Sure Road? Nationalisms in Angola, Guinea-Bissau and Mozambique African Social Studies Series

Editorial Board Martin R. Doornbos, International Institute of Social Studies, The Hague Carola Lentz, University of Mainz John Lonsdale, University of Cambridge

VOLUME 28

The titles published in this series are listed at brill.nl/afss Sure Road? Nationalisms in Angola, Guinea-Bissau and Mozambique

Edited by Eric Morier-Genoud

LEIDEN • BOSTON LEIDEN • BOSTON 2012 This book is published with the support of the Association des chercheurs de la revue ‘Lusotopie: Recherches politiques internationales sur les espaces issus de lʼhistoire et de la colonisation portugaise’.

Cover illustration: Mural “Terra de Cabral” by artist Joel Bergner on the walls of the Amílcar Cabral Foundation in Praia, Cape Verde. © Joel Bergner, 2011.

Library of Congress Cataloging-in-Publication Data

Sure road? Nationalisms in Angola, Guinea-Bissau and Mozambique / edited by Eric Morier- Genoud. p. cm. -- (African social studies series ; v. 28) Includes bibliographical references and index. ISBN 978-90-04-22261-8 (pbk. : alk. paper) 1. Angola--Politics and government. 2. Nationalism-- Angola. 3. Guinea-Bissau--Politics and government. 4. Nationalism--Guinea-Bissau. 5. Mozambique--Politics and government. 6. Nationalism--Mozambique. 7. Africa, Portuguese- speaking--History--Autonomy and independence movements. 8. Portugal--Colonies--Africa-- History. 9. Decolonization--Africa, Portuguese-speaking--History. I. Morier-Genoud, Eric. II. Series: African social studies series ; v. 28.

DT1348.S87 2012 320.540967--dc23 2012007361

This publication has been typeset in the multilingual “Brill” typeface. With over 5,100 characters covering Latin, IPA, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities. For more information, please see www.brill.nl/brill-typeface.

ISSN 1568-1203 ISBN 978 90 04 22261 8 (paperback) ISBN 978 90 04 22601 2 (e-book)

Copyright 2012 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Global Oriental, Hotei Publishing, IDC Publishers and Martinus Nijhoff Publishers.

All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher.

Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change.

This book is printed on acid-free paper. CONTENTS

List of Illustrations ��vii List of Contributors �� ix

Introduction. Thinking about Nationalisms & Nations in Angola, Guinea-Bissau and Mozambique ��xiii Eric Morier-Genoud

I Anticolonialism & Nationalism: Deconstructing Synonymy, Investigating Historical Processes. Notes on the Heterogeneity of Former African Colonial Portuguese Areas �� 1 Michel Cahen

II Virtual Nations and Failed States: Making Sense of the Labyrinth ��31 Philip J. Havik

III The Social Origins of Good and Bad Governance: Re-interpreting the 1968 Schism in Frelimo �� 79 Georgi Derluguian

IV Writing a Nation or Writing a Culture? Frelimo and Nationalism During the Mozambican Liberation War ��103 Maria-Benedita Basto

V ‘An Imaginary Nation’. Nationalism, Ideology & the Mozambican National Elite ��127 Jason Sumich

VI UNITA and the Moral Economy of Exclusion in Angola, 1966–1977 ��149 Didier Péclard

VII Angola’s Euro-African Nationalism: The United Angolan Front ��177 Fernando Tavares Pimenta vi contents

VIII Changing Nationalisms: From War to Peace in Angola ��199 Justin Pearce

IX Is ‘Nationalism’ a Feature of Angola’s Cultural Identity? ��217 David Birmingham

X Nationalisms, Nations and States: Concluding Reflections ��231 Gavin Williams

Thematic Bibliography ��251 Index ��265 LIST OF ILLUSTRATIONS

Maps

1. Map of contemporary Guinea-Bissau. �� 30 2. Map of contemporary Mozambique. �� 78 3. Map of contemporary Angola. ��176

Figures

1. First edition of the anthology Poesia de Combate, 1971, one of the first two books published by FRELIMO. ��107 2. Cover page of Os Heróicos, one of the guerrilleros’ newspapers. �� 111 3. Cover of the pamphlet resulting from the first seminar on culture, 1971/72. ��115

LIST OF CONTRIBUTORS

Maria-Benedita Basto is Associate professor (Maître de conférences) at Paris-Sorbonne, University Paris IV, Department of Iberian and Latin- American studies. She works on colonial and postcolonial issues, on memory, identity and nationalisms in the Lusophone world, as well as on the internationalist movements of the 1960s and 1970s. She has published A guerra das escritas. Literatura, nação e teoria pós-colonial, Lisbon: Vendaval, 2006 and edited Enjeux littéraires et espaces démocratiques en Afrique Subsaharienne, Paris: Editions EHESS, 2007. Her latest article is entitled “Corps poétique et critique démocratique: Vico et l’humanisme engagé chez Said”, Tumultes, n°35, 2010/2, pp.103–117. She has recently done research on representations of history at the intersection between literature/writings and documentary cinema.

David Birmingham was educated in Switzerland and took a history degree at the University of Ghana in West Africa. His doctoral thesis, published in both English and Portuguese, considered the relations between Portugal and Angola in the sixteenth and seventeenth centuries. In addition to teaching in universities in Africa and Europe, he has given lectures across the world from California to China and published some thirty essays on Angola and its neighbours. He retired in the year 2001 but many of his books are still in print at Ohio University Press and his history of Portugal lives on in several editions and languages.

Michel Cahen is a political historian of modern colonial Portugal and contemporary Portuguese-speaking Africa. He is a Senior Researcher of the Centre national de la Recherche Scientifique (CNRS) at Bordeaux Political Studies Institute, where he is the deputy-director of the Centre “Les Afriques dans le monde”. From 1994 to 2006, he was editor-in-chief of Lusotopie, a journal devoted to the political analysis of spaces stemming from Portuguese Colonization and History. His main interests relate to Marxism and nationalism, identity and citizenship, political identity of the margins, coloniality and globalization. His most recent books are: Os outros. Um historiador em Moçambique, 1994, Basel: P. Schlettwein Publishing Foundation, 2003; Le Portugal bilingue. Histoire et droits politiques d’une minorité linguistique: la communauté mirandaise, Rennes: Presses Universitaires de Rennes, 2009. x list of contributors

Georgi Derluguian is a sociologist whose monograph Bourdieu’s Secret Admirer in the Caucasus: A World-Systems Biography (Chicago University Press, 2005) received many awards including the Norbert Elias Prize. He studied African history and languages at Moscow State University in the 1980s. During this period, he also spent two years in Mozambique. After 1990 Derluguian worked at the Fernand Braudel Center for the Study of Economies, Historical Systems and Civilizations in Binghamton, New York. He presently teaches evolutionary macro- historical sociology at Northwestern University and New York University, USA.

Philip J. Havik (PhD, Leiden University, The Netherlands) is currently researcher at the Institute for Research in the Tropics (IICT) in Lisbon, while also teaching at universities in Portugal and Brazil. His multidisci- plinary research centers upon the study of state formation, taxation, colonial administration and entrepreneurship in sub-Saharan Africa, with a special emphasis on West Africa and the Guinea Bissau region. His recent publications include: Orlando Ribeiro: Cadernos de Campo Guiné 1947, Oporto: CEAUP/Humus, 2011, in collaboration with Suzanne Daveau; and “Direct or Indirect Rule? Reconsidering the roles of appointed chiefs and native employees in Portuguese West Africa”, Africana Studia, n°15, 2010, pp. 29–56.

Eric Morier-Genoud (PhD State University of New York at Binghamton) is a lecturer in African and Imperial history at Queen’s University Belfast, Northern Ireland. He has written extensively on religion and on politics in Southern Africa, and he works presently on missionaries and transnational sciences and politics as well as war, memory and memorial in contemporary Mozambique. He is editor of the journal Social Sciences and Missions (Leiden: Brill); a co-author, with Caroline Jeannerat and Didier Péclard, of Embroiled. Swiss Churches, South Africa and Apartheid, Münster: Lit Verlag, 2011; and a co-editor, with Michel Cahen, of Imperial Migrations. Colonial Communities and Diaspora in the Portuguese World, London: Palgrave, 2012.

Justin Pearce has been writing on Angola since 2001, as a journalist, researcher and academic, and has closely followed the political and social changes that have accompanied the transition from war to peace. He received his doctorate from Oxford University in 2011 on the local list of contributors xi politics of the civil war in the Angolan Central Highlands, based largely on interviews conducted in the region over the course of the previous two years. He currently holds an ESRC Postdoctoral Fellowship at the School of Oriental and African Studies (SOAS) in London, where he continues to research and publish on Angolan politics.

Didier Péclard is senior researcher at the Swiss Peace Foundation (swis- speace) in Bern and lecturer in political science at the University of Basel. He has worked and published extensively on Christian missions and nationalism as well as on the politics of peace and transition in Angola. As a fellow of the Swiss National Centre of Competence in Research (NCCR) North-South, his current main research focus is on the dynamics of statehood after violent conflicts. He is the author, together with Caroline Jeannerat and Eric Morier-Genoud, of Embroiled. Swiss Churches, South Africa and Apartheid, Münster: Lit Verlag, 2011; and co-editor with Tobias Hagmann of Negotiating Statehood. Dynamics of Power and Domination in Africa, Chichester: Wiley-Blackwell, 2011.

Fernando Tavares Pimenta (PhD, European University Institute) is a researcher at the Centre for Interdisciplinary Studies of the 20th Century, at the University of Coimbra, Portugal. He is the author of the following books, among others: Brancos de Angola. Autonomismo e Nacionalismo, 1900–1961, Coimbra: Minerva, 2005; Angola no Percurso de um Nacionalista. Conversas com Adolfo Maria, Oporto: Afrontamento, 2006; Angola. Os Brancos e a Independência, Oporto: Afrontamento, 2008; Portugal e o Século XX. Estado-Império e Descolonização, 1890–1975, Oporto: Afron tamento, 2010; and Storia Politica del Portogallo Contemporaneo, 1800–2000, Florence: Mondadori, 2011.

Jason Sumich studied anthropology at the London School of Economics and his work examines nationalism, legitimacy, democratisation and social stratification in Mozambique. He currently works as a Research Fellow at the Department of Anthropology, NTNU and as a fellow of the Human Economy Research Programme, University of Pretoria. His latest publications include: “Does all that is Solid Melt into Air? Questioning ‘Neo-Liberal’ Occult Economies in Mozambique”. Kronos: Southern African Histories, Vol. 36, pp.98–113 and “The Party and the State? The Ambiguities of Power in Mozambique”, Development and Change Vol. 41(4), pp.679–698. xii list of contributors

Gavin Williams is South African. He studied at the Universities of Stellenbosch and Oxford. He lectured at Durham University before being elected to a Fellowship in 1975 at St Peter’s College, and Lecturer at the University of Oxford. He has taught and examined for Stellenbosch, Rhodes, Natal and Witwatersrand Universities. He has supervised a number of masters and doctoral theses, of which four have been on Mozambique and Guinea-Bissau. He has published extensively on politics, political economy, land and agricultural policies in Nigeria and South Africa. His primary interests are in social theory and the empirical study of politics and societies. He is currently engaged in research on the history of the South African wine industry. INTRODUCTION

THINKING ABOUT NATIONALISMS & NATIONS IN ANGOLA, GUINEA-BISSAU AND MOZAMBIQUE

Eric Morier-Genoud

Our sure road is pain and blood Straight road to the sun The sun of our freedom.1

Nationalism is back in force in Angola, Guinea-Bissau and Mozambique. While the years 1990s saw a rise in public debates about the liberation struggle and national heroes, the years 2000s saw a flurry of official com- memorations and the building of many monuments to celebrate nationalist achievements. In Angola, three new memorials to national heroes and martyrs have been built by the Ministry of Public Works since 2003 as well as a Cultural Centre to the memory of Angola’s first president Agostinho Neto. In Mozambique, all formers liberation war’s military bases have been turned into monuments, and a National Liberation Museum has been created in the northern village of Chai in 2005. Smaller memorials to specific nationalist figures have been inaugurated all over the country as well as many monuments to the first postcolonial president, Samora Machel. In Guinea-Bissau, the on-and-off civil war since 1998 has absorbed most of the government’s energies. But an interesting initiative has still emerged in 2004 joining former African nationalist guerrillas and former Portuguese soldiers to hold a conference, restore a military base and establish a museum at Guiledje – the conference was held in 2008 with the support of the Mario Soares Foundation.2 In Cape Verde, the government and the Amílcar Cabral Foundation were active the same year in celebrating the memory of the national hero of Guinea- Bissau and Cape Verde who should have turned 80 that year. In all these

1 Sampadjudo, “Our sure road” in Margaret Dickinson (ed.), When Bullets Begin to Flower, Nairobi: East African Publishing House, 1980 [1972], pp. 106–107. 2 See the website: http://www.guiledje.org/ (accessed 3 August 2010).

© koninklijke brill nv, leiden, �� | doi ��.��/��_�� xiv eric morier-genoud countries, the inauguration of monuments, museums and centres, and the celebrations of anniversaries, have been accompanied by a bout of nationalist discourse geared towards today’s youth who is asked to cher- ish the past and faithfully continue the work that heroes are said to have begun. Needless to say, these nationalistic activities have led to much contro- versy. In Mozambique, the hottest discussion has been about who is, and who should be, in the crypt at Heroes Square. Individuals buried there before 2003 were all Frelimo liberation fighters and this came to be seen as problematic, even within the party in power, for two main reasons. First, because the heroes were all armed fighters. There was no civilian, and particularly none of the artists who had done so much to create a national culture and consciousness. Second, the individuals in the crypt were all members of Frelimo, the party in power since independence. There were no nationalists not affiliated with the party in power, no armed fighters from other Mozambican liberation movements, and none of the Frelimo dissidents who played a fundamental role in the foundation of the front. In 2003 and 2004, President Joaquim Chissano addressed the first line of critique by including two more personalities in the crypt: the poet José João Craveirinha and the composer of the first national hymn, Justino Chemane. But Chissano did not tackle the partisan accusation, nor has his successor since 2005. This has led some opposition parties, such as Renamo, to boycott official ceremonies at Heroes Square which they argue is a partisan locale. Is this the case? Is Heroes Square merely a Frelimo place? Are Mozambique’s official heroes only from one political party? By extension, is today’s official history and commemoration selective and biased? If so: how, why, and since when? What is being left out, what kind of nationalism is being promoted (and obscured), and what impact does this all have on the national imagined community? To put it in broader terms, what is the history of the formation and politics of nationalisms and nations in Mozambique and, by extension, Angola and Guinea-Bissau? There is today a strong interest in nationalisms and nations in Portuguese-speaking Africa, both in society and in academia. This renewed interest can be traced back to the 1990s, and its cause to three main factors. First, the authoritarian decompression which took place in the late 1980s-early 1990s led to more open and pluralist political systems, something which permitted the emergence in the public realm of new voices and new testimonies about the war of liberation and the history of nationalisms. Voices dissonant from the official scripts begun to be heard introduction: thinking about nationalisms and nations xv publicly in the 1990s and this fostered much curiosity. As founding myths were shattered, a demand for “truth” emerged. Second, time has passed since the liberation struggles of the 1960s and 1970s and a younger generation has come of age with no direct experience of the period while the men and women who lived through these events are gradually disappearing. This results in the old generation wishing to leave a testimony of its experience, in particular war veterans, while the new generation wishes to know more about a period which is clearly fundamental yet not properly recounted. A third and final factor is the emergence of new state nationalisms in Portuguese-speaking Africa. Indeed, while the authoritarian decompression fractured national myths, democracy and neo-liberalism simultaneously forced states and their elites to develop new forms of “patriotism”.3 These new patriotisms generate their own set of discussion and debates (about what the nation is, what it should be, where it comes from, and where it should go) and this sparks ever more interest in the topic within society. Surprisingly academic research on nationalisms and nations in Angola, Mozambique and Guinea-Bissau has remained limited during the last twenty years – particularly in English. Few academic works have revisited the liberation struggles, have focused on the history of nationalisms, or analysed the formation and make-up of contemporary Portuguese- speaking African “imagined communities”. There are of course some wor- thy exceptions in English,4 as well as in French and Portuguese.5 Still, one can say that, overall and comparatively to the interest developing in society, little has been published on the subject, and we are still very

3 Among others, see Sara Dorman, Daniel Hammett & Paul Nugent (eds), Making Nations, Creating Strangers. States and Citizenship in Africa, Leiden: Brill, 2007 and the special issue “New Nationalism and Xenophobia in Africa”, Africa Spectrum (Hamburg, Germany), Vol.41, No. 1, 2009. 4 Mustafa Dhada, Warriors at Work: How Guinea Was Really Set Free, Niwott: University Press of Colorado, 1993; Patrick Chabal (ed.), A History of Postcolonial Lusophone Africa, London: Hurst & Co, 2002; Marissa Moorman, Intonations: A Social History of Music and Nation in Luanda, Angola, from 1945 to Recent Times, Athens: Ohio University Press, 2008; and Paolo Israel, “Utopia Live: singing the Mozambican struggle for national liberation”, Kronos (Cape Town), Vol.35, No. 1, 2009, pp. 98–141. 5 Christine Messiant, “Sur la première génération du MPLA: 1948–1960. Mário de Andrade, entretiens avec Christine Messiant (1982)”, Lusotopie 1999, Paris: Karthala, pp. 185–224; Jean-Michel Mabeko-Tali, Dissidências e poder de estado. O MPLA perante si próprio (1962–1977), Luanda: Editorial Nzila, 2001, 2 vols; Michel Cahen, Les bandits: Un historien au Mozambique, 1994, Paris: Centre culturel Calouste Gulbenkian, 2002; Luís Reis Torgal, Fernando Tavares Pimenta & Julião Soares Sousa (eds), Comunidades Imaginadas. Nação e Nacionalismos em África, Coimbra: Imprensa da Universidade de Coimbra, 2008. xvi eric morier-genoud dependent on the material produced in the 1970s and 1980s on, if not by, the MPLA (Movimento Popular de Libertação de Angola), FRELIMO (Frente de Libertação de Moçambique) and the PAIGC (Partido Africano da Independência da Guiné e Cabo Verde). Yet very important new archives have opened in the last twenty years in Portugal as well as the Portuguese- speaking African countries: the PIDE secret police archives, the colonial military archives, Frelimo’s archives (partially), MPLA’ archives (partially), etc. Moreover, the worldwide historiography on nationalisms and nations has evolved considerably and the theoretical approaches on the subject have changed greatly (see more below). There is therefore not just a quantitative lack of investigation on nationalisms and nations in Portuguese-speaking Africa, but also a qualitative issue. There is an urgent need to revisit the subject of nationalisms and nations in Angola, Guinea- Bissau and Mozambique to enrich and verify empirically as well as theoretically what we already know of the subject. The present collection of essays aims to move in this twofold direction: bringing in new sources and renewing our theoretical approaches. Doing research on nationalisms and nations in Lusophone Africa today is a complicated affair. States and parties-in-power, as we have noted, are very active in developing and deploying new forms of patriotism and nationalisms. Linked to this, there is a spell of new public history projects on the “liberation struggles” and there is a flurry of publications by former actors of the liberation/colonial wars about “decolonisation”. Historical projects have been launched in the last ten years by no less than the Comunidade dos Países de Língua Portuguesa (CPLP), the Southern African Development Community (SADC), the Mozambican Ministry for the Affairs of Former Combatants, the MPLA and the Frelimo parties, among others – see more in table 1 and 2 below. On the former colonial side, some military scholars have published analysis of the African “campaigns” and former soldiers are investing ever more in the writing of their memoirs (in paper as well as online).6 While this spate of research and writings is most welcome and useful, it brings a range of problems connected to the particular vantage point and objectives of the authors and researchers involved (e.g. veteran fighters) and their often uncertain

6 For military scholarship, see for example Aniceto Afonso & Carlos de Matos Gomes, Guerra Colonial, Lisbon: Notícias, 2000; Carlos de Matos Gomes, Moçambique 1970. Operação Nó Górdio, Lisbon: Prefácio, 2002; and Rogério Cardoso Teixeira, Angola (n’gola). História do Batalhão de Caçadores 109, Lisbon: Quarteto, 2008. For soldiers’ memoirs, see the very rich website of the “Overseas wars” at http://ultramar.terraweb.biz/ introduction: thinking about nationalisms and nations xvii academic method and rigor. From the material already published, we can say for example that a majority of the work reads back into the past and produces a teleological narrative of “decolonisation” or “national liberation”.7 Yet we know that there never was, nor ever is, any “sure road” or “straight road” to national formation and national liberation (the alleged “sun of our freedom” as the opening militant poem states) or to “decolonisation”. The academic scholarship has already deconstructed much of the official nationalist narratives and paradigms in relation to Angola, Mozambique and Guinea-Bissau.8 The purpose of the present volume is therefore not to repeat or continue such deconstruction, but to open up instead the subject by asking new questions and pointing new directions for research and debate. By looking at the nature and dynamics of “marginal” nationalisms, the role of culture, the way nationalist movements fitted into international networks, or how particular nationalist movement failed, we hope to restore the uncertain, convo- luted and conflictual trajectories of nationalisms and nations before and after independence.

Table 1. International research projects on nationalism. Institution Project name Begun in Nordic Africa “Nordic Documentation on the Liberation 2003 Institute Struggle in Southern Africa Project” [to “fill the gaps that might exist in the search for a new liberation history”] ALUKA “Struggles for Freedom in Southern Africa” 2003 SADC Hashim Mbita Project [“to document the 2004 history of the liberation struggles”] CPLP Joint Historical Archive of the National 2005 Liberation Struggles

7 See for example Lourenço do Rosário (ed.), II Congresso sobre a Luta de Libertação Nacional-Guerra colonial. 27 anos depois, a reflexão possivel, Maputo: Edições ISPU, 2004; Helder Martins, Porquê Sakrani? Memórias dum Médico duma Guerrilha Esquecida, Maputo: Editorial Terceiro Milénio, 2001. 8 Among others, see Carlos Pacheco, MPLA. Um nascimento polémico, Lisbon: Vega, 1997; Michel Cahen, “The Mueda Case and Maconde political ethnicity. Some notes on a work in progress”, Africana Studia (Oporto), No.2, 1999, pp.29–46; and Christine Messiant, “‘Chez nous, même le passé est imprévisible’. L’expérience d’une recherche sur le nationalisme angolais, et particulièrement le MPLA: sources, critique, besoins actuels de la recherche”, Lusotopie 1998, Paris: Karthala, pp. 157–197. xviii eric morier-genoud

Table 2. National research projects on nationalism. Country Project run by Begun in Angola Arquivo Histórico de Angola ? Angola MPLA party 2004 Mozambique Ministry for the Affairs of 2001/2008* Former Combatants Mozambique Frelimo party 2000s Guinea-Bissau Various 2004 * 2001 is the date of a new state policy which aimed at fostering investigation; 2008 is the date of the creation of the Centro de Pesquisa da História da Luta de Libertação Nacional

Analysing Nationalisms and Nations

The historiography of nationalism and nations has evolved importantly over the last two decades. The 1980s and 1990s saw intense debates between “primordialist” and “modernist” authors, the first arguing that nations were “ancient” and “natural” while the second put forward that nations were “modern” and “imagined”, if not “invented”.9 The debates drew much interest and have led to a growth of the field as well as the development of new institutions and journals – the primordialist launched for example the Association for the Study of Ethnicity and Nationalism in 1990 and the journal Nations and Nationalism in 1995. Debates bore many intellectual fruits as new questions were raised, much research undertaken and new answers proposed. By the 1990s already the primordialist and the modernist approaches had absorbed some of the critique made against them and had evolved as a result. Primordialists recognised an element of imagination and invention in nations and nationalisms and they now argued that nations were imagined, if on the basis of something natural such as “ethnic cores”.10 Modernists, on their side, conceded that there could be no pure invention and went on to argue in favour of an idea of a “construction”, “imagining” and “formation” on the basis of history and pre-existing identities (thus still arguing these processes were genuinely modern). By the years 2000s, the two schools

9 For the classic primordialist approach, see Anthony D. Smith, The Ethnic Revival in the Modern World, Cambridge: Cambridge University Press, 1981. For the classic modernist approach, see Benedict Anderson, Imagined communities. Reflections on the Origin and Spread of Nationalism, London: Verso, 1983 and Eric Hobsbawm & Terence Ranger (eds), The Invention of Tradition, Cambridge: Cambridge University Press, 1983. 10 Anthony D. Smith, The Ethnic Origins of Nations, Oxford: Blackwell, 1986. introduction: thinking about nationalisms and nations xix still differed over the amount of imagining/primordiality they saw in nations and nationalisms, but they reached a consensus that imagination was at work and that the latter was not performed “out of the blue” – there is always much history and politics involved. Coming from subaltern studies, Partha Chatterjee made an important contribution in 1993 when he criticised the above-mentioned approaches for giving too passive a role to non-elite actors and non-Europeans.11 He attacked Benedict Anderson in particular for claiming that the European model of modern national imagined communities had been merely exported and reproduced overseas. Chatterjee argued in favour of an analysis of the specific historicities of national imaginations, at the intersection of a hegemonic Western normalizing project, various frontal and fragmented resistances, and diverse dynamics of re-appropriation. The aim, he claimed, was to reclaim the “freedom of imagination” of those once-colonized in the making of their own identities.12 Chatterjee was rapidly joined in his deconstruction and complication of the history of nationalism(s) by other authors. Coming from a gender perspective, Susan Geiger showed importantly in 1997 that not only had women been active in nationalism in Tanganyika, but that they had often set the pace which educated male politicians had followed (if not taken advantage of).13 Authors like Elizabeth Schmidt pushed this line further and went on to study nationalism “from the bottom up” – how the mobilization of the masses shaped elite nationalism.14 Such “decentring” was reinforced in the 2000s with a recourse to the anthropology of consumption which argues that consumption is part the total production of commodities (because consumption is an act of appropriation and alteration). Authors like Kelly Askew thus looked at how the nation was performed in Tanzania, that is formed through (in her case musical) performance.15 By 2012, the study of nationalisms and nations has swung from an analysis of export and imposition to one of diffusion and re-appropriation, from an analysis of Western and elite nationalism to investigations from below and the periphery, and from an analysis of production to that of

11 Partha Chatterjee, The Nation and its Fragments. Colonial and Postcolonial Histories, Princeton: Princeton University Press, 1993. 12 Partha Chatterjee, The Nation and its Fragments, op.cit., p. 13. 13 Susan Geiger, TANU Women. Gender and Culture in the Making of Tanganyikan Nationalism, 1955–1965, Oxford: James Currey, 1997. 14 Elizabeth Schmidt, “Top Down or Bottom Up? Nationalist Mobilization Reconsidered, with Special Reference to Guinea (French West Africa)”, American Historical Review, No.110, October 2005, pp. 975–1014. 15 Kelly M. Askew, Performing the Nation: Swahili Music and Cultural Politics in Tanzania, Chicago: Chicago University Press, 2002; see also Moorman, Intonations, op.cit. xx eric morier-genoud reception and performance. While authors might prefer one approach to another (and there are disciplinary, theoretical as well as political issues here), a full understanding of the “historical formation” of nations and nationalisms will eventually need to encompass all angles.16 And the fact is that much still needs to be done in terms of an analysis of nationalisms as social movements from below and of nations as cultural productions, among others, before we can talk of totalities and before we can reconstruct adequately the span of diversities and historicities of nations and nationalisms – and this is particularly true of the cases of Angola, Guinea- Bissau and Mozambique. This said, substantial theoretical lessons have been gained already and it is worth discussing some key advances now so as to enrich our understanding of the subject in general and in relation to the cases which interest us here, as well as to highlight some of the contributions which the present book makes. The list of advances will not be complete, nor tightly reflect the chapters that follow. Instead it will highlight certain aspects in the hope of offering structural foundations to a critical understanding of the subject of nationalisms and nations, and in the hope of providing a stimulating perspective on how to read the chapters that constitute the present volume. The first such advance is conceptual: a clarification of concepts and a realisation that one has to be very careful with definitions and vocabulary. One needs to be attentive not to conflate words and concepts such as “nationalism” and “nationalist movements” for example. The two are not only different, but there is usually more nationalism in any society than what exists in any nationalist movements (as representative as the latter may be). Susan Geiger has shown this, as we mentioned earlier, in relation to Tanganyika where women promoted nationalism before any nationalist movement was formed as well as outside and beyond existing political unions and parties when the latter came into being.17 Similarly, we need to distinguish between the concept of “nation” (a particular form of imagined community) and the concept of “nationalism” (an ideology or a political project which aims at creating, defending or working for this imagined community). The history of a nation is not equivalent to the history of nationalism: nation formation is a long-term society-wide development while nationalism can be short term and can be the affair of a small group only, or even an individual. The difficult part here is not to distinguish and

16 On the concept of “historical formation”, see Jean-François Bayart, The Illusion of Cultural Identity, London: Hurst & Co., 2005, ch.2. 17 Susan Geiger, TANU Women, op.cit. introduction: thinking about nationalisms and nations xxi unpack these concepts: it is to analyse the intricate relation between the two and their evolution over time – how a nation comes into being and how nationalists build on, and contribute to shaping, a community imagined by a much wider and diverse set of people. Still about concepts, one needs also to avoid conflating nations and nation-states as Michel Cahen reminds us in his opening text (chapter I). The two can go together, partly or fully, but they rarely coincide – in fact they sometime do not correspond at all (in spite of what actors may say). In the same vein, anti-colonialism and nationalism need to be distinguished even if, again, the two can go together, partly or fully. Anti- colonialism is more encompassing than nationalism and it is reactionary (a reaction to colonialism) while nationalism can be that too, but it is also and above all a forward-looking social and political project. Last but not least, Cahen introduces in his chapter a new concept, “nationism”, so as to distinguish between a form of nationalism which would build on or reflect an existing nation, and nationism which does not build on or reflect an existing nation but aims at creating a nation (by force if necessary). Such concept and distinction help us think more finely about nationalism and its relation to the nation and other forms of imagined communities. Like the preceding distinctions, the difficulty of these concepts does not lie in their definition, but in their application and our using them rigorously in our analysis – too often authors and theorists slip, conflate or confuse terms and concepts. A second advance made in the last twenty years relates to how we understand the history of nationalisms and nations. Much early literature focused on elite and the Western origins of nationalisms and nations as we have seen, and it tended to view change in an evolutionist fashion. The understanding of the history of nationalisms and nations was progressive and teleological: a succession of stages which inevitably led to the liberation of all un-free nations. History was seen as linear and causal: change was produced by antecedent causes, outcomes seemed by and large determined, and progress followed a straight and sure road between all points.18 Nations were seen as actors of history who aimed at their own liberation and one could as a result practice history as an investigation into the origins of a pre-determined outcome, reading back into the past the origins and development of nations. The post-structuralist and

18 For a good deconstruction of such linear nationalist history, see Prasenjit Duara, Rescuing History from the Nation. Questioning narratives of Modern China, Chicago: University of Chicago Press, 1995. xxii eric morier-genoud postcolonial critique of this type of history has been very perceptive and its lessons need to be integrated into our analysis. Without obligatorily adopting the postcolonial approach, we need to reject linear, evolutionist and teleological narratives and restore diversity, complexity and uncertainty. This means, first, to integrate all currents of nationalisms in our analysis, not just the dominant ones or those who won historically. Taking an extreme case, Fernando Tavares Pimenta analyses in this volume (chapter VII) the case of “leuko-nationalism” in Angola. After showing how nationalism emerged within some sectors of the white population, he discusses the formation of a white nationalist movement whose outcome was failure, but whose history tells us much about Angola and African nationalisms in that territory. In his contribution, Didier Péclard (chapter VI) analyses the roots of UNITA’s nationalist movement. He shows how a particular social and historical trajectory led not just to a particular form of nationalism in central Angola, but also to a paradoxical success for the party which emerged there, i.e. its collapse at independence allowed it to rebuild more successfully after independence. A third advance in the analysis of nationalisms and nations in the last twenty years has been in relation to culture. Culture is not understood today as a founding-block of nationalism anymore nor as a mere trait of a people. Culture is seen as a human production which has its own history which needs to be unpacked and analysed. Nationalism in turn is now seen as a cultural production among others. Following this approach, Maria-Benedita Basto analyses in her text (chapter IV) how Frelimo, the dominant Mozambican nationalist movement, related to culture. She reveals spontaneous and organised literary production in the “liberated areas” during the war and she proceeds to analyse how the Frelimo leadership tried to establish a nationalist canon through selection, correction and exclusion. Dealing with the same period, Georgi Derluguian revisits Mozambique’s famous intra-nationalist conflict around 1968 (chapter III). He looks at the cultural background, and the habitus, of the two factions of Frelimo at the time and how these tied, successfully or not, into the global network and structures of power of the time. Finally, dealing with a different period, Jason Sumich (chapter V) looks at the meaning and form which nationalism has taken after independence in Mozambique. He shows how it has changed among the elite and how the latter uses it today as a sign of distinction and a mean to continue their domination. A final contribution in the last twenty years has to do with the history of nations, as opposed to nationalisms, that is the history of “imagined communities”. Michel Cahen in his opening text (chapter I) shows the introduction: thinking about nationalisms and nations xxiii very diverse historical trajectories of the nations in Angola, Guinea- Bissau, Mozambique and other Lusophone countries and the implication that each route has had for the formation of a particular imagination. Following on this macro-analysis, David Birmingham looks in his contribution (chapter IX) at the particular case of Angola. He analyses the historical development of culture and identity in that territory and he wonders thereafter, considering present connections and imaginations, if today’s culture does no contradict, and possibly even undermine, any national, nationalist or nationist agenda. More pessimistically (or realistically) yet, Philip J. Havik considers (in chapter II) the state of the nation in Guinea-Bissau, a country marred by civil war since 1998. While the liberation war in that territory was the most successful among the Portuguese colonies in the 1970s, the national imagination seem to have receded to the point that the author asks whether it is not a myth or even a curse! Finally (in chapter VIII) Justin Pearce looks at the latest development of nationalisms and nations in Angola. While he concurs with most authors that the nation in Angola is weak and nationalism is dominated by the party in power, he reveals lively debates under the surface of formal politics, showing that the idea of the nation is still alive, debated and a bone of contention and competition. In the last chapter of the book, Gavin Williams provides an overview of the chapters and cases dealt with in the book. He offers a comparison of the dynamics of nationalisms and nations in Angola, Guinea Bissau and Mozambique, and he brings the state and the economy back into the discussion. This is particularly useful since structural dynamics, notably economic, always constrain historical developments and the action of individuals. Bringing the state back in the discussion is similarly important since it is a fundamental actor in all the developments related to nationalisms and nations. The state may be an idea, even a collective misrepresentation of capitalist societies, but it still generates practices and institutions which are central in shaping people’s thoughts and actions, not least because it holds the monopoly of legitimate use of force over a territory. These elements and the analysis which Gavin Williams brings help move the discussion towards a greater level of generalisation and, at the same time, bring us back to fundamental issues of the social sciences. To conclude, the chapters of this edited volume will not provide a history of nationalisms and nations in Angola, Guinea-Bissau and Mozambique. They will not provide an encompassing picture on, or an updated systematic view of, the subject. What it will do is reveal new xxiv eric morier-genoud facets and dimensions of the history of nationalisms and nations thanks to access to new primary sources. And just as importantly, it will raise new questions and propose new approaches on the subject, all in the hope of opening up the topic for an informed, critical and open-ended discussion. In opposition to returning “patriotic histories”,19 this book intends indeed to contribute at restoring the complexity of the historical process in question, reveal the various roads which were possible at all times, and show the choices which actors made and bifurcations they took, for better and for worse, within the constraints which existed at the time, in the name of “the sun of our freedom”. Contrary to what the opening poem stated, there never was nor ever will be any “straight” or “sure” road to nation formation and a nation’s independence. There are other legitimate forms of political organisations as well as many hybrid forms; there are many roads as well as some dead end; and all outcomes are eventually the result of historical processes which are contingent, contextual, and fought over – alongside their memory and their history.

Acknowledgment

This collection has its origin in a workshop held at the University of Oxford in December 2007 under the auspices of the Oxford Research Network on Government in Africa (OReNGA), the Department of Politics & International Relations, and St Cross College. Support for the workshop came from British Academy (Grant BGC-47197), the African Studies UK (ASAUK) and diverse sections of the University of Oxford: OReNGA, the Department of Politics & International Relations, the Faculty of History, the African Studies Centre and the Camões Centre for Portuguese Language. I would like to thank, for their contribution during and after the conference: Dr. Jan-Georg Deutsch, Dr. Jocelyn Alexander, Dr. Ike Okonta, Dr. Luisa Pinto Teixeira, Professor Joel das Neves Tembe, Professor Luís de Brito, Professor Marissa Moorman, Professor Patrick Chabal, Professor Gary Littlejohn, Professor Neil McFarlane, and Bryan Mukandi. I am particularly grateful to Gavin Williams with whom I organised the conference and who subsequently provided advice and support for the making

19 Terence Ranger, “Nationalist historiography, patriotic history and the history of the nation: the struggle over the past in Zimbabwe”, Journal of Southern African Studies, Vol. 30, No.2, 2004, pp.215–234. introduction: thinking about nationalisms and nations xxv of the book. Behind Gavin Williams stood the spirit of Thomas Lionel Hodgkin, the pioneer of studies of nationalism in Africa.20 Hodgkin had been Gavin Williams’ mentor, and Gavin was mine when I was a “Thomas L. Hodgkin Research Fellow” in Oxford. Mention needs to be made finally of the School of History and Anthropology at Queen’s University Belfast who provided me the time and means to transform the conference’s proceedings into a book. I am equally grateful to the Association des chercheurs de la revue Lusotopie who supported this publication financially and intellectually.

20 Thomas L. Hodgkin, Nationalism in colonial Africa, New York: New York University Press, 1957.

CHAPTER ONE

ANTICOLONIALISM & NATIONALISM: DECONSTRUCTING SYNONYMY, INVESTIGATING HISTORICAL PROCESSES Notes on the Heterogeneity of Former African Colonial Portuguese Areas

Michel Cahen

In this chapter, the historical connections between anticolonialism and nationalism will require the discussion of universal concepts, while also exploring the particularities of Portugal’s former African Empire. The general idea is to contest the interchangeable use of the words ‘anticolonialism’ and ‘nationalism’, as well as to understand why these words are used so interchangeably. I do not support the theory of the African ‘imported State’, as put forward by Bertrand Badie some years ago,1 because in defining a State the first task is not to describe it, but to explain its functions or “duties”.2 From this point of view, African independent States, even if they are completely neo-colonial, are new States, new historical productions (even if globalized) and they are responsible for new tasks in linking their legitimacy with the world economy. Of course there can be no state without a certain historicity.3 But there was no such thing as a “colonial State transmission,” because there never was a “colonial State” in each of the

1 Bertrand Badie, L’État importé. L’occidentalisation de l’ordre politique, Fayard, Paris, 1992, 334 p. 2 The same debate exists over the characteristics of the Portuguese state under the rule of António de Oliveira Salazar: the methodology of description of the state leads to the conclusion of huge differences with the regimes of Hitler or Mussolini, and therefore to a definition of the Portuguese state as ultra-authoritarian but not ‘fascist’ when the analysis of its functions militates for such a definition, which is my personal point of view. On this specific debate, see Michel Cahen, « Salazarisme, fascisme et colonialisme. Problèmes d’interprétation en sciences sociales, ou le sébastianisme de l’exception », Portuguese Studies Review (Trent University, Canada), vol. 15, no. 1, 2007 [issued 2009], pp. 87–113. 3 Jean-François Bayart, ‘L’historicité de l’État importé,’ in J.-F. Bayart (ed.), La greffe de l’État, Karthala, Paris, 1996, p. 11–39. It is worth noting that Bayart maintains the idea of “imported state”, nevertheless insisting on its historicity, which seems to me quite contradictory …

© koninklijke brill nv, leiden, �� | doi ��.��/��_�� 2 michel cahen colonies.4 What there was is an imperial administrative apparatus of European States. The imperial administrative apparatus was not a State, or even a proto-State, any more than the Nazi police and State were a local State in France during World War II. They were a part of the German State. One may consider the imperial state apparatus as a kind of nation- alization of Chartered Companies, which existed previously, a kind of estate but not a state.5 The colonial administration did not “prepare” the African independent State, except during the very late years, a very short period of time on a historical scale. On the contrary, over a far longer period, the colonial administration first broke the African States, and then went on to weaken the African tradition of the State. The colonial period was a time of State recession in Africa. African countries are still paying for that, but their present states, which succeeded the European administrative apparatus chronologically, are new States, not States in which only the leadership has changed. That said, I am not denying the existence of neo-colonialism, but the neo-colonial nature of the African State stems from the peculiarity of modern integration in the world economy, and not only from the sociology of colonial transmission. Here, the focus is on the very nature of the State. But beyond this level of analysis, when we look at social and cultural history, I am the first to acknowledge that there has been the sociological transmission of a huge range of things, not least ideology, all of which Balandier called a ‘colonial situation’.6 But this “transmission” paradigm applies not only to leaders, generals or social scientists from Africa; but also to European leaders, and to us, European social scientists, when we consider Africa.

4 Examples of the use of the concept of « Colonial State » applied to colonial Africa may be found in, among others, John Lonsdale & Bruce Berman, ‘Coping with the contradictions: the development of the colonial state in Kenya,’ Journal of African History, no. 20, 1979, pp. 487–506; Crawford Young, The African Colonial State in Comparative Perspective, Yale University Press, New Haven, 1994, 356 p.; Frederick Cooper. ‘Grandeur, décadence … et nouvelle grandeur des études coloniales depuis les années 1950,’ in ‘L’État colonial’ (dossier), Politix (Bruxelles, De Boeck), 2004, vol. 17, no. 66, pp. 107–136, pp. 17–48; Pierre Boilley, Jean-Pierre Chrétien & Christine Deslaurier, La transmission de l’État colonial, Karthala, Paris, forthcoming. Out of Africa, see for instance Romain Bertrand, État colonial, noblesse et nationalisme à Java: la Tradition parfaite (17ème-20ème siècle), Karthala, Paris, 2005, 760 p. (« Recherches internationales »). 5 The archetype would obviously be the « État indépendant du Congo » (1885–1908), but its official recognition by the Belgian state in 1908 did not change its nature. It is worth noting that, in the Portuguese empire, Chartered companies existed up until 1941. 6 Georges Balandier, ‘La situation coloniale: approche théorique,’ Cahiers internation- aux de sociologie (Paris, Presses universitaires de France), vol. 11, 1951, pp. 44–79. anticolonialism & nationalism 3

That is particularly true of the nation-state. But, bearing in mind that, for obvious reasons, the view of the nation-state in Africa stems from the European model, I will also explain why a certain kind of Marxism has reinforced this Eurocentrism. I am not saying that “Marxism is not for Africa”, far from it, but there is no doubt that a certain kind of Marxism has increased a local tendency towards Eurocentrism.

Avoiding Confusion of Concepts: State, Nation, Nation-State

Now and then, the nation-state is confused with the state, with the terms sometimes used interchangeably. For example, when former Portuguese colonies gained their independence in 1974 and 1975, it was said that new nations had appeared. Why “nations”? States yes, republics certainly, but why “nations”? Similarly, the anticolonial struggle was widely spoken of as the national Liberation Front. Why “national”? What was liberated? Doubtless liberation from colonialism, but is this automatically national? In the same way, talk of “new nations” led to the use of next step concept, that is to say “nation building processes”. If we refuse this tautological assertion – if there is a state there is a nation, and since there is a nation there is a state – and if we refuse the automatic nature of the creation of the nation from the state, there is a whole range of other questions to be considered. How do we define what a nation is? We will return to this question later. Why is there a need to build a nation in order to stabilise and legitimise a republic? Why should a republic have the status of a nation, or have a “nationalising” policy, in order to apply the state-building agenda? Why should a nation-building process, or even nation formation (as defined by Lonsdale in his discussion of state-building and state formation)7 be the required path to stability for a State? Again, I must insist that this reasoning is not intended to imply that a nation-state is “no good for Africa”. On the contrary, I disagree with phrases like “the crisis of the nation-state in Africa”. No doubt there is a crisis of the state in Africa, and no doubt the ideology of many of these states is a “nation-statist ideology”. But is it because a state is

7 Bruce Berman & John Lonsdale, Unhappy Valley. Conflict in Kenya and Africa. 1. State and Class. 2. Violence and ethnicity, James Currey, London – Heinemann Kenya, Nairobi – Ohio University Press, Athens, 1992, 223+vii p. et 504+viii p. (Eastern African Studies). See too John Lonsdale, ‘Ethnicité morale et tribalisme politique,’ Politique Africaine (Karthala, Paris), no. 61, March 1996, pp. 98–115. 4 michel cahen

“broadcasting” a nation-statist ideology that this state is in fact a nation- state? We must not forget that a nation-state is the state of a nation first. Ideology and political discourse exist and have effects, including those on identity, and these effects must be studied. But there is nothing fated about this process, and a political discourse, an ideology, may provoke support and “membership” as well as reluctance and open resistance. That means a nation-statist ideology will have effects, but theses effects may not lead to the formation of a nation; ideology alone cannot be sufficient to mould the very nature of the state of the nation. Conversly, the crisis of the state in Africa today is very much the crisis of the non- existence of nation-state, the crisis of “land-states” without a nation “fill- ing” the whole area of the former colonial space, without a “sense of nationhood” throughout the land – in short, the crisis of the “gatekeeper state” as Frederick Cooper has called it.8 The problem stems not just from the colonial “artificial borders” which mark unbelievable territories, but also, and above all, from the very nature of colonisation, from which the lack of historicity of the borders and territories derives.9 From this point of view, it is obvious that the issues of the nation are not the same in Latin America and in sub-Saharan Africa. But is this the case just because Latin-American states gained independence 150 years before Africa? This is one factor, but probably not the most important one.

At Least Three Categories of Colonisation

Former colonies have experienced three kinds of evolution in relation to their former “mother countries:” (1) independence without decolonisation, (2) decolonisation without independence, and (3) independence with decolonisation.

8 Frederick Cooper, Africa Since 1940: The Past of the Present, Cambridge University Press, Cambridge, 2002, 230 p. 9 In any case, focusing the discussion on the “artificiality” of the borders is not very relevant, since all borders are essentially “artificial”. The French Revolution tried to defend the concept of “natural borders” (seas, rivers, mountains, etc.), because, besides the Enlightenment ideology, it was in its own interest: the Pyrenees as a border with Spain preserved the French ownership of Northern Catalonia, the Alps allowed for the annexation of Savoy from the Kingdom of Piemont, and the river Rhine allowed for the annexation of the area which now forms Belgium. The issue is whether a border is social, human, that is to say historically produced. For research on African borders, the African Borderlands Research Network () is an invaluable resource. anticolonialism & nationalism 5

Let Us Stop Confusing Independence and Decolonization The first category is most widespread in the former Spanish and Portuguese Americas. Independence was a political break, but in no way an anticolonial movement. What defines a colony is not primarily the fact that the subject territory is part of an imperial state. The Ancient Greek colonies were independent from their former mother cities, yet they nevertheless retained a clear identity as colonies in Sicily, Southern Italy and the Mediterranean Sea and Black Sea. What defines a colony is the nature of the society produced by a conquest – not all conquests produce colonies, but there is no colony without a conquest. Indeed, in South America the independence revolt was not only led by the settler class, but it also included (more or less) the colonial society as a whole, with all its social milieus, including black slaves: but did independence put an end to the appropriation of Indian land and destruction of their political structures? In such cases the new country is independent but it remains a colony, only now it is a self-defined colony. This is what Abel Quijano calls the coloniality of power,10 which seems quite different from postcolonial theory (even if this author is frequently associated with that approach). This process was most extreme in Portuguese America as it was the metropolitan state which decided to stay in its colony, finding refuge during the Napoleonic wars, but afterwards refusing to return to its mother country. The “grito de Ipiranga” (1822)11 never meant for the state to stop being Portuguese, but that from then on it would build the Império not from Lisbon but from Rio. It was Portugal’s colony, but it became a

10 Aníbal Quijano, ‘Colonialidad y modernidad/racionalidad’, Péru Indígena (Lima), vol. 13, no. 529, 1992; same title, in Heraclio Bonilla (ed.), Los Conquistados. 1492 y la Población Indígena de las Américas, Tercer Mundo Editores, Bogotá, 1992, pp. 437–447; ‘Colonialité du pouvoir et démocratie en Amérique latine,’ in A. Alvarez Béjar et alii, Amérique latine, démocratie et exclusion, L’Harmattan, Paris, 1994, pp. 93–101 (Futur Antérieur); ‘Colonialidad del poder y classificacion social,’ in ‘Festchrift for Immanuel Wallerstein/Part I’ Journal of World-Systems Research, vol. 6, no. 2, Special Issue, 2000, pp. 342–386; ‘Colonialidad del poder, eurocentrismo y América Latina,’ in E. Lander (ed.), La colonialidad del saber, op. cit.; ‘Colonialidad del Poder, Globalización y Democracia,’ Trayectorias (Monterrey, Nuevo León, Mexique), vol. 4, no. 7–8, 2001–2002, 23 p.; ‘¿ Qué tal raza ?,’ Rio Abierto (Lima), 11, 2004 ; ‘Coloniality of Power, Eurocentrism, and Social Classifiction,’ in Mabel Moraña, Enrique Dussel & Cazrlos A. Jáuregui (eds), Coloniality at Large. Latin America and the Postcolonial Debate, Durham, London, Duke University Press, 2008, 630 p.: 181–224 ; ‘La revanche des Indiens ? Le contexte global de l’Amérique latine. Réponse aux questions de Yann Moulier-Boutang,’ Multitudes (Paris), no. 35, 2009, pp. 97–102. 11 On the banks of the River Ipiranga, the Portuguese regent Pedro (later Emperor Pedro I) issued the Grito do Ipiranga (« Independence or Death ! »), the declaration of Brazil’s independence from Portugal, on September 7, 1822. 6 michel cahen self-defined Portuguese colony. This is why the first name of the state was Império brasílico, not brasileiro; they were simply building the same Empire from a new location. Indeed, the moment we might consider to be the birth of Brazil in 1889 (with the end of the Empire and the birth of First Republic) did not mean that the country was no longer a colony: since the national identity had slowly been transferred, but not the social colonial formation, it was no longer a self-defined Portuguese colony, it had become a self-defined Brazilian colony. It must be stressed that this idea applies not only to the ruling class, but to the whole of colonial society. For example, while one may concede that Bolivarian or Brazilian independences took on a strong elitist character, the Cuban revolution of 1898–1901 was a popular revolt which was by no means restricted to an elite uprising. But it was a colonial society (not just the colonial ruling class) which took power, expelling the Spanish state, within a framework shaped entirely by colonial history, since the Indian population had disappeared.12 So when Brazilian colleagues say “We were colonised by Portugal,” I answer, “Begging your pardon, no, you were the colonisers for Portugal,” because a large majority of today’s Brazilian population comes from colonial society (and I include Black and mixed-race Brazilians, as although slaves were the most exploited social class, it was a class within the colonial society, not within colonised societies, which were limited to the indigenous Indians). In Latin America, to a greater (Argentina, Chile, Brazil, most of Mexico) or lesser (Paraguay, Peru, Bolivia, Ecuador) extent, it was the coloniser who fashioned society. In sub-Saharan Africa, the situation is completely different and belongs to the third category (independence with decolonisation). African societies were defeated and conquered, exploited and humiliated, partially acculturated and socially “reshaped,” but these African societies continued to exist. The coloniser did not bear society, and only represents a small minority of the total population. The domestic mode of production (referring to Claude Meillassoux’s concept)13 was not destroyed, but was subjected to capitalism. The new states of today may be perfectly neo- colonial, but they are no longer colonies, meaning that true decolonisation occurred.

12 I do not discuss here the fact that Cuba was almost recolonized (by occupation) by the United States of America up until 1934. At the moment of the uprising, the American intervention did not change the popular and colonial nature of the movement. 13 Claude Meillassoux, Maidens, Meal and Money: Capitalism and the Domestic Community, Cambridge University Press, Cambridge, 1981, 212 p. [translated from the French ed., 1975]. anticolonialism & nationalism 7

Let me touch briefly on the second category, that of decolonisation without independence. With respect to Portugal, we have two examples: Goa and Macao, which were decolonised by integration into another country. Furthermore, General Costa Gomes (the second Portuguese president after the Carnation Revolution of April 1974) thought the same would occur with East Timor – which actually happened between 1975 and 1999, by integration into Indonesia. Another significant case in this category is that of the French “Départements d’Outre-Mer,” where there is the complete institutional assimilation of the territories into the French Republic. Going back to the first and third categories, what are the consequences for a nation? They are enormous on at least two levels.

Colonisation Categories and Nations

Firstly, imagining the nation is a completely different process when the coloniser was the producer of the society or when the coloniser “merely” invaded an existing foreign society. Obviously the borders which shape the country are just as artificial in both cases. But in the former (independence without decolonisation), the colonisers are the ones who “shape” the territory, their territory, the society, their society. They are no longer foreigners. They went and settled there as Spanish or Portuguese people, with an already formed identity (it is debatable whether the identification process was the same at the time of the Kings or at the time of national revolutions: but they settled with a kingdom/national identification that had already been formed, as a part of a formerly imagined community, and the shift was much more political and economic than national). This process is one of the duplication of the former nation, accompanied by a slow process of differentiation. In the second scenario, African societies were not destroyed (at least, not as in the Americas), and the coloniser did not build a settler society (with the partial exception of Algeria, South Africa and, to a lesser extent, Kenya, Southern Rhodesia, Angola and Mozambique).14 The settlers remained foreigners within African societies. But these colonized African societies were placed in, and divided by, territories decided by the

14 Even in these cases of settler colonization in Africa, the African societies have survived and were not exterminated (except in a few cases), or completely marginalized as in North and South America. The settler societies have remained more or less tiny minorities. 8 michel cahen coloniser outside the context of a colonial society (like in Latin America). Spaces like “Angola” and “Mozambique” (as well as “Gambia” or “Senegal”) have only a colonial relevance, not a colonial and national one as in the Americas. Why would the Ndaus of Núcleo Negrófilo de Manica e Sofala, in the fifties, or the Macondes of Mueda in 1960, fight for “Mozambique”, something almost completely foreign to them? They wanted to free the land, their land, and had no reason to accept the piece of land the coloniser had outlined on the map.15 Secondly, there is the question of the mode of production. When settlers and other segments of the colonial population came to America with their slaves, they brought with them a way of production which was already part of the first capitalist world-system to which their former nations belonged. I would not deny the existence of differences: for example, in the Portuguese world, the capitanias and sesmarias in Brazil were not exactly the same as those from Portugal at the time of the Reconquista,16 or similar to the prazos of Mozambique,17 and although at that time there were many Black slaves in Portugal itself,18 their numbers would never reach the same level as in the Americas. But on the grand scale of the Empire, it was all part of the same slavery and long distance trade-based economy; there was systemic unity between colonies and mother land. In Africa, the landscape became completely different after the Berlin Conference when effective conquest of the land became necessary in order to legitimise colonial sovereignty. Traditional social relationships were not totally destroyed but were integrated within the new economy of imperialism. This is what Marxists called the “articulation of modes of

15 On these cases of African anticolonial nationalisms, see Michel Cahen, ‘The Mueda Case and Maconde Political Ethnicity. Some notes on a work in progress,’ Africana Studia (Oporto), no. 2, 1999, pp. 29–46; M. Cahen, ‘L’anticolonialisme identitaire: conscience ethnique et mobilisation anti-portugaise au Mozambique (1930–1965),’ in Colette Dubois, Marc Michel & Pierre Soumille (eds), Frontières plurielles, Frontières conflictuelles en Afrique subsaharienne, L’Harmattan, Paris, 2000, 462 p., pp. 319–333. 16 António Vasconcelos Saldanha, As Capitanias. O Regime Senhorial na Expansão Ultramarina Portuguesa, Centro de estudos de História do Atlântico, Funchal, 1991, 343 p. 17 Allen F. Isaacman, Mozambique. The Africanization of a European institution: the Zambesi prazos, 1750–1902, University of Wisconsin Press, Madison, 1972, xviii-260 p. ; Malyn D.D. Newitt, Portuguese Settlement on the Zambesi. Exploration, Land tenure and Colonial Rule in East Africa, Longman, London, 1973, 434 p. 18 José Ramos Tinhorão, Os Negros em Portugal. Uma presença silenciosa, Caminho, Lisbon, 1988, 460 p. (Universitária, 31); Didier Lahon, O negro no coração do Império: uma memória a resgatar, séc. xv-xix, Secretariado Coordenador dos Programas de Educação Multicultural, Lisbon, 1999, 103 p. ; Jorge Fonseca, Escravos no Sul de Portugal: Séculos xvi- xvii, Vulgata, Lisbon, 2002, 261 p. (Rota do escravo. Estudos, 2) ; Isabel Castro Henriques, A Herança Africana em Portugal, CTT Correios de Portugal, Lisbon, 2009, 238 p. anticolonialism & nationalism 9 production,” “domestic mode of production” being globally used for the greater profit of colonial capitalism.19 In Angola and Mozambique, this was evident in the system of forced labour: for six months of the year (Mozambique), or for one year out of every two (Angola), male indígenas [natives] had to work for the coloniser because their traditional and non- monetised economic activity was not recognised as work (therefore they were considered to be lazy vagrants). But this forced labour was not a modern process of proletarianisation: African workers were compelled to return home, and in order to oblige them to do so, part of their already low wages were only paid after they had returned home rather than at the end of the contract, at the workplace. Why? Officially, it was to allow them to “rest”. In reality, it was to allow them to help their wives restore domestic cultures, because these cultures were essential to colonial capitalism, enabling the payment of Africans below the cost of their social reproduction. Proletarianisation would have implied a sharp increase in the salaries, allowing Africans to live only from them. Highly gendered forced labour system was far more profitable for modern capitalism. Turning to Latin American and Sub-Saharan African societies, we can now easily understand that the state in the former will correspond immediately to the kind of economy that the colonial society produced; and when it was necessary to adapt something, the mother state could be expelled to create a self-defined colonial state. This state is essential to the functioning of its economy, and for the functioning of its society. Even if the colonial society is strongly segregated into various social classes and milieus, it remains one society, not several, and the real other society, the colonized Indian society, is comprehensively marginalised. The state, and indeed the country itself, are defined by the colonial society. It allows for a strong process of identification among the majority of the population, and this majority may bring with it the majority of the rest: national relevance is therefore hegemonic, even if there are always some margins. In Sub-Saharan Africa, the domestic mode of production survived throughout the 20th century, though combined with capitalism. It is now disappearing through the acceleration of the processes of urbanisation. There has been no unification of the market, no unification of space, and many different migration trends. In Angola and Mozambique railways are

19 Bruce Berman & John Lonsdale, Unhappy Valley … 1. State and Class, op. cit.; Claude Meillassoux, Maidens, Meal and Money …, op. cit. ; Pierre-Philippe Rey, Colonialisme, néo- colonialisme et transition au capitalisme, Maspéro, Paris, 1971, 527 p. (Économie et socialisme, 15). 10 michel cahen east-west, not north-south, and link the two countries to the British territories and economies of the interior only. What is this “space”? Any independent state built on the basis of such a space will face huge problems to be seen and felt by heterogeneous peoples as representative. It does not represent the peasant economy because that is not its function. It had to accept the relevance of its borders without any consultation of the nations it embraced in its colonially-defined land, and even if these borders cut across long-established African nations. Consequently, it cannot perceive the new nation as the necessary result of a very slow, multi-century process of merging from the former nations without destroying them, because it needs to achieve contemporary legitimacy by building a new nation quickly; it needs to become a colonial-based physical nation immediately. This is why the nation-statist ideology is always, in this context, a paradigm of authoritarian modernisation: this modernisation may become effective through unbridled capitalism or so-called “Marxism- Leninism”, but, as a model of development, it always contains dense, high-technology islands that are quite unable to breathe dynamism into the surrounding peasant economy. This by no means prevents ethno- clientelist behaviour by those in power, but the state ideology has to be that of an already existing nation, with great consequences for day-to-day policies that cannot be based on the cultural and social demands of its populations. If this state, despite all of the weaknesses that have been pointed out, can guarantee social, economical and cultural progress for the people of the first nations (the so-called “ethnic groups”) integrated within the colonial space – even with a paradigm of authoritarian modernisation, a kind of enlightened despotism as it were – this guarantee of progress could create an identification process. For example, it was better for the Germanic people of Alsace to be French rather than Prussian because, in the context of post-revolutionary France, it was socially better to be citizens of the Latin French state than subjects of the Prussian king. Therefore, despite being ethnically Germanic, Alsatians preferred to become French. The matter of social progress is a fundamental one in the processes of identification. But is the state in the capitalist ultra-periphery able to guarantee this progress? I think not. For an identification process to work within a still virtual community of human beings, it requires the pre-existence of more or less similar social relationships among a population group. The pre-existence of a certain kind of social relationship in the Basque Country came long before the feeling of being ‘Basque’. If a state is able to modify and unify these anticolonialism & nationalism 11 social relationships in a manner to which people respond positively, it will lead to a process of identification. But it is not nationalism per se which creates the nation. Ernest Gellner argued, in his most famous works,20 in favour of such a thesis, but without, in my opinion, resolving the problem: if nationalism, per se, creates the nation, who creates nationalism? And why does this nationalism run in some sections of the elite and the population, and not in others? In fact, nation and nationalism are the same dialectical process which one may call the political expression of a social movement of identities, closely linked with long term trajectories of social life. Nations are not a programme,21 but a long historical process.

What Peculiarities Does the Portuguese Empire Present?

In this global context, what are the peculiarities of the Portuguese Empire? Since the publication of Gervase Clarence-Smith’s wonderful book The Third Portuguese Empire,22 I consider the debate about the “economic” or “uneconomic” nature of Portuguese imperialism to be over.23 It is not necessary to create a special category for Portuguese imperialism. It was an economic imperialism as a whole, in which the links with Brazil were far more profitable than the ones with Africa up until the 1950s, although obviously its political and “mental” aspects should never be underestimated. Portuguese imperialism in Africa is part of the diversified family of European imperialisms. Therefore, what is specific and relevant to our discussion? Far from a culturalist approach, and far from insisting on the unique nature of the motherland of Luís de Camões, I will focus on three specific characteristics, which are nuances rather than huge differences:

20 Ernest Gellner, Nations and nationalism, Blackwell, Malden (Mass.), 2006, liii-152 p. (New perspectives on the past) [1st ed. 1983]; Ernest Gellner, Nationalism, Phoenix, London, 1998, x-114 p. 21 Unlike what seems to be thinking Eric J. Hobsbawm, Nations and Nationalism since 1780: Programme, Myth, Reality, Cambridge University Press, Cambridge, 1992 [2nd ed.], 214 p. 22 Gervase Clarence-Smith, The Third Portuguese Empire 1825–1975. A Study in Economic Imperialism, Manchester University Press, Manchester, 1985, X-246 p.; see my comments: ‘Lénine, l’impérialisme portugais, Gervase Clarence-Smith,’ Cahiers d’études africaines (Paris, EHESS), no. 107–108, 1987, pp. 435–442. 23 Gervase Clarence-Smith was explicitly answering to the book by Richard James Hammond, Portugal and Africa, 1815–1910: a study in uneconomic imperialism Stanford University Press, Stanford (Ca.), 1966, xv+384 p. 12 michel cahen

Portugal, the First to Go First, we have to remember that Portugal was the first European state to travel all the way around Africa. This did not lead to “five centuries of colonisation,” since the effective conquest of 95 % of the colonial area occurred in the ten years or so following the Berlin Congress of 1885. But in tiny areas, small old colonial elite social milieus emerged, often mixed- race and comprising Portuguese, African, Arabic, Indian, Shirazi and Chinese people. Although very narrow demographically, the old Creole social milieus were very important in the genesis of the anticolonial movement, even if the majority of these milieus strongly identified themselves with Portugal. The Portuguese Empire of the 20th century is not the only such case: outside of it, but also from luso-brazilian history, we can observe examples in Lagos (Nigeria), Benin, the Agudas;24 we have famous cases in Liberia and Sierra Leone, although these Creolities are “imported” (the kriol elite came from the USA or England and were not locally produced as was the case in the Portuguese empire); we have some cases on the coast of Senegal, where there are some small areas of former colonisation but officially and definitely recognized by French Third Republic only in 1879 (the ‘Quatre Communes’). But it is within the Portuguese Empire that old Creolities have been most important.25 Thus, when Portugal actually conquered colonies that it believed it had owned for a long time, what occurred was not a turning point between “pre-colonial” and colonial stages, but the turning point between ages of colonisation. Even if there was no permanent Portuguese presence in these lands, there were former trade roads of lançados (Senegambia) or pombeiros (Angola), a former commercial presence and, in Mozambique, the famous Prazos

24 The Agudas (from the name of [S. João Baptista de] Ajuda, or Ouiddah) are a creole milieu stemming from slave traders and slaves from the Coast of Guinea, settled in Brazil and returning to Africa during the last period and after the end of the trade. Andrzej Krasnowolski, Les Afro-Brésiliens dans les processus de changement de la Côte des Esclaves, Wroclaw, Zaklad Narodowy im. Ossolinskich [Polish Science Academy], 1987, 197 p., [16] p. of plates; Kadya Tall et al., 1995, Le comptoir de Ouidah, une ville africaine singulière, Karthala, Paris, 191 p. ; Olabiyi Babalola Yai, ‘Les “Aguda” (Afro-Brésiliens) du Golfe du Bénin. Identité, apports, idéologie: essai de réinterprétation,’ Lusotopie (Paris, Karthala), vol. 3, 1997, pp. 275–284; Milton Guran, Agudás: os “brasileiros” do Benim, Editora Nova Fronteira/Editora Gama Filho, Rio de Janeiro, 2000, xv-290 p. ; Nancy Priscilla Naro, Roger Sansi-Roca & David Treece, Cultures of the lusophone Black Atlantic, Palgrave MacMillan, New York, 2007, ix-256 p. 25 I call Creolities those social milieus produced by, or at the close periphery of, the imperial state apparatus. Therefore it is not a racial category, but a socio-cultural one. Creoles may be white, mixed-race or completely black, but obviously not native (in the English meaning of Native, that is to say traditional, indigenous). anticolonialism & nationalism 13 da Coroa.26 But the geopolitics of the first age of colonisation were not the same as those of the new capitalist colonialism. Again, it was in Mozambique that the difference was most prominent: while the former colonisation was centred between the Zambezi river and the Querimba islands, with Moçambique island as the colony’s capital, the new colonialism migrated further to the South, in order to connect the new colonial economy to South-African capitalism. This change in domestic politics had wide-ranging social effects, with the marginalisation of most of the former elites increasing regional heterogeneity in the country, and heterogeneity in the relationship towards the modern state, a trend which has continued right up to the present day. It is not surprising that segments of these marginalised elites were often in favour of Renamo.27 In Angola, the effects of the two ages of colonisation were different, because Luanda remained the capital city, which explained the very important standing of Creole families in the genesis of the MPLA.28 But there was a comparable marginalisation of the south-western part of Angola, in the Benguela and Namibe areas, after the decline of links with Brazil. In São Tomé e Príncipe, the former elite of the Filhos da Terra (Sons of the land), or Forros, which Portugal never succeeded in subjecting to forced labour, and whose refusal to work in agriculture became a factor of identity, managed to control the independence processes and nationalise the roças (cocoa plantations) in order to prevent their appropriation by the workers who actually worked on them, that is to say Mozambicans, Angolans or Cape Verdeans. These are three different cases, but nevertheless three cases where understanding the weight of former elites stemming from the first age of colonisation is crucial to understanding the way the present day nation sees itself.

No Social Space for an African Elite Secondly, even if Portuguese imperialism was economic on the whole, it was also certainly very bureaucratic, the colonialism of poor capitalism.

26 A. Isaacman, op. cit. and M. Newitt, op. cit. 27 Renamo, Resistência Nacional de Moçambique, Mozambique National Resistence, a guerrilla movement backed by apartheid South Africa in order to fight “Marxist Frelimo”. Frelimo, Frente de libertação de Moçambique, Mozambique Liberation Front, created in Dar-es-Salaam in 1962, which gained independence for the country in June 1975 and is still at power today. 28 MPLA, Movimento Popular de Libertação de Angola, Angolan People’s Liberation Movement, created in 1960 in Tunis (oficially in 1956 within Angola), which gained independence in November 1975 and is still at power. 14 michel cahen

I am not speaking about bureaucracy within the administration, for example, but about social milieus which can be characterised by their bureaucratic functions. Portuguese settlers were first of all civil servants, trade employees, skilled workers in state-owned ports and railways, soldiers or priests. This hypertrophy of the tertiary sector led to a very small part of the colonial population having close links with commodity production – either as bourgeois, or white workers as in South Africa. The myth about the “lusitanian peasants within the Tropics” was just that; more than 80% of the colonial population was crammed into the main cities, and the most rural segments of the colonial population were Indian, Chinese and mixed-race people, not white people. In this context, there was almost no social space for the emergence of a new urban (as well as rural) African elite, because white people occupied the whole sphere of wealth creation. The tiny new African elite occupied the same jobs as lower-level whites, only they were socially bureaucratic jobs. This elite had very few links with the rural economy and traditional peasant hierarchy; in short, it was entirely the product of 20th century Portuguese colonialism. Once again, this is more evident in Mozambique than in Angola. In Angola, as the late Christine Messiant demonstrated, it was more a slow process of differentiation among the assimilados,29 which may in part explain the Coup attempts by Nito Alves in May 1977. In any case, we never or rarely see an African bourgeoisie, or African trading petty bourgeoisie appearing, but a social milieu of tertiary-sector employees entirely produced by the modern colonial stage. I do think that this constitutes a sufficient explanation for the kind of Marxism of the MPLA and Frelimo, or “revolutionary democracy” that the PAIGC and MLSTP30 advocated as a possible structuring discourse that could be understood within the elite, expressing its desire for subaltern westernisation. The third characteristic, and this time not a structural but a contextual one, was the need for an armed struggle. Portugal was not alone in this case, since the United Kingdom experienced the Mau-Mau revolt in Kenya, and France the war in Algeria, not to mention Dutch Indonesia and French Indochina. But with Portugal, the whole of the decolonising

29 C. Messiant, 1961. L’Angola colonial, histoire et société. les prémisses du mouvement nationaliste, P. Schlettwein Publishing, Basel, 2006, 444 p. 30 PAIGC, Partido Africano da Independência da Guiné e Cabo Verde, African Party for the Independence of Guinea and Cape Verde, founded in 1959 by Amílcar Cabral, launched a successful guerilla war in Guinea, obtaining independence for the two countries in 1974 and 1975 ; MLSTP, Movimento de Libertação de São Tomé e Príncipe, Movement for the Liberation of São Tomé and Príncipe, in power in the archipelago from 1975 to 1991. anticolonialism & nationalism 15 process had to be a violent one. Obviously, the intensity and duration of the war served to radicalise the situation. This brings us back to the links between nationalism and Marxism.

Nationalism and Marxism

As we have seen, not only was the mindset of this very small elite shaped by its origins within or close to the state imperial apparatus, but Marxism also encouraged a certain kind of functionalist view of the nation. More specifically, this was the kind of Marxism which seemed relevant to the small elite who had studied in France, Switzerland, Portugal or Eastern European countries. This Stalinised Marxism advocated a one-party system with the state as the main actor in the economy, homogeneity of the nation with only one language, corporatist labour unions and authoritarian management of workers and peasants; none of which was foreign to the small, new, colonially-shaped African elite, educated at the time of Salazarist Portugal. If there was a will for political change, the social model remained “very Portuguese.” But this Marxism also offered an explanation for the nature of the revolution to be achieved, which could solve the problem of the illegitimacy of a nation conceived within a merely colonial space. In the Marxist tradition, bourgeois revolution is closely linked with national revolution. Indeed, Marx analysed capitalism as the crucible of the modern nations through the unification of a market. We could have a long discussion as to whether capitalism created this unified space, or if it slotted into a space that had already been shaped by the late feudal state.31 But nobody will deny that market unification had major consequences for the mindset of a more modern state and nation. In France, it has often been said that it was the Revolution that created the nation, and in Germany the nation that created the state. There is no space here to enter in the criticisms of both theses, but in both cases,32 there is no doubt that the bourgeoisie played an important role in the shaping of the nation. To varying extents, the bourgeois revolution was also a national revolution at the same time.

31 I discuss this issue in my article ‘Lusitanidade, “lusofonidade” e modernidade. Uma exploração nos conceitos de identidade e de nação,’ Episteme. Revista interdisciplinar da Universitade técnica de Lisboa (UTL Lisbon), 2004, vol. 5, no. 13–44, pp. 123–139. 32 See my book Ethnicité politique. Pour une lecture réaliste de l’identité, L’Harmattan, Paris, 1994, 176 p. 16 michel cahen

But we then have two quick shifts amongst many Marxists epigons: first, the bourgeois revolution was no longer occurring only at the same time as a national revolution, the bourgeois revolution was the national revolution and the national revolution was the bourgeois revolution. Secondly, what occurred (or seemed to occur) in Europe became universal: everywhere, the national revolution was the stage of the bourgeois revolution. Therefore, for many Marxists, the decolonising process in Africa was a late stage of the bourgeois revolution, after the end of “feudalism,” even if this feudalism was subjected to imperialist capitalism. But there was a problem: the Soviet Union could not tell its African friends they were reaching the stage of the bourgeois revolution. Soviet theoreticians therefore applied to Africa and some Asian countries an actualized version of the thesis of “non-capitalist development” stemming from the Sixth Congress of Comintern (1928),33 that is to say neither capitalist nor socialist. And the more radical experiments were qualified as National Democratic Revolutions (NDR), or, with a more Maoist accent, People’s National Democratic Revolutions (PNDR). Even if it was not acceptable for the anticolonial fighters to be characterised as agents of a bourgeois national revolution, this global Stalino-Marxist paradigm stressed the idea that the anticolonial revolution was a national revolution per se, either by liberating the nation or by creating a new nation. Creating the state was creating the nation: the problem of the modern legitimisation of a state rooted in a colonial space was thus solved. This was not just the application of theoretical cosmetics to justify things to the outside world. It was primarily an explanation for the narrow milieu of the radical elite itself, allowing them not to question the colonial relevance of the territory, not to have to recognise the relevance of other nations (the first nations, generally spoken of as “ethnic groups”), or of other legitimacies produced outside the imperial state apparatus which was familiar to the new elite. This was an ideology adapted to a habitus rather than a political discourse, even if it was also a political discourse denouncing all kinds of “separatism,” “tribalism” and “obscurantism”… What we have to remember here is that this kind of Marxism, or the influence of the kind of Marxism that captured or influenced the elite,

33 Although it does not cover all Marxist criticisms of the soviet concept of “Non- capitalist way of development,” the article by Hooshang Amirahmadi (‘The Non-Capitalist Way of Development,’ Review of Radical Political Economics, 1987, no. 19, pp. 22–46) may be useful. anticolonialism & nationalism 17 increased confusion within the state and the nation, and within the nation and the party, the crucible of the nation. But it was not the “fault” of Marxism if there was this confusion: it was because the elite needed this confusion that this kind of Marxism was an operational tool. It was not because they were taken by this Marxism that there was a one-party state; it was precisely because the elite wanted a one-party state to build a modern and homogeneous nation quickly that it chose to use that kind of Marxism. Marxism was thus, in a certain context of bureaucratic colonisation and the need for armed struggle, the way the elite found to express its desire for a nation. It also explains why it was, at the next step, very easy to abandon this Marxism, which was a tool and not an identity.

Creating the Nation through Armed Struggle? But did armed struggle not have “nationalising” effects, merging fighters of several ethnic origins? Yes and no. Firstly, the same could be said for the colonial army which recruited soldiers from all over the colony during the Africanisation process of the colonial war: the idea that the war spread a sense of nationhood by way of the draft could be applied to Guinea or Angola, as well to as Portugal. Secondly, it takes more than a ten year period for a sense of national identity to be created; nation formation is a historical process of the crystallisation of an identity, rooted in the pre- existence of common social relationships. The will for a new State, for a “government of our colour”, may crystallise within ten years, but will not produce a national identity per se. Thirdly, if the war of liberation had unifying effects, it also had dividing ones: the first age of colonisation of African society was not a peaceful one, and it is unsurprising that one could find African people who preferred European power to an African power. Why, for example, would the Bitonga people of Inhambane see Gungunhane as an anticolonial hero, when he was a resolutely predatory king whose dynasty had arrived after the Portuguese?34 How could the Bacongos, during the great uprising of 1961 in Angola, see Ovimbundu people as their allies, when the latter were working in coffee plantations for the settlers who had seized their land one generation before?35 And so on.

34 Malyn D.D. Newitt, A History of Mozambique, Indiana University Press, Bloomington, 1995, xxii-679 p. 35 John A. Marcum, The Angolan revolution. Vol. 1. The anatomy of an explosion (1950– 1962), MIT Press, Cambridge (Mass), London, 1969, xiv-380 p. 18 michel cahen

If we want to be precise in the analysis of historical processes, the war of liberation expresses the desire and need for a new government of the territory, and it is easily understandable that this should mean a new state covering the colonial area. But that is all! The same is not true, for example, of Poland divided between Russian, Prussian and Austrian empires during the 18th century. Poland was already a nation which, in spite of being invaded, continued to exist under foreign rule, with its nobility, Catholic Church and Jewish Yiddish villages. Polish nationalism expressed the will to restore sovereignty politically: nationalism was the expression of a nation, nationalism was produced by the social movement of the nation. In the cases of the PAIGC, MPLA or Frelimo, the wish to expel the colonisers, to have new governments and ultimately to have new states, was made synonymous with new nations. But the desire for a nation was not produced by the social movement, it was (and has remained) a project, it was proclaimed, it stemmed from the political elite and was imposed to the social movement to deny the first nations’ relevance and promote the homogenised New Man.

Shall We Call It Nationalism?

Shall we call this phenomenon nationalism? The term nationalism has traditionally been used to express two broad categories of political expression of the nation: On the one hand, it can express the resistance of a nation which is submitted to a political context which is not perceived as being positive by all or part of its people: it is the nationalism of, or within, oppressed nations. I have already mentioned occupied Poland, but other examples include Arab nationalism (though not necessarily limited to a specific post-colonial country), Kurdistan, the Basque Country, Ireland, Tibet, Chechnya, and so forth. On the other hand, it may express the nationalism of oppressive nation- states: Nazism, pan-Sinism, as well as French or Portuguese colonialism which are also a form of nationalism since they attempted to impose their nation (not only their domination but also their identity) on other peoples.36

36 Michael Löwy, Patries ou planète ? Nationalismes et internationalismes, de Marx à nos jours, Éditions Page Deux, Lausanne, 1997, 158 p. anticolonialism & nationalism 19

It is obviously unsatisfactory to use the same concept for two such different political expressions. But they still have a point in common: they are the product of an existing nation or of the nation-state of that nation. They are the product of the social movement of the nation, or of the capacity of the nation-state to integrate the social movement of the nation. If we now turn to Angola, Mozambique or Guinea, we see strongly socially-rooted movements against colonialism. Opposition to the coloniser produces a semblance of identity – the community of those who want to expel the coloniser. Is that enough to give root to a national feeling, which is to say a feeling of identity which is able to overcome local identities and be perceived as being more important? Identities are always mixed; a person may be a Kongo and an Angolan person, a Maconde and a Mozambican, a Balante and a Guinean, a Basque and a French person. However, a Scottish person may be British, but will never be English, a Catalan person may be Spanish, but never Castilian. This points to the problem of the “nation-statist state” which is identified only with one of the ethnicities of the state and places a brake on identification processes. If the Mozambican or Angolan states are identified by large sections of the population as a mainly Changane or Mbundu/Creole state, it will slow down the identification process with this state. The economic dimension of that includes populations feeling that all the wealth is given only to one ethnic group. There is then little sense in speaking about a nation if the people’s principal loyalty is not to that nation. For instance, in the time of the Soviet Union, it could be said that there was a kind of “Soviet identity”, but the Soviet Union was never defined as a nation by its leaders: they spoke about the “Motherland of Socialism”, motherland but not nation. The nations remained Russian, Ukrainian, Armenian, without a nation of nations in the same way as the British identity is a pan-identity of English, Scottish, Irish and Welsh identities. “Sovietness” was a kind of “lesser identity” that could be an accepted identity but was never embraced as a national identity. It could perhaps have been transformed into a feeling of a nation in five centuries, but History decided otherwise. If a local identity within a country37 is far more important in day-to-day life than the so-called nation of a nation-statist state,

37 It is worth noting that, if one does not oppose nationhood and ethnicity, a local identity may also be analyzed as national (why could a nation not be very small, given that Luxembourg is more than five times smaller than Mueda district in Mozambique?). It is not a question of size, but of history. 20 michel cahen this state is not a nation-state. That does not mean that this state is weak, or collapsed or illegitimate – I said at the beginning of the paper that the nation was not an historic necessity for a State to be stable and legitimate. But this state will be a “no nation-state” and, to be stable, will have to invent politics of greater respect for all local identities – national or not. This kind of state, a common home for different peoples, is yet to be invented.

The State Does not Produce the Nation. But if It Did, Which One Would It be? Can a state produce a nation ex-nihilo? As I have stated elsewhere several times, the answer is “no”,38 but that depends on the stance we take on national consciousness. If we no longer accept the confusion between state, republic and nation – a very French or Portuguese confusion – it is clear that we are speaking about imagined realities;39 that is, something which actually exists but on a subjective level. No “objective definition” of the nation, as Stalin attempted,40 can work. Who are the members of a nation? Only those who feel they are members.41 What is a nation ? Only the community defined by its members. Who is French, Jew, Zulu ? Those who feel they are. This process of imagination stems from a long historical process of the crystallization of identity, and cannot be reduced to the relevant role of an elite.42 It is a dialectic process within and between all the social milieus of a society which perceive that there are enough shared social relationships to fell a sense of community. Still, we should not deny that the very existence of a state, even the neo-colonial state, or of

38 M. Cahen, ‘Mozambique: histoire géopolitique d’un pays sans nation,’ Lusotopie (L’Harmattan, Paris), vol. 1, no. 1–2, June 1994, pp. 213–266 ; Ethnicité politique, op. cit.; ‘Desigualtats, Etnicitats, Democratització: de la construcció de l’Estat sense nació a la nacionalizació del món,’ Studia Africana (Centre d’Estudis Africans, Barcelone), 8, March 1997, pp. 143–160 (french version: ‘De la construction de l’État sans nation à la nationalisation du monde,’ pp. 19–44, in M. Cahen, La nationalisation du monde. Europe, Afrique, l’identité dans la démocratie, L’Harmattan, Paris, 1999, 256 p.) ; ‘Lusitanidade,“lusofonidade” e modernidade …,’ op. cit. 39 Benedict Anderson, Imagined communities: reflections on the origin and spread of nationalism, Verso, London, 1983, 160 p. 40 Joseph Stalin, Marxism and the national question, Foreign Languages Publishing House, Moscow, 1945 [1913], 79 p. 41 Michel Cahen, ‘El potencial revolucionario de una categoria desechada: la etnia y las ciencias sociales aplicadas en África,’ in Albert Roca Álvarez (ed.), La Revolución pendiente. El cambio político en el África negra, Edicions de la Universitat de Lleida, Lleida, 2005, pp. 107–152. 42 That is the feeling one may get when reading B. Anderson, op. cit., and even more Eric J. E. Hobsbawm, Nations and Nationalism since 1780: Programme, Myth, Reality, Cambridge University Press, Cambridge [England], 1990. anticolonialism & nationalism 21 boundaries, even if they are “artificial,” induces processes of identity. The question is: to what extent? In censuses within Tito’s Yugoslavia, citizens had to indicate their identity: Serbian, Slovene, Croatian, Bosnian, and so on. But they could also refuse to pick one of these, simply identifying themselves instead as “Yugoslavian.” This is most interesting because we know that the existence of a Yugoslavian state for seventy years had “produced” Yugoslavians who no longer felt that they were members of their primary nations. In the last census held within this Yugoslavia, however (with Tito already dead, but several years before the civil war), less than 10 % of the citizens said that they were Yugoslavian. Seventy years of the Yugoslavian state had therefore produced Yugoslavians, and a Yugoslavian nation existed, but it was a minority in Yugoslavia.43 This example is useful for our examination of Portuguese-speaking Africa, and for Africa more broadly. I have no doubt about the existence of Mozambican or Angolan or Guinean nations – that is to say, national feelings delineating a community –, but the question is, what proportion of the population holds this sense of nationhood to be the primary element of their identity? Yugoslavia was a European state, able to promote social progress and therefore to encourage the identification processes. The situation is worse on the capitalist periphery. And in Portuguese- speaking Africa, particularly in Mozambique and the early post-independence history of Guinea (up to the 1979 coup) or Angola (up to the beginning of open clientelism and corruption in the middle of the eighties),44 the paradigm of authoritarian modernisation, closely linked with the nation-statist ideology, slowed down the identification process. When Frelimo imposed “communal villages”, or banned traditional chiefs or rain rituals, when it prohibited African funeral associations, when it organised literacy campaigns only in Portuguese, including for old people, when it nationalised all schools, and so on, it was not primarily engaged in a socialist process, but rather attempting to launch a “nationalising” process, designed to impose the transformation of all inhabitants into modern European-model Mozambicans. This Frelimo state was so strange for a large part of the inhabitants that it seemed entirely foreign.

43 Michel Cahen, Ethnicité politique …, op. cit. 44 Christine Messiant, L’Angola post-colonial. 1. Guerre et paix sans démocratisation. 2. Sociologie politique d’une oléocratie, preface by Georges Balandier, Karthala, Paris, 2008, 420 p. + 432 p. (« Les Afriques »). 22 michel cahen

But when this authoritarian policy does not bring social and cultural progress to these inhabitants, not only does it not produce the nation, it also provokes anti-statist reactions. The Renamo rebellion45 may also be seen in this way: a reaction to the modern authoritarian state. Mozambicans will all recognise the fact that Armando Guebuza is the President and none of them will have any difficulty in recognising that they are “Mozambican.” But as social scientists, we must understand the exact meaning of such a word. Recognising political legitimacy, or at least the acceptance of a state, of a President, of a master, is not synonymous with the existence of a sense of nationhood. They may all be “Mozambican,” but that does not prevent people in Mambone or Mossuril from speaking of the “nation” when they speak about Maputo: “Dear comrades, may I introduce you this comrade coming from the nation?”, as they say when they introduce a leader coming from Maputo. Likewise, when peasants or urban people in Northern provinces are angry at a visiting leader, they may say, “Let him come back to Mozambique!”,46 acknowledging that being a member of a country is not, in itself, indicative of a national feeling, but may be the consequence of authoritarian policies which fall far from respecting the demands of local populations. From this viewpoint, the “Marxist-Leninist” paradigm and the neoliberal philosophy are not so different: peasants are always to be “organised,” to be “modernised,” to be “integrated,” or to be “forgotten” as in Angola! They are a problem for the state. This project of the nation as opposed to the first nations, and as opposed to the original social relationships held by the peasant people, does not correspond to either category of nationalism that is laid out above. This project does not reflect the nation nor does it respect the long-term processes of identity crystallisation as they exist in History. It is always closely linked to a paradigm of authoritarian modernisation. It is striking to note that what has always remained, in spite of the many political turns of Frelimo or the MPLA, is the idea of nation-building: the ideology of the nation-building process is the key line of continuity in these parties and stalinized Marxism was only one stage in it.

45 Renamo was backed by South Rhodesia and apartheid South Africa, but was able to express, in a certain way, the deep crisis of the peasantry that Frelimo’s policy of authoritarian modernization had provoked. Michel Cahen, Les Bandits. Un historien au Mozambique, 1994, Publications du Centre culturel Calouste Gulbenkian, Paris, 2002, 354 p. 46 I am indebted to Joana Pereira Leite (Lisbon), for this last example. anticolonialism & nationalism 23

If we accept that this vision of the nation-building process characterizes a state as a nation-state, we make defining what nationalism is – already a complex range of political expression – even more difficult. It gives rise to confusion between the republic and the nation, and therefore the disregard for the former nations and all that is “not modern” from a modernist, statist point of view. In Mozambique’s Frelimo, this disregard began very early on, and is clear even in the way it has written its own history. This is a very func- tional story, in which three “regional” movements (meaning “not national”), Manu, Udenamo and Unami,47 merged into a front and therefore created the national movement, that is to say a movement covering the whole colonial area. Any attempt to continue these former movements would be ipso facto an attempt against the “nation.” This idea of the nation imposed the one-party system (the “One people, one nation, one party” slogan, based on the “For the Nation to live, the Tribe must die” precept). We know today that this merger of three movements did not occur; that Frelimo is a new organization stemming from strong pressure by Tanzanians and Americans. But the history of Udenamo, Manu and Unami is still to be written. Nevertheless, the findings of my research do not correspond with the official history: the problem of Udenamo was not at all a “regional characteristic”, but was its leadership, since its president, Adelino Gwambe, was very young, radical and immature. Udenamo was not a “regional movement” of the centre and south of the country. The founding members were from these regions, but Udenamo grew strongly and included many Maconde people.48 Indeed, Manu was a Maconde organization, but not the largest (thus not the most representative): the principal Maconde organization was the Mozambican African Association, MAA, which refused to merge with TMMU (Tanganhica Mozambique Maconde Union, close to Tanu49 and was transformed into Manu) and, after the Mueda events (June 1960)50 in which its leadership was arrested, joined Udenamo and not Manu. Two years after the foundation of Frelimo, this maconde base of Udenamo, the former MAA, joined Frelimo, the only movement

47 Manu, Mozambique National Union; Udenamo, União democrática nacional de Moçambique, Mozambique National Democratic Union; Unami, União nacional de Moçambique Independente, Independent Mozambique National Union. 48 M. Cahen, ‘The Mueda Case…,’ op. cit. 49 TANU, Tanganyika African National Union, J. Nyerere’s party which achieved independence in 1961. 50 M. Cahen, ‘The Mueda Case…,’ op. cit. 24 michel cahen capable of making war. But the point is that at the very moment of the foundation of Frelimo, Udenamo was already an inter-ethnic movement with a modern program. And Unami could never be considered representative of Tete and Zambezia (thus “regional”) feelings, since the main African associations of these long-colonized regions never had any link with it:51 Unami was a very small group rooted in the border area south of Milange. But Frelimo created the story of the history of its nation in order to establish its sole legitimacy.52 Nevertheless, from the historical rather than the social sciences viewpoint, the key question is not whether this story is true or false, but whether or not it works as a nation-building process. What I am sure of, is that this global ideological context caused deep disregard for African society, for the first African nations, for regional balance, and a bureaucratic and technocratic approach to development and urbanization – this is what I have called the paradigm of authoritarian modernisation and strong paternalism.53 Along with other serious forms of social suffering, this legacy continues to create ethnic tensions in the country and counteract any feeling that the existence of this Republic is a guarantee of progress and that it could be good to identify with it. Even if Frelimo won the last elections (October 2009) with its best figures ever, this result is more a reflection of allegiance to a now well-stabilized power, within a context of state neo-patrimonialism and absence of any credible opposition parties, than evidence of a reinforced national feeling – which has been a major concern of President Armando Guebuza since his first election in 2004.54 Nonetheless, the project of a nation does exist and, after Luís de Brito,55 I suggest that we stop calling it “nationalism” and begin calling it nationism. This is not for the sake of inventing a new word, or copyrighting a new concept. It is to express a very different political and social process.

51 Sérgio Chichava, Le ‘Vieux Mozambique’: l’identité politique de la Zambézie, PhD thesis, Political Studies institue of Bordeaux, 2007. 52 Lorenzo Macagno, ‘Fragmentos de uma imaginação nacional,’ Revista Brasileira de Ciências Sociais (São Paulo), vol. 24, no. 70, 2009, pp. 17–35. 53 Michel Cahen, « Mozambique: une impossible alternative dans la culture politique ? », in António Romão, Manuel Ennes Ferreira & Joaquim Ramos Silva, Homenagem ao Professor Doutor Adelino Torres, Almedina, Coimbra, 2010 (Económicas, 14). 54 M. Cahen, ‘Mozambique: une impossible alternative …,’ op. cit. 55 Luís Cerqueira de Brito, Le FRELIMO et la construction de l’État national au Mozambique. Le sens de la référence au marxisme (1962–1983), Ph.D Thesis, Université de Paris 8, Paris, 1991, 350 p. ; L. de Brito, ‘Une relecture nécessaire: la genèse du parti-État Frelimo,’ Politique Africaine, no. 29, 1988, pp. 15–28. anticolonialism & nationalism 25

Nationism is not the political expression of a nation, but the nation-statist ideology of an elite that is opposed to the existing (first) nations, seeking to mimic a centuries-long European process in a few years, and to build the new nation regardless of the desires of other inhabitants.

The Heuristic Usefulness of Portuguese-Speaking Africa

All of this could apply to other countries, including some outside Africa. But the case of the former Portuguese empire, in particular Angola and Mozambique, is very useful because it was a more “radical case”: the expansion of a very old merchant nation, colonialism by weak capitalism in which the new colonial capitalism did not succeed in deleting the first age of colonisation, very bureaucratic colonisation, an old nucleus of Creole elites, a very small and socially bureaucratic new African elite, an armed struggle, a weighty paradigm of authoritarian modernisation, post-independence civil wars, the “structuring” role of stalinized Marxism among modern elites, the strong influence of the Portuguese model in imagining the nation, etc. These are the reasons that led to my announce- ment in an article published in the French journal Revue Historique, that I would no longer use expressions such as “national liberation wars”, “national liberation movement”, “nationalist fronts” or “nationalism”, preferring only “anticolonial movement” or “anticolonial liberation wars”, except when the above terms are related to political expression of indisputable nations.56

Cape Verde as a Counterexample I have made no reference so far to the Cape Verde islands. It is not possible here to go through the history of Cape Verdean, but it is necessary to explain why Cape Verde was not included in the discussion. For Guinea, Angola and above all Mozambique, I have explained how and why the acceptance of the relevance of colonial territorial space required a process of modern legitimisation by the rapid production of a nation, imply- ing a paradigm of authoritarian modernisation. I am not discussing here whether or not it was possible to do otherwise, I am only studying what

56 ‘Lutte armée d’émancipation anticoloniale ou mouvement de libération nationale ? Processus historique et discours idéologique. Le cas des colonies portugaises, et du Mozambique en particulier,’ Revue Historique (Paris, PUF), vol. 315–1, no. 637, 2006, pp. 113–138. 26 michel cahen actually occurred. I have explained how and why this paradigm slows down the nation identification process and provokes a founding crisis of nation-statist states which are not nation-states. At the beginning of the article, I classified the end of European imperial power into three categories: independence without decolonisation (the case of Latin America), decolonisation without independence (the case of Goa, Macao, or French Départements d’outre-mer) and independence with decolonisation (the case of Africa and parts of Asia). The case of Cape Verde does not quite fit into any of these categories. As in some countries in Latin America, society in Cape Verde was a complete “production” by Portugal, it was not an African society invaded by a foreign country. But after the 18th century decline studied by António Carreira,57 the great majority of white settlers left the archipelago, leaving slaves, mixed-race foremen and the Catholic Church. So what is a slave society, colonial or colonized? Obviously slaves were colonised, since they were abducted and personally subjected to European civilization, but coming from a wide range of geographical locations and ethnic backgrounds, and continuing under European domination, they were unable to rebuild African societies. I am not saying that no African culture survived, I merely note that the social structure of Cape Verdean society is completely different from that of any society on the continent: no lineage kinship, no tribes, no traditional chiefs, no clans and their food taboos, no ethnicities, no age classes. It is a society made up of colonised persons but it is not a colonised society. It is a special case of colonial society in which the huge majority of the population is part of the most exploited social class of the colonisation, that is to say the slave class. In Brazil this same class existed, but not as huge a majority of the population. Yet, it must be considered that slaves had been colonised as individuals, but individuals remaining within the colonial society, not forming a colonised society. Even quilombos, mocambos or candomblés never rebuilt African societies, but built small slave republics or historically slave-rooted religious groups.58 From this point of view, Cape Verde could

57 António Carreira, Cabo Verde. Formação e extinção de uma sociedade escravocrata (1460–1878), Instituto do Património Cultural, Praia, 2000 [3rd ed.], 548 p. See too Maria Emília Madeira Santos (ed., with Luís de Albuquerque for Vol. I), História Geral de Cabo Verde, 3 vols, Instituto de Investigação Científica Tropical, Lisbon / Instituto Nacional da Cultura, Praia, 1991, 1995, 2002, 523+642+540 p. 58 Kadya Tall, 2002, ‘Comment se construit et s’invente une tradition religieuse ? L’exemple des nations de candomblé de Bahia,’ Cahiers d’études africaines (Paris), vol. 167, no 3, 2002, pp. 441–461; K. Tall, Le candomblé de Bahia, miroir baroque des mélancolies post- coloniales, forthcoming, Paris, 165 p. anticolonialism & nationalism 27 be compared with Haiti, another country of former slaves, but the comparison is difficult because the 1804 revolution in Haiti killed almost all of the whites, derailing the genesis of a mestizo Creole society. Paradoxically, the slave history of Cape Verde engendered positive conditions for a nation formation process. The mode of production of the whole society was “automatically” that required by the state and the same as that of the colonial mother land, which provided a link to the world economy (though social formations were different in Portugal and in the archipelago, its slave economy was directly integrated into the merchant economy of the empire). There was no resistance of indigenous societies organised around a domestic mode of production which was neither mer- cantile nor capitalist. People were “automatically” organised as part of the world economy, even if it was the result of profound marginalisation; they did not constitute another system and there was no need for an imposed articulation of separate modes of production. In this situation, the colonial apparatus of the European dominant state fitted in with the society: it exploited and dominated the people, but it was not a foreign power ruling over an indigenous society – everybody and nobody is a foreigner in such a situation. It was the local administration of a slave state over its slaves. The functions of the state were coherent with society. After the slow decline of slavery and extensive spread of Creolity, and above all after independence, the coherence of the state was reinforced within its population, thereby allowing an identification process. In other words, the problem here was not of an authoritarian modernisation process, imposed by nation-statist ideology, which actually weakened the identification process.59 Here, the modern state did not have to do that since the society had been entirely shaped by the coloniser – a situation which was never contradictory to the existence of resistance and revolts. The authoritarian modernisation process had already been performed by slavery which broke native social relationships and ties of lineage or ethnically-based horizontal solidarities, creating conditions, after slavery, for more individual autonomy, which is to say exactly what capitalism needs for its labour market. It also explains the fact that Creolity is not confined to a separate social milieu (as in Guinea-Bissau or

59 Obviously, this statement deals only with a general framework, since there were some events in Cape Verde archipelago where PAIGC tried an authoritarian modernization reform, as in the case of the land reform in Santo Antão island during the 1980s, which provoked violent peasants demonstrations. 28 michel cahen

Angola), but is a feature of the whole society, from the “blackest” and poorest peasant to the “lightest” and wealthiest student, even if there are social tensions between these social milieus and class struggle. This explains why the politics of the same party, the PAIGC, were so different in Guinea-Bissau and Cape Verde.60 Not only were there two different orientations, but the societies were completely different. In Guinea- Bissau, Creolity is not a national feature, but the distinctive characteristic of a socio-cultural elite. In Cape Verde, there are some insularities but no ethnicities. This Republic is one of the few cases of the nation-state in Africa. That does not mean that it is a paradise, for it is not at all – except for tourists of course.

60 Cape Verde and Guinea-Bissau were a unique case in Africa up until 1979, with the same party holding power in two different states. The coup d’état by Nino Vieira in November 1979 put an end to this experiment dreamt up by Amílcar Cabral.

Map 1. Map of contemporary Guinea-Bissau. © Maura Pringle, Queen’s University Belfast. CHAPTER TWO

VIRTUAL NATIONS AND FAILED STATES: MAKING SENSE OF THE LABYRINTH

Philip J. Havik

The literature on colonialism and the post-colony has tended to emphasize the importance of nationalist movements and their icons for an understanding of political change in Africa and beyond. Over the last decades nationalism in its various forms has become an integral part of the debate on civil society in Africa on a par with ethnicity and modernity. Nationalism has long been approached as a question of power and resistance, thus subsuming the historical moment of independence into a teleological process that had a clear starting point and, with the benefit of hindsight, a pre-determined outcome. By way of a binary discourse, the struggle that lay in between was essentially reduced to one in which colonial powers were rudely awakened and subsequently challenged – and defeated – for having conveniently ignored or simply forgotten their subjects’ past and present predicaments. In what appeared to be a metamorphosis, buildings, monuments and streets that once evoked the glory of occupation and its protagonists, were repainted, removed or renamed in order to celebrate the new-found freedom and its heroes. As one collective memory was replaced by another, a discourse built round unity and progress shifted the focus from struggle to a brighter future. National independence and unity as well as social justice replaced metropolitan dependency and colonial differentiation by appealing to new civic values and moral precepts that were to guide peoples in their quest for viable and lasting independence. Much quoted scholars such as Anderson have described nations as ‘imagined political communities’ emphasizing not only their apparent ‘sovereignty’ and their inherent ‘limitations’, but also their association with certain narratives.1 Whilst on the one hand, the author juxtaposed

1 Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism, London: Verso, 2006, p. 204.

© koninklijke brill nv, leiden, �� | doi ��.��/��_�� 32 philip j. havik nationalism’s ‘philosophical poverty and even incoherence of ideas’ with its sheer ‘political power’, he also evoked its spiritual connotations and the remarkable sacrificial aspects of its impact.2 Although some ‘religious’ aspects are inferred, by artificially separating cosmologies from nationalisms the approach fails to take into account the strong linkages between these diverse concepts and sentiments. In addition, the emphasis upon the printed word tends to ignore an entire universe of oral traditions, which are of particular importance in the case of sub-Saharan Africa Other key texts3 which associate political discourse with social and political change in sub-Saharan Africa provided novel perspectives on the cultural roots of such diverse notions as tradition, modernity, ethnicity and autochtony thereby broadening the debate on state and nation, their historical evolution and political legitimacy. Over the last two decades these concepts have become the subject of intense debate, not in the least for having induced a now almost standard and widely accepted framework for the perception of colonial and post-colonial culture. Rather than simply regarding state and nation as institutions or concepts, it is important to recognize their existence as ongoing narratives or transcripts in civil societies,4 for example as cosmologies and mythologies.5 Thus ‘biographical sketches’ of nations emerged, reproduced in official accounts as well as versions rooted in popular culture. Nationalism has been described as ‘a system for organizing the past that depends upon certain narratives, assumptions, and voices, and that continues to have important stakes throughout the social and political order’.6 Indeed, ‘nationalism was not simply a movement that took place to gain freedom’, but a ‘set of meanings which translated everyday struggles and evoked multiple social histories’.7 In this sense, it is one amongst a body

2 Anderson, Imagined Communities, pp. 5–7. 3 Eric Hobsbawn & Terence Ranger (eds), The Invention of Tradition, Cambridge, Cambridge University Press, 1983. Richard Werbner & Terence Ranger (eds), Post-Colonial Identities in Africa, London: Zed Books, 1996; Bruce Berman, Dickson Eyoh & Will Kimlicka (eds), Ethnicity & Democracy in Africa, Oxford/Athens, James Currey/Ohio University Press, 2004; Joshua B. Forrest, Sub-Nationalism in Africa: ethnicity, alliances and politics, Boulder: Lynne Rienner, 2004. 4 James Scott, Domination and the Arts of Resistance: hidden transcripts, New Haven & London: Yale University Press, 1990. 5 Achille Mbembe, “Provisional Notes on the Postcolony”, Africa, 62, 1, 1992, pp. 3–4; Achille Mbembe, On the Postcolony, Berkeley: University of California Press, 2001, p. 111. 6 Nicholas Dirks (ed.), Colonialism and Culture, Cambridge: Cambridge University Press, 1995, p. 14 7 Dirks, Colonialism, pp. 14–17. virtual nations and failed states 33 of competing and complementary narratives, some static and others dynamic, some public others hidden, which interact over time within a given space. Thus, colony and post-colony have come to be seen as a historical and cultural continuum in which the question of the nation is but one amongst many ideational threads that can be identified. Scholars such as Ekeh, Davidson, Bayart, Geschiere, Werbner, Chabal & Daloz, Mbembe, Das & Poole, De Boeck & Honwana, Broch-Due, Richards and Kaarsholm,8 amongst others, have contributed to broaden our understanding of the ‘post-colonial situation’, by proposing new frameworks for addressing African societies and politics. Notably, these approaches have one particular feature in common, i.e. a tendency to take into account the plurality of civil societies’ voices. Rather than limiting themselves to well worn theories and concepts, these authors attempted to transcribe the diversity of local narratives, memories and cosmologies. The overriding publicly diffused image of the African continent is that of societies in constant turmoil, affected by conflict and poverty, which may erupt at any moment. Violent conflicts that haunted African countries and populations over the last decades have served to highlight the importance of listening to these voices ‘from below’ and understanding their significance for an understanding of citizenship and survival in rapidly changing environments.9 Scott employed the term infrapolitics in order to capture forms of disguised and surreptitious resistance produced by

8 Peter Ekeh, “Colonialism and the Two Publics in Africa: a theoretical statement”, Comparative Studies in Society and History, 17, 1, (1975), pp. 91–112; Peter Ekeh, “Social Anthropology and Two Contrasting Uses of Tribalism in Africa”, Comparative Studies in Society and History, 32, 4, 1990, pp.660–700; Basil Davidson, The Black Man’s Burden: Africa and the Curse of the Nation State, Oxford: James Currey, 1992; Jean-François Bayart, The State in Africa: the politics of the belly, London & New York: Longman, 1993; Peter Geschiere, The Modernity of Witchcraft: politics and the occult in post-colonial Africa, Charlottesville & London: Virginia University Press, 1997; Richard Werbner (ed.), Memory and the Postcolony: African anthropology and the critique of power, London: Zed Books, 1998; Patrick Chabal & Jean Pascal Daloz, Africa Works: disorder as political instrument, Oxford: James Currey, 1999; Mbembe, “Provisional Notes”; Mbembe, On the Postcolony; Veena Das & Deborah Poole (eds), Anthropology in the Margins of the State, Santa Fé & Oxford: School of American Research Press/James Currey, 2004; Filip De Boeck & Alcinda Honwana (eds), Makers and Breakers: children and youth in Postcolonial Africa, Oxford: James Currey, 2005; Vigdis Broch-Due (ed.), Violence and Belonging: the quest for identity in post-colonial Africa, London: Routledge, 2005; Paul Richards (ed.), No Peace, No War: anthropology of contemporary armed conflicts, Athens: Ohio University Press, 2005; Preben Kaarsholm, Violence, Political Culture and Development in Africa, Oxford/Athens/Pietermaritzburtg: James Currey, Ohio University Press & University of KwaZulaNatal Press, 2006. 9 Mariane C. Ferme, “Deterritorialized Citizenship and the Resonances of the Sierra Leonean State”, in Das & Poole (eds), Anthropology, pp. 81–116. 34 philip j. havik subaltern groups organized in kinship, community and client groups that constantly test and probe the “boundaries of the permissible”.10 Others focused on the ‘neo-colonial’ nature of (nation) states and statist nationalisms, their ‘failure’, their arbitrary and predatory nature, regarding them as a curse and a threat to progress.11 Both however tend to underline the considerable diversity and dynamics of societies and states, and their extraordinary capacity to move between extremes in the epistemological spectrum: ‘peoples can never for long be confined to cages of any neat scenario’.12 Over the last decades, scholars have begun to question the notion of ‘stateness’,13 which, responding to certain ‘Western’ paradigms of political organization of society, became the standard, exclusive reference for polities globally, to the detriment of societies and non-statal institutions. The recent debate regarding the situation in countries that experienced liberation and/or civil wars such as Mozambique, Angola, Zimbabwe, Sierra Leone and Guinea Bissau has shown to what extent the study of war and conflict is challenging existing mind frames on sub-Saharan African societies.14 Despite several decades of independence and the arrival of new generations on the scene, the fuelling of narratives and memories in periods of crisis has acutely illustrated the contemporary search for meanings in order to make sense of the post-colony transformed into a theatre

10 Scott, Domination, pp. 199–200. 11 Davidson, Black Man’s Burden, pp. 206–209. 12 Davidson, Black Man’s Burden, p. 293. 13 Joel Migdal, Strong Societies and Weak States: state-society relations and state capitalism in the Third World, Princeton: Princeton University Press, 1988; Joel Migdal, Atul Kohli & Viviennne Shue (eds), State Power and Social Forces: domination and transformation in the Third World, New York: Cambridge University Press, 1994; Daniel Lambach, “The Perils of Weakness: failed States and perception of threat in Europe and Australia” (unpubl. conference paper), London: King’s College, 2004; Bernard Helander, “Who Needs a State: Civilians, Security and Social Services in North-East Somalia”, in Richards (ed.), No Peace, No War, pp. 193–202; Charles T. Call, “The Fallacy of Failed States”, Third World Quarterly, 29, 8, 2008, pp. 1491–1507. 14 Mariane C. Ferme, The Underneath of Things: violence, history and the everyday in Sierra Leone, Berkeley: University of California Press. 2001; Harry Englund, From War to Peace on the Mozambique-Malawi Borderland, Edinburgh: Edinburg University Press/ International African Institute, 2002; Alcinda Honwana, “The Undying Past: spirit possession and the memory of war in Southern Mozambique”, in Birgit Meyer & Peter Pels, Magic and Modernity: interfaces of revelation and concealment, Stanford: Stanford University Press, 2003, pp. 60-.80; Inge Brinkman, “War and Identity in Angola: two case studies”, Lusotopie, 2003, pp. 195–221; Jocelyn Alexander & Jo Ann McGregor, “Hunger, Violence and the Moral Economy of war in Zimbabwe”, in Broch-Due (ed.), Violence and Belonging, pp. 75–90; Henrik Vigh, Navigating Terrains of War: youth and soldiering in Guinea Bissau, New York: Berghahn Books, 2006. virtual nations and failed states 35 of seemingly endless conflicts and struggles. The idea of a ‘no peace, no war’ continuum has emerged as a framework for analysis of a twilight zone in which many African countries and populations find themselves locked into.15 This has heightened the need for grasping the social relationships that lie at the heart of these upheavals and linking ‘local’ narratives to ‘national’ ones in order to compare and distinguish perceptions of events, for example in the case of liberation and civil wars16 as well as looking at African nationalisms and ‘sub-nationalisms’ from a comparative point of view.17 However, scholars have rightly cautioned the over- emphasis upon the ‘powerful rhetoric’ of anti-colonialism and nationalism and the subsequent neglect, loss or exclusion of information on certain aspects that go beyond the underlying concept of the nation thereby de- legitimizing alternative discourse.18 The case of Guinea Bissau plainly justifies a reassessment of nationalism and nationhood in the light of popular narratives. Few countries can boast a successful nationalist armed struggle (1963–1974) as well as a ‘failed state’ to the degree Guinea Bissau appears to have achieved over the last decades.19 The genealogy of the nation was deeply rooted in and identified with armed resistance and the anti-colonial struggle, while being directly associated with a single nationalist liberation movement

15 Paul Richards, “New War: an ethnographic approach”, in Richards (ed.), No Peace, No War, pp. 1–21. 16 Inge Brinkman, “Ways of Death: accounts of terror from Angolan refugees in Namibia”, Africa, 70, 1, 2000, pp. 1–24; Harry Englund, “Conflicts in Context: political violence and anthropological puzzles”, in Broch-Due (ed.), Violence and Belonging, pp. 60–74; Ferme, “Deterritorialised citizenship”; Alexander & McGregor, “Hunger, Violence”. 17 Crawford Young, The African Colonial State in Historical Perspective, New- Haven-London, Yale University Press. 1994; Frederick Cooper, Decolonization and African Society: the labor question in French and British Africa, Cambridge-New York: Cambridge University Press, 1996; Frederick Cooper, Africa since 1940: the past of the present, Cambridge-New York: Cambridge University Press, 2002; Forrest, Sub-Nationalism in Africa; Alexander Keese, Living with Ambiguity: integrating an African elite in French and Portuguese Africa, 1930–61, Stuttgart: Franz Steiner Verlag, 2007. 18 Frederick Cooper “The Dialectics of Decolonization: nationalism and labor movements in colonial French Africa”, in Frederick Cooper & Ann Laura Stoler (eds), Tensions of Empire: colonial cultures in a bourgeois world, Berkeley: University of California Press, 1997, especially p. 408. 19 According to the Ibrahim index of African governance, Guinea Bissau is rated as one of the ‘most corrupt’ countries (out of a total of 48) in sub-Saharan Africa. The ‘Failed States’ index of 2011 lists Guinea Bissau as a state in danger of ‘failing’ (18th position), suffering from a significant measure of ‘de-legitimization’ and ‘factionalized elites’. It is flanked by states that are seen as ‘stable’ (i.e. Senegal) and ‘critical’ (i.e. Guinea). See index compiled by Foreign Policy and Fund for Peace at http://www.foreignpolicy.com/ articles/2011/06/17/2011_failed_states_index_interactive_map_and_rankings 36 philip j. havik and post-independence ruling party and its leadership.20 Colonial resistance grafted the nation so deeply onto political discourse, thereby engendering a veritable cult of the ‘nation-party-state’ as the sole pillar and symbol of authority. It is significant that despite the historical importance of the nationalist movement and its successful quest for independence, its leaders have rarely publicized their version of events or provided accounts of certain episodes that occurred during the struggle or in government. When they did, their accounts tend to belittle, disregard or silence events that challenged their or the party’s notion and purported practice of ‘struggle’ while ignoring or demonising other nationalist movements, personalities or ideas that played a role in it.21 However, what had long been muted or un-challenged is now up for discussion: the utterance of ‘heterodox’ comparisons between the pre- and post-1974 era has almost become a recurrent habitus of politicians, scholars and civil society alike. The multiparty experiment accelerated an all-enveloping political and ideological fragmentation which, deepened by armed conflict and economic crisis, imported the spectre of breakdown of state authority in the form of a gamut of topical labels such as ‘weak’, ‘failed’ and ‘shadow’ state. This begs the question what has become of the nationalist project which once attained mythical proportions and subsequently shipwrecked, and, importantly, how ‘civil society’ has since tried to make sense of its demise. Indeed, Guinea-Bissau’s accident-prone post-colonial course raises a number of questions. First, the idea that nationalist discourse is central to an understanding of the post-colony and independent states should be reexamined in the light of its reification before independence and its latter-day fragmentation and marginalization in public debate. Secondly, the idea that nationalist struggles forge unitary nations or anything like nation-states also requires reevaluation, e.g. in the light of the

20 Lars Rudebeck, Guinea Bissau. A Study in Political Mobilization, Uppsala: SIAS, 1972; Stephanie Urdang, Fighting Two Colonialisms: women in Guinea Bissau, New York: Monthly Review Press, 1979; Amílcar Cabral, Unity and Struggle, London: Heinemann, 1980; Patrick Chabal, “Party, State and Socialism in Guinea Bissau”, Canadian Journal of African Studies, 17, 2, 1983, pp. 189–210; Rosemary Galli & Jocelyn Jones, Guinea Bissau: economy, politics and society, Boulder, Westview, 1987; Carlos Lopes, Etnia, Estado e Relações de Poder na Guiné-Bissau, Lisbon: Edições 70, 1982; Carlos Lopes, A Transição Histórica na Guiné Bissau: do movimento de libertação nacional ao Estado, Bissau: INEP, 1987; Mustafah Dhada, Warriors at Work: how Guinea was really set free, Niwot: University Press of Colorado, 1993. 21 See the accounts of the ousted president of Guinea Bissau, Luis Cabral, Crónica da Libertação, Lisbon: O Jornal, 1980; and of the long-time secretary-general of the liberation movement and first Cape Verdean president, Aristides Pereira, Uma Luta, Um Partido, Dois Paises, Lisbon: Ed. Notícias, 2002. virtual nations and failed states 37 trans-national ‘one nation, two states’ ideology proposed by the liberation movement, its failure and profound ambivalence. Thirdly, if states can be built and fail, one could argue so can nations; the issues are interrelated and should be regarded as processes of construction and deconstruction. Fourthly, the banner of poder popular or popular power erected by the liberation movement and the latter’s fetishisation of state power beg the question of (non-)sharing and how it is locally perceived. Fifth, ‘public’ and ‘hidden transcripts’ should be taken into account when attempting to signify local and national events, institutions and sentiments in order to illustrate their particular socio-historical context. And finally, in the case of Guinea Bissau the linguistic context needs to be considered given that Guinean Creole or Kriol was crucial for the production and diffusion of nationalist discourse. The narratives thus generated illustrate the entanglement of ‘local’ and ‘national’ metaphors on society and politics and the diversity of meanings within a dynamic cultural space. The time span of the essay encompasses the period from the anti- colonial war (1963–1974) to the 1998–99 guera di Bissau, i.e. the Bissau war. In order to analyse Guinean nationalism and situate it within the context of popular discourse and memory, data have been included from published literature, as well as from archival and oral sources. After dealing with the specificities of nationalist meanings and metaphors, and placing them in a historical context in which they emerged, it focuses on the legitimization and reification of the nation, discusses its post-colonial fragmentation and present insignificance. It does so by looking at key turns in the country’s short history, including the nationalist struggle, the 1980 coup, multi-party elections and the 1998–1999 war. Finally, it addresses a number of key questions concerning Guinea Bissau’s anti- and post-colonial experience, that require urgent reassessment in the light of the paradigm shifts that occurred over the last decades as a result of political and economic reforms, continuous political infighting and violent conflicts and social (re)stratification. Thus, it proposes a framework for the analysis of these dynamics based upon narratives and sequences that take into account the complementary worlds of harsh realities one the hand, and sensorial experiences on the other that some have called ‘labyrinths of memory’.22

22 Marea C. Teski & Jacob J. Climo (eds), The Labyrinth of Memory: ethnographic jour- neys, Westport/London: Bergin & Garvey, 1995. The authors define the concept as “the changing, elusive locale of our past, present, and future lives (…) that takes different turns every time we go back it” (p. 1). 38 philip j. havik

Unity & Nation: The Making of Myths and Heroes

The issue of nationalism and culture has over the last decades been reformulated into one that contrasted religious and nationalist discourses, i.e. sacred languages with nationalist consciousness.23 By evoking the simultaneous emergence of nationalist and religious discourse in the age of enlightenment, their entanglement implied a cross-fertilization of terminology and slogans. Nevertheless, the ‘magic’ located in nationalism’s ability ‘to turn chance into destiny’24 reveals strongly spiritual overtones by associating it with credence and promise that were to result in national ‘salvation’ and a ‘limitless future’. Nationalist discourse stressed an unas- sailable truth, centered on the (inherent) superiority of leadership and enmeshed the genealogy of culture with that of its protagonists. Viewed from the vantage point of its encoding, there appears to be very little difference between colonial and post-colonial ‘nationalist’ tropes given that both appeal to ‘sacred national ideals’.25 These patriotic sentiments found expression in slogans stressing ‘national unity’ and ‘national solidarity’ that were inter-changeable and multi-purpose.26 The creed of the liberation movement, PAIGC (Partido Africano para a Independência da Guiné e Cabo Verde) built around ‘unity, struggle and progress’ turned them into national values of the ‘new order’. The recourse to ‘sacred language’ was a common tool in this respect and pretended to forge a sense of community amongst receptors in order to overcome the obstacles of social and political diversity and tensions. Pan- Africanism and its Lusophone pendant27 associated with the nationalist liberation movements operating in Portuguese Africa, pretended to create a community of interests across geographical and political dividing lines, while playing on common historical genealogies. Seeing the denial of history as a denial of culture, the erstwhile founder and leader of the PAIGC, Amílcar Cabral (1924–1973)28 in a famous lecture delivered to an

23 Anderson, Imagined Communities, pp. 9–46. 24 Anderson, Imagined Communities, p. 12. 25 António E. Duarte Silva, A Independência da Guiné Bissau e a Descolonização Portuguesa, Oporto: Afrontamento, 1997, pp. 34–37. 26 Other parallels have also been identified between colonialist and nationalist projects; see for example Galli & Jones, Guinea Bissau, pp. 45–50. 27 The Lusophone version of Pan-Africanism was formalised in 1961 in the form of CONCP (Conferência das Organizações Nacionais das Colónias Portuguesas), which brought together the nationalist movements in Portugal’s African colonies. 28 Patrick Chabal, Amílcar Cabral: revolutionary leadership and people’s war, London: Hurst & Co, 2002; Jock McCulloch, In the Twilight of Revolution: the political theory of virtual nations and failed states 39 academic audience at Syracuse University in New York 1970 on ‘National Liberation and Culture’ argued in favour of a process of re-Africanisation which he referred to, significantly, as one of reconversion’. 29 This in turn implied a communion of sacrifices on the part of the people and its leaders if they were to be ‘victorious’ and ‘create progress’. Given that culture also contained the elements of stagnation, he warned his constituency for the risks of confusing, e.g. in the case of racialist ideas, ‘objective and historical reality’ with ‘what appeared to be a ‘spiritual creation or the result of a special nature’.30 What we wish to suggest here is that the ‘sacred language’ used by the nationalist movement before independence ended up by conflating one with the other for reasons of expediency, thereby incit- ing its supporters to move beyond the confines of the village and ‘accept sacrifices and even do miracles’.31 Thus he identified the lack of ideology as one of the major weaknesses of African liberation movements.32 Revolution without revolutionary theory was unthinkable and unfeasible. A general ignorance of social and political realities had weakened their efforts, just like the absence of a revolutionary consciousness of the mass of the people. Both had to be nurtured for not being spontaneous: hence the emphasis on ‘correct guid- ance’.33 Some revolutions foundered, not in the least because of the ‘can- cer of betrayal’ that led to their leaders’ demise, as in the case of Nkrumah, while others were successful such as the Cuban revolution.34 In both instances however, a measure of heroism was attributed to leaders transforming them into figureheads, icons of the new age of independence.

Amílcar Cabral, London: Routledge, 1983; António Tomás, O Fazedor de Utopias: uma bio- grafia de Amílcar Cabral, Lisbon: Tinta da China, 2007; José Julião de Sousa, Amílcar Cabral 1924–1973: vida e morte de um revolucionário africano, Lisbon: Nova Veja, 2011. 29 Cabral, Unity and Struggle, p. 145. For an analysis of Cabral’s theory of liberation and the issue of (re-) africanisation, see Reiland Rabaka, “Amílcar Cabral: using the weapon of theory to return to the source(s) of revolutionary decolonization and revolutionary re- africanization”, in R. Rabaka (ed.), Africana Critical Theory: reconstructing the black radical tradition, from W.E.B. Dubois to C.L.R. James to Frantz Fanon and Amílcar Cabral, Lanham, MD: Lexington Books, 2009, pp. 227–83. 30 Cabral, Unity and Struggle, p. 150. 31 Cabral, Unity and Struggle, p. 151. 32 Cabral, Unity and Struggle, p. 122. 33 Galli & Jones, Guinea Bissau, p. 48. 34 Cabral, Unity and Struggle, pp. 114–18 & 119–21. What Cabral was unable to foresee at the time, was that although independence would eventually be achieved as a direct outcome of the struggle, his own destiny was to be decided by betrayal and assassination in 1973 at the hands of members of the liberation movement he helped create, an ominous precedent; see José Pedro Castanheira, Quem Mandou Matar Amílcar Cabral?, Lisbon: Relógio d’Água,1999. 40 philip j. havik

Thus, Nkrumah is posthumously placed by Cabral on the ‘highest peak of the Kilimanjaro of the African revolution’ in order to ‘immortalise (…) the ‘pioneer of Pan-Africanism’.35 In this vein, Nkrumah – who is compared to Christ – is thought ‘to rise again each dawn in the heart and determination of freedom fighters’, while his ‘immortal spirit presides and will pre- side at the judgment of history at this decisive phase in our peoples’ lives’.36 The carefully chosen transcendental metaphor of resurrection and its biblical apotheosis is given a topical turn with the example of the ‘Cuban miracle’ which proved that the ‘glorious struggle’ against a common enemy could be crowned with success while allowing the party to prepare the way for the advent of a reborn ‘new Man’.37 In addition, the same speeches consolidated the position of the ‘vanguard’ and its ‘correct interpretations of objective conditions and specific demands’, bestowing an aura of un-questionability upon the party’s intervention while linking such national conquests to international ‘bonds of history, blood and culture’.38 The skirmishes and battles that resulted from the strategy of confrontation undertaken by the anti-colonial resistance movement PAIGC from the mid 1950s which it claimed as ‘momentous events’, like the Pidjiguity dock-workers strike in 1959 or the battle of Como-Caiar in 1964, figure prominently in the movements’ history as turning points in the ongoing anti-colonial campaigns.39 While singing the praise of the ‘acts of heroism and sacrifices’ these events and the actors involved took on new form by means of the reification of ‘the party’, written in capital letters in commu- niqués and publications. Explicitly, the idea of ‘duties of the party and the struggle’ became watchwords of a movement that exulted the shedding of

35 Cabral, Unity and Struggle, p. 115. 36 Cabral, Unity and Struggle, p. 117. 37 Cabral, Unity and Struggle, pp. 119–20. Such references clearly refer to one of Cabral’s theoretical mentors, Frantz Fanon, and his emphasis on the emergence of the ‘new (history of ) man’ from the ashes of empire; see Frantz Fanon, The Wretched of the Earth, New York: Grove Press, 1963, p. 315. But in his speech/essay Weapon of Theory, Cabral proposes a novel solution for the tendency of African elites to transform themselves into a ‘national pseudo-bourgeoisie’, stressing the need for the petty bourgeoisie to (be able to) ‘commit suicide as a class’, in order to be ‘reborn’ or ‘restored to life’ (‘ressuscitar’) as a revolutionary worker, who completely identifies with the most profound aspirations of the people to which it belongs.’ (p. 136). 38 Cabral, Unity and Struggle, p. 121. 39 PAIGC, História da Guiné e as Ilhas de Cabo Verde, Oporto: Afrontamento, 1974, pp. 148–55; Leopoldo Amado, “Guiné-Bissau: 30 anos de independência”, Africana Studia, 8, 2005, pp. 109–35. virtual nations and failed states 41 blood while pledging brotherhood and a sense of community in the course of a national and Pan-African crusade. We all know that the African peoples are our brothers. Our struggle is theirs. Every drop of blood that falls from us equally from the body and heart of our African brothers, these African peoples.40 The use of universal terminology that evoked blood ties and kinship unperturbed by frontiers was carefully woven into a political message on the national dimension of the struggle and the movement that spear- headed it.41 In the case of Guinea Bissau, a number of specific factors imprinted themselves upon the campaigns led by the PAIGC and the discourse it employed. The objective of the creation of a ‘one nation, two states’ configuration already inscribed in the movement’s name and superimposed upon the struggle for independence, proposed an innovative vision of a bi-cephalous nation which would span geographical and cultural divisions. The party’s official history associates this objective of the struggle with the ‘close and intimate ties between two peoples united as in one body and soul’.42 Thus the party motto Unidade, Luta e Progresso (unity, struggle and progress) suggests a set of less obvious and ‘hidden’ meanings which should be put into context. The fact that the armed struggle was fought on Guinean soil, that Capeverdeans formed a significant part of the leadership and that unity did not exclusively refer to Guinea Bissau but also to a broader, trans-national political unity with Cape Verde, precludes any simplification of the analysis of ‘unity’ and ‘nation’. The leader of the movement and principal proponent of this objective himself symbolised the ambivalence of such a national identity: born of Cape Verdean parents in Portuguese Guinea (Bafatá, 1924) he claimed to straddle that divide. Although some have maintained that he was essentially a Cape Verdean in outlook43 whilst others emphasise his belief in a “spiritual africanisation”, his genealogy allowed for an evocation of mixed transnational roots and local knowledge, further reinforced during his work on the agricultural census in Guinea in the 1950s.

40 Cabral, Unity and Struggle, p. 256. 41 PAIGC, História, pp. 140–47. 42 PAIGC, História, p. 140. 43 Chabal argues that Cabral was a Cape Verdean educated in Portugal who assimilated Portuguese culture; see Chabal, Amílcar Cabral, p. 168. On the other hand, Tomás suggests that Cabral’s studies in Portugal, strongly influenced his search for an African – pan-Afri- can - identity; see Tomás, O Fazedor, p. 65. 42 philip j. havik

However, the translation of these multicultural antecedents into the stated political aims of a resistance movement demands closer scrutiny. While the fact that Guinea is inhabited by a large number of ‘ethnic’ communities already posed a major challenge to any nationalist platform, the incorporation of two far flung territories within a future ‘nation state’ further added to its already formidable complexities. The colonial context in which Capeverdeans were recruited into the administration of mainland Guinea and other colonies such as Angola – of which his own parents were an obvious example – clearly provided a convenient political setting. Whereas Cabral himself rejected the idea that his call for unity was a personal ‘caprice’,44 his arguments were easily refutable: the idea that the struggle was not feasible if conducted in either one of the territories separately was the case for Cape Verde but not for Guinea.45 Curiously, the claim to the historicity of unity was shared with colonial officials, themselves of Cape Verdean origin,, whose writings were used in order to prop up Cabral’s case.46 The argument that Portuguese colonialism had pur- posefully followed a policy of divide and rule with regard to the two communities, setting one against the other by using Capeverdeans to govern Guineans, shows that the unity issue was essentially a counter-stratagem to thwart colonial rule. Indeed, the idea that Guinea could become independent without Cape Verde was by no means unthinkable despite Cabral’s eloquent defense, and can only be understood in the light of Portugal’s ‘special relationship’ with the archipelago which was advocated and even cultivated by certain factions of the New State, and even considered at a late stage in the war as a possible alternative by its opponents, such as Spínola.47 He admitted as much when stating that ‘only someone who understands nothing about strategy’ could contemplate Guinea’s independence whilst Cape Verde was still colonized.48 More than thirty years on, Aristides Pereira, Cabral’s brother in arms and successor after his assassination in 1973, long-time secretary general of PAIGC as well as the first president of independent Cape Verde, confirmed the

44 Cabral, Unity and Struggle, pp. 40–44. 45 Dhada, Warriors, p. 130. According to this author, “the PAIGC’s commitment to secure the freedom of Guinea and Cape Verde was crucial, not in the nationalist history of Guinea before or after its independence, but for Cape Verde’s liberation”. 46 See António Carreira, “A Guiné e as Ilhas de Cabo Verde: a sua unidade histórica e populacional”, in Ultramar, VII, XIII, 32, 1968, pp. 70–98. 47 José Vicente Lopes, Cabo Verde: os bastidores da independência, Mindelo: Instituto Camões/Centro Cultural Português, 1996, p. 434. 48 Lopes, Cabo Verde, p. 43. virtual nations and failed states 43

‘tactical’ dimension of the slogan, i.e. outwardly towards the colonial camp and internally within the movement, which in his view obtained the desired results.49 Indeed it has been argued that the idea of transnational unity was the only way in which the presence of Capeverdeans in the leadership of a Guinean liberation movement could be justified.50 Interviews with these and other party leaders of Capeverdean descent in the 1990s evoke the existence of a deep-seated insular sentiment, in reaction to suggestions by some party cadres and by the movement’s allies such as Sékou Touré, in neighbouring Guinea Conakry, that Cape Verde should be integrated into the independent state of Guinea Bissau.51 The virulent language they deployed which included metaphors deriding the idea as sheer ‘stupidity’ required a radical stance, i.e. ‘cutting off the pro- posal’s head’ in order to nip the question of the simultaneous independence of Guinea and Cape Verde in the bud during negotiations with Portugal in 1974/75. At the time, Capeverdean cadres argued that the two territories were different countries and only if the respective peoples agreed would unity be declared,52 thereby underlining the optionality of unification. By admitting the uniqueness of the situation – but for a passing reference to the unification of Zanzibar and Tanganyika in the state of Tanzania – while failing to address the political and economic viability of a future ‘mini-state’, Cabral brought the issue into the realm of the un- disputed. His appeal to the self-evidence of the notion whilst attributing any doubts in this respect to sheer ignorance, clearly points in that direction.53 Pereira acknowledged that the ‘issue of unity had been hotly disputed and given rise to misunderstandings, without their having been any interest in seriously and thoroughly debating such a key question for the peoples of Guinea and Cape Verde’.54 Indeed, while still defending the idea of unity, he went as far as stating that in this case ‘the language used did not exactly reflect what we intended’, but that ‘our conviction was absolute’ and that ‘reason was on our side’.55 Those who did not agree, and saw the idea of unity as utopian, were ‘racists’, ‘assimilated pseudo

49 Pereira, Uma Luta, um Partido, p. 104. 50 Chabal, Amílcar Cabral, p. 138. 51 Lopes, Cabo Verde, pp. 317–22. 52 Lopes, Cabo Verde, p. 319. 53 Cabral, Unity and Struggle, p. 42. 54 Pereira, Uma Luta, um Partido, p. 102. 55 Pereira, Uma Luta, um Partido, pp. 105–06. 44 philip j. havik elites (…) full of complexes created by the colonial system’, who demonstrated a ‘malicious ignorance’ and ‘limitless bad faith’.56 The ‘they’ in this case, were Guinean factions within the party which hotly contested the continuation of the Capeverdean presence in Guinean government after independence. Their position was ‘irrational’ for failing to understand that the ‘basic principle’ of the party was based upon ‘the historical communion of the two countries’.57 But in the same breath Cabral himself is quoted as saying that the goal of unity was ‘idealist’ and represented a ‘magical mentality’,58 thereby defying the exclusivity of reason. The coup staged in 1980 in Guinea Bissau by a former guerilla commander, “charismatic popular cadre”59 and Armed Forces commander at the time, João Bernardo ‘Nino’ Vieira, ended all prospects of unity by provoking a split out of which two separate parties emerged, i.e. in a Guinean (PAIGC) and a Cape Verdean (PAICV) organization.60 Allegedly moti- vated by anti-Cape Verdean sentiments, the coup was qualified by its Guinean leaders as a reaction to the ‘narrow nationalism’ of some Capeverdean politicians,61 but as a hostile act by the Capeverdean leadership carried out by ‘false nationalists’.62 The latter line is also taken by Luís Cabral, Amilcar’s half-brother – born in Cape Verde – who was ousted as president by the aforementioned coup. He testifies to the movement’s subtle but significant change of position which replaced ‘two peoples’ by one unified people once the victories on the battlefield combined with favourable results on the diplomatic scene in 1966, i.e. following the Tri-continental conference in Cuba.63 In the latter’s opinion, it is at this juncture that the ‘theory of a nation’, based upon the alleged ‘popular acceptance of a common political authority’ began to take shape.64

56 Pereira, Uma Luta, um Partido, p. 103–06. 57 Pereira, Uma Luta, um Partido, p. 108. The author here quotes an excerpt from Manuel Duarte & Renato Cardoso, Cabo-verdianidade e Africanidade e Outros Textos, bases da unidade preconizadas pelo PAIGC (n.d.). 58 See Duarte & Cardoso cited in Pereira, Uma Luta, um Partido, p. 111. 59 Mustafah Dhada, “The Liberation War in Guinea Reconsidered”, Journal of Military History, 62, 3, 1998, p. 577. 60 Fafali Koudawo, Cabo Verde e Guiné-Bissau: da democracia revolucionária à democracia liberal, Bissau: INEP, 2001. 61 See Relatório do CNG do PAIGC, 1° Congresso Extraordinário, 8–14/11/1981. 62 Pereira, Uma Luta, um Partido, p. 112. 63 Luís Cabral, Crónica, p. 282. 64 The speech Cabral made at the conference, i.e. The Weapon of Theory, became the centerpiece of the movement’s political platform and his principal legacy to liberation theory; see Cabral, Unity and Struggle, pp. 119–37. virtual nations and failed states 45

Popular Power, Magic and Legitimacy

The notions of forging an independent nation-state by rebellion that are contained in and to a large extent structure the language and metaphors employed by actors on the political stage point towards a close but complex relationship between nationalist tropes, political legitimacy and violence. The use of violence in terms of the foundation, legitimation and maintenance of the state65 not only left a deep mark on the political debate, but also indissolubly linked it to the building of the post-colony. The question here is how armed conflict shaped nationalist thinking and discourse, and how local cosmologies occupied centre stage in the course of political mobilization. And also, which meanings they espoused before and after independence, and how the ‘magic’ of popular resistance during the struggle turned into a ‘curse’ – and with it the nation – after 1974. The exercise of violence which was already directly associated with governance both in a physical and a structural sense during the colonial period became part and parcel of the nationalist creed. In this respect it took the high ground of legitimacy, distancing itself from colonial practice in which (‘structural’) violence in the form of forced labour, forced crop cultivation and direct taxation played a central role in order to justify its use.66 At the same time, violence was seen as a strategy to maximise political benefits, a ‘negative heritage’ from the period of the struggle that played a decisive role in the power struggles of the post-colony and the antagonistic social relations that underpinned them.67 But its apparent self-evidence – which was openly legitimized in a discourse built around ‘the ends justifying the means’ – was only part of the picture. For this nationalist ‘vanguard’ was composed of a curious amalgam of

65 Mbembe, On the Postcolony, p. 25. 66 Philip J. Havik, “Estradas Sem Fim: o trabalho forçado e a política indígena na Guiné (1915–1945)”, in CEAUP (ed.), Trabalho Forçado Africano: experiências coloniais comparadas, Oporto: Campo das Letras/Centro de Estudos Africanos (CEAUP), 2006, pp. 229–47; Philip J. Havik, “Ilhas Desertas: impostos, comércio, trabalho forçado e o êxodo das Ilhas Bijagós (1915–1935)”, in CEAUP (ed.), Trabalho Forçado Africano: articulações com o poder político, Oporto: Campo das Letras/CEAUP, 2007, pp. 171–89. 67 Joshua B. Forrest, Lineages of State Fragility: rural civil society in Guinea Bissau, Athens/Oxford: Ohio University Press/James Currey, 2003, pp. 222–32; Amado, Guiné Bissau, p. 121; Ricardo Jacinto Dumas Teixeira, Tiro na Democracia: uma análise sobre o processo de transição democrática na Guiné Bissau, 1994–2007 (2007), published on-line at: http://www.didinho.org/TIRONADEMOCRACIA.htm; Ricardo Jacinto Dumas Teixeira, “Consciência nacional, democratização e conflito político: semelhanças e diferenças entre Guiné-Bissau e Moçambique”, conference paper 2010, published on-line at: http://www .didinho.org/ArtigoCEIA7[1].pdf 46 philip j. havik

Capeverdean Creoles and Christianised Guineans, locally called Kriston, who had for centuries been active commercial brokers on the Guinean mainland. The relay trade which emerged soon after first contact in the fifteenth century, and the export of cash crops (such as peanuts) from the nineteenth century, had long constituted a way of life for these strata. Neither were these Creole speaking strata that Cabral referred to as a local ‘petit bourgeoisie’ new to armed struggles: the ‘pacification’ campaigns waged against Guinean communities from 1890 to 1915 testify to their role – often on both sides – as purveyors of arms and of their use on the battlefield,68 and as political actors.69 In the process, local cult heroes were born, as in the case of João Okica de Sá, a Kriston with Mancanhe roots whose daring exploits are still celebrated in oral traditions. Interestingly, his position was quite ambivalent, being a Kriston-auxiliary for the Portuguese while at the same time a pre- tender to the Mancanhe chiefdom or regulado of Có, eighty kilometers North of Bissau. He received support for his claims from the Liga Guineense (1910–15), an association of Cape Verdean Creoles and local Kriston elders or ‘grandes’ which shared a ‘nativist’ platform with proto-nationalist elements.70 In a situation marked by extreme violence and the clash between Portuguese colonial aspirations and rapidly shifting local alliances, his miraculous ascent to heaven has distinct messianic dimensions which are reinforced by a sense of shared experience and memory.71 These magical moments attenuated the loss of autonomy these strata suffered under a colonial regime which thereafter implanted itself in the territory allocated to Portugal during the Berlin conference (1884–85). At the same time, these oral accounts illustrate the existence of a hybrid spiritual realm which projected highly charged images of popular resistance.

68 Réné Pélissier, Naissance de la Guiné: Portugais et Africains en Sénégambie, 1844–1936, Orgéval, 1989. 69 Philip J. Havik, “Esta ‘Libéria Portuguesa’: mudanças políticas e conflitos sociais na Guiné (1910–1920)”, in Philip J. Havik, Clara Saraiva & José Alberto Tavim (eds), Caminhos Cruzados em História e Antropologia: ensaios de homenagem a Jill Dias, Lisbon: Instituto de Ciências Sociais (ICS), 2010, pp. 123–41. 70 See for example Pélissier, Naissance, pp. 299–300 on Okica’s request for support from Portuguese authorities for his pretensions to settle an internal ‘ethnic’ dispute‘. In a letter to the governor of Portuguese Guinea, he requested arms and munitions in order to remove a rival who had in his view illegitimately occupied the position of régulo. In it, he professes to be Portuguese ‘in heart and soul’ but also states that he cannot abdicate from his customary rights. The authorities’ refusal to concede led him to turn to his compatriots in the Liga Guineense who backed his claims. He was killed during clashes in the Pepel tchon 1915. 71 Interview with a Kriston/Pepel informant, Bissau, Guinea Bissau, 1991. virtual nations and failed states 47

Portuguese officials emphasised the “almost invisible” presence of the native “enemy” which regarded itself as “invincible”, hidden in the thick forest, the corners of which the latter knew like the palm of their hands, whose numbers could not be ascertained with any certainty while those fallen in battle were rarely found.72 The reenactment of this alliance between Capeverdean Creoles, mostly employed in public administration, and the Kriston who traditionally earned their livelihoods as intermediaries in the riverine relay- and urban trade, in the 1950s and 1960s during the nationalist mobilization drives and armed struggle was therefore highly significant. It is by no means coincidental that its revival took place in the same location where it was originally conceived, i.e. in the trading town of Bissau. However, whereas at the time of ‘pacification’ this alliance found its temporary expression in the Liga Guineense73 which essentially had an embryonic nativist – or proto-nationalist for some character and was centred in the Bissau-Boimbo area, from the mid 1950s when first PAI (Partido Africano da Independência) and then PAIGC was established, its ramifications extended far beyond this perimeter and embraced a post WWII nationalist creed. Despite the affirmation of ‘modern’ nationalism in Guinea in the 1950s, the resurgence of mysterious phenomena and heroic exploits during the luta suggests that local narratives retained their importance. The notion

72 Frederico Pinheiro Chagas, Na Guiné, 1907–1908, Lisbon: J.F. Pinheiro, 1910, pp. xii-xvi. The reports by military officials involved in the ‘pacification’ campaigns, contain a large number of narratives regarding unusual situations or magical events that were attributed to superstition; the use of amulets (mezinhos), way in which native auxiliaries approached and honoured the dead bodies of their compatriots, the attacks by bees (baguéra), the natives seeking protection behind baobab trees (polón), or advancing heroically and impetuously with swords drawn in the direction of colonial positions; see Chagas, Na Guiné p. xxvi; p. 70; p. 96; p. 105; p. 108, p. 113; p. 118. Also see Eric Gable, “Conclusion: Guinea Bissau yesterday… and tomorrow”, African Studies Review, 52, 2, 2009, pp. 165–79 and p. 169. Other contemporary accounts, refer to the savage cruelty of natives; the fétiche or witchraft associated with firearms; natives’ sudden attacks of madness, etc.; see Luiz Nunez da Ponte, Campanha da Guiné (1908), Oporto: Guedes, 1909, p. 19, p. 41, p. 42 & p. 104. Similar references can be found in the accounts of the successive military campaigns against the inhabitants of the island of Kañabak from 1917 to 1936. 73 James Cunningham, “The Colonial Period in Guiné”, Tarikh, 6, 4, 1980, pp. 31–46 and p. 35; Peter K. Mendy, Colonialismo Português em África: a tradição de resistência na Guiné Bissau (1879–1959), Bissau: INEP, 1994, pp. 329–35; Wilson Trajano Filho, Polymorphic Creoledom: the “creole” society of Guinea Bissau (unpubl. PhD diss), Philadelphia: University of Pennsylvania, 1998, pp. 228–301; Philip J. Havik, “Mundasson i Kambansa: espaço social e movimentos politicos na Guiné Bissau (1910–1994)”, Revista Internacional de Estudos Africanos, 18–22, 1999, pp. 115–67 and pp. 120–25. 48 philip j. havik of ‘força e fama’ (strength and fame)74 lie at the heart of these transcripts which combine Christianised with ancestral African culture. During the crucial battle waged in 1964 on the islands of Como/Caiar close to the border of then independent Sékou Touré’s Guinea, one of the PAIGC commanders, Agostinho Sá, who staged a successful surprise attack on Portuguese troops by becoming invisible, is a another example of the celebration of ‘magic’ powers in epic battles.75 This time the apotheosis was a victorious one given that the nationalist movement gained control of these islands and thereby of a strategic foothold in the South, where its popular support was greatest amongst younger generations in the rural areas. The evocation of the figure of the trickster well illustrates the continuum which links popular with political performance. The party’s first congress held at Cassacá in the first ‘liberated territories’ in the South, illustrated how different cultures clashed: i.e. party- nationalist with local ‘ethnic’ cosmologies. By leveling accusations of witchcraft at villagers, a group of young un-initiated Balanta commanders had, without the movement’s consent, carried out witch-hunts in Balanta villages as well as retaliatory actions against local farmers.76 The party leadership decided to put the said commanders on trial accusing them of ‘barbarous crimes’, some of whom were subsequently executed, apparently, in order to avoid any further erosion of the movement’s recently acquired popular base and strategic gains.77 The continuing silence on the Congress proceedings and the simultaneous celebration of its major

74 For an account of this terminology and its signification, derived from the Balanta- Brassa, rice cultivators par excellence in Guinea Bissau, see Roy van der Drift, Arbeid en Alcohol: de dynamiek van het gezag van de oudste bij de Balanta-Brassa in Guinee Bissau, Leiden: CNWS, 1992, p. 95. For an account of the adherence of Balanta to the PAIGC movement, see Roy van der Drift, “Birds of Passage and Independence Fighters: an anthropological analysis of Balanta migration to southern Guinea-Bissau and mobilization for the liberation war”, in G. Gaillard (ed.), Migrations anciennes et peuplement actuel des côtes guinéennes, Paris: L’Harmattan, 2000, pp. 151–164. 75 See Trajano, Polymorphic Creoledom, pp. 507–09. This supernatural tale or misteriu was reproduced by L. Azevedo & M.P. Rodrigues, Diário da Libertação (A Guiné Bissau da Nova África), São Paulo: Versus, 1977. 76 This witch-hunt conducted by members of a secret society (fierapte) was led by Balanta healers or djambakós who identified the witches accusing them of causing the death of villagers in the war at the hands of the colonial troops. See Joop T.V.M de Jong, A Descent into African Psychiatry, Amsterdam: Koninklijk Instituut voor de Tropen (KIT), 1987, p. 78. 77 Cunningham asserts that witchcraft accusations by non-initiates were mainly directed against the wealthier Balanta compounds in the Tombalí region, precisely those elders among whom the movement was trying to mobilize; see Cunningham, Guinea Bissau,1956–1974, p. 5. virtual nations and failed states 49 importance for the history of the struggle, again evokes the movement’s deeply rooted ambivalence. The recurrence of such an anti-witchcraft ‘fierapte’ movement led by young women in the early 1980s amongst Balanta-Brassa communities in the South, caused the then government headed by Nino Vieira to intervene and clamp down on its leaders.78 It is significant that official reports and subsequent accounts by party leaders did everything but demystify the events preceding the congress and the decisions taken at the time to suppress alleged abuses. Again, silences reigned and continue to do so, also amongst party leaders and between them and their popular base, and whenever expressed by the former are systematically put into the binary straightjacket of modernity and tradition, excluding all other views. In this respect, the answer Luís Cabral received from the sector commander upon visiting the ‘scene of the crime’ in the South during the anti-colonial war is particularly elucidating: The people are difficult. We cannot always trust them, because it is they who set them up against you. (…) The witch-doctors or those who bring them in will not mess things up.79 The same party leader blamed a ‘magic mentality’ for these events which ‘with the collaboration of the people and those who propagated the idea of witchcraft in the villages’ had spun out of control and provoked the ‘terrible abuses’ committed by guerilla base commanders.80 In fact, these situations repeated themselves throughout the territory during the war causing ‘embarrassing scenes’. What most surprised party leaders was the fact that the perpetrators far from being ‘peasants’ originated from ‘mod- est families in Bissau’ many of them having been educated at missionary schools and in the party’s school in Conakry, and within a short period of time had transformed themselves into ‘bloodthirsty commanders’

78 De Jong, A Descent; Carlos Cardoso, “Ki Yang-Yang: uma nova religião dos Balantas?”, Soronda, 10, 1990, p. 3–15; Roy van der Drift, Arbeid en Alcohol, pp. 67–70; Marina Themudo, “From the Margins of the State to the Presidential Palace: the Balanta case in Guinea Bissau”, African Studies Review, 52, 2, 2009, especially p. 55. Significant in this respect are the links that developed between the Kriston as petty traders and planters, and the Balanta-Brassa as producers, during the colonial era and above all in the South, i.e. the Quínara and Tombalí regions. See Drift, Arbeid en Alcohol, p. 5–76. One of the most famous of the PAIGC commanders, João Bernardo ‘Nino Vieira - who later became president of independent Guinea Bissau by means of a coup in 1980, had his base on the Southern ‘front’ where the Balanta had migrated to and settled in large numbers during the colonial period and were mobilized to the nationalist cause - boasted strong Kriston- Pepel roots. 79 Cabral, Crónica, p. 158. 80 Cabral, Crónica, pp. 176–77. 50 philip j. havik terrorizing entire populations.81 But, as noted above in the statement made by a PAIGC leader, Cabral himself had given one of his maquisards money to buy a mezinho or gris-gris prepared by an Islamized healer or muru which all of them wore under their uniform to protect them against enemy fire. His brother, Luís Cabral, commented that with time as fighters equipped with guarda di kurpu fell in battle, they would end up by only trusting modern weapons rather than succumb to such ‘weaknesses’.82 One of the main party cadres at the time remarked that such phenomena would gradually disappear with the creation of regular armed forces under a central command. Such measures were meant to combat tendencies towards ‘localism’, ‘tribalism’ and ‘militarism’ which clashed with national goals and reinforce the movement’s transformation into a party and liberated areas into an embryonic state.83 Other – Capeverdean – party leaders maintained that the day to day interaction between them and Guineans caused the latter to “change their habitus and shed certain beliefs”.84 These events and the deliberations held at the time have been regarded as one of the most decisive moments of the war, given that they triggered the radical modernization, reorganization and ideological shift of a political movement molding it into a ‘national party’ aspiring to governance. But as Cunningham noted ‘the extent to which one can speak of a national consciousness seems ‘problematical’ given the ‘ethnic’ and age based vec- tors that characterized local adherence to the movement. The same author suggests that by using age sets as vehicles for mobilization and ‘mimicking’ and appropriating the division between age groups and insti- tutionalizing it within the movement, the latter actually incorporated or subsumed Balanta-Brassa social structures within its ranks.85 As a result

81 Cabral, Crónica, p. 183. The measures taken by the movement to end these practices and the punishment meted out to the alleged culprits still remains largely untold. The accused of which only a few have since been identified were apparently imprisoned or executed. See Galli & Jones, Guinea Bissau, pp. 58–61; Lopes, A Transição, pp. 50–4; Dhada, Warriors, p. 20; Pereira, Uma Luta, um Partido, pp. 172–76; Álvaro Nóbrega, A Luta pelo Poder na Guiné Bissau, Lisbon: Instituto Superior de Ciências Sociais e Políticas (ISCSP)/ Universidade Técnica de Lisboa(UTL), 2003, pp. 199–204. 82 Cabral, Crónica, pp. 166–67. 83 Cabral, Unity and Struggle, pp. 86–89 & 224–26; Chabal, Amílcar Cabral, pp. 77–83; Pereira, Uma Luta, um Partido, pp. 172–76; José Julião de Sousa, “O fenómeno tribal, o tribalismo e a construção da identidade nacional no discurso de Amílcar Cabral”, in Luís Reis Torgal, Fernando Pimenta & Julião Soares Sousa (eds.), Comunidades Imaginadas. Nação e nacionalismo em África, Coimbra: Imprensa da Universidade de Coimbra, especially p.171. 84 Lopes, Os Bastidores, p. 149. 85 Cunningham, Guinea Bissau, 1956–1974, p. 5. virtual nations and failed states 51 conflicts inherent to those communities in the areas in question were also imported within the party. One may therefore conclude that rather than fighting ‘localisms’ and ‘tribalisms’ the party was actually exploiting these identities and practices for political ends while at the same time becoming part of them. Clearly then the ambivalence of nationalist discourse and practice mediating between different ‘ethnic’ cultures and national ‘publics’, made it permeable to ancestral perceptions of the spiritual world. In the case of the Balanta-Brassa spirit shrines or iran in their northern homelands were important sites of protection, divination or conflict resolution (e.g. theft of cattle), also during the war, which required the mediation of elders.86 Other instances of these ad-hoc alliances were recorded for the Northern Cacheu region where the Mama Jumbo cult based in the Caboi region amongst the Manjaco was ‘crucial to the success of nationalist forces during the Independence war’.87 Again, the leadership was seen to exploit ritual mechanisms of rallying local support by authorizing the purchase of sacrificial animals and libation for a local shrine in order to ‘bolster the confidence and courage of fighters’. Indeed it was believed to have decisively contributed to the failure of Portuguese incursions in the area which became a no-go area for them: Cabral himself regarded the Cobiana spirit shrines in the Manjaco tchon as ‘nationalist’.88 In the process, besides increasing local adherence, some of the movement’s leaders also became ‘regular clients’ of the shrine, which thereby aug- mented its ‘symbolic appeal’ and succeeded in gaining a wider reputation beyond the Caboi region. Given that the main body of support for the movement came from ‘animist’ and Christianized communities in coastal areas, rather than from Islamized groups in the eastern savannah regions, the formers’ mobilization through the forging of ‘magical alliances’ was indispensable.89 Thus, in this perspective, national liberation did not merely imply the defeat of colonialism, but above all ‘the triumph of coastal spiritists’, also over their Islamized neighbours, the Fula and Mandinga.90 Negotiating contracts with the spirits then not only signified an alliance between

86 Interview with Balanta informant, Quínara, Guinea Bissau, 1989. 87 Eve L. Crowley, Contacts with the Spirits: religion, asylum and ethnic identity in the Cacheu region of Guinea Bissau (unpubl. PhD diss.), Ann Arbor: Yale University/UMI, 1990, p. 573. 88 Cabral, Unity and Struggle, p. 59. 89 On these strategies of spiritual alliances and witchcraft accusations during the anti- colonial war in Zimbabwe, see Alexander & McGregor, “Hunger, Violence”, pp. 78–79. 90 Crowley, Contacts with the Spirits, p. 575. 52 philip j. havik different social strata but also ‘between two cosmological orders and the spiritual resources they controlled’.91 The special powers that they conferred upon fighters reinforced the movement’s dependency upon shrines and elders’ support, as a ‘strategic asset’ in campaigns, while they also facilitated the establishment of affinities with other communities (e.g. Pepel, Mancanhe and Felupe/Djola) given their location close to ‘ethnic’ frontiers. In addition, these ‘public spirit shrines’ which were also accessible to strangers allowed for the inclusion of marginal outsiders, such as combatants.92 In fact, this strategy facilitated the reification of filhos da terra like Amílcar Cabral, who projected himself as a national and international leader, into cross-cultural icons. His assassination in 1973 – the circumstances and reasons for which are still the subject of dispute as well as the inquiry carried out by party cadres and its aftermath – further contributed to sanctify the leader’s aura and fuel the posthumous personality cult surrounding his leadership.93 These events and the silences and taboos surrounding them helped create the conditions for the longevity of the ‘linha de Cabral’ or ‘Cabral’s programme’ as a key reference of the erstwhile liberation movement and Guinea-Bissauan politics in general until the late 1990s.

Civil Society and State-Building: The Fetishization of State Power

The changes that occurred during the war point towards a number of important shifts in the relations between a liberation movement with nationalist aspirations and its popular base, as well as between the different strata and communities involved. Having opted for rural rather than an urban mobilization after the Pidjiguity dockworkers’ strike in Bissau in

91 Crowley, Contacts with the Spirits, p. 575. 92 Crowley, Contacts with the Spirits, p. 583. 93 Castanheira, who attempts to reconstruct the events prior and after Cabral’s death, reveals that the documents of the two committees which investigated the assassination in Conakry could not be traced, describing Bissau as an ‘absolute void’ in this respect. He added that very little cooperation was forthcoming from informants and witnesses, probably owing to the fact that the killing was, at least in part, the result of an internal plot and carried out by party members. There is also a wider Lusophone, background to these committees which included high ranking cadres from other liberation movements; see for example the assassination of Eduardo Mondlane in Dar-es-Salaam in 1969; Amílcar Cabral delivered the first Eduardo Mondlane lecture entitled ‘National Liberation and Culture’ at Syracuse University in 1970. Castanheira, Quem Matou Amílcar Cabral, pp. 277–81. Also see Rudebeck, Guinea Bissau, pp. 112–17; Chabal, Amílcar Cabral, pp. 132–43; Pereira, Uma Luta, um Partido, pp. 210–22; Tomás, O Fazedor de Utopias, pp. 265–85. virtual nations and failed states 53

1959, the leadership consciously opted for an ‘ethnically oriented’ campaign, rather than following a ‘foco’ strategy as applied in the Cuban case which had dispensed with a prior ‘political’ mobilization campaign.94 This in turn created the conditions for the ‘translation’ of nationalist concepts of struggle into local ‘ethnic’ idiom, and concomitantly the reverse, i.e. the inscription of popular cosmology into the movement’s genealogy. In fact, rural communities’ initial reticence to contemplate armed resistance had to be overcome, given that they were by no means convinced of the possibility and success of openly challenging colonial authority.95 As a result, spiritual protection was a sine qua non for the adherence of populations to the nationalist cause and their taking an active part in the resistance. Indeed, the struggle actually contributed to accelerating the diffusion of mesiñu amongst communities, to the use of guarda di kurpu (also in Kriol: figa) from different sources crossing ‘ethnic’ lines, to the increasing ‘national’ character of ‘local’ shrines which gained a multi-ethnic clientele and to the spread of Guinean Creole or Kriol as a vehicle for contact and trans-culturation of ‘local’ into ‘national’ culture, and vice versa.96 Cabral’s crafting of metaphor when referring to the power of the forests which had been turned from much feared locations into the bases of the movement’s campaigns, thereby ‘transforming weaknesses into strength’,97 illustrated this strategy of ‘nationalising’ hitherto sacred places.98 What is interesting about Cabral’s philosophy of turning ‘weakness into strength’ was the underlying intention to ‘hide’ this inherent and apparent contradiction

94 See testimony of Cau Sambú who took part in the first mobilization campaigns, in Urdang, Fighting Two Colonialisms, pp. 86–92. On mobilization strategies, also see Basil Davidson, Révolution en Afrique: la libération de la Guinée Portugaise, Paris: Éditions du Seuil, 1969, pp. 49–66; Rudebeck, Guinea Bissau, pp. 230–48; Lino Bicari, “Relação Massas Populares – Poder no Processo de Libertação da Guiné Bissau” (unpubl. conference paper), Rome, 1983; Chabal, Amílcar Cabral, pp. 90–131; Forrest, Lineages, pp. 181–204. Also see Patrick Chabal, “National Liberation in Portuguese Guinea, 1956–1974”, African Affairs, 80, 318, 1981, pp. 75–99. For a critical assessment of the military aspects of the nationalist struggle/colonial war, see Dhada, “The Liberation War”. Cunningham argues that the role of the PAIGC political leadership in the August 1959 dock strike has been considerably overestimated; see James Cunningham, “Guinea Bissau,1956–1974: a reassessment”, (unpubl. conference paper), London: Institute of Commonwealth Studies, 1980, p. 3. 95 Interview with Balanta informant, Quínara, Guinea Bissau, 1991. 96 Interview with Biafada informant, Quínara, Guinea Bissau, 1989. 97 Amílcar Cabral, “Introduction”, in Davidson (ed.), Révolution en Afrique, p. 15. “C’est cela la lutte: transformer les faiblesses en forces.” 98 See Davidson, Révolution en Afrique, p. 80. When asked by Davidson whether the PAIGC appealed to magic, Amílcar Cabral answered ‘Yes, in the beginning, the amulets and charms, many practiced that kind of [magic]. But now they have understood that it’s better to have a good shelter and fire with precision’. 54 philip j. havik between the dependency of a nationalist movement upon the mentalities and practices it was supposed to combat. Notwithstanding, the curbing and suppression of ‘excesses’, such as the witch-hunts in the South for allegedly thwarting mobilization and the use of local cosmologies was a widely accepted – and inevitable – means of conducting the nationalist struggle and broadening its popular base. In fact, this form of support was an extension of the proverbial hospitality extended to stranger-fighters who also benefited from the food provided by local hosts as well as shelter and cures for treating illnesses and wounds.99 Given the internationalization of the movement especially after the Tri-Continental Conference in Havana in 1966 and its increased standing on a global scale, the gap between official discourse and local practice, and therefore between public and hidden transcripts, widened considerably. The process of creating ‘liberated areas’ with their own system of governance (i.e. the village committees) meant that local practice became part and parcel of political culture, in which envy and rivalries were regu- lated by taking recourse to healers, i.e. djambakós (from non-Islamized coastal groups) and murus (from Islamized groups in interior) and personal power and ambitions countered by witchcraft accusations. While mesiñu were commonly used to increase a person’s protection against futis or curses, contracts with the spirits could also be employed to accu- mulate wealth and conspicuous consumption or ronku. The spread of violence caused the use of ritualized spiritual protection to include modern firearms, which just like their wielders could be ‘blessed’.100 Such customs showed to what extent magic, witchcraft and ‘modernity’ were and are intertwined, and above all how popular culture appropriated and reinterpreted the symbols of modern warfare.101 As a result, ‘ritual

99 On the relationship between guerrillas and communities, see Jocelyn Alexander, Jo Ann McGregor & Terence Ranger, Violence and Memory: one hundred years in the dark forests of Matabeleland, Oxford: James Currey, 2000, pp. 75–79. 100 Interview with Biafada informant, Quínara, Guinea Bissau, 1989. 101 See Alexander & McGregor, “Hunger, Violence”; Brinkman, “Ways of Death”; and Sverker Finnstrom, “For God and My Life: war and cosmology in Northern Uganda”, in Richards (ed.), No Peace, No War, pp. 98–116. Also, see Geschiere, The Modernity of Witchcraft, ch. 4, especially p. 97–110; Cyprian F. Fisiy & Peter Geschiere, “Witchcraft, Development and Paranoia in Cameroon: interactions between popular, academic and state discourse”, in Moore, Henrietta L. & Sanders, Todd (eds), Magical Interpretations, Material Realities: modernity, witchcraft and the occult in Postcolonial Africa, London: Routledge, 2004, pp. 226–46 for a discussion of the changing discourse regarding djambe (witchcraft) among the Maka in Cameroon and the impact of multiparty elections and rivalries upon the ambivalence of popular and official perceptions towards accumulation virtual nations and failed states 55 rebellions’102 were imported into party politics, thus becoming part of Guinea’s ‘national(ist)’ culture. But rather than balancing each other out and strengthening the movement, the building of a state apparatus after independence, which was largely inherited from colonial times and grew alongside the military command structure built during the war, brought these existing, latent tensions to the centre of the post-colony. The establishment of the popular armed forced (FARP, Forças Armadas Revolucionários do Povo) after the Cassacá conference in 1964 served as a transmission belt for ‘local’ tensions to the ‘national’ level. At the same time it was not clear where the party ended and the state began; during the war, the party itself appeared to function as state.103 In the areas where they had been established during the war, local village committees, the so called ‘comités de tabanka’, which symbolized the notion of poder popular or popular power were neglected and ignored in favour of non-elected sector committees, regional councils and central governing bodies.104 The first elections held in liberated areas under UN supervision in 1972 confirmed the tendency towards popular consultation rather than active participation while illustrating the different levels of regional popular support.105 By settling accounts with régulos or appointed chiefs – some of whom were executed – in the East and North of the country, these differences were exacerbated as the power vacuum was filled with party cadres.106 Rather than sharing power and resources, as it had done during the war in liberated areas, the leadership opted for an authoritarian, one party system which increasingly distanced of wealth and power. Geschiere described how political struggles in Cameroon are translated into local discourses, e.g. on witchcraft, and thereby associated with the world of kinship and family; Geschiere, Modernity, p. 126. 102 See Max Gluckman, “Rituals of Rebellion in South-East Africa”, in M. Gluckman, Order and Rebellion in Tribal Africa, London: Cohen & West, 1963, pp. 110–38. 103 Rudebeck, Guinea Bissau, pp. 105–48; Lars Rudebeck, “Political Mobilisation for Development in Guinea Bissau”, Journal of Modern African Studies, 10, 1, 1972, especially p. 5. 104 Galli & Jones, Guinea Bissau, pp. 65–6 & pp. 81–2; Lars Rudebeck, “Kandjadja, Guinea Bissau, 1976–1986 Observations on the political economy of an African village”, Review of African Political Economy, 41, 1988, pp. 17–29. Rudebeck remarks that with the first elections for the national assembly held in liberated areas in 1972 under UN supervision, the ‘idea of a state separate from the party’ emerged; see Rudebeck, Guinea Bissau, pp. 150–56. For Rudebeck, the notion of poder popular grew out of political mobilisation, and was countered if not contradicted by the practice of democratic centralism at the top of the party. 105 Philip J. Havik, “Mundasson I Kambansa”, pp. 137–39. 106 Raúl M. Fernandes, “Partido Único e Poderes Tradicionais”, Soronda, INEP, Bissau, 16, 2003, especially pp. 44–45; Forrest, Lineages, p. 212. 56 philip j. havik itself from its original power base, i.e. rural communities.107 ‘The democratic principles of PAIGC were systematically disregarded in favour of internal conflicts of interest between militants from different social strata’.108 The ‘Movimento Reajustador’ led by the self-appointed president Nino Vieira in 1980 did not carry out any significant administrative reforms, but did alter popular perceptions of them.109 Significantly, at its first extraordinary congress after the coup held in 1981, the movement’s leadership declared that the restructuring of the party was to be carried out under its new designation as a ‘national party’.110 Locally, this ‘bonapartist’ concentration of powers111 and the distancing between state and society that resulted from it, was interpreted as non-sharing, as appropriation, and a violation of a contract made with the ‘partidu’ during the ‘luta’. The statement by a Balanta elder in the South of Guinea Bissau eloquently reflects this sentiment: We know that since it was established, the State stole the power from our committees. During the guerilla war, the school, the health post, the shop, the militia, all depended upon us, committees. But now, to the contrary, any small civil servant can do what he wants and we cannot inter vene (nõ boca ká está lá = our voice is not heard there). After the 14th of November (the 1980 coup) we heard on the radio that it is necessary to reorient the party according to Cabral’s views, because it strayed,

107 Lopes, Etnia, Estado e Relações de Poder, pp. 69–97; Lars Rudebeck, Problèmes de Pouvoir Populaire et de Développement: transition difficile en Guiné Bissau, Uppsala: SIAS, 1982; Rudebeck, “Kandjadja”, and Lars Rudebeck, Conditions of People’s Development in Post-Colonial Africa, Uppsala, AKUT, 1990; Bicari, “Relação Massas Populares-Poder”; Galli & Jones, Guinea Bissau, pp. 72–108; Lopes, A Transição Histórica; Havik, “Mundasson i Kambansa”, pp. 280–84; Nóbrega, A Luta pelo Poder, pp. 213–39; Forrest, Lineages, pp. 205–21. 108 Lopes, Etnia, Estado e Relações de Poder, p. 73. 109 Lopes, Etnia, Estado e Relações de Poder, p. 71. The said ‘readjustment movement’ whose leadership was organized in the ‘Council of the Revolution’ aimed at ‘safeguarding the revolutionary conquests of the armed struggle’, ‘revitalizing the party’, its ‘reaffirma- tion as the sole governing party’, the ‘reinforcing of its internal unity and cohesion’, and ‘the defense of its ideological purity’. It therefore proposed establishing a ‘revolutionary national democracy’ and the intensification of ‘politico-ideological work’ amongst party militants in accordance with ‘revolutionary morality’. It focused on a strategy that had agriculture as its priority and the driving force of the economy, in contrast to the failed industrialisation projects of the pre-coup regime, symbolising a return to the ‘linha de Cabral’, i.e. Amilcar Cabral’s original programme. I Congresso Extraordinario do PAIGC, Bissau, 8–14 November, 1981. 110 I Congresso Extraordinario do PAIGC, Bissau, 8–14 November, 1981. 111 Raúl M. Fernandes, “Processo Democrático na Guiné Bissau”, Soronda, 17, 1994, pp. 31–43. The author identifies a centralised, ‘bonapartist’ party (the PAIGC) and a con- comitant ‘bonapartist’ presidency, tendencies that became pronounced with one party rule after independence. virtual nations and failed states 57

which is true, but do you wish to follow Cabral’s line? Then, return the power that the State took from us: if we recover it, then we will also have responsibility.112 The negation of sharing and reciprocity or torna profoundly rooted in African culture provoked a deep resentment amongst the party’s erstwhile supporters, of which the ‘antigu kombatenti’ (former guerillas) – who had their own ministry – were the most visible and vociferous.113 Their prestige was not only based upon their having ‘liberated’ the country, but more importantly upon ‘having shared’ their lives and livelihoods with the population during the struggle. But at the local level, the battle- ground of the luta, the idea of change or kambansa (Kriol metaphor for change; from crossing a river) which was associated with independence and the promise of a better future in the context of popular democracy or ‘poder popular’ never materialized.114 Thus the notion of sharing which was deeply inscribed in the movement’s programme and formed one of the main pillars of the idea of the nation as a political brother hood and community, shipwrecked on the ‘democracy of the possible’.115 The first signs of this shift were already visible before independence, e.g. during the party’s second congress in Madina do Boé in 1973, when Cabral’s assassination was invoked in order to, significantly, justify the change of the party’s motto into ‘Unity, Struggle and Vigilance’.116 Given that no support was forthcoming from the party/state while sacrifices (i.e. the National Reconstruction Tax), low producer prices and other obligations (such as an increase in productivity and delivery of crops) were demanded from populations which were left to fend for themselves, rural communities refused to ‘share’ their excess produce with the state controlled trade outlets, the People’s Stores.117 The fact that a legal framework for local councils was not created and local elections were not held is a clear indication of this one-way policy. As such, the post-colony appears as a continuation of its colonial predecessor, in its failure to redis- tribute resources and to integrate the population within a participative

112 Bicari, “Relações Massas Populares-Poder”, p. 22. 113 For a discussion of the notion of torna, see Trajano, Polymorphic Creoledom, pp. 548–9. 114 Rudebeck, Problèmes de Pouvoir Populaire, pp. 63–5. 115 Quote from interview with former president Luís Cabral, Visão, 13–1-1994. 116 See PAIGC, II Congresso do Partido, 18–22 Julho 1973, Regiões Libertadas do Oeste, p. 14. 117 Galli & Jones, Guinea Bissau, pp. 109–121; Rosemary Galli, “On Peasant Productivity: the case of Guinea Bissau”, Development and Change, 18, 1, 1987, pp. 69–98. 58 philip j. havik political culture, despite these goals forming an integral part of its political platform. The appropriation and mediation of cultural symbols and practices which had long been the almost exclusive domain of Kriston strata, had by that time become a generalized feature of Guinea’s emerging ‘civil society’ as a result of the joining of large numbers of rural cadres with a variety of cultural roots into the party.118 The above mentioned authors have eloquently indicated to what degree the views of the latter clashed with those of urban based ‘intellectuals’ and political commissars that had concentrated power in the capital Bissau.119 Some have also pointed to the role of a ‘Creole elite’ acting as an intermediary between the intellectual and the military, rural base of the party.120 Although the perspective of the ‘peasantry’ differed from the Kriston and Capeverdean cadres or ‘elites’ that had pioneered the luta, there were other distinctions that marked their different strands of nationalism and political culture.121 Once aborted, the ‘two countries, one nation’ project laid the foundations for a ‘coming of age’ of latent Guinean nationalism and its aspirations

118 For an overview of the role of the Kriston stratum during the (early) colonial era, see Trajano, Polymorphic Creoledom, pp. 228–313, and Philip J. Havik, “Planters, Traders and Go-betweens: The Kriston in Portuguese Guinea”, Portuguese Studies Review, 19, 1 & 2, 2011, pp. 197–226. The Kriston’s important role as mediators between the rural and urban areas in the context of the anti-colonial resistance and the PAIGC in particular, has so far not been the subject of sustained research; see Fernandes, “Partido Único”, pp. 40–1. For a recent attempt at delineating the involvement of these strata in the process of anti-colonial resistance, see Wilson Trajano Filho, “From an exclusive elite to nation-builders: the various faces of creoledom in Guinea-Bissau”, (unpubl. Conference paper), European Association of Social Anthropologists, Lubljana, 2008. 119 Lopes, Etnia, Estado e Relações de Poder, and A Transição Histórica; Rudebeck, Problèmes de Pouvoir Populaire; Rudebeck “Kandjadja”; and Rudebeck, On Democracy and Sustainability: transition in Guinea Bissau, Stockholm: SIDA Studies, 4, 2001; Patrick Chabal, “Revolutionary Democracy in Africa: the case of Guinea Bissau”, in P. Chabal (ed.), Political Domination in Africa: Reflections on the Limits of Power, Cambridge: Cambridge University Press, 1986, pp. 84–108; Galli & Jones, Guinea Bissau, pp.78–90; Forrest, “Guinea Bissau since Independence” and Lineages. 120 Fernandes, “Partido Único”, p. 41; Trajano, Polymorphic Creoledom, p. 84; Carlos Cardoso, A Formação da Elite Politica da Guiné-Bissau, Lisbon: Centro de Estudos Africanos (CEA)-ISCTE, 2000; João Ribeiro Butiam Có, “Representação e confinação de estruturas sociais na Guiné-Bissau: uma abordagem sobre conflitos e consensos”, Working Paper 10, Lisbon: Instituto Superior de Economia e Gestão (ISEG), 2010 121 The Movimento para a Libertação da Guiné (MLG) founded in 1958 – also known as the União Patriotica da Guiné – later merged into the UPANG (União Patriotica Anti- Neo-Colonialista da Guiné Bissau). In 1977 it was strongly imbued with this ‘autochtonous’ current of Guinean nationalism, presenting itself as truly Guinean (‘filhos autenticos da Guiné’) and firmly opposing what it saw as the Cape Verdean ‘re-colonisation’ of the country; see União Patriótica Anti-Neocolonialista da Guiné Bissau (UPANG), Manifesto, Bissau, 1976. virtual nations and failed states 59 for effective political independence which was seen by some as a step on the road to becoming a ‘movement of maturity’.122 But the period of one-party rule (1974–1994) was characterized by a number of trends which link the projects of the nation with a certain form of authoritarian rule and belie attempts to brand it as a function of Capeverdean-Guinean antagonism. The tendency towards denial and secrecy which epitomized the ruling party and its cadres imbued with an autocratic ‘governmentality’ has been characterized as a cosmology of unaccountability.123 While the roots of the concentration of power are already present during the luta, the fetishization of state power and the monopolisation of national values by the single party which characterized the first decades of independence until the multiparty elections in 1994124 dislocated the notion of nationhood from its erstwhile popular base. Party congresses were little more than rituals of representation, akin to Mbembe’s ‘simulacre’, 125 that indissolubly linked the idea of nation to the party-state while the poder popular was reduced to ‘ideological fiction’.126 Thus the party’s claim to ‘national liberation’ which had been laboriously forged during the war became orphaned from its historical base precisely because of the state’s growing autonomy.127 Being a colonial creation, the post-independent variant of the state soon revived memories of a recent past. The authoritarian, one-party regime as well as the personalisation of governance after independ ence transformed the state into a curse or futis, which contaminated party

122 Basil Davidson, No Fist is Big Enough to Hide the Sky, London: Zed Books, 1981; Robert Buijtenhuys, “People’s War in Africa: the quest for ‘movements of maturity”, Africa, 59, 3, 1989, pp. 381–90. 123 Mbembe, On the Postcolony, p. 111. 124 For an assessment of PAIGC as single party, see Lopes, Etnia, Estado e Relações de Poder; Rudebeck, Problèmes de Pouvoir Populaire; Patrick Chabal, “Party, State and Socialism in Guinea Bissau”, Canadian Journal of African Studies, 17, 2, 1983, pp. 189–210. Joshua B. Forrest, “Guinea Bissau since Independence: a decade of domestic power struggles”, Journal of Modern African Studies, 25, 1, 1987, pp. 95–116; and Nóbrega, A Luta pelo Poder. See also Rui Jorge da Conceição Gomes Semedo, PAIGC: a face do monopartidar- ismo na Guiné Bissau (1974–1990), MA thesis, São Carlos (Brazil): Universidade Federal de São Carlos, 2009. 125 Mbembe argues (On the Postcolony, p. 111) that the relations between rulers and rules are enacted through the simulacrum, whereby the popular challenge to governmental power and its hold over society (commandement) demystifies it and its legitimacy by using the “signs and language of officialdom”, thus creating “potholes of indiscipline”, without however undermining the material base of state power. 126 See Buitenhuijs, “Peoples’ War in Africa”, p. 384. 127 See Joshua B. Forrest, “The Quest for State ‘Hardness’ in Africa”, Comparative Politics, 20, 4, 1988, pp. 423–42. 60 philip j. havik and its leaders and effectively neutralised the ‘national-liberationist magic’. The largely fictional post-colonial ‘nation-state’ thus came to be deconstructed by the very state and party which invoked it.128 The idea of nationhood perfunctorily reproduced in official discourse and legalist for- mula, became marginalized, retaining only its connection with the shared struggle and sacrifice from the period of the luta relegated to an almost religious mysticism in the haunted present. By obsessively cultivating the silence of power while personalizing its exercise, the regime developed a statist cosmology based upon secrecy and exclusion that, devoid of charismatic rule, undermined its own legitimacy. The increased presence of urban cadres in the party and the national assembly associated with public administration, the armed forces and the secret services129 contributed to the institutionalization of this cultivation of state power. Its arbitrary exercise and obsessive ritualization were likened by Guinean opponents of the one-party state to a pathological frame of mind.130 Attempts to reconcile ‘the party’ with ‘the people’ during the first decades of single party rule and to redeem the former’s political distance by renewing the ‘national contract’ with the ‘popular masses’ that had emanated from the war of liberation were manifold. The best known examples are the opening of mass graves after the 1980 coup and the search for ancestral relics buried during the colonial wars of ‘pacification’.131 However, continued authoritarian rule, the change of policies towards economic liberalization and a series of high profile events called the notion of ‘national reconstruction’ into question, the most significant of which was the crisis resulting from the 17th October 1985 when a purported coup attempt was foiled and its purported leaders imprisoned, tortured and executed. The crucial question was not from what ‘ethnic’

128 See Iain Walker, “What Came First, the Nation or the State? Political process in the Comoro Islands”, Africa, 77, 4, 2007, pp. 582–605, on the process of (de)construction of the nation in the case of the Comoro Islands, where the state is a largely irrelevant entity. 129 Forrest, Lineages, p. 229; Havik, “Mundasson i Kambansa”, p. 153. 130 União Patriótica Anti-Neocolonialista da Guiné Bissau (UPANG), Manifesto, Bissau, 1976, p. 15. Bayart refutes the idea of the pathological nature of power in the context of the seemingly incompatible combination of the ‘unfinished’, ‘weak’ state with the intricate maze of social networks that underpins its centralization; see Jean-François Bayart, The State in Africa, pp. 261–2. 131 See the case of the search for the ‘navio espada’, a canoe, filled with treasures, which had supposedly been buried in the Pepel tchon during the ‘pacification’ campaigns against the Portuguese which lasted from 1890 to 1915; about these wars, see Pélissier, Naissance de la Guiné. Given that the then president had Pepel roots, the gesture was interpreted as an ‘ethnic’ one and an attempt to seek an entente with the djagras, i.e. chiefly authorities representing ruling lineages; also see Tomás, O Fazedor de Utopias, pp. 305–6. virtual nations and failed states 61 group they originated (i.e. the Balanta), an issue largely exploited by the party and state apparatus, but that the victims were former guerilla fighters who had earned their reputation in the bush. According to the then president, the coup was a challenge to the ‘glorious party’ and Guinea’s ‘national independence and identity’,132 which exposed serious contradictions within the one party state. However, the accusation of political treason was justified on the grounds of the surreptitious ‘ethnifi- cation’ of relations between party and state, figuring as an attempt to instrumentalise the notion of ‘ethnicity’ in the sense of ‘political tribalism’.133 The plotters had allegedly undermined the ‘cohesion of the armed forces’ as well as bringing to the fore the ‘negative aspects of the relations of ethnic solidarity that were obstructing the socio-economic development of the country’.134 At the same time, these ‘ethnic’ tendencies were associated with the Yang-Yang movement, an anti-witchcraft cult which had simultaneously emerged amongst mostly young women in Balanta-Brassa communities in the South of Guinea Bissau and had allegedly ‘corrupted’ the minds of some personalities in party and state who had then attempted to exploit the unrest amongst rural populations.135 The events that preceded the imprisonment of members of the PAIGC leadership at the highest level show that the growing following in the South was regarded as a matter of ‘national’ security. The supernatural powers attributed to the cult’s female leaders, its rapid expansion towards the Centre and North, and its purported penetration of state and party translated into the rounding up of its supposed protagonists and the mass arrest of alleged coup conspira- tors.136 It was precisely at this point – where the contradictions between public discourse and hidden meanings of power were exposed – that the single-party regime decided to seek internal support as resistance grew. Elders and chiefly lineages in the North and East were approached and

132 PAIGC, IV Congress, Bissau,1986, Nô Pintcha, 16–7-1986 & 19–7-1986. 133 See John Lonsdale, “Moral Ethnicity and Political Tribalism”, in Preben Kaarsholm & Jan Hultin (eds), Inventions and Boundaries: historical and anthropological approaches to the study of ethnicity and nationalism, Roskilde: Roskilde University, 1994, pp. 131–50. 134 See the reports to the IV Congress of PAIGC which blamed ‘ideological failings, opportunism and ambition’ for the 1985 crisis; Mensagem das Forças de Segurança, Nô Pintcha, 6–9-1986. 135 For an analysis of the Yang-Yang (meaning shadow in Balanta) in the tradition of ‘ritual rebellions’ as proposed by Gluckman (1963), see De Jong, A Descent, pp. 81–3. See Inger Callewaert, “O Surgimento da Prática ritual Ki-Yang-Yang na Forma de Vida Balanta”, Soronda, Nova Série, 1, 1, 1997, pp. 79–120. 136 Fernandes, “Partido Único”, p. 47; Zamora Induta, Guiné: 24 anos de independência, 1974–1998, Lisbon: Hugin, 2001, pp. 26–33. Geschiere illustrates the post-independent state’s heightened sensitivity of witchcraft and its proliferation by quoting a Cameroonian 62 philip j. havik subsequently reinstated with official consent.137 The ‘shadow’ had broken the spell of silence.

Pillars of Meaning: Making Sense of Crises

Coming to terms with a situation of economic and political failure the one party regime was obliged to introduce reforms that unlocked the hitherto state controlled economy to private actors.138 It is significant that this policy change occurred immediately after the violent repression of the Yang- Yang cult and the resurgence of PAIGC’s familiar anti-tribalist rhetoric and the renewed emphasis on the need for national unity. Smuggling or clandô in Kriol vernacular had become a common phenomenon after 1974 in response to state controls over the economy, reflecting strong popular resistance to the party-state. However, the policy shift towards economic liberalization initiated in the mid 1980s also created the conditions for a rapid spread of the consultation of healers and spirit shrines (irã or china) throughout the country as well as in the capital Bissau.139 Guinean Creole idiom well illustrates this cosmological context. The ‘magical market’ proved to be extremely flexible and quick to adapt to economic monetization and modern consumption, as accusations of corruption (suku di bas) and ostentation (ronku) proliferated. Soon, the informant who states that: ‘Now we Africans are in charge, we know that witchcraft is real’ (Geschiere, The Modernity of Witchcraft, 1997, pp. 216–17). 137 The societies whose chiefly authorities or régulos were reinstated were all centralised/cephalous or stratified, such as the Manjaco and Fula, while the segmentary or state- less societies such as the Balanta were, obviously, excluded from these arrangements. For an account of the re-investiture of Manjaco chiefs, see Eve L. Crowley, Contracts with the Spirits, and Eve L. Crowley, “Regional Spirit Shrines and Ethnic Relations in Guinea Bissau”, Africana Journal, XVII, 1998, pp. 27–39; Clara Carvalho, Ritos de Poder e a Recriação da Tradição: os régulos Manjacos da Guiné Bissau (unpubl. PhD diss.), Lisbon: Instituto Superior de Ciencias do Trabalho e da Empresa (ISCTE), 1998; Clara Carvalho, “A revital- ização do poder tradicional e os regulados manjaco da Guiné-Bissau”, Soronda, Nova Série, 7, 2003, pp. 7–44; Clara Carvalho, “Local Authorities or Local Power?”, Soronda, special issue, 2008, pp. 39–56. Also see Eric Gable, “The Decolonization of Consciousness: local skeptics and the will to be modern in a West African village”, American Ethnologist, 22, 2, 1995, pp. 242–57, and Eric Gable, “Manjaco Rulers after a Revolution”, Africa, 73, 1, 2003, pp. 88–112. And also Forrest, Lineages, ch. 10. 138 On these reforms and their impact, see Faustino Imbali (ed.), Os efeitos sócio- económicos do programa de ajustamento estrutural na Guiné-Bissau, Bissau: INEP, 1993; and Isaac Monteiro (ed.), O Programa de Ajustamento Estrutural na Guiné Bissau: análise dos efeitos socio-económicos, Bissau: INEP, 1996. 139 Mamadú Jao, “Os Poderes Tradicionais no Período de Transição”, in Fafali Koudawo & Peter K. Mendy (eds), Pluralismo Politico Político na Guiné Bissau: uma tran- sição em curso, Bissau: INEP, 1996, pp. 123–33. virtual nations and failed states 63 programme to privatize the public sector and axe a few thousand jobs in the civil service supported by the World Bank led officials who felt threatened to counteract any attempt by seeking ‘magical’ protection.140 Resistance and magical counter strategies to the policy – which led to the cutting of social benefits exclusive to the civil service such as the well- known monthly income supplements in the form of bags of rice – played an important role in reducing its impact.141 On the other hand, these measures encouraged large scale rural-urban migration as well as out- and emigration amongst younger generations seeking their fortunes not only in neighbouring countries such as Senegal but also beyond, in Europe.142 The idea of ‘dúbria’ (to get along) emerged as a response to increasing hardships and poverty, replacing the certainty of state jobs and the security of kinship as popular slogans by the skills associated with ‘creative survival’143 and ‘social becoming’.144 Originally limited to young generations, it became an urban pastime for everyone trying to it underlines djiresa (cunning and trickery) and maçundadi (courage), while acting as a counterpoint to the national expression ‘djitu ka ten’ (there’s no way out) which expressed resignation and distance.145

140 On this issue, see for example Carlos Cardoso & Faustino Imbali, “As Questões Institucionais e o Programa de Ajustamento Estrutural na Guiné”, in Imbali (ed.), Os Efeitos, p. 58. The end result was that little was actually achieved as the programme soon ground to a halt; another problem was the lack of reliable data on (the flux of ) state sector jobs. For a discussion of the relationship between national politics and the occult, see Geschiere, The Modernity of Witchcraft, pp. 97–130. 141 Interviews with informants in Bissau, 1988–1995, and personal observation. 142 Since the 1800s when export crops such as peanuts (mankara) were introduced in the Senegambian region, the (male) out-migration of the navetanes (navet: wind) towards Senegal and the Gambia became a seasonal flux, above all for accumulating resources for ritual obligations such as the brideprice. For the migration of the Mancanha, see Mamadú Jao, “Origem Étnica e Migração entre os Mancanha da Guiné Bissau”, Soronda, Nova Série, 6, 2003, pp. 107–17. 143 For the concept of dubriagem, see Vigh, Navigating the Terrains of War, pp. 128–39, and Wilson Trajano Filho, “O Precário Equilibrio entre Improvisação e Regras: reflexões sobre a cultura política da Guiné Bissau”, Revista de Antropologia, 51, 1, 2008, especially pp. 250–51. Originating from the French se débrouiller, it is defined locally by Kriol speakers as getting by through shrewdness and skill in order to extricate oneself from a difficult situation, for example poverty or war, and counter mufunesa or bad luck and unhappiness while improving one’s prospects. But as Trajano points out, it also implies that those that act within this framework, the dubriaduris, place themselves ‘beyond the system’ by emi- grating and seeking their fortunes elsewhere. For the question of ‘creative survival’, see A.B. Zack-Williams, “Crisis, Structural Adjustment and Creative Survival in Sierra Leone”, Africa Development, XVIII, 1, 1993, pp. 53–65. 144 See Vigh, Navigating, ch. 5. 145 Themudo reports from the South of Guinea Bissau how a popular Creole saying Nô sinta, nô cala, nô péra son ! (i.e. we sit, we remain silent and we just wait!) was commonly 64 philip j. havik

The main reason for the emergence of creative survival as a modern way of life is precisely the weakening of the reciprocity inherent in the bonds of kinship and relatedness between rural societies and urban areas.146 There is a spatial and social differentiation at work here, given that the philosophy of getting by is typically associated with young males in an urban environment, while mbida, i.e. making a living, is related to women’s strategies for income generation.147 The rapidly growing urban populace (with about one third of the Guinean population of just over 1,500.000 living in the capital Bissau) has dented families’ traditional hospitality or osprindadi, to the extent that mutual-aid (djunta-môn) – common between members of lineages, compounds and communities – has been increasingly overshadowed by monetization and the privatization of interests.148 Despite the large numbers involved, the mandjuandadi networks (based upon age sets and rites of passage) proved to be particularly resilient in this respect.149 In rural areas, people had become accustomed and adapted to self-reliance strategies in response to the lack of aid and support from Bissau: “We have no state here. We ourselves are the state”.150 The most “striking feature” of such rural communities was the “actual absence of party and state” and the

used to express people’s attitude towards the events in the capital, Bissau; see Marina Themudo, “Cultura, Agri-Cultura e Cultura Política no Sul da Guiné Bissau”, Lusotopie, 8, 2, 2006, especially p. 142. On the role of youth in urban areas, see Silvia Roque & Katia Cardoso, “Entre a Marginalização e a Securitizaçaõ: jovens e violências em Cabo Verde e na Guiné Bissau”, conference paper, 7° Congresso Ibérico de Estudos Africanos, 2011, accessed at http://www.ces.uc.pt/myces/UserFiles/livros/716_Roque_Cardoso%202011 .pdf 146 On the question of the marginalization of youth and social change in the Bijagós Islands, see Lorenzo I. Bordonaro, Living at the Margins: youth and modernity in the Bijagó islands (Guinea Bissau), unpubl. Ph.D diss, Lisbon: ISCTE, 2006. 147 See Philip J. Havik, “Female Entrepreneurship in a Changing Environment: gender, kinship and trade in the Guinea Bissau region”, in C. Risseeuw & K. Ganesh (eds), Negotiation and Social Space: a gendered analysis of changing kin and security networks in South Asia and Sub-Saharan Africa, New-Delhi: Sage, 1998, especially pp. 205–25. The female variant of dubria is typically associated with prostitution. 148 Carlos Cardoso & Faustino Imbali, “A Família”, in Monteiro (ed.), O Programa de Ajustamento, especially p. 231. 149 Trajano, Polymorphic Creoledom, pp. 314–405; Fafali Koudawo, “Sociedade Civil e transição pluralista na Guiné Bissau”, in Koudawo & Peter Karibe Mendy (eds), Pluralismo Politico na Guiné Bissau: uma transição em curso, Bissau: INEP, especially pp. 108–109; Fafali Koudawo, “La Guerre des Mandjuas: crise de gouvernance et implosion d’un modèle de résolution de crises”, Soronda, INEP, Bissau, Número Especial 7 de Junho, especially pp. 168–72; Mamadú Jao, “Uma Leitura do Conflito Guineense”, Soronda, Número Especial 7 de Junho, 2000, especially pp. 116–21. 150 Rudebeck, “Kandjadja”, p. 18. virtual nations and failed states 65 notable authority exercised by local institutions and lineages as well as their cultural autonomy.151 The attempt of the leadership to capitalize upon social networks such as the mandjuandades, reinstate régulos or chiefs and allow public discussion and legalise political movements (from 1991), had also highlighted the crisis of political authority at national level, as well as its need to rally public support.152 These events served to project dissension and rivalry at the highest political level onto the public screen ending the hitherto dominant policy of ritualised nationalist discourse occasionally interrupted by political quarrels, violent incidents and purported coups. The contested and often confused run-up to the multiparty elections and the con- comitant proliferation of popular associations and political parties and slogans was conducive to the bringing of hitherto marginalized or ignored issues and memories into the public domain, as the grip of secrecy loos- ened. Thus, it was by no means surprising that the multi-party (and presidential) elections in 1994 and the ‘Bissau war’ of 1998–99 would reopen Pandora’s box of djanfa (trickery).153 Whereas the first parliamentary and presidential elections in 1994 had confirmed the prominence of the former single party and its leader, it also gave rise to new alliances and divisions in a climate of expectation.154 However, it soon became clear that despite the profusion of political manifestos and alternatives, elections did not imply the announced – and much hoped for – kambansa, i.e. change, but rather a gradual paralysis of the state apparatus and more intense political infighting and internecine conflict between an increasing number of rival parties and factions, that, but for a few exceptions emanated from the former single party.

151 Basing himself on the observation of local Manjaco informants, Gable finds that the kambatch initiation ceremony amongst the Manjaco of Basserel is compared to PAIGC party congresses; see Gable, “The Decolonization of Consciousness”, p. 244. 152 This transformation of traditional institutions into tools for state-led patrimonialism, were summarised by Koudawo as having produced a “republic of the mandjuas”; see Koudawo, “La Guerre des Mandjua”, p. 170. Similar strategies (such as the reinstatement of chiefs) were put in place, for example, by the governing FRELIMO party in Mozambique following the 1992 peace treaty which ended the protracted civil war (1975–1992); see Harry West, Kupilikula: governance and invisible realm in Mozambique, Chicago: University of Chicago Press, 2002, pp. 195–206. 153 ‘O PRS é agora um clube de feiticeiros’, i.e. ‘the PRS (the Partido de Renovação Social), is now a club of witch doctors’; Sambuia Caeiro Nevada, “A Luta de Espetezas no seio do Partido do Dr. Kumba Yala”, Foros de Debate, 19/4/2007 (www.guiné-bissau.com). Djanfa appears here as the opposite of filanta (to agree, to compromise). 154 Carlos Sangreman, Fernando Sousa Jr, Guilherme Severino & Miguel Barros, A Evolução Politica Recente na Guiné Bissau: as eleições presidenciais de 2005, os conflictos, o desenvolvimento a sociedade civil, Lisbon: Centro de Estudos sobre África (CESA), 2006. 66 philip j. havik

The wave of discontent soon spread to the Armed Forces which building upon their reputation and as ‘Combatentes da Liberdade da Pátria’ whose legitimacy derived from the ‘Luta de Libertação Nacional’, challenged the sitting president. The military Junta which emerged during the war and eventually ousted the then president Nino Vieira from power after a number of long drawn out battles, claimed to avert a repetition of previous crises which led to the 1980 coup and the alleged 1985 conspiracy.155 Although the outbreak of the ‘Bissau war’ in 1998 elicited already faint memories of the colonial war, these were soon overshadowed by the apparent senseless nature of the conflict fought between personalities and factions of the former ruling party. A clear distinction was made between the nationalist luta and the Bissau guera.156 On the ground’ the urban exodus also provoked a massive recourse to the hospitality and mutual aid of kin networks as well as spiritual aid in order to pinpoint problems, settle differences and seek protection against attacks by purported enemies and rivals. The displacement of more than three hundred thousand inhabitants of the capital Bissau who sought safety and shelter in the interior, reconnected the capital with its hinterland but also highlighted strains on the capacity for providing care and sharing resources. The warm reception of refugees in Guinea emulated the ‘generosity of villagers [which was] stretched to the limit’ that Ferme witnessed during the urban exodus in war-torn Sierra Leone.157 These events would not only further weaken the already ‘weak state’ and relocate the armed forces at the centre of events, but also produce a shift in official political discourse and public perception. The Guinean armed forces (FARP) which had gained direct access to state power with the 1980 coup but whose links with the PAIGC were, at least formally, severed with

155 On the 1998–99 war, see Roy van der Drift, “Democracy: Legitimate Warfare in Guinea Bissau”, Lusotopie, 1999, pp. 225–40; the special issue of Soronda, dedicated the war; and Induta, Guiné, (p. 47). The book by Zamora Induta, who joined the Junta Militar, and which is the first publication by a member of the armed forces on the conflict, provides an account of the 1999 war and the events that preceded it from independence onwards, includes a number of facsimile documents, e.g. the report of the commission of the Guinean National Assembly (ANP) on the arms trafficking with the Casamance MDFC rebels that precipitated the outbreak of fighting (Bissau, 8–6-1998). 156 Themudo, “Cultura”, p. 139; Vigh, Navigating, pp. 68–73; Henrik Vigh, “Conflictual Motion and Political Inertia: on rebellions and revolutions in Bissau and beyond”, African Studies Review, 52, 2, 2009, especially pp. 144–45. 157 See Ferme, The Underneath of Things, p. 228. Rudebeck, On Democracy, p. 94, argues for ‘innovative measures’ which would include ‘support to self organization in the countryside’ and ‘community based self-help projects in education, health and economic improvement’ in order to ‘make democracy work’ at ‘the grassroots’. virtual nations and failed states 67 the new constitution of 1991, now, just as in 1980, momentarily, posed as the saviours of the country.158 The parliamentary report about the illegal arms trade between Guinea Bissau and the Casamance independence movement in southern Senegal which formed a pretext for the initiation of hostilities provided a novel perspective of political transparency legitimized by popular representation.159 By implicating the then president Nino Vieira as well as high ranking politicians and members of the armed forces in arms trafficking and their cover-up, the report publicly exposed and confirmed political contentious political issues and suspicions that had been that had been circulating for quite some time in the form of rumours or bokasiñu and newspaper reports.160 The ‘historical’ taboos that long continued to determine political and academic discourse, including the role of the armed forces as the (last) heirs of the revolution and the guardians of Guinean nationhood, were eventually challenged with multi-party politics and the 1998–99 war. Rather than guardians of the nation, they ‘officially’ emerged from the ‘guera di hermonia’ or ‘brotherly war’161 as a deeply divided group, affected by numerous scandals and personal rivalries, i.e. as a source of conflict rather than unity. In this respect, two metaphors associated with the conflict stand out in terms of their heavily charged symbolic character in which public and hidden transcripts are intertwined, i.e. the notion of firkidja and the symbolism of the polón di Brá. Firkidja was one of the political platforms to

158 Not only the military are deeply divided, but so is public opinion which reflects this division. On the one hand the main role of the military is considered to act as the guardians of the nation, while on the other their strict non-intervention is the norm. “[the armed forces] are obliged to defend the aspirations and interests of the people and the survival of the nation” (Jose Guiné Sá, 31–01–2005), whilst others oppose any intervention of the military in politics: “it is unacceptable that the military intervene in the country’s politics (S.S. 1–11–2004, Sede www.guiné-bissau.com, Foros de Debate). 159 For an account of the conflict in the Casamance region between the MFDC (Mouvement des Forçes Democratiques de Casamance) and the Senegalese state, see Michael Lambert, “Violence and Conflict and the War of Words: ethnicity vs. nationalism in the Casamance”, Africa, 68, 4, 1998, pp. 585–602; Jean Claude Marut, “The Problème Casamançais: est-il soluble dans l’État Nation?”, in Momar C. Diop (ed.), Le Sénégal Contemporain, Paris: Karthala, 2002, pp. 425–58; and Pierre Englebert, “Compliance and Defiance to National Integration in Barotseland and Casamance”, Afrika Spectrum, 39, 1, 2005, pp. 29–59. 160 In the months preceding the war, the Grupo de Antigos Combatentes da Liberdade da Pátria sent a letter (28–2–1998) to the Parliamentary Committee of Inquiry, extracts of which were published in the press, revealing the serious problems that were affecting the Armed Forces. 161 Vigh, Navigating, pp. 64–85. 68 philip j. havik emerge from the war as a result of the factional struggle within the PAIGC in opposition to the reformist or ‘renovador’ faction and the Junta Militar, led by a government minister who proposed a return to the barracks in order to save democracy. It has since become an often used keyword in current day Guinea Bissau on the political stage, associated with multiparty democracy, peace and the state of the nation. Since the war firkidja has taken on a new life as a ‘national’ slogan reifying and legitimizing a variety of initiatives and ideas, e.g. an internationally funded schooling project as well as the support for human rights and democracy.162 The fork-like shape of the wooden strut upholding the roof is a familiar object to all Guineans, being associated with the traditional djemberén (hut) and the residential compound or morança. Although it implies sharing the same roof, the use of this metaphor is double-edged by using an ‘idiom of ambiguity’. On the one hand it expresses the claim of the urban political strata to form a bridge between different sectors of society, cultivating their role as political mediators or kambantadur. However, its employment for political ends also illustrates the introduction of an object representing a cherished collective moral value and invested with strong symbolic and spiritual connotations into the arena of divisive conflict, reinforcing its contextual ambivalence.163 The fact that a pillar can support very different roofs and uphold solutions allows for the broad ‘national’ appeal of the term, but also for the dilution of meanings and domains. The main significance behind its use as a widely understood metaphor which also implies the need for settling quarrels is that of sharing the same ideal in a divided and conflict ridden country, whether in time of peace or ‘fraternal’ warfare. The tree that marked the front line in the outskirts of the capital during part of the war, the polón di Bra, functioned both as a geographical and spiritual marker during the ‘guerra di Bissau’. Not only is it located close

162 The project providing support for primary education is led by Nelvina Barreto, daughter of the aforementioned minister, Nicandro Barreto, who formed part of the first groups of Guinean nationalists in the 1950s. 163 The use of the term by descendants of an influential trading – and political – lineage or gan that was established in Cacheu in the early 1800s, is by no means accidental, given the importance of the funereal ceremony of Finka Firkidja amongst the Manjaco, the dominant ethnic group in the region between the island of Bissau and the Cacheu River. The symbolically charged ritual not only functions as a rite of passage but also as a means of conflict resolution amongst Manjaco communities, given that any existing tensions and quarrels amongst the members of the defunct’s lineage need to be resolved before the ceremony; see A. Idrissa Embaló, “Religious and spiritual means of conflcit resolution”, Soronda, special issue, 2008, especially p. 314–16. virtual nations and failed states 69 to strategic munitions storage depots164 and the bus station, but it also evoked an association with ancestral cosmologies in general and with the Pepel tchon in which the city lies.165 The Junta Militar was quick to occupy this emblematic site right from the start of the conflict, without relin- quishing it, deploying it as a talisman in its confrontation with government forces. Given that the Junta was to a large extent composed of troops from Islamized areas and the government troops generally of Kriston and ‘animist’ origin,166 it is a particularly significant cross-cultural framework of reference. Even more so, because the leader of the Junta, Ansumané Mané, nicknamed Brik-Brak, a hero of the luta, was regarded as a skilled fighter and commander and whose reputation was on a par with that of the then president and the erstwhile leader of the liberation movement.167 The reification of the shrine was therefore meant as a legitimization of antigus combatenti (former freedom fighters) and the higher moral – national – ground they so conspicuously occupied, to the point of proposing to include the tree in the constitution as a ‘national symbol’.168 The use of the polón in a conflict with the elected government and the president himself – who was of Pepel extraction – and as a location for mediation and negotiation between the parties in the conflict, also illustrates its symbolic political charge. Moreover, the invocation of sacred transcripts was also relevant to the mediation between the visible and the invisible, between the world of the living and the dead, between ancestors and their heirs in the present. This imagery is aptly transcribed by the author of the comic strip Lutu na Polón di Brá who transforms the setting of the war and its main protagonists to that of a quarrel in a Guinean village or tabanka involving chiefs, elders, healers, and the local population, but also includes caricatures of foreign politicians, diplomats and soldiers.169

164 Induta, Guiné, p. 119. 165 João Carlos Gomes, Polón di Bra, Bissau: INACEP, 1999. 166 Vigh, Navigating, p. 233. 167 Induta, Guiné, p. 145 168 Gomes, Polón, p. 143. In this respect, it is interesting to note the ambiguity of meanings attributed to the polón or baobab tree in the allegorical Kriol folktales or stória. For political connotations, see for example the case of the polón in the Bambadinca area close to the road that connects the town to Bissau, where African commandos from Islamized communities accused of collaborating with colonial troops were executed after independence by the victorious liberation movement. 169 Fernando Júlio, Lutu na Polón di Brá, Bissau: INACEP, 1999. The same author was already well known for his entertaining cartoons portraying everyday life in Guinea Bissau, with unforgettable figures such as Ntori Palán and the Tris N’Kurbadu, whose exploits were published in Bissau newspapers. 70 philip j. havik

Conclusions

Given that the national resistance and struggle against colonialism became the standard by which identity and agency was judged in Guinea Bissau, this experience and the collective memory of it initially tended to exclude all other forms. As such, the signification of state and nation hinged upon ‘sacred languages’ that were contained in public discourse; its monopolisation by the single party and its leaders after the war gave rise to initially encoded but increasingly audible voices of dissent. Whereas the ‘magic’ of struggle had been cultivated by the movement its attempt to mobilize local support, the party-state systematically discour- aged or demonized such narratives after independence, thereby relegat- ing them to the status of rumours and innuendo. The strategy for hegemony was rooted in the historical claim of cultural mediation and political leadership made by Christianised Guineans and Cape Verdean Creoles acting as privileged intermediaries for an embryonic civil society. Once the political high ground and the control over the political arena was openly disputed within the ruling party in the course of the 1980s and later between it and rival political movements in the run up to multiparty elections in 1994, the notions of ‘two countries, one nation’ and the ‘nation-party-state’ which had served as political slogans were publicly exposed as a figment of the political imagination. In the post 1998–1999 war period, the nation has all but disappeared from public discourse, except for perfunctory phraseology reproduced by the erstwhile independence movement and its offshoots.170 Given that ‘nationalism was no longer the motivating ideology’171 and associated with a certain historical context contributed to its marginalization and subsequent transformation into a residual category in the public sphere. However, the co-existence of different narratives with regard to violent conflicts needs to be addressed: those which emanate from the stark reality of war and destruction, and others that are superimposed on the former which reside in the realm of imagination and the supernatural. They represent distinctive experiences each with their own momentum, locations, threads and meanings, but nevertheless complementary: it is impossible to understand crisis situations without taking account of the way certain events are interpreted by those caught up in them and

170 See for example the idea of a ‘national pact’; “Nino Vieira defende Pacto Nacional”, Jorge Neto, Espaço África, Nov/Dec, 2005. 171 Alexander & McGregor, “Hunger, Violence”, p. 84. virtual nations and failed states 71 therefore imperative to recognize the existence of multiple narratives and how they interact with each other. While the magic of the imagined universe appears to, at least temporarily, cancel out the other, cruel world of conflict, the former, notwithstanding its ambiguity, also renders the latter more meaningful and palatable by setting new challenges while projecting desires on a troubled horizon. The idea of nation – and with it a nation state – which was meant to underpin a sense of change, of a shared destiny and bolster hopes for a brighter future, has proved to be fata morgana in the case of many countries in sub-Saharan Africa. As a result, the lack of a sense of community with a common cause prompted citizens to identify with an elsewhere, another place, where a collective sense of belonging could become tangi- ble. The nation as a common goal or rallying cry thus ceases to exist as it is neutralized and reconfigured into a bad dream, a tale of woe and suffering, invaded by the realities of everyday life. Over time, the differ ent narratives become intertwined as conflict and violence superim pose themselves upon the imagined narrative, thus establishing inter- connections and contradictions that break the spell of the latter with tragic results. The case of Guinea Bissau well illustrates this process by which the utopia of a nation nurtured in the course of an apparently successful liberation struggle proved, with hindsight, to be a particularly deceptive image. The ‘magic’ associated with national liberation and the imagery used to underpin it, lost its appeal as a result of the glorification of political power by state and party. Or concomitantly, paraphrasing Das, the state failed to uphold its ‘magical’ mode of being, principally achieved by means of non-written means of communication, as distinct from its rational-bureaucratic logic of control, thus demonstrating its fragility and lack of legitimacy. The state’s “uncanny presence” in communities and their everyday life as poignantly observed by Das172 in the case of India, was compromised in Guinea Bissau not only in the course of single party state but in an accelerated pace also after its transformation following multi-party elections in 1994. Subsequently, as a result of economic liber- alisation and political reforms toward multi-partyism, what was left of the idea of nation became subaltern to other more pressing factional political concerns and interests. The 1998–99 war served to momen tarily revitalize and reassemble some of the remnants of nationalist

172 Veena Das, “The Signature of the State: the paradox of illegibility”, in Das & Poole (eds), Anthropology in the Margins of the State, especially pp. 225–30. 72 philip j. havik discourse only to dash any hopes of re-connecting it with local realities (hence the ‘Bissau-war’), failing to reestablish the link between magical and rational realms. Rather than providing a sense of belonging, the nation proved to be founded on the quicksand of slogans, an empty shell, a remedy incapable of generating cross-cultural consensus and collective solidarities. At the same time, the state deconstructed the imagined nation, as conflict and war rudely invaded peoples’ daily lives and expectations. Rather than just being a ‘failed state’, it turned out to be a ‘virtual nation’ forming part of a dreamlike narrative which some have termed a “messianic fiction of a golden age”173 overwhelmed by the stark realities of the post-colony.174 The contrast between the latter and the mixture of pessi- mism and resignation summarized in the Creole epithet djitu ka ten is a striking feature of the profound ambivalence of Guinea-Bissauan ‘national consciousness’ or ‘guineidade’, i.e. being guinean.175 The two threads constitute parallel but complementary post-colonial narratives that became progressively disentangled as a response to ongoing crisis, failure and decline. The popular evocation of the ‘sociedade do cacre’, i.e. the ‘crab society’ that neutralizes all individual initiative and success conditioning its social acceptance upon a ‘visible’ redistribution of resources amongst kin and friends176 finds its counterpoint in the perspective of hope and aspirations contained in the expression djitu ten ke ten.177 Post-colonial

173 Davidson, No Fist, p. 111. 174 Some authors, following Anderson’s lead, have concluded that “failed states are also failed nations” (Gable, “Conclusion”, p.168), quoting from a survey that demonstrates a markedly depreciatory popular perception of Guineans in relation to government, the Armed Forces and political parties. The survey data in question were assembled in 2005, i.e. after the 1998–99 war, and in an election year. See for details and analysis of survey results that show markedly different perceptions between respondents in rural and urban areas, E. Gacitua-Mario, S. Aasland, H. Nordang & Q. Wodon, “Institutions, Social Networks and Conflicts in Guinea Bissau. Results from a 2005 survey”, in Boubacar-Sid Barry, Edward G. E. Creppy, Estanislao Gacitua-Mario, and Quentin Wodon (eds), Conflicts, Livelihoods and Poverty in Guinea Bissau, Washington D.C.: World Bank, 2007, especially pp. 30–40. Published on-line at: http://siteresources.worldbank.org/EXTDEVDIALOGUE/Resources/ Conflicts_Livelihoods_GB.pdf. 175 On the term ‘guineidade’, see Moema Parente Augel, “Desafios do Ensino Superior na África e no Brasil: a situação do ensino superior no Brasil e na Guiné Bissau e a construção da guineidade”, Estudos de Sociologia, 15, 2, 2009, pp. 137–59. 176 See Paula Pinto, Tradição e Modernidade na Guiné Bissau: uma perspectiva interpretativa do subdesenvolvimento, unpubl. MA thesis, Oporto: Faculdade de Letras, Universidade do Oporto, 2009, pp. 10–11. 177 Expression used as title of panel on Guinea Bissau at the AEGIS/ECAS Conference, Leiden, 11–14, 2007: there must be a solution, djitu ten ke ten. See comment ‘A Guiné-Bissau não é apenas instabilidade. Temos também coisas boas que devemos dar a conhecer ao virtual nations and failed states 73 narratives of a conflict ridden country show how the progressive fragmentation of a national political ideal which never actually materialised was above all associated with the loss of its magic – and thus its spell – that was replaced by notions of ‘self reliance’ and ‘living with conflict’ imbued with a ‘modern’ sense of creative survival and the particular skills it requires. As Das has shown, the ‘magic’ presence of the state in communities is located in its capacity to centre local discourse on the state – its ‘signature’ – as rumours are generated and spread on its margins.178 Once the notion and daily practice ‘getting along’ took over, the state’s elusive ability to (re)focus and reshape popular narratives that challenge and contest it was fatally undermined and its signature relegated to the past. Although most reports remain speculative on the underlying reasons for the eruption of the conflict in June 1998, it clearly marked collective memory which distinguished the unexpected cruelty inflicted during eleven months of the guera di Bissau from the eleven years of anti- colonial luta.179 The war and its immediate aftermath would mark a notable change in Guinean politics as the PAIGC which had led the anti- colonial struggle and governed since independence, was ousted from power for the first time in the elections held in November 1999.180 The subsequent instances of political violence which eventually culmi- nated in the successive assassinations of the Armed Forces’ chief of staff, Tagme Na Waie, and the president Nino Vieira in March 2009,

mundo’ (Guinea is not just instability. We also have good things to show to the world); Intumbo Incanha, 27–12–2007, Foros de Debate, www.guine-bissau.com. The author refers here to the multi-cultural aspects of Guinea-Bissau’s Creole language. 178 Das, “The Signature of the State”, pp. 244–5. 179 Jao, “Uma Leitura”, p. 106. For the distinction in popular narratives between colonial and post colonial violence in the case of Zimbabwe and Angola, see Jocelyn Alexander, Jo Ann McGregor & Terence Ranger, Violence and Memory; Alexander & McGregor, “Hunger, Violence”, pp. 80–4; Brinkman, “Ways of Death”, p. 10 and p. 20; and Inge Brinkman, “War and Identity in Angola: two case studies”, Lusotopie, 2003, pp. 195–221. Quoting local Balanta informants in the South of Guinea Bissau, distant from the Bissau upheavals and crises, Themudo transmits their assessment of ‘war’, in opposition to the luta: ‘guera ka bali’ (war is of no use, war is not worth it); see Themudo, “Cultura”, p. 135. 180 See Cardoso, A Formação, p. 25; Amado, Guiné-Bissau, p. 130; Sangreman et al., A Evolução, p. 26. Djaló suggests that the fact that the newly elected president originated from the Balanta ethnic group – while the party he led, the Partido para a Renovação Social (PRS) with strong support in Balanta areas - which “had never had any hegemonic pretensions for power”, implied a “profound change of Guinean society”, bringing to the fore the already existing contradictions between [luso-christianised] elites and societies once regarded as indigenous”; Tcherno Djaló, “Lições e Legitimidade dos Conflcitos Políticos na Guiné Bissau”, Soronda, special issue, December 2000, p. 25–35. 74 philip j. havik appear to close a long chapter in Guinea’s history.181 The apotheotic conclusion serves to illustrate the latent and lasting personal and polit ical rivalries that are associated with the country’s agitated postcolonial course in the 1980s, and also with the 1998–99 war. The gruesome pictures of the victims that were widely circulated on the internet soon after the events, do however provide an unusual public transcript of the killings, clearly distinguishable from previous cases involving high ranking figures.182 Indeed, such events end up by feeding into the exist ing clichés regarding African states and their backward ‘slide’ into political violence, while providing the backdrop for painting a graphic picture of the “bleakness of an entire nation”.183 Popular perceptions of the killings however expressed the notion of a ‘hidden legacy’ of the luta, that began with the assassination in 1973 of the PAIGC leader Amílcar Cabral.184 This spectre of apparently senseless violence spiraling out of control contrasted with the “national dream of freedom, justice, peace and progress” that had been kindled by the erstwhile founder of the nation.185 What could have been a ‘magical moment’ ended up as a ‘ritual rebellion’ that although temporarily galvanising communities and projecting a sense of national belonging, left the statu quo largely unchanged as communities ‘waited’.186 As a sense of urgency associated with survival and making ends meet took over, state and nation became increasingly distant concerns, fading and contested memories of a collective experience that belonged to another generation, only to reappear as an unfulfilled promise when

181 The chief of staff, Ansumane Mané who commanded the Armed Forces against the then president Nino Vieira during the 1998–99 war, was assassinated in 2001 during the presidency of Kumba Iála. The latter was removed by the Armed Forces chief, Verissimo Seabra, in 2003, only to be assassinated the following year, whereafter Tagme Na Waie took over as chief of staff, who in turn was killed in a bomb attack in 2009. For a background on these events, see Trajano, “O Precário Equilíbrio”, pp. 256–9. 182 The images of death and destruction that were publicized in the course of the 1998– 99 war initiated this shift in perceptions. But the public display of assassinated political and military figures which began, albeit tentatively, with those of Ansumane Mané in 2001, was a novel marker in the process. Nonetheless, the importance of this shift has been largely ignored in the subsequent mediatic coverage. However, popular opinion had it that the real circumstances of the killings would remain a ‘military secret’; see Themudo, “From the Margins”, p. 61. 183 Gable, “Conclusion”, p. 167 184 Themudo, “From the Margins”, p. 62. 185 See the memorandum of the Junta Militar para a Consolidação da Democracia, Paz e Justiça, Praia, 25–8-1998, in Gomes, Polón, especially p. 137. 186 See Annemarie Hochet, Paysanneries en Attente, Dakar: Enda, 1983. virtual nations and failed states 75 battles were fought.187 The example of Guinea Bissau to a certain extent illustrates this apparent ‘no state, no nation’ limbo marked by “incessant rebellions” rather than revolutions.188 At the same time the process of dis- aggregation and deconstruction, also denotes the emergence of new popular slogans that translate contemporary concerns by means of the Creole language or Kriol into commonly understood – but not necessarily ‘national’ – keywords and catchphrases.189 The curious combination of resignation on the one hand and survival against all odds on the other into a special symbiotic brand of a collective, shared sense of pride – not uncommon in many African countries – highlights the need for diversify- ing and deepening our understanding of societies and how they process, contextualise – and become – social realities. Penetrating and mapping these ‘labyrinths of memory’ that are in constant flux in the ongoing process of remembering and forgetting, and finding out how they are shared is a major challenge for researchers. As some scholars have observed, despite the fact that recent research has focused upon the deciphering of the causes and dynamics of conflict and conflict management in Guinea Bissau and beyond, many contributions tend to centre their attentions on the role of state institutions rather than on local communities and society at large.190 Rather than revolving round (virtual) nations and (failed)

187 Some authors have referred to the use of places of collective memory (i.e. the ‘lieux de mémoire’ as proposed by Pierre Nora) in political discourse that explore symbols of the past, for example those associated with the master-narratives of the liberation war; see Maria da Conceição das Neves Silva, “Lógicas pessoais & Estado burocratizado alargado: dinâmicas de transição na Guiné Bissau”, in Óscar Soares Barata & Sónia Frias (eds), Populações, Ambiente e Desenvolvimento em África, Lisbon: Instituto Superior de Ciências Sociais e Políticas (ISCSP), 2001, pp. 309–24. 188 Vigh, “Conflictual Motion”, p. 159. See the headline in the Guinea-Bissauan weekly Última Hora, immediately after the killings of the then president Vieira and the chief of staff Na Waie: “Guinea Bissau: empire of Kalasknikov and machetes”. 189 See Bordonaro, Living at the Margins, p. 209 on the ‘phantom-’ or ‘skeleton state’ in Guinea Bissau. Recent developments have caused the country to be labeled as a ‘narco- state’ (see e.g. Ed Vulliamy, “How a tiny West African country became the world’s first narco state’, Observer, 9–3-2008). See also Dirk Kohnert, “Democratisation via elections in an African narco state? The case of Guinea Bissau”, in 50 anos das Independências Africanas: desafios para a modernidade, Proceedings of the 7° Congresso Ibérico de Estudos Africanos, Lisbon, 2010 – Access Proceedings on-line at: http://repositorio.iscte.pt. These epithets however clash with the idea of a ‘failed’, ‘phantom’ or ‘empty’ state, requiring a careful rethinking of such, often misleading, categories. 190 See G. Klute, B. Embaló, A.K. Borszik & I. Embaló, “Local Experiences of Conflcit Management”, in Soronda, special issue, p. 15–33, and also Có, “Representação e Confinação”, for a discussion of the issue of conflict management and local experience in Guinea Bissau. 76 philip j. havik states, it is urgent for scholars to identify how public codes associated with the ‘nation-state’ relate to and run parallel to other ‘hidden’ and innovative forms of signification expressing shared meanings that emerge against a background of ongoing political conflict.191 These ‘languages of life’192 are composed of practices, narratives, images and signs rooted in everyday life and experiences such as ‘getting along’ and ‘creative survival’. In order to do so, the misteriu or mystique contained in the dynamics of popular narratives and experiences should be addressed in order to make sense of the many challenges posed by the present day quandary of seemingly endless conflict and violence that is seen to characterize societies and social change in sub-Saharan Africa.

191 See for example Paul Richards, “New War”, and Vigdis Broch-Due, “Violence and Belonging: analytical reflections”, in Broch-Due (ed.), Violence and Belonging, pp. 1–40. 192 Mbembe, On the Postcolony, p. 15.

Map 2. Map of contemporary Mozambique. © Maura Pringle, Queen’s University Belfast. CHAPTER THREE

THE SOCIAL ORIGINS OF GOOD AND BAD GOVERNANCE: RE-INTERPRETING THE 1968 SCHISM IN FRELIMO

Georgi Derluguian

Good governance became the catch phrase of international development agencies in the late 1990s. It originates in the crisis of neoliberal Washington consensus that had once prescribed overcoming the accumulated contradictions of bureaucratic developmentalism by switching to the business-friendly policies. When the promised market growth failed to arrive in so many poor countries, the standard agenda of neoliberal reforms could not be revised due to the heavy ideological and institutional investments made by Western donor agencies. Instead, criticism was shifted to the implementation process. This is why, as David Woodruff masterfully explains, in the late 1990s the neoliberal agenda was extended to stress the fight against corruption, institution building, rule of law, accountability, transparency and other now familiar measures under the rubric of good governance.1 Mozambique in the early 2000s was hailed as an encouraging example of better governance. Granted, not an entirely ideal example, for some degree of corruption was obviously present. Yet, Mozambique became an African country that international donors could cite among those making progress. Curiously, Mozambique had been also regarded a more encouraging example during the previous epoch of socialist developmentalism. Moreover, from the global epoch of socialist developmentalism to the epoch of market globalization Mozambique continued under the same ruling party, Frente de Libertação de Moçambique (FRELIMO), and even under the same political personnel. Such continuity in power smacked of political opportunism to many commentators.2 It is not, however,

1 See David Woodruff, “Rules for Followers: Institutional Theory and the New Politics of Economic Backwardness”, Politics & Society 28:4 (December 2000), pp. 437–482. 2 A good example would be Alice Dinerman. Revolution, Counter-revolution and Revisionism in Postcolonial Africa. The Case of Mozambique, 1974–1994, London and New York: Routledge, 2006.

© koninklijke brill nv, leiden, �� | doi ��.��/��_��5 80 georgi derluguian analytically very interesting to ponder the awkward dilemmas arising from the shifting ideological discourses. There is little new here. Paris is worth the mass, in the notorious statement of King Henry IV. Generally, political elites should be expected to cling to power as long as they can manage. Ever pragmatic, the leaders of FRELIMO sacrificed ideological faith from one period for the sake of staying a recognized actor in the world-system in another period. Let us posit the question differently: What made FRELIMO such a successful power elite? Since its creation in 1962, this organization has with- stood long wars, ruinous economic crises, the global wave of ideological extinction, and it has twice survived the violent deaths of its charismatic leaders: Samora Machel in 1986 and Eduardo Mondlane in 1969. My explanation proceeds from three theoretical bases. The first is the world- systems perspective of Immanuel Wallerstein and Giovanni Arrighi. It helps to appreciate the dilemmas of power in the periphery. The second is the movement mobilization and state-building theory of Charles Tilly. The third theoretical source might sound surprising for it is usually considered a sociology of cultural production and it was rarely applied to Africa – I mean the sociology of Pierre Bourdieu.3 This theoretical combination allows to sketch an explanation of FRELIMO’s politics without falling into the usual partisan polemic. My claim is that FRELIMO was able to achieve and maintain over different epochs the position of key mediators between core and periphery. The shifting political ideology was not merely opportunism. This reflected the opportunities at the level of geopolitics and geoculture which the leaders of FRELIMO rather aptly exploited to establish their credentials as national state-builders, whether socialist or neoliberal. In the process, the social network of FRELIMO’s historical leaders, almost entirely hailing from the colonial intelligencia of Lourenço Marques (Maputo), had to marginalize and overcome their less cosmopolitan competitors whose social capital was invested in local patronage and therefore could be branded “tribalist”. The crucial difference between the “progressive” national leaders and the “tribalist reac- tionaries” was in the kinds of social capital, habitus and power bases. For this article, we take just one although, in retrospect, crucial episode from FRELIMO’s earlier history. It is the legendary schism occurring around the symbolic date of 1968 in which the “progressive” cosmopolitan

3 It should be, however, reminded that Bourdieu first made his name in the study of late colonial Algeria. See Pierre Bourdieu, The Algerians (Translated by Alan C. M. Ross. With a pref. by Raymond Aron), Boston: Beacon Press, 1962. the social origins of good and bad governance 81 faction of Eduardo Mondlane found themselves pitted against the “obscurantist” traditionalist faction of local “tribal chiefs”. The ideological labels are conditionally bracketed here because they emerge from FRELIMO’s official discourse. Our task is rather to excavate what social structures underpinned the contemporary ideological oppositions. In doing so, we might be able to elucidate the origins of personalistic power strategies commonly called state corruption and the different, more bureaucratic and impersonal power strategy which is called today good governance. The dichotomy of “bad” corrupt ethnic patronage and “good” impersonal rule of law arguably applies to the entire postcolonial Africa if not universally (to which I can attest as someone living in Chicago and, incidentally, teaching at the very same Northwestern University where Eduardo Mondlane had been a graduate student back in the 1950s). This makes the FRELIMO example eminently generalizable and theoretically important because, in this case, good governance somehow prevailed over bad. Why?

The Standard Legend

Let us first recap the officially consecrated version of events. FRELIMO was formed in 1962 outside the Portuguese – controlled Mozambique. It merged three émigré “proto-parties” that had appeared slightly earlier in the neighboring British colonies with milder political climates. Their leaders, it is claimed, enjoyed only limited bases of support among the migrant workers, missionary pupils and refugees from the adjacent Mozambican provinces. In addition, a small but far more cosmopolitan network of Mozambican émigré intellectuals coalesced in Western Europe and in the United States. The American-educated sociologist Eduardo Mondlane became FRELIMO’s founding father and foremost theorist.4 Two years later, in September 1964, FRELIMO officially launched its

4 Mondlane was born in 1920 in southern Mozambique and, with combination of luck and inordinate talent, obtained education from Swiss missionaries and the University of Witwatersrand in South Africa. When apartheid policies made impossible his further studies in South Africa, the same Swiss missionaries helped Mondlane to transfer to Oberlin College and later to Northwestern University in the United States. In 1960 Mondlane received Ph.D. in sociology and job offers from Syracuse University and the United Nations. Instead, he returned to Africa to lead the FRELIMO. He was accompanied by the White American wife who also became an activist. During his years as FRELIMO’s first president, Mondlane wrote numerous articles, theatrical plays, and a brilliant book called the Struggle for Mozambique, London: Penguin, 1969. Mondlane’s powerful and eloquent analysis shaped the radical ideology of FRELIMO and much of subsequent 82 georgi derluguian guerrilla campaign against Portuguese colonialism. Almost from the out- set FRELIMO achieved a very impressive international recognition as the sole legitimate alternative to the Portuguese rule in Mozambique. This recognition carried considerable political and material advantages. Improbable as it may sound, FRELIMO’s fighters received training at the same time in the Arab countries and in Israel; aid was offered by the Soviet Union, Maoist China, and, covertly, by the United States under the Kennedy administration. Last but not least, even the Vatican defied Lisbon by establishing contacts with its African nationalist opponents.5 The euphoria of initial successes did not last. Defying the international trend of decolonization, Salazar’s Portugal built up its colonial armies and police, established special ties with the apartheid South Africa and the white minority regime in Rhodesia, and undertook an extremely ambitious dirigiste program of economic investment intended to create the settler-dominated industrial economies in Angola and Mozambique. For a while it seemed that Portugal, might become an exception to the general trend of post-1945 decolonization. Like the fellow anti-colonial movements in Guinea-Bissau, Angola, Zimbabwe, Namibia, South Africa, FRELIMO underwent a period of military setbacks and political disorien- tation. The years preceding the FRELIMO’s Second Congress in 1968 were marked by inchoate factional struggles, riots in refugee camps and the movement-established schools, a succession of splits and defections, and finally by the mysterious assassination of Eduardo Mondlane in February 1969.6 The period of internal crisis ended in 1970 when FRELIMO was purged of oppositionists and reconsolidated under the leadership of charismatic guerrilla commander Samora Machel. The movement ideology and propaganda at this point crystallized in the typical Third World commentary on the situation in Mozambique by sympathetic Western observers and solidarity campaigners. See the anonymous biography Eduardo Mondlane, London: Pan- Africa Great Lives Series, 1972; Barry Munslow, Mozambique: the Revolution and its Origins, London: Longman, 1983; Allen Isaacman, Mozambique: From Colonialism to Revolution, 1900–1982, Boulder: Westview Press, 1983; John S. Saul (ed.), A Difficult Road: the Transition to Socialism in Mozambique, New York: Monthly Review Press, 1985. 5 Foreign assistance to FRELIMO is documented in the special chapter of Thomas Henriksen’s Revolution and Counterrevolution. Mozambique’s War of Independence, 1964– 1974, Westport: Greenwood Press, 1983. We may expect further detail from the archival disclosuress regarding the history of Cold War. 6 The parcel bomb that killed Mondlane almost certainly was made by the Portuguese secret police, but it was widely speculated at the time that the device had to be delivered to FRELIMO’s offices in Dar-es-Salaam by an oppositional faction or defector. There is, however, no evidence to the latter accusation. the social origins of good and bad governance 83 quasi-Marxist terms. The guerrilla campaign was re-launched with new military equipment and skill. The population of refugee camps in Tanzania and the liberated zones in Mozambique received new administrative apparatus which prominently featured the institutions of popular democratic deliberation, artisanal and agricultural cooperatives, as well as education and social services. FRELIMO became a de facto state with an explicitly progressive ideology and policies credibly challenging the reactionary Portuguese regime in the rest of Mozambique. In 1973 FRELIMO pushed its military operations into the economically important central highlands to the south of river Zambeze. The strategic breakthrough of guerrillas frightened the white settlers and convinced the Portuguese military of the futility of the decade-long war effort. In April 1974 the leftist military coup in Lisbon toppled the antiquated dictatorship paving the way to a rapid decolonization. In the ensuing panic, the majority of Mozambique’s 220,000 white settlers fled abroad. The People’s Republic of Mozambique, proclaimed by FRELIMO in June 1975, automatically nationalized their abandoned properties which instantaneously created a sizable state sector. FRELIMO had few doubts regarding how to use this historic opportunity. In 1977 FRELIMO’s Third Congress triumphantly acknowledged the commitment to the Marxist-Leninist path of development including the official establishment of one-party rule, the beginning of rapid indus- trialization, and collectivization of peasantry. Yet at the same time the Rhodesian Central Intelligence Organization (CIO) began foment ing internal revolt in Mozambique, hoping to prevent it from providing bases for the Zimbabwean guerrillas. The shadowy political front that emerged out of foreign subversion was baptized RENAMO (Resistência Nacional Moçambicana, or Mozambique’s National Resistance). When the Rhodesian white minority regime surrendered power in 1980, the South African military adopted RENAMO. In the meantime, the economy of Mozambique was falling apart under the combined impact of severe disorganization after the withdrawal of settlers and the embargo imposed by South Africa (Mozambique’s real economic metropole). The Eastern bloc technical aid and the dispatch of Soviet specialists, as confused and inefficient as it was, could not match the scale of the dam- age to Mozambique’s economy. Instead of the proclaimed Decade of the Victory over Underdevelopment, the eighties brought economic ruin, widespread famine amidst the rural populations, and atrocious internal war that spread to the entire territory of Mozambique except the few large towns. The misery and destruction soon became so massive that, 84 georgi derluguian demonstrating remarkable adaptability, FRELIMO at its Fourth Congress in 1983 abandoned its erstwhile socialist pretensions. Subsequently it accepted the standard IMF program of structural adjustment and began seeking political accommodation with the vastly much wealthier and militarily stronger South Africa. Nonetheless RENAMO continued its dif- fuse but immensely destructive war. Only in 1992, with the looming end of apartheid regime in South Africa, a peace settlement was finally brokered by the Catholic Church and the United Nations. In the competitive elections that followed, RENAMO narrowly lost with some forty per cent of the vote. RENAMO (unlike Angola’s UNITA) was starved of resources and therefore had no choice but to concede the defeat and become an oppositional party. FRELIMO thus continued to rule into the new period of peace and market recovery. In the late 1990s Mozambique even became one of the World Bank’s showcases in Africa.

“Third World Liberation” as Global Network

In 1962 the founding congress of FRELIMO only nominally brought together the aspiring Mozambican politicians of very disparate social and ethnic backgrounds. The three “proto-nationalist” movements grew among the Mozambican labor migrants who regularly (and mostly ille- gally) traveled across the colonial borders. The earliest activists in the migrant communities usually were the Protestant missionary converts whose beliefs and social mobility aspirations clashed with the Iberian Catholic identity of Portuguese colonizers. The three organizational precursors of FRELIMO corresponded to the three migratory loops originating from very different parts of Mozambique: the central districts of Manica and Sofala connected to Rhodesia (Zimbabwe); the north-central Zambezia and Tete adjacent to Nyasaland (Malawi); and the far northern Makonde people who regularly gravitated towards Tanganyika (Tanzania) and Kenya. All three “proto-nationalist” movements were in effect spin offs closely emulating the African independence parties of their host countries. Unsurprisingly, their moderate political programs and the petitioning tactics were copied from the contemporary examples found in the British territories. Their leaders were either preachers or simply “big men” and “chiefs”, most of them with only basic missionary education and provincial outlook. On the surviving photographs these men look stolid, buttoned up Africans dressed in their Sunday church suits and hats. the social origins of good and bad governance 85

Due to the hostile climate of police repression, no mass African movement could emerge in the more developed southern Mozambique and in the colonial capital of Lourenço Marques (renamed Maputo after independence). Since the formation of colonial economy during late 19th century, southern Mozambique received the role of labor preserve and transportation node tied to the gold-extracting economy of Transvaal. The South African mining boom even induced the Portuguese to move the capital of its colony to the extreme south. The orientation towards Transvaal profoundly separated Lourenço Marques from other parts of Mozambique that in their turn evolved into the suppliers of African labor and transportation services to the neighboring economies of British Rhodesia, Nyasaland, and Tanganyika. The Portuguese never possessed the resources to integrate Mozambique economically and socially. This difficult task fell to the anti-colonial nationalists. The repressive corporatism of Salazar’s Estado Novo severely limited the venues of social mobility and civil association. What remained open in southern Mozambique to the majority of Africans officially classified indigena, or “uncivilized native”, was self-advancement through the Pro testant missionary schools (Mondlane provides an outstanding example) or the fluid, highly charismatic and mostly clandestine Afro-syncretist congregations Christianity which had been spreading from South Africa through labor migrants and itinerant African prophets. Machel is a good example of the latter pattern. The future guerrilla commander and first president of Mozambique did not represent the first generation of leaders in his family. In the 1910s-1940s his father became a self-made rural patriarch after he had invested the money earned at the mines of Transvaal into purchasing a herd of cattle, the European-style plow and the obliga- tory gifts to villagers. The elder Machel consolidated his social prominence by establishing a new congregation of African Zion church whose adept he had become while in South Africa.7 Jumping a little ahead, let us notice that the oppositional Makonde “chiefs” whom Samora Machel would later crush, in fact were almost exactly like his own father: the self- made patriarchs in rural African communities who combined entrepreneurial acumen and resources earned in labor migrations with the social capital offered by religious conversion. The difference was in time period.

7 Iain Christie, Machel of Mozambique, a Biography, London: Panaf, 1989. Regarding Samora Machel’s father, the obituaries and quite reverent memoirs that appeared around the time of his death in June 1984 clearly implied that the old patriarch was not exactly the poor African peasant as suggested by the official biographies of his President son. 86 georgi derluguian

This typical social personage of late colonial period appeared in the remoter Makonde highlands only in the late 1950s where it came in conflict with both the declining colonial hierarchy and soon with the rising nationalist state-builders as well. The majority of Mozambican national state-builders, however, originated in the subordinate colonial elites, in the various ranks of “honorary whites”: mixed race, Asian or Indo-Portuguese from Goa, the officially cer- tified “civilized” urban Africans whose parents had managed to escape the “native” status. Significantly, unlike the rest of Mozambique the associa- tive experience in Lourenço Marques bore a decidedly secular bent. The pre-political organizing among the capital city urbanites typically took the form of various benevolent associations and social clubs ranging from charities to dances, poetic circles and the enthusiasts of European cinema. It is this Europeanized multiracial middle-class milieu of the colonial capital that would eventually dominate in FRELIMO’s revolutionary leadership. A systematic reconstruction of network ties within the colonial elites remains a future task. Yet it can fairly securely be said that a disproportionate number of Mozambique’s top appointments after independence were former students at Licéu Salazar, the exclusive high school in Lourenço Marques. This group can be characterized as the nascent intelligencia of Mozambique.8 They were the locally-rooted modern technical specialists, educators, functionaries, artists, and aspiring professionals whose major form of capital was accumulated not in economic assets but rather in their social networks and education credentials, what Bourdieu famously called social capital. As with all intelligencias around the world, the

8 The word intelligencia appeared in the late 19th century Russia (or perhaps Poland, as the perennial competitors of Russians for the leadership in Slavdom would claim.) At first its usage was pejorative: the conservative Russian publicist Boborykin who is usually cred- ited with inventing the word, called intelligencia the first generation of university educated commoners and petty gentry that has appeared in Russia as the result of the abolition of serfdom and the other so-called Great Reforms of the 1860s. The first generation of university graduates hardly fit into the aristocratic-agrarian social structure of imperial Russia which produced a panoply of oppositional ideologies and strategies ranging from the intellectual crusades to the early nihilism (famously decried by Turgenev and Dostoevsky) and eventually the narodniki terrorism against the aristocratic officialdom. As it often happens with timely invented words, the pejorative soon became transformed into a badge of honor (like the British satirical coinage meritocracy during the 1960s.) We consider the spelling intelligencia with “c” instead of the originally Russian “ts” more appropriate because the root and the suffix are obviously Latin, therefore the copious transliteration of Cyrrilic back into Latin script looks too awkward. The analytical applica- tions of the word intelligencia, as we shall argue later, by far transcends the boundaries of Russian history and culture, which is an even heftier reason for a more universal spelling. the social origins of good and bad governance 87 group sense of cohesion, moral mission and eventually their political radicalization were a pretty direct result of the discrepancy between the perceptions of social worth and the actual rigidity of the existing rank hierarchy compounded by the dearth of career opportunities in liberal professions. Adding to that was, of course, the race classification. Even if the official ideology of Portuguese colonialism did not proclaim racism as overtly as in South Africa, the reality wasn’t too different because the Portuguese settlers recently arriving from Europe a priory enjoyed access and affinity to the colonial administration. With the multiplying examples of decolonization in Asia and the 1957 breakthrough of Ghana in Africa, the situation of the colonial intelligencia of Lourenço Marques became, to say the least, politically very ambiguous. If they needed any reminder, Salazar’s secret police PIDE and the hostile outbursts from many paranoid whites would go a long way towards fostering the sense of embattled group unity. The question still remained, whom did this intelligencia represent beyond their narrow circle? One can see in the classical Russian pattern of politicized intelligencia that there could be no intelligencia without a “People”. For the excessively educated Russians of Tolstoy’s generation, the “People” became peasants. For the intelligencia of Lourenço Marques, it had to be the Nation. Under a less reactionary political regime than that of Salazar’s Portugal, the upwardly mobile African intelligencia could have gradually advanced into the positions of bureaucratic authority and professional prestige. This might have secured a moderately conservative road to Mozambique’s independence following the much-touted multiracial “Brazilian way”. In fact, the tendency towards a conservative compromise did emerge later in the 1960s when the Portuguese military command following the French and American strategies of “hearts and minds” reigned in the ter- roristic secret police and when the combat duty opened to Africans the road to Portuguese citizenship and veteran benefits. Evidently quite a few among the middle class Mozambican youths responded to the opening of the war-time social mobility. In some well-known instances even the younger brothers of prominent guerrillas chose to enlist in the Portuguese commando units which offered faster promotions. But at the turning moment of 1959–61 Mozambique was still ruled by the reactionary colonial regime that regarded as dangerously suspicious the films of Buñuel, Bergman, and Fellini or the prose of Albert Camus. The bloodshed in the former Belgian Congo and especially the 1961 African revolt in Angola induced a veritable paranoia in the Portuguese colonial apparatus and among ordinary settlers. The rising wave of terror sent many Mozambican 88 georgi derluguian enthusiasts of European cinema and existentialist philosophy escaping into the larger world to which their aspirations belonged. A mundane question would help us to visualize the implications of what the prominent American researcher Sidney Tarrow calls the “capil- lary work of movement organizing”9 Where in a big Western metropolis during the late 1950s/early 1960s might go a poor refugee student from Africa who had just experienced the dangerous and romantic escape from the fascist secret police, who had read Camus, Hemingway and perhaps a smuggled copy of the Communist Manifesto? Here are the old instruc- tions, given with a chuckle by veterans of African liberation movements: in London, call on Basil Davidson or someone at New Left Review; in New York, from the UN headquarters take the subway up to Columbia University and seek there Prof. Immanuel Wallerstein; in Paris, Jean-Paul Sartre himself will get on the phone for you if his own famous sofa in the living room is currently occupied. The European and American metropo- lises thus became the sites of encounters with the Western New Left theorists and the fellow refugees from around what has just been dubbed the Third World. With the intellectual predispositions and connections of this kind came the heated all-night debates and the common set of readings regarding the strategies of revolutionary transformation: Marx or Rostow? Lenin or Gandhi? Stalin or Trotsky? Mao, Fanon, or Ché Guevara? These were the crucial strategic choices that account for much of what would happen all over the Third World in the aftermath of decolonization.10 The exuberant feeling which animated this spontaneously emerging transnational network was borne by precisely what Benedict Anderson called the “print” effect on the formation of modern national identities: the feeling of simultaneousness, the public sharing of parallel individual experiences within a big imagined community.11 Joining this transnational network could offer the hefty advantages of the rising global identity as

9 See Sidney Tarrow, Power in Movement. Social Movements, Collective Action and Politics, Cambridge: Cambridge University Press, 1994. 10 A very good reconstruction of the ways in which the Third World ideology and social network emerged is found in Vijay Prashad, The Darker Nations: A People’s History of the Third World, New York: The New Press, 2007. 11 Benedict Anderson, Imagined Communities, London: Verso, 1983. Even closer to our discussion, see the most recent work of Benedict Anderson, Under Three Flags: Anarchism and the Anti-Colonial Imagination, London: Verso, 2005. A very germane analysis of transnational social movement can be found in the classic study of the Second International by George Haupt, Aspects of International Socialism, 1871–1914: Essays. With a Preface by Eric Hobsbawm, Cambridge: Cambridge University Press, 1986. the social origins of good and bad governance 89 anti-colonial activists. It also brought the resources of international solidarity and the shared understanding of political situation, goals and action strategies. The decision to launch the guerrilla campaigns back in the home countries logically flowed from the recent guerrilla successes in Indonesia, Vietnam, Algeria and, above all, Cuba.12 Not a small consideration, the guerrilla expedition also seemed the best escape from the infamous pressure cooker of émigré circles. The world geopolitics at the time seemed exceptionally conducive to such decisions. The direct encouragement to return to Africa and fight came from the radical anticolonial leaders like Egypt’s Nasser, Tang anyika’s Nyerere, Ghana’s Nkrumah and Algeria’s Ben Bella. Next came the less radical in their forms yet quite substantial aid from the diverse West European and North American citizen groups, solidarity campaigns, churches, and Social Democratic governments. Even Israel (at least before the war of 1967) tried to break out of its international isolation by actively seeking a role in African decolonizations. It is tempting to name the Cold War rivalry between the two superpowers as the major condition conducive to the success of radical state- building projects in the periphery of the world-system. A word of caution is therefore in order here. To a sober observer at the time it was already quite clear – and today, after the opening of Moscow archives, it also obtained documentary proves – that the Soviet Bloc served a romanti- cized example rather than providing active help. Khrushchev’s Kremlin certainly felt the international pressure to live up to its projected image of the cradle of revolution. Besides, the competing activism of Maoist China was a growing worry in Moscow which was not a small consideration at the time. After 1955 the grim Stalinist isolationism and repudiation of “ultra-left adventurism” were abandoned for the newly proclaimed alliance with all “progressive anti-imperialist forces”.13 Yet Moscow felt bound by the Cold War compact and feared that an active involvement in the Third World conflicts could escalate into another proxy confrontation

12 Guerrilla war, as Eric Hobsbawm reminds us, was not in the tool kit of revolutionary movements until the 1959 triumph of Fidel Castro. The communist parties which by necessity engaged in what they called partisan wars, nevertheless did not theorize this strategy. Even Mao became regarded a grand guerrilla strategist only retrospectively, after the Cuban “revolution en la revolution”. The timing of FRELIMO’s guerrilla couldn’t be more prescient. See Chapter 2 “World Revolution” in Eric Hobsbawm, The Age of Extremes: A History of the World, 1914–1991, New York: Vintage, 1996. 13 For an excellent discussion of this transition in the Soviet foreign policy, see Ted Hopf, Social Construction of International Politics: Identities & Foreign Policies, Moscow, 1955 and 1999, Ithaca: Cornell University Press, 2002. 90 georgi derluguian with the United States like the Korean War or Cuban missile crisis. Besides, the USSR simply lacked the resources and expertise to control its political investments in the unexpectedly active Third World. On its side, the United States felt analogous pressures to live up to its hegemonic project of post-1945 world order based on modernization and the Wilsonian vision of national self-determination. No less importantly, the American political establishment felt the growing urgency to find an answer to the domestic challenge of the Negro civil rights movement. At the level of policy conduct, the Third World decolonization was continuously forcing upon Washington the dilemma of supporting the declining European powers or the anti-colonial militants who could become governments next day. These were not easy choices. Vietnam eventually proved to be the major price for the blunder of siding with the defeated colonialists. This brief survey of 1960 world politics suggests that the guerrilla war was an indigenous choice of the Third World political forces. The insur- gent road to decolonization was not a mere peripheral reflection of Cold War geopolitics. Rather, it was imposed on the global calculus of superpowers by the anti-colonial movements acting with unexpected vigor. The periphery in the world-system of the 1950s-1960s obtained and vigorously pursued the chance to dramatically upgrade the collective bargaining positions. In the international atmosphere of the 1950s-1960s, the pursuit of national sovereignty and social reform by armed insurrection presented eminently acceptable credentials for the international recognition even before the actual assumption of state power. This meant that guerrilla movements were recognized not as a military force (in this, they had only to be credible enough) but rather as truly modern states in statu nascendi, progressive alternatives to colonial state-building which after 1945 so suddenly became untenable. On a serious consideration, Geoffrey Barraclough still today seems right in claiming that this represented perhaps the main theme of the twentieth century politics, the totally unexpected beginning of a truly world-historical reversal of fortunes.14

Cosmopolitans versus Provincials

When in 1962 Eduardo Mondlane and his comrades arrived in Dar-es- Salaam for the founding congress of FRELIMO, the three existing African

14 Geoffrey Barraclough, An Introduction to Contemporary History, London: C.A. Watts, 1964. the social origins of good and bad governance 91

“proto-parties” were offered little choice but to get unified under the leadership of these cosmopolitan intelligencia who carried such an impressive social capital of international prestige and connections. Nevertheless, elites being elites, the locally based African politicians were not at all prepared to completely surrender their own ambitions nor their material privileges and clientilist bases bounded by African “tribal” ethnicities. The unification under the brand name of FRELIMO in mundane reality opened a long series of infighting. It continued until the expulsion of the last “tribalist” elements from the leadership in 1970, shortly after Mondlane’s assassination and the internal take-over by the radical guerrilla wing of the organization. This explains why the intelligencia leaders of FRELIMO insistently talked about the “struggle for the correct definition of enemy” and why their political vocabulary featured the expressions reaccionário, obscurantista, tribalista, elemento feodal and simply ambicioso even more prominently than fascista or colonialista. Inciden tally, the ubiquitous term racista could be applied equally to the Portuguese rulers, their apartheid allies, and to the Pan-Africanist militants opposed to the universalism of FRELIMO’s leadership. Within the fields of ideological discourse and internal organizational politics the local oppositions had to mobilize on particularist basis against the modernist universalism espoused by the vanguard intelligencia. Aside from the charges of usurpation and dictatorial methods, Mondlane was typically accused of disrespect towards chiefs and elders with the attendant “gift-giving”, hostility to African traditions especially to polygamy, surrounding himself with fellow southerners from the capital of Lourenço Marques, and all the way to the accusations of reading too many foreign books and being married to a white American woman (who indeed sought a political role of her own in mobilizing Mozambican women). Taken without polemical bent, much of these charges were actually true. What is interesting, however, is precisely the provincial, “backward” and simply self-defeating discursive form in which these accusations were expressed. Couldn’t these challengers come up with something better, more legitimate beyond their small parochial world? Yet the grievances of FRELIMO’s “backward” oppositionists correspond to a fundamental social dynamic that can be politically repressed and further marginalized but it cannot be derisively dismissed or analytically ignored. This is indeed a central point in the contemporary predicament of African continent. Understanding the indigenous anti-state dynamic would help us to explain rationally, without falling into common racist stereotypes, the downward trajectory of most African states after the achievement of political sovereignty. This social dynamic found 92 georgi derluguian everywhere in peripheral villages and colonial slums constituted an indigenous alternative to nation building. Perhaps it was not a very promising alternative in global perspective. But the stolid Big Men from small villages, the missionary catechists and carpenters, petty clerks in the colonial apparatus, or simply the enterprising youngsters from the indigenous outskirts of colonial towns neither knew nor could afford any different ways of achieving vertical mobility in post-colonial societies and institu- tionalizing their social capital. The peripheral African politicians and entrepreneurs were fundamentally opposed to the FRELIMO’s project of building an efficient modern state for a good reason. Their skills and credentials were unsuited for the bureaucratic pattern of governance. Instead they favored patrimonialist and ethnic patronage over the modern bureaucratic form. These anti-state aspirations would prevail in many postcolonial African states. Even within FRELIMO these men were numerous, though disunited, and they were playing on their home ground against the intelligencia who were very few and essentially alien. The intelligencia could prevail only when tightly organized into an ideologically driven political party or state bureaucracy that furthermore could rely on external resources for its statist mobilization. In the geopolitics and geoculture of the 1960s this was eminently possible. It is actually true that Mondlane and most other leaders of FRELIMO at popular meetings often spoke through a chain of interpreters: from English or Portuguese into the East African lingua franca of Swahili which sometimes had to be further rendered into the local languages of northern Mozambique. This inevitably resulted in linguistic and cultural mis- translations. For example, the word “imperialism” usually was translated into Swahili as ubeberu. This is simply a pejorative, literally meaning “goat-ness” (compare to the popular Portuguese and Spanish term of abuse cabrões/cabrones). Agricultural cooperatives and collective farms in the 1960s Swahili usage were shamba la ushiriki or shamba la ujamaa, i.e. the communal (-ity) field. In the context of African villages, this reso- nates more with traditional institutions of collective reciprocity. The examples can go on. But we are not interested so much in the ironies of mis-translation as in the political effect of FRELIMO’s official discourse on African peasants. Eduardo Mondlane wrote the marvelously lucid monograph the Struggle for Mozambique. Its analysis of colonial Mozambique and FRELIMO’s politics deeply impressed a wide range of Western-educated readers ranging from communists to mainstream liberals. On that side, the discursive fit seems almost perfect. But it remains to be explained how the social origins of good and bad governance 93 a small group of activist intelligencia managed to create a loyal and enthusiastic popular base among the peasants of the northern Mozambique highlands. Bear in mind, for the population of guerrilla “liberated zones” this involved direct personal risks. Moreover, why would these mostly illiterate people eventually turn against their purported traditional chiefs? A more typical outcome should be the fate of Russian narodniki when during the 1870s they went into villages to conduct revolutionary propaganda among the peasants – and were promptly delivered to the Tsarist police by the same frightened peasants. Or recall Ché Guevara’s heroic death in Bolivia. The official histories of FRELIMO are mute on this puzzling point. The peasant grievances against their colonial oppressors and native African exploiters were taken for granted. This simple explanation is no longer tenable. Recent social science research on the dynamics of mass mobilization and revolutions compellingly demolished the “relative depriva- tion” (which is a fancier expression for grievance) model of rebellion.15 To begin with, had oppression been the cause of popular mobilization, strikes, protests and revolutions would be much more common in this imperfect world. For all their variety, the new theories have in common the agreement that political contention is a difficult process which criti- cally depends on the availability of mobilization resources: social networks, organizing nuclei, external alliances, channels for transmitting information, ideological frames and, yes, guns when it comes to it. One of the common shorthands for the new theories of contention is indeed “resource mobilization” paradigm. Let me sketch now how this theoretical understanding of social movements could be applied to explain just one among several factional schisms within FRELIMO in the late 1960s. It would be the faction of Makonde “chiefs”. This group, one of the three “proto-parties” incorporated into FRELIMO at its founding congress in 1962, had originally coalesced among the Makonde diaspora in Kenya and Tanganyika. Typically of many diaspora associations in the world, in the beginning the Makonde Union was a mutual-help association of migrant laborers that primarily aided its members in need or on such life-cycle occasions as weddings and funerals. In 1961 the Makonde activists followed the

15 Charles Tilly, European Revolutions, 1492–1992, Oxford: Blackwell, 1993. Randall Collins offers a sweeping theoretical synthesis in the chapter called the “Maturation of the State-Centered Theory of Revolution and Ideology” in Collins’s recent book Macrohistory: Essays in Sociology of the Long Run, Stanford: Stanford University Press, 1999. 94 georgi derluguian sweeping contemporary trend in declaring their union a political party under the name of MANU.16 The name literally copied the names of Kenya’s African National Union (KANU) and Tanganyika’s TANU. Both parties at the time were nearing power as Britain was preparing for a negotiated decolonization of its East African territories. The name MANU thus expressed not just organizational imitation but also a clear and admittedly rather naïve expectation of spearheading the presumably imminent decolonization of Mozambique. MANU possessed at least two political advantages. First, it operated from the friendly territory of Tanganyika whose ruling party TANU was expressly supportive of Mozambican brethren. Secondly, MANU could derive its political legitimacy from the violent episode which came to be internationally known as the Mueda massacre. In its wake, large numbers of Makonde activists and common population fled across the border into Tanganyika where they were gathered in refugee camps – a ready-made base for mobilization. The Mueda massacre happened at the tense moment when Africans began to anticipate and colonialists began to fear the looming independence. Earlier in 1960 the scheduled meeting of African petitioners with the authorities at Mueda, the far-off seat of Portuguese administration in the Makonde highlands, turned violent after the irate colonial official ordered to arrest the petitioners. Soldiers opened fire on the crowd who tried to liberate the prisoners.17 To the extent of what we know, the petitioners at Mueda were advancing three typical demands: an end the unpaid forced labor to which all “natives” were subjected by the colonial law, allowing Africans to own shops and agricultural cooperatives, and better purchasing prices for their cotton which had to be sold to the government-controlled acquisition board.18 It also matters that among the petitioners were the Makonde laborers who had been recently forced to return to Mozambique due to the economic crisis which adversely affected the sisal plantations in British East Africa with the spread of new synthetic fibers on world markets.

16 The acronym contained what seems an intentional ambiguity: its letter M could be interpreted as either Makonde or Mozambique African National Union. 17 The French researcher Michel Cahen (personal communication) who spent much time disentangling the empirical details of these events concluded that the number of casualties was greatly exaggerated. Still, for the political mobilization let alone human concerns it makes little difference whether there were a ‘mere’ dozen deaths or, as it was claimed, several hundreds. 18 História de Moçambique, Vol. 3 Moçambique no Auge do Colonialismo, 1930–1961, Maputo: Universidade Eduardo Mondlane, 1993, pp. 241–243. the social origins of good and bad governance 95

These men had already witnessed pro-independence rallies in Kenya and Tanganyika. The peasants gathered in Mueda at first were patiently wait- ing for the response to their petition. In a common psychological pattern, after several hours their leaders decided to cheer them up by chanting Uhuru! – the Swahili word for freedom which had just acquired the new meaning of independence. The shooting at Mueda followed the familiar pattern of so many contemporary massacres in colonial Africa and Asia. It started as a local agrarian protest caused by fluctuations in the world markets. Yet it soon acquired the political overtones provoking the irrationally coercive response of colonial state authorities which, in turn, provoked political radicalization. In the words of Charles Tilly, coercion is so often found in the repertoire of state power because it usually works. But violence works only when there is no escape. In 1960 shooting African pro- testors brought radicalization instead of submission because the world geopolitics and geoculture became radically different. By 1960 colonialism became illegitimate and even the illiterate African peasants, observing the political developments in Tanganyika, knew this just as well. When FRELIMO was formed two years later, the Makonde refugees gathered in Tanganyika made up the bulk of its rank–and-file members. In a common African political practice of those days, the Makonde “chiefs” from the MANU ranks were incorporated as the chairmen of FRELIMO’s local chapters. The notable man called Lazaro Nkavandame became appointed FRELIMO’s Head of Civil Administration which was considered institutionally equal to the guerrilla structures. This position might not seem very prominent politically. Yet it turned out to be important and potentially very lucrative when FRELIMO, a nation-state in the making, acquired in the refugee camps and the guerrilla liberated zones inside Mozambique its own economic infrastructure of supply warehouses and peasant cooperatives. It was through this economic infrastructure that the arriving foreign aid was supposed to get to the recepients. His new prominence allowed “chief” Nkavandame to make a rapidly growing private fortune which also served to consolidate his private clientele of corrupt subordinates spread across the various civilian positions within FRELIMO’s emergent state. The story is arguably common in many Third World countries. The organizational and political dilemma now facing the intelligencia leaders of FRELIMO was also quite common. Should they turn a blind eye to Nkavandame’s corruption for the sake of preserving indirect control over the Makonde populations? This strategy might look distasteful yet it had its advantages. As long as Nkavandame and his cronies ensured a 96 georgi derluguian degree of internal compliance, the cosmopolitan intelligencia were free to pursue their political ambitions at the diplomatic front and just wait until international pressure on Portugal would make them rulers of an independent state. In fact, this compact would have served well Nkavandame’s ambitions and thus FRELIMO could preserve internal peace and a semblance of unity. This kind of indirect rule behind the façade of national liberation is what in reality ensured so many “moderate” transitions from colonialism to independence: the aspiring national intelligencia replacing the foreign managerial personnel of colonial apparatus and ruling through local notables. Let me stress, corruption was not anything endemically inscribed in cultural patterns or psychology. It is a political strategy of control based on the negotiated toleration of ineffi- ciency and private abuse of office for the sake of clientilistic indirect rule. This is an appealing choice for many politicians in weak and not so weak states. The activist combative approach is so much rarer because it is more dangerous and difficult. Confrontation with the corrupt subalterns requires great ideological confidence and strong political organization. By the mid-1960s the leaders of FRELIMO had acquired both conditions. In this, they could rely on the flows of resources delivered to them by the contemporary world-system which Mondlane and his comrades had masterfully learned to use to their advantage. Indeed, the world structural conditions in the 1960s were uniquely advantageous to the national state-building projects. The structural advantages, however, needed to be identified and exploited which took some learning process. Admittedly in the beginning Mondlane lacked political experience and made serious mistakes. Understandably enough, for nearly a year he seemed vacillating between his American university job and the career of full-time African revolutionary. More embarrassing is the fact that during this year Mondlane was represented at the FRELIMO offices in Dar-es-Salaam by a shady African-American adventurist who claimed to be a native Mozambican. This scandal might be mentioned to remind us that the struggle within FRELIMO was less clear-sided than later claimed. To his credit, Eduardo Mondlane chose to abandon Western academe and return to Africa to fight Portuguese colonialism – and no less the impostors or corrupt chiefs. Had he lived longer, Eduardo Mondlane would likely become one of the greatest African statesmen. This grandiose statement should not be perceived as an indictment of so many Third World intellectuals today who seek refuge in Western academe. We live in a very different age. Back in 1963 Mondlane’s decision to sacrifice an American academic career and join the armed struggle was the social origins of good and bad governance 97 arguably noble yet it was also rationally political. The FRELIMO was rid- ing the historical wave of the intelligencia-led national state building. It could rely on established revolutionary statist models (socialist or nationalist), international solidarities, low barriers to collective actions and the widely accepted legitimacy of the cause. Feeling the mounting political pressure, Nkavandame sought personal alliances with those FRELIMO members who saw their ambitions stymied by the newly enforced organizational discipline. The oppositionists, logically enough, sought to capitalize on their very provincialism and parochialism. The narrow localism of their cultural practices and social networks was advanced as expressing genuineness, the deep roots in native African soil – unlike Mondlane’s worldly cosmopolitan ism. Mondlane’s dominant faction answered by turning the charge on its head and branding the critics obscurantist, tribalist and feudal istic. In effect, Mondlane acknowledged that his critics were tradi tional personages. Here we encounter another instance of discursive misrecognition. Despite his proud claims, Nkavandame was not a tribal chief. He simply could not be one. The Makonde populations never possessed a centralized tribal organization nor chiefs. To use the jargon of cultural anthropologists, the traditional pattern of Makonde social organization was lineage segmentary and matrilocal.19 To put it in simpler descriptive terms, the Makonde were small-scale subsistence farmers spread across their realm in small villages fitted into the productive capacity of their environment. The nodes of power in the form of an even small-scale chieftaincy could not have emerged in the absence of easily monopoliz- able resources such as cattle herds, warrior bands, extensive ritual networks or valuable trade routes.20 Things began to change only very late, really in the 1950s when the remote lands of Makonde were penetrated by the colonial networks of trade and labor migrations. It is more than a little ironic that, looked closer, Nkavandame represented new, not old life. He squarely belonged to the new rising class of African rural capitalists. The ostensibly traditional institutions of polygamy, ethnic clientism and the binding “gift” exchange in fact represented rational investments in social status with the attendant claims to special

19 Jorge Dias, Os Macondes de Moçambique, 4 vols, Lisbon: Junta de Investigações do Ultramar, 1964–70. 20 Timothy Earle, How Chiefs Come to Power. The Political Economy in Prehistory, Stanford: Stanford University Press, 1997. 98 georgi derluguian rights and, more tangibly, in the household and community labor. The Makonde cotton-growing cooperatives which in the 1950s were fostered by missionaries and colonial authorities became a vehicle for consolidat- ing the emergent social inequality and claiming the exclusive possession of land and crops. The new market-oriented practices ran against the Makonde traditions of lineage land tenure and circulation of goods. Numerous wives, debtors, clients and retainers as well as strategic manipulation of village feasts and gift-giving represented investments in the new forms of social capital. The conversion to Christianity and later the political affiliation with national liberation movements served exactly the same purpose. (Recall the example of Samora Machel’s own father who had pursued essentially the same social strategy – albeit a generation earlier and at the opposite side of Mozambique much closer to the centers of colonial economy.) These African village capitalists were fundamentally insecure in their native social environments because they lacked traditional sanction for social inequality derived from private wealth accumulation. Which is why the fellow Makonde often regarded Nkavandame with unconcealed scorn for his behaving like a white man or Indian trader.21

Achieving Utopia in the Periphery

We are now equipped to disentangle FRELIMO’s key foundational epic: progressive intellectuals united with their people emerging victorious over the foreign and native exploiters. Like all war epics, it served to legitimate the victors. This does not mean, however, that the epic was entirely fictional even if it was shrouded in ideological misrecognition. In the process of pre-independence struggles, FRELIMO did achieve a remarkable degree of organizational cohesion and effectiveness. The afterglow of this success lasted until very recently enabling Mozambique’s ruling party to survive and even prosper in the dramatically different environment of neoliberal globalization. In the case of FRELIMO, the balance of bureaucratic and patrimonial power – which is found in any complex organization – turned out to be less patrimonial and therefore collectively more effective in what regards the manipulation of world-systemic flows.

21 Yussuf Adam and A.M. Gentili, “O movimento dos Liguilanilu no planalto de Mueda, 1957–1962”, Estudos Moçambicanos, 4 (1983). the social origins of good and bad governance 99

This, I claim, was the lasting institutional result of political decisions taken back in the mid-1960s. The “traditionalist” faction of Nkavandame found themselves squeezed by two social forces. From the one side, it was the increasingly effective organizing pattern of FRELIMO which learned and adopted the best contemporary practices of guerrilla armies. From the other side, it was what the leaders of FRELIMO took for granted as the popular support for their correct line, but which proved, in fact, a political manifestation of what James Scott famously called the moral economy of the peasant.22 Under Mondlane, FRELIMO was rapidly evolving into the Bolshevik-type organization which the American Weberian scholar Stephen Hanson has inven- tively characterized as a hybrid “which Max Weber himself could not have imagined: a charismatic bureaucracy”.23 One cannot attribute FRELIMO’s Marxist-Leninist turn to Soviet influence. It was always very weak. FRELIMO’s version of Leninism was rather a common Third World modification which looked loosely Maoist – and certainly nothing like the Stalinist industrial design. The ideological evolution was part of organizational adaptation. The generic Maoist (or Vietnamese) organizational pattern proved most effective in conducting protracted guerrilla wars in the Third World countries. This kind of war was less a military campaign than the process of building an alternative state from the margins of retreating colonial authority. The key was to convince the masses of peasants to accept as legitimate government the radical intelligencia who organized along the lines of modern military command. This tremendous organizational task could not be solved without a strong and necessarily modern ideological platform simultaneously straddling the global discourse of national liberation and the local desires for a social justice. Besides, in Mozambique the leadership was culturally and ethnically even further removed from the popular base than in other Third World countries. Leninism – with or without Marxism – in such situations presented the hefty advantage of formulating the national development project in terms of universal justice and without much reference to ethnicity. At the ideological level, the war for the control of post-colonial state was becoming transformed into the epic class struggle against foreign imperialism and colonial exploitation waged by the

22 James C. Scott, The Moral Economy of the Peasant, New Haven: Yale University Press, 1976. 23 Stephen Hanson, Time and Revolution: Marxism and the Design of Soviet Institutions, Chapel Hill: University of North Carolina Press, 1997, p. 26. 100 georgi derluguian peasants and their intelligencia vanguard and simultaneously by all the like-minded forces in the world. The peculiar international climate of the 1960s-1970s allowed FRELIMO the incredible luxury of war communism turned upside down. The guerrilla forces extracted recruits, porters and food supplies from peasants. Yet they also delivered to peasants the industrially produced goods (blan- kets, cloths, tools), medicines, and school supplies that FRELIMO was receiving from a whole panoply of foreign donors. When Nkavandame and his clique of chairmen were forcibly removed (their militias were no match to the trained guerrillas armed with automatic rifles), the Makonde peasants embraced the FRELIMO military as the givers of social justice. The pragmatic communism of intelligencia vanguard thus joined with the moral communism of peasants.

Conclusion

This story, let me repeat, is only a sketch intended to suggest signposts for new research. The application of Pierre Bourdieu’s notions of habitus and social capital to the study of power practices in the world-system’s periphery promises to explain something previously overlooked in the standard studies of corruption which gravitate either towards narrow empiricism or the excessively abstract economistic models. Instead, we might try to envision corruption as elementary power strategies often pursued by the actors whose skills and endowments are below the levels of more durable and effective bureaucratic governance. We might also get a better understanding, both theoretical and in empirical instances, regarding what exactly is wrong with neoliberal utopia. It is not only socially unjust, it is also destructive of societies and states – in exactly Karl Polanyi’s sense of destroying the substance of human life.24 Put it this way: could have FRELIMO ever reigned in the corrupt neopatrimonial tendencies if the world geoculture prescribed individualism and personal wealth as it does today? The irony is that FRELIMO proved to be good rational bureaucrats in the world’s poorer zone because they were also good utopian egalitarians. Should we then feel nostalgic for the 1960s? Certainly not in sense of producing a huge ideological misrecognition again. Charles Tilly in his beautifully crafted essay “Democracy Is a Lake” wisely warned us against

24 Karl Polanyi, The Great Transformation, New York: Rhinehart, 1944. the social origins of good and bad governance 101 the delusion of detecting one and the only right pattern in the historical formation of institutions.25 Tilly, of course, also made his name in social science by arguing that more effective states which we call modern have emerged from the long-running processes of warfare and popular contention over its costs in the form of taxation. FRELIMO’s trajectory pretty well fits the model. It was once an organization seriously engaged in conflict which it was determined to win. The organizational adaptations found in the course of conflicts served to strengthen the organization. Does this mean, however, that today we should wish for more guerrilla wars? Or, as Tilly suggests, other cumulatively beneficial patterns of organizational growth could be devised? And last, but not least, FRELIMO’s organizational success was clearly related to the inordinate ability of its intelligentcia leaders to benefit from global alliances. Put differently, a progressive transformation of the world’s periphery has to rely on the flow of resources (ideological as well as material) from the core. Such flows could not be directed by markets alone because markets, left to their logic, flow where power and resources are located – and not in the opposite direction. What could be the patterns of global solidarity in the future? This is the big question which we must still keep on our minds as we ponder African politics of the last generation.

25 Charles Tilly, “Democracy Is a Lake”, in Charles Tilly (ed.), Roads from Past to Future. Lanham, Maryland: Rowman & Littlefield, 1997, pp. 193–215.

CHAPTER FOUR

WRITING A NATION OR WRITING A CULTURE? FRELIMO AND NATIONALISM DURING THE MOZAMBICAN LIBERATION WAR

Maria-Benedita Basto

There now exists a very extensive bibliography on the subject of nations and nationalism, with prominent names including Anderson, Gellner, Hobsbawm, Nairn, Smith, Ranger, as well as Chatterjee, Bhabha and Lloyd in the field of postcolonial theory. Yet none of these studies seems to offer a sustained reflection on the complex internationalism of the 1960s and 1970s which sustained the liberation movements of the former Portuguese colonies in Angola, Mozambique and Guinea and (de)formed the traditional European models which were supposed to be transmitted by the colonial system. A reason for this absence might lie in the fact that Anglophone research was not confronted with similar late-revolutionary liberation movements in colonies like India or Indonesia, while in the 1950s ideologies created by the nationalist struggles in colonies like Ghana, Kenya or Tanzania used references to a particular Pan-Africanist, “socialist” strand of internationalism. Another explanation may be related to the auxiliary or transitory status of nations and nationalism within the Soviet, Marxist-Leninist internationalist movement of the 1920s and 1930s, which may deter researchers from focusing on this issue. Placed within a historical dynamic of class-struggle, at least in theory, nations would eventually be dissolved in a centrifugal movement towards the final victory of the proletariat, while the centripetal forces of socialist homogenization lead to the exclusion of the bourgeois enemies of the revolution. With its Universalist outlook Soviet internationalism therefore can be conceived as a form of “civic nationalism”, rooted in the French Revolution, which finds its bourgeois counterpart in the organic, ethno-linguistic model of the nation exemplified by German Romanticism.1

1 See Hobsbawm, Eric, Nations and Nationalism since 1780, Cambridge: Cambridge University Press, 1990.

The internationalism of the 1960s and 1970s, however, presents a more complex picture than the classical Soviet model. Related to anti-colonial and anti-imperialist movements, such as the Afro-Asiatic movement, the Chinese cultural revolution, the group of non-aligned countries, the events of May 1968 and Havana and the Latin-American Tricontinental movement, this internationalism is associated with the introduction of the idea of “revolutionary national culture” as discussed by authors such as Frantz Fanon and Amilcar Cabral, and finding a parallel in Latin American literature in the reflections of writers like Roberto Fernandez Retamar on the revolutionary potential of aesthetics.2 On the one hand, FRELIMO would inherit the difficulties which the Soviet system had with the national question. On the other hand, it would borrow from the internationalism of the ‘60s and ‘70s that makes popular culture the central motor of the revolution. Could FRELIMO not exploit precisely this ambivalence in order to cope with the nationalism question?3 What are the consequences of the use of the internationalist idea of (revolutionary/ popular) culture for the way in which nationhood is constructed by FRELIMO? How do answers to this question translate in terms of theories of civic and ethno-linguistic nationalisms? In this article I propose to provide some preliminary responses to these questions by examining the case of the construction of the Mozambican nation by FRELIMO, the Mozambican Liberation Front. In this respect, FRELIMO had to solve two problems. Firstly, the movement was confronted with the practical necessity of legitimizing the unity and identity of a territory against the hegemonic claims of the colonial empire. Secondly, FRELIMO needed to distinguish itself from other political factions, which had already elaborated their own visions of Mozambican nationalism. Against these movements, which it considered to be colonial and bourgeois,4 FRELIMO’s revolutionary project sought to abolish class inequalities and racism within a broad internationalist movement. In the present chapter, I argue that the solution to the problem of both sustaining a discrete political entity and being internationalist

2 Of course the problem of culture is also present within the Pan-africanist current which advocated an internationalist project preceding all those cited above. 3 In the second chapter of the second part of my Ph.D. thesis I deal with the way in which the FRELIMO and CONCP elites, and the liberation press, related to these different kinds of internationalisms. 4 As Anthony Smith observes, from the revolutionary perspective nationalism always remains suspect as it may be used by the bourgeoisie to cover up class contradictions (Nationalism in the Twentieth Century, New York: NY University press, 1979, p. 142). writing a nation or writing a culture? 105 revolutionaries was to avoid the nation by substituting it with the idea of culture, opting for “writing a culture” rather than “writing a nation”. However, providing the idea of culture with a concrete historical sense and legitimacy meant dealing with a fundamental ambivalence, a tension that I associate with the idea of an “impossible space”.5 In this space, the desire, the dream and the ideal of a new nation (which, in the case of FRELIMO, correspond to a new society characterized by the birth of a “New Man”) are confronted with the difficulty of constructing the instruments and devices necessary to constitute a feeling of belonging which embraces the socio-cultural particularities of the future citizens of the nation, taking into account their practices and real experiences as historical subjects. This “impossible space” did not give rise to a movement of negation or rejection. On the contrary, it provoked a mythical hyper- affirmation of the space, a result of the desire to establish a nation/ culture/people based on a perfect concurrence of discourse and place, what Jacques Rancière calls a “utopia”.6 In the following pages I will approach this idea of writing culture by considering two paradigmatic cases: the preparation of an anthology of poetry and the organization of the “1st Seminar on Culture”. I suggest that in both instances the writing of culture resulted in the creation of a link with the past through what Eric Hobsbawm and Terence Ranger call “an invented tradition”.7 In the present context, the term relates to “an unchanging, invariant tradition”8 in which a common past is created to be perpetuated and anchored in the experience of the liberated zones and liberation struggle. This leads me to consider some further questions: what were the characteristics of this diversity of experiences before their reduction? What does the writing culture/writing nation contrast mean from this localized angle? Can we be sure that, from a grassroots perspective, Mozambican nationalism was not, as Partha Chatterjee9 suggests, a mere “derivative” discourse reproducing the homogenizing and dichoto- mizing colonial model it rejected?

5 For a more detailed discussion see Basto, Maria-Benedita, A Guerra das escritas. Literatura, nação e teoria pós-colonial em Moçambique, Lisbon: Edições Vendaval, 2006, part 2, chapter 1. 6 Rancière, Jacques, La chair des mots, Paris: Galilée, 1998, p. 27. 7 Hobsbawm, Eric and Ranger, Terence (eds.), The Invention of Tradition, Cambridge: Cambridge University Press, 1983. 8 Idem, p. 2. 9 Chatterjee, Partha, Nationalist Thought and the Colonial World. A Derivative Discourse? Minneapolis: University of Minnesota Press [1986] 1993. 106 maria-benedita basto

Printing Books: Why Poetry in Times of War?

In 1971, students from Finland donated a modern printing machine to FRELIMO. That same year FRELIMO published its first two books: a thin pamphlet featuring an essay from President Samora Machel Produzir é aprender. Aprender para produzir e lutar melhor (Producing is Learning. Learning to Produce and Fight Better), which launched the collection “Estudos e Orientações (Studies and Orientations)”, and an anthology of poems bearing the title Poesia de Combate (Poetry of Combat). One wonders why the guerrilla movement published a collection of poems instead of something more “useful” related to the practical problems of the struggle. But perhaps this is a wrong question, and we should rightly ask: why is this anthology “useful”? Given that FRELIMO had successfully weath- ered operation “Gordian Knot” (No Górdio), a major offensive launched by the Portuguese colonial army the previous year, it might seem that the time had come to satisfy the familiar need of nationalist movements to associate proof of the existence of a people with the presence of a national literature. Literature can provide the nation with a poetry, a style and a pedagogy which may appeal to the nation’s “aesthetic imagination”.10 From this perspective, culture is political because it fulfils a necessary role of guaranteeing the cohesion of the nation-state. As Ernest Gellner puts it: “Modern man is not loyal to a monarch or a land or a faith, whatever he may say, but to a culture”.11 However, I would argue that the “use” of culture as represented by the anthology derives from an inversion in which it is not the struggle for national sovereignty which becomes the foundation of a political culture, but a cultural struggle which constitutes the base of the (revolutionary) state. As I mentioned earlier, such a strategic move from “writing a nation” to “writing a culture” is inspired by the ideas of the anti-colonial, internationalist movement of the 1960s and 70s. Its characteristic reasoning is exemplified in Amilcar Cabral’s famous phrase: “(…) if imperialist domination has the vital need to practice cultural oppression, then national liberation is, necessarily, an act of culture”.12 The struggle as cultural act, which is here represented by Poesia de Combate, not only becomes a political underpinning of the State, but also

10 Smith, Anthony, Nationalism and Modernism. A critical survey of recent theories of nations and nationalism, London and New York: Routledge, [1998] 2003, p. 234. 11 Gellner, Ernest, Nations and Nationalism, Oxford: Blackwell, 1983, p. 36. 12 Cabral, Amílcar, Unity and Struggle: Speeches and Writings, London: Heinemann, 1980, p. 143. writing a nation or writing a culture? 107

Figure 1. First edition of the anthology Poesia de Combate, 1971, one of the first two books published by FRELIMO. 108 maria-benedita basto the foundation of an invented tradition associated with the liberated zones, providing an appropriate past that can be repeated in the present. In this respect, the way scientific research dealt with Poesia de Combate supported such an invention by, albeit unintentionally, sustaining the idea of the “unchanging and invariant”13 nature of this revolutionary literature. The practical and contextual characteristics of Poesia de Combate set it apart from “high” literature, thus making it unattractive to literary scholars. As a result, even after an initial period of politically inspired sympathy for the revolutionary perspective as represented by the notion of “poetry of combat”, the question of whether material which was mani- festly circumstantial, dated and defined by conventions of Socialist real- ism could raise any important empirical or theoretical issues was not addressed. Moreover, because the 1971 edition of the anthology became rare after independence, general readers and researchers mostly referred to this anthology through a second edition published in 1979 by FRELIMO (edited with the title Poesia de Combate 1).14 However, what was held to be identical to the first edition was actually not the “same” book.15 The two

13 Hobsbawm, Eric and Ranger, Terence (eds.), The Invention of Tradition, p.2. 14 Sometimes Poesia de Combate 1 and Poesia de Combate 2 (a book that FRELIMO published in the wake of its Third Congress of 1977 and which included authors such as the established poets Marcelino dos Santos and Sérgio Vieira, accompanied by José Craveirinha and Rui Nogar) are mistaken for each other. The mix-up between the two works, very different in terms of content and function (one should say three because, as I have already shown, the two editions of the first volume are not identical) eliminates the stakes related to the first anthology and therefore an analytic interest. In this field the most important references are Ferreira, Manuel, No reino de Caliban III – Moçambique, Lisbon: Plátano Editora, 1985; Mendonça, Fátima, Literatura moçambicana. A história e as escritas, Maputo: UEM, 1988; Hamilton, Russel, Literatura africana, literatura necessária II, Moçambique, Cabo-Verde, Guiné-Bissau e São Tomé e Príncipe, Lisbon: Ediçoes 70, 1984; Chabal, Patrick, Vozes Moçambicanas. Literatura e nacionalidade, Lisbon: Vega, 1994; Chabal, Patrick, The Postcolonial Literature of Lusophone Africa, London: Hurst & Company, 1996; Laban, Michel, “Écrivains et pouvoir au Mozambique”, Lusotopie, Paris: Karthala, 1995, pp. 171– 180; Laranjeira, Pires, De letra em riste. Identidade, autonomia e outras questões na literatura de Angola, Cabo Verde, Moçambique e São Tomé e Príncipe, Oporto: Afrontamento, 1995; Saúte, Nelson, Nunca mais é sábado. Antologia de poesia moçambicana, Lisbon: Dom Quixote, 2004. No reference is made to the anthology Poesia de Combate or to any other literary document produced during the armed struggle in Gomes, Aldonio and Cavacas, Fernanda, Dicionário de autores de literaturas africanas de língua portuguesa, Lisbon: Caminho, 1997. 15 In A New Bibliography of the Lusophone Literatures of Africa (London: Hans Zell, 1993, revised ed.), Moser and Ferreira point out a difference between the two editions concerning the number of poems. According to them, the 1979 edition includes an additional poem: “A Luta Armada” by Mahasule. For the rest the authors consider the editions to be identical (idem, p. 268). This claim however is not exact because the poem in question is just missing from the table of contents and not from the collection. writing a nation or writing a culture? 109 editions differ in terms of layout, chosen illustrations (real photos instead of drawings) and the order of the poems. The new ordering of the 1979 edition tells a different history. Briefly: the first poem “O guerrilheiro” is placed in the middle of the book; the poem “Venceremos”, originally in the middle of the first part of the book, is now the opening poem; the poem which was last, “Morrer para patria”, now comes near the beginning – in fourth place; and the last poem in the new edition, “Guerrrilheiro, guia do povo”, was originally in the middle of the collection. While the first edition starts naturally with the figure of the guerillero, author of the poems and actor of the struggle and leaves open the outcome of the liberation struggle at its end, insisting above all on the necessary sacrifice of one’s life, the second edition affirms the legitimacy of FRELIMO’s rule, beginning with the victorious outcome and concluding with the figure of the guerillero as “guide of the people” and model for society after independence. Contrasting with the uncertainty inherent in the desired future evoked in the first edition, in Poesia de Combate 1 the poems of the desired future now justify the victorious present and ascribe to it an unequivocal foundational root: the liberation war.16 Why did FRELIMO not simply reprint the book? Why not leave the order of the poems unchanged? Reading the 1979 edition while believing that it is just a reprint of the book of 1971 opens up the possibility of posit- ing the struggle as a foundation myth. This new order corresponds to a linear writing of history. Such linear narrative supported the idea that Poesia de Combate represented the written result of a homogeneous community ready to become the affirmed model of independent Mozambique. Departing from this idea of homogeneity and coincidence of texts and realities, the necessity to investigate where the poems had actually come from, and where, when and by whom they had been written was not felt. Because this was not done, it was furthermore not possible to know that the poems of both editions of the anthology were not exactly the poems originally written by the guerrilleros. Before dealing with the making of Poesia de Combate, I would like to begin with a study of the provenance of the writings of the guerilleros and then briefly present some reflections on the creation of a canon which opened up the possibility of governing a territory through what I call “writing culture”.

16 Basto, Maria-Benedita, A Guerra das escritas, pp. 122–125. 110 maria-benedita basto

Poesia de Combate: Where Did the Poems Come from?

Concerning the guerilleros’ writings, I conducted an enquiry in the archives of FRELIMO and the Mozambican National Archives between 2000 and 2005, which revealed that the poems in the anthology actually came from newspapers produced by FRELIMO militants. My investigation exposed a network, or what one might describe as a decen- tralized press, produced during the liberation struggle as a “local press”. The writings were thus published in politico-military preparation camps in Tanzania, in bases in the North and the East of Mozambique, as well as in FRELIMO schools – a press which should be considered separate from the publications produced in Dar-es-Salaam by the central power structures.17 The writers were grass-root level and intermediary level officials, students and fighters. All of these men and women were subsumed under the category “guerrilleros” or “militants”, even if they were not actually engaged in combat. The writers were predominantly men, but female exceptions were found in the newspaper 25 de Setembro. In this newspaper, one of the oldest, writers described themselves as “reporters” and “correspondents”, thus emphasizing their journalistic function. The press discussed here includes eight newspapers.18 They can be divided broadly into three categories: those published in military training camps, those published in schools, and those published in military bases.19

17 Colin Darch (“As publicações da FRELIMO: um estudo preliminar”, Estudos Moçambicanos, No. 2, Maputo: Centro de Estudos Africanos da UEM, 1981, pp. 105–120) and Ilídio Rocha (A Imprensa de Moçambique. História e Catálogo (1854–1975), Lisbon: Edições Livros do Brasil, 2000) have previously produced inventories of the Mozambican press, which are either restricted to the liberation press of Dar-es-Salaam or only report the existence of a very limited number of journals. In Basto (A Guerra das escritas) I provide a contribution to a history of the local press during the Mozambican liberation war. 18 25 de Setembro, Os Heróicos, Jornal do Centro, Rasgando as Trevas, O Camarada, A Luta Continua, 3 de Fevereiro, Jornal Semanal de Tunduru. 19 25 de Setembro was launched in November 1965, and not in 1966 as Rocha suggests (2000: 376). It was published until 1979. Together with Jornal do Centro, which started 6 years later in 1971 and ended in 1975, it was edited in Nachingwea, FRELIMO’s most reputed training camp. Jornal Semanal do Tunduru, beginning in 1973 and ending in 1974, as the name indicates, was published in the camp of Tunduru. Published from 1969 until after independence Rasgando as Trevas and O Camarada, which started in 1973, were the result of the work of students in the schools of FRELIMO, respectively the Educational Camp of Bagamoyo and the Secondary School of Bagamoyo. Finally, Os Heróicos (1970– 1975), A Luta Continua (1971 - ?) and 3 de Fevereiro (1971-?) are related to bases of FRELIMO on Mozambican territory, respectively Cabo Delgado, Niassa Oriental and Tete. The paper of these publications is rough and yellow. The journals were typed and then reproduced writing a nation or writing a culture? 111

Figure 2. Cover page of Os Heróicos, one of the guerrilleros’ newspapers.

with stencils and in some cases with rudimentary manual printing machines. The average number of pages is between six and eight, but they can range from one (the daily issue of Jornal do Centro) to thirty-four (commemorative numbers of 25 de Setembro). The format is between A4 and A5 and there are no photographs. However, we do find various illustrations in the form of hand-made drawings and sometimes cartoons. Journals often are monthly, but in some instances, as in the case of the Jornal Semanal de Tunduru, they are weekly, or daily, as in the case of Jornal do Centro. 112 maria-benedita basto

The newspapers were considered “local” by their writers as well as by FRELIMO leaders.20 Remarkably, we read in various texts published in different newspapers by these local writers about the importance of the fact that this “local” press covered all provinces where combat zones and liberated zones existed, and that together they formed a kind of network which contributed to the production of a unified “national consciousness”. Put differently, a typographical space covered a cartographical one and therefore delimited a “new” territory which should correspond to a unity of belonging. Let me give some examples of how the network operated and worked to raise consciousness. The first is from the newspaper 25 de Setembro, which is one of the most well-known of these publications – its distribu- tion spread to expatriates living in Zambia and Malawi as the exchange of letters among militants suggests. In a short article, an account is given of the first edition of the journal A Luta Continua in November 1971. The author notes that the new newspaper is rich in information concerning “our” war, “our” social life and programs of “national reconstruction” in Niassa Oriental, but also provides news about other provinces such as Cabo Delgado and Tete.21 This is done through an exchange among the three provinces in struggle: “this is the way the uniformization of information on Mozambique is guaranteed”.22 The article ends by lauding the editors of the newspaper who confront a lack of material and staff and it stresses the fact that “we now have newspapers in all three provinces in combat: Heróicos from Cabo Delgado, 3 de Fevereiro in Tete and A Luta Continua in Niassa Oriental”.23 The second example is from an editorial in the same newspaper in which the authors seek to provide an overview of the activities and functions of their own periodical among which they cite the “raising of political consciousness” by reading and by direct participation through “writing articles”, namely on the “formation of a national consciousness”.24 This points to the fact that, in the context of the revolutionary struggle, national consciousness was seen as political and vice-versa.

20 Jorge Rebelo, who was responsible for the Department of Information during the struggle, told me that “local information organs” emerged because FRELIMO did not have the logistic means to transport information material to the war zone. The idea initially came up when military bases won a battle and wanted to spread the news; an action which was encouraged by FRELIMO, who wanted to mobilize people (interview, Maputo, September 2001). 21 25 de Setembro, No. 71, year 7, 30–12–71, p. 15. 22 Idem. 23 Idem. 24 25 de Setembro, No. 56, Feb./March 1970, p. 4. writing a nation or writing a culture? 113

Finally, another example comes from Os Heróicos in which a text congratulates another periodical, Rasgando as Trevas, for publishing information about the creation of 3 de Fevreiro in Tete. The directors of Os Heróicos conclude the article by recommending that the news organs with “local character” make an effort to build a unity of action. This, according to them, could be done by exchanging articles and staff.25 These statements show a process similar to the one described in Benedict Anderson’s Imagined Communities. Emerging communities of writers and readers produced a shared ideal of belonging. This emerged through local experiments that were interlinked in a loose network. However, as I suggested above, in the early 1970s FRELIMO reached a point where it sought to consolidate its identity both within the country and with respect to the internationalist movement by establishing canonical cultural rules which had as their corollary canonical forms of citizenship. The organization of FRELIMO’s “First cultural seminar” exemplifies this approach, which I suggest finds a parallel in the publication of Poesia de Combate.

Establishing a Canon: Frelimo’s First Cultural Seminar, 1971–72

This seminar, which took place between 30 December 1971 and 21 January 1972 and which has not yet been studied, represents a very important moment in FRELIMO’s work as well as a major effort if we consider that it was held under very hard conditions during a war. In 25 de Setembro, the presentation of the seminar underlines both its relevance and national character: There has never been a similar occasion in which Mozambicans of different parts of the country have found themselves united in a conference on National Culture. In the history of almost eight years of armed liberation struggle in Mozambique, this seminar marks an important stage of our development. The seminar has clarified the essential questions: ‘What is Mozambican Culture? What in our culture has been preserved? How should we develop our Culture?26 In my research in the archives of FRELIMO, I found several documents concerning the event. Two reports help us to understand its important political and historical dimension. The first, a document produced by the

25 Os Heróicos, No. 14, 01–10–71, pp. 3–4. 26 25 de Setembro, No. 72, 30–01–1972, p. 14. 114 maria-benedita basto

Department of Defence in the Camp of Nachingwea, and dated 31st of December 1971, states: The first cultural seminar opens in Tunduru. By telegraph our Center of Politico-Military Preparation received a memorandum sent by the team of the “Jornal Diário” of the Educational Centre of Tunduru. It informs us that the First National Cultural Seminar was opened on Thursday the 30th of December 1971 at 9 a.m. According to the memorandum this seminar seeks to study and analyse in depth the various aspects of our life concerning the development of our society, in particular the cultural aspect of our country, as it is expressed by our revolutionary life.27 The document also provides information about the identity of the participants and the fact that they were intended to reflect the territorial representation of a true Mozambican culture: Armando Guebuza, National Political Commissary, a high FRELIMO official; delegations of the Centres of Nachingwea, Tete, Eastern Niassa, Western Niassa, Bagamoyo, Rutamba, and Tunduru; students, militants and militants’ wives. The document makes reference to a more recent press release received in the camp and signed by Armando Guebuza, who is today Mozambique’s elected President, which reads: This event marks an important phase in the history of the liberation struggle of the Mozambican people, in the creation of a new man and in the improvement and unification of our culture.28 Several commissions were created to analyse the different themes into which the seminar was divided: Working Commission on Dances and Songs, Working Commission on Drama and Poetry, and Working Com mission on Miscellaneous Forms of Cultural Expression. The other archival document, a brochure of 18 pages entitled “1° Seminário Cultural”, presents the report and final resolutions of the seminar. It declares on page 1 that: Through our social and practical life, one realizes that there is no people without culture. It is a significant task of the Mozambican Revolution to inspire the development of a Mozambican culture, making it national, popular and revolutionary.29

27 Relatório de Actividades Nachingwea/Activity Report Nachingwea – Doc. DD, n. 35, ARQFRE 1971, p. 1. 28 Idem, p. 2. 29 Brochure “1° Seminário Cultural, Dez. 71/Jan. 1972”, 1972, Caixa DEC, doc., p. 1 writing a nation or writing a culture? 115

Figure 3. Cover of the pamphlet resulting from the first seminar on culture, 1971/72.

With regards to the different orientations identified, it is interesting to see what is said concerning dance, song and storytelling. Here one can detect a familiar approach, inspired by the conventions of Socialist real- ism, in which the whole question of tradition is relegated to a secondary 116 maria-benedita basto level that only concerns the sphere of form. Within the exterior “enve- lopes” of traditional oral stories, as well as those of songs and dances, form is preserved while content is altered. The secondary role of tradition also concerns the difficulty of integrating diversity beyond the “formal” level. Here young people are encouraged to become agents of transformation of tradition. However, the measure of this transformation is less the dynamics of the population than the projected development of “the new revolutionary modern society” with which Mozambique should be identified. As a result, FRELIMO ends up struggling to produce an inclusive and negotiated vision of nationalism. The project of the anthology Poesia de Combate may be seen to form part of the same trend of centralization and standardization. In a certain way the compilation of the anthology opened up the path to the cultural seminar. This was primarily due to the fact that this project involved a close reading of the poetry written in the local press and thus provided FRELIMO with an insight into the cultural dimension and practices of its militants, as well as ideas concerning the necessary corrections towards building its new canon. The preface of the anthology presents this poetic work as homogeneous. But did the poetical experiments of the guerilleros correspond to what FRELIMO presented as the anthology of “our” “Mozambican” poets?

The Making of the Anthology

The anthology is entitled Poesia de Combate and its subtitle (lost in the 1979 version) is “Poemas de militantes da FRELIMO. Caderno n°1 [Poems from Frelimo militants. 1st notebook]”.30 The spirit of the book is similar to a very simple object, a notebook, where the “militants”,

30 Mutimati Barnabé João, the guerillero’s heteronym of António Quadros/Grabato Dias deeply understood this kind of manufactured object, when he used as subtitle “second notebook” in the manuscript of Eu, o Povo he left to Rui Oliveira. The book of the same title was the national poem he dedicated to Mozambique’s independence. FRELIMO published it on the eve of independence in 1975 and the publisher, the same Rui Oliveira, changed the subtitle to “poems of revolution” (see pp. 190–195, Basto, Maria-Benedita, “Enjeux, double je(ux): hétéronymie, genre et nation dans Eu, o Povo de António Quadros/ Mutimati Barnabé João” in Maria-Benedita Basto (dir.), Enjeux Littéraires et construction d’espaces démocratiques en Afrique subsaharienne. Littérature, politique, identités, Paris: EHESS, 2007, pp. 183–216). As I have already mentioned, the 1979 edition of Poesia de Combate completely changed this “handmade” aspect of the book. The latter version also erases the subtitle and uses more “real and scientific” photos instead of “fictional” drawings. writing a nation or writing a culture? 117 meaning those who are engaged in the struggle, can build a new Mozambican writing together. The anthology is illustrated with naive drawings, resembling the comics or cartoons one finds in local publications and diaries. The anthology contains 23 poems by 20 poets.31 In my investigations I succeeded in retracing the origin of 18 of these poems, all drawn from 25 Setembro.32 The preface of the anthology presents the theoretical lines defining the project, the basic idea being that the revolution gave birth to a new (homogeneous) canon. Significantly, this contrasts with the findings of my research which suggest that the anthology did not provide a mirror reflecting the variety of existing poetic experiments. The following analysis starts out with the title of the collection by situating it in relation to ideas circulating within the internationalist movements. As I argue elsewhere, in this respect the appropriation of Frantz Fanon is essential: This may be properly called a literature of combat, in the sense that it calls on the whole people to fight for their existence as a nation. It is a literature of combat, because it molds the national consciousness, giving it form and contours and flinging open before it new and boundless horizons; it is a literature of combat because it assumes responsibility, and because it is the will to liberty expressed in terms of time and space.33 In accordance with Fanon’s argument, the choice of the title of this anthology already presupposes a poetry/nation relationship, given that “poetry of combat” is what “informs” and “gives form” to a “national conscience”. By this, one should also understand that the poetry of this

31 “O guerrilheiro” (Daniel Cosme), “Irmãos que esperam” (A. Rufino Tembe), “Moçam bique chorou e chora” (Jackson), “A luta armada” (Mahasule), “Até ao fim” (Malido), “Venceremos” (Alfredo Manuel), “Foi o que disse mamã” (Xicalavito), “Minha mãe” (Kantumbyanga), “É de madrugada” (Comodoro), “Moçambique diz” (Alfredo Manuel), “Guerrilheiro Guia do Povo” (Ngwembe), “Mãe e filho” (Domingos Sávio), “É nosso dever” (Atumbwidao), “O guerrilheiro em marcha” (Damião Cosme), “Creio em ti herói” (Omar Juma), “A luta justa” (Domingos Sávio), “Farol da liberdade” (Rafael Bobo), “Plantai árvores” (Manuel Gôndola), “Mamã” (Maguni), “O meu tombar no combate” (Polvo Cheirinho), “Moçambique” (Kumwanga), “Canto de um militante” (Djakama), “Morrer pela Pátria” (Luchwacha). 32 Poems came from the local press, but most of them were written between 1965 and 1970. In fact, the beginning of the seventies marks a change in 25 de Setembro (most of the other journals did not exist for as long): from being “the voice of all militants” it was transformed into “the organ of the Political Commissar”, mostly featuring political discourses or documents and losing its spontaneity. The newspapers of military bases that were created in the beginning of the 70s tried to renew this spontaneity by publishing literary texts, but they were unstable and usually re-published older poems. 33 Fanon, Frantz, Wretched of the Earth, New York: Grove Press, 1963, p. 240. 118 maria-benedita basto anthology is defined in the introduction as “a slogan” corresponding to the third and highest level of the Fanonian classification of poetry, defined as a true realization of national poetry.34 Fanon’s model allows us to see how the introduction clearly situates the anthology with respect to the construction of a national literature. The first three paragraphs of the introduction present the capitalism/ revolution and capitalism/people oppositions, placing colonialism into a direct relationship with the capitalist system. Thus what is at stake here is not just an anti-colonial struggle, but a fight against capitalism and imperialism of which the key word is Revolution. Revolution is actually one of the first words of the text: “Graças à Revolução em Moçambique, a poesia…”/“Thanks to the Revolution in Mozambique, the poetry…”.35 This way of arranging the facts gives rise to an idea of the people as horizon and origin. As mentioned above, here the people – whose creative energy is liberated by the revolution – finally gain access to poetry, a space and an art which the capitalist-colonialist society reserved to intellectuals and the bourgeoisie.36 The first three paragraphs of the introduction, which represent half of the text, create the context which governs and regulates the Mozambican struggle and its new poetry. It is an international context which directly legitimates the new poetry through extra-local categories, without relying on the Mozambican historical or social background. Only afterwards does the text deal specifically with the anthology and the writings it includes, by beginning with a definition of the new poet: who are these poets? Who can be a poet? The poets of the new Mozambican poetry are, as the text specifies, FRELIMO militants who are directly “engaged” in the struggle. This is the condition which both characterizes and legitimizes their writing. It is also the aesthetic base of this poetry: it can only be truthful because the poets are directly engaged in the struggle. Correspondingly, the next phrase begins with the expression: “[b]ecause here is the essential characteristic of today’s Mozambican poetry: there is an absolute identification between

34 The following citation gives a brief idea of Fanon’s classification of anticolonial poetry: “The lament first makes the indictment; then it makes an appeal. In the period that follows, the words of command are heard. The crystallisation of the national consciousness will both disrupt literary styles and themes, and also create a completely new public. (…) It is only from that moment that we can speak of a national literature.” (Fanon, Frantz, Wretched of the Earth, pp. 239–240). 35 FRELIMO, Poesia de Combate. Poemas de militantes da FRELIMO, Caderno No.1, Dar- es-Salam: FRELIMO/Departamento de Educação e de Cultura, 1971, p.1. 36 Idem, p. 1. writing a nation or writing a culture? 119 revolutionary practice and the sensitivity of the poet”.37 This principle, which establishes a strict homology between biography and literature, also provides support for the assertion that this kind of poetry “does not speak of myths, of abstract things”.38 So what does this poetry speak about? The answer to this question provides the first opportunity for a massive appearance of deictic forms per- taining to the affirmation of belonging to a community. Accordingly, in the course of the last three paragraphs of the introduction, the pronoun “our” is present sixteen times: this poetry speaks about “our life of struggle, our hopes and certainties, our determination, our love for our comrades, (…) our country”.39 Finally, the last paragraph affirms that this volume of poetry is an “anthology” which shares the experiences of [our] poets of today.40 Thus it is by establishing a link between “anthology” and “our” that the national space of this book is outlined. One may notice that apart from the expression “armed struggle of national liberation”,41 the word “national” is never used and “country” is mentioned only once. Therefore the “national” is delimited and defined through the common belonging to the space-time of the armed struggle of liberation. The latter is the referent and the content of the nation to be built, the referent and content of the poetry to be written. Reading the anthology, it becomes clear that no space for personal experiences or diverse “styles” – things that one would normally expect from a collection – exists. Among the texts included, there is no conflict, and within each of the poems no tension is present between an inside and an outside. The anthology seeks to display itself as homogenous texture, as mature fruit of a uniform “common thought”. But a collection made up of twenty-three poems written by twenty poets, claiming to be a variation without variety on the same theme, with the same “style”, the same relationship to the past, present and future, the same “new” voice constructed by “totally” “new” men, transformed by a struggle which becomes the sole emotional referent, certainly requires some reflection. Do we not detect here a certain incapacity to think “decolonization” as a process marked by conflict,

37 FRELIMO, idem., p. 1. 38 Idem, p. 1. 39 Idem, pp. 1–2; my italics. 40 Idem. 41 Idem, p. 1. 120 maria-benedita basto various ambiguities and ambivalence? Should anti-colonial nationalism necessarily take the form of a “derived discourse,” as Partha Chatterjee has asked?42 A comparison between the body of the poetry published in the anthology and the poems printed in local newspapers leads to the following conclusion: rather than presenting a sample of the guerillero’s poetic work, the anthology works as an “invention of tradition” along the lines of Eric Hobsbawm and Terence Ranger’s thesis.43 “Invented” does not mean unreal: our analysis shows how the invention of a past through the present, which in the case of the anthology works as a memory of the future, institutes and orders society. This invention can be defined by two strategies: exclusion and correction.

Inventing by Exclusion and Correction

Concerning exclusion, four types of poems were left out of the collection: poems written in local language; poems written in a pan-African nationalist tradition drawing on the heritage of Noémia de Sousa and José Craveirinha (authors who had been excluded from FRELIMO’s revolutionary canon); love poems; and finally, a number of poems which were the product of a rewriting of Portuguese canonical texts, literary texts or popular songs. Significantly, here, in the excluded poems writing involves coming to terms with the past and its memory in a way that negotiates an identity by subverting and dislocating a preceding order without dichotomising it. A citation of a guerrillero drawn from an article entitled “Culture” in the first issue of 25 de Setembro illustrates the process. In this instance, we encounter the negotiated and inclusive spirit of “concerts”44 in which the revolutionary does not replace the traditional but includes it in militant practice (the brackets used by the author stress this approach):

42 Chatterjee, Partha, Nationalist Thought and the Colonial World. A Derivative Discourse?, p. 30. 43 Hobsbawm, Eric and Ranger, Terence (eds.), The Invention of Tradition. 44 “Concerts” was the name given to the cultural activities at bases or military training camps which normally took place on weekends and consisted of theater, dance, songs and the reading of poems. writing a nation or writing a culture? 121

(…) I am very happy for this opportunity to write about Mozambican culture. (…) In my point of view in the future our system will be adequate for the masses. (…) Our (revolutionary) traditional concerts should be reborn, consolidated and developed by the youth, outdoors, popularly, in school buildings, in public buildings, in suburbs and rural camps with the objective of making our country into a real one.45 Let me provide another example for such a strategy of rewriting the past, in this case represented by one of the categories of excluded poems, those dealing with the imperial Portuguese canon. It concerns a poem called “Caminho percorrido”/“The way left behind”, written by Polvo Cheirinho, in which an initially epic diction rapidly descends into a colloquial style. It mixes three types of language coming from three texts that exemplify what I call an Imperial Library:46 the Lusiadas, by Luis de Camões, the Mostrengo, by Fernando Pessoa, and possibly the Auto das Barcas by Gil Vicente. The poem describes the path of a guerrillero in his fight from the moment when he leaves to join the liberation struggle until his possible death and posterity. It opens with an evocation of the difficulties along the path: the lack of water in the zones through which the guerrilleros have to pass. Observing attentively the gourds Of the water of our path, one knows Easily what are the obstacles of Our Revolution and its partisans In the second strophe, those who do not abandon are guided by the “ship of the revolution” which is opposed to the “caravella of the discoveries”. So these sailors sit down on the beach of the Indian Ocean, meaning they leave Tanzania where the training camps are, and enter Mozambique which, like Portugal, is represented by a synecdoche in relation to the Ocean.

45 25 de Setembro, No 1, ano 1, 21–11–65, p. 5. 46 The concept of Imperial Library points at the constitution by the colonizer of a literary corpus which has an ideological finality diffused through didactical devices. This finality develops on three levels – the historical, the anthropological and the juridical – which respectively enounce the value of the Lusitan race, its civilizing mission and the legitimization of the empire (Basto, Maria-Benedita, A Guerra das escritas, p. 191). 122 maria-benedita basto

Well adventured are those who Are guided by the ship of the Revolution, taking leave Joyously from their dear brothers Bye, Bye my dear brothers I sit down On the beaches of the Indian Ocean Here, at the point of encounter between the Ocean and Mozambique comes an encounter with a “giant”. An incongruous giant: living on the bottom of the sea, where he comes from in the canonical texts, advising the fighter to follow his steps on his carriage. Said the Giant which was In the deep waters of his Ocean: Yes, from there you will follow In my path on my chariot This same giant reappears in the poem at the moment of the hypothetical death of the soldier, victim of the struggle, who remains “at the side” of “his” giant. I am proud because I am the son of pride I will be victim of this ship And they will bury me next to my giant. We must note that with the expression “my giant” the poet appropriates a figure which was part of the Portuguese mythology and thus a symbol of the dominant culture and makes it “his” friend. For this the author explores a suppressed side of the Imperial library because the giant was actually the mythical Adamastor, the first enemy of the Portuguese expansion, which dared to invade its territory in the quest for wealth and glory. The Imperial Library uses the Adamastor episode to emphasise the force of the Portuguese race. Polvo Cheirinho, in contrast, chooses a different option, “retelling the story” and thus making use of the liberty to dis- identify with the imperial narrative. Meanwhile, this poem operates a second process of disidentification, this time in relation to the Manichean identity of the Revolutionary “new man” in FRELIMO’s latter discursive phase. In this respect, the exclusion of poems which rewrite the imperial canon from the anthology gives a clear indication of the centralized dynamic to which FRELIMO was now committed. The exclusion of the kinds of poems discussed above was accompanied by correction of the poems that were included in the anthology. When I compared the selected poems with the originals, I soon realised that they were not exactly the same. With changes to spelling and style and writing a nation or writing a culture? 123 erasure or replacement of words, verses, or even strophes, the modifica- tions I discovered yielded “cleaned up” canonical versions of poems that had previously been characterised by the spontaneous register of the immediate experience of the author, complete with his or her doubts and ambiguities. These poems sometimes express a certain innocence; they convey a sense of revolutionary ardour, without necessarily being strate- gically ideological. Creating a patchwork which crosses codes and influ- ences, they represent performative, inclusive practices. They are written in a Portuguese language which is appropriated without complete mas- tery of the official grammar and mixing esthetical and linguistic codes. An appropriation of the present revolutionary time is combined with a non- radicalised relation to the past. Let me give two short examples: The well-known poem “Plantai Arvores” (Let’s plant Trees), by Manuel Gondola, celebrated as a revolutionary appeal to a construction of the new way of life, was originally a call addressing young boys and children. The poet writes “Plantai árvores, crianças/Plantai árvores, meninos” (Let’s plant Trees, children / Let’s plant trees, little boys), which are then replaced in Poetry of Combat by “Plantai árvores, camaradas/plantai árvores camaradas” (Let’s plant trees, comrades). In the poem “Moçambique diz” (Mozambique says), by Alfredo Manuel, where the author writes: “Your death is my happiness/Merits my future liberty”, in Poetry of Combat we find: Your combat is my happiness/Is my future liberty”. In the next strophe, A. Manuel says:

So that I can be free, you have to die And you spread your heroic blood day after day Adapt to circumstances Because that’s the Wealth of the world So that I may become joyful You have to fade away My existence is only for you My son. In Poetry of Combat we can now read:

So that I can be free you have to stand up And you have to spread your heroic blood day after day Adapt to the circumstances as wealth depends on the future So that I may become joyful You should not hesitate My existence is only for you My son. 124 maria-benedita basto

These corrections and the exclusions noted earlier confirm the constructed character of the time-space which the liberated zones are held to represent in history, hiding, according to Partha Chatterjee47 an unequal power relation. In this respect, the construction of the anthology reflects a situation in which the citizen-militants are not really accepted as they are, neither are the alternatives which they point at. Ignored here is what Fanon calls “the zone of occult instability”,48 a concept which refers to the ambiguities and ambivalences arising from the colonial or postcolonial situation in the individual and collective construction of identity. In the making of the anthology, all that does not fit the chosen model is excluded. For example, love poems not only do not represent the male ideal of the new man, but the feeling of love can block rational behaviour, dulling reason and the motivation for combat. Meanwhile, an important influence on poetry such as pan-africanism is submitted to a narrow reading, identifying it with a negritude close to Senghor and the neo-colonialism the latter represents, without any real comprehension of the complexity it encompasses. Used in everyday life as well as in the press, African languages are left aside in favour of the exclusive usage of Portuguese. Finally, in the organisation of the anthology, the rewriting of the Imperial Library is not understood as a subtle way of negotiating the past but as a form of alienation from revolutionary practices of (national) culture.

Conclusion

In this chapter I have considered the making of the anthology Poesia de Combate and observed a process of national construction which involved writing culture and an idea of citizenship through exclusion and correction. The result of this process was an “invented tradition” grounded in a historical experience, which provided the building blocks for an invented homogenous unity and offered the language and cultural codes which would provide “the cement of society” after independence.49 In this sense, the anthology allowed for the creation of a “primordial tie”,50 both invented and real, which through a sedimentation into the people’s collective imagination could construe a necessary common past.

47 Chatterjee, Partha, Nationalist Thought and the Colonial World. A Derivative Discourse? 48 Fanon, Frantz, Wretched of the Earth, p. 227. 49 Smith, Anthony, Nationalism and Modernism, p. 27. 50 Nash, Manning, The Cauldron of Ethnicity in the Modern World, Chicago and London: Chicago University Press, 1989. writing a nation or writing a culture? 125

This process meant obliterating the creative diversity of local journalistic and literary production during the war. In the anthology, FRELIMO replaced the negotiation of meaning and identity occurring in a heterogeneous time of experience in which pre-colonial, colonial and revolutionary referents are interwoven, with the homogeneous and discontinuous space of the “new” revolutionary culture. Drawing a line between the colonial and traditional past and the revolutionary present, FRELIMO’s categorisations constitute a “derivative discourse” in as far as they involve Manichean colonial representations of culture. Correspondingly, one could argue that its model of national culture reproduced a dichotomy between civic and ethnic nationalism in which the former was valued to the detriment of the latter. Meanwhile, as the poetic production in the local press suggests, FRELIMO militants sometimes created a form of national revolutionary culture which cut across this dualism. Taking up an idea from Walter Mignolo51 and Beck,52 it is possible to argue that the revolutionary representations in these local spaces reflected the necessary historical and cultural focus on situation to which any kind of universalism is subject. From this perspective even the “purest” and most rational civic nationalisms would more or less implicitly be subject to a conditioning by local references and values when considered as existing in the time/space of practice. As the brief reference to the differences between the first and second editions of Poesia de Combate suggests, there is great interest in studying even those kinds of revolutionary nationalism that at first sight seem to be extremely one-dimensional. I called this contribution “writing a nation or writing a culture?” Indeed it seems that the displacement of nationalism towards culture – what FRELIMO called “our poets” or “our culture” – allowed the liberation movement to obliterate the local and traditional dimension of national culture, the ambivalent balance of memory and identity, and avoid confrontation with, and subtle appropriation of, a colonial past. This understanding of culture made a selective use of ideas in circulation within the anti-colonial revolutionary internationalism of the 1960s and 1970s. Reflecting the heterogeneity of these currents, FRELIMO could sometimes combine socialist-realist ideas with avant-gardist or pan-africanist references.

51 Mignolo, Walter, “Géopolitique de la connaissance, colonialité du pouvoir et dif- férence coloniale”, Multitudes Web, 2001, http://multitudes.samizdat.net/Geopolitique-de -la-connaissance (accessed 10/02/2009). 52 Beck, Ulrich, Qu’est-ce que le cosmopolitisme? Paris: Aubier, 2006. 126 maria-benedita basto

FRELIMO’s replacement of the concept of “nation” by that of “revolutionary culture” illustrates not only the importance of taking into account the contribution of internationalism, but also points at the limitations of analyses of African nationalism which concentrate on identifying a problematic tension between an imitation of the European model of the nation brought by colonialism and African modes of life. Opting for a solution that differed from the colonial project of the nation as well as from the idea of Africanization, FRELIMO’s approach represents a particular kind of decolonizing experiment. The dilemmas it faced reflect fundamental contradictions and ambiguities within internationalist currents, which all ended up having difficulties in confronting diversity and a past deemed to be feudal or capitalist, and also remind us of a particular moment of “high modernism” in which strongly centralized States expressed grand narratives of progress based on the contributions of science and technology.53 In such context, “writing (a) culture” – as in the case of FRELIMO’s “1st Seminar on Culture” and the compilation of the anthology Poesia de Combate – can be seen as an effort to construct a time-space, a people and a nation which, as I have argued, fits in with Jacques Rancière’s concept of “utopia”. Utopia here is not a faraway place, but the coincidence of a discourse with a place, a territory, and a reality. In the words of Rancière: “if modern utopia has a meaning, it is certainly not in the myth of an island that is nowhere, but rather in the possibility of identifying, in any place, the correspondence of text and reality”.54 In this sense, the liberated zones and the liberation struggle are not myths created only after independence. They were already a myth in the act itself of their constitution, from the moment when the diversity of their writings was erased in the search of this correspondence.

53 Scott, James C., Seeing like a State: How certain Schemes to Improve the Human Condition Have Failed, New Haven: Yale University Press, 1998. 54 Rancière, Jacques, Courts voyages au pays du peuple, Paris: Seuil, 1990, p. 48. CHAPTER FIVE

‘AN IMAGINARY NATION’ NATIONALISM, IDEOLOGY & THE MOZAMBICAN NATIONAL ELITE

Jason Sumich*

Between 2002 and 2004 I conducted research investigating the formation and social reproduction of the ruling elite in Maputo, the capital of Mozambique.1 One day I was talking with Josina, whose parents became high-ranking members of the Frelimo party after independence in 1975.2 Although her family had risen to prominence in a movement that was once socialist, Josina described herself as a capitalist who backed ‘neoliberal’ reforms. When I expressed doubts that this model would really help the poor, Josina told me that I had seriously misunderstood the nature of Mozambican society. According to her the inequalities that have arisen are not because the poor are denied opportunities, but because they did not want them: There is a huge difference here that I do not think you understand. You spend all of your time with people like us, we are educated and westernised.

* This research was undertaken with the assistance of the Department of Anthropology, London School of Economics, Research Student Grant, The British East Africa Institute Small Grant, and the Radcliffe Brown Award from the Royal Institute of Anthropology. Supplemental Research and the writing of this article were possible due to my postdoctoral position at the Crisis States Research Centre, London School of Economics. I owe special thanks to CEA (Centre of African Studies) at Eduardo Mondlane University in Maputo for providing me an institutional home during my fieldwork. I am indebted to Eric Morier-Genoud, Jo Beall, Ed Simpson, Fraser McNeill and Casey High for their generous comments and I would like to especially thank Sandy Robertson and James Sumich for their much needed editorial help and comments. 1 My data is derived from almost two years of ethnographic research followed by an additional month in 2006. The core of the research was based on participant observation, the collection of life histories and semi-structured interviews of 23 elites and their families and 16 members of the middle class and their families. My information was bolstered by many additional interviews among both elites and the middle class who did not constitute core informants. All names have been changed in this paper. Interviews were conducted in Portuguese and English and all translations are mine. 2 Frelimo (the Portuguese acronym for Mozambican Liberation Front) fought an eleven year liberation struggle and has ruled continuously since independence.

Those of us who are privileged have tastes and desires that are very different from everyone else. It is really a question of interest. Most people in this country are peasants, they have a machamba (a small plot of land) and they are interested and satisfied with that. They do not really need education or anything more, in fact they do not really want it. For instance my father comes from a poor, rural background. He liked to read, but he was not that interested in education.3 He did not become interested until he saw the Portuguese and how much they had in comparison to himself. Most people want to be left alone to farm their machambas. It is those of us who are privileged that want and need all of these things. What interested me about Josina’s response was not only it resemblance to some of the colonial justifications of inequality, but that it was not uncommon amongst those connected to the Frelimo based elite in Maputo. There was often an implicit assumption for members of this group that because they are educated and ‘modern’ they are fundamentally different from the vast majority of the population in important ways and this is their claim to power. Ironically, this sense of difference appeared to have its foundation in a vision of nationalism that once promised a revolutionary form of unity for all of the nation’s inhabitants. In this chapter I trace the trajectory of the nationalism championed by the ruling Frelimo party to understand how an ideology that once promised radical unity also came to be a powerful symbol of social difference. Mozambique has experienced a whirlwind of political and social change since independence in 1975. In 1977, Frelimo declared itself a Marxist- Leninist vanguard party; early efforts towards a Mozambican ‘Perestroika’ were introduced in 1983 during a brutal foreign-imposed civil war; and from 1989 the party began to move to ‘free market’ democracy. However, underlying these dramatic changes has been a persistent, if evolving, sense of nationalism with an ideological bedrock based on a vision of modernity held by the ruling elite.

Elites, Nationalism and Modernity in Mozambique

Many studies of nationalism have pointed out that ruling elites often attempt to create legitimacy for their project by basing it on myths of some sort of primordial connection, the idea that the nation has ‘always’

3 During the colonial period the vast majority of the population was illiterate. The fact that her father could read and had access to books hints that his background was actually rather privileged. nationalism, ideology & the mozambican national elite 129 existed and its inhabitants have self-identified themselves as its members since time immemorial. In sub-Saharan Africa this legitimising project has often been more difficult since the vast majority of the continent’s nations can be so obviously traced back to their not to distant origins, but the straight lines and fertile imaginations of imperial cartographers. This does not mean there have not been attempts, such as TANU/CCM’s ideology of African Socialism based on the return to a supposed moral purity that was corrupted by colonial rule or ZANU PF’s incorporation of the symbolism of the trading empire of Great Zimbabwe into the new national name, flag and currency. Frelimo also adopted elements of this by incor- porating various forms of early resistance against the Portuguese as nationalist figures, even if it is doubtful that those in question ever conceived of the territory as anything resembling a potential nation-state as it is currently understood. Nonetheless, there was a significant difference in Frelimo’s ideology of nationalism. Instead of being based on some sort of primordial connection that bound Mozambique’s diverse peoples together, it was a vision of a nation as of yet uncreated, a nation not built on its glorious past but on a radiant future still to come. I argue that the nationalism held by the Frelimo elite cannot be understood without understanding its modernist orientation; in fact they are part and parcel of the same thing. The modernist orientation of the Frelimo elite is not so much temporal but, to paraphrase James Ferguson and Frederick Cooper, is a ‘local’ category that has its origins in their experience of colonialism.4 In this case, it operates as an ideology to understand and redress the injustices of colonial rule and to explain their role within it. It was a way to create a new nation, different than what came before that was as strong as the Europeans, but also unique and responsive to its inhabitants. The discourse of this ideology filtered throughout nationalist rhetoric and programmes. It promised a sense of unity and potential based on something to come, something that would be inherently ‘modern’. However, as the promised brave new world never seemed to arrive, except for a chosen few, it has increasingly become a symbol of the powerful, a sign of their right to status and privilege. Frelimo was able to come to power in a classically populist fashion,5 as a primarily urban-based intellectual leadership formed an alliance

4 F. Cooper, Colonialism in Question: Theory, Knowledge and History, Berkeley and Los Angeles: University of California Press, 2005; and J. Ferguson, Global Shadows: Essays on Africa in the Neoliberal World, Durham: Duke University Press, 2006. 5 A. Robertson, People and the State: an Anthropology of Planned Development, Cambridge: Cambridge University Press, 1984. 130 jason sumich with the northern peasantry and tapped into their desire to liberate the country and improve the miserable conditions of poverty, illiteracy and exploitation they suffered. However, independence was to be only the first step in a far more ambitious social revolution that increasingly drew its rhetoric from the Marxist-Leninist model. In a manner similar to what Donham described for the Derg in Ethiopia, this undertaking drew its inspiration not from an eventual communist utopia but from the concrete examples of the Soviet Union and the People’s Republic of China.6 Frelimo also hoped to take a ‘backward’, humiliated, agrarian population and reshape them into an industrial nation that could no longer be exploited by foreign powers. The party leadership envisioned the complete transformation of the population, with peasants becoming workers, scattered machambas giving way to agro-industries, factories churning out goods and growing prosperity. On a social level the transformation was to be no less thorough. The brutalised, fractured and localised population would be transformed into a subject that was universally modern with science and rationality replacing ‘feudalism’, ‘superstition’ and exploitation, and also simultaneously nationalist, loyal to Mozambique and transcending ties of village, region, ethnicity and race. For the most devoted members of Frelimo’s Marxist intellectuals even this stage would eventually be surpassed as a communist society would arise from the ashes of colonialism sometime in the future. Modernism, for the Frelimo elite, was a fundamental part of the nationalist project as it aimed not to give voice to what always was, but instead to create the foundations for what will be. In a recent book ‘Global Shadows: Africa in the Age of Neo-Liberalism’, Ferguson makes an important contribution to the vast literature on modernity when he describes it as a local category for his Zambian informants.7 Modernity in this case is a way of describing the global pecking order and it is the ‘lack’ of it that is the cause of the gross material inequalities that characterise their lives.8 Ferguson argues that when people in Africa speak of their ‘backwardness’ they are not referring to a sense of supposed cultural inferiority, but instead the socio-economic conditions the vast majority are forced to suf- fer.9 There is much to commend this argument and Frelimo was speaking

6 D. Donham, Marxist Modern: An Ethnographic History of the Ethiopian Revolution, Berkeley, Los Angeles and Oxford: University of California Press and James Currey, 1999. 7 J. Ferguson, Global Shadows. 8 ibid, p. 186. 9 ibid, p. 33. nationalism, ideology & the mozambican national elite 131 very clearly about the exploitative practices that brought so much poverty and suffering when it made the case for independence. However, the party’s ideology was also deeply concerned about what it felt was Mozambique’s cultural inferiority.10 It was this inferiority and weakness that allowed them to be conquered by one of the weakest European powers in the first place (or as a common saying in Maputo would have it: ‘Our tragedy is that we were colonised by the Africa of Europe’). The construction of the new nation also depended on the construction of a new mentality. In many ways Frelimo’s projects resemble the description made famous by James C. Scott, with its combination of modernisation from an authoritarian government and an ill-disguised contempt for ‘local practices’.11 This did not necessarily make the project illegitimate though and the party tapped into a widespread desire to create a new and different world throughout the country. The initial project of modernist nationalism, in its revolutionary guise, failed in part as noted by Scott, because it was coercive and contemptuous of local practices, thus engendering opposition. Just as importantly though, it failed because, through a variety of causes such as terrible planning mistakes, the naivety of the leadership, foreign aggression and a brutal civil war from 1977 to 1992, the promised modernity never appeared.12 In fact, for the majority, life became even more arduous. Revolutionary modernist nationalism collapsed with the end of the civil war and has been theoretically replaced by a new liberal ideology. It is undeniably true that tactics have changed and revolutionary zeal has often been replaced by cynicism and corruption on the part of elites and disillusionment and alienation among the wider population. However, even the new liberal state still claims legitimacy through its history as the liberators and creators of the nation and those who can bring about ‘development’. I argue that Frelimo’s nationalist ideology has not been discredited so much as reconceptualised by the national elite. What was

10 M. Hall and T. Young, Confronting Leviathan: Mozambique since Independence, London: Hurst & Co, 1997. 11 J. C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed, New Haven and London: Yale University Press, 1998. 12 H. West, ‘Creative Destruction and Sorcery of Construction: Power, Hope and Suspicion in Post-War Mozambique’, Cahiers d’Etudes Africaines, vol.147, 1997, pp. 675– 698; and H. West, ‘Sorcery of Construction and Socialist Modernization: Ways of Understanding Power in Postcolonial Mozambique’, American Ethnologist, vol. 28, 2001, pp. 119–150. 132 jason sumich once an authoritarian, but potentially emancipatory, project based on the mass mobilisation of the populace has increasing become an indicator of status and power for the elite. The ideology of modernist nationalism however, serves as a ‘unifying field’ that creates unity and an internal legitimacy among elites.13 They rule and occupy positions of status and power because they created the nation of Mozambique and they are the only ones with the ability to bring a better future. Furthermore it is a claim to power to an increasingly alienated population as they can argue with some success they are the only ones with sufficient skills to fulfil the nationalist promise and bring a better, more prosperous future. Even as this future’s arrival date seems to recede ever further. However, the failure of socialism and, more recently, capitalism to bring the brave new world any closer have created doubts among the national elite who are becoming increasingly uncertain about how to realise this project. Is local culture a powerful unifying field that creates national identity; is it still Frelimo’s role to educate the populace towards a new mentality, or is it some combination of the two? Debates rage within the party as to the proper way forward, but in practice the modernity at the heart of the nationalist message serves as a symbol of everyday claims to social power in a nation they are still trying to build. This symbolism may be disputed, but it is generally understood by the wider population, at least in Maputo, the area I know best. I begin by explaining the social origins and transformation of the elite and the origins of the development of modernist nationalism as a local category and how it has transformed through Mozambique’s changing political landscape. I will then discuss how this ideology is inculcated into younger generations of elites through education and expressed by self- presentation and consumption. I will end the paper with a brief discussion concerning what the Mozambican case has to offer to the wider anthropological arguments concerning modernity and nationalism in Africa.

The Origins and Transformation of Revolutionary Nationalism

In Mozambique there are multiple groups that can claim the title of elite, from régulos (‘traditional’ authorities), to religious leaders, foreigners

13 J. Gledhill, ‘The Power Behind the Masks: Mexico’s Political Class and Social Elites at the End of the Millennium’, in C. Shore, and S. Nugent (eds), Elite Cultures: Anthropological Perspectives, London and New York: Routledge, 2002, pp. 39–60. nationalism, ideology & the mozambican national elite 133 attached to multinationals or powerful international agencies, the top members of the Indian merchant class, various local level Frelimo authorities and high-ranking members of Renamo (Mozambican National Resistance), the former rebels and now the official opposition party. This paper focuses on a specific elite group that appears to be socially dominant, if not uncontested. It is based on members of the ruling Frelimo party in Maputo, their families and close business associates. I do not mean to claim that this group is completely homogenous; multiple factions and social cleavages exist. These divisions encompass the various social backgrounds of members of the elite, including race, religion, role (or lack of one) in the liberation struggle and level of education. There are fierce disagreements within the party concerning the future of the nation. However, the central goal of somehow building a modern nation still remains, and it creates an overall unity for the elite while attempting to legitimise their positions of power and material interests. To understand why this is so, we have to examine the social origins of the dominant section of the Maputo elite from 1930 until independence. Much of the leadership comes from the south among those who used to be members of a former official colonial category of assimilados (the assimilated), a group that proved to be extremely influential, despite the smallness of their numbers. One of the more common estimates puts them at around 5000, out of an African population of around 8,200,000 in 1960.14 To be an assimilado one had to fulfil certain legal criteria: to swear loyalty to the colonial state, speak only Portuguese at home, adopt ‘European’ habits, abandon ‘heathen’ beliefs and have a Portuguese official vouch for one’s character. If one fulfilled these criteria one was theoretically granted full legal equality with the Portuguese.15 While equality was not the case in practice, assimilados were granted a range of privileges, such as freedom from forced labour, better access to education, employment and urban residence, and jurisdiction under civil law.16

14 K. Sheldon, Pounders of Grain: A History of Women, Work and Politics in Mozambique, Portsmouth NH: Heinemann, 2002. 15 E. Mondlane, The Struggle for Mozambique, London: Penguin Books, 1969. 16 B. O’Laughlin, ‘Class and the Customary: The Ambiguous Legacy of Indigenato in Mozambique’, African Affairs, vol. 99, 2000, pp. 5–42; J. Penvenne, ‘The Unmaking of an African Petite Bourgeoisie: Lourenço Marques, Mozambique’, Working Papers in African Studies, no. 57, African Studies Center, Boston University, 1982; and J. Penvenne, ‘“We are all Portuguese!” Challenging the Political Economy of Assimilation: Lourenço Marques, 1870–1933’, in L. Vail (ed), The Creation of Tribalism in Southern Africa, Berkeley and Los Angeles: University of California Press, 1989, pp. 255–288. 134 jason sumich

Assimilados could gain preferential employment in the highest bastions of the colonial economy that a black person could realistically aspire to, such as nursing, teaching, railway administration and low level positions in the civil service. Due to the way the southern region was incorporated into the Portuguese empire, and to the proximity of the capital and the mines and markets of South Africa, a particular kind of assimilado elite developed in Maputo: One that had weak ties to ‘traditional’ authority, but greater opportunities to engage in commercial farming, labour migration and the urban-based government sector.17 However, their relative privileges could not mask the fact that they would remain second class citizens at best and were subjected to systematic discrimination and humiliation.18 In many ways they were aspirants to a vision of modernity that under colonialism would always remain just out of reach.19 The most alienated of this group later became well represented in the leadership of the liberation struggle. The coalition that formed the leadership of the Frelimo party at independence was based on southern urban assimilados, more rural, mission- educated aspiring elites from the north and various other disaffected urbanites from Portuguese, Indian and Mulatto (mixed race) backgrounds. Assimilação (assimilation), along with most other colonial practices, was anathema to all that the new society was supposed to represent. The word assimilado began to take on connotations of selling out or collaboration. Once when I was arranging an interview with a man who had been an assimilado during the colonial period I was warned by his niece not to use that word or ‘he will throw you out of his house’. Nevertheless, as noted by Fry, socialism in Mozambique had many surprising similarities with assimilação: In spite of the anti-colonial discourse of the center and Frelimo in general, it is impossible not to observe that the socialist project in Mozambique was if anything more ‘assimilationist’ than the Portuguese ever dared to imagine and it is tempting to suggest that this is one of the reasons why the Mozambican elite found the socialist program so attractive. Structurally speaking there was little difference between an authoritarian capitalist state run by a small body of ‘illuminated’ Portuguese and assimilados and an

17 A. Pitcher, Transforming Mozambique: The Politics of Privatization, 1975–2000, New York: Cambridge University Press, 2002. 18 J. Penvenne, ‘“We are all Portuguese!”. 19 M. Hall and T. Young, Confronting Leviathan: Mozambique since Independence. nationalism, ideology & the mozambican national elite 135

authoritarian socialist state run by an equally diminutive and equally enlightened vanguard party.20 Frelimo’s attempt was far more ambitious than the Portuguese ever dared dream. Instead of being restricted to a few chosen sons of colonialism, it should expand to include the entire nation. The party’s attempts at social engineering drew from the leadership’s own background. The goals of the nationalist project were deeply intertwined with elite modernism: the nation was the instrument that could bring the ideology to fruition. As the following example will show, the appeal of this project for many of Frelimo’s cadres was based on the fact that it appeared to be the best way to alleviate the misery suffered by the majority of the population, while having an added benefit of being socially familiar. Sergio Abdul comes from an Indian/mixed race family and grew up on a plantation in the south during the colonial period with his step- father who had been active in an early nationalist movement. Sergio described his upbringing as ‘urban in a rural area’, the house was always filled with literature and they kept in touch with the intellectual currents of the capital. Sergio managed to attend secondary school and then moved to the provincial capital where he worked as mechanic until liberation. He had been a Frelimo sympathiser and joined the party after independence because he was deeply attracted to the goal of the socialisation of the countryside. He grew up among peasants, knew first hand the misery of the rural areas and strongly supported a movement that was dedicated to alleviating their hardships. He later left the party when a combination of mistakes, changing policies and the civil war meant this goal became impossible. Sergio continues to support Frelimo and although he is critical of the current policies, he feels the party is the only one who can eventually bring a better life to the people. The nationalist vision of the party is based on people like Sergio, those whose background influenced a very particular, future oriented view of the nation that they would create. The party’s goals, however, went much further. In the leadership’s view, building a better world meant that the nation’s citizens had to be completely transformed. This goal though, was deeply influenced by the social background of the leadership. For the leadership, ‘traditional’ culture was not a base of ‘local’ knowledge, but the cause of the nation’s ‘backwardness’, which resulted in defeat and humiliation during the colonial period.

20 P. Fry, ‘Cultures of Difference: The Aftermath of Portuguese and British Colonial Policies in Southern Africa’, Social Anthropology, vol. 8, 2000, pp. 117–143, p. 129. 136 jason sumich

As observed by Hall and Young: ‘The Frelimo elite and the social strata to which it appealed were profoundly convinced of the superiority of modern civilisation and the need to ‘catch up’ with it. The only way to resolve these dilemmas was to see the “people” as empty vessels, but possessing the potential to develop’.21 Obviously the ‘people’ were not in fact empty vessels, but for Frelimo the creation of its state depended on destroying traditional structures and the party was willing to use state power to do so. This is not to say that every decision was based simply on abstract ideological goals. Policies were put into place for a variety of reasons, including security concerns, responses to adverse conditions, or simply the continuation of former Portuguese projects. However, even pragmatic choices made in response to a changing social environment were seen through the lens of the modernist ideology central to the nationalist project, and were attempts to buttress these overall goals. Shortly after independence the institution of régulos (‘chiefly’ leadership) was abolished, lobola (bridewealth) was outlawed and polygamous men were denied entry to the party. Major religious institutions, were under outright attack by the party from 1977 to 1979 and certain ancestral and rain-making ceremonies were banned. Practitioners of ‘sorcery’ could be sent to re-education camps and attempts were made to continue the Portuguese counterinsurgency strategy of moving the peasantry from their former scattered hamlets to centralised communal villages, which would become ‘cities in the bush’. Intense effort was focused on combat- ing what was termed superstition or ‘obscurantism’ and replacing them with rationality and scientific socialism.22 This was combined with almost messianic appeals from the first president, Samora Machel that through the determination of the party, ‘underdevelopment’ would be overcome in ten years. Many members of the Frelimo elite were privately skeptical, especially as this plan depended on a bankrupt country achieving an annual economic growth rate of 17.5%. Much of the rank and file though was convinced and enthusiastic. As I was told by a lecturer at the national university: ‘It was amazing, many of the party members actually believed and would tell you with complete sincerity that everything will change and the country will be utterly unrecognisable in ten years’. One of the foundations of Frelimo’s programme to create this modern nation during socialism was the creation of the Homem Novo (new man).

21 M. Hall and T. Young, Confronting Leviathan, p. 65. 22 ibid, p. 86. nationalism, ideology & the mozambican national elite 137

According to one of Frelimo’s leading theorists, Sergio Vieira, the new man would provide the decisive break from the past.23 The new man would be based on science, ‘rationality’ and collective labour, but it was still in the process of being born.24 Assimilados were modelled on the Portuguese national identity, but the new man would be both a universal subject and the embodiment of an emerging Mozambican personality and citizen. Whereas the enthusiasm among the population for a complete re-creation of the Mozambican personality probably varied, its allure to Frelimo militants was quite genuine. I was told by one former party member: ‘To be called a new man by Samora Machel was truly thrill- ing, we were going to build a new nation, part of a new world, everything was urgent … we lived in a permanent state of exultation’. The new man was to be the concrete expression of the great nationalist project, and a new type of person for a new nation was Frelimo’s ideology made flesh. While the new man was to embody Frelimo’s modernist nationalism, the party hoped that in the meantime the experience of colonial oppression could create the basis for a shared sense of citizenship.25 However, just as the nationalism of Frelimo flowed from the social experiences of the leadership, reactions to it were influenced by a variety of local conditions. When Frelimo took power it had a tremendous degree of popular support and it counted on its widespread acclaim to mobilise the population for the heroic effort of building the nation. While much of the population was thrilled to achieve independence and a desire for change was widespread, their relationship to the evolution of Frelimo’s modernising nationalism was often more ambivalent. Frelimo had a strong social base that believed in its goals and this legitimised the party’s power; the young residents of the south and far north, and ironically the better off who tended to benefit the most.26 The rural population of the centre and parts of the north of the country though, largely became alienated from the government due to the party’s programme.27 Just two years after independence Mozambique was caught up in another war. Frelimo

23 S. Vieira, The New Man is a Process. A Speech by Sergio Vieira, Member of the Central Committee of Frelimo, to the Second Conference of the Ministry of Education and Culture, held in December 1977, p. 3. 24 ibid, p. 25. 25 E. Mondlane, The Struggle for Mozambique; and A. Pitcher, Transforming Mozambique. 26 M. Hall and T. Young, Confronting Leviathan. 27 M. Cahen, ‘Check on Socialism in Mozambique – What Check? What Socialism?’, Review of African Political Economy, vol. 57, 1993, pp. 46–59; and C. Geffray, A Causa das Armas, Oporto: Edições Afrontamento, 1991. 138 jason sumich supported the liberation movement in neighbouring Rhodesia and its white minority regime retaliated by forming a rebel army, Renamo, composed of Mozambican dissidents. After the Rhodesian regime fell and the newly named Zimbabwe became independent, South Africa became Renamo’s sponsor and formented the civil war that devastated Mozambique. Many of the Frelimo government’s real gains in education and health care were destroyed. Despite their brutality, Renamo thrived where the population had been deeply alienated by Frelimo’s social revolution. In parts of the centre and the north Renamo was initially greeted as liberators and it could count on the population’s overall indifference towards which side won the war.28 Frelimo may have tapped into a desire for thorough change when it came to power, but the actual programme was of course far more difficult, especially considering the weakness of the party’s control over its desperately poor nation. As the next example will demonstrate, many of the party leadership’s assumptions were based on wishful thinking. Paulo was born in Maputo (then Lourenço Marques) in the 1920s. He came from a Portuguese working class family. His father was strongly opposed to the dictatorship of António Salazar and Paulo became radicalised early in his life. As he told me: ‘As a child I would see people send their servants to the police for some minor thing and the servant would have to hand a note to the police which would tell the police how many times to flog the poor person. It was not hard to see that the system was wrong’. In the 1940s he became a member of the communist party and later joined Frelimo during the liberation struggle. In his opinion Frelimo’s grand dream of creating a completely new and modern nation went wrong because Mozambique did not have the sufficient conditions to realise it: Frelimo was not really a communist movement; it was a nationalist one, although progressive. The goal was not simply to free the nation, but also to transform it. For this the party wanted to create a modern socialist state but we simply did not have the conditions. We did not have the social preparation to make a revolution. At independence most people were peasants, even those who lived in the urban areas were psychologically peasants. The cities did not fulfil their historic role by making them workers, most people were servants and now they wanted to be masters. We tried to turn peasants into workers, but we failed. We lived in an imaginary nation, the dream of the nation we wanted to create, not the flesh and blood nation we actually had.

28 C. Geffray, A Causa das Armas, and M. Hall and T. Young, Confronting Leviathan. nationalism, ideology & the mozambican national elite 139

The amount of social preparation the population needed to achieve Frelimo’s goals is debatable. Nevertheless, the party was surprisingly ambitious considering that they came to power with a handful of guerrilla soldiers in a country that was largely illiterate and surrounded by powerful, hostile white minority regimes. The revolutionary programme collapsed; however some form of modernisation is still central to elite efforts to legitimise themselves as the only ones who can continue to build the nation. Even Paulo, an orthodox Marxist, still supports Frelimo as a progressive force, the embodiment of Mozambican nationalism.

Liberal Modernist Nationalism and the Persistence of the Past

Today the discourse of modernist development remains ubiquitous in nationalist rhetoric. If socialism failed, perhaps liberalism will succeed. Thus, in Maputo, one is constantly confronted with the government’s exhortations to join the struggle against absolute poverty and even musi- cians when interviewed on national television are routinely asked how their music will contribute to the development of the nation. However the failures of modernist nationalism have caused a sense of disquiet among some members of the elite. As Varyna, a young woman from an elite family explained, in her opinion, many of the central problems facing Mozambique do not arise from overriding political or economic factors, but within the population itself. The problem is we blacks are still two hundred years behind. You can see it everywhere. When we get power it just becomes about lining your own pockets. Everything is about who you are and who you know. When something goes wrong we just blame the outside. It’s the west or spirits, its always something. If something goes wrong they go and find a witchdoctor to find out what it is. Its not just disease, I mean they know something about herb- alism and that might work. They do it for stupid things, like relationship trouble. People think magic can stop someone from cheating on them. You would be surprised about the people I am talking about; I mean people who have been to university, they are educated and should know better. Her comments recall the former idea of the ‘new man’, where social problems stem from the fact that the nation’s inhabitants are ‘superstitious’ instead of rational. In fact, those who bear the brunt of her attack are the elite, the people who should be the closest to the modern ideal as they are educated and should know better. Other members of the leadership feel that the continuing failure of the modernist ideal to transform the country is caused by political drift. As I was told by a former minister: ‘Yes, 140 jason sumich there is the struggle against absolute poverty, but that is a slogan not a plan. How do we actually make this happen?’ Another, more junior, Frelimo party member declared: ‘What we need is a revolution inside of Frelimo. We just keep recycling old ideas and everything seems to just get worse.’ While the self-confidence of this elite may have been damaged by their continuing failure to completely transform the nation, their legitimacy is based on the fact that they can bring some kind of development to the nation, even under liberalism. Plans are no longer as ambitious as in the revolutionary period, but some form of development still equals modernisation in the minds of the elite. The transformation of Mozambique from a poor agrarian backwater into something else has become the justification of the nationalist project. What the end result of the modernisation process is though, outside bolstering the power and status of elites, is becoming increasingly uncertain. The following section will examine how the local category of elite modernism is becoming ever more exclusive. What once signified the goal of populist nationalism has now become a symbol of one’s fitness to hold positions of power. In the following section I examine how this ideology is inculcated through educational practices and how claims to social status are displayed by forms of consumption, especially among a new generation of elites who were born after independence and for whom socialism is only a childhood memory. I argue that the changing forms of the nationalist goals have been instrumental, not only in re-enforcing overall elite unity, but also by allowing a group whose power stemmed from their membership in the party, and by extension, the control of the state, to expand into new opportunities.

Education and Distinction

Education in Mozambique is held in high regard, perhaps, in part, due to the colonial state’s dismal record for its provision. Portugal largely out- sourced education provision for the African population to the Catholic Church.29 At independence in 1975 ‘one percent of the population – about

29 T. Cruz e Silva, ‘Identity and Political Consciousness in Southern Mozambique, 1930– 1974; Two Presbyterian Biographies Contextualised’, Journal of Southern African Studies, vol. 24:1, 1998, pp. 223–236; T. Cruz e Silva, Protestant Churches and the Formation of Political Consciousness in Southern Mozambique (1930–1974), Basel: P. Schlettwein Publishing, 2001; M. Buendía Gómez, Educação Moçambicana: História de um Processo: 1962–1984, Maputo: Livraria Universitária/Universidade Eduardo Mondlane, 1999; and A. Pitcher, Transforming Mozambique. nationalism, ideology & the mozambican national elite 141

80,000 people – had completed more than four years of school, and most were Portuguese settlers; in 1973 only forty of the 3000 university students were African’.30 Estimates put the illiteracy rate at independence at over 90%.31 Access to a decent education became a mark of distinction as it was so rare and served as a symbolic entrance to a wider idea of modernity. This is not surprising as access to education was generally restricted to those from privileged families and it was this that allowed these families to bolster their position. Education allowed one to obtain jobs that through their relative importance and symbolic accompaniments (such as a suit and tie, residence in a better provisioned assimilado area of town or, for the highest of all, a car) displayed one’s lofty place in the social hierarchy for all to see. Furthermore it was education that allowed the chosen few to advance (as far as was possible) in the predominantly urban sector of the colonial economy. One former assimilado told me that when he went to elite Portuguese schools during the colonial period, some of his class- mates no longer saw him as Black. Education gave one status and authority both among the general population, and as much as was possible, among the Portuguese. When Frelimo took power it was determined to greatly overhaul the educational system of its new nation. This was both because the vast majority of the Portuguese population had fled following independence, virtually decapitating Mozambique’s professional and managerial class, and also because education would be one of the foundations of a new sense of national identity. Here the ideals of Frelimo could be spread to sections of the population which had little to no experience of the liberation struggle, and a new type of citizen, based on science and ‘rationality’ could be formed.32 Furthermore, education would also respond to the demands from below and bring increased opportunities to the populace. Efforts were made to expand the educational system throughout the country, but a lack of resources and trained personnel made it difficult to reach the rural areas, outside of periodic campaigns.33 It was in the cities that the government’s attempts were much more focused, especially

30 W. Minter, Apartheid’s Contras: An Inquiry into the Roots of War in Angola and Mozambique, Johannesburg: University of Witswatersrand Press, 1996, p. 22. 31 J. Hanlon, Mozambique: The Revolution Under Fire, London: Zed Books, 1990; and B. Munslow, Mozambique: The Revolution and its Origins, London, New York and Lagos: Longman, 1983. 32 M. Buendia Gómez, Educação Moçambicana. 33 M. Hall and T. Young, Confronting Leviathan. 142 jason sumich

Maputo. Naema, a secondary school student immediately after liberation and now a teacher at a private school, compared the revolutionary system to that of today in the following way: It was all very exciting. We did not know much about Frelimo before liberation. I had seen the word Frelimo written on wall, but my uncle told me never to speak about it or we could get into trouble. When Frelimo came and got rid of the Portuguese we were all rejoicing, especially the young. We were going to build a new Mozambique. They (Frelimo) would come to our schools and teach us about our country. We would do voluntary cleaning in the city on Sundays; they wanted us to be good citizens. That was really good. Now the city is filthy and no one does anything. Many of my students do not even know the words to the national anthem. Patriotism does not mean anything anymore and they do not try and teach it like before. Naema greatly appreciated the attempts towards nation building and, as a teenager herself, she was caught up in the euphoria of the times. Others, though, cast doubts as to the quality of the greatly expanded education system, even in Maputo where facilities were generally superior. Another former primary school student, now a resident in Portugal described her schooling shortly after independence as follows: It was a strange time. Former guerrilla soldiers would just come to school and start classes. Many of them did not have any education, but they had just freed the nation and they thought that was enough. They would sit in front of the class and talk about their lives. They thought that was teaching. Even the other teachers had almost no experience. Many of them were very young and they had just a few years of school themselves. My teachers and I learned to read together. Members of the Frelimo elite also shared this rather bleak assessment. A former high-ranking officer told me: Those were very exciting days. We were sure we could do a better job than the Portuguese had. They were foreigners and they wanted to exploit our country for their own purposes. We were going to make things better for our people. We were very ambitious and very young; we had immense responsibilities for people our age. That was part of the problem I guess. We wanted to educate the nation, but we were barely educated ourselves. Frelimo’s social revolution initially intended to modernise the entire nation and education was a cornerstone of this policy. However this tre- mendously ambitious plan had to be undertaken with minimal resources and little in the way of personnel. What began as a great attempt for the nation soon began to be scaled back to urbanites generally and to the elite in particular. nationalism, ideology & the mozambican national elite 143

The growing social stratification was initially offset by an upsurge in social mobility, especially in Maputo, that followed liberation. The exodus of the Portuguese meant that almost all of the professional and managerial positions in the country were vacant and Mozambicans were promoted to the colonialists’ former places.34 With the opening of the education system for both children and adults, it did seem as if the opportunity to gain distinction and social power through schooling had been expanded to at least some of the wider urban population.35 Yet, due to the economic crisis and the growing civil war, an economy of privation soon developed and the role of education as an entrance to the ‘modern’ economy, distinction and a position of responsibility in the new nation began to falter. This was brought home to me during a dinner I attended with a relatively well-off family (by Mozambican standards). After the meal the host- ess pulled out a photo album filled with pictures of establishing their family shortly after the revolution. The children were amused by the large Afro hairstyles and the outmoded fashions of their parents, but they were also shocked by how thin everyone was and the shoddy quality of their clothes and furniture. The host tried to explain that there was nothing else available during what she described as the tempo da fome (the time of hunger). I heard similar comments in debates concerning the legacy of the socialist period throughout my fieldwork. Some claimed that things had been better during o tempo de Samora (the presidency of Samora Machel) as at least there was order, purpose and little crime, while cynics responded that was because there was nothing to steal. Others claimed negative solidarity, stating that everyone had been poor together, instead of the current situation where the rich supposedly monopolise everything. However even the most dedicated defenders of the revolutionary period conceded those despite the solidarity and euphoria, times were very hard and most were hungry. Although Frelimo had managed to broaden the system, the distinction and the status that modernity symbolised remained elusive, despite the promise, even for the privileged urbanites of the capital. With the transition to capitalism, status differences are hardening and once again changing forms of education provision are signalling wider

34 J. Sumich and J. Honwana, ‘Strong Party, Weak State?: Frelimo and State Survival through the Mozambican Civil War’, Crisis States Research Centre Working Papers, no. 23 (series 2), London School of Economics, 2007. 35 B. O’Laughlin, ‘Contingent Identity and Socialist Democracy in the Port of Maputo’, Unpublished Paper, The Hague, 2002. 144 jason sumich changes in the ideology of modernist nationalism. A short stroll through contemporary Maputo is usually enough to reveal the differences in the educational facilities patronised by elites and those used by the rest of the city in the current era. There are many state schools, but they are frequently in poor repair, with peeling paint and broken windows. Wave after wave of students issue forth from them throughout the day as overcrowding necessitates that schools work in multiple shifts. The difference between public institutions and private and international schools is striking. For the latter, the paint is fresh, the windows are intact and students are dropped off by an array of expensive cars and SUVs. The facilities are far better, the teachers are among the best trained, and tuition (in 2002–2004) ranged from $100 a month, which can be difficult even for a middle class family in Maputo, to $1,000 a month, far beyond the means of most. While many of the children of the elite study abroad, with South Africa, Swaziland, Brazil, Portugal and the UK being the most common destinations, they complete their initial schooling through private and international schools. The majority of the younger generation of the elite bypass state schools entirely and class differences within Maputo are indicated by those who have to rely on the public system and those who do not. Education was meant to transform the populace, responding both to the party leadership’s plans to create a new kind of citizen because of demands from below for a better life and to increased opportunities. How ever, like many aspects of modernist nationalism, high quality education has become a mark of privilege, largely monopolised by a small elite. As education is one of the central planks of the elite ideology of modernity. Privileged access to it implies both symbolic and real power. The elite obtain their power through their connection to the state and the Frelimo party. In this respect they have developed some of the characteristics of what has been called a state bourgeoisie.36 However the fall of socialism has opened new opportunities for members of this elite and they are increasingly able to move beyond their Maputo base and expand into international networks. The privatisation of state assets has allowed Frelimo families to own businesses and acquire land and houses, and the influx of multinational businesses and aid agencies attached to the

36 C. Leys, ‘The Kenyan Bureaucracy’, in C. Allen and G. Williams (eds), Sociology of “Developing Societies”: Sub-Saharan Africa, New York and London: Monthly Review Press, 1982, pp. 176–178; M. Cohen, ‘Public Policy and Class Formation’, in C. Allen and G. Williams (eds), Sociology of “Developing Societies”: Sub-Saharan Africa, New York and London: Monthly Review Press, 1982, pp. 179–182. nationalism, ideology & the mozambican national elite 145 international community have also opened new, well-remunerated opportunities for current and former officials and their families. Members of the elite often spend their professional lives circulating between the government, the international community and private businesses and thus control access to the major economic sectors within the country. The superior quality of education gives the elite the qualifications to take advantage of these opportunities in a way very difficult for anyone else, while at the same time portraying the system as based on merit. They can argue that their high positions are justified because they are the only ones with the skills and experience to run a modern nation. This creates a ‘unifying field’ which holds together different factions of the elite and forms the core of their group identity.37 Education and status tend to reinforce each other with education, giving the outward qualification and membership in the elite, which allows access to extremely powerful social networks. This position of dominance is expressed by members of the elite through patterns of consumption and self-presentation. This became clear to me once during a conversation with a woman whose parents are high-ranking Frelimo members. She felt that Renamo could not run the country as they were, in her view, unedu- cated peasants and she illustrated this point to me with a story. I remember in 1992 when peace was declared and Renamo came out of the bush. They were given houses, at least the big guys in the party were. It was one of the conditions of peace. When they (Renamo) came here they had no idea how to live in a city. They used to wash their clothes and leave them to dry on the front lawn! Can you believe that? These people think they could run a country. It’s a joke; they had never been out of the bush before. I asked her if it had not been similar when Frelimo first emerged from ‘the bush’ after the liberation struggle. She replied: ‘Of course not, Frelimo fought in the bush but they knew how to live in a city, they were not igno- rant’. Thus self-presentation through the consumption of high status goods, such as cars, western clothes and expensive, high quality education are crucial factors to express modernity and social power.38 It is a claim to superiority over much of the population and to equality with

37 J. Gledhill, ‘The Power Behind the Masks: Mexico’s Political Class and Social Elites at the End of the Millennium’. 38 P. Bourdieu, Distinction: A Social Critique of the Judgement of Taste, Cambridge MA: Harvard University Press, 1984; and G. vom Bruck, ‘The Imagined “Consumer Democracy and Elite (Re)production in Yemen’, Journal of the Royal Anthropological Institute, vol. 11, 2005, pp. 255–275. 146 jason sumich global elites and the western world. This claim is recognised in Maputo, even if the justice of it is disputed. Once in a café I was approached by a man who felt as a foreigner I was duty bound to help him fund his revolt against what he felt was the current unjust state of affairs. When I asked him what the outcome of his revolution would be, he smiled and replied: ‘I will be the one driving the Mercedes’. Although he claimed he wanted to wage a revolution he was displaying a logic similar to that of the elites. He made no arguments concerning redistribution, but instead argued that the markers of social power (a Mercedes, the car issued to government ministers) are being held by the wrong people. His revolution would make sure the proper people, in this case himself, possessed these objects.

Conclusion: Modernity, Nationalism and Power

In this paper I have argued that an ideology of modernity held by a politically based dominant elite in Mozambique is crucial both to structuring power relationships and to the form of nationalism Frelimo has championed. Although it has appeared in different guises since independence it continues to take symbolic form as a claim to legitimacy, and gives shape to a sense of coherent elite unity insofar as it enables elites to monopolise opportunities both within the state and in the expanding market economy, and also in more intangible ways. Through their educational and cultural opportunities and the fact that they are fluent in Portuguese and often English as well, they can more easily ‘fit in’ with the foreigners who run the multinational businesses and the agencies of the international community, while being an embodiment of the nation at the same time. By understanding this ideology and its historical development this paper hopes to demonstrate the connections between ideas of a very particular kind of modernisation and nationalism in Mozambique. Nationalism, as has often been argued, is always a modernist project. This was demonstrated in Benedict Anderson’s famous argument that nationalism is an ‘imagined community’ brought into being primarily through print capitalism and widespread education. While the ‘modernity’ of nationalism is not in any doubt, nationalists often try to clothe their goal in the garments of an imagined tradition of primordial and organic unity that has ‘always’ bound the nation’s people together. In late colonial Mozambique, where neither print capitalism nor education were particularly widespread, nationalism took a slightly different trajectory. Frelimo did not attempt to disguise its project with the trappings of a nationalism, ideology & the mozambican national elite 147 mythic past, but explicitly stated that it would build a nation where none had existed before and unity would be based on shared progress, not a primordial connection. For Frelimo, nationalism was part and parcel of its vision of modernity, however including Mozambique’s diverse peoples within this vision is a continuing struggle. In Mozambique, modernist nationalism operates as a local category growing from the social experiences of many members who went on to form the nationalist leadership and it has survived the fall of socialism and provides a sense of mission for elites and a justification of their status and power. Despite the advent of capitalism, many elites throughout Africa and the wider world are expected to be ‘developmental’ and they attempt to derive legitimacy as those who can provide (or can claim to) some form of betterment for their impoverished people, if for nothing else, than as a way to access funds provided by donors. The premise of modernisation has been attacked from both the right and the left, but it remains a foundation of nation-building, even under capitalism. The real danger, as in the case of Frelimo, is what happens when nationalism is not based on a supposed primordial past but a glorious future that for many, fails to appear. I have argued that the idea of modernity is becoming a more distinctive feature of the Frelimo elite, a marker of power, as opposed to a premise for mass mobilisation. However, the steady erosion of older certainties has been accompanied by schisms and growing demoralisation for the members of the Maputo elite I know. They still maintain a strong hold on power, both because power, or access to it, is often the foundation for everything else they have managed to build, and because many sincerely believe there is no one else who could run the country. There is however the growing danger that Mozambique is starting to resemble Arthur Koestler’s famous aphorism concerning Stalinism: “Revolution betrayed, tradition decayed and utopia, yet once again delayed”.

CHAPTER SIX

UNITA AND THE MORAL ECONOMY OF EXCLUSION IN ANGOLA, 1966–1977

Didier Péclard

After fourteen years of independence struggle, Angola entered a phase of formal decolonisation in early 1975. Despite profound divergences between the three nationalist movements fighting to take over power from the colonial administration, the Popular Movement for the Lib eration of Angola (MPLA), the National Front for the Liberation of Angola (FNLA), and the National Union for the Total Independence of Angola (UNITA) signed in January the Alvor agreement in Portugal. Amongst other things, the Agreement saw the formation of a transitional govern ment made out of ministers from the three competing movements as well as the election of a Constituent Assembly in charge of elaborating the contours of the future Angolan State and of nominating its President. Independence itself was scheduled for November 11 of the same year. The Alvor agreement was a very complex mechanism. Designed in a very tense political context, where none of the three movements seemed to genuinely envisage any power sharing, it did not make illusion for long, and a few months after it has been signed, the question was no longer whether or not violence was going to break out again, but when. At the time, UNITA was by far the smallest of the three competing nationalist movements. Founded in 1966 in Muangai, to the East of the country in an area close to the Zambian border, it still only counted a few hundred guerrilla fighters1 eight years later when it signed a cease-fire

1 They were between 600 and 800 according to the US Consul then in place in Luanda. See Piero Gleijeses, Conflicting Missions. Havana, Washington, and Africa, 1959–1976, Chapel Hill (NC) & London: The University of North Carolina Press, 2002, p. 241. In 1973, according to the Portuguese political police (PIDE/DGS), UNITA had 300 guerrilla fighters, against 3000 for the MPLA and 2000 for the FNLA (see Dalila Cabrita Mateus, A PIDE/DGS na Guerra Colonial (1961–1974), Lisbon: Terramar, 2004, p. 236). Information given by UNITA deserters or prisoners between 1971 and 1974 when interrogated by PIDE/DGS tend to con firm this. See: Instituto do Arquivo Nacional/Torre do Tombo (Portuguese National Arquive, hereafter: IAN/TT), Serviços de Centralização e Coordenação das Informações sobre Angola (SCCIA), Caixa n°266, Proc. 34.69.77 (Organização militar da UNITA).

© koninklijke brill nv, leiden, �� | doi ��.��/��_�� 150 didier péclard with the Portuguese army shortly after the ‘Carnation revolution’ that put an end to Salazarist rule in Lisbon on 24 April 1974. The few military bases that it managed to create and maintain during those years were all in the marginal and sparsely populated Moxico district, and despite various attempts, it did not manage to extend its guerrilla activities to the West of the Kwanza river, in the highly strategic Bié and Huambo districts.2 Despite initial support from China, where some of its leaders were trained in the late 1960s, and a small grant from the World Council of Churches (given in the early 1970s to nationalist movements throughout Southern Africa as part of WCC’s Programme to Combat Racism), it did not have strong international support either. Besides, in 1974–75, UNITA still did not have a large social base, and was, comparatively to the MPLA and the FNLA, still relatively unknown inside Angola. Yet, despite being severely defeated on the battlefield in the first post- independence phase of the civil war, by the end of the 1970s UNITA had grown to become a full-blown guerrilla movement with support from within the country as well as from strong allies in the Western block of the Cold War. In 1976, when the MPLA, supported by Cuban troops, took over the Central Highlands3 and the city of Humabo, where UNITA and the FNLA had declared the birth of a short-lived ‘Democratic Republic of Angola’ as an alternative to the MPLA’s Popular Republic, UNITA was force to retreat into its “Long Walk”. This would eventually bring it to the Terras do Fim do Mundo, the far south-eastern corner of the country close to the Namibian (then still South-West African) border, where it later established its “capital” Jamba. But UNITA’s loss of Huambo and the way in which the MPLA took over power in the area actually turned out as what could be called a “successful defeat”. Indeed, it was, as I will argue here, instrumental for the movement’s legitimacy in the eyes of those who felt excluded by the new rulers of Angola and who therefore came to see UNITA as their champion. To explain this turn of event, two main arguments have been advanced. The first puts a rather obvious stress on the international dimension.

2 IAN/TT, SCCIA, Caixa n°266, Proc. 34.69.77 (Organização militar da UNITA), fls. 119–129: report from the PIDE/DGS delegation in Silva Oporto (Bié), 04.03.1972. 3 The Central Highlands, or planalto central, is a plateau in the centre of the country, covering roughly the present-day provinces of Huambo and Bié, as well as part of the prov inces of Benguela, Kwanza-Sul and Huíla. It is the historical home of the Ovimbundu, the largest ethnic groups which makes up to approximately 35% of the population. Today, large numbers of Ovimbundu live in cities outside of the planalto, mainly in the capital Luanda. unita & moral economy of exclusion in angola, 1966–1977 151

Caught up by cold-war dynamics, the conflict that opposed the three Angolan nationalist movements turned after independence into an inter nationalised civil war.4 In this context, the support that UNITA received from South Africa and the USA was of course crucial to the development of its small guerrilla movement into a full-fledged “rebel” army, the FALA (Forças Armadas de Libertação de Angola). But, whatever the crucial importance of foreign involvement in the Angolan civil war, especially in terms of money, weapons and direct military support, it is clear that the roots of the war between the Soviet backed MPLA-dominated govern ment and UNITA’s “freedom fighters” were first and foremost internal.5 This is where the second main argument comes in. According to it, if UNITA was able to grow fast in numbers and to broaden its social base in those crucial years, moving from a “party of cadres”6 to a socially represen tative guerrilla movement, it is mainly because it was the “natural” – and sole – champion of the Ovimbundu, the main ethnic group in the Angolan Central Highlands. In this chapter I address specifically this second aspect of the “de- marginalisation” of UNITA shortly before and after independence. There is little doubt that ethnicity played an important role in the Angolan civil war, and that it proved to be a very potent mobilising tool for UNITA. What I would like to argue here, however, is that, whatever the relative weight of ethnicity in the history of the movement, there was nothing “natural” or self-evident in the adhesion to UNITA of (sections of) the ovimbundu populations of the Central Highlands – and especially of its elites – shortly before and after independence. The main question that needs to be addressed here is whether or not, until the end of the liberation struggle in 1974 at least, the historical dynamics of the political

4 On the international dimension of the civil war in Angola, see P. Gleijeses, Conflicting Missions, op.cit.; George Wright, The Destruction of a Nation. United States’ Policy Toward Angola since 1945, London: Pluto Press, 1997; Fernando Andresen Guimarães, The Origins of the Angolan Civil War. Foreign Intervention and Domestic Political Conflict, Basingstoke: Macmillan, 2001; William Minter, Apartheid’s Contras. An Inquiry into the Roots of War in Angola and Mozambique, Johannesburg: Witwatersrand University Press/London, Zed Books, 1994. 5 To date, the most comprehensive and subtle analysis of the internal roots of the Angolan civil war is without doubt the two-volume collection of essays by Christine Messiant, published two years after her death: Christine Messiant, L’Angola postcolonial. 1. Guerre et paix sans démocratisation; 2. Sociologie politique d’une oléocratie, Paris: Karthala, 2008. See also Patrick Chabal et al., A History of Postcolonial Lusophone Africa, London: Hurst & Co., 2002 (especially David Birmingham’s chapter on Angola). 6 Christine Messiant, “Angola, les voies de l’ethnisation et de la décomposition, I”, Lusotopie. Enjeux Contemporains dans les Espaces Lusophones, 1–2, 1994, p. 168. 152 didier péclard societies living on the planalto central offered the right context for the development of political contestation in the name of the suffering and interests of the Ovimbundu. Whether, in other words, the historical stage was set for the emergence of an ovimbundu ethnic conscience in the name of which the nationalist struggle could have been led. I argue that the response to this question is negative, and that it is the civil war con text of 1975–1977 that gave ethnicity a new political meaning which UNITA managed to instrumentalise at its own advantage and thereby turn its military defeat of 1976 into a political success. More importantly, I argue that ethnicity was just one of several sym bolic repertoires that UNITA drew upon in order to attract followers. What sustained the growth of the Galo negro7 after its defeat and near extinction in 1976 is a broad narrative of exclusion according to which UNITA stood for those who had not only been unjustly excluded from power at independence but also pushed (again) to the margins of society in much the same way as had been the case during colonial times. This narrative was imbued with immediate political meaning when UNITA and its cadres were driven out of their central highlands stronghold to the Terras do fim do Mundo in the first months of 1976, since UNITA’s military defeat was propagated by Savimbi and his followers as the “second colo nisation” of Angola by the “portugalised” elites of the capital and their Cuban allies. But the narrative was also deeply rooted in the social history of the planalto, and it rested upon the idea that the “true nature” of Angola was to be found in its African, that is black, and rural heritage and not in the mestiço urban communities out of which, according to UNITA rheto ric, the MPLA was born. Developed within Christian (mainly Protestant) mission networks and taken over by the first generation of Angolan pas tors, teachers, qualified workers and clerks this particular vision of what I call the “culturalist modernisation” of Angolan society constituted the symbolic repertoire upon which UNITA later built its legitimacy. The chapter starts with a brief critical overview of the historiography of Angolan nationalism, showing how dominant interpretations of the role of ethnicity and religion in explaining the divisions within Angolan nationalism have provided a reductionist picture. It then goes on to anal yse the social and political meaning of ethnicity in the late colonial phase before offering some thoughts on the moral economy of exclusion upon which UNITA built its growth in the late 1970s.

7 The Black Cockerel (Galo Negro) is one of UNITA’s symbols, also regularly used as synonym for the party or for Jonas Savimbi. unita & moral economy of exclusion in angola, 1966–1977 153

Ethnicity, Religion and Divided Nationalism

One of the characteristic features of Angolan nationalism is its division in various movements. Indeed, there never was in Angola a broad national ist front that would assemble the different movements willing to fight for independence, as was the case in Mozambique or Guinea-Bissau. From its beginnings in the 1950s, then, Angolan nationalism was divided – between the FNLA and the MPLA in a first phase, and later in three competing movements after the creation of UNITA in 1966 as the result of a split within the FNLA.8 In the analysis of these divisions, ethnicity has held centre stage. John Marcum9 and René Pélissier10 for instance, authors of the two classical references on Angolan nationalism, argued that the emergence of the three movements could only be understood by refer ence to their bakongo (FNLA), mbundu (MPLA) and ovimbundu (UNITA) roots. And even if Pélissier distinguishes between what he calls “modern ist nationalism” of which the MPLA would be the expression, and bakongo and ovimbundu “ethno-nationalism”, the ethnic variable remains key in explaining the divisions between the three movements, while other elements such as the social history of colonialism or regional and socio- economic differences between various social groups are ignored. Recently, this “ethnicist” vision has been however criticised, most notably by Christine Messiant who showed how the history of the three movements was embedded in the late colonial history and the dynamics of elite formation in the country,11 or by Patrick Chabal who on his part moved from a strict ethnic categorisation to a more structural one between

8 The FNLA was created in 1957 as Union of the Populations of Northern Angola (UPNA), before changing its name one year later to Union of the Populations of Angola (UPA). After merging with another bakongo party it became the FNLA in 1962. The MPLA on its part was founded in 1960, although according to the official version its creation goes back to December 1956 already, which would make it the eldest of the Angolan nationalist movements. On the controversies about the birth of the MPLA, see Carlos Pacheco, MPLA. Um nascimento polémico (as falsificações da história), Lisbon: Vega, 1997 and on the politi cal (ab)use of the movement’s history, see Christine Messiant, “‘Chez nous, même le passé est imprévisible’. L’expérience d’une recherche sur le nationalisme angolais, et particu lièrement le MPLA: sources, critique, besoins actuels de la recherche”, Lusotopie, 1998, pp. 157–197. 9 John Marcum, The Angolan Revolution I: the Anatomy of a Revolt (1950–1962), Cambridge (Mass.): MIT Press, 1969 and The Angolan Revolution II: Exile Politics and Guerilla Warfare (1962–1976), Cambridge (Mass.): MIT Press, 1978. 10 René Pélissier, La colonie du Minotaure. Nationalismes et révoltes en Angola (1926– 1961), Orgeval: Pélissier, 1978. 11 Christine Messiant, 1961. L’Angola Colonial, Histoire et Société. Les Prémisses du Mouvement Nationaliste, Basel: P. Schlettwein Publishing, 2006 [1983]. 154 didier péclard

“modernisers” (MPLA), “traditionalists” (UNITA) and “ethno-nationalists” (FNLA).12 However, the ethnic reading has had a strong and lasting influ ence on popular and academic discourse, not just on the history of nation alism, but also of the post-independence civil war. The second explaining variable, after ethnicity, that one finds in the literature is religion. But here also, research has proceeded by a double simplification. Firstly, a clear distinction was established in the literature between the Catholic Church on the one hand and Protestant missions on the other as to their respective attitudes towards Portuguese colonialism as well as regarding their role in the construction of the colonial state and its eventual demise with the nationalist struggle. According to this vision, the Catholic Church would have walked hand in hand with Portuguese colonialism, especially during the time of the Salazar dictatorship (1930– 1974), and it would have been a docile auxiliary of colonialism. Protestant missions, on the contrary, which were barely tolerated – but also regularly combated – by the colonial regime, would have been the cradle of nation alist thinking.13 Secondly, as a corollary to the ethnicist vision about the division of Angolan nationalism, a direct link was established between the influence of the three main Protestant mission societies in Angola and the building of the ethnic consciousness which framed the expression of nationalist thinking and conditioned its institutionalisation in political movements. At the basis of this view is the fact that, as in many other cases throughout Africa, the three main Protestant missions in Angola divided up the work between them according to regional and cultural criteria. Thus, British Baptists were in charge evangelising the Northern part of the country, Luanda and its hinterland up to Malanje was left to

12 Patrick Chabal, “Lusophone Africa in Historical and Comparative Perspective”, in P. Chabal et al., A History, op. cit., pp. 5–16. 13 In the past few years, this simplistic version has been criticised and nuanced. On Angola, see Benedict Schubert, Der Krieg und die Kirchen. Angola, 1961–1991, Luzern: Edition Exodus, 1997; Emmanuelle Besson, “Autour du procès de Joaquim Pinto de Andrade. L’Eglise catholique et l’Angola colonial, 1960–1975”, Le Fait Missionnaire. Histoire et Héritages, n°12, July 2002. Didier Péclard, Etat colonial, missions chrétiennes et nationalisme en Angola, 1920–1975. Aux racines sociales de l’UNITA, Paris: Institut d’études poli tiques, PhD Thesis, 2005. On Mozambique, see Eric Morier-Genoud, “Y a-t-il une spécificité protestante au Mozambique ?”, Lusotopie, 1998, pp. 407–420 ; Religious Orders and the Making of Catholic Politics. The case of the diocese of Beira, 1940–1974, Binghamton: State University of New York, PhD Thesis, 2005. Zélia Pereira, “Les Jésuites et la formation d’élites au Mozambique, 1961–1974”, LFM. Missions et sciences sociales, n°14, July 2004, pp.75–116. For a general perspective on the social and political positionning of Protestant missions in Portuguese colonies, see Christine Messiant, “Protestantismes en situation coloniale. Quelles marges ?”, Lusotopie, 1998, pp. 245–256. unita & moral economy of exclusion in angola, 1966–1977 155

US Methodists, while the central highlands came under the responsibility of two North American (US and Canadian) Congregationalist mission societies. This is what led authors to establish a direct correlation between Protestant missions, the invention of ethnicity and the eventual emer gence of ethno-nationalist movements. In short, the following triple equa tion has been advanced as an explanatory grid:

Baptist Mission => Bakongo => FNLA Methodist Mission => Mbundu => MPLA Congregationalist Mission => Ovimbundu => UNITA.14

The mains problem with this vision is the sense of historical inevitability that it conveys and its strong teleological undertones. The history of nationalism in Angola and its divisions, far from any ethnical, cultural or even regional determinism, is first and foremost a social history, as Christine Messiant showed in her study of the “premises of the nationalist movement”.15 Indeed, on the eve of the uprisings of 1961 in the North of Angola and in Luanda which marked the start of the nationalist struggle, the three main Angolan “ethnic blocks” were far from being socially, cul turally and historically homogeneous. This means that, to the exception maybe of the Bakongo area,16 the ethnic repertoire alone was not enough to form the cultural vehicle of political grievances. In other words, there was no bakongo, mbundu or ovimbundu response to Portuguese colo nialism, but a great variety of social, economic, religious and political tra jectories within the three main geographical and cultural spheres and between them. When applied to the history of nationalism in the central highlands, the ethnic-cum religious paradigm leads moreover to a paradox. On the one hand, UNITA is presented as the political expression of ovim bundu nationalism, as the “natural” ideological and political extension of

14 This is by and large the interpretation put forward by Marcum (The Angolan Revolution, I, op. cit.), Pélissier (La colonie du Minotaure, op. cit.), Lawrence Henderson (The Church in Angola. A River of Many Currents, Cleveland (OH): The Pilgrim Press, 1992) or Linda Heywood (Contested Power in Angola, 1840s to the Present, Rochester: University of Rochester Press, 2000). 15 C. Messiant, 1961. L’Angola colonial, op.cit. 16 According to Messiant, “bakongo society was […] at the same time strongly frag mented on the basis of traditional cleavages and strongly united against the exterior due to its particular features and to a feeling of belonging to the same ethnic group”. (1961. L’Angola colonial, op.cit., p. 362. My translation). For the history of the FNLA, see in par ticular J. Marcum, The Angolan Revolution, I & II, op. cit. 156 didier péclard a culturally and ethnically homogeneous unit. On the other hand, although Ovimbundu were, with up to a third of the Angolan population, by far the biggest ethnic group and although tens of thousands of young ovimbundu men had a very violent experience of colonialism as forced labourers sent to the coffee plantations of the North of the country, the central highlands remained somewhat cut off from the nationalist strug gle, and few Ovimbundu joined in. This is what led Pélissier to argue that, when the nationalist struggle started in 1961, the central highlands were “a great blank spot […] on the Angolan geopolitical map because ethno- nationalism there was structured much later than elsewhere in the coun try”, and that this “delay” was “surely one of the major weaknesses of the 1961 revolts”.17 In a similar vein Linda Heywood also admits that “up to the mid-sixties […] the majority of Ovimbundu […] did not challenge the [colonial] regime”, but that after its creation in 1966, “UNITA became a forum for rural, regionalist and ethnic sentiments”.18 According to Heywood, Protestant missions were instrumental in forging this unity and this ethnic pride. The Protestant communities of the planalto, because of their own cohesion and as a result of the hostility of the Portuguese authorities towards them, served as havens where a common “ovimbundu experience” of Portuguese colonialism and political opposition to the colonial state developed.19 Indeed, UNITA did draw on Protestant networks to build up support. In January 1975, Savimbi made an important visit to Dondi, the symbolic heart of the Congregationalist movement in the central highlands,20 where he had studied in the 1940s. Presenting himself as the prodigal son returning home, he insisted, in front of a “cheering audience”, that he “still believed in the one God who sees everything that we do”,21 adding that he had come to account for the work he had done during his years in exile.22

17 R. Pélissier, La colonie du Minotaure, op. cit., p. 291. My translation. 18 Linda Heywood, Contested Power in Angola, op. cit., pp. 151–152. 19 Ibid., pp. 104–117, passim. See also, by the same author: “Unita and Ethnic Nationalism in Angola”, The Journal of Modern African Studies, 27 (1), 1989, pp. 47–66. 20 On the role of Dondi in the history of the central highlands, see D. Péclard, Etat colonial, op. cit., chapter 3. 21 Quoted in Lawrence Henderson, The Church in Angola. A River of Many Currents, Cleveland (OH): Pilgrim Press, 1992, p. 346. 22 Jonas Savimbi left Angola in the late 1950s to study in Lisbon, from where he had to flee in 1959. He eventually settled in Switzerland where he became very active within Angolan student circles. According to an account he published in a book written in hon our of one of his former professors at the University of Lausanne, Henri Rieben, the foundations of UNITA were laid in a chalet in Champex, in the Swiss Alps (Jonas Savimbi, “L’Angola de demain”, in Hommage à un Européen offert à Henri Rieben à l’occasion de son unita & moral economy of exclusion in angola, 1966–1977 157

Indeed, it was “as a son rather than as a politician” that he had returned to this “place where he had played football”, where he “had had moments of joy when he was awarded distinctions at school”: I still consider myself as a child willing to accept his parents’ and elders’ advices. Each of my former professors, each of my former school mates can come and sit at my side, not in order to talk to the President of UNITA – indeed I do not like to be called that way –, but simply in order to sit down with Jonas. [Because] I am [still] the same Jonas.23 After this visit, a certain collaboration was put in place between sectors of the church and UNITA. Many Congregationalist pastors took position in favour of Savimbi, and part of the church leadership started to envisage a state where the church could not only work without the type of adminis trative and political hurdles that the Portuguese administration had deployed against it, but also be fully and officially acknowledged and sup ported for its social work, especially in the health and education sectors.24 In other words, the Congregationalist church started to dream of a UNITA- dominated state where it would have the privileges of a state church. In February 1977, one year after the takeover of the planalto by the MPLA and Cuban troops, a part of the Congregationalist leadership even left Dondi in order to join UNITA in the bush,25 the church being from then on separated between an urban church based at Huambo and a rural one which eventually settled with UNITA when the movement’s capital was founded at Jamba.26 At first sight then, the history of UNITA’s development between 1974 and 1977 looks very much like “the chronicle of an alliance foretold”: the

70e anniversaire, Lausanne: Ecole des HEC & Fondation Jean Monnet pour L’Europe, 1991, p. 127). Henri Rieben, a man with many connections in the Swiss business world, was one of Savimbi’s staunches supporters throughout the 1980s. He also defended UNITA’s posi tion after the elections in 1992. Rieben is even said to have had the University of Lausanne award his former student a Doctor honoris causa, but this has not been confirmed by the alma mater (see François Ruchti, “Seigneur de guerre lausannois”, L’Auditoire, 176, February 2007 – www.auditoire.ch/spip.php?article516, accessed on 6 May 2010). 23 Alocução do Presidente da UNITA, Dr. Jonas Savimbi, a 20 mil pessoas na Missão do Dondi. Bela Vista, 28 de Janeiro de 1975. A copy of this document was given to me by the then representative of UNITA in Lisbon, Carlos Morgagdo. On Savimbi’s visit to Dondi and his comeback to the planalto, see Marco Vinicios & Maria João Saldanha, Jonas Savimbi. Um desafio à ditadura comunista em Angola, Lisbon: Armasilde, 1977, pp. 98–110. 24 B. Schubert, Der Krieg, op, cit., pp. 201–206. 25 The missionary station of Dondi, one of the biggest in the whole country, which had become the headquarters of the Angolan Congregational church in the 1950s was taken over by government troops after the flee of the church’s leader and it was turned into a military garrison. 26 For a UNITA insider’s account on the flight of Congregationalist leaders out of Dondi, see Samuel Chiwale, Cruzei-me com a história, Lisbon: Sextante Editora, 2008, chapter XVI. 158 didier péclard gradual move towards the defence of the values and interests of ovim bundu ethnicity; arrival of numerous Ovimbundu new recruits as the movement strive to extend from a “party of cadres” to a “mass organisa tion”; collaboration with the Congregationalist church which contributes to broadening the movement’s social basis. All seems to indicate that UNITA was indeed the “natural” expression of a “Holy Alliance” which emerged during the colonial period between Protestant missions, ovim bundu societies from the planalto and ovimbundu nationalism. Yet, if one looks beneath the surface, things were not as linear and pre-determined as they seem to be.

Late Colonialism and (Forced) Social Diversification

Between the military conquest of the planalto central in the first two decades of the 20th century and the launch of the nationalist struggle in 1961, ovimbundu societies of the planalto went through a series of pro found changes. From the late 18th century up until the 1910s, they were organised and structured around long-distance trade. Out of the dozen of ovimbundu independent ‘kingdoms’27 and subsidiaries, two, Bailundu and Bié, took up a particular importance thanks to their commercial rela tions with the Portuguese. After serving as intermediaries in the slave trade,28 they dominated commerce in beeswax and ivory until natural rubber became the main commodity exported from the Angolan coast during the last quarter of the 19th century. Between 1874 and approxi mately 1912 up to 50’000 people participated every year in rubber collec tion campaigns that could last for up to several months. During this rubber boom, which was remembered half a century later as the period when the Ovimbundu “had more clothes”29 than they ever had later, the whole social system evolved around long-distance trade and the social institution of caravans.

27 On the history of the ovimbundu kingdoms, see Gladwyn Murray Childs, Umbundu Kinship and Character, London: Oxford University Press, 1949, Chapter XII. L. Heywood, Contested Power, op. cit., Chapter 1. Maria da Conceição Neto, “Entre a tradição e a mod ernidade: os Ovimbundu do planalto central à luz da história”, Ngola, Revista de Estudos Sociais, Luanda I (1), 1997, pp. 193–215. 28 Joseph Miller, Way of Death. Merchant Capitalism and the Angolan Slave Trade, 1730– 1830, London: James Currey, 1988. 29 According to an old man who had participated in the rubber trade, interviewed in the 1950s by Adrian Edwards, a Catholic priest who wrote one of the two standard ethno graphic studies of the Ovimbundu (The Ovimbundu under Two Sovereignties, London: Oxford University Press, 1962). unita & moral economy of exclusion in angola, 1966–1977 159

The involvement of Bailundo and Bié kings in the rubber trade had, however, contrasted consequences. On the one hand, the profit made in the barter of rubber against cloth and weapons were key in establishing the two kingdoms’ authority over other ovimbundu communities. On the other hand, the rubber boom was also a time when new economic entre preneurs emerged and established themselves beyond the direct control of ovimbundu rulers. By trading their newly acquired wealth against social and political capital they contributed to undermining the power of the olosoma.30 The strategy of extraversion31 which brought Ovimbundu kingdoms an unprecedented level of prosperity and political independence through out most of the 19th century rested on a strategic alliance with Portuguese traders and authorities. As the rubber trade was booming however, the weight of the alliance started to shift. With the Berlin conference and especially the British ultimatum of 1890, the Portuguese crown was forced to extend its presence on the ground beyond the trading posts and military garrison it had had in the central highlands since the late 17th century to a policy of actual control over the territories it claimed to be hers. In this new context, several elements combined to put an end to the indepen dence of the ovimbundu kingdoms. The military conquest, although very long and costly for the Portuguese authorities, was achieved when the kingdom of Bié was defeated in 1890 and that of Bailundo around 1905.32 The rubber trade started to decline roughly at the same time, when cheaper rubber was discovered and commercialised in Asia and in Brazil. Finally, with the downfall of the Portuguese Crown and the advent of the First Republic in 1910, the first premises of a colonial state apparatus were to be developed based on the actual occupation of the territory. The colonial conquest and the end of the rubber trade marked the advent of a new era for ovimbundu societies. The most dramatic change was the passage from long-distance trading to cash-crop agriculture, with the development of maize culture on the planalto.33 In the first two

30 L. Heywood, Contested Power, op. cit., pp. 16–23. 31 Jean-François Bayart, “Africa in the World: A History of Extraversion”, African Affairs, n°99, 2000, pp. 217–267. 32 Douglas Wheeler & C. Diane Christensen, “To Rise with One Mind: The Bailundo War of 1902” in Franz-Wilhelm Heimer (ed.), Social Change in Angola, Munich: Weltforum Verlag, 1973, pp. 53–92. 33 Hermann Pössinger, “Interrelations Between Economic and Social Change in Rural Africa: The Case of the Ovimbundu of Angola”, in F-W. Heimer (ed.), Social change, op. cit., pp.31–52. M. da C. Neto, “Entre a tradição”, op. cit. 160 didier péclard decades after the military conquest, the rapid intensification of cash-crop agriculture opened up opportunities for some ovimbundu entrepreneurs. Due to the still rather scant presence of Portuguese settlers on the high lands until the 1940s, cash-crop agriculture became a viable alternative to long-distance trading. Besides, many ovimbundu social entrepreneurs successfully invested the new institutions, such as Christian missions, that the advent of the colonial order had brought in its wake much in the same way as they had invested in their dependent relationship with Portuguese traders. However, the overall picture of the history of planalto in the 20th century is that of a slow but steady disintegration and atomisa tion of ovimbundu societies. The first casualties of the colonial conquest were ovimbundu tradi tional power structures. Partly weakened, as we saw, by the rubber trade, they hardly survived their military defeat, and their legitimacy was severely damaged when many were integrated in the colonial apparatus, especially as local suppliers of manpower for the forced labour system that was so crucial to the development of the colonial economy. As Gladwin Murray Childs, Congregationalist missionary and ethnographer, put it, by the middle of the 20th century, ovimbundu traditional authori ties were “hardly more than living museum pieces”.34 Contrarily to what happened in the North of Angola, where the memory of the ‘glorious days’ of the great Kongo empire served as a potent symbolic repertoire for modern nationalists in the 1950s,35 there was no social institution within ovimbundu societies whose memory could be appropriated and instru mentalised in order to support a nationalist project. This is partly due to the history of the ovimbundu kingdoms which was marked by fierce eco nomic competition since their insertion in long-distance trading, and which prevented the emergence of a strong political centre, or even of a pan-ovimbundu feeling of belonging. Of the three main ethnic groups in Angola, the Ovimbundu were the ones who were most affected by the Portuguese forced labour system. Every year, thousands of young ovimbundu men were taken away from their community and sent for long periods of work in the coffee planta tions of the North, the great sisal farms off the coast from Benguela or the fisheries around the coastal town of Namibe (Moçâmedes), to the South – Bailundo becoming thereby a derogatory word for a crude manual worker without much education. The forced labour system was by far not a factor

34 G.M. Childs, Umbundu Kinship, op. cit., p. 20. 35 See for instance J. Marcum, The Angolan Revolution, I, op. cit. unita & moral economy of exclusion in angola, 1966–1977 161 of integration to the colonial economy of course, but a factor of exclusion while at the same time accelerating the dissolution of traditional social networks.36 Although the central highlands were considered by the Portuguese authorities as their main wheat granary, there was no experience of mas sive land expropriation in favour of white settlers as had been the case in the coffee growing areas in the North. Whereas this common experience of land dispossession played an important role in the rising opposition to colonialism in the bakongo area after 1945, the slow but steady impover ishment of ovimbundu farmers had opposite consequences. Indeed, cash crop agriculture had been the principal means of adaptation of ovim bundu societies after the colonial conquest, thanks notably to the efforts provided by Christian missions in terms of rural development.37 In the mid 1950s, despite the growing presence of white settlers and commercial farmers, it was still cash-crop agriculture which “incorporated villagers into the commercial economy of modern Angola”.38 In this context, indi vidual strategies in order to “get by” and find one’s way into or along the colonial world made much more sense than common claims against the colonial oppressor, especially as social institutions such as Christian mis sions did provide for some possibilities of upward mobility. A further important element of change in the mid-twentieth century was urban growth and development. As of 1940, the city of Nova Lisboa, by then the second Angolan city in size although it had been founded in 1912 only, went through a spell of rapid economic development and was a strong magnet for rural populations of the highlands, especially for the new elites educated within Protestant and Catholic missions. With the unprecedented economic boom that characterised Angola between the late 1950s and the early 1970s, the planalto was not the only region where such rapid changes were under way. In the North, coffee growing was also in full swing. The exponential development of the coffee industry however served in a first instance the interests of the big white fazendeiros and was to the detriment of the majority of the population, except for a small group of kulak entrepreneurs.39 To the contrary, in

36 C. Messiant, 1961. L’Angola colonial, op. cit., pp. 212–250. Jeremy Ball, ‘The Colossal Lie’: The Sociedade Agrícola do Cassequel and Portuguese Colonial Labour Policy, 1899–1977, Los Angeles (CA): University of California, PhD Thesis, 2003. 37 This is an argument I develop in D. Péclard, Etat colonial, op. cit., Chapter 3. 38 A. C. Edwards, The Ovimbundu, op. cit., p. 74. 39 David Birmingham, Frontline Nationalism in Angola & Mozambique, London: James Currey, 1992, p. 46. 162 didier péclard the central highlands, those among the African population who could hope to profit from the economic boom were the small elite of qualified workers, artisans, technicians and clerks who had been educated in secondary schools, very often within Protestant or Catholic missions, and for whom the gradual, albeit very restricted, opening of the colonial world that marked the 1950s and early 1960s was a new window of opportunity. In that sense, Christian missions, whether Protestant of Catholic, did not necessarily contribute to nationalism. Being a member of a particular religious or denominational community did not have any direct bearing on one’s position in the political sphere, contrarily to well established cli chés. Christian missions were the only institution where the young African elites, those whom Christine Messiant aptly described as “new assimilados”, 40 could seize the opportunity to enter into the colonial world and make their way up the social ladder. These dynamics of social mobil ity were of course different on both sides of the confessional barrier, but fundamentally the processes at work were similar. Within Catholic mis sions, young boys and girls would generally move from their village of ori gin to a mission station and, at one point or another, to an urban centre where most of the secondary education institutions (whether academic, technical or professional) were located. They thus contributed to a cer tain integration of Angolan elites into the colonial world – to the very limited extent that this happened of course. For Protestant new assimilados the situation was more ambiguous. On the one hand, Protestant mis sions promoted a very strong sense of belonging to a special culture, made out of a complex mix of Protestant values and a reinvention of what Protestant (foreign) missionaries and local church dignitaries defined as ‘true’ African ideals, a culture that was against any attempt at integrating the colonial world whom most saw as inherently corrupt and not repre sentative of African-ness, and that preached a spirit of self-sacrifice for the interests of Protestant rural communities. On the other hand, it was clear to young Protestant new assimilados that education and training within Protestant missions was a possible support for social upward mobility, and that social upward mobility, especially from the 1950s onwards, meant attempting to get a position (as a clerk in the local admin istration or in a small commercial venture, for instance) in an urban context, at the immediate margin or slightly within the colonial world.

40 C. Messiant, 1961. L’Angola colonial, op. cit., pp. 345–349, passim. unita & moral economy of exclusion in angola, 1966–1977 163

In this context, opting for engaging with a nationalist movement poten tially had a high social and economic cost in the particular context of eco nomic boom in the central highlands. A last distinctive feature of Angolan societies on the planalto at the eve of the nationalist upheaval is the fact that there was no black or mestiço urban bourgeoisie, contrarily to Luanda. In the capital indeed, the 20th century marked the slow but steady drop in status of the creole elite of mixed Angolan, Portuguese, Brazilian and even Dutch extraction that had ruled the city – and the colony by and large – until the 1920s approxi mately. This relegation of creole bourgeoisie to the margins of a colonial society of whom they had been the masters played a crucial role in the advent of anti-colonial politics in and around Luanda.41 In the central highlands, the advent of the colonial state in the first half of the 20th cen tury went hand in hand with growing racial exclusion and segregation, especially in the city of Nova Lisboa. The crucial point here is that social and racial barriers were such that local elites had no way to meet and share common or similar experiences. This again stands in stark contrast to Luanda, where one of the great strengths of the MPLA was the success ful alliance between old assimilated elite from the capital and new assimilados of the hinterland, many of whom Protestant. Such social mixing or strategic alliance was impossible in the central highlands of the 1950s and early 1960s. Put together, these various historical, social and cultural factors may explain why, in terms of participation to the nationalist uprising, the planalto was this “blank spot” that so puzzled Pélissier. When the anti- colonial war began, the African societies of the highlands were “dispersed by migrant labour, atomised, with no possibility to flee or fall back on an autonomous social organisation, without any real force of opposition in other words”.42 Umbundu elites, most of whom were new assimilados of rural origin and socialisation, but who started to find their way into the urban colonial world, had neither the same reason as degraded old assimilados families of Luanda or dispossessed Bakongo who had not profited from the coffee boom to radically question a colonial system within which, despite all its inequalities and injustices, they had managed to find

41 C. Messiant, 1961. L’Angola colonial, op. cit. Marcelo Bittencourt, ‘Estamos juntos’ O MPLA e a Luta Anticolonial (1961–1974), Rio de Janeiro: Universidade Federal Fluminense, 2002, PhD Thesis. Jean-Michel Mabeko-Tali, Dissidências e poder de Estado. O MPLA perante si próprio (1962–1977). Ensaio de História política, 2 Vols., Luanda: Nzila (Coleção Ensaio 3), 2001. 42 C. Messiant, 1961. L’Angola colonial, op. cit., p. 373. 164 didier péclard a somewhat ‘privileged’ place as compared to the mass of indígenas. Nor did their societies, more importantly maybe, offer the sort of symbolic repertoires in the name of which the struggle could be led and a national ist project articulated (as was the case with the Kongo empire in the North), or, because of the extent of social atomisation and disintegration, even the social networks and structures, a minimal sense of common des tiny necessary for federating social and political grievances in a single movement. The fourteen years of nationalist struggle did not fundamen tally change this. Throughout the anti-colonial war (1961–1974), the central highlands remained sealed off from most of the actual fighting. This was partly the result of the absence of instiutionalised nationalism in the region and of the structural weaknesses of the nationalist movements and their armies. But it was also the result of a military strategy of containment on the part of the Portuguese authorities who did not want the most populated and at the time economically prosperous part of the country to risk falling into the hands of the ‘rebels’. The Portuguese counterinsurgency policy had two elements. Firstly, Salazar’s security and intelligence apparatus functioned at full swing throughout the conflict. In the first months after the launch of the war in February-March 1961 especially the infamous PIDE/DGS, the political police that Salazar brought to Angola in the late 1950s,43 launched an offensive against African elites of the planalto. Many were arrested and sent to one of Angola’s prisoners’ camps. This resulted in further damaging the coherence and leadership of ovimbundu societ ies, especially as this first act of repression was followed by other such waves of arrests of Angolan elites, especially Protestant,44 through out the 1960s. The second part of the Portuguese counterinsurgency plan was a sort of ‘winning the hearts and minds’ of the local population campaign. It rested upon and was made possible by the rapid economic development of the post-WWII period, and led to important social change on the planalto. Up until the mid 1950s, Salazar’s economic policy regarding Portugal’s colony

43 On the history of the PIDE/DGS in the Portuguese colonies, see Dalila Cabrita Mateus, A PIDE/DGS na Guerra Colonial (1961–1974), Lisbon: Terramar (Arquivos do Século XX), 2004, Chapter 1. 44 See for instance the case of Jesse Chipenda, the first Angolan Secretary General of the Congregationalist Church in Angola, who was arrested in 1967 and taken to the infamous camp of Sau Nicolau, where he died two years later (Lawrence Henderson, Development and the Church in Angola. Jesse Chipenda the Trailblazer, Nairobi: Acton Publishers, 2000). unita & moral economy of exclusion in angola, 1966–1977 165 amounted mainly to supporting the metropolitan wine and textile indus tries by providing them with exclusive export rights and preferential tariffs.45 In the 1950s however, the Portuguese economy started to shift gradually away from its colonies towards Europe (Portugal joined EFTA in 1959). In that new context, the barriers to the industrialisation of Angola were gradually lifted. Indeed, with the exponential growth in settler population as a result of Salazar’s policy of colonisation, the development of a local industry in Angola started to make sense. Two successive development plans (1953–1958 and 1959–1964) thus allowed important investments in Angola, especially in the extraction and transforma tion industries. In the early 1970s, industry made up to 41% of Angola’s GDP, against a mere 26% in 1960.46 These changes in the economic structure of Angola had a direct impact of course on the need for manpower to feed the growing industry. Even if sisal and coffee planta tions continued to employ thousands of unqualified workers, the grow ing industry in the 1960s offered new opportunities to young qualified Angolans. Another aspect of Portuguese colonial reformism in the 1960s is a reform of the education system. Following the abolition in 1961 of the legal distinction between natives (indígenas) and assimilados, racial seg regation in the school system was abolished up to a certain extent. Between 1960 and 1973, the number of pupils in primary and secondary schools were quintupled – but the situation was so catastrophic in the 1950s that, despite this sharp rise, hardly more that half of school-age chil dren were enrolled in the early 1970s.47 Despite their obvious limitations, the reforms in the education sector contributed, with economic develop ment, towards opening up the colonial world to African elites. The main consequences of this move were a relatively greater social and racial diversification within colonial societies of the central highlands, notably in Nova Lisboa. Besides, there were more possibilities for young people

45 William Gervase Clarence-Smith, The Third Portuguese Empire, 1825–1975. A Study in Economic Imperialism, Manchester: Manchester University Press, 1985. Valentim Alexandre, “Ideologia, economia e política: a questão colonial na implantação do Estado Novo”, Análise Social, vol. XXVIII (123–124), 1993, pp. 1117–1136. 46 Valério, Nuno & Fontoura, Maria Paula, “A evolução económica de Angola durante o segundo período colonial – uma tentative de síntese”, Análise Social, Vol. XXIX (129), 1994, p. 1203. 47 Eduardo Sousa Ferreira, O fim de uma era: o colonialismo português em África, Lisbon: Sá da Costa Editora, 1977. Elisete Marques da Silva, “O papel societal do sistema de ensino na Angola colonial (1926–1974)”, Revista internacional de estudos africanos, 16–17, 1992– 1994, pp. 103–130. 166 didier péclard coming from very different social and cultural background to meet and socialise in schools. In other words, the colonial world of the planalto was more integrated and less segregated in 1974 than it had been 14 years earlier. The history of the integration of the societies of the Angolan central highlands to Portuguese colonialism between the late 19th century and the end of the independence struggle leaves little room for the develop ment of political grievances in the name of the suffering and interests of the Ovimbundu as well as for the emergence of an ovimbundu ethnic con sciousness on the basis of which a nationalist movement could have been structured. One of the key elements in the history of the twelve ovim bundu kingdoms was their early insertion into long distance trade (in slaves, but also wax, ivory and rubber). This favoured both competition between the various kingdoms over the control of major trade routes and access to resources, and the individual initiative of political and economic entrepreneurs, thus leaving little room for the development of a pan- ovimbundu identity. And when ovimbundu societies were forced, during the first quarter of the 20th century, to convert to cash-crop agriculture following the end of long distance trade and the military conquest of the planalto, this reinforced individual initiatives and strategies of adapta tion, rather than nurturing a sense of common destiny in the face of colo nial rule. The social diversification which accompanied the partial opening up of the colonial economy and society in the late 1950s and 1960s went in the same direction. It was the beginning of the civil war around independence, and especially the way it played out in the Central Highlands, that was to change this and give political sense to a common ovimbundu narrative.

The Moral Economy of Exclusion

The war of 1975–76 brought back to the fore a central feature of the his tory of societies on the planalto in the 20th century, namely that there was very little, if any, social interaction between the different groups of elites that emerged at the margins of the colonial system. These divisions, as we just saw, had somewhat decreased with the economic boom of the 1960s and the (small and reluctant) opening of the colonial world to educated indígenas, but this move ultimately proved to be too short-lived to resist the consequences of the civil war. In order to understand the origins of unita & moral economy of exclusion in angola, 1966–1977 167 these divisions, we now turn to look at the dynamics of elite formation on the planalto.48 Due in good part to the weakness of the Portuguese education system in Angola, and the lack of will, on the part of the colonial authorities, to invest in the basic training of the indígenas since what the colonial plan tation economy needed were docile, disenfranchised workers, Christian missions played a crucial role in the formation of African elites on the planalto. For the vast majority of the population indeed, access to primary and secondary education, to vocational and technical training, or to health and other social services, was only possible through Christian mis sion societies. This is not the place to go into the detail of the education system that Protestant and Catholic missions developed, but one central feature of this project needs to be underlined here. Christian missionaries in Africa always had an ambiguous relation towards African cultures. On the one hand, most African cultural habits and religious beliefs represented for them all that they had come to fight in order to bring Africans towards the “lights” of Christian civilisation. On the other hand, however, the 19th century missionary revival, especially within Protestant circles, was born out of a strong criticism of European and North American societies in the age of the Industrial Revolution, which was seen as responsible for the growing secularisation of Western societies and for the “triumph” of materialism. In that vision, Africa repre sented the pristine continent where a new Kingdom of God could be built. Colonial societies in Africa, and colonial cities in particular, were for their part considered as the vehicles of the social ills that missionaries condemned in their own metropolitan societies: rampant secularisation, materialism, alcoholism, prostitution, individualism, extreme poverty, etc. Therefore, the role of mission societies was to “protect” the natives against the dangers of a colonial society that risked annihilating their communities.49

48 This section draws upon – and summarises dramatically – chapters 3 and 4 of D. Péclard, Etat colonial, op. cit. 49 See for instance Patrick Harries, “The Anthropologist as Historian: the Work of H-A. Junod”, Journal of Southern African Studies, Vol. 8(1), 1981, pp. 37–50. Having wit nessed the rapid social changes brought about by colonisation in Southern Mozambique and the effects of migrant labour to the Rand mines, missionary Henri-Alexandre Junod conceived of his duty as anthropologist as having “to document a civilisation about to disappear, to record what already existed and hence was ‘normal’, rather than that which was innovative and hence an exception” (p. 40). On the ways in which this ideology of 168 didier péclard

This sense of “protective segregation” was very present in the way Christian missions developed on the planalto. In a colonial context that was extremely hostile to traditional African societies, and within a system of plantation economy that threatened to dismantle the local communi ties where forced labourers were recruited, the interest of indigenous societies lay in their being allowed to develop in their “natural” environ ment, that is far away from the cities and other “dangerous” expressions of European materialism. As one Congregationalist missionary put it: Much has been written about the new ‘Tribe of God’ which the Church is calling into being in Africa. My observation is, however, that unless the new tribe is rooted firmly in the soil of its own social heritage, it will not grow into a tribe at all, but rather will fall apart into its components of déracinés, literally uprooted people.50 In other words, for Christian missionaries, the modernisation of Angolan societies – the creation of the “new Tribe of God” – went through the reinvention of traditional African cultural values within a Christian mould. The prime objective was therefore to built strong, cohesive Christian communities as much separated from the colonial world as was possible, where Angolans could develop in security and harmony, that is at their own pace and according to their own standards (or rather to the stan dards set for them by missionaries and Church leaders). This culturalist perspective on modernisation influenced the whole missionary project in the planalto: from school curricula to the structure of vocational training, the education and role of women in society and rural development. It bor rowed partly from the work of Booker T. Washington and his insistence on the role of vocational training in the education of Black Americans, but it also echoed some of the discussions going on within liberal circles in South Africa in the 1940s and early 1950s about the definition of an ‘appro priate’ education system for the African population. As an American Congregationalist missionary summarised it in a long report about the creation of a “Rural Life School” destined to train the community leaders that ovimbundu societies needed:

“protective segregation” was drawn upon by the ideologues of apartheid, see Hermann Giliomee, “The Making of the Apartheid Plan, 1929–1948”, Journal of Southern African Studies, Vol. 29(2), June 2003, pp. 373–392. 50 Gladwyn Murray Childs, Umbundu Kinship and Character, London: Oxford University Press, 1949, p. 69. In the 1950s, Childs, professor of theology at Dondi, taught a course on “aspects of ovimbundu culture that could serve as preparation for understanding and accepting the Gospel” (American Board of Commissioners for Foreign Missions Archives, Houghton Library, Harvard University, Boston - ABC 15.1. Angola Mission, 1950–1960, Box 2: 20, Report of Emmanuel Seminary and Bible School for the year 1955). unita & moral economy of exclusion in angola, 1966–1977 169

Because Africans have a different history and a different cultural back ground, because they are for the most part rural dwellers, because their present-day needs are different from those of present-day Europeans, differ ences in curricula are not only desirable but imperative. […] The objective of education in Africa is to produce the good African – the Native who is proud to be an African, appreciative of the finer elements in his culture, willing and anxious to accept European culture in so far as it is complemen tary and supplementary to his own, quite unwilling to be an imitative and unoriginal White man. Good native education would have its roots in African culture with leaves and branches of the best that the West has to offer.51 For American missionaries in the planalto, the main danger related to the ‘de-tribalisation’ of ovimbundu societies due to colonisation was that they become “imitative and unoriginal White men”, i.e. that they be left with no other choice than to become uprooted objects of the Portuguese colonial project. Therefore, their mission was conceived in many ways as that of instilling a culture of distinction within ovimbundu societies. The missionary project, in other words, was about laying the basis for a renewed sense of “honour”52 in ovimbundu culture – or rather, a Christian re-invention of this culture – that had been severely damaged after the defeat against colonial troops. Because they saw ovimbundu societies as fundamentally rural, their ‘long walk’ towards Christian modernity had to go through the constitution of rural bastions solid enough to resist the assaults of Portuguese colonialism and to serve them as a shield behind which they could develop in peace and without the negative influence from the exterior world. This ideological discourse was sustained by visible transformations in the material culture of ovimbundu communities. Christian villages (Protestant in particular) were known (and even envied by many Portuguese officials) for their orderliness and their tidiness.53 Protestant missionaries were also very proud of the fruit orchards that they planted in their mission stations, and they were the first to cultivate strawberries in the region. As John T. Tucker, one of the main figures of

51 ABC 15.1 Angola Mission, 1940–49 Box 3: 9, [notes by Carl Dille on education], [1948]. Dille was referring explicitly to C.T. Loram, South African educationist whose work (The Education of the South African Native, 1917) was very influential. Whether this is a direct quote from Loram is however unclear. 52 John Illife, Honour in African History, Cambridge: Cambridge University Press, 2005. 53 On the role of Protestant missionaries in the transformation of habitat and village landscapes see Jean & John Comaroff, Of Revelation and Revolution, Vol. II: The Dialectics of Modernity on a South African Frontier, Chicago & London: University of Chicago Press, 1997, Chapter 6. 170 didier péclard the Congregationalist missions in Angola recounts in his memoirs, a British journalist who visited the mission in 1925 was surprised when he was served strawberries for desert, and literally astonished when he found out that these strawberries had been sold to the Tuckers by a farmer from a nearby village who would then use the money to pay his taxes. A few days later, the Times published the story under the headline: “Taxes paid by natives with strawberries in Angola”.54 Soy beans were another identity marker of Protestant communities on the planalto. Introduced by Congregationalists in order to diversify the staple food of the populations and as a source of proteins, it came to be known as feijão protestante, liter ally Protestant beans, feijão being the Portuguese staple food. These few examples are important well beyond their anecdotic charac ter. Christian mission laid the ground for the adaptation of ovimbundu societies to the new economic system dominated by cash-crop agricul ture, and this on several levels. Their programmes in rural development (new types of culture, fight against erosion, help with acquiring new tech niques such as the plough), which were repeatedly hailed as success by the Portuguese authorities despite the official distrust of those very authorities towards Protestant missions, were one of the factors that helped account for the successful “adaptation” of ovimbundu societies to their new socio-economic environment. This material culture of life under colonialism also had an important symbolic component: ploughing one’s fields, cultivating new fruits and vegetables, living in “modern vil lages” were important identity markers of the new rural elites that Christian missionaries had intended to build. Finally, with their educa tional system, Christian missions were also a place where an elite of new assimilados could emerge without having to turn its back to its culture of origin. In a colonial system that despised “traditional” African cultures, this proved to have a lasting impact. Indeed, the missionary perspective on the culturalist modernisation of Angolan society was not only deeply influential on missionaries and on the institutions they created, the school curricula they drew, etc. Most importantly, it also left a deep mark on the first generations of Angolan converts and on the elite of teachers, nurses, clerks and qualified workers that emerged within the bosom of Christian missions. They appropriated

54 J. Tucker, A Tucker Treasury. Reminiscences and Stories of Angola, 1883–1958. Selected and prepared by Catherine Tucker Ward, Winfield: Wood Lake Books, 1984, pp. 192–193. unita & moral economy of exclusion in angola, 1966–1977 171 this perspective and made it their own as a way to cope with the military defeat against the Portuguese. They integrated it into the new “modes of self-writing”55 that they were inventing under colonial rule, in order to both find a new sense of pride and find ways up the social ladder (or out of the far margins of society where many had retreated after the colonial conquest and the end of the rubber trade). In other words, this model was successful because it combined the two registers within which the colo nial encounter was played out: the register of “radical individualism”, which “had to do with the construction, by the Europeans, of the modern ist African subject” and the register of “primal sovereignty”, which “defined people by virtue of membership in ‘customary’ political communities”.56 The crucial point here is that whether or not the model of culturalist modernisation made sense politically and socially depended strongly on the historical context. In the Angolan central highlands, three moments need to be differentiated. Firstly, up until the 1950s approximately, as I indicated earlier, this model was instrumental in facilitating the conver sion of ovimbundu societies to cash-crop agriculture and in helping them cope with the new political order. Secondly, with the relative opening up of the colonial world that the changes in the economy and the needs of Portugal’s counter-insurgency campaign permitted, the model was chal lenged by a generation of young clerks, teachers and nurses who did not see, like their parents had, much sense in returning back to their commu nities of origin after completing their education in order to support the cohesion of the Christian rural bastions they had mainly come from, and for whom modernity clearly meant life in the city and working within the colonial world (rather than at the far margins, as the missionaries and their parents would have it). As we saw, this resulted in a greater social diversification and integration, especially in and around the city of Nova Lisboa (Huambo). Thirdly, with the first phase of the civil war, this changed again dra matically. After the takeover of Huambo by the MPLA and Cuban forces,

55 A. Mbembe, “A propos des écritures africaines de soi”, Politique Africaine, 77, 2000, pp. 16–43. 56 J. & J. Comaroff, Of Revelation and Revolution, op. cit., pp. 369–370. Emphasis original. Whereas the Comaroffs see the two registers as antagonistic, I see them as complementary, since it was by combining the two that African elites managed to fully enter modernity by resorting to (re-invented) tradition, as John Peel has shown so convincingly in the case of the Yoruba (Religious Encounter and the Making of the Yoruba, Bloomington & Indianapolis: Indiana University Press, 2000). 172 didier péclard a new administration was put in place in order to fill all the posts that had been left vacant with the departure of the Portuguese at independence. This is when the type of socialisation that many ovimbundu elites had had through the mission system, and beyond that the whole symbolic world that had sustained their communities on the basis of the mission ary model of culturalist modernisation again became an obstacle to their integration into independent Angola. Indeed, there was a wide social gap between local elites from the rural planalto and the young men and women who came as the ‘new forces of the nation’ in the wake of the MPLA’s military victory. Although both groups partly came from the same Central Highlands, their life experiences had been very different. Whereas the first had been educated and socialised mostly within networks that centred on the planalto, the second had all left the region to acquire sec ondary education in Luanda or Lisbon. It is also in Luanda and Lisbon that the second group were socialised and grew into the MPLA. Their out look on what it meant to live in an independent Angola and, more impor tantly, their perspectives on the modernisation of Angolan society were way apart, when not in direct opposition. Since, because of the history of social segregation on the planalto and in the city of Nova Lisboa in par ticular, they had had virtually no opportunity to meet and share experi ences, either at school, in sports club or through music and dance, they had not grown to know and respect each other. As already indicated, the limited trend towards social inclusion that had marked the late 1960s and early 1970s had not been enough to change this. All this meant that the encounter between the MPLA cadres that took over the state administra tion and the new assimilados of the Central Highlands was simply not possible. As a result, many of those new assimilados, despite their educa tional background and, for some of them at least, years of experience as teachers, schoolmasters, nurses, etc., were not integrated into the local administrative structures. Or when they were, it was rarely in a leading position.57 In other words, the very values around which part of ovimbundu soci eties had re-built themselves after their defeat at the hands of the colonial armies and on the basis of which they had struggled to develop a sense of

57 This aspect of the social history of the elites on the Central Highlands still remains largely to be written, especially for the very final phase of the colonial period. I am aware of the fact that, for both lack of space and of research on the matter, the picture I am draw ing here is too broad and rapid. It should therefore be taken more as a hypothesis or sug gestion for further research. unita & moral economy of exclusion in angola, 1966–1977 173 social exclusiveness tended to be relegated, in the particular conditions of Angola’s access to independence, to signs of social and cultural backward ness that were incompatible with the MPLA’s modernising project. The project of culturalist modernisation had lost and was in the process of being sidelined. In these circumstances, UNITA’s military defeat in 1976 was experienced by many not just as the loss of a military battle, but as the defeat of a social and political imaginaire. This is what UNITA managed to draw upon after its forced retreat into the Terras do Fim do Mundo and what made this defeat a “successful” one. This is also what formed the symbolic repertoire out of which UNITA drew the narrative of exclusion that was to form the core of its social legitimacy until the early 1990s at least.

Conclusion

What I have argued here is that, due to the particular political, social and economic conditions of the “colonial moment” on the planalto, as well as to the history of ovimbundu kingdoms, the ovimbundu imagined community only had very limited political meaning or expression as a response to colonial rule. The articulation of ovimbundu societies to colonial modernity followed various routes, thereby reinforcing cen trifugal tendencies between them. Christian missions, especially after military conquest, played a crucial role as cultural and social brokers between ovimbundu societies and the colonial world, and they offered some of the very few avenues of social upward mobility for so-called indígenas in a colonial system which had been built on their exclusion. But they also largely contributed to social compartmentalisation in the planalto in their attempt at creating closed-in, “gated” communities where “true” Africans, or at least a Christianised reinterpretation of what they saw as the “genuine” identity of Africans could be educated away from the “dangers” of the colonial world. As a consequence, there was until the end of colonial rule very little interaction between the various elites that emerged, very few possibilities of socialisation and common experiences. This started to change slowly in the course of the 1960s and early 70s when, partly as a response to the nationalist struggle, partial integration into the colonial world became possible. But this movement came to a halt with the beginning of the civil war, which crystallised what was by then a very moving and open social setting, and which gave new political meaning to regional, racial, social and ethnic identities. 174 didier péclard

Although until independence appropriating (ethnic) tradition for political action hardly made any sense in the social context of the Angolan central highlands, this changed dramatically with independence. And the growth of UNITA in those crucial years can be explained, partly at least, by the way in which Savimbi managed to rewrite the stories of the various ways in which ovimbundu societies articulated to colonial modernity into a single grand narrative of common exploitation and suffering on the part of the Ovimbundu, first in the hands of Portuguese colonialists, and sec ond in the hands of the “new colonisers” who had conquered Huambo in the MPLA’s name. In the process, the different forms of “moral ethnicity” which had developed within ovimbundu societies during colonial times, not least in the bosom of Christian missions and churches, gave way to the “political tribalism”58 that fuelled part of the civil war until 2002.

58 John Lonsdale, “Ethnicité, morale et tribalisme politique”, Politique Africaine, 61, mars 1996, pp. 98–115.

ANGOLA’S EURO-AFRICAN NATIONALISM: THE UNITED ANGOLAN FRONT

Fernando Tavares Pimenta

This chapter deals with the political identity of the white settler community in Angola. More precisely, I analyze the emergence of a form of Euro- African nationalism among Angolan-born whites during the Portuguese colonial dictatorship of Salazar, the Estado Novo (“New State”). Firstly I will define Angola’s Euro-African nationalism. Then I will proceed by describing the features of the Portuguese colonization in Angola and the conditions from which there emerged a kind of Angolan white settler nationalism, its evolution and failure to achieve the control of the Colonial State. Euro-African nationalism was the anti-colonial political protest carried out by a part of the Angolan white community (and some mulattoes) whose aim was the acquisition of Angola’s independence from Portugal. Euro-African nationalism conceived the Angolan nation as the product of the cultural interaction between the European and the African social spheres and it considered that the whites had a political dynamic role in the conquest of independence and in the construction of the Nation State in Angola. Euro-African nationalism was – in a certain sense – the Angolan equivalent of the Creole (white) nationalism in Brazil, in the 18th and 19th centuries, and it also had some parallels with the white nationalism in South Africa. Indeed Brazil and the British dominions (and self governing colonies) served as political models for the Euro-African nationalists in Angola. As a consequence, Euro-African nationalism was distinct in character from African nationalism, since the latter considered Angola fundamentally an African (black) nation where the European (white) element was politically peripheral.1 The history of Euro-African nationalism is deeply connected to the features of Portuguese colonization in Angola. Indeed during the

1 Fernando Tavares Pimenta, Angola. Os brancos e a independência, Oporto: Afrontamento, 2008

twentieth century the Portuguese practised in Angola a subtype of colonialism that we can classify as settler colonialism. Settler colonialism is characterized by the existence of a third force in the colonial situation, the settler community. Indeed, what distinguishes settler colonialism from other forms of colonialism is that whereas in the majority of colonial situations there are two major groups involved – the colonial authorities (and some temporary migrants from the colonizing country, such as military personnel and missionaries) and the indigenous people –, in settler colonies the settlers constitute a third group.2 Cleary the existence of a settler community permanently established in a colony has strong implications for its economic, social and political life, since settlers compete with the Motherland government for the control of the Colonial State3 and with the colonized elites for the internal hegemony in the colonial society, especially for the control of both the land and the indigenous labour. Actually, white settlers wanted to control the political colonial apparatus in order to protect their economic interests from adverse foreign economic interests (of the Motherland bourgeoisie or others) and from competition of temporary migrants from the metropolis and non-white producers and traders. That was also important in order to assure the preservation of their reserves of African cheap labour. During the twentieth century the number of settlers in Angola grew steady, although slowly, until World War II and then more quickly until 1974. White settlers were 9.000 in 1900, 12.000 in 1910, 20.700 in 1920, 30.000 in 1930, 40.000 in 1940, 78.000 in 1950, 172.000 in 1960, 290.000 in 1970 and 335.000 in 1974. By the end of colonial rule, in 1974/1975, the settlers represented more than 5% of the total population in Angola and they were the second largest white settler community in Africa, which was only overcome by South Africa.4

2 See David Prochaska, Making Algeria French. Colonialism in Bône, 1870–1920. Cam bridge: Cambridge University Press, 2002; Udo Krautwurst, “What is settler colonialism?”, History and Anthropology, vol. 14, n.° 1, 2003; Caroline Elkins, Susan Pedersen (Edited by), Settler Colonialism in the Twentieth Century: Projects, Practices, Legacies. New York/ London: Routledge, 2005. 3 The problem was that the economic policy dictated by the Motherland government was frequently an expression of the needs of the metropolitan bourgeoisie or even of the interests of foreign capital. However, the needs of the metropolitan bourgeoisie or the interests of foreign capital often clashed with the economic interests of settlers. Thus settlers demanded the control over the colonial State and Angola’s political autonomy. See Paul Mosley, The Settler Economies. Studies in the Economic History of Kenya and Southern Rhodesia 1900–1963. Cambridge: Cambridge University Press, 1983, pp. 170–194. 4 Fernando Tavares Pimenta, Brancos de Angola. Autonomismo e Nacionalismo, 1900– 1961. Coimbra: Minerva, 2005, pp. 191. See also Cláudia Castelo, Passagens para África: o angola’s euro-african nationalism 179

In the first decades of the twentieth century, Portuguese settlers developed a sort of autonomist political protest, which advocated the economic and political autonomy of Angola, as the first step for the achievement of self-government. Thus, settlers demanded the control over the colonial State and the non interference of the metropolitan government in the internal political and economic issues of the colony. White settlers in Angola were politically influenced both by the republican ideas from the Motherland and by the self-government policy applied by Britain in Southern Africa. Indeed white settlers expected to construct a new “home” in Angola in the sense that they considered Angola a “white man country” or, in other words, a place where they would construct a new nation, as European settlers had previously done in America, Australia and South Africa. More precisely, Portuguese settlers in Angola saw themselves as the constructors of a sort of “second Brazil” or even a “New Lusitânia”, which would be part of a larger Portuguese Commonwealth (Comunidade Lusíada) made up by Portugal, Brazil, Angola and Mozambique.5 This means that white settlers supported the idea that Angola should evolve into an independent country like Brazil, under the political leadership of the whites, but with the political participation – in a subordinate position – of the mulatto and the Europeanized black elites.6 This idea was also supported by some colonial administrators, namely Norton de Matos (the 1st High Commissioner of Angola) and Vicente Ferreira (the 3rd High Commissioner of Angola), who believed that Angola’s highlands could become the home of a new “white nation” in Africa.7 Initially the white settlers thought that the Portuguese government would back up their political aspirations. Indeed, during the Portuguese Democratic Republic (1910–1926), the settlers expressed freely their political demands through their Economic Associations and additionally through the colonial press, such as the newspapers Jornal de Benguela

povoamento de Angola e Moçambique com naturais da metrópole (1920–1974), Oporto: Edições Afrontamento, 2007. 5 See, for example, Jornal de Benguela, Ano VIII, n.° 7, de 14/02/1919, p. 1; n.° 9, de 28/02/1919, p. 4; n.° 17, de 25 de Abril de 1919, p. 1; n.° 32, de 8 de Agosto de 1919, p. 1; n.° 25, de 20 de Junho de 1919, p. 17; Jornal de Benguela, Ano XIII, n.° 33, de 15 de Agosto de 1924, p. 2; Jornal de Benguela, Ano XV, n.° 17, de 23 de Abril de 1926, pp. 1–2. 6 José de Macedo, Autonomia de Angola. Estudo de administração colonial. Lisbon: Edição do Autor, 1910, pp. 83, 209–215. 7 Norton de Matos, A Província de Angola. Oporto: Edição de Maranus, 1926, pp. 11–13; Vicente Ferreira, A política colonial portuguesa em Angola. Coimbra: Imprensa da Universidade, 1932, p. 59; Vicente Ferreira, A “Nova Lusitânia” – Angola. Separata do Boletim da Sociedade de Geografia de Lisboa, n.° 7 e 8, Julho-Agosto de 1948, p. 32 180 fernando tavares pimenta and A Província de Angola. Their strongholds were the freemason civic associations in Luanda (Grémio Português), Benguela (Grémio Lusitânia) and Bié (Grémio Pátria Nova), which were a sort of “interest groups” reuniting the elites of the white settler community. Settlers also formed some autonomist parties, such as the Partido Reformista de Angola, Partido Republicano Colonial and Centro Republicano de Benguela, and the more radicals Partido Pró-Angola and União dos Defensores de Angola.8 However, despite of the political claims of the white settlers, the Portuguese democratic government denied settlers the control over the Colonial State, which functioned under the control of a High Commissioner, who was nominated by Lisbon. Nevertheless Lisbon gave a form of limited autonomy to the colony and it accepted the creation of a Legislative Council in 1922, where settlers elected representatives and could express and defend their political views and interests.9 The political situation changed drastically with the establishment of a military dictatorship in Portugal (1926), which evolved into the civil dictatorship of Salazar, known as the “New State” (Estado Novo, 1933–1974). The dictatorship reinforced the Motherland’s control over the colony, suppressed the Legislative Council and repressed the autonomist political activities. The white settlers tried to resist the colonial repression and they even raised in rebellion against Portugal in March 20th 1930.10 During the rebellion some influential settlers from Benguela proposed that Angola should declare itself an independent Republic, following the example of Brazil.11 But this was a complete “madness”, because the settlers did not have the numbers and the political, military and economic organization to make a unilateral declaration of independence. The fact was that settlers were few (they were still less than 1% of Angola’s total population) and they did not have the strength to defeat the Portuguese government. Consequently settlers were forced to submit to Salazar, who already was the most influential member of the dictatorship government.12

8 Associação Comercial de Benguela, Projecto de Autonomia de Angola. Benguela: [s.p.], 1919; J. Camilo Rodrigues, Projecto de Autonomia de Angola. Luanda: [s.p.], 1919. 9 Fernando Tavares Pimenta, op. cit., 2005, pp. 77–103. 10 Arquivo Nacionais Torre de Tombo (AN/TT), Fundo do António Salazar (AOS), AOS/ CO/PC – 3, Pasta 1 – Acontecimentos revolucionários em Angola (1929–1930). Cf. Helena Pinto Janeiro, “Os 21 dias que abalaram Angola”, História, ano XX (nova série), n.° 1, Abril 1998, pp. 26–35; Fernando Tavares Pimenta, op. cit., 2005, pp. 105–110. 11 Public Record Office (PRO), Foreign Office (FO) 371/15030, Portugal, 1930 (Consul- General Smallbones to Mr. A. Henderson, 10/05/1930). 12 PRO, FO 371/15030, Portugal, 1930 (Consul-General Smallbones to Mr. A. Henderson, 29/07/1930). angola’s euro-african nationalism 181

Salazar’s reaction to the rebellion was hard: white settler autonomists were arrested, deported or exiled; freemasonry was prohibited; almost all political and civic freedoms were suppressed.13 With the promulgation of Colonial Act of 1930 (and subsequent legislation), Salazar concentrated all the power in the hands of the Lisbon government; centralized the administration and excluded completely the settlers from the decision making process in Angola.14 Salazar also enforced the execution of its economic legislation favouring metropolitan interests as opposed to settlers’ interests. From 1930 onwards the white settlers were obliged to buy almost all the things they needed from Portugal alone and, at the same time, they were obliged to sell their products to Portugal at low prices. Portuguese economic legislation also favoured the interests of foreign capitals (American, Belgian, British, French and South African), which controlled the colony’s economy. For example, a substantial part of the coffee production, the entire cotton production and the mining industry – especially diamonds – were controlled by foreign capital, sometimes with the participation of Portuguese capital. Foreign capital also controlled transport and communications, especially the navigation companies and the Benguela railway, which connected the Katanga Coperbelt (in the former Belgian Congo) to the international Atlantic port of Lobito.15 As a consequence, settlers resented what they called the economic exploitation of Angola by foreigners (including metropolitan Portuguese) and the close control of all business and administration in the colony by the Motherland. They also resented the fact that almost all senior officials of the colonial administration were recruited from Portugal.16 Salazar’s colonial dictatorship also practised a sort of racial discrim ination against Angolan born whites, who were racially classified as

13 José de Sousa e Faro, Angola. Como eu a vi em 1930–31. Lisbon: Imprensa da Armada, 1932, pp. 12–16; Fernando Tavares Pimenta, op. cit., 2005, pp. 111–114. 14 Ministério das Colónias, Acto Colonial, decreto n.° 18.570, de 8 de Julho de 1930. 15 PRO, FO: 371/39583 – Angola reports on political and economic conditions, 1944 (From “Donald – Anglo-Portuguese Club Luanda, to the Rev. B. F. Chambers, 11, Vicarage Gate, Kensington, London, 28/10/1943); Associações Económicas de Angola – Considerações sobre o problema das transferências de Angola. Lisbon: Imprensa Libânio da Silva, 1932. See also Adelino Torres, O Império Português entre o real e o imaginário. Lisbon: Escher, 1991. 16 PRO, FO: 371/125894 – Internal political situation: Angola, 1957 (Grant Purnes, British Consul General, Luanda, to British Embassy, Lisbon, 23/04/1957). AN/TT, AOS/CO/UL – 8B, Pasta 3 – Crise de Angola, 1932; AOS/CO/UL – 8B, Pasta 12 – Situação política em Angola (1934); AOS/CO/PC – 3C, Pasta 17 – Situação em Angola (1936); AOS/CO/UL – 8B, Pasta 18 – Situação económica e social em Angola (1937–1941) 182 fernando tavares pimenta

Euro-Africans. Indeed, the Angolan colonial situation was permeated by heavy racial tensions and – contrary to official Lusotropicalist assump- tions17 – racism was at the basis of Portuguese colonialism in Angola. Racial discrimination legitimised the official demographic segmentation of Angolan population in three groups: whites, mulattoes and blacks, the former divided in Europeans and Euro-Africans,18 the later divided in “civilized” and “uncivilized” or, in other words, assimilados and indígenas. Until the abolition of the indigenous status or Estatuto do Indigenato19 in 1961, only “civilized” black or assimilados (never more than 1% of the entire black Angolan population) had the right to claim full Portuguese citizenship. “Uncivilized” blacks or indígenas were considered merely as colonial subjects and they could be recruited by the colonial authorities as forced labour.20 In the case of the white population, the colonial regime promoted racial prejudice against Angolan born whites, who were considered second class citizens, as well as the mulattoes. There are many historical evidences and statements of the existence of this kind of discrimination against white Angolans. For example, Thomas Okuma, an American missionary in Angola during the 1950s, stated that: Discrimination by the Portuguese against Angolan Europeans over a period of many years has made the ties to the mother country weaker for the Angolan Europeans than for the new colonos or settlers from Portugal. Angolan Europeans resent the fact that metropolitans consider them second-class Portuguese. Prior to 1950 the bilhete de identidade (identity card) of Angolan Portuguese was not valid in the homeland. Restrictions on

17 See Cláudia Castelo, O modo português de estar no mundo. O Luso-tropicalismo e a ideologia colonial portuguesa (1933–1961). Oporto: Afrontamento, 1998. 18 Direcção dos Serviços de Economia. Repartição de Estatística Geral da Colónia de Angola, Censo Geral da população 1940. Luanda: Imprensa Nacional, 1947; Afonso Costa Valdes Thomaz dos Santos, Angola. Coração do Império. Lisbon: AGC, 1945, pp. 41–42; João Pereira Neto, Angola: meio século de integração. Lisbon: [s.p.], 1964, p. 241; Alberto Lemos, Nótulas históricas. Luanda: Fundo de Turismo e Publicidade de Angola, 1969, p. 196; Fernando Tavares Pimenta, Angola no percurso de um nacionalista. Conversas com Adolfo Maria. Oporto: Afrontamento, 2006, pp. 32 e 37–40. 19 Ministério das Colónias, Estatuto Político, Civil e Criminal dos Indígenas de Angola e Moçambique, decreto n.° 12.533, de 23 de Outubro de 1926; Ministério do Ultramar, Estatuto dos Indígenas Portugueses das Províncias da Guiné, Angola e Moçambique, decreto-lei n.° 39.666, de 20 de Maio de 1954, Ministério do Ultramar, Revogação do decreto-lei n.° 39666, que promulga o Estatuto dos Indígenas Portugueses das Províncias da Guiné, Angola e Moçambique. Decreto-lei n.° 43893, de 6 de Setembro de 1961. Lisbon: AGU, 1961. 20 Ministério das Colónias, Código do Trabalho dos Indígenas nas Colónias Portuguesas de África, decreto n.° 16.199, de 6 de Dezembro de 1928; Adriano Moreira, “As elites das províncias portuguesas de indigenato: Guiné, Angola e Moçambique”, separata da revista Garcia da Orta, vol. 4, n.° 2 (Lisbon, JIU), 1956. angola’s euro-african nationalism 183

travel to Portugal applied to them as it did to non-Portuguese residents in Angola.21 In 1961, the United Nations Sub-committee on the situation in Angola also stated that: It was stated that though the major line of distinction in social practices has been between the não-indígenas and the indígenas and in spite of the objectives of Government policy regarding a multi-racial society, in Angola race and place of birth had come to determine, in practice, many rights and privileges. It was said that in Angola there were in practice five categories of inhabitants. First the Portugal-born Portuguese; second, the Portuguese actually born in Angola; third in line was the mestiço (mulatto); next was the African assimilado; and finally, the great majority of the Africans.22 As a consequence of Salazar’s policy of discrimination against Angolan born whites, the white community suffered heavy social and political tensions between two groups of settlers: on the one hand, an older settler generation and the Angolan born whites; on the other hand, a new generation of Portuguese settlers which established itself in the colony after 1945.23 This division was stated (and denounced) by a Euro-African nationalist movement in the 1960s, the Angolan United Front (Frente de Unidade Angolana, FUA): Nowadays, Angola’s white population is divided in two different and even antagonist groups, which are: 1) The Africanised whites, formed by settlers and their descendants, whose economic interests and national feelings are entirely in Angola; 2) European whites, formed by those emigrants who came to Angola only in search of self-enrichment – once they fulfilled this aim they go back to their Homeland; and by civil servants and clerks of big international companies, whose presence in the country is only temporary. The motive of their antagonism is quite clear since their interests are divergent. The Africanised white farmer, industrial or trader feel that its interests are damaged by the interests of foreign economic forces; the Angolan born white worker, civil servant or intellectual feel that they are put apart by the Portuguese government, who chose only Portuguese born whites to decision positions in colonial administration, in spite of the professional merits of each one. The first group is nationalist by its attachment

21 Thomas Okuma, Angola in Ferment: the Background and Prospects of Angolan Nationalism. Boston: Beacon Press, 1962, p. 59. 22 United Nations General Assembly, 16a Session, Agenda Item 27 – Report of the Sub- committee on the Situation in Angola (22/11/1961). 23 Interview to Adolfo Maria, Angolan born white nationalist, member of MLNA, FUA and MPLA (Lisbon, January 2004). See Fernando Tavares Pimenta, Angola no percurso de um nacionalista. Conversas com Adolfo Maria. Oporto: Afrontamento, 2006. 184 fernando tavares pimenta

to the country; the latter is colonialist, since it represents and defends colonial interests.24 Simultaneously, probably as a reaction to Salazar’s discriminatory policy, white settlers – especially the Angolan born whites – developed a strong political attachment to Angola. This deep identification of settlers with Angola was pointed out by the British Consul General of Luanda, J. C. Wardrop, in 1962: It is not generally realised how deep are the roots of the European population in Angolan soil. Many were born here, many have come in the present generation with the intention of staying for good. You find them not only in the larger towns and plantations but dotted all over the map in innumera- ble tiny and remote villages. The majority are humble folk who could not afford to visit Portugal even if they wanted to. They belong here; they know no other home; to them Angola é nossa (“Angola is ours”)! They have no parallel in any British colony that I know of. In our former West African possessions the British were administrators, soldiers or business men, the great majority of whom were based on, and retired to Britain. In Kenya and Rhodesia we have, it is true, settlers of longer standing. But in the main they are relatively well-to-do and still have their links with the home country. Only Algeria and South Africa are comparable in this respect with the Portuguese African Provinces.25 Salazar’s authoritarian and racist government increased the settler political and economic dissatisfaction with the Portuguese colonial rule and it contributed to the emergence of anti-colonial nationalist feelings among the white settler community. Indeed, Euro-African whites felt that they were being victims of the colonial policies of the Portuguese government and they even felt that they were being colonized by Portugal, as the mulatto and black populations. Because of this, the settlers political identity changed in the sense that whites started to consider themselves as Angolan nationals and not as Portuguese expatriates. White settlers also began to reject the metropolitan colonial rule and to support the idea of Angola’s secession. Thus the Salazar’s colonial dictatorship caused the metamorphosis of the settler autonomist protest into a more radical political protest, the Euro-African nationalism which emerged in the 1930s and the 1940s.26 Euro-African nationalists did not have, however,

24 Frente de Unidade Angolana, “A população branca no contexto nacional”, Kovaso. Órgão da FUA, Fevereiro de 1963, p. 2. 25 PRO, FO: 371/161626 – Internal Political Situation: Angola, 1962 (J. C. Wardrop, British Consul General, Luanda, to British Embassy, Lisbon, 16/04/1962, p. 3). 26 Fernando Tavares Pimenta, op. cit., 2005, pp. 119–136; Afonso Costa Valdes Thomaz dos Santos, op. cit., 1945, p. 54. See Sócrates Dáskalos, Um testemunho para a História de angola’s euro-african nationalism 185 the strength to face openly Salazar’s colonial dictatorship, at least in normal political conditions. That is why Euro-African nationalists challenged the colonial dictatorship and expressed publicly their grievances and political aspirations only in moments of crisis of the colonial regime, namely during World War II (1939–1945), the Portuguese presidential elections of 1958, and in the first years of the Angolan independence war (1961–1974). But let us see in detail the political evolution of Euro-African nationalism from 1939 onwards. The beginning of World War II in Europe (and its extension to Africa) made Salazar’s colonial dictatorship more vulnerable – politically and military: the communications between Portugal and its colonies became weaker, while the Motherland’s military capacity to react to a settler rebellion in Angola became very limited. White settlers tried then to take advantage of the Portuguese vulnerability and of the international political situation to obtain Angola’s secession from Portugal. Indeed, in exchange for Belgian, French and especially South African political (and military) support to their secessionist project, the white settlers promised to fight against any attempt to seize power by the pro-German Portuguese authorities in Angola. The fact was that the pro-German Portuguese current was very influential both in the Motherland and in the colony, a problem that preoccupied the Allied Forces, especially the colonial authorities of French and Belgian Congo and, of course, South Africa. The leading figures of the settlers conspiracy were the clergyman Monsenhor Alves da Cunha, who was the second most important figure of the Catholic Church hierarchy of the colony and the leader of the Angolan freemasons (Kuribeka), and Venâncio Henriques Guimarães, who was a wealthy farmer of Lubango, in Southern Angola. Euro-African nationalists were able to establish some secrets contacts with South Africa, which was interested in extending its own influence to the Portuguese colonies, and also with the Belgian and French colonial authorities in Leopoldeville and Brazzaville. Plans were made for a rebellion of white settlers against Portuguese rule, followed by a military invasion of South African, French and Belgian troops. But the British government opposed the invasion of Angola by the Allied Forces and the Portuguese colonial authorities discovered the white settler conspiracy in mid-1941. As a consequence, the Allied invasion plans were put aside under British pressure, while some of

Angola. Do Huambo ao Huambo. Lisbon: Vega, 2000. Américo de Carvalho, Angola. Anos de esperança. Coimbra: Minerva, 2001. 186 fernando tavares pimenta the leading figures of the settler conspiracy were arrested and deported to Portugal, notably Monsenhor Alves da Cunha.27 Simultaneously a new generation of Angolan born whites begun to rise in politics. In 1940 this new generation of whites created – with some mulattoes – a political movement in Huambo, in the central highland, known as Angola’s Socialist Organization (Organização Socialista de Angola, OSA), whose aim was to attain the independence of Angola. OSA was the first Euro-African nationalist party and its leadership was made of Angolan born whites, such as Sócrates Dáskalos, and some mulattoes. This movement claimed full political independence and not only some kind of autonomy. OSA also demanded an end to discrimination against Angolan born whites – who were classified as second class citizens – and against mulattoes and the end of forced black labor. But OSA was elimi- nated by Portuguese colonial repression in mid-1941.28 The victory of the Allied Forces in World War II obliged Salazar to enact some changes in Portuguese colonial policy and to make some political concessions to the settlers in Angola. Thus Salazar had to allow Legislative elections and also the constitution of an opposition movement in 1945, the Democratic United Movement (MUD). The MUD was based in the Motherland, but it had a delegation in Angola, which was formed by several Euro-Africans nationalists. Indeed, many Euro-African nationalists supported politically the MUD, because they expected to gain self government if the dictatorship would be replaced by a democratic regime in the Motherland. But a few days before the Legislative elections, the MUD decided to withdraw from the ballot because the government did not assure the holding of democratic and free elections. This was the reason why the elections were won by the National Union, the single party of Salazar, who went on to remain in power until 1968. For the white settlers it was a remarkable political disappointment, which generated a lot of political frustration among younger generation of Angolan born whites.29

27 AN/TT, AOS/CO/UL – 8F, Pasta 5 – Incidente entre o Vigário-Geral de Angola, Monsenhor Alves da Cunha, e o Governador Geral, Dr. Marques Mano; AOS/CO/UL – 62, Pasta 12 – Situação em Angola (1941); AOS/CO/UL – 62, Pasta 13 – Situação política em Angola (1941). PRO – FO 371/26847, Angola and Union of South Africa, 1941; PRO, FO 371/26841, British troops movements on frontier of Angola, 1941; PRO, FO/371/31120, British interests in Angola, 1942; PRO, GFM 33/507, German Legation Lisbon – Consulate Luanda – 1941, Serial 1094, 1095. 28 Ibid.; Socrates Dáskalos, op. cit., 2000, pp. 46–70; Américo de Carvalho, op. cit., 2001, pp. 17–45. 29 AN/TT, AOS/CO/PC – 3I – Democratas de Angola e as eleições legislativas (1945); PRO, FO 371/67859 B, Economic Situation in Angola (1945–1946), 1947. See also Henrique angola’s euro-african nationalism 187

Politically Salazar also changed the official status of Angola, which evolved from a colony to an Overseas Province of Portugal, in 1951. But there were no actual substantial political changes – this was part of a strategy to avoid the anti-colonial pressure of the United Nations. Only in 1955 did Lisbon make a first move towards colonial decentralization, but not to self-government, in the sense that Salazar authorized the creation of a new Legislative Council. This new Legislative Council gave a symbolic political representation to that part of the Angolan population which was officially classified as “civilized”, in other words, the white settlers, the majority of mulattoes and a tiny minority of Europeanised blacks, known as assimilados (a sort of Portuguese évolués). But the Legislative Council had only a consultative role and the degree of political participation of white settlers in the colonial government remained rather low. Thus for the Euro-African nationalists the creation of the Angolan Legislative Council was not enough and their grievances against Portuguese rule continued throughout the 1950s.30 Economically Salazar authorized some industries in Angola and Mozambique, which were controlled by the metropolitan capital. At the same time, the colonial products – especially coffee – registered a boom of prices and thousands of new settlers arrived to Northern Angola searching for a piece of land upon which to grow coffee. Salazar also started to support the establishment of new settlers in the rural areas of Central Angola and many of them were filled with racial prejudices against the local population – blacks as well as mulattoes and Angolan born whites. Thus “racial” tensions between Angolan born whites and the new settlers aggravated, while competition for land and labour between elements of the two groups became harder. In this respect, it should be underlined that the settlers main economic activities differed from region to region. In Eastern Angola, settlers were predominantly engaged in the internal trade with the black population. In the Huambo high plateau, and in the areas crossed by the Benguela railway, settlers were involved in farming, especially corn, sisal, fruit and sugar. In Northern Angola, settlers developed a coffee plantation economy based on the exploitation of cheap black labor.31

Galvão, Por Angola (Quatro anos de actividade parlamentar). Lisbon: Edição do Autor, 1949. 30 Ministério do Ultramar, Nova Legislação Ultramarina. Lisbon: AGU, 1953; Ministério do Ultramar, Estatuto da Província de Angola. Lisbon: AGU, 1955. PRO, FO 371/125894, Internal Political Situation in Angola, 1957. 31 The majority of black population only practised a subsistence economy, but it provided an important contribution to agricultural production, namely in corn and coffee. See Franz-Wilhelm Heimer, Social Change in Angola. Munchen: 1973. 188 fernando tavares pimenta

The fishing industry was important in the coast South of Benguela, mainly in Moçâmedes, but it also depended heavily on cheap black labour. Raising cattle was very important in the South, near Lubango. In Luanda and Lobito, which were the main ports of the country, there were small textile, alimentary and construction industries belonging to the white settlers. Nevertheless, and in spite of the growth of the Angolan economy after 1945, restrictive Portuguese economic legislation continued to condition the economic development of Angola and to limit the profits of the white settlers, especially the Euro-African whites. For example, the huge mineral resources of the colony – diamonds and, after the 1950s, oil – were directly controlled by the metropolitan and foreign capital. This was a source of latent tension between Lisbon and the white settlers in Angola.32 Meanwhile, as a consequence of the economic and demographic changes that took place in the colony after 1945, white society changed quickly and class became an important factor of differentiation among whites. Between 1945 and 1960 the bulk of the elite of the white settler community was centered in Luanda, where there was a strong import- export commercial class. Indeed, Luanda’s import-export elite, the big commercial and industrial businessmen – the patrões (“bosses”) – and the upper echelon of colonial administration formed the top of the Angolan social structure. They shared their social prestige – but not their economic power – together with physicians, lawyers, engineers, high school professors, other professionals and almost everyone who possessed a university diploma. Yet their hegemony was contested by the new wealthy “coffee barons” of the North-west, and by some very rich merchants, farmers and cattle owners of the Centre and South. However, the latter never succeeded in imposing their leadership. The middle class was formed by civil servants of minor (but important) level clerks, small city traders, the majority of farmers and the upper echelon of the working class, particularly some skilled railway and port workers. Normally, this middle class tended to integrate with the descendents of the once powerful nineteenth century mulatto and Europeanised black elite (now reduced to a civil servant and clerk middle class) or with the upper echelons of the group of new urban Europeanised black Angolans, known as assimilados. In the rural areas, especially in the high plateau, there was also a very large group of bush traders – the comerciantes do mato –, which formed a sort of rural middle class, even if in fact they were no more than petits blancs.

32 Fernando Tavares Pimenta, op. cit., 2005, pp. 41–62. angola’s euro-african nationalism 189

At the bottom of the white social hierarchy was the large group of poor settlers: unskilled workers (such as domestic and commercial servants), proletarian craftsmen (such as shoemakers), taxi drivers, ruined farmers, unemployed settlers and even beggars. The majority of this predominantly urban proletariat lived in the “best slums” of Luanda, Lobito and other urban centres, side by side with their black and mulatto neighbours (including some assimilados). In addition to this, many of the ruined farmers and other rural poor settlers lived in the government sponsored settlements of the high plateau (as for example Cela, in the Kwanza Sul district, not far from Huambo).33 How then did white settlers – especially the Euro-African nationalists – react to the political, economic and social changes that took place in Angola after 1945? There were two main reactions. On the one hand, a small part of the new generation of Angolan born whites engaged in anti-colonial politics, side by side with some mulattoes and Europeanised blacks. Almost all of them were left-wing intellectuals and some were in touch with the (clandestine) Portuguese Communist Party (PCP) and with nationalists from other Portuguese colonies – Cape Verde, Goa (India), Guinea Bissau, Mozambique and São Tomé. Many were also members of some cultural and academic associations, that were formed by Angolans both in the colony – the Angolan Cultural Society (Sociedade Cultural de Angola), the Angolan Natives Association (Associação dos Naturais de Angola) – and in the Motherland – the Angolan Students’ House (Casa dos Estudantes de Angola), the Empire Students’ House (Casa dos Estudantes do Império). In the protected political environment of these associations, Angola’s new white (and mulatto) generation collaborated for the first time with the Europeanised black younger generation, by essaying through literature an Angolan national identity. They created the idea of angolanidade or in other words “the idea that Angola had an individual cultural identity not only distinct and independent from the Portuguese one, but also free from any kind of racial, ethnic or religious prejudices”.34 Politically, they defended the independence of Angola under (black) majority rule and, in a certain sense, they were entirely African (and not Euro-African) nationalists. During the 1950s,

33 Isabel Castro Henriques, “A sociedade colonial em África. Ideologias, hierarquias, quotidianos”, in História da Expansão portuguesa (volume V, Dir. by Francisco Bettencourth & Kirti Chauduri). Navarra: Temas e Debates, 2000. 34 See Alfredo Margarido, Estudos sobre literaturas das nações africanas de língua portuguesa. Lisbon: A Regra do Jogo, 1980. 190 fernando tavares pimenta these left-wing whites organized themselves into several nationalists groups of Marxist inspiration, the most important of which was the Angolan National Liberation Movement (Movimento de Libertação Nacional de Angola, MNLA). The MNLA united whites, mulattoes and some Europeanised blacks of Luanda, including an engineer and a high school professor, and some prestigious writers and poets (such as Luandino Vieira) of the Sociedade Cultural de Angola. But, in 1959 the Portuguese political police (PIDE) crushed Luanda’s nationalist groups, namely the MLNA, while a number of African nationalists (whites, mulattoes and blacks) were arrested and/or deported to Portugal.35 Later on, some of these white nationalists supported the independence war which was conducted by the black guerrillas, especially the MPLA. On the other hand, Euro-African nationalists continued to feed the dream of a “new Brazil”. The supporters of this idea were concentrated in the district of Benguela, which included the cities of Benguela and Lobito, and in the remaining districts of Central and Southern Angola (Bié, Huambo, Huíla and Moçâmedes) where settlers had more roots to the land and there were many 2nd and 3rd generation Angolan born whites. As in 1945, Euro-African nationalists thought Angola could attain self government or even independence under white settler leadership if Salazar dictatorship was replaced by a democratic regime in Portugal. Thus, Euro- African nationalists sought a political alliance with the (non-communist) Portuguese Democratic Opposition and they provided massive political support for the Opposition candidate, General Humberto Delgado, in the Portuguese Presidential elections of 1958. The elections results were a fraud and the opposition candidate lost to the regime’s official candidate, Admiral Américo Tomaz. Nevertheless, Humberto Delgado obtained a significant result in Angola, especially in the Central and Southern districts, and he actually won in the Benguela district, which was the political bas- tion of Euro-African nationalists. Humberto Delgado had 66,7% of the ballot in the Benguela district while Admiral Américo Tomaz had only 33,3%. It was a defeat for Salazar without parallel in the Motherland and in the remaining Portuguese Empire. It was also a clear sign of the political strength of Euro-African nationalism in Central and Southern Angola.36

35 AN/TT, Arquivo PIDE/DGS, Movimento de Libertação Nacional de Angola (MNLA), Processo 3474/59, 2968; AN/TT, Arquivo PIDE/DGS, Casa dos Estudantes do Império, Processo 3529/62, 3352. Interviews with Alfredo Margarido, Fernando Falcão, Adolfo Maria, Fernando Martinho. 36 AN/TT, AOS/CO/PC – 51A, Pasta 8 – Relatórios sobre a campanha eleitoral na Guiné e em Angola (1958). PRO, FO 371/131637, Effect on Angola of Election Held in Portuguese angola’s euro-african nationalism 191

The leader of Benguela’s Euro-African nationalists was Fernando Falcão, an Angolan-born white (of 2nd generation), who was an engineer and an important businessman of Lobito. He was also an elected member of Lobito’s city council where he defended the economic interests and the political aspirations of the Euro-African nationalists. During the Portuguese Presidential elections of 1958, Fernando Falcão led with success the political campaign of the opposition candidate in Central and Southern Angola. The elections gave him the opportunity to create a political network involving some of the most influential businessmen and intellectuals of the white settler community (and also some mulattoes) under his leadership. In 1959 Fernando Falcão presented his own candi dature to the Angolan Legislative Council. But the Portuguese colonial authorities excluded him from the ballot because they feared a crushing victory for him. Fernando Falcão concentrated then in the rise of a political movement capable of unifying all the Euro-African nationalists in Angola. In the beginning of 1961, Fernando Falcão founded the Angolan United Front (Frente de Unidade Angolana, FUA) which was the most important party in the history of Euro-African nationalism. The FUA was led by a political commission composed by Fernando Falcão (who was the movement’s President) and several other Euro-African businessmen and intellectuals from Central and Southern Angola, including Socrates Dáskalos, who had been the leader of the OSA in 1940.37 The FUA was partially the response of Euro-African nationalists from Central and Southern Angola to the beginning of the war of independence initiated by the black armed movements – UPA/FNLA and MPLA – in Northern Angola in February/March 1961. These guerrillas had been created by some exiled blacks and mulattoes Angolans: the UPA had been formed in 1958 by a group of Angolans from Northern Angola in the Belgian Congo, while the MPLA had been formed in 1960 fundamentally by an intellectual elite of mulattoes and some blacks from Luanda and other urban areas who lived in Europe and North Africa. In 1961, after Portugal’s categorical refusal to give independence to Angola, UPA/FNLA and MPLA separately launched wars of independence, which would last

Africa on 8 June 1958, 1958; PRO, FO 371/131635, Internal Political Situation in Portuguese Africa, 1958. 37 AN/TT, AOS/CO/PC – 77, Pasta 70 (sbd.) – Criação em Angola da Frente de Unidade Angolana; AN/TT, AOS/CO/UL – 30D, Pasta 5 – Diversos, actividade da FUA; AN/TT, Arquivo PIDE/DGS, Frente de Unidade Angolana (FUA), Processo 515/61, 3059; AN/TT, Arquivo PIDE/DGS, Delegação de Angola, Fernando Gonçalves Magalhães Falcão, Processo 993, 1021 (1.° volume). 192 fernando tavares pimenta until 1974, when the Portuguese dictatorship was overthrown by a left wing revolutionary military movement, who agreed to negotiate Angola’s independence with the Angolan armed movements.38 The FUA tried to represent an intermediate political position between the Portuguese colonial dictatorship and the black armed movements, in the sense that it condemned the use of violence and demanded a peaceful resolution to the armed conflict, as well as Angola’s immediate political autonomy, which would prepare the country for independence. This independence would signify the recognition of the political participation of the mulatto and black population in the government of the country, but, at the same time, it would assure the political rights and the economic interests of the whites in a future independent Angolan State. The FUA was able to mobilize the majority of the white settlers, a substantial part of the mulattoes and some Europeanised blacks of Central and Southern Angola. It also seems the FUA had the sympathy of the Brazilian authorities, since some contacts were made between the FUA leadership and a delegate of the Brazilian Embassy at Lisbon.39 It failed, however, in mobilizing the white elites of Northern Angola, especially those in Luanda, who preferred to negotiate a compromise with the Portuguese dictatorship, through its new Colonial Minister, Adriano Moreira. Indeed, Luanda’s white elite was frightened by the massacres of settlers perpe- trated by the UPA in Northern Angola. This elite was also afraid of loosing its privileges and would not accept a black political leadership. The Portuguese government took advantage of the divisions among Angolan white settlers and launched a large scale repressive operation in the beginning of June 1961, arresting and deporting the leaders of the FUA to Portugal, namely Fernando Falcão and Sócrates Dáskalos.40 Yet, in 1962 some of the FUA’s members, who were deported to Portugal, were able to escape to Paris (France), where they re-organised the movement in exile under the leadership of Socrates Dáskalos. Later on, in 1963, the FUA’s exiled committee moved to Algiers (Algeria). The FUA

38 See John Marcum, The Angolan Revolution. I: The Anatomy of an Explosion (1950– 1962). Cambridge: MIT Press, 1969; John Marcum, The Angolan Revolution. II: Exile Politics and Guerrilla Warfare, 1962–1976. Cambridge: MIT Press, 1978. 39 However, by the end of May 1961 the Brazilian President Jânio Quadros was dealing with strong internal criticisms of his Third World external politic and little support could be offered by Brazilians to Angolan white nationalists. 40 AN/TT, AOS/CO/UL – 30D, Pasta 5 – Diversos, actividade da FUA; AN/TT, Arquivo PIDE/DGS, Frente de Unidade Angolana (FUA), Processo 515/61, 3059; AN/TT, Arquivo PIDE/DGS, Delegação de Angola, Fernando Gonçalves Magalhães Falcão, Processo 993, 1021 (1.° volume). angola’s euro-african nationalism 193 exiled committee published its own political statutes and programme, as well as a political journal named Kovaso; it also presented a petition to the United Nations General Assembly and wrote letters to several Heads of State in Europe and Africa denouncing the Portuguese colonial repression in Angola.41 The FUA presented itself as a non-racial nationalist movement, opened to the militancy of all the Angolans, but with a special attention on the problems and position of the whites in Angola. The FUA also changed its political strategy and accepted the use of violence to defeat colonialism. Therefore they tried to mobilize, politically speaking, the whites settlers to fight against the Portuguese colonial rule. The FUA tried to re-establish itself in Angola, through some contacts in Brazil. Fernando Falcão was able to return to Angola in December 1962 and several others members of the movement created some political clandestine groups in Central and Southern Angola. But in May 1963, the Portuguese political police arrested almost all FUA members of Huambo, Lobito and Benguela and the movement was crushed definitively.42 Meanwhile, the FUA had also tried to engage in talking with the Ango lan black armed movements, the MPLA and the UPA/FNLA. Its aim was to create a vast nationalist front involving all the Angolan anti-colonial parties and movements. But these movements rejected the FUA’s pro- posal. Actually, both the UPA/FNLA and the MPLA denied whites the right to fight for Angola’s independence and assumed hostile political positions in the eyes of FUA. Indeed the black guerrillas did not recognise the right of whites to Angolan citizenship, in the sense that they considered that the only true Angolans were the blacks and their mulattoe descendants. Thus, the UPA/FNLA and the MPLA considered that all the whites – including those born in Angola of 2nd and 3rd generation – were foreigners and that they should be excluded from the country after independence.43 This means that the black guerrillas had a racial

41 Frente de Unidade Angolana (FUA), Programa e Estatutos. Paris: FUA, 1962; Frente de Unidade Angolana (FUA), Comunicado à Imprensa: “O que é a FUA ou Frente para a Unidade Angolana?”. Paris: FUA, Setembro de 1962; United Nations General Assembly – Petition from Socrates Mendonça de Oliveira Daskalos, Chairman Frente de Unidade Angolana Concerning Angola, A/AC. 109/PET. 53, March 7th 1963. 42 Frente de Unidade Angolana (FUA), Carta Aberta aos Brancos de Angola. Paris: FUA, Agosto de 1962; Frente de Unidade Angolana (FUA), “A população branca no contexto nacional”, Kovaso. Órgão da FUA, n.° 1, Argel, Fevereiro de 1963; AN/TT, Arquivo PIDE/ DGS, Frente de Unidade Angolana (FUA), Processo 515/61, 3059. 43 Movimento Popular de Libertação de Angola (MPLA), Statuts et Programme. Mouvement Populaire De Libération De L’Angola. [S.l.]: MPLA, 1962; Movimento Popular de Libertação de Angola (MPLA), First National Conference of the Peoples’s Movement for the 194 fernando tavares pimenta understanding of the meaning of nation; for them nation and race were concepts politically equivalents and independence would mean the end of the white presence in Angola. As a consequence of its internal failure and of the political hostility by the black guerrillas, the FUA exiled committee dissolved itself in August 1963. Its disappearance signalled the end of Euro-African nationalist hopes, since there remained no group or party in the interior of Angola courageous enough to challenge the Portuguese colonial repressive dictatorship. Differently from Southern Rhodesia, the white settlers in Angola were never able to attain control over the Colonial State or self government. This was a direct consequence of the administrative centralization and of the political repression of the Portuguese colonial dictatorship of Salazar. But it must be said that this situation did not change with the overthrown of the dictatorship in Portugal (1974), in the sense that the Portuguese revolutionary government rejected the political participation of the white settlers’ representatives (such as Fernando Falcão) in the decolonisation process of Angola. Indeed, Lisbon accepted to negotiate Angola’s independence directly with the three black armed liberation movements (MPLA, FNLA and UNITA), but it excluded the Euro-African nationalists from the independence agreement. The white settlers had no choice but to support one of the three armed movements, or leave the country. This last option became almost compulsory when a civil war broke between the three movements and violent struggles damaged the major urban centres where the whites lived. As a result the great majority of the whites – approximately 300.000 people – hurriedly left the country throughout 1975. Only 30.000 (10% of) Angolan whites remained in Angola. By the end of 1975, Angola had ceased to be a settler society and Euro-African nationalism had ceased to exist.44 In this chapter I have demonstrated how the political identity of the white settlers in Angola evolved along the twentieth century. More precisely, I have demonstrated the emergence and evolution of a local form of nationalism developed by a part of the white settler community, the Euro-African nationalism, which conceived Angola as a “new Brazil”. However, the degree of political participation of the white settler in the

Liberation of Angola (MPLA). [S.l.]: MPLA, December 1962; Sócrates Dáskalos, Mukanda para Luandino Vieira. Argel: [s.p.], 4 de Junho de 1965. 44 The majority of the white exiles went to Portugal, but many established themselves in Brazil, South Africa and Southern Rhodesia. Their exodus was almost the repetition of the pied noir exodus from Algeria in 1962. See Fernando Tavares Pimenta, op. cit., 2008, pp. 342–427. angola’s euro-african nationalism 195

Table 1. Angola’s demographic evolution between 1900 and 1974.45 Year Whites Mulattoes Blacks Total 1900 9.000 0,4 7.000 0,3 2.700.000 99,3 2.716.000 1910 12.000 0,4 9.500 0,4 2.900.000 99,2 2.921.500 1920 20.700 0,6 10.500 0,4 3.100.000 99,0 3.130.200 1930 30.000 0,9 13.500 0,5 3.300.000 98,6 3.343.500 1940 44.083 1,2 28.035 0,8 3.665.892 98,0 3.738.010 1950 78.826 1,9 29.648 0,7 4.036.687 97,4 4.145.266 1960 172.529 3,6 53.392 1,1 4.604.362 95,3 4.830.449 1970 290.000 5,1 (?) – (?) – 5.673.046 1974 335.000 – (?) – (?) – (?) colonial administration was always low and their political movements had been continuously repressed by the Portuguese colonial dictatorship since the mid-1920s. Yet, the victory of the Allied Forces in World War II obliged Salazar to enact some changes in the Portuguese colonial policy and to make some political concessions to the white settlers, such as the participation of the opposition in the electoral processes and the creation of an Angolan Legislative Council. Thus the period subsequent to the war was marked by important transformations in the white settler society caused by the economic growth of the colony and by the demographic increase of the settler population. Despite these changes, Salazar’s colonial dictatorship continued to promote a sort of racial discrimination against Angolan born whites, who were racially classified as Euro-Africans and were considered second class citizens, such as the

45 Until 1930 see Walter Marques, Problemas do desenvolvimento económico de Angola. Luanda: Junta de Desenvolvimento Industrial, 1962, pp. 40–42; A. H. Oliveira Marques; Joel Serrão, Nova História da Expansão Portuguesa. O Império Africano, 1890–1930 (vol. XI). Lisbon: Estampa, 2000, p. 309. 1940: Direcção dos Serviços de Economia – Repartição de Estatística Geral da Colónia de Angola, Censo Geral da população 1940. Constituição das famílias. Fecundidade e sobrevivência. Casais segundo a permanência na colónia. Quadros resumos (Volume XI e XII). Luanda: Imprensa Nacional, 1947, pp. 185–218. 1950: Direcção dos Serviços de Economia e Estatística Geral da Província de Angola (Portugal) – Recenseamento geral da população, 1950 (volumes IV e V). Luanda: Imprensa Nacional, 1955–1956, pp. 84–116 e 173. 1960: Direcção dos Serviços de Economia e Estatística Geral da Província de Angola (Portugal), op. cit. (1° volume), 1964, p. 16. 1970 and 1974: Gerald Bender e P. Santey Yoder, “Whites in Angola on the Eve of Independence: the Politics of Numbers”, Africa Today, 21 (Fall 1974), p. 126; Franz-Wilhelm Heimer, op. cit., 1980, p. 106. See also: Instituto Nacional de Estatística, Delegação de Angola, Anuário Estatístico. Portugal, Estado de Angola. Luanda: 1972; Estado de Angola, Informações Estatísticas, 1973. Luanda: 1974. 196 fernando tavares pimenta

Table 2. Demographic evolution of Africa’s white settler colonies (1920– 1940). 1920 – Population46 1940 – Population47 Country White Settlers Total White Settlers Total Population Population South Africa 1.521.000 21,9 6.926.000 2.732.000 23,2 11.775.000 Algeria 791.370 13,8 5.714.556 946.013 13,2 7.147.457 Zimbabwe 33.620 3,8 884.736 68.954 4,7 1.467.106 Angola 20.700 0,6 3.130.200 44.083 1,2 3.737.947 Mozambique 11.000 0,4 3.120.000 27.438 0,5 5.086.000 Kenya 9.700 0,2 3.835.000 22.800 0,5 4.884.000 Zambia 8.765 0,5 1.753.000 13.000 0,6 2.099.000 mulattoes. These racial tensions were aggravated by the establishment of thousands of new settlers in Angola, who competed, for land and labour, with the Euro-Africans. As a consequence, the Euro-African nationalism became politically stronger, especially among settlers of Central and Southern Angola. But the Euro-African nationalists were incapable of taking control of the Colonial State; their movement (the Angolan United Front, FUA) was crushed by the Portuguese political police and their

46 South Africa (data of 1921): Christopher Saunders et al., Historical Dictionary of South Africa. London: The Scarecrow Press, 2000, p. xxxv. Angola: Walter Marques, Problemas do desenvolvimento económico de Angola. Luanda: Junta de Desenvolvimento Industrial, 1962, pp. 40–42. Algeria (data of 1921): Brune Étienne, Les européens d’Algérie et l’indépendance algérienne. Paris: Éditions du Centre National de la Recherce Scientifique, 1968, p. 18. Mozambique: (data of 1918) Gervase Clarence-Smith, The Third Portuguese Empire, 1825– 1975. A Study in Economic Imperialism. Manchester: Manchester University Press, 1985, p. 134. Kenya (data of 1921) Alison Smith, “The immigrant communities (1): the Europeans”, in History of East Africa (Dir. D. A. Low; Alison Smith – Vol. III). Oxford: Clarendon Press, 1976, p. 576. Zambia: Donald George Morrison et al., Black Africa: a Comparative Handbook. New York: The Free Press, 1972, pp. 14 e 483. Zimbabwe (data of 1921): Rita Cruise O’Brien, “White society in Africa”, Tarikh, vol. 6, n.° 2. Lagos: University of Lagos, 1977, p. 20. The black population of Zambia and Zimbabwe were consistently under-evaluated until independence was achieved. 47 South Africa (data of 1946): Christopher Saunders et al., op. cit., 2000, p. xxxv. Angola: 1.° Recenseamento da População de Angola, 1940. Algeria (data of 1936): Denise Bouche, Histoire de la colonisation française. Flux et reflux (1815–1962) (Tome 2). [Sl.]: Fayard, 1991, p. 337. Moçambique: 1.° Recenseamento da População de Moçambique, 1940. Kenya (data of 1941): Alison Smith, op. cit., 1976, p. 576. Zambia (data of 1939): Shula Marks – “Southern Africa”, The Oxford History of the British Empire. The Twentieth Century (Dir. Judith M. Brown; WM. Roger Louis). Oxford: Oxford University Press, 1999, p. 553. Zimbabwe (data of 1941): Rita Cruise O’Brien, op. cit., 1977, p. 20. angola’s euro-african nationalism 197

Table 3. Demographic evolution of Africa’s white settler colonies (1960– 1970). 1960 – Population48 1970 – Population49 Country White Settlers Total White Settlers Total Population Population South Africa 3.008.000 18,8 16.002.000 3.773.000 17,3 21.794.000 Algeria 1.050.000 9,7 10.850.000 – – – Zimbabwe 221.500 5,8 3.790.000 271.000 4,5 5.971.000 Angola 172.529 3,6 4.830.449 290.000 5,1 5.673.046 Mozambique 97.245 1,5 6.578.569 200.000 2,4 8.234.000 Kenya 53.000 0,6 8.833.000 – – – Zambia 75.000 3,3 2.200.000 – – –

leaders were arrested and deported to Portugal. The black guerrillas also refused the participation of whites in the war of independence and they did not recognize their right to Angolan citizenship. Nevertheless, more important than the success or failure of the Euro-African political protest, the analysis of the political behaviour of Angolan whites indicates that settlers were much more than simply agents of European colonialism and that they actually created a local form of nationalism, which opposed the continuation of European colonial rule – an aspect which might be important for the understanding of the white settler political identity in colonial Africa.

48 South Africa: Christopher Saunders et al., op. cit., 2000, p. xxxv. Angola: 3° Recenseamento da População de Angola, 1960. Algeria: Bruno Étienne, op. cit., 1964, p. 18. Mozambique: 3.° Recenseamento da População de Moçambique, 1960. Kenya (data of 1961): Alison Smith, op. cit., 1976, pp. 576–577. Zambia: D. Abshire; M. Samuels, Portuguese Africa: a Handbook. New York: Praeger, 1969, pp. 205–206. Zimbabwe (data of 1961): Larry Bowman, Politics in Rhodesia: white power in an African state. Massachusetts: Harvard University Press, 1973, p. 13. 49 In 1962 Algeria achieved independence under non-European majority rule. The majority of settlers left the country before or soon after independence. In 1964 Kenya and Zambia achieved independence under non-European majority rule and many settlers also gradually left the countries. South Africa: Christopher Saunders et al., op. cit., 2000, p. xxxv; Zimbabwe (data of 1973): Patrick O’Meara, Rhodesia. Racial Conflict or Coexistence?. London: Cornell University Press [s.d.], p. 1; Angola: Gerald Bender and P. Santey Yoder, “Whites in Angola on the eve of independence: the politics of numbers”, Africa Today, 21 (Fall 1974), p. 126; Mozambique (data of 1974 for the white population): Malyn Newitt, A History of Mozambique. London: C. Hurst, 1995, p. 476.

CHAPTER EIGHT

CHANGING NATIONALISMS: FROM WAR TO PEACE IN ANGOLA

Justin Pearce

Angola is not unique in having witnessed the rise of several different conceptions of national identity within its borders in the period before independence, nor is it unique in that these different conceptions were associated with rival political movements. Angola is unusual, however, in that no one strand of nationalism became hegemonic in the decades following independence. This fact is inseparable from the reality that control of the Angolan population was divided between the MPLA state and the Unita rebels from 1975 until 2002. In this paper I will argue that the 27 years of near-continuous war that followed Angolan independence served to entrench the contradictions between different strands of nationalism, as separate ideas about the nature and character of the Angolan nation were propagated in the zones controlled respectively by the MPLA and Unita. The military victory by the MPLA government in 2002, which has since been consolidated by the MPLA’s overwhelming success in the 2008 parliamentary elections, provided an opportunity for the government to reinforce the position of its own version of nationalism. At the same time Unita, still negotiating its role as a political player following its military defeat, has stopped asserting its claims to an alternative nationalism. Debates around Angolan national identity nevertheless remain alive, albeit not in the arena of formal politics, and include more or less explicit challenges to the MPLA’s claims to exclusive nationalist legitimacy. In considering the origins and the development of different nation alisms in Angola, we need to bear in mind the characteristics that nationalism had already assumed elsewhere in Africa at a time when anti- colonial activity in Angola had barely begun. The first of these is the fact that African nationalism was, and still remains, bound up with ideas of anti-colonial liberation. For years after independence, the memorialisa- tion of independence struggles has continued to be an important political tool for governments to assert their nationalist legitimacy. Second, in the African context, nationalism almost always refers to an ideology that asserts the primacy of identities based on a state that has taken its borders

© koninklijke brill nv, leiden, �� | doi ��.��/��_�� 200 justin pearce from a former colony. In independent Africa, we might further see nationalism as an ideology that puts a high value on national unity, and which may be suspicious of ethnic or regional claims to identity, or which might seek to accommodate these claims within a broader definition of nationalism. Nationalist movements present themselves as the guardians of the unity of a nation against forces – be they real or imagined – that would seek to divide it. The strong association in Africa between nationalism and anti-colonial liberation has given nationalism a normative character that is hard to attack. Moreover, when we do consider challenges to nationalism, we need to be mindful of whether what is going on is indeed the promulgation of ideologies other than nationalism, or rather a challenge to the nationalist claims of the incumbent government: not attacking nationalism, but continuing to assert the positive value of nationalism while undermining the claims of others to be nationalist, by accusing supposedly nationalist incumbents of being in fact the representatives of frac- tional interests. I would suggest that the latter response is the more useful reference point when we seek to understand the political role of nationalism in Angola throughout the civil war, and still today. We need to look at different claims to nationalism – the different imaginings of the community called the nation, as Benedict Anderson would have it1 – and to what extent these nationalisms assert their exclusivity, or admit the possibility of a more plural, accommodating kind of nationalism. Elsewhere in this book, David Birmingham notes that today, none of the three former liberation movements in Angola seems to embody the concept of nationalism, at least in the public consciousness: “Each was associated with the greed of a small élite”.2 In the introduction to another recent volume on Angola, Patrick Chabal notes that nationalism in Angola today is less important than the realities of incumbency: The MPLA benefits today from the considerable advantages of having held power since independence, rather than from any legitimacy attached to its success in having defeated its anti-colonial rivals.3 In a chapter in the same volume, Christine Messiant suggests that there exists in Angola a new sort of nationalism centred on the MPLA:

1 B. Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism, London: Verso, 1991. 2 See Birmingham in this volume, p. 228. 3 P. Chabal, “Introduction: E Pluribus Unum: Transitions in Angola” in P. Chabal and N. Vidal (eds), Angola: The Weight of History, London: Hurst, 2007 p. 7. changing nationalisms: from war to peace in angola 201

[The] ideological vacuum is filled with talk of ‘sovereignty regained’, a nationalist discourse that is primarily geared to reject outside interference with the current political ‘transition’. The emphasis on unity and reconciliation … is … meant to call all groups to rally behind the banner of the MPLA, still today conceived as synonymous with the Angolan nation.4 It is difficult to take issue with Birmingham’s scepticism about the nationalist credentials of the former liberation movements, and likewise with Chabal’s assertion that power, and the associated ability to dispense patronage, are more important than nationalism (or indeed than any other ideology) in maintaining hegemony. In a country which with the outbreak of civil war in 1975 was denied the post-independence euphoria that much of Africa enjoyed in the 1960s, and where many people over the age of fifty speak of “o tempo colonial” with some nostalgia, narratives of the independence struggle would not appear to be the most effective rallying-call. Yet, as I shall discuss later, this has not prevented the MPLA from continuing to invoke the independence struggle in support of its own legitimacy. Even though it is questionable to what extent we can see nationalism as a programme of action in Angola today – or, indeed, at any point in the country’s history – the fact remains that nationalism as an idea has been prominent in Angolan political discourses from before independence and throughout the civil war. Most of the historical literature on Angola from the mid-20th century onwards identifies three main strands of nationalism, which gave rise to the three nationalist movements that sought liberation from Portuguese colonialism.5 Briefly: the MPLA and FNLA both emerged in the early 1960s. The MPLA leadership came from the assimilado and mestiço populations of the coastal cities, including a strong element among Angolan students in Portugal. The FNLA organised among Bakongo exiles in the then Zaire. These two movements vied for the support of independent African states for recognition as the legitimate

4 C. Messiant “The Mutation of Hegemonic Domination” in P. Chabal and N. Vidal (eds), Angola: The Weight of History, London: Hurst, 2007 p. 7. 5 F. Heimer, The Decolonization Conflict in Angola 1974–1976: An Essay in Political Sociology, Geneva: Institut Universitaire de Hautes Etudes Internationales, 1979; F. Guimarães, The Origins of the Angolan Civil War: Foreign Intervention and Domestic Political Conflict, Basingstoke: Macmillan, 2001; D. Birmingham, “Angola” in P. Chabal et al, A History of Postcolonial Lusophone Africa, London: Hurst, 2002; P. Chabal (2002), “The limits of nationhood” in P. Chabal et al, A History of Postcolonial Lusophone Africa; C. Messiant, 1961: L’Angola colonial, histoire et société. Les prémisses du mouvement nationaliste, Basel: P.Schlettwein, 2006. N. Pestana, L’état en Angola: Discours et practiques, PhD thesis, Paris, 2006; also Péclard in this volume. 202 justin pearce representative of the Angolan people. Later, Jonas Savimbi broke away from the FNLA, taking with him his constituents, whose origins were in the Central Highlands, to form Unita. The suppression of nationalist activity by the colonial authorities in Angola ruled out whatever possibility of national consensus may otherwise have existed. Nevertheless, the late decolonisation of the Portuguese territories meant that by the time independence was in sight, the principle of independent states retaining their colonial boundaries had become a reality across most of Africa and was no longer a matter for debate. Rather than looking for partition, all three anti-colonial movements laid claim to the mantle of Angolan nationhood. The precipitous departure of the colonial authorities, who bestowed independence on “the Angolan people” at a time when there was no consensus of who the representatives of the Angolan people were, created the conditions for the civil war that lasted until 2002.6 The FNLA ceased to be a significant player following its military defeat on the outskirts of Luanda in November 1975. The rest of this paper will look at the rival nationalist claims of the MPLA and Unita.

Rival Nationalist Claims

The MPLA adopted a modernising discourse which sought to erase ethnic and regional difference and which positioned the MPLA as the sole representative of the Angolan people. The principal theme in Unita’s political discourse was that the Ovimbundu people, and blacks and southerners more generally, had suffered the worst of colonial rule and that the MPLA, representing specific regional and class interests, was in no position to represent Angola at large.7 Some writers have sought intrinsic differences in the content of the MPLA’s and Unita’s nationalism. René Pélissier, for instance, characterises the MPLA as a “modernist” movement in contrast to an “ethno-nationalist” UNITA, while Anthony Pereira contrasts Unita’s “blood and soil” nationalism with the MPLA’s notion of citizenship based on rights and responsibilities.8 Yet the distinction is not as clear as this.

6 Birmingham, “Angola”. 7 See F. Bridgland, Jonas Savimbi: A Key to Africa, London: Hodder & Stoughton, 1986, pp. 67–68. 8 R. Pélissier, “Part two” in D. Wheeler and R. Pélissier, Angola, London: Pall Mall Press, 1971, pp. 161–166; A. Pereira, “The Neglected Tragedy: the Return to War in Angola 1992–3”, Journal of Modern African Studies, Vol. 32 (1), March 1994, pp. 1–28. changing nationalisms: from war to peace in angola 203

Reading Agostinho Neto’s poetry suggests that the MPLA’s ideologues were not averse to the kind of “blood and soil” appeals more commonly associated with Savimbi,9 while Unita, like the MPLA, sought legitimacy in the eyes of its followers by creating notions of rights and responsibilities: Unita at least tried to offer education and health services in the peasant farming areas that it controlled, in return for extracting taxes in the form of food for its soldiers.10 Even if Unita never enjoyed the command of an internationally-recognised state, its efforts to establish legitimacy in the eyes of its adherents had something in common with the workings of a state, most remarkably so during the existence of Unita’s “capital” at Jamba in Cuando Cubango province throughout the 1980s.11 If the presence of the state is what puts the nation in a position of primacy among “imagined communities”,12 then any intrinsic differences in the content of the MPLA’s and Unita’s nationalisms was less important than the fact that both movements made nationalist appeals backed up by the legitimacy of statehood on the part of the MPLA, and of a simulacrum of statehood on the part of Unita. What created the conditions for civil war in Angola, then, was not so much the differing content of the rival nationalisms as the mutual mistrust between the rival leaders, while material support from the Cold War adversaries allowed this mistrust to escalate into armed conflict. The division of the country into separate zones of control in turn reinforced separate concepts of nationalism, particularly since each party sought legitimacy by positioning itself as the defender of Angolan interests, and,

9 A. Neto, Sagrada Esperança, São Paulo: Editora Ática, 1985. 10 Author’s interviews with Unita members in Huambo province, 2008. Unita’s relationship with the people it controlled during the civil war is discussed further in J. Pearce, “L’Unita à la recherche de « son peuple ». Carnets d’une « non-campagne » sur le planalto”, Politique Africaine, No.110, June 2008, pp. 47–64. 11 My thinking here is influenced by theorists who have suggested that the idea of the state, and the languages through which the state (or quasi-state) is represented may be more significant than the material existence of the state. T. Mitchell “Society, Economy and the State Effect” in G. Steinmetz (ed.), State/culture: state-formation after the cultural turn, Ithaca: Cornell University Press, 1999 writes of “the state effect”. T. Hansen and F. Steputtat, States of Imagination: Ethnographic Explorations of the Postcolonial State, Durham and London, Duke, 2001 use the concept of “stateness”. See also V. Das and D. Poole, Anthropology and the Margins of the State, Oxford: James Currey, 2004, and A. Gupta “Blurred Boundaries: The Discourse of Corruption, the Culture of Politics, and the Imagined State”, American Ethnologist, Vol. 22 (2), May 1995, pp. 375–402. 12 Anderson, Imagined Communities, p. 7 suggests that though the nation is only one among several “imagined communities” to which we may belong, the nation nevertheless, historically, occupies a special status among the other possible imaginings since “nations dream of being free [and] [t]he gage and emblem of this freedom is the sovereign state”. 204 justin pearce through political education, portraying the other side as the puppet of foreign invaders. MPLA historiography claims the prison break of 4 February 1961 as “the day on which the Angolan people, under the leadership of the MPLA, took the initiative in rising against Portuguese rule, opened up a strategic challenge to the colonial system”.13 Critical histori- ans have seen these events rather as an initiative by “ill-equipped hot- heads” that was claimed only retrospectively by the MPLA.14 Accounts favouring the MPLA tend to emphasise Unita’s collaboration with the colonial authorities against the MPLA during one phase of the independence war, and to position Unita as essentially a creation of the United States and South Africa.15 Unita, for its part, was never as assertive as the MPLA in claiming the status of sole liberation movement. Savimbi staked Unita’s claim to legitimacy on the grounds of being based in the Angolan countryside since the days of the anti-colonial struggle, in contrast to the MPLA and FNLA whose leadership remained based outside the country for much of the duration of the independence struggle. Yet Savimbi, in writings published in 1979, appears to acknowledge Unita as only one among three liberation movements: It was this capacity for resistance, which the enemies ought to have the hon- esty to recognise publicly, which gave credibility to [Unita’s] participation in the decolonisation process, on the basis of equality with the MPLA and the FNLA, and often in a leading position.16 But even if Unita accepted the MPLA as a nationalist movement, Unita’s political narratives nevertheless portray the MPLA as the creation of a coastal elite who were the heirs, both genetically and culturally, of the Portuguese colonisers, and whose roots in Africa were questionable. For Unita, the presence of the Cuban soldiers who provided the MPLA with its military support constituted a foreign invasion, against which the duty

13 M. Andrade and M. Ollivier, The war in Angola: a socio-economic study, Dar-es- Salaam: Tanzania Publishing House, 1975, p. 119. 14 D. Birmingham, Angola, p. 36. See also Guimarães, Origins, pp. 44–45. 15 See W. Minter, Operation Timber: Pages from the Savimbi Dossier. Trenton NJ: Africa World Press, 1988; W. Minter, Apartheid’s contras: an inquiry into the roots of war in Angola and Mozambique. London: Zed, 1994; G. Wright, The Destruction of a Nation: United States Policy towards Angola since 1945, London: Pluto Press, 1997. MPLA supporters whom I interviewed frequently emphasised Unita’s links with the colonial regime and with the South African and United States governments as evidence of what they saw as Unita’s anti- nationalist character. 16 J. Savimbi, A resistência em busca de uma nova nação, Lisbon: Edição da Agência Portuguesa de Revistas, 1979. Online at: http://petrinus.com.sapo.pt/consequencias.htm changing nationalisms: from war to peace in angola 205 fell upon Unita to defend the nation, and which provided further weight to Unita’s argument that the MPLA was not indigenously Angolan.17

Elections and Return to War

The cessation of hostilities that preceded the 1992 elections was not enough to dissuade each of the warring partners of its exclusive ownership of Angolan nationhood.18 The internationally brokered plan that led to those elections assumed the possibility of a national consensus which, in reality, was lacking. The MPLA campaigned on a nationalist platform, its slogan “Angola na coração” (Angola in the heart) perhaps a rejoinder to a Unita whose claims to authenticity were linked to being rooted in the heart of Angola. Unita’s campaign was less obviously nationalist: Linda Heywood argues that Unita orientated its election message not in terms of an encompassing vision of Angolan nationhood, but rather in terms of offering a rural Ovimbundu constituency a chance to share in the power and resources of the state, and alienated many voters in the process.19 Around the same time, a notorious episode revealed how illusory was the doctrine of non-tribalism espoused by the MPLA: the mass killing of Ovimbundu people in Luanda, apparently by the government’s civil defence units, at the end of October 1992. Nevertheless, in public the government continued to speak the language of inclusion. Its position was strengthened by its election victory, which strengthened its claims to legitimacy both at home and abroad as the democratic representative of

17 Those former Unita adherents whom I interviewed in 2002 and 2008 almost invari- ably cited the Cuban presence in Angola among their reasons for fighting with Unita “to defend my country”. This appears to have been an important theme in the political education received by Unita members during the years of bush warfare. 18 Messiant, “Mutation”, p. 100 characterises the Bicesse peace agreement that set the terms for the 1992 elections as a “winner-takes-all” model that inevitably would fail to meet the aspirations of the party that lost the elections. 19 L. Heywood, Contested power in Angola: 1840s to the Present. Rochester: University of Rochester Press, 2002, p. 218: “Although both parties spread party and ethnic propaganda, in the rural areas of the central highlands UNITA’s candidates presented the campaign as a contest between an MPLA party dominated by a corrupt urban Afro-Portuguese elite and UNITA, a genuinely rurally-based, African party which they argued represented the only hope for Africans. In densely populated Ovimbundu centers, UNITA’s candidates increasingly geared their message to and ethnic Ovimbundu electorate. … Although this strategy strengthened UNITA’s support among committed Ovimbundu, it alienated the non-Ovimbundu, especially urban Afro-Portuguese who were alarmed by the racial rhetoric that came through.” Ibid, p. 224: “[Unita] believed that opposition status deprived them of the power over the state that they had come to believe was their right.” 206 justin pearce the Angolan people. At least at the international level, this was extremely successful. The Clinton administration became the first US government to recognise the MPLA’s authority, and campaigns for economic sanc- tions against Unita became increasingly effective throughout the 1990s.20 This new-found use of democracy to bolster nationalist claims would not have been sustainable, however, if the MPLA’s discourse had sought to delegitimise the other political parties that had emerged at the time of the elections. Hence a new national narrative emerged from the side of the government, one which allowed political pluralism but which cast Savimbi as a terrorist with no political agenda. The smaller political parties posed not the slightest threat to the MPLA, serving only as a useful reminder that the days of one-party rule were officially over. To complete the illusion of inclusiveness, the government funded the establishment of a breakaway Unita faction, Unita-Renovada. The plan was to isolate Savimbi: to write him out of the new democratic nationalist narrative, and thus to delegitimate his claims to national leadership.21 Unita, for its part, continued to make nationalist claims during the war years, though with diminishing confidence as the war progressed and the movement’s military position weakened. On the defensive, Unita seems to have realised that the best it could hope for was equal status with the government in peace talks. Recalling the period in 1998 when Unita was driven from its last urban strongholds on the central plateau, the Unita official Alicides Sakala writes: In effect, the various cyclical conflicts that Angola has known since independence had their roots in the profound contradictions that emerged during the struggle between the three national liberation movements, and others. A reflection of the strong contradictions in the heart of the classes of

20 See C. Messiant, “Why did Bicesse and Lusaka fail? A critical analysis.” Accord, No.15, pp. 16–23. Online at: http://www.c-r.org/our-work/accord/angola/bicesse-lusaka .php 2004. 21 The Angolan government statement to the UN Security Council on 15 August 2001, following an attack on a train at Zenza do Itombe in which 200 people were killed, emphasised the non-political nature of the attack, calling it “an act devoid of any military objective whose aim was simply to spread death, suffering and destruction of our people in the most atrocious way”. Significantly, this statement makes no mention of UNITA, only of Savimbi. See http://www.reliefweb.int/rw/rwb.nsf/db900SID/ACOS-64CMB2?Open Document&cc=ago&emid=ACOS-635NGV (accessed 10 June 2011). Later, the word “terrorism” became prominent in government statements on UNITA, apparently in an appeal to solidarity with the US after the events of 11 September. See “Letter from Angola to the UN President of the Security Council S/2001/958, 10 October 2001” online at: http://www .reliefweb.int/rw/RWB.NSF/db900SID/ACOS-64DFH4?OpenDocument&rc=1&cc =agoaccessed 10 June 2011. changing nationalisms: from war to peace in angola 207

Angolan society, exacerbated by colonialism, and springing essentially from ethnic, racial and regional factors, this further enlivened the ideological differences between Angola’s nationalist movements, the MPLA, the FNLA and Unita.22 Sakala, while making no exclusive claims for Unita as the inheritor of Angolan nationhood, nevertheless asserts that innate differences lie at the heart of the Angolan conflict. By 2001, Savimbi himself was calling for nothing more than talks with the government on equal terms: terms which the government, having gained the military upper hand, was in no mood to grant.23

The 2002 Peace Process

Once Savimbi was dead, the ideology of nationhood that the MPLA had established in wartime set the scene for a depoliticised peace process. The peace accord that followed Savimbi’s death was constructed as a military, not as a political affair. The Unita and government military commanders signed the deal establishing the modalities for the cessation of hostilities in Luanda on 4 April 2002, with President dos Santos presiding over the occasion as though he were a mediator rather than a party to the conflict. The chief of the armed forces, Armando da Cruz Neto, hailed President Dos Santos as “the architect of peace”, an epithet that was repeated over and over again in the state media.24 Central to media representations of the immediate post-war period was a scheme called “ponto de encontro” (meeting point) whereby people who had become separated from relatives during the war were encouraged to come to designated public places in the main cities, and either be lucky enough to run into their loved ones on the spot, or enter their names in a register and record announcements that were broadcast on radio and television. It was billed as “o reencontro da grande família angolana” (the reunion of the big Angolan family). While most of the tracing work was done behind the scenes by the Red Cross, the government made much of

22 A. Sakala, Memórias de um guerrilheiro: Os últimos anos de guerra em Angola, Lisbon: Dom Quixote, 2006. 23 “Angolan rebels submit ‘peace plan’”. BBC News, 24 August 2001. http://news.bbc .co.uk/2/hi/africa/1507869.stm accessed 10 June 2011. 24 The phrase “architect of peace” originally formed part of the MPLA’s 1992 election campaign, and was revived in 2002. I am grateful to Rafael Marques for pointing out this detail to me. 208 justin pearce the publicity value of these highly emotional events. The meeting point in Luanda was at the Primeiro de Maio square on the edge of central Luanda, beneath a newly restored statue of Agostinho Neto. At the base of the statue were mosaics with the iconic images of anti-colonial struggle – a muscular African man tearing apart the chains of oppression, for example. One panel bore the words of Neto’s most famous poem, “Havemos de voltar” (“We will return”).25 When it was written, the poem appeared to express the longing of exiles to return home, and to assert indigenous ownership of the land. But now it seemed to take on a new resonance: A state that had been confined to a few urban enclaves could now reassert its claim to the whole of Angola’s national territory. The MPLA was in this way able to portray itself in a nation-building role, rather than as a party to the conflict. The new national discourse was about an inclusive, but depoliticised sense of nationhood, and was silent on the issues of identity – region, class and race – which had underlain the emergence of rival nationalist movements some forty years earlier. As Messiant notes, this new discourse with its emphasis on the government and/or the MPLA as the creators of peace may be “primarily geared to reject outside interference with the current political ‘transition’”.26 In this respect, however, it is interesting to note the continuities with the MPLA’s wartime discourses which, as noted earlier, sought to delegitimise its main nationalist rival, Unita, by emphasising Unita’s foreign links. In the context of the “new” post-war nationalism, the twentieth anniversary of the Battle of Cuito Cuanavale, in 2008, provided an opportunity for the MPLA to reassert its claim as the defender of the Angolan nation, by reviving memories of a time when it was at war not only with South Africa, but with Unita. As the weekly paper Semanário Angolense noted: “The battle of Cuito Cuanavale took place 20 years ago, but the battle for Cuito Cuanavale looks as though it’s still to come”.27 The symbolism of the commemorative events betrayed a confusion – usefully so, from the MPLA’s perspective – between the roles of party and state. MPLA flags were prominently displayed, while President dos Santos declared: The battle, won by Angolan forces, gave rise to profound changes in Southern Africa, namely the application of UN Security Council Resolution

25 Neto, Sagrada Esperança. 26 Messiant, “Mutation”, p. 120. 27 Semanário Angolense, year 4, edition 256, 15 March 2008. changing nationalisms: from war to peace in angola 209

435/78, thus opening new perspectives for the fall of the apartheid regime in South Africa and the independence of Namibia.28 Dos Santos’s emphasis on the Angolan forces both cast the MPLA’s victory as a victory for Angola, and played down the decisive role of Cuban forces in the outcome of the battle. During the parliamentary election campaign later that same year, the MPLA appeared sufficiently confident in its claim to be Angola’s sole liberation movement that it could evoke its role in the independence struggle in its efforts to win votes. At an MPLA rally in Uige on 1 September 2008, President dos Santos declared: “It was the MPLA who said in 1956 that we must struggle against colonialism”. “Who won independence?” the president asked rhetorically. “The MPLA!” roared the crowd in response.29 More recently, President dos Santos has again returned to narratives of national liberation in an attempt to head off new challenges in the form of popular protest inspired by events in north Africa. Addressing the MPLA’s Central Committee in April 2011, Dos Santos contrasted what he portrayed as the authentically indigenous movements against colonial oppression with the criticism, muted as it may be, of today’s government. When we were young, in the time of colonialism, we knew that the struggle for people’s emancipation was conducted through trade union movements, political parties or national liberation movements […] By contrast, according to Dos Santos, today’s critics not only lack popular support – they are in fact part of a neo-colonial conspiracy. Dos Santos alluded to plans to put puppets in power, who would obey the will of foreign powers who want to return to plunder our riches and send us back to the misery from which we have liberated ourselves through much sacrifice.30

28 Quoted in Semanário Angolense, year 4, edition 256, 15 March 2008. The article also quoted Unita’s secretary general, Kamalata Numa, as taking exception to Dos Santos’s version of events. 29 Broadcast on Televisão Pública de Angola, “Jornal de Noite”, 2 September 2008. 30 Angop, 15 April 2011. “Discurso do Presidente José Eduardo dos Santos na reunião do Comité Central”, online at: http://www.portalangop.co.ao/motix/pt_pt/noticias/politica/ 2011/3/15/Discurso-Presidente-Jose-Eduardo-dos-Santos-reuniao-Comite -Central,e545cbc5-1a17-43e7-a00a-f7f3d1f3dbd4.html accessed 24 May 2011. 210 justin pearce

Unita and Nationalism Today

But if ideas of national freedom are once again being politicised on the side of the MPLA, there is little evidence that Unita is managing to turn ideas of emancipation to its own advantage. As it seeks to redefine its political role after the death of Jonas Savimbi and the relocation of its current leadership to Luanda, Unita has concentrated more on issues of material inequality than intrinsic identity: its public discourse is not explicitly about north and south, or mestiço and black, but about rich and poor. Does Unita still assert a history as a liberation movement? The answer to this question will vary depending on whether one is talking about Unita’s leadership, or its rank and file members. Many of the former Unita followers I interviewed in 2008 justify their continued adherence to Unita on the basis that they had originally joined the movement in order to defend Angola against a Cuban invasion, and to defend themselves against an MPLA leadership that did not have the interests of southern Angola at heart. From the side of the leadership, however, the claims to nationalist authenticity have been muted since the end of the war. In 2003, a year after Savimbi’s death, the Unita member of parliament Jaka Jamba spoke at a conference in Huambo on the subject of Angolan identity.31 Jamba, admittedly, was not speaking in his Unita capacity on this occasion, and the word “Unita” did not appear once in his address. The narrative that he presented can be summarised as follows: In colonial times, Angolan identities were suppressed by the imposition of “portugalidade” (Portuguese-ness) and assimilation. During the liberation struggle, the guerrillas – whom Jamba does not name – adopted Mao’s “fish in water” approach. The anti-colonial struggle “presented to the freedom fighters the challenge of launching the bases of building a new nation from cultural, ethnic, linguistic, racial and religious diversity”. Jamba contrasts this with the period of one-party rule when “political and philosophical life was dominated by Marxist-Leninist ideology, the objective being the emergence of the ‘new man’”. Again, Jamba does not name the Marxist- Leninist party to which he refers – but we do not need to know much about Jamba’s own politics in order to be able to read between the lines. The unnamed guerrillas can only be Unita since they are set up in opposition both to colonial rule and to the one-party system that succeeded it.

31 Conference paper delivered in Huambo, September 2003. changing nationalisms: from war to peace in angola 211

Jamba here asserts a nationalist history and a liberal democratic history for Unita without mentioning it by name. Unita has tried to reclaim the history of anti-colonial struggle from the MPLA, but this has been more a matter of denying the MPLA’s claims to nationalist exclusivity than of asserting Unita’s claims to have helped liberate Angola from colonialism. In his address to mark Independence Day in 2005, Unita’s leader Isaias Samakuva declared: 11 November does not indicate the victory of one or other party. It is not the property of any leader, and is no one’s monopoly. 11 November is a date for all Angolans, for independence was a conquest for all Angolans.32 Yet, in Angola, 11 November 1975 was not only Independence Day, but also the day that pitched Unita and the MPLA headlong into war, with the result that any evocation of that period means negotiating a path through uncomfortable memories. In a later Independence Day address, Samakuva portrayed it as follows: About 32 years ago, Angola began a period of serious legal and political disruption that shook the unity of the nation, subverted the role of the state and promoted bad governance. This period is coming to an end.33 The picture that Samakuva painted was one of a situation without agency. Neither Unita nor the MPLA plays a role in the “period of disruption”. Instead, Samakuva spoke of Angolans calling for change, and went on to present a lengthy manifesto that promised an Angola with properly functioning health and education services. He implied that it fell on Unita to make that change a reality on behalf of the Angolan nation. Why has the Unita leadership appeared reluctant to assert its old nationalist appeals? Why has a party that traditionally had a strong regional support base, and which used to talk about racial exclusion, been reticent about playing the identity card in the last few years? In 2002, Unita accepted a role as an opposition party in a nominally multiparty system that is dominated overwhelmingly by the MPLA. As I have suggested, de jure multipartyism is important to the MPLA’s claims to national and international legitimacy. For Unita, it is difficult to reconcile

32 “Discurso a proferir por ocasião das celebrações do 30° aniversário da independência nacional”, delivered 9 November 2005. Online at: http://www.samakuva.com/index.php ?option=com_content&task=view&id=67 accessed 24 May 2011. 33 “Carta do Presidente Samakuva aos Quadros da Nação”, delivered 11 November 2007. Online at: http://www.kwacha.net/samakuva_huambo.htm (accessed November 2007); this website was no longer functioning at the time when this article was being finalised. 212 justin pearce its role as a subordinate loyal opposition party with claims to a nationalist history that brought Unita into head-to-head conflict with the MPLA. All this means that at least within the realm of formal politics, the MPLA is seldom challenged either on its implicit claims to be the sole liberation movement in Angola, or on the specific content of the nationalism that underlies the MPLA’s political discourse. Echoes of the wartime discourse that cast Unita as a terrorist force were heard in the months preceding the 2008 elections, as the reconciliatory tone adopted by the MPLA at the time of the 2002 peace agreement disappeared amid attempts to remind the electorate of Unita’s violent history, while portraying itself as reconciliator.34 The 2008 election results in the former Unita heartlands of the Central Highlands suggest that the MPLA’s efforts in portraying itself as the bringer of peace and claiming credit for post-war reconstruction trumped the identity-based claims upon which Unita had gained a majority in the region in the 1992 election.35 Unita appears to have lost the majority of the rural Ovimbundu vote in 2008, and to have retained the support only of its most loyal followers: those Unita soldiers and their families who remained with the movement until its disarmament in 2002. The sentiments of exclusion voiced by these rump Unita supporters in their provincial bairros and villages seldom find voice at the level of national politics.

Alternative Visions

If we are to find challenges to the dominant vision of Angolan nationalism, we need instead to look to the blogs and newspaper opinion articles produced by Angolan intellectuals, some of whom are closely associated with Unita while others are politically independent. Here we can see claims that a coastal, mestiço, Portuguese-speaking elite continues to dominate public life in Angola. A blogger originally from the Central Highlands, who goes by the name of Angolainterrogada – Angola questioned – has written about a phenomenon that he calls caluandismo – a

34 In the most widely-reported incident, Defence Minister Kundi Paihama claimed that Unita had retained arms caches since some of its leaders planned to return to war: an accusation vigorously denied by Unita. Lusa news agency, 12 February 2008. 35 According to official figures for the 2008 election, in Huambo province the MPLA won 82.05% of the vote to Unita’s 13.51%, and in Bié, the MPLA won 74.93% to Unita’s 18.25%. See online: http://www.cne.ao accessed 10 June 2011. By contrast, Unita won both provinces in the 1992 elections. The MPLA and Unita’s campaign strategies in the Central Highlands before the 2008 elections are discussed in Pearce (2008). changing nationalisms: from war to peace in angola 213 caluanda being someone from Luanda. Caluandismo, he says is what keeps Luanda in a dominant position in the country’s political and cultural life, and what leads to a narrow and exclusive definition of Angolan nationalism. The politicisation of Angolan society and the long conflict made Luanda the realm of the true patriots and the seat of pure Angolanness and of authentic nationalism. In other words, the very concept of “Angolan” nationalism was deplorably caluandised.36 Caluandismo is not simply about being born in Luanda – one can possess caluandista attitudes without being born in Luanda, and conversely, being born in the capital does not necessarily make one a caluandista. The writer defines as crucial the racial, political and geographical elements of caluandismo. For blacks to be accepted as Angolans under the criteria of caluandismo, they have to prove their status as patriots by demonstrating that they are MPLA members, that they come from the “north” (defined as the provinces close to Luanda) and have Portuguese rather than African surnames – African surnames being associated with the less assimilated black people of the south. The effect of this critical vision is to subvert the MPLA’s claim to be the keeper of Angolan nationalism by suggesting that Angola is actually controlled by a racially and regionally exclusive elite associated with the ruling party. The question of race is further highlighted in a newspaper article by the writer Sousa Jamba: a long-time exile and younger brother of Jaka Jamba. In the article, Jamba recalls a conversation with another Angolan living abroad, who said that racism was the main reason why he did not want to return to his home country. Angola, according to him, was a highly racist country because mulatos and whites controlled the most influential ministries, like the finance ministry, and the banks which only gave jobs to their cousins, brothers and friends. He thought Angola needed a kind of Robert Mugabe to transform the inequalities inherited from the colonial system.37 Jamba continues: Racism, for many Angolans, has come to be a highly sensitive and delicate subject, something which is sad and dangerous. One of the causes of this lies

36 “Um Obstáculo Chamado Caluandismo”. Online at: http://angolainterrogada .blogspot.com/ 29 August 2007. 37 Undated article originally from Angolense, posted on the Os Luenas internet forum, http://groups.msn.com/OsLuenas/ 214 justin pearce

in the fact that the debate around race and identity has become highly politicised. And this, by all means, is understandable. In a country like Angola, which inherited rifts that manifested themselves violently, much care is needed for there not to be more contradictions. But this attitude can also have the effect of the ostrich which, when confronted by danger, puts its head in the sand. In my experience, Angolans rarely discuss the question of race honestly. Sousa Jamba goes on to discuss mutual suspicions between black and mixed-race people in Angola. While he acknowledges that these suspicions are largely a matter of perception, he ends with a caution: Nevertheless, perceptions persist and carry much weight, as was amply demonstrated in Rwanda in 1994. The counter-argument is less prevalent, but is also on record: a complaint that mestiços, coastal people and Portuguese-speakers are excluded by a discourse that puts too heavy an emphasis on indigeneity and blackness. An anonymous article in the Angolense newspaper complains about the prevalence of an idea that “some are more Angolan than others” on the basis of skin colour, cultural preferences and so on. More Angolan than all the others will be the poor, the darkest-skinned, those who participated in the national liberation struggle, who in relation to the others have suffered little or no influence, who have only Angolan nationality …38 The writer sees these prejudices revealed in phrases such as “creoles from the coast” and “deep Angola” used by those who assert Angolan indigeneity. The tone of the piece borders on satire, written ironically in a voice that embodies the presumed anxieties of the mestiço and assimilado elite. But it seems more likely to have been written by a member of that elite, setting up a straw man that represents the supposed ascendancy of a culture in which blackness and cultural authenticity are valued over lighter skins and assimilation. Either way, the piece speaks of battle for the meaning of Angolanness, and indicates where the battle lines are situated. In so doing, it gives voice to things that are often left unsaid. Further challenges to the dominance of the MPLA’s nationalist narrative can be discerned in the historical revisionism that has begun since

38 “Quem é o Angolano de gema?” Anonymous article originally from Angolense, posted on the Os Luenas internet forum, http://groups.msn.com/OsLuenas/. Accessed 3 December 2007. changing nationalisms: from war to peace in angola 215 the turn of the present century: the challenges to the MPLA versions of history surrounding the storming of the Luanda prisons on 4 February 1961, and the attempted coup of 27 May 1977 and subsequent reprisals.39 This historical revisionism is notable in that it presents a challenge to the MPLA’s preferred view of its historical role and to its version of nationhood. The reassessment of the prison attack undermines the MPLA’s position as the progenitor of the Angolan nationalist struggle. The re-evaluation of the 27 May uprising challenges the MPLA’s long-held assertion that race was not an issue in Angola. Finally, we may note attempts by organisations and individuals to reclaim the very idea of “the Angolan people”, which hitherto had been contested only by the former liberation movements. The campaigns for a peaceful solution to the civil conflict, taken up by churches and other organisations in Angolan civil society from 1999 until the end of the war, explicitly addressed “the Angolan people” in a way that sought to construct “o povo angolano” as an entity with common interests that were at odds with the interests of the rival elites of the MPLA and Unita.40 Similarly, the hip-hop artist MCK begins one of his raps with the words “Cidadão angolense acorda! (Angolan citizen, wake up!)” before proceed- ing to call attention to the material inequalities and social exclusion that exist under the current administration.41 Even more provocatively, the opposition Youth, Workers’ and Peasants’ Alliance Party of Angola (Pajoca) ended each of its allocated television broadcast slots during the 2008 election campaign with an ironic dedication to one or other hero of the Angolan liberation struggle, whose legacy, according to Pajoca, had been betrayed by the current political leadership: for example, “to António Agostinho Neto, who said ‘most important is to solve the people’s problems,’ and who was betrayed by his companions”.42

39 See Birmingham in this volume. L. Pawson, “The 27 May in Angola: a view from below”, Relações Internacionais, Lisbon: Instituto Português de Relações Internacionais, June 2007. Online at: http://www.ipri.pt/publicacoes/revista_ri/pdf/RI14_LPawson_Eng .pdf (2007) has written of the difficulties that surround discussion of the events of 27 May 1977 in Angola. 40 See for example “O Calvário do povo Angolano”, Episcopal Conference of Angola and São Tomé. 28 January 1999. Online at: http://propace.br.tripod.com/texto_9.html; and the press release issued by Pro Pace, September 2001 accessed 10 June 2011: “Up until now, international support has been directed either towards the MPLA government or the UNITA rebels – never towards the Angolan people, whose interests are not represented by those parties […] This is what we are trying to change.” 41 “Têknikas, kausas e konsekwências”, from the album A Trincheira de Ideias by MCK. 42 Broadcast by PAJOCA on Televisão Pública de Angola, “Antena aberta” 22 August 2008. 216 justin pearce

Conclusion

Implicit in these observations is the fact that what it means to be Angolan remains contested just as it was before and during the civil war, but the end of the war has brought about a change both in the content of nationalist ideologies and in the way they are contested. The peace settlement of 2002 represented the triumph of one particular vision of Angolan nationalism, not by consensus but by the superiority of military force. Nationalism is no longer being contested on the battlefield, but the nature of the settlement is such that it created no space for nationalism to be contested within the sphere of parliamentary politics. Other versions of nationalism, which assert a plurality of Angolan identities and which seek to expose the MPLA as fractionalist rather than as truly nationalist, have emerged in elite public discourse, due in part to the somewhat more open media environment that has developed in Angola thanks to the advent of the internet, and the easing of the repressive political climate that prevailed in the late 1990s. But the MPLA’s current dominant position has allowed its version of nationalism to become normalised, with the party’s long-held claims to be Angola’s only liberation movement reinforced by new claims of having brought peace through its victory over Unita. CHAPTER NINE

IS ‘NATIONALISM’ A FEATURE OF ANGOLA’S CULTURAL IDENTITY?

David Birmingham

In September of 2006 President José Eduardo dos Santos made a speech in Luanda to a conference on the national culture of Angola. In it, to the surprise of foreigners such as myself, he said ‘our national language is Portuguese’.1 African vernaculars are not ‘national’ languages in the eyes of the head of state. Not Umbundu which has been so widely used as the language of command in the rival armed forces. Nor Kimbundu which is the language of the historic heartland of Angola between Luanda and Ambaca. And certainly not Kikongo which, shock-horror, is spoken by aliens across the border in Congo. Not even the old imperial language of the Lunda spoken in eastern Angola but also – and here lay the president’s problem when searching for a pure form of identity – among foreigners in Zambia. When a Zambian football team arrived for a championship game in Luanda their local hosts were amazed to hear them speak to one another in Bemba rather than English, bush people using a barbarous tongue whereas Angolans spoke fluent Portuguese. And when some time later the Angolan football team was drawn to play Portugal in a world cup fixture they were not at all fazed to be facing the old imperial enemy. That is a fine encounter they said: these people are our cousins, we all speak Portuguese. The problem which Angola faced in finding a language which would enhance its sense of national identity and national pride was not a new one. Kwame Nkrumah, Africa’s premier statesman, chose English and insisted that the language become the vehicle of education from the very beginning of primary school classes. Haile Sellassie, the imperial war lord

1 ‘We must have the courage to assert that the Portuguese language, adopted as the official language of the cournty since independence, is to-day the mother tongue of one third of Angolan citizens and must therefore be confirmed as the national language of Angola’. Speech on ‘cultural identity’ by Eduardo dos Santos addressing the Third Symposium on National Culture and cited in the Luanda week-end papers of Saturday 16 September 2006, for example Folha 8 (Luanda), p. 32. The president did go on to say that the Angolan vernaculars, hitherto known as ‘national languages’, should be preserved as part of the country’s historical heritage.

© koninklijke brill nv, leiden, �� | doi ��.��/��_�� 218 david birmingham of Ethiopia, chose Amharic, the language of a northern aristocracy but one not understood by the peasants and pastoralists outside the wall of Addis Ababa. Julius Nyerere, the Christian proponent of social equality in Tanzania, chose Swahili, a language of the northern coast of Kenya which had been adopted as the language of high command during the thirty year rule of the country by Germany. And Eduardo dos Santos opted for Portuguese, a language initially introduced 500 years ago by conquistado- res and merchants of blood. So emphatic was the decision to reinforce Portuguese as the language of the Angolan nation that several hundred members of the press were required to present themselves to the Catholic University of Luanda and take examinations to prove their competence in the ‘national language’. Journalists protested vainly but editors were scan- dalised to the point of incandescence at the indignity of having to sit on student benches beside their underlings and be tested for their fluency. The question of language is far more than a question of efficiency, of having a lingua franca which will lubricate administration and commerce. It is not even to do with the rule of law in a country deeply enmeshed in the bogus legalism of Portuguese administrative practice. It is to do with status. It has been suggested, without good evidence in a country devoid of any reliable statistic and in which even the size of the population is quite unknown, that forty per cent of Angolans speak Portuguese as their first language. And virtually all descendents of the thirteen ruling family dynasties, and all associates of the kleptocratic élite, and most members of the administrative middle class, proudly bear Portuguese names, usually in the Roman style with a first name, a middle name, and a last name. Only the members of the old opposition, a rival group of middle class administrators and entrepreneurs once based on the rival highland kingdoms of the Ovimbundu plateau, pride themselves on having African names, Bantu names. To the sharply dressed citizens of Luanda the term ‘Bantu’ is one of disdain, a term almost as opprobrious in Angola as it once was among the white élite of South Africa. The use of language, and the choice of names, is not the only sign that ‘national identity’ in Angola is rooted in imperial experience rather than in historic life styles. Ten years ago an international congress was held in the old Imperial Cinema at Luanda, by then smartly refurbished to house the parliament which had been elected in Angola’s one-and-only general election. At the end of the week-long debate a grand picnic to an historic site was laid on with executive jets and tureens of soup flown up from Cape Town. It was expected that the historic site would be Mbanza Kongo, medieval capital of a prestigious African kingdom. But no, the picnic was is ‘nationalism’ a feature of angola’s cultural identity? 219 held at the mouth of the Zaire river, at the very spot where – joy of joys – the first Portuguese had made the first landfall on Angolan soil. A replica monument of stone commemorated the arrival of Diogo Cão in 1483. The imperial date, event, and place had become a shrine to the founding of the nation and all previous history, culture and experience was eclipsed.2 A subject which is controversially related to the question of imperial memory and colonial language is the assumption that ‘nationalism’ must be related to a deeply felt ‘African’ identity. Does Angola, or its chattering class and its financial elite, feel itself to be truly African. Probably not. Angola is a semi-detached member of the African Union, a country which is somehow superior to the other fifty, rather more backward, members. Indeed Angola sees itself as more akin to Brazil than to any of the patrioti- cally nationalist post-colonial republics of Africa. One symptom of this alien mind-set is to be seen in the field of planning. Angola wants to model its political future not on the aspirations of its people but on utopian visions either inherited from the settler empire or from the bureaucrati- cally ordered schemes applied to parts of Brazil. Brasilia, for instance, is an utterly soulless town which is zoned in every which way to create a tidy image, but it has none of the vitality of Rio or Bahia. Administrators in Angola aspire to similar zoning and planning regardless of human needs and aspirations. In 2006, for instance, the ministry of internal administration decided that the great market at Huambo was untidy and inappropri- ately located and so they had it cleared and burnt. Women traders and their customers had to move five kilometres out of town to a ‘properly’ designated market site which they had to reach expensively in fleets of minibuses. Human beings, the core of any nation, were arrogantly brushed aside by a state apparatus which had no sensitivity to people.3 Rumour has it that a similar threat hangs over Roque Santeiro, the world’s largest, most colourful, most varied and most energetic enterprise zone and one which serves the needs of Luanda’s four million citizens, squatters, refugees and other residents. The Brazilian dimension of the ‘national’ aspirations of Angola’s princely court circle goes beyond the bureaucratic mind-set of the plan- ners. True nationalists would be investing their earnings from the colonial style-export of raw materials – five billion dollars of oil and one billion

2 Maria Emilia Madeira Santos (ed.), Actas da III Reunião Internacional de História de África, Lisbon: Centro de Estudos de História e Cartografia Antiga, 2001 3 The ruins of the old Huambo market were seen by a British parliamentary delegation visiting the city in September 2006. 220 david birmingham dollars of diamonds – in economic diversification and in social well-being. The Angolan state has no such national aspirations and Angola remains one of the very poorest countries of the entire world with 250 per thousand infant mortality, a total paucity of doctors and nurses outside the quasi-autonomous metropolis of Luanda, and a network of provincial schools which struggles to attract teachers who remain deprived of books and school-rooms. If nationalism is about civic pride and social achievement then nationalism is perhaps confined to the city of Luanda. And perhaps also to the restless trans-Congo province of Cabinda whose off- shore oil-wells provide 40 per cent of Angola’s mineral revenue but some of whose people cannot understand why it was that the Portuguese granted independence to Angola, south of the river, but not to Cabinda, north of the river. A ‘nationalist’ identification with the Luanda metropolis is also ambiguous in many other of Angola’s seventeen provinces – do I want to flee the countryside to the bidonville shanties of the city or do I want to cut free of urban drift and enhance the well-being of Uige, or Huambo, or Lubango. Cabinda, however, is an especially difficult case where journalists who try to enquire into peoples’ aspirations are constantly liable to get caught by the security police. So far little is being done to meet provincial aspirations in any province and although no figures are available it is generally believed that much of the ‘national’ wealth is being invested abroad. Brazil is cited as the destination of much capital which could be otherwise used for development. Not only is Angola looking to Brazil for models and links, it also pro- tecting itself from its greatest enemies, the economic giants of Africa. The most feared enemy and rival is South Africa and the joy felt in the ruling circles when Angola defeated the Afro-Saxons of both South Africa and Nigeria to win the right to host the Africa football cup was intense. Indeed less than serious observers such as the present writer assume that hosting the cup in 2010 will absorb so much administrative energy that there is no prospect of the state being able to organise a national parliamentary election – let alone a more distant presidential one – until the dust has settled on the football stadia which the country plans to build. Mussolini, after all, kept the minds of his people off politics by winning the world football cup twice, in 1934 and in 1938. With such dreams is national pride cemented. Angola may not yet be producing as much oil as Nigeria but it will surely catch up and already it produces as many diamonds as South Africa. Membership of the southern African economic block appears to be of no significance to Angola while its real economic partners are in Brazil and Portugal. is ‘nationalism’ a feature of angola’s cultural identity? 221

When independence swept through West Africa in 1960 the premier statesman of the region, Kwame Nkrumah, called a great pan-African conference of nationalists from all countries, both colonised and decolonised. One of the orators said: ‘the greatest of the freedoms to which we must aspire is the freedom from fear’.4 This is an aspiration to which Angolan nationalists must still aspire. The government dreams of Salazarian authoritarianism, of a state in which decrees poured unhin- dered from the desk of the chairman of the cabinet and in which the whip was cracked by a Gestapo-trained secret police. Citizens of Angola – in contradistinction to politicians – are not proud of their country, nor of their government. The peoples’ only serious relationship with authority is with the police, most notably the ‘fiscal police’. In a country where survival meant scheming, trading, bartering, exchanging wads of local currency notes for hard dollar currency each night and exchanging it back over the next days as liquidity was required for street business, the fear of the fiscal police was a constant preoccupation. Yet the dollar was necessary to survival and returning to the barter economy of the 1980s when the six-pack of lager was the most stable form of ‘currency’ was not an option.5 And yet being dependent on the dollar for medium-term business trans- actions somehow minimised the status of the nation, a reminder of the way in which the early nationalists mocked the independent élite of Liberia which had ruled their republic for a century but did not have any currency other than the American greenback. Fear, and the memory of fear, remains a feature of present day Angola and minimises the development of a single, unifying, nationalist culture. Fifty years ago, in 1957, the political police which was so ubiquitous in ‘fascist’ Portugal was introduced to Angola to try to curtail the emergence of a white nationalist movement which might aspire to a Brazilian-style unilateral declaration of independence or a South African-style régime based on local white supremacy. Four years later, when black nationalist hot-heads tried to spring some of their heroes from the São Paulo prison in Luanda, revolutionary enthusiasm seized ambitious young Angolans

4 The present author was a delegate to this ‘Positive Action’ conference, Accra, June 1960. 5 David Birmingham, “Angola Revisited”, Journal of Southern African Studies, 15(1),1988, pp.1–14. For work on the parallel currencies of Luanda see, inter alia, Alain Morice “Commerce parallèle et troc à Luanda”, Politique Africaine, No.17, 1985, pp.105–120 and Jean-Paul Azam, Paul Collier and Andrea Cravinho, “Crop Sales, Shortages and Peasant Portfolio Behaviour: an Analysis of Angola”, Journal of Development Studies, 30(2), 1994, pp.361–379, 222 david birmingham and the white community panicked. Vigilantes, with at least tacit official support, roamed the streets bludgeoning any black or brown person thought to be educated enough to support an anti-colonial political move ment. As the uprising spread to the countryside, enflamed by long delays in paying arrears of wages to coffee pickers, the government began issuing weapons which frightened settlers used to kill thousands of demonstra- tors and their sympathisers in the north. The culture of fear was not only black against white or white against black but class against class, land- owners against land-workers, immigrants against indigenees. The hostility between the northerners, whose land had been seized by shop-keepers turned planters, and the southerners, Gastarbeiter who had been rounded up and forcibly transported to work on the seized lands, had little to do with deep and historical cultural differences – ‘tribalism’ in the cheap par- lance of the world media – and everything to do with the economics of despair. The 1961 phase of despair greatly inhibited the possibility of a national political culture emerging in Angola and over the next five years two rival groups of muscle men calling themselves the FNLA and the UNITA sought to mobilise that despair in different constituencies.6 Fear, the enemy of national pride and cohesion, afflicted not only the rural provinces of north and south but came also to vitiate the emergence of participatory politics in the city. Angola’s second bout of blood-letting occurred in 1977. It is still remembered with trepidation by a generation reaching fifty years of age. When no one is listening young men and women will ask ‘where were you, mother, on May 27’ or even ‘father, how did you manage to survive the purges of 27 May?’ So powerful are the memories that when a book was published – apparently during November 2007 – on the political purgings of 1977 it immediately vanished from the Lisbon book trade. Shop-keepers assumed that dark forces had managed, officially or officiously, to impose a form of censorship. The 1977 massacres created an on-going crisis. One real difficulty in the political culture of Angola is the problem of the missing generation, the generation that was either lost in war or fled into exile. The grey beards remain in charge but the young men, now rising forty, are not trusted to take over. No middle generation exists to smooth the transition.7 And fear continues to

6 The classic study is John Marcum, The Angolan Revolution. Exile Politics and Guerilla Warfare, 1962–1976, Cambridge, Mass.: MIT Press, 2 vols, 1969 and 1978. 7 For an example of interesting new work on Angolan politics, see Jon Schubert, Dez jogadores contra um guarda-rede. Angola: The Dynamics of State and Civil Society 2002– 2007, MA thesis, Basel University, 2007. is ‘nationalism’ a feature of angola’s cultural identity? 223 permeate urban society and impede the process of political learning, experimentation, and conflict solving. Those who do not remember the great purge of 1977, let alone that of 1961, certainly do remember the third great outburst of urban warfare in Luanda, the war of October 1992. As Angola basked in peace, as displaced persons loaded their possessions onto their heads and set off to their homelands, the politicians – far from preparing for a national future of peace and reconciliation – sharpened their weapons and trained up ever more fearsome units of black-clad police and battalions of urban guerrillas. Street killing, and the long-term culture of fear, once more inhibited the process of nation-building. In the city many people fear the activities of their national government. In the countryside government is more remote, at times even invisible. Other preoccupations are far more important. Should one be so bold as to travel inland to say Cazombo, and to cross the upper Zambezi by canoe under the stone arches of the long demolished road bridge, and arrive in a village with two school rooms, each holding a hundred or so children, not far from a nearby colony of leprosy sufferers, one will be taken to the mais velho, the village elder. He is not the agent of government. He does not speak the ‘national’ language, Portuguese. He is the village pastor, a follower in the tradition of John Nelson Darby who led a break-away group of brethren in Plymouth in the mid-nineteenth century. These Derbyites, whose self-reliant farming communities spread quite widely across Europe, also reached the heart of Africa and there they still survive.8 For most Angolans the church is a far more important, real, com- forting and sustaining institution than any government agency or political party. Some churches, with external ties, can make good some government deficiencies, supply chalk for the school blackboard, measles vac- cine for the town clinic, mosquito nets for the orphanage, a pump for the village well. But in a world of fear, insecurity, disease and the sudden death which takes away one in every four children before the age of five, it is the spiritual comforts of the church which are most cherished. Some of the thatched chapels round which life revolves in lieu of politics are ones, such as the Derbyite ones, which are based on old colonial mission traditions. Angola’s first president, Agostinho Neto, was the son of Methodist minister who got sufficient educational support from the church to graduate in medicine. And Angola’s most fearsome guerrilla leader, Jonas Savimbi, received Presbyterian funding and also entered

8 The British parliamentary delegation of 2003 visited villages around Cazombo. 224 david birmingham medical school, though with only very limited scholastic success. Ironically both the Methodist and the Presbyterian turned to atheistic and alien traditions of political ideology when seeking to create ‘nationalist’ political movements, one following the Stalinist tradition and the other the Maoist one. In the longer term, however, the religion which eclipsed politics in Angola was not based on the ancient Protestant chapel, but a new one broadly called ‘Pentecostal’. In Luanda it is said that the Pentecostal churches are opening one new place of worship every week. Abandoned hotels, empty warehouses, dis- owned shop-fronts, all are being bought up by rich organisations such as the Church of God and turned into vibrant centres of community dynamism. In a land where national politics have nothing to offer but interference, this explosion of religious passion was already observed as early as 1992 by Angola’s foremost social scientist, the former minister of education, one-time professor of architecture, and prize-winning novelist, Artur Pestana. Writing under the pen-name Pepetela, he wrote a fictionalised account of the first cohort of idealistic nationalists, The Generation of Utopia, in which one disillusionned leader moves from politics into the world of Pentecostal high finance.9 The government of Angola does not collect income tax from its people but the Church of God does. And not only does it collect tithes from its massed adherents, it also receives large subsidies from both the United States of America and from the United States of Brazil. The ‘cathedral’, in one of the richer quarters of Luanda, has apparently to be seen to be believed. But even this palatial edifice is not sufficient to meet the demand for very loud, very public, forms of worship and the Pentecostals rent whole football stadia for their great set- piece services. Once upon a time it was nationalist politicians who could fill a stadium to capacity: now it is preachers who speak in tongues. Even the Catholic Church, once the mainstay of empire and later the theologi- cal prop adopted by the Angolan president when he abandoned Marxism, had to struggle to match the popular appeal of the new churches fuelled from Brazil. As the Pentecostals filled a football stadium the Catholics called for a pilgrimage to the seventeenth-century shrine of Muxima, a hundred miles up-river. The political tycoons heard the call and went on pilgrimage in their helicopters, the well-heeled middle class crowded the roads with their BMWs, and the rest of the city population – imitating the pilgrims who flood to Fatima in Portugal – boarded fleets of buses.

9 Pepetela, A Geração da Utopia, Lisbon: Dom Quixote, 1992. is ‘nationalism’ a feature of angola’s cultural identity? 225

The political class played it both ways, supporting the Pentecostal establishment but also taking part in the Catholic pilgrimage.10 More threaten- ing to the ‘nationalist’ politicians, perhaps, were the smaller and older independent churches. The Kimbanguist church, with several million tithe-paying adherents in Congo, still exists in northern Angola and perhaps among northerners who have gravitated to Luanda. A similar independent church is the Tokoist church and the loyalty of its members to the memory of their founder, Simon Toko, has caused past nationalist leaders to be highly suspicious of it. In Africa nationalism, when it took shape, was circumscribed by the artificial boundaries of 1885. In Angola the occupying forces of colo nialism had reached the prescribed borders by 1925. Thereafter such ‘nationalism’ as evolved across Africa was essentially ‘anti-colonialism’. Journalists, across Africa and also in Angola, were often at the forefront of the demand for greater rights and opportunities for subject peoples. In much of Africa, though not in Angola, these rights had been won by 1960 and the nationalists turned their attention to the problem of neo-colonial, as opposed to old colonial, foreign domination. The neo-colonial option, so shrewdly pursued by France which Balkanised its territories into tiny fragments tightly bound to Paris by fiscal, commercial and military ties, was not readily available to Portugal. When independence was reluctantly granted to Angola – the last governor escaped on a gunboat under cover of darkness on the night of 11 November 1975 – the country fell prey to the proxy conflict which pitted the superpowers against each other. Six months earlier the Americans had lost one war in Vietnam and were in urgent search of a territory on which to mount a return match. For the next fifteen years, until 1990, the Portuguese were squeezed out as the ‘popular party’ sold its oil to America and bought its guns from Russia while the ‘opposition party’ bought its guns from America and sold its diamonds to Russia. The end of the Cold War brought on two civil wars, in 1992 to 1994, and in 1998 to 2002 and only then did the government have to face the question of who its neo-colonial partners should be. Who would be the most likely financial interlocutor to allow a national political agenda to be determined in Angola by nationalists politicians rather than by global forces. The two neo-imperial giants which straddled Africa and tried to win the favour of Angola’s ‘nationalists’ were Bretton Woods and Shanghai.

10 See the weekly Agora (Luanda) of 16 September 2006 for a picture of one of the two stadia in which the Igreja Universal claimed to have preached to 100,000 people. 226 david birmingham

Bretton Woods was represented not by the World Bank – which did not offer loans to oil-rich republics – but by the International Monetary Fund. The Fund’s agenda gradually changed from an insistence that impoverished southern debtors should pay any disposable state income to wealthy northern creditors to a new policy of suggesting that local resources be invested in health and education as a starting base for ‘development’ opportunities. The suggestion that Angola should adopt such a development policy was poorly received in Luanda and seen as an impertinent intrusion into national independence of action. And when the IMF asked questions about the thousand million dollars which had somehow leaked out of the public purse and into private off-shore bank accounts, this was deemed the height of impertinence. Pressure to open the books to ‘trans- parent’ inspection or to assess the meaning of ‘corruption’ in the Angolan context was robustly rebuffed in the name of national autonomy and pride. Instead Angola chose the Shanghai economic option, orchestrated by the political masters in Beijing. Shortly after the end of the last Angolan civil war China began to buy more oil from Angola than it did from Saudi Arabia. Equally it began to offer large loans on terms quite different from those which the IMF might have required. No questions were asked about embezzlement, about the closed nature of the entrepreneurial class tightly linked to the ruling political families, about the absence of the rule of law in a country awash with laws but devoid of judges or tribunals, about the human and civic rights of citizens brushed aside when the powers that be wished to bulldoze whole segments of old Luanda for grandiose plans based on five billion dollars of Chinese credit. The strings attached to Chinese money did not concern Africa, and the politics of nationalism, but concerned Asia and the Chinese need to ensure that no voice should ever be heard which questioned Beijing’s policy towards Taiwan. Chinese neo-imperialism, however, was much more than a deal about oil and politics. It was also a deal about markets for manufactures. Indeed it was quintessentially a deal like the neo-colonial arrangements of 1960, a deal which precluded any growth of local industrialisation, a deal which supplied manufactures so cheaply that no Angola textile man- ufacturer could hope to revive the mills and workshops of the late colonial period. China would gradually dominate the market for everything from tea-spoons to brassières, from sewing machines to malaria drugs. The nationalist politicians – like the old comprador bourgeoisie of nineteenth-century Portugal – appreciated the revenues, licit and illicit, which they could syphon off from the import-export trade. By contrast the national citizens, deprived of job opportunities in the productive sector, is ‘nationalism’ a feature of angola’s cultural identity? 227 were more ambivalent. They appreciated the cheap shirts and shoes but feared the steam rolling effect of a new colonial age orchestrated in Asia. When the leaders of Angola’s parliamentary opposition were asked by visiting British parliamentarians about their attitude to the Chinese invasion they said: ‘We love it. If anything can bring defeat to the government in the event of an election it is the sight of Chinese navvies doing work on the roads and railways which underemployed Angolans would be keen and able to do themselves’.11 No one knows how many Chinese are currently working in Angola. The present-day opposition in Angola is mainly UNITA, with a plethora of one-man political bands and a scattering of former FNLA fragments around its fringes. The question must therefore be asked: was UNITA a ‘nationalist’ movement. Or is it better described as ‘sub-nationalist’. What were the unifying features of the liberation movement founded by Jonas Savimbi between 1964 and 1966. It was a liberation movement rather than a political party as it had but one plank to its political agenda, total independence. The media of the time assumed that it was a ‘tribal’ association appealing to the so-called ‘Ovimbundu’ nation or sub-nation. But Ovimbundu was a difficult concept. The planalto, before the colonial conquest, was deeply divided by a dozen antagonistic and milita- rised kingdoms. Even the languages were localised until the Protestant missionaries – led first and foremost by the Swiss polymath Héli Chatelain – tried to unify them into a single written form that could be used for Biblical teaching. When in 1926 the colonisers adopted the Portuguese variant of Mussolini’s fascism they were afeared that vernaculars would undermine the ‘loyalty’ of subject peopled to the empire and so they insisted that all missionaries, French, German, Canadian and the rest, should learn Portuguese and teach only in the Lusitanian tongue. Even as early as 1912, when a homeland for Europe’s unwanted Jews was being abortively planned in Angola, the republican empire had insisted that any immigrant should be taught in Portuguese and would be subject to Portuguese military conscription. Unification of political sentiment on the highland was slow but was reinforced during the colonial war. The Portuguese army feared that the guerrillas of the liberation movements might aspire to infect the densely populated heartland of the colony with concepts of freedom. Taking a leaf from the British in Kenya and the

11 Author’s interview with UNITA leaders, Luanda, May 2006. See also my Empire in Africa: Angola and its neighbours, Athens: Ohio University Press, 2006, ch.11. 228 david birmingham

Americans in Vietnam they decided to round up peasants into concentrated villages which could be overseen by soldiers. The resulting economic hardship incurred by farmers who were accustomed to living on their land was severe enough to make the policy totally self-defeating. But dismayed as they were the highlanders were not immediately keen on the political body which tried to mobilise them, let alone keen on identifying with a ‘nation’ embracing the old imperial boundaries of Angola.12 UNITA was a movement of modernisers, not dissimilar to the MPLA of the Luanda hinterland, but with a much smaller base of merchants, entrepreneurs and university graduates. Its leader, Jonas Savimbi, adopted the style of Dr (commonly used by Portuguese holders of a bachelor’s degree) but although he had switched his course from medicine to politics while in Switzerland he never appears to have finished and graduated. Indeed if he had a degree at all is was probably an honorary one, much like the honorary degree which his alma mater, the University of Lausanne, had given to Mussolini in the 1930s. Without a secure cadre to form a broadly- based leadership, and without a clientele to foster a spirit of ‘nationalism’, UNITA became a one-person dictatorship. Terror was its main tool and execution the main punishment for any sign of disloyalty. Loyalty was rewarded by the gift of conscripted young women as sex-trophies. The personality cult of the leader reached extremes which were disseminated by the Black Cockerel radio transmitter. But when, in 1992, it came to the one-and-only popular test of electoral popularity, the strident leader of UNITA was more feared than the grey machine man of MPLA. Neither seemed to embody the concept of ‘nationalism’. Each was associated with the greed of a small élite. And when, ten year later, the fourth and last Angolan war ended UNITA did not emerge from the bush with any political policies, national or otherwise, but only with a thirst to replace the old guard who had held the pork barrel all to themselves for the first 30 years of independence. If it is hard to associate UNITA with ‘nationalism’ it is even harder to associate FNLA with an identity which embraces the whole national territory. Indeed the movement began as the union of northern peoples and was rooted in a community of Angolans which, ever since 1900, had seen Kinshasa as their urban eldorado, not Luanda. The Angolan expatriates of the north gained their education through a cross-border network of

12 The classic fictional account of these times is ‘Colonialism in Angola: Kinyama’s Experience’ in David Birmingham, Portugal and Africa, Athens: Ohio University Press, 1999. is ‘nationalism’ a feature of angola’s cultural identity? 229

Baptist chapels and schools. Many had escaped the rigours of forced labour in the logging camps of Cabinda. Once fluent in French they aspired to find higher status jobs, but as non-Belgian subjects they were precluded from entry into the huge colonial civil service of Congo and had to enter the private sector. The result was that the ideology of the FNLA was a capitalist one, very different from the socialist ideology of the MPLA in Luanda whose members could enter the Portuguese colonial civil service and therefore identified themselves with the state, the potential nation, with all its opportunities for educated cadres. The Francophone exiles found their role in Congo enhanced by the outbreak of war in northern Angola in March 1961. Although the oft-quoted figures for refugees sometimes outnumber the total population of the two northern provinces of Angola, many people did flee, settle, marry and raise a generation of French-speaking Angolans who sought to return to Angola in 1974. They were not immediately welcomed into any ‘nationalist’ big tent. Nationalism in Angola was fundamentally fractured when the colonial war of 1961 to 1974 was gradually transformed into the civil war of 1975 to 1991. UNITA turned for help to Washington’s Gurkha’s, the Dutch- speaking shock troops of apartheid, in order to advance its claims to wealth and power. FNLA borrowed General Mobutu’s army to conquer Luanda and make sure that a left-wing régime antagonistic to both Zaire and the United States would not inherit power from the Portuguese. To hold off both assaults on the capital the MPLA hired ‘Moscow’s Gurkhas’ who were flown in from Cuba and paid with oil revenues coming – to the utter dismay of Kissinger’s Washington – from Texas. The Cubans claimed that their aim was pure liberation and had nothing to do with Soviet aspirations to world domination. They succeeded in holding the capital city until the Portuguese had slipped away. But they did not succeed in creating a unified, integrated nation. The followers of FNLA were, however, gradually incorporated into Angola’s urban sector. With the departure of all of Portugal’s shop-keepers, and while the MPLA membership was struggling to replace the vanished civil service, FNLA members used their commercial skills to build an informal, parallel, economy which became the Roque Santeiro market, the essential supplement to the struggling planned economy based on Soviet models. The traders, however, remained aliens in the eyes of both the people and the politicians. As late as 2003 the technical colleges of Luanda were teaching their business administration courses in French to reach the second generation of northern returnees who had not yet learnt Portuguese. When the going got tough for consumers the government permitted vigilantes to hunt down 230 david birmingham the so-called Zairotas and burn their market stalls. When the going got tough for government they licensed the fiscal police to harry the merchant sector. National integration was not in evidence. Meanwhile, down the hill at the prince’s court – in the heavily fortified Futungo palace – the mind-set seemed to remain firmly Luso-Brazilian. Is that compatible with African ‘nationalism’?

Post scriptum: Since this paper was written Angola has changed and is now awash with more oil revenue than Nigeria, and the Chinese line-of-credit has risen from $5 billion to $15 billion. As the economy of Portugal declines the old mother-country replaces Brazil as the preferred haven for off-shore investment – as witnessed by the purchase of a Portuguese bank by the president’s daughter at the knock-down price of $50 million, one fifth of its value before the outbreak of the 2011 Euro-crisis. CHAPTER TEN

NATIONALISMS, NATIONS AND STATES: CONCLUDING REFLECTIONS

Gavin Williams*

The study of Africa cannot be insulated from studies of processes and social and political phenomena elsewhere. Interpreting African histories and societies can identify or focus attention on questions arising in other contexts. This is not, as Eric Morier-Genoud explains in introducing the volume, to search for overarching generalisations or to provide an “updated systematic view.” We “need to reject linear, evolutionist and teleological narratives and restore diversity, complexity and uncertainty”. The timely aim of the contributors is to rescue histories from the authority of the victors, or from their denunciation, and to inquire into “historical processes which are contingent, contextual, and fought over”, of which these histories are themselves a crucial and revealing part.

State and Nation as Ideas and Practices

The first task of a study of ‘nations’ and ‘nationalisms’ is to interrogate their meanings, as defined both by historical actors and those who study them. Michel Cahen instructs us here to stop confusing “State, Nation, Nation-state”. What is ‘the state’? The ‘state’ is clearly a ‘social fact’, as described by Emile Durkheim: “… exercising … an external constraint” and “which is general throughout a given society while… existing in its own right, independent of its individual manifestations”.1 Contra Durkheim, the state is not a thing. It is, Philip Abrams explained, “an Idea”. It is the

* This chapter has benefited from the close editorial attention of Rosa Williams. I wish to thank several people for providing me with some understanding of the histories and politics of Angola, Mozambique, and Guinea-Bissau. Among them are my former students Jocelyn Jones, M. Anne Pitcher, João Gomes Cravinho, and Alice Dinerman. I am also grateful to Rosa Williams, Ricardo Soares de Oliveira, Michel Cahen and Eric Morier- Genoud, to the participants in the Oxford workshop on “The Politics of Nations and Nationalisms in Lusophone Africa” in 2007, and the Oxford workshop on “The Study of Angola” in 2011, and to my friend and mentor, David Goldey. 1 Émile Durkheim, Rules of Sociological Method, New York: Free Press, 1982 [1895], p. 52.

“collective misrepresentation of capitalist societies”. “There is a state- system: a palpable nexus of practice and institutions centred in government…”.2 Taken together, they exercise “the monopoly of the legitimate use of physical force over a given territory”;3 the right and capacity to collect revenue and rents; the institutions for classifying people and ordering society; the means for providing resources; and enabling its intermediaries to control the gate4 through which they are distributed. The State may be an ‘Idea’, but it is real enough in its consequences.5 The ‘Nation’ is itself an ‘Idea’, the collective mis/representation, to which nationalists or ‘nationists’,6 attach substantive meanings formed and reformed from disparate materials. It is necessary to the ideologies and projects of nationalism and the purposes for which they can be deployed.7 Nationalisms in Africa had origins and took forms, which were often dissimilar, and even contradictory. Nationalist political discourses were modern and modernist, constructed within political discourses to contest and to legitimate public values and national citizenship.8 They converged on the demand for control of state power which provided instruments for realising the coupled ‘nationalist’ project of bringing political and economic resources under the control of a new “political class”9 and ‘developmental’ project of large-scale manufacturing and farming and providing services to the people.10

2 Philip Abrams, “Notes on the difficulty of studying the state”, Journal of Historical Sociology, Vol. 1, No. 1, p. 48. Linzi Manicom, “Ruling relations: rethinking state and gender in South African history” Journal of African History, Vol. 33, p. 455, cited by Alice Dinerman, Counter-Revolution and Revisionism: the case of Mozambique, 1975–1994, London: Routledge, 2006, p. 47. 3 Max Weber, “Politics as a vocation” in H.H. Gerth and C. Wright Mills (eds), From Max Weber: Essays in sociology, London: Routledge, 1998 [1919], p. 78. 4 Fred Cooper, Africa since 1940: the past in the present, Cambridge: Cambridge University Press, 2002, pp. 5, 156–60 (cited by Cahen, p. 4). 5 W.I. Thomas and D.S. Thomas: “If men define situations as real, they are real in their consequences”, in The Unadjusted Girl: behaviour problems and programs, New York: Knopf 1928, pp. 511–512. Timothy Mitchell, “Society, economy, and the state effect”, in George Steinmetz (ed.), State/Culture, State formation after the cultural turn, Ithaca, NY: Cornell University Press, 1999 (cited by Justin Pearce, p. 203). 6 Luís Cerqueira de Brito, “Une relécture nécessaire: la genèse du parti-État Frelimo”, Politique Africaine, No. 29, 1988, pp. 15–28 (cited by Cahen, p. 24). 7 Patrick Chabal, A History of Lusophone Africa, London: Hurst & Co., 2002, pp. 37–58. 8 See the chapters by Sumich and Pearce in this volume. 9 Ruth First, The Barrel of a Gun. Political Power in Africa and the Coupd d’État, London: Allen Lane The Penguin Press, 1970, p. 104. See Michel Cahen, “Check on socialism in Mozambique – what check? What socialism?”, Review of African Political Economy, No. 57, 1993, p. 50. 10 Gavin Williams, “Reforming Africa: continuities and changes” in Africa South of the Sahara 2004, 33rd ed., London: Europa Publications, 2003, p. 3. nationalisms, nations & states: concluding reflections 233

The nationalist and the developmental projects could be brought together in a common ideology of ‘socialism’, whether of the ‘African’, ujamaa, or Marxist-Leninist varieties.11 In some ways, socialism in Africa was another word for nationalism. Marxism, preferably with Leninism, provided a language for nationalisms that were radical in their demands, actions, and intentions. Nationalism, socialism, and the revolution of the oppressed could all be taken into a teleological idea of a ‘National Democratic Revolution’, directed by a Marxist-Leninist party, which could define the movement for, and also after liberation (Cahen, p. 16). In the 1980s, the idea of Marxism-Leninism and the proclaimed movement towards socialism could be set aside. The ideas of the nation and of the state could not be.

Elites and Progress

Nationalists first needed a nation and the idea and apparatuses of the state, which the colonial power brought into being. Colonial conquest defined the territories within whose boundaries nationalists would ‘imagine’ and give political meaning to the nation and stake their claims to legitimate state authority. In sub-Saharan Africa, only in Tanzania has Swahili been established as an official language which defines a national identity. Elsewhere the national language of élites and political movements and of official discourse and state institutions is inherited from the colonial rulers, whether or not it is understood by the population. In Mozambique, Angola, and Guinea-Bissau only a small per cent of the people speak Portuguese or are able to read and write it (Birmingham, p. 218).12 In Angola, President dos Santos surprised David Birming ham in 2006 when he declared, “Our language is Portuguese.” (Birmingham, p. 217) Nationalists could not write their texts on tabulae rasae. African societies had and have common, similar, diverse and conflicting kinship systems, language, education, religions, social relations, political arrangements, and historical experiences. They engage people in multiplex relationships, which acquire significance in times and places. Within the space of colonial territories, and across their boundaries, they are constructed into pan-ethnic identities. These may be a foundation for,

11 See Chabal, History, pp. 58–64. 12 Cahen, “Check on socialism in Mozambique,” pp. 49–50. 234 gavin williams be invoked, aligned with, assimilated to, or denied or suppressed by nationalist movements but they always raise questions for and offer a challenge to the nationalist project. In the nineteenth and early twentieth centuries, literate coastal élites referred to the liberal traditions of England, France and Portugal in claiming the rights of citizens. The commonalities among them were relational and locational and lay in their own claims to status. The ubiquitous application of ‘Creole’ for different cultural, linguistic and political communities obscures the historical and sociological contexts and the “discursive processes” “within which people identified themselves and one another”.13 In Angola, “the filhos da terra … spoke of themselves in terms of both their civilised practices and their filial relationship with Africa, distinguishing themselves from the ‘corrupt’ white and the ‘primitive’ black”.14 In colonial Guiné-Bissau, as Havik tells us, the Kriston were commercial, military, and cultural intermediaries linking the Portuguese with producers and traders beyond the coast and taking with them Kriol as the language of commercial intercourse (Havik p. 58). In Moçambique, the administrative capital moved south to Delagoa Bay [Lourenço Marques/Maputo], in response to the conjunctures of the opening of the Suez Canal and reduction of coastal shipping, and exports by rail of gold from the Z.A.R. [Zuid- Afrikaansche Republiek], thereby displacing the established élites of the north with a modern élite in Lourenço Marques.15 The filhos da terra, or the Kriston, or the forro of São Tomé, or the élites of Moçambique depended on the Portuguese and vice versa, yet the relations between them and with peoples in the ‘interior’ were always potentially adversar- ial (Cahen, p. 13; Derlugian, pp. 86–7; Sumich 127–8, 133–4). To identify them as proto-nationalists would be anachronistic. Until 1961, the Portuguese administrators defined almost all the people of the colonies as indigenas (natives), subject to customary law and xibalo

13 Rosa Williams, “Migration and miscegenation in the narrative of the Angolan state, 1875–1912,” in Philip J. Havik and Malyn Newitt (eds), Creole Societies in the Portuguese Colonial Empire, Lusophone Studies 6, University of Bristol, 2007, pp. 169–70. Jacopo Corrado, The Creole Elite and the Rise of Angolan Proto-nationalism 1870–1920, Amherst NY: Cambria Press, 2008, p. 236. 14 R. Williams, “Migration and miscegenation”, p. 157. Corrado, The Creole Elite. Corrado, “The fall of a Creole elite? Angola at the turn of the twentieth century: the decline of the Euro-African urban community”, Luso-Brasilian Review, Vol. 47, No. 2, 2011, 100–119. (Fernando Pimenta refers to a quite different category of ‘Euro-Africans’ in his contribution to this volume). 15 Cahen, “Check on socialism in Mozambique”, pp. 49–50. nationalisms, nations & states: concluding reflections 235

(forced labour). Africans could acquire the legal status of assimilados if they had sufficient education and could demonstrate that they had adopted the civilised way of life of the Portuguese colonisers. In between were the ‘mestiço’, neither black nor white and often from the families of the nineteenth century elites, the colonial Europeans, and above them the authentic Portuguese (Pimenta, pp. 183–4). David Birmingham writes: In the last phase of colonial rule the old Angolan Creoles were challenged not only by white immigrants but also by a new class of young men. (…) These ‘new assimilados’ still belonged to African society and spoke Kimbundu in their leisure time, but they had been absorbed into the lower ranks of the colonial establishment by being educated to speak and write in Portuguese.16 In the later colonial period, nationalists identified themselves as the party of progress, overthrowing the colonial masters and bringing the benefits of reason and enlightenment to the people. In Guiné-Bissau, for example, they fought for ‘Unidade, luta e progresso’ (Unity, Struggle, and Progress) which, with no evident irony, became the “national values of the ‘new order”’ (Havik, p. 38) taking in the themes of Brasil’s national motto of ‘Ordem e Progresso’. 17 In Moçambique, Angola, and Guiné-Bissau, local or even regional rejection of, or resistance to, the party of progress could be ascribed to tribalisms and feudal elements; ignorance, obscurantism, and backward traditions; and to racism.18 As Havik observes, “The strategy for hegemony was rooted in the historical claim of Christianised Guineans and Cape Verdean Creoles as privileged intermediaries” (Havik, p. 70). Many of FRELIMO’s secular nationalists were drawn from “subordinate colonial elites”, of assimilados, mestiço, Asian, or Goan origin (Derlugian, p. 86). Sumich emphasises that FRELIMO’s nationalism was a modernist ‘vision of a nation as of yet uncreated, a nation not built on its glorious past but on a radiant future still to come.’ (Sumich, p. 129)

16 David Birmingham, “Angola” in Chabal, History, pp. 147–8. 17 The motto was explicitly derived from Comte’s ‘Positive philosophy’, which connects the development of society from past to present and to a future guided by the sure trusteeship of an ‘industrial patriciate’ [try substituting ‘national liberation movement’!] M. P. Cowen and R.W. Shenton, Doctrines of Development, London: Routledge, 1995, pp. 27–35. For Comte, see Gertrud Lenzer (ed., with Introduction), August Comte and Positivism: the essential writings, New York: Transaction Books, 2nd ed. 1997. 18 For example, Aristides Pereira, Uma Luta, Um Partido, Dois Paises, Lisbon: Ed. Notícias, 2002, pp. 102–12 (cited by Havik, pp. 50; see also Derlugian, p. 91). 236 gavin williams

Nationalisms and Liberation

Political hegemony did not come to the revolutionary leadership of its own accord. The struggles for national liberation from colonial oppression were more than that, and also less than that, differing in their histories and geographies, the character and salience of issues, in their forms of action, and in their local and even national political allegiances. The national political movements were able to respond to specific grievances and experiences of colonial rule. Nationalist movements trace their own origins to specific events, the strike in Pidjiguity docks in 1959 (Havik, p. 40), the firing on petitioners at Mueda in northern Mozambique in 1960, and the Angolan prison-break of 4 February 1961, “the day on which the Angolan people, under the leadership of the MPLA, took the initiative in rising against Portuguese rule” (cited Pearce, p. 204). Derlugian lists the demands of the petitioners who were fired on after they tried to release prisoners at Mueda in northern Moçambique in 1960: an end to xibalo (forced labour), the right to own shops and agricultural cooperatives, and better prices from government for their cotton (Derlugian, p. 94). These atrocities facilitated mobilization but they all preceded the formation of the liberation movements. The national liberation movements sustained and expanded their struggles against colonial armies, conscripted in Portugal and recruited from the colonised population itself. The government of Marcelo Caetano could not sustain the fiscal, military, and political costs of colonial wars nor dislodge the guerrilla armies. Its colonial failures were an immediate reason for the coup by the Armed Forces Movement on 25 April 1974 and Portugal’s ‘Carnation revolution’. Portugal transferred power to independent governments in 1974 and 1975. PAIGC and FRELIMO could take over the capitals and free people from colonial exactions and repression. In January 1975, even before the Portuguese reluctantly conceded Angolan independence, contending parties fought for the prize of the capital city. MPLA was able take control of Luanda when a South African flying col- umn was repelled with Cuban troops and Soviet arms. This did not bring them control of the countryside. The movement of the MPLA and Cuban troops into the planalto gave UNITA the scope to capture political support and make this a ‘“successful” defeat’ (Péclard, p. 173). Péclard argues that UNITA was not an ‘ethnic’ or ‘religious’ movement. He explores carefully the historical intersection of ovimbundu élites and Protestant missions and “how with their educational system, Christian missions were … a place where an élite of new assimilados could emerge without having to turn its back to its culture of origin” (Péclard, p. 150). nationalisms, nations & states: concluding reflections 237

In all the Portuguese colonies and the liberated Lusophone states, the victorious parties claimed their exclusive right to represent all the people, to the extent that they could exercise the administrative capacities and military power to do so. They alone were authentic nationalists, who gave substance to the ‘nation’ and to ‘nationalism’. They displaced the demand for ‘national self-determination’ from the liberal discourse of rights to the claims of nations and their own rights to define and speak for the nation and for the people. This excludes from the public memory predecessors who do not fit the current political mythology and dismisses or marginal- izes rivals or dissidents, past or present (Havik, pp. 43–4; Derlugian, pp. 90–1; Basto, p. 120). MPLA and UNITA contested one another’s conceptions of ‘the nation’. MPLA positioned itself as the sole national party standing above ethnic and regional interests; UNITA claimed to speak for “the Ovimbundu people, and blacks and southerners more generally, who had got the worst of colonial rule, and [claimed] that the MPLA, representing specific regional and class interests was in no position to represent Angola at large” (Pearce, p. 202). Both could be right. MPLA’s claim to be a truly national party gave it African and international credibility that UNITA lacked. The monopoly of representation was secured, first by the PAIGC, and then by FRELIMO and MPLA by taking armed liberation movements to independence and getting Portuguese and international recognition. At the international level, the claims of the official liberation movements went relatively unchallenged. Derlugian concludes: “FRELIMO’s organizational success was clearly related to the inordinate ability of its intelligencia leaders to benefit from global alliances.” (Derlugian, p. 101). PAIGC, FRELIMO and MPLA were able to appeal to the admiration and solidarity of the global left, whose language of national liberation and resistance to imperialism the nationalists could speak fluently. The liberation wars were struggles against the last remnants (almost) of colonialism in Africa, of fascism in Portugal, and its imperialist alliance with NATO and for international solidarity with the liberation of oppressed people. They were able to align with the Soviet bloc while gaining the active support of ‘Africanist’ radicals, the New Left, Trotskyist political factions, and liberal sympathizers. MPLA fought a war with Soviet-bloc arms and Cuban troops to repel the people against an illegitimate rival and the South African armed forces, occupying Namibia, with not-so- covert support from the U.S.A. FRELIMO defended itself from RENAMO, which had been brought into existence by the Rhodesian Special Forces and then armed by the South African military. The choice, not only on the left, was self-evident. 238 gavin williams

Nationalisms need to define the present by narratives that re/present their past origins and project the ‘sure road’ to the future. Integral to the imagining of the nation is to imagine a ‘national culture’. Maria-Benedita Basto brings out what Frantz Fanon calls the “Reciprocal Bases Of National Culture And The Fight For Freedom”.19 Fanon conceived of “the fight for national existence which sets culture moving and opens it to the doors of creativity”.20 In similar vein, Amilcar Cabral stated that “the liberation struggle is, above all, a struggle both for the preservation and survival of the cultural values of the people and for the harmonization and development of these values within a national framework”.21 For Fanon, “The new humanity cannot do otherwise than to define a new humanism both for itself and for others”.22 FRELIMO defined its task in 1972 ‘in the creation of a new man and in the improvement and unification of our culture’.23 In this volume, Maria-Benedita Basto examines Poesia de Combate, first published during Mozambique’s war of liberation in 1971. When it was repub- lished in 1979 the order and even the wording of the poems had changed along with national priorities and the move from a revolutionary party to a national state-party. ‘Venceremos’ replaced ‘O guerrilheiro’ as the opening to the volume; the guerrilheiro was now configured as ‘guide of the people and model for society after independence’ (Basto, p. 109). Nationalists had to find ways to finesse the awkward distance between the “revolutionary élites” and the people they sought to lead.24 Fanon concluded at the end of his scathing analysis of “The Pitfalls of National Consciousness” that “The nation does not exist except in a programme which has been worked out by revolutionary leaders and taken up with full understanding and enthusiasm by the masses…”.25 Amilcar Cabral recognized with unintended foresight, that “the only social sector capable of having consciousness in the first place of the reality of imperialist domination and of handing the State apparatus inherited from that domination is the native petty bourgeoisie. (…) The colonial situation … offers … the

19 Franz Fanon, The Wretched of the Earth, Harmondsworth: Penguin Books, 1967, p. 190, (cited by Basto, p. 117). 20 Fanon, Wretched, p. 193. 21 Amilcar Cabral, Unity and Struggle: speeches and writings, New York: Monthly Review Press, 1979, p. 153. 22 Fanon, Wretched, p. 198. 23 Frelimo, ‘Brochure “1° Seminário Cultural, Dez 71/Jan 72’, 1972 (signed by Armando Guebuza), (cited by Basto, p. 114). 24 Fanon, Wretched, p.161. 25 Fanon, Wretched, pp.163–4. nationalisms, nations & states: concluding reflections 239 petty bourgeoisie the historical opportunity of leading the struggle”.26 Cabral explained that “to lead a people to liberation and progress, the fundamental need was a vanguard” and that the application of party principles required “collective leadership” in the form of ‘democratic centralism’.27 Both Fanon and Cabral, despite themselves, resolve the problem into one of political and moral commitment, from which nationalisms cannot extricate themselves. As Michel Cahen writes in this volume, ‘This project of the nation is always closely linked to a paradigm of authoritarian modernization.’ (Cahen, p. 22).

Party-States and Economies

At independence, parties translated the liberation war into a party-state, ruling for and in the name of the people. In taking over the central institutions of the state, parties separated themselves from the people in whose names they claimed to rule (Havik, pp. 55–7).28 At the time of independence, PAIGC did not command uncontested allegiance among the Fula. In Mozambique, Malyn Newitt writes, “FRELIMO … had no organized presence in most of the country and had already experienced opposition amongst the largest ethnic group, the makua-speaking people”.29 They had to be brought into the nation by the powers of the state. The successful liberation movements could mould a public representation of nationhood and citizenry but could only fail to eliminate local particularity in its post-independence pursuit of administrative uniformity. The liberation wars in Guinea-Bissau and Mozambique engaged the guerrilla movements with local people, and also with social and political structures and the colonial administration with which they were often unfamiliar. Pursuit of revolution from rural bases required a degree of local accountability. The exercise of state authority did not. The Guinean regime “stole the State from us … now … our voice is not heard there”.30 In 1977, the parties in power in the Lusophone countries all declared themselves Marxist-Leninist in their political systems, the economic strategies, and to a greater (Angola) or lesser (Guinea-Bissau) extent, their

26 Cabral, Unity and Struggle, pp. 134–6 27 Cabral, Unity and Struggle, pp. 85, 246. 28 Rosemary Galli and Jocelyn Jones, Guinea-Bissau: Politics, Economy, Society, Boulder CO: Westview Press, 1987. Forrest ‘Guinea-Bissau’ in Chabal, History, pp. 236–251. 29 Malyn Newitt, “Mozambique” in Chabal, History, p. 196. 30 Balanta elder, cited by Lino Bicari (in Havik, p. 56). 240 gavin williams international alignments. Ideology did matter. It legitimated and also gave meaning to the party’s monopoly of office, privilege, international recognition, and use of violence. It gave direction to and justified state- directed industrial policies, abortive attempts at collective farming, and control of currency exchange, foreign imports, and food prices. The consequences of these policies were extremely repressive, most visibly in Angola in the purge and massacres that ensued after the failure of Nito Alves brutal and unlikely ‘leftist’ attempt to seize power (Birmingham, pp. 222–3),31 and under the dictatorial regime of ‘Nino’ Vieira in Guinea- Bissau (Havik, pp. 55–6). Nationalist economic strategies, pursued in the name of socialism, involved extensive state ownership, controls of prices, imports and foreign exchange. Economically, they were predictably disastrous. They con- tracted the rural and urban economies and increased dependence on imports of subsidised grain from the U.S.A.32 Agricultural policies were met by evasion and obstruction. Price controls reduced prices to producers and increased the prices paid by consumers and were only mitigated by black markets and smuggling. Industrial policies increased imports and reduced production.33 As Manuel Ennes Ferreira argues, Angola’s ‘Strategy of Industrial Transformation’ to “substitute for imports … transformed itself exactly into its opposite, giving origin to a policy, de facto, of import promotion”.34 In 1977, FRELIMO committed itself to a “democratic centralism … extended to all the society” and to state developmentalism.35 Its policies followed colonial precedents, with admixtures of Tanzanian villagization (Cahen, p. 21),36 and Leninist conceptions of the transformation of

31 Also Birmingham, “Angola”, pp. 163–4. For a detailed contemporary account of the politics of the coup, but not of its consequences, see Paul Fauvet, “Nitista coup in Angola”, Review Of African Political Economy, No.9, 1978, pp. 85–104. 32 Gunilla Andrae and Björn Beckman, The Wheat Trap: Bread and Underdevelopment in Nigeria, London: Zed Books, 1985. Philip Raikes, Modernizing Hunger, London: James Currey, 1988. 33 Manuel Ennes Ferreira, A indústria em tempo de Guerra (Angola, 1975–91), Lisbon: Edicões Cosmos, 1999, p. 468, Fig. 9 & 10; Anexos VI. 34 Ennes Ferreira, A indústria, p. 485. 35 Cahen, “Check on socialism in Mozambique”, p. 48. 36 Newitt, “Mozambique”, pp. 200–6. Andrew Coulson, “Agricultural policies in mainland Tanzania,” Review of African Political Economy, No.10, pp. 74–100, and in Judith Heyer, Pepe Roberts and Gavin Williams (eds), Rural Development in Tropical Africa, Basingstoke: Macmillan, pp. 52–89 (see also the contributions on other African countries). nationalisms, nations & states: concluding reflections 241 industry and agriculture.37 FRELIMO moved peasants into communal villages (aldeias communais), thereby reducing production and sales. They went back to the colonial strategy of requiring peasants to deliver their cotton quotas to the cotton buyers, João Ferreira dos Santos and state co-operatives.38 Party officials did not succeed in getting peasants to do what they wanted but they did, predictably, reduce agricultural production and alienate the peasant producers from FRELIMO’s new state. Repressive one-party and disastrous state-controlled development strategies were not peculiar to the Lusophone countries. One-party states were ubiquitous in post-colonial Africa, whether they were of ‘conservative’, ‘socialist’, or ‘Marxist’ varieties. They brought opposition politicians and parties into the fold, and repressed or excluded outsiders, if only pour encourager les autres. For thirty years, they enabled a number of presi- dents, usually of a venerable age, to oversee the management of political institutions and contests for political office and its rewards. Party and military elites appropriated the rewards of power, on a quite extraordinary scale in oil economies such as Nigeria and Angola. Office-holders at national, provincial, and local levels could act as ‘gatekeepers’39 of international, public and private resources, taking most for themselves and leaving hardly anything over for anybody else. Their dependence on these ‘gatekeepers’ at higher levels precluded independent or collective action on their own part.40 These policies were adopted widely though not universally in Africa for some of the same reasons and with most of the same consequences. State finances imploded in the 1980s, as a result of perverse policies, unpayable debts from foreign banks, high real interests, and global economic recession. International agencies responded by insisting on structural adjustment policies as conditions of loans. Imposing the conditions and putting

37 Was FRELIMO’s Marxism ‘Stalinized’ as Cahen describes it? Lenin’s Marxism might do as well (see Derlugian, p. 99). For the Bolshevik vision of the future, see N. Bukharin and E. Preobrazhenskii, The ABC of Communism, London: Penguin, 1969 [1922]. 38 Dinerman, Counter-Revolution and Revisionism, pp. 96–106, 201–12. Cahen, “Check on socialism in Mozambique,” p.34. João Gomes Cravinho, “Modernizing Mozambique: Frelimo ideology and the Frelimo state”, D Phil. Thesis, University of Oxford, 1995. 39 Cooper, Africa since 1940, p. 5. Chris Allen, “Understanding African politics”, Review Of African Political Economy, No. 65, 1995, pp. 301–320. 40 Sam Nolutshungu, “Fragments of a democracy: reflections on class and politics in Nigeria”, Third World Quarterly, Vol. 15, No. 1, 1992,101. See also the debates excerpted in Chris Allen and Gavin Williams (eds), Sociology of Developing Societies: Sub-Saharan Africa, Basingstoke: Macmillan, 1982, pp. 161–202. 242 gavin williams the policies into practice was not as straightforward as agencies and their critics assumed. Nationalist policies gave way to economic reforms in 1983 and 1987 in Mozambique, 1986 in Guinea-Bissau, 1987 in São Tomé and Principe, and 1988 in Angola (partially implemented and mainly postponed), each with the support of the IMF and, except in Angola, under its supervision.41 ‘Liberal’ economic reforms created new channels for profit-making by people with overlapping political and economic connection to parties, governments, and aid agencies.42 Aid flows from government and non- governmental organisations created new ‘gates’ and a wider class of political beneficiaries. The policies raised the stranglehold of state control on economic activity. Predictably, they did not attract direct foreign investment. They also raised living costs and left African producers unprotected from international competition and grain and rice imports subsidised by the United States and the European Union. The benefits and beneficiaries of adjustment loans have depended on continuing financial transfers to meet current expenditures, locking donors and beneficiaries into their mutual relationships. By 1990, one-party and military regimes across Africa faced public opposition to their rule, economic collapse, resistance to the costs of adjustment for ordinary people, and the withdrawal of support from or the desertion of their international patrons. Their loss of credibility opened space for competitive party politics, although the informal rules of the political game, the skills to play it and to adapt the rules to purpose continued. These processes took place over briefer periods in Lusophone countries and in the contexts of civil war in Angola and Mozambique. Military superiority and control of oil revenues put MPLA in political dominance in Angola. ‘Liberation movements’ overthrew African regimes in Ethiopia, Eritrea, Uganda, and Rwanda, following a ‘Lusophone model’ of authoritarian political rule by leaders of the liberation movements. Competitive politics could open the way to political violence and civil war in, for example, Côte d’Ivoire and Guinea Bissau, part of the interlinked political violence of west African countries.43

41 Chabal, History. 42 M. Anne Pitcher, Transforming Mozambique: the politics of privatization, Cambridge: Cambridge University Press, 2002. Pitcher, “Conditions, commitments and the politics of restructuring in Africa”, Comparative Politics, Vol. 37, 2004, pp. 375–98. 43 Chabal, History, pp. 88–136. Michael Cowen and Liisa Liaakso (eds), Multi-party Elections in Africa, Oxford: James Currey, 2002. A.R. Mustapha and Lindsay Whitfield (eds), Turning Points in African Democracy, Oxford: James Currey, 2009. For an earlier, nationalisms, nations & states: concluding reflections 243

The same political elites who assumed power after the liberation could first prescribe Marxism-Leninism and then neo-liberalism. In Angola and Mozambique, “The state itself – site of its own reproduction – had a central economic role”.44 It continued to do so in a liberal economic and ‘dominant party’ political system.45 State developmentalism can give way to capitalist developmentalism. Social networks and political connections gave access to education and status. Money became a more direct means for acquiring and reproducing economic position, status and styles of life (Sumich, pp. 133–4).

Successes and Failures

If the liberation movements were so successful in their battles for independence, why was it so difficult for them to hold on to power? The sim- plest reason, valid as far as it goes, is that they were subject to armed foreign intervention. An alternative explanation, also valid, is their disastrous economic policies, which forced the citizens to bear the scarcities of the controlled economy and then the costs of economic liberalization. The contributions to this book show that it was more complicated than either. A second question is why, if their rule had such devastating consequences, did the MPLA and FRELIMO though not, ultimately, the PAIGC hold on to power? Again there are simple, valid, reasons. They managed the state administration and controlled its security services; they had international recognition; and they could impose military force, their own and that from allies such as Cuba in Angola, and Zimbabwe in Mozambique. Again, as we see from this book, there is more to it than that. In answering them, we must address the further question of the relevance of ‘Lusophone’ Africa as a framework for and subject of analysis. How far are the experiences of Angola, Mozambique and Guinea- Bissau comparable to one another, other than in their shared participation in Comunidade dos Países de Língua Portuguesa (CPLP)? Apparent

very condensed view, see Gavin Williams, “Democracy as idea and as process”, Journal of African-American History, Vol. 86, No. 4, 2003, pp. 343–351. 44 Cahen, “Check on socialism”, p. 50. 45 Eric Morier-Genoud, “Mozambique since 1989: Shaping democracy after Socialism” in A.R.Mustapha & L.Whitfield (eds), Turning Points in African Democracy, Oxford: James Currey, 2009, pp.153–166. 244 gavin williams similarities may hide divergences, parallels, and convergences in institutions and processes. Institutions have a remarkable capacity to perpetuate their forms, across changes in ideologies, policies, and international political alignments. State institutions regulate the ways in which formal and informal official business is conducted and thereby teach the state’s servants that these are how things can be and are to be done. It is easier to change institutional forms than to transform them in substance and, anyway, new ‘states’ face many of the problems of the old ones in administering the system of state institutions. Angola, Mozambique and Guinea-Bissau shared a common heritage of Portuguese colonial rule, wars of national liberation, and dates of independence. The Estado Novo imposed a similar system of administrative rule, political repression, legal and social stratification, and an imperial economic strategy on all three colonies. The ruling parties followed the Salazar state in applying the use of force, legitimately or illegitimately, to manage or suppress any real or imagined threats to the enforcement of their authority. The capital cities from which the colonial power adminis- tered the colonies and imposed its rule became the centres over which the nationalist revolutionaries first had to take control and from which, like the Portuguese government before and after 1900, they had to extend the authority of the ‘state of the revolution’ across the national territory and to sustain it through the exercise of force. There were important differences in the geographies and histories of the three colonies. In very simple terms, Angola had an established European community and subsequently extensive Portuguese displacement of people from the land to make way for Portuguese settlers. Railways linked the coast to interior and to the copper mines in the Congo. There was a much higher level of industrial development and agricultural exports than in Mozambique. Mozambique was integrated into the economies of South Africa and Rhodesia through labour migration and rail connections. There had a smaller number of settlers, a small but significant Indian merchant community, fewer exports than Angola, and a lack of transport connection from the north to the south of the Colony. It is hard to see what the Portuguese were doing in Guinea-Bissau, other than asserting their credibility as a colonial power, and why they held on to it at such cost. Instead of keeping Angola and Mozambique in Por tuguese hands, the war for Guinea-Bissau was ironically crucial to the Portuguese coup and revolution in 1974. nationalisms, nations & states: concluding reflections 245

In both Angola and Mozambique, African social and political elites were defined and also stratified by the status arising from their occupation, their competence in Portuguese, and their social origins and geographical locations. In Guinea-Bissau, the local elites were Cape Verdeans, recruited to fill minor administrative positions. Comparisons may be found in the coastal elites in Francophone and Anglophone colonies and in South Africa, in their claims to political voice and status recognition, and the subsequent prominence of later generations in nationalist movements. In some respects, what the three colonies shared, Angola and Mozambique in particular, was not what they inherited from the colonial period but from the centralized political and economic institutions and their ideological underpinnings after 1974 and their abandonment in form and to a much lesser extent in practice at the end of the 1980s. Civil wars and external interventions, from South Africa and Rhodesia in particular, threatened the security of both Angola and Mozambique but this alone does not explain why UNITA and RENAMO, despite their own violence towards local populations, should have been able to claim so much support in extensive and politically and militarily strategic parts of the two countries. Ruling parties created the conditions for their own opposition. National political elites, centred in the capital cities, better educated and at least in their own view better qualified to rule, left or placed their provincial counterparts in a secondary position. Generation divided the older and new party elites rather than ‘traditionalism’ and ‘modernism’. Between 1975 and 1979, FRELIMO abolished the colonial regulos and replaced them with party secretaries, though the shift in generation did not necessarily replace networks of chiefly authority, which were revalidated as bearers of cultural values and administrative intermediaries.46 Péclard explains that after MPLA and Cuban forces captured Huambo, MPLA brought in younger cadres as the “new forces of the nation” rather than incorporate the local mission-educated “cultural nationalists”, clerks, school teachers and nurses into the local administration (Péclard, p. 172). They were both ‘modernizers’ but their conceptions of a modern nation and a modern society grew from their different origins, situations and, as Péclard shows in the case of the Ovimbundu, cultural orientations.

46 See, for example, Dinerman, Revolution, Counter-Revolution and Revisionism, pp. 193–234. 246 gavin williams

All three countries were involved in civil wars, extending beyond their own boundaries, but the timing of the conflicts differed and the participants were not the same apart from the engagement of South African armed forces in both Angola and Mozambique. Here too, there were important differences. The South African government wanted to establish its hegemony over the southern African region but its primary concern in Angola was to ensure its occupation of Namibia. The turning point in the war came in 1988 when the Cuban and Angolan forces repelled the South African offensive at Cuito Cuanavale leading to a peace agreement at the end of that year. In Mozambique, the apartheid regime took over Rhodesia’s alliance with its client RENAMO, and continued its support even after the Nkomati accord with Mozambique in 1983. It did not engage its own forces in any direct battles nor did it need to do so. In both Angola and Mozambique, civil conflicts acquired their own dynamics so that internal peace agreements only took place in 1992 (Derlugian, pp. 83–84; Pearce, p. 205).47 PAIGC ruled Guinea-Bissau under Luís Cabral and, after a coup in 1980, under João Bernardo ‘Nino’ Viera, until he was himself deposed in 1998. Subsequent contentions for power and office were decided by military coups, armed conflicts, troops from Senegal and Guinea-Conakry, arms smuggling and its proceeds, abortive or completed elections, personal and ethnic rivalries and political assassinations, the last of a sequence being of Vieira himself, who had returned to be elected to the presidency in 2005 (Havik, p. 73).48 What the disparate histories of all three countries have in common is the decisive place of military force in deciding the battles for political office, and the different capacities of the parties of liberation to sustain their control over the machinery of government, thanks to or despite external interventions. Two decades and one, or two, civil wars later, FRELIMO and MPLA remain in power. They have confirmed and even legitimated their political domination and kept the military forces strictly under their authority. In Angola, the 1992 elections divided town and country. According to David Birmingham, “The countryside voted for the opposition, for Savimbi and for change, while the towns voted for the government, for preferential economic treatment, and for armed protection from the hungry raid- ers out in the rural areas.” The outcome of the elections reopened armed

47 Birmingham, “Angola”, p. 164–172. Newitt, “Mozambique”, pp. 229–234. 48 Joshua Forrest, “Guinea-Bissau” in Chabal, History, pp. 236–263. Africa South of the Sahara, 2012. nationalisms, nations & states: concluding reflections 247 conflicts, which could and did only come to an end with the death of Jonas Savimbi and the public display of his corpse in 2002 bringing to an end the “depraved conflict between a corrupt government mesmerized by wealth and inhuman opposition obsessed by power”.49 Both Angola and Mozambique may be described as ‘dominant-party’ systems. They are able to disperse patronage, on a far greater scale in Angola than in Mozambique, and bring the political process itself within their own political orbits. The regime in Guinea-Bissau not only had few resources but also suffered under a party dictatorship which took more and more of less and less until there was nothing much left except the proceeds of cross-border arms trading. FRELIMO and MPLA share many of the features of the more successful single-party states of the period from 1960 to 1990 and of more recent states with multi-party elections, where legislative elections accompanied presidential rule, with or without elections for the presidency. Since 2002, both MPLA and FRELIMO have extended and consolidated their legislative domination. MPLA secured an overwhelming majority of votes cast for the legislature in 2008. Elections are scheduled for 2012. FRELIMO were surprised at the fact that RENAMO was able to get more votes than FRELIMO in five of the country’s provinces in 1994 and six in 1999, but probably should not have been. FRELIMO’s lack of a political presence in the central provinces, as Newitt remarks, preceded independence. In subsequent elections, FRELIMO got an electoral stranglehold, so that it defeated RENAMO in all eleven provinces in 2009, with some unnecessary help from the administration.50 In any political system, the most revealing question is always where does the money come from and where does it go? In Mozambique, the relations among the political classes and with national and international aid donors became more complex under a ‘liberal’ political and economic dispensation. The Mozambican state and the political elites depend on aid from numerous foreign governmental and non-governmental agencies to pay for the recurrent budget and development projects. Aid transfers were 23 per cent of Mozambique’s gross national income in 2004, of which 80% were grants, and debt relief made up 22%. Mozambique’s economic recovery has made it one of the star pupils of the syllabus

49 Birmingham “Angola”, pp. 177, 182. 50 Newitt, “Mozambique”, pp. 224–4 (1994), 232–3 (1999). Electoral Institute for the Sustainability of Democracy in Africa, http://www.eisa.org.za/WEP/mozelectarchive.htm and http://www.eisa.org.za/ WEP/angola.htm. 248 gavin williams prescribed by the international agencies. As in many other African countries, external advisers engage directly in the policy-making process. Nevertheless, its international funders cannot choose to withdraw and just dictate policies unilaterally. Both sides must engage in games of strategic bargaining.51 Party and state institutions act in combination as national and international gatekeepers. Factional politics and contests for spoils take place within FRELIMO. Political and economic interests have to be accommodated and resources more widely shared in Mozambique than in Angola. This is possible because, as Sumich comments, “Education and status tend to reinforce each other” in Mozambique “… which allows access to extremely powerful social networks” (Sumich, p. 145). Angola is more straightforward in its sources of revenue: oil, and more recently Chinese lines of credit (Birmingham, p. 226).52 Here too rivalries among the political elites for economic shares take place within the ruling party and state. The scale, though, is of a quite different order to Mozambique or Guinea-Bissau. Angola paid for its civil war with revenues from oil extracted from Cabinda enclave under the protection of Cuban troops to Gulf Oil, later Chevron-Gulf (Birmingham, p. 229). In Angola, oil revenues are controlled from the centre. They make it possible for President Dos Santos to create a ‘presidential state’ and avoid direct elections, and for the close-knit ruling elites to avoid any accountability to clients, let alone voters, postpone economic reforms, appropriate public funds, export their winnings to foreign banks, and import food cheaply for urban consumers.53 Angola’s ruling elites face outwards, economically, culturally, and in their conceptions of their places in the world. Yet, they fit too well Frantz Fanon’s characterization of the ‘national bourgeoisie’ as “completely canalized into activities of the intermediary type. Its innermost vocation seems to be to keep in the running and be part of

51 Paolo di Renzo and Joseph Hanlon, “Mozambique: contested sovereignty? The dilemmas of aid dependence” and other chapters in Lindsay Whitfield (ed.), The politics of Aid: African strategies for dealing with donors’, Oxford: Oxford University Press, 2009. Paul Mosley, Jane Harragin, and John Toye, Aid and Power: the World Bank and policy-based lending, 2nd ed. 1995, London: Routledge, 1995. Graham Harrison, The World Bank and Africa: the construction of governance states, London: Routledge, 2004. 52 Manuel Ennes Ferreira, “China in Angola: just a passion for oil?” and Soares de Oliveira, “Making sense of Chinese oil investment in Africa”, in Chris Alden, Daniel Large, and Ricardo Soares de Oliveira (eds), China Returns to Africa: a rising power and a continent embrace, London: C. Hurst, 2009, pp. 295–318, 83–110. 53 Ricardo Soares de Oliveira, Oil and Politics in the Gulf of Guinea, London: C. Hurst, 2007, esp. pp. 49–62, 123–59. Birmingham, “Angola”, pp. 162–3, 181–3. nationalisms, nations & states: concluding reflections 249 the racket”.54 Ricardo Soares de Oliveira identifies Angola as one of the “successful failed” petro-states in which the “political process will be unstable and fragmentary, but the structure of politics itself will be stable and viable – as long as oil lasts”.55

Lusophone Africas in Comparative Perspectives

There are naturally continuities in the distinctive histories of each country before, at the time of, and after independence, but many of these have at different times taken participants and observers by surprise. Similar institutional forms and historical processes are common to all three and more particularly to Angola and Mozambique. These are as much or more than the result of changes after independence as attributable to their colonial legacies. Their time sequences differ and their apparent similarities may be misleading. The political institutions and economic strategies that the Lusophone countries adopted and their unsustainabilities cannot be attributed just to the malign influence of Marxism-Leninism. They were shared with many other African countries, whose politics were more or less stable and economic policies less or more disastrous. Tanzania, Mozambique’s closest neighbour and ally, exemplified both political stability and economic failure. Mozambique benefited from the same structural adjustment policies as most African countries. There was more scope for economic growth after recovering from a catastrophic war and more international budgetary support than in other African countries. Angola’s kleptocracy did not depend on foreign aid to keep the President and his revolving circle of clients in the money they became accustomed to. It obviously shared some features with the political economy of Nigeria, except in being more centralized and not under the same constraints to share the spoils with state-based power holders in a federal state. African parallels are to be found in different ways for the similar and diverse histories of each of Angola, Guinea-Bissau, and Mozambique, even if Angolans, do not always want to be told so. Corrupt politics is a feature of all of these though with far greater scope and concentration in Angola. Centralization of power and deference to those who exercise it are hardly exclusively African. Corruption is embedded in many political

54 Fanon, Wretched, p. 122. 55 Soares de Oliveira, Oil and Politics, pp. 61–62. 250 gavin williams systems and links oil corporations blatantly with African governments and political elites. Put simply, divergencies, parallels, and convergencies can be found where we look for them. Things are never quite the same from one case to another. The revealing examples may appear where we do not expect to find them. THEMATIC BIBLIOGRAPHY

All Three Countries Abshire, David and Samuels, M.A., Portuguese Africa: a Handbook, New York: Praeger, 1969. Afonso, Aniceto and Gomes, Carlos de Matos, Guerra Colonial, Lisbon: Notícias, 2000. Anders, Armelle, Histoire de l’Afrique lusophone, Paris: Chandeigne, 1994. Andrade, Mário de, Origens do nacionalismo africano: continuidade e ruptura nos movimentos unitários emergentes da luta contra a dominação colonial portuguesa, 1911–1961, Lisbon: Dom Quixote, 1997. Antunes, José Freire, O império com pés de barro: colonização e descolonização: as ideologias em Portugal, Lisbon: Dom Quixote, 1980. Birmingham, David, Portugal and Africa, London: Macmillan 1999. Buijtenhuijs, Robert, “People’s War in Africa: the quest for ‘movements of maturity’”, Africa, Vol. 59, No. 3, 1989, pp. 381–90. Cahen, Michel, Bourgs et Villes en Afrique lusophone, Paris: L’Harmattan, 1989. ——. “Sur quelques mythes et réalités de la colonisation et de la décolonisation portugaise” in Ch.-R. Ageron & M. Michel (eds), Les décolonisations comparées. Actes du Colloque d’Aix-en-Provence, Paris, Karthala, 1995, pp. 333–351. ——. “Salazarisme, fascisme et colonialisme. Problèmes d’interprétation en sciences sociales, ou le sébastianisme de l’exception”, Portuguese Studies Review, Vol. 15, No. 1, 2007, pp. 87–113. Campbell, Bonnie K., Libération nationale et construction du socialisme en Afrique: Angola, Guineée-Bissau, Mozambique, Montréal: Editions Nouvelle Optique, 1977. Cann, John P., Counterinsurgency in Africa: The Portuguese way of war 1961- 1974, Westport: Greenwood, 1997. Castelo, Cláudia, Passagens para África: o povoamento de Angola e Moçambique com naturais da metrópole, Oporto: Edições Afrontamento, 2007. ——. O modo português de estar no mundo. O Luso-tropicalismo e a ideologia colonial portuguesa (1933–1961), Oporto: Afrontamento, 1998. Chabal, Patrick and al. (eds), A History of Postcolonial Lusophone Africa, London: Hurst, 2002. Chabal, Patrick, Newitt, Malyn and McQueen, Norrie (eds), Community and the State in Lusophone Africa, London: King’s College London, 2003. Chilcote, Ronald H., Emerging Nationalism in Portuguese Africa: Bibliography of Docu mentary Ephemera Through 1965, Stanford: Hoover Institute Press, 1969. ——. Emerging Nationalism in Portuguese Africa: documents, Stanford: Hoover Institute Press, 1972. Clarence-Smith, Gervase, The Third Portuguese Empire, 1825–1975. A Study in Economic Imperialism, Manchester: Manchester University Press, 1985. Cornwall, Barbara, The bush rebels: a personal account of black revolt in Africa, New York: Holt, Rinehart and Winston, 1972. Duffy, James, Portuguese Africa, Cambridge: Harvard University Press, 1959. ——. Portugal in Africa, Cambridge: Harvard University Press, 1962. Ferreira, Eduardo Sousa, O fim de uma era: o colonialismo português em África, Lisbon: Sá da Costa, 1977. Gowan, Susan Jean, Portuguese-speaking Africa 1900–1979. A Select Bibliography, Braamfontein: South African Institute of International Affairs, 1982–1983, 4 vols. 252 thematic bibliography

Hamilton, Russell G., Voices from an Empire: A History of Afro-Portuguese Literature, Minneapolis: University of Minnesota Press, 1975. Havik, Philip J. and Newitt, Malyn (eds), Creole Societies in the Portuguese Colonial Empire, Bristol: University of Bristol (Lusophone Studies 6), 2007. Hanlon, Joseph, Beggar your neighbours: apartheid power in Southern Africa, London: James Currey, 1986. Henriksen, Thomas H., “People’s War in Angola, Mozambique, and Guinea-Bissau”, Journal of Modern African Studies, Vol.14, No.3, 1976, pp. 377–399. ——. “Some Notes on the National Liberation Wars in Angola, Mozambique, and Guinea- Bissau”, Military Affairs, Vol. 41, No. 1, 1977, pp. 30–37. Keese, Alexander, Living with Ambiguity: Integrating an African Elite in French and Portuguese Africa, 1930–61, Stuttgart: Franz Steiner Verlag, 2007. Kitchen, Helen (ed.), Angola, Mozambique, and the West, New York: Praeger, 1987. Lloyd-Jones and Costa Pinto, Antonio (eds), The Last Empire: Thirty Years of Portuguese Decolonisation, Bristol: Intellect, 2004. MacQueen, Norrie, The Decolonization of Portuguese Africa: Metropolitan Revolution and the Dissolution of Empire, London: Longman, 1997. Mateus, Dalila Cabrita, A Luta pela Independência: A Formação das Elites Fundadoras da FRELIMO, MPLA e PAIGC, Lisbon: Editorial Inquérito, 1999. ——. A PIDE/DGS na Guerra Colonial (1961–1974), Lisbon: Terramar, 2004. Meyns, Peter, Demokratie und Strukturreformen im portugesieschsprachigen Afrika: die Suche nach einem Neuanfang, Freiburg: Arnold Bergstraesser Institut, 1992. Minter, William, Apartheid’s contras: an inquiry into the roots of war in Angola and Mozambique, London: Zed Books, 1994. Newitt, Malyn, Portugal in Africa: The Last Hundred Years, London: C. Hurst, 1981. Pélissier, René, Le naufrage des caravelles. Études sur la fin de l’empire portugais (1961–1975), Orgeval: éditions Pélissier, 1979. ——. Angola, Guinées, Mozambique, Sahara, Timor, etc.: une bibliographie internationale critique (1990–2005), Orgeval: Editions Pélissier, 2006. Pitcher, Anne, Politics in the Portuguese Empire: The State, Industry and Cotton, 1926–1974, Oxford: Oxford University Press, 1993. Rosas, Fernando (ed.), A Transição Falhada: O Marcelismo e o Fim do Estado Novo, Lisbon: Editorial Notícias, 2004. Schneidman, Witney W., Engaging Africa: Washington and the fall of Portugal’s colonial empire, Lanham: University Press of America, 2004. Smith, Alan K., “Antonio Salazar and the Reversal of Portuguese Colonial Policy”, Journal of African History, Vol. 15, No. 4, 1974, pp. 653–67. Torgal, Luís Reis Pimenta, Fernando Tavares, and Soares Sousa, Julião (eds), Comunidades Imaginadas. Nação e Nacionalismos em África, Coimbra: Imprensa da Universidade de Coimbra, 2008.

Guinea-Bissau Amado, Leopoldo (2005), “Guiné-Bissau: 30 anos de independência”, Africana Studia, No. 8, 2005, pp. 109–35. Andrade, Mario de, Amilcar Cabral. Essai de biographie politique, Paris: Maspero, 1980. Andreini, Jean-Claude and Lambert, Marie-Laure, La Guinée-Bissau: d’Amilcar Cabral à la la reconstruction nationale, Paris: L’Harmattan, 1978. Augel, Johannes and Cardoso, Carlos, Transição democrática na Guiné-Bissau e outros ensaios, Bissau: INEP, 1996. Bordonaro, Lorenzo Ibrahim, Living at the Margins: youth and modernity in the Bijagó islands (Guinea Bissau), PhD thesis, Lisbon: Instituto Superior das Ciências de Trabalho e da Empresa (ISCTE), 2006. thematic bibliography 253

——. “Modernity as Marginality: The making and the experience of peripherality in the Bijagó islands (Guinea-Bissau)”, Cadernos de Estudos Africanos, No. 18/19, 2010, pp. 117–137. Cabral, Amilcar, Unity and Struggle: speeches and writings, New York: Monthly Review Press, 1979. Cabral, Luís, Crónica da Libertação, Lisbon: O Jornal, 1984. Cardoso, Carlos, “A Transição Democrática na Guiné Bissau: um parto difícil”, Soronda (INEP, Bissau), No. 17, 1994, pp. 5–29. ——. A Formação da Elite Politica da Guiné-Bissau, Lisbon: Centro de Estudos Africanos do ISCTE, 2000. Carvalho, Clara, Ritos de Poder e a Recriação da Tradição: os régulos Manjacos da Guiné Bissau, PhD thesis, Lisbon: Instituto Superior de Ciencias do Trabalho e da Empresa (ISCTE), 1998. ——. “A revitalização do poder tradicional e os regulados manjaco da Guiné-Bissau”, Soronda (Nova Série; INEP, Bissau), 7, 2003, pp. 7–44. ——. “Local Authorities or Local Power?” in Local Experiences of Conflict Management, Soronda (INEP,Bissau), special issue, 2008, pp. 39–56. Castanheira, José Pedro, Quem Mandou Matar Amílcar Cabral?, Lisbon: Relógio d’Água, 1999. Chabal, Patrick, “The Social and Political Thought of Amilcar Cabral: a reassessment”, Journal of Modern African Studies, No. 19, 1981, pp. 31–56. ——. “Party, State and Socialism in Guinea Bissau”, Canadian Journal of African Studies, Vol. 17, No. 2, 1983, pp. 189–210. ——. “Revolutionary Democracy in Africa: the case of Guinea Bissau”, in P. Chabal (ed), Political Domination in Africa: Reflections on the Limits of Power, Cambridge: Cambridge University Press, 1986, pp. 84–108. ——. Amílcar Cabral: revolutionary leadership and people’s war, London: Hurst & Co., 1983. Challiand, Gérard, Armed struggle in Africa: with the guerrillas in ‘Portuguese’ Guinea, New York/London: Modern Reader, 1969. Có, João Ribeiro Butiam, Representação e confinação de estruturas sociais na Guiné-Bissau: uma abordagem sobre conflitos e consensos, Lisbon: Instituto Superior de Economia e Gestão (ISEG), Working Paper No. 10, 2010. Crowley, Eve, Contacts with the Spirits: religion, asylum and ethnic identity in the Cacheu region of Guinea Bissau, PhD thesis, Ann Arbor: Yale University, 1990. ——. (1998) “Regional Spirit Shrines and Ethnic Relations in Guinea Bissau”, Africana Journal, No. 17, 1998, pp. 27–39. Cunningham, James, “The Colonial Period in Guiné”, Tarikh, Vol. 6, No. 4, 1980, pp. 31–46. ——. “Guinea Bissau, 1956–1974: a reassessment”, London: Institute of Commonwealth Studies, 1980. Davidson, Basil, The Liberation of Guiné: Aspects of an African Revolution, Harmondsworth: Penguin, 1969. ——. No Fist is Big Enough to Hide the Sky. The Liberation of Guinea-Bissau and Cape Verde, London: Zed Books, 1981. Davidson, Joana, “Plural Society and Inter-ethnic Relations in Guinea-Bissau”, in R.A. Shweder, M. Minnow and H.R. Markus (eds), Engaging Cultural Differences: The Multicultural Challenge in Liberal Democracies, New York: Russell Sage Foundation Press, 2002, pp. 417–431. ——. “Native Birth: Identity and Territory in Postcolonial Guinea Bissau, West Africa”, European Journal of Cultural Studies, Vol. 6, No. 1, pp. 37–54. Dhada, Mustafah, Warriors at Work: how Guinea was really set free, Niwot: University Press of Colorado, 1993. 254 thematic bibliography

Drift, Roy van der, Arbeid en Alcohol: de dynamiek van het gezag van de oudste bij de Balanta-Brassa in Guinee Bissau, Leiden: CNWS, 1992. ——. “Democracy: Legitimate Warfare in Guinea Bissau”, Lusotopie 1999, Paris: Karthala, 1999, pp. 225–40. ——. “Birds of Passage and Independence Fighters: an anthropological analysis of Balanta migration to southern Guinea-Bissau and mobilization for the liberation war”, in G. Gaillard (ed.), Migrations anciennes et peuplement actuel des côtes guinéennes, Paris: L’Harmattan, 2000, pp. 151–164. Fernandes, Raúl Mendes, “Partido Único e Poderes Tradicionais”, Soronda (INEP, Bissau), No. 16, 1993, pp. 39–50. Forrest, Joshua B., “Guinea Bissau since Independence: a decade of domestic power struggles”, Journal of Modern African Studies, Vol. 25, No. 1, 1987, pp. 95–116. ——. Guinea-Bissau: Power, Conflict, and Renewal in a West African Nation, Boulder: Westview Press, 1992. ——. Lineages of State Fragility: rural civil society in Guinea Bissau, Athens & Oxford: Ohio University Press & James Currey, 2003. Gable, Eric, “The Decolonization of Consciousness: local skeptics and the will to be modern in a West African village”, American Ethnologist, Vol. 22, No. 2, 1995, pp. 242–57. ——. “Manjaco Rulers after a Revolution”, Africa, Vol. 73, No. 1, 2003, pp. 88–112. ——. “Conclusion: Guinea Bissau yesterday … and tomorrow”, African Studies Review, Vol. 52, No. 2, 2009, pp. 165–79. Gacitua-Mario, Estanislao, Aasland, Sigrun, Nordang, Hakon & Wodon, Quentin, “Institutions, Social Networks and Conflicts in Guinea Bissau. Results from a 2005 survey” in Boubacar-Sid Barry, Edward G. E. Creppy, Estanislao Gacitua-Mario, and Quentin Wodon (eds), Conflicts, Livelihoods and Poverty in Guinea Bissau, Washington D.C.: World Bank, 2007, pp. 23–42. Galli, Rosemary, Guinea Bissau: economy, politics and society, Boulder: Westview, 1987. ——. “On Peasant Productivity: the case of Guinea Bissau”, Development and Change, Vol. 18, No. 1, 1987, pp. 69–98. Galli, Rosemary and Jones, Jocelyn, Guinea-Bissau: Politics, Economy, Society, Boulder CO: Westview Press, 1987. Gomes, João Carlos, Polón di Bra, Bissau: INACEP, 1999. Havik, Philip J., “As Sociedades Agrárias e a Intervenção Rural na Guiné Bissau: uma revista da literatura pós independência”, Revista Internacional de Estudos Africanos, No. 14–15, 1991, pp. 279–310. ——. “Mundasson i Kambansa: espaço social e movimentos politicos na Guiné Bissau (1910–1994)”, Revista Internacional de Estudos Africanos, No. 18–22, 1999, pp. 115–67. ——. “Estradas Sem Fim: o trabalho forçado a a política indígena na Guiné (1915–1945)” in CEAUP (eds.), Trabalho Forçado Africano: experiências coloniais comparadas, Oporto: Campo das Letras, 2006, pp. 229–47. ——. “Ilhas Desertas: impostos, comércio, trabalho forçado e o êxodo das Ilhas Bijagós (1915–1935)” in CEAUP (eds.), Trabalho Forçado Africano: articulações com o poder político, Oporto: Campo das Letras, 2007, pp. 171–89. ——. “Esta ‘Libéria Portuguesa’: mudanças políticas e conflitos sociais na Guiné (1910– 1920)” in Philip J. Havik, Clara Saraiva & José Alberto Tavim (eds), Caminhos Cruzados em História e Antropologia: ensaios de homenagem à Jill Dias, Lisbon: Instituto de Ciências Sociais (ICS), 2010, pp. 123–41. Induta, Zamora, Guiné: 24 anos de indepência, 1974–1998, Lisbon: Hugin, 2001. Jao, Mamadú, “Os Poderes Tradicionais no Período de Transição” in Fafali Koudawo and Peter K. Mendy (eds), Pluralismo Político na Guiné Bissau: uma transição em curso, Bissau: INEP, 1996, pp. 123–33. ——. “Uma Leitura do Conflito Guineense”, Soronda, Número Especial 7 de Junho, 2000, pp. 105–23. Jong, Joop T.V.M de, A Descent into African Psychiatry, Amsterdam: Koninklijk Instituut voor de Tropen (KIT), 1987. thematic bibliography 255

Klute, Georg, Embaló, Birgit, Borszik, Anne-Kristin and Embaló, A. Idrissa, “Local Expe riences of Conflict Management”, Soronda (INEP, Bissau), special issue, 2008, pp. 15–33. Kohl, Christoph, “National Integration in Guinea-Bissau since Independence”, Cadernos de Estudos Africanos, No. 20, 2011, pp. 86–109. Koudawo, Fafali, “Sociedade Civil e transição pluralista na Guiné Bissau” in Fafali Koudawo and Peter Karibe Mendy (eds), Pluralismo Politico na Guiné Bissau: uma transição em curso, Bissau: INEP, 1996, pp. 67–120. ——. “La Guerre des Mandjuas: crise de gouvernance et implosion d’un modèle de résorp- tion de crises”, Soronda, Número Especial 7 de Junho, 2000, pp. 153–74. Lobban, Richard and Forrest, Joshua, Historical Dictionary of the Republic of Guinea-Bissau, Metchuen: Scarecrow Press, 1988 (2nd ed). Lopes, Carlos, Etnia, Estado e Relações de Poder na Guiné Bissau, Lisbon: Edições 70, 1982. ——. Guinea Bissau: from liberation struggle to independent statehood, London: Zed Books, 1987. ——. Para uma leitura sociológica da Guiné-Bissau, Lisbon: Edições ES, 1988 Mendy, Peter Karibe, Colonialismo Português em África: a tradição de resistência na Guiné Bissau (1879–1959), Bissau: INEP, 1994. Monteiro, Isaac (ed.), O Programa de Ajustamento Estrutural na Guiné Bissau: analise dos efeitos socio-económicos, Bissau: INEP, 1996. Nóbrega, Álvaro, A Luta pelo Poder na Guiné Bissau, Lisbon: Instituto Superior de Ciências Sociais e Políticas/Universidade Técnica de Lisboa, 2003. Pereira, Aristides, Uma Luta, Um Partido, Dois Paises, Lisbon: Notícias, 2002. Pierson-Mathy, Paulette, La Naissance de l’État par la guerre de libération nationale: le cas de la Guinée-Bissau, Paris: UNESCO, 1980. Rudebeck, Lars, Guinea Bissau. A Study in Political Mobilization, Uppsala: Scandinavian Institute of African Studies, 1972. ——. “Political Mobilisation for Development in Guinea Bissau”, Journal of Modern African Studies, Vol. 10, No. 1, 1972, pp. 1–18. ——. Problèmes de Pouvoir Populaire et de Développement: transition difficile en Guinée Bissau, Uppsala: Scandinavian Institute of African Studies, 1982. ——. “Kandjadja, Guinea Bissau, 1976–1986. Observations on the political economy of an African village”, Review of African Political Economy, No. 41, 1988, pp. 17–29. ——. On Democracy and Sustainability: transition in Guinea Bissau, Stockholm: SIDA Studies, 2001. Sangreman, Carlos, Sousa Jr., Fernando, Severino, Guilherme and Barros, Miguel, A Evolução Política Recente na Guiné Bissau: as eleições presidenciais de 2005, os conflitos, o desenvolvimento, a sociedade civil, Lisbon: Centro de Estudos sobre África (CEsA), Documento de trabalho No. 70, 2006. Silva, António E. Duarte, A Independência da Guiné Bissau e a Descolonização Portuguesa, Oporto: Afrontamento, 1997. Sousa, Julião Soares, “O fenómeno tribal, o tribalismo e a construção da identidade nacional no discurso de Amílcar Cabral” in Luís Reis Torgal, Fernando Pimenta and Julião Soares Sousa (eds), Comunidades Imaginadas. Nação e nacionalismo em África, Coimbra: Imprensa da Universidade de Coimbra, 2008, pp. 161–72. Teixeira, Ricardo Jacinto Dumas, “Tiro na Democracia: uma analise sobre o processo de transição democrática na Guiné Bissau, 1994–2007”, 2007, published on-line at: http:// www.didinho.org/TIRONADEMOCRACIA.htm ——. “Consciência nacional, democratização e conflito político: semelhanças e diferenças entre Guiné-Bissau e Moçambique”, conference paper dated 2010, published on-line at: http://www.didinho.org/ArtigoCEIA7[1].pdf Themudo, Marina P., “Cultura, Agri-Cultura e Cultura Política no Sul da Guiné Bissau”, Lusotopie, Vol. 8, No. 2, 2006, pp. 127–54. ——. “From the Margins of the State to the Presidential Palace: the Balanta case in Guinea Bissau”, African Studies Review, Vol. 52, No. 2, 2009, pp. 47–67. 256 thematic bibliography

Trajano Filho, Wilson, “Invisiveis e Liminares: a sociedade crioula e seus heróis”, Soronda, No. 18, 1994, pp. 77–113. ——. Polymorphic Creoledom: the ‘creole’ society of Guinea Bissau, PhD thesis, University of Pennsylvania, 1998 ——. “Outros Rumores de Identidade na Guiné Bissau”, Working Paper, Série Antropologia, Brasília: Universidade de Brasília, 2000. ——. “O precário equilíbrio entre improvisação e regras: reflexões sobre a cultura política da Guiné-Bissau”, Revista de Antropologia, Vol. 51, No. 1, 2008, pp. 233–266. Urdang, Stephanie, Fighting Two Colonialisms: women in Guinea Bissau, New York: Monthly Review Press, 1979. Vigh, Henrik, “Conflictual Motion and Political Inertia: on rebellions and revolutions in Bissau and beyond”, African Studies Review, Vol. 52, No. 2, 2009, pp. 143–64. ——. Navigating the Terrains of War: youth and soldiering in Guinea Bissau, Oxford: Berghahn Books, 2007.

Mozambique Adam, Yussuf, “Mueda, 1917–1990: resistênca, colonialismo, libertação e desenvolvimento”, Arquivo: Boletim do Arquivo Histórico de Moçambique, N° 14, 1993, pp. 9–101. ——. Escapar aos Dentes do Crocodilo e Cair na Boca do Leopardo. Trajéctoria de Moçambique Pós-Colonial (1975–1990), Maputo: Promédia, 2005. Adam, Yussuf and Dyuti, Hiláro Alumasse “O massacre de Mueda: falam testemunhas”, Arquivo: Boletim do Arquivo Histórico de Moçambique, No. 14, 1993, p. 117–128. Adam, Yussuf and Gentili, Ana Maria., “O movimento dos Liguilanilu no planalto de Mueda, 1957–1962”, Estudos Moçambicanos, No. 4, 1983, p. 41–75. Alden, Chris, Mozambique and the Construction of the New African State, Basingstoke: Palgrave, 2001. Alpers, Edward A., ‘“To seek a better life’: The implications of migration from Mozambique to Tanganyika for class formation and political behavior”, Canadian Journal of African Studies, Vol. 18, No. 2, 1984, pp. 367–388. Arthur, Maria José, “Mozambique: Women in the Armed Struggle” in Patricia McFadden (ed.), Gender in Southern Africa: A Gendered Perspective, Harare: SAPES Books, 1998, pp. 67–82. Basto, Maria-Benedita, “A formação das elites da FRELIMO e os lugares de produção da figura do escritor e do papel da literatura na luta internacionalista anti-colonial” in Maria-Benedita Basto, A Guerra das escritas. Literatura e nação em Moçambique, PhD thesis, Lisbon: Universidade Nova, 2004, pp. 122–226. ——. A Guerra das escritas. Literatura, nação e teoria pós-colonial em Moçambique, Lisbon: Edições Vendaval, 2006. ——. “Enjeux, double je(ux): hétéronymie, genre et nation dans Eu, o Povo de António Quadros/Mutimati Barnabé João” in Maria-Benedita Basto (ed.), Enjeux Littéraires et construction d’espaces démocratiques en Afrique subsaharienne. Littérature, politique, identités, Paris: EHESS, 2007, pp. 183–216. Borges Coelho, João Paulo, “As Fontes Coloniais Escritas no Estudo da Luta Armada de Libertação Nacional: Sobre uma Experiência de Investigação no Arquivo Histórico”, Arquivo: Boletim do Arquivo Histórico de Moçambique, No. 1, 1987, pp. 35–39. ——. O Início da Luta Armada em Tete, 1968–1969. A primeira fase da guerra e a reacção colonial, Maputo: Arquivo Historico de Mocambique, Estudo 7, 1989. ——. “Entrevista com a Associação dos Antigos Combatentes de Tete: Uma Conversa Sobre a Luta de Libertação Nacional na Frente de Tete”, Arquivo: Boletim do Arquivo Histórico de Moçambique, No. 13, 1993, pp. 99–137. ——. “A investigação recente da Luta de Libertação Nacional: Contexto, práticas e per- spectivas”, Arquivo: Boletim do Arquivo Histórico de Moçambique, No. 17, 1995, pp. 159–179. thematic bibliography 257

Bowen, Merle L., The State Against the Peasantry: Rural Struggles in Colonial and Postcolonial Mozambique, Charlottesville: University Press of Virginia, 2000. Brito, Luís de, “Mozambique: guerre et nationalisme”, Politique Africaine, No. 29, 1988, pp. 15–28. ——. Le FRELIMO et la construction de l’État national au Mozambique. Le sens de la référence au marxisme (1962–1983), PhD thesis, Paris : Université de Paris 8, 1991. Brito, Luís de, Castel-Branco, Carlos, Francisco, António and Chichava, Sergio (eds), Southern Africa and Challenges for Mozambique, Maputo: IESE, 2009. ——. (eds), Cidadania e Governação em Moçambique, Maputo: IESE, 2009. Cabrita, João, Mozambique: The Tortuous Road to Democracy, New York: Palgrave, 2000. Cahen, Michel, Mozambique, la révolution implosée. Études sur douze années d’indépendance (1975–1987), Paris: L’Harmattan, 1987 ——. “Check on socialism in Mozambique: what check? What socialism?” Review of African Political Economy, No. 57, 1993, pp.46–59. ——. “Mozambique : histoire géopolitique d’un pays sans nation”, Lusotopie 1994, Paris: L’Harmattan, 1994, pp. 213–266. ——. “Entrons dans la nation. Notes pour une étude du discours politique de la marginalité. Le cas de la Renamo du Mozambique”, Politique africaine, No. 67, 1997, pp. 70–88. ——. “The Mueda Case and Maconde Political Ethnicity. Some notes on a work in progress”, Africana Studia (Oporto), No. 2, 1999, pp. 29–46. ——. “De la construction de l’État sans nation à la nationalisation du monde”, pp. 19–44, in M. Cahen, La nationalisation du monde. Europe, Afrique, l’identité dans la démocratie, Paris: L’Harmattan, 1999. ——. “L’anticolonialisme identitaire : conscience ethnique et mobilisation anti-portugaise au Mozambique (1930–1965)” in Colette Dubois, Marc Michel & Pierre Soumille (eds), Frontières plurielles, Frontières conflictuelles en Afrique subsaharienne, Paris: L’Harmattan, 2000, p. 319–333. ——. “Nationalisms and Ethnicities. Lessons from Mozambique” in Einar Braathen, Morten Bøås and Gjermund Saether, Ethnicity Kills? The Politics of War, Peace and Ethnicity in Subsaharian Africa, London: Macmillan, 2000, pp. 163–187. ——. Les Bandits. Un historien au Mozambique, Paris: Calouste Gulbenkian, 2002. ——. “Lutte armée d’émancipation anticoloniale ou mouvement de libération nationale ? Processus historique et discours idéologique. Le cas des colonies portugaises, et du Mozambique en particulier”, Revue Historique, Vol. 315, No.1, 2006, pp. 113–138. ——. “Moçambique: o ‘fim da história’… única. Trajectórias dos anticolonialismos em Moçambique”, Africana Studia, No. 15, 2010, pp. 195–240. Cahen, Michel, Waniez, Philippe and Brustlein, Violette, “Pour un atlas social et culturel du Mozambique”, Lusotopie, Vol. 9, No. 1, 2002, pp. 305–362. Chichava, Sérgio, Le ‘Vieux Mozambique’: l’identité politique de la Zambézie, PhD thesis, Bordeaux: Institut d’études politiques, 2007. Chingono, Mark F., The State, Violence and Development: The Political Economy of War in Mozambique, 1975–1992, Aldershot: Avebury, 1996. Christie, Iain, Machel of Mozambique, a Biography, London: Panaf, 1989. Cruz e Silva, Teresa, “Igrejas protestantes no Sul de Moçambique e nacionalismo: o caso da ‘Missão Suíça’ (1940–1974)”, Estudos Moçambicanos, No. 10, 1992, pp. 19–39. ——. “Identity and Political Consciousness in Southern Mozambique, 1930–1974. Two Presbyterian Biographies Contextualised”, Journal of Southern African Studies, Vol. 24, No. 1, 1998, pp. 223–236. ——. Protestant Churches and the Formation of Political Consciousness in Southern Mozambique (1930–1974), Basel: P. Schlettwein Publishing, 2001. Darch, Colin, “As publicações da FRELIMO: um estudo preliminar”, Estudos Moçambicanos, No. 2, 1981, pp. 105–120. Darch, Colin (with Calisto Pecheleke), Mozambique, Oxford: Clio Press (World Biblio graphical Series No. 78.), 1987. 258 thematic bibliography

Darch, Colin and Hedges, David, “‘Não temos a possibilidade de herdar nada de Portugal’. As raízes do exclusivismo e vanguardismo político em Moçambique, 1969–1977” in Glaucia Villas Bôas (ed.), Territórios da língua portuguesa – culturas, sociedades, políti- cas: anais do IV Congresso Luso-Africano-Brasileiro, 1 à 5 de setembro de 1996, Rio de Janeiro: IFCS, 1999, p. 135–149. Derluguian, Georgi, “Les têtes du monstre: du climat social de la violence armée au Mozambique”, L’Année Africaine 1989, Bordeaux: Coédition CEAN/A.Pedone, 1990, pp. 89–128. ——. “Social Decomposition and Armed Violence in Post-Colonial Mozambique”, Review (F. Braudel Center), Vol. 13, No. 4, 1990, pp. 439–464. Dinerman, Alice, Counter-Revolution and Revisionism: the case of Mozambique, 1975–1994, London: Routledge, 2006. ——. “Independence Redux in Postsocialist Mozambique”, Relações Internacionais, No. 15, 2007 (online at http://www.ipri.pt). Englund, Harry, From War to Peace on the Mozambique-Malawi Borderland, Edinburgh: Edinburg University Press, 2002. Ergero, Bertil, A Dream Undone: the political economy of democracy 1975–1984, Upsala: Scandinavian Institute of African Studies, 1987. First, Ruth, Black Gold: The Mozambican Miner, Proletarian and Peasant, Basingstoke: Palgrave Macmillan, 1983. Fry, Peter (ed.), Moçambique: ensaios, Rio de Janeiro: Editora UFRJ, 2001. Geffray, Christian, “Fragments d’un discours du pouvoir (1975–1985): du bon usage d’une méconnaissance scientifique”, Politique Africaine, No. 29, 1988, pp. 71–85. ——. A Causa das Armas. Oporto: Edições Afrontamento, 1991. Gomes, Carlos de Matos, Moçambique 1970. Operação Nó Górdio, Lisbon: Prefácio, 2002. Gomes Cravinho, João, Modernizing Mozambique: Frelimo ideology and the Frelimo state, D Phil. Thesis, Oxford: University of Oxford, 1995. Gómez, Miguel Buendía, Educação Moçambicana: História de um Processo: 1962–1984, Maputo: Livraria Universitária, 1999. Hall, Margaret and Young, Tom, Confronting Leviathan: Mozambique since Independence, London: Hurst & Co., 1997. Hanlon, Joseph, Mozambique: the revolution under fire, London: Zed Books, 1984. ——. Mozambique: who calls the shots?, London: James Currey, 1991. Harrison, Graham, The Politics of Democratization in Rural Mozambique, Lampeter: Edwin Mellen Press, 2000. Henriksen, Thomas H., Mozambique: a history, London: Rex Collings, 1978. ——. Revolution and Counterrevolution. Mozambique’s War of Independence, 1964–1974, Westport: Greenwood Press, 1983. ——. “Marxism and Mozambique”, African Affairs, Vol. 77, No. 309, 1978, pp. 441–62. Honwana, Raul, The Life History of Raul Honwana: An Inside View of Mozambique from Colonialism to Independence (edited by Allen Isaacman), Boulder: Lynne Reinner, 1988. Isaacman, Allen, Mozambique. The Africanization of a European institution: the Zambesi prazos, 1750-1902, Madison: University of Wisconsin Press, 1972. ——. From Colonialism to Revolution, 1900–1982, Boulder: Westview Press, 1983. ——. “Colonial Mozambique, an Inside View: The Life History of Raul Honwana”, Cahiers d’Etudes Africaines, No. 109, 1988, pp. 59–89. Israel, Paolo, “Utopia Live: singing the Mozambican struggle for national liberation”, Kronos (Cape Town), Vol. 35, No. 1, 2009, pp. 98–141. ——. “The Formulaic Revolution: Song and the ‘popular memory’ of the Mozambican Liberation Struggle”, Cahiers d’Etudes Africaines, No. 197, 2010, pp. 181–216. José, Alexandrino and Meneses, Paula Maria G. (eds), Moçambique – 16 anos de historio- grafia: focos, problemas, metodolgias, desafios para a década de 90, Maputo: Notícias, 1991. thematic bibliography 259

Kruks, Sónia, “From nationalism to Marxism: the ideological history of Frelimo, 1962–1977” in Irving L. Markowitz (ed.), Studies in power and class in Africa, New York: Oxford University Press, 1987, pp. 241–242. Macagno, Lorenzo, “Fragmentos de uma imaginação nacional”, Revista Brasileira de Ciências Sociais (São Paulo), Vol. 24, No. 70, 2009, pp. 17–35. ——. “Multiculturalism in Mozambique? Reflections from the field”, Vibrant, Vol. 5, No. 2, pp. 223–246. Manghezi, Nadja, O meu coração está nas mãos de um negro: uma história da vida de Janet Mondlane, Maputo: Centro de Estudos Africanos/Livraria Universitária, 1999. Martins, Helder, Porquê Sakrani? Memórias dum Médico duma Guerrilha Esquecida, Maputo: Editorial Terceiro Milénio, 2001. Mazula, Brazão, Educação, Cultura e Ideologia em Moçambique: 1975–1985, Oporto: Afrontamento, 1995. ——. (ed.), Mozambique: Elections, Democracy and Development, Maputo: Inter-Africa Group, 1996. Meyns, Peter, Befreiung und nationaler Wiederaufbau von Mozambique, Hamburg: Institut für Afrika-Kunde, 1979. Morier-Genoud, Eric, “Mozambique since 1989: Shaping democracy after Socialism” in Abdul Raufu Mustapha & Lindsay Whitfield (eds), Turning Points in African Democracy, Oxford: James Currey, 2009, pp. 153–166. ——. “Y a-t-il une spécificité protestante au Mozambique? Discours du pouvoir post-colonial et histoire des églises chrétiennes”, Lusotopie 1998, Paris: Karthala, 1998, pp. 407–420. ——. Religious Orders and the Making of Catholic Politics. The case of the diocese of Beira, 1940–1974, PhD Thesis, Binghamton: State University of New York, 2005. Minter, William, Apartheid’s Contras: An Inquiry into the Roots of War in Angola and Mozambique, London: Zed Books, 1996. Mondlane, Eduardo, The Struggle for Mozambique, London: Penguin Books, 1969. Munslow, Barry, Mozambique: The Revolution and its Origins, London: Longman, 1983. ——. (ed.), Samora Machel: An African Revolutionary. Selected speeches and writings, London: Zed Books, 1985. Newitt, Malyn, A History of Mozambique, London: Hurst & Co., 1995. Ncomo, Barnabé Lucas, Uria Simango: um homem, uma causa, Maputo: Edições Novafrica, 2003. O’Laughlin, Bridget, “Class and the Customary: The Ambiguous Legacy of Indigenato in Mozambique”, African Affairs, No. 99, 2000, pp. 5–42. Opello Jr., Walter C., Internal war in Mozambique. A social-psychological analysis of a nationalist revolution, PhD thesis, Boulder: University of Colorado, 1973. ——. “Pluralism and elite conflict in an independence movement: FRELIMO in the 1960s”, Journal of Southern African Studies, Vol. 2, No. 1, 1975, pp. 66–82. Panzer, Michael G., “The Pedagogy of Revolution: Youth, Generational Conflict, and Education in the Development of Mozambican Nationalism and the State, 1962–1970”, Journal of Southern African Studies, Vol. 35, No. 4, 2009, pp. 803–820. Pélissier, René, Naissance du Mozambique.Résistances et révoltes anticoloniales (1854–1918), Orgeval: éditions Pélissier, 1984. Penvenne, Jeanne, “‘We are all Portuguese!’ Challenging the Political Economy of Assimilation: Lourenço Marques, 1870–1933” in Leroy Vail (ed.), The Creation of Tribalism in Southern Africa, Berkeley and Los Angeles: University of California Press, 1989, pp. 255–288. ——. African Workers and Colonial Racism. Mozambican Strategies and Struggles in Lourenço Marques, 1877–1962, Portsmouth: Heinemann, 1995. ——. “A Tapestry of Conflict: Mozambique 1960–1995” in David Birmingham and Phyllis Martin (eds), History of Central Africa. The Contemporary Years, London: Longman, 1998, pp. 230–266. 260 thematic bibliography

Pereira, Zélia, “Les Jésuites et la formation d’élites au Mozambique, 1961–1974”, LFM. Missions et sciences sociales, No. 14, 2004, pp. 75–116. Pitcher, M. Anne, Transforming Mozambique: the politics of privatization, Cambridge: Cambridge University Press, 2002. ——. “Conditions, commitments and the politics of restructuring in Africa”, Comparative Politics, No. 36, No. 4, 2004, pp. 375–98. ——. “Forgetting from above and memory from below: strategies of legitimation and struggle in postsocialist Mozambique”, Africa, Vol. 76, No. 1, 2006, pp. 88–112. Power, Marcus, “Post-Colonial Cinema and the Reconfiguration of Mocambicanidade”, Lusotopie 2004, Paris: Karthala, 2004, pp. 261–278. Reis, João and Muivane, Armando Pedro (eds), Datas e documentos da história da Frelimo, Maputo: Imprensa Nacional, 1975. Rennie, Keith John, Christianity, Colonialism and the Origins of Nationalism among the Ndau of Southern Rhodesia, 1890–1935, PhD thesis, Evanston: Northwestern University, 1973. Rocha, Ilídio, A Imprensa de Moçambique. História e Catálogo (1854–1975), Lisbon: Edições Livros do Brasil, 2000. Rosário, Lourenço do (ed.), II Congresso sobre a Luta de Libertação Nacional-Guerra colonial. 27 anos depois, a reflexão possivel, Maputo: Edições ISPU, 2004. Saul, John S. (ed.), A Difficult Road: the Transition to Socialism in Mozambique, New York: Monthly Review Press, 1985. Sheldon, Kathleen, Pounders of Grain: A History of Women, Work and Politics in Mozambique, Portsmouth: Heinemann, 2002. Sopa, António (ed.), Samora, homem do povo, Maputo: Maguezo, 2001. Sousa Santos, Boaventura de and Trindade, João Carlos (eds), Conflito e Transformação Social: Uma Paisagem das Justiças em Moçambique, Oporto: Afrontamento, 2003, 2 vols. Souto, Amélia de, Caetano e o ocaso do ‘Império’: administração e guerra colonial em Moçambique durante o Marcelismo (1968–1974), Oporto: Afrontamento, 2007. Sumich, Jason, “Contruir uma Nação: Ideologias de Modernidade da Elite Moçambicana”, Análise Social, Vol. 43, No. 2, pp. 319–345. ——. “Nationalism, Urban Poverty and Identity in Maputo, Mozambique”, Crisis States Research Centre Working Papers, No. 2, 2010. Urdang, Stephanie, And still they dance: women, war and the struggle for change in Mozambique, New York: Monthly Review Press, 1989. West, Harry, “Creative Destruction and Sorcery of Construction: Power, Hope and Suspicion in Post-War Mozambique”, Cahiers d’Etudes Africaines, No. 147, 1997, pp. 675–698. ——. “Girls with Guns: Narrating the Experience of War of FRELIMO’s Female Detachment”, Anthropological Quarterly, Vol. 73, No. 4, 2000, pp. 180–194. ——. “Sorcery of Construction and Socialist Modernization: Ways of Understanding Power in Postcolonial Mozambique”, American Ethnologist, No. 28, 2001, pp. 119–150. ——. Kupilikula: Governance and the Invisible Realm in Mozambique, Chicago: University of Chicago Press, 2005. História de Moçambique, Maputo: Universidade Eduardo Mondlane, 1983–1993, 3 vols.

Angola Andrade, Mario de, and Mark Ollivier, The war in Angola: a socio-economic study (translated by Marga Holness), Dar-es-Salaam: Tanzania Publishing House, 1975. Barradas, Acácio, Agostinho Neto: uma vida sem tréguas, 1922–1979, Lisbon/Luanda: [s.n.], 2005. Bhagavan, M.R., Angola’s Political Economy, 1975–1985,Uppsala: Scandinavian Institute of African Studies, Research Report No. 75, 1986. Bender, Gerald J., Angola under the Portuguese: The Myth and the Reality, Berkeley: University of California Press, 1978. thematic bibliography 261

Besson, Emmanuelle, “Autour du procès de Joaquim Pinto de Andrade. L’Eglise catholique et l’Angola colonial, 1960–1975”, Le Fait Missionnaire. Histoire et Héritages, No. 12, 2002. Birmingham, David, “The Twenty-seventh of May: An Historical Note on the Abortive 1977 Coup in Angola”, African Affairs, Vol. 77, No. 309, 1978, pp. 554–564. ——. “Angola revisited”, Journal of Southern African Studies, Vol. 15, No. 1, 1988, pp. 1–15. ——. “Carnival at Luanda”, Journal of African History, No. 29, 1988, pp. 93–103. ——. Frontline Nationalism in Angola and Mozambique, London: James Currey, 1992. ——. Empire in Africa: Angola and its neighbours, Athens: Ohio University Press, 2006. Bittencourt, Marcelo, “A criação do MPLA”, Estudos Afro-Asiáticos, No. 32, 1997, pp. 185–208. ——. Dos jornais às armas: trajectórias da contestação angolana, Lisbon: Vega Editora, 1999. ——. ‘Estamos juntos’: O MPLA e a Luta Anticolonial (1961–1974), PhD Thesis, Rio de Janeiro: Universidade Federal Fluminense, 2002. ——. “A historia contemporanea de Angola: seus achados e suas armadilhas” in Construindo o passado angolano: as fontes e a sua interpretacao - actas do II seminario internacional sobre a historia de Angola, Lisbon: Comissão Nacional para as Comemorações dos Descobrimentos Portugueses, 2000, pp. 161–185. Bridgland, Fred, Jonas Savimbi: A Key to Africa, London: Hodder & Stoughton, 1988. ——. “Savimbi et l’exercice du pouvoir: un témoignage”, Politique Africaine, No. 57, 1995, pp. 94–102. Brinkman, Inge (ed.), Singing in the bush: MPLA songs during the war for independence in south-east Angola (1966–1975), Cologne: Rüdiger Koeppe, 2001. ——. “War, Witches and Traitors: Cases from the MPLA’s Eastern Front in Angola (1966– 1975)”, Journal of African History, Vol. 44, No. 2, 2003, pp. 303–325. ——. “War and Identity in Angola: two case studies”, Lusotopie 2003, Paris: Karthala, 1993, pp. 195–221. ——. A war for people: civilians, mobility, and legitimacy in South-East Angola during MPLA’s war for independence, Cologne: Rüdiger Koeppe, 2005. Chabal, Patrick and Vidal, Nuno (eds), Angola: The Weight of History, London: Hurst, 2008. Chiwale, Samuel, Cruzei-me com a história, Lisbon: Sextante, 2008. Ennes Ferreira, Manuel, A indústria em tempo de Guerra (Angola, 1975–91), Lisbon: Edições Cosmos, 1999. Corrado, Jacopo, The Creole Elite and the Rise of Angolan Proto-nationalism 1870–1920, Amherst: Cambria Press, 2008. Gleijeses, Piero, Conflicting Missions. Havana, Washington, and Africa, 1959–1976, Chapel Hill: University of North Carolina Press, 2002. Guimarães, Fernando Andresen, The Origins of the Angolan Civil War: Foreign Intervention and Domestic Political Conflict, Basingstoke: Macmillan, 2001. Heimer Franz-Wilhelm, The Decolonization Conflict in Angola 1974–1976: An Essay in Political Sociology, Geneva: Institut Universitaire de Hautes Etudes Internationales, 1979. ——. (ed.), Social Change in Angola, Munich: Weltforum, 1973. Henderson, Lawrence W., Angola: Five Centuries of Conflict, Ithaca: Cornell University Press, 1979. ——. The Church in Angola. A River of Many Currents, Cleveland: The Pilgrim Press, 1992. ——. Development and the Church in Angola. Jesse Chipenda the Trailblazer, Nairobi: Acton Publishers, 2000. Heywood, Linda, Contested power in Angola: 1840s to the Present, Rochester: University of Rochester Press, 2002. ——. “Unita and Ethnic Nationalism in Angola”, Journal of Modern African Studies, Vol. 27, No. 1, 1989, pp. 47–66. ——. “Towards an understanding of modern political ideology in Africa: the case of the Ovimbundu of Angola”, Journal of Modern African Studies, Vol. 36, No. 1, 1998, pp. 139–167. 262 thematic bibliography

Hodges, Tony, Angola. From Afro-Stalinism to Petro-Diamond Capitalism, Oxford: James Currey/Bloomington: Indiana University Press, 2001. ——. Angola: Anatomy of an Oil State, Oxford: James Currey/Bloomington: Indiana University Press, 2004. James, W. Martin, Historical Dictionary of Angola, Lanham: Scarecrow Press, 2004. ——. A political history of the Civil War in Angola, 1974–1990, Brunswick: Transaction Publishers, 1992. Mabeko-Tali, Jean-Michel, “Tribalisme, ‘régionalisme’ et lutte de libération nationale: la question ‘tribale’ et ‘ethnico-régionale’ dans la dissidence au sein du MPLA dans l’est angolais en 1969–1974”, L’Année Africaine, Paris: Karthala, 1993, pp. 463–486. ——. Dissidências e poder de Estado. O MPLA perante si próprio (1962–1977). Vol. 1. Ensaio de história política. 1962–1974, Luanda: Editorial Nzila (Coleção Ensaio 3), 2001. ——. Dissidências e poder de Estado. O MPLA perante si próprio (1962–1977). Vol. 2. Ensaio de história política. 1974–1977, Luanda: Editorial Nzila (Coleção Ensaio 3), 2001. ——. “La question de Cabinda: séparatismes éclatés, habiletés luandaises et conflits en Afrique centrale”, Lusotopie 2001, Paris: Karthala, 2001, pp. 49–62. ——. “Entre économie rentière et violence politico-militaire : la question cabindaise et le processus de paix angolais”, Politique Africaine, No. 110, 2008, pp. 65–83. Malaquias, Assis, Rebels and Robbers: Violence in Post-Colonial Angola, Uppsala: Nordiska Afrikainstitutet, 2007. Marcum, John A., The Angolan Revolution, 1: The Anatomy of an Explosion (1950–1962), Cambridge: MIT Press, 1969. ——. The Angolan Revolution, 2: Exile Politics and Guerrilla Warfare (1962–1976), Cambridge: MIT Press, 1978. Martin, Phyllis M., Historical Dictionary of Angola, Metuchen: Scarecrow Press, 1980. Mateus, Dalila Cabrita, Purga em Angola, Lisbon: Texto Editores, 2009. Mateus, Dalila Cabrita, and Mateus, Álvaro, Angola ’61-Guerra Colonial: Causas e Consequências, Lisbon: Texto Editores, 2011. Medina, Maria do Carmo, Angola: processos políticos da luta pela independência, Coimbra: Almedina, 2005. Messiant, Christine, “Angola, les voies de l’ethnisation et de la décomposition, I. De la guerre à la paix (1975–1991). Le conflit armé, les interventions internationales et le peuple angolais”, Lusotopie 1994, Paris: L’Harmattan, No. 1–2, 1994, pp. 155–210. ——. “Angola, les voies de l’ethnisation et de la décomposition. II. Transition à la démocratie ou marche à la guerre ? L’épanouissement des deux ‘partis armés’ (mai 1991-septembre 1992)”, Lusotopie 1995, Paris: Karthala, 1995, pp. 181–212. ——. “Sur la première génération du MPLA, 1948–1960. Mário de Andrade, entretiens avec Christine Messiant (1982)”, Lusotopie 1999, Paris : Karthala, 1999, pp. 185–221. ——. “Angola: the challenge of statehood” in David Birmingham & Phyllis M. Martin (eds), History of Central Africa. The contemporary years (since 1960), London: Longman, 1998, pp. 131–165. ——. “‘Chez nous, même le passé est imprévisible’. L’expérience d’une recherche sur le nationalisme angolais, et particulièrement le MPLA : sources, critique, besoins actuels de la recherche”, Lusotopie 1998, Paris: Karthala, 1998, pp. 157–197. ——. “Why did Bicesse and Lusaka fail? A critical analysis”, Accord, No.15, 2004. ——. “Viriato da Cruz em Pequim: as provações de um revolucionário angolano. Esboço de um percurso e tentativa de interpretação” in Michel Laban (ed.), Viriato da Cruz: Cartas de Pequim, Luanda: Chá de Caxinde, 2004, pp. 215–360. ——. 1961: L’Angola colonial, histoire et société. Les prémisses du mouvement nationaliste, Basel: PSP Verlag, 2006. ——. “The Mutation of Hegemonic Domination” in Patrick Chabal and Nuno Vidal (eds), Angola: The Weight of History, London: Hurst, 2007, pp. 93–123. ——. L’Angola postcolonial, Paris: Karthala, 2008, 2 vols. Minter, William, Operation Timber: Pages from the Savimbi Dossier, Trenton: Africa World Press, 1988. thematic bibliography 263

Mohanty, Susanna, Political Development and Ethnic Identity in Africa: a study of Angola since 1960, London: Sangam Books, 1992. Moorman, Marissa J., “Of Westerns, women, and war: re-Situating Angolan cinema and the nation”, Research in African Literatures, Vol. 32, No. 3, 2001, pp. 103–122. ——. “Putting on a pano and dancing like our grandparents: Nation and dress in late colonial Luanda” in Jean M. Allman (ed.), Fashioning Africa: power and the politics of dress, Bloomington: Indiana University Press, 2004, pp. 84–103. ——. “Duelling bands and good girls: gender, music, and nation in Luanda’s musseques, 1961–1974”, International Journal of African Historical Studies, Vol. 37, No. 2, 2004, pp. 255–288. ——. Intonations. A Social History of Music and Nation in Luanda, Angola, from 1945 to Recent Times, Athens: Ohio University Press, 2008. Muekalia, Jardo, Angola - A Segunda Revolução. Memórias da luta pela Democracia, Lisbon: Sextante, 2011. Neto, Agostinho, Sagrada Esperança, São Paulo: Editora Ática, 1985. Neto, Maria da Conceição, “Ideologias, contradições e mistificações da colonização de Angola no século XX”, Lusotopie 1997, Paris: Karthala, 1997, pp. 327–359. ——. “Angola no século XX (até 1974)” in Valentim Alexandre (ed.), O Império Africano (séculos XIX e XX), Lisbon: Colibri, 2000, pp. 175–195. Nunes, António Lopes Pires, Angola 1961. Da Baixa do Cassange a Nambuangongo, Lisbon: Prefacio, 2005. Okuma, Thomas, Angola in Ferment: the Background and Prospects of Angolan Nationalism, Boston: Beacon Press, 1962. Pacheco, Carlos, MPLA. Um nascimento polémico (as falsificações da história), Lisbon: Vega, 1997. Paredes, Margarida, “Deolinda Rodrigues, da Família Metodista à Família MPLA, o Papel da Cultura na Política”, Cadernos de Estudos Africanos, No. 20, 2011, pp. 11–26. Pawson, Lara, “The 27 May in Angola: a view from below”, Relações Internacionais, No. 14, 2007, pp. 1–18. Pearce, Justin, An Outbreak of Peace: Angola’s Situation of ‘Confusion’, Cape Town: David Philip, 1995. ——. “L’Unita à la recherche de ‘son peuple’. Carnets d’une ‘non-campagne’ sur le planalto”, Politique Africaine, No. 110, 2008, pp. 47–64. Péclard, Didier, “Religion and politics in Angola: the church, the colonial state and the emergence of Angolan nationalism, 1940–1961”, Journal of Religion in Africa, Vol. 28, No. 2, 1998, pp. 160–186. ——. Etat colonial, missions chrétiennes et nationalisme en Angola, 1920–1975. Aux racines sociales de l’UNITA, PhD Thesis, Paris: Institut d’études politiques, 2005. ——. “Missions et nationalisme: hypothèses théoriques et historiques autour du cas angolais” in Véronique Faure (ed.), Dynamiques religieuses en Afrique australe, Paris: Karthala, 2000, pp. 299–318. ——. “Les chemins de la ‘reconversion autoritaire’ en Angola”, Politique Africaine, No. 110, 2008, pp. 5–20. Pélissier, René, “Evolution des mouvements ethno-nationalistes Bakongo d’Angola avant la révolte du nord-ouest (1961)”, Revue française d’études politiques africaines, Vol. 10, No. 111, 1975, pp. 81–103. ——. La colonie du Minotaure. Nationalismes et révoltes en Angola (1926–1961), Orgeval: Pélissier, 1978 Pereira, Anthony W., “The Neglected Tragedy: the Return to War in Angola 1992–3”, Journal of Modern African Studies, Vol. 32, No. 1, 1994, pp. 1–28. Pestana, Nelson, L’état en Angola: Discours et practiques, PhD thesis, Paris, 2006. Pimenta, Fernando Tavares, Brancos de Angola. Autonomismo e Nacionalismo, 1900–1961, Coimbra: Minerva, 2005. ——. Angola no Percurso de um Nacionalista. Conversas com Adolfo Maria, Oporto: Afrontamento, 2006. 264 thematic bibliography

——. Os Brancos e a Independência, Oporto: Afrontamento, 2008. ——. Portugal e o Século XX. Estado-Império e Descolonização, 1890–1975, Oporto: Afrontamento, 2010. ——. “Angola’s whites: political behaviour and national identity”, Portuguese Journal of Social Science, Vol. 4, No. 3, 2005, pp. 169–193. Sakala, Alcides, Memórias de um guerrilheiro: Os últimos anos de guerra em Angola, Lisbon: Dom Quixote, 2006. Saldanha, Maria João, Jonas Savimbi. Um desafio à ditadura comunista em Angola, Lisbon: Armasilde, 1977. Savimbi, Jonas Malheiro, A resistência em busca de uma nova nação, Lisbon: Edição da Agência Portuguesa de Revistas, 1979. Schubert, Benedict, A Guerra e as Igrejas: Angola 1961–1991, Basel: PSP Verlag, 2000. Soares de Oliveira, Ricardo, Oil and Politics in the Gulf of Guinea, London: Hurst & Co., 2007. Teixeira, Rogério Cardoso, Angola (n’gola). História do Batalhão de Caçadores 109, Lisbon: Quarteto, 2008. Tvedten, Inge, Angola: Struggle for Peace and Reconstruction, Boulder: Westview Press, 1997. Vidal, Nuno and Andrade, Justino Pinto de (eds), O processo de transição para o multipar- tidarismo em Angola, Lisbon: Firmamento, 2006. ——. (eds), Civil Society and Politics in Angola, Lisbon: Firmamento, 2008. Wheeler, Douglas L., “‘Angola Is Whose House?’ Early Stirrings of Angolan Nationalism and Protest, 1822–1910”, African Historical Studies, Vol. 2, No. 1, 1969, pp. 1–22. Wheeler, Douglas and Pélissier, René, Angola, London: Pall Mall Press, pp. 159–239. Wright, George, The Destruction of a Nation: United States Policy towards Angola since 1945, London: Pluto Press, 1997. INDEX

Abrams, Philip 231 Brazil 5n11, 6, 8, 11, 12, 13, 26, 87, 144, 159, African-American/Black Americans 96, 163, 177, 179, 180, 190, 192, 193, 194, 219, 168 220, 221, 224, 230 Afro-Saxons 220 Bretton Woods 225–6 Ages of colonisation 12, 13 Britain 19, 84, 94, 181, 184, 185, 227; see also Agudas 12 United Kingdom Algeria 7, 14, 80n3, 89, 184, 192, 194n44, Brito, Luís de 24 196, 197 Alves, Nito 14, 240 Cabinda 220, 229, 248 America 6n12, 8, 23, 81, 87, 88, 89, 91, 96, Cabral, Amílcar xiii, 14n30, 28n60, 38–46 99, 155, 167, 168, 169, 179, 181, 182, 221, 224, passim, 50–53 passim, 56, 57, 74, 104, 225, 228; see also United States 106, 238, 239 Anderson, Benedict xix, 31, 72n174, 88, 103, Cabral, Luís 44, 49, 50, 246 113, 146, 200, 203n12 Cacheu 51, 68n163 Anti-colonialism xxi, 1, 3, 5, 7, 35, 40, 49, Caetano, Marcelo 236 58n118, 82, 104, 106, 125, 163, 177, 184, 189, Caluandismo 212–13 199, 200, 210, 211, 225 Cameroon 54n101 Arab 12, 18, 82 Cape Verde xiii, 13, 25–28, 41–44, 46, 47, Archives xvi, xvii, 82n6, 89, 110, 113 50, 58, 59, 70, 189, 235, 245 Armed forces 44, 50, 60, 61, 66, 67, 72n174, Casamance 66n155, 67 73, 74n181, 207, 236, 237, 246 Cassacá 48, 55 Armed struggle 14, 17, 25, 35, 41, 46, 47, Catholic 18, 26, 84, 140, 154, 158n29, 56n109, 96, 119 161–162, 167, 185, 218, 224, 225 Artists xiv, 86, 215 Changane 19 Assimilados 14, 133–34, 137, 141, 162, 163, Chatterjee, Partha xix, 103, 105, 120, 124 165, 170, 172, 182, 183, 187, 188, 189, 201, Chemane, Justino xiv 214, 235, 236 Chiefs 21, 26, 55, 60n131, 62n137, 65, 69, 81, Authoritarian xiv, xv, 1n2, 10, 15, 21, 22, 24, 84, 85, 91, 93, 95, 96, 97, 136, 245; see also 25, 27, 55, 59, 60, 131, 132, 134, 135, 184, Traditional chiefs and Régulos 221, 239, 242 China 12, 14, 82, 89, 104, 130, 150, 226, 227, 230, 248 Bailundo 159, 160 Chissano, Joaquim xiv Bakongo 17, 153, 155, 161, 163, 201 Citizen/citizenship 10, 21, 33, 71, 87, 89, Balandier, Georges 2 105, 113, 124, 134, 135, 137, 141, 142, 144, 182, Balanta 19, 48–49, 50, 51, 56, 61, 62n137, 186, 193, 195, 197, 202, 215, 217n1, 218, 219, 73n180 221, 226, 232, 234, 239, 243 Baptists 154, 155, 229 Civil war xi, xiii, xxiii, 21, 25, 34, 35, 128, 131, Bayart, Jean-François 33, 60n130 135, 138, 143, 150–54 passim, 166, 171, 173, Belgium 2n5, 4n9, 87, 181, 185, 191, 229 174, 194, 200–03 passim, 216, 225, 226, Benguela 13, 160, 179, 180, 181, 187, 188, 190, 229, 242, 245, 246, 248 191, 193 Clarence-Smith, Gervase 11 Berlin conference 8, 12, 46, 159 Class struggle 6, 9, 26, 28, 99, 103 Bié 150, 158, 159, 180, 190, 212n35 Cold War 82n5, 89, 90, 150, 151, Boundaries 21, 202, 225, 228, 233, 246 203, 225 Bourdieu, Pierre 80, 86, 100 Colonialism xxi, 1, 2, 3, 13, 14, 18, 19, 25, 31, Bourgeoisie 14, 15, 16, 40n37, 46, 103, 104, 42, 51, 70, 82, 87, 95, 96, 118, 126, 129, 130, 118, 144, 163, 178, 226, 238, 239, 248 134, 135, 153–56 passim, 158–166, 169, 170, 266 index

178, 182, 193, 197, 201, 207, 209, 211, Democratic United Movement 225, 237 (MUD) 186, 190 Colonial situation 2, 33, 124, 178, Durkheim, Emile 231 182, 238 Colonial wars xvi, 17, 53n94, 60, 66, 227, Eastern Europe 15, 83 229, 236 Egypt 89 Comintern 16 Elites xv, xix, xxi, xxii, 6, 11–20 passim, 25, Communal villages 21, 136, 241 28, 40n37, 44, 58, 73n180, 80, 86, 91, Communist party and ideas 88, 89n12, 92, 104n3, chapter 5; Elite and moder- 100, 130, 138, 189, 190 nity 128–32, 139–42; Elites and Comparison chapter 1, 36, 120, 150, 243–50 education 140–47; 151, 152, 153; New CONCP 38n27, 104n3 elites 161–67, 170–73 passim, 178–80, Congo 2n5, 87, 181, 185, 191, 217, 225, 229, 188, 191, 192; 200, 204, 205n19, 212–20 244; see also Zaire passim, 228; Elites and progress 233–36, Congregationalist 155–8, 160, 164n44, 241, 243, 245, 247, 248, 250 168, 170 Estado Novo 85, 177, 180, 244 Consumption xix, 54, 62, 132, 140, 145 Ethiopia 130, 218, 242 Cooper, Fred 4, 129 Ethnicity xviii, 10, 16, 17, 19n27, 24, 26, 28, Cosmopolitanism 80, 81, 90–91, 96, 97 31, 32, 42, 46n70, 48, 50, 51, 52, 53, 60–61, Cosmologies 32, 33, 45, 48, 52, 53, 54, 59, 73n180, 81, 84, 91, 92, 97, 99, 130, 151, 152; 60, 62, 69 Ethnicity, Religion and Costa Gomes, Francisco da 7 Nationalism 153–58; 160, 166, 173, 174, Cotton 94, 98, 181, 236, 241 200, 202, 205n19, 207, 233, 236, 237, 239, Counter-insurgency 136, 164, 171 246 Coup d’état 14, 21, 28n60, 37, 44, 49n78, Europe (EU) 2, 16, 63, 81, 87, 88, 89, 131, 56, 60, 61, 65, 66, 83, 215, 236, 240n31, 165, 185, 191, 193, 223, 227, 242, 244 244, 246 CPLP xvi, xvii, 243 Falcão, Fernando 191–94 passim Craveirinha, José João xiv, 108n14, 120 Fanon, Franz 40n37, 88, 104, 117, 118, 124, Creole 12, 13, 19, 25, 27, 37, 46, 47, 53, 58, 238, 239, 248 62, 70, 72, 75, 163, 177, 214, 234, 235; see Ferguson, James 129, 130 also kriol, mestiço, mulatto FNLA 149, 150, 153–5, 191, 193, 194, 201, 202, Cuba 6, 39, 40, 44, 53, 89, 90, 150, 152, 157, 204, 207, 222, 227, 228–29 171, 204, 205n17, 209, 210, 229, 236, 237, Football 157, 217, 220, 224 243, 245, 246, 248 Forced labour 9, 13, 45, 94, 133, 156, 160, Culture xiv, xvii, xxii, xxiii, 9, 26, 32, 35, 38, 168, 182, 229, 235, 236 39, 40, 41n43, 48, 51, 53–58 passim, 80, France 2, 7, 10, 12, 14, 15, 20, 26, 87, 103, 181, 92, 95, 100, chapter 4, 132, 135, 137, 162, 185, 192, 225, 227, 234 167–171 passim, 214, 217, 219, 221, 222, Frelimo xiv, xvi, xviii, xxii, 14, 18, 21, 22, 23, 236, 238 24, 65n152, chapter 3, chapter 4, chapter Cunha, Monsenhor Alves da 185, 186 5, 235–41 passim, 243, 245–48 passim Cuito Cuanavale 208, 246 Frente de Unidade Angolana (FUA) 183, 191–94, 196 Dar-es-Salaam 13n27, 52n93, 82n6, Fula 51, 62, 239 90, 96, 110 Funeral associations 21, 93 Dáskalos, Sócrates 186, 191, 192 Davidson, Basil 33, 88 Gambia 8, 63n142 Decolonisation xvi, xvii, 4, 5–7, 26, 82, 83, Geiger, Susan xix, xx 87, 88, 89, 90, 94, 119, 149, 194 Gellner, Ernest 11, 103, 106 202, 204 Germany 15, 208 Decolonisation without independence 4, Ghana 87, 89, 103 7, 26 Goa 7, 26, 86, 189 Domestic mode of production 6, 9, 27 Guebuza, Armando 22, 24, 114 index 267

Guevara, Ché 88, 93 Latin America 4, 5, 6, 8, 9, 26, 104 Guiledje xiii Legitimacy 1, 10, 15, 22, 24, 32, 45, 59n125, Guinea Conakry 43, 49, 52n93, 246 60, 66, 71, 94, 97, 105, 109, 128, 131, 132, Gungunhane 17 140, 146, 147, 150, 152, 160, 173, 199–211 Gwambe, Adelino 23 passim Liberalism 139, 140 Heroes xiii, xiv, 17, 31, 38, 46, 69, 215, 221 Liberia 12, 221 Heróicos (newspaper) 110n19, 111, 112, 113 Liga Guineense 46, 47 Highlands (Angola) 150, 151, 152, 155, 156, Lineage 26, 27, 60n131, 61, 64, 65, 68n163, 159–65 passim, 171, 172, 174, 179, 186, 202, 97, 98 212, 218, 227; see also Planalto Lourenço Marques 80, 85, 86, 87, 91, 138, Historiography xvi, xviii, 152, 204 234; see also Maputo Hobsbawm, Eric 103, 105, 120 Luanda 13, 150n3, 154, 155, 163, 172, 180, 184, Hodgkin, Thomas Lionel xxv 188–92 passim, 202, 205, 207, 208, 210, Huambo 150, 157, 171, 174, 186, 187, 189, 190, 213, 215, 217–21 passim, 223–29 passim, 193, 203, 210, 212n35, 219, 220, 245 236 Lubango 185, 188, 220 Ideology xx, 2, 3, 4, 10, 16, 21, 22, 25, 27, 37, 39, 70, 80–83 passim, 87, 127–47 Machel, Samora xiii, 80, 82, 85, 98, 106, passim, 199, 200, 201, 207, 210, 224, 229, 136, 137, 143 233, 240 Macao 7, 26 IMF 84, 226, 242 Maconde (Makonde) 8, 19, 23, 84, 85, 86, Imperialism 8, 11, 13, 92, 99, 118, 226, 237 93, 94, 95, 97, 98, 100 Imperial library 121, 122, 124 Magic 38, 44, 45–52, 53n98, 54, 60, 62, India 71, 103, 189 63, 70–74 passim, 139; see also Indians 12, 14, 98, 133, 134, 135, 244 witchcraft Indígenas 9, 164–67, 173, 182, 183, 234 Malawi 84, 112 Indigenous societies 27, 168 Mandiga 51 Indonesia 7, 14, 89, 103 Manica & Sofala 8, 84 Intellectuals 58, 81, 96, 98, 118, 129, 130, 183, Manjaco 51, 62n137, 65n151, 68n163 189, 191, 212 MANU 23, 94, 94, 95 Internationalism 103, 104, 113, 117, 125, 126 Mao (Maoism) 16, 82, 88, 89, 99, 210, 224 Islam 50, 51, 54, 69 Maputo 22, 80, 85, 127, 128, 131–34 passim, Israel 82, 89 138, 139, 142, 143, 144, 146, 147, 234; see also Lourenço Marques Jamba (town) 150, 157, 203 Marx, Karl 15, 88 Jamba, Jaka 210–11, 213 Marxism 3, 10, 14, 15–17, 22, 25, 99, 224, Jamba, Sousa 213, 214 233, 241n37 Junta militar 66n55, 68, 69 Marxism-Leninism 10, 22, 83, 99, 103, 128, 130, 210, 233, 239, 240, 243, 249 Kennedy administration 82 Massacre 94, 95, 192, 222, 240 Kenya 7, 14, 84, 93, 94, 95, 103, 184, 196, 197, Mbembe, Achille 33, 59 218, 227 Meillassoux, Claude 6 Kimbundu 217, 235 Memory xiii, xxiv, 31, 37, 46, 70, 73, 75, 120, Kikongo 217 125, 140, 160, 219, 221, 225, 237 Kingdom 7, 158, 159, 160, 166, 167, 173, Messiant, Christine 14, 151n5, 153, 155, 162, 218, 227 200, 208 Kissinger, Henry 229 Mestiço 152, 163, 183, 201, 210, 212, 214, 235; Kongo empire 19, 160, 164, 218 see also creole, kriol, mulatto Kovaso 193 Methodist 155, 223, 224 Kriol 12, 37, 53, 57, 62, 75, 234; see also Migration 9, 63, 85, 97, 134, 244 creole, mestiço, mulatto Missionary 49, 81, 84, 85, 92, 98, 157n25, Kriston 46, 47, 49n78, 58, 69, 234 160, 162, 167–172 passim, 178, 182, 227 268 index

Mixed-race 6, 12, 14, 26, 86, 134, 135, 214; Neo-colonialism 1, 2, 20, 34, 124, 209, see also race, creole, kriol, mulatto, 225, 226 mestiço Neto, Agostinho xiii, 203, 208, 215, 223 MLSTP 14 Neto, Armando da Cruz 207 MNLA 190 Newspapers 67, 110–12, 117n32, 120, 179, Mobilisation xix, 45, 47, 50–54 passim, 212, 213, 214 55n104, 80, 92, 93, 94, 132, 147, 236 Nigeria 12, 220, 230, 241, 249 Mobutu, Sese Seko 229 Nkavandame, Lazaro 95–8, 99, 100 Moçambique island 13, 234 Núcleo Negrófilo de Manica & Sofala 8 Moçâmedes 160, 188, 190; see also Namibe Nyerere, Julius 89, 218 Modernisation 10, 21, 22, 24, 25, 27, 131, 139, 140, 146, 147, 152, 168, 170–73 passim Oil 188, 219, 220, 225, 226, 229, 230, 241, Modernism (high modernism) 126, 130, 242, 248, 249, 250 135, 140, 145 One-party rule 15, 17, 23, 59, 60, 83, 206, Modernist nationalism 23, 91, 129, 131, 132, 210, 241, 242; see also Party-state 136, 137, 139, 144, 146, 147, 153, 171, 202, Opposition xiv, 24, 82, 84, 91, 97, 133, 232, 235 156, 186, 190, 191, 195, 210, 211, 212, Modernity 31–32, 49, 54, 128–147 passim, 215, 218, 225, 227, 239, 241, 242, 245, 169, 171, 173, 174 246, 247 Mondlane, Eduardo 52n93, 80, 81–82, 85, Organização Socialista de Angola 90, 91, 92, 96–97, 99 (OSA) 186, 191 Monuments xiii, xiv, 31, 219 Ovimbundo 17, 150n3, 151–61 passim, 164, Mozambican African Association 23 166, 168–74 passim, 202, 205, 212, 218, MPLA xvi, xviii, 13, 14, 18, 22, 42, 149–55 227, 236, 237, 245 passim, 157, 163, 171–74 passim, 190–94 passim, 199–216 passim, 228, 229, 236, PAICV 44 237, 242, 243, 245–47 passim PAIGC xvi, 14, 18, 27n59, 28, 38, 40, 41, 42, Mueda 8, 19n37, 23, 94–95, 236 44, 47, 48, 49n78, 50, 53n94, 56, 58n118, Multiparty elections/system 36, 59, 65, 68, 59, 61, 62, 66, 68, 73, 74, 236, 237, 239, 70, 211 243, 246 Mulatto 134, 177, 179, 182, 183, 184, 186–193 PAJOCA 215 passim, 195, 196; see also creole, kriol, Pan-Africanism 38, 40, 41, 91, 103, 104n2, mestiço 120, 124, 125, 221 Music xix, 139, 172, 215 Party-state 36, 59, 62, 70, 239–243; see also One party rule Nachingwea 110n19, 114 Paternalism 24 Namibe 13, 160; see also Moçâmedes Patrimonialism 24, 65n152, 92, 98, 100 Namibia 82, 150, 209, 237, 246 Patriotism, xv, xvi, xxiv, 38, 142, 219 Narrative xvii, xxii, 31–35 passim, 37, 47, PCP 189; see also Communist party and 70–73 passim, 76, 109, 122, 126, ideas 152, 166, 173, 174, 201, 204, 206, 209, 210, Peasant 10, 14, 15, 22, 27n59, 28, 49, 58, 83, 214, 231, 238 85n7, 87, 92, 93, 95, 99, 100, 128, 130, 135, Nation-building 3, 22, 23, 24, 147, 136, 138, 145, 203, 215, 218, 228, 241 208, 223 Pentecostals 224–5 Nationhood 4, 17, 19n37, 21, 22, 35, 59, 60, Pepetela 224 67, 104, 202, 205, 207, 208, 215, 239 Pereira, Aristides 36n21, 42, 43 Nationism xxi, 24–25 PIDE xvi, 87, 164, 190; see also Secret Ndau 8 police New Man 18, 40, 105, 114, 122, 124, 136, 137, Pidjiguity 40, 52, 236 139, 210, 238 Planalto (Angola) 150n3, 152, 156–61 Neo-liberalism xv, 130, 243 passim, 163–70 passim, 172, 173, 227, 236; Niassa 110n19, 112, 114 see also Highlands Nkomati accord, 246 Poder popular 37, 55, 57, 59 Nkrumah, Kwame 39, 40, 89, 217, 221 Poetry chapter 4, 203 index 269

Portugal xvi, 1, 5, 6, 7, 8, 11, 12–15, 17, 26, 27, Samakuva, Isaías 211 42, 43, 46, 82, 87, 96, 121, 140, 142, 144, Santos, José Eduardo dos 207–09, 217, 218, 149, 152, 164, 165, 171, 177, 179–87 passim, 233, 248 190–92 passim, 194, 197, 201, 210, 217, 220, São Tomé e Principe 13, 189, 234, 242 221, 224–30 passim, 234, 236, 237 Saudi Arabia 226 Portugalidade 210 Savimbi, Jonas 152, 156, 157, 174, 202, Prazo 8, 12 203, 204, 206, 207, 210, 223, 227, 228, Presbyterians 223–24 246, 247 Primordialism xviii, xix, 124, 128, 129, Scott, James C. 33, 99, 131 146, 147 Secret police/political police xvi, 82n6, 87, Proletariat/proletarianisation 9, 103, 189, 88, 164, 190, 193, 196, 221, 228; see also Propaganda 82, 93, 205n19 PIDE Protestantism 84, 85, 152, 154–58, 161–64 Senegal 8, 12, 63, 67, 246 passim, 167, 169, 170, 224, 227, 236; see Settlers 5, 7, 8, 14, 17, 26, 82, 83, 87, 141, 160, also Congregationalists, Baptists, 161, 165, chapter 7, 219, 222, 244 Methodists, Missionaries, Presbyterians Shrines 51, 52, 53, 62, 69, 219, 224 Sierra Leone 12, 34, 66 Quijano, Abel 5 Sisal 94, 160, 165, 187 Slaves 5, 6, 8, 12n24, 26, 27, 158, 166 Race 6, 12, 14, 26, 87, 121n46, 122, 130, 133, Social milieus 5, 12, 14, 20, 28 183, 194, 208, 213, 214, 215; see also Socialism 19, 129, 132, 134, 136, 139, 140, 144, mixed-race 147, 233, 240 Racism 87, 104, 150, 182, 213, 235 South Africa 7, 13, 14, 22n45, 81n4, 82, 83, Radio 56, 207, 228 84, 85, 87, 134, 138, 144, 151, 168, 177, 178, Rancière, Jacques 105, 126, 179, 181, 184, 185, 194n44, 196, 197, 204, Ranger, Terence 103, 105, 120 208–09, 218, 220, 221, 236, 237, 244, 245 Red Cross 207 Soviet Union 16, 19, 82, 83, 89, 90, 99, 103, Refugees 66, 81, 82, 83, 88, 94, 95, 219, 229 104, 130, 151, 229, 236, 237; see also Russia RENAMO xiv, 13, 22, 83, 84, 133, 138, 145, Spínola, António 42 237, 245, 246, 247 Spirits 40, 51, 52, 53, 54, 62, 66, 68, 139; see Régulos 46n70, 55, 62n137, 65, 132, 136, 245; also Witchcraft and Magic see also Chiefs and Traditional chiefs Stalin (stalinism) 15, 16, 20, 22, 25, 88, 89, Revolution 14–16, 27, 39, 40, 56n109, 67, 99, 147, 224, 241n37 75, 88, 89, 93, 97, 103–06, 114, 116–26, 130, Strike 40, 52, 53n94, 93, 236 131, 138–150 passim, 221, 233, 236, 238, Swaziland 144 239, 244 Switzerland 15, 156n22, 228 Revolution (Bourgeois) 15, 16 Swahili 92, 95, 218, 233 Revolution (Carnation) 7, 150, 192, 194, 236, 244 Taiwan 226 Revolution (Cuban) 6, 39 TANU 23, 94, 129 Revolution (French) 4n9, 10, 15 Tanganhica Mozambique Maconde Union Revolution (Haiti) 27 (TMMU) 23 Revolution (National) 7, 15, 16, 125 Tanganyika xix, xx, 43, 84, 85, 89, 93, 94, Rhodesia (Southern) 7, 22n45, 82, 83, 84, 95; see also Tanzania 85, 138, 184, 194, 237, 244, 245, 246; see TANU 23, 94, 129 also Zimbabwe Tanzania xix, 23, 43, 83, 84, 103, 110, 121, Rituals 21, 51, 54, 59, 68n163, 74, 97 218, 233, 240, 249; see also Tanganyika Russia 18, 19, 86n8, 87, 93, 225; see also Taxation 45, 101, 170, 203 Soviet Union Television 139, 207, 215 Terras do Fim do Mundo 150, 152, 173 Sakala, Alcides 206, 207 Tete 24, 84, 112, 113, 114 Salazar, António 1n2, 85, 87, 138, 150, 154, Touré, Sékou 43, 48 164, 165, 177, 180, 181, 183, 184, 185, 186–87, Tradition 2, 8, 9, 14, 21, 26, 32, 46, 49, 190, 194, 195, 221, 244 65n152, 81, 91, 92, 93, 97, 98, 99, 105, 108, 270 index

115, 116, 120, 124, 125, 135, 136, 146, 147, Vieira, Sergio 108n14, 137 154, 160, 161, 168, 170, 171n56, 174, 234, Vieira, João Bernardo ‘Nino’ 28n60, 235, 245 44, 49, 56, 66, 67, 73, 74n181, 75n188, Traditional chiefs 21, 26, 65n152, 81, 93, 240, 246 132, 134, 136, 160; see also Chiefs and Vietnam 89, 90, 99, 225, 228 Régulos Villagization 240 Tribalism 16, 50, 51, 61, 174, 205, Violence 45, 46, 54, 71, 73, 74, 76, 95, 149, 222, 235 192, 193, 240, 242, 245

Udenamo 23, 24 War of liberation xiv, 17, 18, 60, 238 Uige 209, 220 Washington 79, 90, 229 Umbundu 163, 168n50, 217 Washington, Booker T. 168 UNAMI 23, 24 Witchcraft 48, 49, 54, 61; see also Magic Union of the Populations of Angola and Spirits (UPA) 153n8, 191–93 passim, 228 Whites 14, 27, 86, 87, chapter 7, 213 UNITA xxii, 84, chapter 6, 194, 199, 202–09 Women xv, xix, xx, 9, 49, 61, 64, 91, 110, passim, 210–12, 215, 216, 222, 227, 228, 168, 172, 219, 222, 228 229, 236, 237, 245 World Bank 63, 84, 226 Unita-Renovada 206 World Council of Churches (WCC) 150 United Kingdom (UK) 14, 144; see also World-System 8, 80, 89, 90, 96, 98, 100 Britain United Nations 84, 183, 187, 193 Youth xiv, 87, 121, 215 United States (USA) 6n12, 12, 81, 82, 90, Yugoslavia 21 151, 204, 224, 229, 237, 242; see also America Zaire 201, 229; see also Congo UPANG 58n121, 60n130 Zambezi (river) 13, 83, 223 Utopia 43, 71, 98–100, 105, 126, 130, 147, Zambezia 24, 84 219, 224 Zambia 112, 130, 149, 196, 197, 217 ZANU 129 Vanguard 40, 45, 91, 100, 128, 135, 239 Zanzibar 43 Veterans xv, xvi, 87, 88 Zimbabwe 34, 82, 83, 84, 129, 138, Vieira, Luandino 190, 194n43 196, 197, 243