recreated the caliphate of the nineteenth century, using Islam to unite the North under one political party. 35 The North has been the most successful in the use of ethnicity, combined with Islam, to dominate power, both during civilian and military rule. The second is that there were always pockets of resistance to Awolowo among the Yoruba, an avenue that was exploited by politicians and parties from other regions to justify electoral malpractices, build alliances, and actually recruit genuine supporters. The resistance among the Yoruba, largely concentrated among the Oyo-Yoruba groups in Modakeke, Oyo, Ibadan and Ogbomoso, revealed old rivalries between groups and the consequences of the rift that split the Action Group in the 1960s. There was, for instance, the hostility between two powerful kings, the Ooni of Ile-Ife and the Alaafin of Oyo. The AG had regarded the Ooni as the supreme Yoruba king, in line with the politics of using the myth of Oduduwa as a political ideology. The Alaafin, regarded by the British until the 1930s as the supreme Yoruba king, felt slighted. If the Ooni has traditionally sought alliance with the party in power among the Yoruba, the Alaafin has had to seek allies in the north. Both try not to keep the same powerful friends. Old and new rivalries have often been invoked to justify the formation of alliances between Yoruba and other ethnic groups, as in the case of prominent members in the National Party of Nigeria in the 1970s and 1980s. Here is the fault line that weakens the ethnic solidarity: there are those who do believe that the Yoruba cannot control federal power unless they form alliances with other ethnic groups, notably the northerners. Chief M.K.O. Abiola subscribed to this position; he made many friends in the north, used Islam as an opportunity to reach out, won an election, but never assumed power. The Abiola case demonstrates that the strategy of alliance, too, is not foolproof.

Yoruba Ethnicity in the 1990s: The Revival of 'Oduduwa Kingdom5 The failure of Abiola, like the failure of Awolowo, was devastating to the ideologues of Yoruba unity and their claim that the Yoruba have the strongest credentials to have a son as the country's President. The way in which the elections were annulled in 1993 was interpreted to mean that the vested interests in the North did not want a southerner and a Yoruba in power. Abiola became a hero, although he was despised in the 1970s and 1980s for being anti-Awolowo and pro-Sharia. His support for the Sharia in the 1980s alienated him from the Christians, but they too chose to forgive him. As the Yoruba grew louder in crying injustice, Abiola's supporters in the north began to abandon him. The 35 See, for instance, Toyin Falola and Hassan Matthew Kukah, Religious Militancy and Self-Assertion: Islam and Politics in Nigeria. London: Avebury, 1996.

The Past in the Yoruba Present

161

military regimes of both Babangida and Abacha cleverly condemned the proAbiola rally and protest as the excessive ambition of the Yoruba to control the federal government. The challenges to the Abacha regime have generally been described under the rubric of pro-democracy movements, and the resurgence of Yoruba ethnicity has gone un-captured. Yoruba ethnicity received its biggest expression in the 1990s, mainly in reaction to the annulment of the elections won by M.K.O. Abiola. In the expression of Yoruba ethnicity, three approaches were prominent. The Democratic Option There was a political path, pursued by the established politicians, which ultimately led to the formation of the Alliance for Democracy (AD), the party that now controls power in all of the Yoruba-speaking states. The activities of the pro-Abacha movements were concentrated in the southwest, among the Yoruba. Many members of the National Democratic Coalition (NADECO) were leading Yoruba politicians. The Yoruba also formed an association, the Afenifere, a panYoruba gathering which brought within its ranks people from different political parties and places. Strong members of the Afenifere later became influential in the AD. Up until the death of Abacha and Abiola in 1998, the focus of the politicians was on demanding the mandate of Abiola and a national conference to discuss the future of the country. When Abiola died, the politicians joined in the transition program. The anti-Abacha, pro-Abiola associations were transformed into a political party controlled by the Yoruba. So powerful did the AD become that it was difficult for another party to penetrate the areas of Yorubaland. Chief Olu Falae was the presidential flag-bearer, in a two-way race with Obasanjo, a member of the People's Democratic Party. 36 Once again, as with the formation of the Action Group and the Unity Party of Nigeria in the 1970s, Yoruba ethnicity found expression in the formation of a political party. The Autonomy or Secession Option There have been various groups, within and outside Nigeria, seeking the autonomy of the Yoruba either as a separate nation or a strong one within the federation. The denial of Abiola5s presidency was regarded as sufficient reason to create a new Yoruba nation or seek autonomy in a weakened Nigeria. So important did this option become that it was widely discussed in the mid-1990s, and a few organizations devoted time and attention to its articulation. I want to review the statement and argument of one such association, the Egbe Omo Yoruba (EOY). The EOY was an amalgamation of many Yoruba associations in the United States and Europe, with endorsement from leading Yoruba politicians. The EOY has published a handbook and a magazine, in addition to maintaining a web page. 37 Using the language of redemption, agony and hope, pro-autonomy groups envisaged an end to the Yoruba unless they pulled out of Nigeria or attained sufficient independence in a new political arrangement. Membership in Nigeria is regarded as a curse, and the Yoruba cannot 'entrust their destiny in the hands of others'. 3 8 According to the pro-autonomy groups, except for a 36

For an elaboration of the politics of this period, see Toyin Falola, The History of Nigeria. Westport, CT: Greenwood, 1999, chap. 14. Yoruba Autonomy Alert; Yoruba Family Handbook and National Blueprint; and http: //www.yorubanation.org/. 38 Yoruba Family Handbook, p. 1. 37

Toyin Falola

162

brief period under Awolowo, the Yoruba have suffered enormously during the twentieth century. In the first half of the century, the British took away power from their chiefs and kings, forced them to become part of Nigeria, and 'the main instruments of British colonial control, the army and the civil service, were set up and used mostly against Yoruba interest'. 39 This period was followed by one of devastation, according to the pro-autonomy groups. Described by the EOY as the 'years of occupation' by the Hausa-Fulani, the Yoruba were destroyed. The itemization of the consequences reveals how they perceive ethnic competition, regarding the Yoruba as the victims and the Hausa-Fulani as the oppressors. 40 The document goes further to discuss the implications for the Yoruba: i. Destruction of legitimacy. Since we have been having military regimes that have no roots among the people, the concept of legitimate leadership is gradually being eroded among the people. Might is now right. The so-called governors or administrators, holding their positions at the behest of the occupying power, behave as if they own [sic] no responsibility to the people. ii. Destruction of heritage. Having no emotional attachment or intellectual understanding of the heritage of the people, the so-called rulers posted from the center continue to waste even the scarce resources and heritage that they met on the ground. As far as they are concerned, the market would soon be over and the earlier and faster they grab what they can, the better. iii. Lack of leadership focus. Since those who are ruling are not the elected leaders of the people, they do not share the people's hunger for education, technology, science, cultural and artistic liberty, afreeand socially responsible press, full employment, democracy and social justice. They see any emerging pattern of independent leadership as a direct challenge to their illegitimate privilege. * Arising from these depressing conditions, the Yoruba are enjoined to assert their independence and explore the options of creating autonomy within Nigeria and using their skills to develop themselves. 42 Pro-autonomy and secessionist groups, like Awolowo in the years before them, have accepted the historical evidence of Yoruba unity, and have disregarded any data on division. They regard the Yoruba as victims, and not one of the 'big three' (Hausa-Fulani, Igbo and Yoruba) who dominate the country. The failure of Awolowo and Abiola to attain power was read not as a personal tragedy, but as an ethnic tragedy in which a collective nation was robbed of its well-deserved power. In the view of these groups, the victims must empower themselves: ethnicity must not just involve a struggle with others, but an independence movement. I want to turn to yet another group that has pushed the demand to the zone of violence. The Oodua People ys Congress Now a radical and an 'anarchist' organization, the Oodua People's Congress (OPC) started as an association to advocate unity and autonomy for the Yoruba. Founded by Dr Fasheun, a medical doctor and politician, its mission was to seek self-determination for the Yoruba, protect their interests and fight for their rights. Its agenda included the funding of a research center in Ile-Ife, occasional public lectures, and massive support for a leader who stands for the Yoruba. 39 40 41 42

Ibid., Ibid., Ibid., Ibid,

p. 4. pp. 5-6. pp. 6-7. pp. 7-8.

The Past in the Yoruba Present

163

Indeed, a number of prominent Yoruba have honored the OPC with lectures and funding. The OPC split in 1998, over the issue of the transition program announced by General Abubakar who replaced Abacha. Dr Fasheun argued that the transition should be supported, since a Yoruba would become President, but that the organization should not identify with any political party. Another faction, led by his deputy, Ganiyu Adams, was against the transition. Although Fasheun was in his sixties and a veteran politician, Adams, a 29-year-old carpenter and a newcomer to politics, created a faction that became more popular. The Adams' faction became militant and anti-government. A number of young men joined the organization, turning it from a political interest group into a radical youth movement. Adams re-defined the agenda of the OPC to include fighting criminals, thus bringing the OPC into confrontation with gang members, known as the 'Area boys'. To strengthen its own power as well, Adams' OPC also recruited a number of 'Area Boys' to its fold, thereby transforming the organization into a Vigilante group'. In Lagos, Ibadan and Ilesa, Adams' OPC confronted criminal organizations, relying on mob popularity. According to Adams, fighting criminals is part of defending the interests of the Yoruba. 43 With Obasanjo in power and the AD in control of the Yoruba-speaking states, the belief is that either the activities of the autonomy/secessionist groups should end or that they should become less combative, for fear that the army will overthrow a government headed by a Yoruba. Those who opted for the democratic option now have the upper hand; many are in power and profiting from it, and groups such as the OPC are no longer regarded as allies but as enemies. In 1999 and 2000, the OPC has been credited with killing police officers, creating havoc in different parts of Lagos, causing a market crisis that led to the death of over a hundred people, and house burning, armed robbery and insecurity. Regarding the OPC as a successful organization, young people in other parts of the country formed a few similar associations, such as the Arewa People's Congress, comprising northerners, and the Ijaw Youth Congress, all to constitute pressure groups or to champion the ethnic cause. The scenario of widespread violence all over the country became frightening to the government. So bad did things become in Lagos State that the President threatened, in January 2000, to declare a state of emergency if the state government could not check the activities of the OPC, whom he described as a group of criminals. As Adams has been able to recruit many young men and purchase weapons, it is unclear who is financing them and what the true missions have become. The police and the OPC are now enemies, and Adams a scapegoat of a state riddled with insecurity.44 The members of the AD are now put in the uncomfortable position of dissociating themselves from an organization that they endorsed when Abacha was in power, but which is now stigmatized as a gang. The OPC continues to use the ethnic banner and the agenda of protecting Yoruba interests. One thing that is clear is that, whenever the OPC becomes a useful tool, politicians can be trusted to move closer to its restless members.

43

Interview with Ganiyu Adams, Tell, 31 January 2000, p. 26. See, for instance, a detailed report on Adams and the police search for him in The News, 31 January 2000. 44

164

Toyin Falola

Conclusion Yoruba ethnicity has evolved over time, reaching its peak in the 1990s with the vigorous demand for autonomy. In their long precolonial history, the Yorubaspeaking people did not constitute a single, socio-political unit. Their different groups and states were autonomous units. However, the groups and states established relations with one another. Each city-state forged an identity through the use of myths, and war, trade, diplomacy and marriages fostered inter-group relations. The Yoruba also developed relations with their neighbors, through commerce, war, diplomacy, inter-marriage and cultural borrowing. Thus, history provides evidence of separate development and regional co-operation, thus providing modern political actors with historical materials and multiple conclusions to draw from. The consciousness and manipulation of a pan-Yoruba identity began during the nineteenth century and ethnicity was consolidated during the twentieth century. The consciousness and consolidation were due to the emergence of a new elite, the rapid changes that accompanied colonial rule, and the incorporation of the Yoruba into modern Nigeria. A new intelligentsia, led by Awolowo, called on the Yoruba to ignore the evidence of past autonomous development and choose instead that of the creation myth, similar culture and inter-group relations. In other words, politics demanded that the Yoruba submerge the parochialism of their many city-states and sub-groups and choose the nationalism of one Yoruba. A pan-Yoruba identity works when the Yoruba are involved in the broader arena of Nigerian politics where they have to compete. However, as they also have to organize their space and share resources, an ethnic fault line rears its head: the interests of different city-states and groups (for example, Ekiti, Ibadan, Egba) can be activated. Thus, one can discern multiple layers in the construction and use of Yoruba identity: (i) at the national level, a pan-Yoruba identity is necessary in order to present a united front in a competitive system; (ii) at the regional/state level, a group identity becomes important. Here the Yoruba accept their partition into the various states of Lagos, Ogun, Oyo, Osun, Ondo, and Ekiti; (iii) at the group/state level, cities and villages have to compete, thereby creating the need to foster 'township identities'; (iv) at the township level, lineages and wards create boundaries of exclusion and inclusion. All these variations and multiple identities shape the relevance and use of history. They subject history to creativity, propaganda and abuse. Policy options have to consider the reality of multiple identities. For instance, suggestions on confederacy ignore the fact that the values and interests of a sub-ethnic group can assume dominance when an ethnic group acquires more power, or that new minorities can emerge in the process. Federalism, on the other hand, enables pan-Yoruba consciousness to flourish. The emergence of a relevant ideology that would transform Nigerian society must grapple with the problems of building a pan-Nigerian consciousness. More importantly, it has to focus on values that cut across ethnic boundaries. Ethnicity will continue to shape the outcome of democracy. For half a century, the Yoruba have always stood behind a leader and party that represents regional-cum-ethnic interests. Awolowo's Action Group and the Unity Party of

The Past in the Yoruba Present

165

Nigeria, and the current AD, have had to seek alliances in the quest to win the control of the center. In all the alliances, they have always demanded that a Yoruba should be the presidential flag-bearer. In other words, alliances have been formed to ensure Yoruba hegemony or interests. Abiola and Obasanjo, who have succeeded in winning elections, have done so by presenting themselves as the representatives of interests bigger than those of the Yoruba. Abiola consistently fell on the use of extensive connections with northerners and with Islam. Obasanjo, on the other hand, presents himself as a 'nationalist5, as a Nigerian before being a Yoruba. Abiola was denied power partly because of a perception that he would not serve northern interests; Obasanjo obtained power partly because of a belief that he would serve northern interests. Events since 29 May 1999, when Obasanjo assumed power, have shown that the forces that created the problems of the 1990s are still in place, that the role of ethnicity and communities in politics remains strong, and that the interests of the various nationalities have not been discussed, to say nothing of being addressed. Ethnicity remains important. The self-determination of each group has to be considered, which is why the country opted for a federal system in the first instance. The center has to decentralize power for ethnic groups to attain some of their aspirations. The more centralized and autocratic the center is, the more assertive the ethnic groups become. Prolonged military rule has undermined federalism and is actually responsible for the demands to restructure the basis of the federation. Economic decline and worsening living standards have called into question the relevance of the state and the need to seek alternative models for the Nigerian federation. Ethnic and religious nationalism has become powerful in the light of the failure of the Nigerian state to meet the expectations of its people. With Obasanjo in power, the Yoruba stand to be implicated in his administration. Whether he fails or succeeds, they have achieved their long-standing dream of producing a President. If Obasanjo represents their chance for a long time to come, should he, then, govern as a Yoruba with Yoruba interests in mind? This option is not open to him at a time of rising communal, religious and ethnic nationalisms in various parts of the country. The Yoruba will continue to speak of the 'Oduduwa kingdom' and the possibility of autonomy, thereby retaining the relevance of ethnicity in a new millennium.

Ill

Ethnicity & the Politics of Democratization

10

E. S. ATIENO ODHIAMBO

Hegemonic Enterprises & Instrumentalities of Survival: Ethnicity & Democracy in Kenya Now the concept of 'social change' is more than a mere umbrella for several parallel, probably somehow related changes in diverse aspects of social life; it denotes the systematic transformation of a particular society. But at what level do we set 'society'? . . . It was much easier as late as the 1940s, to speak of local social systems like those of the Asante or the Luo as being societies [rather] than whole colonies like the Gold Coast or Kenya. (Peel, 1984: 142) We all come from one womb, the common womb of one Kenya. The blood shed for our freedom has washed away the differences between that clan and this one. Today there is no Luo, Gikuyu, Kamba, Giriama, Luhya, Maasai, Meru, Kalenjin or Turkana. We are all children of one mother. Our mother is Kenya, the mother of all Kenya people. (Ngugi wa Thiong'o, 1983: 234-5)

Democracy for Africa? Thinking about a possible African variant of democracy is sometimes reminiscent of the 'discovery' of the okapi by Sir Harry Johnston in central Africa a century ago. As Richard Sklar re-tells the story, the okapi was an animal resembling a giraffe, a deer and a zebra. It was unknown to the European zoologists of the time, but they were nevertheless reluctant to acknowledge its existence as a distinct genus (Sklar, 1996: 708-9). Likewise, the genealogies of liberal democracy in the West have been multiple, as have been the prescriptions for its attainment. Its historiography is of relevance to African discourses to underscore its peculiarity in the Atlantic tradition, as well as to assert the possibility of uniqueness in Africa (Ake, 1993). Alexis de Tocqueville early on observed its oddity even within the Atlantic world. Democracy, he observed, had been a truly revolutionary impulse in France, overthrowing the existing feudal-aristocratic order and setting up a whole new stratum of bourgeois class rule, whereas the contrary had happened in America, where democracy had been appropriated by the Republican Party for the pursuit of essentially conservative goals. In the middle of the nineteenth century Lord Acton and John Stuart Mill 167

168

E. S. Atieno Odhiambo

advanced rival theories about the relationship of democracy and culture. 'Acton thought that a civilized state should contain as many different cultural groups as possible to prevent egalitarian democracy from creating an intolerant nationalism; Mill, that where two national groups of roughly the same size exist within a single state, free institutions are unlikely to flourish because the ballot box will be used to capture the state for the majority group - therefore, in such circumstances, partition along national lines will best serve the democratic cause' (Mayall, 1997: 571). The Kenyan scholar Michael Chege has often invoked a third mantra on the political culture of democracy originally asserted by Barrington Moore III, to wit, 'No bourgeois, no democracy' (1994: 74). I understand this to mean that historically it has been the bourgeois stratum of society that has elaborated the idea of a political nation of citizens. The mid-twentieth-century claims for African socialist variants of democracy by rulers ranging from Leopold Senghor to Julius Nyerere have been deleted from the discourses of the 1990s. Yet it would deny them justice to regard their efforts at self-prescription during the early 1960s as merely self-serving and therefore insincere. Also, the failure to recognize the potential originality of Ujamaa as a social theory in the current discussions excludes an important segment of intellectual engagement by Africans in imagining a post-colonial society. Nor should the persistent reiteration of democracy as an alternative metaphor of power for veteran nationalists like Oginga Odinga, throughout the open season of one-party dictatorships in Africa in the 1970s and '80s, be treated, as it so often was by the ruling classes, as a mere idiosyncrasy of the individual losers. The West African, or specifically, the Akan variant of democracy grassroots consensus about chiefly rule in the villages - recently advocated by some African sociologists deserves interrogation as well (Owusu, 1996, 1997). Equally remarkable is the total absence of the radical Left and the 'super-Left' of the 1970s from the prevailing discussions, as well as their revisionism regarding theory and practices (Saul, 1997). This revisionism is best typified by President Yoweri Museveni's rediscovery of Adam Smith and by Professor Peter Anyang'-Nyong'o's move from popular struggles for democracy to liberal democracy and a preference for 'a party of issues', shorthand for reasoned elite politics of the professoriate disciplined by the political sciences {sciences politiques) and comprehensible to the international financial institutions and liberal donor agencies like the Friederich Naumann Foundation. This posture conforms to G. K. Chesterton's description of laissez-faire as 'the philosophy in office' (Joyce, 1998: 208).

The 1990 Conjuncture in Kenya The Kenyan political story over the past century has been one of a rapid march from the creation of the conquest state, to its high noon of settler ascendancy during the interwar years, to the deep colonial crisis precipitating the Mau Mau war between 1952 and 1956. This was followed by the brief period of mass nationalisms between 1957 and independence in 1963, which was succeeded by a contested statehood whose future continues to be uncertain. Throughout the century four strands or historical themes have crisscrossed this political terrain. The first is the high politics of the state, which turned on the issue of state power and who should control that power. Its subtext was racialism until 1963, and tribalism subsequently. In the current political valency it has been rebaptized 'ethnicity'. The second is the 'tyranny of property', pitting

Ethnicity & Democracy in Kenya

169

the haves against the have-nots and informing the nature of class formation. Third is the deep politics of the clan and tribe, pitting insiders against outsiders, clansmen against foreigners, and original landowners against sojourners. This has been translated once again into the idiom and practice of ethnic cleansing. Finally there is the theater of world citizenship, which links the individual and the state to an international discourse on democracy, and a desired world moral order, established by modern protocols on human rights and international laws against all forms of discrimination. The current struggles for a new and democratic political order in Kenya are part of 'the third wave of democratization' that has swept through the African continent and beyond (Huntington, 1991). The following forces have been marshalled against the authoritarian single-party regime of Kenya President Daniel Toroitich arap Moi: the original radicalist tradition of dissent sustained by Oginga Odinga for three decades (Atieno Odhiambo, 1998); the active pulpit opposition maintained by the Reverend Henry Okullu since 1974 and joined by mainline Protestant and Catholic churches in the mid-1980s (Cohen and Atieno Odhiambo, 1992: 30-32; Throup and Hornsby 1998: 54-91; Lafargue, 1996; Sabar-Friedman, 1997); and a tradition of the protest phenomenon sustained by fractions of intellectuals and students at university campuses since the 1960s (Savage and Taylor, 1991). Since the mid-1980s these constituencies have been boosted by a group of reformist-constitutionalist lawyers. More poignantly the debate has been sustained by that crucial ingredient, the urban crowd, a nonelite available to recently disgruntled economic-cum-ethnic power barons like Kenneth Matiba, to the thrice-detained leftist Raila Odinga in Nairobi, and to the volatile Islamic preacher Sheikh Balala in Mombasa and his Islamic Party of Kenya (IPK). The year 1990 was critical in the construction of alliances that brought the debate about democracy to the streets of Nairobi for the first time since the opposition Kenya People's Union (KPU) was banned in 1969. The international context was auspicious as well, in that the sole imperialist superpower, the United States, as well as the international financial institutions, in particular the International Monetary Fund, were calling for accountability and good governance in a triumphant imperialist setting. The period marked the third partition of Africa by US 'special interests' and an intensified 'conditionally' as a clientelist requirement of the Bretton Woods system. 'Rogue ambassadors' like Smith Hempstone felt free to patronize and fraternize with the opposition forces (Hempstone, 1997). Rhetorically it was a season for the promotion of democracy and of a possible New World Order (Khadiagala, 1992). However, the decision by the trading partners and donor agencies - Canada, the United States, Britain, Germany, Denmark, Finland, France, Italy, Japan, the Netherlands, Sweden, Switzerland, the African Development Bank, the IMF, and the United Nations Development Program - to back the forces for change in Kenya in November 1991 by denying aid to the Moi regime was not based on altruistic motives. Rather they wanted to guarantee that the incumbent or incoming regime would meet its commitments toward debt repayments. These organizations were acting in their own self-interest (Makinda, 1996, 565-6; Grosh and Orvis, 1997; Gordon, 1997: 158). The local slogan was a call for multipartyism. Undergirding the summons was the desire for liberal democracy as characterized by free competition among political parties, periodic elections, and respect for the fundamental freedoms of thought, expression, and assembly (Atieno Odhiambo, 1987). These expecta-

170

E. S. Atieno Odhiambo

tions were to be forced onto an authoritarian regime that had been a de facto one-party state from 1969 and a de jure one-party state since 1982.

Repression and Dissent in Moi's Kenya, 1978-90 Immediately after coming to power in 1978 the Moi regime drastically curtailed freedom of expression. In October 1979 students at the University of Nairobi demonstrated in the streets against the decision by the government to bar Oginga Odinga and three other past opposition leaders from standing in the general elections. The university was closed for a month, the students' organization SONU was proscribed, and the student leaders were expelled. In May 1980 riots broke out at the same university to protest against the barring of outside speakers from delivering lectures on campus. The cause celebre was an invitation to speak on campus issued to Edgar Tekere, the fiery former chief of the ZIPRA guerrilla forces in Zimbabwe. The invitation was issued by a Faculty of Arts seminar committee comprising Michael Chege, Atieno Odhiambo, and Peter Anyang'-Nyong'o. As a result of the riots, the passports of twelve faculty members regarded as critical of the government were seized. Among those affected were Peter Anyang'-Nyong'o, Michael Chege, Atieno Odhiambo, Micere Mugo, Ooko Ombaka, Mukaru Ng'ang'a, Shadrack Gutto, and Okoth Ogendo. University Deputy Vice-Chancellor Philip Mbithi stated that foreign money had been poured into the campus to create chaos. In late July 1980 Moi accused the academic staff* union of hiding machetes (pangas) in the students' residences at the university. A denial by union official Peter Anyang'-Nyong'o brought his swift arrest and disappearance for three days. (Coincidentally, his brother mysteriously drowned at the Likoni ferry in Mombasa even as Anyang'-Nyong'o was under incarceration, and his remains were never found.) The academic staff union was banned as was, inexplicably, the Kenya Civil Servants Association. The faculty remained under pressure for two years, with party faithful such as Kariuki Chotara calling to say that the government should move in and arrest the 'troublesome lecturer' known as 'Karo Maksi', by whom he meant Karl Marx. In May 1982 Oginga Odinga, on a visit to Britain, announced the possibility of a second political party. Politician George Anyona, Atieno Odhiambo, Raila Odinga (son of Oginga Odinga), and journalist Patrick Onyango Sumba had been drawing up a constitution for a second party and openly seeking support from a wide spectrum of the Kenyan public, including academics and students. The second party had also been seeking support from the seven sitting recalcitrant legislators dubbed 'the Bearded Sisters' by strongman minister Charles Njonjo, including James Orengo, Abuya Abuya, Lawrence Sifuna, Koigi wa Wamwere, and Mwashengu wa Mwachofi; from the trades unions; from marginalized politicians such as Bildad Kaggia and Msanifu Kombo; from Mau Mau veterans such as Kihara Young and Muigai Lumumba; from landless ahoi such as Benson Manyara; and from liberal journalists such as Salim Lone and Rosemary Mbugua. The writer Ngugi wa Thiong'o had three times declined to be named party secretary-general, as he was then suffering from a 'spiritual freeze' following the banning of his narodnik village theater experiment at his rural Kamirithu home. On 24 May 1982 George Anyona was detained before he could announce the intention to register the party. It was believed at the time that his arrest was set up by a journalist he trusted or whose phone was tapped. The following week

Ethnicity & Democracy in Kenya

171

parliament enacted the amendment better known as Section 2 [a] of the constitution, which made Kenya a dejure one-party dictatorship. In June 1982, the state's paranoia had reached a dangerous level. Addressing a rally in Nakuru: President Moi revealed a plot by university lecturers to arm school and university students to cause chaos in the country. Their action was intended to coincide with a call by some politicians for the establishment of a second political party to rival KANU [the Kenya African National Union]. He singled out six university lecturers who he said were teaching nothing but 'politics of subversion majoring in violence'. The president vowed to root out plotters and agitators from within the Kenya nation. (Ogot, 1995: 199) There were no further moves by faculty or anyone else to organize, as the whole country was in fright and panic. Between 4 and 17 June six lecturers were arrested and subsequently detained without trial: Al-Amin Mazrui, Edward Oyugi, George Mkangi, Kamonji Wachira, Willy Mutunga, and Mukaru Ng'ang'a. Several lecturers, such as Atieno Odhiambo and Ooko Ombaka, went into hiding, while others such as Micere Mugo and Shadrack Gutto fled into exile. At the same time history lecturer Maina wa Kinyatti was jailed for six and a-half years on trumped-up charges of sedition. Previous rumors of plots against the state led by Minister Charles Njonjo's cousin Andrew Muthemba, and an actual attempt at its overthrow by at least two factions of its own air force on 1 August 1982, led to more paranoia and random arrests, including those of university lecturers Atieno Odhiambo and Otieno Malo for brief periods. There were charges of treason against Raila Odinga and journalist Otieno Makonyango, and misprision of treason against the university's Dean of Engineering Alfred Otieno. Odinga, Makonyango, and Otieno were subsequently detained. The student leader Titus Adungosi was sentenced on 24 September 1982 to a ten-year term for celebrating the coup attempt; he died in jail on 27 December 1988 under suspicious circumstances. Six others were jailed for five to six years. The most memorable of the student trials was that of a University of Nairobi freshman Onyango Oloo, who eventually defended himself in court, having rejected plea bargains for a short sentence from his attorney Richard Kwach on the principled basis that he was not guilty of any crime. Sixty-seven other students were held in remand prison from August 1982 until March 1983. In May 1983 Moi finally fell out with his friend Charles Njonjo and subjected him to a 102-day scrutiny at a judicial commission of enquiry. In what became the celebrated msaliti (traitor) affair Moi revealed at a public rally in Kisii that a member of his own cabinet was plotting treason in concert with foreign forces. Cabinet Minister Elijah Mwangale pointed an accusing finger at Njonjo in parliament. Chief Justice Cecil-Miller subsequently headed what amounted to an expose of the regime's arrogance of power. At the end of it, on 12 December 1984 Moi pardoned Njonjo because of his age and length of service. In 1985 Moi set up a special unit of the intelligence detail, answerable directly to the president, to fight political dissent. It arrested hundreds of people. In 1986 the state detained more lecturers and students, businesspeople, and ordinary peasants for allegedly belonging to an underground revolutionary movement, mwakenya. They were subjected to untold cruelties: physical assault, water torture, and mental humiliation in the basement of Nyayo House, an adminis-

172

E. S. Atieno Odhiambo

trative and security building at the heart of downtown Nairobi; summary late afternoon trials and judgements in the courts; and harsh prison sentences averaging five years for people like Dr Adhu Awiti. Some, like Mbaraka Karanja, died at Nyayo House. The unbroken ones were shunted to detention, including Israel Otieno Agina, Mukaru Ng'ang'a, and Dr Katama Mkangi. The state was likewise at odds with Presbyterian and Anglican clergy for preaching against corruption. The plight of the churches was made worse by their opposition to the open queuing system for electoral voting that the regime passed in 1988. Also in 1988 Raila Odinga, released in March from five and ahalf years of detention, was again detained for allegedly plotting a coup on the verandah of the Norfolk Hotel with Atieno Odhiambo. Several fearful people fled into exile, including this author. Following the fall of President Nicolae Ceausescu in Romania in 1989, the Reverend Timothy Njoya called on the nation to examine its conscience, which infuriated the leadership. There followed efforts by Kenneth Matiba, Charles Rubia, and Raila Odinga to form a political party. They were swiftly detained in July 1990. Then followed the Sabasaba (seventh of July) rioting and the police shooting of protestors. The avalanche of demands, demonstrations, and organizations for reform over the next year, as well as the thwarted attempts by Oginga Odinga to register the National Democratic party (NDP), eventually coalesced into the populist Forum for the Restoration of Democracy (FORD) led by Oginga Odinga. His back to the wall, Moi conceded to pressure and allowed an amendment to the constitution that opened the way to multipartyism at the end of 1991 (Throup and Hornsby, 1998: 54-88). It was a move he did not believe in. Moi warned ominously that it would lead to ethnic clashes; this prophecy soon came to pass. To ethnicity therefore this chapter must next turn.

Constructing Ethnicity 'In the years to come, ethnicity, in whatever concrete forms and under whatever name, will be so important a political resource and an idiom for creating community that today's social scientists and anthropologists have no choice but to confront it' (Lentz, 1995: 303). George Carew adds, c The tension between ethnicity and the State in the newly independent states of Africa has all but derailed the transition to democratic governance' (1996: 3). For the Kenyan population, Lentz's prediction has already come true. Since its independence the country has nurtured the political culture of what used to be called 'tribalism' and is now called 'ethnic clashes'. These ethnic clashes have occurred for much of the 1990s: between 1991 and 1994 and again in 1997-8. It is important therefore to begin this discussion with a disclaimer. Two simultaneous discourses have been in vogue in Kenya: tribalism as a lived reality or a practice of daily life, and ethnicity as a generic topic for academia. The two terms may refer to the same epiphenomenon, but they do not necessarily bear the same meaning. While for academia ethnicity may be a distant 'dazzling, and ambiguous category, at once descriptive and value-normative' (Lentz, 1995: 304), Kenya Africans do not talk of ethnicity in their offices, on public platforms, or in whispers along the streets. They talk and think about tribalism as the regular experience of their everyday lives, in its many enabling capacities, its incapacitating impediments on the hopes of individuals, and its blocking of opportunities for whole communities. They use tribalism as a practical vocabulary of politics and social movements.

Ethnicity & Democracy in Kenya

173

Thus tribalism is a term of the everyday acknowledgement of experiences, contextual practices, and political fact, shorthand for the welter of the simultaneously modernizing and recrudescently violent political cultures that have come to typify the post-colony of Kenya. In the context of a sharply conflicted and volatile political arena it has often and recurrently become a dependable locus for the articulation of legitimate interest. In this sense it at least entertains potentially democratic grassroots possibilities or may facilitate egalitarian demands for ethnic representation at the lattices of state power. The narrative of this chapter at one level is about how certain tribalisms came to exist in Kenya. Here it will be argued, with particular reference to the Luo and the Gikuyu, that the deep roots of their colonial ethnicity lie in the idea of citizenship, 'an analytical lens that provides a precise way of theorizing the relationship between ethnic identity, authority and legitimacy' (Ndegwa, 1997, 599). Clear distinctions have been made between ethnic citizenship as a matter of inclusive birthright that confers recognition, identity, and patrimony, and state citizenship, which is a matter of kipande, a bureaucratic obligation. Thus the individual can be a Luo citizen, a reluctant subject of the Kenya state, and a citizen of the free world (Cohen and Atieno Odhiambo, 1989: 26-8). The second level concerns how tribalism has been seen or is regarded as an attribute of state power. This section deals with how ethnicity has been appropriated for state and regime building in the interests of the incumbent regime, Gikuyu under Jomo Kenyatta, as part of the 'instrumentalities of survival' (Young, 1982) in the uncertain social formation that is the post-colony of Kenya.

The Colonial Roots of Luo and Gikuyu Ethnicity The popular social models of the industrious Kikuyu and the lazy Luo were the results of colonial productions of knowledge; they have continued to inform the constant creation and re-creation of ethnic categories in the post-colony. Let us move beyond structure into agency by considering the making of ethnicity by the Luo and Kikuyu leaders Oginga Odinga and Jomo Kenyatta. The Luo preserved their relative autonomy from the state during the colonial period. The Luo sat on land that the British had little need of; it was not suitable for European settlement. Conversely, the British had no developmental agenda or project for the Luo. Previous efforts to make them into cash-crop cotton growers had foundered again and again. A backward subsistence peasant economy persisted, with maize, millet, sorghum, and ghee surpluses being marketed, following good harvest seasons, along with hides and skins. Within the wider political economy the Luo were a manual pool, categorized and labeled 'Kavirondo labor' at the workplaces in the settler plantations and the Mombasa dockyards. In the Central and South Kavirondo reserves the Luo constituted a 'tribe' with an autonomous way of life. 'This autonomy was multifaceted: the tribal economy was a source of livelihood, tribal ideology a source of identity and common purpose' (Mamdani, 1991: 91). This was a poor but free peasantry. 'The [District Commissioner] does not close my gate for me', every male Luo homeowner {won dala) could proudly assert. Conflicts with colonial authority were usually limited to the local level. Personal protest was largely about the seasonal arrests for failure to pay the poll tax. Sublocational and locational politics were characterized by competition among clans for nomination to, and perpetual appropriation of, colonial chieftainships within specific clans. The deep politics of the clan meant that most members wanted their own sons appointed

174

E. S. Atieno Odhiambo

chief, headman (miruka), or sub-chief (mlango). In a nutshell the struggles were about the accumulation of local power (Lonsdale and Whisson, 1976; Cohen and Atieno Odhiambo, 1989: 25-30). During the interwar years, the Central and South Nyanza Local Native Councils provided arenas for fierce competition between the appointed chiefs and the rival elected elites. The concerns of this popularly elected local leadership were largely economic; their grievances were about the uneven playing field brought about by the licences and trading opportunities granted to the local Indian merchants. This set of grievances crystallized into Oginga Odinga's Luo Thrift and Trading Corporation (LUTATCO). Formed in 1945, LUTATCO was a movement of the modernizing local notables who wanted economic opportunities in the world of commerce dominated by Indian merchant capital. Its ethnic dimension turned on the call for unity of the Luo. 'Persistence Is Strength, Unity Is Strength' (Kinda e Teko, Riwruok e Teko) was its motto. It was not until the bench terracing controversies of the late 1940s that a common cause for continuous opposition to colonialism was galvanized by Oginga Odinga and turned into the local expression of the colony-wide anticolonial movement. This unity was enormously fortified through the election of Oginga Odinga as ker, paramount leader of the Luo Union (East Africa), the umbrella cultural organization of the Luo nation that had its origins in the urban world of the 1930s. It was this combination of economic and cultural leadership that entitled Oginga Odinga to the honorific cognomen Jaramogi, meaning the reincarnation of the eponymous Luo founder, Ramogi. It was with this mantle of Luo leadership that Odinga entered into nationalist politics in 1957 (Atieno Odhiambo, 1975). Once elected, Oginga Odinga carved unrivalled credentials for himself as a Kenyan nationalist by declaring in the Kenya Legislative Council that Jomo Kenyatta, the ostracized and incarcerated leader of the Mau Mau movement, was not a convicted criminal but the true leader of the Kenya Africans. In one stroke Odinga released Kenyatta politically and reinstated him at the apex of Kenyan African political leadership. 'Without Odinga our Jomo would have died in prison' (Dine onge Odinga, nyithiwa di ne Jomo otho ejela). Mama Aboge and her Luo market women sang at Chiro Mbero in Kisumu town. This assertion also provided the battle cry for nationalism: Freedom and Kenyatta (Uhuru na Kenyatta). With Odinga as the rural leader and the trade unionist Tom Mboya as the workers' leader the Luo occupied center stage in the politics of nationalism. They thought they had a stake in the outcome of independence, but they were proved wrong by the Kenyatta regime (Atieno Odhiambo, 1998). There was a stark contrast between the Luo trajectory of social change and aflFairs in Kikuyuland, the home base of Jomo Kenyatta. By the time the British arrived as conquerors in the late nineteenth century, this was a land- and wealthconscious community that linked wealth to virtue, and virtue to a sense of history that regarded ownership of land and goats as a trust (Lonsdale, 1992; Kershaw, 1997). Various frontier landholding units (mbari) in southern Kikuyuland lost 30-70 per cent of their land to British settlers in 1902-3, which led to direct confrontation with the alliance of the European settler and the colonial state. The story of Gikuyu land hunger had begun. The dispossessed moved to the Rift Valley to become squatters. Their sense of security there soon came to an end with the promulgation of the 1937 Resident Laborers Ordinance. The battleground for the contestation of land thus became extended from mbari ya Igi at Komothai in southern Kiambu to the Powys Cobb farm at Mau Narok,

Ethnicity & Democracy in Kenya

175

better known in subsequent colonial historiography as Olenguruone. Squatters stood against settlers regarding the rights to Gikuyu labor, time, and accumulation of wives, goats, and sheep - in other words, regarding the futures of the white man, muthungu, and the black man, mundu muiru (Kanogo, 1987; Throup, 1987; Furedi, 1989; Kershaw, 1997). The Gikuyu who remained in the southern Kiambu reserve likewise had to rework new property relations among themselves. The few rich landowners who were also the chiefs, court clerks, interpreters, and teachers tightened their hold on the land, eventually shedding their tha, their traditional gift-giving obligations, such as giving their poorer kin access to land. The middle-class and poor peasants kept hoping for redress from the British, a return of their 'stolen lands'. Their expectations were particularly raised by the Carter Land Commission, sent out in 1933 to look into the land question colony-wide. The commission's report, however, failed to address Gikuyu grievances. Fear and anger filled the land, with the Gikuyu nation on one side of the divide and the settlers on the other. The two nations were not communicating. No colonial explanations would satisfy an increasingly politically articulate and cohesive marginalized society. Conflict would come (Throup, 1987; Kershaw, 1997). Senior Chief Koinange wa Mbiu contrasted the situation, speaking the idiom of accumulation to the volatile Nairobi crowd of the Luo, Akamba, and Gikuyu that oscillated between Kiburi House, headquarters of the Kenya African Union (KAU), and his Kiambaa seat and oathing outpost. 'The muthungu [white man] was the well fed Friesian herd, leisurely ambling up the pasture ridge from the waterhole; the African was the thirsty shenzi cow stampeding down the cliff in search of water' (interview with Okola Owor, village of Liganua, 9 August 1997). The Koinange oath of unity was administered to all who could get to Kiambaa from late 1946 onward, and to trusted Africans in the quarries of Kariobangi outside Nairobi (ibid.). No Gikuyu leader could sidestep or ignore the issue of 'stolen lands.' This was especially true of Jomo Kenyatta, who returning to Kenya in 1946 after fifteen years of sojourn in Europe, found a restless community of young and poor awaiting him. He was the man who would bring deliverance, the very embodiment of Gikuyu power. In response to a hero's welcome at the home of Senior Chief Koinange in Kiambu, Kenyatta made an impassioned plea for unity. His words were published in the 21 October 1946 edition of the Nairobi newspaper Mwalimu. 'I have nothing, not even a cent to give you', he said. 'But with UNITY even the atomic bomb cannot defeat us.' In the subsequent six years Kenyatta consolidated his alliances with the Gikuyu notables (andu arume), marrying the daughters of Chiefs Koinange and Muhoho, but primarily advocating the right of the Agikuyu to freedom and independence from British oppression. The tree of freedom would be watered by blood, he said. He was already holding the head of the lion; were the masses going to help him tame it? he asked. 'Kenyatta is already holding the lion's tail, straining it, this oddity of a man' (Kenyatta omako yiwe yuayo, mbaklo), sang his Luo political sisters-in-law such as Omuya chi Kenya ('the wife of the incipient Kenya nation'). In preparation for conflict varieties of local oaths were designed and administered by Agikuyu leaders to the Gikuyu masses. The common denominator in all the oaths was the call for unity and commitment and, conversely, calls for death for those who betrayed the oaths. The Governor of Kenya issued a Declaration of Emergency on 20 October 1952 that led to the arrest of Jomo Kenyatta and the entire Gikuyu political machine hith-

176

E. S. Atieno Odhiambo

erto associated with the Kenya African Union (KAU). The longer lasting result of Kenyatta's arrest and conviction for 'managing Mau Mau' was to reinforce the myth of Jomo Kenyatta as the leader of the entire Gikuyu nation and the father of Kenya's nationalism (Atieno Odhiambo 1995). Kenyatta was sorely tried {Kenyatta aliteswa sana), all African neophytes knew in the late 1950s. He returned from prison early in August 1961 and assumed the leadership of the territorial nationalist movement, but more significantly of a Gikuyu society reeling from the effects of colonial emergency and intra-Gikuyu civil war. His was a desperate community hopeful that he would get power and freedom for Kenya under Kiambu leadership, and give land and hope to the poor. He did (Hyden, 1994: 80-81).

Hegemonic Enterprises: Kenyatta's Use of Ethnicity The independence project was grim and urgent in the early 1960s. Jomo Kenyatta had been convicted by settler (in)justice for 'managing Mau Mau' in 1953. In 1962 the same settlers, recognizing the inevitability of the transfer of state power, recaptured Jomo Kenyatta and made him guarantee them a safe future. At an anxious moment, he reassured the assembled settlers in Nakuru in August 1963 that he would respect and honor their property rights in land: If I have done wrong to you it is for you to forgive me. If you have done wrong to me it is for me to forgive you. We want you to stay and farm this country (Ngugi, 1981: 88). This was a major capitulation in the eyes of political nationalists like Paul Ngei and Bildad Kaggia, both of them convicted and jailed with Kenyatta as Mau Mau leaders. It was, of course, treason in the eyes of the Mau Mau. They began taking their oaths again; they re-constituted their underground movement, Kiama kia Muingi; and they revisited the forests under Field Marshals Baimuinge, Mwariama, and Wanjiru, vowing to fight to the end and never surrender to Kenyatta's compromise over land (Omosule, 1976). For his part, Kenyatta began to gain the 'respect' of settlers and international capital once he accepted that settler farms would have to be purchased by the Africans on a willing buyer-willing seller basis. The trouble, though, lay in the fact that the neediest Africans, the former Mau Mau, had no capital with which to buy the land. Empire and international finance came quickly to the rescue: the British Government and the World Bank would loan the Kenya(tta) government funds with which to buy out the settlers. Kenyatta would appease the land hunger of the former Mau Mau by settling them successively into the Million-Acre, Haraka, and Harambee settlement schemes on soft loan terms. The ex-Mau Mau would have to pay gradually. Within the space of a few years, from 1960 to 1966, the former White Highlands were re-peasantized. The Agikuyu came back to the Rift Valley, and the Gikuyu reserves of Kiambu, Murang'a, and Nyeri were now extended into the Nakuru, Laikipia, and Nyandarua districts, which became fully Gikuyu reserves, and also into the eastern Nandi, eastern Kericho, and southern Uasin Gishu districts (Leo, 1984). In those six years Kenyatta proved himself by saying and doing (kuga na gwika) and rightly earned his honored title of muthee, or 'old man'.

Ethnicity & Democracy in Kenya

177

'Othering 5 the Luo 'Culture, race and gender become primary sites for plotting the processes of othering, the extent to which they converge and the way they are traversed. Culture is the slate upon which exchange is inscribed' (Darby, 1998: 243). The era of independence facilitated the consolidation of Gikuyu dreams as power was transferred from the British initially to an alliance of the Kenya African National Union (KANU) tribes that had won the 1963 elections and formed the government with Kenyatta as Prime Minister and later President. But power soon devolved into the hands of a decidedly Kiambu cabal with Kenyatta, Mbiyu Koinange, Njoroge Mungai, Charles Njonjo, and James Gichuru constituting the inner core. Marginalizing the two erstwhile Luo allies, Oginga Odinga and Tom Mboya, was an early political project of this group. Its long-term effect was to alienate the Luo from the inner sanctum of power for the following three decades. Our interest lies in the link between structure and agency in this most obdurate of Kenya's political cleavages, and in cultural agency as a possible explanation, with a focus on culture as 'the slate on which exchange is inscribed' (ibid.). The parting of the ways between Odinga and Kenyatta was ideological, but it was also intensely local and reflected their different understandings and dreams. Their positioning represented conflicting understandings of the African past, because both of them had been immersed in the inventions of that past. Both men brought with them an ethno-cultural understanding of politics. Each understood only too well the demands of democratic citizenship, 'a common membership of a shared and imminent community' (Stewart, 1995: 75, original emphasis). Both men were deeply cultural and espoused values that were locally rooted, Kenyatta in Gikuyu individual enterprise and personal virtue, Odinga in clan-based communocratic and achieving values plus a tradition of resistance to authoritarianism of any sort. Both of them understood the link between individual and community, the potency of the emphasis on hard work and unity, and the force and power behind the developmental roots of ethnicity. But in contradictory ways. Kenyatta went for the already successful entrepreneur: he mobilized the Kikuyu elite around him and linked up with the local and international allies in capitalism. The old Anglo-Kikuyu alliance made sense, and it stood for continuity and stability (Rogers, 1979; Clough, 1990). The new Kikuyu-Kalenjin alliance forged with Daniel Toroitich arap Moi from late 1964 became a categorical imperative for Kenyatta in the 1960s. Kenyatta had to solve the perennial agrarian problem for the Agikuyu. The frontier of opportunity lay, as it had since the early 1900s, in the Rift Valley. So he made a pragmatic alliance with Moi, a member of the conservative wing of the Kalenjin, to enable the Gikuyu to settle in the former White Highlands as freeholders (Throup and Hornsby, 1998). Jaramogi Oginga Odinga, on the other hand, went in for populist politics that brought to the fore the national underdogs: the ex-Mau Mau fighters who should be rewarded, the landless who should get free land, the shopkeeper who should get a loan for business expansion, and the small man in general who should be uplifted. Kenyatta chose the muthuuri - the big man; Odinga's advocacy was for the common man - the mwananchi. The organizing principle was actually the tyranny of property, a Kenyan concern from the beginning. Property had to be protected using power; in turn, power itself had to be consolidated. In Kenyatta's view it was best protected within the Gikuyu nation, in

178

E. S. Atieno Odhiambo

the House of Gikuyu and Mumbi. So ultimately the social struggle for the political future was turned around and re-baptized Kikuyu-Luo rivalry. Ethnicity won over ideology. The acrimonious campaigning against the opposition party Kenya People's Union (KPU) led by Oginga Odinga from 1966 to 1969 turned ethnicity against the socialist ideology that the KPU espoused. As the date for local and general elections loomed in 1968 (they were subsequently postponed for a year), truckloads of Agikuyu went voluntarily, or were coerced into going, to Gatundu, Kenyatta's country seat, to take oaths to guarantee that the Kenya flag would never leave the House of Mumbi. Often the participants took the oath on a flag of Kenya spread on the ground (Ochieng', 1995: 102). Subsequent opposition political debates have articulated the issues as a dialectic between the following: ethnicity versus representation, ethnicity versus equitable resource allocation, ethnicity versus popular participation and legitimacy, and ethnicity versus class conflict. Ultimately they have also been a test of the forms of loyalty to be entertained or to be confronted by the varying ethnic or class claims (Ndegwa, 1997: 600). By the end of 1969, however, Kenyatta's type of ethnicity had won. The Luo found themselves excluded from the 'common good' - the right to power and access to developmental resources. The assassination of Tom Mboya in July 1969, the banning of the KPU, and the detention of Oginga Odinga marked the ascendancy of Kenyatta's hegemonic enterprise. Freed from both strong contenders over policy and power, Kenyatta was finally free to pursue his own agenda. Is the practice of 'legitimate' ethnicity expected of the African political elites as sons of patrimonial villages? 'Not ideas', wrote Max Weber, 'but interests material and ideal - directly govern men's conduct. Yet very frequently the "world images" that have been created by "ideas" have, like switchmen, determined the tracks along which action has been pushed by the dynamic of interest' (Weber cited in Brubaker, 1992: 17). Peter Ekeh has often argued for Africa's two publics: the one at the secretariat or state capitol, the other at the feet of the village elders; the former a judgmental forum on one's citizenship, the latter a resource for fulfilling the demands of filial duty (Ekeh, 1975: 1990). Kenyatta was obligated to the Agikuyu in specific ways, but to Kenya in general ways. He chose to be self-regarding and inclusive towards all Gikuyu, the limit of his moral ethnicity (Lonsdale, 1992), and to be other-regarding towards the rest of Kenyan society, his subjects. 'Neo-patrimonialism after independence represented in many ways a reconstruction of the principles of "moral ethnicity"' (Throup and Hornsby, 1998: 41). As argued elsewhere, Kenyatta had a fervent vision of the Gikuyu future but no mental map of Kenya beyond a territory to be governed much as the colonial authorities had done (Atieno Odhiambo, 1995). His regime made clear distinctions between the homeboy, muru wa mucii, and the Luo westerners,, waruguru, the ultimate 'other' in the regime's political lexicon. Specifically, he chose to exclude the Luo as a cultural 'other' beyond the bounds of Gikuyu civil society, which in any case was coterminous with the Kenyatta state. This post-colonial discourse took the form of village-degrading utterances which 'animalize the Other in their midst' (Werbner and Ranger, 1996: 20), as the Luo were regularly referred to as kinyamu giiki or kihii giiki - little animals or little boys. The state became the primary site for plotting the processes of this othering. In terms of political sociology Kenyatta's ethno-cultural exclusion of the Luo fits Max Weber's notion of social closure. The notion has its classical exposition in the

Ethnicity & Democracy in Kenya

179

opening pages of Economy and Society>> where Weber distinguishes between open and closed social relationships, and regards citizenship as a salient social and cultural fact that occupies a central place in the political culture of the modern nation-state (Brubaker, 1992: 23). Under Kenyatta the Luo had legal citizenship of the Kenya state but were subjected to many tacit, uncodified, internalized classificatory schemes and ethno-cultural markers in the political and economic arenas. 'Andu waruguru' - foreigners from the West - they were called. 'These people are finishing us' (Jogi tiekowa), they cried in the early 1970s as they ran into one blocked opportunity after another. But the regime was not listening. Current discourses have privileged Western conceptualizations of the Napoleonic construct with its state-civil society distinctions, often confining civil society to 'that aspect of social interaction that lies between the state and the individual and that is not focused on acquiring power at the level of the state' (Chege and Hyden, 1998: 598). This conceptualization tends to ignore local notions of citizenship in Africa that are organized around rituals, especially those that elaborate identity, legitimacy, and authority, a 'vital element in the processes that make and remake social facts and collective identities' (Comaroff and Comaroff, 1993: xvi). Gikuyu rites of passage from childhood to manhood and womanhood are signified by male circumcision and cliteridectomy, the sine qua non for entry into civil society for men and the route by which women could 'become beautiful in the ways of the tribe' (Ngugi, 1965). This Gikuyu notion of civil society was extended by Kenyatta to the political arena of the state in 1966-9 when he accused the KPU opposition of being chameleons - definitely not part of civil society, and by extension therefore not legitimate citizens of the Kenya state that he ran. The Luo were targeted for this rhetorical exclusion ostensibly because they did not practice male circumcision. This specifically central-Kenyan discourse on being cut - 'the narcissism of small differences' as Freud once spoke of it, the tendency to think of ourselves as superior to others because of some laughably superficial and non-essential feature - resurfaced in 1992 as two Gikuyu barons, Kenneth Matiba and Mwai Kibaki, bid for the presidency against Oginga Odinga. It was widely asserted that Odinga ought not to be elected because he was not circumcised. Odinga understood its potential damage, and raised it as a debating issue at a rally in Kiambu in late 1992. In Meru the Ford-Kenya party secretary and parliamentary candidate Gitobu Imanyara was severely ridiculed for fronting for Odinga, an uninitiated 'boy'. Thus the ball set in motion by Kenyatta found its everyday life extended in the bid for a post-Moi state. The issue of circumcision also confronted the N D P presidential candidate Raila Odinga in 1997, again in central Kenya. Raila Odinga treated it as a case of false consciousness, bantering facetiously that the women were not complaining, and calling for a focus on the real issues of the campaign. But the issue would not go away, for the politics of the state had become routinized around an exclusive ritual. 'Such rituals establish a hierarchy of power within the community that, when combined with other icons of power and status in the modern state, enables elites to mobilize within the ethnic community for the interests of the secular state' (Ndegwa, 1997: 602). The point is that civil society melds into political society and the state in uneven ways, rendering the Napoleonic distinction irrelevant at the foothills of Mount Kenya. Language is therefore a crucial criterion of identity and a readily available symbol of ethnicity with prescriptive power for legitimacy or exclusion in the existing or putative ethnic-state. Its power lies in its facilitation of the articula-

E. S. Atieno Odhiambo

180

tion of popular forms of consciousness. In the heyday of the Kenyatta regime it was assumed that the people within the corridors of his power would speak Gikuyu. Shadrack Ojudo Kwassa, a Luo former Chief of Protocol, recalled the surprise of First Lady Mama Ngina Kenyatta at his inability to speak Gikuyu at an official encounter over tea one afternoon in Gatundu, President Kenyatta's country fiefdom. He was out of the protocol office the following day, his job being assigned to a more appropriate Mogikoyo, Daniel Gachukia (interview with Wod Nam, University of Nairobi Senior Common Room, 12 July 1997). Our interest in this anecdote lies in the instrumentality of language in the affirmation of power and powerlessness at the strategic time and place, and not in its accessibility, for virtually anyone can learn language or languages in context. Indeed the average urban Kenyan is adroit at code-switching and multilingualism. 'The more general analytical point', writes Rogers Brubaker, 'is that cultural idioms are not neutral vehicles for the expression of pre-existing interests: cultural idioms constitute interests as much as they express them. These culturally mediated and thereby culturally constituted interests are not prior to, or independent of, the cultural idioms in which they are expressed. Thus it becomes necessary to study the social production of political languages themselves' (1992: 16-17). The rhetorical question may now be posed in stark terms: when will an uncircumcised Luo presidential candidate be voted for by subscribers to a socially produced political language, organized around the idea of manhood citizenship, constructed within the cultural geography of circumcision, while facing Mount Kenya?

References Ake, Claude. 1993. 'The Unique Case of African Democracy', International Affairs 69: 239-44. Atieno Odhiambo, E. S. 1975. 'Seek ye first the economic kingdom': A history of the Luo Thrift and Trading Corporation 1945-1956', in B. A. Ogot ed. Hadith 5: The economic and social history of East Africa. Nairobi: East African Publishing House. — 1987. 'Democracy and the Ideology of Order in Kenya', in M. G. Schatzberg, ed. The Political Economy of Kenya. New York: Praeger. — 1995. 'The Formative Years, 1945-1955', in B. A. Ogot and W. R. Ochieng, eds, Decolonization & Independence in Kenya. London; James Currey and Athens, OH: Ohio University Press. — 1998. Jaramogi Oginga Odinga. Nairobi: East African Educational Publishers. Brubaker, Rogers. 1992. Citizenship and Nationhood in France and Germany. Cambridge, MA: Harvard University Press. Carew, George. 1996. 'Developmental Democracy and Postcolonial Polities', Quest 10 (2): 3-36. Chege, Michael. 1994. 'The Return of Multiparty Polities', in Joel Barkan, ed. Beyond Capitalism vs. Socialism in Kenya and Tanzania. Boulder, CO: Lynne Rienner. Chege, Michael, and Goran Hyden. 1998. 'Research and Knowledge: Social Sciences', in Encyclopaedia of Africa south of the Sahara. Vol. 3. New York: Charles Scribner & Sons. Clough, M. S. 1990. Fighting Two Sides: Kenyan chiefs and politicians, 1918-1940. Niwott, CO: University of Colorado Press. Cohen, D. W. and E. S. Atieno Odhiambo. 1989. Siaya: Historical anthropology of an African landscape. London: James Currey; Athens, OH: Ohio University Press. — 1992. Burying SM: The politics of knowledge and the sociology of power in Africa. Portsmouth, N H : Heinemann Inc.; London: James Currey. Comaroff, Jean and John Comaroff, eds, 1993. Modernity and its Malcontents. Chicago: University of Chicago Press. Darby, Philip. 1998. 'Taking Fieldhouse Further: Post-colonializing imperial history', The Journal of Imperial & Commonwealth History 26, (2): 232-50. Ekeh, P. P. 1975. 'Colonialism and the Two Publics in Africa', Comparative Studies in Society and History 17, (1): 91-112. — 1990. 'Social Anthropology and the Two Contrasting Uses of Tribalism in Africa', Comparative

Ethnicity & Democracy in Kenya

181

Studies in Society and History 32, (4): 660-700. Furedi, Frank. 1989. The Mau Mau War in Perspective. London: James Currey; Athens, OH: Ohio University Press. Gordon, David F. 1997. 'On Promoting Democracy in Africa: The international dimension', in Marina Ottaway, ed. Democracy in Africa: The hard road ahead. Boulder, CO: Lynne Rienner Publishers. Grosh, Barbara, and Stephen Orvis. 1997. 'Democracy, Confusion, or Chaos? Political conditionality in Kenya', Studies in Comparative Political Development 31, (4): 46-65. Hempstone, Smith. 1997. Rogue Ambassador: An African Memoir. Sewanee, TN: The University of the South Press. Huntington, Samuel P. 1991. The Third Wave of Democratization in the Late Twentieth Century. Oklahoma City: Oklahoma University Press. Hyden, Goran. 1994. 'Party, State and Civil Society: Control versus openness', in Joel Barkan ed., Beyond Capitalism vs. Socialism in Kenya and Tanzania. Boulder, CO: Lynne Rienner Publishers. Joyce, Patrick. 1998. 'The Return of History: Postmodernism and the politics of academic history in Britain', Past and Present 158: 207-35. Kanogo, Tabitha. 1987. Squatters and the Roots of Mau Mau. London: James Currey; Athens, OH: Ohio University Press. Kershaw, Greet. 1997. Mau Mau from Below. Oxford: James Currey; Athens, OH: Ohio University Press. Khadiagala, Gilbert. 1992. 'Thoughts on Africa and the New World Order', The Round Table 324: 431-50. Lafargue, Jerome. 1996. 'Dieux et la ruse ont la parole. La protestation publique au Kenya et en Zambie', Politique Africaine 64: 41-51. Lentz, Carola. 1995. '"Tribalism" and ethnicity in Africa', Cahiers des Sciences Humaines 31 (2): 303-28. Leo, Christopher. 1984. Land and Class in Kenya. Toronto: University of Toronto Press. Lonsdale, John. 1992. 'The Moral Economy of Mau Mau', in B. Berman and John Lonsdale, eds, Unhappy Valley, book 2: Violence & ethnicity. London: James Currey; Athens, OH: Ohio University Press. Lonsdale, John, and Michael Whisson. 1976. 'The Case of Jason Gor and Fourteen Others: A succession dispute in historical perspective', Africa 45, (1): 50-65. Makinda, Samuel. 1996. 'Democracy and Multi-party Politics in Africa', The Journal of Modern African Studies 34, (4): 555-73. Mamdani, Mahmood. 1991. 'Social Movements and Constitutionalism in the African Context', in Issa G. Shivji, ed., State and Constitutionalism: An African debate on democracy. Harare: SAPES Books. Mayall, James. 1997. 'Democratic en Afrique?', The Round Table 344: 571-3. Ndegwa, Stephen N. 1997. 'Citizenship and Ethnicity: An examination of two transition moments in Kenyan polities', American Political Science Review 91, (3): 599-615. Ngugi, James. 1965. The River Between. London: Heinemann. Ngugi wa Thiong'o. 1981. Detained: A writer's prison diary. London: Heinemann. — 1983. Devil on the Cross. London: Heinemann. Ochieng', W. R. 1995. 'Structural and Political Changes', in B. A. Ogot and W. R. Ochieng' eds, Decolonization & Independence in Kenya, 1940-93. London: James Currey; Athens, OH: Ohio University Press. Ogot, B. A. 1995. 'The Politics of Populism', in B. A. Ogot and W. R. Ochieng', eds, Decolonization & Independence in Kenya. London: James Currey; Athens, OH: Ohio University Press. Omosule, Monone. 1976. 'Kiama kia Muingi'. Paper presented at the annual conference of the Historical Association of Kenya, Nairobi. Owusu, Maxwell. 1996. 'Democracy in Africa - a view from the village', in Julius E. Nyang'oro, ed., Discourses on Democracy: Africa in comparative perspective. Dar es Salaam: Dar es Salaam University Press. — 1997. 'Domesticating Democracy: Culture, civil society and constitutionalism in Africa', International Quarterly 39 (1): 120-52. Peel, J. D. Y. 1984. 'Social and Cultural Change', in M. Crowder, ed., The Cambridge History of Africa. Vol. 8. Cambridge: Cambridge University Press. Rogers, Peter. 1979. 'The British and the Kikuyu 1890-1905: A reassessment', Journal of African History 20: 255-69. Sabar-Friedman, Galia. 1997. 'Church and State in Kenya, 1986-1992: The churches' involvement in the "game of change"', African Affairs 95: 25-52. Saul, John S. 1997. '"For fear of being condemned as old fashioned": Liberal democracy vs. popular

182

E. S. Atieno Odhiambo

democracy in sub-Saharan Africa'. Review of African Political Economy 73: 339-53. Savage, D. C , and Cameron Taylor. 1991. 'Academic Freedom in Kenya', Canadian Journal of African Studies 25 (2): 308-20. Sklar, Richard L. 1996. 'Duty, honor, country: Coping with Rhodesia's unilateral declaration of independence', The Journal of Modern African Studies 34, (4): 701-14. Stewart, Angus. 1995. 'Two Conceptions of Citizenship', British Journal of Sociology 46: 63-78. Throup, David. 1987. Economic and Social Origins of Mau Mau 1945-53. London: James Currey; Athens, OH: Ohio University Press. Throup, David and Charles Hornsby. 1998. Multi-party Politics in Kenya. Oxford: James Currey; Athens, OH: Ohio University Press. Werbner, Richard and Terence Ranger, eds, 1996. Postcolonial Identities in Africa. London: Zed Books. Young, Crawford. 1982. 'Patterns of Social Conflict: State, class and ethnicity', Daedalus 3, (2): 71-98.

11

SHULA MARKS

'The dog that did not bark, or why Natal did not take off5: Ethnicity & Democracy in South Africa KwaZulu-Natal

O

NE of the many paradoxical features of the 'new' South Africa has been the virtual disappearance of the political violence associated with the Inkatha Freedom Party and its founder and leader, Chief Gatsha Mangosuthu Buthelezi, since the 1994 elections. Widely described in the media as 'black on black violence' and ascribed by many a pundit to deepseated ethnic or tribal conflict, the violence, at least in its political guise, seems to have vanished almost unnoticed by those who were most inclined to proclaim its innate propensity.1 If, in the ten years prior to the elections, the questions were why the violence and what are its implications for democracy in South Africa, the question now is how and why has that violence come to such a remarkable end? What, if anything, can the strange death of ethnic violence tell us about the way in which 'ethnic diversity can become part of the democratization process and be accommodated within democratic institutions'? Has the 'imprint' left by this process furthered or hindered democratic processes? The answer to these questions encompasses and illuminates in part at least three broad themes addressed in this volume: (i) the ways in which ethnicities are constituted and transformed through confrontation or struggles over the organization of political community, power and resources in colonial and post-colonial state formation; (ii) the imprints of ethnicity on nation-state building and politics, state-society relations, and popular perceptions of the interrelationship of class and communal differences; and (iii) the challenges posed to the search for more legitimate, inclusive and democratic forms of governance.

Historical Context In a world awash with violent 'ethnic' conflict, it is already difficult to recall the intensity of the violence in South Africa in the decade before the 1994 elections, the threat to the peace process it posed and how close to civil war and anarchy South Africa seemed. 'Blood Set to Flow as Zulus Talk War' screamed the Sunday Times headlines in December 1993. 2 'Emergency plans are being made to airlift up to 350,000 Britons out of South Africa should the country slide into chaos after this month's elections' announced the Daily Mail three 1 For an excellent account and critique of the long history of over simple labeling of violence in KwaZulu-Natal as 'tribal' or as 'faction fighting', see Dennis Jabulani Sithole, 'Land, Officials, Chiefs and Commoners in the Izimpi zemibango in the Umlazi Location of the Pinetown District in the context of Natal's changing political economy, 1920-1936', unpublished MA thesis, University of Natal, Pietermaritzburg, 1998, chap. 1. 2 Cited in Anthony Sampson, Mandela. The Authorised Biography. London, 1999, p. 479.

183

184

Shula Marks

weeks before the elections. 3 Between 1985 and the elections in April 1994, over 8,000 people were killed in internecine warfare in KwaZulu and Natal, 4 ostensibly between Zulu members of the Inkatha Freedom Party (IFP) and their political opponents in the African National Congress, most of them also Zulu. Thousands more were injured and tens of thousands displaced from their homes. There were said to be some half a million internal refugees from the violence in Durban alone. Nor was the violence restricted to KwaZulu-Natal. The transformation of Inkatha Yenkululeko Yesizwe (the Zulu Cultural National Movement, henceforth Inkatha), founded by Chief Mangosuthu Buthelezi in 1975 as a 'cultural liberation movement', into the nation-wide Inkatha Freedom Party (IFP) in July 1990 extended the attempt to mobilize Zulu-speakers along ethnic lines beyond the confines of KwaZulu-Natal. With this, the battle was taken from Natal to South Africa's industrial heartland, where more than 500 Africans were killed in a matter of days. In the next four years in the run-up to South Africa's first non-racial democratic elections in 1994, nearly 5,000 more Africans lost their lives. On the eve of those elections and for a short time thereafter, Buthelezi seemed to be holding the country to ransom as violence threatened to turn into what journalists feared would be a Bosnian-type conflagration. Until the IFP decided in the last week to enter the elections, South Africa seemed at the edge of the abyss, and even after the elections there were widespread and well-authenticated reports that the IFP was training special armed units on the farms of conservative whites and in the game reserves. In August 1995 the party published its plans for a provincial constitution for 'The Kingdom of KwaZulu-Natal', which demanded what was in effect an independent army and police force and exclusive powers over all constitutional, legislative, judicial and financial matters. Buthelezi followed this up at 'national gathering' in Durban, on 20 August, in which he exhorted all Zulu 'to support a "covenant" pledging [themselves] to battle for an autonomous kingdom'. 5 Since 1996, however, this highly charged situation has simmered down. Not only has political violence greatly diminished; although the IFP's provincial constitution was disallowed by the Constitutional Court, one hears little today of the separatism which was so long a feature of the region. 6 While there were still sporadic killings and considerable tension, the 1999 national elections saw little of the bloodshed which marred the run-up to the 1994 election. Despite rumblings from powerful chiefs in the region in the first months of the new millennium over government plans to change municipal frontiers and deprive rural chiefs of some of their powers in local government, fears that this would lead to 3

3 April 1994. Cited in Sampson, Mandela, p. 486. KwaZulu was then an 'independent' homeland, and Natal one of the four provinces in the Republic of South Africa. In terms of the new Constitution, the renamed KwaZulu-Natal is one of South Africa's nine provinces. I have used 'KwaZulu-Natal' in what follows, unless the text refers specifically to historic Zululand, Natal or Kwazulu. 5 Laurence E. Piper, 'The Politics of Zuluness in the Transition to a Democratic South Africa', D.Phil, thesis Cambridge University, 1998, p. 206. The original version of this chapter was written before I read Piper's fascinating thesis which asks a similar question - in his words, why Zulu nationalism 'did not explode with a bang but ended with a whimper' (p. 258) - and covers political developments from 1990 to 1997 in far greater depth than I do here. I am enormously grateful to John Lonsdale for drawing this thesis to my attention. I have indicated in my footnotes where I have used the thesis to further substantiate or sharpen my argument. 6 For the long history of secessionist threats from this region, see below. 4

Ethnicity & Democratization in KwaZulu-Natal

185

another 'wave of violence, reminiscent of what took place in the 1980s and 90s', proved unfounded. 7 Why, then was the ethnic separatist card played with such devastating success in KwaZulu-Natal in the run-up to the 1994 election and its immediate aftermath? After all, under apartheid, it was the avowed intention of the government to encourage ethnic identification among all the so-called 'black nations' of South Africa - Xhosa, Sotho, Tswana, Venda, Tsonga, Ndebele; yet they had relatively little success. After February 1990 when the goverment unbanned the ANC, the state-created, ethnically-defined 'Bantu homelands' virtually all collapsed like a stack of cards; their rulers were for the most part discredited, their people anxious to claim their rights in a unified South Africa. What, then, is peculiar about the KwaZulu-Natal region, that led apparently to so successful an ethnicization of its politics? And - equally important - why has it now all but disappeared as a threat to the civic nationalism of the state? The first point to make is perhaps a simple demographic one. At some 7 million, Zulu-speakers are the single largest 'ethnic' group in South Africa, and have relative linguistic and cultural homogeneity. With a decentralized federal state and proportional representation, an ambitious ethnic entrepreneur like Chief Mangosuthu Buthelezi had everything to play for. Following the example of ethnic nationalists the world over (and of Afrikaner nationalists closer to home), he realized that if he could get the Zulu to identify first and foremost as 'Zulu' (not as Zulu-speaking Africans, or as Zulu-speaking workers, or as Zuluspeaking social democrats) he would have an impressive power base in any postapartheid dispensation. In this political project, Buthelezi had two advantages to draw on, apart from his own very considerable personal political skills: the availability of an already mythologized history of the heroic Zulu people, which gave him an unrivalled ideological resource, and the apartheid state's creation of balkanized 'Bantu homelands' in the 1960s and '70s based on the authority of 'traditional rulers', which gave him a political base.

The Political Mobilization of Zulu Ethnic Identity Originally simply one of the small, closely related chiefdoms in the coastlands of South East Africa, the Zulu rose to prominence in the second and third decades of the nineteenth century, under their revolutionary leader, Shaka, who became king of the Zulu people in 1816 and created an empire by conquering his neighbours. It was during Shaka's reign that white traders from the Cape Colony arrived in Natal, harbingers of the missionaries, settlers and colonial rulers who were to establish themselves between the Thukela and Umtamvuna rivers in the succeeding two decades. And it was during this period that the legend of Shaka and his victorious armies was produced both by African communities and the literate whites who encountered them, and provided some of the raw material for Inkatha's heroic version of Zulu history.8 Although Natal became a British colony in 1843, north of the Thukela River the Zulu Kingdom remained independent of white rule until 1879, periodically 7 Sicelo Dladla, 'Another Wave of Violence Threatens to Engulf KwaZulu Natal, warn traditional leaders angered at the government's plans for new demarcations of municipal boundaries', Land and Rural Digest, March/April 2000. 8 For a brilliant account of this interactive process, see Carolyn Hamilton, Terrific Majesty. The Powers of Shaka Zulu and the Limits of Historical Invention. Cambridge, MA, 1998.

186

Shula Marks

inspiring terror among neighbouring white settlers with their collective memories of Zulu military might. Their image of the Zulu as a powerful 'warrior race' became as crucial for later Zulu self-representations. The Anglo-Zulu war of 1879 with its initially spectacular Zulu victories further confirmed this stereotype. 9 Hence Buthelezi's rallying call to the chiefs in KwaZulu Natal as 'Zulu brothers born out of Zulu warrior stock . . .' 10 As Mare has commented, 'This idea of "manhood" permeates the vision of the essence of Zuluness.' 11 For much of the nineteenth century the colonial state south of the Thukela in Natal was extremely weak, and was forced to come to terms with the existing structures of African society; as elsewhere in colonial Africa, this was the basis for what Mahmood Mamdani has termed the 'bifurcated state'. 12 Natal's form of governance also provided a model in the twentieth century for broader South African policies of rural segregation and later apartheid. Between 1846 and 1864 eight reserves were set aside in Natal for the sole occupation of Africans, some one-eighth of its land area. In the absence of administrative or financial resources, Africans on the reserves were reconstituted into 'tribes' and governed through chiefs, appointed by the government. African customary law - as interpreted by the colonizers - was recognized, and the British Governor took the position of 'Supreme Chief, with powers over the African population that were supposedly modelled on those of the Zulu kings. After its annexation by the British in 1887, and incorporation by Natal in 1897, similar policies were applied to Zululand, north of the Thukela river. Initially Africans in Natal appear to have acquiesced in colonial rule. So long as they had access to land, they had little need to work on the white-owned sugar plantations that by 1860 had become the most important sector of the colonial economy. As a result, Indian indentured labourers were brought to Natal in large numbers: by 1900 they slightly outnumbered the white settlers. Despite the increasing involvement in the world economy of both white settlers and black peasants, these policies tended to conserve - rather than to restructure African society, albeit in deformed ways, and to give the indigenous population a strong sense of identity against the Indian 'other'. While the lands reserved for Africans were regarded as adequate when they were set aside (although even then the missionaries argued that they were 'fit only for the eagle and the baboon'), through the last century they became increasingly overpopulated. By the early decades of the twentieth century land shortage and increasing impoverishment were already fuelling so-called 'faction fights' between the chiefdoms and fractions of chiefdoms which had been given a continued material reality by the colonial state. The result in both Natal and Zululand was to perpetuate parochial group loyalties and residual land claims. A century-long struggle for control over an ever-shrinking territorial base and its boundaries resulted and was conducted in the idiom of kinship loyalty, which, 9 As late as 1999, ex-President F.W. de Klerk still could not resist the cliche, and writes in his autobiography of the Zulu as 'the proud descendants of fierce warriors' The Last Trek. A New Beginning. London and Basingstoke, 1999. 10 Buthelezi's speech, 13 September 1990, cited in Gerhard Mare, Ethnicity and Politics in South Africa. London, 1993, p. 68. King Goodwill Zwelethini used the identitcal phrase in addressing 'the Zulu nation and . . . all South Africans' on 26 May 1991. Mare published his book in South Africa in 1992 under the title, Brothers Born of Warrior Blood. Politics and Ethnicity in South Africa. 11 Mare, Ethnicity and Politics, p.68. 12 Citizen and Subject. Contemporary Africa and the legacy of late colonialism. Cape Town, Kampala, London, Princeton, 1996, p. 25 and passim.

Ethnicity & Democratization in KwaZulu-Natal

187

whether real or Active, was still probably the most salient identity in rural KwaZulu-Natal, at least for the older generation. 13 In Zululand, as in Natal itself, colonial control and the extraction of rent, tribute, tax or labour power depended on using the powers of chiefs and homestead heads who became the subordinate agents of the state. To this day, about one-third of the 'traditional' chiefs in South Africa are located in KwaZulu-Natal alone, and some 10,000 'traditional' headmen or izinduna\ together they constitute a formidable body of local authority in the rural areas, with serious consequences for regional - and gender - politics, as we shall see. 14 The Zulu monarchy, by contrast, posed a problem for the state of a wholly different order, representing, as it did, the social cohesion of the Zulu nation and the focus of resistance to British rule. Both the imperial and the colonial authorities were determined on its destruction. As a result, the last independent Zulu king, Cetshwayo ka Dinuzulu, was exiled to St Helena for ten years by the British when they annexed Zululand in 1889; he was exiled again by the Natal authorities in the wake of a major poll tax rebellion in 1906 for which he was widely believed to be responsible. The position of the Zulu king was not recognized by the state until after the Afrikaner Nationalists came to power in 1948. The existence of a cohesive kingdom, under threat from the mid-nineteenth century, and under a king who claimed to represent all the people, gave Zululand a capacity for constructing a nationalist identity probably unmatched in the rest of South Africa. This was the great advantage the Zulu north of the Thukela had over their neighbours. And this was the advantage which the first Inkatha movement (founded in the early 1920s) had over other ethnic associations which flourished in other parts of South Africa in the inter-war years. The history of the kingdom has provided a ready and rich resource for Zulu culturebrokers of the twentieth century. So rich a resource was it that Africans in Natal who had never formed part of the Zulu kingdom, who had actively resisted it or who had served in colonial armies fighting it, came to identify with it as an alternative to colonial oppression. At the beginning of the century, for example, Mark Radebe, a prosperous landowner and founder of the Natal Native Congress, responded to the question 'Are you a Zulu?' put to him by a South African government commission with a resounding 'Yes' - despite the fact that his next answer revealed that he was of 'Sesutho [sic] origin'. 15 As Helen Bradford, describing the influence of parochial identities on the rhetoric of Industrial and Commercial Workers Union (ICU) intellectuals in the 1920s, has remarked: 'In Natal in particular, 13 As John Lambeth points out, 'Although kinship was only one of the ties holding people together, it was important ideologically in the maintenance of social and political cohesion', and even chiefdoms defined themselves genealogically {Betrayed Trust. Africans and the state in colonial Natal, Pinetown, Natal, 1995, p. 24). Nevertheless, Jonathan Clegg has shown how in feuds people frequently placed affiliation to their neighbours over familial or clan loyalty. ('Ukubuyisa Isidumbu 'Bringing back the Body': an examination into the ideology of vengeance in the Msinga and Mpofana rural locations, 1882-1924', in P. Bonner, ed., Working Papers in Southern African Studies, Vol. II. Johannesburg, 1981, pp. 190-1. 14 According to the Draft Discussion Document towards a White Paper on Traditional Leadership and Institutions issued by the Department of Provincial and Local Government, 11 April 2000, there are 277 chieftaincies (including 38 vacancies) in KwaZulu-Natal, compared with approximately 460 in the rest of the country. 15 South African Native Affairs Commission 1903-5, Cape Town, 5 vols. 1905 vol. 3, p. 531.

188

Shula Marks

the ruins of South Africa's most powerful precolonial kingdom haunted the present.' 16 Moreover, state attempts to destroy the Zulu monarchy as the pivot of Zulu national unity, even after colonial Natal became part of the Union of South Africa in 1910, merely served to heighten its legitimacy. A new Zulu ethnic identity was forged based on a reinterpretation of past history, and 'custom'. This new Zulu ethnic nationalism was not an unmediated transmission of inherent and immutable past values and culture, however. The redolent historical and cultural tradition of the Zulu did not in itself predict that a Zulu ethnic nationalism would be constructed, although it undoubtedly assisted its creators who formed the first Inkatha movement in the early 1920s, the Zulu Cultural Society which followed on its demise in the 1930s, and the revived Inkatha movement in the last quarter of the twentieth century. Ironically, but not uniquely in the annals of ethnic mobilization in Africa as elsewhere, the first Inkatha was created by an alliance of seemingly incompatible interests. Natal African Christians, many of whose forebears had fled persecution in the Zulu kingdom in the nineteenth century, joined with the Zulu royal Family, anxious to mobilize popular support in order to gain state recognition, and white ideologues of segregation. The latter saw a refurbished traditionalism based on the monarchy as a bulwark against the more radical class-based and nationalist politics; in the words of George Heaton Nichols, their key ideologue, 'Bantu communalism' was the answer to 'Bantu communism'. 17 At a time of considerable social change, the new Zulu ethnic nationalism resonated with popular experience. Historically, economic development in South Africa has been based on the constant oscillation of migrants between town and countryside, dependent on urban wages yet subsidized by women's production in the rural reserves. Paradoxically, migrant labour acted both to conserve many of the political powers of chiefs, who retained control over rural resources, and to discredit them as subservient to the state. Yet if many of the subordinate chiefs became known as 'government's boys', others retained the support of their followers by resisting the excesses of colonial overrule, overtly or covertly, and by continuing to provide their followers, many of whom were migrant labourers, with a degree of security over their lands and, crucially, over women in the rural areas. As a result popular chiefs were able to mobilize considerable numbers of young men in support of their aspirations for authority, even when this involved mortal combat with their neighbours in so-called faction fights. 18 At the same time growing dispossession and exploitation led many people to turn to the Zulu king, precisely because of the state's refusal to recognize his authority. Through their allegiance to the Zulu royal family, they expressed their resentment against white rule and their dreams for a future in tune with their 16

Helen Bradford, A Taste of Freedom. The ICUin Rural South Africa. New Haven, C T and London, 1987, p. 100. 17 S.Marks, The Ambiguities of Dependence in South Africa. Class, nationalism and the state in twentieth-century Natal. Baltimore, M D and Johannesburg, 1986, p. 97. For the alliance between the amakholwa elite, the Zulu monarchy and white segregationists, see my 'Natal, the Zulu Royal Family and the "Ideology of Segregation'", Journal of Southern African Studies, 4, 2, 1978, pp. 172-94 and N. Cope, To Bind the Nation. Solomon ka Dinuzulu and Zulu Nationalism, 1913-33. Pietermaritzburg, 1993. 18 See Sithole, 'Land, Officials, Chiefs and Commoners', pp. 56, 103-5, 125-6; and Debby Bonnin, Georgina Hamilton, Robert Morrell and Ari Sitas, 'The Struggle for Natal and KwaZulu: workers, township dwellers and Inkatha, 1972-1985', in Robert Morrell ed., Political Economy and Identities in KwaZulu-Natal. Historical and Social Perspectives. Durban, 1996, p. 147.

Ethnicity & Democratization in KwaZulu-Natal

189

past: the king was to be their redeemer, representative of a heroic past in strong contrast to their impoverished and humiliating present as farm labourers, miners and 'houseboys'. This allegiance was not confined to the dispossessed or those in the historic Zulu heartland. In his autobiography, Chief Albert Luthuli, at that time President of the African National Congress and Nobel prize-winner, wrote 'Respect for the Paramount is in my bones'. Despite the fact that his Christian community lived on the Groutville mission reserve well to the south of Zululand, he recorded the saying of his people: 'Our doors face in the direction of Zululand!' 19 This support, however, was not unconditional. Luthuli registered his own dilemma as the ' "Paramount" . . . seemed to fall in lamely with the manoeuvres of the Government'. 20 Thirty years before, in 1932, a dissatisfied commoner reminded the scion and heir to the Zulu monarchy, Solomon ka Dinuzulu, even more pointedly at a public meeting: The king was subject to the will of the people. Tshaka became powerful through the efforts of his people - not through his own efforts . . . In ancient times an unworthy king lost his throne at times . . . Nor should the nature of this ethnic nationalism be romanticized - either then or now. 'Zuluness' was, and is, conceived in intensely masculine and hierarchical terms. 22 In the 1930s, one of the key objectives of the Zulu Cultural Society was to restore male patriarchal control over women and youth, as its charter revealed: there was fear that the 'departure from wholesome Zulu traditions' meant a lack of discipline in the homes, and the loss of control over women and youngsters, and this has been a powerful motor of the contemporary Inkatha movement. 23 For African men fears about the loss of control over women were deeply rooted in the role which women played as the producers of labour power and as the source oilobola (bridewealth), while for the state control over female mobility was central to their attempts to slow down the processes of African proletarianization and urbanization. At the same time, farmers were concerned to retain women and young people on the land at a time of intense competition for labour, and used male heads of households as their agents in doing so. The patriarchal collusion between chiefly authorities, household heads, employers and the state was a crucial element in white rule and notions of 'African culture'. 24 Zulu ethnic nationalism was never uncontested, however, whether in the interwar period or later. Many of the same forces which led the more conservative intelligentsia to construct Zulu ethnic nationalism were also responsible for the rise of more radical pan-South African movements, which found a ready 19

Albert Luthuli, Let my People Go. An Autobiography. London, 1962, p. 74. Ibid. 21 Natal Archives CNC 57/7/4 Nl/1/3 (32), minutes of a meeting of Chief Native Commissioner with Solomon ka Dinuzulu and one hundred representatives of the Usuthu chiefdom at Nongoma, 18 February 1932. 22 Mare, Politics and Ethnicity, p. 68. 23 See my 'Patriotism, Patriarchy and Purity: Natal and the politics of Zulu ethnic consciousness', in Leroy Vail ed., The Creation of Tribalism in Southern Africa. London, 1989, pp. 215-40; for the more recent position, see L. Segal, 'The human face of violence. Hostel dwellers speak', Journal of Southern African Studies (JSAS), 18, 1, 1992; Cathy Campbell, 'Learning to Kill? Masculinity, the Family and Violence in Natal', JSAS, 18, 3, 1992; and Shireen Hassan, 'Inkatha's backbone. An Analysis of the Women's Brigade and its role in Inkatha's polities', Paper No 23, Workshop on Regionalism and Restructuring in Natal, Durban, 1988. 24 See my 'Patriotism, Patriarchy and Purity'. 20

190

Shula Marks

response in Natal, although all political organizations in the region had to come to terms in different ways with a powerful undertow of Zulu cultural identity. Even in the 1950s when the African National Congress expanded to claim the allegiance of a far wider constituency in Natal for the first time, there were tensions, particularly over the ANC's non-racial policies and its collaboration with the South African Indian Congress, born in the aftermath of the 1949 antiIndian riots in Natal. Nevertheless, for much of the 1950s Natal was firmly in the ANC fold, under the leadership of Luthuli. In the second half of the century, several processes intensified the struggles in Natal. The processes of economic change and the social dislocations of the interwar period were greatly exacerbated by the state's policies of apartheid between 1948 and the early 1980s. Apartheid involved even greater control over African influx into the urban areas, the extension of the migrant labour system and the reinvigoration of chiefly authority in the reserves through the 1951 Bantu Authorities Act and the 1959 Bantu Self-Government Act. In the hope of satisfying African political aspirations without undermining white political power, the government was even prepared to recognize the Zulu monarchy. 25 The use of 'tribal authorities' to govern the rural areas, and thus consolidate the power of the chiefs, culminated in 1972, with the establishment of Kwazulu as a 'Bantu homeland' (or Bantustan). The establishment of Kwazulu gave the inchoate but still resonant Zulu cultural nationalism embodied in Inkatha a territorial base, even if the highly fragmented 'state' relied on South Africa for more than three-quarters of its revenue and most of its investment. At the same time, it also gave its chiefly political leaders far greater material resources and patronage. These were particularly important at a time when in other ways Kwazulu, like the other Bantustans, was quite incapable of providing subsistence for the vast majority of its inhabitants. As elsewhere on the South African periphery, dire poverty and dependence had led to and been aggravated by the constant drain of the able-bodied and energetic from the rural areas to South Africa's white-controlled mines, farms and industries, whose wealth has been built on their labour. No-one used the space provided by the state's establishment of so-called 'national homelands' or Bantustans more inventively than Chief Mangosuthu Buthelezi, 'uncle/cousin' of the Zulu king, scion of powerful nineteenth-century royal advisers (the basis for his claim to be the king's hereditary prime minister) and Premier of the Kwazulu Territorial Authority which later became the Kwazulu Legislative Assembly. In 1975, at a time when the ANC and other radical non-racial political organizations were banned, Buthelezi filled the political vacuum by resurrecting Inkatha with the covert support of the ANC. Combining an appeal to a heroic Zulu past with the ANC colours, songs and slogans of the 1950s, he used his Kwazulu base and resources to build up a powerful constituency. The Kwazulu government and Inkatha rapidly became synonymous, as Inkatha members filled all the seats in the Legislative Assembly and dominated the executive and the bureaucracy. 26 Initially Buthelezi was brilliantly successful in working within the system, while distancing himself from it. As he advised the Zulu people, they 'should 25

Bill Freund, 'Confrontation and Social Change: Natal and the forging of apartheid 1949-72,' in Morrell, Political Economy and Identities in KwaZulu-Natal, p. 132. 26 Truth and Reconciliation Commission of South Africa. Report. Cape Town, 1998, p. 454, calls them 'interchangeable concepts'.

Ethnicity & Democratization in KwaZulu-Natal

191

take as much as they could within the system5.27 Through the 1970s, with the blessing of the ANC, Buthelezi was able to walk this tightrope, although he faced mounting criticism from members of the Black Consciousness Movement who accused him of colluding with the apartheid state. Thereafter, with the rise of alternative nodes of resistance, Buthelezi began to paint himself into a corner, because of his increasing dependence on the apartheid state. Effectively Kwazulu became a one-party mini-state and he found himself having to put down a rising tide of popular resistance. By 1979, Buthelezi had broken with what he termed contemptuously 'the ANC in exile', while portraying himself as the true heir to its legitimate tradition. By opposing its armed struggle and sanctions campaign, he now projected himself nationally and internationally as the 'moderate' leader whom local whites and international capitalists could trust and fund. The internal image was rather different. Seriously opposed by the surging internal resistance, Buthelezi retreated to his local constituency and its ethnic symbolism. Portraying the rising tide of anti-apartheid militancy as anti-Zulu, between 1980 and 1985 the Inkatha movement acquired a reputation for violent vigilante action, against schoolchildren, university students, workers and community leaders, while the Chief himself uttered veiled threats against his critics, especially those who were non-Zulu. With his close ties to business, Buthelezi was particularly angered by the launch of the radical Congress of South African Trade Unions in Natal in 1985, and the violence escalated. With the unbanning of the ANC in February 1990, this reached a new pitch of intensity, as people had to choose between competing loyalties. As we now know, from the mid-1980s as the internal resistance to apartheid mounted, close collusion also developed between the KwaZulu premier and South Africa's security forces, and in 1986 Buthelezi sent 200 Inkatha soldiers to the Caprivi strip to train as an offensive paramilitary force to be deployed against the ANC and its allies in the community. 28 The second Goldstone Report in late 1995, a series of highly incriminating court cases and the subsequent hearings before the Truth and Reconciliation Commission confirmed beyond any shadow of doubt the extent of the security force's complicity in Inkatha violence, and the existence of a Third Force which continued to destabilize the country during the negotiations. 29

27

Mzala, Gatsha Buthelezi: Chief with a Double Agenda. London, 1988, p. 73, cited in Freund, 'Confrontation and Social Change', p. 133. 28 TRC Report, vol. 2, pp. 453-77. See, also, Weekly Mail and Guardian, 16 January, 1998; Electronic Mail & Guardian, 29 October 1998; and Daily Mail and Guardian, 27 November 1998. According to his long-time right-hand man, Walter Felgate, who defected to the ANC in 1998, Buthelezi's association with BOSS went back, in fact, to 1973. 29 According to Sampson, Mandela (fh. on p. 439), 'The expression "third force" had been used by President Botha's State Security Council in November 1985, when they discussed setting up a separate paramilitary unit to enforce internal security. The police and defence could not agree about the control of the unit and both later created separate organisations which effectively did the job. Ministers could therefore claim that no "third force" existed.' As I write, Ferdi Barnard, a self-confessed member of the 'Third Force', has given evidence of the arming of Zulu hostel-dwellers by the Third Force on the eve of the so-called Boipatong massacre on the Rand (Guardian, 3 October 2000). It should be noted that neither Buthelezi nor members of the IFP sought amnesty by appearing before the TRC, though individual members were subpoenaed to do so.

192

Shula Marks

Aftermath Important as the concrete evidence is of Inkatha, SADF and police culpability in fomenting the political violence, it does not fully explain the form, nature, or timing of the ethnic mobilization involved. Laurence Piper correctly sees in this mobilization an essentially political project, 'which strategically invoked Zulu ethnicity, [rather than being] the organic political expression of a national sentiment which had been constructed culturally'. 30 In pursuing his political ambitions, Buthelezi was able to tap into the deep reservoir of Zulu ethnic identification, and was assisted by the availability of those handy building blocks for a sense of nationhood: history, tradition and a common language. 31 Thus, in his attempts to galvanize his constituency Buthelezi laid great stress on Zulu history - with frequent invocations of the glorious Zulu past at Inkatha rallies and political events - and on traditional virtues embodied in the 'Ubunthu-Botho' (Good Citizenship) school syllabus. This, like the Charter of the Zulu Cultural Society in the 1930s, stressed the need for discipline and obedience from the youth, and deference from women. The ideology was calculated to appeal to older men from the rural areas who felt most threatened and alienated by the corruptions of modernity. At the same time Buthelezi could not ignore his youthful constituency. On the contrary, the Inkatha Youth Brigade was Buthelezi's direct responsibility, and in the 1980s it became the largest section of Inkatha. In the hostels on the Rand young workers took the lead in organizing the violence, largely displacing the former control by elders. Yet if this suggests that there was an ambiguity at the heart of the discipline of the elders, there was an even greater contradiction at the heart of Inkatha's espousal of 'tradition'. 32 This contradiction was shown very clearly in the rift that opened up between the Zulu King, Goodwill Zwelithini, and Buthelezi soon after the 1994 election. Inkatha's mobilization of Zulu ethnic nationalism had depended increasingly in the months before the election on an aggressive defence (and manipulation) of the monarchy, despite Buthelezi's earlier strategy of marginalizing the political authority of the king. 33 Freed for the first time from his dependence on Buthelezi for his stipend and body-guard, and with his constitutional position guaranteed by the new government, in September 1994 the king attempted somewhat belatedly to unify his divided people by inviting President Mandela and by extension, Zulu-speaking ANC members - to the annual commemoration of the Zulu founder-king Shaka's death. Buthelezi (now Minister for Home 30

Piper, 'The Politics of Zuluness', p. 259. Piper points out that Buthelezi barely mentions the issue of language in his many speeches invoking Zulu identity. However, he did not have to: this was the one thing his constituency had in common. Nevertheless in the 1990s, one way in which Inkatha warriors distinguished their opponents, the amaqabane - the ANC youthful comrades - was apparently by the less 'pure' Zulu they spoke. 32 Segal, 'The human face of violence'. 33 The relationship between Buthelezi and the Zulu king was frequently strained and took care to ensure that the king would be relegated to a purely symbolic role. In 1972, for example, Buthelezi prevented the king from exercising executive powers in the KwaZulu Legislative Assembly, or even personally attending its proceedings, on the grounds that the monarch should not be embroiled in party politics. The tension between Buthelezi and the king surfaced intermittently through the years, but from 1985 the king became increasingly important to Inkatha in its intensified struggles with its political and trade union opponents. (See Bonnin et al. 'The struggle for Natal and KwaZulu', pp. 148, 173; Piper, 'The Politics of Zuluness', pp. 110-11; M.Z. Shamase, King Zwelithini. A Zulu Reign in Crisis, Empangeni, 1997, pp. 7-8, 13-21.) 31

Ethnicity & Democratization in KwaZulu-Natal

193

Affairs in South Africa's first post-apartheid government) objected that he should have been consulted as 'hereditary Prime Minister' - a position Zwelethini now repudiated - and maintained that he could not ensure the safety of the State President. Amidst rumours that he intended replacing Buthelezi on his royal council, the king expressed fears for his life while Buthelezi conducted the Shaka Day ceremonies in defiance of a royal ban and in the absence of Goodwill; as one journalist remarked it was like Hamlet without the Prince. 34 In fact, Buthelezi's espousal of 'traditionalism' served to disguise the radicalism of his support for free enterprise, and the consequences of his alliance with the apartheid state and capital in the 1980s. By the late 1970s he had emerged as the most articulate black protagonist of capitalist development, wooing the international business community and persuading some of its maverick rightwing millionaires to fund him. In the 1980s the consequent, highly uneven economic development in Kwazulu intensified the stratification of the local populace into a select band of entrepreneurs associated with the top echelons of the 'state' and party, who managed to accumulate considerable wealth through their ties with the bureaucracy, through sugar farming and the Kwazulu Development Corporation, and a vast mass of people who were marginalized and displaced by the same processes of 'enterprise capitalism' - and who found their way into the squatter townships of Natal. 35 These processes of class formation, proletarianization and urbanization were accompanied by enormous social upheaval and conflict. The enterprise capitalism which Buthelezi applauded involved the breaking up of community in Natal and Kwazulu through the allocation of communally held lands for individual tenure; together with drought in the 1980s, it allegedly put Durban and Pietermaritzburg among the fastest growing cities in the world. There was acute competition over resources, as huge numbers of the landless flooded into shacklands around the major towns, forcing the repeal of the pass laws. In Molweni in 1989, for example, battle raged between 'a settled community afraid of sharing resources with impoverished newcomers . . . victims of flooding and the construction of the Inanda Dam. The refugees felt a sense of powerlessness and loss while older Molweni residents resented the pressure on local resources through crowding in a time of growing impoverishment and unemployment.' 36 At the same time, generational and gender conflict intensified as young women and men came to assert their political agency, and male heads of families attempted to reassert their authority. In a situation of rapid social change in which the power of adult men in the family was being constantly eroded, both by their lowly position in the race-class hierarchy and by the growing number of female-headed families, high rates of unemployment and 'catapulting' of the youth into political prominence, there was a reassertion of patriarchal power, which encapsulated not only tensions in family life but also many of the urban and rural struggles in South Africa.37 In many ways the violence which characterized South Africa during the transition resulted as much from the insecurities 34

See Weekly Mail and Guardian, 9, 16, 23 and 30 September, 1994 and John Carlin's accounts in The Independent, 21 and 22 September 1994 for an account of these events. 35 D.J. Tilton, 'Writing the script for the future. Inkatha and the role of development in Kwazulu', D. Phil, thesis, Oxford University, 1991. 36 S.Stavrou and A. Crouch, 'Violence on the Periphery: Molweni', unpublished paper presented at the Twentieth Annual Congress of the Association for Sociology in South Africa, University of Witwatersrand, July 1989. 37 See, especially, Campbell, 'Learning to Kill?'

194

Shula Marks

- and unfulfilled but suddenly no longer wholly unattainable aspirations released by reform and the disintegration of 'the spatial order of apartheid' as from apartheid itself.38 How better to blur the explosive consequences of class formation and potential class conflict than by calling for cross-class Zulu identification - and blaming the impoverishment of the Zulu not on apartheid, or the accumulation of the new middle class, but on the youth and their 'disrespect for authority', on the ANC and its call for economic sanctions, and on the unions and their 'wildcat' strikes, which deprived Zulu workers of employment. Thus the most desperately impoverished of South Africa's black population - the migrants living in the anomic single-sex hostels on the Rand - were enlisted in a vicious war on neighbouring squatter camps through appeals to their Zulu identity and their ancient military glory, while in KwaZulu-Natal, too, many of the battles were between the most recently proletarianized of its refugees and migrants and the more established and stable urban communities. As elsewhere, broader processes of 'globalization' lay behind ethnic revival.39 At least as important as the party differences, behind the violence lay generational cleavage and class differences: as many observers pointed out at the time, very often '. . . party affiliation was shadowy and all but the dimmest knowledge of national politics rare'. 40 Often it was the local politics of dispossession and perhaps more historic tensions over chieftaincy succession disputes and land that were played out in the growing violence, not the wider national political issues.

Why the Dog Did Not Bark There can be little doubt that the conflicting political ambitions of the ANC and IFP are of considerable importance for the future stability of post-apartheid South Africa, as demonstrated both by the violence in the run-up to the election in Natal in 1994 and the results in both the 1994 and 1999 national elections and in the 1996 local government elections, in which Inkatha retained a hold over the majority of rural (but not urban) Zulu-speakers. In response to what is perceived as the power of ethnic identification, Zuluspeaking members of the ANC have begun to develop their own versions of Zulu 'tradition'. Increasingly, the political battle between Inkatha and the ANC, like that between the king and his uncle, was framed as an ideological struggle over what it meant to be 'Zulu'. By 1993 ANC leaders were attending meetings in 'traditional' Zulu dress. Since 1994 the KwaZulu-Natal Arts and Culture Council has encouraged the revival of Zulu culture and in 2000 produced a calendar illustrated with pictures of the Zulu kings which was completely sold out. Finally, in a truly post-modern twist, it was reported that '. . . the youth [formerly those most hostile to Inkatha's version of 'tradition'] are talking to traditional leaders about setting up cultural events and starting heritage sites that will 38 M. Morris and D. Hindson, 'Political Violence and Urban Reconstruction in South Africa', unpublished paper for Economic Trends Meeting, 7 July 1991. (I am grateful to Mike Morris for sending me this paper.) See also Bill Freund, 'The Violence in Natal, 1985-90', in Morrell, Political Economy and Identities, pp. 188-90. 39 For 'elsewhere' see, for example, Karl Cordell ed., Ethnicity and Democratisation in the New Europe. London and New York, 1988, p. 4 and passim. 40 Segal, 'The human face of violence.' Stavrou and Crouch, 'Violence on the Periphery: Molweni', also argue that the political definitions are too clear-cut and that these identities cohere around different factions in a struggle for scarce material resources, rather than out of the different ideologies of the organizations per se.

Ethnicity & Democratization in KwaZulu-Natal

195

form the basis of a new tourism industry in their areas. Both parties are saying "Hey! There are new businesses to be had", and are beginning to co-operate around their common culture.' 41 At the same time, since coming to power, the ANC has been forced to take the powers not only of the king, but also of the hundreds of chiefs and thousands of headmen seriously. A recent draft discussion paper intended to 'affirm and define' 'traditional leadership' and 'clarify' its role in 'democratic governance' carries a foreword from the Minister for Provincial and Local Government, Fholisani Sydney Mufamadi, arguing that: Traditional leadership embodies a system of discourses, which describe Africa's earlier forms of societal organisation. In this sense it is a monument to our past and is a true icon of our identity as Africans ... The call for an African Renaissance, therefore, is a call for the coming of age of Africa's institutions, and not for a perfect mimicry of the world's powerful societies. After all [he concludes] it is hard to conceptualise African culture without any reference to the institution of traditional leadership and customs. The complexity of these events warns against their easy labelling either as some kind of return to an atavistic past - or, its obverse, a demand for more democratic local autonomy. As in most, if not all, ethnic nationalist movements, Zulu history and the demands for an autonomous Zulu kingdom entered Inkatha's configuration of events as an invented, reworked and contested past, not an accurate portrayal of reality. Moreover, for Buthelezi the national stage was always the ultimate goal. The ethnicizing of political conflict and its eruption into brutal killing was a product of present political interests, as Piper has so clearly shown, not of past culture. Behind the violence lay not deep cultural or psychological traits, some kind of 'Zulu essence', but a political strategy which found inflammable material in the intensification of rural poverty, migrancy, unemployment, and urban overcrowding as apartheid began to unravel - the deliberate actions of agents provocateurs, and the unpredictable outcomes of a culture of resistance and ungovernability. That it took an ethnic form was the result of a constant reworking of ethnic divisions and the deliberate glorification of Zulu military 'traditions' by politicians and culture-brokers, both black and white, in pursuit of political objectives. Ultimately Chief Buthelezi's success in this depended on his ambition and astuteness, and the support he had from the South African state, its security forces and Western governments, rather than on any tug of primordial emotion. Some of the clues to the diminution of political violence in Natal can also be discerned in this narrative. First, it should be noted that not all violence in KwaZulu-Natal has died down. A good deal of what used to be called oldfashioned 'faction fighting' which long preceded the 1980s continues, expressing much older political rivalries. Levels of criminal violence are high, fuelled by appalling levels of poverty and unemployment and as former warlords and security agents turn to new ways of making a living as the new state takes control of law and order. Political tensions continue to smoulder in pockets of the province, which remains divided into Inkatha and ANC no-go areas. It is also no coincidence that the most heated resistance to the government's plans for new demarcations of municipal boundaries came from KwaZulu-Natal. 43 41 42 43

Eddie Koch, 'A Zulu renaissance could bring peace'. Electronic Mail and Guardian, 23 July 1997. Draft Discussion Document, p. 3. See above, and footnote 7.

196

Shula Marks

Nevertheless, at least for the time being any breakdown of social order in the name of Zulu separatism seems remote. Part of the answer is undoubtedly also to be found in the very careful handling of Buthelezi and Inkatha, and especially the successful wooing of the Zulu king, by an ANC leadership only too well aware of the consequences of politicized ethnicity backed by international right-wing forces. The resemblance to Savimbi and UNITA in Angola was only too chillingly evident. Mandela's own earlier friendship with, and perseverance in conciliating, Buthelezi, 44 Mbeki's renowned 'pragmatism', and the role of Jacob Zuma in attempting to mollify Buthelezi and draw him into the national government both as Minister of Home Affairs and even on occasion as Acting President have all been crucial. 45 Major concessions on issues of local autonomy and 'traditional' leadership and institutions have also played their part. The death of Harry Gwala, leader of the ANC in Natal and a hard-line opponent of peace with Inkatha, was a further factor facilitating peace. But, to put it crudely, it takes two to tango. And certainly until late in 1996 there was little sign that Buthelezi was prepared to drop his apparent commitment to Zulu autonomy and invocations of violence. So, what has changed? Several shifts in the political terrain help explain this turn. According to Piper, the more extreme demands for Zulu autonomy were always part of Buthelezi's brinkmanship, £a defensive reaction from a political party on the retreat', a way of asserting the IFP's right to inclusion as a major player in the politics of negotiation, rather than a gambit for secession. Once the party achieved political power at the regional level, and Buthelezi himself was rewarded with a ministerial position and even Acting President status nationally, both the party and its leader had a stake in the new dispensation. While Zulu ethnicity is real, Zulu nationalism in the sense of a demand for Zulu political autonomy is not. 4 6 This is undoubtedly largely correct, 47 although Buthelezi's brinkmanship cost thousands of lives and very nearly ended in disaster. In the end, as so often in the past when the province threatened to 'take off', its centrifugal tendencies proved weaker than the centripetal ones. There is, after all, a long history of secessionist threats from Natal which go back to the time of the Union, when English-speaking Natalians feared being 'swamped' by the Cape 'liberal' tradition (with its non-racial franchise) and Afrikaner nationalism. The threat of secession was invoked again in the 1930s at the time of the Smuts-Hertzog pact, and intensified once more after the National Party electoral victory in 1948. On each occasion white Natalians' desire for autonomy was outweighed by their fear 44

Sampson, Mandela, pp. 436-7, 470, 485. Sampson, Mandela, p. 512. 46 Piper, 'The Politics of Zuluness', p. 259-60. 47 While I find myself in broad agreement with Piper's study, I believe the distinction he draws between cultural and political nationalism, the first an 'organic and spontaneous' expression of 'a culturally constructed nationhood', the second a purely instrumental response to party interest ('The Politics of Zuluness', p. 201), is too inflexible. I suspect that his argument on p. 172 that because Zulu-speakers divided 40: 60 between the ANC and IFP 'the Zulu "nation" did not properly exist' and that it was the result of an 'invention from above', rather than of 'popular imaginings from below', the work of party agency rather than supposed social identities', is too stark. The argument seems premised on an ideal type of nationalism which has rarely if ever existed. The reworking of ethnicity into nationalism is always a political project - in Bosnia and Kossova no less than in KwaZulu-Natal; and its success is always dependent on the cultural material available for the political brokers, the capacities of the cultural entrepreneurs and the precise political configuration in a given region. 45

Ethnicity & Democratization in KwaZulu-Natal

197

of being 'swamped' by the Zulu, and their recognition that without the Rand Natal's economy could not survive.48 Nor have these threats come only from the white community with its demands for federalism. Throughout the twentieth century Zulu separatism within the ANC has been almost as marked. Although the Reverend John Dube from Natal was the first president of the South African Native National Congress (the precursor of the ANC), he was ousted in 1917 because he seemed to accept the state's policy of segregation. Like Buthelezi, he was forced back onto his Zulu constituency, so that until the Second World War there was a dissident Natal African Congress in existence, in opposition to the national organization. Even after Dube's death in the early 1950s, there were continued strains between the ANC nationally and the ANC in Natal, while the Natal Indian Congress was also frequently at odds with the national South African Indian Congress. 49 By the 1990s, however, the people of KwaZulu-Natal, white and black, were simply far too dependent on the South African economy for a secessionist project to have popular appeal. Also important was Buthelezi's loss of support from his erstwhile international backers. The fall of Margaret Thatcher, Ronald Reagan and Chancellor Kohl, and the internationally accepted legitimacy of the post-apartheid state, have been extremely important in this; at least as significant for the recognition of the new South Africa's legitimacy has been the acceptance by the ANC government of 'free enterprise' and the market economy. 50 Quite simply, international capital found it had nothing to lose from an ANC government, now that the Cold War was over and the Communist bogey gone. A key moment in this was Mandela's conversion at the Davos World Economic Forum. As he told the ANC and the country on his return, 'I came home to say: "Chaps, we have to choose. We either keep nationalization and get no investment or we modify our own attitude and get investment'". 51 Clearly the West can do business with an ANC government as well as, if not better than, with the volatile Buthelezi and his flirtation with secession. All this has meant an end of externally generated funds for Inkatha which seems to be in some financial difficulties. Nor have local capitalists who seemed at one stage to be backing Inkatha delivered since 1994; as in the past, local businessmen have recognized their dependence on central government and a unified South African state. Financial difficulties may also have concentrated the mind of an ageing and ailing Buthelezi, when the choice was between government office and funds, or the bunker and bankruptcy. The UNITA option was simply not there. At the same time, it is possible that there is another powerful pressure on Buthelezi to lie low. As became increasingly clear in a series of trials since 1994 and during the TRC hearings, the evidence showing Buthelezi's close relationship with the security forces over many years and incriminating him in 'gross 48 White Natal's secessionist tendency has been brilliantly satirized by Anthony Delius in his The Day Natal Took Off. Cape Town, 1963 as well as Tom Sharpe's, Riotous Assembly. London, 1971 and Indecent Exposure. London, 1972, and has been treated more seriously in Arthur Keppel-Jones's partly prophetic When Smuts Goes, 1947. 49 I am grateful to Dr Heather Hughes for this point. 50 For the importance of the international right, and for Thatcher in particular, see Sampson, Mandela, pp. 440-1 and passim. 51 Sampson, Mandela, p. 435. As Sampson sees it, 'The settlement with international capitalism was as important as the settlement with de Klerk.' Ibid. p. 451.

198

Shula Marks

human rights abuses' was overwhelming. The chances of his emerging unscathed from a judicial process was uncertain, to say the least. Under the circumstances, it would surely seem far safer to remain in government than risk further revelations in a possible trial. Both the shift to the right in ANC policy and Buthelezi's weakened position, as well as a recognition by Inkatha party members that development resources for the desperately poor province depend on state patronage, mean that there has also been a considerable rapprochement between Inkatha and the ANC at a local level. This has progressed so far that in KwaZulu-Natal there is now a coalition ANC-Inkatha government and there are sporadic rumours of an Inkatha-ANC merger, around a new, 'non-partisan' Zulu culture. 52 This brings me to the final paradox. In the past the contest between Inkatha and the ANC shaped the constitution and ensured a degree of federal autonomy, while the threat of violence from an unsatisfied Inkatha leadership continued to menace the very democracy it espoused. Today, however, it is the talks of a merger between an already powerful ANC and the only other black party which has been able to mobilize significant numbers of votes which could jeopardize the expression of popular dissent, limit the extent of democratic choice in South Africa and check the government's commitment to gender equality. The talks are yet another example of the 'climate of conciliation and cynicism' which, according to Chris Gilligan 'provides the context within which democratization is being discussed today' internationally. 53 Such a merger would almost certainly move the ANC further centre-right and increase tensions in the relationship between the ANC, the South African Communist Party and the Congress of South African Trade Unions (COSATU). In his Citizen and Subject, Mahmood Mamdani recognized that in South Africa, as in the rest of Africa, 'an overall democratization that would transcend the legacy of a bifurcated power' requires what he terms 'the detribalization of customary power' and the dismantling of the old 'Native Authorities' of the colonial or apartheid state. The immediate danger from a militant Inkatha may have been resolved through these processes of co-option and conciliation. Failure to dismantle the old 'Traditional Authorities', however, means that the danger Mamdani feared - that the non-racial South African state will reproduce 'the legacy of apartheid' in the countryside by allowing the continuance of nondemocratic ethnicized rural institutions - remains a distinct possibility.54 The continuance of these 'ethnicized rural institutions' - chieftaincy, customary law, 'tribal authorities', legitimated through 'discourses of "tradition", "custom" and "African culture"' - also have extremely serious implications for women, as Cherryl Walker noted as early as 1995. As she showed then, the 'tradition' which was being negotiated with the Zulu king in the run-up to the first democratic elections was 'fundamentally patriarchal', despite the ANC commitment to gender equality. 55 Walker stresses that 'the Zulu king is a representative and symbol not only of a Zulu ethnic identity, but of a patriarchal social order, which he himself upholds and promotes'. 56 Thus, the discourse of culture and tradition is not 'gender-neutral. Men and women do not stand in the same rela52

See, for example, Eddie Koch and Enoch Mtembu, 'KwaZulu: the centre may just hold', Electronic Mail and Guardian, 6 June, 1997. 53 Chris Gilligan, 'Citizenship, ethnicity and democratisation after the collapse of left and right', in Cordell, Ethnicity and Democratisation, p. 38. 54 Mamdani, Citizen and Subject, pp. 25, 26 and 32. 55 Cherryl Walker, 'Women, 'Tradition' and Reconstruction', South Asia Bulletin, Comparative Studies of South Asia, Africa and the Middle East, XV, 1, 1995, p. 58.

Ethnicity & Democratization in KwaZulu-Natal

199

tion to "tradition" and do not necessarily agree on what is valuable or significant or on what practices should be retained today, and in what form.' 57 It also has extremely serious implications in a time of rampant HIV/AIDS - nowhere more rampant than in 'patriarchal' Natal. If the discourse of tradition is profoundly ambiguous for women, it is also profoundly ambiguous for the poor. As Ari Sitas has remarked, while a renaissance in Zulu culture might bridge the gap between the 'youth and traditional leaders' around a celebration of heritage and history, 'healing through heritage' is unlikely to do much for 'the poverty-stricken shantytowns around Durban, where the realities of penury and neglect continue to make life harsh and brutish' - and, since the advent of HIV/AIDS, short. 58 56 57 58

Ibid. p. 68. Ibid. p. 60. Cited in Koch, 'A Zulu renaissance'.

12

G I T H U MUIGAI

Jomo Kenyatta & the Rise of the Ethno-Nationalist State in Kenya

T

HIS chapter sets out to explore the role played by Jomo Kenyatta in establishing the place of ethnicity and ethnic sub-nationalism in the politics of modern Kenya. The chapter traces the historical events that established the political culture within which Kenyatta and other African leaders in Kenya came into political consciousness, and the influence that this had on their perception of the ethnic question in African politics. Secondly, it investigates Kenyatta's personal response to the challenge of ethnic nationalism. It endeavors to demonstrate that, whereas Kenyatta himself may have had a nascent commitment to Kenyan nationalism, he inherited from the colonial government a political culture and a political class that placed ethnicity at the heart of politics and conspired to derail the nationalist agenda at every turn. Finally, the chapter seeks to show that, while Kenyatta's personal charisma and autocratic grip on power served to conceal the growing significance of ethnicity as the basis of political mobilization and organization, his own captivity by the Kikuyu power elite served to fuel a deadly ethno-nationalism and to expand the potentially destructive nature of this force. Most traditional scholarship on the issue of ethnicity and politics in Africa has tended on the whole to be ahistorical accounts seeking to demonstrate how primordial forces continue to dominate contemporary political processes. We contend that several issues remain underinvestigated. Three are particularly important. The first is the enduring legacy of the colonial state as the cradle of ethnic consciousness and politics. The second is the character of contemporary ethnicity as a response to powerful forces of social change. The third is the complex issue of the making of ethnicity by political leaders like Jomo Kenyatta. The state, as constituted at independence, was essentially a compromise of ethnic claims to power and resources. It bore little relevance to the reality of the political lives of the majority of the people. For most people the focus of political obligation and loyalty remained the ethnic group. For most of the leaders as well, in the absence of any coherent ideology, the ethnic group remained the basis for and the focus of political mobilization. Jomo Kenyatta started off unpretentiously as a reluctant leader of the Kikuyu people. He was catapulted into the leadership of the Kenya nation on the back of a broad nationalist coalition that subsumed ethnic divisions in the fight for political independence. The nationalist alliance failed to survive the post-independence struggle over power and resources (Nyong'o, 1986). Ethnicity re-emerged as the dominant political ideology of the day. Not surprisingly, Kenyatta's response to the issue of ethnic nationalism defined the character of the state that he created and the state he bequeathed to his successors. Yet Kenyatta's response was often confused. He constantly sought to reconcile the dual role of being a 'tribal chief with that of being 200

Jomo Kenyatta & the Rise of the Ethno-Nationalist State of Kenya

201

President of the nation. Faced with the twin problem of forging a nation from the diverse ethnic communities of Kenya and placating the Kikuyu masses., to whom he felt he owed a historical duty, Kenyatta decided to pursue both goals simultaneously. In many ways his strategy was unlikely to resolve the issue of historical ethnic claims. First, by choosing to surround himself with an inner circle of Kikuyu advisers, he was seen as having created a Kikuyu government within the government, to the exclusion of other ethnic groups. Secondly, by choosing to champion the deep-seated land and power grievances of the Kikuyu, he was perceived as having consented to be the Kikuyu paramount chief. Thirdly, by co-opting the power elite of other ethnic nationalities into his ruling coalition, he set himself up as the ultimate patron in the neo-patrimonial state he presided over, without placating the poor and dispossessed. This complex client-patron network within which Kenyatta set up 'ethnic chiefs', mostly sons of colonial chiefs who owed direct loyalty to him and who assured him of the support of their various ethnic groups, was also a way of broadening the class base of his regime. In Kenyatta's Kenya ethnic and class politics were intertwined. Without doubt Kenyatta's rule had a great and abiding influence on the politics of Kenya. He made Kenya in his own image. His legacy was largely a divided one, however. This chapter seeks to understand part of that legacy.

The Political Rise of Jomo Kenyatta Johnstone Kamau Ngengi, later Jomo Kenyatta, was a reluctant politician. After completing his early education at the Church of Scotland mission headquarters near Nairobi, he worked briefly as an interpreter at the Supreme Court in Nairobi, and as a meter reader with the City Council. He was enjoying the lifestyle of the new urbanized Kikuyu at Dagoretti near Nairobi when he was drawn into nascent African politics by pure happenstance. James Beauttah, a telephone operator with the colonial Post Office and who served as secretary to the Kikuyu Central Association (KCA), had been transferred to Kampala and the KCA needed a new secretary (Brown, 1972: 104) The KCA was among the very first African political forums in Kenya. While another organization, the Kikuyu Association, existed in southern Kikuyu country its political significance was negligible. The KCA was set up by Fort Hall (Muranga)-based Kikuyu Christian men, like Joseph Kangethe and Jesse Kariuki, who had been demobilized from the African carrier corps after the First World War. They were African proto-capitalists who were unhappy about the treatment they had received at the hands of the colonial government, compared with that of their white colleagues. In their political aspirations they saw themselves as the successors to Harry Thuku's East African Association which had come to an end in 1922. Kenyatta came to KCA politics without any illusions. For him, it was just a better job. His European detractors were later to suggest that he took the job because he had been dismissed by the City Council. The truth is that he was able to extract better terms from an expanded and more financially stable KCA (Brown, 1972: 105-6). Initially, his political contribution was minimal. He merely took the minutes of meetings, translated and drafted letters for the semiilliterate leadership and generally assisted as directed. The KCA was itself a moderate body intent on working with and not against the colonial government. Its slogan was cpray and work'. Over time, however, its demands assumed a

202

Githu Muigai

more political tone. By 1925, inspired by James Beauttah's experience in Uganda, of seeing the progress made by the Baganda under their king, it came up with a number of clearly political demands: first, the release of Harry Thuku, secondly the repeal of the 1915 Crown Lands Ordinance, thirdly the right to grow coffee and cotton, and finally the election of a Kikuyu paramount chief (Kyle, 1999: 26). By 1928 it had started publishing the Kikuyu-language paper Muigwithania, of which Kenyatta was the proud editor. Its stated objective was quite modest: to reconcile Kikuyu traditions with biblical teachings. Its tone was moderate. Its substance was parochial. The unity of the Kikuyu, which it sought, was a cultural and not a political unity. Kenyatta was keen to avoid ending up in detention like Harry Thuku before him. As secretary of the KCA Kenyatta learned four important lessons. One was the value of controlling a medium through which to address the masses. The second was the power of myth. The third was the influence of ideas. The fourth was the importance of building alliances. As a propagandist, Kenyatta was able to use the paper skillfully to create a Kikuyu sub-nationalist ideology that legitimized the accumulation of land and capital by the proto-capitalists of the KCA within a framework of revitalized traditional mythology (Throup, 1987: 33). To his credit, once he warmed to his job, he appears to have recognized the potential of the KCA as a political forum and a vehicle for his ambitions. He threw himself into the job with vigor traveling all over Kikuyu country setting up branches of the association and even going into Meru and Embu country. He wrote to the Governor pleading for the release of Harry Thuku. Kenyatta understood his own limitations as a leader, especially his lack of serious formal education. He aspired to more learning, which he knew would not be available locally, and he longed to go abroad. As fate would have it, by 1929 the KCA, alarmed at the failure of the Hilton Young Commission, before which it had made representations on native land rights, to appreciate the objections to the proposed legislation, had decided to send one of its own to London to make direct representations to the Colonial Office. Kenyatta was not the KCA's first choice. The leadership did not trust him; they were not satisfied as to his commitment. Oscar Beauttah, in whom the association had greater faith, cited family reasons and declined to go. However, he recommended Kenyatta who, though married with children, had no hesitation in accepting the offer. Kenyatta therefore sailed to England in 1929 as the representative and spokesman of the Kikuyu people in their fight against the colonial government. This was the beginning of the myth of Kenyatta as a champion of Kikuyu nationalism. While Kenyatta was in Britain, he became the subject of hero worship among Kikuyu youth, most of whom supported the KCA and who had been radicalized by the position taken by the Church and the government on female circumcision. While Kenyatta himself remained fairly moderate in his politics, his name featured prominently as the model of the Kikuyu leader who was responding directly to white domination, in contrast to the colonial government-imposed collaborating Kikuyu chiefs like Waruhiu and Koinange (Kyle, 1999: 32). Even from abroad, Kenyatta carefully fed the myth of his political significance and achievements. Not surprisingly, over time he became more important in Kikuyu politics than any of his KCA bosses and even threatened to eclipse Harry Thuku who was still in detention. How he had so easily and quickly won acceptance as the true leader of the Kikuyu people is truly phenomenal, especially since he was already in Britain when the emotive issue of female circumcision first broke out.

Jomo Kenyatta & the Rise of the Ethno-Nationalist State of Kenya

203

In Britain, Kenyatta did not achieve much; he met resistance everywhere he went. The imperial government was no different from the colonial government. However, he used his time well. He familiarized himself with British bureaucracy, made important contacts with sympathetic anti-colonial individuals and groups, and traveled widely. Though he was in Europe for only eighteen months, a definite change occurred in his politics. Some of his critics thought Kenyatta had been radicalized by his exposure to the Russian communists, while his sympathizers took the view that he was now better focused and more articulate (Murray Brown, 1972: 130-1). On the whole, he kept his thoughts and impressions to himself. Surprisingly, while he continued to bask in the myth of being the champion of Kikuyu rights, he was not eager to return home. It took the efforts of his English sympathizers to encourage him to return, including buying the ticket to take him back. But even as he set sail for Kenya, one thing was clear to Kenyatta; he intended to return to Britain as soon as possible. Kenyatta's return to Kenya was pleasantly surprising. No sooner had he landed in Mombassa in September 1930, than word went round the Kikuyu reserves that the 'king of the Kikuyu' had returned. Kenyatta basked in the aura of his newfound fame. He plunged with vigor straight into the controversies of the day. But he was soon torn between his role as a modernizer, emphasizing education and modern agriculture, and his support for Kikuyu culture and pride epitomized by female circumcision. This desire to bestride two worlds and his inability to be decisive until the last minute were to dog him for the rest of his political life. On the political front, Kenyatta's claim to be the 'king of the Kikuyu' was complicated three months after his return by the release of Harry Thuku, the real 'king of the Kikuyu'. As if this was not enough trouble, his own position within the KCA was becoming weaker. As fate would have it, the KCA again wanted to make further representations to the colonial government's Parliamentary Select Committee on the future of an East African union. Kenyatta enthusiastically accepted the mandate. Whereas this time the KCA leadership was happy to send him, they still did not trust him sufficiently and somebody else had to be sent with him. In May 1931 Kenyatta returned to London ostensibly to represent the views of the KCA before the Parliamentary Select Committee, despite the presence in London at the time of Chief Koinange, the most senior Kikuyu chief. In the event, Kenyatta's memorandum to the committee was rejected. This does not appear to have discouraged him. He spent the next sixteen years in Europe, travelling and living as far afield as the Soviet Union. He took time off to study at the London School of Economics under Professor Malinowski and to write his seminal work on Kikuyu customs, practices and culture, Facing Mount Kenya. This book was acknowledged as a serious and uncompromizing stand for tribal integrity and an important contribution to Kikuyu cultural nationalism (Rosberg and Nottingham, 1985: 130). He also wrote a pamphlet entitled My People the Kikuyu in which he again expounded his Kikuyu nationalism. As part of his evolution as a thinker and politician, he changed his name from Johnstone to Jomo to signify his commitment to his mission of championing Kikuyu rights against colonial rule. During his stay in Europe there can be no doubt that Kenyatta continued to perceive his role as that of spokesman for Kikuyu interests. His marriage to an English woman did not appear to influence his politics. However, his socialization with Africans from across the Diaspora and his introduction to the idea of Pan-africanism appear to have affected him profoundly. He began to see the

204

Githu Muigai

larger problem of colonial rule as affecting the entire people of Kenya. He now wrote in the newspapers and petitioned the Colonial Office on the colonial problem in Kenya. He was slowly converting into a Kenyan nationalist. In September 1946 Kenyatta returned to Kenya for the second time. It was a return to a country that had been transformed dramatically in his absence. The KCA and other African political organizations had been suppressed by the colonial government as far back as May 1940, as a way of containing growing collaboration between the KCA and other African associations (Berman, 1990: 244). With the extinction of the KCA Kenyatta's political base had been substantially eroded. Some progress had been made, however, in the political field. Eliud Mathu had been nominated as the first African to the Legislative Council and an organization called the Kenya African Union (KAU) had been formed. The composition of this group was significant. For the first time in Kenya, the membership of a political association cut across ethnic boundaries. The leading members were Harry Thuku (Kikuyu - chairman) Francis Khamisi (Coast secretary) Albert Owino (Luo - Treasurer). The rest of the committee consisted of John Kebaso (Kisii), James Gichuru (Kikuyu), F. M. Nganga (Kikuyu), J. Jeremiah (Taita), Simeon Mulandi (Kamba), Harry Nangurai (Masaai) and S. B. Jackayo (Luhya). Thuku, under pressure from more assertive members, resigned in 1945 and James Gichuru took over as chairman. Kenyatta received a hero's welcome. His name had been kept alive during his absence and he had been romanticized as the hero who went to England to confront the colonial government. On his return he was confronted by two immediate dilemmas. One was how to find an entry point into Kikuyu politics, which was still his main frame of reference. The other was how to relate to, and dominate, the new intra-ethnic political formation represented by the KAU. By this time it was clear to Kenyatta that the future lay in articulating a broad-based nationalist political agenda, as opposed to narrow Kikuyu sub-nationalism. But he was shrewd enough to appreciate that one could not have a national following without a strong domestic base. He was therefore not ready to abandon Kikuyu sub-nationalism. To gain an inroad into the radically transformed Kikuyu politics, Kenyatta decided to cultivate the patronage of the old Kikuyu chiefs whom the KCA had long condemned and vilified as colonial stooges. He started by mending his fences with senior chief Koinange. He married Koinange's daughter as his third wife and thereby gained acceptance in the traditional leadership of the Kiambu Kikuyu. Chief Koinange's son Mbiyu Koinange, whom Kenyatta had met in London, was to become a life-long political adviser to Kenyatta. Nearer his home in Gatundu, Kenyatta married the daughter of Chief Muhoho as his fourth wife. His place in the Kikuyu leadership was now assured; he was now finally part of the Kikuyu aristocracy. At the national level, Kenyatta soon stage-managed a takeover of KAU. In June 1947 James Gichuru stepped down as chairman in favour of Kenyatta. This may sound like an incredible feat, but one needs to remember that the bulk of the KAU membership was made up of the remnants of KCA members, in whose eyes Kenyatta still had more stature than most of the leadership of the organization at the time. To his credit, Kenyatta set out to establish KAU as a truly national political forum and a mass movement. But this was not easy. To begin with, there were serious disparities in political awareness, and widespread ethnic parochialism largely fostered by the colonial government. The pastoral communities of the Rift valley showed little interest in national politics. In Western and Nyanza, while political consciousness was higher, there was no

Jomo Kenyatta & the Rise of the Ethno-Nationalist State of Kenya

205

realistic basis for political organization as parochial clan issues continued to predominate (Rosberg and Nottingham 1985: 218). That notwithstanding, Kenyatta forged alliances with leaders like Walter Odede, Achieng Oneko and Oginga Odinga among the Luo and Joseph Otiende, John Adala and W. W. W. Awori among the Luhya, in the hope that these alliances would form the basis of a nationalist coalition that would lead Kenya to independence. While the KAU sought to project a nationalist image, its links with other communities were often strained because of what was perceived as Kikuyu domination of the organization. Many suspected that the KCA still lurked under the wing of the KAU. In the Kikuyu reserves this was indeed true (Berman, 1990: 325). For his part Kenyatta did his best to forge a broad-based executive and to conduct business in Swahili, but the suspicions lingered. At its height, the KAU claimed to have more than 100,000 fully paid up members. In reality it remained at best a 'coalition of Bantu speakers'(Rosberg and Nottingham, 1985: 217-20). But the KAU is not without historical significance. One of its greatest achievements was to recast African politics within a national framework. Between 1947 and 1953 when it was proscribed, the KAU under Kenyatta engaged in a long, albeit futile, struggle to achieve political change by constitutional means. At best, its impact on government policy was negligible. But, whatever its limitations, the KAU provided an open and legitimate national organization pursuing socio-economic and political reforms. While convinced of the ultimate necessity of African self-rule, Kenyatta led the KAU towards the path of moderation in the hope of establishing a constitutionalist political culture. Younger men, more radical and impatient, did not buy into this gradualist plan. This was Kenyattta's third problem; how to reconcile the moderates and the radicals without losing control of the KAU. The membership of the KAU was largely drawn from the African educated classes, civil servants, schoolteachers and other petty-bourgeois elements and this gave the organization its elitist orientation. Many of the educated Africans, while opposed to the color bar and the domination by the settlers, were in fact tied to the colonial state by both their occupations and their aspirations. Herein lay another problem for Kenyatta; how to create a truly mass movement without alienating any of the class factions inherent in contemporary African politics. Despite all its imperfections, the KAU represented the 'pan-tribal' coalition that the colonial authorities had feared and worked to prevent for over a generation. It ensured that African political activity shifted definitively towards colony- wide issues and demands for access to the central institutions of the colonial state (Berman, 1990: 323). All this notwithstanding, by 1950 the KAU was largely moribund. Kenyatta had over time withdrawn from active participation, leaving much of this to Tom Mbotela, his Luhya vice-president. The moderate constitutionalist ideology had run out of steam. Among the Kikuyu urban poor, the landless peasants in the reserves and the squatters in the Rift Valley, the moderate politics of the KAU was increasingly irrelevant. A new crop of leaders was emerging, and the focus of political activity was shifting. Trade unions emerged as more viable vehicles for political agitation. Union leaders such as Fred Kubai, Bildad Kaggia and Makhan Singh were suspicious of the KAU leadership, who they believed were proteges of the Governor (Kaggia, 1975: 79). The growing militancy of African politics coincided with greater repression and coercion on the part of the colonial government. The KAU's inability to effect gradualist reforms shifted the nature of the political struggle from ame-

206

Githu Muigai

liorating specific grievances to an outright challenge to the entire basis of colonial rule. This radicalization of African and particularly Kikuyu politics served to render the KAU less and less relevant and to put Kenyatta's leadership in question. The young radical leaders revived the KCA apparatus and began large-scale oathing of the Kikuyu in Nairobi, Kiambu and the Rift Valley, the home of Kenyatta's father-in-law Chief Koinange being used as a base for this purpose. The Mau Mau rebellion was already under way. In many places where oathing took place, it was associated with Kenyatta's name. Kenyatta had initially encouraged oathing in the 1940s as a way of giving the KAU wider appeal among the Kikuyu as the genuine heir of the KCA. (Kyle, 1999: 49). There is little evidence that he personally sanctioned the second wave of oathings, and there are no reports of his having administered the oath himself. But there can be little doubt that in the early days he must have supervised the same from his safe house at Githunguri teachers' college. After the oaths were declared illegal, it is doubtful that he was ever connected with them. At his subsequent trial, for being a leader of Mau Mau, this was one of the charges that the prosecution was unable to prove (Macharia, 1991: 68). Urban oathing in Nairobi was no doubt conducted under the auspices of the central committee of the KAU. Significantly the central committee, as opposed to the main committee, was made up entirely of Kikuyus, a legacy of KCA history. Kenyatta was genuinely caught between the two factions. While he did not expressly discourage the mass oathings and may indeed have given tacit consent, he continued to pursue the moderate constitutionalist politics of the KAU and seemed to believe it would work. For example in May 1951 when the Labor Secretary of State, James Griffiths, visited Kenya, Kenyatta led the KAU in petitioning for 12 African elected leaders in the Legislative Council and seeking redress for Kikuyu land grievances. He also persuaded the radicals that Achieng Oneko, a Luo and an associate of Oginga Odinga, should join Mbiyu Koinange in London in one last attempt to save talks with the British government on the land question. In response to the open leadership challenge posed by the militants who had now taken charge of the Nairobi KAU office, Kenyatta assumed a more radical stance. In August 1951 he made a speech in Thika in which he declared that blood might have to be shed in the nationalist cause. He definitely appreciated that African politics was at a turning point and desperately attempted to revamp the KAU to position it for the coming challenges. At a meeting held in November 1951 (the first the KAU had held since 1949), Kenyatta was reelected president without opposition, a demonstration of the grip he still held over national politics. However, he had to use all his persuasive skills to stop Bildad Kaggia, one of the major radicals, from being elected secretary-general. He argued that to have two Kikuyus in the top leadership would give the organization the wrong image. On the whole, his plea to the radicals to retain the multi-ethnic nature of the executive and to refrain from making the KAU too much of a 'Kikuyu affair' went unheeded, and Tom Mbotela, Ofafa and Joseph Katithi lost their positions. The new national executive was both more Kikuyu and more militant. Bildad Kaggia, Fred Kubai, James Beauttah, James Njoroge and Charles Wambaa represented the militants. The rest of the committee consisted of Ex-chief Koinange, Jesse Kariuki, Jesse Kitabi (Kamaba), Jonathan Njoroge, J. D. Otiende as secretary-general (Luhya), Paul Ngei (Kamba), Peter Okondo (Luhya) and H. L. Nangurai (Masai). The stage was set for more radical politics. At the policy level the KAU under its new leadership was more

Jomo Kenyatta & the Rise of the Ethno-Nationalist State of Kenya

207

bold and assertive, and immediately declared its commitment to the independence of Kenya under African rule (Kyle, 1999: 50). A meeting held in Nyeri in July 1952, which was attended by over 25,000 people, demonstrated Kenyatta's personal hold over the Kikuyu masses but also the precariousness of the political situation in the country. The crowd was clearly in favor of more action and less talk. Soon thereafter the colony exploded and a state of emergency was declared. The declaration of a state of emergency on 20 October 1952 was a turning point in the history of Kenya. It confirmed both the inability of the colonial government to hold the colony together and therefore the need for direct imperial intervention, and the victory of the radicals over the moderates in shaping the future of African politics. The assassination of senior chief Waruhiu on 7 October was the long awaited excuse to declare the emergency. British intelligence persuaded the Governor that Kenyatta knew of the assassination and although he had counseled against it, Koinange and others had decided to proceed (Kyle, 1999: 52). The arrest, charge and conviction of Jomo Kenyatta for masterminding Mau Mau were farcical to say the least and a great travesty of British justice. There can be no doubt that Kenyatta did not control Mau Mau. Indeed, there is evidence that the central committee of Kubai, Kaggia, Mutonyi and Gathanju did not trust Kenyatta. At some point it even summoned him to warn him to stop the denunciation of Mau Mau, which he was undertaking at public meetings at the behest of the colonial authorities and African chiefs (Kaggia, 1970: 113-14). While Kenyatta was jailed together with several non-Kikuyu KAU operatives like Achieng Oneko and Paul Ngei, this did not, however, detract from the fact that the focus of the entire judicial farce was to legitimize a blow at Kikuyu nationalism. A discussion of the several levels of conflict within Mau Mau and the impact of this on Kikuyu society lies beyond the scope of this chapter. Suffice it to say that Mau Mau presented Kenyatta with yet another dilemma, the genesis of which lay in earlier controversies among the Kikuyu community. This was the expanding gap between the Christian, Westernized middle class and the rural peasants and squatters. From the days of the KCA Kenyatta's sympathies lay with the former; Mau Mau brought home to him the futility of challenging the latter, if he wanted to remain a credible leader of the Kikuyu people. Kenyatta was therefore a reluctant Mau Mau. Yet Mau Mau presented another long-term problem for the Kikuyu political leadership and for Jomo Kenyatta in particular. The colonial government had waged an intense and successful propaganda campaign to alienate other communities from the Kikuyu. The oathing ceremonies were depicted as primitive and atavistic and the Kikuyu as a people were depicted as dangerous to the rest of society. This was baggage that Kenyatta would have to carry with him into national leadership. The declaration of the emergency and the detention of almost all the leaders of the KAU did not lead to its immediate demise; in fact it lasted until June 1953. For some months, the government permitted its existence in the belief that the absence of the Kikuyus would make it less radical. A provisional executive was appointed. F. W. Odede was appointed acting president. Joseph Murumbi acting secretary and W. W. W. Awori acting treasurer. They immediately reiterated the KAU's previous demands and further demanded the immediate release of all those arrested under the emergency regulations. This confirmed that the core objectives of the KAU had support among leaders who

208

Githu Muigai

were non-Kikuyu. It did not, however, prove that the KAU could operate without Kikuyu patronage and influence. While Kenyatta languished in jail and the full might of the imperial army was brought to bear against the rag-bag Mau Mau army, more significant changes were taking place on the constitutional front. The first African elections were held in 1957 under the Lyttelton multiracial Constitution, which had come into force in March 1954. For the first time, there were African politicians in the Legislative Council who could truly claim to represent the interests of the African majority. Daniel Moi represented Rift Valley, Oginga Odinga, Nyanza, Bernard Mate, Central (Kikuyu and Meru), James Nzau, Ukambani, Masinde Muliro, Western and Ronald Ngala, Coast. By colonial design, all the constituencies represented by the new members were essentially ethnically based. All nationwide political parties had been banned alongside the KAU in 1953. Since 1955, when the onslaught on the Mau Mau army appeared to be working, the government had began to encourage district-wide and therefore ethnic-based parties in what was obviously a deliberate effort to undermine African nationalism (Muigai, 1995: 165). Consequently, of all the elected leaders only Tom Mboya could claim to have been elected on a multi-ethnic ticket in Nairobi. Surprisingly for the colonial government, the African elected leaders rejected the proposed multi-racialism of the Lyttelton Constitution and supported African nationalism. With the exit of the Kikuyu from the national political stage, due to the banning of the KAU and the subsequent detention of its leadership, the Luos and Luhyas moved into the political center-stage. Oginga Odinga and Tom Mboya were the most eminent of the new leaders. Odinga came from a Local Native Council background, and Mboya from a trade union background. The rise of trade unions as key vehicles for ventilating African grievances in the absence of political parties was a significant development of the time. Both Odinga and Mboya were to play a crucial role in the construction of the Kenyatta State. The demand for Kenyatta's release was perhaps the most dramatic achievement of the elected Africans. In June 1958, in the Legislative Council, Odinga spoke of Jomo Kenyatta and those detained with him as the 'political leaders' of the African people. Not to be upstaged, in September the same year Mboya called for the day the leaders were arrested to be observed as a day of fasting. The legendary rivalry between Mboya and Odinga, which was to continue until the former's death, was about to erupt. Inevitably, a major political rift soon emerged among the African leaders. Some of those who thought they had a better chance without Kenyatta were vehemently opposed to his leadership being linked to future political and constitutional reforms. The majority saw Kenyatta's stature as a leader and ex-political prisoner as an asset in the nationalist struggle and were equally opposed to any independence talks without him. Kenyatta's years of positioning himself were paying off.

The Demise of the Colonial State and the Search for a Nationalist Alliance From 1957 African nationalism had grown under very difficult circumstances. There was no national party that could enforce discipline. Personality and leadership conflicts were the order of the day. The Constituency Elected Members Organization (CEMO) which had been formed as an umbrella body to press for

Jomo Kenyatta & the Rise of the Ethno-Nationalist State of Kenya

209

a constitutional conference soon collapsed under the weight of the squabbles. Efforts to present a united front in the pre-independence negotiations were unsuccessful. In July 1959 the Kenya National Party (KNP) was formed with the support of eight of the fourteen African elected members led by Muliro, Ngala, Moi and Towett. A month later Odinga, Mboya and Kiano formed the Kenya Independent Movement (KIM). The polarization of African politics was complete. Yet the polarization in the politics of Kenya was not confined to the Africans. It was equally bad among the European settlers. Two groups had emerged. There were the conservatives who formed the Convention of Associations (CA), made up of small-scale farmers and recent settlers, and who were initially led by Group Commander Llewelyn Briggs and later by Sir Ferdinald Cavendish Bentick. Their agenda was simple; to maintain the status quo. The Liberals, mainly among the professional and business groups, were led by Sir Michael Brundell and were organized under the New Kenya Group (NKG). They were for multi-racialism and worked at broadening their appeal by cultivating the African middle class. The settlers, both liberal and conservative, sought to marginalize Kenyatta by forming a government without him. Inasmuch, therefore, as the political history of Kenya had been the history of adjusting the claims of the various racial and ethnic groups to power and influence, the politics of decolonization followed the same pattern (Ghai and McAuslan, 1972: 25-33). When the first Lancaster House Conference was convened in 1960, the Africans tried one more time to present a united front. Ronald Ngala was appointed leader and Tom Mboya secretary of the African delegation. This alliance proved short-lived. As soon as the delegation returned home the cleavages endemic in African politics re-emerged. Many scholars and commentators have explained the basis of these cleavages as 'radicalism' and 'conservatism' (Leys, 1975: 58). This, in our view, overstates the case. The differences were probably less ideological and more personal. The new political organizations were built on the groupings already formed in 1959. James Gichuru, Tom Mboya and Oginga Odinga formed the Kenya African National Union (KANU) in May 1960. It was understood that Gichuru would be president until Kenyatta was released (Odinga, 1967: 194) Ronald Ngala was elected in absentia as treasurer and Daniel Moi as assistant treasurer. The idea appeared to be to build a broadbased coalition of ethnic leaders to facilitate independence without any hitches. In the event, both Moi and Ngala declined to join KANU. History now records that they were under tremendous pressure from the white settlers, who were the ideologues behind the Kenya African Democratic Union (KADU), to distance themselves from KANU. In the short run, Moi went on to form the Kalenjin Political Alliance (KPA) which had no pretensions to being a national party. Its immediate action was to lay claim to all land in the western highlands. Ngala, for his part, proceeded to form the Coast African People's Union (CAPU). The fragmentation did not stop there. Masinde Muliro formed the Kenya African Peoples Party (KAPP). The Maasai United Front was also set up to represent the Maasai and the Somali National Association was formed to represent the Somalis. In June 1960 a month after KANU was established the five ethnic-based organizations united to form the Kenya African Democratic Union (KADU), with Ngala elected president, Muliro vice president and Moi chairman. Paul Ngei, who felt that he had not been accorded due recognition within KANU and who did not wish to join KADU, went on to form the African Peoples Party (APP) through which he sought to mobilize the Kamba people

210

Githu Muigai

and negotiate a place for himself in the post-independence power structure. The real basis for ethnic politics had been laid. In terms of its ethnic profile, KANU brought together the big ethnic groups of the Kikuyu (including their cousins the Embu and Meru) and the Luo. This alliance was at best an uneasy one. KANU had been formed at a time when many Kikuyu leaders, including Kenyatta, were still in prison, and while Luos and Luhyas dominated the political scene. Among the new KANU leaders only James Gichuru was a Kikuyu. This was a matter of grave concern to the Kikuyu leadership who worried that they might be marginalized within the new party. Their fears were not unfounded. For example, Dr Njoroge Mungai, Kenyatta's nephew, was not elected to any office, although he had been nominated for all except the presidency (Kyle, 1999: 116). The Kikuyu leadership particularly resented the election of Tom Mboya to the post of Secretary-General and in order to contain him they plotted to create the post of organizing secretary to be filled by one of their own. Whatever grievances the Kikuyu had within KANU, it was, as far at its opponents were concerned, a Kikuyu and Luo party. KADU, on the other hand, brought together the smaller, mostly pastoral ethnic groups. Since its inception it had been under great settler influence. Much of its ideology, such as there was, was supplied by settler ideologues. Indeed, for some time, KADU labored under the belief that it could form a multiracial government together with the small political groupings of European settlers and Asians (Odinga, 1970: 197). In reality, neither KANU nor KADU was a genuine grassroots political party. Both were loose alliances of the pre-existing district political organizations, which had been fostered with great care by the colonial government as vehicles of ethnic sub-nationalism. At another more important level, they were coalitions of ethnic elites. The masses of the people were mere pawns in the game of ethnic politics between the power elites. Their rhetoric notwithstanding, there were no major ideological differences between the two parties. In reality ethnicity became the defining issue. The second attempt at territorial nationalism was therefore no more successful than the first. In 1961 the two parties faced each other in the first general elections. Predictably, outside Nairobi ethnic loyalties dominated the elections. Interestingly, this was not the case in Nairobi. Tom Mboya, for example, was able to beat Munyua Waiyaki in a Kikuyu-dominated constituency in Nairobi. The electoral system produced strange results. While KANU received 67.4 per cent of the votes and won 19 seats, KADU received 16.4 per cent of the votes and won 11 seats. KANU refused to form a government without Kenyatta, who, while no longer in prison, was still under restriction in Lodwar. KADU formed a coalition with the settler-based NKP and formed a government that lasted for the next eleven months. Kenyatta was finally released in August 1961. His attempts to mediate a truce and possible merger between KANU and KADU proved futile. Why he wanted a single party for the Africans at this time is an important issue. In all probability he was more concerned at this point that several parties might jeopardize the constitutional negotiations and thereby delay independence, than he was that ethnic sub-nationalism might threaten national unity. Having finally accepted KANU's presidency and through a bi-election joined the Legislative Council, luckily for him he did not have to make difficult choices at this stage. He found that prospects for a multiracial government had long disappeared and KADU's support from the settlers had waned. The mood was right for a more aggressive

Jomo Kenyatta & the Rise of the Ethno-Nationalist State of Kenya

211

nationalist political platform. He nonetheless remained a master schemer. With his eyes always on the goal, he was willing to make short-term concessions. In 1962, he led KANU into a coalition government with KADU, which lasted until independence in 1963. But not surprisingly, this joint government did little to bring the two parties closer. Indeed, after independence KANU proceeded to form a government on its own, with Jomo Kenyatta as Prime Minister.

Kenyatta's State - Ethnic Coalitions and Ethnic Nationalism The most immediate task facing Kenyatta and KANU was how to address the crisis of expectations generated by independence. So much depended on how this was done. Several forces were in play. On the one hand, Kenyatta had to address the immense expectations of the Kikuyu masses who had borne the brunt of colonial occupation and who formed the core of his and KANU's support. Unfortunately for him, Kikuyu society was not homogenous. Several factions existed, the most prominent being the 'royalists' and the Cex-MauMau', both of whom expected Kenyatta's support and reward. Elsewhere, Kenyatta was called upon to deal with the legitimate expectations of the various other factions of Kenya's ethnic sub-nationalist elites who had supported KANU's bid for power and whose continued support was necessary for the survival of the regime. In order to cement Kikuyu sub-nationalism, which he saw as the basis of his national support, Kenyatta began to cultivate the image of the Kikuyu paramount chief. He was already married into the families of Chief Koinange and Chief Muhoho, thereby giving him wide acceptance as part of the Kikuyu aristocracy in Kiambu. He now proceeded to cultivate with care the loyalty and allegiance of the sons of former colonial chiefs. Mbiyu Koinange, the son of Chief Koinange, became his chief political adviser. Charles Njonjo, son of Chief Josiah Njonjo, became his Attorney General and one of the chief architects of the Kenyatta State. He also courted Dr Munyua Waiyaki from the large Waiyaki family and in fact used him to negotiate with the last remnants of Mau Mau fighters in the forests. Later he was to court Arthur Magugu whose father was a chief in Githunguri. Unfortunately for him, his success in establishing himself as the undisputed chief of the Kiambu Kikuyu, created a separate problem. He alienated the Nyeri, Muranga, Ndia and Rift Valley Kikuyus who felt marginalized in the new configuration of power. This was a problem that Kenyatta never truly solved in his lifetime. His response was to try to establish linkages through leaders he could control. For example, he set up Julius Kiano as his link with the Kikuyu in Muranga and Mwai Kibaki as his link with the Nyeri Kikuyu. His kitchen cabinet, however, remained largely drawn from the group that came to be known as the Kiambu mafia or the inner court (Leys, 1975: 246). For all his political acumen, Kenyatta never really achieved complete political autonomy from this group. Indeed, in his last years this group largely ran the state without reference to him. At another level, Kenyatta's alliance with former collaborators within the Kikuyu community, such as Eliud Mahihu, Isaiah Mathenge and John Michuki, to whom he entrusted the provincial administration, seriously alienated the ex-Mau Mau and their sympathizers. This opened another cleavage that he never quite sorted out. Until his death many of the Kikuyu poor considered that Kenyatta had betrayed Mau Mau. His inability to deliver free land to the Kikuyu peasantry only reinforced this view. His support of the million acre scheme and other land-buying efforts was perceived as

212

Githu Muigai

support for the Kikuyu 'haves' against the 'have nots'. Tragically for him, among other less affluent ethnic groups the land-buying programs were perceived as partisan in support of the Kikuyu. Kenyatta was convinced that, in order to present himself as the leader of Kenyan nationalism, he had to foster a secure Kikuyu ethnic sub-nationalism. To secure the latter he was willing to jeopardize the former. Herein lies history's great injustice to Kenyatta. For once in his political life he could not ride both horses. In no area was his partisanship as evident as that of the distribution of civil service jobs and land redistribution. Of all the offerings on the carcass of the colonial state the most important was land. Unfortunately for Kenyatta, it was not only the Kikuyu who were hungry for land. The Kalenjin, the Luhya and the Luo were similarly affected. Indeed, the fear of a mass migration of the Kikuyu into the white highlands had partly inspired KADU's federalist demands. On this most critical issue, Kenyatta put Kikuyu interests first. In order to contain the sub-nationalism of the other communities and provide legitimacy for his regime, Kenyatta set up an elaborate patron-client system, with himself as the chief patron. From the various communities he identified and politically supported individuals who owed personal allegiance to him. On the other hand, he systematically eliminated from the political field those who were opposed to him or his proteges. His complete domination of the civil service and the provincial administration made this easily possible. The demise of KADU in 1964 was a godsend in this regard. Within a single party, which was the sole avenue to legitimate political activity, Kenyatta was able to determine who remained politically relevant and who did not. No politician had any doubts about this. Having made 'ethnic arithmetic' the linch pin of his regime's legitimacy, Kenyatta played the ethnic game very shrewdly. For example, in recognition of the significance of the Luo as the second most important block within KANU, he appointed Oginga Odinga as vice-president. Tom Mboya, on the other hand, was appointed to the high-profile office of Minister of Justice and Constitutional Affairs. In the Rift Valley Kenyatta had identified Daniel Moi as an ally even before the demise of KADU in 1964. Moi was instrumental in containing Kalenjin objections to the resettlement of the Kikuyu in the Rift Valley, a key plan in Kenyatta's redesign of the state. As a quid pro quo, Kenyatta threw his weight behind the Kalenjin in their acquisition of land in Transzoia and Uasin Gishu, in opposition to the Luhya. This helped Moi to eclipse Towett and Seroney as the one politician who could claim true leadership of the Kalenjin (Throup, 1987: 46). When Moi was finally appointed vice-president in January 1967, after Odinga had been jettisoned within KANU, he brought with him Kalenjin support for the ruling coalition. The Kalenjin had replaced the Luo in the shifting coalitions that made the Kenyatta state, the stability and legitimacy of which depended on the existence of these coalitions. Consequently, the state was a continuously shifting series of coalitions both within and outside Kikuyuland. Several factions nonetheless existed within these coalitions. To maintain his dominance of the state, Kenyatta had therefore to play certain Kikuyu factions off against each other and to incorporate non-Kikuyus into his coalition (Throup, 1986). The collapse of KANU as a nationalist alliance, within which ethnic claims could be resolved, and the formation of the Kenya People's Union (KPU) as a predominantly Luo party in 1966, were a major challenge to the stability of Kenyatta's state. The ethnic dimension overshadowed the ideological and per-

Jomo Kenyatta & the Rise of the Ethno-Nationalist State of Kenya

213

sonality fights between KANU 'A' and 'B' that had been simmering since KADU joined KANU. The collapse of the Kikuyu-Luo alliance that had been the backbone of KANU and the rise of an entirely new alliance between the Kikuyu and the former KADU communities illustrate the highly fragile ideological basis of the respective groups. For example, Moi's Kalenjin supporters, hitherto die-hard KADU supporters, were instrumental in carrying out Kenyatta's machinations at the Limuru conference that resulted in the humiliation of Odinga and his subsequent exit from KANU (Leys, 1975: 224). After this, not surprisingly Moi was appointed vice-president and Ronald Ngala, the former chairman of KADU, entered the cabinet. The KPU, which was formed in April 1966, was a loose coalition of KANU 'B' 'radicals' and trade-union leaders. Although the party was supported by a fifth of the existing Members of Parliament, it was widely perceived as a Luo party. Kenyatta and his cohorts, using the state media, waged a deadly and highly effective propaganda war against the KPU. In addition, Mboya and Njonjo then engineered constitutional amendments that virtually emasculated the infant party by punishing defectors (Muigai, 1992). After the 'little general election' of 1966, the KPU had only 9 Members of Parliament, all of them from Nyanza. By all accounts the election had been neither free nor fair, especially in Kikuyu country where Kenyatta had employed extensive state resources and exerted tremendous ethnic pressure. Whenever Kenyatta's grip on the nation was threatened, he retreated into the safety of the ethnic cocoon provided by the Kiambu-Gatundu group. Several incidents illustrate how much he came to rely on the advice and support of his inner circle. The first was the assassination of Tom Mboya in 1969. The second was the oathing of the Kikuyu in the same year. The third was the assassination of J. M. Kariuki in 1975 and the fourth was the rise of the GEMA-inspired Change the Constitution group. Whatever may have been the intention of the people who assassinated Tom Mboya, his death was extremely disruptive of the last modicum of ethnic harmony within KANU 'A'. Between independence and the fall-out with Odinga and the 'radicals' - the so called KANU 'B' - in 1966, Mboya had become Kenyatta's most important non-Kikuyu confidant. Kenyatta used Mboya to deal with the ideological problems posed by the Odingas, Onekos, Kaggias and Kalis. Mboya had been so successful that he became the victim of his own success. He was perceived as posing a threat to the presidency. His ambivalent relationship with the Kiambu-Gatundu group is evidenced by the change introduced into the Constitution to raise the minimum presidential age to 40, as Mboya was 37 at the time (Goldsworthy, 1989: 270). The assassination of Mboya in 1969 was the first serious manifestation of the inherent weakness of the Kenyatta state and the fragile nature of the ethnic alliance that was KANU. Ironically, Mboya was deemed to be a threat precisely because he appealed to Kenyan national solidarity. He had won election consistently in a constituency where the Kikuyus outnumbered the Luos. That he had been killed by a Kikuyu gunman, apparently with the connivance of the authorities, drove a deep wedge into the Kikuyu-Luo alliance, which was never healed. The oathing of the Kikuyu in 1969 in the aftermath of Mboya's death, with Kenyatta's knowledge and connivance, was perhaps one of his worst political mistakes. The oathing was justified to the ordinary Kikuyu folk on the basis that they had to be prepared to protect the presidency in the 'house of Mumbi'. Politically, the oathing achieved nothing. Yet it served to unmask Kenyatta as an opportunistic and shameless manipulator of ethnic sentiment. It also served to

214

Githu Muigai

create even greater suspicion against the Kikuyu among other ethnic groups in Kenya. The KPU was banned in the same year. The Luo nation had therefore suffered a double tragedy. The 1970s were to witness greater efforts on Kenyatta's part to retain control at all costs. The formation in 1973 of the Gikuyu, Embu and Meru Association (GEMA), with Kenyatta's consent, had a two-pronged mission: to strengthen the immediate ethnic base of the Kenyatta state by incorporating the Embu and Meru into a union with the Kikuyu, and to circumvent KANU's party apparatus in the mobilization of political support among these groups. GEMA was therefore confirmation that KANU was no longer deemed a viable vehicle for the consolidation and expansion of Kikuyu hegemony. While posing as a cultural organization, GEMA all but replaced KANU as the vehicle for all political activity for most of the Kikuyu power elite. The immediate result of the rise of GEMA was that almost all other ethnic groups formed 'cultural groups' of their own. For example the Luo formed the powerful Luo Union, while the Kamba formed the New Akamba Union. The facade of nationalism within KANU had broken down irretrievably. KADU-style federalism was back in vogue (Widener, 1992: 118). The assassination of J. M. Kariuki, a vocal Kikuyu critic of Kenyatta, in 1975 showed further serious cracks in the Kenyatta state. J. M. was assassinated because he attempted to subvert Kikuyu sub-nationalism from within (Throup, 1987). An ex-Mau Mau detainee, a great orator and a great populist, Kariuki appealed directly to the Kikuyu peasantry and was able to rouse considerable opposition to the policies of the Kenyatta government, especially over the equity of land redistribution. He was seen as drawing attention to the hegemonic position of the Kiambu-Gatundu group and the privileged position of the Kikuyu capitalists. He personified Kenyatta's worst nightmare - opposition from within. The death of Kariuki further served to drive a wedge between the Nyeri Kikuyu and Kenyatta's immediate circle. On the other hand, it also served as a lesson to all politicians who might otherwise have attempted to forge cross-regional coalitions based on common economic problems (ibid.: 111). The Change the Constitution group was a GEMA-inspired plot to block vicepresident Daniel Moi from acceding to the Presidency on Kenyatta's demise, which at the time appeared imminent. The intention was to preserve Kikuyu hegemony and by extension preserve Kikuyu capitalism (Throup, 1987: 52). With Kenyatta fairly sickly, several factions within the ruling class were jostling for power. Kenyatta's close circle and some members of his family seemed naively to believe that if Moi was stopped in his bid to become president after Kenyatta's death, then it would be possible to organize for one of their own to assume power (Ochieng and Karimi, 1980). It is hard to tell the extent of Kenyatta's involvement with the group. It is, however, inconceivable that such a vocal group could have emerged without some form of tacit approval from Kenyatta himself. Yet the fact that the movement did not get far, must also mean that Kenyatta at some point decided against the idea. This was vintage Kenyatta. He played both sides for as long as he could. At some point, he must have decided that a non-Kikuyu president could best safeguard the future of his family and his cronies. He instructed his Attorney General to stop any further public discussions on the succession issue. In this way he partially redeemed himself, but the damage was done. The GEMA challenge to a Moi presidency had the unintended consequence of causing Moi to seek ethnic alliances across the country in much the same way as Kenyatta himself had done, thereby setting the tone for the Moi state.

Jomo Kenyatta & the Rise of the Ethno-Nationalist State of Kenya

215

By the time Kenyatta died in August 1978 he had not been in control of the state for a number of years, and the struggle for its control among the power elite threatened its very stability. The Kikuyu power elite were faced with the prospect of winning all or losing all. This may be one explanation of why the Njonjo-Kibaki axis saw the imperative of going along with a peaceful transition to a Moi presidency. It must have been understood at the time that this was to be the case as long as Moi toed the line. Little wonder that Moi's avowed 'philosophy' became one of Nyayo, following in Kenyatta's footsteps. Unbeknown to his assumed handlers, Moi had his own plans on how to put in place a new ethnic configuration - in Kenyatta's tradition.

Kenyatta's Legacy Kenyatta's legacy was a divided one. On the one hand, he was able to hold together the state, while almost all the post-independence African governments were toppled by military coups, and civil wars were waged relentlesly. Comparatively, Kenyatta's state was reasonably stable and prosperous. By his personal charisma and political genius, he oversaw the transformation of colonial institutions into indigenous institutions and the Africanization of commerce and agriculture. He cultivated national pride and patriotism and helped to decolonize the minds of the Kenyan people. On the other hand, he totally failed to address adequately the problem of ethnic politics. In fact, he served to entrench ethnicity as the dominant basis of political mobilization. By his complex web of ethnic coalitions he created a false sense of both nationhood and political stability, which he carefully grafted onto simmering ethnic tensions based on unfulfilled ethnic claims to power and resources. This illusion of stability is still in place today, while the underlying ethnic discontent remains unresolved. Furthermore, his autocratic grip on power and his complete domination of politics served to conceal the growing significance of ethnicity as the basis of political identity. In this most basic sense, Kenyatta failed in his stated goal of creating a sense of nationhood. Inasmuch as he made reasonable efforts, history must judge his success as dismal. He no doubt approached the problem of nation-building with a somber mind. His experience in the KAU had demonstrated how difficult it was to suppress or contain the sub-nationalism of all of Kenya's ethnic groups. But his efforts to retain the broad ethnic alliance within KANU without losing Kikuyu support must have left him completely frustrated. Increasingly therefore he sought to subjugate and control ethnic sub-nationalism by using the apparatus of the state. In the end, and perhaps in spite of his better judgement, Kenyatta settled into the easy role of Mzee (Old Man) which could mean many things to many people, as opposed to Baba wa Taifa (father of the Nation), which is what he had always aspired to be. From this position he was both the undisputed paramount chief of the Kikuyu and the chief patron of the Kenya nation, to whom all the leaders of the ethnic groups, but not necessarily the people themselves, owed allegiance. Paradoxically, therefore, while largely beholden to the Kikuyu power elite, Kenyatta still endeavored, and to some extent managed, to extend his neo-patrimonial system to almost all leaders of Kenya's ethnic sub-nationalist movements (Throup and Hornsby, 1998: 23). There were a number of weaknesses in Kenyatta's strategy of dealing with the force of ethnic sub-nationalism. One was that, in the absence of an individual

Githu Muigai

216

with Kenyatta's personal charisma and historical claim to power, the issues of ethnic equity long swept under the carpet of the authoritarian state, would sooner or later re-emerge with perhaps greater intensity, for what Kenyatta had done was to postpone and not to solve the problem (Muigai, 1995). The second lay in the fact that Kenyatta's model would operate only for as long as the state was authoritarian and dictatorial, dominating civil society and denying it all autonomy. The single-party state therefore afforded the ideal environment, with its demise, the issues would resurface. In all probability, the organization of the new political parties would be based on ethnicity and some ideology of ethnic nationalism. A third weakness is the fact that the economy needed to keep growing at a rate that outstripped population growth and was therefore able to accord a certain measure of national prosperity, which in turn would provide a modicum of political stability and social contentment. In the absence of viable economic opportunities for individuals and groups, the stability of the political order would be jeopardized alongside all its assumptions. A fourth weakness lay in the fact that the neo-patrimonial political structure, through which patronage in the form of jobs, loans, development schemes, etc., flowed from the leaders of ethnic groups in return for political support, could only function for as long as the state dominated most economic activity. Fifth, Kenyatta's model would work only for as long as the power elite received the co-operation of the factions controlling the dominant economic activities in the country. This, as Moi was to find out, could never be assured at all times. Finally, the model never foresaw the changing nature of ethnic identity and solidarity. Education, inter-marriage, migration and other social forces would over time undermine the capacity of ethnic political elites to purport to negotiate on behalf of entire groups whose interests would often conflict. Herein lay the real weakness of Kenyatta's legacy, and it could provide no serious long-term answer to the problem.

References Berman, Bruce. 1990. Control and Crisis in Colonial Kenya; the dialectic of Domination. Nairobi: EAEP; London: James Currey; Athens, OH: Ohio University Press. Brown, Jeremy Murray. 1972. Kenyatta. London: George Allen & Unwin. Ghai, Yash and J. P. W. B. McAuslan. 1970. Public Law and Political Change in Kenya. Nairobi: Oxford University Press. Goldsworthy, David. 1989. Tom Mboya; The Man Kenya wanted to Forget. Nairobi and London: Heinemann Educational Books. Kaggia, Bildad. 1970. Roots of Freedom. Nairobi: EAPH. Kenyatta, Jomo. 1938. Facing Mount Kenya. London: Seeker and Warburg. — 1942. My People the Kikuyu and the Life of Chief Wangombe. London: United Society for the Propagation of the Gospel. Kyle, Keith. 1999. The Politics of the Independence of Kenya. Basingstoke: Macmillan. Leys, Colin. 1975. Capitalism and Undervelopment in Kenya, The Political Economy of Neo-Colonialism. Nairobi: Heinemann Educational Books. Macharia, Rawson. 1991. The Truth about the Trial of Jomo Kenyatta. Nairobi: Longman. Muigai, Githu. 1992. 'From the Governor to the Imperial President: Constitutional Transformation and the Crisis of Governance in Kenya'. Ford Foundation Governance in East Africa Research paper. Muigai, Githu. 1995. 'Ethnicity and Democracy in Kenya' in H. Glickman, ed., Ethnic Conflict and Democratization in Africa. Atlanta, GA: ASA. Nyong'o, P. Anyang'. 1986. 'Struggles for Democracy: State and Popular Alliances in Africa', Eastern African Social Science Research Review, pp. i-x. Ochieng, Philip and Joseph Karimi. 1979. The Kenyatta Succession. Nairobi: Transafrica. Odinga, Oginga, 1970. Not Yet Uhuru. Nairobi and London: Heinemann.

Jomo Kenyatta & the Rise of the Ethno-Nationalist State of Kenya

217

Rosberg, Carl and John Nottingham. 1985. Nationalism in Colonial Kenya, the Myth of Mau Mau. Nairobi: Transafrica. Throup, David. 1987. 'The Construction and Deconstruction of the Kenyatta State' in M. G. Schatzberg, ed., The Political Economy of Kenya. New York: Praeger. Throup, David and Charles Hornsby. 1998. Multi-Party Politics in Kenya. Oxford: James Currey; Athens OH: Ohio University Press. Widner, Jennifer. 1992. The Rise of a Party-State in Kenya; From Harambee to Nyayo. Berkeley, CA: UCLA Press.

13

MAMADOU DIOUF

Between Ethnic Memories & Colonial History in Senegal: The MFDC & the Struggle for Independence in Casamance Translated from the French by Jonathan M. Sears

F

or the past twenty years, Senegal has faced an armed struggle for independence in the southern region of the Casamance, a struggle led by the local Joola (or Diola) ethnic group. Leaders of the Joola, and of some other local ethnic groups, claim to be ignored or oppressed by the central government, which is dominated by the northern Wolof group. Since the outbreak of the rebellion in 1982 and the reactivation of the Democratic Forces Movement of Casamance (MFDC), these new divisions and designations reflect a constant concern of the Senegalese government: to avert 'war' and to accord this major crisis the status of a 'problem', a crisis manageable at the regional level, and of no consequence at the national level. Treating the worst crisis that Senegal has known since independence as a regional and not a national crisis seems to be an approach broadly shared by the political actors, public opinion, and the Senegalese media. This contains the Casamance crisis in an isolated region, little known by the 'nordistes' (northerners, in particular the Wolof and the Haalpular). It also avoids its being exploited politically by the opposition or their trying to find solutions in a zone where persistent defiance is manifested by votes against the party in power. The opposite camp favours mobilizing the Casamance 'elites' to intervene from the clientelistic networks of the regime, in favor of the ruling party. The presidential election campaigns of 1993 and February-March 2000 best illustrate the discourse about the Casamance in the Senegalese political arena. President Abdou Diouf began his re-election campaign in Ziguinchor in 1993, at a time when the insecurity had reached an intolerable threshold for the population, who were victims of a war waged with determination, constancy, and violence by the rebels and the army. The president-candidate's presence in the Casamance bore witness both to the state's consummate understanding of the situation, and to a reaffirmation of the authorities' firm commitment to guarantee peace and the territorial integrity of the Senegalese nation. All of the other candidates passed through Casamance to put forward their solutions, highlight their contacts with some presumably influential members of the rebellion, and accuse the Senegalese government of being unable to find a political solution to the crisis. Abdoulaye Wade, in particular, played this card insistently. Despite this, the question of the southern region has not avoided being cast as a local and regional one. After appearing at Ziguinchor or Bignona, the candidates returned to concerns of the center such as youth employment, democratic institutions, and especially the unbridled search for support from the marabouts of the 'pays du diggeV} the heart of Senegalese political power. The increasingly 1

The expression designates the Wolof regions of the groundnut Basin where the Mouride and

218

The MFDC & the Struggle for Independence in Casamance

219

lethal character of the rebellion's activities since the 1993 elections (mines, armed attacks on public transport vehicles, etc.); the increasing number of civilian, military and rebel victims; the increasingly blind repression by the Senegalese security forces; and the near-total paralysis of production activities and economic transactions 2 all amply justified a more muscular rhetoric during the electoral campaign of February-March 2000, and stronger denunciations of the management of the crisis by the regime in place. 3 An attentive reading of the press shows the trend of the positions taken by the candidates and their parties, and their analyses of and proposals for the Casamance. For the candidate Iba Der Thiam, 'there is a Casamance specificity . . . intellectuals of the region have exploited the situation to develop the idea of a Casamance where the autochthonous people are excluded. This idea has helped turn an economic, social, and cultural problem into a political problem.' 4 Professor Thiam's position has evolved considerably since his 1984 writing on Casamance 'separatism': we would be mistaken to see there the consequence of censure by government policy, and even less the expression of a supposed neglect of the Casamance . . . In 20 years, the government has done more for this part of Senegal than the different colonial systems did since 1645, in other words during 315 years.5 For his part, President Diouf continues to cloak himself in solemnity as the rigorous guardian of the republican ideal: the territorial integrity of a single and indivisible nation. He accompanies this with paternalistic manifestations of magnanimity, a spirit of forgiveness, and proclamations of his commitment to peace. During his day of campaigning in Ziguinchor, he promised, once reelected, to make peace in Casamance one of his first priorities. To help his candidacy and ensure a peaceful campaign in the southern region, the president went to Ziguinchor on 28 January 1998 to meet the leader of the MFDC, a Catholic priest, Abbe Augustin Diamacoune Senghor, while encouraging his government to make contact with the MFDC. Two meetings took place in Banjul, the Gambian capital, in December 1999 and January 2000. During the second meeting, Banjul II, the Senegalese government recognized the MFDC 'as an interlocutor in the search for a definitive solution to the Casamance problem', seeking to consolidate the constantly violated cease-fire accord of 8 July 1993. 6 Like Iba Der Thiam in the 1980s, Abdou Diouf forcefully rejected the idea that the Casamance had been neglected by the post-colonial state as a myth invented for political ends. In an account with a revealing title and subtitle, a former prefect who had served in the region set about the meticulous dismantling of the 'neglected Casamance' thesis.7 The President did not convert to Professor Thiam's new philosophy; rather (* contd) Tidjani brotherhoods developed, under the leadership of an incontestable general Khalif who gave political, economic, and social orders and electoral recommendations (diggel) to the disciples according to the circumstances. See M. C. Diop and M. Diouf, Le Senegal sous Abdou Diouf. Paris: Karthala, 1990, p. 47. 2 See 'Les Femmes et la Crise en Casamance', Wal Fadjri, 22 December 1999, which uses different available sources to paint the macabre picture of the victims of war. 3 One of the candidates for president of the republic did not hesitate to predict the disappearance of the Joola if the war continues. 'Iba Der au Forum Civil. Un risque genocide Diola', Wal Fadjri> 26 January 2000. 4 Ibid. 5 Iba Der Thiam, 'La Casamance en question', Le Soleil, 411,9 January 1984. 6 'Resolution' adopted at the close of the meeting in Banjul between the Government of Senegal and the MFDC. Sud Quotidien, 27 January 2000. 7 Abdourahmane Konate, Le Probleme casamangais. My the ou Realites ? Dakar, June 1993.

220

Mamadou Diouf

his speech has more nuance. It shares Thiam's argument on the economic and social nature of the crisis. The President renews his commitment to 'discuss everything with the Democratic Forces Movement of Casamance', with the unavoidable and non-negotiable obligation to recognize that 'Senegal [is] one and indivisible . . . and its territorial integrity could never be sold'. 8 Further, he also says he understands 'the retreat into one's own identity' of the Casamance, brought on by globalization which forces the regions of the world, according to him, 'to live for themselves'. His plan is to 'return to the Casamance its dignity' through 'total de-mining' of the zones infested by anti-personnel mines, 'the insertion of the rebels' into economic life, and 'the establishment of industries', and 'to undertake these initiatives with appropriate urgency'. 9 The other candidates took up the dual thesis of rejecting the idea of the Casamance's neglect and the non-recognition of its distinct character in the processes of Senegalese state formation and of imagining the 'master fiction* of the Senegalese nation. The same proposals were put forward; most of them related to improving the region's economic and political integration into the Senegalese plan. Opposition leaders denounced the government's inability to find solutions other than repression, 10 or ineffective palliatives such as administrative decentralization and regional autonomy. 11 Furthermore, they also claimed to be the only group endowed with the personal skills and political and economic resources needed to end the war quickly.12 This survey allows us to highlight certain constants in the discussion of the Casamance problem. The treatment it has received remains local and regional, not national. This approach opens up a host of possibilities for political and clientelistic manipulation, and for claiming supposed capabilities to resolve the crisis, mainly on the basis of Casamance integration into political networks and economic circuits, through forms of integration that do not clash with the system. Nevertheless, by synthesizing the outcomes of the different meetings between the Senegalese government and the MFDC - in particular Banjul I and II - and the positions of the different political actors, we see a slight shift in the treatment of the crisis towards the center of the Senegalese scene: even if, as the Senegalese journalist Moussa Paye has noted, this shift signals danger for national cohesion. Evaluating the results of the most recent negotiations, he states: Are decentralization and regional autonomy a response to the demands of the MFDC? Surely not, unless the Senegalese State revises its constitution and admits the possibility of particular organizations campaigning for election in a regional context . . . Thus one of the important conditions could be met: freedom of speech and movement for the MFDC - another euphemism for its legal recognition. An agreement between the Senegalese State and the rebellion is inadmissible except through this plan for a large sector of the Senegalese population. For a particular status to be given to the Casamance just because some of its sons took up arms would be setting the wrong example.13

The balance between local and regional treatment and national treatment is not the only reasoning that causes tensions and confusions; there is also the sub8 'Ziguinchor. Abdou Diouf demande a etre decharge apres un septennat', Sud Quotidien, 10 February 2000. 9 Ibid. 10 See the positions of the AFP candidate in Oussouye, Sud Quotidien, 8 February 2000 and in Ziguinchor, Sud Quotidien, 7 February 2000. 11 For example, the proposition of Mademba Sock, Wal Fadjri, 9 February 2000. 12 See the propositions of the URD candidate in Ziguinchor, Sud Quotidien, 7 February 2000. 13 Moussa Paye, 'Esquisses Senegalaises', Sud Quotidien, 26 January 2000, emphasis added.

The MFDC & the Struggle for Independence in Casamance

221

regional dimension, and the hesitancy and difficulties of Senegalese diplomacy, 14 notably with Guinea-Bissau 15 and Gambia. 16 This chapter is concerned neither with the genesis and chronology of the 'Casamance problem', nor with the motivations given by the actors, 17 but instead with the different 'ingredients' that feed the discourse of independence demands. It attempts to track very closely the production of a memory of suffering, of struggle, and of distinctness that appeals to demography, geography and history to legitimize separation from Senegal. This history is based on the distinct colonial moment and a post-colonial compromise that the Senegalese state has not respected. The approach taken here favours the identification of a discursive formation that supports a social and cultural 'imaginaire'18 produced by a specific memory. The interaction between the discursive formation 19 and social imaginaire, on the one hand, and the memory and the truth regime such a process initiates on the other, aims at the following: to create a reality disconnected from Senegal as it is invented, and to establish the Casamance as periphery, against the islamo-wolof nature of the Senegalese state, all the while aspiring to a modernity that avoids being accused of tribalism. The memory of Casamance identity thus produced is systematically dissociated from that of a single Senegal; an alternative history-in-the-making is establishing itself against the islamo-wolof history of the colonial and post-colonial state. Against the accommodation and complicity of the nordistes during the colonial period, this history opposes the secular resistance of the Casamangais. To the nordistes' cruelty and malice, the Casamangais show simplicity, honesty, and gentleness. Calling upon the idea that the Casamance was neglected 20 since independence, this history forcefully emphasizes the colonial nature of the 14 See the excellent dossier drawn up by Momar C. Diop, Le Senegal et ses voisins. Dakar, Societes Espaces Temps, 1994. 15 On the role of Guinea Bissau, see also J.-C. Marut, 'Le mythe. Penser la Casamance', pp. 19-26 and 'Les dessous des cartes casamancaises. Une approche geopolitique du conflit', pp. 193-211, in F-G. Barbier ed., Comprendre la Casamance. Chronique d'une integration contrastee. Paris: Karthala, 1994. Also 'Solution militaire en Casamance?', Politique africaine, 58, 1995, pp. 163-9 and 'Apres avoir perdu Test, la Guinee Bissau perd-elle aussi le nord?', Lusotopie, 1996, pp. 81-92. 16 See Arnold Hugues, 'L'effondrement de la Confederation de la Senegambie' and E. Sail and H. Sallah, 'Senegal and Gambia: The Politics of Integration', in Diop, Le Senegal et ses voisins. 17 Among the important studies on the Casamance see D. Darbon, 'Le culturalisme bascasamancais', Politique africaine 14, 1984, pp. 125-8, 'La voix de la Casamance . . . une parole diola', Politique africaine 18, 1985, pp. 125-38; and Uadministration et le Paysan en Casamance. Paris: Pedone, 1988; p. Geschiere and J. van der Klei, 'La relation Etats-paysans et ses ambivalences: modes populates d'action politique chez les Maka (Cameroun) et les Diola (Casamance)', in E. Terray (sous la direction), UEtat contemporain en Afrique. Paris: Karthala, 1987, pp. 297-340 and 'Popular Protest. The Diola of South Senegal', in P. Quarles van Ufford and M. Schoffeleers eds, Religion and Development Towards an Integrated Approach. Amsterdam: Free University Press, 1988, pp. 209-29); Diop and Diouf, Le Senegal sous Abdou Diouf, A. Diaw and M. Diouf, 'Ethnic Group versus Nation: Identity discourses in Senegal', in O. Nnoli ed., Ethnic Conflicts in Africa, Dakar: Codesria Book series, 1998, pp. 259-85. The most complete and up-to-date chronology is by F. de Jong, 'The Causes of Conflict in Africa. The Casamance Conflict in Senegal', unpublished, April 1998. 18 This notion is understood here in the sense given by Craig Calhoun, Nationalism. Minneapolis: University of Minnesota Press, 1997 as an imaginaire that is not opposed to reality but, by the complex historical processes thus created, imagines the world that corresponds to the demands and expectations of a group, a community and/or a society (chap. 2 in particular^. 19 See Michel Foucault, Uarcheologie du savoir. Paris: Gallimard, 1969, and VHistoire de la sexualite. Paris: Gallimard 1984, which shows the slide in Foucault's reflection from a strictly 'archeological' to a genealogical approach, which insists on discourse as a site of the production of reality. In 'La Poussiere et le Nuage', in M. Perrot ed., Vimpossible prison. Paris: 1980, he explains at length the 'statements' and rhetorical devices which are among the modes of the construction of reality. 20 Darbon, 'La voix de la Casamance'.

222

Mamadou Diouf

nordistesy migration into the region 21 , with the increasing soil aridity and barrenness of their homelands. 22 Its inhabitants have become subjects, and its resources the object of systematic looting that respects neither the sites dedicated to the memory of 'Joola identity' nor the shrines of the sacred forests. In the light of these considerations, we are trying to answer the following question: are not the geography and history mobilized by the M F D C regional precisely because they are intended to mask the primordial object of the claim that would necessarily be ethnic? To repeat the title of an article by Dominique Darbon, it is not to give voice to a multi-ethnic Casamance, but to the Joola. The Casamance discourse would be nothing other than a mask to cover up the affirmation of its radically Joola identity, thus removing the claim for ethnic and religious pluralism and constructing an essentialized Casamangais who is inevitably Joola.

An Unstable and Political Geography: The Territory against the Terroir 23 The approach adopted in this section contrasts the territory, which is of necessity national, and the terroir which is defined spatially, and which carries a sense of community and spirituality and a close, discriminatory group identity. This approach favours the colonial and post-colonial moments. Because of their genealogical relationship, both of these moments lay down a logic that integrates and registers the communities, their lands, and their history in the account, the imaginaire, and the territory of the colony and of the post-colonial state. This process of territorial integration has always attempted to subordinate all other logics to its own totalizing grammar. Faced with such syntaxes of authority, authorization and/or accommodation, the communities produce formulas of withdrawal, retreat, direct confrontation and feigned submission. These formulas combine to produce ideological texts of resistance, controlled assimilation or total submission. Conflicts described as ethnic in Africa are constructed within the unstable, interactive production of the national territory. For our purpose, Cosgrove's approach that the nation-state constantiy tries to promote a single identity within a single territory, and uses procedures that often encounter resistance in the regions where subaltern symbols and alternative versions of history are prevalent, provides an interesting framework for analyzing such a dynamic of antagonism.24 21 'The Senegalese are imperialists', according to Salif Sadio, one of the military leaders of the rebellion. Interview with Diane Galliot (1994). 22 See one of the foundational texts of the rebellion, the conference of Abbe Diamacoune at the Chamber of Commerce in 1980, and published as 'Message de la reine Aliin Siitowe Diatta ou hommage a la resistance casamancaise'. 23 In his 'Translator's note' to Michel de Certeau, Luce Girard and Pierre Mayol, The Practice of Everyday Life, Vol. 2, Living and Cooking. Minneapolis: University of Minnesota Press, 1998, Timothy J. Tomasik insists on the difficulty in translating the concept of terroir by writing: 'According to its etymology, terroir is rooted in the popular Latin territories, referring to earth, land, or soil, which is a change of territorium, referring more specifically to territory. Terroir is often employed in the context of food products that come from or have a flavor unique to a particular region.' Also in his 'Certeau a la Carte: Translating discursive Terroir' in The Practice of Everyday Life: Living and Cooking', SAQ, 100, Spring 2001, pp. 519-42), he writes '... Borges, in his celebrated history of impossible cartography, adduces that a terroir is a sort of movable mental and physical feast, a selection of geographic and sensory shapes that carries a strong affective charge. Terroirs...are sites where words and things acquire uncommon aroma, taste and gustatory savor', pp. 520-21. 24 D. Cosgrove, 'Then We Take Berlin: Cultural Geography 1989-90', Progress in Human Geography, 14, 1990, pp. 560-8, p. 564.

The MFDC & the Struggle for Independence in Casamance

223

The Casamance is a good illustration of a region's historical production in the context of colonial expansion, and in the construction of the Senegalese nationstate in 1960. Casamance geography, population, ethnic composition, the nature of its territory, and its administrative and political management are becoming controversial issues. This controversy has been expressed since 1983, on the one side, by the reactivation of the M F D C and the resort to armed violence to gain independence, and, on the other, by the strong desire to safeguard territorial integrity, the use of military and police forces included. The Casamance has always been constructed as a geographical entity in contrast to the northern region of Senegal, which we have described earlier as 'pays du diggeV. This entity is solidly based on the Senegalese islamo-wolof model imposed ever since the establishment of the colonial political economy, which inevitably serves as the reference point for Senegalese modernity or Senegalne, to use Leopold Sedar Senghor's word. Thus it creates a center and peripheries according to the degree of integration into the colonial and post-colonial political, social, and economic system. This contrast is admirably brought home by the images used by Abbe Augustin Diamacoune Senghor, General Secretary of the MFDC, the most provocative of which is that of the Senegalese baobab as compared with the Casamance palm tree.25 These strong images associate the destruction of the environment and Senegal's open spaces with inequality and violence as opposed to the 'harmonious symbiosis of the land's rich hummus and the fertile wind of the open sea'; 26 to the forest, the solidity, and the rituals and initiations of sacred forest shrines through which 'the Senegalese did not pass'; and to Casamance democracy and egalitarianism. The geography of Casamance is marked by controversies, not only about the boundaries of the region, the ethnic composition of its population and the stages in their formation and settlement in the region, but also the culture (s) that has (have) flourished there, and their degree of integration in the colonial and postcolonial images. Each of these aspects plays a particular role in the rhetorical devices and the discursive regime of truth of the parties involved. The partisans of the M F D C identify the Casamance as the area between the Atlantic Ocean to the West and the Faleme, an effluent of the Senegal River to the East: 'an intermediary zone between the Sahelian and the Guinean zone . . . a site of inestimable meteorological value, a strategic point of prime importance . . . a great crossroads where different migratory waves meet, mingle and mix . . . a hospitable land'. 27 The territorial claim thus expressed and its characteristics have many functions: to reject the accusation of 'ethnicism' and tribalism made against the M F D C , to specify the multi-ethnic and multi-confessional character - hence the insistence on the crossroads site and the notion of mixing - of Casamance independence demands, and above all to model the struggle with indisputably modern nationalist images. The militant and political geography of the M F D C , as it is produced in the texts of its main spokesman Abbe Augustin Diamacoune Senghor, has a history and a complex 25

Letter from Abbe Augustin Diamacoune Senghor to President Abdou Diouf, 12 May 1982. The other images are those of '. . . a Casamance stick [if 'submerged 2000 years in the lather of the'] Senegalese lion [will never become] a Senegalese caiman', ibid., and 'Casamance, Pays du Refus' (November 1994) by Abbe Diamacoune Senghor which is the response to the document developed by Jacques Charpy, General Conserver of Culture, 'Casamance et Senegal au temps de la colonisation francaise', (November 1993), in the context of French mediation, in which is included a M F D C declaration dated 21 December 1993. 26 Declaration, 1993. 27 Ibid.

224

Mamadou Diouf

network of symbolic, discursive and landmark features that abolishes the boundaries of ethnicity. The palm tree displays precisely the Casamance national image, and at the same time confers immediacy and radical difference from the Senegalese baobab with the attributes and images associated with these two trees. 28 While Abbe Diamacoune Senghor refers to 'the geographical, historical, and political Casamance', 29 Jacques Charpy puts forward the following geographical definition of the region: The territories of the Casamance today form two regions of the Republic of Senegal: the regions of Ziguinchor and Kolda. They take their names from the 300-kilometerlong river of the same name. Bordered to the West by the Atlantic Ocean, to the North by the Gambia, to the East by the Koulountou, an effluent of the Gambia, to the South by Guinea Bissau and Guinea Conakry, the Casamance covers 28,350 km2, or oneseventh of the Senegalese territory. Historians generally distinguish the upper and lower Casamance, separated by the Sougrougrou River, whereas geographers include the greatest part of the region in the 'Soudanese' zone, with only the shore and the lower rivers from the river systems of the South.30 It is interesting to stress the contrast between the two geographical propositions. Charpy's definition is, of course, rejected by the M F D C . The plural territories of the Casamance correspond to the M F D C notion of a single space, and retain the reference to the historians' emphasis on ethnic and cultural diversity, as if to validate its neutral and scientific character. This difference between the upper and lower Casamance is reinforced by geographical characteristics: the Soudanese zone (which recalls the Sahel), habitat of the Peul and Manding, and the rivers of the South, habitat of the Joola and other Balant. A vision of geographic, climatic, religious, and ethnic diversity, and economic pluralism is contrasted with the territorial unity and mixing claimed by Abbe Diamacoune Senghor. 31 Partisans of the Senegalese nation also claim to be objective; their geographical definition of the Casamance takes its contours from the colony of Senegal. The testimony of Abdourahmane Konate, former District Officer (Prefet) who served in the Casamance, is instructive on this subject. He considers the Casamance first of all 'from the geographical point of view' and sees that there are if not 'two, even three Casamances: the maritime Casamance (Ziguinchor region or lower Casamance), and the continental Casamance composed of two sub-elements: the districts of Kolda and Velingara (in Upper Casamance) and the district of Sedhiou (in middle Casamance) that constitute the Kolda Region'. 32 His approach further fragments the Casamance entity by proposing three regions in the place of the two adhered to by Jacques Charpy, and changes the land area from 28,350 km 2 to 21,110 km 2 . Deliberately or not, it mingles geographical criteria and administrative delimitation, while leaving implied but unspoken that the spatial division corresponds to a diversity of ethnic groups, ways of life and religious beliefs. 28

On the visualization of the nation, see Beth Baron, 'Nationalist Iconography', in James Jankowski and Gershoni Israel, eds, Rethinking Nationalism in the Arab Middle East. New York: Columbia University Press, 1997, pp. 105-24), p. 105. 29 Abbe Diamacoune Senghor, 'Casamance, Pays du Refus', p. 17. 30 Charpy, 'Casamance et Senegal au temps de la colonisation francaise', p. 3. 31 This interpretation follows the same structure as section 3 of Charpy's report 'Resistance' and chapter 2 (3.2), 'Les ethnies de Casamance et leurs rivalries', p. 10. 32 A. Konate, Le Probleme casamangais, p. 7.

The MFDC & the Struggle for Independence in Casamance

225

To complicate matters, the various communities living in Casamance have been pitted against each other by historians and ethnologists. Tables 13.1 and 13.2 present the ethnic distribution and composition of the population in each of the districts that constitute the two new regions of the Casamance: the region of Ziguinchor, with the districts of Bignona, Oussouye and Ziguinchor (Table 13.1) and the region of Kolda, with the districts of Kolda, Sedhiou and Velingara (Table 13.2). The region of Ziguinchor has a population of 398,387 inhabitants, distributed among its three districts as follows: Bignona, 184,807 , Oussouye, 37,098 and Ziguinchor, 176,432. Table 1 presents the ethnic distribution of the population according to the four large groups, Joola, Manding, Peul/Pular, and Wolof. Table 13.1 Ziguinchor: Ethnic distribution of population (%) Ethnic groups Joola Manding Peul/Pular Wolof

Bignona 80.6 5.3 6.1 1.8

Oussouye 82.4 4.7 1.5 4.8

Ziguinchor

Total

34.5 13.5 14.4 8.2

60.7 8.8 9.3 4.8

Source: Direction de la Prevision et la Statistique. Population Census. May 1988.

The region of Kolda has a population of 591,833 inhabitants, distributed among its three districts as follows: Kolda, 183,716, Sedhiou, 281,604 and Velingara, 126,513. Table 13.2 presents the ethnic distribution of the population according to the four large groups, Joola, Manding, Peul/Pular, and Wolof. Table 13.2 Kolda: Ethnic distribution of population (%) Ethnic groups

Kolda

Sedhiou

Velingara

Total

Joola Manding Peul/Pular Wolof

1.6 73.5 9.7 7.6

10.9 39.5 19.9 1.6

0.7 8.3 80.0 1.2

5.9 49.3 23.6 3.4

Source: Ibid.

The two tables show, from the perspective of the partisans of national unity, that the ethnic diversity of the Casamance is expressed by one ethnic group dominating in each of the regions: the Joola in the three districts constituting the Ziguinchor region (Lower Casamance) while being a minority in the districts of Kolda, and the Peul being the majority in the district of Velingara and the Manding in Sedhiou. This is one way of rejecting with strong scientific and objective evidence the M F D C claim to lead a region, the Casamance, which exists only in the imagination of its militants and sympathizers. The nationalist geography and demography insist especially on the weak presence of the 'nordistes\ migrants coming from the northern region of Senegal, which allowed the construction of twin notions: the 'neglect' and the 'invasion' of the Casamance as the exclusive two forms of the central government's intervention in the region. The only common point that these three strongly ethnographic approaches share (in terms of cultures, ethnicity and religions), is that the particularity of

226

Mamadou Diouf

the Casamance is its territorial discontinuity with the rest of Senegal; the Gambia divides Senegal along nearly 300 kilometers and across a width of 50 kilometers. This discontinuity is reinforced by the common ethnic kinship between the populations of this country and those of Guinea Bissau.

Autochthonous and First Settlers: Contested Histories and Ethnology of the Casamance peoples In large measure, the controversies about the history of the Casamance population lie at the heart of the debate on the methods of ethnic construction in the context of the Casamance rebellion, and they locate the construction of the truth regime that forms the basis of secession and the modes of recovering and revealing Joola identity in the region - an identity that certainly cannot be exhibited but whose discrete but forceful presence is critical. In this subtle and difficult process, Abbe Diamacoune Senghor is a passed master. His expert work is magnificent, compared with the objectivity and scientific or political rationality sought by the other actors. Thus he uncovers the opposition between memory and history. Abdourahmane Konate relies on the authority of Louis Vincent Thomas 3 3 to identify 'nine ethnic groups . . . separated both by their language and their location . . . on the right bank of the River (La Casamance): les bliss-karone, the joola fooni) and the joola juugout. On the left bank, xhzfloup at Oussouye, xhzjaamat at Yutu and at Effoc, the juuat at Jimberin, the her at Kabrouss, the joola of Point Saint Georges (which constitutes the villages of Mloump and Kanut) and the joola brin-seleky'.34 The list of the nine ethnic groups plays the same role in Thomas' ethnographic approach and in Konate's politico-administrative work: to demonstrate the non-existence of a Casamance people, the ethnic fragmentation of the population, and the absence of homogeneity among the Joola group. The function of this description differs, of course, for Thomas and Konate. My presentation of Thomas is of the 'romanticization/exoticization'35 of the segmentary groups' resistance to colonization, and their stubborn retreat into their identity of distinctness and difference from the societies of northern and western Senegal. Thomas' work thus plays a by no means negligible role in the construction of the forms of identification produced by the MFDC. It is crucial to note that Thomas' identification of the different Joola groups has developed a great deal. At the beginning of his career he recognized fifteen, then he adhered to nine (those mentioned by Konate), then eight.36 Insisting constantly on the cultural and linguistic diversity of the Joola - 'one cannot speak of the Joola except in the plural' the French ethnologist would finally retain the following groups as Joola: settled on the left bank of the Casamance are 'the Flup, the Diamat, the Bayot, the Her, and the Dyaiwat; on the right bank, the Fogny, the Blis, the Karone, the Kambo, the Narang, and the Kadiamoutay. Each has appreciable socio-cultural and linguistic differences, as well as obvious antagonisms.' 37 Recognizing nevertheless 33 Louis Vincent Thomas is 'the doyen of the Diola ethnographers', according to R. M. Baum, 'The Emergence of a Diola Christianity', Africa, 60, 3, 1990, pp. 370-98, p. 370. He is the author of Les Diolas. Essai dyanalyse fonctionnelle sur une population de Basse Casamance. Dakar: IFAN, 1959. 34 A. Konate, Le Probleme casamancais, pp. 12-13. 35 The most eloquent illustration of this romantic vision is found in Thomas' 'Foreword' to Peter Mark, The Wild Bull and the Sacred Forest. Forms, Meaning, and Change in Senegambian Initiation Masks. Cambridge: Cambridge University Press, 1992, p. ix. 36 See Mark, The Wild Bull and the Sacred Forest, p. 22. 37 Thomas, 'Foreword', p. ix.

The MFDC & the Struggle for Independence in Casamance

227

that the shared consciousness of belonging to the same group prompts the Joola to regroup in a common front against the outside threat, Thomas confirms once again the difficulty of answering the following question: 'Are the Joola an ethnic group in the process of creation or dissolution?' 38 Since then it is understood why the different ethnic groups identified by Konate are not necessarily Joola, or at least he does not specify it. This vagueness is accentuated by his labelling certain Joola - thefooni and xhtjuugat - while the others are either not labelled or are located by their villages (Yutu or Effoc) or by their region (Ousseye, Point Saint Georges). He does not assign them indigenous origins, and maintains that they came from elsewhere, from Gabou, today in Guinea Bissau. They settled 'in Casamance where they met populations from the northern regions such as the "Nyanthio" or the "Guelevars" of Sine Saloum, to whom they are related, and who came themselves from the Empire of Gabou after the wars and political rivalries between the "Nyanthio" who commanded the provinces of this empire'. 39 The Casamance is thus a meeting place of different population groups with certain particularities: 'some are in the forest, others on the maritime fringe and in the less wooded zones that evoke their origins, namely the forest, the maritime coast and the savana. This refers especially to the Joola, the 'Lebous' (Lebu), the 'Sereres' (Sereer), and the Saint Louis fishermen.'40 Interestingly, Konate is concerned neither with the chronology nor the settlement of successive waves of ethnic groups who populate the Casamance today in his approach to the migration question. For example, the Lebu probably started to inhabit the region at the beginning of the twentieth century and the cGuet Ndariens' (Saint Louis fishermen) in the middle of the same century. With his chronological telescoping, Konate reinforces not only the fragmentation, but also the lack of cohesion and the weakness of a genealogical claim to construct an autochthony of exclusion. Jacques Charpy is, in contrast, more precise. He relies on documents from the French administration and the works of French historians, in particular Christian Roche. He adopts a less complicated typology. He gives (in section 3, 'Resistance') the paragraph on the Casamance populations a revealing title: 'The ethnic groups of Casamance and their rivalries'. It emphasizes the differences between the groups, and the reciprocal hostility that each showed to the others. The sub-title of this paragraph, 'The colonial conquest complicates the ethnic struggles', 41 specifies that the ethnic conflicts existed before colonial domination. The French archivist considers that 'the Diolas bear the constraints of the administration with difficulty and react with vigor. The 1886 assassination at Seleke in Bayotte country of Lieutenant Truche and his escort gave them a great notoriety. Troubles will perturb the Diolas' territory for a long time yet.' 42 The Bainouk come next, 'the most ancient people of the Casamance are in the process of disappearing. They were pressed in the eighteenth century by their bellicose neighbors, then decimated by the Mandingues to the East, repelled by the Diolas to the West and the Balants in the South, and occasionally assimilated by the Portuguese.' 43 Charpy provides the very important information of the Bainouk's precedence in the Casamance territory; the Peuls/Pulaar 'who occu38 39 40 41 42 43

Ibid. Konate, Le probleme casamangaise, pp. 12-13. Ibid. Charpy, 'Casamance et Senegal au temps de la colonisation francaise', p. 10. Ibid. Ibid., p 11.

228

Mamadou Diouf

pied the Fouladou in Upper Casamance, and who liberated themselves from the Mandingues yoke at the call of Alfa Molo, were supported by the almamys of the Fouta Djalo in return for their joining Islam . . . After his death and in agreement with Moussa Molo, his son, the French penetrated into the Fouladou in 1883.' 44 The Manding were 'settled since the seventeenth century between the Diolas of Fogny in the West, the Balants to the South-West and the Peuls in Firdou to the East. Although unable to subdue the Peuls, the fervent Mandingue agents of Islamization were encouraged by the marabout Fode Kaba to extend themselves on the right bank of the Soungrougrou; they pushed back the Bai'nouks, and from 1877 to 1893 struggled against the Diolas,, settling Mandingue colonies in the middle of the Diolas' villages. Fode Kaba signed an agreement in 1893 with the governor (of Senegal - M. Diouf) and evacuated Fogny; this indirectly saved the Diolas . . . '. Finally, the Balant 'retreated back on the left bank of the Casamance to the south of the Bai'nouks5; the Manding 'were introduced by the Portuguese to Ziguinchor'; the 'Toucouleurs, discrete and isolated, settled in the East of the Casamance'; and the Ouolofs, 'some of whom arrived individually as agents of the commercial houses or assistants of the administration, were sometimes treated as agents of French imperialism.' 45 This description reveals, with its ethnographic taint of colonial images and paternalism, the colonial vision of the Casamance peoples. In this sense, Charpy's testimony acts as that of the ex-colonial power in the Casamance crisis, and has an inestimable value. At one and the same time, it signals a genealogy that refutes the precedence of the Joola in the region, celebrates the role of French pacification in the preservation of the Joola ethnic group, and emphasizes the violence and generalized war in the precolonial period, the civilizing role of the Wolof collaborators, and the marginal position of the haalpular (Toucouleurs) who were settled in Casamance. Interestingly, Charpy's chronology is precise on certain questions and silent on others, for example the arrival of the haalpular in Casamance. Following Louis Vincent Thomas, he insists forcefully that the major demographic and cultural phenomenon of the Casamance is the meeting of several populations. In large measure, he renews the ethnological fascination that the Joola societies have had and continue to have on the French, notably the former colonial administrators and social scientists. They have played a significant role in the invention of the Joola ethnic group's qualities, especially in contrast to the Wolof. The best illustration is Hubert Deschamps' 46 reaction in his memoirs when he discovered the Lower Casamance: the beginning of Guinean Africa, the hot humidity and the wooded savanna: higher trees, palm trees, rich undergrowth, the play of shadows and light; rich but not crushing vegetation, suitable for man. The European sheds erected along the river were a proud line of rusted sheet metal. From them we saw exiting, like ants, long lines of women carrying baskets on their heads; they went thus, along rudimentary wooden jetties to load the sea vessels. All this honored neither the technological innovations nor the esthetic of our beloved France. Rather, the indigenous village was charming, with its woven straw fences, its kapok trees, palm trees, its turbulent life to the rhythm of the rice pestles and the tam-tams. Since a year ago attention was on the Floup. They inhabited the left bank in the maritime region, dense with astonishing vegetation, run through by brackish marshes, bordered by mangroves and soft clays. In this twilight 44

Ibid. Ibid. 46 Former governor of the colonies and first Chair of African History at the Sorbonne. 45

The M F D C & the Struggle for Independence in Casamance

229

world between the land, the sea and the soft waters, we saw fish climb into the trees, evoking the good old days some hundreds of millions of years ago, when their curious fellows left the water to become batrachians, the first terrestrial animals, our own ancestors. The Floup, naked, anarchist, and egalitarian always refused all dependency, all slavery, and hierarchy; whence, between villages in times past, came a fair margin of insecurity and of vendettas, yet which gave life its price, and which were settled by ceremonies and honorable compensations. This frequent type of organization in Africa which I later called 'balanced anarchy', with a lot of admiration on my part, has never worked well in Europe. Anarchy perhaps contributed to stimulating initiative; in any case the Floup had triumphed over so foreign and hostile a natural site with immense original works: building dams, desalination of the silts by the rains, and growing irrigated rice. The farming instrument, the kayendo, was also peculiar: a sort of an oar with a long handle. The Floup houses improve on the Senegalese straw huts with their quasi-roman plan: walls of clay, entry columns, and a central impluvium surrounded by rooms and granaries.47 Deschamps' testimony contains the entirety of the ethnological cliches and the construction modes of Joola identity, as distinct from that of the Senegalese: a vision sympathetic to pure populations, close to nature, 'naked'. The parallel with the genesis of humanity is arresting. Enterprizing and ingenious, they knew how to create a civilization comparable to that of ancient Rome. This tender regard of the colonial administrator recreates a situation from 1941, less than a year after the revolt led by Alin Sitoe Diatta. Peter Mark is one of the rare anthropologists who has dedicated his career to the populations of the Casamance, especially the Joola. He has tracked with historical precision this essential element: the Joola identity as revealed by French ethno-colonial writing. He combines historical, anthropological, artistic, and technological elements on the longue duree, thanks to a systematic collection of oral tradition, use of European documents since the Portuguese period (fifteenth century), and an approach that combines the results of linguistic, historical, and archeological research. His portrait of the Casamance societies depicts them as of the utmost complexity.48 The first conclusion brought out by Mark's contribution is the great diversity of the peoples that occupy the Casamance between the Gambia River and the Rivers of the South (the region of Cacheu in Guinea Bissau). He notes, first, the variety of the political and social governance structures between the stateless and segmentary Joola societies, on the one hand, and the Manding state-organized societies, which are strongly stratified both on the social (castes) and the political levels (hierarchical orders). This diversity, in a space that does not totally coincide with the geography of Abbe Diamacoune - who never mentioned the Gambian Joola regions, probably for strategic and diplomatic reasons - should not mask the similarities in institutions, culture, religions, and farming techniques. This diversity is the result of an historical trajectory described as follows: For centuries, however, these groups have had extensive contact with one another. This contact has led to the sharing of social institutions. Although the horned masks and their associated symbolism are not the only cultural traits that link these groups, the use of common imagery in initiation ritual graphically illustrates the strength of the cultural ties between Mandinka, Joola, Bagnum (Bainuk), and Balant.49 47 Hubert Deschamps, Roi de la Brousse. Memoires d'autres Mondes. Paris: Berger Levrault, 1975, p. 16. 48 His books, A Cultural, Economic, Religious History of Basse Casamance since 1500. Wiesbaden: Franz Steiner, 1985, and The Wild Bull and the Sacred Forest, are indispensable. 49 Mark, The Wild Bull and the Sacred Forest, p. 1.

230

Mamadou Diouf

From Peter Mark's perspective, diversity and similarity have resulted in a permanent fluidity of ethnic groups and an instability of ethnic identifications in the formation and distribution of the population of the Lower Casamance. Abdourahmane Konate describes in amazement this dead end 50 by writing: the region has no possibility of expansion. Driven back against the sea, the Joola is among the 'laterally suppressed'. This forces him to live in a recess, whence the implacable territoriality and the frequent vendettas.51 The American anthropologist undermines the thesis of the enclosure of the non-Manding societies in a jealously defended autarchy and primitivism manifest both in their social and institutional and also in their religious organization. On the contrary, he shows that the oral traditions point to population movements across the whole region from the seventeenth century. The ethnic groups of the region, the Joola and the Balant in particular, entered into commercial relationships with Portugal in the sixteenth century and with the French and English in the seventeenth centuries. The exchanges with the Europeans, from their arrival on the Atlantic coast, superimposed themselves on to the indigenous trade. The commercial intensity varied according to the zone and the periods, and gave rise to new forms of inter-ethnic cultural hybridization. According to Mark, the intense transactions between the Joola and Bainuk in the seventeenth century, in the regions of Buluf and Fogny, are two illustrations of this hybridization: traces are still visible in contemporary Joola culture, and in the strong interactions, especially at the military level, between the Joola and Manding in the western and northern border zones of the Casamance. 52 He defends the idea of an ethnic group in the process of formation, insisting on the constant construction and reconstruction of the peoples and cultures of the Casamance during the colonial period, and puts forward a typology and a history of the different groups inhabiting the region. He reduces the number of Joola groups from fifteen, nine and eight derived from the different typologies of Louis Vincent Thomas, to four: Fogny, Buluf (Jugut) and Karone, on the north bank of the Casamance river, and Kasa, on the south bank. 53 He confirms Charpy's thesis that the Bainuk are the oldest settled ethnic group in Casamance by relying on the Joola and Bainuk oral traditions, and on the European, Portuguese and French, colonial sources. 54 The Bainuk lived in Lower Casamance with the Cassagas, a related group that would have played an important economic role in Atlantic commerce, notably the slave trade and wax commerce. 55 Organized in centralized states, the Bainuk were very different from the acephalous Joola societies. The name of the Casamance is derived from that of the sovereign monarch of the Kassas or Kassangas, 56 Kasa Mansa. They occupied Fogny in the eighteenth century and made it the center of a powerful 'Bainuk empire' that controlled the trade routes between the Cacheu and Gambia rivers. At the end of the same century, they lost their control of these routes and faced, without success in the mid-nineteenth century, the dual pressure of the Balant in the East and South 50

Ibid, and p. 16. A. Konate, Le probleme casamancaise, p. 14. Mark, The Wild Bull and the Sacred Forest, p. 3. 53 Ibid., p. 17. 54 Ibid., p 22. 55 Ibid., p. 28. 56 Ibid. See also Olga Linares, 'Deferring to Trade in Slaves: the Joola of Casamance in Historical Perspective', History in Africa, 14, 1987, pp. 113-39. 51

52

The MFDC & the Struggle for Independence in Casamance

231

and the Joola who were settling with extraordinary speed in Fogny and Buluf; the Balant expelled them from their former territory, and the Joola managed to assimilate them culturally, thus erasing their autonomous signs of identification in these two regions. Today the Bainuk are estimated at 10,000 persons. 57 Compared to the Bainuk whom they expelled by force from the south bank of the Casamance river East of the town of Ziguinchor, the Balant are the new arrivals. After destroying the kingdom of Kasa, they renamed the region all along the south bank the Balantacounda, the land of the Balant. According to Mark, they inspired more fear than respect and friendship, compared with the Bainuk. 58 The last group, the Manding, occupy the north part of the Casamance. They would have come to the region from Mali in the thirteenth and fourteenth centuries, and probably assimilated a population related to the Sereer-Niominka, who live today in Senegalese territory north of the Gambia river. The SereerNiominka's presence in this Casamance region, and the contacts they would have had with the different peoples, would explain, according to Mark's hypothesis of a pre-Manding culture 59 , cultural traits common to the Joola, Sereer, Sonike, and Manding. 60 This culture was destroyed by contacts and intermarriage between Joola and Manding, the development of slavery and the slave raids, and, during the nineteenth century, the Islamization of the Joola and the Balant by Manding marabouts. 61 In light of these three perspectives - politico-administrative, archival-colonial, and academic - Abbe Diamacoune Senghor's response to Jacques Charpy's testimony clarifies the means by which a truth regime becomes irrefutable. Senghor's narrative constructs a reality through discourse and discursive formation that imposes itself incontestably on the set of actors. Senghor postulates the cultural unity of the Casamancais., a unity forged by the region's geography, which is 'an intermediary zone between the sahelian zone and the Guinean zone . . . a great crossroads where different migratory waves meet, mingle and mix'. 62 At this stage, he does not stray far from the theses presented above. While acknowledging this great diversity, he reduces its scope, first by placing it in the distant past, and then by integrating it into a unitary matrix that is produced by a continuous process of assimilation. The concept of mixing and migratory waves' confluence into a single great river helped the Abbe Diamacoune as a first step 63 to mask ethnic diversity and merge difference: he took the Joola culture as a common denominator without saying so. He pays this price in order to produce the Casamancais, unique, essential, and radically different from the Senegalese and from his own remarkable ancestors. Senghor's introductory section subtly challenges the thesis of Bainuk 57

Mark, The Wild Bull and the Sacred Forest, p. 29. Ibid. 59 Ibid., p. 30. 60 Ibid., p. 31 and G. Brooks, 'Historical Perspectives on the Guinea-Bissau Region: 15th to 19th century', in A. Texeira Da Mota. In Memoriam, Lisbon, 1987. 61 Mark, The Wild Bull and the Sacred Forest, pp. 31-2. 62 Diamacoune Senghor, 'Casamance, Pays du Refus', p. 32. 63 Declaration of the MFDC, Ziguinchor, 21 December 1993. p. 67. The second step consists of postulating the existence of generalized resistance movements in Casamance, which is the manifestation of the existence of a single and distinct people. This approach recalls the historical work of J. Michelet which traces and celebrates the part played by the French people as actors in the history of France, the founding act being the taking of the Bastille in 1789. 58

232

Mamadou Diouf

autochtony prior to the arrival of the Joola 64 by deconstructing the different presentations of the history of the region. He revisits the real etymological analysis ('in reality', he says) of the name Casamance which 'means "the country of rivers", "the country of waters'", 'the land that emerges from a great expanse of water formed by the ocean and the waterways', 'great river'.65 Reconstructing the name reinforces the regional geography drawn by the Joola priest from the Atlantic to the Faleme (the Ocean, the rivers and the vast river, the Senegal), expels the Bainuk, and introduces at the same time the other large ethnic group of the region, the Manding. Because the alliance with the Manding is crucial to the success of the MFDC rebellion, they become the second focal point of an ethnic history resulting from a critical reflection on the present: a second price to pay in the production of the Casamance. Even so, Abbe Diamacoune keeps the idea of a Bainuk - 'Jamaat' 66 - origin for 'Kasama aku', insisting that 'in Jamaat, the word "husamu" means "river". "Kasamu" is an augmentative that means "great river" or rather a great stretch of water or a great stretch of water from rivers or the ocean or the two together.' He draws from this the diminutive 'kassa' that designates a portion of 'Kasamu' and yet recalls the 'khasso' country, a zone between the Senegal River and the Niger River.67 Situating the Bainuk in the confluence between the Senegal and the Faleme, Abbe Diamacoune thus ties the foundational alliance between the Joola and the Manding. The name Joola, 'which means "those (from the country) of the waters" is of "Malinke" origin, which is to say Manding. The Manding lived in the eastern region, called "Gaabu", which means "those (from the country) distant"' 68 from the mainly Jamaat west. In this last sense, Jamaat no longer specifically signifies Bainuk, but a collection of ethnic groups around the Joola. The construction of a larger collection around the Joola extends the small groups demographically relative to the Manding and Fulani (Peul), and illustrates Abbe Diamacoune's keen sense of the vitality of ethnic competition and the depth of suspicion between the ethnic groups of 'his Casamance nation'. The second stage in the Casamance priest's undertaking is the analysis of the migratory waves, which, according to him, 'are composed of multiple stratifications that are overcome and eventually covered over and mixed'. 69 On this question, he does not depart from the positions of the researchers but, once again, his chronology relies on the prior character of the Joola. The Joola are the absolute reference point, and the ultimate conjunction point of the Casamance ethnic groups and their metamorphosis - certainly through resistance - into a single people. He writes: 'the "Joola" wave, the oldest, the largest even, and longer than the territory that extends from the Faleme to the Atlantic, can be considered as the 'the shared trunk of several branches'.70 The branches of the Jamaat group cited are the Joola, the 'Bainuka (Bainuk) branch of the kassa . . . Majuka (Manjak) Mankanga (Mancagne), Balanta (Balant)'.71 Following this first population, 'the "Malinke" wave subsequently covered the Casamance territory, more or less from the Faleme to Soun Grougou [a deformation of San 64

Ibid. Diamacoune Senghor, 'Casamance, Pays du Refus', p. 7. The 'Jamaat are Bainuka, a branch of the Kassa', ibid. 66 Ibid. 67 Ibid. 68 Ibid. 69 Ibid. 70 Ibid., emphasis added. 71 Diamacoune Senghor, 'Casamance, Pays du Refus' 'Introduction'. 65

The MFDC & the Struggle for Independence in Casamance

233

Gregorio], the Upper and Middle Casamance' and 'finally the Tula' wave, which extends from the Faleme all the way to engulf an eastern part of the Middle Casamance'. 72 Closing the list of ethnic groups present in the Casamance region points to the process of excluding the 'northerners'. Neither the Wolof nor the Halpular are mentioned by the priest. In our interpretation, the geography and the history of the Casamance region are a deliberate construction, and center on the imagining of a Casamance people with reference to a core cultural matrix and a Joola essence, the discovery of which in statements and grammar cannot help but be concealed. Abbe Diamacoune's specific understanding of this concealment reinforces our interpretation. Remaining in this place of shadows and light, he observes with false naivety: Thus it goes so far as to wish to make the Casamance problem simply a 'Joola' affair; this is inaccurate, it could be rigorously explained with reference to the extent and precedence of the ethnic stratifications, or even by the decisions taken by the Assembly before 25 December 1982, with people of all ethnic groups, religions, backgrounds, and political views together, which marked the start of the current events in the Casamance. The demonstrators had set off from Faleme to participate, at Ziguinchor, in the march by more than a hundred thousand Casamangais on 26 December 1982, with people of all ethnic groups, religions, backgrounds, and political views together. Those who have gone through and followed closely the genesis and the progress of the events in Casamance know very well why the burden of this struggle rests especially on the Diola (Joola) country, and also why this war of national liberation was at first restricted voluntarily to the Lower Casamance Zone . . . With its propagandistic manipulations of local and outside opinion, Senegal spoke of a supposed demand for a 'Diola Republic' by the demonstrators of 26 December 1982. The Senegalese operation tended to get the Casamance into trouble with the surrounding countries. Let no one be mistaken about it.73 Abbe Diamacoune's third stage is the invention of a golden age of the Casamance civilization that, while reinforcing the central image of the Joola, blends and unites the memories into a single one, and, at the same time, postulates the unity and particularity of the Casamance populations. This specificity combines an egalitarian democracy; a rice-cultivating technology that assures food self-sufficiency; and a traditional religion whose rituals, symbolism, and initiation ceremonies ensure a remarkable cohesion of society and civilization. Building on the work of Peter Mark, we have already noted not only the ethnic diversity of the Casamance, but also the fragmentation of the population called Joola. Not only were there frequent wars between different groups, but also between different neighborhoods or villages of the same group. The alliances between factions and groups were made and could be expressed by combining religious and ritual resources, and the organization of collective prayers; this situation also illustrates the absence of religious unity and cohesion among the groups. The importance of competition among Joola groups for access to environmental resources or those of Atlantic commerce, especially the slave trade, contradicts the idyllic vision of unity, stability and ethnic cohesion that broke with the societies of northern Senegal. Peter Mark writes: From the early eighteenth century, if not earlier, an important factor in continued interethnic contact in the Gambia-Casamance region was the domestic slave trade. The Mandinka obtained Jola and Bagun captives, some of whom were victims of 72 73

Ibid., pp. 7-8. Ibid., p. 8.

234

M a m a d o u Diouf

warfare in Fogny. Other captives, victims of Jola slave raids, were traded north to the Mandinka, and still others were local criminals sold into captivity. The descendants of many of these captives were eventually incorporated into Gambia society. This process of Mandingization also served to introduce aspects of the captives' own culture into the Gambia. 74 T h e final stage in constructing vestiges of Casamangais identification is expressed under the heading of religion. We have already mentioned that the traditional religion of most of the Joola groups is solidly rooted in the cult of the ancestors, with specific rites and initiation ceremonies, and whose site is the sacred forest, which plays a crucial role in the invention of the contemporary Joola identity. 7 5 According to L. V. T h o m a s , P. Mark and R. M . Baum, there is not only an often considerable variation in the nature of the sacred forests, shrines, and rituals from one village to another, b u t added to this diversity is the importance of conversion to the Catholic or Muslim religion within the Joola population at the beginning of the twentieth century. T h e authors note that Islam dominates among the Joola of the N o r t h bank of the Casamance River and Catholicism on the South bank. 7 6 Granted, this conversion does not necessarily mean the erasure of indigenous religions, or attendance in the sacred forests and at the cult of the ancestors. M a r k and B a u m interpret the Joola's belonging to one or other of the revealed religions as a response to the crises of being colonized, participating in the Senegambian monetary economy, notably groundnut farming, the intensifying urban migration, and the search for autonomy by youth and women in Joola communities experiencing harshly changing economic and social conditions. 7 7 Mark writes: Between 1900 and 1940, the establishment of French colonial administration and the introduction of cash crop agriculture led to far-reaching social and economic changes among the Diola of Boulouf. The growth of urban migration and trade, together with the sale of peanuts as a cash crop, brought a degree of financial autonomy to the young men who participated in these activities. Many individuals converted to Islam to achieve status not accessible to them in the traditional structure. 78 B a u m paints the following picture of the situation to interpret the conversion to Catholicism: In the late nineteenth century, Diola religion confronted the perpetual problems of rain, fertility, and the securing of good fortune and good health for the believer, but it also faced other problems of a more recent origin. Increased warfare and opportunities to gain wealth through trade had aggravated the problems of quarter, township and regional unity. The desire of the wealthy to control major religious shrines threatened the religious influence of the less privileged. A growing population of slaves and newcomers threatened the egalitarian ideology of a relatively homogenous society. The quest for spiritual unity had become a central problem of awasena religion as missionaries prepared to bring the new religion to Esulalu. 79 74

Mark, The Wild Bull and the Sacred Forest, p. 32 On this theme, see F. De Jong, Playing with Politicians. The Hague: Wotro, Working Paper 6, 1998; R. M. Baum, 'The Emergence of a Diola Christianity', and Shrines of the Slave Trade. Diola Identity and Society in Precolonial Africa. Oxford: Oxford University Press, 1999; Mark, The Wild Bull and the Sacred Forest. 76 Ibid. p. 23; P. Mark, 'The Rubber and Palm Oil Trade and the Islamization of the Joolah Boulouf, 1890-1940', Bulletin de LTFAN, 39, 1977, pp. 34-61; and 'Urban Migration, Cash Cropping and Calamity: The Spread of Islam among the Diola Boulouf, African Studies Review, 1978, pp. 91-9; and Baum, 'Emergence of a Diola Christianity', p. 378. 77 Ibid. 78 Mark, 'Urban Migration', p. 1. 79 Baum, 'Emergence of a Diola Christianity', p. 378. 75

The MFDC & the Struggle for Independence in Casamance

235

Abbe Diamacoune Senghor minimizes this insistence on religious diversity by returning to an essence that serves as the soul of the Casamangais. He forcefully notes, 'Religiously, despite the influence of Islam, Christianity and modern contributions, the Casamangais remains, in general, profoundly rooted in his traditional values'. He imagines that the work of the so-called revealed religions is to 'take on, inform, and sublimate these inherited values of his honored ancestors'. 80 Thus the priest manages not only to downplay the religious diversity among the Joola, but also to deny the mass scale and sometimes strongly proselytizing character of the Islam of the Manding and Peul. Following his ethnological reconstruction, Abbe Diamacoune's second reference point in constructing the Casamangais identity is, on the one hand, the resistance to colonization and, on the other, the colonial government of the Casamance. This resistance is an axis along which to arrange the Casamance groups into one single people; thus he completes his portrait by highlighting the Joola core.

The Production of Joola Identity: Colonial History and the Demands for Autonomy History is one of the main instruments of legitimization used by the MFDC. The choice of Portuguese history instead of French history shows the artificiality of the ethnic discourse, the manipulation used in constructing an historical memory, and the creation and production of a country that has significance only in terms of the current difficulties. The history serves to justify the ideologicopolitical relevance of the terminology of the 'independentists'. The work of the presumed leader of the M F D C expresses this perspective. The systemic rejection of the term 'separatism' and the radical affirmation of the right of the Casamance to self-determination are founded on the hard fact that the Southern region had never been included in the French colony of Senegal: '. . . according to Honore Pereira Barreto, Governor of Portuguese Guinea, based on documents he consulted, the Casamance was the first river and the first country the Portuguese crossed on the western side of black Africa . . . The Portuguese administration was established at Ziguinchor in 1645, fourteen years before the foundation of Saint-Louis in Senegal by the French in 1659. It remained there until 1888, or 243 years of Portuguese presence. This is not forgotten.' 81 This precedence of the Portuguese presence in Casamance is confirmed by Christian Roche in his History of the Casamance: In 1641 the king of Portugal Don Joao IV named Congalo Gamboa Ayala the first captain of this trading post, who founded Farim in 1645 on the upper Cacheu and Ziguinchor on the Lower Casamance. He wanted to assemble the Portuguese merchants who were scattered the length of the main kingdoms of Guinea.82 The separatists' appeal to Portuguese history has a dual interest. On the one hand, it allows them to anchor the legitimacy of the nationalist demand to the administrative exclusion/non-inclusion of the Casamance in the colony. Whether autochthonous territory or protectorate under the supervision of the colony of Senegal, the Casamance was ceded to France by Portugal by the 80 81 82

Diamacoune Senghor, 'Casamance, Pays du Refus', p. 7. SudHebdo, No. 90, 1 February 1990. C. Roche, Histoire de la Casamance. Conquete et resistance: 1850-1920. Paris: Karthala, 1985, p. 67.

236

Mamadou Diouf

Convention of 12 May 1886. Moreover, the shift from one legal status to another is explained by the French indecision about how to manage the Southern region, whether as an entity independent of the Senegal colony or as an integral part of it. The use of the term 'colony of Senegal,' 'territory of Casamance' in official administrative documents is one proof of the juridicopolitical state of flux. We can decode from the use of history the desire for scientific objectivity inherent in the legitimization of the demand, which rather poorly obscures plagiarism of the Joola nationalist discourse; this discourse in no way relies on the rejection of French colonial history but rather on its timely and selective use. As an example, we can cite Abbe Diamacoune Senghor's reference to the placard wavers who welcomed the Governor General of the AOF (French West Africa) William Ponty on 20 March 1914, demanding financial, territorial, and administrative autonomy for the Casamance; this reference would be taken up again and published by Politique Africaine83 Because this allusion resonates with French colonial history (General de Gaulle welcomed by placard wavers in 1958), it becomes a way of almost explicitly disqualifying the Senegalese position. In addition, the reference to Portuguese history carves out a territorial space. The map of the Joola Republic, drawn by ATTIKA (the army branch of the MFDC), 8 4 included the Lower Casamance and a part of the Republic of Guinea Bissau, namely, the province of San Domingo, in the Cacheu region. This ethnic label for the Republic and the space it constitutes might be surprising in this multi-ethnic region. 85 It excludes all the other ethnic groups and gives only the Joola a right to Casamance space. Whatever the cause, the validation of a Joola identity with the social imaginaire that is rightfully its own is rooted in a country circumscribed by a particular memory. This memory is compiled from a long experience of resistance which can be summed up by the words of Governor General Van Vollenhoven to Lieutenant-Governor Levecque of 17 November 1917: 'We are not the masters of the Lower-Casamance, we are only tolerated there . . . The Casamance must no longer be a kind of wart on the colony for which it should be the jewel.' 86 The mechanism to affirm Joola identity and exclude other ethnic groups is not based entirely on the historical reference, which is an artifice used by the nationalist discourse to legitimize the demand for independence and an ethnic layout for the future 'Republic'. Religious and sociological criteria, strongly colored by criticism of the regional economic policy of the central government, are also used. The secession fought for by the M F D C is, in a certain way, sustained by the rupture with the Islamo-Wolof model, while at the same time making it operational. This dual rupture relies on a linguistic, sociological, and religious triptych. If language is both an indispensable instrument of exclusion in the creation of all 'imagined community', 87 and the expression of an identity, it is also a supposedly natural exclusion factor. Affirmation of the Senegalese nation operates around the reterritorialization of identity values borrowed from the Wolof ethnic 83

Politique Africaine, No. 18 June, 1985, Paris: Karthala, p. 129. Le Temoin, 6 November 1990. Roche, Histoire, p. 22. 86 Archives Nationales du Senegal 13G 384. Le Gouverneur General au Lieutenant Gouverneur du Senegal, 17 November 1917. 87 B. Anderson, Imagined Communities. Reflections on the Origin and the Spread of Nationalism. London: Verso, 1983, p. 122. 84

85

The MFDC & the Struggle for Independence in Casamance

237

group. This wolofization of the populations in the construction of the nation is evident in the figures from the National Census Bureau: the Wolof represent 43.7 per cent of the population and the Joola 5.5 per cent. This distribution is still more instructive once is it made on the basis of spoken languages: Wolof's percentage as a language (70.9 per cent) exceeds that of the ethnic group, in contrast to the Joola which is characterized by a balance between the language and the ethnic group. 88 What conclusion can we draw? We could follow Souleymane B. Daigne's reflection on the future of tradition when he discusses the process of wolofization of society '. . . the wolof is, in fact, less and less a well defined ethnic group, and merely a language spoken by 80 per cent of the population'. 89 In other respects, the wolofization process is the most obvious sign of continuity between the colonial administration and the post-colonial state. With the choice of urban centers as political sites (the Four Communes whose inhabitants enjoyed French citizenship were all located in Wolof territories) and the maraboutic network's coverage of the rural world, the whole process centers on the groundnut sector, once the nerve of the colonial economy and today the economic bedrock of political clientelism. This continuity is also found in the role of Islam in 'deterritorializing and reterritorializing' (90 per cent of the Senegalese are Muslim) communities through the operations of conversion and religious transactions. 90 The dual task of Islam is explained by the need to merge the specific regions from both the ethnic and the spiritual points of view, and to embark on the constitution of a homogenous space where the Senegalese universe, which is Islamic and Wolof, can be read. If the identity discourse of the post-colonial state borrows from these different elements, we must recognize the excesses of a unitary state logic: 'if tying Senegal to the global market by means of groundnut farming contributes to a tendency for the islamo-wolof model to dominate, this also gives rise to its opposite, which is the ethnic claim'. In the Casamance case, the connection between the territorial rupture and the rupture with the islamo-wolof model must be read in this context. 91 The Senegalese state was not able to express its national integration project by anchoring its own peripheral zones to the center. By playing on multiple logics and legitimacies (including ethnic ones) for its own consolidation, founded on political clientelism, the state found itself opposing a resistance born of an affirmation of Joola identity (for example) based on socio-cultural specificities resistant to bureaucratic and administrative solutions. Elsewhere we have shown how the 'modernizing complex' 92 had its efficiency undermined in the peripheral zones generally, and particularly in the Casamance. To illustrate, we can cite the impact of the co-operative movement in the groundnut basin and its lack of effect in Casamance. The southern populations think that the administrative directives are aimed at allowing the nordistes a new colonization of the Casamance. This view is based especially on the fact that Joola society has always lived under the rule of multiple decision-making centers. This situation ensures a more egalitarian or less hierarchical context compared with the northern societies. A highly intelligent combination of 88 D. Cruise O'Brien, 'Langue et nationality au Senegal: l'enjeu politique de la wolofisation', VAnnee Africaine, 1979, pp. 319-35. 89 Diop. Le Senegal et ses voisins, p. 283. 90 Ibid., p. 287. 91 Ibid., pp. 286-7. 92 Ibid., p. 255.

238

Mamadou Diouf

liberty, independence and solidarity, Joola individualism has always opposed foreign invasion, once Portuguese and French, now Senegalese, to use the words of the separatists. Because of this, it seems that the state's authority in the Southern region has always been resented as a graft from a foreign body. The integrationist alchemy of Islam could never affect this Joola particularity. In fact, until the nineteenth century, local beliefs and spiritualities exercised a monopoly on Joola beliefs. The practice of Bukut at the beginning of the nineteenth century replaced the Kahat, and is not unrelated to the important breakthroughs of Islam and Christianity at the start of the century. Roche sees in the Bukut 'the characteristic of a society worried about its collective survival'. 93 This survival reflex manifest in the Bukut is explained differently after the First World War with the prophetesses or queens, the most famous of whom was Alin Sitoe Diatta, symbols of a new form of resistance. According to A. Mbembe, the renewed prominence of the pagan spirit should be situated in relation to 'the historical capacity of societies for unruliness'. 94 In Casamance, the sacred forest crystallizes this unruliness, and establishes itself as a factor for identification and contestation. The political charge of the symbolic arena disqualifies the state's totalizing islamo-wolof project.

Conclusion The armed struggle for Casamance independence marks the limits, and even the failure, of the identity discourse used since independence by the state to legitimize its instrumental function in the elaboration of the Senegalese nation. By taking as its own the islamo-wolof model produced by colonial policy, the postcolonial state was unable to create new spaces and production sites for this 'imagined community' with a real identity, born of the ethnic differences and souls of the Senegalese nation. The state discourse has never been able to capture the ethnic discourse, but it has never stopped trying, albeit clandestinely. Casamance separatism sets its own historical trajectory in this current national historical process that it does not see as its own. The sense of exclusion is based on several material elements: the isolation of the southern region, the massive arrival of 'nordistes^ who clear and expropriate the local lands for their own profit, the regional wealth that does not benefit the autochthones, etc. These elements express the irrelevance of an economic policy that is driven by the center's obsession with the groundnut basin and the Dakar region, and that increasingly marginalizes the peripheries, the Casamance in particular. The special features of the Casamance explain its irredentism, and are based on the combination of two essential factors. Its marked isolation has accentuated its exteriority relative to the Islamo-wolof model. Also, the fraction of the ecclesiastical elite present in the region has known how to amplify the Casamance problems greatly, producing a literary imagination 95 to forge an 'imagined community' and above all its own memory to anchor Joola identity. Much as it is 93 The Kahat is a public initiation with circumcision as an important element. Roche speaks of the Kahat as being the oldest form of paganism in Lower Casamance. With the malinke penetration and the extension of the colonial order (at the beginning of the nineteenth century), which saw the joola world opening up to the outside, the Kahat yields place to Bukut (more esoteric, more secret and aimed at the affirmation of masculine sexuality). 94 A. Mbembe, Ajriques Indociles-Christianisme, Pouvoir et Etat en societe postcoloniale. Paris: Karthala, 1988, p. 23. 95 Adrian Hastings, The Creation of Nationhood. Cambridge: Cambridge University Press, 1988.

The M F D C & the Struggle for Independence in Casamance

239

possible to speak of a Casamance socio-cultural particularity, the ethnic discourse secreted by separatism is artificial. A manifest recourse to Portuguese colonial history, a clandestine solicitation of French colonial history, an affirmation of an identity both pagan and Catholic, a recall of the 'long-lasting' history of resistance against the invader . . . , these elements have meaning only in the way that they are used by the discourse to legitimize a Joola country where there would be no place for any other ethnic group. The Casamance crisis is also a demand for territorial rupture, to create a new state for the Casamance nation. It has now lasted too long. What solution can be envisaged? The ethnic argument used by the M F D C is less an end in itself than a rhetorical and political instrument to express unease within the national entity and in the model of development put into practice. The colonial heritage of state centralization and the Islamo-wolof model in which the trajectory of the Senegalese state is expressed have never succeeded in creating a space for the ethnic cultures to combine and enter into dialogue. The sense of frustration and the cultural contempt for the Joola are indeed more than just subjective realities or epiphenomena. What is at stake is not simply a peaceful solution to the crisis, but also the necessary challenge to the center-peripheries model on which the political practice of the state rests. The crisis of the peripheries, as much in the Casamance as in the Senegal River valley region, shows that no democratic vision of society can function exclusively at the center without the risk of imploding the whole system. If a break with the center-peripheries model is imposed as a solution, it will be effective only inside new regional and subregional political entities, thus recreating the 'natural' countries where the populations could live with their roots. The Casamance is bound to head in this direction, that is to say, in the direction of history.

14

B O G U M I L JEWSIEWICKI & LEONARD N ' S A N D A BULELI

Ethnicities as 'First Nations' of the Congolese NationState: Some Preliminary Observations Translated from French by Jonathan M. Sears

W

ithin the context of the political arena's institutionalization by the nation-state, the management and scholarly construction of ethnicity strip this modern solidarity of its constituent subjectivities. The nation-state, while setting itself the task of merging ethnicities within the nation, institutionalizes and perpetuates ethnicity. This ethnicity becomes a form of unfulfilled nationalism, a nationalism without a state, or even an initial element of the nation, necessary but insufficient in the absence of the state. We intend in this chapter to analyze the use of ethnicity to frame contemporary movements that challenge the state monopoly of sovereignty, movements that, in striving to achieve a right to self-governance in the political arena, eventually help civil society to emerge from this arena and rebuild the state. A few general introductory comments are necessary before presenting the transformation of student organizations, in Lubumbashi of the 1980s, into openly political movements, and the transformation of ethnic associations, in Kivu of the 1990s, into movements for territorial self-defense of the locally absent nation-state.

On State and Ethnicity Our account stresses three characteristics of the nation-state that are particularly important in the post-colonial context: (i) Even if the nation, in the historical and Western sense of the term, does not necessarily constitute the operational requirements of the contemporary state, the internal criteria of the global inter-state system introduce an equation of the state and the nation as a foundational criterion of legitimacy. A state able to demonstrate this equation enjoys greater legitimacy both with international organizations that make the system function, and with international public opinion. The modern nation is a realm of memory (Nora, 1997), European first, Western later. The memory of modern politics cannot escape the nation represented as people, as race, as language, the uniqueness of which gives the state its singularity in its universality. (ii) The standards of the inter-state political system make (or at least made until a few years ago) stability and the precise definition of borders a condition of the full enjoyment of sovereignty. A state exercises its prerogative within a territory delimited by internationally recognized borders and populated predominantly by a single nation. The primacy of the nation over ethnic groups thus constitutes the condition of state sovereignty. (iii) The international preference given to electoral democracy and decisionmaking by majority vote (at least as a fagade) presupposes that individual lib240

Ethnicities as 'First Nations' of the Congolese Nation-State

241

erties are privileged to the detriment of public liberties, even though the effective exercise of the former clashes with the absence of the latter. At Independence, this model of democracy was handed down to the post-colony (Mbembe, 2002) by the colonial administration, and then reinforced by the international system and by international public opinion (which is above all produced by Western media and popular culture). The pressure thus exercised on African states, at least on their decorum, has not received the attention it deserves. In the case of Central Africa formerly under Belgian administration, three abrupt changes marked political memory at independence. When the electoral system was introduced to Kasai, the support of the colonial administration was given to the 'autochthonous' Lulua to the disadvantage of the Luba, who had previously enjoyed favors from this same administration. The reversal of the administration's position in Rwanda was meant to emancipate the majority Hutu, seen as 'native' Bantus, in order to limit the political influence of the minority Tutsi, Nilotic 'immigrants' who also were formerly privileged by the colonial administration. At independence, the administration was also seeking a majority in Katanga that would stabilize the political system in the form of either a federal or an independent entity. The colonial administration, worried about the success of candidates not native to Katanga in the first municipal elections (in three bourgmestres out of four), helped to construct a Kantangese majority (natives of the territory) within the secessionist state. In each of these three cases, the colonial administration supported the native populations of a territory, those whose ethnic group was then considered as the prior occupant. Bruce Berman and his colleagues rightly stress in the Introduction to this volume that the ideal transition from the colony to the independent state did not succeed, even in the case of the British system, despite its having been better prepared historically for the transfer of power. The transfer of knowledge and management skills was better instituted because of the more widespread practice of indirect rule, and especially because of the university education that had been accessible for a longer period of time, and was more generally accessible to the colonized populations. The last-minute ambition in the Belgian colonies was also to hand over the political institutions to elites educated in the spirit of continuity. All over Africa for at least three decades, a university education has been the key to entering the management of the state and its specialized institutions. Students often felt called to participate more in the management of political affairs than in a practical apprenticeship in social or economic realities. Learning the theory and ideology was more attractive than getting the experience. What resulted was an 'ethnologically' informed mediation between state management of the social and political practice of wealth and power, between the individual and the community, between collective identities that legitimize solidarity practices and their instrumentalization in the political sphere. Ethnographic monographs, most often written by colonial agents, became an ethnic 'bible'; a territorialized ethnicity prepared the way for the notion 'native of a postcolonial transformation of the colonial legal category of native {indigene). The anthropology of Africa was taught in secondary school as normative knowledge divorced from the urban practice of kinship. Scholastic university knowledge became an indispensable mediation between the individual and the institution, between daily practice and political action, between everyday life, considered unauthentic, and the life decreed in schools as fully authentic, and between village tradition and urban modernity.

242

Bogumil Jewsiewicki & Leonard N'Sanda Buleli

Paradoxically, the everyday construction of the real life was decreed to be, and practically accepted as, unauthentic because it operated between two ideal poles: on the one hand, the precious, authentic heritage of ancient Africa and, on the other hand, modernity, represented as a foreign intrusion, and thus unauthentic. At least from the scholastic point of view, that of Mobutu's intellectuals, the reality was metisse, impure, while ideology and scholastic discourse were models of the future. Scholastic mediation devalued experience, actor subjectivity and memory, and privileged ideological discourse. It is not, however, only an issue of the discourse produced and supported by the state, for example that of ethnicization, which transformed ethnicity into race in Rwanda and Burundi. Having become guides for reading social and political realities, ethnohistory and ethnography acquired a political importance. University discourse took on an unequalled power in Central Africa. A mere master's thesis or an honors essay for the graduation certificate took on the quality of a decree guaranteeing the legitimate collective presence of a group which aspired to the status of an ethnic polity. At another time and place it would have remained unpublished, buried in a university library, or perhaps even lost or all its copies destroyed. The following example clearly demonstrates that this race to establish collective belonging to the sphere of legitimate politics is no trivial matter, and cannot be seen as merely flirting with identity. In 1988 at the Bukavu Institute of Higher Learning (Institut Pedagogique Superieur), then the main university institution in Kivu, a Munyamulenge student (before this term acquired its scholarly and political significance) wanted to present a bachelor's dissertation in history on the emergence of this social entity. The teachers steadfastly refused to admit this subject as legitimate, even though, at the time, the Banyamulenge (mu - singular, ba - plural) political problem was not yet a national issue. 1 Certainly, the deteriorating cattle farming economy, and, especially, the cessation of widespread self-sufficiency since the end of the early 1960s' rebellions, pitted the cattle farmers living on the high plateau of South Kivu against their agricultural neighbors living below. The memory was undoubtedly still fresh of military conflicts during the eradication of the rebellions. These conflicts had pitted the Bemba militias against the young cattle farmers, who were armed by the Mobutu regime, and were protecting their herds or avenging their destruction. It was, nevertheless, a conflict between groups of individuals for whom the issue was land, rather than a conflict defined in ethnic terms. The return of the first generation of Banyamulenge youth to have finished school, and later university, introduced a mediation of identity and a politicization of practices. As actors in local struggles for land, the cattle farmers were forced into the realm of the nation-state as a collective entity, and thus not only received their own collective name but also inherited a duty to remember and a cultural legacy which transformed them into bearers of a 1 In his bachelor's dissertation the student wanted to prove that the Banyamulenge were present in their current location before the arrival of Belgian colonists. This demonstration was seen as intending to justify a right to Congolese citizenship and to affirm the legitimacy of the occupation of lands where the Banyamulenge lived. The dissertation was presented before a jury composed of a member of the Shi ethnicity (chef de travaux Kazunguzibwa Nyenyezi) and another from the Hunde ethnicity (chef de travaux Hangi Shambamba), and was rejected for obvious political reasons, despite the protests of the dissertation supervisor (chef de travaux Habimana Munyarinaje), a Congolese Tutsi from Rutshuru. The Institute's academic authorities were forced to send the dissertation to another jury at the National Learning Institute of Kinshasa, since the problem had stirred up such emotion in Bukavu.

Ethnicities as 'First Nations' of the Congolese Nation-State

243

shared destiny. Their identity, and because of this their interests, were institutionalized in the state political sphere; at a stroke this changed the nature of the conflict, opening it both to political and to expert scholastic interventions. The professors of the Bukavu Institute wanted to prevent the young Banyamulenge intellectuals from assisting this entry into the state sphere by making Banyamulenge identity an academic issue.

Ethnicity in Congo during the Statization (Etatisation) of the Political Arena The objectification and institutionalization of ethnicity do not exhaust its political dimensions. Political, social and economic practices have been interwoven with subjective expressions and conjunctural constructions of solidarity that claimed to represent a sense of belonging to a specific entity defined in terms of descent and native territory. There has been a conscious alignment of ethnicity and the nation in Central Africa since at least 1990, assigning to each ethnic group a native territory where its members would enjoy their ethnic, and thus economic, sovereignty to the full. Elsewhere in the national territory they would at best be hosts, at worst illegal immigrants, except in Kinshasa with a population of more than 3 million inhabitants. A brief summary of the relationships between the Mobutu party-state and local ethnic identities is in order before we present examples that will show how, in the context of state institutionalization and scholarly mediation, social and political practices that depend on ethnic solidarity can force an opening up of the political arena. Peter Ekeh has rightly reminded us that ethnic mobilization is under two constraints in the post-colony. First of all, the leader of an ethnic community cannot maintain mobilization unless he can effectively redistribute goods that have currency in social and political transactions: economic, political, or symbolic (cultural). The capitalist economy endangers his position, threatened by challengers who are able to redistribute as many or more goods acquired through activities outside of the moral community, whether it is the ethnic group or the ethnic nation. In the case of Congo/Zaire under Mobutu, when the nation was politicized and defined in ethnic terms, the circulation of goods was strictly confined to the political sphere. Economic or religious entrepreneurs who were able to accumulate economic or spiritual goods outside of the party-state's control potentially threatened the monopoly of the President, the head of the ethnic nation. The administrative destruction of the Bank of Kinshasa and the co-optation of the Kimbanguist church, followed by all the Protestant churches, well illustrate the attention accorded them by the regime. Mobutu endeavored to construct, on the scale of the state he controlled in the late 1960s, an ethnic nation with himself as the sole chief. While the ideology of unity, recalling Lumumba, was shown to the outside world, the domestic representation of Mobutu was that of a generous and peacemaking chief. He liked to emphasize that no village had ever needed two chiefs at the same time. Nationalization of land and mining resources (the so-called Bakajika law) was going to be the basis of a common heritage in the hands of the leader. Ideologically, the incessantly recalled memory of four years of rebellion gave Mobutu the aura of the warrior who brought the new civilization. The fall of international copper prices and the progressive exhaustion of industrial capital, which was built up during the colonial period on the work of the Congolese,

244

Bogumil Jewsiewicki & Leonard N'Sanda Buleli

brought about a change at the beginning of the 1980s in the circulation modes of the goods controlled by the leader. The concept of the nation nonetheless stayed unchanged, and its ethnic character was even reinforced by its racialization. ZaTrianization (confiscation of businesses owned by ethnic non-Zai'rians) gave the President all of the goods which, once withdrawn from the economic sphere, could be put into political circulation. Meanwhile, by giving himself the power to allocate these expropriated (zai'rianized) businesses, Mobutu began the process that transformed him into an 'ethnic' leader among others, but certainly the most powerful at that time. Thus, during the 1980s, his base of support was reduced first to one region, l'Equateur, then to one group, the Ngbandi. Thus ethnicity ended up at the disposal of the political class, despite having been declared illegitimate in the state sphere because it potentially competed with the ethnic nation (remember that a law proclaimed each Zairian a member from birth of the single party - the MPR). Decentralization of the state administration in the 1980s broadened the application of the principle that each political entrepreneur must generate his own resources, on condition that he paid tribute to Mobutu. Outside the capital city, where political life and military power remained centralized, even local arbitration ended up conceded to regional leaders in order to escape the influence of Mobutu. The political system reached the point of no return in the 1990s. The concessions to political entrepreneurs in the central state arena - especially after two waves of plundering in the early 1990s - no longer generated resources for redistribution, and resulted in the abandonment of Kinshasa by Mobutu. A vain effort to restructure the political arena attempted to form a mosaic of the spaces that had been delimited by the clash of urban residence and the origin principle (a territorialized filial relationship that defined belonging by the native village of the last male descendant from the rural milieu). Contrary to every expectation of experts and Congolese politicians, the sense of national belonging proved to be a real stumbling block for the initial successes of regional ethnic cleansing. In the 1990s, belonging to the nation was nevertheless indirect: one was Congolese to the extent that one belonged legitimately to an ethnic group that was part of the nation. Thus, in the late 1990s, the high degree of support for the idea that a Tutsi could never be Congolese was explained, since, historically, his ethnic group was not part of the nation. The explanation rests in part on the fact that ethnic community is experienced in terms of shared space (heritage) within which ethnic rules are applied, rules that have currency in the private sphere (family or lineage). This space only partially defines an exclusive territory, and penetrates public, particularly urban, space that is experienced as amoral and separated from the private sphere. For this reason, a non-native is always suspected of mismanagement, and consequently a member of a non-Congolese ethnic group could never lead the country. Mobutu's famous precept, which occurs moreover only in the Lingala version of his well-known speech on the Zairian problem, is the accepted behavior in the public realm: it recommends stealing the state, but intelligently, little by little. This realm is open to generalized predation, and the specific interests of diverse moral communities (whether ethnic or religious) meet and enter into competition there, without being able to tame the public space. The solution to the problem could only come from a federalization that would unite the ethnic groups that exercise authority on their territory instead of uniting autonomous territories, and thus allow for a moralization of public space, to make it into a common good.

Ethnicities as 'First Nations' of the Congolese Nation-State

245

Native and Congolese Ethnicity and Nationality A final fundamental element in the context of political will is representing the Congolese state - especially the Zairian state - as a nation-state. This was so obvious that all observers of the Congolese political situation (myself, BJ. included) neglected to include it in their analyses. It took the political crisis, provoked by the Banyamulenge demand for recognition of their native nationality, for the practical, lived reality of politics and socio-political analysis finally to converge. Now, we must discover the lessons. The following is a brief historical account. Congolese nationality is initially difficult to define because the colonial regime was preceded by the independent and internationally recognized Congo Free State. In this state 'nationality' status existed and was enjoyed by duly registered natives before the period of subjugation of all Congolese as indigenes. Even if registration was not abolished, its application was suspended, and it was only in the 1950s that the civic merit card and the renewal of registration began a return to the recognition of individuals' particular status with regard to the state. The vast majority of Congolese were thus subjugated to the colonial administration as indigenes (natives) governed by customary law or by a system derived from customary law (extra-customary centers). The independence Constitution voted by the Belgian Parliament did not define native Congolese nationality. It was the first Constitution developed by Congolese politicians, from 1 August 1964 (the constitutional reference point since the end of the Mobutu regime), that defined native nationality in terms of patrilineal descent from an ancestor who was a member of a tribe, or a fraction thereof, situated on the colonial territory in 1908. Not until 1981 did Law 72/002 of 5 January recognize the bilateral transmission of Congolese nationality, with reference to the founding of the Congo Free State (1885) rather than that of the colony. The 1999 constitutional project together with the Lusaka Accords maintained this assignment of nationality by belonging, which is situated between the two principles regularly used: those of jus sanguinis and jus soli. In fact, it is descent and residence that establish nationality, but with prior reference to ethnic membership - which is defined nowhere. A government decree (of 26 March 1971) added another criterion for a particular category of the population. Natives of Rwanda-Urundi based in the Congo on 30 June 1960 are recognized as native nationals. The law of 1971 pushed this date back to 1 January 1950 while conserving the reference to territory ('Natives of Rwanda-Urundi, who were in the Province of Kivu before 1 January 1950, and who continued to reside there ever since . . . ' ) . In this way, between 1971 and 1999, two types of Congolese nationality co-existed, each acquired by virtue of a different criterion: one nationality defined by ethnic belonging, and another (for the Rwanda-Urundi natives) defined by individuals' residence at a precise moment in history. Government decree 197 of 29 January 1999 returns to the Constitution of 1964 and standardizes the status of native Congolese nationality by extending to all the principle of assignment by belonging ('Is native Congolese, from 30 June 1960, every person for whom one of his ancestors is or was a member of a tribe based in the territory of the Democratic Republic of Congo, as delimited on 1 August 1885 . . . ' ) . The decree rules out assignment of native nationality by virtue of personal legal status, as proposed in the foreword to the March 1998 Constitution (persons whose ancestors, at least one of them, were on the DRC territory on 30 June 1960,

246

Bogumil Jewsiewicki & Leonard N'Sanda Buleli

with the status of indigene (native) or assimile (assimilated)). Nevertheless, when Article III, section 16 of the Lusaka Accords reaffirms the principle of assignment by ethnic belonging, it again confuses the maps by suggesting that ethnic territory constituted the state space at the time of the Congo's independence. With this brief summary, it becomes clear that the demand for native Congolese nationality by the Banyamulenge was nothing other than a symptom of a more general and fundamental problem. The recognition of native nationality (meanwhile the law imposed several legal impediments on persons who acquired nationality by naturalization) by virtue of ethnic belonging constitutes the real political issue, which refers to the conflict between individual rights and public or community rights taking their place. Decentralization will make this issue particularly interesting and will lead to a more precise definition of the relationship between an ethnic group and a territory, going as far back as possible into the colonial past, if not to 1885. A definition is also needed of the relationship between the individual - who has often settled for two or three generations outside of the region where 'his' ethnic group lived in the rural milieu - and this same ethnic group which has since then become a community that shares the national territory with other similar communities. Finally, a definition is needed of the relationship between the individual and public space, on the one hand, and to the common good of the nation, on the other; this can be done by drawing a distinction between the sphere of private morality and the sphere of amoral renown. The progressive retreat of the state from the public domain, and then its delinquency, would make ethnicity (defining nationality) the main context of political activity, both within the political participation mechanisms authorized by the party-state, and with those against the state. This is the same period when the urban definitions of ethnicity have taken root in the rural milieux; these definitions claim ancient legitimacy (whether from 1908 or from 1885) at the expense of the transformations of the twentieth century. An extremely positivistic perspective sees the ethnic situation during state formation as a truer reality than contemporary practices. This situation is well illustrated by the tribute paid to Leon de Saint Moulin's (1998) cartographic work by Noel Obotela Rachidi (2000: 145): 'The clarity of the maps will allow most of our compatriots to learn more about the ethnic composition of the DRC. The recent conflict in our country has shown that the average Congolese is ignorant of the geography and the ethnic groups within the national borders.' Two further examples focus attention on the dual political usage of the ethnic framework. Within the state sphere, recourse to ethnicity superimposes the political participation of the ethnic elements on to the dictatorial power. Outside this sphere, ethnicity currently serves in the fight for the nation-state, or at least in favor of this symbol of national unity encompassing different ethnic groups.

Student Organizations at the University of Lubumbashi The evolution of student organizations on the University of Lubumbashi campus is an interesting case of a feeling of ethnic belonging that led to the proclamation of territory as a common good and to the acceptance of ethical rules applying to everyone (let us not forget that belonging alone bestows native nationality). 2 Throughout the 1980s, the organizations - some defined region2

The Lega Lusu offers a good example. This organization has existed on all the university campuses of Congo-Kinshasa since 1976, where it reinforces solidarity among Lega students. I was myself (L. N'S. B.) among the founders of the first Lusu of the University of Lubumbashi. This movement sub-

Ethnicities as 'First Nations' of the Congolese Nation-State

247

ally, but experienced through their ethnic dimensions as moral communities were established on all university campuses in the country as sites where the autonomy of student political life could be enhanced, challenging the national organization imposed by the party-state: the Youth of the Popular Movement of the Revolution (JMPR). The organizations were a practical delivery mechanism for the JMPR's activities, and imposed themselves progressively on this organization as a largely autonomous framework of competition for leadership positions. Not all candidates elected by the students were accepted by the authorities, but it became more and more difficult during the 1980s to impose student leaders without taking these elections into account. In fact, the JMPR sadly failed to take a hand in student life and in the political guidance of students at a time when access to university and the conditions for success were strongly defined by ethnic criteria. The proportion of certain ethnic groups in the student body was to be limited, and the Luba in particular were considered over-represented. Regionalization of admissions would in theory favor members of ethnic groups native to the region where the university institution was located. In the end, the precariousness of student life and the exponential increase in corruption within the universities (considered a public good), accentuated the importance of solidarity, defined by a combination of regional and ethnic origins. In contrast to the amoral nature of the public sphere, a moral community's solidarity is the ultimate recourse for survival in a system of generalized predation of the public good. Solidarity also becomes the tool for regenerating the public good into a common good. Thus the university campus prefigures the territory of the nation. Here we resume briefly a detailed presentation of the situation, given recently by a student leader at the University of Lubumbashi, Muela Ngamulume Nkongolo. He refers explicitly to the organization of the policing contingent of the JMPR, the Student Brigade (de facto a police and intelligence service), to explain the birth of the student organization. Arising from the regionalization and tribalization of the Student Brigade, the kasapards (the university campus was situated in Kasapa and the students gave themselves this collective identity) who had no means at their disposal to change the course of events, responded by transforming their various sociocultural organizations into ethno-tribal and regional organizations. The organizations at their inception were particularly concerned about questions relating to the survival of their members. Over the course of the years, however, they gained a very strong influence among the students, and ended up imposing themselves and acting as pressure groups, opposition groups - a concept to view in light of the times - and as real traditional chieftaincies. Their leaders were also considered by the kasapard students as customary chiefs: belonging to the organizations implied at the time a relatively high level of political awareness as well as an ongoing concern to see equality among the kasapards materialize. These organizations expressed the unwavering will of the kasapard students to sequently extended, first to the University and the higher institutes of Kinshasa and then to the higher institutes of Bukavu. These student Lusu are also a site of meditation on the future of the ethnic group, and sometimes a site of debate with other Lusu in the area. Teachers from the respective universities occasionally participate in these Lusu sessions and help expand their activities. In the current struggle, the Lusu that exist outside the territories occupied by the DRC rebellion have turned themselves into a lobby group to support the mai-mai' combatants in the field, notably in the territory of Mewnga, Shabunda and Walikale.

248

Bogumil Jewsiewicki & Leonard N'Sanda Buleli

resist the dictatorship as a group rather than as individuals. (Muela, 2000: 35, emphasis added)

The number of student organizations at the University of Lubumbashi varied during the 1980s between twenty and thirty, while the number of regional organizations regrouping certain ethnic organizations (a region is an administrative subdivision of the country) varied, by virtue of the number of regions, between 8 and 10. Thus the regional organizations, particularly during campaigns for the leadership positions in the JMPR, formed into two cartels reflecting the political division of the country into two blocks, that of the West, made up of the regions of Bandundu, Lower-Zaire and Equateur, considered loyal to Marshal-President Mobutu, and that of the East, the one opposed to him, regrouping other regions of the country. The stance of Upper-Zaire and Kinshasa fluctuated in response to electoral issues. Between 1985 and 1990, the East cartel won all the elections for leadership positions in the JMRP with the exception of the final year. Conversely, the Student Brigade was always led by a student from the West, who was usually Ngbandi (Mobutu's ethnic group). Not all students were members of the organizations, but some of them were members of several organizations at the same time, mainly associations for former secondary school students, and usually designated by the name of the town where the school was located. Because of this, these associations often brought together certain factions of one ethnic organization or another. They were nevertheless more selective, and benefited from the prestige of the former student associations in certain missionary schools, several of which were the breeding grounds from which emerged the political activity and the political parties of the 1950s. Created faculty by faculty, the cultural societies represented the ultimate subdivisions of the organizational structure both for the material survival and for the social and political life of the students of the University of Lubumbashi. The most active and best known society, the Balance, was created by Law faculty students at the start of the 1980s. Its leaders transformed hosting cultural and social reflection into systematic political opposition. Full of the experiences of their organizations, the students reacted publicly to the regime's about-turn in the spring of 1990 - the announcement of democratization, followed by its quasi-cancellation. Muela presents the main causes of the students' desire to take the continuation of democratization from the state in the following way: 'We were ashamed to see our country losing its dignity . . . having no voice in the concert of nations . . . we were conscious of living in a jungle republic . . .' (Muela, 2000: 138-9) The students formed a National Solidarity Union, exercised a sovereign authority on the campus, and put an end to mobutism there from 24 April to 10 May 1990. The 'customary chiefs' (organization presidents) elected a union president to head the movement. The students began cleaning up the campus, repainting the fronts of certain buildings, giving new names to the roads, in short, organizing a universe in which they had a legitimate sovereignty in their eyes, by their own collective will as a community. At this level their movement strongly resembled that of the Dakar youths called setsetal. They also created their own tribunal in Lubumbashi to judge traitors, spies from the regime's security services, who had infiltrated among them. In all these positive expressions of the sovereignty of political will, the delegation of authority passed through the mediation of the organizations for which belonging, like that of the nation, was defined by the dual principle of descent and territoriality.

Ethnicities as 'First Nations' of the Congolese Nation-State

249

The massacre of the night of 11 to 12 May 1990 put an end to this political experiment with student democracy. It is worth mentioning in conclusion that the commandos who attacked the students saw their ranks swell quickly with a number of students natives of Equateur, especially Ngbandi, joining in. Among the attackers themselves identification was by the words lititi . . . mboka which means literally grass, village, but more precisely means village fetiches, and recalls once again the principle of belonging as the key to political identity.

Kivu: Ethnic Organizations and Mai-Mai' Warriors In the Congo, each time the political or economic order creates a general uneasiness, each time the state fails in its mission, ethnic organizations emerge spontaneously, organize themselves into a system of self-defense, and create stabilizing conditions. In urban areas, where 25 per cent of the Congolese live (in cities with more than 25,000 inhabitants according to Saint Moulin, 1998), organizations are born when 'things go badly'. The novelty of the 1980s and 1990s was the organization of political life, and especially the defense of rural land, by these ethnic organizations. These organizations often stem from nongovernmental organizations (NGOs) founded initially for economic improvement: promoting and marketing export agriculture. Their initial objective was to stop the exploitation or to prevent the marginalization of the local populations by businessmen, or by capitalist enterprises on a national or international scale unfamiliar to the ethnic group or the region. We see a convergent movement of organizations in Kivu in the 1980s that leads inevitably to armed confrontation between the big cattle farmers, who were producing butchery cattle for the Kinshasa market, and the local populations who were victims of massive land spoliations. L'ACOGENOKI (the Cooperative Association of Cattle Farmer Groups of North Kivu) was thus dominated by the big cattle farmers, 'Tutsis' of Congolese origin, and was perceived as a Banyarwanda 'organization'. Its members had privileged relationships with the Zairian state and especially with the security apparatus. The explosive land question in North Kivu became the thorniest issue in this region; on both sides it was understood in terms of a political confrontation of an ethnic nature, referring back to nationality. Several large herd owners, often former barons of the Mobutu regime, were accused of not being Zairian, and of acting not only as strangers to the local ethnic groups, whose rights they were flouting, but also as strangers who were pillaging the entire country. In the Masisi, Rutshuru, Walikale and Lubero territories in North Kivu, and those of Fizi, Uvira and Mwenga in South Kivu, the ethnic organizations were born in response to conflicts over land ownership, which were expressed on a political level mainly as conflicts related to ethnic membership and to nationality. The task of these organizations was to defend the rights and interests not only of their members, but of the whole ethnic community. Among the most important organizations were the following: • Kayanda, among the Nande in the Lubero territory. This association, encouraged and supported by the customary nobility, aimed at countering the migration of 'Tutsi' cattle farmers towards Kanyabayonga and further into the region of Masereka. • Bushenge Hunde developed among the Masisi Hunde in struggles against the

250

Bogumil Jewsiewicki & Leonard N'Sanda Buleli

'Tutsi' cattle farmers, notably at Mwesso, Birambizo, Kirotshe and Gungu, where there were many victims of violence. 3 • Magrivi is an association of Hutus of the Rutshuru territory, where a large Tutsi community also live3 notably at Jomba, Katale, Rumangabo and Kibumba. The thinking of Hutu intellectuals 4 and its dissemination aimed to counter the growing influence of Tutsi intellectuals from the Rutshuru territory who were supported by businessmen from Goma. • Umoja was an attempt by the Hutu and Tutsi of Goma to manage relations between the two communities after the conflicts that occurred in Masisi. Its inspiration seems to have been the political authorities, the churches, and the customary chiefs. Several development NGOs - such as Campaign for Peace - promoted their own activities through this organization. • Acuba of the Hunde, Nyanga and Tembo was a coalition of these three communities determined to fight against the 'Rwandan immigrants' of every ethnic group (the Hutu and Tutsi) who were eating away at their land. Acuba aimed at repelling these 'invaders' from their lands. This preceded the mai'ma'i phenomenon. The Banyamulenge, a Rwandaphone Tutsi population of the Itombwe plateau, established the Milima association.5 Created by Monseigneur Jerome Gapangwa, a Catholic bishop from the Uvira diocese, and some intellectuals, among them Muller Ruhimbika and John Mutambo, it aimed to defend Banyamulenge interests. Initially, this N G O promoted potato cultivation - the only profitable agricultural activity in this high altitude zone. The Milima group, as it is usually called, subsequently became the conceiver of the Banyamulenge ideology, and played the role of political representative for this population for a period of time. Side by side with these recent associations, the main objective of which is to promote local interests and self-defense, there are a number of older ethnic organizations. Their political activities extend beyond the homeland of their members, but several are today actively engaged in the mai'-ma'i warrior movement. The organization of the Lega of Kivu-Maniema, called Lusu, brings together in each village and urban center the Lega natives from the Mwenga and Shabunda territories in South Kivu, and from Pangi in Maniema and Walikale in North Kivu. Lusu recalls the colonial period when its ancestor - Unerga or Union of the Warega - attempted to regroup all the Lega around a sociocultural 3 The mwami Bashali of the Hunde attempted without success to calm things down. Meanwhile, Mobutu had sent a military contingent to 'subdue' the insurrectional movement, which was starting to grow in size, and thus radicalizing the opposition (L. N'S. B's interview with mwami Bashali in Nairobi, September 1999). 4 The core of Magrivi was constituted in Bukavu around Sekimonyo, the President of the Provincial Assembly and professor at the Institut Pedagogique Superieur of Bukavu. A number of Hutu personalities from Rutshuru united around him, with a few higher institute teachers among them. With the co-operation of the Nande leaders, including Nyamwisi Muvingi, Magrivi participated in creating a political party, the D C F - Christian Federalist Democracy. A leadership conflict within the association split it into two factions, one of Hutu allegiance, with Sekimonyo as its national president, and the other, of Nande allegiance, presided over by Nyamwisi Muvingi. 5 Milima means mountains or hills. In the cattle herders' culture of the Great Lakes region its symbolic meaning is close to that of the terms lusu among the Lega or lokoshi among the Zimba: signifying hearth, meeting place, a place characteristic of a group, a place of belonging. Nevertheless, the literal meaning of the term describes their habitat, and informed the Banyamulenge's political adversaries that they were claiming the Itombwe plateau.

Ethnicities as 'First Nations' of the Congolese Nation-State

251

structure. Unerga was transformed into a political party in I960, and played an important role in the political demands of this ethnic group, notably when there was an attempt to bring all the Lega from the three provinces together. Lusu is not a political party but intervenes regularly in the political arena. From it originated a political party of the Mobutu transition (see below), the Movement of Solidarity for Development (MSD). 6 The Lega political leaders use this structure either to get elected or to establish their political party among the Lega. 7 The Lusu involvement in the current mai-mai' struggle is evident especially in the territories of Shabunda and Mwenga, which border the Itombwe plateau. 8 The Zimba (or Babinja) of Maniema are regrouped in an organization called Lokotshi, which also has its origins in an older organization, the National Alliance of the Wazimba (ANWA), created in Bukavu in 1960. Officially apolitical, Lokotshi nonetheless plays a very important role in the Zimba ethnic group. It participates in the country's political life through opportunistic alliances to gain advantages for the Zimba. The Zimba currently participate in the mai-mai struggle against the rebel movement of the DRC supported by Rwanda. The Bembe of Fizi currently have several local associations. The Bembe created an apolitical association in 1960 called UNEABFI - Economic Union of the Bembe of Fizi - which played an important role in the political choices of the Bembe populations, and in the success of the Bembe candidates of MNC/L (National Congolese Movement/Lumumba) in the parliamentary and local elections of 1960. It was Bembe warriors who spearheaded the Olenga troops who penetrated into Maniema during the 1964 rebellion. Today, the population is actively involved in the struggle in Kivu, notably in the mai-mai movement, which benefits from the active support of many small associations.9 The first and only attempt to create an ethnic association among the Shi of South Kivu dates from the colonial period and was a failure. L'ESSOBA - the Social Expansion of the Bashi - attempted without success to create a political party: the Union of the Bashi. Today, with the effects of the decay of the 6

The founders of this party were all Lega. In particular, there were Mutuza Kabe (Pangi), Lubaga Wendisoga (Shabunda), Matenda Kyelu (Pangi), Isidore Kampene (Pangi), and Polydore Mamangu (Pangi). The reason for creating the political party seems to be, according to the founders, the participation of Lega representatives in power. In fact, this party produced two ministers: Mutuza Kabe in the Mulumba Lokoji government and Lubaga in that of Tshisekedi from the sovereign National Conference. 7 We cite, among others, Kabongelo who bid for a number of political mandates as a parliamentarian for the Pangi territory. His candidacy was always the object of debate within the Lusu section in the Pangi territory. He also attempted, at the time of the Mobutu transition, to impose the MPR as the political option among the Lega of Pangi. Closer to us is the case of Z'Ahidi Ngoma. When he was imprisoned by Kabila, all the Lusu mobilized to alert the whole Lega community about what would happen to him, and to assure him of their support. His entry into the DRC rebellion, however, confused the country's Lusu and Lega communities. When he left the rebel movement, the Lusu welcomed him and reintegrated him back into the community. He was given a glowing reception by the general Lusu of Kinshasa to mark the return home of this 'lost child'. 8 The involvement of the Lega of Mwenga in the mai-mai struggle recalls their involvement in the resistance to the 1964 rebellion. It is as if each time the spatial entity that the ethnic group considers its own is threatened, forms of resistance are born. 9 There is a large Bembe community at Bujumbura in Burundi and at Kigoma in Tanzania. In these countries the Bembe have developed numerous small-scale economic activities. Moreover, many think of never returning to the Congo. In the Bembe's support for the mai-mai, some see the hidden hand of d'Anzuluni Bembe, political leader of Fizi and a former president of the Congolese Parliament in the MPR period. His visits to Kigoma and his contacts with the Bembe exile community in Tanzania may justify this hypothesis (interviews by L. N'Sanda Buleli with Musemakweli in Nairobi, September 1999).

252

Bogumil Jewsiewicki & Leonard N'Sanda Buleli

Congolese state, and often with the support of numerous development NGOs, several small associations have been born. Even if the Shi political resistance has remained vigorous for nearly ten years, there is nothing to confirm that they are physically engaged in the mai-mai movement, which is an ethnic phenomenon primarily associated with the Bembe, Lega, Zimba, Nyanga and Hunde. The ethnic association of the Kusu-Tetela, ANAMONGO, is one of the most solid and best structured of Kivu-Maniema. The term ANAMONGO means child of Mongo, but the prefix ana (child of) is also the acronym of the National Alliance of the Mongo, an association founded in Lodja in March 1960. ANAMONGO played an important role in Congolese politics in 1960-65, notably by its support of the MNC/L and its local leaders from Sankuru, Maniema and Bukavu. The involvement of the Tetela in the mai-mai armed opposition to the DRC rebel movement and the Rwandan army seems to have begun at the time when these lines were drawn. Another recently created association in Maniema, Association of Natives of Kibombo, Samba and Lubunda (ADERKISAL), is evolving parallel to ANAMONGO. Created in 1993 in Kindu, its goal was to support the Kusu leaders Thambwe Mwamba and Ngongo Luwowo. The Maniema Kusu do not seem to participate in the mai-mai struggle. 10 It is generally admitted that the mai-mai movement is a phenomenon specific to Kivu, in particular the conflict zones of Kivu-Maniema affected by the land and nationality struggles. There the mai-mai warriors clash with Rwandan troops of the APR (Rwandan Patriotic Army) and Rwanda-supported Congolese rebels of the DRC/Goma. Although the mai-mai phenomenon 11 may suggest a homogenous movement united around a nationalist ideal of fighting 'strangers' who have invaded the national territory, on closer examination we see that there is no structured or unified movement. It is above all an ethnic resistance that regroups many local elements, and where each one reproduces its own means of mobilization, often linked to local history and memory. The current movement originated in the defense of 'ethnic' lands and was shaped at the political level by the organizations at the end of the 1980s and the start of the 1990s. The movement also recalls the 1960s rebellion as much as the older tradition of local resistance at the end of the nineteenth and beginning of the twentieth centuries - a resistance that was supported ideologically and ritually by political and ritual ethnic authorities ('healers'). The most recent movement of this type seems to be starting in Sankuru, and probably expresses it best. The movement is called AHOKA WA N G O N G O LETETA (the warriors of Ngongo Leteta, a Tetela late nineteenth-century chief). The father of the leader (a local intellectual whose profession cannot be divulged without putting his family at risk) was a colonel during the rebellion of 1963-64, who was subsequently assassinated by 10 The participation of Thambwe Mwamba in the DRC rebellion and his capacity to block the struggle against this rebellion could be the main explanation for this. On the other hand, many local leaders who make up the civil society of Maniema exercise political responsibility within the rebel government in Goma. 11 Water (mai, in Swahili, a common language of East Africa, means water) plays an important part in the symbolic construction of opposition movements in East Africa and in eastern Congo. Today the mai-mai" use the ritual of sprinkling water containing protective substances to persuade the combatants that they are invulnerable to their enemy's bullets, which are transformed, it is said, into water. Mai-mai is sometimes used as a war cry that should also produce this effect. The ritual and the term mai-mai make appeals to collective memory and situate the struggle for ethnic sovereignty in the tradition of the previous movements.

Ethnicities as 'First Nations' of the Congolese Nation-State

253

Mobutu's men. All the local chiefs bear the names of the former officers of Ngongo Leteta (Lopaka, Kandolo, Mulosa, Kipoke, etc.). 'They make life difficult for the occupiers. Nevertheless, they have no modern weapons, otherwise the situation would have changed already in Sankuru, since the enemy is very isolated' (Ndjate Alphonse, 2000). Kabuya-Lumuna's remark about the rebellion is still relevant: 'all these insurrections, that we call "rebellions" against the established order, happened with and within the mobilization of tribal nationalism reorganized by calling upon discourses of universalist mobilization'' (Kabuya-Lumuna, 1986: 13, emphasis added). The mai-mai reacted at first against the threat to their 'national' tribal space. Taking this objective into account in the definition of the conditions for acquiring native Congolese nationality provides a better understanding of the apparent U-turns of the mai'-mai' in 1996-2000; they were allied at times with the central authority, and at others with the movements fighting against it. The mai'-mai movement of North Kivu was born as an initiative of the customary chiefs, the mwamis of the Nyanga and the Hunde in the Walikale and Masisi territories. In an interview with Laurent Kabila, the mwami Akilimali told him: It was my friend Bashali here and I who created our own mai-mai groups to resist the invasion of our lands by the Banyarwanda, when President Mobutu let us down. We mobilized our youth to take up arms and we asked our magicians and fetishers to make some 'dawa' to help them fight better against the invaders. Today our young combatants are fighting to the limit of their powers and receive nothing from you. Look, in February our young mai'-mai occupied the city of Goma for seven whole hours. We thought that you would send us some reinforcements and we waited for your men, but in vain. And when the Rwandan forces attacked with their tanks our young combatants retreated to avoid sustaining losses. 12

Ethnic leaders from the old rebellion of 1964-7 organized the struggle in South Kivu, notably in the territories of Fizi, Mwenga, and Shabunda. There are in particular among the Bembe 'generals' Padiri, Dunia, Lwetsha and Sikatenda. After the defeat of the Rebellion, all four participated in the Itombwe underground, then led by Kabila. With the arrival of the AFDL (Alliance of Democratic Forces for the Liberation of Congo) in 1996, only Sikatenda rejoined Kabila, while the others, notably Dunia, continued to fight against the 'anti-Bembe' forces in Itombwe. 13 Among the Zimba in the Kasongo territory it was 'general' Raphael Milambo, native of the Mulu collectivity, who led the 12 Interview between the civil society delegates of Kivu-Maniema and President L.-D. Kabila during an audience accorded to them on Sunday 9 October 1999 in Lumumbashi. This information was received from Gaspard Mugaruka, charge d'affaires for the Congo in Kenya, who was present at the interview. 13 After the conquest of Maniema by the AFDL in April 1997,1 (L. N'Sanda Buleli) accompanied the new Governor of Maniema, Pierre Lokombe, on a trip to the Kambambare territory. We heard complaints from the Buyu about the nibbling away at and the state of their lands following the clash between the Bembe of 'general' Dunia and the Banyamulenge. The Buyu are militarily opposed to the Bembe, who, suffering from the pressure of the Banyanmulenge in Itombwe, began descending towards the Buyu lands. The wariness of Kabila towards the Bembe 'generals', his former underground companions, makes the Congolese central government reluctant to support the Bembe maimai fighters. Kabila's nomination of 'General' Lwetsha to the head the general staff of the Congolese Armed Forces (FAC) is more a matter of political maneuver than of concern for efficacy in the struggle against the rebellion supported by Uganda and Rwanda.

254

Bogumil Jewsiewicki & Leonard N'Sanda Buleli

mai-mai movement. 14 Following the example of the Bembe leaders, he also participated in the rebellion of 1963-4. In 1967 he started an underground in the Mulu forest, between the villages of Kampene and Bikenge, that was quickly reabsorbed following negotiations with Mobutu's soldiers. Among the Lega the mai-mai resistance was organized around the customary chiefs: the mwami Longani, in the collectivity of the Basimwenda (Mwenga country) and the mwami Moligi, in the collectivity of the Basile (Kalole country). In May 1999, a combined offensive of Zimba and Bembe mai-mai approached Kindu in Maniema after having invested the Lega groups from Kampene, Kayuyu, Kingombe-Bari and Samweli. Initiated by the Zimba and Bembe combatants, the Lega created their own mai-mai resistance movement, which is operational to this day in the collectivities of Babene and Bakabangu. Each ethnic group has its own zone of operations that corresponds to its 'ethnic' territory, defined with reference above all to the 'ethnographic' studies and maps established by the former colonial administration. Thus the Bembe are fighting to liberate and defend their native Fizi and the Itombwe plateau, just as the Lega are doing in Mwenga and Shabunda, the Zimba in Kasongo and Maniema, the Nyanga in Walikale, and the Hunde in Masisi. The mai-mai do not always act in a scattered manner. Without speaking of co-ordinated actions, there are nevertheless means of communication that allow them to lead certain conceited actions: for example, the large offensive towards Kindu in May 1999. This was led by Zimba and Bembe combatants joined by deserters from the Congolese Armed Forces dispersed in the Maniema forest (poorly equipped and often unpaid, many leave the army to live off the land). Common objectives and co-ordinated actions are certainly the fruit of the relationship with the urban ethnic organizations, several members of which live outside the country. Use of cellular phones and the Internet now allow relatively regular communication between these diverse elements of the ethnic 'nation'. Their objectives are not necessarily in perfect harmony, and the warrior movement does not seem to be subject to a political leadership located outside the combat zones. Nevertheless, we can note the effects that matching military actions has on political objectives. This is fairly clearly the case with military actions that aim to upset the artisanal exploitation and marketing of alluvial gold, since it only profits the enemy.

Conclusion Neither the central Congolese government of Kabila, nor the successive opposition movements (the AFDL in the 1990s or the DRC in 1998 and 1999), have been able to impose leadership on or to unify the mai-mai. Only an incorporation of individuals in one or other of the structured armed forces would have succeeded in this unification. In any event, this warrior movement of the maimai, as opposed to what was perceived as 'foreign' control of the 'national' ethnic territory (the central government could also be seen as foreign as could a neighboring state), would lose all local legitimacy if it were integrated with a military force under an authority other than that of the ethnic group (customary chief, ethnic organization, local military leader). The current attempt by Kabila 14 I (L. N'Sanda Buleli) found 'General' Milambo in Mingama in 1997 during a visit by Governor Lokombe. In the unrest provoked by the arrival of the AFDL, which drove away the Mobutu army, he ran as a candidate for the position of chief of the Mulu collectivity, but failed to get elected. Disappointed, he allied himself to his former companions Dunia and Lwetsha and returned to the underground.

Ethnicities as 'First Nations' of the Congolese Nation-State

255

to reunite the mai-mai under the umbrella of Popular Forces of Self-defense is undoubtedly doomed to fail. Let us end by drawing attention to the desire to exercise state-like sovereignty on territory that is perceived as a base of political and economic autonomy by a group defining itself in ethnic terms. These terms are undoubtedly expressed first and foremost in the language of kinship (without necessarily referring to kinship in fact, since many native Katangese do not share any kinship ties). With the apparent use of tradition to legitimize contemporary political solidarity, we see instead a return to what the 'modern' constitution defines as a principal political link. The same goes for the desire to affirm the political appropriation of territory, an indispensable heritage for the exercise of the sovereignty prerogatives needed to ensure the autonomous economic survival of a group. We note in conclusion that even students at the University of Lubumbashi defined themselves in group terms made up of an alliance of 'ethnic' assemblies — the kasapards led by 'customary chiefs' - that delimited the territory on which it demanded its own sovereignty: control of entry, control of traffic, changing the street names, etc. Since its emergence is already a fact, are we witnessing the affirmation of ethnic groups as 'first nations' of the statized nation? What would be the new rights and new duties of these entities, which make demands that the state and the nation should respect, since they gave the nation birth?

Bibliography Braeckman, C. 1999. Uenjeu congolais. UAfrique centrale apres Mobutu. Paris: Fayard. Bucyalimwe Mararo, S. 1993. 'Land Conflict in Masisi, Eastern Zaire : the Impact and Aftermath of Belgian Colonial Policy (1920-1989)', Ph.D. dissertation, Indiana University, Bloomington, IN. Chretien, J.-P. 2000. UAjrique des Grands Lacs. Deux milles ans d'histoire. Paris: Aubier. Dibwe dia Mwembu. 1999. 'L'epuration ethnique au Katanga et ethique du redressement des torts du passe', Canadian Journal of African Studies, 33 (2-3): 483-99. Jewsiewicki, B. 1984. 'Les pratiques et l'ideologie de l'ethnicite au Zaire: quelques reflexions', Canadian Journal of African Studies, 18: 103-7. — 1989. 'The Formation of the Political Culture of Ethnicity in the Belgian Congo 1920-1959', The Creation of Tribalism in Southern Africa, ed. L.Vail, London and Berkeley, James Currey/University of California Press: 324-49. — 1999. 'Afrique centrale francophone entre au XXIe siecle: Citoyens et Etats, nations, regions face au monde', Canadian Journal of African Studies, 33 (2-3): 221-30. Kabamba Nkamany, A. 1997. Pouvoirs et ideologies tribales au Zaire. Paris: L'Harmattan. Kabuya-Lumuna, S. 1986. 'Les aspects ethniques des rebellions' in H. Weiss and B. Verhaegen (eds), 'Les rebellions dans l'est du Zaire' (1964-1967). Cahiers du CEDAF, 7-8. Kapeta Nzovu Ilunga. 2000. 'La question de la nationality congolaise et les perspectives de paix en R . D . C , in Sabakinu: 117-35. Lovens, M. 1974. 'Le revoke de Masisi-Lubutu (Congo-Beige, January-May 1944)'. Cahiers du CEDAF, 3-4, Brussels. Mahano ge Mahano. 1998. Existe-t-il des Rwandais Congolais} Kinshasa: Editions Sophia. Mathieu, P., P.-J. Laurent, T. Mafikiri et S. M. Mugangu. 1997. 'Competition fonciere, confusion politique et violences au Kivu: des deribves irreversibles', Politique africaine 67: Mathieu, P. et Mafikiri Tshongo. 1998. 'Guerres paysannes au Nord-Kivu (Republique democratique du Congo), 1937-1994', Cahiers d'etudes africaines, 150-152: 385-416. Mbembe, A. 2002. On the Postcolony. Berkeley, CA: University of California Press. Muela, Ngalamulume Nkongolo. 2000. Le campus martyr. Lubumbashi 11-12 mai 1990. Paris: L'Harmattan 'Memoires lieux de savoir' series. Mulambu Mvuluya. 2000. 'Democratic et paix face aux conflits interethniques', in Sabakinu: 95-102. Mwena, Batende. 1978. Mouvements messianiques et protestation sociale. Kinshasa: Editions sociales. Ndaywel e Nziem, 1.1998. Histoire generate du Congo, de Fheritage ancien a la Republique democratique.

256

Bogumil Jewsiewicki & Leonard N'Sanda Buleli

Gembloux-Paris: Duculot. Ndjate Alphonse. 2000. 'Le reveil des Batetela' in www.congo2000.com/forum, rubrique 'Congo agresse', posting of 21 March. N'dom, C. 1979. 'Utilisation des procedes d'initiation et d'imunisation a caractere magique par les "Simba" du Zaire en 1964' in H. Weiss and B. Verhaegen, eds, 'Les rebellions dans Test du Zaire' (1964-1967). Cahiers du CEDAF 7-8: 79-86. Nora, P., ed. 1997. Les lieux de memoire, Paris: Gallimard, 'Folio'. Translated into English as Realms of Memory. Rethinking the French Past. New York: Columbia University Press. Ntumba Luaba. 2000. 'Grandes puissances et avenir de la Republique Democratique du Congo', in Sabakinu: 219-38. Obotela Rashidi, N. 2000. 'Les derives identitaires: un danger pour la paix et la democratic en Republique democratique du Congo', in Sabakinu: 137-47. Pabanel, J.-L. 1991. 'La question de nationality au Kivu', Politique africaine, 41: 32-40. — 1993. 'Conflits locaux et strategies de tension au Nord-Kivu', Politique africaine, 52: 132-34. Reyntjens F. et S. Marysse, eds. 1996. Conflits au Kivu, antecedents et enjeux. Antwerp: University of Antwerp. Ruhambika Manase. 2001. Les Banyamulenge entre 2 guerres, Paris: L'Harmattan, 'Memoires lieux de savoir' series. Sabakinu, Kivilu, ed. 2000. Democratic etpaix en Republique democratique du Congo. Kinshasa: Presses de l'Universite de Kinshasa. Saint Moulin, L. 1998. 'Conscience nationale et identites ethniques. Contribution a une culture de la paix', Congo-Afrique, 330: 587-622. Sikitele Gize. 1983. 'Histoire de la revoke de Pende. 1931', Ph. D. dissertation, University of Lubumbashi. Verhaegen, B. 1984. 'Les associations congolaises a Leopoldville et dans le Bas-Congo, de 1944 a 1958'. Etudes africaines du CRISP, Brussels, No. 112-13. Verhaegen, B. 1967. Rebellions au Congo T. II Maniema. Brussels-Kinshasa: CRISP-IRES. Willame, J. C. 1997. Banyarwanda et Banyamulenge. Violences ethnies et gestion de Videntitaire au Kivu. Cahiers africains, Brussels-Paris: Institut africain-L'Harmattan. — 1997. 'Zaire: Etat de crise et perspectives futures (fevriel)' in www.unhcr.ch/refworld/country/ writenet/wrizairf. htlm. — 1998. 'Kivu: la poudriere' in C. Braeckman et al, Kabilaprendlepouvoir. Brussels: GRIP-Editions Complexe.

15

A. RAUFU M U S T A P H A

Ethnicity & the Politics of Democratization in Nigeria

Introduction

A

s Obasanjo was being sworn in as President of Nigeria on 29 May 1999, Ijaw and Itsekiri ethnic nationalists were engaged in an orgy of inter-ethnic warfare resulting in the death of 50 people. Between May 1999 and February 2002, over 10,000 people are estimated to have lost their lives in 33 reported ethnic, religious and regional conflicts across Nigeria. Clearly, the politics of identity has become a central feature of Nigerian democratization. Analytically, we can isolate three strands in this interaction between ethnicity and democratization in Nigeria. First we can examine the complex relationship between ethnicity, the military institution, and the authoritarian regime. A second strand focuses on the actual processes of democratization between 1986 and 1999, including the various reversals therein. A third strand concentrates on the impact of ethnicity on the consolidation, or destabilization, of the democratic experiment initiated on 29 May 1999. This chapter concentrates on the third strand, while drawing on aspects of the first two. It is obvious that the escalating level of sectarian violence since May 1999 poses a serious threat to democratic rule and national unity in Nigeria. On the face of it, it is puzzling that there should be such a resurgence of sectarian animosity under the democratic dispensation. Some suggest that democracy creates a Vent' for the pent-up frustrations accumulated under 15 years of military authoritarianism. While there is some merit in this argument, it obscures, however, the fact that many of these violent conflicts predate the democratization process. What we need to explain in these cases, is why democracy has failed to provide non-violent avenues for conflict resolution. Other conflicts, such as those involving the Yoruba militia, the O'odua Peoples Congress (OPC), and the ethno-religious conflict over the extension of the Islamic Sharia legal system, are directly traceable to the process of democratization. Here, we are faced with the possibility that democratic openings may give rise to the emergence of anti-democratic forces. Ethnic sectarianism is a real threat, not only to Nigerian democracy, but also to the territorial integrity of the country. But this threat is little understood and all too often sensationalized. In this analysis, I argue that it is one-sided to emphasize the inter-ethnic sectarian processes that threaten Nigerian democracy and unity. For a balanced view, we also need to pay attention to intra-ethnic disagreements and confrontations. While not underestimating the threats, we should at least aspire to a comprehensive and dynamic analysis. In the following section, I set out a brief political and historical context for the transition from military authoritarianism in 1999, while the third section looks at the patterns of 257

258

A. Raufu Mustapha

inter-ethic mobilization that has taken place since then. I then examine the core issues in contention in inter-ethnic mobilization., followed by an explanation of the theme of intra-ethnic divisions. The final section concludes the analysis by focusing on the connection between the state, hegemonism, xenophobia and democracy in Nigeria.

Trajectories and Transformations Certain patterns of inter-ethnic conflict had emerged in colonial and post-colonial Nigeria, the best known of which are the ethno-regional conflicts which emerged from the decolonization period of the 1950s. These conflicts were between political blocs constructed around specific ethnic, religious and regional elements: from the Hausa-Fulani, parts of the Kanuri, and other Muslim communities of the North; the Igbo in the mainly Christian East; and the Yoruba in the West. If the main conflict revolved around the hegemonic ambitions and fears of these three political blocs, the sub-text was the conflict between the 'majority' ethnic group and the numerous 'minority' ethnic groups within each region. Beyond these two fault-lines, it is possible to identify several others, all shaped by changing constitutional arrangements and territorial divisions. It was the recognition of the primary hegemonic confrontation that led to the adoption of federalism in the terminal colonial period, with minority interests and demands being largely ignored. Though each region had substantial powers, the competition between them for the control of a fragile centre generated enormous tensions. It was the inherent instability of this federal system that led to the collapse of the First Republic in 1966. Since then, attempts at political engineering have sought to reconstruct the foundational architecture of Nigerian federalism. The three regions of 1939-64 and the four regions after 1964 have been successively broken up by military governments into 12, then 19, then 21, then 30 and now into 36 states. The federal character principle, which makes it mandatory to reflect the territorial and ethnic composition of the country in state institutions, has become entrenched in the constitution. Other nationalistic policies, such as the creation of the National Youth Service Scheme and the Nigerian Television Authority are expected to complement the structural changes. Post-1966 Nigeria also witnessed an intense process of political and fiscal centralization, leading to the subversion of the very federalism to which military reformers had committed themselves. The oil boom from 1970 meant that the central government could be freed from reliance on agricultural resources that were closely connected to regional sources of power. A rentier state emerged, freed from the constraints of fiscal and political accountability. The central state increasingly dominated other levels of governance; the creation of numerous states reliant on federal grants meant that no state government could assert itself against an over-bearing federal government. The accretion of federal power at the expense of the states was completed by the centralized and hierarchical nature of the military institution; state-level military governors were under the direct control of their superiors at the centre. These changes had important consequences for the organization of the state and the development of ethnic identities. Awash with oil revenue, the central state became the centre for accumulation and class formation - a prize to be fought over within the distributional logic of the rentier state. The extant ethnoregional majority blocs reconstituted themselves around new symbols. The

Ethnicity & the Politics of Democratization in Nigeria

259

tripodal ethno-regional blocs persisted, even when their institutional anchors in the regional governments had disappeared. With the return to electoral politics in 1979, the old patterns of party formation emerged, complete with their erstwhile ethnic colorations. However, two important changes took place after 1966, developing more rapidly in the course of the 1990s. Firstly, minority ethnic groups increasingly asserted their interests and separate identities. In some instances, it was a case of minority assertion against majority tutelage; in others, it was a case of minority assertion against another minority. In yet others, as in the Niger Delta, minority ethnic groups reviewed and reversed their erstwhile cosy clientelism with the central state. Under the conditions of military authoritarianism, this assertion of minority rights increasingly assumed a violent character. From the 1970s we can chart an increase in the incidences of communal clashes amongst ethnic minorities in the Middle Belt, the southeast and the Niger Delta. A second post-1966 change was the acceleration of the concentration of power at the centre. The powers of all state governments were virtually collapsed into those of the federal government, while those of the central government became increasingly collapsed into the military institution at the head of the government, Furthermore, the military was increasingly seen by its southern opponents as a 'northern' or 'Hausa-Fulani' institution. When the Babangida junta annulled the presidential elections of June 1993, it was interpreted as the thwarting of a southern presidential candidate by the northern military and political establishment. Even in the early 1980s, Nigerian federalism still had some life to it; the civilian government in Bendel State took the federal government to court over the revenue allocation formula and won. After the 1983 military takeover, the centre ruthlessly imposed its will on the states; and this centre was increasingly seen as northern-dominated. One southern commentator alleged that many important parastatals became northern-controlled {Guardian (Lagos), 30 January 2000), while a northern commentator pointed out that over the 15 years of military dictatorship: . . . the North simply went berserk in terms of appropriation of government machinery for itself. It was so clear and brazen . . . The whole of the South had one grouse or the other, and they tried to express themselves but they were continuously being denigrated as troublemakers. (Ahmadu Abubakar, Tell, 6 September 1999)

The process of democratization was influenced by the counter-claims of the southern elite - particularly the Yoruba elite. Though this was largely a confrontation between the majority ethnic blocs, it had important minority subcurrents. Democratization became marked by the vociferous demand for a 'power shift' to the south and the related demand for a Yoruba presidency. When Nigerians finally went to the polls in 1999, they were confronted with two presidential candidates, both of whom were Yoruba. Obasanjo, the candidate supported by the wider ethnic alliance, won. Lack of consensus, growing conflict, and continued polarization are the context for this process of democratization. Tables 15.1 and 15.2 illustrate the ethno-regional mobilization and voting patterns in the two major elections held in the course of the transition. In both 1993 and 1999, the Hausa-Fulani in the northwest and the Yoruba in the southwest tended to concentrate their votes for those parties and candidates closely associated with their respective hegemonic blocs.

A. Raufu Mustapha

260 Table 15.1 June 1993 presidential election: party support by ethno-regional/geo-political zone. Zones Northwest Northeast Northcentral Southwest Southeast Southsouth F C T (Abuja)* Grand Total

Reg. Voters 9,549,115 5,460,079 7,743,176 4,673,175 5,886,294 152,686 38,353,662

% Turnout

SDP Votes (%)

NRC Votes (%)

25.4 42.3 47.1 32.7 43.3 25.1 37.0

42.7 44.3 56.1 84.9 48.3 51.3 52.1 58.0

57.3 55.7 43.9 15.1 51.7 48.7 47.8 41.9

Source: Adapted from Akinterinwa, 1997, who composed the figures from the National Electoral Commission, SDP Secretariat, and international observers' figures.

Table 15.2 February 1999 presidential election: party support by ethno-regional/geo-political zone. Zones Northwest Northeast Northcentral Southwest Southeast Southsouth F C T Abuja Grand Total

Reg. Voters 13,494,968 7,992,211 7,645,845 11,918,421 7,658,424 8,843,677 385,399 57,938,945

% Turnout 42.5 63.3 66.6 46.3 43.3 61.8 25.7 52.3

AD-APR (%) PDP (%) Falae Obasanjo 29.2 27.9 27.9 79.2 29.5 22.1 40.2 37.2

67.8 70.2 71.0 19.8 69.5 77.4 59.8 62.7

Source: Constructed from official data held by the Independent National Electoral Commission (INEC), Abuja, August 2000.

Democracy and Inter-Ethnic Conflict Widespread disenchantment with the Abacha regime called for serious reform of governance, and in the course of his campaign for office, Obasanjo promised to fight corruption, promote due process and the rule of law, and eradicate the culture of impunity. He started a process of re-professionalizing the armed forces by dismissing all those with a past record of involvement in political offices. A good number of these are from the north. He sought to appoint professional-minded officers to the leadership of the armed forces, ostensibly based on merit. Though many of these appointees are from the north, none was from the core Hausa-Fulani far north that had been prominent within the military establishment. Obasanjo also shook up the civil service, dismissing scores of officials and promoting many from non-hegemonic areas into prominent positions. The rampant cronyism of the Babangida and Abacha periods was increasingly called into question. Investigative panels and commissions were set up to probe past cases of human rights abuses and corruption. Though much of what Obasanjo has done falls within his election promises, there has been no consensus supporting its actual implementation. Firstly, the

Ethnicity & the Politics of Democratization in Nigeria

261

democratization process was not based on agreed fundamentals about the new mode of governance. Secondly, each ethnic and ethno-regional group interpreted the changes in terms of their own narrow political and economic interests. These reforms have therefore led to the intensification of the ethno-regional polarization already implicit in the voting patterns. As a result, four distinct perspectives on democratization in Nigeria have emerged. The first perspective is from the political bloc constructed around the HausaFulani north. Even before the military had left the stage, squabbles broke out over the sharing of political offices within Obasanjo's People's Democratic Party (PDP). A leading pro-north newspaper, Today, announced that northerners were 'alienated' by the process. Today paraded a list purporting to show the sharing of ministerial, legislative and bureaucratic positions between the six geopolitical zones of the country. Though much of the detail of this list turned out to be false, the impression nevertheless gained ground that there was a plot: . . . carefully designed to frustrate the North and render it politically ineffective . . . What has befallen the North is a carefully crafted agenda aimed at rubbishing the indices of power [of] the North . . . (Today> 25 April - 1 May 1999] Today blamed the situation on the failure of northern politicians to negotiate the terms of the 'power shift'. The stakes were raised when northern politicians from the northwest zone - the 'core' of the core north - insisted on having the Presidency of the Senate conceded to their zone, the Vice-President being from the northeast zone. They argued that since the executive is under Obasanjo, a southerner, the north ought to have the leadership of the legislative arm of government. They pointed out that a similar arrangement was adopted in the last civilian government between 1979 and 1983. The political bloc constructed around the Igbo elite within the party resisted the demand. In their view, the Yoruba had the Presidency while a northerner was effectively the number-two man as his vice-president. The Igbo bloc demanded the number-three slot, the Senate Presidency, as a right, given the extant tripodal framework of hegemonic contestation. In the end, the Senate Presidency was allocated to the Igbo southeast zone, to the chagrin of northern politicians. Obasanjo's reforms were soon confronted by a growing tide of disaffection within northern political circles. He was accused of being in cahoots with the Christian Association of Nigeria (CAN); according to Tanko Yakassai, Obasanjo 'is there to do the bidding of CAN' and 'the government is gradually being handed over to CAN' (Guardian, 27 June 1999). Wada Nas, a controversial minister in the Abacha junta, added his voice to the complaints. Obasanjo was accused of implementing a 'hidden agenda' of his Yoruba ethnic group to take over political power. Nas also pointed out that the Yoruba already controlled a disproportionate number of federal bureaucratic positions and more than their fair share of the national economy. Nas therefore raised the spectre of a totalizing Yoruba hegemony. The two zones of the core north, the northwest and northeast, were said to be deliberately disadvantaged by appointments favouring Yorubas and Christians, some of whom are from the third northern zone, the northcentral. Another influential voice from the north, Mahmud Tukur, joined the fray, arguing that Nas 'essentially reflected public opinion in much of the cultural North'. He added that: . . . Northerners were initially of the view that whoever they supported would show appreciation . . . This, I am afraid, is not the impression they get from the initial activities of the Obasanjo administration. (Today, 11-17 July 1999)

262

A. Raufu Mustapha

This disaffection gained hold on northern society as violent sectarian confrontations increased in occurrence. What are the central elements of the north bloc's complaints? Firstly, that Obasanjo obtained the Presidency on the strength of northern votes, but has since failed to 'show appreciation'. Obasanjo's Yoruba brethren have often denounced him as a 'northern stooge' during his leadership of the country in the period 1975-79. This was the reason for their rejection of him in 1999. Secondly, it is alleged that Obasanjo's reforms have disproportionately disadvantaged the north. It is claimed that nine out of eleven permanent secretaries initially retired were from the north (Albashir, Today, 25-31 July 1999); that the purge of political officers in the armed forces was aimed at the north, there being only 5 'northerners' left out of the 45 brigadier-generals in the army [Balarabe Musa, Post Express, 13 October 1999); that the retirements are 'aimed at removing all officers of northern origin so as to cripple the political power of the north' (Hassan Sani Kontagora, Sunday Guardian, 27 June 1999); that it is unbelievable that no core northerner is in operational control of any arm of the armed forces; and that many northern heads of parastatals have been removed from office. Thirdly, it is alleged that Obsanjo was pandering to the Yoruba who had spurned him at the polls. This accusation has since congealed into the idea that Obasanjo is a willing part of a wider Yoruba agenda to add political power to their established dominance in the bureaucratic and economic spheres (Nas, Abuja Mirror, 20-26 October 1999). The transfer of the head offices of four maritime parastatals back to Lagos on the coast from Abuja is held up as an attempt to subvert the transfer of the capital to Abuja. Obasanjo is also accused of not cracking down on the Yoruba militia, the OPC, blamed for instigating riots against northerners in Lagos. Fourthly, attention is drawn to alleged anti-northern statements from sections of the southern press. For example, when sections of the southern press accused Buhari, first Speaker of the House of Representatives, of falsifying his academic credentials, it is claimed that 'the orchestrated media attack on Buhari was a ploy by the southern press to rubbish the former speaker, a northerner'. (Post Express, 26 July 1999). It is pointed out that when similar allegations surfaced in connection with the Yoruba Governor of Lagos, Ahmed Tinubu, the southern press was lukewarm in its condemnation and campaigning. The differential treatment was held up as a sign of unyielding hostility. The second perspective on the democratization process is from the political bloc built around the Igbo. The main complaint of this bloc is that the Igbo have been 'marginalized' from power since the end of the civil war in 1970. Because Igbo officers were either killed in the July 1966 counter coup, or joined the secessionist Biafran army in 1967, the number of Igbo officers in the federal armed forces fell drastically. Since the end of the war, only a few Igbos have risen to senior command positions and they are often retired before this. Within the military institutional dynamic, the Igbo have been largely underrepresented. On the whole, civilian regimes have mainly been based on an alliance between northern and Igbo political forces, while military regimes have often been based on an alliance of northern and Yoruba officers. Prolonged military rule from 1983 therefore accentuated Igbo feelings of'marginalization', and democratization was welcomed as an antidote. Though Igbo politicians were disappointed by the PDP's failure to endorse Ekwueme's candidacy, they still threw their support behind his rival, Obasanjo, garnering massive votes for him in the southeast in the 1999 election.

Ethnicity & the Politics of Democratization in Nigeria

263

When Obasanjo began establishing his government, the Igbo bloc became despondent because they felt that their marginalization had continued, despite their overwhelming support. It is argued that inconsequential ministerial posts were given to the Igbo, while the 'juicy ones' went to the Yoruba and the north. Even the highly respected Ekwueme added his voice to these complaints, pointing out that the distribution of posts in Obasanjo's cabinet was not equitable and that it did not reflect the decisions of the PDP leadership (Post Express, 7 July 1999). This particular development is reminiscent of the military era when Igbo ministers were regularly shunted into the Information and Culture, Justice, and Science and Technology portfolios with little power of patronage. It is even speculated that newly appointed Igbo ambassadors are being largely assigned to war-ravaged countries! Igbo leaders ask why no Igbo has been put in command of any arm of the armed forces and why there is no Igbo on the National Defence Council. The Igbo bloc increasingly see the Obasanjo presidency as the prize 'won' by the Yoruba bloc for their tenacity and courage in the face of the 'Hausa-Fulani' military state (Agbagha, Post Express, 8 July 1999). The 'lesson' drawn was made clear by a controversial minister of Information under Abacha: Mr. Uche Chukwumerije has said that unless the Igbos come together to articulate a common position, there will be no end to their cry of marginalisation . . . He remarked that nothing has changed in the present Nigeria political dispensation, rather what has happened . . . was that power has been transferred from the Hausa-Fulani to the Yoruba hegemonists. (Post Express, 15 December 1999) In a similar vein, the former leader of secessionist Biafra, Ojukwu, claimed that: Nothing has changed . . . At the end of the war, we were a caged people . . . Today, it is not like that. We are no longer in the sealed cage but somehow a glass cage has been put around us . . . (Guardian, 13 January 2000) The Igbo bloc's response has therefore been the attempt to develop the elite pan-Igbo institution, the Ohaneze, into the umbrella organization for Igbo interests. The third perspective is that of the political bloc constructed around the Yoruba elite. Without doubt, this grouping suffered most under Abacha's tyranny on account of its unequivocal opposition to continued military rule after the annulment of the 12 June 1993 presidential election, won by Abiola, a Yoruba. Abiola had been a sworn political enemy of this bloc, but when he was denied his mandate, the bloc rallied solidly behind him, claiming that the annulment underlined the fact that other Nigerians were being denied equal access to power because of northern control of the military and the government. Though Yoruba support for Obasanjo has increased since 1999, there are still many powerful elements within this bloc who continue to regard him with deep suspicion. Many disagree with some of his policies, particularly his refusal to call a Sovereign National Conference, while others regard him as being a hostage of the northern forces that promoted his candidature. This bloc often points out that the dark days of military authoritarianism may be gone, but Nigeria has yet 'to reach Jerusalem'. Central to the view of this bloc, therefore, is the perception that the inner workings of the Nigerian state continue to be structurally rigged in favour of the north and against them. Regional autonomy in a re-worked federation is a fundamental demand. A second grouse of the Yoruba bloc is that the oppressive statism of the centrist state is blocking their capacity for self-realization. According to a leading

264

A. Raufu Mustapha

Yoruba intellectual and former Finance Minister, Onaolapo Soleye: As a people, our interest is to have maximum opportunity for individual and community development without let or hindrance. We need an enabling socio-political environment that provides the best opportunity to individuals for self-actualization and fulfilment so as to unleash their creative energies for the development and transformation of the local and regional communities and the national polity as a whole . . . We no longer wish to be part of an arrangement that discriminates against our children. (Guardian: 19 May 1999) This fundamental aspect of Yoruba disaffection is little understood by others, who tend instead to concentrate on the fact that the Yoruba already dominate much of the economy and the bureaucracy. Finally, there is a third, historical, grievance felt by this bloc; the majority of them come from the political stable of Obafemi Awolowo, the dominant Yoruba political leader from the 1940s to the 1980s. That Awolowo never fulfilled his burning ambition to lead Nigeria was seen by this bloc as a travesty of reason and decency - a clear manifestation of the disregard and contempt with which the enterprise and 'enlightenment' of the Yoruba are held by others. The reaction to the annulment of Abiola's mandate flowed precisely from this long-running sense of 'abuse'. While the end of military authoritarianism brought some relief, and Obasanjo's reforms have placated many, important sections of the Yoruba bloc continue to regard the Nigerian state with deep suspicion, tempered only by their grudging acceptance of Obasanjo. The demand of this bloc is for a Sovereign National Conference (SNC) at which the ethnic nationalities in the country can negotiate the basis of their continued partnership. The fourth and final perspective on democratization comes from ethnic minorities, particularly from the Niger Delta (southsouth) and the Middle Belt (northcentral). Their complaints are that, despite democratization, they remain powerless in the wider structural workings of the Nigerian state. They remain victims of a long-running neglect in the distribution of facilities, appointments and resources. Their oil resources continue to be controlled by the central government, beholden to the needs of the hegemonic blocs. On the other hand, they continue to bear the brunt of the environmental fallout of oil extraction. This perspective also demands a Sovereign National Conference, but it is politically distinct from the demands of the Yoruba bloc. However, while minority mobilization in the Middle Belt is largely against the core north, mobilization in the Niger Delta is largely against the central state.

Three Bones of Contention The list of ethno-regional grievances and allegations cited above is not exhaustive, but they are illustrative of the tone of ethno-regional mobilization accompanying democratization. These grievances also point us in the direction of three major contentious issues threatening the process of democratization: firstly, the claim that the Obasanjo administration is actively promoting Yoruba hegemonic ambitions; second, the shaky constitutional foundation of the democratization process and the persistent demands for a Sovereign National Conference; and third, the adoption of Sharia criminal law in some states in the north. The allegation of promoting Yoruba hegemony started as a trickle and then turned into a flood. Initially, only a limited set of appointments and nominations were under scrutiny. In January 2000, 40 of the 58 Senators from the states of

Ethnicity & the Politics of Democratization in Nigeria

265

the north met and denounced Obasanjo's style of administration and his alleged favouritism towards the Yoruba: We have observed that the lopsided and unfair manner of appointments into key offices of government . . . since the advent of this administration has continued unabated despite the hue and cry from other sections of the country. (Vanguard, 18 January 2000) A committee of the Senate was consequently constituted to look into the distribution of the major appointments made by Obasanjo. The criteria used were those established by the Federal Character Commission, charged with ensuring that the composition of public bureaucracies in Nigeria reflects the 'federal character' of the country. Of the 208 appointments reviewed, the two 'core' northern zones had the highest shares; northwest zone 40, and northeast zone 37. The third northern zone, the northcentral, had 33. Of the three southern zones, Obsanjo's home zone, the southwest, had 35, followed by the southern minorities' southsouth zone at 32, and finally the Igbo southeast at 31 (Guardian, 18 February 2000). Based on the Senate figures, we may deduce that the accusations of favouritism are coming from a perception of being disadvantaged, rather than the reality of it. Secondly, we may be dealing with a sense of relative deprivation by which some core northern elite groups see their earlier pre-eminent position worsening relative to others; a similar review exercise, using data from the Babangida and Abacha periods, would be instructive in this regard. A third way of reconciling the accusation of favouritism with the Senate figures is by looking, not just at the numbers of 'top jobs' counted, but at the distribution of what these worldly-wise Nigerian politicians refer to as 'juicy appointments'. But it might also be that the resentment is being fuelled by wider strategic changes within the bureaucracy. According to Asaph Zadok, an official of the northern Arewa People's Congress (APC): We are not the least surprised that the Inspector-General of Police, the Minister of Internal Affairs, the Minister of Police Affairs, the Director of Defence Intelligence Service, the Director of Military Intelligence and the Director-General of State Security Service all come from Yoruba-speaking tribes, thereby forming a 'concentric ethnic ring'. (Guardian, 15 May 2000) A similar nomination of a Yoruba Auditor-General of the Federation was vetoed by the Senate because the Governor of the Central Bank and the AccountantGeneral are both Yorubas. Are these appointments meritocratic or are they part of a wider strategic calculation to entrench Yoruba interests in the state system? As far as the northern political bloc is concerned, Obasanjo is using his powers of appointment firstly to favour the Yoruba, and secondly to instigate minority agitation in the northcentral zone. There is no doubt that the Yoruba, as an ethnic group, have a disproportionate share of the formal economy and the federal bureaucracy. This is as a result of long-run historical advantages which have led to higher levels of economic infrastructure and human capital in the Yoruba southwest. In a divided society like Nigeria, such a structure of ethnic advantage, even when won through the 'normal' run of historical development, is bound to create political problems. Obasanjo's reforms, with their emphasis on merit and experience in bureaucratic appointments, are likely to exacerbate the structural imbalances. Furthermore, the cumulative consequence of the changes under Obasanjo may also have been seen as disrupting the 'Platonic' division of power in Nigerian society credited to a northern politician, Maitama Sule. Under this scheme, it is suggested that,

266

A. Raufu Mustapha

while the Yoruba are good in diplomacy and so should control the bureaucracy, the Igbo are good in business and should therefore control commerce. The Hausa are seen as good in governance, and, by implication, they should control the government. Obasanjo's presidency, particularly given his apparent independence of his northern allies, might then be seen as a structural anomaly. On the other hand, Obasanjo could simply be promoting Yoruba hegemonism. Everything considered, I doubt that Obasanjo would embark on such a project. This is not to suggest that there is no hegemonic impulse in Yoruba politics; the argument is that Obasanjo is an unlikely carrier of such an impulse. Firstly, it runs counter to what he has stood for since he entered public life in 1966. Politically, Obasanjo has always been nationalistic, authoritarian, paternalistic, and against Yoruba sectarian mobilization. It is my opinion that he has not shifted much from these tendencies, the only notable change being his increased willingness to rely on external forces. Secondly, some of the 'losses' of the north since the inception of his administration can be justified as creating a level playing field which gives others a chance after the wanton distortions of the Babangida and Abacha years. Thirdly, the division between the Yoruba 'mainstream' and the minority of the Yoruba who actively support Obasanjo continues to be substantial. These arguments notwithstanding, the 'threat' of Yoruba hegemonism under Obasanjo has become a controversial factor in the democratic process. The controversy over the convocation of a Sovereign National Conference has a long history. In 1990, two groups started canvassing for the convocation of such a conference, a group of left-of-centre activists led by the radical lawyer and Bar Association President, Aka-Bashorun, and a second group of retired but influential federal bureaucrats. Concerned by increasing signs that Babangida was not committed to his own transition agenda, the Aka-Bashorun group sought to convoke the conference to force the issue of democratization and regional autonomy. The bureaucrats, on the other hand, were said to have canvassed the idea at the behest of Babangida, their objective being the establishment of a 'French-type' constitutional arrangement under which Babangida might continue in office as President, albeit with an elected Prime Minister in charge of day-to-day government. The incompatibility of their respective objectives led to a parting of ways and the call for the SNC became an anti-military, pro-democracy preserve. It was quickly picked up by restive ethnic blocs amongst the Yoruba, in the Niger Delta, and generally in the south. The SNC became the antidote to a perceived northern-inspired military authoritarianism. By the time Abacha took over in 1993, the idea of the SNC was so entrenched in opposition demands that he promised to convoke it as a manoeuvre for winning over critical support in the south. In the end, he convoked a conference with limited powers, and foisted on it a constitution tailor-made for the perpetuation of an Abacha presidency, albeit in civilian garb. The convocation of a 'proper' SNC has since become the litmus test of genuine democratization amongst the Yoruba and minority Niger Delta blocs. Amongst the Yoruba, most hope that such a conference will provide the basis for negotiating a balanced federal structure. There is no doubt, however, that a significant minority see it as the prelude to the declaration of the O'odua Republic (ie. an independent Yoruba state). The Abdusalami transition programme which ushered in the Obasanjo presidency rejected, but could not put to rest, the demands for the SNC. Shortly after the installation of democratically elected governments the six state governments of the Yoruba southwest demanded the convocation of the

Ethnicity & the Politics of Democratization in Nigeria

267

SNC. Their argument is that the 1999 Constitution produced by the Abdusalami regime and its associated transition programme did not reflect the aspirations of the people. Nevertheless, both were tolerated because of the need to remove the military from government. Since this 'transition to the transition' has been completed, the real transition should start with a SNC of all ethnic nationalities to examine all long-standing grievances as a prelude to a comprehensive constitutional settlement. Specific areas for attention are revenue allocation, residual powers for the states, separate state police forces, as well as the need for states to have their own constitutions. Though the argument is frequently fought out on the terrain of federal-state relations, at real issue is the demand that the federation be re-organized to take account of its ethnic composition. It is the 'liberation' of the Yoruba from the straitjacket of the 'northerndominated' Nigerian state that is the objective, rather than the rights of, say, Lagos state per se. Many Yoruba leaders insist that this demand for restructuring the Nigerian federation is 'not negotiable' (Fasanmi, Guardian, 16 November 1999). Supporting the call for the SNC are the ethnic minorities of the Niger Delta who enjoy the backing of the influential Guardian: It is now quite clear that the country cannot make progress unless and until the fundamental basis of its existence and unity is re-examined and renegotiated. The present system is fatally flawed in several respects. For example, a country that is supposed to be a federation has been run as a unitary system for the better part of 30 years. The overcentralization favours parasitic groups and individuals who manipulate all governments and thus prevent the prosecution of a programme of modernization and development. The exploitation and plunder of resources of parts of the country are consequences of this inequitable arrangement. The condition of internal colonialism will not change unless there is an open platform for all nationalities and interest groups to discuss and agree on the best way of living together in unity and harmony. (Guardian, 6 December 1999) Predictably, this call for the SNC has been rejected by many northern politicians and academics, fearful of the withdrawal of oil revenues from the north and the potential for Yoruba succession. Instead of the SNC, Obasanjo has set up a Presidential Technical Committee (PTC), composed of representatives of the three registered political parties, and charged it with reviewing the Constitution in consultation with the public. It is expected that, at the end of the exercise, a bill on constitutional reform will be sent to the National Assembly. Subsequently, the division between the SNC and the PTC has been blurred by the demands put forward by a group called the Patriots, made up of 16 senior lawyers and politicians from across the three zones of southern Nigeria. In a memorandum submitted to the PTC, the Patriots call for the creation of six regions, based on the current six geo-political zones. They also advocate that substantial powers be given to these regions, restoring effective federalism. They support the principle of derivation in the sharing of mineral revenue from onshore, but not offshore, oil operations. They argue for the reorganization of the Nigerian police along the lines of the six regions proposed, with an Assistant Inspector General of Police answerable to the regional governor. Finally, they recommend the convocation of a National Conference, but without sovereign powers, where the fine details of the new arrangement can be negotiated. A predecessor of the Arewa Consultative Forum (ACF), the Northern Elders Forum (NEF) dismissed the Patriots' proposals, claiming that the restructuring of the country had been canvassed and rejected before for lack of merit. Secondly, it advocates 'able leadership and good governance' as the real needs

268

A. Raufu Mustapha

of the country. Thirdly, it cites the 1995 Constitutional Conference convoked by Abacha as one instance when Nigerians discussed their problems, pointing out that there is no need for drastic measures. Finally, the NEF has vowed to 'uphold the unity of the landmass called Northern Nigeria' (Vanguard, 3 March 2000). For its part, the umbrella organization of Niger Delta politicians, the Union of Niger Delta (UND), has insisted on the convocation of the SNC. It wants '100 per cent resource control by ethnic groups', and condemns the distinction between on-shore and off-shore oil drawn by the Patriots. It accepts the six-region formula 'as the ONLY acceptable basis for the continued existence of the Nigerian polity' (Vanguard, 6 March 2000). There is no doubt that Nigerians seriously need to talk to each other. This is made all the more pressing by the inadequacies of the military-imposed 1999 Constitution. But it can be argued that the SNC is not the only venue for such a discussion. What if some ethnic and religious extremists take advantage of the SNC to plunge the country into anarchy? If so many ethnic and religious communities are already resorting to force, what assurance do we have that the same murderous intent will not be brought to the negotiating table? That the 1999 Constitution is inadequate is not in doubt. Secondly, that the unitarist military state must be decentralized for meaningful progress to be made is also not in doubt. In my view, however, to insist that these tasks can only be done through the SNC is to make the country hostage to fortune. There are indications that the Obasanjo administration and representatives of political forces in the three southern zones and the northcentral zone have essentially agreed to a National Conference without sovereign powers. The third contentious issue is the controversy surrounding the extension of the scope of Sharia law. The Sharia legal system was established in northern Nigeria with the foundation of the Sokoto Caliphate in 1804. When the British took over the region in 1904, the legal system was maintained but reformed to suit British sensibilities. Things remained much as they had been, largely thanks to the close alliance between the colonial administration and the local aristocracy. This cosy situation was called into question in the terminal colonial period when a man sentenced to death in a Sharia court appealed successfully to the magistrate's court. The resulting crisis of confidence in 1957/8 was resolved by the establishment of a committee made up of British, Sudanese and Pakistani legal experts along with representatives of local interest groups, including a northern Christian. A penal code, based on an earlier version developed in the Anglo-Egyptian Sudan, was introduced, incorporating the basic tenets of the Sharia, but shorn of those aspects considered to be repugnant to natural justice. Since then, a form of Sharia law has been part of the Nigerian legal system, regulating civil matters between consenting Muslims. From the 1970s, there has been a tendency for politicians from Muslim constituencies to demand the extension of this Sharia court system to the federal level. This demand has always been opposed by politicians from Christian areas, fearful of the 'Islamization' of the country, particularly under Muslim military leaders. It was within this context of the increasing politicization of religion that Governor Ahmed Sani of Zamfara State in the northwest zone promised to implement '100 per cent Sharia\ and not the '70-80 per cent' embodied in the penal code. It must be pointed out, however, that this is a new position, radically different from the demands for Sharia since the 1970s. From the 1970s, advocates of Sharia wanted the extension of the '80 per cent' penal code based on the Sharia to the federal level. This would still have been limited to civil matters

Ethnicity & the Politics of Democratization in Nigeria

269

between consenting Muslims. What Ahmed Sani proposed amounted to the Islamic equivalent of 'socialism in one country'. He would introduce '100 per cent' Sharia, covering both civil and criminal matters, but only in Zamfara. As far as he was concerned, 'whoever administers or governs any society not based on Sharia is an unbeliever' (Vanguard, 3 March 2000). State resources were deployed to meet this new objective and a Ministry of Religious Affairs was created. 62 Sharia courts were built and staffed. This system is to run parallel with the old penal code system for Christians. Ahmed Sani defended his position by claiming that his state was 99.9 per cent Muslim, and that this was what the electorate wanted. Secondly, he had promised to implement the Sharia during his electioneering campaign. Thirdly, he claimed that the 1999 constitution allowed Zamfara to exercise its rights within the federation by having a legal system which accorded with the wishes of its citizens. Finally, he claimed that since the Sharia was limited only to Muslims, followers of other faiths had nothing to fear. Sharia was declared in an unkempt public square in October 1999, with full implementation coming into effect in January 2000. Various organizations, including the Zamfara Youth Council, a vigilante group, were formed to ensure the implementation of the new system. Zamfara has no police force and the federal police have declined to enforce the directive coming from the new legal order. There are many political and constitutional problems with the declaration of '100 per cent' Sharia. In the first place, it outlaws social activities, such as drinking alcohol and attending cinemas, which are matters of individual choice in the Nigerian Constitution. Secondly, by demanding the segregation of men and women, boys and girls, in public places, it challenges the right to free association and discriminates against women's access to public facilities. Many who are not culturally disposed to this style of behaviour find themselves compelled to follow it by the vigilante group. Thirdly, the claim that the Sharia is superior to the Constitution sets the stage for potential conflict, if not immediate anarchy. Fourthly, the extension of Sharia to criminal law means that different punishments for a particular crime could be imposed on citizens, depending on their religion. Though advocates of the Sharia often point out that states within a federation can have different laws, they fail to note that those laws must be equally binding on all citizens within the boundaries of the given state. Fifthly, punishments such as the cutting off of limbs and the stoning to death of some category of offenders are allowed, contrary to the constitutional protection against cruel and inhuman punishment. Sixthly, since apostasy is punishable by death in Islam, freedom of conscience for those born into Muslim families will be circumscribed by the full Sharia. The enacted Sharia codes have been silent on this issue. Finally, the stipulation that the evidence of two women is equivalent to that of one man contradicts the constitutional guarantee against discrimination on grounds of sex. Above all else, Christians, inside and outside Zamfara, fear that the unilateral extension of the Sharia is a calculated challenge to the multireligious status of the Nigerian state. From his Zamfara base, Ahmed Sani took to proselytizing. Emissaries were despatched to other parts of the north. In Bauchi, scene of numerous previous violent clashes between Christian ethnic minorities and the dominant Muslim Hausa-Fulani: The Zamfara State Governor's emissary told the Bauchi monarch that his mission was to remind him 'to ensure that, as both Bauchi and Adamawa [another mixed and volatile area] were historically strong pillars of the old caliphate order, destiny requires

270

A. Raufu Mustapha

that they should play their historic role again as a new caliphatic order is about to be restored'. He affirmed that it would be the happiest moment for the government of Zamfara, if all the states in Northern Nigeria which were part of the great dynasty of Usman Dan Fodio could adopt and implement the new Sharia law as part of their legal system. Arguing that the much vaunted differences among minorities, Christians and Moslems in the north were a myth, he stressed that the north is a strong Moslem enclave, and that 'as true Moslems, Sharia law is inevitable'. (Guardian, 8 August 1999) Clearly, the 'export' of Sharia was going to be just as problematic as the export of the 'socialist revolution'! Regardless of the potential problems, the Sharia craze soon spread to other core northern states, as political entrepreneurs and genuine believers struggled to promote this Islamic virtue. As was to be expected, reactions to the unilateral extension of Sharia came thick and fast from the Christian community nationwide, some based on genuine fears, others fuelled by ignorance and prejudice. A member of the Cross River State House of Assembly opined that Islam was based on 'violence and discrimination'. The Zamfara move was denounced as 'secessionist'. Were the federal government to fail to call Zamfara to order, Cross River State would declare a 'Christian State'. The resolution was duly passed. For most people in southern Nigeria, the unilateral advocacy of Sharia at this juncture in the democratization process is politically motivated. For the Christian communities of the north, both indigenous and non-indigenous, the threat of Sharia looms larger still, since their daily lives are at stake. Belligerent speeches were followed by frantic mobilization from both sides of the Sharia divide, leading to violent conflicts. The charge of political motivation is certainly not without foundation. Former Head of State, General Buhari, seizes every opportunity to call on Muslims to vote only for those who will 'protect' their religion. Even left-of-centre northern politicians like Balarabe Musa have joined the Sharia bandwagon, claiming that the Sharia is superior to the Constitution, since 'the constitution has failed to address fundamental issues which affect the people'. He justified the adoption of Sharia because 'there are vocal Nigerians who are saying that we should have ethnic republics' (Guardian: 26 October 1999). In the words of a leading northern journalist, Kabiru Yusuf, 'You have all sorts of these [ethnic] movements appearing across Nigeria and I guess in some ways sharia is our response from the north' (Reuters, 29 November 1999). It must, however, be stated categorically that we are not dealing here with a group of historical romanticists trying to recapture the past glory of a long-gone caliphal order. Neither are we dealing with a movement led by mullahs or clerics, sincerely bent on recasting society. What is involved is the intensified politicization of religion by a section of the northern elite. The reason for this apparent provocation may not be unconnected with the disillusionment with Obasanjo. As for the masses in the core north, they support the extension of the Sharia because they believe it will bring them the justice which the current legal system has failed to deliver. The irony of the situation is that it is the same elite that bankrupted the existing system that is now spearheading the theocratic alternative. It would be wrong, however, to imagine that the north is split neatly along religious lines, Muslims confronting Christians. Some predominantly Muslim states like Kwara and Adamawa have continued to use the penal code. And some northern leaders have publicly expressed reservations about the Zamfara declaration. As far as the fiery preacher, El Zak-Zaky, is concerned, the Zamfara

Ethnicity & the Politics of Democratization in Nigeria

271

declaration is bound to fail. He expressed the fear 'that the Sharia legal system under the current political dispensation would become an instrument of oppression' (Vanguard, 3 November 1999). He has also denounced the Zamfara declaration as 'a political campaign strategy' (Guardian, 2 March 2000). Similarly, an elder statesman in Kano State, Ali Abdullah, warned that the Sharia would not work in the state because of its heterogeneity, while another elder statesman, Lawal Danbazau, warned that 'it is not possible to operate more than one law in the country except if you want chaos'. He argued that the declarations of Sharia were politically motivated: I am of the opinion that some wealthy Nigerians who are not comfortable with the actions currently being taken by the present administration have gone out to collaborate with foreigners to use religion as a tool against the President, simply because he is a Christian and again because of the investigations currently being carried out. I suspect that they intend to use the jobless youths and cause religious chaos which will not do the country any good. (Vanguard, 16 November 1999) Between the numerous allegations of the different political blocs and the three thorny issues raised by democratization, the impression is often created that Nigeria is at breaking point. The repeated violent crises have consolidated this view. Such a one-sided view needs to be tempered by other equally important factors. Firstly, in Nigeria, we are clearly not dealing with the visceral sort of ethnic hatred common in the recent history of the Balkans; there is no example of systematic ethnic cleansing in Nigeria. Secondly, in the immediate climate created by ethno-religious posturing and mobilization, it is often forgotten that many things continue to bind Nigerians together. For example, all the major political forces in the country are agreed that the return of the military is not the way forward. Thirdly, images of the secession of discrete sections of the country fail to recognize how fragmented these ethno-regional blocs are from within.

Democracy and Intra-ethnic Conflict That democracy has fanned inter-ethnic conflict is not in doubt. But there is an equally important intra-ethnic dynamic in play. Though the ethnic blocs aspire to building monolithic constructions said to represent all members of their purported constituency, the reality is often different. Starting with the ethnoregional construction around the north, an important internal debate has emerged on the relationship between the 'core5 north of the Muslim communities and the 'lower' north of the Middle Belt. When people like Mahmud Tukur claim to represent the disgruntlement of the 'cultural north' with Obasanjo's appointments, they are immediately challenged by others, like Moses Adah: Talking about the North, even the most backward mind will find it difficult to understand what they mean. Are Benue, Kogi, Plateau, Adamawa and Kwara states no longer part of the North? Maybe they mean that Christians of the North are not northerners. Or probably that when it comes to appointments some states in the North are not to be considered as being in the North . . . How come that in times of political campaigns and elections, the North is promoted as one incongruous whole but when it comes to appointments, the same people tell us that some people are more Northern than the others? (Vanguard, 24 June 1999) It is not surprising that some political movements within the ethnically heterogeneous Middle Belt have pursued their political interests with little, if any, regard for the monolithic construct of the 'north'. For example, the main zonal

272

A. Raufu Mustapha

political organization within the Middle Belt, the Middle Belt Forum (MBF), has tended to argue its case, separate from those of the larger north. Furthermore, it challenges its inclusion in the 'north' (Vanguard, 26 October 1999). The division is, however, more complex than a simple divide between the 'core' and the 'lower' north - a division based on the long-standing 'minority question' in the old Northern Region. Within the 'lower' north, there are those, such as the Middle-Belt Youth Consensus Forum, who still assert that they remain firmly 'Northerners' (Today, 21-27 November 1999). Similarly, some 'core' northern analysts have argued for tolerance and restraint in the interaction between the two 'halves' of the north (Lamido Sanusi, Guardian, 26 July 1999). And the influential core northern politician, Lawal Kaita, has forcefully defended the interests of the Middle Belt within the north (Guardian, 14 July 1999). Even within the core north, the construction of a monolithic Hausa-Fulani position is constantly being challenged. Many younger analysts have tended to criticize the proponents of northern disgruntlement as self-serving political operators concerned only with their narrow individual or class interests (see Lamido Sanusi, Guardian, 26 July 1999; Ahmadu Abubakar, Tell, 6 September 1999; and Plain Truth Movement, Vanguard, 24 June 1999; Governor Ahmed Makarfi, Guardian, 17 July 2002; Musa Rabiu, Guardian, 26 February 2002). They argue that during the long period when this northern elite was dominant in Nigerian politics, its members ignored the development of human capital in the north, while concentrating on their individual and collective political and bureaucratic advancement. While the northern elite prospered, the northern masses wallowed in the worst poverty, illiteracy and disease in Nigeria. How, then, could this same elite now claim to champion 'Northern' interests when they patently failed to do so under the best of circumstances? Even amongst northern 'elders', there is disagreement. Wada Nas illustrates this tendency in northern politics when he: . . . attributed the apparent backwardness of the North to the activities of some of its elders whom he accused of seeking recognition and power through autocracy. (Guardian, 14 February 2000) The quest for a united Igbo construction has also prompted an internal debate about the causes of Igbo marginalization. Some, like Ugochukwu (Guardian, 27 July 1999), have blamed Igbo marginalization on excessive individualism leading to anarchic leadership. Nwabuikwu (Guardian, 27 October 1999) develops on this theme when he states that: 'If Igboland were a limited liability company, the characters who have been posturing as its directors would have been sacked with ignominy long ago'. However, the most trenchant critique of Igbo politicians came from Mma Agbagha who wrote: This motley crowd of 'movers and shakers' was created by military dictatorships over the years . . . Their modus operandi has always been to idolize any ruling clique in Abuja and intimidate locals through name dropping and blackmail. . . For these men and women, contracts, money and exhibition of wealth are the ultimate end of politics . . . If Igbos . . . must be relevant in . . . Nigeria . . . a new crop of leadership will be inevitable . . . The so-called Ohaneze, touted as an umbrella organization is nothing else but a platform of renegade traditional rulers and political boot lickers. (Post Express, 27 June, 1 July, 8 July 1999) Even the Yoruba, who are generally seen as having the most vigorous ethnoregional mobilization, have not escaped from internal dissension. At one level,

Ethnicity & the Politics of Democratization in Nigeria

273

the division is between the Yoruba 'mainstream', which represents the 'core' of the Yoruba construction around such organizations as the Afenifere, the OPC, and the political party, Alliance for Democracy (AD), and opposed to this core, a pragmatic minority to be found largely in Obasanjo's PDP. Separating the two is the fact that the minority is more inclined towards inclusive national politics, and is not committed to the agitation for a Sovereign National Conference. Though the mainstream is grudgingly supportive of Obasanjo, serious tensions remain between the Afenifere leadership and Obasanjo, and between the AD and the PDP in the southwest. Even the Yoruba 'mainstream' is not immune from dissension. It is seriously split between the supporters of the influential politician, Bola Ige, and those of the gerontocratic leadership of Afenifere. While both are committed to the promotion and defence of Yoruba interests, the Ige group has a more national political orientation. It also wants the organizational separation of the Afenifere and the AD. Thirdly, on the all-important question of the SNC, it is prepared to consider the realization of Yoruba objectives by other means, such as the review of the Constitution through the National Assembly (Vanguard, 1 February 2000). The AD is fundamentally split between these two warring groups, and intra-Yoruba conflict came to a head in December 2001 with the assassination of Ige who was Justice Minister in Obasanjo's government. Finally, the ethnic minorities are equally riven by divisions. The pattern of violent communal clashes between minority communities in the Middle Belt and the Niger Delta persists. Other divisions, such as the factionalization of the Movement for the Survival of Ogoni People (MOSOP), have emerged. The other major ethnic minority movement, the I jaw National Council (INC) is also factionalized. From this review of the internal conflict within the various identity constructions in Nigeria, we can begin to see that instead of having distinct identity blocks intent on dismembering the country, what we have are overlapping claims and counter-claims. Disentangling these is virtually impossible. It is highly unlikely that the so-called Oduduwa, Usmanniya or Biafran Republics of the hegemonic blocs can ever be realized. Were any aggrieved party to insist on pushing its case to the maximum, anarchy is the most likely result. To avoid that potential anarchy, it is imperative that we use democracy as the institutional and ethical framework for negotiating compromises for the thorny problems confronting Nigeria. Democracy matters.

Conclusion: State3 Hegemonism, Xenophobia and Democracy There is some truth in the claim by Stewart and O'Sullivan that democratic institutions are not sufficient in themselves to prevent conflict in sharply divided societies. Even when these democratic institutions are redesigned to reduce conflict, as in the case of Sri Lanka, they may fail to do so (Stewart and O'Sullivan, 1998). They advocate the promotion of inclusive government, even when this depends on some undemocratic structures and processes. But to what extent can we discuss ethno-regional conflict in Nigeria in terms of fundamental differences in value? If the answer is that there are indeed fundamental differences in value systems, then we must consider Rawls' argument that societies need to be founded on certain agreed values. Since these values are contested and difficult to determine, Rawls suggests that we find a solution through a 'sufficient consensus', enough to hold the society together and provide a foundation for the

274

A. Raufu Mustapha

rules by which society arranges its internal relations, while not imposing uniformity and leaving the composite members free to pursue their own agendas. Claude Ake makes the same point when he defines democracy in Africa as 'negotiated compromise' (Ake, 2002). However, I remain sceptical about the fundamental nature of the differences in value systems across Nigerian societies. That notwithstanding, it must be pointed out that there is a significant difference in the perceptions of, and the attitudes towards, the state amongst different political formations across Nigeria. I argue that it is the nature of the state and the varying attitudes towards it that lie at the root of the Nigerian problem. This difference in attitude is best illustrated by the demands of the Patriots for restructuring, and the defence of the unitarist status quo by the NEF. Though it is methodologically imprecise to extend the actions of an individual or a group to the wider community they are supposed to be representing, there are often valid bases to so do (Kaviraj, 1997). We must, however, remain aware of the fuzziness of our characterizations. In its most crude form, this difference can be characterized in the following way. As a result of the accident of history and geography, but also as a result of conscious purposive decisions, there is a major divide in the level of development of human capital in Nigeria. In the 1920s and 1930s, the operative consciousness of this divide was between the southwest, considered as privileged and more enlightened, and the southeast. In this period, the Igbo in the southeast got together in an internal process of collective action which narrowed, if not eliminated, some of the perceived 'gaps' between them and the Yoruba southwest. The attitude to the state, as an avenue for competitive self- and collective advancement, is common to these two groups. In the north, however, a different trajectory took place. To begin with, when the even wider 'gap' between the south and the north became politically salient from the 1940s, the initial northern reaction was quite similar to those of the Igbo at an earlier period. Under the leadership of the regional Premier, Ahmadu Bello, the north pursued an aggressive policy of building up its social institutions and expanding its human capital. Unlike in the south, however, the northern project was largely statist, local communities and individuals played only a minor role, such as the public collection to enable the northern government to send students to Britain. The northern elite that is rightly castigated for ignoring the development of the human capital of its constituencies grew up in this allembracing statism and is largely unfamiliar with the sort of communal self-help drive common in the south. But such was the size of the gulf, that the north also adopted a second strategy for the defense of its interests within the evolving Nigerian state. It sought to convert its regional numerical strength into political capital. The internal statist posture in the northern region was now coupled with a wider federal posture predicated on northern control of the federal government as a means of containing the perceived threat of southern domination. After 1983, northern strategy shifted decidedly in favour of political capital, as the 'private motivation' of northern officers and bureaucrats began to supersede 'group motivation' (Stewart, 2002) as the primary motor of northern politics. Northern presence in the upper echelons of the bureaucracy was promoted, yet little effort was made to bridge the gap in human capital development. In 2002, Vice-President Atiku pointed out that seven southern organizations won licences to open private universities, while there were no applications from the north; another prominent northerner, Liman Chiroma, added that northern

Ethnicity & the Politics of Democratization in Nigeria

275

youths preferred politics and business to educational pursuits (Guardian, 26 February 2002). The statist orientation, and the emphasis on political capital, predisposes political forces in the north towards not just the centralization of the Nigerian state, but also the quest for hegemonic control of that state in the face of overwhelming southern 'hordes' thrown up by a more developed human capital base. On the other hand, the southern elements who already control more than their fair share of the resources of the system are resentful of constraints imposed upon them by this northern strategy. The zero-sum political logic of the rentier state exacerbates these tensions. While northern centralizing and hegemonic instincts are dysfunctional and ultimately unsustainable, southern demands for ethnic federalism and its associated xenophobia, are in my view, ahistorical and impractical (Mustapha, 1999). There is therefore no future for Nigeria outside of a 'negotiated compromise'. How to develop such a 'sufficient consensus' with minimal concessions to 'undemocratic' practices and structures remains the central task of nation-building in Nigeria. Politicians and their ethno-regional blocs will ultimately have to choose between such a negotiated compromise and anarchy.

References Ake, Claude. 2002. The Feasibility of Democracy in Africa. Dakar: CODESRIA. Akinterinwa, B. A. 1997. 'The 1993 Presidential Election Imbroglio', in L. Diamond et al., Transition Without End: Nigerian Politics and Civil Society Under Babangida. Boulder, CO: Lynne Rienner. Kaviraj, Sudipta, 1997. 'Collective forms in modern polities', in K. Dean, ed., Politics and the Ends of Identity. Aldershot: Ashgate. Mustapha, A. R. 1999. 'Back to the Future? Multi-ethnicity and the State in Africa', in L. Basta and J. Ibrahim, eds, Federalism and Decentralisation in Africa. Fribourg: Institut du Federalisme. Stewart, Frances. 2002. 'Root Causes of Violent Conflict in Developing Countries', British Medical Journal^ 9 February. Stewart, Frances and Megan O'Sullivan. 1998. Democracy, Conflict and Development - Three Cases, Working Paper Number 15. Oxford: Queen Elizabeth House.

IV

Ethnicity & Institutional Design in Africa

16

RICHARD S I M E O N & CHRISTINA MURRAY

Multi-Level Governance in South Africa Introduction

N

the word 'federal' nor 'federalism' appears in the 1996 Constitution of South Africa. Yet Section 40 states that: cIn the Republic, government is constituted as national, provincial and local spheres of government which are distinctive, interdependent and interrelated.' Section 43 vests legislative authority in the national parliament, in the provincial legislatures, and in municipal councils. Under section 104 (3), 'A provincial legislature is bound only by the Constitution.' Thus, provinces (and local governments) derive their status directly from the Constitution. They are independently elected, and exercise significant legislative powers. The Constitution also provides for intergovernmental fiscal arrangements, and sets out principles ('co-operative government') for the conduct of intergovernmental relations. The Constitutional Court is the final arbiter of disputes among the spheres of government. By most definitions,1 then, South Africa can be considered a federal country, albeit highly centralized. But federal, or federal-type arrangements, have been and remain deeply problematic in South Africa. During the years of apartheid, the concepts and discourse of federalism and confederalism were used to justify the existence of the Bantustans. 2 Whether or not a democratic South Africa was to be a federal South Africa was deeply contested throughout the transition and the negotiation of the new Constitution. 3 It remains contested, as provinces and municipalities EITHER

1

Ronald L. Watts, Comparing Federal Systems, 2nd edn. Kingston, Ontario: Institute of Intergovernmental Relations, Queen's University, 1999, p. 7. See also his 'Is the New Constitution Unitary or Federal?' in Bertus De Villiers ed., Birth of a Constitution. Kenwyn: Juta and Company, 1994, pp. 75-88. 2 Otherwise known as 'homelands'. 3 See Jenny Robinson, 'Federalism and the Transformation of the South African State,' in Graham Smith, ed., Federalism: The Multiethnic Challenge. London: Longman, 1995, pp. 255-78. She describes the Interim Constitution as 'protofederalism'.

277

278

Richard Simeon & Christina Murray

struggle to establish and consolidate entirely new political institutions and processes. Although multi-level government is embedded in the constitutional design, it has yet to become internalized as a basic fact of governance in the eyes of leaders and citizens. In this chapter we explore the debate on federalism and decentralization as it developed during the South African transition to democracy and the writing of its 1994 Interim and 1996 final Constitutions. We explore the constitutional design of the new system of 'multi-sphere' governance and examine its implementation in the ensuing years. The fundamental concern of this book is the recognition, accommodation and management of ethnic differences and cultural diversity in contemporary Africa. Federalism, or other forms of decentralization, is frequently advocated as one of the primary institutional means to achieve a balance between unity and diversity, £self-rule' and 'shared rule,' and thus promise reconciliation among diverse groups. It is just as frequently derided as divisive and ineffective. But decentralized government can also be assessed from other viewpoints. First is the democratic perspective. To what extent does federalism promote or hinder democratic governance, including participation, transparency, accountability and effective links between government and civil society? Second is a functional perspective focused on governance and the making and implementation of public policy. Does decentralization promote or hinder socio-economic transformation, choosing and setting priorities and delivering services such as pensions, housing, education, health and economic growth? We shall engage these questions as we examine the ability of the new institutions of multi-level government to carry out their assigned roles and responsibilities. In particular, if the system is to work effectively - with respect to any of the overriding goals of conflict management, democracy and good governance - then the institutions, in particular the new provinces, must develop a number of basic 'capacities'. Our description of the working of the system will thus be framed in terms of: • political capacity: have provinces been able to establish themselves as autonomous governments, with an independent political base and a presence in the minds of the population? • legislative capacity: have provinces and their legislatures developed the ability to formulate and enact their own legislation? Have they been able to influence national legislation that affects them, in the institution designed specifically for that purpose, the National Council of the Provinces (NCOP)? • administrative capacity: have the provinces been able to manage the civil service to ensure that it is efficient, uncorrupt, and able to deliver services effectively? This is especially important in South Africa because, as in Germany, the provinces administer most national legislation. • fiscal capacity: do the provinces have access to sufficient resources to carry out their responsibilities, and have sufficient financial flexibility to set their own priorities? • intergovernmental capacity: how effective is the machinery designed to encourage co-operative relationships among the spheres? How, and how well, do governments work together to provide a coherent framework for governance? This perspective suggests that multi-level governments, like other institutional forms, cannot be fully assessed in the abstract; their interaction with their eco-

Multi-Level Governance in South Africa

279

nomic, social and political contexts - and with their inherited historical legacies - is crucial. In South Africa, this means that the multi-sphere regime must deepen the democracy embraced in 1994, must contribute to socio-economic transformation and effective service delivery, and must contribute to the management of diversity in a divided society. It is obviously premature to suggest firm conclusions; instead, we shall concentrate on raising issues, questions, and possible directions for the future.

The Federalism Debate The advantages and disadvantages, costs and benefits, of federalism and related institutions have generated a lively and unresolved debate in recent years, both with respect to advanced industrial countries and to developing countries like South Africa. The literature on both sides of the question suggests that federalism is Janus-faced. Under some circumstances, it may facilitate conflict resolution, democracy and good governance; in other circumstances, it is seen as their enemy. Firm generalizations are few and far between, and almost always contested. In recent years, the decentralists have led the discussion, reacting against the dominance of majority ethnic groups, concentration of power in central governments, and the policy distortions inherent in over-centralized regimes. As Yash Ghai puts it, 'Autonomy has become the most sought-after and resisted device for conflict management.' 4 This is a major shift from the emphasis on the need for strong unitary states that dominated the literature in the postwar period. Federalism and Diversity Given a society deeply divided along ethnic, cultural, religious or linguistic lines, and in which the relevant groups are territorially concentrated, the argument for federalism is straightforward. It enhances the ability of groups to live together in relative harmony through a strategy of 'disengagement'. While sharing in a national government making decisions for the whole ('shared rule'), each group will be free to make its own decisions, according to its own values, needs and preferences through state, provincial or local governments that they elect ('selfrule'). This will ensure that the minorities will not be frustrated by a national government in which they constitute a majority, and gives them protection against minority views being imposed on them. The creation of a protected political space will increase the sense of security of minorities; and this, in turn, will help reconcile minority members to continued membership in the larger system. Recognition and accommodation in multi-national states - and hence peace and stability - are possible only if we set aside the notion of a homogeneous, unitary nation-state. 5 The counter-arguments are also powerful. The worry is that federalism institutionalizes, exacerbates, perpetuates, and entrenches the very conflicts it is designed to mitigate. Equipping minorities with the institutional armoury of constitutional powers and fiscal and bureaucratic resources provides minority 4

Yash Pal Ghai, 'Constitutional Assymetrics: Communal Representation, Federalism, and Culturural Autonomy', in Andrew Reynolds, ed., The Architecture of Democracy: Constitutional Design, Conflict Management and Democracy. Oxford: Oxford University Press, 2002, p. 155. 5 Michael Keating, 'So many nations, so few states: Territory and nationalism in the global era', in Alain-G. Gagnon and James Tully, eds, Multinational Democracies. Cambridge: Cambridge University Press, 2001, pp. 39-64.

280

Richard Simeon & Christina Murray

elites with the tools to nurture and enhance regional identities and grievances and to challenge national identities. At the limit, empowered local governments can become the launching pad for secession. Federalism also potentially threatens liberal values, through the possible discrimination against 'minorities within minorities', raising the spectre of ethnic cleansing to ensure the ethnic homogeneity of each province. Conflicts may be further intensified when there are major disparities in wealth and resources among the constituent units. Such concerns have led some theorists to suggest that arguments for federalism as a device for conflict management need to be qualified. For example, the rights of individuals wherever they live need to be guaranteed by a nation-wide Bill of Rights. The devolution of power outwards to constituent units needs to be balanced by equal consideration of ways of cbuilding-in' - ensuring that local minorities are represented in national institutions. The divisive effect of regional disparities needs to be counteracted by strong programmes to ensure equalization and sharing among the units. A particularly difficult issue is the design of sub-national boundaries. Is the principle to be 'one group, one province', with each unit constituted as ethnically homogeneous? Or, is conflict mitigated when each of the units is internally heterogeneous? Writers such as James Tully and Arend Lijphart who stress the virtues of self-government and self-determination for minorities stress the former; writers such as Daniel Elazar and Donald Horowitz stress the latter. In one case accommodation among different ethnic groups will occur within each unit; in the other, it will take place in relations between them. In Africa, Nigeria and the recently established Ethiopian federation lean strongly towards an ethnically based federalism; South Africa has sought to avoid it. Federalism and Democracy Similar debates take place with respect to decentralization and democracy. The democratic virtues of decentralization lie in the possibility of bringing government closer to the citizen, with greater opportunities for participation, and the ability to turn to another level of government when the first proves unresponsive. Federalism promises a better 'fit' between citizen preferences and policy outputs when these preferences or values vary across different groups. Some also argue that accountability and transparency are more easily achieved when political units are smaller and closer to the people. Local governments are less remote, and more accessible than different national governments. The counter-arguments are that decentralization may advantage local minorities, but frustrate national majorities; that corrupt, tyrannical and elitist governments are just as likely to be found at local levels as at the national level; and that the operation of multi-level government is itself complex, making it difficult for citizens to understand who is responsible for what, or who to hold accountable, thus leading to a 'democratic deficit'. Again, federalism appears to be Janus-headed. Federalism and Effective Government Finally, the arguments for decentralization suggest that such regimes are more likely to produce effective governance. It allows policies to be more closely tailored to local needs; it offers opportunities for innovation and experiment; it avoids the dangers of 'one-size-fits-all' policies for diverse communities. It frees national governments to concentrate on nation-wide priorities. On the other side are arguments that decentralization may frustrate the development of national

Multi-Level Governance in South Africa

281

solutions to national problems; that responsibilities and the resources to meet them are frequently mismatched; that different jurisdictions may pursue competing, contradictory and mutually defeating policies; and that the need for complex intergovernmental machinery to manage the relationships leads to excessive 'co-ordination costs' and unnecessary delay. All these issues have been played out in the African context. In the postcolonial period, the emphasis was on the need to establish strong, effective, unitary governments. Only they, it was argued, would be competent to forge national unity out of the ethnic fragmentation and competition that had been fostered by the colonial powers. Only they would have the resources to engage in the massive transformation of society and economy that confronted the newly independent regimes. By the 1990s, there had been a considerable shift in thinking. Now it was the costs and failures of over-centralized regimes that were more apparent. With respect to ethnic divisions, too often central governments acted not as unifiers but as agents of a dominant group, suppressing minorities. With respect to democracy, too often central regimes concentrated power and subverted democracy. And with respect to policy and governance, too often central governments proved remote, corrupt and incompetent. Hence, there has been a renewed interest in the virtues of decentralization, whether from international agencies such as the World Bank, focused on governmental effectiveness, or from citizen-based NGOs focused on citizen participation. All these issues were prominent in the debates in South Africa: was it to have a federal system; and, having chosen one, what would be its consequences?

The South African Background Federalism has a long history in South Africa, but it has little relevance to the contemporary debate. The 1909 Union of South Africa established a partially federal regime, but it was significant only for white citizens. Under the apartheid regime, power was increasingly concentrated in an all-powerful central state. The four traditional provinces 6 became merely administrative units. In addition, the apartheid regime established the Bantustans, made up of four 'independent states' - Transkei, Ciskei, Bophuthatswana, and Venda - and six 'self-governing territories'. These were constituted on tribal lines, and justified by appeals to federal and confederal values. This experience fundamentally de-legitimated the idea of federalism, especially in a form that institutionalized ethnic difference, in the eyes of the African National Congress (ANC). For the ANC, the new South Africa should be a unitary state for a non-racist society. There was a deep antipathy to institutionalizing ethnic or tribal differences in a federal constitution. 'Proposals for strong regional government were seen as a form of neo-apartheid.' 7 Moreover, the ANC and its allies - schooled both in Marxist ideas and in the principles of British parliamentary government - believed that only a powerful centralized state would be able to engage in the massive process of social and economic transformation that lay ahead. Fragmenting and dispersing authority would 6

The Cape, Orange Free State, Natal, and the Transvaal. Richard Humphries et al., 'The Shape of the Country: Negotiating Regional Government', in Steven Friedman and Doreen Atkinson, eds, The Small Miracle: South Africa's Negotiated Settlement. Johannesburg: Ravan Press, 1994, pp. 148-81. They quote Albie Sachs: 'We spoke of a unified country and the sovereignty of the people to emphasize the fundamental and equal citizenship of all', p. 149. 7

282

Richard Simeon & Christina Murray

make decision-making more difficult, undermining the capacity to achieve reconstruction and development. The National Party (NP) was the chief advocate of a federal regime. This stemmed partly from its fear of majority government. Federalism, along with the much more important Bill of Rights, could be an important check on the power of the majority, in an American-inspired system of checks and balances under a liberal constitution. In earlier periods, the NP also sought to promote a vision of South African society constituted not as a black majority and a white minority, but as a collection of tribal groups that would be represented both in provincial and national institutions providing group representation, along with a consensus-based consociational pattern of decision-making. Some Afrikaners, indeed, believed that it might be possible to create a Volkstaat, a homeland with an Afrikaner majority.8 Among black Africans, the demand for federalism came mainly from Chief Mangosuthu Buthelezi and his Inkatha Freedom Party (IFP), claiming to speak for the Zulu people, and demanding a highly autonomous KwaZulu-Natal, with the right of self-determination9 - confederalism, or 'extreme federalism'. 10 Support for federalism by the N P and IFP, of course, further discredited the idea in the eyes of the ANC. Despite these profound differences, a multi-sphere regime eventually became part of the 'pacted' constitutional settlement. A key element in reaching agreement was a confidential Report to the Political Parties in March 1993, prepared for the Consultative Business Movement. It argued that the imperatives of building one nation and achieving reconstruction and equity could be reconciled with regional powers that 'would provide additional checks and balances and that degree of security for different communities which is essential for constitutional order, social stability and economic growth and development . . . The danger lies in the extremes. An over-powerful center prevents at inception the forging of one nation - over-powerful regions on the other hand can lead to disruption and the ultimate destruction of one nation.' 11 The group sought a balance, much of which was embodied in the '34 Constitutional Principles' agreed to in 1993, and attached to the 1993 Interim Constitution. These Principles were to bind the democratically elected Constitutional Assembly when it drew up the 'final' constitution. Principle XIX stated that each level of government (national and provincial) should have both exclusive and concurrent powers; and XXI stated the principle of subsidiarity - decisions should be taken at the level that is most 'responsible and accountable.' Kay Hailbronner and Christine Kreuzer conclude that these and other elements in the binding principles 'provide in essence for a federal structure of the Republic of South 8 See Allister Sparks, Tomorrow is Another Country: The Inside Story of South Africa's Negotiated Revolution. Sandton: Struik Book Distributors, 1994, pp. 191, 204-6. Some of these ideas suggested that others would not have the vote; others held that only Afrikaners would be allowed to live in such a homeland. Some Afrikaners still maintain hopes of a homeland. 9 See Humphries et al., 'The Shape of the Country'. 10 Heribert Adam and Kogila Moodley, The Opening of the Apartheid Mind: Options for the new South Africa. Berkeley: University of California Press, 1993, p. 156. The extent of the autonomy Buthelezi had in mind was reflected later in KwaZulu-Natal's abortive attempt to write its own provincial constitution. The Constitution was rejected by the Constitutional Court, which deemed it inconsistent with the national constitution. See In re: Certification of the Constitution of the Province of KwaZuluNatal, 1996, 1996 (11) BCLR 1419; 1996 (4) SA 1098 (CC). 11 Consultative Business Movement, Regions in South Africa: Constitutional Options and Their Implications for Good Government and a Sound Economy, Confidential Report for Discussion with Political Parties. Johannesburg: 1993, pp. vi-vii.

Multi-Level Governance in South Africa

283

Africa'.12 In part, the ANC had to accept a multi-level system as part of the bargain, but as Richard Humphries and his colleagues point out, some ANC leaders had come to see advantages in effective regional governments both for the delivery of services and for the empowerment of citizens. Their exposure to foreign models of federalism, especially in Germany, convinced them that regional governments could be combined with strong leadership from the center. 13

The South African Institutional Design Thus, South Africa was to be a decentralized, perhaps even federal, system of multi-level government. But what form would it take? There were many variants from which to choose. Participants in the Constitutional Assembly studied several federations, including Canada, India, Germany, Switzerland, and the United States. The design chosen followed the German model of shared or integrated federalism most closely. Its pattern of national leadership, with framework legislation fleshed out and implemented by the Lander, and with tight integration between central and Land governments achieved through the Bundesrat, was deemed more appropriate to South African needs than the more divided, competitive, and decentralized Canadian model, to take just one example.14 The Provinces The new Constitution created nine provinces, 15 varying enormously in size, population and wealth. Gauteng is the most wealthy province, followed by the Western Cape. The Eastern Cape is the poorest. 16 As measured by the United Nations Human Development Index, Western Cape offers the highest quality of life, scoring 0.826; the Northern Province scores the lowest of the nine provinces with just 0.470. 17 Some provinces (for example, the Free State) follow the old boundaries; others were newly created. Several of them were required to incorporate all or parts of the pre-existing homelands. One of their central challenges has been to integrate the old - often much inflated, corrupt, and under-trained - civil services into the new South Africa. As we have seen, the institutionalization of ethnic differences was anathema to most of the designers of the new constitution. As a result, provincial boundaries are not designed to coincide with racial or tribal boundaries; indeed, given the distribution and mobility of the population, that would have been very difficult. South Africa explicitly rejected the models adopted by some other federations, which are designed to empower distinct national or ethnic groups with their own political institutions. As Regions in South Africa states: 'given the legacy of racial and ethnic division, 12 Kay Hailbronner and Christine Kreuzer, Implementing Federalism in the Final Constitution of the Republic of South Africa. Johannesburg: Konrad Adenauer Stiftung, 1995, p. 7. 13 Humphries et al., 'The Shape of the Country', pp. 152-3. 14 Richard Simeon, 'Considerations on the Design of Federations', SA Public Law 13, 1998, pp. 42-71. 15 Unlike Canada, the United States or Australia, the system was not constituted by previously existing units coming together, as in confederation. 16 Department of Finance, Republic of South Africa, Intergovernmental Fiscal Review 2000. Pretoria: October 2000, p. 8. 17 Department of Finance, Republic of South Africa, Intergovernmental Fiscal Review 1999. Pretoria: September 1999, p. 17. The scores range between 0 and 1 and this index seeks to measure quality of life.

284

Richard Simeon & Christina Murray

it would be unfortunate if a new regional policy and set of regional boundaries were to have the effect of promoting these divisions . . . The overriding goal in the demarcation of regions should not be the creation of exclusive ethnically homogeneous units.' 18 Ethnicity and provincial boundaries do largely coincide in three provinces - KwaZulu-Natal, the Eastern Cape, and the North West - in each of which more than two-thirds of residents speak a single language. But only in KwaZulu-Natal has this resulted in an ethnically based demand for more autonomy. 19 The Western Cape has also called for greater power, but this is based less on ethnic appeals than on the fact that it is the only province to have elected a national opposition party as its government. Thus, South Africa is a highly diverse society, but that diversity is not encapsulated in the demarcation of federal boundaries. Once apartheid was dismantled, the artificial homelands failed to produce any lasting or politically significant identity. Except for KwaZulu-Natal, and to a lesser extent the Western Cape, none of the provinces have articulated their concerns in ethnic terms or stimulated mobilization along ethnic lines. There are many debates about decentralization in contemporary South Africa, but issues of identity and sub-national political communities play virtually no role in them. This is partly the result of a deliberate decision to downplay ethnic difference in the institutional design; and partly the result of the fact that the fundamental divide remains that between black and white South Africans. As David Welsh puts it, 'race trumps ethnicity'. Cultural differences are indeed recognized in the Constitution. The Bill of Rights protects 'language and culture,' and the rights of 'cultural, religious and linguistic communities'. Section 6 on National Languages acknowledges 11 official languages, and establishes the 'Pan-South African Official Languages Board' to promote and protect these languages. Section 185 establishes an independent Commission for the Promotion and Protection of the Rights of Cultural, Religious and Linguistic Communities, with a mandate to ensure their rights. Section 235 recognizes the right of cultural and linguistic communities to 'self-determination', to be determined by national legislation. The key point, however, is that these cultural identities are not empowered by federalism; nor are the constituent groups allocated autonomous decision-making authority. Instead, the commitment to cultural pluralism is found primarily in individual rights, and in the realm of civil society. Thus, federalism in South Africa is not primarily about the resolution and accommodation of ethnic difference; indeed, fear that it would entrench and exacerbate ethnic divisions was a major reason for the ANC's doubts about federalism. Many of the provinces are themselves ethnically diverse, with the result that much of South Africa's ethnic politics will be played out within the provincial sphere. This approach is carried through in all the other elements of the constitutional design. Co-operative Government It begins with the concept of 'co-operative government', set out in Chapter Three of the Constitution. This requires the three spheres of government to function as a single, unified system, collaborating rather than competing. They must 'respect the constitutional status, institutions, powers, and functions of government in the other spheres'; act 'in a manner that does not encroach on the geographical, func18

Consultative Business Forum, Regions in South Africa, p. 17. In KwaZulu-Natal the Inkatha Freedom Party governs in coalition with the ANC. All other provinces are governed by the ANC. 19

Multi-Level Governance in South Africa

285

tional or institutional integrity of government in another sphere'; and 'cooperate with each other in mutual trust and good faith' through 'fostering friendly relations', ensuring communication and co-ordination, and avoiding taking their disputes to court. 20 They share responsibility for preserving 'the peace, the national unity, and the indivisibility of the Republic' and for providing 'effective, transparent, accountable and coherent government for the Republic as a whole' (Section 41). Thus, interdependence and overlap are at the centre of the design. Provinces are not set up to be rival centres of power and authority. The Division of Powers The allocation of responsibilities embodies a strong commitment to concurrent or shared roles. All spheres have the authority to legislate. However, the Constitution gives sweeping powers to the national government to set national standards and norms and to override provincial legislation that is believed to threaten national unity or national standards. There are no bright lines of divided jurisdiction; rather there are soft boundaries. There is a short list of 'exclusive' provincial powers, 21 but these are subject to the power of the national government to legislate to maintain national security, economic unity, or national standards, or to prevent 'unreasonable action' by a province that is 'prejudicial to the interests of another province or the country as a whole'. 22 These provisions reinforce the commitment to preventing the provinces' emergence as rival power centres, and to the clear predominance of the values of unity and nation-wide standards. Provinces can also establish their own constitutions, subject to the Constitution itself. KwaZulu-Natal and the Western Cape have exercised this power. The Western Cape constitution successfully passed constitutional muster; that of KwaZulu-Natal did not. 23 It failed mainly because the draft constitution asserted the autonomy of KZN as the homeland of the Zulu people. Most critical powers are concurrent. In these areas, national law prevails in a wide variety of circumstances. Under Section 146, national law will prevail if it deals with a matter that provinces cannot effectively regulate on their own or if it is necessary for 'national security', 'economic unity' and the common market, the promotion of equal opportunity and equal access to government services, or the protection of the environment. Such legislation must operate uniformly across the country. National law will also prevail if it is designed to prevent 'unreasonable action' by a province that harms others or impedes the implementation of national economic policy. In the few areas of exclusive provincial jurisdiction, Section 44(2) states that national legislation can still trump provincial initiatives if it can be shown to be necessary to maintain national security, economic unity, and 'essential national standards'; or to 'prevent unreasonable action taken by a province which is prejudicial to another province or the country as a whole'. These are sweeping federal powers, but they are not unlimited. They must be linked to clear national purposes or to clear evidence of the negative effects of a 20 Elsewhere, however, it is made clear that the Constitutional Court is the ultimate arbiter of relationships among the three spheres. 21 Including abattoirs, provincial planning, cultural matters, recreation, roads and sport, and veterinary matters. 22 Section 44 (2). 23 In re: Certification of the Amended Text of the Constitution of the Western Cape, 1997, 1997(12) BCLR 1653; 1998 (1) SA 655 (CC); In re: Certification of the Constitution of the Province of KwaZulu-Natal, 1996 (11) BCLR 1419; 1996 (4) SA 1098 (CC).

286

Richard Simeon & Christina Murray

province's actions. There is no blanket paramountcy for national legislation; it must be justified according to criteria set out in the Constitution. In a form of subsidiarity, the burden of proof is on the national government to show why its legislation should prevail. If not, 'provincial legislation prevails over national legislation'. Moreover, if the Constitutional Court is asked to resolve a conflict, it must 'have due regard to the approval or the rejection of the legislation by the NCOP'. Much, therefore, depends on how courts and others interpret these key provisions. The residual power is left to the national government. Sections 44 (2) and 146 can be read in a number of ways. One reading suggests that the criteria are so broad that it is difficult to imagine any provincial law surviving a national challenge. Another suggests a stricter test that could open up considerable space for provincial legislative initiative, given sufficient political will, independence and capacity. Following the German model, the Constitution also empowers the national government to enact legislation setting norms and standards regulating the conduct of provinces in taxation, budget preparation, management of provincial treasuries, and the structure and management of the public service. 24 This authority gives the centre considerable power to shape, supervise and monitor provincial institutions and processes. Fiscal Arrangements The dominant position of the national government extends into the financial arena. It monopolizes all major revenue sources. Provinces have very limited revenue-raising and borrowing capacities, and even these are subject to national regulation. The result is that fiscal federalism is highly centralized. This is despite the fact that provinces account for about 57 per cent of spending, compared with 42 per cent for the national government. 25 The rules are set out in the Constitution (Sections 214-216) and in the Intergovernmental Fiscal Relations Act, passed in 1997 and implemented in 1999. There is to be a single pool of revenue from which each sphere of government is entitled to an 'equitable share'. The distribution must take into account several criteria, including the national interest and the needs of the national government, the ability of the provinces to perform their tasks, and the need to combat economic disparities 'within and among provinces'. 26 An independent commission representing all three spheres, the Finance and Fiscal Commission (FFC), is designed to make recommendations on the distribution of the budget, and the national Minister of Finance must take explicit account of its recommendations in his annual Division of Revenue Bill.27 In addition to the unconditional equitable share of national revenue, the national government may make conditional and unconditional grants to the provinces. Provinces thus have little fiscal autonomy. They have no direct access to the major revenue sources, though the Constitution does contemplate the possibility of their laying a surcharge on the national income tax. 28 Local governments 24

Sections 228 (2) (b), 216 and 197 Department of Finance, Intergovernmental Fiscal Review 1999, p. A.2. Formally all revenue raised nationally is subject to equitable division under the Constitution. In practice, however, a 'top slice' is first taken off to service debt and provide a pool for contingencies. The remainder is shared amongst the national, provincial and local spheres of government with local government receiving just over 1 per cent. 26 Section 214 (1) and (2). 27 Inter-Governmental Fiscal Relations Act 97 of 1997, Section 10 (5). 28 Section 228. 25

Multi-Level Governance in South Africa

287

have somewhat more autonomy - at least in theory -because, subject to a regulatory power of the national government to impose maximum and minimum limits, they may impose taxes on property. Executive Intergovernmental Relations The interdependence envisioned in co-operative government places intergovernmental relations at the heart of the South African system. In addition to the principles discussed earlier, the Constitution instructs governments to make 'every reasonable effort to resolve any disputes though intergovernmental negotiation' and to exhaust 'all other remedies' before approaching the courts to resolve them. If a court is not satisfied that such efforts have been made, it can refer the matter back to the governments. Legislative Intergovernmental Relations29 South Africa again follows the German model by establishing a strong second chamber to represent provincial interests in the national Parliament. The National Council of Provinces is to 'ensure that provincial interests are taken into account in the national sphere of government', 'by participating in the national legislative process and by providing a national forum for public consideration of issues affecting the provinces'. 30 The logic is that, because national legislation imposes wide obligations on provinces and determines their resources, it is crucial that they have an effective voice in its development. Provincial delegations are (at least nominally) headed by the provincial premier, and include six permanent delegates selected by the legislature, and three other 'special delegates', who may rotate according to the issue under discussion. Permanent members cannot sit in the provincial legislature, but they are subject to recall, and their term ends when the legislature ends. 31 The NCOP's powers vary according to the legislation in question. If a Bill does not directly affect the provinces, NCOP members vote as individuals (actually along party lines). They may support, amend, or reject it, but can be overridden by a simple majority of the National Assembly. The NCOP has more power on Bills respecting concurrent jurisdiction (Section 76 legislation) or areas of provincial jurisdiction. Here, each province casts a single vote, under instruction - 'mandate' - from the provincial legislature. If the two houses disagree, a mediation committee is set up; and if the disagreement persists, the National Assembly can still overrule the NCOP, but now it requires a two-thirds majority. The N C O P is also the vehicle through which provinces participate in constitutional amendment. Amendments affecting them require the support of six of the nine provinces.

The System in Practice: Assessing Capacities This bare-bones sketch of South Africa's multi-level constitution says little about how the system actually works. Many other factors - such as party systems, economic considerations, the interaction of federal institutions with other elements of the institutional structure, and patterns of identity - affect the 29

See Christina Murray and Richard Simeon, 'From Paper to Practice: the National Council of Provinces after its First Year', SA Public Law 14, 1, 1999, pp. 96-141. 30 Section 42. 31 This differs crucially from the German Bundesrat, which represents provincial executives, thus unifying legislative and executive federalism.

288

Richard Simeon & Christina Murray

operation of federations. This is especially so in a brand-new dispensation because the details of how the institutions were to operate could not be spelled out or even anticipated in advance. In this section we examine some of the emerging dynamics of multi-level government in terms of the 'capacities' listed at the outset. Political Capacity Provinces were created from scratch, with little previous presence in the minds of citizens or government leaders. Consolidating them as viable political institutions, capable of policy-making and implementation, and with a firm political base, has proved a difficult task. Provincial legislatures are elected at the same time as the national Parliament, according to the same list system of proportional representation. National parties prepare the electoral lists for both levels. Provincial elections are not fought on the basis of regional interests, led by regional leaders. Instead, they are part of the national election campaign in which provincial interests are relatively unimportant. Hence there is a common perception that provincial legislators are elected on the coat-tails of national politicians. Service in a provincial legislature is reserved for distinctly lower-ranking party figures. Provincial premiers are not elected on their personal political standing within the province. Rather, they are 'deployed', in the case of the ANC, by the party's National Executive Committee. 32 In 1998, in order to counteract creeping provincialism, the party decided that future premiers would be appointed by national structures; provincial ANC chairpersons would no longer automatically become premier. In effect, provincial premiers are more accountable and responsible to the national party and its leadership than they are to their own legislatures or voters. Thus, national politics dominates the provincial political leadership, preventing the emergence of an autonomous, politically dynamic provincial sphere and the development of fully responsible cabinet government in the provinces. 33 In Gauteng, the premier and two members of cabinet (MECs) are on the National Executive Committee of the ANC, and two other MECs are on the executive of the South African Communist Party, which is part of the governing coalition. In other words, they are more a part of the national than the provincial political scene. This, in turn, limits the role of individual members of provincial legislatures (MPLs), since they do not have a direct role in selecting the executive, as in other parliamentary systems. Their role is further undermined because they have few direct links to their constituencies, though in most provinces an effective informal system of assignment to constituencies has been established. The lack of provincial political autonomy extends into the NCOP, where delegations often act more as party representatives than of distinct regional interests. In addition, provincial civil servants tend to see their role as implementers of national legislation, rather than as servants of a provincial agenda. 32

Extensive reshuffling of ANC provincial cabinets following the second set of elections in 1999 may also suggest a lack of political autonomy. See Richard Humphries, 'Little Continuity in Provincial Posts', Business Day, 7 July 1999, p. 11. 'If anything, the newness of so many MECs to these positions indirectly strengthens the hand of the national government as it interacts with provinces in the MinMECs'. 33 As provincial executives gain more confidence, and provincial branches of the ANC become more assertive, there may be resistance to such domination by the national party. Thus, in Gauteng, the national and provincial parties differed in their candidates for premier. The national executive won the battle, but a sense of provincial autonomy and distinct provincial interests was present.

Multi-Level Governance in South Africa

289

This picture of central dominance over provincial affairs must be qualified. The ANC is not a monolith; it is internally divided along many dimensions. Internal dynamics within the ANC vary considerably among provinces, leading to wide variations in the style and form of political life. Local ANC organizations in Gauteng, the Northern Province, and the Free State have on occasion vigorously opposed national dictation. One observer has suggested that provincial premiers have 'been demonstrably keen to extend the boundaries of their autonomy in policy matters'. 34 On this view, such provincial assertiveness is likely to increase as provincial executives exercise greater responsibilities, as they come to enjoy the perquisites of power, and as they perceive a narrowing of opportunities to move to the national level. Others are more struck by the control that the national government has been able to maintain. Nevertheless, the long-term hegemony of the national party over its provincial branches is by no means guaranteed. As Tom Lodge observes, 'In a political culture historically shaped by patron-client relations, politicians who fail to develop local bases and personal networks may simply end up pleasing nobody.' 35 Provincial leaders will no doubt learn this lesson, but it will require that they begin to see their careers playing out at the provincial level, rather than seeing provincial office simply as a stepping-stone to the national arena. The reputation of provinces in the eyes of their citizens remains poor. One 2000 survey found that only 28 per cent trusted their provincial government, a substantial decline from the 1998 figure of 49 per cent. Local government fared even more badly, with only 23 per cent of respondents asserting trust in local government. 36 These figures suggest that provinces continue to lack legitimacy. This, in turn, encourages provincial politicians to focus their attention on national rather than provincial politics and ensures that they will have little leverage in any disagreement with the central government. Legislative Capacity Three aspects of the legislative competence of provinces need to be assessed. First is their ability to participate in the national legislative process, through the NCOP. Second is their ability to monitor and scrutinize the provincial executive. Third is their ability to initiate legislation in their assigned exclusive and concurrent jurisdiction. In none of these areas have the provinces become centres for legislative innovation and experiment. There are a number of reasons for this. First, provincial legislative attention is so taken up with carrying out mandates imposed from above that there is little room to initiate new activities. Indeed, the demanding task of responding to national legislation, and developing the mandates to be exercised by provincial delegations in the NCOP, leads some observers to suggest that legislating at the provincial level is not the primary job of legislatures. Rather, 'the system of cooperative governance centralizes the law-making function in the . . . national parliament, and hence that it is through the N C O P that provinces exercise their legislative 34

Tom Lodge, South African Politics Since 1994. Cape Town: David Philip, 1999, p. 16. Ibid., p. 24. 36 Robert Mattes, Yul Derek Davids and Cherrel Africa, Views of Democracy in South Africa and the Region: Trends and Comparisons. Johannesburg: Southern African Democracy Barometer, October 2000. The polling results are difficult to assess. In 1999, in a Human Sciences Research Council poll, 46 per cent of respondents 'approved' of the performance of their provincial government (up from only 32 per cent the previous year), and 54 per cent reported that they trusted their provinces. Somewhat larger percentages reported approval and trust of the national government. (Human Sciences Research Council, Pretoria, Media Releases, 18 May 1999, January 2000). 35

290

Richard Simeon & Christina Murray

roles'. 37 But this is an enormous task that is seldom performed well. 38 Provincial legislatures are ill equipped in terms of information and expertise to pass judgement on national legislation. This problem is exacerbated by poor communications between the National Assembly and the N C O P and between N C O P delegations and their provincial legislatures. Draft bills are often provided to the N C O P with little time for provinces to respond. 39 N C O P members are supposed to provide a bridge between the national and provincial legislatures, but their links with both are often weak and ineffective. Individual N C O P members, shuttling between Parliament in Cape Town and remote provincial capitals, are placed under enormous strain. Another problem is that there is little linkage between the exchange of information and ideas in the processes of executive intergovernmental relations, and at the parliamentary level through the NCOP. Indeed, provincial executives take little interest in NCOP matters. This differs greatly from the operation of the German Bundesrat, whose members are provincial executives, thus integrating legislative and executive intergovernmental relations. Occasionally, the N C O P is able effectively to articulate provincial interests. As one member of the Mpumalanga legislature puts it: 'There are many examples where ANC-dominated provinces make amendments to legislation that, in principle, went against what was decided in the NA, or by the ANC for that matter, placing firmly the interests of the province first.'40 But this appears to be an optimistic view. There is little evidence that the party leadership takes the N C O P seriously, except as a minor chamber of second thought. A more likely and useful role for provincial legislatures lies not in law-making, but in the monitoring and oversight of provincial executives and in developing avenues of communication with citizens. Provincial responsibility for implementing national laws means that they have major executive responsibilities. National government supervision can help ensure that these are carried out according to national norms, but only effective scrutiny by the provincial legislature can ensure that they are carried out in ways sensitive to local concerns. In addition, South Africa's history of deep alienation between citizens and government makes the legislature's role in building effective linkages to individuals and civil society especially important. Much needs to be done in these areas, and some provinces such as Gauteng have taken important steps in this direction. 41 As for the third dimension of legislative capacity - the potential scope for provincial initiative - we have seen that the national government predominates over the provinces in most areas. Yet this dominance is not unlimited. It does not prevent provinces from pushing the envelope to test their powers, filling in spaces that the national government has not occupied and supplementing 37 Firoz Cachalia, 'Provincial Legislatures do not Warrant being Maligned', Business Day, 17 November 1999, p. 14. 38 Murray and Simeon, 'From Paper to Practice'. 39 See Walter Sehurutshe, 'NCOP in North West rises above (some) problems' in Provincial Whip. Cape Town: Institute for Democracy in South Africa, 1998. The N C O P must usually deal with Bills referred to it from the National Assembly within four weeks. See Murray and Simeon, 'From Paper to Practice' pp. 105-6. 40 Ebrahim Fakir, 'Mpumalanga still beset with technical and communication glitches' in Provincial Whip. Cape Town: Institute for Democracy in South Africa, 1998. 41 Firoz Cachalia, 'Public Participation in Legislative Processes', in Gregory Houston et al., Public Participation in Legislative Processes, Work in Progress Seminar Paper. Pretoria: Human Sciences Research Council, 1999. Mpumalanga and the Eastern Cape are in the process of introducing public petitions processes modeled on that of Gauteng.

Multi-Level Governance in South Africa

291

national legislation to address local needs. Provinces have yet to exploit these possibilities. They have initiated little legislation on their own. 42 Rather, they appear to believe that their major role is to implement and deliver the services required by national legislation. Provincial legislative potential will be realized only if provincial leaders change this ingrained tendency to defer to the centre. Much will also depend on the interpretation of jurisdictional conflicts by the Constitutional Court. Will it be sympathetic or hostile to provincial legislative activism, sympathetic or hostile to national intervention? Only a few relevant cases have been decided so far, but the Court appears to be reaching for some kind of balance. In earlier cases, in which the Inkatha Freedom Party claimed sweeping powers, the Court emphasized the unitary aspects of the system. 43 A more recent case, focused on the division of powers, upheld national powers over the economy. The Court saw the national government as paramount, with the provinces largely limited to the 'delivery of services in everyday [!] matters such as health, housing, and primary and secondary education'. 44 However, the Court has also asserted that the unitary character of South Africa 'is not absolute, since it must be read in conjunction with the further provisions of the Constitution, which show that governmental power is not located in national entities alone'. 45 Even as it supported the national government in the case of national public service legislation, the Court warned that the 'power of the national legislature is one which needs to be exercised carefully . . . to ensure that the national legislature does not encroach on the ability of the provinces to carry out the functions entrusted to them by the constitution'. 46 In its first certification judgement, the Court emphasized that the Constitution embodies not 'competitive' but 'co-operative government'. This implies that disputes should, where possible, be resolved politically rather than through litigation. The Court has permitted provinces some room in developing their own constitutions, but has been vigilant to keep them within national constitution norms. As noted, it rejected a proposed KwaZulu-Natal constitution, and rejected the Western Cape's proposal to introduce a mixed, rather than purely proportional, electoral system. 47 Thus, the Court is sensitive to federalist values, but in the context of the centralized co-operative model set out in the Constitution. Critically important for the future will be the tests that the Court develops to decide which government will prevail in cases concerning concur42 Between 1994 and 1999, the annual average number of laws passed by the provincial legislatures was eight. This estimate does not take account of the North West Province because at the time of publication the Province had not yet collated the relevant information. See also Nico Steytler, 'Concurrency and Co-operative Government: A South African Case Study'. Paper presented at the Conference on Reform of German Federalism: An International Comparative Approach, Bertelsmann Stiftung, Berlin, May 2000, p. 5. 43 In re: Certification of the Constitution of the Province of KwaZulu-Natal, 1996, 1996 (11) BCLR 1419(CC); 1996 (4) SA 1098 (CC); In re: Certification of the Constitution of the Republic of South Africa, 1996, 1996 (4) SA 744 (CC); In re: Certification of the Amended Text of the Constitution of the Republic of South Africa, 1996, 1997 (2) SA 97 (CC). 44 Ex parte President of the Republic of South Africa: In re: Constitutionality of the Liquor Bill, 2000 (1) SA 732 (CC), para. 76. 45 Ibid, para 20. 46 Premier, Western Cape v President of the Republic of South Africa and another 1999 (4) BCLR. 382; 1999 (3) SA 657 (CC), para 60. 47 This is unfortunate because it seems to close the door on the provincial experimentation that federalism makes possible. Given the problems associated with the purely proportional electoral system, it would have been salutary to permit provinces to experiment with alternatives. See Christina Murray, 'Provincial Constitution-Making in South Africa: The (Non) Example of the Western Cape', Jahrbuch des Offentlichen Rechts der Gegenwart, Neue Folge/Band 49, 2001, pp. 481-97.

292

Richard Simeon & Christina Murray

rent powers and exclusive provincial powers. This requires it to give more precise operational meaning to Sections 44(2) and 146 of the Constitution. Administrative Capacity Perhaps the greatest challenge facing the provinces has been to develop efficient, honest, and effective public administration. The provinces inherited three sets of problems. First, in a number of cases, they were required to absorb the civil servants who had previously worked for the deeply flawed homeland governments, 48 and amalgamate them with provincial administrations. Second, large numbers of formerly national public servants were transferred to provincial payrolls. Third, wages and working conditions continue to be negotiated between the national government and civil service unions, further limiting the provinces' ability to manage their large work forces. 49 Provinces have little autonomy in managing provincial public servants. The Presidential Review Committee of 1998 reminded South Africans that the public service is 'essentially a unitary entity operating at both national and provincial levels'. This view has been affirmed by the Constitutional Court, which in a dispute between the Western Cape and the national government over amendments to the Public Service Act, 'found in favour of the central government, ruling in effect that parliament has the competence to prescribe how provincial administrations may be restructured'. 50 Provinces have also found it difficult to recruit top-level senior public officials. There are more opportunities and more prestige at the national level, and many senior public servants prefer to live and work in Pretoria or Cape Town, than in remote Kimberly or Bisho. Personnel are not the only problem. Provinces have had difficulty building effective administrative structures and processes for management and delivery of services. A 1997 report 51 on provincial administrations found that provinces were insufficiently focused on delivering services to citizens; management was over-centralized; there was too much political interference in administration; strategic planning and organizational design were inadequate; lack of discipline and misconduct were rife; financial management and information systems were inadequate; and the civil service was still unrepresentative in terms of race and gender. There was a critical need to increase staff skills, increase accountability, and improve control of misconduct. Some observers may regard this litany as a decisive reason why provinces should be reduced or eliminated. However, the Provincial Review Report also found that many of the provincial problems originated in the national government. The centre often set new policies without considering their financial and service delivery implications for the provinces. Current public service regulations 'have encouraged a culture of lack of responsibility', and national depart48 Tom Lodge notes that the provinces that have been relatively successful in building administrative capacity - Gauteng, Western Cape, Northern Cape, and the Free State - are those that were least encumbered by the legacy of Bantustan regimes. Lodge, South African Politics Since 1994, p. 14. 49 The problem here is not simply one of under-trained black civil servants, but also of the inheritance of a white-dominated public service, both nationally and in the homelands, which was in many respects under-skilled, imbued with highly traditional models of administration, and often deeply committed to the old regime. 50 Robert Cameron and Chris Tapscott, 'The Challenges of State Transformation in South Africa,' Public Administration and Development 20, 2000, p. 86. 51 Minister for the Public Service and Administration, Provincial Review Report. Pretoria: Department of Public Service and Administration, 1997.

Multi-Level Governance in South Africa

293

ments provided inadequate support for provincial capacity-building. 52 Moreover, none of these problems is unique to the provinces. Most would still exist if all the provinces were abolished and their responsibilities fell to the national government. 53 The fact is that the rapid devolution of responsibilities placed enormous demands on provinces that they did not have the infrastructure to fulfill. Administrative capacity is a profound problem at all levels of South African government. Finally, there are some areas, notably financial management, where provincial capacities have improved in recent years. The Department of Finance, in its Intergovernmental Fiscal Review for 1999, reported that while many problems remain, 'The emphasis on cooperative governance and on laying a sound foundation is paying dividends, as shown by a dramatic turnaround in provincial financial affairs, and improvements in provincial service delivery'. 54 Fiscal Capacity55 Provinces raise only about 4 per cent of their revenues on their own. They are thus entirely dependent for their constitutionally mandated 'equitable share' of the national budget on the quality of the work done by the FFC and the national Ministry of Finance - and on the influence they can bring to bear through the Budget Council and the NCOP. Ideally, the allocation process should be sensitive to provincial needs and driven by objective rather than partisan political criteria. 56 The basic vertical division of revenues, however, remains fundamentally a political decision by the national government, and the influence of the FFC potentially critical for protecting provincial interests - is declining. Equalizing provincial revenues is a mammoth task in light of the stark disparities between historically privileged and historically deprived regions. An equalization programme is being introduced over a five-year period, based on complex measures of need, fiscal capacity, and the like. 57 In the first years under the Interim Constitution, committees within the national government largely determined provincial spending. Beginning in 1997-8, however, the equitable share was transformed into an unconditional grant. This provided somewhat more provincial discretion, shifting the allocation of funds from a process of competition among provinces in each functional area to competition between functional areas within each province. 58 Conditional grants continue in a number of areas - provoking the complaint, familiar in other federations, that they distort priorities and disrupt service delivery. 'Unfunded mandates' through which the national government sets service-delivery standards that provinces cannot support within existing budgets, or transfers functions or responsibilities to provinces with no addi52 Section 125(3). National support has been inadequate partly because the national administration has its own problems of building capacity to make the new system work. 53 Indeed, many of the observations in the Provincial Review Report could apply equally to the national administration. 54 Department of Finance, Intergovernmental Fiscal Review 1999, 'Introduction'. 55 See, for more detail Joachim Wehner, 'Fiscal Federalism in South Africa', Publius: The Journal of Federalism, 30, 3, Summer 2000, pp. 47-72 and Christina Murray and Richard Simeon 'South Africa's Financial Constitution: Towards Better Delivery', South African Public Law, 15, 2, 2000, pp. 477-504. 56 For an excellent survey, see Joachim Hans-Georg Wehner, Intergovernmental Fiscal Relations in South Africa. Cape Town: Applied Fiscal Research Centre, October 1999. 57 Department of Finance, Intergovernmental Fiscal Review 1999, p. 2.11. 58 Ibid., p. A.2.

294

Richard Simeon & Christina Murray

tional resources, continue to be a major problem. 59 Provincial executives are often required to carry out tasks that have not been conferred on them by their own legislatures. Dependence on the national budget, limited revenue-raising capacities, and limited discretion in spending reduce both provinces' effective autonomy and their accountability to citizens. For example, national direction requires provinces to spend 85 per cent of their revenues on health, education, and welfare, necessitating large cuts in other areas, such as transportation. As Thabo Rapoo points out, the centralized taxation and revenue-sharing power 'creates a situation not unlike that in the former "homelands" system in which, because allocations are decided by the center, provinces are not responsible to their electorates for spending - shortfalls can simply be blamed on the central government'. 6 0 The ability of provincial governments to collect the few sources of revenue now available to them is also in question. 61 This creates a chicken-andegg problem: without their own fiscal autonomy, provinces will be slow to develop effective and accountable administrations; but without that capacity, it is more difficult for them to claim a greater fiscal role. There has been considerable debate about the possibility of increasing provincial fiscal autonomy by allowing the provinces to impose a surcharge on the national income tax. The FFC has recommended it as a way to improve provincial fiscal capacity and accountability; others, notably in the Ministry of Finance, 62 worry about interprovincial tax competition, increasing provincial disparities, and administrative difficulties. The Finance Minister has, however, indicated his openness to proposals to increase provincial revenue-raising capacity in other areas. 63 Intergovernmental Capacity Given the interdependence among governments built into the constitutional design, it is not surprising that a dense network of intergovernmental mechanisms began to develop almost immediately. 64 They did not wait for the constitutionally mandated Act of Parliament that would 'promote and facilitate executive intergovernmental relations' and dispute resolution. 65 Approximately 65 intergovernmental bodies had been created by 2000. 66 The national govern59 For example, provinces were directed recently to monitor liquor licences and establishments, but the national government continued to collect the relevant fees, even though liquor licensing is one of the few areas of exclusive provincial jurisdiction. 60 Thabo Rapoo, 'A System in Dispute,' Policy Issues and Actors 8, 10, 1995, pp. 13-14. See also his 'Fertile Feuding: Conflict and Reform in South African Intergovernmental Relations', Policy, Issues and Actors 12, 2, 1999, pp. 15-16. 61 Farouk Chothia, 'Provincial Tax Collection Declines', Business Day, 25 February 2000, p. 3. 62 See also Seventh Interim Report of the Commission of Inquiry into certain Aspects of the Tax Structure of South Africa - Synthesis of Policy Recommendations with regard to Provincial Taxation, Pretoria: 14 July 1998, and see summary in the Department of Finance, Intergovernmental Fiscal Review 1999, p. 2.8. 63 Work on legislation to permit provincial surcharges on a variety of other revenue sources is under discussion. Linda Ensor, 'Council rules out levy on Income tax', Business Day, 22 October 1999, p. 5. 64 Bertus de Villiers, 'Intergovernmental Relations in South Africa', SA Public Law, 12, 1, 1997, pp. 197-213. 65 Section 41(2). Section 41(3) calls on government to exhaust all other avenues of dispute resolution before having recourse to the courts; section 148 calls on the courts to 'prefer any reasonable interpretation ... that avoids a conflict over any alternative resolution that results in a conflict'. 66 'An Overview of Intergovernmental relations in South Africa in the Year 2000,' Ministry of Provincial and Local Government, IGR Focus, 1,1, 2000, p. 4.

Multi-Level Governance in South Africa

295

ment's relations with the provinces are under the broad supervision of the Department of Provincial and Local Government. The Department of Finance also plays a central role through its control of fiscal relations. At the provincial level, most relationships are managed through the premiers' offices, and by individual line departments working with their national counterparts. Among the first new bodies to emerge was the Intergovernmental Forum (IGF) to bring provincial premiers and national ministers together quarterly for policy discussion. 67 However, it proved too large and unwieldy for effective coordination, and in 1999 was replaced by a new President's Co-ordinating Council designed to 'formulate, co-ordinate and implement policy,' with 'integrated approaches to planning, implementation and the allocation of resources'. An annual intergovernmental relations conference was also planned. 68 At the official level, the Forum for South African Directors General (FOSAD) was created in 1998. The national and provincial ministers of finance meet regularly as the Budget Council (known by its members as 'Team Finance') which was mandated by the 1997 act governing fiscal relations. 69 About 20 other ministerial forums (MinMECs) 70 have been formed to discuss policy in specific areas. While these mechanisms are essential to making the system work, their ad hoc status raises a number of questions. Are they arenas for collective decisionmaking or devices through which national ministers can issue instructions to the provinces? Chapter 3 of the Constitution seems to define co-operative government as a partnership among equals, but law and practice make it clear that it is primarily a top-down, paternalistic relationship. As intergovernmental relations become deeper and more complex, South Africa, like other federations, will need to strike a balance between creating highly formalized, institutionalized intergovernmental machinery and maintaining the flexibility to adapt to new conditions. 71 Chris Tapscott notes that attempts to codify the mechanisms and processes of intergovernmental relations may add greater legal certainty to the process, but may also aggravate tensions. 72 South Africa, therefore, must worry about the 'democratic deficit' that arises when governments interact behind closed doors, potentially freezing out the views of citizens. 73 This problem is compounded when the design of the Constitution generates broad areas of shared responsibility, making accountability and clarity to citizens more difficult. 'Central and provincial governments are as quick to claim credit for successes as they are to blame each other for failure, and ... it is often difficult to establish who is accountable for what. So long as each government has the other as an excuse for non-delivery, so long will 67 Permanent members include the Ministers of Provincial and Local Government, Finance, and Public Service and Administration; the Director-General of the President's Office; the nine provincial premiers; and ex officio the President and Deputy President. Chris Tapscott, 'Intergovernmental Relations in South Africa', Public Administration and Development, 20, 2000, p. 124. 68 Xolani Xundu, 'Tuning Govt from Top to Bottom', Business Day, 17 December 1999, p. 2. 69 The Council is supported by prior meetings with the largest spending departments, known as 'four by fours'. 70 Apart from the Budget Council, only the Committee of Education Ministers has a statutory basis. 71 The Constitution calls for national legislation to govern intergovernmental relations. It has not yet been developed, even as the day-to-day operation of IGR continues. 72 Tapscott, 'Intergovernmental Relations', p. 127. See also Ronald L. Watts, Intergovernmental Relations. Pretoria: Department of Constitutional Development and Provincial Affairs, 1999. 73 See David Cameron and Richard Simeon, 'Intergovernmental Relations and Democratic Citizenship', in B. Guy Peters and Donald Savoie, eds, Governance in the Twenty-first Century, Ottawa and Montreal: Canadian Centre for Management Development and McGill-Queen's University Press, 2000, pp. 58-118.

296

Richard Simeon & Christina Murray

the country be held to argument and indecision.' 74 Bertus de Villiers adds that 'More will have to be done to broaden the scope of legislative involvement in intergovernmental relations.' 75 Despite these challenges, the development of a working system of intergovernmental relations is one of the successes of South African multi-level government. De Villiers concluded that by 1996, 'Amazing progress has been made in a very short time in structuring intergovernmental relations ... One cannot but be struck by the pragmatism and urgency that have led all those involved to establish a myriad of intergovernmental forums. Compared to other developing countries this is even more remarkable.' 76 Other analyses are less sanguine. Tapscott concludes that the present, largely ad hoc system of intergovernmental relations is 'in a state of flux and, at times, operate[s] dysfunctionally'.77 Similarly, a substantial study of intergovernmental relations commissioned by the national government identifies serious shortcomings and makes extensive recommendations for improvement. 78 Our judgment is that South Africa has a functioning system of intergovernmental relations, but one far more concentrated on process than on policy content and delivery. Legislative Intergovernmental Relations: The NCOP The N C O P is designed to act largely as a conduit between national and provincial governments. The process is almost impossibly complex. The NCOP's agenda is driven chiefly by legislation coming to it from the National Assembly, though increasingly Section 76 legislation is first introduced in the NCOP. There is little time for Bills to be communicated to the provinces, and for the provinces to carry out their own deliberation before their 'mandates' are returned. In some cases, no mandates are effectively generated, and in virtually all cases, provincial legislatures have neither the expertise nor the information to make considered judgments on national legislation. Intergovernmental relations at the executive level are far more effective than at the legislative level through the NCOP. This is in large part a function of the co-existence of multi-level government in South Africa with a Westminster-style parliamentary system that concentrates power in the hands of executives; 'executive federalism' characterizes all these cases. 79

Conclusion This survey of multi-level government in South Africa has provided a mixed picture. In each of the areas examined, significant governance 'deficits' have been identified. But these should not blind us to the progress that has been made. The 1993 and 1996 Constitutions initiated a new regime in South Africa, an important part of which was a system of multi-sphere government. Institutions and 74

Steven Friedman, 'Why We Need Provincial Governments', Business Day, 9 March 1998, p. 6. De Villiers, 'Intergovernmental Relations in South Africa', p. 212. Ibid., p. 212. 77 Tapscott, 'Intergovernmental Relations', pp. 126-7. 78 Department of Provincial and Local Government and the Ministry for Local and Provincial Government in collaboration with the School of Government, University of the Western Cape and the Institute of Government of the University of Fort Hare, The Intergovernmental Relations Audit: Towards a Culture of Co-operative Government: Final Report of the Intergovernmental Relations Audit. Pretoria: 1999. 79 Intergovernmental relations in Canada and Australia are also almost exclusively conducted at the executive level. 75

76

Multi-Level Governance in South Africa

297

practices had to be constructed and implemented in a very short time in the face of extraordinary challenges resulting from the legacy of apartheid, and from powerful fiscal and other constraints. The surprise is not that South Africa has fallen so short, but rather how successful it has been in establishing a working system of multi-level governance. Yet much about multi-sphere governance in South Africa remains deeply problematic. The fundamental question is whether the provinces are to exist simply as 'a cog in the conveyor belt', 80 as administrative arms of the central government, responsible to its directives and undertaking few initiatives of their own. If this is to be the case, then it may be a fiction to have separately elected provincial governments with their own legislatures. Or, are the provinces to be alternative centres of citizen representation, responsible government, legislative initiative, and oversight of the bureaucracy? Significant improvement in all the dimensions of capacity we have discussed will be necessary to achieve these goals. Despite the constitutional model of three spheres as 'distinctive, interdependent, and interrelated', the first years have shown the overwhelming dominance of the national government. Provinces have taken only small steps to emerge from its shadow, and these steps have been focused much more on service delivery than on autonomous policy-making. As Nico Steytler puts it, it is 'a case of overgrowth by the national (federal) tree, smothering the young provincial saplings. Provinces have yet to come into their own right.' 81 This is partly a consequence of the constitutional design itself, with the predominance of legislative power assigned to the centre, with a centralized system of public finance, and with considerable central powers to oversee provincial performance. However, several other factors have shaped the system, and how they evolve will do much to determine its future. These include: • The limited acceptance of the idea of federalism, as a way of organizing governance, in the eyes of both citizens and elites. The dominant perceptual framework remains focused on South Africa as a single unitary system. As long as such perceptions exist, provinces will be seen as subordinate institutions at best, and as expensive, redundant barriers to good governance at worst. Similarly, at the mass level, provincial autonomy will depend on their success as an alternative source of identification and democratic responsiveness for citizens. Federal institutions must be underpinned by a federal society. • The nature of the national party system. The ANC dominates both national politics and politics in seven of the nine provinces. It is highly centralized. If provinces are to develop the capacity to act more autonomously, then either the ANC itself must become more regionalized, with regional executives able to challenge the national executive, or new provincial party formations need to develop. • Provincial administrative and fiscal capacity. Improvement along both these dimensions is crucial if provinces are to be more effective. • The continuing evolution of judicial interpretation. Uncertainty about each of these factors makes it impossible to predict the future of multi-sphere government in South Africa. One of the greatest unknowns is the 80

A comment by KwaZulu-Natal Premier Lionel Mtshali. See Farouk Chothia, 'Provinces are Powerless Cogs', Business Day> 31 March 1999, p. 5. 81 Nico Steytler, 'Concurrency and Cooperative Government: A South African Case Study'. Unpublished Paper, 2000, p. 15.

298

Richard Simeon & Christina Murray

future of local government, and its relationship to provinces and the national government. Local government is undergoing radical change. The previous 843 municipalities have been consolidated into 284 councils, and in 1996 local government was extended to some rural areas where it had never existed before. Major urban areas, such as Johannesburg and Cape Town, are being reconstituted as six 'Metropolitan Municipalities', or 'Unicities', with strong 'executive mayors'. Outside the Unicities, 46 'district' municipalities, each overarching a number of local municipalities, are to be created, forming yet another tier of governance. For some, especially in the ANC, these developments foreshadow an 'hourglass' system in which the national and local spheres become the dynamic elements, and the role of provincial governments is reduced, if not eliminated. The argument is that three spheres of government are one too many, and that most of the benefits of multi-level government could be achieved by a strong local sector working closely with the national government, bypassing the provinces. Such developments are attractive to the governing party because they seem to address national frustration over provincial capacities and its long-standing hostility to federalism. On the other hand, this view may be wishful thinking, as local government is currently in disarray, under-resourced, and enormously lacking in capacity. As Elke Zuern puts it, 'Political change at the level of municipal government has been halting and difficult; . . . local level transitions [to democracy] have taken place much more slowly than at the national and provincial level'. 82 There remains the more fundamental question: was it a mistake to adopt this system? Would South Africa be better off as a unitary state, or one with just two spheres of government, national and local? Let us return to the broad criteria suggested at the outset. Does Multi-Sphere Government Serve Democracy? We have noted some ways in which multi-sphere governance gets in the way of democracy, posing issues of transparency and accountability. But it seems hard not to agree with Steven Friedman's analysis that provinces do provide an 'alternative site of representation', 83 and that grass-roots citizen groups are likely to have much more chance of being heard at the provincial level than at a distant centre. 'If provinces did not exist, there would be far less opportunity to check remote, top-down government.' Genuine participation is far more likely to develop at provincial and local levels than it is in Pretoria or Cape Town. In post-apartheid South Africa, the strengthening of formal political institutions, and the hegemony of the ANC, have resulted in a severe weakening of the strength and dynamism of civil society organizations - 'formal democracy has been created at the expense of participatory democracy'. 84 The democratic potential of local and provincial government depends greatly on their ability to build strong linkages with a reinvigorated civil society. This, in turn, will strengthen democracy in the country as a whole. Effective provincial governments can also give voice to smaller and poorer regions, whose voices could well get lost at the national level. This perspective suggests that provinces, especially through their legislatures, should work far more diligently at building linkages with civil society and at providing citizens greater access to government. If they cannot play such a role, then provincial legislatures may 82 Elke Zuern, 'The Rise of Formal Democracy and the Demise of Participatory Community Structures in Post Apartheid South Africa', Unpublished paper, 2000. 83 Steven Friedman, 'Why We Need Provincial Governments', Business Day, 9 March 1998, p. 6. 84 Zuern, 'Rise of Formal Democracy', p. 8.

Multi-Level Governance in South Africa

299

indeed be redundant. 85 Moreover, the more provinces are focused on their national legislative role and are caught up in ongoing intergovernmental relations, the less time, energy, and inclination for building a vibrant provincial political system. Provinces will contribute to democratization in South Africa, not as subordinate appendages to the national government, but as articulators of regional interests and links to locally based civil society. Does Multi-Level Government Improve the Capacity to Deal with Ethnic and Cultural Conflict? As discussed at the outset, a common justification of federalism is that it allows regionally concentrated minority groups an autonomous political space in which to promote their own community and defend themselves against domination by the majority. Although this was the objective of some early advocates of federalism in a democratic South Africa, it was not the chosen model. South African multilevel governance does not institutionalize and entrench ethnicity, tribalism, or racialism.86 It thus seeks to avoid the common worry that in so institutionalizing such divisions, federalism can create a 'disbuilding' dynamic, from which minority groups can seek ever greater powers, perhaps even leading to secession.87 At the same time, in both KwaZulu-Natal and the Western Cape, multisphere government has allowed important minorities to have a measure of selfgovernment. It may thus have helped reconcile these minorities to the new regime. 88 The South African model of multi-level governance helps 'bind in' such minorities to the larger system. 89 Indeed, it can be argued that the existence of a provincial government, and the need for the closely matched IFP and ANC in KwaZulu-Natal to work together on national issues, helped bring an end to the political violence in KwaZulu-Natal. In addition, the dispersal of political authority may well mean that some ethnic tensions may be managed at the provincial level, and thus not threaten national political stability. As Donald Horowitz argues, federalism can help mitigate such conflict by providing an arena in which new leaders can be educated and socialized into the complexities of governing in a divided society, and can disperse conflict 'by proliferating the points of power, and can support democracy by making one-party hegemony more difficult to achieve'. 90 85 For a discussion of provincial efforts in this area, see Gregory Houston et al., Public Participation in Legislative Processes, Work in Progress Seminar Paper. Pretoria: Human Sciences Research Council, November 1999. 86 Arguing that South African governance should give a greater role to traditional leaders, the deputy president of the NCOP noted that 'as we carry out this work with traditional leaders, we should also remain vigilant that we do not encourage a retrogressive ethnicity and the re-invigoration of backward customs and habits'. Speech, NCOP, 18 December 1999, distributed by Government Communication and Information System, http: //www.gcis.gov.za. 87 Richard Simeon, 'Federalism and the Management of Conflict in Divided Societies', in Alain-G. Gagnon and James Tully, eds, Democracy in Plurinational Societies, Cambridge: Cambridge University Press, forthcoming. 88 Heribert Adam and Koglia Moodley, The Negotiated Revolution: Society and Politics in PostApartheid South Africa. Johannesburg: Jonathan Ball, 1993. Writing early in the negotiations, they suggest that, were the constitution to impose centralism on KwaZulu-Natal, the result could be secession or civil war. A federal constitution guaranteeing 'meaningful regional autonomy' could avert such outcomes. 89 It may be no accident that the government of KwaZulu-Natal has been (along with Gauteng) among the strongest contributors to the work of the NCOP. On the other hand, the Western Cape has played little role in the NCOP. 90 Donald Horowitz, A Democratic South Africa? Constitutional Engineering in a Divided Society. Berkeley: University of California Press, 1991, p. 217.

300

Richard Simeon & Christina Murray

It is too early to make definitive predictions about the future development of South Africa's multi-level government, much less to come to settled judgements about its contribution to democratization and transformation. However, our analysis suggests some modest conclusions. First, South Africans have succeeded in putting into place the basic institutions and processes of multi-sphere government, and making them work. A framework born of a political settlement is now an established part of the governing structure. Although the system has many shortcomings and difficult challenges to overcome, especially in the development of capacities for co-ordination and for service delivery, and although it continues to evoke considerable criticism, it appears that it is sustainable. It is not about to collapse or disappear. Second, despite the difficulties, there is little reason to believe that multi-level government is a substantial barrier to the wider goals of democratization, socio-economic transformation, and conflict management. The chief obstacles to these lie elsewhere in South Africa's political economy, not in the difficulty of managing multi-sphere government. But, third, it is equally difficult to make a confident argument that multi-level government has proved, at least so far, to be a positive and powerful force for deepening democracy or promoting transformation. That must remain a potential for the future.

17

JOHN BOYE EJOBOWAH

Liberal Multiculturalism & the Problems of Institutional Instability

T

HE last decade of the twentieth century was marked by the global resurgence of particularist claims, especially national claims, and the seeming inability of even well settled democratic states to accommodate these claims. This has triggered academic debate about the relevance of traditional liberal theory for multinational states and the ways in which it could be revised to make it more fitting. On one side is the conventional liberal argument that acknowledges moral and concrete identities but excludes these pluralisms from political life by constructing what it considers to be a just and stable constitutional order from the universal qualities of individuals. At the heart of this argument is the political liberalism of John Rawls. On the other side are the multicultural liberal arguments that recognize and accommodate identities in the construction of a just and stable political order. This second set of arguments engenders institutional arrangements that pay attention to national/ethnic groups. I intend to take up the problems associated with institutions of this type and to discuss possible solutions to them.

Political Liberalism and its Correlative Institutions As a doctrine, liberalism appeals to the universal moral qualities of human beings and uses these as the basis for treating everyone equally. This gives rise to a conception of citizenship that is undifferentiated and expressed in identical legal and political rights. This conception receives an excellent elaboration from John Rawls who develops a political liberalism that acknowledges the pluralism of society but does not recognize it. Rawls develops his theory by using the 'original position' as a hypothetical device to model a condition in which representatives of free, rational and equal beings negotiate fair terms of social co-operation by which none has bargaining advantage over the other. The 'original position' eliminates contingencies of the social world, making it possible for representatives to be symmetrically situated behind a Veil of ignorance'. So situated, representatives cannot foretell who will occupy the top or lower positions of the society being formed. They therefore single out primary goods - things that are generally necessary as social conditions and are means for the pursuit of good life - and adopt principles that guarantee them. The five, in order of priority, are: Earlier versions of this paper were presented at the eighteenth World Congress of the International Political Science Association, Quebec City, August 1-5, 2000 and at the Queens Conference on Ethnicity and Democracy, March 17-21, 2000. It was revised to take account of written comments by Richard Simeon, Jacques Bertrand, and by Rotimi Suberu who was not in either of these conferences but to whom I mailed a copy at the Woodrow Wilson Centre in Washington, DC. 301

302

John Boye Ejobowah

• basic liberties (freedom of thought and liberty of conscience); • freedom of movement and free choice of occupation against a background of diverse opportunities; • powers and prerogatives of offices and positions of responsibility; • income and wealth as means to a wide range of ends; and • the social bases of self-respect. The principles adopted to guarantee and assign priority to these goods specifically regulate political, social and economic institutions, that is, the basic structure of society. Also, they are derived behind a Veil of ignorance', without reference to any of the diverse non-liberal moral doctrines in society. Therefore, they stand independently and command overlapping consensus in society for their fairness (Rawls, 1993: Part One Lecture 1, Part Two Lecture 4, Part Three Lecture 8; 1971: chapter 3). Political unity finds its basis in citizens' understanding that principles that govern the basic structure of society are fair and are independent of diverse moral doctrines and commitments. The constitutional order that arises is one in which citizens share a common identity as bearers of the same rights in the public sphere, but have different identities as members of diflferent doctrinal groups in the private sphere. Thus liberalism is at the political level, for that is where citizens are autonomous and stand as equals. In the private sphere citizens are members of this group or the other, governed by comprehensive doctrines operative in each. The constitutional order generated by the political liberalism of Rawls has no recognition for moral, religious or philosophical pluralisms. These are relegated to civil society where their doctrines are not expected to conflict with the rule of law whose fairness is the basis for the overlapping consensus and unity of citizens. So, the critical underpinnings of the constitutional order are the uniform rights and rules of the public sphere and official neutrality with regard to the pluralism of civil society. The political community that emerges is the single undifferentiated type. Institutionally, this requires the adoption of a unitary form of government, but a federal arrangement is not inconsistent, in so far as there is no deviation from universal citizenship rights. Indeed, in political federations like the United States, Germany, Austria, Australia, Argentina and Brazil whose constitutions provide uniform citizenship rights, the boundaries of the constituent units are difference-blind. In these territorial federations power is dispersed from the center to the diflferent territorial units equally (Elazar, 1987: 128-9; Resnick, 1994: 71). In the units, citizens enjoy rights by virtue of residence and identities do not count. Regardless of the powers they may posses, the units are not expected to violate individuals' rights in any way. Politically and constitutionally, the center and the units are required to enforce the rights of individual citizens and any deviation from universal equality is a violation of the law (Stepan, 1999: 31). The difficulties with Rawls' political liberalism are well known and need not be recited here. What needs to be discussed is the basis for the overlapping consensus and unity of citizens who belong to diflferent moral and philosophical groups in society. The core of Rawls' theory is the original position where the Veil of ignorance' prompts rational individuals to single out and give priority to five primary goods. Self-respect - the fifth primary good - is defined by Rawls to mean a secure sense of our own value and of the worth of our choices of life. This definition is not framed to mean the worth of cultural ways of life, as Jonathan Seglow (1998: 965) has rightly shown. But it should, for 'our sense of self-value stems not merely

Liberal Multiculturalism & Institutional Instability

303

from our role as authors of right but also from the regard shown our ways of life against the background culture of society' (Seglow, 1998: 965). A person whose cultural life is denigrated will not receive respect even as a right bearer and, in fact, will lack the confidence to exercise his/her rights. Rawls' primary good of selfrespect requires that cultural condition be the basis for right in the public sphere, but accommodating this will corrupt the impartial principles (the rule of law) by opening them to the contingencies of the social world. So, in political liberalism there is no respect for identities; instead the latter is relegated to civil society.

Multicultural Liberalism Political theorists have tried to deal with this difficulty in Rawls' argument by presenting identity as the basis for self-respect/autonomy and grounding it on liberal foundations. Will Kymlicka argues that the autonomy of individuals to make and revise meaningful choices and the sense that the choices are worth pursuing require a secure cultural membership. So, culture is the foundation of the autonomy to make life plans and of the self-respect that those plans individuals have made are worthy. But Kymlicka (1995: 76) points out that only cultures that have institutions can provide the choice-context required to exercise autonomy. In multinational states, he argues, public policy supports, in fact, generalizes, the majority culture as the national one. Equality will require that minority cultures be granted special rights that include the right to internal selfgovernment, special representation in intergovernmental bodies, and veto rights over issues that affect them. However, it is only those minority cultures that have claims to a definite territory and have institutions that can provide the structure for freedom that qualify for these rights. Charles Taylor also establishes a direct relationship between cultural membership and human agency. According to him, the individual's identity comes from his/her inner self, but the realization rests on recognition received from others. This engenders individual demands for equal recognition. But Taylor (1994: 32) equates individual recognition with that of the identity group, because the cultural community provides the moral language we use to describe who we are, how we see others, what life encounters mean to us, how we evaluate available options and the course of action we take. So, cultures are entitled to equal respect, and those that are threatened have to be secured. There are differences between these theories of multiculturalism, but what is most important for the project at stake is their point of unity, which is a revision of the rules and rights of liberalism to include cultural pluralism as a value worth defending (Requejo, 1999: 262). Indeed, what emerges from the theories is the inclusion of identity in the conception of citizenship rights and the plurality of these identities in the constitutional order. Whereas the political liberalism of Rawls gives rise to constitutions that recognize individuals as bearers of equal rights and entitlements, multicultural liberalism invites a political community in which disadvantaged identities receive recognition. Historically the institutional arrangements for granting recognition have included federal arrangements that permit oppressed minorities to exercise powers of internal self-government, and power-sharing that allows minorities to be represented in intergovernmental bodies. The 1993 South African arrangement provides a good example. The institutional arrangements have also included language recognition, as done by South Africa, Canada, and Spain, and an electoral system that makes for group representation in the legislature, but coupled with a co-operative inducement mecha-

304

John Boye Ejobowah

nism as used by Lebanon under its National Pact of 1943 that endured until 1975. Of all the above, federalism is most fundamental because of its institutional expression of pluralism and also because its grant of internal self-determination allows groups to govern themselves and remain within their own bounds. Although a decentralized unitary system can also be used to accommodate diversities, the devolution of power does not entail the constitutional grant of internal self-determination, as is the case with Uganda. This limits the system as an arrangement for accommodating groups. An exception might be the British unitary system that has devolved significant powers to Scotland to the point of giving the latter its own legislature. Some will regard the present British arrangement as a shift from a unitary to a federal system. The Rawlsian model of political community certainly requires a unitary constitution or a territorial federal model that has no space for groups. On the other hand, the differentiated political community of the multiculturalists requires what Ivo Duchacek (1970: 348-9) has referred to as polyethnic federalism, meaning a federation whose sub-units coincide with identity groups. This form of federalism could also be asymmetrical, meaning that some of the sub-units have to be granted different competencies and group-specific rights (Stepan, 1999: 21; Leslie, 1994: 37). Historic examples of such federations are Spain, Canada, India and Belgium. Nigeria's federalism is polyethnic but is symmetrical, though its federal credentials are questionable. 1 The Ethiopian federalism is also polyethnic and symmetrical, but there is suspicion about its federalist claims. In the asymmetrical federations, members of disadvantaged groups enjoy citizenship rights by virtue of special powers and collective rights granted to their groups. Certainly, the Catalans of Spain would not have enjoyed full citizenship rights, had it not been for the public status given to their language. South Africa's territorial federalism grants constitutional recognition to practically all its native languages and there is de facto retention of the 1993 consociational agreement regarding the formation of government. On the whole, several multinational states have either used polyethnic federalism or group-specific rights or both of these devices to extend and deepen citizenship within their political boundaries. There is therefore historical evidence to support the arguments of multicultural political philosophers that liberalism could meet its claims to equality by recognizing and defending identity as a primary good. But problems could arise and I identify four: one is endless recog1 Some might argue that the boundaries of Nigerian states are not drawn around ethnic groups and that it will be a mistake to characterize the country's federalism as polyethnic. I shall respond to this by noting that most, if not all, the states primarily take on ethnic identities, however the latter might be defined. For example, Ibo states in the east and Yoruba states in the west do not contain other ethnic groups, rather they contain subgroups of the Ibo and Yoruba respectively. The same goes for the Hausa/Fulani states, excepting those in which other language groups are mixed and could not be neatly separated (e.g., Niger, Bauchi and Kaduna States). The exception might be the minority groups that share several states with other minority groups. Here it has to be noted that where there are multiple minority groups, it is practically impossible for each to have its own state. What the Nigerians have done is avoid splitting groups into two, unless it is unavoidable, as is the case with the Ijaws and Ibos of Delta State. The false argument that state boundaries crosscut identities has its origin in Nigerian policymakers' denial of the explicit use of ethnicity as a basis for drawing state boundaries for fear that it will generate political instability. For this reason, they used colonial administrative boundaries that were actually drawn by colonial anthropologists for the purposes of demarcating ethnic groups for indirect rule. In resorting to old administrative boundaries, Nigerian policy-makers were actually using identities as the criterion. Those who argue that identities were constructed in colonial times will insist that these administrative boundaries gave shape to ethnic groups that exist today.

Liberal Multiculturalism & Institutional Instability

305

nition and institutional instability; the second is political corruption; the third is conflict associated with group control of resources in their domain; the fourth is rights violation. In the section that follows, I discuss these potential difficulties and suggest possible solutions to thfem.

Potential Risks and Possible Solutions a. Group Proliferation and Institutional Instability Firstly, constitutional structures generated by, or consistent with, the equality argument of the multicultural theorists have to accommodate explicitly, at the minimum, all disadvantaged cultures. These are the minority ones that suffer domination by majority culture, though there are historic cases where the reverse is true. Each and every such group has to be recognized by being allowed to govern its own area within the polity. If there are several minority groups and one or two major ones, as is often the case, each has to be separated for special recognition. Disentangling the groups to establish clear-cut boundaries could be an inconclusive business because of the indeterminacy of identities and internal counter-claims, as Abdul Raufu Mustapha has rightly noted in a different context in his chapter in this volume. The separation of one group will be the beginning and starting point for another. As soon as one is recognized another will arise and the process continues ad infinitum. Continuous recognition means that institutions associated with internal self-government have to be created continuously. These institutions define those special rights that create conditions for equality and are not supposed to be regarded as material benefits. But the reality is that they are tangible rights because institutions such as legislative and executive arms of government have to be created and filled, and political representation in intergovernmental bodies has to be made. These automatically open the door to new claims to minority status even from within the majority group(s). So the inevitable outcome of group-expressive arrangements is group proliferation and institutional instability. Duchacek seemed to have recognized this difficulty when he made a related remark about polyethnic federalism. According to him: 'No polyethnic system in the world can integrally implement the federal ethnoterritorial principle according to which every ethnic community, however microscopic it may be, should administer its own self-contained autonomous area' (Duchacek, 1991: 30). The group proliferation problem assumes that federal decentralization is the only mode of recognizing and accommodating groups, which is not true. Unitary devolution could also be used, except that it might not provide the institutions of self-government. In which case it ceases to be helpful as an instrument for creating the equality condition for oppressed minorities. But the British form of unitary devolution comes close to meeting the equality condition, for Scotland is permitted to exercise internal self-government through its own legislative and executive institutions. This approaches the federal option, and it brings back the problem of endless demands but in the form of greater powers. Indeed, current debate in the United Kingdom is on the issue of whether greater devolution to Scotland will eventually lead to more demands and eventually to secession. In Nigeria the problem of federal ethnoterritorial recognition is one of the emergence of new groups with new claims to recognition. Kymlicka provided a rule for limiting recognition when confronted with the cases of the United States and Canada where there are innumerable ethnic com-

306

John Boye Ejobowah

munities. His rule is that only identities that are tied to homelands and have culture-sustaining institutions should be given special rights. This rule limits recognition to non-immigrant groups such as the Aboriginals, Hispanics and French-speaking Quebecois and excludes several immigrant groups in those two countries. This will not work in regions like Africa that is predominantly nonimmigrant. There, groups are made up of subgroups and each has its own piece of territory and cultural institutions. Recognition will be an endless business and continuous fragmentation or duplication of state institutions will make the concern for stability a primary issue. The dialectic is such that the concern for stability and order will trump recognition claims. In this respect, Alan Cairns' argument that 'federalism can contribute to interethnic harmony or civility only when the ethnic groups in question are territorially concentrated and thus capable of . . . exercizing limited powers of self-government in provinces or states', might not be entirely correct in the case of Africa (Cairns, 1992: 112). Should polyethnic federalism - symmetric or asymmetric - be abandoned for an arrangement that has no room for identities, given the difficulties discussed above? From the viewpoint of the multicultural theorists, an abandonment of group-expressive arrangements on account of instability would amount to a triumph of injustice over justice. While stability is important and indeed a prerequisite for any human endeavor, pursuing it without concern for justice might be self-defeating, as Hobbes' theory proved to be. 2 Empirically, we have seen this in most African states where concern for stability has been invoked by both civilian and military dictators to justify their centralization and concentration of power. Perhaps a unique case is the Sudan where the Sharia was adopted by the military regime of General Numeiri in 1983 and has been vigorously implemented as the supreme law of the land since 1989 when the fundamentalist regime of General Hassan El-Beshir came to power (Fluehr-Loban, 1990: 617; Oyemu, 1999). Southern rebellion against the powers in Khartoum, and the civil war that has raged in that country since 1989, are representative of the intense violence experienced in other parts of Africa on account of political intolerance, group-suppressive arrangements, and centralization of power. So, suppressing the moral claims of groups through the adoption of non-expressive arrangements is more likely to increase political and social instability than recognition will do. The fundamental issue, then, is not the denial of group recognition but finding a solution to the problem of internal differentiation and endless fragmentation of political units. 3 One solution is to listen to claims and limit recognition to those which the functioning federal and state institutions have failed to accommodate. If the existing constituent units do not have the capacity to accommodate some groups within them, then recognition would be justified. The advantage of this option is that it steers the state away from the authoritarian policy approach to ethnic demands by providing institutional mechanisms for the expression and peaceful resolution of grievances.4 This gives a feeling of security to groups and confers 2

Hobbes' preoccupation with stability yielded political institutions that concentrated executive and legislative powers in a ruler that was also sovereign and self-perpetuating. 3 The next three paragraphs are drawn heavily from my book, Competing Claims to Recognition in the Nigerian Public Sphere: A Liberal Argument about Justice in Plural Societies. Lanham, M D : Lexington Books, 2001, pp. 146-7. 4 In the 1960s and 1970s authoritarian African regimes developed an avoidance policy for dealing with ethnic claims. The policy entailed insulating the state from groups, and in most cases this took the form of outright repression or the imposition of a one-party or no-party system, as was the case in Sudan, Chad, Ethiopia, Ghana, and Sierra Leone.

Liberal Multiculturalism & Institutional Instability

307

legitimacy on the system. However, it does not resolve the problem at stake because the readiness of regimes to listen and evaluate claims for their merit provides opportunities for rival entrepreneurs from the same ethnic group to mobilise local support for political recognition. It was precisely the willingness of successive Nigerian regimes to receive and consider new claims that produced an endless state-creation exercise in the country. Another solution is to adopt a policy that prohibits further fragmentation once devolution occurs. This might be considered a reactive and vexatious policy and new regimes might set it aside, as was the case during the Alhaji Shehu Shagari regime in Nigeria and during the military regime of General Ibrahim Babangida. 5 A prohibitive policy may not endure because different regimes will regard it as not reflecting the social circumstances that confront them. A third alternative is to emphasize a viability requirement. This is the requirement that any sub-unit that is being demanded is able to stand on its own economically. Duchacek made this point when he noted that for a new state to be created it must have the necessary economic means to provide for its economic existence (1970: 239-42 and 314-15). Mexico offers a good example of a country that has a stringent economic condition for the creation of new states within the national territory. The marginal attention paid by successive Nigerian regimes (since the 1970s) to the notion of economic self-sustenance in their treatment of group claims may have been responsible for the dramatic increases in the number of units from 12 to 19 and then to 36. This viability option seems reasonable from the purely economic perspective, but would be fruitless because, in countries like Zaire, Angola and Nigeria where rich minerals are concentrated in one part of the country, every village in such an area could easily meet the requirement. In this case, every village having natural resources under its soil will be mobilized by its local elites to make claims to difference and statehood. To grant such claims would not only trivialize federalism, but would also be unjust to millions of people living in the rest of the country. At another level, the normative argument for political recognition of groups collapses if the economic requirement is taken seriously. In a context where revenue-yielding resources are concentrated in one part of the country, it is likely that polyethnic federalism will not work if economic viability is stressed. If Nigeria's political rulers had placed much emphasis on self-sustenance as a condition for political recognition, it is unlikely that the nineteen-state structure would have emerged as it did in 1976, and the adoption of federal character would have been severely compromised. Emphasizing it would seriously question the multi-state structure that has been used to separate groups and has served as the basis for nurturing difference in politics. Perhaps a better alternative is to accompany recognition and institutionalization of difference with policies and institutions that bind groups together. Federalism or a highly decentralized unitary system of the British type has to be understood as not depending on an exclusive identification with either the whole country or the separate parts. Rather, it implies and also requires a balance 5

In Nigeria, the 1975/76 Constitution Drafting Committee that produced the 1979 Constitution reacted to pressure for new states by drafting constitutional provisions requiring certain procedures to be followed. The Committee technically prohibited further multiplication of units by making the procedures difficult and complex, yet the civilian regime that was installed in 1979 tried to undermine the Committee's goal by slowly navigating its way around them. Similarly, in 1986 a government commission - the Political Bureau - prescribed a 25-year moratorium on the creation of states, but the prohibition was set aside by the very regime that set up the commission.

308

John Boye Ejobowah

between national and local identities. Adopting policies and institutions that bind the separate parts together, as argued by Horowitz (1991), could maintain a balance of some sort. These will include a just resource-sharing regime, an electoral formula that induces political co-operation across group lines, and a constitutional requirement that the federal ethnoterritorial units should stabilize once they have come into being. Since continuous multiplication undermines the recognition project, the separation of groups should be done only once and the units be given time to operate intact so that they can consolidate. A policy that requires the existing federal ethnoterritorial units to remain intact for some decades would lead to an adjustment process whereby ethnic elites and their followers become accustomed to operating within the framework of the units in which they are grouped. Over time, they will adjust to accommodate themselves and build up networks of political, social and economic exchanges. To the extent that they get used to living in, and develop a sense of attachment to, the units in which they are grouped, continuous demands for new ones will fizzle out and the system will stabilize. b. Political Corruption One important prescription of the multiculturalists is group representation (at least of the disadvantaged ones) in national institutions. Apart from the right to self-government already discussed, groups have to be represented in the central government and in national bodies. This prescription does not differ from the consociational arrangement in which groups are proportionally represented in high offices of state. In a deeply divided society, group representation could certainly be a fair way of ensuring equal participation in government or sharing the rewards of citizenship, but the practical operation of this form of representation raises the problem of nepotism and corruption, which I explain below. Group representation is no more than sharing political offices and national resources among identity groups. Universal standards (the merit criterion) have to count less; if not, disadvantaged groups will suffer injustice until they are wiped out of existence through assimilation. In the multiculturalists' arguments, identity emerges as a key variable in the distribution of offices and resources. The import of this is that access to public office and resources will have to be governed not only by universal standards but also by ethnic membership. In turn, formal bureaucratic relations that govern the public domain will open up to particularistic informal relations or will be infected by the form of politics that obtains in most deeply divided societies. The politics that obtains in most of these societies is one that is built around ethnic-based patron-client relationships or what Richard Joseph has referred to as prebendalism. In the relationships, factional and internally competitive ethnic entrepreneurs maintain a dominant position in their community by patronizing an array of clients who, in turn, act as patrons by maintaining their own local clienteles whose tentacles reach every village and residential area. So, support is mobilized from bottom to the top and, in return for this political service, patronage has to flow from top to bottom. The 'Big Man-Small Boy' politics that Bruce Berman referred to earlier on in his chapter in this volume becomes very pervasive and, indeed, felt everywhere.6 Now, group representation in national institutions as prescribed by the multi6 'Big-Man' is a word used to designate a patron, while 'Small-Boy' refers to a client who offers his/her services in exchange for patronage that might range from periodic monetary reward to bags of rice or periodic assistance in gaining access to public institutions.

Liberal Multiculturalism & Institutional Instability

309

culturalists has the potential of grafting these ethnic notables into higher offices of state and providing them with resources for servicing their hierarchy of clients below. Indeed, in countries that have devised institutional arrangements to accommodate groups, particularistic informal relations have permeated the public domain to generate the practice whereby clients short-circuit formal procedures by using their patrons to gain favoured access to the state. As Berman and Lonsdale have shown in their respective chapters, any official decision, no matter how inconsequential, provides opportunity for the big man to appropriate resources for himself, his 'boys,' and his 'people'. Political tribalism (interethnic competition for state resources) thrives in this setting and the 'who you know' syndrome displaces informal rules that govern conduct in public institutions. This has certainly been the case in Nigeria since the adoption of the federal character policy in 1979. It is also best exemplified by the Middle Eastern state of Lebanon, where the National Pact of 1943 produced a consociational arrangement that shared offices between Christians and Muslims and among the various sects within each. Between the independence era and 1975 when the arrangement remained in force (apart from 1958 when there was a brief civil crisis), particularistic informal relations so permeated the bureaucratic structure that the latter simply collapsed and its rational norms ceased to be operative.7 Michael Johnson, who conducted an extensive field research in the country, noted that: Clientilism pervaded many aspects of economy, society and politics in independent Lebanon. Relations between employer and employee, and between landlord and tenant often took a clientilist form . . . In order to find employment, to be promoted, to enter one's children in school, to gain admittance to a hospital, to obtain a passport or a trade licence, or to have any dealings with the state administration which were not of the most routine and trivial kind, it was necessary to make use of an intermediary or wasita (pronounced wasta). This could be a friend, a relative, a neighbour, a co-religionist or a za'im. The most important thing was that he or she was in a position to assist in providing the service. The wasita mentality was so widespread that it was often impossible to perform the simplest tasks, without an intermediary. (Johnson, 1986: 97)8 Indeed, factional quarrels among powerful politicians at various levels of the clientelist hierarchy partly accounted for the breakdown of law and order and the fighting of 1975-6 (Johnson, 1986: chapter 7). 9 This indictment is worrisome more because consociationalism best embodies group representation and the Lebanese case was presented by the institutional theorist Arend Lijphart (1977) and others like Eric Nordlinger (1972) as a successful application of power-sharing that should be copied by other fragmented societies. What should be done, given the problem of political corruption that goes with group representation in government? Perhaps one response is to reject the prescription altogether and go by the Rawlsian universalist principles that recognize individual abilities rather than group membership. The problem with this response is that, in a multi-ethnic society, a commitment to equality cannot be met by the sole use of the universalist principles that are known to promote the good of some sections of the country. What is required is a set of principles that take account of the interest of all, not some. In this respect, universalism has to be balanced with group representation. Perhaps, one way of creating a balance is to regard the problem of political 7

For a discussion of consociationalism in Lebanon and why it failed see Michael Hudson (1988: 224-39). 8 Substitute the name Nigeria for Lebanon and the description will fit perfectly. 9 For a different interpretation see Hudson (1988: 224-39).

310

John Boye Ejobowah

corruption as an ailment that can be minimized if the political will is mustered. Having the political will means that group representation has to be understood not as amounting to the relaxation of rational legal rules to accommodate unknown particularist rules, or the use of public resources to patronize a hierarchy of clients below, but rather as meaning that public institutions have to be representative of, and responsive to, the various cultural segments in the country. This should not be translated to mean that public resources can be used by occupants of higher roles to maintain themselves and their personal followers, or that rational procedures can be subverted to satisfy personal interests and those of one's political clients. A commitment to minimizing the use of offices for private purposes will require a clear separation of public offices from the personal pursuits of their occupants and the rigorous enforcement of a strict regime of laws that are designed to maintain that separation. Nigeria exemplifies this where the elected President Olusegun Obasanjo has tried to enforce adherence to bureaucratic norms and standards without compromizing the policy on group representation. However, Obasanjo remains encumbered by the very patron-client networks that brought him into office. c. Rich Natural Resources Arguments for group-specific rights that include rights to internal self-determination require that identity groups have control of revenue-generating resources in the territorial space where they exercise self-rule. Jurisdictional right within one's territory is a central element of self-government, and this jurisdictional right includes the right to independent economic sustenance. In genuinely federal arrangements, whether symmetric or asymmetric, the units have independent sources of revenue, which are usually within their territory. These include income taxes (personal and corporate), commodity taxes, rent exacted from enterprises that are given the right to extract resources, and outright production and marketing of rich natural resources. In the asymmetric federalism of Canada, the provinces control and develop resources within their territory, and in the symmetric federalism of the United States, the states have taxation powers that are considerably less than what obtains in the provinces north of the border (Whyte, 1983: 206-17). These should be seen not as models, but as empirical illustrations of the economic rights exercised by autonomous units. The rights co-exist with that of the centre to its sources of revenue. However, a balance between national and local rights is also established through intergovernmental transfer from richer to poorer regions. A constitutional grant of internal self-determination to groups will inevitably include the right to resources in the area. From the multiculturalists' viewpoint, resource control is not only consistent with the exercise of internal autonomy but also necessary for groups to maintain their distinctive ways of life (Barry, 1998: 315). Such control should not be a problem, especially if the national economy is industrialized and wealth is not concentrated in one part of the country. However, difficulties arise if the economic well-being of the entire country is dependent on revenue-yielding resources concentrated in one area. In this scenario, the control of resources by ethnic units will put at risk the economic survival of the greater part of the country that is economically disadvantaged. Such control will be considered by the have-nots as excessive privilege, and will definitely be resented. Take the case of East Africa, where the Kingdom of Buganda fought to resist political incorporation into a united Uganda during the late 1950s. A constitu-

Liberal Multiculturalism & Institutional Instability

311

tional arrangement that was negotiated at independence granted a federal status to Buganda and a semi-federal status to four other kingdoms, while the rest of the country was given a unitary status. Thus, the country had units that were federal and semi-federal in their relationship with the central government, and several others that had a unitary relationship. With its special autonomy, Buganda had independent sources of revenue, like most provinces or states in federal systems, and enjoyed enormous financial power relative to the rest of the country. For example, in 1964-5 its revenue was 10.3 per cent of that of the central government, while, on the other hand, the combined revenue for all the semi-federal units in relation to that of the central government was 6.58 per cent. In addition, it received federal grants from the centre. Buganda's wealth, which was directly linked to its special position in the country, was considered to be excessive privilege. In fact, other regional negotiators at the constitutional conference that conferred special status on the kingdom had warned that the special rights would not last, and that they amounted to unequal treatment of the 'tribes' in Uganda. The four semi-federal units, the parts of the country under unitary arrangements, and political leaders on the national stage all resented the privileged status and looked for ways of abolishing it. Buganda's demands for control of police, courts, and other demands that were considered excessive presented the opportunity for Milton Obote's government to destroy its autonomy. 10 Although Buganda was the most populous region and generated its revenue from taxes and levies, its conflict with the rest of the country and the eventual abolition of its autonomy demonstrate that the control of resources by self-governing groups does not make for peaceful coexistence. What this case suggests is that the right to resources, as part of the groupautonomy package, does not make for political unity and stability. Instead, it generates resentment and antagonisms. There is always the feeling on the part of the have-nots that the country belongs more to the haves and that they are being neglected by the government, or that the haves are enjoying a privileged status that must be abolished. 11 What should be done, given a situation in which resources are concentrated in a few ethnoterritorial units? Daniel Weinstock (2001) has shown that countries whose economies are centred on resources derived from one part of the country would be better sustaining a unitary system because they need a strong central government capable of distributing the wealth. One option, therefore, is to give top priority to political unity and stability by centralizing resource ownership. This will entail making the central government the main beneficiary of revenue derived from resources in the country. This is what Uganda with its unitarian but decentralized political arrangement did in 1999 when oil was discovered in the north-western part of the country. The government quickly brought in a law declaring state ownership of all natural resources under the soil. This was following in the footsteps of Nigeria where petroleum is located in the minority Niger Delta area and the state has declared its ownership of the rich resource since the late 1960s, with a view to distributing the revenue equally among all groups in the country. After its three-year civil war, the country's political rulers opted for equal distribution of wealth as a way of ensuring 10

For further discussion of this see Augustine Wamala (1994: 255-7). Thus, in Nigeria a 100 per cent derivation principle for revenue-sharing that was negotiated in 1954 hurt Ibos who were not endowed with export crops. It had to be tempered in 1959 with a new formula by which 50 per cent of revenue went for derivation and the other 50 per cent went into a national Distributive Pool Account that was shared equally. 11

312

John Boye Ejobowah

harmony. Certainly, the country would not have been able to sustain its federal policy without the central control and distribution of oil wealth derived from the Niger Delta. Indeed, consociational writers present equal sharing of resources as one of the principles that make for political co-operation. The problem with the centralization option is that it denies decision-making powers to units that ought to be self-governing. This denial is also inherent in the prescriptions of consociational theory that require proportional sharing of resources among federal ethnoterritorial units. This prescription grants selfgoverning rights to groups but denies them the powers to make decisions about revenue-yielding resources within their domain. Thus, in monocultural countries the burden of political unity has to be borne by those few groups in whose domain resources are concentrated. Very often, those groups are minorities. In theory they are sub-units, but in practice they have no decision-making powers regarding revenue control. The point of granting internal self-government is meaningless, as effective decision-making powers are lacking. So, centralizing resources is another way of allowing the concern for unity to trump the concern for group justice. It is this denial of justice that has bred extreme violence in the Niger Delta of Nigeria and turned the region into one of the world's major hot spots. The conflicts in Sierra Leone and Angola are not ethnic, but are driven by resource wealth controlled by the central government. The same could be said of Zaire where the conflict has more of an ethnic flavour than in the other two countries. Another alternative is to strike a balance between centrist and autonomist options by concentrating ownership in the centre but returning some proportion of the wealth to the units where they are derived, as well as establishing institutions to enable the centre to respond to the developmental needs of the resource-bearing areas. This is what Indonesia has done recently. Nigeria has also been exploring this approach since the second half the 1990s, but the problem here is that the approach is paternalistic. It treats the resource-rich areas like infants that have to be guided and looked after by a knowledgeable and caring centre. This type of arrangement compromises their autonomy as independent units of government. Perhaps a better alternative is to promote stability through justice by requiring the federal level to control resources in federal lands while the units control those in their own lands. This obtains in the United States where the states also receive part of the royalties from minerals produced on federally owned lands within their boundaries, but have no right to what is found in the outer continental shelf. However, the federal government levies taxes (like the windfallprofit tax) on minerals produced on non-federal lands (Mieszkowski and Toder, 1983: 68 and 76-7). A different, and perhaps opposite, form of this obtains in Canada where the provinces own public lands and have the right to the resources of those lands. It is uncertain if the federal level exercises any right, other than the power to impose taxes and to make laws for peace, order and good government of the country (Whyte, 1983: 206 and 214). In any case, there is a fiscal equalization program by which the federal government transfers fiscal resources to the have-not provinces in order to raise their revenue to the average obtainable across the provinces. 12 In Spain, the transfer of wealth to poorer regions is done under the 'principle of solidarity' (Agranoff, 1994: 77). 12 For an individualist and market-oriented examination of the concept of equalization payments see Buchanan (1980) and Scott (1980).

Liberal Multiculturalism & Institutional Instability

313

Applying what obtains in the United States to African countries might be problematic because it requires differentiating federal lands from those of the constituent units. It will be difficult, if not self-contradictory, to map out parts of ethnoterritorial units as federal lands. It might be easier to demarcate the national capital and the continental shelf (for non-land-locked countries) as federal territories, but beyond these there will be difficulties as to how it should be done. In short, federal lands would amount to no more than the continental shelf and the national capital. The Canadian approach does not help because states/provinces with rich resources will have everything, while the rest of the country will be left to itself. Equalization payments could offset the financial pains of the have-nots if they are high enough, but this will be in the context of a highly diversified economy in which the entire country is not dependent on the concentration of resources in a few provinces. In a context where there is such concentration, the efficacy of equalization payments will be dependent on how much the federal government can extract in taxes. It will also depend on how much 'owners' of the rich areas see themselves as members of a national community, which entails an obligation to share. On balance, a possible best alternative is to strike a balance between national and local rights by recognizing the rights of ethnoterritorial units to control resources in their lands without compromising the economic survival of the entire country. This will require that: • the units control resources in their territorial space; • revenues derived therefrom are heavily taxed by the federal government to fund an equalization program that transfers wealth to economically disadvantaged regions; • the federal level spends less on the provision of infrastructure in the rich units. The positive aspect of the option is that the federal level could obtain through taxation what it would have obtained through direct control of resources, unless there is lack of trust that the rich areas will pay up. In his chapter in this volume Bruce Berman has offered both a theoretical and a historical explanation for the virtual absence of 'broad and abstract collaborative trust' in the economy and institutions of contemporary African states. True, the problem of trust is real and is best exemplified by the non-acceptance of cheques in business transactions or the prevalence of the cash-and-carry syndrome. However, the trust problem should not arise between levels of government if there is a constitutional agreement on resource-sharing, unless parties to the constitutional agreement do not intend to abide by what they set out to negotiate; in which case, conflict will reign. Taxation by the centre and control by the rich areas as spelt out above is the possible best option. This is consistent with Kymlicka's (1996) argument that national minorities which are vulnerable to political decisions made by the majority should be granted substantial control over the use of resources in their own land. Individual Rights Reconciling individual rights of citizens with particularistic claims of ethnonational groups is another possible challenge that might arise if institutions are constructed to take account of difference. Most of the existing unitary states with some sort of decentralization are already experiencing real contestation over the meaning of citizenship in the regional/provincial/local communities. Even though these decentralized political sub-units do not coincide with ethni-

314

John Boye Ejobowah

city, the 'native-born' make proprietory claims to them by virtue of ancestry. 'Blood and soil' or 'son of the soil' is evoked to control political positions and other privileges, to the exclusion of 'strangers' who may have been born in their area of residence, lived all their lives there, and contributed immensely to its development. This denial of citizenship rights to internal immigrants occurs in a context in which the constitutional order is not designed to institutionalize difference. In his chapter in this volume, Dickson Eyoh shows that, in the case of Cameroon, the so-called strangers challenge the particularization of rights by invoking residence and national citizenship rights, and that contest over the meanings of citizenship goes all the way down to the clan and village levels. This is what obtains in many African states. It is even more intense in Nigeria where the boundaries of the thirty-six states and the 774 local government areas follow ethnicity closely, and the country's federal policy requires parental origin for appointments to political positions in government at each of these two levels. In these states and local governments, the federal character of policy makes it possible for sons of the soil to deny political rights and privileges to internal immigrants. How, then, can individual rights be respected in this context? How can the stranger become a citizen or the immigrant a native? The answer is straightforward: residence as the criterion for definition of membership of the political sub-units. This has been a standard liberal condition for the extension of rights to immigrants. Although immigrants might periodically visit their original homelands and even think of going back permanently some day, the reality is that they do not go back to stay. Their newly chosen places turn out to be their permanent homes. They speak the local language fluently, contribute to the economic and social development of their places of settlement, become bonded to the communities in which they live, and get disoriented when separated from their adopted homes. Justice requires that they be treated equally in the assignment of rights. Very often, original natives harbour the fear of political domination if 'outsiders' are given political rights. Indeed, in multi-ethnic societies, the question of who owns the land (this could be the province/region, city, town, or village) is always at the centre of politics, and 'indigenes' respond by excluding resident 'strangers' from political offices. However real this fear of domination might be, denying political rights to immigrants is a response taken too far. Immigrants do not constitute a majority in the communities where they live and are not in a position to dominate both elective and appointed positions. Where they constitute a significant percentage of the population, the fear of domination can be addressed by placing a limit on the public positions they can fill. In effect, political recognition of groups is not completely at odds with individual rights; where there is a clash, the former can be reconciled with the latter.

Conclusion To sum up, while ethnic identity has political salience in some African states, it might not be so in others. At one end are states like Rwanda, Burundi, Nigeria, Ethiopia, Kenya and Sudan that are prone to intense ethnic conflict, and at the other pole are states like Mali, Ghana, Botswana, Mauritius, Tanzania and Zambia where relative harmony prevails among groups. The rest are somewhere between these two poles. In South Africa the post-apartheid constitutional arrangement went far in minimizing racial conflict but, as Cheryl Hendricks

Liberal Multiculturalism & Institutional Instability

315

shows in her paper, recognition claims remain, especially among the Coloureds. The country would not fall at the Nigerian end of the continuum; rather it will be in the middle. In those states where identity claims have political salience, constitutional arrangements consistent with the prescriptions of the liberal multiculturalists might do justice to marginalized groups and make for peaceful coexistence. However, such arrangements could generate institutional instability and ethnically based patron-client networks that elevate personal ties over standard rules in the conduct of public affairs. They could also generate conflict over ownership rights to rich natural resources, as well as give rise to practices that elevate group rights over individual rights. Constitutional designers have to be alert to these difficulties and be prepared to provide remedies or reconcile inconsistencies where they are likely to arise.

References AgranofF, Robert. 1994. 'Asymmetrical and Symmetrical Federalism in Spain', in Bertus de Villiers, ed., Evaluating Federal Systems. Cape Town: Juta. Barry, Brian. 1998. 'The Limits of Cultural Politics,' Review of International Studies 24: 307-19. Buchanan, James. 1980. 'Federalism and Fiscal Equity', in Bhajan S. Grewal, Geoffrey Brennan and Russel L. Mathews, eds, The Economics of Federalism. Canberra: Australian National University. Cairns, Alan C. 1992. Charter Versus Federalism: The Dilemmas of Constitutional Reform. Montreal: McGill-Queen's University Press. Duchacek, Ivo. 1991. 'Comparative Federalism: An Agenda for Additional Research', in Daniel J. Elazar, ed, Constitutional Design and Power-Sharing in the Post-Modern Epoch. Lanham, MD: University Press of America. — 1970. Comparative Federalism: The Territorial Dimension of Federalism. New York: Holt, Rinehart and Winston. Ejobowah, John Boye. 2001. Competing Claims to Recognition in the Nigerian Public Sphere: A Liberal Argument about Justice in Plural Societies. Lanham, MD: Lexington Books. Elazar, Daniel. 1987. Exploring Federalism. Tuscaloosa, AL: University of Alabama Press. Fluehr-Lobban, Carolyn. 1990. 'Islamization in the Sudan: A Critical Assessment', The Middle East Journal^ (4). Horowitz, Donald. 1991. A Democratic South Africa? Berkeley, CA: University of California Press. Hudson, Michael. 1988. 'The Problem of Authoritative Power in Lebanese Polities', in Nadim Shehadi and Dana Haffar Mills, eds, Lebanon: A History of Conflict and Consensus. London: The Centre for Lebanese Studies in association with I. B. Tauris and Co. Johnson, Michael. 1986. Class and Clientilism in Beirut. London: Ithaca Press. Kymlicka, Will. 1996. 'Concepts of Community and Social Justice', in Fen Osier Hampson and Judith Reppy, eds, Earthly Goods: Environmental Change and Social Justice. Ithaca, NY and London: Cornell University Press. — 1995. Multicultural Citizenship: A Liberal Theory of Minority Rights. Oxford: Clarendon Press. — 1989. Liberalism, Community and Culture. Oxford: Clarendon Press. Leslie, Peter. 1994. 'Asymmetry: Rejected, Conceded, Imposed', in Leslie Seidle, ed., Seeking a New Partnership: Asymmetrical and Confederal Options. Montreal: Institute for Research on Public Policy. Lijphart, Arend. 1977. Democracy in Plural Societies: A Comparative Exploration. New Haven, CT: Yale University Press Mieszkowski, Peter and Toder, Eric. 1983. 'Taxation of Energy Resources,' in Charles E. McLare and Peter Mieszkowski, eds, Fiscal Federalism and the Taxation of Natural Resources. Lexington, MA: Lexington Books (D.C. Heath and Company). Nordlinger, Eric. 1972. Conflict Regulation in Divided Societies. Occasional Paper, Centre for International Affairs, Harvard University. Oyemu, John. 1999. 'The African-Arab Conflict in the Sudan: How Can it be Resolved?' A seminar paper delivered for Pol 488F, University of Toronto. Rawls, John. 1993. Political Liberalism, New York: Columbia University Press. — 1971. A Theory of Justice. Cambridge, MA: Harvard University Press. Requejo, Ferran. 1999. 'Cultural Pluralism, Nationalism and Federalism: A Revision of Democratic Citizenship in Plurinational States,' European Journal of Political Research 35: 255-86. Resnick, Philip. 1994. 'Toward a Multinational Federalism: Asymmetrical and Confederal

316

John Boye Ejobowah

Alternatives.' in Leslie Seidle, ed., Seeking a New Partnership: Asymmetrical and Confederal Options. Montreal: Institute for Research on Public Policy. Scott, A. D. 1980. 'Notes on Grants in Federal Countries.' in Bhajan S. Grewal, Geoffrey Brennan and Russel L. Mathews eds, The Economics of Federalism. Canberra: Australian National University. Seglow, Jonathan. 1998. 'Universals and Particulars: The Case of Liberal Cultural Nationalism,' Political Studies XLVI: 963-77. Stepan, Alfred. 1999. 'Federalism and Democracy: Beyond the U.S. Model,' Journal of Democracy 10 (4): 19-34 Taylor, Charles. 1994. Multiculturalism: Examining the Politics of Recognition. Edited by Amy Gutmann. Princeton: Princeton, NJ: University Press. — 1991. 'Shared and Divergent Values.' in, Ronald Watts and Douglas M. Brown eds, Options for a New Canada. Toronto: University of Toronto Press. Wamala, Augustine. 1994. 'Federalism in Africa: Lessons for South Africa.' in Bertus de Villiers, ed., Evaluating Federal Systems. Cape Town: Juta. Weinstock, Daniel. 2001. 'Towards a Normative Theory of Federalism/ International Social Science Journal 53 (167): 75-83. Whyte, John D. 1983. 'A Constitutional Perspective on Federal-Provincial Sharing of Revenues from Natural Resources.' in Charles E. McLure Jr. and Peter Mieszkowski, eds, Fiscal Federalism and the Taxation of Natural Resources. Lexington, MA: Lexington Books (D.C. Heath and Company).

18

BRUCE BERMAN 3 D I C K S O N EYOH & WILL KYMLICKA

Conclusion African Ethnic Politics & the Paradoxes of Democratic Development

I

T is a necessary truism that, in understanding the development of democratic polities, issues of theory, both normative and empirical, and of institutional design cannot be dealt with in a historical and political vacuum. For Africa this involves recognizing the specificity of the context of the development of ethnic communities and identities, particularly over the past century. The fundamental characteristics of this development were outlined in the introduction to this volume and investigated in the empirically focused chapters that followed. These characteristics, and the idiosyncratic forms they assume in particular states, shape African responses to efforts at political reform and institutional engineering in ways that may produce results strikingly different from those experienced in other polities where such institutions and practices were first successfully introduced. In concluding this volume, we shall briefly reiterate the characteristics of the African ethnic context and note their relationship to the various forms and levels of democratic institutional development. • First, contemporary African ethnicities are modern, not primordial survivals of some primitive tribal past. As the chapters of this volume have shown, they are relatively recent and dynamic responses to the political, economic and cultural forces of Western modernity as introduced to Africa during and after the epoch of European colonialism. Ethnicity is the contested outcome of intense political processes through which the boundaries, political ethics and moral economies of African communities have been continuously fought over and redefined in response to powerful and often disruptive forces of social change. These political processes have been expressed in an idiom of reinterpreted meanings of custom and tradition, including the syncretic incorporation of diverse elements of other African and European cultures. • Second, contemporary African ethnicities are intimately linked to the processes of colonial and post-colonial state formation and the development of capitalist market economies. They are grounded in, and express, in particular, the inequalities of, economic development and access to state resources both within and between ethnic communities. • Third, the distinction between the internal and external dimensions of ethnicity is critical for understanding the relationship between ethnic communities, capitalism and the state. Ethnicity is never just a matter of culture and tradition, but is also about competition for wealth and power. The conflicts within ethnic groups over social obligations, authority and the boundaries of community, and the conflicts between ethnic groups over access to markets and control over the state, are fundamentally about the control of the sources of material prosperity and political power. However, the nature of the competition is quite different in the two cases. 317

318

Bruce Berman, Dickson Eyoh & Will Kymlicka

• Fourth, the internal contestations are also moral conflicts over fundamental issues of social responsibility, solidarity, and the collective moral economy of ethnic communities. The term 'moral ethnicity' denotes the web of moral obligations emerging from these conflicts that tie together members of an ethnic group and provide them with some degree of protection and solidarity. The hierarchical and conservative values and the clientelistic relationships expressed in the internal processes of ethnic development, moreover, may conflict with the liberal individualist values and moral economy of the market and liberal democracy. • Fifth, the external confrontations between ethnic communities over access to and control over state institutions take place in an amoral free-for-all pervaded by ethnically based patron-client networks. In the clashes of 'political tribalism' the formal rules of the political process mean little, and control over parts of the state apparatus mean a great deal. This process both expresses and reinforces the wariness, distrust and opportunistic materialism that characterize the view of the state in African societies. • Sixth, ethnic communities in Africa shape and are also shaped by other bases of social differentiation and conflict. Internally, conflicts of both gender and generation are recurrent foci of clashing interpretations of custom and tradition. The conflicts between developing classes over issues of moral economy and political authority, in particular, are a crucial component of the politics of moral ethnicity. Externally, the clashes of political tribalism are also usually intra-class confrontations between competing ethnic factions of national political elites struggling for position within the state, and, at the same time, inhibit the mobilization of other classes across ethnic lines. • Seventh, and finally, these factors must also be placed within the particular national context of the widely varying total number and size of ethnic communities in African states. This ranges from unipolarity, where a single ethnic group comprises the majority of the total population, to deeply fragmented multipolarity, where many small ethnic groups, none of which constitutes a large proportion of the total population, co-exist. The relative size and number of ethnic communities shapes the dynamics of inter-ethnic relations and conflict.l The impact of all these factors has been heightened by the contemporary political and economic crises of sub-Saharan Africa. The combination of economic decline and state failure, exacerbated by the disruptive effects of neo-liberal reforms, has led to increasing conflict, insecurity and distrust which has, in turn, increased individual and collective reliance on clientelistic networks and the solidarities of ethnic communities. This, then, is the political, social and cultural context that has shaped and will continue to shape the response to political reforms and institutional engineering that are intended to produce stable democratic nation-states in African countries, and raises crucial questions regarding the objectives and effects of such innovations. In the dynamic circumstance of African ethnicity, where new ethnic communities are in the process of formation and older ones are disappearing, 2 any institutional intervention has potential 1 Yusuf Bangura, Ethnic Structure and Governance: Reforming Africa in the 21st Century. Geneva: United National Research Institute on Social Development, Issues Paper, 2000, pp. 16-32. 2 As in Kenya, where the Moi regime has tried, as yet unsuccessfully, to promote the emergence of another ethnic coalition, the Kamatusa (Kalenjin, Maasai, Samburu and Turkana) that will be larger and more broadly based than Moi's own Kalenjin group, itself a relatively recently created coalition of several smaller groups that is also wracked with internal dissension, including the threatened

Ethnic Politics & the Paradoxes of Democratic Development

319

implications for the continued growth or decline of various ethnic communities as well as the distribution of wealth and power within and between them. Where and how, then, should political reforms express or be blind to ethnic differences? Paris Yeros has noted that theories of ethnicity in Africa contain often hidden assumptions about the legitimacy and viability of ethnic communities and identities in politics, and yet constitute normative interventions into the political process that need to be acknowledged. 3 Those which adopt a view of ethnic communities as 'primordial' survivals of some primitive past and ethnic politics as essentially reactionary anti-modernism, will tend to look for institutional reforms that discourage the expression of ethnic differences and interests, and look toward their disappearance and replacement with the politics of a secular national modernity. The 'instrumentalist' approaches which overlap to some degree the constructivist analyses broadly represented here, tend to see ethnic politics as elitist manipulations of a form of false consciousness or ruling class hegemony that is similarly lacking in legitimacy or valid claim to a place in a democratic future in African societies. The dominant perspective that informs this volume, in contrast, by more openly accepting the modernity and legitimacy of ethnic identities and communities, as well as the important moral and political issues they express with all their positive and negative consequences, involves recognizing them as necessary and continuing features of African socio-cultural, economic and political development. Doing so, however, involves as well recognition that, within the complexities of African ethnicity outlined above, institutional reforms dealing with issues of group and individual rights, citizenship, political access and participation, and the probity and competence of government will paradoxically require both expressing and ignoring ethnic differences. The dilemma is: how can both be achieved in the same institutional context without mutual contradiction? We can look briefly at how this dilemma affects the various levels of institutional reform discussed in the earlier chapters. Solutions to the problem of how to eliminate the ethnic bias of centralized states have focused on the alternatives of consociational sharing of power and office among ethnic groups or federal decentralization of a significant degree of autonomy to formal sub-state units formed around particular spatially concentrated minority ethnic communities. In the African context, however, these constitutional structures accord recognition to some individual and ethnic rights and interests, but leave others unresolved. First, federal units do not resolve issues of the internal conflict of sub-groups and class within larger ethnic communities, and may ensure local domination by elites and the larger sub-groups. Where there are very large numbers of small ethnic communities, functional federal units cannot be created for all of them without falling into a political reductio ad absurdum. Second, there is the problem of ethnic differences and minority groups within the federal units, an issue likely to grow with internal migration and the increasing ethnic diversity of regions and urban centers as a consequence of economic development. The numerical predominance of particular ethnic groups in 'their' federal unit is likely to decline and exacerbate the clash of the core rights of universal citizenship within the nation with the insti(2 contd) withdrawal of the Marakwet. Meanwhile, as many as 16 small ethnic communities and their languages are threatened with disappearance as they are culturally and linguistically assimilated into larger, culturally related groups. ('Kenya May Lose 16 Tribes,' Daily Nation, 6 March 2002.) 3 Paris Yeros, 'Towards a Normative Theory of Ethnicity: Reflections on the Politics of Constructivism', in Paris Yeros, ed., Ethnicity and Nationalism in Africa: Constructivist Reflections and Contemporary Politics. Basingstoke: Macmillan, 1999, pp. 101-27.

320

Bruce Berman, Dickson Eyoh & Will Kymlicka

tutional ethnic bias in favor of 'natives' of local communities or federal units. The problem of rights becomes unavoidably multi-layered. Thirds the consociational reservation of powers and offices not only does not resolve the problem of patronage networks, but also formally allocates the resources of patronage to particular ethnic elites and inevitably shapes the balance of power in the internal politics of moral ethnicity. The design of electoral systems is another primary instrument of institutional reform. 4 The various alternatives from first-past-the-post and majoritarian systems to strict proportional representation have been promoted because of their supposed effects on either the expression or the muting of group differences by assuring the access of ethnic minorities to elective office or forcing the formation of trans-ethnic alliances in multi-party coalitions or 'national' coalition parties. All of these electoral systems are premised on the existence of various types of multi-party systems as the fundamental mediating institutions between citizens and the state. In the African context, group expressive electoral systems such as PR can foster extreme party fragmentation, splitting larger ethnic communities or organizing numerous small ones. Coalition formation in such circumstances can be highly unstable, as parties jostle for the best deal that formation of a government by a winning coalition can give them. Where losing groups are entirely isolated from the benefits of state office, the politics of political tribalism can become intense and destabilizing, spilling out of the arena of electoral institutions at either central or federal levels. Ethnically expressive electoral and party systems also do not mute but actually intensify patronage politics, as patronage resources are focused on securing electoral victory and enhancing the usual advantages of incumbency. Furthermore, in light of the weak ideological differences between ethnically based parties, other than securing the group's 'just' share of the national cake, party and electoral systems can simply formalize the 'politics of the belly' and confirm public understanding of the political process as a struggle solely for the material division of the spoils. Electoral party systems also constrain the organization, political expression and institutional access of groups expressing other bases of social cleavage, ideology and interest, particularly of class. In such circumstances, the construction through the political process of a broader public interest or project of national development is exceptionally difficult. Neither constitutional nor electoral systems can effectively deal with the most mundane and yet most important level of the citizens' relations with the state, namely the tense interaction between them and the bureaucrats of the state apparatus. The typical mixture of hope and fear, opportunity and threat, that characterizes the relationship between the state and ordinary people in Africa, their knowledge that it can either help them to 'eat' or allow them to be 'eaten', to use the common metaphors, blocks development of any significant trust in political institutions and public officials as disinterested arbiters of conflict or competent and equitable implementers of public policy. The issue of reform of the state administrative apparatus is much more than a matter of anti-corruption measures, which will have little effect where both officials and citizens remain tied to patronage politics. Civil service reform involves the often contra4 For useful discussions of the possibility of 'electoral engineering in divided societies', see Bangura, Ethnic Structure; Benjamin Reilly, Democracy in Divided Societies: Electoral Engineering for Conflict Management. Cambridge: Cambridge University Press, 2001; and Donald Horowitz, A Democratic South Africa: Constitutional Engineering in a Divided Society. Berkeley and Los Angeles: University of California Press, 1991.

Ethnic Politics & the Paradoxes of Democratic Development

321

dictory objectives of generating a bureaucracy that is both representative and competent. Ethnic allocation of offices, through quotas or assignment of particular positions, simply formalizes the prevailing practices of patronage in African politics, unless combined with politically isolated agencies of civil service recruitment that focus on universalistic criteria of competence and probity. But each principle of recruitment threatens to undermine the others. At the level of policy implementation and service delivery, proportional allocations of office and programs to ethnic communities through such instruments as regional development or affirmative action policies treat individuals differentially as members of particular communities, as clients of the state, in contrast with their shared rights with members of other communities to equal access and treatment as citizens. Patron-client politics is hardly unique to Africa, and may, indeed, be the most globally widespread form of informal authority relations and political connection. Patronage politics was and is significant in Western liberal democracies, as any student of 'pork' in American politics knows well. In Western states, however, there also developed a level of hegemonic civic politics in which the electoral institutions of liberal democracy achieved widespread legitimacy as disinterested arenas for the peaceful resolution of secular dispute, and the administrative apparatus of the state was similarly accepted as providing competent and honest implementation of public policy. This was imbricated with institutions of political mobilization - notably parties, unions and diverse pressure groups - that pursue through the political institutions the representation of variously defined groups and communities of collective interest and principle, whether class, religion, region, gender or ethnicity. The institutions and procedures of the liberal democratic state take on not simply legitimacy but also are publicly valued in themselves as the functional embodiment of a civic national community. The degree to which this has occurred varies from nation to nation and over time within particular nations as shifts in levels of public trust and institutional efficacy shape the ongoing character of liberal democracies as works in progress. It has been a process marked in all liberal democracies by continual and often intense conflict over the political institutions themselves and the ability in particular nations to amend and revise them peacefully in light of shifting constellations of political organization and interest. The central dilemma in Africa is how to reform political institutions in a setting in which the introduced institutions of the modern state have been the focus not of civic loyalty but of the materialistic opportunism of the largely ethnically defined patronage networks. As the Ghanaian professor of public administration, Samuel Woode, has noted in a comment on Ghana that applies throughout sub-Saharan Africa, the lack of conceptions of public service or conflict of interest amongst the political elite and bureaucracy, and their tendency to ignore institutional procedures and treat public property as a 'free good' to be appropriated wherever possible, destroy public trust, contribute to the low moral tone of the civil service, and undermine the development of national cohesion.5 Democratic development thus faces the challenge of generating public trust in reformed political institutions, whether expressive of group or individual rights and interests, and providing concrete reasons for letting go of clientelistic networks in favor of open civic politics. Such an achievement, however, is inex5 Samuel Woode, Values, Standards and Practices in Ghanaian Organizational Life. Accra: Asempa Publishers, 1997, pp. 32-3, 54-60.

322

Bruce Berman, Dickson Eyoh & Will Kymlicka

tricably tied in the context of contemporary Africa with the success of development efforts, whether state- or market-led, in producing a significant improvement in the material prosperity and security of the populace. Democratic reform cannot survive repeated economic failures and social decay. And this means recognition of the contradiction between the pressures by the international financial institutions and major bilateral aid donors for 'democratization 5 and their continued insistence on structural adjustment and neo-liberal reforms which consistently undermine African economies and lead to the deteriorating circumstances of ordinary people. Two further issues remain. The first is the relationship between democratic reforms and the continuing dynamism of ethnic development. Political reforms centered on the recognition of existing ethnic communities shape the process of ethnic development in ways that not only can permanently fix ethnic boundaries and institutional relations and empower particular ethnic elites, but also, by failing to recognize both the internal cleavages of the communities and the potentially unanticipated impact of social change on the formation of new communities and identities, lay the basis for future conflicts. Many commentators talk about such 'freezing' of identities as if it were a blanket objection to state policies of recognition. We do not argue, however, that this necessarily constitutes a bad thing. To do so would imply that there is some 'natural' or 'authentic' process of 'ethnic development' into which the state intervenes. Instead, the approach adopted here emphasizes that ethnic development was always a result of political interventions, most notably by the colonial state over the past century, and that there is no 'natural' process of such development and no 'authentic' pre-political identities to be either recognized or ignored. The fact that recognizing identities and communities is an 'intervention' by the state is not in itself a problem or an objection; there is, in fact, no way for the state to avoid this sort of involvement. The sole question is which sort of state intervention is appropriate to recognize what kinds of identities. From the liberal-democratic point of view, the answer, presumably, is that we want state policies of recognition to be consistent with liberal-democratic values of individual freedom, social justice and political democracy. We want the state to recognize identities in such a way that will enhance individual freedom, distributive justice and democracy. This can be quite consistent with institutionalizing identities in a more or less permanent way, as indeed many Western countries do (for axample, in Quebec, Catalonia, Flanders and with regard to indigenous peoples). The result of state interventions and institutional reforms may well be to 'permanently fix' certain ethnic identities, but this is acceptable in the process of democratization so long as the institutionalizing of those identities does not lead to restrictions on individual rights, or the eroding of redistributive policies, or the inability to sustain democratic procedures. Second, should ethnic expressive reforms take precedence over democratic institutions that permit the expression and resolution of conflicts based on different grounds of social cleavage, particularly class? If capitalist development succeeds in transforming African states from societies of peasants into societies predominantly of waged workers, the salience of class-based confrontations in the political arena can only increase and intersect with ethnically based institutions and practices that tend to produce elite domination within particular communities. In this context, the crucial issue is also the protection of the basic rights of organization and activity of parties, unions and interest associations expressing non-ethnic collective identities and solidarities, whether based on

Ethnic Politics & the Paradoxes of Democratic Development

323

internal mobilization within ethnically defined sub-state units or on trans-ethnic mobilization at the national level. Fundamental democratic rights and procedures need to be ensured across the political spectrum of interests of a diverse civil society. African states thus present a series of distinctive challenges to the achievement of multicultural liberal democracy that defy any single, formulaic solution particularly to the development of a civic national allegiance and consciousness, above ethnic identity and community. The latter can grow only out of the shared experiences of political participation and conflict resolution within the reformed institutional means of co-existence. In addition, the long historical sequence of economic and political development in Western liberal democracies simply cannot be reproduced in Africa. Instead, we must recognize the necessary simultaneity of democratization, economic growth and increasing resources for human development in African societies more and more acutely aware of the yawning disparity between their poverty and the wealth of the West. This would require, as Richard Sandbrook has argued, a 'social democratic' reform of the forces of globalization that accords African states the political space and resources for managing their economies and distributing more equitably the results of growth. 6 Such space and resources would make possible the formulation and implementation of distinctive African solutions out of the syncretic integrations of indigenous and foreign cultural and institutional forms worked out within the particular circumstances of each national arena. It is a process that demands the full and active involvement of African scholars and intellectuals in a dialogue with both political leaders and citizens of their own societies and with the global community of Africanist scholars, development activists and policy-makers in Western states and international institutions. The present volume is hopefully a contribution to that essential dialogue. 6 Richard Sandbrook, Closing the Circle: Democratization and Development in Africa. Toronto: Between the Lines; London and New York: Zed Books, 2000, pp. 142-8.

Index

Abacha, General 148, 161,163, 260, 263, 265, 266, 268 Abdullah, Ali 271 Abdurahman, Abdullah 116-17 Abernathy, David B. 29 Abiola, M.K.0.148,149, 160-2 passim, 165, 263,264 aboriginals 57, 58n2, 306 Abubakar, Abdusalami 163 Abubakar, Ahmadu 259, 272 Abuya,Abuya 170 accountability 4,12, 43, 75, 80, 81, 88, 90, 94, 95, 99,169, 258, 278, 280, 294, 295, 298 Acton, Lord 167-8 Achu,Achidi 110 Adah, Moses 271 Adala, John 205 Adamolekun, Ladipo 19 Adams, Ganiyu 163 adjustment, structural 21, 51, 322 administration 5,105, 212, 292-3, 324 Adungosi, Titus 171 Afghanistan 74, 75 Agbagha, Mma 272 age-grades 153 Agina, Israel Otieno 172 AgranofT, Robert 312 agriculture 7, 249 Ahidjo,Ahmadou 99-101 aid 21, 47, 169 Aka-Bashorun 266 Ake, Claude 167, 274 Akilimali, mwami 253 alliances, 9, 10,100,104-5, 117,150,154-6 passim, 160, 164, 165, 169,177, 188, 200, 205,208-13,262 Anderson, B. 131 Anderson, L-G. 143

Angola 20, 66,196,307,312 anthropology 6, 24-5, 27-9 passim, 32, 46, 75, 88, 241 Anyona, George 170 appointments 100,106, 212-13, 260-5 passim, 271, 308-10, 314, 321 Arabs 28, 30, 34 Argentina 302 armed forces 14, 223, 260, 262, 263 see also military regimes Ashanti 28, 77, 79, 81 Asia/Asians 70, 74, 186,190, 197, 210 assimilation 46, 59-60,101, 115-16, 131, 137,144,222,308 associations, civic/professional/voluntary 1,33,51,139,163 Atieno, Alfred 171 Atieno Odhiambo, E.S. 6-8,10-12 passim, 100,167-82 Atiku,Vice-President 274 Australia 59, 302 Austria 20, 61, 66, 302 authoritarianism 1, 4, 5, 8, 11, 15, 45, 47, 48, 52, 56, 64,100, 169-72, 216, 257, 259, 263-6 see also military regimes; anti102-7 passim autochthony 11,13, 99, 103-5 passim, 109, 111,152,219,226-35,241 autonomy 58, 67, 68, 99,105, 110, 161-5 passim, 173,184,195, 196, 279, 285, 303,311,312 regional 57, 58, 66, 67,100, 220, 263, 266 Awiti,Adhu 172 Awolowo, Obafemi 149,150,158-60, 162, 164,264 Awori,W.W.W.205,207 Azikiwe, Nnamdi 150, 159,160 325

326 Baartmann, Saartjie 123 Babangida, Ibrahim 148, 161, 259, 260, 265, 266,307 Balala, Sheikh 169 Bangura, Yusuf 97 Barreto, Honore Pereira 235 Barry, Brian 310 Basques 17, 57, 58 Baum, R.M. 234 Bayart,J-E 47, 75-6,101,113 Beauttah, James 201, 202, 206 Beauttah, Oscar 202 Belgium 19, 20, 58, 66, 241, 304 Bello,Ahmadu 159,274 Bello,Bouba 103 belonging, communal 11,19, 104—6, 111, 242-9 Berman, Bruce 1-21, 38-53, 98, 139, 204, 205, 241, 308, 309, 313, 317-23 Biya,Paul 10,101-3,110 Bloch, Marc 26-8 passim Bohannan, Laura and Paul 29 borders 13, 14,19, 240 Botswana 13, 70, 129-47, 314 Balope Commission 142,144 Constitution 134-5,142-4 Kamanakao 138, 143-4 Kgalagadi 138, 141 parties 140-2; BCP 141; BDP 133, 140; BNF 140-1 Sarwa 136, 138 SPIL 138,139 Tswana/Tswanification 130-2, 136, 140, 142,144 Yei 143-4 boundaries, communal 4, 12, 19, 98—9, 101, 110,135,149,151,152,156,164, 280, 283, 284, 304, 305, 317 Bourdieu, P. 136 Bradford, Helen 187-8 Brazil 302 Bretons 17, 57 Briggs, Llewelyn 209 Britain 19, 25, 28, 31, 39, 43-4, 55, 80-1, 84-9 passim, 99,115, 151,157-8, 162, 169,173,176,186,187, 202-3, 230, 241, 268, 304, 305, 307 Brown, Jeremy Murray 201, 203 Brubaker, Rogers 178-80 passim Brundell, Sir Michael 209 Buganda 310-11 Buhari, General 270 Buhari, Speaker 262

Index Buleli, Leonard N'Sanda 240-56 bureaucracy 5, 8, 38, 43-7, 50-3 passim, 100,102,130, 265-6, 271, 320-1 Burundi 20, 66, 242, 245, 314 business 7, 14, 41, 50,102,108,193, 197, 244, 249, 266 Buthelezi, Gasha Mangosuthu 183-6,190-9 passim, 282 Cairns, Alan 306 Caliguire, D. 122 Cameroon 10,11, 67n8, 70, 99-112, 314 Anglophones 101-10 passim Bamileke 102,103,109 Beti 10, 101-3 passim, 106 Collectif Changer Le Cameroon 105 Constitution 105,110 Duala 109 Francophones 106,107 Hausa-Fulbe 100—3 passim parties; CDU 103; CNU 100; CPDM 101-3 passim, 105, 107,110; M D R 103; NUDP 103; SDF 103.105,107, 109,110 SAWA/SWCC/SWELA 107,109 Canada 19, 56-9 passim, 169, 303-5, 310-13 passim Canovan, Margaret 55 capitalism 3, 6, 7, 39, 41-5 passim, 47, 50, 74,193,197,317,322 Carew, George 172 Casamance 218-39 Joola 222, 224-38 passim Kolda 218, 224-5 Manding 224, 225, 228-32 passim, 235 MFDC 218-20, 223, 224, 232, 233, 235, 236, 239 Sedhiou 224, 225 Velingara 224, 225 Ziguinchor 218, 219, 224, 225 cattle farmers 242, 249-50 Catalonia/Catalans 57, 58, 67, 304, 322 Cavendish Bentinck, Sir Ferdinald 209 centralization 9,100,109, 152,153, 258-9, 275,281-2,306,311-12 Chabal, Patrick 75 Charlton, R. 130 Charpy, Jacques 224, 227-8, 230, 231 chauvinism 56, 76, 78, 94,105,106 Chege, Michael 168,170, 179 chiefs 5-7 passim, 13, 45, 78,123,134-5, 137,142,144,173-4,184-8 passim, 190, 195,198, 201-4 passim, 211, 243,253,254

Index Chiroma, Liman 274-5 Chotari, Kariuki 170 Christianity 5,149,160, 188, 234, 238, 261, 268—70 passim Chukwumerije, Uche 263 churches 18, 51, 93-4, 169,172, 243 circumcision 179-80; female 179, 202, 203 citizenship 4, 9, 12-13,17, 36, 43, 55, 60-1, 63, 91, 96-113,115,117,129-47,152-3, 169,173,177,179, 301-4 passim, 313-14,319 city-states 152, 164 civil service 29, 43-4, 47, 212, 260, 278, 283, 288, 292, 320-1 civil society 1,18-19, 40, 42, 43, 51, 94, 97, 179,240,278 clans 104,169,173-4,205 class 3, 6, 7, 43, 48, 49, 76, 115,121,131, 158,193-4, 201, 205, 258, 259 cleansing, ethnic 82, 92,169, 244, 271, 280 clientelism 1-3 passim, 5, 9, 45, 47, 51, 53, 66, 76, 78-80 passim, 82, 218, 237, 259, 308,309,318,321 closure, social 178-9 Clough,M.S. 177 Cohen, D.W. 169,173,174 colonialism 3, 5-9 passim, 22, 24-5, 27, 29-32, 45-7 passim, 76, 78, 80, 85, 90, 115,116,139,157-8,173-4,186-8, 200-11, 221-3,227-8,238, 241,258, 317 post- 1, 3, 8,10-12 passim, 18, 22, 25, 29-33 passim, 39, 45, 47-8, 66, 80, 97, 98, 100, 107-8, 139, 221-3 passim, 238, 240-1, 243, 258, 281, 317 pre- 3, 4, 46, 48, 77-8,150,151, 228 ComarofF, J.L. 138,140; and J. 135, 137,179 competition 1, 6, 7, 9-12 passim, 45, 46, 50, 78, 96-100 passim, 103,104, 109, 111, 169,317 conditionality 169 conflict 6, 7, 43, 242-3, 318, 321 ethnic 9-15, 66, 69,129,158,172, 222, 227, 257-75 passim, 280, 299-300, 318-22 passim Congo 240-56 AFDL 253, 254 Banyamulenge 242-6 passim, 250 Bembe 251, 253-4 Constitution 245 D R C 254 Free State 31, 245 Hutus241,250 JMPR 247, 248

327 Kivu 249-54 Kusu-Tetela 252 Lega246n2, 250-1, 254 Luba241,247 Lubumbashi University 247-9, 255 Lulua 241 mai-mai movement 250—5 passim Milima group 250 parties; MNC/L 251, 252; MSD 251 Shi 251-2 Tutsi 241, 244, 249, 250 Unerga 250-1 Zimba 251, 253-4 consociationalism 20-1, 66, 69, 308, 309, 312,319,320 contracts 35, 40-2 passim, 50, 83 corruption 39, 44, 51, 52, 97,102,103,106, 129, 247, 260, 305, 308-10, 320 Cosgrove, D. 222 Crawhall, N. 123 Cromwell, Oliver 25 culture 3-6 passim, 10,11, 13,16-18 passim, 40, 43, 46, 54-69, 64-8, 89-91 passim, 94,109,115,116,122,124,133, 138-9,156,177-80 passim, 194-5, 198-9, 200, 205, 284, 298, 302, 305, 317 multi- 59-60,114,126,139, 303-6, 308, 310 custom 4-6 passim, 59,188,198 see also law Dahomey 28 Daigne, Souleymane B. 237 Danbazau, Lawal 271 Darby, Philip 177 Datta, K. 137 Davidson, Basil 28 debt 169 decentralization 12, 52, 97—9, 104—7 passim, 110, 111, 220, 244, 246, 268, 278-81 passim, 284, 304-7 passim, 313, 319 decolonization 96,158, 209-11, 258 De Klerk, F.W. 121 democratization 9-13, 23-37, 50-3, 56, 167-275 passim, 280-1, 297-8, 322, 323 denationalization 62 Deschamps, Hubert 228-9 detribalization 88,139,198 development 1, 6, 7, 21, 56, 73-167, 317-23 passim

Diatta, Alin Sitoe 229, 238 differentiation 4, 6-7, 46, 96-112, 306, 307, 313-14,318

328 Dike. K. Onwuka 28 Dinuzulu, Cetshwayo ka 187 Dinuzulu, Solomon ka 189 Diouf,Abdou 218-20 passim Diouf, Mamadou 4, 5, 7,11,17, 218-39 diplomacy 154,155, 164, 266 donors 169,322 Dube, Rev. John 197 Duchacek, Ivo 304, 305, 307 Dudley, B.J. 159 Dunia, 'General' 253 Dunn, John 44 Durham, D. 133,136, 138 East Africa Association 201 Ebousi-Boulaga. E 105 economy 1,15,102, 173, 216, 237, 243, 307,310-12,317 education 5, 7, 52-3, 55-7 passim, 83-4, 88, 108,142,149,216,241-2,278 Ejobowahjohn Boye 11,12, 20, 97, 301-16 Ekeh, Peter 5, 8, 11,12, 22-37, 47,113, 139,178,243 Ekwueme 262, 263 Elazar, Daniel 280, 302 El-Beshir, Gen. Hassan 306 elders 7, 78, 84, 272 elections 1, 2, 9, 20, 43, 51, 52, 103, 105, 107-10 passim, 113,121,126,129,141, 148,159-60,169,183-4,192,194, 208, 210, 213, 218, 219, 241, 251, 259-60, 262, 264, 288, 320 elites 2-4, 9-12, 48-52 passim, 64, 75-7 passim, 80, 90, 92-3, 96, 99-110 passim, 132,150,156-8,174, 200, 210, 211, 215, 216, 218, 241, 259, 263, 272, 274, 307, 308, 318-22 passim employment 7,106,118, 132 equalization/equality 280, 309 payments 312, 313 Eriksen, Thomas Hyland 11,18, 65, 96, 98 Eritrea 67 Estonia 14 Ethiopia 17,19, 67, 79, 280, 304, 314 Ethiopianism 157 ethnology 226-35, 241 Europe 6, 7,15, 16, 22-3, 25-8, 34, 40, 43, 52, 69, 70, 74, 80 Eastern 14-15, 96 Evans-Pritchard, E.E. 25, 29 Eyoh, Dickson 1-21, 49, 96-112,139, 314, 317-23

Index FageJ.D. 30 Falae,01u 149,161 Falola,Toyin 4, 8-11 passim, 100,148-65 Fanon, Frantz 88 Fasheun, Dr 162-3 federalism 13,19-20, 58-9, 67-9 passim, 107,165,197, 244, 258-9, 266, 267, 277-81, 284, 297, 299, 302, 304-8 passim feudalism 26-8, 34, 40, 88-92 passim Finland 19,169 Finnegan,W. 120 fiscal affairs 286-7, 293-4 Flanders/Flemish 57, 58, 67, 322 Fluehr-Loban, Carolyn 306 Forde, Daryll 28 Fortes, M. 25.29 Fox, Richard 27-8 France 17, 31, 55, 57, 81, 99,106, 107, 167, 169, 228, 230, 234-6 passim, 239 Fraser, Nancy 137 freedom, of information/speech etc. 48, 169,302 Fried, Morton 27 Friedman, Steven 298 Fulani 22, 28 see also Hausa Fulbe 100-2 passim Furedi, Frank 175 Gabasiane, O. 140 Gachukia, Daniel 180 Gambia 221, 226, 229, 234 Gapangwa, Jerome 250 Gathanju 207 Geertz, C. 139 gender issues 3, 6, 7, 13, 78, 86,193, 198-9, 269,318 generation issues 3, 6, 7, 79,193,194, 318 Germany 31, 39, 55, 60, 61, 99,169, 283, 287, 290, 302 Ghai,Yash 209, 279 Ghana 30, 50, 314, 321 Gichuru, James 177, 204, 209, 210 Giddens, Anthony 41 Gikuyu see Kenya, Kikuyu Gilligan, Chris 198 GladneyD. 130 Goldin,I. 117 Goldsworthy, David 213 Good, K. 129 Gordon, David F. 169 governance 97, 169, 260, 266, 277-300 local 298 Greece 63

Index Griffiths, James 206 Grosh, Barbara 169 group proliferation 305-8 growth, economic 7, 132, 216, 278, 323 guest-workers 60-1, 69 Guinea-Bissau 221, 226, 229, 236 Gurr,Ted58,66,70 Gutto, Shadrack 170,171 Gwala, Harry 196 Hailbronner, Kay 282-3 Haile Selassie 17, 67 Hausa see Cameroon; Nigeria headmen, village 5-7 passim, 45,123,187, 195 health care 56, 278 Hempstone, Smith 169 Hendricks, Cheryl 4, 5,10,113-28, 314-15 Hendrickse, Rev. Alan 118 Herbst, J. 96 Hintze, Otto 26, 27 Hispanics 57, 306 history 150-3,185-7, 218-29, 245-6 HIV-AIDS 199 Hobbes, Thomas 23-6, 306 Homans, George 22 Hornsby, Charles 169,172,177,178, 215 Horowitz, Donald 280, 299, 307 Humphries, Richard 283 Hungarians 34 Huntington, Samuel P. 169 Hutterites 56 Hutus see Congo Hyden, Goran 176, 179 Ibrahim, Jibrin 18 Iceland 13 identity 2-9, 11-13,16-18, 38-40 passim, 46, 50, 51, 55, 65, 73, 96, 98,101,104, 113-31,137,149, 151,156-65,185-91 passim, 221, 226, 229-30, 235-8, 257, 301,304,308 ideology 2,10, 11, 49, 82, 87, 99, 101, 111, 114,116,154,159,164,178,194, 202, 210, 242, 243 IFIs97,102,169,322 Ige, Bola 273 Igbo see Nigeria Ikime, Obaro 28 Imanyara, Gitobu 179 IMF 51,169 immigrants 59-61, 63, 65,103, 314 India 19, 304

329 indigenous people 10, 64, 70, 99, 101, 104, 109, 111, 245, 314, 320 see also natives individuals 22-37; and kinship 31-6 Indonesia 312 institutions 2, 5—9 passim, 12—13, 18, 27, 35, 38-9, 41-4 passim, 97, 277-317 passim, 320, 321 intelligentsia 5, 6, 45, 151, 154,164 Inuit 70 investment 11, 21, 74,106 Ireland 25, 26 Northern 14,66 Isichei, Elizabeth 29 Islam 101,102,124, 160, 231, 234-8, 268, 270 Israel 14, 79 Italy 19, 55, 58 Jackson, R. 96 Janson,T. 143 Jewsiewicki, Bogumil 240—56 Johnson, Michael 309 Johnson, Samuel 157 Johnston, Sir Harry 167 Jones, G.I. 28 Joseph, Richard 308 Joyce, Patrick 168 Kabila, Laurent 253-5 passim Kabuya-Lumuna, S. 253 Kaggia, Bildad 170, 176, 205-7 passim, 213 Kaita, Lawal 272 Kangethe, Joseph 201 Kanogo, Tabitha 175 Karanja, Mbaraka 172 Karimi, Joseph 214 Kariuki, Jesse 91, 201, 206, 213, 214 Kaspin, Deborah 69nl4 Katithi, Joseph 206 Kaviraj, Sudipta 274 Kenya 8,10,11, 38, 51, 70, 75-95, 100, 167-82,200-17,314 Asians 174,210 Change the Constitution group 213, 214 GEMA 214 Hilton Young Commission 202 Kalenjin 10, 88,177, 212, 213; KPA 209, 212 Kikuyu 8,10, 79, 82, 84-9, 92-4,174-6, 200-15 passim, KCA 201-6 passim', Muigwithania 202 Law Society 93 Luhya 205, 208, 210, 212

330 Luo 10, 82, 84, 86, 94, 173-4, 177-80, 205, 208, 210, 212-14 passim, LUTATCO 174; Union 214 Maasai 70, 88 Mau Mau 79, 82-9 passim, 168,176, 206-8 passim, 211 parties 169,170, 209-15; APP 209-10; CAPU 209; IPK 169; KADU 79, 209-13 passim; KANU 79, 93, 177, 209, 209-15 passim; KAPP 209; KAU 82-6, 89, 176, 204-6 passim, 215; KIM 209; KNG 209; KPU 79, 88, 169,176, 178, 179, 212-14 passim Kenyatta, Jomo 8, 78, 82-92,173-9, 200-17 passim

Kenyatta, Ngina 180 Kershaw, Greet 174, 175 Khadiagala, Gilbert 169 Khamisi, Francis 204 Kiano,Julius209,211 Kibaki,Mwai 179,211,215 Kikuyu see Kenya Kimathi, Field-Marshal Sir Dedan 86-7, 90 kinship 4, 5,10-12 passim, 23, 25-36 passim, 40, 46, 49, 50, 98,101,104,109,186-7, 226,241,255 Kinyatti, Maina wa 171 Kirdi 70,101 Kitabi, Jesse 206 Kofele-Kale, N. 106 Kohl, Helmut 197 Koinange, wa Mbiu 175, 202-4 passim, 206, 207,211 Koinange, Mbiyu 177, 204, 206, 211 Kombo, Msanifu 170 Konate, Abdourahmane 224, 226, 227, 230 Kontagora, Hassan Sami 262 Korea 13 Kreuzer, Christine 282-3 Krieger,M. 102 Kubai, Fred 205—7 passim Kwassa, Shadrack Ojudo 180 KwaZulu-Natal 20,183-5,190-9, 282, 284, 285,291,299 Inkatha Freedom Party 183-4, 188,192, 194-9 passim, 282, 291, 299 Inkatha movement 184,187-92 passim Inkatha Youth Brigade 192 Zulus 183-99 passim, 282, 285; Cultural Movement 184,189 Kyle, Keith 202, 206, 207, 210 Kymlicka,Will 1-21, 54-71, 96, 97,130, 134,135, 303, 305-6, 313, 317-23

Index labor 4, 5, 7, 42, 46, 78, 85, 86, 88,108, 115, 173, 188-90 passim Lafargue, Jerome 169 Laitin, David 17 land 4, 5, 33, 46, 85, 87, 88, 92, 94, 111, 140,151,154,174-7,186,193, 201, 202, 206, 211-12, 214, 242, 243, 249, 252 Carter Commission 175 language 5,15-18 passim, 29, 54-9, 64-8 passim, 77, 79, 99,106,115,122, 129, 133, 136-9 passim, 142,149,156, 157, 179-80,192,236-7,284,303 Larbi, G. 97 Latin America 14-15, 56 Latvia 14 law 42,103, 268-71, 285-6, 304 see also Sharia

customary 186,198,245 Law, Robin 28 leadership 4, 7, 8, 10-12 passim, 86, 90,101, 105,149-50,195 Lebanon 304, 309 Lemarchand, Rene 66, 68 Lentz, Carola 172 Leo, Christopher 176 Leslie, Peter 304 Leteta, Ngongo 252-3 LeVine,V96 Levtzion, Nehemia 30 Lewis, G. 117 Leys, Colin 209, 211, 213 liberalism, cultural 303-5 political 301-3 liberalization 10, 11,15, 96-112 Liberia 20, 66, 75 Lijphart,Arend280,309 lineage societies 28-9,104, 109,153-5, 244 literacy 32, 76, 78, 79, 82, 85, 87 Little, Joe 123 Little, Kenneth 33n4 Lodge, Tom 126,289 Lone, Salim 170 Lonsdale, John 4, 5, 7, 8,10—12 passim, 73-95,100,113,139,174,178, 309 loyalty 13, 20, 40, 49, 50, 70, 80, 150, 200 Lugard, Sir F. D. 28 Lumumba, Muigai 170 Lumumba, Patrice 243 Luo see Kenya Lusaka Accords 245, 246 Lustick, Ian 14 Luthuli, Albert 189,190 Luwowo, Ngongo 252

Index Macharia, Rawson 206 Machel, Samora 8 Mafumadi, Fholisani Sydney 195 Magugu, Arthur 211 Mahihu,Eliud211 Maine, Henry Sumner 26, 27 majorities 19, 64-7,130-6 passim, 280, 305 Makarfi, Ahmed 272 Makinda, Samuel 169 Makonyango, Otieno 171 Malawi 69nl4 Malaysia 14,19 Mali 19, 30, 314 Malinowski, Bronislaw 7, 203 Malo, Otieno 171 Mamdani, Mahmood 12,113, 144,173, 186,198 Mandela, Nelson 192, 196, 197 Mannathoko, C.E. 139 Manning, Patrick 28 Manyara, Benson 170 Mare, Gerhard 186 marginalization 8, 21, 56, 106, 135, 136, 262-3, 272 Mark, Peter 229-31 passim, 233-4 Markakisjohn 18,64-5 markets 4, 41, 42, 45, 46, 48, 97, 154, 317 Marks, Shula 4, 5, 7,10,11,13,119,138, 183-99 marriage 78, 81,155,164,216 Marx, A. 120 Marx, Karl/Marxism 7, 23-5 passim, 42, 82, 170 masquerades 154 Mate, Bernard 208 Mathenge, Isaiah 211 Mathenge, Stanley 86 Mathu, Eliud 204 Matiba, Kenneth 93, 169,172,179 Mauritius 18,65,314 Mayall, James 168 Mazrui,Alia and Alamin 65n4, 68nl3,171 Mbeki, Thabo 196 Mbembe, A. 101,238,241 Mbithi, Philip 170 Mbotela,Tom 84, 205, 206 Mboya,Tom 174,177, 178, 208-10 passim, 212,213 Mbugua, Rosemary 170 McCaskie,Tom 77 McAuslan,J.P.W.B.209 media 6, 18,142,262 Melone, S. 105

331 memories 218-43 passim metics60-l,64,70 Mexico 60, 307 Michuki, John 211 Middletonjohn 32 Mieszkowski, Peter 312 migration 6,10,12, 19, 32, 33, 69-70, 78, 98,104,108-9,132,151-4,188,190, 194, 212, 216, 222, 226-35, 249, 314, 319 Milambo, 'General' Raphael 253-4 military regimes 48, 56, 153, 161, 165, 257-9, 262-4 passim Mill, John Stuart 167-8 mining 7,132n4, 243, 307 minorities 10, 14, 17-19, 55-9, 63-70 passim, 96, 98, 129-47 passim, 258, 259, 264, 266, 267, 272, 273, 297-80, 299, 303,305,313,319,320 missions/missionaries 5, 16, 156-7 Misztal, Barbara 44 Mkangi, George 171 Mkangi, Katama 172 mobilization, ethnic 192, 238, 243, 264 political 8, 10, 56, 62, 64, 77, 98,100, 117,200,214,215,264,321 Mobutu, Sese Seko 242-4 passim, 248 modernity 3-5 passim, 7,12, 39-46 passim, 51,96,130,137,242,317,319 modernization 7,149,159 Mogae, President 135 Moi, Daniel arap 10, 82, 87, 90-3, 169-72 passim, 177, 208, 209, 212-16 passim Molokomme,A. 140 Molomo,M.G. 135 Molutsi,P. 142 Moore III, Barrington 168 Mpho,M.K. 130 moral ethnicity 4—12 , 46—51, 76—88 passim, 91, 94, 95,113,126,178, 318, 320 Morrison, J.H. 29 Muela, Ngalamulume Nkongolo 247—8 Mufson, Steven 119 Mugabe, Robert 74 Mugo,Micere 170,171 Muhoho, Chief 175, 204, 211 Muigai, Githu 8,10,11,100, 200-17 Muliro, Masinde 208, 209 Mungai, Njoroge 177, 210 Murray, A. 137 Murray, Christina 277-300 Murumbi, Joseph 207 Musa, Balarabe 262, 270 Museveni, Yoweri 168

Index

332 Musonge, Peter 110 Mustapha, A. Raufu 4, 8-11 passim, 19-20, 100, 257-75, 305 Muthembe, Andrew 171 Mutunga, Willy 171 Mwachofi, Mwashengu wa 170 Mwamba, Thambwe 251 Mwangale, Elijah 171 myths 18,149,152-4, 157,164, 202 Nangurai, H.L. 204, 206 Nas,Wada261,262,272 Natal 186-90 passim, 196-7 see also KwaZulu-Natal nation-building 1-21, 53-71 passim, 75, 82, 96-102,114,115,121-6,130,156-7, 183, 200-1, 204-16 passim, 275 nation-states 38-9, 42, 45, 50, 54-6, 64, 73-4, 96, 97, 222, 223, 245-56 nationalism 6, 8, 11, 44, 47, 56—8 passim, 65, 76, 82, 86, 88, 96,150,157, 165, 168, 174,204,200,208,212,240 ethnic 62, 67-8, 85, 187-90, 192,195, 196,200-17 nationality 245-6, 249, 253 nationalization 243 natives 10,101,104, 111, 245, 314, 320 naturalization 246 Ndegwa, Stephen N. 173, 178, 179 Ndi, Fru 103 Ndjate Alphonse 253 Nengwekhulu, R. 135 neo-patrimonialism ln2, 2, 97, 99, 178, 201, 215,216 nepotism 39, 308 Netherlands 20, 66, 169 neutrality, ethno-cultural 54-5, 65, 66, 96, 132 New Zealand 58 Ngala, Leonard 209 Ngala, Ronald 208, 209, 213 Ng'ang'a, Mukaru 170-2 passim Ngei, Paul 176, 206, 207, 209 NGOs 249, 250, 252, 281 Ngugi waThiong'o 167,170,176 Ngugi, James 179 Nichols, George Heaton 188 Niger 70 Nigeria 9-11 passim, 20, 29, 33-5 passim, 50-1, 67, 68,100,108,148-65, 257-75, 280, 304, 305, 307, 309-14 passim

Abeokuta 153

Awori 153 Biafra Republic 273 Constitution 258, 266—70 passim, 273 Dumu 152 Egba 151—3 passim, 158,164; Egbe Omo Oduduwa 158; Egbe Omo Yoruba 161-2 Ekiti 151,152, 164 Hausa 67, 68, 150, 156, 158,162, 258, 259,261,263,266,269,272 Ibadan 153-5 passim, 163, 164 Igbo 29, 34,148, 150,158,160,158,160, 162, 258, 261-3 passim, 266, 272, 274; Ohaneze 263, 272 Ijaw 257; National Council 273;Youth Congress 163 Ijaye 152,153, 155 Ijebu 151,153 Ikale 152 Ile-Ife 151—3 passim, 160 Ilesa 152,163 Itsekiri 257 Iwo 154 Iyegba 152 Lagos 156,163,262,267 Middle Belt 160, 259, 264, 271-3 passim; Forum 272 Niger Delta 160, 259, 264-8 passim, 273, 311-12; UND 268 North 9,10, 160, 261-72 passim, 274-5; Arewa People's Conference 163, 265, Arewa Consultative Forum 267; Northern Elders Forum 267-8, 274; Jam'yyar Mutanen Arewa 158 Ogoni 67nl0; MOSOP 273 Oke-Odan 153 Ondo 152,153,164 Ooni 151 Owo 153 Oworo 152 Oyo 28,151-6 passim, 160,164 parties 150, 158, 259; AD 161,163, 165, 273; NADECO 161; PDP 161, 263 Patriots 267, 268, 274 Presidential Technical Committee 267 Sovereign National Conference 263, 264, 266-8, 273 Usmanniya Republic 273 Yoruba 68, 79,148-65, 257-67, 272-4 passim; Action Group 151,158-60 passim, 164;Afenifere 161,273; Alliance for Democracy 161, 163, 165,

Index 273; O'odua People's Congress 162—3, 257, 262, 273 Zamfara State 268-71 passim Njonjo, Charles 170, 171,177, 211, 213, 215 Njoya, Rev. Timothy 172 Njoroge, James 206 Njoroge, Jonathan 206 Nkongolo, Muela Ngumulume 247 Nnoli, Okwudiba 158 Nora, P. 210 Nordlinger, Eric 309 Nottingham, John 203, 205 Numeiri, General 306 Nwabuikwu 272 Nyamnjoh, F. 104 Nyati-Ramahobo, Lydia 143 Nyerere, Julius 168 Nyong'o, Peter Anyang' 168,170, 200 Nzau, James 208 oathing 85, 175-6,178, 206, 207, 213-14 Obasanjo, Olusegun 148,149, 159,160, 163, 165, 257, 259-67 passim, 270, 273, 310 obligation 4, 7, 8,11, 12, 35, 46, 76, 200, 317 Obote, Milton 311 Obotela Rashidi, N. 246 Ochieng, Philip 214 Ochieng',W.R. 178 Odeda, Walter 205, 207 Odinga, Oginga 84, 93,168-70,172-4, 177-9 passim, 205, 208-10 passim, 212, 213 Odinga, Raila 169-72 passim, 179, 213 Oduduwa 149,152-4,158-60 passim, 165, 266, 273 children of 154,158 Ofafa 206 Ogendo,Okoth 170 Ogot,B.A. 171 oil 9,102,106, 258, 264, 267, 268, 311, 312 Ojukwu, Col. 263 Okoh, Peter 65n4 Okondo, Peter 206 Okullo, Rev. Henry 169 Oldfield, S. 140 Oloo, Onyango 171 Ombaka, Ooko 170,171 Omosule, Monone 176 Oneko, Achieng 205-7 passim, 213 OnomoJ. 102

333 Orengo, James 170 organizations, ethnic 240, 249-54 students' 240, 246-9 Orvin, Stephen 169 O'Sullivan, Megan 273 'other'/'othering' 115, 116, 177-80 Otiende, Joseph 205, 206 Otieno, S.M. 94 Owino, Albert 204 Owor,Okola 175 Owusu, Maxwell 168 Oyemu,John 306 Oyovbaire, S. Egite 27 Oyugi, Edward 171 Pan-Africanism 203 parastatals 259, 262 Parsons, N. 130-2 passim participation 11,13, 99,104,113,178, 278, 280,281,298,308,319,323 parties, political 1, 9, 20, 66, 96,100, 320, 322 see also under individual countries

Pashukanis, Evgeny 42 patriarchy 13,189,193,198 patron-client networks 8, 39, 40, 45-51 passim,129,133, 201, 212, 308-10 passim, 315,318,320,321 patronage 1—3 passim, 5, 7, 8,10,12,15,19, 39, 45-52, 66, 76, 79-82 passim, 85, 86, 90-2,100-2,105,107,129,133, 201, 216, 308, 320, 321 Paye, Moussa 220 PeelJ.D.Y. 167 Peters, P 135 Phillips, Anne 133 Phillpotts, Bertha Surtees 28 Pieterse, Edgar 119-20 Piper, Laurence 192,195, 196 pluralism 12,13,15, 222, 224, 281, 284, 301—3 passim

ethnic 3, 52, 96—110 passim, 222 police 14,100,192,223,267 politics 1-21, 38-53, 73-6, 79-112, 120, 140-2,148-217, 243-6, 257-75, 288-92,317-23 ofbelly8,46,49,75,320 one-party 90,100, 170-2, 204-17 passim Ponty, William 236 Portugal 13, 55, 230, 235-6, 239 poverty 140,195 prebendalism 308 proletarianization 189, 193, 194 protection 8, 26, 27, 30-4 passim, 49

334 public, civic/primordial 47, 113, 178 Puerto Rico 57, 58

Index

Quebec/Quebecois 57, 58, 67, 306, 322

Russia 19 Rwanda 20, 34, 66, 74, 93, 96, 241, 242, 245,251,252,314 Ryder, A.EC. 28

Rabiu, Musa 272 race 77,113-16 passim, 118-22,124-6 passim, 168,242,299 RadclifFe-Brown, A.R. 24-5 Radebe,Mark 187 Ranger, Terence 178 Rapoo, Thabo 294 Rasool, Ebrahim 126 Rawls, John 273-4, 301-4, 309 Reagan, Ronald 197 rebellion 218, 226, 242, 243, 251-4 passim, 306 reciprocity 7, 8,11, 35-6, 40, 48, 49, 77, 79, 80 recognition 99,110,136-7, 303-8 passim, 314,315,319,322 reconciliation 13-21 reform 2, 5,11, 39, 43-4, 50-3, 84,107, 205, 260-2, 264, 265, 317-23 passim refugees 69,184 religion 1,13,16, 54, 56, 59, 68, 76, 78, 93-4,124, 234-5, 237, 257, 268-71 passim representation 10, 12, 20, 99, 104, 133, 136-7,178, 280, 308-10, 321 proportional 20, 320 Requejo, Ferran 303 reserves, African 186,205 residence 11,12,104, 245, 314 Resnick, Philip 302 resources 5, 7-12 passim, 104, 106, 111, 178, 264, 267-8, 278, 307-15 passim, 317 responsibilities 4, 5, 47, 318 Reynolds, A. 126 rights 4, 7, 35, 46, 47, 54-71, 99,105,110, 111, 119,131,133-7,169, 246, 284, 302-5, 310-15 passim, 319, 321 Bill of 280, 282, 284 Lawyers Committee for 33 riigi 86-9 passim riots 170,172,190, 262 risk 42, 43 Roche, Christian 227, 235, 238 Rogers, Peter 177 Rosberg, Carl 203, 205 Ross, Mervyn 122 Rowland, M. 104,105, 109 Rubia, Charles 172

Sabar-Friedman, Galia 169 Saint Moulin, Leon de 246, 249 Samatar,A. 140,144 Sami 70 San 70 Sandbrook, Richard 323 Sani, Governor Ahmed 268, 269 Sanusi, Lamido 172 Saul, John S. 168 Savage, D.C. 169 Scandinavia 28, 58 Scarritt, J. 98 Schapera, Isaac 130 Schilder,K. 101 Scotland 58, 304, 305 secession 20, 57, 67,122,142,148,150, 161-3,184-5,196-7, 221, 226, 235-9 passim, 271, 280, 305 security 5,11,12,20,22-37 Seglow, Jonathan 302—3 segregation 61, 62, 116,121 self-determination 142, 235, 280, 282, 284, 304, 310 self-government 56—8 passim, 63, 66, 68, 240, 278, 280, 299, 303, 305, 308, 310, 312 self-respect 302-3 Senegal 11,218-39 Senghor, Leopold Sedar 168, 223 Senghor, Abbe Augustin Diamacoune 219, 223-4, 226, 229, 231-6 passim Seroney 212 settlers 174-6 passim, 186, 209, 210 Shagari, Alhaji Shehu 307 Shaka 185,192 Sharia 10,148,160, 257, 264, 268-71, 306 Sierra Leone 156, 312 Sifuna, Lawrence 170 Sikatenda, 'General' 253 Silver,Alan40,42 Silver, Kalman 42 Simeon, Richard 10,12, 20, 97, 277-300 Singh, Makhan 205 Sitas,Ari 199 Sklar, Richard 167 slaves 61-2,114,115,123-4, 231 trade 22, 25, 27-32 passim, 230, 233-4 Smooha, Sammy 14

Index socialism, African 88, 168 sociology, political 22-3, 25, 29,178-9 Sokoto Caliphate 22, 268 Soleye, Onaolapo 264 solidarity 4, 7,17, 20, 39, 46, 77, 80,109, 158,243,247,312,318,322 Solway, Jacqueline S. 4, 5, 13,129-47 Somalia 34, 67n9, 75, 93 Songhai 30 'sons of the soil' 10, 19, 68nl2,104, 314 South Africa 10, 13, 50, 67n7,113-28, 183-99,277-300 apartheid 10, 190,194, 277, 281 APO 116 Bantustans 184,185,190, 277, 281, 284 Black Consciousness Movement 118, 191 Cape Coloureds 10,113-28, 315; CLPP 114,117,120 Constitution 13, 277-8, 282-7, 295, 296; Court 291, 292; Principles 282-3 CONTRALESA 125 Cultural Heritage Development Council 122, 123 December 1st Movement 122-4 passim Finance and Fiscal Commission 286, 293,294 Intergovernmental Forum 295 Khoisan 115,122,123,125 Kleurling Weerstand Beweging 122, 123 parties 288, 297; ANC 10, 113,119-21 passim, 125,126,184, 185,190, 191, 194-9, 281-3 passim, 288, 289, 297, 299; NP 113, 117,120,121,196, 282; UDF 118-20 passim provinces 278, 283-94, 297-9; NCOP 278, 286-90 passim, 293, 296 Truth and Reconciliation Commission 122, 191 South West Africa 31 sovereignty 151-3, 240, 243, 255 Soviet Union 96 Spain 17,19, 56, 58, 60, 303, 304, 312 Spinner, Jeff 62 squatters 174-5 Sri Lanka 273 state and society 15-19, 22-37, 45-50, 179, 183 statization 243-6, 263-4, 274, 275 Stepan,Alfred302,304 Stewart, Angus 177 Stewart, Frances 273, 274 Steytler, Nico 297

335 'strangers' 10,12, 41, 81, 94, 104, 109, 249, 252,314 students 169-71, 241, 246-9 passim, 255 Sudan 67, 68, 93, 306, 314 Sule, Maitama 265—6 Sumba, Patrick Onyango 170 Switzerland 19, 58, 61 Tait, David 32 Takougang,J. 102 Tanganyika 31 Tanzania 314 Tapscott, Chris 295, 296 taxation 45, 92,187, 286-7, 294, 310, 312, 313 Taylor, Cameron 169 Taylor, Charles 56,136-7, 303 Tekere, Edward 170 Thatcher, Margaret 197 Thiam, Iba Der 219-20 Thomas, Louis Vincent 226-8 passim, 230, 234 Throup, David 169, 172,175,177, 178, 202, 212,214,215 Thuku, Harry 201—4 passim Tinubu, Ahmed 262 Tocqueville, Alexis de 167 Toder, Eric 312 Towett,Taita arap 209, 212 trade 7, 21, 30, 39, 74, 75, 78, 81, 154,156, 164,174,230,234,266 trade unions 18, 51,194,198, 205, 208, 321, 322 transparency 43, 278, 280, 298 tribalism 5, 8, 74, 84, 85, 88-90 passim, 95, 134-5,142, 144, 168,169,172-3,186, 198,299,317 political 5-8 passim, 11-12, 19, 20, 46-53, 76-83 passim, 85, 88, 91, 92, 113,126,309,318,320 trust 5, 7, 11, 17, 18, 20, 38-53, 94, 98,121, 313,320,321 Tuareg 70 Tukur, Mahmud 261, 271 Tully, James 280 Turks 60 Tutsi see Congo Tyrol, South 58 Uganda 202, 304, 310-11 Ugochukwu 272 Ujamaa 168 UNDP 169

Index

336 unemployment 89, 140, 193, 195 United Nations 107 United States 2, 53-63 passim, 167,169, 302,305,310,312-13,321 African Americans 61-3 Amish 56 civil rights movement 62 universities 51, 52,170-1, 241-2, 246-9, 274 urbanization 132, 139,141, 189, 193 Vail, L. 138 value systems 6, 40, 273-4 Van Binsbergen, William 136, 138,139 Van der Ross, R. 117 Vikings 34 Villiers, Bertus de 296 violence 10,12, 31, 34, 97, 105,113, 163, 183-4,191, 192, 194,195, 223, 228, 257, 259 Vossen, Rainer 136, 143 Wachira, Kamonji 171 Wade, Abdoulaye 218 Waiyaki, Munyua 210, 211 Wales 58 Walker, Cherryl 198 Walloons 58 Walzer, Michael 60, 61 Wambaa, Charles 206 Wamwere, Koigi wa 170 war 75, 154, 155,164 Anglo-Zulu 186 civil 9, 16, 34, 148,157, 159,176, 215, 262, 263, 306 Mau Mau 85-91 passim, 168,176 of Religion 16 Second World 7 Ward, K. 124 Waruhiu, Chief 202, 207

Weber, Max 39, 41, 178-9 Weinstock, Daniel 311 Welsh, David 284 Werbner, Richard 178 Whisson, Michael 174 Whyte,JohnD310 Widener, Jennifer 214 Wilks, Ivor 28 Wilmsen, Edwin 136 Winstanley, G. 129 Wiseman, J. 140 women 6-7,13,18, 78, 81, 86, 91, 135, 188, 189,192,193,198-9,269 Woode, Samuel 321 Worden, N. 124 World Bank 51, 176, 281 xenophobia 56, 275 Yakassai, Tanko 261 Yeros, Paris 319 Yoruba see Nigeria Young, Crawford 96, 135,139, 173 Young, Kihara 170 youth 6, 7,138,163,189,192-4 passim, 202,275 Yugoslavia 73, 75 Yusuf, Kabiru 270 Yusuf, Sheikh 124 Zadok, Asaph 265 Zaire/Za'irianization 244, 307, 312 Zak-Zaky, El 270-1 Zambia 314 Zimbabwe 74 Zuern, Elke 298 Zulus see KwaZulu-Natal Zuma, Jacob 196 Zwelithini, King Goodwill 192—3