African Political Systems Revisited: Changing Perspectives on Statehood and Power 9781800734739

Citation preview

African Political Systems Revisited

INTEGRATION AND CONFLICT STUDIES

Published in Association with the Max Planck Institute for Social Anthropology, Halle/Saale Series Editor: Günther Schlee, Arba Minch University, Ethiopia, and Director emeritus at the Max Planck Institute for Social Anthropology Editorial Board: Brian Donahoe (Max Planck Institute for Social Anthropology), John Eidson (Max Planck Institute for Social Anthropology), Peter Finke (University of Zurich), Jacqueline Knörr (Max Planck Institute for Social Anthropology), Bettina Mann (Max Planck Institute for Social Anthropology), Ursula Rao (Leipzig University), Stephen Reyna (Max Planck Institute for Social Anthropology) and Olaf Zenker (Martin Luther University, Halle-Wittenberg) Assisted by: Viktoria Giehler-Zeng (Max Planck Institute for Social Anthropology) The objective of the Max Planck Institute for Social Anthropology is to advance anthropological fieldwork and enhance theory building. ‘Integration’ and ‘conflict’, the central themes of this series, are major concerns of the contemporary social sciences and of significant interest to the general public. They have also been among the main research areas of the institute since its foundation. Bringing together international experts, Integration and Conflict Studies includes both monographs and edited volumes, and offers a forum for studies that contribute to a better understanding of processes of identification and inter-group relations. Recent volumes: Volume 26 African Political Systems Revisited: Changing Perspectives on Statehood and Power Edited by Aleksandar Bošković and Günther Schlee Volume 25 Entrepreneurs of Identity: The Islamic State’s Symbolic Repertoire Christoph Günther

Volume 21 Space, Place and Identity: Woɗaaɓe of Niger in the 21st Century Florian Köhler Volume 20 Mobile Urbanity: Somali Presence in Urban East Africa Edited by Neil Carrier and Tabea Scharrer

Volume 19 Volume 24 Playing the Marginality Game: Identity Politics in After Corporate Paternalism: Material Renovation West Africa and Social Change in Times of Ruination Anita Schroven Christian Straube Volume 18 Volume 23 The Wheel of Autonomy: Rhetoric and Ethnicity in Lands of the Future: Anthropological Perspectives the Omo Valley on Pastoralism, Land Deals and Tropes of Felix Girke Modernity in Eastern Africa Volume 17 Edited by Echi Christina Gabbert, Fana Bishkek Boys: Neighbourhood Youth and Urban Gebresenbet, John G. Galaty and Günther Change in Kyrgyzstan’s Capital Schlee Philipp Schröder Volume 22 On Mediation: Historical, Legal, Anthropological and International Perspectives Edited by Karl Härter, Carolin Hillemanns and Günther Schlee For a full volume listing, please see the series page on our website: http://www.berghahnbooks.com/series/ integration-and-conflict-studies

African Political Systems Revisited Changing Perspectives on Statehood and Power Edited by Aleksandar Bošković and Günther Schlee

First published in 2022 by Berghahn Books www.berghahnbooks.com © 2022 Aleksandar Bošković and Günther Schlee

All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher.

Library of Congress Cataloging-in-Publication Data Names: Bošković, Aleksandar, editor, author. | Schlee, Günther, editor,   author. Title: African political systems revisited : changing perspectives on statehood   and power / edited by Aleksandar Bošković and Günther Schlee. Other titles: Integration and conflict studies ; v. 26. Description: New York : Berghahn Books, 2022. | Series: Integration and   conflict studies; v. 26 | Includes bibliographical references and index. Identifiers: LCCN 2021052747 (print) | LCCN 2021052748 (ebook) |   ISBN 9781800734722 (hardback) | ISBN 9781800734739 (ebook) Subjects: LCSH: African political systems. | Political anthropology--Africa. |  ­Indigenous peoples--Africa. | Tribal government--Africa. Classification: LCC GN492 .A37 2022 (print) | LCC GN492 (ebook) |   DDC 306.2096--dc23/eng/20211028 LC record available at https://lccn.loc.gov/2021052747 LC ebook record available at https://lccn.loc.gov/2021052748

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library

ISBN 978-1-80073-472-2 hardback ISBN 978-1-80073-473-9 ebook

Contents

List of Illustrations

vii

Forewordix Adam Kuper Chapter 1.  The Right Book at the Right Time: Early Reactions and Continuing Debates Aleksandar Bošković Chapter 2.  African Political Systems and Political Anthropology Herbert S. Lewis Chapter 3.  Complementary Segmentary Opposition, Early Kingship and the Looming State: Bridging the Dichotomy of African Political Systems Simon Simonse

1 15

47

Chapter 4.  The Shilluk Reth: Early King or Head of State? An Inter-Nilotic Exploration79 Simon Simonse Chapter 5.  A New Approach to the Political Anthropology of Africa: From African Political Systems and Tribes without Rulers via The Early State  Petr Skalník

102

vi  Contents

Chapter 6.  Retaliation, Mediation and Punishment in Ankole: Revisiting the Chapter by Oberg 122 Günther Schlee Chapter 7.  Beyond ‘African Political Systems’? The Relevance of Patrilineal Descent in Moments of Crisis in Northern Somalia 139 Markus V. Hoehne Chapter 8.  Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan 163 Munzoul Assal Chapter 9.  The Nkandla Controversy: Insights from African Political Systems Robin Palmer

180

Chapter 10.  Rethinking Tswana Kingships and Their Incorporation in Modern Botswana State Formation Ørnulf Gulbrandsen

206

Afterword241 Bilinda Straight Index247

Illustrations

Figures 7.1  Map of political divisions, northern Somalia  7.2 Map of territories of clan-families and some clans in northern Somalia  7.3  Simplified genealogical sketch of the Dhulbahante clan  7.4 Elders under the ‘peace tree’; second from left, with white hat: Caaqil Shiine (Samakaab Cali)  7.5 The members of our delegation; third from the left (standing): Ayaanle, the target of the killers; on his left: Axmed ‘Daakir’; on his right: Maxamuud Ilweyn 

145 149 151 154 156

Tables 3.1  Types of precolonial political organization prevalent in Africa 3.2  Complementary segmentary opposition 3.3 The king as a segment in a political field structured by complementary segmentary opposition 4.1 Periods during which recorded observations were made of the autonomous operation of royal power in the three studied types of kingdom before their transformation by colonial incorporation

48 55 56 80

Foreword Adam Kuper

The three most influential contributions made by ‘British social anthropology’ to the literature on African affairs – even to the study of politics – were African Political Systems, E.E. Evans-Pritchard’s The Nuer and Max Gluckman’s Analysis of Social Situation in Modern Zululand. The theoretical perspective that informed them was laid out by A.R. Radcliffe-Brown in a presidential address to the Royal Anthropological Institute, ‘On Social Structure’ (Radcliffe-Brown 1940a). Remarkably, all these publications appeared in 1940. One could hardly imagine a less auspicious moment. The Battle of Britain was raging. Britain was in existential danger. The future of the British Empire was precarious. On 24 October 1939, a few weeks after war was declared, the Council of the Royal Anthropological Institute ‘agreed that the Institute would be prepared to act as an intermediary to make available to the government the expert knowledge of anthropologists and others, which might be of use to the national war effort’.1 And yet there are no indications that the book was conceived and written under the shadow of a world-shattering conflict. At the same time, British colonial policy in Africa was in crisis. The future of Indirect Rule was in question. Yet today it requires a leap of imagination to see that African Political Systems does touch here and there on urgent debates on colonial administration. More immediately, the editors were preoccupied with parochial, academic concerns, notably the smouldering feud between two factions of anthropologists, the followers of Bronislaw Malinowski at the London School of Economics and the followers of A.R. Radcliffe-Brown at Oxford. ‘British social anthropology’ was

x  Adam Kuper

still a small, insecure field. In the early 1920s, Malinowski and Radcliffe-Brown displaced the old-school evolutionists and diffusionists, and recast anthropology as a social science. In the 1930s, they became rivals, scrapping with each other over jobs and grants, and becoming increasingly divided on questions of theory. The Oxford School emphasized the comparative study of social groups and roles, while the Malinowskians gave priority to ‘cultural’ processes and to individual strategies in competition for power and resources. Radcliffe-Brown had been appointed to the Oxford chair of social anthropology in 1937. He became the revered clan elder. His only colleagues were two much younger men, Edward Evans-Pritchard and Meyer Fortes, both appointed as ‘research lecturers in African sociology’. They were firm allies – Fortes described Evans-Pritchard as his ‘elder brother in anthropology’ (Goody 1995).2 Two other close associates were fellow South Africans and friends of Fortes. Isaac Schapera had studied under Radcliffe-Brown in Cape Town. A younger man, Max Gluckman, once a student of Schapera, had thrown in his lot with the Oxford men. ‘Radcliffe-Brown has often told me that he considers Fortes and Evans-Pritchard the two best (far and away) of all the younger British anthropologists’, Gluckman wrote to a friend, ‘I think Brown’s judgment here (unlike his account of S. African politics) quite correct’ (Gordon 2018: 153). African Political Systems was, among other things, a statement by this Oxford clique. It provided a showcase for their comparative, structural anthropology. The Oxford men were in reaction against Malinowski, who had dominated the field for the better part of two decades. His influence had been exercised in part through his weekly seminar at the London School of Economics, which was attended by virtually all the younger scholars who came into the field in the 1920s and 1930s. But Malinowski’s influence also had its material side. From 1933 to 1939, the Rockefeller Foundation funded research fellowships, administered by the International African Institute in London, to study ‘race relations’ in British African colonies. Malinowski was effectively put in charge of recruiting and training these research fellows. A generation of Africanist anthropologists was forged in Malinowski’s seminar. (In 1938–39, three of these young anthropologists, Fortes, Gluckman and Oberg, were candidates for a single research position at the newly formed Rhodes Livingstone Institute in Northern Rhodesia, directed by another former Rockefeller research fellow, Godfrey Wilson.) Seventeen research fellowships were awarded. All the eight case studies in African Political Systems were written by veterans of Malinowski’s seminar who had also been Rockefeller research fellows. To put it another way, just under half of the Rockefeller research fellows were recruited to write contributions to this volume. But, with the sole exception of Audrey Richards, they were no longer Malinowskians. The small circle around Radcliffe-Brown took charge of the production of African Political Systems. Radcliffe-Brown wrote the preface, which presented

Foreword  xi

his positivist, comparative, structural programme for ‘social anthropology’ (as did his presidential address to the Royal Anthropological Institute, also published in 1940). Fortes and Evans-Pritchard were the editors, and they wrote the ­introduction – Fortes apparently doing most of the work. Fortes, Evans-Pritchard, Gluckman and Schapera contributed half of the eight case studies. The other four case studies were very different in tone and content, and did not always conform to the Oxford programme. Audrey Richards was the only loyal Malinowskian (and the only woman) to contribute a case study to African Political Systems. She was already a stand-out critic of Evans-Pritchard and the Oxford men. A staunch, if liberal, supporter of British colonialism, Richards was shortly to join the Colonial Office, where she served for the duration of the war (Kuper 1996). Her chapter on the Bemba is the only one in the volume that contains a policy-oriented discussion of Indirect Rule in a particular colonial province. The remaining three contributors, Siegfried Nadel, Günter Wagner and Kalervo Oberg, had attended Malinowski’s seminar, but they were by no means down-the-line ‘functionalists’ or ‘structural-functionalists’, in fact, they were hardly to be classed as ‘British social anthropologists’ at all. Nadel and Wagner were Central Europeans. Oberg was a Canadian, the son of Finnish immigrants. All would make their careers largely or entirely outside Britain. Siegfried Nadel, who became a close friend of Fortes, was a graduate of the University of Vienna, where he specialized in the psychology of the perception of sound. For several years, he pursued a career in music and ethnomusicology. He was then (with Fortes and a Dutch scholar, Sjoerd Hofstra) appointed one of the first three Rockefeller research fellows. Following initial fieldwork in Nigeria, Nadel became actively engaged in British colonial administration. He was appointed Government Anthropologist of the Anglo-Egyptian Sudan in 1938. During the Second World War, he became a senior officer in the British colonial forces, was Secretary of Native Affairs in the British Military Administration of Eritrea and in 1945 became Secretary of Native Affairs in the British Military Administration of Tripolitania. In 1950 he was appointed the first Professor of Anthropology at the Australian National University, where he remained for the rest of his career. Despite his own experience as an administrator, Nadel did not write in any depth about British colonial rule. However, his contribution to African Political Systems, a study of the Kede riverain trading empire, a sort of state within a state, in the shadow of the Nupe polity, can be read as an implicit critique of the generalizations about African states in the editorial introduction by Fortes and Evans-Pritchard. Kalervo Oberg wrote a dissertation on the social organization of the Tlingit Indians under the supervision of Radcliffe-Brown at the University of Chicago. In 1934 he spent a post-doctoral year under Malinowski at the London School of Economics (LSE) and was given a Rockefeller fellowship to carry out research

xii  Adam Kuper

among the Ankole in southwestern Uganda, which resulted in his chapter in African Political Systems. He later made a career as an applied anthropologist in Canada and the United States (McComb and Foster 1974). Oberg’s account of the ‘kingdom’ of Ankole was also not obviously shaped by the preoccupations of the editors. As Günther Schlee argues in his chapter in the present volume: Oberg’s Ankole kingdom is classified as belonging to Group A in the introduction to APS … but on a closer look, it straddles the dichotomy between this group, the state-like societies with central authority, and Group B, the societies with segmentary lineage systems. There is no monopoly of power. Clans and lineages have to sort out internal affairs themselves and interclan violence needs to be authorized by the king, but the authorized party would have to carry it out by itself. The Banyankole therefore provide a model case for the interpenetration of different logics of action. The third of these outsiders, Günter Wagner, gave an account of local political systems in western Kenya. These appeared to be, in many crucial respects, similar to the systems on the other side of Lake Victoria, in western Uganda, described by Oberg. Tribal identities were uncertain, clan affiliations ambiguous, and political roles situational and only roughly defined. Here was another case study, based on excellent ethnographic research, which did not fit the models of ‘state’ and ‘stateless’ lineage-based societies presented by the editors of African Political Systems. But that was not the only incongruity in a classic, foundational text of ‘British social anthropology’. As Jan de Wolf (2003) remarks, the life of Günter Wagner was ‘disconcerting’.3 Born in Berlin in 1908, Wagner began his anthropological studies in Freiburg and Hamburg. His professor sent him to study for a year with Franz Boas at Columbia University. Directed by Boas, he made a field study of the Peyote cult among the Yuchi and completed a thesis on the topic for his doctorate at Hamburg University. He then carried out further fieldwork in North America, among the Comanche, and spent some time at Berkeley with Kroeber and Lowie. He was therefore a fully accredited Boasian. In 1933, he was appointed to an International African Institute Rockefeller-funded fellowship. He attended Malinowski’s seminar at the LSE and then carried out two years of fieldwork in Kenya. During a break from fieldwork, Wagner visited Evans-Pritchard at Oxford, and he wrote to Boas that he hoped to take part in the new research programme that Radcliffe-Brown was starting up there.4 He participated alongside Radcliffe-Brown, Evans-Pritchard and Fortes in an Oxford University Summer School for Colonial Administration. Then Britain declared war. Wagner returned

Foreword  xiii

to Germany, He was now technically an enemy alien, and Fortes had to apply to Oxford University Press to get special permission for Wagner’s chapter to be published in African Political Systems.5 In December 1939, Wagner joined an organization run by the Nazi Ministry of Propaganda called the Antisemitische Aktion. He later joined the Nazi Party and was appointed to the colonial policy departments of the Propaganda Ministry and of the Nazi Party. He completed his Habilitation under two well-known ethnologists who were deeply implicated in Nazi colonial projects, Richard Thurnwald and Dietrich Westermann. (In a letter to me, Andre Gingrich writes: ‘I tend to qualify both Westermann and Thurnwald as leading figures of German “Völkerkunde” under Hitler. Neither was an active Nazi party member – but you didn’t have to be one in order to support the Nazi terror regime and make it “work” in theory and practice as long as it lasted’) (Gingrich 2010).6 After the war, Wagner was designated a ‘fellow traveller’ of the Nazi Party. This made it difficult for him to get an academic appointment in Germany, although, according to Udo Mischek, Wagner’s main problem was that he was too closely identified with the British anthropologists!7 In any case, de Wolf writes that ‘Wagner appealed against this decision and got many of his colleagues, among them Evans-Pritchard, Fortes, Forde and Nadel, to write testimonials confirming the impartiality of Wagner’s ethnological work before the war’ (2003: 468). In 1947, Radcliffe-Brown invited him to apply for a post at Oxford, but Godfrey Lienhardt was appointed. In 1949, Wagner was appointed Assistant Government Ethnologist at the Department of Native Affairs in Southwest Africa (modern Namibia, which had been a German colony until the First World War and then came under South African administration). He served as an apparatchik of the apartheid system until his death in 1952. The ‘Introduction’ to African Political Systems by Fortes and Evans-Pritchard classified African polities into two types: ‘Group A’ and ‘Group B’, state and stateless political systems. This distinction was clearly relevant to the system of Indirect Rule, which depended on ‘chiefs’. However, the broad characterizations of the two types derived from established, broadly evolutionist ideas about clans and early states – ideas associated with Morgan, Robertson Smith, Maine and Durkheim, with perhaps a sidelong glance at Marx, at least in the contribution of Gluckman. Some of the assumptions made about African states drew on the work of C.G. Seligman (1934) on ‘divine kingship’, which had particularly influenced his two most loyal students, Evans-Pritchard and Schapera. The emphasis on lineages seems to have its origins in a suggestion of Radcliffe-Brown. ‘I was present on this occasion’. Fortes recalled. ‘Evans-Pritchard was describing his Nuer observations, whereupon Radcliffe-Brown said, as he stood in front of the fireplace: “My dear Evans-Pritchard, it’s perfectly simple, that’s a segmentary lineage system, and you’ll find a very good account of it by a man called Gifford”’ (Fortes 1979: viii)

xiv  Adam Kuper

This classification of political systems into states and the lineage-based stateless societies has been subjected to generations of commentary and criticism. As Aleksandar Bošković notes, the first wave of reviews of African Political Systems highlighted several of the issues that were to feature in later debates. Theories of ‘early states’ and the ‘conquest theory’ of state formation are expertly reviewed here by Petr Skalník. Other contributors tackle questions of the nature and even the reality of ‘tribe’, ‘clan’ and ‘lineage’. Perhaps the most powerful early critique, though largely implicit, was Edmund Leach’s Political Systems of Highland Burma. Published in 1954, this was (among other things) a counterblast to the whole genre of ‘tribal’, timeless, equilibrium models. Leach showed that the gumlao and gumsa political systems in the Burmese Highlands were intricately interconnected. Moreover, they were in a constant state of flux. And they were not to be divided into two enduring types – the states and the lineage-based systems – rather, leaders of dominant ‘lineages’ manipulated marriage alliances to construct small-scale ‘states’, but these statelets regularly collapsed back into rivalrous kinship networks. There were echoes here of the political processes sketched for East Africa by Oberg and Wagner in African Political Systems. However, Leach did not relate the petty politics of the Highland tribes to the imperial ambitions of their Burmese and Chinese neighbours, or to the machinations of British colonial officers, but then ethnographies that were written on political structures during colonial times usually left the colonial structure itself out of the picture. The system of Indirect Rule in British African colonies was designed to deal with ‘chiefs’ and ‘tribes’. It was based on the premise that African populations were divided into geographically demarcated and culturally homogeneous tribes. These tribes were presumed to represent the primary focus of individual identity and loyalty. They were ruled by ‘chiefs’ and ‘headmen’ who supposedly commanded unquestioned respect and obedience from their followers. It followed that the colonial administration should divide colonies into tribal districts and rule through the chiefs. As a Provincial Commissioner in Tanganyika summed up the policy in 1926: ‘Each tribe must be considered a distinct unit. Each tribe must be under a chief. Each tribe must be entirely within the borders of a district’ (Graham 1976: 4). This administrative set-up seemed to offer rich opportunities to the anthropologists, who could offer expertise on ‘tribes’ and ‘chiefs’. Skimming through Lord Lugard’s Dual Mandate, the foundational text of Indirect Rule published in 1927, Malinowski scribbled a triumphant note to himself: ‘[Lord Lugard’s] Indirect Rule is Complete Surrender to the Functional Point of View’ (Cell 1989: 483). Obviously, problems arose when administrators had to work with ethnically diverse, politically amorphous populations. Confronted with what seemed like anarchy – even ‘ordered anarchy’ – colonial administrators were inclined to invent ‘chiefs’ and ‘tribes’. It helped that anthropologists could describe how

Foreword  xv

these systems worked, as Evans-Pritchard did for the Nuer, Meyer Fortes did for the Tallensi and Günter Wagner did for the ‘Bantu Kavirondo’. But by the late 1930s, the Colonial Office was coming to recognize that Indirect Rule was a drag on economic development and that its focus on rural, ‘traditional’ populations could not be sustained as the towns grew, and an educated, Christian elite pressed for political representation. A colonial grandee, Sir Malcolm Hailey, was commissioned by the Colonial Office to conduct a thorough survey of Britain’s African colonies, with instructions to come up with recommendations for administrative reform. Assisted by Audrey Richards, Lucy Mair and Marjorie Perham, three women who had been stalwarts of Malinowski’s seminar and beneficiaries of the Rockefeller programme, Hailey delivered his report to the Colonial Office in August 1938. Published by Oxford University Press under the title An African Survey: A Study of Problems Arising in Africa South of the Sahara, running to over 1,800 printed pages, this report formed the basis of the Colonial Development and Welfare Act 1940, and five years later, at the end of the Second World War, the Colonial Development and Welfare Act 1945. These established a new framework for research and administration in the African colonies. If African societies were in the throes of change and if administrative priorities were shifting, perhaps functionalist anthropologists were no longer going to be very helpful. As Hailey remarked: ‘The problem of the maladjustments in African society created by the extension to it of Western economic or political institutions is no more amenable to treatment by the anthropologist than by anyone else’ (1938: 59–60). Critics pointed out that static equilibrium models could not account for social change, that they downplayed conflicts within communities and political systems, and that they assumed, against all the evidence, that social and political interactions took place only within strict geographical boundaries. Malinowski had anticipated this concern with social change. In 1929, he published a paper entitled ‘Practical Anthropology’ in the journal Africa, calling for ‘an anthropology of the changing native’ (Malinowski 1929). The ‘changing native’ was to be understood as a product of ‘culture contact’. In 1938, introducing a series of essays entitled ‘Methods of Study of Culture Contact in Africa’, Malinowski insisted that it was impossible to recover the precontact ‘baseline’ African cultures. The investigator was faced rather with a process, in which three foci could be identified: a complex of traditional institutions, beliefs and practices, that were, however, probably far removed from the preconquest institutions; the powerful, intrusive Western culture; and a hybrid culture that was emerging in the cities and that would spread into the country districts, fostering a new way of life for the whole society.8 This cultural model was challenged by Fortes, Schapera and Gluckman, three South Africans. In South Africa, the conflicts between the white ruling caste and the rest of the population was more stark, extreme and explosive than in British

xvi  Adam Kuper

African colonies. And political and academic debates were fraught and urgent. In the 1920s, a leading historian, W.M. Macmillan, attacked local anthropologists as ‘paralysed conservatives’, unable to see that Southern African ‘tribes’ could only be properly understood as creations of the South African state, their people living as deprived minorities within an unequal, oppressive and modern ‘common society’. There was no way back to ‘traditional’ tribal life (Macmillan 1989). In 1921, in his inaugural lecture as Professor of Social Anthropology at the University of Cape Town, Radcliffe-Brown took a similar radical line. The social systems of the African peoples in South Africa had been transformed by European interventions: ‘we inaugurated something that must change the whole of their social life’. From the principles of structural-functionalism, an ineluctable conclusion followed: ‘Segregation is impossible’ (Gordon 1990).9 As undergraduates at the University of Cape Town, Schapera and Fortes had taken social anthropology courses from Radcliffe-Brown. They agreed with Macmillan and Radcliffe-Brown that a ‘common society’, shot through with conflicts, had come into being in South Africa. The lives of ‘tribal’ populations were being shaped by the overarching institutions of a powerful state and an industrializing economy. As Schapera wrote: The missionary, administrator, trader and labour recruiter must be regarded as factors in the tribal life in the same way as are the chief and the magician. Christianity, in so far as it has been accepted, must be studied like any other form of cult … So, too, the trading store, the labour recruiter and the agricultural demonstrator must be considered integral parts of the modem economic life, the school as part of the routine educational development of the children, and the Administration as part of the existing political system. (Schapera 1935: 315) In 1940, in his presidential address to the Royal Anthropological Institute, ­Radcliffe-Brown made the argument in more general terms. The situation in African colonies could not be understood in terms of ‘culture contact’, as Malinowski proposed. It was imperative to think in terms of social structures: Let us suppose that we wish to study and understand what is happening in a British or French colony or dependency in Africa, at the present time. Formerly the region was inhabited by Africans having their own social structure. Now a new and more complex social structure has been brought into existence. The population now includes a certain number of Europeans – government officials, traders, missionaries and, in some instances, settlers. The new political structure is one in which the Europeans have a large measure of control, and they generally play an important part in the new economic structure. The outstanding characteristic

Foreword  xvii

of this kind of social structure is that Europeans and Africans constitute different classes, with different languages, different customs and modes of life, and different sets of values and ideas. It is an extreme example of a society compounded of heterogeneous elements. (Radcliffe-Brown 1940a: 10) The only contribution to African Political Systems to follow through on some of the obvious implications of the ‘common society’ model was Gluckman’s chapter on the Zulu. Best read as a companion piece to his ‘Analysis of a Social Situation in Modern Zululand’ (Gluckman 1940),10 the presentation was organized historically, setting out various ‘stages’ of Zulu history, and in a quasi-Marxist manner, it indicated strains and conflicts in each stage. (The history was drawn from a single source, A.T. Bryant’s Olden Times in Zululand and Natal, which had just been published, in 1938.)11 Gluckman insisted on the oppressive nature of South African rule over the Zulu, but he set out the ways in which brute force was embedded in more accommodative, and legitimate, structural relations. As Radcliffe-Brown noted in his preface to African Political Systems: ‘Dr Gluckman’s essay on the Zulu shows how the former system of a balance between the power of the chief, on the one side, and public sentiment, on the other, has been replaced by one in which the chief has to maintain as best he can some sort of balance between the requirements of the European rulers and the wishes of his people.’ This was in keeping with his insistence that ‘the political constitution must also be studied as an equilibrium system’ (Radcliffe-Brown 1940b: xxiii). In his account of the feud among the Nuer, Evans-Pritchard claimed that it sustained a balance between rivalrous tribes and clans. But Gluckman later confessed: ‘I was still thinking in crude functional terms of institutions – even civil war, which after all can be an institution – contributing to the maintenance of a rather rigidly conceived social structure’ (Gluckman 1963: 20). Evans-Pritchard would himself write a powerful, historical account of a segmentary society that was united only in its struggle against the Italian colonial regime (The Sanusi of Cyrenaica, published in 1949). The most common criticism of African Political Systems has been that it largely failed to confront the crisis in colonial government and did not address the sweeping, destabilizing social and economic changes that Africans were experiencing. The International African Institute had promised that the Rockefeller research fellows would concern themselves with colonial policy and social change (Richards 1944). In 1932, the Institute set out an ambitious programme in the form of a ‘Five Year Plan’ (bizarrely echoing the Soviet jargon). As Daryll Forde glossed the prospectus: ‘This was the first study, at once large-scale and intensive, to be undertaken of the structure of African societies and the changes taking place in them, particularly under the impact of western ideas, techniques, and economic forces’ (Forde 1951). In practice, however, the contributions to African

xviii  Adam Kuper

Political Systems were only occasionally relevant to debates about ‘race relations’ and colonial policy. Political and social change was generally dealt with briefly, if at all, in the concluding sections of the ethnographic chapters. According to Max Gluckman, Evans-Pritchard ‘had sworn he would never touch “Culture Contact” or “Culture Change”’ – but he [Gluckman] himself found modern politics more interesting and important’. Gluckman commented to Fortes that ‘other anthropologists were blind’ (Gordon 2018: 116–17). ­Evans-Pritchard was not blind, but he did argue that the study of change was less interesting, even less urgent, than studies of traditional systems. There were also divisions in principle over engagement in ‘applied anthropology’. The ­Radcliffe-Brown faction at Oxford was inclined to take a purist line on ‘science’, while the Malinowskians were ready, even willing, to work with the Colonial Office. ‘The racket here is very amusing’, Evans-Pritchard wrote to Fortes at the end of the Second World War: It would be more so if it were not disastrous to anthropology. Everyone is advising government – Raymond [Firth], Forde, Audrey [Richards], Schapera. No one is doing any real anthropological work – all are clinging to the Colonial Office Coach. This deplorable state of affairs is likely to go on, because it shows something deeper than making use of opportunities for helping anthropology. It shows an attitude of mind and is I think fundamentally a moral deterioration. These people will not see that there is an unbridgeable chasm between serious anthropology and Administration Welfare work. (Goody 1995: 73)12 This had not always been Evans-Pritchard’s view. When he was preparing to study the Nuer in the field, their prophets were being hanged by the British, and Nuer resistance was ruthlessly being suppressed. Soon, as Douglas H. Johnson records, ‘a vigorous new attempt to administer the Nuer would begin, with the building of roads, dispensaries, and administrative centres, and the organization of the Nuer into a new administrative system under government appointed chiefs’ (Johnson 1982: 231). Evans-Pritchard offered his services to the Sudan administration, writing to a senior colonial official in 1929: ‘If I can be of any assistance in furthering this work by making investigations amongst the Nuer I shall be glad to do so’ (ibid.: 232). A decade later, Evans-Pritchard was still sometimes willing to consider applied research projects. Robert Gordon has discovered that Evans-Pritchard and Fortes submitted an unsuccessful funding application in March 1940, with a strong covering letter from Radcliffe-Brown, bidding for £6,000 to study modern political development in Africa: ‘This project took as its basis the approach they had adopted in the soon-to-be-published African Political Systems, which

Foreword  xix

emphasized that the British colonial policy of “indirect rule” required a comprehensive knowledge of indigenous political institutions’ (Gordon 2018: 152). If that was really the goal of the editors of African Political Systems, the book was not an unqualified success. But, in any case, it is not read today as a guide to the stresses and strains of Indirect Rule. So far as the editors themselves were concerned, the volume had a wider purpose. In their ‘Introduction’, Fortes and Evans-Pritchard wrote: ‘We believe that the eight societies described will not only give the student a bird’s-eye view of the basic principles of African political organization, but will also enable him to draw a few, perhaps elementary, conclusions of a general and theoretical kind’ (1940: 4). These were significant claims, and they were largely justified. The editors were right to insist that political theory could not in good faith ignore the real-world variety of political systems, or fall back on the fantasy anthropology of Thomas Hobbes, Jean-Jacques Rousseau or Karl Marx. The comparative ambition of Radcliffe-Brown was also vindicated, to some degree at least. The typology set out by the editors has been very influential, albeit subjected to serious criticisms. Perhaps most importantly, the editors brought together a set of pioneering, authoritative, ethnographic accounts of African political traditions. Remarkably, some of these case studies are helpful in understanding contemporary African politics. As Simon Simonse and Robin Palmer demonstrate in their contributions to the present volume, these range from the civil war in Southern Sudan to the position of ‘traditional authorities’ in South Africa. But the most enduring achievement of this volume was that it has helped to launch an ethnographically informed political anthropology and, as a consequence, to broaden the horizons of political theory. Adam Kuper has taught anthropology in universities in Uganda, South Africa, the United Kingdom, the Netherlands and the United States. He was most recently Centennial Professor of Anthropology at the LSE and a visiting professor at Boston University. His latest books are The Chosen Primate: Human Nature and Cultural Diversity; Culture: The Anthropologists’ Account; and Incest and Influence: The Private Life of Bourgeois England (all published by Harvard University Press). He is now writing a history of museums of anthropology.

Notes  1. https://www.therai.org.uk/archives-and-manuscripts/archive-contents/census-of-brit​ ish-anthropologists-a71 (retrieved 3 September 2021).   2. See Chapter 4, ‘Personal and Intellectual Friendships: Fortes and Evans-Pritchard’, and Chapter 6, ‘The Oxford Group’.   3. This is an extended review of a doctoral thesis: Mischek (2002).   4. See de Wolf (2003: 465); Mischek (2002: 74).   5. See Wolf (2003:466; Mischek (2002: 77–78).

xx  Adam Kuper   6. On Thurnwald, see Gingrich (2005: 106, 121–23, 130).  7. In Imperfect Interpreters: South Africa’s Anthropologists 1920–1990, W.D. ­Hammond-Tooke cites a letter written to him by Mischek expressing this view (­Hammond-Tooke 1997: endnote, p. 209).   8. A series of papers originally published in the journal Africa were collected and issued in 1938 with an introduction by Malinowski: Methods of Study of Culture Contact in Africa, Memorandum (no. XV) of the International Institute of African Languages and Cultures, London.   9. Inaugural Lecture, Cape Times, 25 August 1921. Reprinted as an Appendix in Gordon (1990). 10. For excellent commentaries on this classic text, see Macmillan (1995); and Gordon (2018). 11. See Wright (1991).  12. Letter written in July 1945. Cited in Goody (1995: 73).

References Cell, John W. 1989. ‘Lord Hailey and the Making of the African Survey’, African Affairs, 88: 481–505. De Wolf, Jan. 2003. ‘A Disconcerting Life’, Africa, 73(3): 461–72. Evans-Pritchard, E.E., and Meyer Fortes. 1940. ‘Memorandum on a Plan of Research into Problems of Modern Political Development in Africa’ (with an introduction by A.R. ­Radcliffe-Brown), Hebdomadal Journal, 175: 185–96. Forde, Daryll. 1951. ‘International African Institute 1926–1951, Report of the Administrative Director’, Africa, 21(3): 226–34. Fortes, Meyer. 1979. ‘Preface’, in Ladislav Holý (ed.), Segmentary Lineage Systems Reconsidered: The Queen’s University Papers in Social Anthropology. Belfast: Queen’s University of Belfast, Department of Social Anthropology, pp. vii–xii. Fortes, Meyer, and E.E. Evans-Pritchard. 1940. ‘Introduction’, in Meyer Fortes and E.E. ­Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 1–24. Gingrich, Andre. 2005. ‘Ruptures, Schools, and Nontraditions: Reassessing the History of Socio-cultural Anthropology in Germany’, in Frederik Barth et al. (eds), One Discipline, Four Ways: British, German, French and American Anthropology. Chicago: University of Chicago Press, pp. 59–153. ———. 2010. ‘Alliances and Avoidance: British Interactions with German-Speaking Anthropologists, 1933–1953’, in Deborah James, Evie Plaice and Christina Toren (eds), Culture Wars: Context, Models and Anthropologists’ Accounts. New York: Berghahn Books, pp. 19–31. Gifford, E.W. 1926. ‘Miwok Lineages and the Political Unit in Aboriginal California’, American Anthropologist, 28: 389–401. Gluckman, Max. 1940. ‘Analysis of a Social Situation in Modern Zululand’: (A) ‘The Social Organisation of Modern Zululand’ and (B) ‘Social Change in the History of Zululand’, Bantu Studies, 14: 1–30, 147–74. ———. 1947. ‘Malinowski’s “Functional” Analysis of Social Change’, Africa, 17(2): 103–21. ———. 1963. Order and Rebellion in Tribal Africa. London: Routledge. Goody, Jack. 1995. The Expansive Moment: Anthropology in Britain and Africa, 1918–1970. Cambridge: Cambridge University Press.

Foreword  xxi Gordon, Robert J. 1990. ‘Early Social Anthropology in South Africa’, African Studies 49: 15–48. ———. 2018. The Enigma of Max Gluckman: The Ethnographic Life of a ‘Luckyman’ in Africa. Lincoln, NE: University of Nebraska Press. Graham, I.D. 1976. ‘Indirect Rule: The Establishment of “Chiefs” and “Tribes” in Cameron’s Tanganyika’, Tanzania Notes and Records, no. 77. Hailey, William Malcolm, 1938. An African Survey: A Study of Problems Arising in Africa South of the Sahara. New York: Oxford University Press. Hammond-Tooke, W.D. 1997. Imperfect Interpreters: South Africa’s Anthropologists 1920–1990. Johannesburg: Witwatersrand University Press. Johnson, Douglas H. 1982. ‘Evans-Pritchard, the Nuer, and the Sudan Political Service’, African Affairs, 81(323): 231–46. Kuper, Adam. 1996. ‘Audrey Richards’, in Edward Shils and Carmen Blacker (eds), Cambridge Women: Twelve Portraits. Cambridge: Cambridge University Press, pp. 221–44. Macmillan, Hugh. 1989. ‘“Paralysed Conservatives”: W.M. Macmillan, the Social Scientists and the “Common Society”, 1923–48’, in Hugh Macmillan and Shula Marks (eds), Africa and Empire: W.M. Macmillan, Historian and Social Critic. London: Institute of Commonwealth Studies, pp. 72–90. ———. 1995. ‘Return to the Malungwana Drift: Max Gluckman, the Zulu Nation and the Common Society’, African Affairs, 94(374): 39–65. Malinowski, Bronislaw. 1929. ‘Practical Anthropology’, Africa 2(1): 22–38. McComb, Marlin R., and George M. Foster. 1974. ‘Kalervo Oberg, 1901–1973’, American Anthropologist, 76(2): 357–60. Mischek, Udo. 2002. Leben und Werk Günter Wagners (1908–1952). Gehren: Escher. Radcliffe-Brown, A.R. 1940a. ‘On Social Structure’, Journal of the Royal Anthropological Institute of Great Britain and Ireland, 70(1): 1–12. ———. 1940b. ‘Preface’, in Meyer Fortes and E.E. Evans-Pritchard, African Political Systems. London: Oxford University Press, pp. xi–xxiii. Richards, A.I. 1944. ‘Practical Anthropology in the Lifetime of the International African ­Institute’, Africa, 14(6): 289–301. Salamone, F. 2000. ‘The International African Institute: The Rockefeller Foundation and the Development of British Social Anthropology in Africa’, Transforming Anthropology, 9: 19–29. Schapera, Isaac. 1935. ‘Field Methods in the Study of Modern Culture Contacts’, Africa: Journal of the International African Institute, 8(3): 315–28. Seligman, C.G. 1934. Egypt and Negro Africa: A Study in Divine Kingship. London: Routledge. Wright, John. 1991. ‘A.T. Bryant and “The Wars of Shaka”’, History in Africa, 18: 409–25.

Map of ethnic groups and kingdoms mentioned in this volume (Jutta Turner, © Max Planck Institute for Social Anthropology)

Chapter 1

The Right Book at the Right Time Early Reactions and Continuing Debates Aleksandar Bošković

Introduction Since its publication in 1940, African Political Systems (APS) has both inspired anthropologists (as well as sociologists like Eisenstadt (1959)), and at the same time provoked some important debates (Gellner 1969; Holý 1979; see also Lewis, Chapter 2 in this volume). In this chapter, I will first analyse the critical responses to the book, as published in the leading anthropological and sociological journals of the time, and will conclude with the general view about the issues that the contributions in this volume address. As an overview of the contributions to the present volume is not the main task of this chapter, readers will be able to find a more detailed overview of particular texts in the Afterword by Straight. In this chapter, I have collected all the available reviews of APS after it was first published. The last (and most recent) review considered here was published only after the Second World War (Paulme 1948). It is difficult to overstate the importance of this book for the history of anthropology and anthropological theory.1 First of all, despite the fact that this was not the first time that ‘political systems’ of traditional African kingdoms had been studied, the influence of this volume has proved to be very strong – considerably stronger than that of the volumes that preceded it.2 Second, along with later work of the French Africanist Balandier (1967), it proved to have a decisive influence on the establishment of the growing field of political anthropology. As the study of kingdoms gradually evolved into the study of state systems, scholars engaged in this field were among the contributors in the special issue of the journal Cahiers

2  Aleksandar Bošković

d’Études Africaines that provided important contributions to the study of state systems (Alexandre 1982). Only a year later, another special issue was published in France, this time of Pouvoirs: Revue française d’études constitutionelles et politiques (Ardant and Conac 1983), where ten scholars focused on power (as well as some contemporary issues like development). Finally, it opened up a wide-­ reaching debate on the actual importance (and, in some cases, of the very existence) and interpretation of the segmentary lineage system, and its relevance for understanding the structure of traditional African societies (Hammond-Tooke 1965, 1985; Subrahmanyam 1999). APS (especially the typology proposed in Radcliffe-Brown’s preface) also provided an important point of reference in the subsequent analyses of the key terms associated with segmentary models that could be used in the studies of Maghreb communities (Salem 1982: 117). Therefore, in many ways, it proved to be ‘the right text at the right time’, as stated by Geertz in another context (Geertz 1997). While it was not the first critical study of the political systems of African peoples (as made clear by Meek and Herskovits in their reviews, for example), it was published at a time of dramatic change in the world, both in a political (the Second World War, with the dawning of the end of colonial empires) and a theoretical (criticism of the still-dominant functionalist approach) sense, and, as such, paved the way for important new interdisciplinary research. This includes, but is not limited to, important new insights in the fields of customary law and its effects on the state institutions (Zenker and Hoehne 2018), as well as about strengthening of the institution of chieftaincy in recent years (Comaroff and Comaroff 2018). In this chapter, I will look at some early reactions to the book – they tell us a lot about the context in which the book was received and they also offer a glimpse into the making of contemporary anthropological debates about politics at all levels, issues of identity and statehood, and ways of dealing with (and writing about) difference. Unfortunately, there is no discussion about the political organization of African societies in some of the more recent studies of the development of ‘complexity in archaic states’ (Adams 2001), but I hope that the present volume can also add to these types of considerations.

First Reviews The first review of APS was published in the Journal of the Royal African Society in July 1940 (Anonymous 1940). It was not signed and it was under the rubric of ‘sociology and anthropology’. In this brief note, the author quoted from the editors’ introduction and listed all the chapters, concluding that: The book, then, is a comparative study of African political institutions by a team of trained investigators, working under the auspices of the International Institute of African Languages and Cultures. An ­Introduction

The Right Book at the Right Time  3

by the Editors explains the aims of the study, and draws such tentative conclusions from it as they regard as so far justified. The painstaking researches represented by this book into African political science, so far as developed in the indigenous institutions of a variety of tribes or nations in different parts of the Continent, are of obvious interest and value. (Anonymous 1940: 284) Another review was published in the same journal a year later by Margaret Wrong (1887–1948), a scholar who was better known for her missionary work, an O ­ xford-educated Canadian and former Secretary of the International Committee on Christian Literature in Africa. In her review, Wrong briefly outlined the contents of each essay and praised them as ‘stimulating’ (1941: 280). She also appreciated the amount of data that the book provides. However, she was critical of the volume’s apparent lack of coherence, pointing out that, for e­ xample: ‘Dr. Richards and Dr. Fortes, writing of the Bemba and Tallensi, supply illuminating material on the weakening of the political system through European conquest and rule, while Dr. Oberg, writing on the Kingdom of Ankole, devotes only one paragraph to this aspect of the subject’ (ibid.: 281). This, she felt, ­diminished ‘the comparative value of the book’, but it seems that she expected more of a comparative political study, even though she quoted the editors of APS in her review when they wrote that they were ‘more interested in anthropological than in administrative problems’ (ibid.). Charles Kingsley Meek’s (1885–1965) review was published in Man in 1941. In this, the author summarizes the individual contributions (especially praising the one by Fortes), stating that: ‘This book is a collection of essays by competent anthropologists dealing with the political structure of eight tribes in various parts of Africa, and should prove of great value both for the new material it provides, and as a stimulus to others to undertake similar descriptive studies’ (Meek 1941: 41). On the other hand, he noted important omissions: It may seem curious that in a book called African Political Systems there is no mention of standard works such as Lord Lugard’s Dual Mandate, Miss Perham’s Native Administration in Nigeria, Rattray’s Ashanti, etc. There are no references to ‘secret’ societies, whose authority, incidentally, may be much wider than that of the political group. And little is said about inter-group or inter-tribal relationships, the use of ‘ambassadors’, arbitrators, and so on. (Ibid.: 42) However, Meek also stated that the editors themselves saw this volume as a beginning of the work and an important aspect of a more general ‘inquiry’. From a more contemporary perspective, it seems quite interesting that he also thought that the book could be of use not only to students, but also ‘to administrative ­officers of the Colonial Service’ (ibid.). Perhaps this could also be seen as a

4  Aleksandar Bošković

r­eference to dealing with the ‘administrative problems’ referred to in the review by Wrong. Or perhaps this is a reflection of Meek’s personal wish, having served as ‘government anthropologist’ in southern Nigeria in the late 1920s, having being appointed by Lord Lugard himself.3

Reviews in American Journals Wilfrid Dyson Hambly (1886–1962) wrote a very positive review for the A ­ merican Sociological Review (Hambly 1940). Hambly was born and educated in the United Kingdom, but moved to the Field Museum of Natural History in Chicago in 1926, where he became Curator of African Ethnology. In his review of the book, he especially praised the fact that the chapters were all based on meticulous ethnographic research. On the other hand, he noted that would have appreciated more historical context, as well as the addition of some other geographical areas: One agrees heartily with the absence of theoretical speculation, such as might have resulted from a pedantic introduction of philosophical intricacies of Rousseau or Hobbes. We have here something for the plain, practical man. Nevertheless, one feels that a historical chapter dealing with such kingdoms as that of early Congo and Dahomey might have had a logical place in this scheme. The historical introduction to Gluckman’s essay on the Zulu is pertinent and helpful. (Hambly 1940: 797) Less generous was the review in American Anthropologist written by Melville J. Herskovits (1895–1963). At the time, Herskovits was establishing himself as the main specialist in African (and African American) studies in the United States in the Department of Anthropology that had been established only a couple of years earlier at Northwestern University. Herskovits immediately recognized the importance of APS, especially ‘for the wider field of comparative studies of political institutions in general’ (1941: 465). On the other hand, he criticized the omission of any reference to the works by Africanists like Roscoe, Meek, Labouret and Gutmann (ibid.: 466).4 Or, as he put it: It is somewhat unfortunate that a few simple additions were not made to the book that would indicate the attention paid to African political organizations by earlier students. It should not have been difficult to include recognition of the existence of such works, and it would have been gracious for some of the authors, at least, to have written without such utter disregard of the research of their older colleagues. (Ibid.: 465) Among the authors mentioned by Herskovits, C.K. Meek (who wrote the review of APS for Man mentioned above) was especially prolific (for example, Meek

The Right Book at the Right Time  5

1925, 1931, 1937). His Northern Tribes of Nigeria became an instant classic in African studies. He was educated in theology. Two other authors mentioned by Herskovits were missionaries, Anglican John Roscoe (1861–1932), and Lutheran Bruno Gutmann (1876–1966). Roscoe was fascinated by (and probably saw himself as an intellectual heir of ) David Livingstone, and his detailed ethnography of what was then the Uganda Protectorate was a major reference work for Africanists for decades (Roscoe 1921), although clearly it was not considered important enough by Wagner to refer to it in APS. Gutmann served between 1902 and 1938 among the Chagga near Kilimanjaro (in what is now Tanzania), fiercely defending their cultural particularity against unstoppable modernization and producing some fine ethnographic material in the process (Gutmann 1909, 1932/38). Finally, French ethnologist Henri Labouret (1878–1959) is the only one of the scholars mentioned whose work continues to enjoy respect and admiration, as well as influence among French Africanists (Labouret 1931, 1937). Labouret held many important posts, including Director of the African Institute (Institut international africain) and Professor of Sudanese Languages at the École nationale des langues orientales vivantes (1926–45), and he was also Professor of African Civilization at l’École coloniale, Paris (1926–45). Although the contributors to APS were probably familiar with the works of these scholars, the focus of most of these works (with the possible exception of Meek’s) seems very far removed from the idea of social anthropology that they were working on. Thus, while it is easy to see how some particular aspects of their work (like Roscoe’s vocal opposition to slavery or Gutmann’s and Labouret’s criticism of colonialism) were very close to those of Herskovits and the idea of cultural anthropology that he was developing, for the contributors to APS, they seemed like fleeting moments that had little relevance in terms of understanding the social and political processes. At the same time, the active involvement of some of them with the colonial authorities (especially Meek, until 1933, but also Labouret) did not help their case either. In retrospect, this also helps us to understand the limited influence that these scholars had later on, especially in comparison with those who contributed to APS. Herskovits was also critical of the fact that the volume only included references to the territories held by the British: Before the war, when this volume was projected, persons in France and Belgium could have been found to write on the political organization of selected tribes in the areas controlled by these other nations, and this would have greatly enhanced the value of the book and substantially reinforced conclusions drawn from its materials. (Herskovits 1941: 466) Herskovits criticized the dichotomy that the editors introduce (APS, p. 5), as he believed that ‘the second type’ (i.e. ‘societies which lack centralized ­authority,

6  Aleksandar Bošković

administrative machinery, and constituted judicial institutions’) was much more present in Africa and that there were other ‘informal control mechanisms’, as he called them, which regulated matters in specific societies. Turning to specific chapters, he saw ‘difficulties’ when ‘analyzing political institutions in “non-­ political” states’ (1941: 467): This is implicit, for example in Wagner’s definition of a political unit (p. 199 [of APS]). Another example is found in Fortes’ summary of the political institutions of the Tallensi (p. 241), while the same problem is inherent in Evans-Pritchard’s attempt to indicate what is meant by the term ‘tribe’ among the Nuer (p. 278). For this is untrod ground, and perhaps this is why two of the three contributors who are concerned with tribes of this kind give the impression of being uneasy over the absence of specific categories. (Ibid.: 467) Herskovits also strongly criticized Wagner’s discussion of ‘internal political structure’ as ‘merely a compilation of comments concerning social organization, gift exchange economics, religion, and … quasi-legal processes’ (ibid.). Evans-Pritchard fared slightly better because of his statement ‘of Nuer “political relativity”’, which made his arguments stronger. According to Herskovits, Fortes and Wagner were too heavily influenced by ‘the functionalist position, which holds that no cultural institution can be understood except in terms of its interrelationship with all other portions of the culture of which it forms part’ (ibid.). The concluding lines of this review forecast a major change (and a paradigm shift) that he claimed would occur in anthropology within a decade: This approach is shown to be impossible when dealing with shadowy, amorphous canons of regulation such as are found in ‘non-political’ ­African groups. And in Evans-Pritchard’s thesis that ‘the consistency we perceive in Nuer political structure is one of process rather than one of morphology’ (p. 296 [of APS]), we see reflected the importance of studying a given aspect of a culture in terms of the special problem it poses. (Ibid.: 467) This criticism was very powerful and detailed, but it also reflects a different view of anthropology. For Herskovits, anthropologists needed to take a much more active (or even activist) role than his colleagues working in Britain were ready to do. Colonialism was still alive and well (at least on the surface) at the time when APS was published, and its publication cannot be abstracted out of the political context of its time. The anthropologists who contributed to APS were engaged in a different type of approach from their colleagues on the other side of the Atlantic, actively trying to understand the societies they studied, but

The Right Book at the Right Time  7

with different emphasis and much more ambiguity. Herskovits did not find their approach very convincing. Of course, this should all also be put into the context of his own position at the time, as he had established the Anthropology Department only a couple of years earlier (1938) at Northwestern University and was in the process of establishing African Studies in the United States (Herskovits 1941a).

French Perspectives The final review that I will discuss here is the one published in L’Année sociologique in 1948. Since this journal was not published during the Second World War, the massive volume in which the review article appeared covered the years between 1940 and 1948. By then, APS was already in its third edition. The review was written by one of the leading French Africanists, Denise Paulme (1909–98), known for her fieldwork among the Dogon in West Africa and also for her work in museums. Between 1937 and 1961, she headed the ‘Afrique noire’ section at the Musée de l’Homme, and in 1957 became Senior Fellow at the École pratique des hautes études. She had studied law, but was attracted to ethnology after attending lectures by Marcel Mauss in 1927. Her work was greatly appreciated by leading Africanists (like Balandier), and Claude Lévi-Strauss helped her to secure a permanent position in 1957. She was one of the most influential people in the founding generation of French anthropology (Héritier 1999). In her review, Paulme first noted that the fact that the book was already in its third edition proved its importance. She summarized the content of all the chapters, focusing on key points, and occasionally adding interesting observations (for example, she felt that Nadel was too pessimistic in his assumption that ‘the Kede have now abandoned all pretention to the existence of an autonomous politics’ (Paulme 1948: 315)). This led her to conclude that the contributors of APS privileged in their studies disintegration processes to the processes of integration – also using Wagner’s chapter as an example (ibid.). Paulme was not satisfied with Evans-Pritchard’s definition of the clan (‘not all African clans are exogamous, as shown by Mrs Kuper among the Swazi’ (ibid.)), but quoted with approval (just like Herskovits before her) his relativistic definition of the ‘tribe’ among the Nuer. Although she was not as critical as Herskovits, Paulme found the distinction made by Fortes and Evans-Pritchard between two different types of societies to be ‘too vague’, as there are societies that have categories that resemble the ones of the state (Paulme 1948: 316). Here she drew on her own research experience in Sudan, but also on her expertise as a legal scholar. Just like Meek (quoted above), she noted with disapproval the absence of reference to ‘secret societies’, which could provide the dominant influence that, according to the editors, was

8  Aleksandar Bošković

missing from what they called the ‘Group B’ societies. She also noted that it was very strange, especially for a volume entitled African Political Systems, not to find any reference to ‘ambassadors’ – trade or political representatives sent to other societies. Finally, she also criticized the opposition of, on the one hand, an administrative organization and, on the other hand, a lineage system, as the lineage could also be seen as a form of administrative organization (ibid.).5 Nevertheless, she praised the book and concluded that it would ‘provide great benefit to ethnographers, students, and [sic] all colonial administrators’ (ibid.: 316).6 Paulme was also friends with one of the most fascinating figures of the last century, Michel Leiris (1901–90), an artist, poet, writer, critic, traveller, surrealist and ethnographer, a true ‘Renaissance Man’ whose friends included Breton, Bataille, Giacometti, Picasso, Césaire and Métraux. He was also a great innovator in modern confessional literature, writing several autobiographies and keeping a diary from 1922 until 1989. This confessional and very personal style influenced the creation of his first major ethnographic work, L’Afrique fantôme (Phantom Africa), published in 1934 and based on the notes and diaries from the Dakar-Djibouti expedition (1931–33 (Leiris 1996)). In the recent European history of ideas, the concept of ‘Africa’ has occupied a very prominent place. The idea of the ‘people without history’ was first formulated by Hegel in his lectures on the philosophy of history in the early nineteenth century. At the turn of the twentieth century, it was Conrad’s Heart of Darkness that captured the imagination of the people, and this image persisted for much of the century. The fascination with the mysterious other that this ‘Africa’ had to offer also influenced first appreciations of African art – from Henry Moore’s sculptures to cubism and Picasso. This fascination also made it into the Parisian surrealist circles of the 1920s. However, with scholars like Leiris, this image became much more complex – he saw (and, in doing so, was much ahead of his time) the artificiality of the whole construct. He realized that Europeans’ fascination with the ‘Dark continent’ was mostly the product of their own fears and insecurities, as well as a consequence of their inability to see others as being just as human as themselves. It was also a useful tool for colonial domination, whose consequences were present in aspects of everyday life (Bošković 2003: 527). Leiris also developed a specific – and quite original – version of functionalism that was completely independent of the forms practised by English-speaking anthropologists (Bošković 2010: 137). This brief excursion was necessary in order to contextualize and historicize one of the reviewers of this volume, but also to point to some of the directions in which studies of African societies were moving in France – and this reviewer’s influence was very strong. Thus, Paulme’s review should also be seen in the context of the emerging debates in French ethnology and anthropology at the time, about the colonial legacy, ‘African-ness’ and the whole concept of négritude.

The Right Book at the Right Time  9

Legacy and Influences The collection of essays edited by Fortes and Evans-Pritchard had a lasting effect. It started as a collective project by eight anthropologists influenced by ­Radcliffe-Brown’s version of structuralism, and offering examples of case studies that combined functionalist insights with regional expertise – although, as noted by Herskovits, Evans-Pritchard was already moving in a different direction.7 ­According to Radcliffe-Brown, this was the opportunity to give ‘the comparative study of political institutions, with special reference to the simpler societies … the attention it deserves’ (APS, p. xi): The task of social anthropology, as a natural science of human society, is the systematic investigation of the nature of social institutions … ­Applied to human societies the comparative method used as an instrument for inductive inference will enable us to discover the universal, essential, characters which belong to all human societies, past, present, and future. The progressive achievement of knowledge of this kind must be the aim of all who believe that a veritable science of human society is possible and desirable. (Ibid.) As some of the authors secured full-time positions in the emerging discipline, APS also provoked important debates, especially regarding kinship: During the great debates on kinship of/in the 1950s and 1960s, the model presented in African Political Systems was subjected to criticism from many quarters. Some felt that it was simply too neat to fit the complexities of real life. Others disparaged it as evolutionism in disguise. Yet others (most prominently Lévi-Strauss) rejected its exclusive focus on descent as the main principle of kinship. (Eriksen and Nielsen 2013: 89) The debates that began in 1940 led to important reconsiderations of kinship and descent in the following decades. The model of descent proposed in this volume was further developed by Evans-Pritchard in his studies of the Nuer (Evans-Pritchard 1940, 1951), and Fortes established himself as the most ­ ­influential scholar in kinship theory (Fortes 1949), distancing himself from ­Radcliffe-Brown (Kuper 2015: 61). APS had a very wide impact immediately after its publication – as noted by Easton (1959), for example. Based on some of his own research in Sudan, Holý (1979) strongly criticized the whole concept of the ‘segmentary lineage system’. Even today, the emphasis that the contributors of this volume place on the aspects of political organization and leadership still resonate as very important – and this is especially relevant with the political shifts taking place, particularly in East Africa, as well as the implications that these shifts have for the processes of

10  Aleksandar Bošković

ethnic alliances and identification (Schlee and Watson 2009a, 2009b; Bošković and Ignjatović 2012; Eidson et al. 2017). As our understanding of others takes different forms (the usefulness of the work for colonial administrators seems to be a relic of a bygone era), it is important to take note of the works that transformed the history of our discipline. And in the case of APS, it was clear from the first reviews that the importance of this work was immediately realized.

Conclusion: The Present Volume The chapters in the present volume address the issues raised by APS in two different ways. All the authors address both the book and the subject matter of the book (African political systems). Some approach it through the impact it had on the history of anthropology (Guldbransen, Lewis, Simonse, Schlee and Skalník), while others (Assal, Hoehne and Palmer) focus on its role in shaping anthropological theory (‘Group A’ vs. ‘Group B’ societies). Of course, as suggested by Günther Schlee, ‘two approaches boil down to the same, if we think of types of theory, how they succeed each other in time and how they are connected to each other’.8 A major area of overlap is that all chapters critically discuss the rigid dichotomy between Group A and Group B societies. Some do it by looking back into the history of anthropology and re-examining the interpretation of the examples discussed in APS and classical monographs, while others apply this dichotomy to the results of their own field research and examine the extent to which it works and also where it does not work. Gulbrandsen begins with certain ‘cultural idioms’, but this discussion of Tswana kingship brings him all the way to modernity and postcolonial Botswana. While Tswana kingdoms served to legitimize the strong state with parliamentarian democracy, they have at the same time produced and influenced various contradictions, as demonstrated in his chapter. Lewis takes as his starting point the debate about the primacy of the social versus the individual and presents a detailed account of different aspects in the formation of APS, as well as subsequent readings of it. This debate is also reflected in Schlee’s chapter. Schlee starts with Oberg’s chapter from APS and, taking into account both ethnographic (Africanist) and comparative historical data, focuses on the process of state becoming a state (‘thinking about the state’). Again, Lewis also notes the fact that the contributors to APS ignored work on similar topics that had already been published before 1940 (as noted above). While also looking at the book from a historical perspective, Simonse presents ‘an ethnographic case where the structural continuity between societies of Group A and Group B is clearly visible’. Skalník takes an even wider approach, comparing the book (as well as its latent evolutionism) to later projects, like Tribes without Rulers, as well as his own work on the early state. Just like Gulbrandsen, Skalník is optimistic about the insights that APS

The Right Book at the Right Time  11

could bring into understanding parallelism of modern forms of state and the various nonstate polities, be it chiefdoms or acephalous communities but also present-day terrorist, ultranationalist, utopian and extremist religious political forms that may or not call themselves states. Höhne critically examines the concept of the segmentary lineage system and, based on his research in northern Somalia, argues that it does not really work (contrary to, for example, Holý (1979)), but that it can be used as a heuristic device. This leads him to caution against the rejection of a model that effectively helps to understand contemporary conflict dynamics in a volatile area far from overarching state control. Assal presents an interesting account of the Tuareg in northern Darfur (Sudan), as well as how they became victims of the recent political crisis in the country. Palmer also uses the ongoing political crisis in his own country (South Africa) in order to highlight some aspects of representation of a postcolonial society (Gluckman’s contribution in APS), as well as to offer anthropological insights into the cultural and ritual role of the homestead and the roots of traditional patriarchy to offer an interpretation of the Nkandla Controversy (a major scandal involving the former South African President). With the combination of historical insights and contemporary ethnographic research, the contributors to this book offer valuable insights into the paradigms and challenges facing societies in the first decades of the twenty-first century. I believe that they also speak loudly and clearly about the importance of anthropological perspectives, and how they could help us understand the challenges of a world with shifting paradigms. Aleksandar Bošković is Senior Research Scientist at the Institute of Archaeology in Belgrade, S­erbia. He has taught at the Faculty of Social Sciences, the ­University of Ljubljana (­Slovenia), the University of Brasília (Brazil), Rhodes University in Grahamstown (South Africa), the University of the Witwatersrand (Johannesburg, South ­Africa), University of Belgrade (Serbia), the University of Donja Gorica (Montenegro) and the University of St Andrews (Scotland). His books include William Robertson Smith (Berghahn Books, 2021), Mesoamerican Religions and Archaeology ­(Archaeopress, 2017), Other People’s ­Anthropologies (Berghahn Books, 2008, editor) and The Anthropological Field on the Margins of Europe (LIT Verlag, 2013, co-editor with Chris Hann).

Notes 1. For more detailed references, see also Lewis, Chapter 2 in this volume. 2. This does not imply anything about the actual quality of the books preceding this one, but the fact is that some authors who wrote in APS had a lasting influence on social and cultural anthropology. Herskovits explicitly mentions omissions of four authors, and I will come back to them when discussing his review.

12  Aleksandar Bošković 3. Lord Lugard (1858–1945) was the Governor General of Nigeria (1914–1919), and was the one who came up with the idea of ‘Indirect Rule’. Kuper (2015: 81) offers an excellent explanation of the position of British social anthropologists within the colonial context, as well as about some common misconceptions about the relationship between anthropologists and colonial administrators. 4. Radcliffe-Brown (APS, p. xiii) expected that the volume would ‘stimulate other anthropologists to give us similar descriptive studies’. So, the authors around APS ­ did not see themselves as producing something ‘definitive’. The editors stated in their ­introduction that their aim ‘was to provide a convenient reference book for anthropologists’ (ibid.: 1). 5. In her own words: ‘Mais dans beaucoup de sociétés africaines, les sociétés secrètes peuvent exercer cette influence dominante. A ce propos, il peut sembler étrange que, dans un livre qui s’intitule Systèmes politiques africains, on ne trouve aucune référence aux sociétés secrètes dont l’autorité peut déborder parfois l’unité politique; peu de choses sur les rapports internationaux, l’emploi d’ “ambassadeurs”, les rapports entre organisation économique et système politique. Il paraît enfin un peu arbitraire d’opposer d’une part une organisation administrative, d’autre part un système de lignages, le lignage étant lui aussi, vu sous un certain angle, une organisation administrative’ (1948: 316). (‘In many African societies, secret societies could exert a dominant influence. That is why it seems strange that in a book with the title African Political Systems there are no references to secret societies whose authority can sometimes overwhelm political unity; little about international relations, the use of “ambassadors”, the relationship between economic organization and political system. Finally, it seems a little arbitrary to oppose an administrative organization, with a kinship system, as the kinship is also being, seen from a certain angle, a form of administrative organization’. 6. Perhaps following up on the sentence from the Editors’ Note: ‘We hope this book will be of interest and of use to those who have the task of administering African peoples. The anthropologist’s duty is to present the facts and theory of native social organization as he sees them’ (ibid.: vii). 7. Of course, as noted by Eriksen and Nielsen, ‘structural-functionalism’ was not ‘a static doctrine; it was a theory that evolved and ultimately transcended itself ’ (2013: 90). 8. Personal communication, 2019.

References Reviews of APS

Anonymous. 1940. ‘African Political Systems by M. Fortes; E.E. Evans-Pritchard’, Journal of the Royal African Society, 39(156) (1940): 283–84. Hambly, Wilfrid Dyson. 1940. ‘African Political Systems by M. Fortes; E.E. Evans-Pritchard’, American Sociological Review, 5(5) (1940): 796–98. Herskovits, Melville J. 1941. ‘African Political Systems by M. Fortes; E.E. Evans-Pritchard’, American Anthropologist, N.S., 43(1) (1941): 465–68. Meek, C.K. 1941. ‘African Political Systems’, Man, 41 (1941): 41–42. Paulme, Denise. 1948. ‘Fortes (M.) et Evans Pritchard (É.). – African Political Systems’, ­L’Année sociologique, 1 (1940–48): 313–16. Wrong, Margaret. 1941. ‘African Political Systems by M. Fortes; E.E. Evans-Pritchard’, Journal of the Royal African Society, 40(160) (1941): 280–81.

The Right Book at the Right Time  13

Literature

Adams, Robert McC. 2001. ‘Complexity in Archaic States’, Journal of Anthropological ­Archaeology, 20(3): 345–60. Alexandre, Pierre (ed.) 1982. Systèmes étatiques africains. Special issue of Cahiers d’Études ­Africaines 22, vol. 87–88. Paris: EHESS. Ardant, Philippe, and Gérard Conac (eds). 1983. Les pouvoirs africains. Special issue of ­Pouvoirs: Revue française d’études constitutionelles et politiques No. 25. Paris: Seuil. Balandier, Georges. 1967. Anthropologie politique. Paris: PUF. Bošković, Aleksandar. 2003. ‘Michel Leiris: Ethnologist in Search of Meanings’, Anthropos, 98(2): 526–29. ———. 2010. Kratak uvod u antropologiju. Zagreb: Jesenski i Turk. Bošković, Aleksandar, and Suzana Ignjatović. 2012. ‘Understanding Ethnic Conflicts through Rational Choice: A Review Article’, Ethnos, 77(2): 289–94. Comaroff, John L., and Jean Comaroff (eds). 2018. The Politics of Custom: Chiefship, Capital, and the State in Contemporary Africa. Chicago: University of Chicago Press. Easton, David. 1959. ‘Political Anthropology’, Biennial Review of Anthropology, 1: 210–62. Eidson, John R., Dereje Feyissa, Veronika Fuest, Markus V. Hoehne, Boris Nieswand, Günther Schlee and Olaf Zenker. 2017. ‘From Identification to Framing and Alignment: A New Approach to the Comparative Analysis of Collective Identities’, Current Anthropology, 58(3): 340–59. Eisenstadt, S.N. 1959. ‘Primitive Political Systems: A Preliminary Comparative Analysis’, American Anthropologist, 61(2): 200–20. Eriksen, Thomas Hylland, and Finn Sievert Nielsen. 2013. A History of Anthropology, 2nd edn. London: Pluto Press. Evans-Pritchard, E.E. 1940. The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People. Oxford: Clarendon Press. ———. 1951. Kinship and Marriage among the Nuer. Oxford: Clarendon Press. Fortes, Meyer. 1949. The Web of Kinship among the Tallensi. London: Oxford University Press. Geertz, Clifford. 1997. ‘The Legacy of Thomas Kuhn: The Right Text at the Right Time’, Common Knowledge, 6(1): 1–5. Gellner, Ernest. 1969. Saints of the Atlas. London: Weidenfeld & Nicolson. Gutmann, Bruno. 1909. Dichten und Denken der Dschagganeger: Beiträge zur ostafrikanischen Volkskunde. Leipzig: Evangelical Lutheran Mission. ———. 1932–38. Die Stammeslehren der Dschagga, 3 vols. Munich: Beck. Hammond-Tooke, W.D. 1965. ‘Segmentation and Fission in Cape Nguni Political Units’, Africa, 35(2): 143–67. ———. 1985. ‘Descent Groups, Chiefdoms and South African Historiography’, Journal of Southern African Studies, 11(2): 305–19. Héritier, Françoise. 1999. ‘Denise Paulme-Schaeffner (1909–1998) ou l’histoire d’une volonté’, Cahiers d’études africaines, 39(153): 5–12. Herskovits, Melville J. 1941. The Myth of the Negro Past. New York: Harper & Brothers. Holý, Ladislav. (ed.). 1979. Segmentary Lineage Systems Reconsidered. Belfast: The Queen’s ­University Papers in Social Anthropology, vol. 4. Kuper, Adam. 2015. Anthropology and Anthropologists: The British School in the Twentieth ­Century, 4th edn. Abingdon: Routledge. Labouret, Henri. 1931. Les tribus du rameau Lobi. Paris: Institut d’ethnologie. ———. 1937. Le Cameroun. Paris: Paul Hartmann. Leiris, Michel. 1996. Miroir de l’Afrique. Paris: Quarto, Gallimard.

14  Aleksandar Bošković Meek, C.K. 1925. The Northern Tribes of Nigeria: An Ethnographical Account of the Northern Provinces of Nigeria Together with a Report on the 1921 Decennial Census, 2 vols. Oxford: Oxford University Press. ———. 1931. A Sudanese Kingdom: An Ethnographical Study of the Jukun-Speaking Peoples of Nigeria. London: Kegan Paul, Trench, Trubner. ———. 1937. Law and Authority in a Nigerian Tribe. Oxford: Oxford University Press. Roscoe, John. 1921. Twenty-Five Years in East Africa. Cambridge: Cambridge University Press. Salem, Lilia Ben. 1982. ‘Intérêt des analyses en termes de segmentarité pour l’étude des sociétés du Maghreb’, Revue de l’Occident musulman et de la Méditerranée, 33: 113–35. Schlee, Günther, and Elizabeth E. Watson (eds). 2009a. Changing Identifications and Alliances in North-East Africa, Volume I: Ethiopia and Kenya. New York: Berghahn Books. ———. 2009b. Changing Identifications and Alliances in North-East Africa, Volume II: Sudan, Uganda, and the Ethiopia-Sudan Borderlands. New York: Berghahn Books. Subrahmanyam, Sanjay. 1999. ‘Les aventures de l’ “État segmentaire”’, Critique Internationale, 3(1): 45–54. Zenker, Olaf, and Markus Virgil Hoehne (eds). 2018. The State and the Paradox of Customary Law in Africa. New York: Routledge.

Chapter 2

African Political Systems and Political Anthropology Herbert S. Lewis

African Political Systems contains within it elements of two great and opposing social science perspectives or orientations that continually oscillate through the intellectual world. One argues that the social system is everything, controlling, forcing or cajoling the individual to conform. This view is represented by ­Auguste Comte, Karl Marx, Herbert Spencer, Emile Durkheim, Alfred R. ­Radcliffe-Brown, Alfred L. Kroeber and Leslie A. White. The other declares that this view is reification, that social systems are abstractions and that individuals and groups of individuals are all that is real. This view is represented by William James, Gabriel Tarde, Max Weber, Franz Boas, Bronislaw Malinowski and Raymond Firth. This chapter will address these paradigms and their oscillation in twentieth-century British and American anthropology.

The Background to African Political Systems African Political Systems (APS) was the work of a very small and select group – all but two of them students of Bronislaw Malinowski or participants in his seminars at the London School of Economics (LSE).1 Edward E. Evans-Pritchard, Isaac Schapera and Audrey Richards were three of his first five Ph.D. students; Meyer Fortes had a doctorate in psychology, but joined Malinowski’s seminar. Nadel received his Ph.D. a few years later, but Max Gluckman received his Ph.D. at Oxford with Evans-Pritchard and Fortes, and also spent time in Malinowski’s seminar. These first Ph.D.s in British anthropology were the cream of a very small but remarkable crop, along with Raymond Firth, Ian Hogbin, M ­ argaret

16  Herbert S. Lewis

Read, Phyllis Kaberry and Hortense Powdermaker. From Malinowski they learned, most importantly, the gospel of Malinowskian fieldwork and also the principles of Malinowskian functionalism. This involved a focus on culture, the ‘functional interrelationship’ of all the elements of a culture and the importance of culture for the maintenance of human life as well as social systems. Above all, ­Malinowski’s ‘scientific theory of culture’ began with the needs of the individual, including the biological ones. In the words of Audrey Richards: ‘Malinowski gave data on individuals as well as groups, on variations in behaviour as well as conformities. His islanders strained at society’s rules, fell in love, committed adultery, jumped off palm trees, bragged, cheated, quarrelled, and were subject to the romantic call of dangerous overseas “Argonauts”’ (1957: 28). (As Malinowski put it, his ‘savages’ were ‘as self-seeking and as self-interested as any Christian’ (1926: ix).) His ethnographies and those of his students were full of actors, activities, action and life. But by the mid-1930s, his holism and his theory of needs had come to seem naïve, unscientific, tautological and unworkable to some of his followers. Enter Radcliffe-Brown, a major influence on English-speaking students of anthropology in several countries, who became accepted as the more sophisticated theoretician. Radcliffe-Brown had taught at the University of Cape Town (1920–25), the University of Sydney (1925–31) and the University of Chicago (1931–37) before his return to become the first Chair in Social Anthropology at Oxford in 1937. Although he did not produce Ph.D.s in the way that M ­ alinowski did, he greatly influenced students, and Evans-Pritchard and Fortes envisioned APS in terms of Radcliffe-Brown’s version of structural-functionalism. By the 1950s, his approach had become dominant in British anthropology. Radcliffe-Brown was fervently devoted to his theory in the way converts to causes tend to be. After studying at Cambridge with Alfred C. Haddon and William H.R. Rivers, who had stressed history, psychology and diffusion, Radcliffe-Brown found God – in the form of Emile Durkheim and his Parisian followers. Durkheim’s models became Radcliffe-Brown’s – from the ‘collective conscience’ imparted by shared symbols and rituals (The Andaman Islanders) to the mechanical solidarity supplied by complex marriage and kinship systems (The Social Organization of Australian Tribes), to the organic solidarity of the division of labour. These ideas were basic to Durkheim’s and Radcliffe-Brown’s central problem: social solidarity and the coherence of societies, which were said to be like biological organisms. Herbert Spencer was clearly another influence on Radcliffe-Brown through his organic simile (society as an organism) and his notions of social statics and social dynamics. In contrast to Malinowski, Radcliffe-Brown would not consider the biological individual and individual psychology for the purposes of analysis and theory. As Phyllis Kaberry put it: ‘If, in Malinowski’s [ethnographic monographs], the people are always with us (and, some would say, too much with us), in ­Radcliffe-Brown’s they are conspicuous by their absence’ (1957: 75).

African Political Systems and Political Anthropology  17

For Radcliffe-Brown and those who would follow him, the solidarity and maintenance of society, social integration and order within society was paramount. Quoting Fortes: ‘His starting-point, the view that “any persisting culture (is) an integrated unity or system in which each element has a definite function in relation to the whole”’ (Fortes 1949: vi). As the dissenting Edmund R. Leach wrote: ‘Social systems were spoken of as if they were naturally existing real entities and the equilibrium inherent in such systems was intrinsic, a fact of Nature’ (Leach 1964: x).2 As Leach depicted Radcliffe-Brown’s paradigm: Individuals were presumed to be born free into a society composed of corporate institutions … The structural shape of such a corporation is intrinsically self-perpetuating and is independent of the individual lifespan of its particular members. Social structure is thought of as a network of relationships between ‘persons’ and ‘roles’. The stability of the system requires that the content of such relationships shall be permanent. In such a society every individual who fills a role finds himself under jural constraint to fulfil the obligations inherent in that role. More crudely, the customs of society are seen as providing a body of moral norms worked out in behaviouristic form; the discrepancies between individual behaviour and customary behaviour are due simply to the inability of the average man to live up to the moral demands of his society. (1961: 296–97) Radcliffe-Brown’s programme was to consist of: (1) comparison; (2) classification of types; and (3) generalization – working towards ‘general laws’ (Kuper 2015). This would be the essence of a ‘scientific’ approach, his ‘natural science of society’. As Radcliffe-Brown put it in the preface to APS: ‘Applied to human societies the comparative method used as an instrument for inductive inference will enable us to discover the universal, essential, characters which belong to all human societies, past, present, and future’ (Radcliffe-Brown 1940: xi). Radcliffe-Brown’s search for the universal and the essential in social structure required stripping away the ‘cake of custom’, i.e. the culture, that hides the comparable, classifiable and generalizable elements in societies. According to Adam Kuper, the shift from Malinowski and the LSE approach to that of Radcliffe-Brown and Oxford (meaning Fortes and Evans-Pritchard) meant a move from functionalism to social structure, from ‘the family, magic, and making a living – to … political and kinship systems and religion’ (Kuper 2015: 62). Furthermore, ‘[a] comparative study of political systems has to be on an abstract plane where social processes are stripped of their cultural idiom and are reduced to functional terms’ (Fortes and Evans-Pritchard 1940: 3).

18  Herbert S. Lewis

The Book The preface by Radcliffe-Brown and the introduction by Fortes and E ­ vans-Pritchard are the best-known parts of APS. Radcliffe-Brown repeated his exhortations regarding the significance of the comparative method for the study of political institutions, ‘by the methods of natural science’ (1940: xxiii) in the interest of a ‘veritable science of human society’ (ibid.: xi). And most famously, he stated the major principle that: ‘In studying political organization, we have to deal with the maintenance or establishment of social order, within a territorial framework, by the organized exercise of coercive authority through the use, or the possibility of use, of physical force’ (ibid.: xiv). ‘In dealing with political systems, therefore, we are dealing with law, on the one hand and with war, on the other’ (ibid.: xiv). And he claimed that: ‘The political organization of a society is that aspect of the total organization which is concerned with the control and regulation of the use of physical force’ (ibid.: xxiii). The key notion that political systems are, and political life is, about social order is the orienting idea of the book as well as the sort of political anthropology that issued directly from it. The problem of how social order could be maintained in social systems that lacked formal governmental institutions would be a major concern of two of the essays in this book as well as of many studies in the following twenty years. Radcliffe-Brown stated another major principle: ‘Social structure is not to be thought of as static, but as a condition of equilibrium that only persists by being continually renewed, like the chemical-physiological homeostasis of a living organism’ (ibid.: xxii). This idea became a major source of contention in the following years.3 The introduction to APS by Meyer Fortes and E.E. Evans-Pritchard is a central statement, an overall view of the elements that they saw as being fundamental to an understanding of the varied range of political phenomena in Africa. Critics pointed out many flaws, blinders and gaps, but it is a useful summary of what they considered to be the lessons of the studies in the book. The editors proclaim in a cavalier fashion that: ‘We have not found that the theories of political philosophers have helped us to understand the societies we have studied and we consider them of little scientific value’ (Fortes and Evans-Pritchard 1940: 4). Thus, despite the current origin myth of the three great founding fathers of anthropology, there is absolutely no hint of two of those great ancestors, Karl Marx and Max Weber. But the introduction is also remarkable for the virtual absence of references to any ethnographic literature outside the essays in the book. As both Melville Herskovits and Charles K. Meek point out in separate reviews, the editors never mentioned the works of others who had written on African political systems, even the books of Meek and Herskovits themselves (which the reviewers were too modest to note). Nor is there any reference to any theoretical anthropological literature or to political systems outside of Africa, except for some of the work of

African Political Systems and Political Anthropology  19

the American Robert H. Lowie. There is just the slightest mention of Durkheim; he is present in spirit, though not in person! For that matter, the ethnographic essays in the book are singularly lacking in either theoretical or comparative references, except the bare minimum related to the ethnographies themselves. The introduction begins with a completely unoriginal ‘classification’ of African political systems into two groups: ‘Group A consists of those societies which have centralized authority, administrative machinery, and judicial institutions – in short, a government – and in which cleavages of wealth, privilege, and status correspond to the distribution of power and authority’ (in other words, ‘primitive states’). ‘Group B, consists of those societies which lack centralized authority, administrative machinery, and constituted judicial institutions – in short which lack government – and in which there are no sharp divisions of rank, status, or wealth’ (in other words, ‘stateless societies’) (Fortes and Evans-Pritchard 1940: 5).4 The editors discuss the role of ‘kinship in political organization’, ‘the influence of demography’, ‘the influence of mode of livelihood’, the conquest theory of state formation, ‘the territorial aspect’, ‘the incidence and function of organized force’, ‘the mystical values associated with political office’ (by far the longest section), ‘the problem of the limits of the political group’ and ‘differences in response to European rule’, a very short section. Critics might take issue with many of their generalizations, but the section that has raised the most problems, with far-ranging ramifications, is the one called ‘The Balance of Forces in the Political System’. In this section, Fortes and Evans-Pritchard adumbrate their claim that political life is about the maintenance of order and the stability of systems. In addition, they are at pains to argue that African states are not sites of rampant exploitation, but are controlled by the checks and balances within the system, by countervailing forces of various types that ‘impose severe restrictions on a king’s authority’ (Fortes and Evans-Pritchard 1940: 11). ‘Looked at from another angle, the government of an African state consists in a balance between power and authority on the one side and obligation and responsibility on the other … The structure of an African state implies that kings and chiefs rule by consent’ (ibid.: 12). They recognize that there are chiefs and kings who would be despots, offering Shaka, the King of the Zulus, as a case in point, and they note that revolts and attempts to restore the balance by overthrowing the despot occur. However, they claim that: ‘It should be remembered that in these states there is only one theory of government. In the event of rebellion, the aim, and result, is only to change the personnel of office and never to abolish it, or to substitute for it some new form of government’ (ibid.: 13). Max Gluckman would later take this topic much further with his version of ‘conflict theory’. At the end of this section, in less than half a page, the editors make the claim that will launch many research projects, papers and books, along with much controversy:

20  Herbert S. Lewis

A different kind of balance is found in societies of Group B. It is an equilibrium between a number of segments, spatially juxtaposed and structurally equivalent, which are defined in local and lineage, and not in administrative terms. Every segment has the same interests as other segments of a like order. The set of inter-segmentary relations that constitutes the political structure is a balance of opposed local loyalties and of divergent lineages and ritual ties. (Fortes and Evans-Pritchard 1940: 13) Here we have a theory of order in nonstate societies inspired by Durkheim’s ‘mechanical solidarity’ and illustrated by (and, indeed, based upon) Evans-Pritchard’s model of the Nuer political system and Fortes’ account of the Tallensi lineage and kinship systems. Fortes would later attempt to extend the model more generally for ‘simple’ societies (‘The Structure of Unilineal Descent Groups’ (1953)). The model of the nonstate as a ‘segmentary system’, with lineages and systems of descent as the all-important segments, would be a major concern for British structural-functional anthropology for the next two decades. It would inspire both original research on such systems as well as reinterpretations of the Nuer and Tallensi ethnographies.

The Essays The first five essays focus on states and the last three on stateless societies. As suggested above, the chapters on the Nuer and the Tallensi bear the burden of their theory of order in stateless (segmentary) societies.5 However, the five that deal with states are less theoretical and hardly revolutionary. Indeed, there is not much to distinguish these chapters from accounts by historians of state systems anywhere, or from earlier good ethnographies, such as Robert S. Rattray’s ethnography of the Ashanti kingdom (1929), Meek’s study of the Ibo and Herskovits’ Dahomey. It is interesting that the chapters on the Zulu (Gluckman), Ngoni (Schapera), Bemba (Richards) and Kede (Nadel) each include discussions of change under colonial intrusion and rule, even though the image of structural-­ functionalism suggests that they would not – even could not – deal with these. Max Gluckman’s discussion of ‘The Period of European Rule’ presents a detailed summary of the situation of Zulu chiefs, the king and other royals in the face of European rule and the European magistrates.6 He also discusses the various competitions among Zulu political figures, leaders and the Zulu civil wars, and had wanted to use his pioneering essay, ‘Analysis of a Social Situation in Modern Zululand’ (1940b) in the chapter. This piece dealt directly with the complex relationships between the whites and the Africans in South Africa. Contrary to postcolonial dogma, Gluckman cannot be accused of ignoring either the contemporary scene or the realities of power in a European-dominated African society.

African Political Systems and Political Anthropology  21

Isaac Schapera’s essay, ‘The Political Organization of the Ngwato of Bechuanaland Protectorate’, introduces ‘the overriding authority of the British administration’ in the sixth line, once again contradicting the belief that the British functionalists ignored the presence of the colonial powers and their effects. For about two-thirds of the chapter, the author presents a rather formal account of the social and ‘governmental’ organization of the Ngwato, but in the last section, the formal account gives way to a discussion of ‘tribal politics’ that is ‘in fact made up to a considerable extent of the quarrels between the chief and his near relatives and of their intrigues against one another to command his favour’ (Schapera 1940: 79). This is Max Weber, not Radcliffe-Brown. Like those by Gluckman and Schapera, Audrey Richards’ chapter on the Bemba is not just an attempt to reconstruct the precolonial system, but also includes a good deal about the changes and the relations with colonial officials and missionaries. Siegfried Nadel hardly speaks of the British administration of the Kede or Nupe states of northern Nigeria, probably because these were under ‘Indirect Rule’ and there was enough to say about the history and administration of that polity in a short essay. But Nadel’s chapter is distinguished by its concluding paragraph, in which he writes of ‘psychological factors – that is, the social motive power that may lie in the temperamental and general psychological dispositions of a group’. This is a very unusual consideration for ‘British’ anthropologists of the time and more like the contemporary Americans Margaret Mead and Ruth Benedict. Furthermore, he writes that his emphasis on environmental factors was not ‘intended to minimize the decisive part that enterprising and far-seeing individuals must have played in the creation of the Kede State’. He speaks of ‘the leadership of certain outstanding individuals’ and asks ‘How much in the gradual social development of the Kede State was due to the selective effects of environment, and how much to the spontaneous actions of exceptionally gifted individuals who, at one point of Kede history, may have shown their people a new way of life’ (Nadel 1940: 195).Would it be a mistake to see, in these rich chapters, the influence of Malinowski’s emphasis on a rounded view of people, their cultures and change? Or is it just because of the better documentation of the historical situations of the state systems? The essay by Evans-Pritchard on the Nuer political system has been recounted and restudied so often that I shall not say more about it, but Fortes’ conclusions regarding the structure of Tallensi society parallel the case of the Nuer: Every region of Tale society, from the joint family to the whole vaguely delimited aggregate known as the Tallensi, exhibits a dynamic ­equilibrium – of like units balanced against one another, of counterpoised ties and cleavages, of complementary institutions and ideological notions … The principal mechanism by means of which this equilibrium is maintained is the balanced distribution of authority and prerogative,

22  Herbert S. Lewis

on the one hand, and of obligations and responsibilities – economic, jural, moral and ritual – on the other. Through this mechanism the component elements of any segment of the society control one another. (Fortes and Evans-Pritchard 1940: 271) This is the basic paradigm for the stateless or ‘segmentary’ society and for decades it obsessed British anthropologists – some building on the notion, others tearing it down.

The Aftermath Working with the Structural-Functional Model (More or Less) Radcliffe-Brown, Fortes and Evans-Pritchard and their students would attempt to reproduce, modify and improve upon the models of the segmentary lineage system presented in APS. The first great comparative effort, a parallel to APS, was African Systems of Kinship and Marriage (1950), edited by C. ­Daryll Forde and Radcliffe-Brown, which featured a major 85-page introduction by ­Radcliffe-Brown expounding and expanding his ideas of the structures and functions of kinship and marriage. However, the volume did not further the segmentary lineage model very much. Six of the nine chapters were written by the same people as APS (Evans-Pritchard, Fortes, Gluckman, Nadel, Richards and Schapera), but five of the essays discussed state societies and the three newcomers were not devoted Radcliffe-Brownians (Forde, Hilda Kuper and Monica Wilson). Even Fortes dealt with a state (that of the Ashanti) and the word ‘­segmentary’ ­appears only six times in the 400-page book. The study of segmentary lineage systems was pursued by a number of scholars, including the ethnographic studies of the Tiv of Nigeria by Paul Bohannan (1957) and I.M. Lewis on the politics of the Somali lineage system (1961). The collection of ethnographies, Tribes without Rulers, edited by John Middleton and David Tait, was a central work. The authors of these essays were explicitly following up and attempting modifications of the segmentary lineage system model: ‘We decided to make a more refined terminology and classification, and also in an introduction to attempt to consider some of the basic assumptions and findings of the research done in Africa since 1940’ (1958: vi). After the preface by ­Evans-Pritchard and a long introduction by the editors, there follow ethnographic essays by Laura Bohannan on the Tiv, Jean Buxton on the Mandari, ­Godfrey Lienhardt on the Western Dinka, Edward Winter on the Bwamba, ­David Tait (posthumously) on the Konkomba and John Middleton on the Lugbara. Meyer Fortes wrote much more on the subject, above all his two volumes on Tallensi kinship and clanship and his generalizing article, ‘The Structure of Unilineal Descent Groups’ (1953).7 Evans-Pritchard amplified his discussion of the Nuer as well (e.g. Evans-Pritchard 1940a, 1951).

African Political Systems and Political Anthropology  23

To follow up the problem of power and exploitation in African states, John Beattie published a supportive work, ‘Checks on the Abuse of Political Power in Some African States’ (1959), and Audrey Richards published ‘African Kings and Their Royal Relatives’ (1961). This subject seemed to be an obsession within the tight little ship of British social anthropology from the 1940s through to the 1960s, with many variations, but with questions about the basic paradigm.

Critical Reactions There were numerous critiques, of course. For example, Peter Worsley published a very critical ‘revaluation’ of Fortes’ accounts of Tallensi kinship and clanship by introducing the economic dimension, change and ‘the world economy’ (1956: 72). Michael G. Smith (1956) produced a long essay reappraising the notion and progress of segmentary lineage theory and Aidan Southall contributed modifications of both classification and process with his Alur Society (1956). The Alur, he contended, had elements of both a segmentary lineage system as well as that of a state – albeit an incipient and perhaps fragile one – and thus was born the concept of the ‘segmentary state’. He also highlighted the factors of ancestor worship and ritual as bases for political action and domination, and was one of the first anthropologists to introduce the ideas of Max Weber into his work.8 It is often said that the very small world of British social anthropology of the 1940s to the 1960s was run as a tight ship, with stern controls by the chaired professors, Radcliffe-Brown and then Evans-Pritchard at Oxford, Meyer Fortes at Cambridge, Isaac Schapera and Raymond Firth at the LSE, and Max Gluckman at Manchester (see Boissevain 1974; Vincent 1990; Barth 2005: 43). But if ­‘Oxbridge’ anthropology during this period had a powerful orthodoxy, centred on the matter of descent, lineages and order, it seems as though the LSE, at least, was the source of a different, if not oppositional, perspective, as were Max Gluckman and his followers at the Rhodes-Livingstone Institute and then at Manchester University. Even without considering the work of the non-Oxbridge anthropologists, an investigation of the works produced during this period (and even in the 1930s) indicates more diversity, if not insurrection, than one might be led to expect from the tight ship stereotype. In addition to the modifications made to ideas from APS, there were a number of works that featured nonstate systems that were not based on segmentary lineages. These include: Kenneth L. Little’s The Mende of Sierra Leone: A West African People in Transition (1951); Monica Hunter Wilson’s Good Company: A Study of Nyakyusa Age-Villages (1951); Paula Brown’s ‘Patterns of Authority in West Africa’ (1951); Jack Goody’s ‘Fields of Social Control among the LoDagaba’ (1957); C. Daryll Forde’s ‘The Governmental Roles of Associations among the Yakö ’ (1961); and, much later, Philip H. Gulliver’s Neighbours and Networks: The Idiom of Kinship in Social Action among the Ndendeuli of Tanzania (1971). In fact, in 1940, Evans-Pritchard himself published a

24  Herbert S. Lewis

paper about a group of stateless peoples that were not organized through segmentary lineages, ‘The Political Structure of the Nandi-Speaking Peoples of Kenya’. It does not seem to have been noticed.9 Two senior members of the establishment wrote their own general books about political anthropology. As Vincent says: ‘It is, indeed, tempting to see [Isaac] Schapera’s Government and Politics in Tribal Societies (1956) and [Lucy] Mair’s Primitive Government (1962), an equally controlled regional comparison of indigenous political organization in eastern Africa, as LSE alternatives to the Oxbridge model’ (1990: 259). Max Gluckman went in another direction, confronting the fact that there is always conflict in African societies, as in any society. Despite his awareness of, and sensitivity to, tensions, competition and domination, he would yet try to save the structural-functionalist notions of homeostasis, equilibrium and the fundamental idea that government and politics serve the function of maintaining order and the system (Gluckman 1956, 1963, 1965, 1968; cf. Leach 1954). Building on the idea that ‘[i]n the event of rebellion, the aim, and result, is only to change the personnel of office and never to abolish it, or to substitute for it some new form of government’ and that feuding is frequently endemic in stateless societies, the Manchester leader argued in numerous places and ways that conflict within a system may, paradoxically, have the effect of reproducing the social order and even strengthening it. It is not clear how much he was influenced by the Georg Simmel’s Conflict Theory. (For a critical view, see Worsley (1961).) Gluckman is credited with originating another significant method, or perspective, bringing Europeans, administrators, settlers and whoever else into the same frame for analysis. His paper, ‘Analysis of a Social Situation in Modern Zululand’ (Gluckman 1940b) is considered a classic for two reasons. First, methodologically, Gluckman described in detail the events he witnessed in a single day and used these as the basis for generalization about a complex plural society. This approach would be known as ‘situational analysis’ or the ‘extended case method’ and was often used by his students. Second, this work presented a pointed critique of segregation in South Africa. The students and colleagues of Max Gluckman, from both the Rhodes-­ Livingstone Institute (1941–47) and the University of Manchester (1947 until his death in 1975), carried out pioneering work in their studies of modern political situations mostly (but not only) in Africa. His followers produced a more dynamic, varied and complex portrayal of contemporary politics than was typical of the Oxbridge anthropologists. To begin with, Gluckman’s followers looked at contemporary situations and modern problems rather than attempting to work with the social structures of ‘tribes’. They were amongst the earliest anthropologists to write in terms of ethnic groups and ethnicity in Africa. Their modus operandi was to include all the personnel and interest groups, urban and rural,

African Political Systems and Political Anthropology  25

white and black, unions and administrators, etc. involved in the social, economic and political situations they studied. In keeping with their guru’s major concerns, they tended to focus on situations of conflict and conflict resolution, and on the economy and modern political interests.10 Among these were Gluckman’s colleague and successor as Director of the Rhodes-Livingstone Institute, Elizabeth Colson, as well as Victor W. Turner, J.A. Barnes, Ian Cunnison, A.L. Epstein, William Watson, J. van Velsen and Fred J. Bailey. Two others, J. Clyde Mitchell and Peter Worsley, would call themselves sociologists. The work of these researchers was indicative of the increasing concern with change and the growth of urban life, migrant labour, the development of labour unions, ethnicity in urban areas, etc.11 It blended smoothly into what I would call ‘the new political anthropology’, joining others who would get there from different starting points. The founder of the Rhodes-Livingstone Institute in Northern Rhodesia in 1938 was Godfrey Wilson, another LSE student of Malinowski’s. He went to the field in Tanganyika in 1934 to study acculturation, the problem Malinowski was then concerned with. He and his wife, Monica Hunter Wilson, studied the effects of employment in mining, urbanization and ‘detribalization’ (see The Analysis of Social Change (1945)) and led a group that carried out research in Northern Rhodesia, Southern Rhodesia and Tanganyika. Godfrey Wilson’s pro-native, leftist and pacifist views got him into trouble with the mining companies and he resigned from the directorship in 1942, but he set the tone that Gluckman, and then Colson and the other researchers, would follow (see Brown 1973; Schumaker 2001). Lyn Schumaker, who has carried out research into the history and ethnography of the Rhodes-Livingstone Institute, points out the varied backgrounds of the researchers and writes, that ‘the Rhodes-Livingstone Institute’s creative mix does not fit the retrospective image of social anthropology at its high point as a discipline dominated by the tribal unit of analysis, with little scope for discussing politics or the agency of the individual’ (2001: 114). The output of studies and ideas from these researchers through several decades cannot be summarized in a brief paper, but three works are worth mentioning here, each for a different reason. Mitchell’s The Political System of the Yao of Southern Nyasaland (1949) presents a picture of a people, once stateless, now under the control of a colonial power structure in which white administrators direct an African civil service that must deal with Yao chiefs, village headmen and ‘villagers in kin groups’. It is worth noting that this work of British social anthropology takes ample note of history – it could not work otherwise – and deals with the realities of white rule.12 A.L. Epstein’s Politics in an Urban African Community (1958) is a study of developing political institutions, class and other distinctions in a ‘Copperbelt’ city of Northern Rhodesia (Zambia). Epstein deals with various activities and

26  Herbert S. Lewis

organizations (unions, councils, welfare societies, municipal and mine administrations, courts, the African National Congress as well as ‘tribal’ concerns). In Schism and Continuity in an African Society: A Study of Ndembu Village Life (1957), Turner builds on Gluckman’s idea of ‘rituals of rebellion’. Gluckman writes: ‘He concludes, beyond the point I reached, that ritual “is not so much a buttress or auxiliary of secular social regularities as a means of restating, time and again, a group unity which transcends, but to some extent rests on and proceeds out of, the mobility and conflicts of its component elements”’ (Gluckman 1963: 48). Once again, the Fortes model of lineage of territory and segmentation is found to be inappropriate for an African society and Turner, using his concept of ‘social drama’, stresses individual and collective interests and motivations in the face of community values and norms: ‘The conflict is between norm and impulse in each individual member of the group’ (Turner 1957: 126). The Rhodes-Livingstone Institute endured despite opposition from the Colonial Office, mine operators and white settlers (Brown 1973). It dealt with sensitive matters, with all manner of conflict, and with social and economic problems such as rural and urban poverty, the consequences of labour migration and African grievances, protests and developing political parties. Gluckman himself was expelled from Northern Rhodesia on political grounds. The Manchester ‘products’ would play an important part in the establishment of the new political anthropology.

The American Scene Part I Lewis Henry Morgan, the pre-Boasian amateur pioneer of American anthropology, devoted two-thirds of his influential book Ancient Society to ‘Growth of the Idea of Government’, focusing on the Iroquois Indians, whom he knew so well, and some other American Indian groups, as well as the Aztec, Greek and Roman political systems. But when American anthropology was founded as a discipline at the start of the twentieth century, there was no particular interest in political systems per se. It was not until the resurrection of cultural evolutionary theory in the 1950s and 1960s that there was significant focus on political systems, and this involved a concern, like Morgan’s, with the evolution of political organizations. Appropriately, the neo-evolutionists were concerned with classifying types and conjecturing on the growth and succession of political forms. Works that stood out in this vein were Morton Fried’s The Evolution of Political Society (1967) (adopting Morgan’s term) and Elman Service’s writings on the succession from ‘band to tribe to chiefdom to state’ (1962). For the most part, this work did not involve intensive fieldwork with a focus on living systems of politics of the sort that had developed in British anthropology. Notable exceptions to the impulse to typologize in this era was the ethnographic study by Douglas Oliver of the political dynamics of the Siuai of

African Political Systems and Political Anthropology  27

­ ougainville, based on intensive fieldwork (1955). Marshall Sahlins, on the other B hand, worked with ethnographic literature and evolutionary typologies to try to extrapolate process. One of his many publications involves speculation about the dynamics of segmentary lineages (1961) and another compares processes of leadership by Melanesian Big-Men with those of Polynesian chiefs (1963). He attempts to understand how several different systems of political organization, leadership and control might operate. Other members of the cohort of neo-evolutionists, who approached their subject from materialist and somewhat Marxian foundations, shifted their concern from ‘traditional’ organization to modern class systems, contemporaneous with the Rhodes-Livingstone researchers. Sidney Mintz’s Worker in the Cane: A Puerto Rican Life History (1960) is a classic miniature of this genre, while Eric Wolf worked on a much greater canvas with his Peasant Wars of the Twentieth Century (1969).

The Newer Political Anthropology in Britain New approaches to political phenomena began to develop in the late 1950s and 1960s in the United States, the United Kingdom and Norway. Although this period coincided with the beginning of the dissolution of the British and French Empires, and the growth of outspoken and well-publicized anticolonialism, it is an ahistorical and polemical mistake to credit the change in political anthropology primarily to these ‘political’ factors. There were also theoretical/analytical reasons. To use Thomas Kuhn’s notion loosely, there was increasing awareness of the weaknesses in the predominant structural-functionalist p ­aradigm as applied to political phenomena. There was widespread dissatisfaction with ­Radcliffe-Brownian structural-functionalism for its typological obsessions and its Durkheimian insistence on synchronic and ahistorical analysis that rendered the approach limited in its ability to deal with change and process. The assumption that social systems are like organisms in homeostatic equilibrium and that the function of political systems is to maintain order and keep social systems stable and persistent was questioned. And we have seen that a number of anthropologists disagreed with Radcliffe-Brown and Fortes’ claims of the centrality, almost inevitability, of lineage and descent as the basis for the organization of acephalous societies. Isaac Schapera, a student of Malinowski, wrote: In studying political organization we therefore have to study more than merely ‘the maintenance or establishment of social order’. We have to study, in fact, the whole system of communal leadership and all the functions (as well as the powers) of the leaders; and in this context such activities as the organization of religious ceremonies or collective hunts,

28  Herbert S. Lewis

or the concentration and redistribution of wealth, are as relevant as the administration of justice and similarly significant for comparative purposes. (Schapera 1956: 219) The view of the political in APS was seen to be too restrictive, leaving out not only leadership of activities but also slighting such obvious elements of political life as the pursuit of power and competition. We might say that there was a ‘return of the repressed’ at the beginning of the new political anthropology in Britain, that is, a newly found influence of the students of Malinowski. Raymond Firth begins his introduction to the edited work Man and Culture: An Evaluation of the Work of Bronislaw Malinowski as follows: ‘This book has been written because some of us have thought for a long while that too little attention has been paid to the work of Bronislaw Malinowski’ (Firth 1957: 1). One key to the new approach comes from Malinowski’s concern with individuals and individual action. In Man and Culture, Edmund Leach wrote: ‘Malinowski’s biggest guns are always directed against notions that might be held to imply that, in the last analysis, the individual is not a personality on his own possessing the capacity for free choice based in reason’ (Leach 1957: 127).13 Firth, Malinowski’s first Ph.D. student, and Leach, one of the last students to know Malinowski, were two who stated basic premises for a new approach to political anthropology.14 In a 1954 article, Firth laid down several major critiques of Radcliffe-­ Brownian ‘structure’, leading to a major statement of a very different Weltanschauung. At the centre of his argument, Firth emphasizes the importance of individual choice in the creation of social/cultural formations. He argues for the primacy of process (‘organization’ in his lexicon) over form (structure) and he ­describes ‘social organization’: as the working arrangements of society. It is the processes of ordering of action and of relations in reference to given social ends, in terms of adjustments resulting from the exercise of choices by members of the society. This is not the same as describing social organization as the working rules of the society, which implies conformity, an imperative, in the ordering of the activities of the members of the society which may be only partly true. People often do what rules lay down, but these rules alone are an incomplete account of their organized activities. (Firth 1964: 45; originally 1954). There are two more relevant passages. The first highlights individual choice: ‘What the concept of social organization does is to focus on those aspects of dynamics or process in which choice is exercised in a field of available alternatives, resources are mobilized and decisions are taken in the light of probable social

African Political Systems and Political Anthropology  29

costs and benefits’ (Firth 1964: 17; originally 1962) and ‘In actual life individuals are continually faced by choices between alternatives for action’ (Firth 1954: vi). Not coincidentally, Raymond Firth came to London from New Zealand with a doctorate in economics, and the notion of scarcity and the need for choice is fundamental to that discipline. His dissertation was entitled ‘The Primitive Economics of the New Zealand Maori’ (1927). The second passage highlights social process: ‘The working arrangements by which a society is kept in being, the ways in which relations between groups are made operative and become effective, rest upon individual choice and decision. Here is our great problem as anthropologists – to translate the acts of individuals into the regularities of social process’ (Firth 1964: 46, emphasis added).15 Additionally: ‘The anthropological contribution to dynamic theory lies in major part in the analysis of structures of interests and of power, of meanings and of values, not only at their ostensible “realistic” level, but also as they express symbolically ideas of group identity and activity’ (ibid.: 29). In 1954, Leach, the former student of Firth and Schapera, published a book that was considered revolutionary at the time: Political Systems of Highland Burma. It is a complex work, making a number of unorthodox claims about ritual, culture, kinship and more, and illustrated with very detailed ethnographic and historical data. His general approach was significant for political anthropology. To begin with, Leach vigorously denied the notion of social systems as ‘naturally existing real entities’ and rejected the organic analogy and the idea that social systems could be in equilibrium. Instead, his book demonstrated processes whereby human action altered sociopolitical systems in historical time – ‘I seek to demonstrate a basic mechanism in social change’ (Leach 1954: 17) – and in order to motivate his model of action and change: ‘I consider it necessary and justifiable to assume that a conscious or unconscious wish to gain power is a very general motive in human affairs. Accordingly, I assume that individuals faced with a choice of action will commonly use such choice as to gain power, that is to say they will seek recognition as social persons who have power; or, to use a different language, they will seek to gain access to office or the esteem of their fellows which may lead them to office’ (ibid.: 10).16 Leach’s introduction of the power motive was new in social anthropology and while it broadened the definition of ‘the political’, it also raised considerable objections. One of the milder ones can be found in the foreword by Firth, who wrote: ‘And yet one feels there is some speciousness in such a monolithic explanation. For the operation of social affairs in Polynesian communities to seem explicable, allowance must be made empirically for notions of loyalty and ­obligation which cut across the narrow confines of group power interests. And in other ethnographic fields it would seem that valuations of a moral and religious order enter and jostle the power and status-seeking elements’ (Firth 1964: viii).17

30  Herbert S. Lewis

Leach’s Pul Eliya, A Village in Ceylon: A Study of Land Tenure and Kinship is another ethnographically dense book that continued his assaults on Durkheimian-Radcliffe-Brownian thought with this conclusion: The social structure which I talk about in this book is, in principle, a statistical notion … It is a by-product of the sum of many individual human actions, of which the participants are neither wholly conscious nor wholly unaware.18 When I started writing this book … I wanted to understand the principles of structural continuity in this small-scale community which lacked any obvious type of exclusive on-going corporation. There were no unilineal descent groups, no secret societies, no sects; what then were the continuing sets of relations which kept the society in being? It was only gradually that I came to realize that this whole formulation was altogether too much in the tradition of Radcliffe-Brown and the Oxford structuralists. Why should I be looking for some social entity other than the individuals of the community itself? (Leach 1961: 300, emphasis added) On the heels of Political Systems of Highland Burma came a book by a student of Leach about patterns of political leadership among the Yusufzai Pathans (Pushtuns) of the Swat Valley of Pakistan. Fredrik Barth drew explicitly upon Max Weber’s approach to political legitimacy and on Bernard de Jouvenal: who sees political activity in a means-to-end framework as directed towards rallying supporters for desired purposes. In such a framework, allegiance is regarded not as something which is given to groups, but as something which is bartered between individuals against a return in other advantages. The system of authority and the alignment of persons in groups is thus in a sense built by the leaders through a systematic series of exchanges. (Barth 1959: 2) Political Leadership among Swat Pathans (1959), a more accessible and shorter book than either of Leach’s ethnographies, stresses the choices of individual actors made in ‘frameworks’ that ‘serve to define and restrict the alternatives which are offered to each actor’ (Barth 1959: 3). The central point of Barth’s presentation of Pathan political life is that groups are formed as the result of the actions and activities of ambitious leaders, both secular and religious, and the free choices of individual men to follow those leaders who promise the best rewards. On the other hand, those choices are, of course, constrained by ‘a matrix of values and social relations’ (Barth 1966: 21) as well as the material resources available to the various actors.19 Barth expanded on ‘the analytical importance of transaction’, which:

African Political Systems and Political Anthropology  31

by helping us to isolate a basic social process … depicts the strategic limitations imposed on persons who engage in social activity with a view to obtaining something of value; simultaneously it shows the compounded effects which multiple independent actors, each seeking to pursue the transactionally optimal course of behaviour, have on each other, and thereby the gross frequentative patterns of behaviour which will tend to emerge in such situations. (Ibid.: 11) Later he wrote of ‘the relationship between codification and praxis, or how forms of understanding arise from experience, and reciprocally how behaviour and experience are predicated by collective representations’ (Barth 1981: 12). A clear pattern from Firth to Barth emerges: ‘structure and agency’ avant la lettre. Lucy Mair stated this developing perspective regarding the individual’s pursuit of ‘power, the ability to control the action of others’ (Vincent 1978: 179) clearly and succinctly: Whatever kind of society we are looking at, we see people facing alternative courses of action and choosing which they will follow. They may be choosing between equally legitimate alternatives; they may decide to break a rule or neglect an obligation and take the consequences, or hope to evade them. They make the choice in accordance with their calculation of relative advantages – one advantage being always that approval of one’s neighbors which is gained by conforming to the rules that are generally accepted. (Quoted in ibid.: 179) In 1969, Fred G. Bailey, a student of Gluckman at Manchester who worked in India, published the first in a number theoretical and generalizing books, Stratagems and Spoils: A Social Anthropology of Politics. ‘To make a beginning, think of politics as a competitive game’, a metaphor he maintains as he writes of the ways and the machinations of ‘political man’, individuals who work hard to maximize their self-interest in political situations. He generalizes about the range of activities, opportunities, tactics, problems and burdens of political ‘entrepreneurs’ in a work that draws its examples from cases in India and England, as well as Barth’s Swat Pathans.20 With Firth, Mair, Leach, Barth and Bailey, we are dealing with a very different ‘political anthropology’ from the one we started with in 1940: 1. The very definition of ‘the political’ has changed from the political as the maintenance of order in a society to politics as search for advantage, as competition between individuals and groups, as the arts of domination, and politics as ‘who gets what, when, how’ as the American political scientist Harold Lasswell, put it (1936).

32  Herbert S. Lewis

2. The central ideas of Durkheimian and Radcliffe-Brownian structural-­ functionalism were abandoned. There was no longer a search for an integrated system, analogous to an organism as an equilibrium system. The adoption of this perspective represents (even if unconsciously) the belated victory of Durkheim’s lesser-known rival Gabriel Tarde, who rejected the notion of society as an organism and made the human actor, making choices, the centre of his theories. For Tarde, the key processes were innovation and invention by individuals, followed by their imitation by others – in other words, the translation of the acts of individuals into cultural and social forms.21 3. Politics was now conceived of as operating in many different spheres, not just tribes or ‘corporate groups’. Politics took place in villages, tribes, castes, bureaucracies, in and between unions, ethnic groups, political parties, colonial and mine administrations – almost anywhere. 4. Central to this approach was the assumption that the acts of individuals exercising choice were always made in the context of values, social relations and resource constraints.22 Among other names, this approach would come to be called ‘action theory’ and the ‘actor-oriented approach’.

The American Scene Part II There were parallel developments in political anthropology across the Atlantic, arising from independent origins. One stream came from Harvard University in the early 1930s, where a group of graduate students were participants in a continuing seminar influenced by the thought of Vilfredo Pareto and by developments in the emerging field of industrial sociology. The senior members of the seminar included Elton Mayo, a founder of industrial sociology, and the dynamic young anthropologist W. Lloyd Warner, today a largely forgotten anthropologist who pioneered studies of American social systems. Among the students were Conrad M. Arensberg, William Foote Whyte, George C. Homans and Eliot D. Chapple. They were particularly impressed by a contemporary industrial relations study, ‘The Bank Wiring Observation Room’, which demonstrated the importance of the informal grouping of the workers in a workroom regardless of the official structure of the system. The project meticulously studied the activities and the interactions among the members of the group. They discovered much about their relationships and the values that developed among them over time. Their attitudes belied the expectations of the expert consultants (Homans 1950). Arensberg had been a researcher in the Warner-led ‘Yankee City’ study of Newburyport, Massachusetts, and he carried out fieldwork in Ireland in the early 1930s (see his The Irish Countryman (1937)). William Foote Whyte did his research in Boston, producing the classic Street Corner Society: The Social Structure of an Italian Slum (1943), and Homans became a leading sociological theorist stressing the importance of small groups for the study of human behaviour

African Political Systems and Political Anthropology  33

(see The Human Group (1950). Chapple, originally working with Arensberg, was the co-author of Principles of Anthropology (with Carleton Coon), a unique work – rarely cited – that stressed the importance of interpersonal relations and demonstrated the kinds of interactions typical of different activity systems and institutions (Chapple and Coon 1942). The work of this group was part of the development of ‘interaction theory’, stressing the importance of following individuals in action with others. Douglas Oliver, whose studies of leadership among the Siuai people of Bougainville was mentioned above, came from this Harvard milieu in the 1930s and Roger Keesing, a student of Oliver’s at Harvard, would later contribute a suggestive paper about ‘decision-making models’ (Keesing 1967). Another source of the new political anthropology came from Yale, where Malinowski had taught for several years before his death in 1942. One of the students he impressed was Ward Goodenough, who wrote: ‘My theoretical approach has been that human behavior is largely aimed at accomplishing purposes, taking care of wants and needs, not only in regard to physical survival but also to social relationships and, perhaps most important of all, to emotional well-being’ (quoted in Borofsky 1994: 274–75). It does not require a great leap of imagination to see some influence of Malinowski and his type of functionalism in this. Goodenough’s dissertation, which was published as Property, Kin, and Community on Truk (1951), begins with a concern for property and then works ­outwards – not unlike Leach in Pul Eliya. Like the Harvard group, he stressed the importance of looking at ‘activity systems’. His paper ‘Residence Rules’ (Goodenough 1956) was a central document in the development of an actor-oriented and processual view in the United States. Goodenough and J.L. Fischer each collected data on kinship in a community on the island of Truk within a three-year period: ‘It was quite a shock, therefore, when I recently found myself differing considerably with John Fischer about the incidence of residence forms’ (Goodenough 1956: 22). Goodenough’s data indicated that the people of the community were ‘essentially matrilocal’, but on the basis of Fischer’s data, he would have to call them ‘bilocal’. Rather than argue which of them was right, Goodenough realized that the people they were dealing with were not choosing postmarital residence according to the anthropologists’ categories, but were doing so on the basis of their own calculations of personal interest, their circumstances at the time, and numerous values, ‘rules’, their interpersonal relations and so on. Thus, Goodenough pointed us away from the attempt to fill the categories found in Notes and Queries and towards the search for what people actually do, the decisions they make in the light of their personal, cultural and social conditions. His discussion of the bases upon which residence decisions are made is very similar to Raymond Firth’s criticism of Radcliffe-Brown’s famous essay on ‘the mother’s brother’ (Firth 1964: 50–53).

34  Herbert S. Lewis

Following Goodenough, Alan Howard, who carried out fieldwork in Rotuma, a dependency of Fiji, continued the critique of the search for more exact ‘models’ of kin groups (such as ‘cognatic descent groups’) that was prevalent at that time: ‘Instead of beginning with a definition and trying to fit indigenous (ideal) group concepts into it, we can proceed empirically and derive our analytical concepts on the basis of the principles that determine group membership, as Goodenough has suggested’ (Howard 1963: 431–32). And enlarging his view beyond residence rules, Howard claimed that the same approach can be used for all activity systems; not merely observing the decisions and choices that the members of a community make, but trying to derive from these the principles, the cognitive bases, for the decisions: ‘It boils down to what one must know to predict behavior accurately’ (ibid.: 435). With his emphasis on activity systems and how they operate he was also echoing – perhaps unwittingly – the ‘principles of anthropology’ of Chapple and Arensberg. In 1967, Roger Keesing, who carried out extensive fieldwork on Bougainville (following his Harvard mentor, Douglas Oliver), published a significant article building on both Goodenough and Howard. ‘Statistical Models and Decision Models of Social Structure: A Kwaio Case’ proposed a ‘decision model’ ‘to denote an ethnographic description that is actor-oriented and based on the categories of the culture under study, i.e. “emic”’ (Keesing 1967: 2, emphasis added). The concept of ‘emic’ had been put forward by Goodenough to denote the perspective of the members of a culture, in contrast to ‘etic’, the view of the outside observer. Keesing’s method: ‘(1) defines the situation or context in a culturally meaningful way; (2) defines the range of culturally acceptable courses of action in the situation; and provides either (3) a set of rules for making appropriate decisions under culturally possible combinations of circumstance, or (4) a set of strategies for deciding among alternatives, i.e., a value-maximization model’ ­(Keesing 1967: 2). Like Howard, he would ‘focus on action groups, not merely jural categories like “descent groups”’. The substance of the paper in which Keesing demonstrated his method involved a model he had constructed of the various bases for Kwaio marriage contributions. Howard and Keesing were thinking in terms of choices, decisions made by members of a society, based on both cultural ideas and values and the individual’s own estimation of his or her interests and capabilities. During the same years, a number of other American anthropologists were reinserting the economists’ ­notion of scarcity and choice into economic anthropology. After a decade in which a ‘substantive’ approach to noncapitalist economics seemed to carry the day, insisting that ‘maximization of value’ is appropriate only to capitalist and ‘market economies’ (for example, Polanyi et al. 1957; Sahlins 1972), several papers appeared to refute this idea. Robbins Burling (1962) and Frank Cancian (1966) made the case for all humans needing to make choices between ends, regardless of what those ends might be – whether material, marital, spiritual,

African Political Systems and Political Anthropology  35

political, ­sexual, or for social status. Each individual would be limited or empowered by the means available to achieve those ends as well as directed by the values, rules and sociocultural conditions prevailing at the time. These notions worked well with transaction theory, exchange theory and interaction theory, and with Fredrik Barth’s view of values and action in ‘Models of Social Organization’ (1966). Barth wrote of values as ‘people’s principles and scales of evaluation’: I hold that these values are empirical facts which may be discovered – they are not an analyst’s constructs, but views held by the actors ­themselves. They are views about significance, worthwhileness, preferences in/for things and actions. I have represented them as being initial to items and sequences of behaviour – they are the criteria by reference to which alternative actions are evaluated, and on the basis of which choice is ­exercised. (Ibid.: 12) Clearly, the same analytical world existed on both sides of the Atlantic at this point.23 The volume Political Anthropology (Swartz et al. 1966) gave the field a name, and the introduction provided a basic statement of the anthropological study of political organization that was very different from APS. Although two of its editors, Marc J. Swartz and Arthur Tuden, were Americans, the third, Victor Turner, a student of Gluckman, was one of a substantial number of British anthropologists who found academic homes in the United States. Political Anthropology was followed two years later by Local-Level Politics, edited by Marc Swartz (Swartz 1967). The two books form a pair – each consisting of extensive introductions followed by individual essays. While most of the authors in Political Anthropology were American, about half of those in Local-Level Politics were British and several of them studied in Manchester with Gluckman. The introduction to Political Anthropology can be seen as a ‘capstone’ as much as a pioneering statement, declaring more explicitly the results of a quarter century of concern about politics since APS. The very image and definition of political had changed – ‘away from the earlier preoccupation with taxonomy, structure and function of political systems to a growing concern with the study of political processes’ (Swartz et al. 1966: 1). ‘The study of politics, then, is the study of the processes involved in determining and implementing public goals and in the differential achievement and use of power by the members of the group concerned with those goals’ (Swartz et al. 1966: 7). In the words of Jeremy Boissevain, there was ‘a new orientation that focusses on people, instead of on institutions and groups; on how individuals try to construct a social reality, instead of on how they are moulded by society; on dynamic configurations, instead of on static forms, on

36  Herbert S. Lewis

the ­essential unity of individual and society, instead of on their polarity’ (Boissevain 1974: 211).24 In her review essay ‘Political Anthropology: Manipulative Strategies’, Joan Vincent wrote of ‘action theory … associated with a range of theoretical frameworks, among them those built around transactions, symbolic interaction, systems analysis, methodological individualism, game theory, interaction theory, and political clientelism’ (1978: 175; emphasis in original). The following quote details her inventory of concepts developed by the ‘earlier formative phase’ of action theory: Some concerned the political forms generated out of the coalescence of individual actors: among these were quasi-group, action-set, clique, gang, faction, coalition, interest group, and party. Others related to modes of political behavior: choosing, maximizing, decision-making, strategizing, interacting, transacting, manipulating, career-building, spiraling, recruiting, excluding, maneuvering, competing, fighting, dominating, encapsulating. Still others related to the context (both spatial and temporal) of political action: chief among these were event, situation, arena, field, political system, environment, and power structure. (Ibid.: 176) Vincent states: ‘If the contribution of African Political Systems to the study of politics was the delineation, in 1940, of non-State political structures, the contribution of action theory since 1960 has been to delineate forms of competitive organization’ (1978: 184). She goes on to review works dealing with political leadership, patronage, factionalism and power-brokers. But the author ends her review on a prophetic note: ‘Political situations and encounters that have long characterized this approach within political anthropology are now meshed with a concern with emergent relations of domination and exploitation within a modern world system’ (ibid.: 190, emphasis added). She states, ‘Today action theory relates most closely to dialectical theory and the general sociology of Marx and Weber.’ Vincent was signalling a major shift that had taken place in American anthropology.

The 1960s and After By early 1970s I had become politicized by my students [at the University of California-Santa Cruz], and my interests shifted from cognition and social structure to more global and political interests, including a belated self-reeducation in Marxism and social theory. In the last twenty years, I have been examining questions I mainly ignored in my early research: class, gender, power, ‘development’ and

African Political Systems and Political Anthropology  37

­ ependency, colonial discourse, cultural nationalism. I draw theoretical d guidance from Marx, and more recently from Gramsci, Foucault, Bourdieu, Hall, Said, Guha, and a range of feminist theorists (notably Rowbotham, Mitchell, Ehrenreich, English, Irigaray, and Wittig). (Keesing 1994: 311) These words of Roger Keesing, a skilled ethnographer and creative contributor to the new political anthropology, perfectly capture what happened to A ­ merican anthropology after the late 1960s. Above all, it was Vietnam, the disastrous American war, its consequences on the home front and especially its impact on students, that had the greatest impact. But there was much more happening in America in the 1960s. There was the excitement of the turbulent and ubiquitous civil rights movements, and the dismay at the violence and assassinations that accompanied the reaction to both the nonviolent demonstrations and the more radical Black Panther movement. It was the time of the growth of the women’s movement and other organized assertions of identity, such as that of the Chicano (Mexican-American) movement ‘La Raza’ and the American Indian Movement (AIM). The gay and lesbian movements made their influence felt somewhat later. And we must not forget ‘the greening of America’, the so-called ‘youth movement’, the sexual revolution, the infatuation with the musical revolution: ‘sex, drugs and rock and roll’. Those were amazing times and the impact on students – and many of their professors – was profound: ‘Everything that was part of the existing order was questioned and criticized’ (Ortner 1984: 138).25 And the developing nongroup, decision-making and actor-oriented approaches – part of an ongoing quest for a ‘scientific’ understanding of culture, society and history – was wiped out. The first wave of critique was set in Marxian terms, which seemed an obvious choice to those furious at America, ‘the West’, capitalism and colonialism. Marxist thinking is hostile to the ‘methodological individualism’ that is said to be at the base of the actor-oriented approach. ‘Pure’ Marxian thinkers insist on materialist foundations and prefer to focus on class and on whole (oppressive) systems, and reject the view that individuals and their choices are significant for the study of politics and history. A prime example of this is Talal Asad’s critique of Barth’s employment of a ‘market model’ in his Pathan study, and Barth’s ignoring the primacy of ownership of land and the power of landowners (Asad 1972; see also Ortner 1984). Joan Vincent remarks, perhaps a bit triumphantly, but accurately, that Asad’s re-evaluation ‘marked the decline of action theory in its several guises’ (Vincent 1990: 360). Alongside Marxian approaches, but not really congruent with them, came Andre Gunder Frank’s ‘dependency theory’ and Immanuel Wallerstein’s ‘World Systems’ approach, each of which would attract fervent devotees for a while. However, the enthusiasm for the Marxian model would soon wane, to be ­replaced

38  Herbert S. Lewis

by a far more sweeping critique of the idea of positivism, of the pretension of science and of everything in ‘Western’ culture. Under the impact of such intellectual movements as feminism, postmodernism, postcolonialism, poststructuralism, and the writings of Michel Foucault, Edward Said and others mentioned by Keesing above, American anthropology was transformed. From a field that was conceived of as a positivistic and ‘objective’ search for understanding of human behaviour, culture and society, it was converted into a field many of whose leading members denied the possibility of objectivity, scoffed at or reacted in fury at the notions of ‘truth’ and ‘science’, and devoted themselves to study situations of domination and exploitation. This is the major obsession of today’s American anthropology and apparently of much of world anthropology as well. This major shift in orientation and character was not due to anomalies in the theoretical approaches of the time, but was, in the most profound way, affected by political factors in the world at large and especially in the United States.26 In the United States today, rather than an attempt at an ‘objective science of culture’, there is, in large part, a politicized field whose subject is domination, victimization and the study of misery. In contemporary critical thinking, there is only the struggle for domination: By the mid-80s, critical anthropology had become mainstream. The goal of mainline anthropology was to critique hidden and open oppressions of Western bourgeois culture: its racism, sexism, nationalism, homophobia, and scientism. The Enlightenment – the historical center of liberal ideas – came to be seen as a well of poison. According to the agenda, the task of the ethnographer was to examine the resistance of non-Western and peripheralized peoples to the Western modernizing forces that oppress them. (D’Andrade 2000: 223) In her famous 1984 paper, Sherry Ortner took ‘practice [theory] as the key symbol of eighties anthropology’ (Ortner 1984: 158) and practice theory should have had considerable kinship with the sorts of approaches that were blooming in the mid-1960s. Although the newer practice theory shared a strong sense of the shaping power of culture/structure with the political anthropology of the 1960s, this shaping power is viewed darkly, as a matter of ‘constraint’, ‘hegemony’ and ‘symbolic domination’ (ibid.: 147). Ortner herself proclaimed that ‘practice theory’ is only appropriate for cases of domination. And this sums up the ethos of post-1960s American anthropology. The field has become thoroughly politicized and lives on borrowed theories from thinkers like Gramsci, Foucault and Said, who had never set eyes on the peoples anthropologists regularly live among and try to understand. According to the compiler of the entry for ‘Political Anthropology’ for the recent Oxford Bibliographies Online, these are the headings for the sections under

African Political Systems and Political Anthropology  39

the larger rubric of Contemporary Trends: Colonialism and Postcoloniality, Subalterns and Social Movements, Critical Development Studies, Anthropology of the State, Socialism and Postsocialism, Neoliberalism, Diaspora and Transnationalism, Human Rights and Humanitarianism, and Secularism and Religion.27 These are, of course, all worthy topics, but there is a frequently repeated cliché in the United States: ‘If your only tool is a hammer, everything looks like a nail.’ I fear for a discipline with only one tool. Herbert S. Lewis is Professor Emeritus of Anthropology at the University of Wisconsin-Madison. He received his BA from Brandeis University in 1955 and his PhD. from Columbia University in 1963. His research interests include the history of anthropology, political anthropology, ethnicity, culture change, Ethiopia, Israel and American Indians. He is the author of A Galla Monarchy: Jimma Abba Jifar, Ethiopia (University of Wisconsin Press, 1965); After the Eagles Landed: The Yemenites of Israel (Westview Press, 1989); Oneida Lives: Long-Lost Voices of the Wisconsin Oneidas (University of Nebraska Press, 2005); and In Defense of Anthropology: An Investigation of the Critique of Anthropology (Transaction Publishers, 2014). He has published numerous papers, including ‘The Misrepresentation of Anthropology and Its Consequences’, American Anthropologist 100: 716–31, 1998; ‘The Passion of Franz Boas’, American Anthropologist 103, 447–67, 2001; ‘The Radical Transformation of Anthropology…’ in Regna Darnell and Fred Gleach (eds), Histories of Anthropology Annual, vol. 5, 2009; ‘Boas, Darwin, Science, and Anthropology’, Current Anthropology 42, 381–406, 2001.

Notes  1. Günther Wagner (Ph.D., University of Hamburg) and Kalervo Oberg (Ph.D., University of Chicago) were not students of Malinowski, but Radcliffe-Brown taught at Chicago during Oberg’s years studying there.  2. Joan Vincent, with her Marxist sensibility, writes that the Radcliffe-Brown programme was ‘in essence a conservative nineteenth-century program, and derivative at that’ – from Maine’s Ancient Law and Durkheim’s Division of Labor in Society (Vincent 1990: 255–56).  3. One interesting point in the preface that I have never seen mentioned is Radcliffe-Brown’s emphatic and unexpected statement against reification: ‘There is no such thing as the power of the State; there are only, in reality, powers of individuals – kings, prime ministers, magistrates, policemen, party bosses [he had recently come from Chicago!] and voters.’ (1940: xxiii).  4. It is rarely noticed that they also mention a third type of political system – what some American anthropologists would later call band societies, ‘very small societies … in which even the largest political unit embraces a group of people all of whom are united to one another by ties of kinship, so that political relations are coterminous with kinship relations and the political structure and kinship organization are completely fused’ (Fortes and Evans-Pritchard 1940: 6–7). The comparative literature on such societies makes it clear that this is a great oversimplification and misses much that might be called ­political in these small societies, such as leadership and the organization of activity. For a fine

40  Herbert S. Lewis example, see Lévi-Strauss’ uncharacteristic early article ‘The Social and Psychological Aspects of Chieftainship in a Primitive Tribe: The Nambikuara of Northwestern Mato Grosso’ (Lévi-Strauss 1977 [1944]).   5. The third essay about a stateless society – that by Guenther Wagner on ‘the Bantu of Kavirondo’ – is not based on the same model.   6. It was reprinted as Chapter 6 in Gluckman’s Order and Rebellion in Tribal Africa (1963: 171–77).   7. Fortes’ student Jack Goody would have much to say on this (some of it critical) in his studies of the related LoWiili (1956) and the LoDagaba (1957).   8. In the same year, Lloyd A. Fallers employed Weber in his study Bantu Bureaucracy (1956). I believe that my 1963 dissertation was the next work on African political systems to apply Weberian ideas to the study of an African state (see Lewis 1965).   9. I described similar patterns in ‘Neighbors, Friends, and Kinsmen: Principles of Social Organization among the Cushitic-Speaking Peoples of Ethiopia’ (Lewis 1974a). 10. Gluckman’s Introduction to his collection Order and Rebellion in Tribal Africa (1963) presents a long account of the development of his ideas and experience in political anthropology, from South African beginnings (including the influence of Durkheim and Radcliffe-Brown), through APS and his work in Zululand, to the development of his ideas of conflict and equilibrium. 11. J.A. Barnes, a member of the Gluckman group, published a paper that belongs in a discussion of responses to the segmentary lineage system model (Barnes 1962). He writes of the inappropriateness of the African-based model of the segmentary lineage for the peoples in the New Guinea Highlands, where postwar conditions had made it possible for anthropologists to have access to many new societies. 12. This is hardly uncommon and should be a reminder to beware of stereotypes, even in relation to anthropology and anthropologists. It is a mystery to me how Johannes ­Fabian’s myth-making book about anthropology and history managed to gain such acceptance. 13. Edmund Leach contended that Malinowski found the appropriate body of theory for his own predilections and preferences in William James’ pragmatism: ‘The heart of the matter is that James was deeply suspicious of any abstraction that could not immediately be referred to directly observable facts; so was Malinowski’ (Leach 1957: 122). ‘It may fairly be said of Malinowski, as it has been said of James, that “he was an individualist, interested in the experiences, perplexities, and satisfactions of individual souls, and anything claiming to be more-than-individual he distrusted from the depths of his soul” [Gallie 1952: 29]’ (ibid.: 127). Lewis (2001, 2015) compares William James and Franz Boas in similar terms. 14. Another important Malinowski student was the brilliant Siegfried F. Nadel, who frequently weighed in on theoretical issues, but whose death in 1956 at the age of fifty-three robbed him of the importance he would undoubtedly have had. He wrote: ‘it seems impossible to speak of social structure in the singular. Analysis in terms of structure is incapable of presenting whole societies … There are always cleavages, dissociations, enclaves, so that any description alleged to present a single structure will in fact present only a fragmentary or one-sided picture’ (Nadel 1957: 153). 15. The American sociologist George C. Homans (1967: 106) wrote: ‘The central problem of social science remains the one posed, in his own language and in his own era, by Hobbes: How does the behavior of individuals create the characteristics of groups?’ 16. Leach had anticipated this approach in an earlier work, Social and Economic Organization of the Rowanduz Kurds: ‘It is true that the structural pattern of a society does impose

African Political Systems and Political Anthropology  41

17.

18. 19. 20.

21. 22. 23.

24. 25.

26. 27.

upon individuals some standardization of behaviour, but this structural pattern affects the i­nterests of different individuals in widely different ways and their resulting reactions differ accordingly. There can never be absolute conformity to the cultural norm … The mechanism of culture change is to be found in the reaction of individuals to their differential economic and political interests’ (Leach 1940: 62). This was the position of Franz Boas in America, though it was not necessarily understood or followed by all of his many students (Lewis 2015). Leach wrote: ‘English social anthropologists have tended to borrow their primary concepts from Durkheim rather than from either Pareto or Max Weber. Consequently, they are strongly prejudiced in favour of societies which show symptoms of “functional integration”, “social solidarity”, “cultural uniformity”, “structural equilibrium” … Systems which display symptoms of faction and internal conflict leading to rapid change are on the other hand suspected of “anomie” and pathological decay’ (Leach 1954: 7). Leach credits the Italian sociologist Vilfredo Pareto with alerting him to the significance of the tendency for people to seek power. Leach had previously written: ‘Custom “makes sense” not in terms of some external, logically ordered moral system, but in terms of the private self-interest of the average man in that particular cultural situation’ (Leach 1961: 298). The matter of relative resources will be central to Talal Asad’s extended critique of Barth’s work; Asad argues the significance of class, which Barth had left out (Asad 1972). Jeremy Boissevain, who studied at the LSE with Lucy Mair and Raymond Firth, was another contributor to the development of actor-oriented (or ‘nongroup’) studies in British anthropology. He published ethnographies of politics and religion, factions and parties in Malta (1964) and patronage in Sicily (1966), and made theoretical contributions to the development of thinking about ‘quasi-groups’ and networks, etc. (1968). Tarde’s best-known book, Les Lois d’Imitation (1890), was translated into English in 1903 by Franz Boas’ colleague Elsie Clews Parsons. Following Bruno Latour’s (2005) discovery of Tarde, he may become better known in the world at large. This approach may be called ‘methodological individualism’ and, whether from Hobbes or not, is normally considered conservative, even reactionary, by Marxists. It is not irrelevant that Barth spent the period 1961–62 at Columbia University with Conrad Arensberg. In his major theoretical statement, ‘Models of Social Organization’, Barth cites Arensberg (1957) with respect to the ‘feedback effect’ between actions on the one hand, and the values, social arrangements and culture patterns that determine the constraints and the incentives available for action on the other hand. These actions, in turn, affect the values and social arrangements themselves (Barth 1966: 15). Boissevain offers a brief history and a ‘sociological’ interpretation of the processes of this intellectual shift. Sylvia Yanagisako, who was younger than Roger Keesing, represents the sort of student who opened his eyes: ‘During my undergraduate and early graduate career at the ­University of Washington in the sixties, my intellectual perspective was shaped less by my professors than by the national and international dialogue generated by the anti-Vietnam war movement, the youth cultural revolt, and the Black civil rights and Black power movements from which the former two drew a good deal of their inspiration, political analysis, and tactics’ (Yanagisako 1994: 200). Anthropology in Germany seems to have gone through a similarly traumatic period (Braukämper 2002). The collection compiled by Nugent and Vincent (2004) emphasizes works in the same vein, as, to a lesser degree, does Vincent (2002).

42  Herbert S. Lewis

References Arensberg, Conrad M. 1937. The Irish Countryman. London: Macmillan. ________ . 1957. ‘Anthropology as History’. in K. Polanyi, C.M. Arensberg and H.W. ­Pearson (eds), Trade and Market in the Early Empires. Glencoe, IL: The Free Press, pp. 97–113. Asad, Talal. 1972. ‘Market Model: Class Structure and Consent: A Reconsideration of Swat Political Organization’, Man, 7: 74–94. Bailey, Fred G. 1969. Stratagems and Spoils: A Social Anthropology of Politics. New York: Schocken Books. Barnes, John A. 1962. ‘African Models in the New Guinea Highlands’, Man, 62: 5–9. Barth, Fredrik. 1959. Political Leadership among Swat Pathans. London: Athlone Press. ———. 1966. ‘Models of Social Organization’. Royal Anthropological Institute Occasional Paper 23. ———. 1981. Process and Form in Social Life: Selected Essays of Fredrik Barth, vol. 1. London: Routledge & Kegan Paul. ———. 2005. ‘Britain and the Commonwealth’, in Fredrik Barth, Andre Gingrich, Robert Parkin and Sydel Silverman, One Discipline, Four Ways: British, German, French, and American Anthropology. Chicago: University of Chicago Press, pp. 3–57. Beattie, J.H.M. 1959. ‘Checks on the Abuse of Political Power in Some African States: A Preliminary Framework for Analysis’, Sociologus 9: 97–115. Bohannan, Paul. 1957. Justice and Judgment among the Tiv. Oxford: Oxford University Press. Boissevain, Jeremy. 1964. ‘Factions, Parties, and Politics in a Maltese Village’, American ­Anthropologist, 66: 125–87. ———. 1966. ‘Patronage in Sicily’, Man, 1: 18–33. ———. 1968. ‘The Place of Non-groups in the Social Sciences’, Man, 3: 542–56. ———. 1974. ‘Towards a Sociology of Social Anthropology’, Theory and Society, 1: 211–30. Borofsky, Robert (ed). 1994. Assessing Cultural Anthropology. New York: McGraw-Hill, Braukämper, Ulrich. 2002. ‘Trauma einer Ethnologen-Generation? Die Tagung der Deutschen Gesellschaft für Völkerkunde in Göttingen 1969’, Zeitschrift für Ethnologie, 127: 301–19. Brown, Paula. 1951. ‘Patterns of Authority in West Africa’, Africa, 21(4): 261–78. Brown, Robert. 1973. ‘Anthropology and Colonial Rule: Godfrey Wilson and the Rhodes-­ Livingstone Institute, Northern Rhodesia’, in Talal Asad (ed.), Anthropology and the Colonial Encounter. London: Ithaca Press, pp. 173–197. Burling, Robbins. 1962. ‘Maximization Theories and the Study of Economic Anthropology’, American Anthropologist, 64: 802–21. Cancian, Frank. 1966. ‘Maximization as Norm, Strategy and Theory: A Comment on Programmatic Statements in Economic Anthropology’, American Anthropologist, 68: 465–70. Chapple, Eliot D., and Carleton S. Coon. 1942. Principles of Anthropology. New York: Henry Holt. D’Andrade, Roy. 2000. ‘The Sad Story of Anthropology 1950–99’, Cross-Cultural Research, 34(3): 219–32. Epstein, A.L. 1958. Politics in an Urban African Community. Manchester: Manchester ­University Press. Evans-Pritchard, E.E. 1940a. ‘The Nuer of the Southern Sudan’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 272–96. ———. 1940b. The Nuer. Oxford: Clarendon Press. ———. 1940c. ‘The Political Structure of the Nandi-Speaking Peoples of Kenya’, Africa, 12: 250–67.

African Political Systems and Political Anthropology  43 ———. 1951. Kinship and Marriage among the Nuer. Oxford: Clarendon Press. Fabian, Johannes. 1983. Time and the Other: How Anthropology Makes Its Object. New York: Columbia University Press. Fallers, Lloyd A. 1956. Bantu Bureaucracy: A Century of Political Evolution among the Basoga of Uganda. Chicago: University of Chicago Press. Firth, Raymond. 1954. ‘Foreword’, in Edmund R. Leach, Political Systems of Highland Burma: A Study of Kachin Social Structure. London: Bell & Sons, pp. v–viii. ———. 1964. Essays on Social Organization and Values. London: Athlone Press. Forde, C.D. 1961. ‘The Governmental Roles of Associations among the Yakö’, Africa, 31: 309–23. Fortes, Meyer. 1940. ‘The Political System of the Tallensi of the Northern Territories of the Gold Coast’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 239–71. ———. 1945. Dynamics of Clanship among the Tallensi. London: Oxford University Press. ———. 1949. The Web of Kinship among the Tallensi. London: Oxford University Press. ———. 1953. ‘The Structure of Unilineal Descent Groups’, American Anthropologist, 55: 17–41. Fortes, Meyer, and E.E. Evans-Pritchard (eds). 1940. African Political Systems. London: Oxford University Press. Fried, Morton H. 1967. The Evolution of Political Society: An Essay in Political Anthropology. New York: Random House. Gluckman, Max. 1940a. ‘The Kingdom of the Zulu of South Africa’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 25–55. ———. 1940b. ‘Analysis of a Social Situation in Modern Zululand’, Bantu Studies 14 (1): 1–30. ———. 1955. Custom and Conflict in Africa. Oxford: Basil Blackwell. ———. 1963. Order and Rebellion in Tribal Africa. London: Cohen & West. ———. 1965. Politics, Law, and Ritual in Tribal Society. Chicago: Aldine. ———. 1968. ‘The Utility of the Equilibrium Model’, American Anthropologist, 70: 219–37. Goodenough, Ward. 1951. Property, Kin, and Community on Truk. New Haven: Yale University Publications in Anthropology, No. 46. ———. 1956. ‘Residential Rules’, Southwestern Journal of Anthropology, 12: 22–37. ———. 1994. ‘Intellectual Roots’, in Robert Borofsky (ed.), Assessing Cultural Anthropology. New York: McGraw-Hill, pp. 274–75. Goody, Jack. 1956. The Social Organization of the LoWiili. London: Oxford University Press. ———. 1957. ‘Fields of Social Control among the LoDagaba’, Journal of the Royal Anthropological Institute, 87(1): 75–104. Gulliver, Phillip. 1971. Neighbours and Networks: The Idiom of Kinship in Social Action among the Ndendeuli of Tanzania. Berkeley: University of California Press. Herskovits, Melville. 1938. Dahomey: An Ancient West African Kingdom. New York: Augustin. ———. 1941. ‘Review of Meyer Fortes and E.E. Evans-Pritchard (eds). African Political Systems’, American Anthropologist, 43: 465–68. Homans, G.C. 1950. The Human Group. New York: Harcourt Brace & World. ———. 1967. The Nature of Social Science. New York: Harcourt Brace & World. Howard, Alan. 1963. ‘Land, Activity Systems and Decision-Making Models in Rotuma’, ­Ethnology, 2(4): 407–40.

44  Herbert S. Lewis Kaberry, Phyllis. 1957. ‘Malinowski’s Contribution to Field-work Methods and the Writing of Ethnography’, in Raymond Firth (ed.), Man and Culture: An Evaluation of the Work of Bronislaw Malinowski. London: Routledge & Kegan Paul, pp. 71–91. Keesing, Roger F. 1967. ‘Statistical Models and Decision Models of Social Structure: A Kwaio Case’, Ethnology 6: 1–16. ———. 1994. ‘Intellectual Roots’, in Robert Borofsky (ed.), Assessing Cultural Anthropology. New York: McGraw-Hill, 310–12. Kuhn, Thomas. 1962. The Structure of Scientific Revolutions. Chicago: University of Chicago Press. Kuper, Adam. 2015. Anthropology and Anthropologists: The British School in the Twentieth ­Century. Abingdon: Routledge. Lasswell, Harold. 1936. Politics: Who Gets What, When, How. New York: McGraw-Hill. Latour, Bruno. 2005. Reassembling the Social: An Introduction to Actor-Network Theory. New York: Oxford University Press. Leach, E.R. 1940. Social and Economic Organization of the Rowanduz Kurds. London: Lund, Humphries. ———. 1954. Political Systems of Highland Burma: A Study of Kachin Social Structure. ­London: Bell & Sons. ———. 1957. ‘The Epistemological Background to Malinowski’s Empiricism’, in Raymond Firth (ed.), Man and Culture: An Evaluation of the Work of Bronislaw Malinowski. London: Routledge & Kegan Paul, pp. 119–38. ———. 1961. Pul Eliya, A Village in Ceylon: A Study of Land Tenure and Kinship. Cambridge: Cambridge University Press. ———. 1964. ‘Introductory Note to the 1964 Reprint,’ Political Systems of Highland Burma: A Study of Kachin Social Structure. Boston: Beacon Press, pp. ix–xv. Lévi-Strauss, Claude. 1977 [1944]. ‘The Social and Psychological Aspects of Chieftainship in a Primitive Tribe: The Nambikwara of Northwestern Mato Grosso’, Transactions of the New York Academy of Sciences, 7(1): 16–32. Lewis, Herbert S. 1965. A Galla Monarchy: Jimma Abba Jifar, 1830–32. Madison: University of Wisconsin Press. ———. 1968. ‘Typology and Process in Political Evolution’, in June Helm (ed.), Essays on the Problem of Tribe. American Ethnological Society 1967: 101–10. ———. 1974a. ‘Neighbors, Friends, and Kinsmen: Principles of Social Organization among the Cushitic-Speaking Peoples of Ethiopia’, Ethnology, 13(2): 145–57. ———. 1974b. ‘Leaders and Followers: Some Anthropological Perspectives’. Addison-Wesley Modules in Anthropology No. 50. ———. 2001. ‘Boas, Darwin, Science and Anthropology’, Current Anthropology, 42(3): 381–406. ———. 2009. ‘The Radical Transformation of Anthropology: History Seen through the Annual Meetings of the American Anthropological Association, 1955–2005’, Histories of Anthropology Annual, 5: 200–28. ———. 2015. ‘The Individual and Individuality in Franz Boas’ Anthropology and Philosophy’, in Regna Darnell et al. (eds), Franz Boas as Public Intellectual: Theory, Ethnography, Activism. Lincoln, NE: University of Nebraska Press, pp. 19–41. Lewis, I.M. 1961. A Pastoral Democracy. Oxford: Oxford University Press. Little, Kenneth. 1951. The Mende of Sierra Leone. London: Kegan Paul. Mair, Lucy. 1962. 1962. Primitive Government. Baltimore: Penguin. Malinowski, Bronislaw. 1926. Crime and Custom in Savage Society. London: Kegan Paul, Trench and Trubner.

African Political Systems and Political Anthropology  45 Marett, Robert R. 1912. Anthropology. New York: Holt. Meek, C.K. 1937. Law and Authority in a Nigerian Tribe: A Study in Indirect Rule. London: Oxford University Press. ———. 1941. ‘Review of Meyer Fortes and E.E. Evans-Pritchard (eds). African Political Systems’, Man, 22–24: 41–42. Middleton, John, and David Tait (eds). 1958. Tribes without Rulers: Studies in African Segmentary Systems. London: Routledge & Kegan Paul. Mintz, Sidney. 1960. Worker in the Cane: A Puerto Rican Life History. New Haven: Yale ­University Press. Mitchell, J. Clyde. 1949. ‘The Political System of the Yao of Southern Nyasaland’, African Studies, 8(3): 141–59. Morgan, Lewis Henry. 1878. Ancient Society. New York: Henry Holt. Nadel, S.F. 1940. ‘The Kede: A Riverain State in Northern Nigeria’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 165–96. ———. 1957. The Theory of Social Structure. London: Cohen & West. Nugent, David, and Joan Vincent (eds). 2004. A Companion to the Anthropology of Politics. Oxford: Blackwell. Oliver, Douglas. 1955. A Solomon Island Society: Kinship and Leadership among the Siuai of Bougainville. Cambridge, MA: Harvard University Press. Ortner, Sherry. 1984. ‘Theory in Anthropology since the Sixties’, Comparative Studies in Society and History, 26: 126–66. Polanyi, Karl, C.M. Arensberg and H.W. Pearson (eds). 1957. Trade and Market in the Early Empires. Glencoe, IL: The Free Press. Radcliffe-Brown, A.R. 1922. The Andaman Islanders. Cambridge: Cambridge University Press. ———. 1931. ‘The Social Organization of Australian Tribes’. ‘Oceania’ Monographs, No. 1. ———. 1940. ‘Preface’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. xi–xxiii. ———. 1950. ‘Introduction’, in A.R. Radcliffe-Brown and C.D. Forde (eds), African Systems of Kinship and Marriage. Oxford: Oxford University Press, pp. 1–85. Radcliffe-Brown, A.R., and C.D. Forde (eds). 1950. African Systems of Kinship and Marriage. Oxford: Oxford University Press. Rattray, R.S. 1929. Ashanti Law and Constitution. London: Oxford University Press. Richards, Audrey I. 1940. ‘The Political System of the Bemba Tribe – Northeastern Rhodesia’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 83–120. ———. 1957. ‘The Concept of Culture in Malinowski’s Work’, in Raymond Firth (ed.), Man and Culture: An Evaluation of the Work of Bronislaw Malinowski. London: Routledge & Kegan Paul, pp. 15–32. ———. 1961. ‘African Kings and Their Royal Relatives’, Journal of the Royal Anthropological Institute, 91: 144. Sahlins, Marshall. 1961. ‘The Segmentary Lineage: An Organization of Predatory Expansion’, American Anthropologist, 63: 322–45. ———. 1963. ‘Poor Man, Rich Man, Big Man, Chief ’, Comparative Studies in Society and History, 5: 285–303. ———. 1965. ‘On the Sociology of Primitive Exchange’, in The Relevance of Models for Social Anthropology. London: Tavistock, pp. 139–236. ———. 1972. Stone Age Economics. Chicago: Aldine-Atherton.

46  Herbert S. Lewis Schapera, Isaac. 1940. ‘The Political Organization of the Ngwato of Bechuanaland Protectorate’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 56–82. ———. 1956. Government and Politics in Tribal Societies. London: Watts. Schumaker, Lyn. 2001. Africanizing Anthropology: Fieldwork, Networks, and the Making of Cultural Knowledge in Central Africa. Durham, NC: Duke University Press. Service, Elman. 1962. Primitive Social Organization. New York: Random House. Smith, M.G. 1956. ‘On Segmentary Lineage Systems’, Journal of the Royal Anthropological Institute, 86: 39–81. Southall, Aidan. 1956. Alur Society: A Study in Processes and Types of Domination. Cambridge: W. Heffer. Swartz, Marc J. 1967. Local-Level Politics: Social and Cultural Perspectives. Chicago: Aldine. Swartz, Marc J., Victor W. Turner and Arthur Tuden (eds). 1966. Political Anthropology. ­Chicago: Aldine. Tarde, Gabriel. 1903. The Laws of Imitation, translated from French by Elsie Clews Parsons. New York: Henry Holt. Turner, Victor W. 1957. Schism and Continuity in an African Society: A Study of Ndembu Village Life. Manchester: Manchester University Press. Vincent, Joan. 1978. ‘Political Anthropology: Manipulative Strategies’, Annual Review of Anthropology, 7: 175–94. ———. 1990. Anthropology and Politics: Visions, Traditions, and Trends. Tucson: University of Arizona Press. ———. (ed.). 2002. The Anthropology of Politics: A Reader in Ethnography, Theory and Critique. Oxford: Blackwell. Wagner, Gunther. 1940. ‘The Political Organization of the Bantu of Kavirondo’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 197–238. Whyte, William Foote. 1943. Street Corner Society: The Social Structure of an Italian Slum. Chicago: University of Chicago Press. Wilson, Godfrey, and Monica Wilson. 1945. The Analysis of Social Change, Based on Observations in Central Africa. Cambridge: Cambridge University Press. Wilson, Monica Hunter. 1951. Good Company: A Study of Nyakyusa Age-Villages. London: Oxford University Press. Wolf, Eric. 1969. Peasant Wars of the Twentieth Century. New York: Harper & Row. Worsley, Peter. 1956. ‘The Kinship System of the Tallensi: A Revaluation’, Journal of the Royal Anthropological Institute, 86: 37–75. ———. 1961. ‘The Analysis of Rebellion and Revolution in Modern British Social Anthropology’, Science and Society, 21: 26–37. Yanagisako, Sylvia. 1994. ‘Intellectual Roots’, in Robert Borofsky (ed.), Assessing Cultural Anthropology. New York: McGraw-Hill, pp. 200–1.

Chapter 3

Complementary Segmentary Opposition, Early Kingship and the Looming State Bridging the Dichotomy of ­ African Political Systems Simon Simonse The Dichotomy of Acephalous Societies and Centralized Societies In their introduction to African Political Systems, the two editors, Meyer Fortes and E.E. Evans-Pritchard, posit a distinction between two types of precolonial political organization prevalent in Africa:1 societies that are defined by a centralized authority and administrative and judicial institutions (labelled Group A in the book); and societies that consist of a number of corporate groups of roughly equal size and power that are defined by relations of agnatic kinship and that lack a central authority and administrative and judicial institutions (labelled Group B in the book). The introduction by Meyer Fortes and Evans-Pritchard is followed by eight case studies by the eminent Africanist anthropologists of the day. Five discuss societies of Group A: the Zulu kingdom (South Africa), the chieftaincies of the Ngwato, a tribe of the Tswana ethnic group (Botswana), the Bemba chiefdoms (Zambia), the Ankole kingdom (Uganda), and the Kede state (Nupe, Nigeria). The other three chapters deal with Group B societies: the Bukusu and Maragoli, the two largest tribes of the Luhya ethnic conglomerate (Kenya), the Tallensi of Ghana, and the Nuer of South Sudan. The case studies are presented independently of one another. In line with the structural-functionalist approach, the emphasis of the studies is on demonstrating the internal coherence of the political systems, not their relationship with the other cases in the book or their insertion in one of the two categories of political systems identified by the editors. The idea that there might be structural

48  Simon Simonse Table 3.1  Types of precolonial political organization prevalent in Africa (Fortes and EvansPritchard 1940: 5–6) Group A

Group B

The sociopolitical cohesion of the society is based on sharing a single central authority

The sociopolitical cohesion of the units that make up the society is based on their opposition to one another, and on their fusion in more inclusive units whose cohesion is based on opposition with groups of the same order

The identity of the sociopolitical unit is defined by its central authority and has territorial boundaries

The identity and the boundaries of the sociopolitical units on different levels of segmentary inclusion (clans, subclans, lineages, etc.) is defined by descent from a common ancestor

Power is played out in the relationship between the central authority and its subjects

Power is played out in the rivalry between complementary social segments

The exercise of power is regulated by notions of complementary obligations and duties between rulers and subjects that include protection on the side of the ruler and payment of tribute on the side of the subjects; abuse of power by the central authority is kept in check by the possibility of rebellion

The balance of power between complementary segments tends towards an equilibrium and is kept in check by the need for joint mobilization and collaboration of minor segments in confrontations at the next higher level of social segmentation, etc.

The capacity to include ethnically heterogeneous groups is expedited by the relative ease with which new groups can attach and subordinate themselves to a central authority and by the interest the recipient authority has to extend its authority to new groups, resulting in a diversification of the population of the kingdom

The capacity to include ethnically heterogeneous groups is impeded by the necessity of a minimum of social integration and cultural assimilation before newcomers can fully participate in the relatively egalitarian decision-making of the recipient communities, resulting in the maintenance of socio-cultural homogeneity

Social cohesion underpinned by common attachment to mystical values and sacred symbols embodied in the central authority

Social cohesion underpinned by pride in one’s group identity and by the traditions and ritual powers vested in the descent groups

continuity between the two types, including the possibility of a society converting from one type of political system to the other is only raised once (Fortes and Evans-Pritchard 1940: 9–10) and then as a process brought about by external factors (conquest). Any further investigation in this direction is discouraged as futile because of a supposed lack of relevant ethnographic information (ibid.: 10). In this chapter, I will present an ethnographic case where the structural continuity between societies of Group A and Group B is clearly visible. We shall see that this

Complementary Segmentary Opposition, Early Kingship and the Looming State  49

continuity lights up when we focus our analysis on the manner in which social consensus is generated and reproduced. It was Evans-Pritchard’s genius to see that Nuer social segments did not have a reality of their own and that they only assumed corporate existence in opposition to segments of the same order of social inclusion. I will call the underlying dynamic of this manner of group formation consensual antagonism and, following Evans-Pritchard, its manifestation in the political reality of a segmentary society balanced opposition. I will argue that the operation of the institution of kingship in the monyomiji cluster in South Sudan can be understood as a balancing of power between the king and the people, driven by consensual antagonism. Second, I will demonstrate that this balancing of power between the king and the people in the monyomiji cluster had the potential of transforming itself into processes of state-formation. I will present Buganda as an example of a state that could have resulted from a transformation of the type of consensual antagonism that made the early kingdoms of the monyomiji cluster tick.

Consensual Antagonism The idea that social groups derive identity and cohesion from their being in opposition to other groups is now common knowledge. René Girard2 has discovered the full anthropological significance of this commonplace knowledge. It is not just a cognitive effect that helps us organize our social world, but an essential moment in the operation of what Girard calls ‘the scapegoat mechanism’ that channels our mimetic proneness to violence in ways that enable us to manage it and stay together in relative peace. Humans have lost the instinctual inhibitions that protect animal species against self-destructive violence. We are not just mimetic in competing for food, territory and mating partners as animals are, but also in our passions, in the very motivation as to what we want and who we want to be. To define our desires and aspirations, we copy others who serve as models. This imitation inescapably nurtures a wish to replace the model and makes human conflict unavoidable, ubiquitous, and more complex and lethal than animal conflict. Once a community suffers aggravated, self-destructive conflict, people may close ranks in blaming a single person for being the cause of their misery and may subsequently eliminate him or her. When that point has been passed, the tension relaxes and a new sense of togetherness, ‘peace’, descends on the group. In retrospect, the expelled victim often assumes a benevolent aspect because of the transformation triggered by his or her death. This is how the scapegoat mechanism allows humans to overcome crisis: an act of limited violence targeting an individual or minority stops and neutralizes an outbreak of larger-scale violence. An alternative to the scapegoating of one of the group’s members is an attack on outsiders or on an enemy. War unites not only the attacking community but also

50  Simon Simonse

the attacked community. Scapegoating carries the risk of creating deep internal divisions in the group, while in warfare, the risk of the use of violence that is disproportionate in relation to the cause of the crisis is greater. Often warrior societies achieve an effective synergy in waging war at a limited cost in terms of human victims. Consensual antagonism is a collective term including both the scapegoating of a single or minority victim and the reciprocal victimizing between enemies (Simonse 2017: 13–23). Sacrifice is the controlled enactment of the scapegoat mechanism: the victim, usually a domestic animal, standing in for the group as a whole, is killed and believed to take the causes of acrimony, illness and other evils with it in its death. This is the oldest religious practice. Girard builds on Durkheim’s idea that the sense of the sacred is the foundation of human society. The sacred consists of all those shared representations that make the members of a particular society distinct and united. Girard transforms Durkheim’s concept of the sacred from within. From a mere ‘representation’ of society, Girard’s sacred carries the traces of the collective expulsion of the ­victim. Girard thus provides a rationale for Durkheim’s observation that the ­sacred is ambivalent in nature, simultaneously auspicious and pure, and dangerous and polluting. The dangerous, inauspicious side of the experience of the sacred evokes the victim charged with the community’s ills before its elimination, while the auspicious experience evokes the purified air of peace after its expulsion. The idea, defended by Durkheim’s pupil and nephew Mauss, that sacrifice is primarily a transaction between humans and a supernatural being in which the human partner expects divine blessings in return for his offering is a late elaboration, in response to the emergence of personal deities. It leaves the question of why the offering is almost always a victim of killing unexplained. Girard also distances himself from Lévi-Strauss’ structuralism, which defines culture as an essentially cognitive construct. For Girard, culture has a practical purpose: that of keeping the human community at peace. Because of this, ­Girard’s approach may not look very different from structural-functionalism. Yet the window that Girard opens on human relations is not reductionist. In our analysis, we are invited to empathize with the actors and the drama in which they find themselves. This drama is never closed to the possibility of crisis. Structural-­ functionalism contents itself in establishing a common-sense practical ­coherence between different social levels – economy, social organization and religion – in a bounded social field at a particular point in time. Conceiving of culture as essentially a means to overcome crisis and intra-­ specific violence, Girard is comfortable with the fundamentals of Darwinism.3 He is not an evolutionist in the sense that he considers the state, for example, as an achievement qualifying its rulers and subjects as representing a ‘higher’ level of human civilization. The outcome of the 7,000-year-old state experiment

Complementary Segmentary Opposition, Early Kingship and the Looming State  51

on whose possible beginnings this article comments is still far from definitive, animated as it is by a potentially destructive dynamic that continues to escape human control (Girard 2007).

The Monyomiji Cluster The ethnographic data I will use for my demonstration come from a group of South Sudanese societies that combine institutions of complementary segmentary opposition with institutions of kingship. I have labelled them ‘the monyomiji cluster’ after their most striking institution: the monyomiji. The approximately twenty ethnic groups practising monyomiji rule inhabit the mountainous east bank of the Nile on the latitude of Juba, the South Sudanese capital.4 The monyomiji are the owners (monye) of the community (miji), the age-grade of middle-aged men that carries the responsibility for public affairs and war. Since they collectively take power, rule and retire, it is customary to call them a generation, though the members of successive generations are not necessarily biological fathers and sons, as is the case among the generation-sets of the Karimojong cluster. The monyomiji stay in power for a fixed number of years. The number of years varies between ethnic groups from as much as twenty-four years in some Lopit communities to twelve or fewer among the Lokoya, Pari and Lulubo. The moment when a new generation-set takes over is also determined by the balance of power between the ruling monyomiji and the junior age-cohorts campaigning to take over. This power balance is determined by the military and political reputation of the sitting generation, and by the numbers and combativeness of the juniors. In English, the monyomiji are commonly referred to as the ruling generation. Each community also has a king. The king is a Rainmaker as well as a political and military leader. His political interests are not necessarily the same as those of the monyomiji. The king and the monyomiji regularly find themselves at opposite ends of the political spectrum, most flagrantly in rain crises, when the monyomiji will blame the king for withholding the rain. The king will retort by insisting that the monyomiji put their house in order, stop their conflicts, settle unpaid debts and take appropriate ritual action on the taboos that have been violated in order for the rain to fall. During the period preceding their accession to power, the would-be monyomiji cultivate a lowly opinion of the achievements of the retiring generation. They will campaign for renewal and revitalization of the country. They will want to imbue neighbouring communities with respect for their power and lobby for one of their age-mates in the royal clan to be made king. A young king is likely to share the ­monyomiji’s interest in establishing a reputation that will be commemorated in stories and songs by future generations.

52  Simon Simonse

Complementary Segmentary Opposition The monyomiji of a particular community relate to monyomiji in neighbouring communities according to the dynamic of complementary segmentary opposition that Evans-Pritchard in African Political Systems described and analysed for the Nuer. In the event of war, the monyomiji of different communities combine so that armies of matching political scope and size will face one another. The difference with the Nuer is that these groupings are not mobilized on the basis of genealogical closeness, but on the basis of neighbourliness, ethnic affinity, historic precedents and political convenience. While there is an awareness of the need to maintain an overall political equilibrium, there are historical cases of deliberately or accidentally unbalanced confrontations that caused powerful polities to collapse.5 One of the measures to manage the balance of forces between neighbouring communities is the synchronization of the handing over of generational power. In the western part of the monyomiji area, it was the rule that the Pari in the north would be the first to hand over power to the junior generation, followed by Liria and the other Lokoya villages, while the Lulubo, furthest south, would be the last to respond to the wave of politico-military change (Kurimoto 1998: 29–50). Evans-Pritchard mentions the synchronization of age-sets among the Nuer. Calling the age-system ‘the most characteristic of all Nuer national institutions’ (­Evans-Pritchard 1940a: 289), he describes how smaller tribes time their initiation to make it coincide with that of the larger tribes. The synchronization of tribal initiations, according to him, serves to maintain an equilibrium between different tribes and tribal coalitions. ‘Tribes’ among the Nuer are defined as the smallest units that wage war. Like ­monyomiji-sets, coeval Nuer age-sets often adopt the same name. Complementary opposition not only defines relations between independent political communities, but also structures the relations between different sections within the same kingdom. The kingdom of Liria, for example, has nine territorial sections that are arranged in a semicircle from west to northeast along the foot6 of Oponi Mountain. As they belong to the same kingdom, sections do not fight to kill. Spears and bloodshed are taboo in intersectional fights; only sticks are allowed. If bloodshed occurs, the fight is stopped and the matter is taken to the monyomiji assembly of the kingdom or to the king. In stick-fights, the sections adjacent to those that started the fighting join their immediate neighbour, ‘to maintain the existing equilibrium’, as one Lirian informant put it. Since the application of this principle is risky for the sections located at the extreme ends of the semicircle, the three western sections concluded a ‘non-aggression pact’ to prevent Okimu, at the extreme west, from being pushed out of Liria. This troika has its own name: ‘Wurewure’. The general expectation is that stick-fights can settle conflicts between sections. When fights escalate, sections further away than the immediate neighbours of the conflict parties move in. In such cases, the fight tends to stabilize in a

Complementary Segmentary Opposition, Early Kingship and the Looming State  53

polarized confrontation between the two moieties, Orinyak and Opwalang. At this level of social segmentation, the problem can only be solved by the overall monyomiji assembly or by the king, or by the two in conjunction. If the king is called in, a tripartite division of Liria, comes into operation: Ovwara, the king’s section, which is also the location of the central ceremonial ground of the kingdom, acts as a separate division (Ovwotong, a mini-section adjacent to Ovwara, being counted as part of the central division). The tripartite division also applies in the competition between the more playful girls’ age-sets. Most forms of competition (sports, cultivation for the king, etc.) are moiety-­ based. The moieties have their own ‘Masters of Bows’ who bless the tools of warfare. Other important offices (Masters of Land, Grain, Fertility and Wind) are centralized like the office of the king, except for the extreme, northeasterly section of Ongole, which is inhabited by the remnants of the people who occupied the mountain before the present invaders. Ongole still has its Masters of Land, Grain, the Mountain and Wind, but the Rain and the rainstones – e­ mblems of royal power – were seized by the leader of the invaders. While the balancing of power that goes on between sections and moieties may look predictable, it is not automatic. There is always the possibility that things will go drastically wrong. At the time of my fieldwork, a gang of youngsters of the section of Ohwa had committed murders in other sections. The m ­ onyomiji of Ohwa were unable to correct the behaviour of these hooligans. Ohwa lost all its friends around the mountain. Its immediate neighbours warned that they were ready to chase the rogue section away from the mountain. In this critical situation, only the overall monyomiji assembly and the king could be expected to impose a solution. If they succeeded in finding such a solution, this would have been counted as a political achievement of the mediators. I left the area before the issue had been resolved.

The King as a Segment in a Field Structured by Complementary Opposition We have seen that in the tripartite division of the Lirian kingdom, the king and his section stand in opposition to the rest of the community. The king offered sanctuary to members of any section that had committed bloodshed and was ­expected to help them find a negotiated solution. Once agreement was reached, the king’s curse was believed to hit any party that would break its terms. The weather was the canvass on which the ups and downs in the antagonism between people and king could be read. In the eyes of the people, drought was a symptom of the king’s ill will. For the king, it was an indication of the level of animosity among his people, or a sign that taboos – especially those regarding the use of violence – had been violated. As the crisis aggravated, the relationship between the king and the people gradually turned sour. The king pressed the

54  Simon Simonse

monyomiji to set things right. The longer a drought lasted, the more people would join the anti-king camp. If no rain fell for a long time, the monyomiji killed the king, saying ‘He is killing us, so why should we not kill him?’, the typical justification for an act of retaliation. His dead body was left in the bush or thrown in a dry riverbed, just like that of an enemy.7 If the monyomiji decided that there was room for negotiation, they would approach the king in the same way that they would approach an enemy: a delegation of blacksmiths or women would be sent to open the negotiation process. If the king allowed himself to be mollified, he demanded signs of a change of heart of his people in the form of demonstrations of respect, gifts of cattle or labour. If pressed hard by the king, the monyomiji could decide to give him a new wife. The suspense of these rain dramas, with the king posturing as the nemesis of his people while risking being scapegoated, unified the people. The relationship between the king and the people was reversible in the same way as relations of power between opposed social segments were reversible. The kings of the societies of the monyomiji cluster were not sovereigns. They were not irreversibly superior to the other members of society, nor were they the passive victims of a sacrificial cult killed like the kings studied by Frazer (1913, Part III), who were killed after their term was over. Kings lived in a state of ‘balanced opposition’ with their people, an opposition that emphatically included the possibility of the use of lethal force. In times of prosperity, the king was the ultimate model of his followers. He was honoured with all kinds of gifts. In times of crisis, the relationship between the king and the people turned sour and often became antagonistic, the king blaming particular groups of people for using inappropriate violence or violating taboos, while the people blamed the king for his ill will. If the crisis persisted, the discontent of the people turned increasingly against the king and ultimately led to his death. In the majority of the twenty-four cases of deliberate regicide that I documented, the victims resisted being killed.8 They tried to escape or fought back. Even after he was killed, the king was believed to take revenge by leaving a curse on his killers. If the drought continued despite his death, the people offered a sacrifice to alleviate the killed king’s anger. Just as people killed their kings, kings killed people. Killing was an integral part of the balancing of power. In the early phase of colonial administration, a frequent reason for members of the royal family to be disqualified from becoming or remaining government chiefs was their homicidal record. In Kings of Disaster, I showed that regicide had a unifying effect on the regicidal community, at least in the run-up to the killing. This was the case when the Pari queen was killed (Simonse 1992: 367–70; Simonse 2017: 389–93). The victimization of the king is the structural equivalent of the killing of an adversary in a confrontation of complementary segments. In both situations, the possibility of reciprocal victimization maintains the boundary between the adversaries and

Complementary Segmentary Opposition, Early Kingship and the Looming State  55 Table 3.2  Complementary segmentary opposition (from Evans-Pritchard 1940a: 282, Diagram II) A The complementary Tribes A and B are united in their opposition to external enemies (the enemies are not represented in the diagram)

B X1 The primary tribal sections X and Y occasionally fight, thus unifying their secondary sections X1 and X2 and Y1 and Y2

The antagonism between Tribes A and B (and others not shown) leads to occasional warfare that keeps the primary sections of both (only X and Y of B X2 shown here) united. The opposition between X2 and X1 keeps the tertiary sections of X2 (not shown) united

Y1 The tertiary tribal sections of Y1 (not shown here) and of Y2 (z1 and z2) remain united because of the opposition between Y1 and Y2

Z1 The tertiary section z1 occasionally fights z2 thus maintaining the cohesion between its composing descent groups (not represented) Z2 The tertiary section z2 occasionally fights z1 thus Y2 maintaining the cohesion between itds composing descent groups (not represented)

enhances the corporateness of the antagonists. The double role of the king as antagonist and unifier of his people can be rendered by the very same diagram that Evans-Pritchard (1940a) used to clarify the operation of complementary segmental opposition (see Table 3.2 above). The applicability of the diagram used by Evans-Pritchard to the unifying dynamic of kingship confirms our presumption that the dichotomy between societies of Group A and Group B may not be so absolute after all. The logic of complementary opposition also explains why relations with other kingdoms should be handled by the king. The king’s structural position as the top-level segment that unifies all the lower-level segments of the kingdom makes him, from the point of view of his fellow kings, the only actor effectively representing the kingdom as a whole and therefore the appropriate person to declare war upon and make peace with. The diagram only models the unifying role of the king in relation to the territorial sections and moieties, and not in relation to his equally important role in bridging the antagonism between generation-sets. Compared to some of the Group A polities treated in African Political Systems, the kingdoms of the monyomiji cluster appear fragile. Its kings lacked a ­monopoly over the use of physical force and their legitimacy largely depended

56  Simon Simonse Table 3.3  The king as a segment in a political field structured by complementary segmentary opposition A King A and People B are united in their opposition to external enemies who may be organized as kingdoms or otherwise (the enemies are not represented in the diagram) The antagonism of King A and People B unites the moieties X and Y into a single people

B X1 The sections X1 and X2 are united as moiety X in their opposition to moiety Y

X2 The opposition between X2 and X1 unites the subsections of X2.

Y1 The opposition between sections Y1 and Y2 unites their subsections Z1 and Z2

Z1 The sub-sections Z1 and Z2 unite when the Y1-Y2 antagonism flares up. Y2 Z2 The antagonism of Z1 and Z2 unites the descent groups constituent of Z2 (not represented).

on their rain-charisma, while rivalry over the succession was a continuous threat to the integrity of the polity. In Kings of Disaster (Simonse 1992: 302–15), I argued that the Lotuho version of the myth of the spear and the bead could be interpreted as a reflection on this fragility. The myth tells the story of the destructive rivalry between two brother-kings and ends with a curse that bans kingship forever from their communities (ibid.: 304–05). Comparing the fortunes of these fragile kingdoms with some of their acephalous neighbours, it is far from clear which of the two should be our preference if we had to look for a safe place to stay. It is likely that societies on the Upper Nile mutated in both directions, not only from acephalous formations into centralized ones, but also the other way around. There is a third way to achieve a stable predictable, political situation: the state. Its primeval protagonist is the king.

The King’s Will to Power: Tipping the Balance Since the king’s life was permanently in the balance, kings devised strategies to maximize their chances of survival and to save their throne. Among the strategies at their disposal were the following.

Generate Wealth and Create Dependents As a general principle, the king had to make sure that the balance of power that defined the relationship with his people was in his favour. For that purpose,

Complementary Segmentary Opposition, Early Kingship and the Looming State  57

he needed supporters in the first place. Counting on clan-brothers and village age-mates was not enough. The royal clan was prone to factionalism when it came to the succession and the monyomiji had a mind of their own and could not always be trusted. The king needed men with stronger loyalties: sons, clients and war-captives. To secure these men, accumulation of wealth was a first step. Building on the compulsory nationwide cultivation day, the surplus in the king’s granaries allowed him to organize more work parties than average community members and thus accumulate even more grain, part of which was exchanged against cattle. The royal herds were likely to be among the largest in the kingdom. They enabled the king and his brothers to enter into multiple marriages and sire many sons. Last but not least, the king received an important share of the booty brought home from warfare, which included captives. Kings exploited their people’s worries about rainfall by using spells of drought to step up their demands for cattle. The earliest reports of travellers and missionaries speak of kings using the rain to blackmail their subjects (Vinco 1940: 307; Simonse 1992: 195; Simonse 2017: 210). For a young king who was popular with the monyomiji, lucky with the rain and calculating in terms of spending his wealth, there were few obstacles to becoming the wealthiest person in the kingdom, even if he had to start from scratch. The king’s wealth enabled him to attract clients. These were young men whose family was unable to raise their bridewealth. They came from within and outside the kingdom. By enabling these men to marry – before or after their ­warriorhood – a corps of elite warriors was formed that had primary allegiance to the king and inspired respect to enemies, royal rivals and monyomiji.

Create a Royal Army Monopolizing the Use of Firearms The kings were the first to acquire firearms on the Upper Nile. They were often first used against the king’s own subjects (Simonse 2017: 209, 219). In the m ­ onyomiji cluster, we only hear of separate king’s armies after the Mahdists had taken control of the army posts on the Nile. King Lomoro of Tirangore had his Awusa, King Ogwok of Padibe his Buchura, Lojele in Lokiliri his Makatub, and the Amakuta in Lafon are to have been on King Alikori’s payroll. It seems that before the Mahdist period, royal armies did not operate independently of the monyomiji. Create Alliances with Neighbouring Kings The king’s position as the social segment that unified all the others turned him into his community’s focal point for foreign affairs and trade. To conduct trade and carry out diplomatic missions, kings spent a great deal of time travelling. This is confirmed by reports of explorers and missionaries. The diary of the missionary Don Angelo Vinco, the first European to travel inland away from the Nile, confirms this. When he arrived in Gondokoro for the second time, in February 1851, he found that King Nyiggilo, whom he had befriended during his earlier

58  Simon Simonse

visit, had travelled to Loudo for business (Vinco 1940: 302). In June of the same year, Nyiggilo and Vinco travelled together to Lafon, where they were the guests of Mucharabong, the King of the Pari. During their stay at the court, they were visited by a number of kings from neighbouring communities who wanted Father Vinco to come and stay with them (1940: 313). On his return from Lafon later in June, King Legge of Liria tried to ambush Vinco – according to Vinco out of jealousy that he had not included a stop in Liria in his first trip. A month later, he paid visits to Legge, to Lado (the king of Longairo) and to Iban (the king of Loudo) in response to these invitations. During this last trip, he also met the Rainmakers of Cecere (Lulubo) and Lyeparang (Bari) away from home, another confirmation that kings spent a lot of time visiting their counterparts. Baker’s journey, twelve years later, in search of the lake that he would name ‘Albert’ after Queen Victoria’s Prince-Consort, followed an itinerary that also corresponded to a royal trade network. He travelled via Legge in Liria and Hujang in Tirangore to Kachiba in Obbo, and from there to Bunyoro on the lake. These royal alliances had a darker side. In the history of the Lotuho and Horiok, there are different examples of kings who used their allies to punish their own monyomiji: Mulak, the queen of the powerful Horiok kingdom of Segele, called on the kings of Longulu and Imatari to help her take revenge on the ­monyomiji of Segele for killing her two sons. King Ngalamitiho of Imatari mobilized not only the monyomiji but also the junior generation-set that was about to take over (Simonse, 1992: 197; Simonse 2017: 211) to destroy Segele. The military superiority meant the end of Segele. The conflict between the two legendary Lotuho princes Facar and Attulang, the protagonists of the Lotuho version of the myth of the spear and the bead, ended when Facar called on the vastly more powerful Toposa to kill his brother. In both cases the norm of proportional response underlying complementary segmentary opposition was broken by royals obsessed by hateful rivalry.

Centralize the Kingdom The Lotuho king had a representative in each village-section, the aboloni hobu. He was a member of the rain-clan and was selected by the monyomiji of his section to serve as their liaison with the king. The king normally had a wife and household in each village. The cultivation day for the king by the monyomiji would be performed in their wives’ location, sometimes complemented by a day in the fields at the king’s main residence. In its initial stage, centralization was about creating a growing number of direct links between the centre and the different villages and village-sections, thus individualizing each community’s contact with the king. This counteracted the possibility of several villages and village-sections building a coalition against the king.

Complementary Segmentary Opposition, Early Kingship and the Looming State  59

Consolidate the Kingdom’s Cohesion by Regularly Waging War on Enemies The consensual benefits of war against external enemies do not need further discussion here. Numerical and territorial expansion is more easily accommodated by the centralism of kingship than by blocks of tribes uniting to face a common enemy. The orbit of centralism is endlessly expandable. There was little therefore that stopped a successful Rainmaker from claiming credit for the rain falling over an ever widening circle of communities – except the claims of rival Rainmakers. Being based on volatile royal charisma and the vicissitudes of the rain, early kingdoms expanded and retracted at exponential speeds. Diversify the Social Composition of the Kingdom while Promoting Forms of Division of Labour Based on Complementarity In their introduction to African Political Systems, Fortes and Evans-Pritchard pointed out the greater heterogeneity of the social composition of Group A formations. They noted the ethnic diversity, the incipient social stratification and the presence of castes. All of these elements were present in the kingdoms of the monyomiji cluster. Castes corresponded to an emerging social division of labour. Their relationship to the society as a whole was different from that of territorial segments. Caste members could not establish matrimonial relations with mainstream members of the society. Their integration in society was by mutual economic interdependence, a form of organic solidarity that was at odds with the mechanical solidarity that made complementary opposition tick (Durkheim 1893). It would seem that the Nilotic kings understood that this kind of interdependence served the long-term interests of kingship. The communities of foreigners, the incipient stratification and the castes worked as ‘circuit-breakers’ in the event of an all-out opposition against the king. Transform the Royal Court into a Long-Distance Trade Hub Before the Egyptian penetration of the Upper Nile, the trade in iron objects played an important role in underpinning the power of kings. The Bekat kings of Bilinyan and Shindiru employed a considerable slave labour force in their iron mines and smithies. The establishment and consolidation of Lotuho kingship was also connected to iron working. The iron trade collapsed when cheap iron became available through the Khartoum traders in the 1840s. As an object of trade, iron was replaced by the luxury objects and guns brought from downstream the Nile. These were exchanged for ivory and slaves. It is likely that there is a relationship between the speed with which trading networks expanded during the first years of Egyptian penetration and the ambition of kings to consolidate their position in relation to their people. The new commodities represented an opportunity for kings to build and widen their networks of allies and clients, and thus strengthen their position in relation to the

60  Simon Simonse

monyomiji. When the sudd was blocked in the early 1880s, the stoppage of the supply of commodities had immediate repercussions on the loyalty to the government of the local kings, some of whom joined the anti-government rebellion in 1884. By the time the Mahdists descended on Equatoria, a multi-ethnic elite had emerged, consisting of allied and intermarrying royals and their middlemen (tarajma), who monopolized relations with the government and the traders. These trade relations were taken over by the Mahdists. By the end of the Mahdist period, many of the communities in the monyomiji cluster were part of a network of which the Lotuho king Lomoro Hujang of Tirangore was the lynchpin.

Concentrate All Ritual Powers into the King’s Hands The smaller societies of the monyomiji cluster (Lulubo, Pari and Lokoya) had a plethora of ritual offices. Each of the Lulubo clans, for example, claimed a special power that protected the community against specific threats (locusts, crop-­eatingbirds, leopards, infertility, particular diseases, etc.). Some of the Lokoya polities had up to six kings (ohobwok), each with a special cosmic domain. Compared to the small-scale kingdoms, the kings of the larger Lotuho and Bari kingdoms had concentrated most of these compartmental powers into their own hands. Among the Bari and the Lotuho, the only offices that continued to be inherited by nonroyal clans were those linked to the original occupation and use of particular stretches of land, mountains and rivers. In almost all cases, the Rainmaker was the ‘king of the kings’. The rivalry for the top office was intense. Except for the Lotuho, most kingdoms had several clans claiming power over rain. Rain-kings sought to centralize the control of rain, often by hook or by crook – by stealing the rainstones of minor rain-clans or by manipulating drought accusations.9

The Looming State The traders allied to the kings arriving from downstream on the Nile and the new underpinnings of kingly power by luxury goods and guns did not automatically translate into a firmer grip of kings over their monyomiji. During a drought in 1903, Lomoro’s international stature and his ‘spick-and-span’ army,10 for example, did not stop the monyomiji of Tirangore, from attacking the king and forcing him to quit his capital Tirangore to Loguruny to stay with his mother Queen Iloyi, who was in charge of the rain-shrine there. The Lotuho king Mayya, who had been Lomoro’s most important rival, had been killed for rain a few years earlier, followed by his wife. In the same period, Kidi, the son of Alikori, the dictatorial king of the Pari, another participant in Lomoro’s network, delivered his father to the Governor of Mongalla Province with the request to exile him

Complementary Segmentary Opposition, Early Kingship and the Looming State  61

because the monyomiji wanted to have him killed and put Kidi on the throne (Simonse 1992: 125–26; Simonse 2017: 143). None of the precolonial kings in the monyomiji cluster established sovereignty, a definitive, irreversible superiority over the people who had become their ‘subjects’. The antagonism in the relationship between the king and the people continued to be played out everywhere as an oscillating balance of power that periodically entered a critical phase when an interruption to the rainfall would give the monyomiji reasons to be suspicious about the king’s loyalty. The power of the king remained embedded in relationships that were handled on the basis of reciprocity, either positive: praise songs, gifts of women and cattle, free labour, as well as all sorts of minor gestures of gratitude (a leg of a hunted antelope, a pot of honey, a calabash filled with termites, etc.) or negative: accusations, acts of defiance, ordeals, beatings and torture, and, ultimately, if the king proved incorrigible, death by mob lynching. Victimization was part and parcel of the relationship between the king and the people, and it was bilateral, just like the feuding of complementary sections in an acephalous society. The king was not the only potential victim. In his moments of glory, he gave free rein to his capacity to victimize. This could take the form of arbitrary killings,11 punitive raids against sections of the populace,12 acts of manslaughter provoked by his anger13 and executions. The executions at the hands of kings have generally been interpreted by early travellers and anthropologists14 as instances of the administration of justice. From the fragmentary accounts at our disposal, it would appear that these executions were more like ‘reality shows’, opportunities for the king to publicly display his power, rather than demonstrations of the force of the law.15 On the other hand, it has also become clear that there were plenty of opportunities for a smart king to enhance the leverage over his people without acquiring a full monopoly of the use of physical force and for a monyomiji-set to cut a pretentious king back down to size. Having established that societies corresponding to Group A and Group B of the binary classification by African Political Systems can be understood as alternative realizations of the same underlying dynamic, we shall now examine the connection between the early kingdoms of the monyomiji cluster and African kingdoms that anthropologists never hesitated to qualify as ‘states’. In what ways are our Nilotic kingdoms prefigurations of these ‘early states’? What are the continuities and what are the discontinuities between the two systems? Do some of the Nilotic kingdoms qualify as ‘early states’? This question seems especially pertinent in the case of the Shilluk kingdom, which has been the subject of considerable anthropological debate. In the next chapter, I argue that the Shilluk kingdom, despite its high population, its historical depth, and the greater stability of its royal succession, had structurally more in common with the monyomiji kingdoms than with kingdoms that anthropologists have labelled ‘early states’.

62  Simon Simonse

Claessen, an international authority on the study of early states worldwide, counts the kingdom of Buganda,16 only 500 km south of the monyomiji area, as one of the most centralized and differentiated state formations that emerged in Sub-Saharan Africa before the colonial period (Claessen and Skalnik, 1978, 1981; Claessen, 1987). Between the first visit of the explorer Speke in 1862 and the establishment of the Uganda Protectorate in 1894, a large number of travellers, missionaries, government representatives and traders visited the kingdom, many of them leaving written accounts of their observations. By cross-checking the various accounts, we are able to obtain a fairly reliable picture of how the still-independent kingdom functioned. Of course, by selecting the Buganda kingdom, we are entering Bantu Africa. The comparison we make with Nilotic polities therefore has to be more global and culturally less subtle. Yet we should also take into account that the boundary between the Nilotic and Bantu peoples has always been porous. The kings of Buganda are a case in point, since they claim Nilotic, Lwoo, ancestry. I know of no study that attributes specific characteristics of the Buganda kingdom to this ancestry. The first eyewitness observations of the Buganda kingdom and its court come from John Hanning Speke, who in July 1862 discovered the point where Lake Victoria poured into the Nile. Before his discovery, Speke had spent four and a half months at the court of king Muteesa of Buganda (1856–84). Muteesa was still a young man ruling under the watchful eye of the Queen Mother. His coronation took place a few weeks after Speke had left the court. The Buganda king was held in fearful awe. Speke relates that when his caravan approached the royal capital and encountered a royal party, the escorts the king had sent to accompany him hid themselves in the roadside from the king’s police. They feared that their mere gaze could be taken as a provocation and as a justification for their execution (Speke 1863: 272). Mock-charges – the usual form of greeting between groups from different communities in the monyomiji cluster – were also performed by visitors to the Buganda court. But before performing the assertive charge, the visitors first prostrated themselves flat on the ground in front of the king, grovelling in subordination and ‘whining after the manner of happy dogs’ (Speke 1863: 256). Only after that did they rise, grabbing their carved sticks ‘and screamed and danced in a mimicry of hostile attack against M’Tsé [Muteesa]’ (Chaillé-Long 1876: 106). While the last part of the greeting was perfectly recognizable to visitors from the monyomiji cluster, the first part must have looked strange, and so did the grotesquely strict court etiquette. Foreign visitors to the precolonial royal courts frequently reported cases of people being sentenced to death for trivial offences (sneezing, laughing, touching the throne, exposing a piece of naked skin, peeping in the direction of the king’s wives, etc.). During the morning audiences that Speke attended, Kunsa, the chief executioner, and Ukunsu, his second-in-rank, were always present, ready

Complementary Segmentary Opposition, Early Kingship and the Looming State  63

to ­implement the king’s execution orders. In his travelogue, Speke relates nine occurrences of the king ordering people in his entourage to be executed. A number of these were multiple executions, among them six cases of women, some of them wives and one a sister of the king. Speke, who only occasionally attended the king’s morning audiences and who may only have related the more flagrant cases in his travelogue, adds that such executions took place on a daily basis. On two occasions, Speke begged the king to suspend a death sentence, which Muteesa consented to. One of these cases regarded the son of Kunsa, the chief executioner, and the other to a wife of the king who had joined the king’s picnic on the lake to which Speke had also been invited. Her offence was to offer the king a fruit that she had picked from one of the trees on the island where they had moored. Her gesture put Muteesa into a rage. When he was about to hit her on the head with a heavy stick, Speke stopped him by restraining the raised arm. The king then relented (Speke 1863: 395). James Grant, Speke’s companion who had stayed behind with the king of Karagwe to recover from an injury, joined Speke late in May 1862. Grant was lodged next to the torture chamber of the king’s ‘Chief Detective’. Screaming was heard day and night (1864: 227). Grant relates the case of an army officer who had the temerity to ask for one more slave after the king had rewarded his bravery with one. He was sentenced to being cut to pieces (ibid.: 230). When Grant one day asked the chief executioner about the well-being of his son (the one whose life had been spared thanks to Speke’s intervention), the executioner informed him that his son had been executed the day before for another offence. The lightness with which subjects were killed is also evident from the following anecdote related by Grant. One day, when the king, who was fond of hunting with the guns he had received from Speke, failed to shoot any game, ‘he shot down many people’ (ibid.: 228). This incident brings to mind the arbitrary shooting into a crowd of spectators by the Bari king mentioned earlier. For observers coming from the kingdoms of the monyomiji cluster, the murderous outbursts of the king would not have come as a complete surprise. Above I noted various cases of kings who were enraged by trivial offences. But the scale and the cold-blooded routine with which the executions were carried out at the Ganda court must have been utterly confusing to observers, as well as the fact there was no noticeable concern about any retaliatory response from the sections and descent groups whose members were the victims of these actions. Had the whole population become enslaved? European observers, including Speke and Chaillé-Long,17 dismissed the executions as mere acts of barbarity. Anthropologists did not know what to make of the extreme violence. This is how Lucy Mair, expressed her perplexity: ‘the question of precisely how the cruelties … by the last independent kings were reconciled with the conception of a “good” king expressed at his accession is one that cannot be answered’ (1934: 177–78). And Audrey Richards remarked that

64  Simon Simonse

‘many African chiefs are formally praised for their ferocity to enemies but the insistence that the Kabaka [the king] can and should destroy his own subjects is, I think, unusual’ (1964: 291). Other travellers and anthropologists tried to understand this violence as the application of some kind of penal law – as ‘punishments’. Punishments, in Africa as well as elsewhere, usually follow a logic of negative reciprocity or of revenge: ‘An eye for an eye, a tooth for a tooth.’ The jurisprudence of traditional African courts is primarily concerned with restorative justice and the resolution of conflict, not with penal law. Lashings as part of court proceedings were a colonial introduction – at least in Southern Sudan. Even there, they often merely served as a demonstration of colonial power (Simonse 1992: 136; Simonse 2017: 152). At the court of the Ganda king, most of the so-called punishments were triggered by what most would consider minor and accidental infringements of court etiquette. Yet these punishments were among the most drastic: death, mutilation or cutting into pieces. The sentences were decreed without any reference to jurisprudence. It is not possible to grasp their significance by considering them as a form of retributive justice administered by the king. To understand their meaning, we should examine them in the context of the wider role that executions played in the Buganda state. Mass executions occurred regularly. They took place at special occasions such as royal funerals. One of the largest such executions occurred at the renovation of the tomb of King Ssuuna (1832–56), the father of Muteesa in 1880. According to the missionary Mackay (1890: 185), 2,000 people were killed at this event. Mass executions were also ordered when the king or the mediums of the lubaale (official divinities) felt that disorderliness in the kingdom was on the rise. Indicators of such disorder were dirty roadsides covered with excrement, young men loitering in the capital, a rise in adultery cases (especially those involving princesses), as well as reports of a planned insurrection. In Mair’s words, they served ‘to set the land right’ (1934: 233). Executions kept everybody on their toes. They counteracted any tendency to slackness or entropy, and they reset the order of the state to its default setting. The frequency at which these mass executions took place before they were abolished was estimated by Mair as once every ten years (1934: 179) and by Wrigley (2002: 244) every five to ten years. The missionary Mackay writes that the massacres had been more frequent during the last years of King Muteesa’s reign and suggests that they were carried out to help restore the king’s health (Ray 1991: 176). His fellow missionary Felkin had a different opinion, arguing that the number of massacres was instead an indicator of the king’s good health. Once Muteesa’s health had recovered, so Felkin had been assured by his informants, the frequency of executions would increase (Wilson and Felkin 1882, vol. 2: 23). The mass executions were called kiwendo, a term that refers to the fact that the number of victims required for this type of execution was fixed in advance.

Complementary Segmentary Opposition, Early Kingship and the Looming State  65

Roscoe (1911: 333), who wrote an extensive monograph on the Baganda following instructions given by Frazer, mentions a number between 200 and 500. The number was fixed by the king, often in compliance with the oracle of a medium of one of the temples of the Ganda gods. Mediums served the king by identifying and alerting him to threats to his kingdom. Frequently these consisted of suspicions of rebellion and often the mediums were able to name suspects. To complete the required numbers, commoners were randomly captured in large numbers by the king’s executioners from the roads leading to the capital. The work of the executioners was supervised by the king’s police. When the quota was full, the king’s police chief would sound the drums to stop the arrests. There were thirteen mass execution sites in the kingdom. Some sites were specific for certain categories of victims: for chiefs and dignitaries, for rebellious princes (who would be burnt or starved to death since royal blood could not be shed), and for wives and friends of the king (a category of victims who would only be executed after a delay of some days to give the king time to change his mind). Other sites catered for a mix of convicted offenders and innocent captives. If Roscoe is to be believed, the demeanour of the victims of these executions was generally cooperative: Those who have taken part in these executions bear witness [to] how seldom a victim, whether man or woman, raised his voice to protest or appeal against the treatment meted out to him. The victims went to death (so they thought) to save their country and race from some calamity and they laid down their lives without a murmur or a struggle. (Roscoe 1911: 338) Before being killed – usually by a spear or club – the victims were made to drink a potion that was believed to give the king control over the victim’s ghost. The bodies were left where they fell for wild animals or birds to prey on. Relatives did not dare bury the corpses because they had been given to the gods (ibid.: 336) or to the king (ibid.: 112). The Ganda kings measured their power in terms of their capacity to victimize subjects. When King Muteesa was shown a photograph of Queen Victoria by the missionary Felkin, he not only asked Felkin ‘how she lived, what she wore, and how many servants she had, but also whether she killed many people’ (Wilson and Felkin 1882, vol. 2: 18). Muteesa’s question confirms the suspicion that the decapitation of the thirty lubaale priests during the audience given to ChailléLong was meant to impress the Khedive in terms of demonstrating the king’s power. Buganda alternated between two contrasting conditions. During an interregnum, it was ‘[a] wild state of disorder … where anarchy reigned, people tried to rob each other, and only chiefs with a strong force were safe, even the smaller

66  Simon Simonse

chiefs being in danger from stronger chiefs, who did as they liked during the short interregnum’ (Roscoe 1911: 103). The other, opposite condition was called mirembe: ‘the king’s peace’. Together with interstate warfare in which tens of thousands of men were mobilized, kiwendo was the principal institution that maintained ‘the king’s peace’, which was believed to be permanently under threat from the forces of disorder. I believe it is only possible to make sense of these massacres if we see them as rites of consensual antagonism. On the one hand, they were a way of dealing with rivals and suspected rebels; on the other hand, they channelled any inarticulate discontent and animosity in the populace in a single direction – away from the king and towards suspected subversive and disorderly elements. The massacres corresponded to sacrifices in the sense given to the term by René Girard. They created a scene in which disorder and the threat of disorder were demonstratively expelled from the kingdom, prompting the survivors to make a fresh start. In the eyes of an observer from one of the kingdoms of the monyomiji cluster, the Ganda kings had accomplished what many of their own kings were trying so hard to achieve: sovereign control over their people. They had irreversibly tipped the balance of power to their advantage. They had obtained a monopoly of the legitimate use of physical force as well as formal immunity to any violence emanating from their subjects. Moreover, the Ganda kings had succeeded with a vengeance: the number of people victimized to maintain order and counteract disaster was immeasurably higher than those in the Nilotic kingdoms, where a single royal victim sufficed.18 Was there any trace left in Buganda of the antagonism that defined the relationship between the king and the people in the monyomiji cluster? Roscoe’s answer to this would have been in the negative. In the passage quoted above, he emphasizes the complete lack of resistance of the victims, as if they agreed to their fate in the interest of the greater good. I believe there is one trace: the fear of the sacrificers/executioners that the victims would take revenge on the king by means of a posthumous curse. This was the rationale for serving them with the potion that ensured that the power released after their death would not threaten the king’s peace (mirembe), but would be to its benefit. The concern about the effect of the victims’ deaths reveals the sacrificial nature of the killing. The forcible ingestion of a king-friendly potion mirrors the precautions taken by regicidal killers in the monyomiji cluster. After the Pari had killed their queen, her tongue was pierced by a thorn at two points so that it could not articulate any curse and her stomach, the organ believed to generate blessings and curses, was opened and treated with a fruit that would blunt any imprecation of her people (Simonse 1992: 370; Simonse 2017: 393). We are now in a position to make a more informed guess as to the meaning of the violent scenes at the Buganda court mentioned above. They mirrored in microcosm the sacrificial violence that maintained order at the level of the state.

Complementary Segmentary Opposition, Early Kingship and the Looming State  67

The royal execution orders seemed to conform to an implicit court protocol that was based on the idea that, in the interests of his kingdom, an effective king should not miss an opportunity to impress upon his subjects that in the intercourse between sovereign and subject, unidirectionality in the use of violence was the best and only way to maintain and reinforce the integrity of the state.

Laying the Foundations of the State Once the king had achieved a monopoly over the use of force and had turned his subjects into ‘victims in suspense’, the way was clear to start building the state. In this final section, I want to briefly touch upon three aspects of this building process as they appear in studies of Buganda history.

The Transformation of the Social Structure of the Kingdom from a Network of Relations Based on Reciprocity into a Comprehensive, Controllable System of Complementary Relationships In an article written in 1960, the American sociologist Alvin Gouldner proposed an important refinement to the concept of reciprocity by, in the first place, distinguishing the practice of reciprocity from the norm of reciprocity and, in the second place, contrasting reciprocity with complementarity. The concept of reciprocity, according to Gouldner, refers to the sequence of interactions identified by Marcel Mauss in his famous Essai sur le don: a gift, recognition and acceptance of the gift by its beneficiary, and the return of a counter-gift by the beneficiary. Each step in this sequence of interactions is contingent on the previous step. Complementarity, by contrast, refers to interactions between partners in a relationship that follow a preset scenario imposed by practical necessity, by tradition, law or by royal decree, and that are dependent on one another.19 The relations between husband and wife are an example. One nurses the baby, while the other goes out fishing. The nursing of the baby is not contingent on the fishing of the husband and vice versa. On the basis of the norm of reciprocity, we are able to make judgements as to whether a complementary role-relationship is fair or unfair on its partners. In the monyomiji cluster, the relationship between the king and his people is one of reciprocity. The king’s rain was a gift that is reciprocated by offerings by his people (wives, cattle, presents, privileges and communal labour). The king’s failure to give rain disrupts the sequence ruled by positive reciprocity and may result in a cycle of negatively reciprocal confrontations that are likely to be justified by the norm of reciprocity: ‘We shall kill you because you are killing us with your drought.’ The negativity is contingent on the perceived performance of the king. The sovereignty of the king means an end to the Maussian reciprocity in the conduct of affairs that are of common concern to the king and the people. The

68  Simon Simonse

actions of the king are no longer contingent on those of the people, and vice versa – as this is overwhelmingly the case in the Nilotic kingdoms where the treatment of the king by his people is contingent on the abundance and timeliness of the king’s rain and where the king’s rainmaking gift is contingent on the love shown to him by his people. In an accomplished state, the relations between the king and the people are ruled by the complementarity of two distinct sets of roles. One set of roles is reserved for the king and his immediate entourage, while the other set defines the behaviour expected from the people. The two sets match like a dovetail joint. They are imposed by the king and their observance is monitored and enforced by the king’s courtiers, including the royal executioners, and lackeys. They constitute a hierarchy, the ground rule of which is that only the king has the right to decide whether, when, how and by whom violence is used. All others have to obey.20 The complementarity of roles opens up the possibility of the organized mobilization of people independently of considerations of reciprocity. It also opens the door for exploitative relationships and for repression. As the complementarity of the king–subject relationship sinks in, the asymmetrical character of the relationship defined by the state becomes an irreversible norm and annexes increasingly larger domains of public life, often to the point that the state’s subjects revolt against this, which may prompt the relationship to be renegotiated. The aspiration to turn the reciprocity of the king and the people into a more predictable complementary role-relationship was already present among the scapegoat kings of the monyomiji cluster. The Lulubo king liked to address his people as ‘my stable’ and ‘my ants’, implying that he was the herdsman and the queen ant. But when the situation required a humbler approach, he used ‘my husband’ as a term of address, suggesting a subordinate, complementary role. Complementary interpretations already hovered over a relationship that was still largely reciprocal. The unilateral complementarity that the monopoly of the use of violence imposed on the intercourse between a sovereign and his subjects not only meant a negation of reciprocity; it also created cultural forms that theatricalized the lack of a common humanity between king and subject – for example, in the funerary arrangements and the installation ceremonies. While among the Bari, the Lotuho and the Shilluk, one or two close associates of the king were buried with the king, in Buganda all the staff in charge of the king’s personal needs – his chamberlain, cooks, firemakers, dairymen and water fetchers, including the wives of these officials – followed the king in his death. But they were not buried alive alongside their dead master as in the Nilotic kingdoms. They were killed at the inauguration of the tomb weeks later, their bodies being left to decompose in the fenced compound surrounding the tomb, like those of the victims of mass executions (Ray 1991: 166). A similar contrast is evident in the installation ceremonies. At his installation, the Nilotic king was confronted with his ultimate victimhood at the hands

Complementary Segmentary Opposition, Early Kingship and the Looming State  69

of the people. Among the Bari, a ritual was performed in which the most feared diseases were transferred to the king in a collectively recited curse. Among the Lulubo, the uncles of the new king demanded payment of damages for putting his sister’s son at the ‘centre of evil’. The Lotuho and the Lokoya started from the premise that their would-be king was a feline monster. Once they had caught him by surprise, they took a lot of sacrificial trouble to humanize and domesticate him into their ruler. Significantly, the installation of the Ganda king followed a diametrically opposed scenario. While he was equated with a leopard like his Lotuho and Lokoya counterparts, the installation rite was aimed at intensifying his feline ferocity, not at taming it. Dressed in a fresh leopard skin, he was given a ceremonial dagger to be able to kill anyone who might resist his power. When the top dignitaries counselled the new king not to be soft on his subjects, they used metaphors that likened the king to a queen termite eating the males that fertilize her since ‘commoners (bakopi) are like sorghum: Whoever judges them owns them’ (Ray 1991: 171). Later, during a nine-day induction tour of the central districts of the kingdom (okukula), the new king, who was often only an adolescent, was made to witness killings, to give orders to kill and even to kill himself (Roscoe 1911: 210–14; Ray 1991: 171–75; Wrigley 2002: 147–54).

The Transformation of the Territorial Units and the Descent Groups That Composed the Original Kingdom of Buganda into a System of Clientelism Depending on the King as Its Chief Patron Earlier I made mention of the liaison officials of the Lotuho king (aboloni hobu) who were present in each village section (amangat). I presented these officials as evidence of a deliberate strategy towards more central control by the king. The Lotuho kingdoms were the largest and most differentiated of the monyomiji kingdoms. In Buganda, similar but far more drastic centralization was already being carried out by the beginning of the nineteenth century. With the mass executions as a menacing backdrop inspiring a ‘centralizing ethos’ (Kodesh 2003: 461), the Ganda kings needed relatively little time to transfer the political and territorial power at the local level from the old-time clan and lineage heads (bataka) to royal appointees, the bakungu and batongole. The clans that still acted as powerful corporate groups in the eighteenth century (Wrigley 2002: 221) were reduced to merely ceremonial institutions lacking the capacity to mobilize their members for political ends. The little land that was left in the custody of the bataka was used for ceremonies and burials. The royal appointees were soldiers and pages of the court who had, in one way or another, distinguished themselves in the eyes of the king. Others, according to Wrigley (2002: 221–25), were originally leaders of armed gangs of young men predating on the local population and later regularizing their status by offering the king a part of their booty as ‘tribute’.21 These gangs were the origin and core

70  Simon Simonse

of the batongole chiefly class.22 The batongole were integrated as another layer of ‘king’s men’ (bakungu) during the rule of Ssuuna and Muteesa. To manage their land, the bakungu and batongole appointed clients of their own who were also bakungu and batongole, but who did not have direct access to the king. Often these second degree bakungu and batongole had their own clients to cultivate the land. By the early nineteenth century, access to land was almost exclusively through the system of clientelism whose ultimate patron was the king (Médard 2007: 224). Buganda may very well have been the most centralized state in nineteenth-century precolonial Sub-Saharan Africa. Because of its exceptional character, it has been the subject of numerous anthropological and historical studies, which are beyond the scope of this chapter to discuss here. All I want to demonstrate is that this extremely centralized kingdom can be understood as the outcome of the evolution of an earlier relationship between the king and the people that functioned as a balance of power. In the case of Buganda, the king succeeded in completely overpowering his opposite number. Only a king who claimed absolute power could have accomplished the comprehensive top-down restructuring of the political economy and sidelined the leaders whose powers derived from constituencies that pre-existed the transformation of the kingdom into a state.

The Control of Succession Rivalry In the monyomiji cluster, rivalry over the succession was the most important threat to the stability of kingdoms, and so it was in Buganda. Eighteenth-century Buganda was torn apart by succession struggles. In the nineteenth century, this problem appears to have been solved in rather drastic ways. The royal clan was dismantled. Henceforth, the king belonged to his mother’s clan, not to that of his father, the king. The number of princes entitled to compete for the throne was thus reduced. Brotherly succession was outlawed. The precedent was set by King Ssemakookiro (ca. 1800), who simplified his succession by killing all his sons except three. Two surviving sons were left to fight it out, with the winner becoming his successor. A similar scenario was supposed to be played out between Muteesa and his brothers after he was selected king, but it was postponed several times. Speke assumed that Muteesa’s brothers, who numbered around thirty, had been burnt at the king’s coronation when he and Grant were in Bunyoro (Speke 1863: 543). It was the responsibility of the queen mother (the Nnamasole, not the biological mother of the king, but an official in her own right with powers over life and death) to arrange for the elimination of potential rivals.23 Gordon’s emissary Linant de Bellefonds reported in 1875, thirteen years later that several brothers had staged a rebellion against Muteesa and warned that this could very well bring about Muteesa’s downfall (Gray 1964: 43–44). It is only at this point that the queen mother, fearing a conspiracy between the brothers and foreign emissaries, locked all the brothers up and starved them to death (Wrigley 2002: 227).

Complementary Segmentary Opposition, Early Kingship and the Looming State  71

Thus, the view shared by Wrigley and Médard that the Ganda court was able to manage successions from one reign to the next peacefully needs considerable qualification.

Conclusions The ethnographic material from the monyomiji cluster suggests a clear structural continuity between the acephalous sociopolitical systems based on complementary segmentary opposition and the polities centred on the king–people polarity. Both systems functioned as a balance of power between antagonists. Both generated social consensus. The similarity in the cultural practices that enabled people to bridge the two kinds of antagonism (the role of women and blacksmiths as intermediaries, the role of marriage in sealing pacts, etc.) indicates that the people involved understood Groups A and B as being driven by the same force. Fortes and Evans-Pritchard were unable to see this because the structural-functional anthropology they practised concentrated its analyses on the consolidated institutions and norm and belief systems, ignoring the forces that generated and undermined these institutions. Our approach removes the mystery with which anthropologists, including Fortes and Evans-Pritchard in African Political Systems, have surrounded the practice of regicide. Regicide as was practised by the communities of the ­monyomiji cluster is the culmination of an escalating, protracted drama in which the community united itself against its king. It was the final step in a real confrontation, not a ritual or human sacrifice. People would have preferred the rain to the death of the king. Once the route to confrontation was taken, there was no way back. Regicide was ‘an inevitable, recurrent tragedy imposed on the society by its antagonistic, centralist structure’ (Simonse 1992: 373; Simonse 2017: 396; 2005). Just as the victimization of one or more adversaries was a condition for the operation of the system of complementary opposition in an acephalous society, so was the possibility of regicide a precondition for the operation of early kingship. Viewing the violence of Ganda kingship as the outcome of a centuries-old antagonism between the king and the people in which the king eventually won the upper hand may remove some of the perplexity that scholars like Mair, Richards and Wrigley felt when attempting to make sense of the mass executions and the violence at the royal court of Buganda. When examining the massacres from the point of view of the kingship experience in the monyomiji cluster, the Buganda scenario must have appeared as a world turned upside down. To the monyomiji, the spectacle of regional Ganda military supremacy upheld by people who were treated as disposable slaves by their king must have had a science-fiction quality to it. However, the same spectacle must have strengthened the belief of the kings of the monyomiji cluster, who

72  Simon Simonse

were tired of bearing the brunt of their people’s violence, in the ultimate feasibility of achieving sovereignty over their people. The Ganda kings did not hide the source of their power. Like their counterparts in the monyomiji cluster, they were totally upfront about the importance of the use of violence. Because of their openness on this point, the kings of Buganda have often been contrasted with other African rulers as being ‘secular’ (for example, Wrigley 2002: 17). The manner in which bodies of the victims of kiwendo and the bodies of palace staff killed at the funeral of the king were disposed of indeed suggests a deliberate effort on the part of the king to play down any religious significance their subjects might attach to the victims, as if the acknowledgement of their victimhood could reduce the stature of the king. This exploration suggests that it is useful to make a distinction between early kingship and the early state as distinct types of political systems. Early kingship can be defined as a form of kingship in which the balance of power between the king and the people is open-ended. This implies not only that the king lacks the monopoly on the use of physical force, but also that the people are free to confront the king and to oust or kill him. In an early state, the king has a monopoly over the use of physical force. The material presented here makes it plausible that early kingship and early statehood are successive configurations in an evolving balance (or should we say ‘imbalance’?) of power between the king and the people. Anthropological studies of the evolution of the state normally examine conditions external to the power balance between the king and the people to explain the origin of the state: the development of technology, agriculture, writing, the production of an economic surplus, conquest, etc. External correlates, such as the climate and the conducive fertility of the land, certainly played an important role in state-formation in Buganda, but they were not the driving force behind the process. With respect to the structural homology between complementary segmentary opposition and early kingship, the material presented here adds plausibility to the assumption that both formations could form an evolutionary sequence. However, the direction of the sequence is not necessarily one-way. In the kingdoms of the monyomiji cluster, both configurations of consensual antagonism existed side by side. The Lotuho myth of the spear and the bead (discussed above) seems to suggest that early kingship was fragile and prone to collapsing into segmentary fragmentation and polarization, especially during periods of rivalry over the succession. From this perspective, it would be worth investigating whether the staunch egalitarianism of the Nuer has had any centralist antecedents. I am not claiming that the trajectory from acephalous society to state outlined here is the only possible evolutionary route. In Alur Society, Southall (1953) described the emergence of segmentary states. A king with a reputation for his rain powers and for fairness in conflict resolution attracted peripheral ethnic groups and clans, which placed themselves completely under the king’s protection. In

Complementary Segmentary Opposition, Early Kingship and the Looming State  73

return for protection, they offered tributes and services, cementing their relationship with the royal core group by contracting marriages and engaging in other forms of exchange. Segmentary states had the capacity to expand very quickly, forming huge pyramidal, bottom-up structures. However, as corporate entities, they remained fragile because their kings had limited control over the use of violence, especially of the more peripheral groups. Under military or political pressure, the segmentary state would simply fall apart into its constituent segments. In contrast, the trajectory covered by the Buganda kingship went straight for the ultimate prize of statehood – sovereignty – dismantling, transforming and subordinating pre-existing descent-based sociopolitical units. In contemporary political crises, solutions are often presented as ‘radical’ when they are open to the relaxation of the constraints many modern states have put on the victimization of particular categories of citizens – such as the reintroduction of capital punishment, the abrogation of the legal limits set to the state’s power to arrest and detain, the designation of particular groups of citizens as targets of persecution, elimination or re-education and such. Ethnographic explorations of the antecedents of the state, such as the comparative exercise undertaken in this chapter, remind us that these beliefs and tendencies are not just figments of a fascist imagination or remnants of barbarism, but that they belong to the default setting of the state in all its historical manifestations, whatever the additional checks and balances that have been brought into play. Simon Simonse is a social anthropologist who trained in Leiden and Paris. He carried out field work on kingship and social cohesion in the societies of the ­monyomiji  cluster in South Sudan while lecturing in the University of Juba. His doctoral thesis ‘Kings of Disaster’ shows how models derived from Girard’s ­mimetic theory account for the dynamism that animates both hierarchical and segmentary social structures in South Sudan. He has also taught in Congo, Uganda, Indonesia, Japan, France and the Netherlands. For the last twenty-five years, he has worked as a conflict transformation expert in the Great Lakes region and the Horn of Africa.

Notes  1. Fortes and Evans-Pritchard mention a third type of society where the largest political unit coincides with a kinship group. This sociopolitical configuration by anthropologists characterized as ‘band-societies’ only prevails in isolated, small-scale groups and is not treated in African Political Systems (Fortes and Evans-Pritchard 1940: 6–7).  2. Girard’s own work is the best introduction to his anthropology: Violence and the Sacred (1977), Things Hidden since the Foundation of the World (1987, especially 59–66 on sacred kingship), The Scapegoat (1989) and Evolution and Conversion (2008, especially Chapters 2 and 3 on hominization). Girard’s contribution to Hamerton-Kelly (ed.), Violent Origins (1987: 73–145) is a good summary of the ideas that concern us in this study.

74  Simon Simonse  3. Hamerton-Kelly (1987, especially 73–145); Girard, Evolution and Conversion (2008); Antonello and Gifford (2015a and 2015b).  4. They include the Lulubo (Olu’bo), the Lokoya, the Pari, the Horiok, the Lotuho, the ­Lomiya, the Ngotira, the Dorik, the Tenet and the Ngaboli (the last five collectively known as ‘Lopit’), the Dongotono, the Logir, the Ketebo, the Lokwa, the Lorwama and the, Imotong (the last six collectively known as ‘Lango’). All are speakers of dialects of the Eastern Nilotic language that is spoken by the Lotuho, the largest group, except the Pari who like the Shilluk speak a dialect of the Western Nilotic Northern Lwoo group, the Lulubo who are Central Sudanic, and the Tenet who are Surma speakers. At the establishment of colonial rule there were more than thirty independent kingdoms, some consisting of a single agglomeration, others counting as many as fifteen large villages.  5. Examples include the downfall of Segele and Imatari (Simonse 1992: 174, 197; Simonse 2017: 191, 211).   6 Before the colonial period, the settlements were higher up the mountain.  7. For an analysis of historical cases of conflict between kings and their peoples escalating to the point of regicide, see Simonse (1992: 199–206 and 343–73); and Simonse (2017: 214–21 and 369–94).  8. Simonse (1992; 2017, Chapter 17) extensively discusses the different anthropological issues raised by the practice of regicide.  9. Simonse (1992; 2017: Chapter 14) presents a number of cases of theft and extortion of rainstones. In Liria, Hatulang, king of the immigrant settlers, stole the famous Mosidik stone as well as Mosidik’s ‘husband stone’ from the neighbouring royal court of Kamuturu, which was later named ‘Langabu’, ‘the place without king’. Later again, he refused to return rainstones he had borrowed from Madhaira, the Rainmaker of Ongole, the section of Liria inhabited by the original inhabitants who had so far been able to keep ‘their own rain’ (Simonse 1992: 294–98; Simonse 2017: 315–18). When the rain-clan of Lokiliri demanded a return of the bridewealth when a sickly bride died soon after her wedding, they insisted that they be given the old rain-clan’s rainstones (Simonse 1992: 293; S­ imonse 2017: 314). When the old Rainmaker of Logopi fell into a hot spring and got stuck in the mud, a passer-by of the rain-family of a neighbouring community ­demanded the rainstones as a reward for saving his life before pulling him out. 10. Donaldson-Smith (1900: 621). 11. King Logunu of Bilinyan insisted on trying out the gun given to him by the Egyptian expedition on the crowd that had assembled to watch the reception of the foreigners. Werne counted eleven Bari killed and many more injured (Werne 1848: 276–77). 12. In 1863, Prince Adang of Tirangore invited Baker to join in a raid on a group of ‘rebellious subjects’ (Baker, 1866, vol. 1: 240). 13. Donaldson-Smith (1900: 621) describes how he stopped King Lomoro of Tirangore from killing one of his aides when the latter brought him the wrong tusk that the King wanted to offer as a present for his American guest. 14. For a discussion of the tendency of anthropologists to present royal acts of victimization as judicial, see Wrigley (2002: 243). 15. The following passage from Werne’s description of the manner in which King Logunu speared criminals evokes the ambiance of a show: ‘He did so very quickly [goâm, goâm] without any fuss. He would be seated under a big tree with a heavy spear in his hand, to administer justice and would exhibit great anger. Maybe, people believed he was inspired by the great spirit in the tree while he was thus presiding over the court, or maybe, it was rather his own feeling of justice that put him into righteous anger and that made him into the chief executioner of the wrong-doer, although normally the latter’s fate had already

Complementary Segmentary Opposition, Early Kingship and the Looming State  75

16.

17.

18.

19.

20.

21.

been sealed by the collective will’ (Werne, 1848: 322). Werne did not know what to think of it. It is strange that a king should act as executioner, especially since Bari kings were kept away from war and the sight of blood. In Luganda, the language of the Baganda, the word ‘Buganda’ refers to the kingdom of the Baganda, which, in the nineteenth century, also included people of other ethnicities. The Swahili word ‘Uganda’ became the name of the protectorate that was established by the British and is now the name of the country. While Buganda was and is the largest political entity within Uganda, Uganda includes other kingdoms and many other ethnic groups. When referring to the ethnicity of the Baganda in an adjectival form, I use ‘Ganda’ without a prefix. Chaillé-Long stayed at Muteesa’s court from 21 June to 9 July 1874 as the envoy of Gordon, the Governor of Equatoria. During his first audience, thirty priests of the lubaale (official divinities) cult were slaughtered, while during his second audience, seven chiefs (batongole) were decapitated. Ernest Linant de Bellefonds who stayed in Muteesa’s ­capital from mid-April to mid-June 1875 to follow up on Chaillé-Long’s mission to bring Buganda within the Egyptian orbit of influence makes mention of only two executions: one of a person who interrupted a conversation between the king and his secretary, and the other of a tobacco smoker whose smoke irritated the king (Gray 1964: 43–44). There is no mention of royal executions in the account of Stanley, whose stay coincided with that of Linant, nor in Emin Pasha’s diary of his short visits in August 1876. From 1876, Muteesa professed to be a Christian and must have abolished the executions during public court sessions. However, the mass executions continued into the early 1880s. In Kingship and State, Christopher C. Wrigley, one of the main authorities on Buganda kingship, shares the following enigmatic intuition for which he does not adduce evidence: ‘the historically known kabakas [Baganda kings] were illegitimate heirs of kings who had not been despots or even rulers, but the suffering servants of their people’ (2002: 246). The known kabakas are qualified as ‘illegitimate’ because of their abuse of power when compared with ancestors who were said to have suffered at the hands of their people. Were the early Ganda kings scapegoats of their people like their counterparts in the monyomiji area? The reader should be made aware that the term ‘complementary’ is here employed in a ­different sense from its earlier use in combination with ‘segmentary opposition’. In the latter case, the use of ‘complementary’ refers to the tendency for symmetrically opposed adversaries to match the segmentary scope when mobilizing for a hostile confrontation. The use of the term in this particular sense was initiated by Evans-Pritchard and was followed by anthropologists working on segmentary societies. Complementarity, as Gouldner defines the term, refers to an asymmetrical role-relationship of which the mutual expectations have been imposed from outside or by the more powerful partner in the relationship. Maybe the king’s violent response to the spontaneous offer of a fruit by one of his favourite wives during the picnic on the lake (discussed earlier) can be understood as a manifestation of this tension between reciprocity and complementarity. The gift could not help but evoke the desire for recognition implicit in the gift. The gesture was at odds with the theatre of complementary roles played by the palace kitchen staff, whose expectations were not tainted by notions of reciprocity. It is likely that the less formal, outdoor ambiance of the boat trip and the picnic made the apparently inexperienced queen forget the rigour of the courtly code of conduct. In fact, Speke was instructed by the official called the ‘Chief-Detective’ by Grant to provision his caravan by confiscating food from neighbouring communities, either with the

76  Simon Simonse c­ooperation of their bakungu or directly. As a guest of the king, any community not feeding his group as well as anyone selling (i.e. not giving it for free) food to him was punishable by the king (Speke 1863: 343). 22. The original meaning of the word is ‘bachelor’ (Médard 2007: 295), which brings to mind the clients of the Nilotic kings who served their patrons to be rewarded with a wife. 23. Speke and Grant note how Muteesa enjoyed spending most of his free time with his brothers, having fun, making music and going out hunting. Grant mentions that some of the brothers were handcuffed and chained and led by servants (1864: 224). Was the queen mother afraid that they would try to conspire against their brother? Grant does not give an explanation, but adds that before he became king, Muteesa also went around chained. This must have been during the period when two parties of provincial chiefs were fighting over which of the two would succeed: Muteesa or his brother Kimera. The succession war was won by Muteesa’s supporters. His brother Kimera and his backers were all executed (Wrigley 2002: 227).

References Antonello, Pierpaolo, and Paul Gifford (eds). 2015a. Can We Survive Our Origins? Readings in René Girard’s Theory of Violence and the Sacred. East Lansing: Michigan State University Press. ———. 2015b. How We Became Human: Mimetic Theory and the Science of Evolutionary ­Origins. East Lansing: Michigan State University Press. Baker, Samuel W. 1866. The Albert N’Yanza: Great Basin of the Nile and Explorations of the Nile Sources, 2 vols. London: Macmillan. Chaillé-Long, C. 1876. Central Africa: Naked Truths of Naked People: An Account of the Expedition to the Lake Victoria Nyanza and the Makraka Niam Niam, West of the Bahr-el-Abiad (White Nile). London: Sampson Low, Marston, Searle & Rivington. Claessen, Henri J.M. 1987. ‘Kings, Chiefs, and Officials: The Political Organization of Dahomey and Buganda Compared’, Journal of Legal Pluralism, 25–26: 203–41. Claessen, Henri J.M., and Peter Skalník (eds). 1978. The Early State. The Hague: Mouton. ———. 1981. The Study of the State. The Hague: Mouton. Donaldson-Smith, A. 1900. ‘An Expedition between Lake Rudolf and the Nile’, Geographical Journal, 16: 600–25. Durkheim, Émile. 1893. De la division du travail social: étude sur l’organisation des sociétés supérieures. Paris: Felix Alcan. Emin Pascha. 1917–27. Die Tagebücher von Dr. Emin Pascha (edited by Franz Stuhlmann), 5 vols. Berlin, Braunschweig and Hamburg: Georg Westermann. Evans-Pritchard, E.E. 1940a. ‘The Nuer of the Southern Sudan’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 272–96. ———. 1940b. The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People. Oxford: Clarendon Press. ———. 1948. The Divine Kingship of the Shilluk of the Nilotic Sudan. Cambridge: Cambridge University Press. Fallers, Lloyd A. (ed.). 1964. The King’s Men: Leadership and Status in Buganda on the Eve of Independence. Oxford: Oxford University Press. Fortes, Meyer, and E.E. Evans-Pritchard (eds). 1940. African Political Systems. London: Oxford University Press.

Complementary Segmentary Opposition, Early Kingship and the Looming State  77 Frazer, James G. 1913. The Golden Bough: A Study in Magic and Religion, 13 vols. London: Macmillan. Girard, René. 1972. La Violence et le sacré. Paris: Grasset; translated as Violence and the Sacred. Baltimore: Johns Hopkins University Press, 1977. ———. 1978. Des choses cachées depuis la fondation du monde, Recherches avec JeanMichel Oughourlian et Guy Lefort. Paris: Grasset; translated as Things Hidden since the Foundation of the World (with Jean-Michel Oughourlian and Guy Lefort). East Lansing: Michigan State University Press, 1987. ———. 1982. Le bouc émissaire. Paris: Grasset; translated as The Scapegoat. Baltimore: Johns Hopkins University Press, 1989. ———. 2004. Les origines de la culture, Entretiens avec Pierpaolo Antonello and João Cezar De Castro Rocha. Paris: Desclée Brouwer; translated as Evolution and Conversion: Dialogues on the Origins of Culture, René Girard, Pierpaolo Antonello and João Cezar De Castro Rocha. New York: Continuum International, 2007. ———. 2007. Achever Clausewitz, Entretiens avec Benoît Chantre. Paris: Carnets Nord. Girard, René, Pierpaolo Antonello and João Cezar De Castro Rocha. 2008. Evolution and Conversion: Dialogues on the Origins of Culture. New York: Continuum International. Gouldner, Alvin W. 1960. ‘The Norm of Reciprocity: A Preliminary Statement’, American Sociological Review, 25(2): 161–78. Grant, James A. 1864. A Walk across Africa: Domestic Scenes from My Nile Journal. Edinburgh: Blackwood. Gray, John. 1964. ‘Ernest Linant de Bellefonds’, Uganda Journal, 28(1): 31–54. Hamerton-Kelly, Robert G. (ed.) 1987. Violent Origins, Walter Burkert, René Girard, and Jonathan Z. Smith on Ritual Killing and Cultural Formation. Stanford: Stanford University Press. Kodesh, Neil. 2003. ‘Review Richard Reid, Political Power in Pre-colonial Buganda’, International Journal of African Historical Studies, 36(2): 459–62. Kurimoto, Eisei. 1986. ‘The Rain and Disputes: A Case Study of the Nilotic Pari’, Bulletin of the National Museum of Ethnology, 11(1): 103–61 (in Japanese). ———. 1998. ‘Resonance of Age Systems in Southeastern Sudan’, in Eisei Kurimoto and ­Simon Simonse (eds), Conflict, Age and Power in North East Africa: Age Systems in Transition. Oxford: James Currey, pp. 29–50. Kurimoto, Eisei, and Simon Simonse (eds) 1998. Conflict, Age and Power in North East Africa: Age Systems in Transition, Oxford: James Currey. Mackay, A.M. 1890. A.M. Mackay, letters compiled by J.W. Harrison (Mackay’s sister). ­London: Hodder & Stoughton. Mair, Lucy P. 1934. An African People in the Twentieth Century. London: Routledge. Médard, Henri. 2007. Le royaume du Buganda au XIXe siècle, Mutations politiques et religieuses d’un ancien État d’Afrique de l’Est. Paris: Karthala and Nairobi: IFRA. Quigley, Declan (ed.). 2005. The Character of Kingship. Oxford: Berg. Ray, Benjamin C. 1991. Myth, Ritual and Kingship in Buganda. Oxford: Oxford University Press. Reid, Richard. 2003. Political Power in Pre-colonial Buganda: Economy, Society and Warfare in the Nineteenth Century. Athens, OH: Ohio University Press. Richards, Audrey I. 1964. ‘Authority Patterns in Traditional Buganda’, in L.A. Fallers (ed.), The King’s Men: Leadership and Status in Buganda on the Eve of Independence. Oxford: Oxford University Press, pp. 256–93. Roscoe, John. 1911. The Baganda: An Account of Their Native Customs and Beliefs. London: Macmillan.

78  Simon Simonse Simonse, Simon. 1992. Kings of Disaster: Dualism, Centralism and the Scapegoat King in the Southeastern Sudan. Leiden: Brill. ———. 2005. ‘Tragedy, Ritual and Power in Nilotic Regicide: The Regicidal Dramas of the Eastern Nilotes’, in Quigley Declan (ed.), The Character of Kingship. Oxford: Berg, pp. 67–100. ———. 2017. Kings of Disaster, etc., revised and illustrated paperback edn. Kampala: Fountain Publishers; also available as an e-book by Michigan State University Press. Southall, Aidan W. 1953. Alur Society: A Study in Processes and Types of Domination. ­Cambridge: Heffer. Speke, John H. 1863. Journal of the Discovery of the Source of the Nile, Edinburgh: William Blackwood & Sons. Vinco, Angelo. 1940. ‘Relazione del viaggio del Reverendo Sacerdote Don Angelo Vinco, Missionario Apostolico fra le varie tribù equatoriali del Fiume Bianco dal principio dell’anno 1851 fino alla metà del 1852’, Annali Lateranensi, IV: 300–28. Werne, Ferdinand. 1848. Expedition zur Entdeckung der Quellen der Weissen Nil (1840–1841). Berlin: G. Reimer. Wilson, C.T., and R.W. Felkin. 1882. Uganda and the Egyptian Sudan, 2 vols. London: ­Sampson Low, Marston, Searle & Rivington. Wrigley, Christopher. 2002. Kingship and State: The Buganda Dynasty. Cambridge: Cambridge University Press.

Chapter 4

The Shilluk Reth Early King or Head of State? An Inter-Nilotic Exploration Simon Simonse

After reading a draft of the preceding chapter, fellow anthropologists asked me how the comparative perspective on kingship presented there would work out for the kingship of the Shilluk. Among anthropologists and historians, the Shilluk king (reth) is one of the better-known cases of kingship in Africa and has been the subject of much anthropological discussion since James Frazer turned him into the centrepiece of the evidence in support of his idea that the drama of the dying god is a central theme in religious belief worldwide. The question asked by my colleagues may have been motivated by the presumption that the Shilluk might present an intermediate case between the early kingdoms of the monyomiji1 and the early state of Buganda. In the first part of this chapter, I will show that the dynamic of complementary segmentary opposition that had relatively free play in the monyomiji societies and that had the potential of leading to all-out confrontations between the community and its king, was absent from Shilluk society by the time the first ethnographic observations were made. Instead, we find two permanently unequal classes: the kwareth (royals) who monopolized all initiatives to use violence and the collo, (commoners), who only used violence as followers of kwareth. The position of reth was frequently contested and was the prize for the prince (nyireth) who had staged the rebellion and won. Consolidation of his power was a central concern of the reth. Applying the list of strategies to consolidate the king’s power that we identified in the previous chapter, we shall discover a great deal of commonality between the mode of operation of the Shilluk king and that of his counterparts in the monyomiji cluster. It will then become clear the Shilluk king was far ahead in neutralising

80  Simon Simonse Table 4.1  Periods during which recorded observations were made of the autonomous operation of royal power in the three studied types of kingdom before their transformation by colonial incorporation Time frame of observations

Authors and nature of accounts

Kingdoms of the monyomiji cluster

1862–1914

accounts by missionaries, explorers, traders, TurcoEgyptian governors; from 1900–1914 intelligence reports by the Uganda Protectorate Administration

Buganda Kingdom

1861–90

detailed, synchronic observations, by explorers, missionaries and emissaries of foreign governments

Shilluk Kingdom

1826–61

impressionistic reports of traders and explorers; documents from literate neighboring communities

rival sources of power while the kingdom had become irreversibly stratified to the king’s advantage. The assumption that the Shilluk practised regicide, or the denial that they ever did so, has been central to the academic debate on the nature of Shilluk kingship. Therefore, I cannot avoid touching on this issue. In the last section of this chapter, I will demonstrate that by adopting an intra-Nilotic comparative perspective, this more-than-a-century-old issue can be definitively laid to rest. There is only one snag to this: in order to make a fair comparison, we need to compare the kingdoms at a point in time when their political systems were functioning autonomously, and when kings could still set their own agendas in dealing with their subjects and with foreigners. The available information about our different cases of kingship is far from equal. Information about Shilluk internal social and political organization before the imposition of foreign rule is almost non-existent.2 The missionaries Banholzer (1902–14) and Hofmayr (1906–16), who edited the ethnographic material found in the late Banholzer’s diaries, the linguist Westermann (in 1911) and the anthropologist Seligman (in 1910) were the first to systematically collect information on the Shilluk, almost half a century after the imposition of foreign rule. In the accounts written by Condominium administrators, it is not always clear whether their description refers to a remembered past that was considered normative by the Shilluk they were dealing with, or to the situation contemporary to their administration.

The Undoing of Complementary Segmentary Opposition The Shilluk kingdom was spread out along the Nile as a 150-kilometre-long ribbon of settlements mainly on the west bank of the Nile. These settlements called podh in the Shilluk language were the largest political units on the level of the kingdom. In this chapter they will be referred to as ‘sections’, a term widely used

The Shilluk Reth  81

for comparable socio-political units in South Sudanese societies. Oral tradition tells us that Nyikang, the first king and leader of the Shilluk migration, divided the land into ten sections. A list made by the colonial administration and published in Pumphrey (1941: 40–-41; see also Howell, 1952: 99) enumerates eighty-nine sections grouped by the colonial administration into twelve or thirteen divisions. The sections were included into two sets of partially overlapping moieties, grouping the whole band of settlements along the Nile into two camps that were mobilized, often on a competitive basis, for nationwide events. One set, Gerr (north) and Luak (south), was politically defined and served during the crisis years of the late nineteenth century as a support base for rivalling princes, while the other set, Gol Nyikang and Gol Dhiang had a more ceremonial character and their interaction played a key role in the royal installation ceremonies. Pumphrey, writing in 1941, pointed out that: ‘There is a natural tendency of contiguous settlements to agglomerate, by reason of a common cause or of the existence of outstanding personality, into loose and impermanent confederacies, described by the term luak.’ (1941: 18). This way of clustering is in line with Evans-Pritchard’s model of complementary segmentary opposition and is identical to what happened in the monyomiji societies. Often laid out in a (semi-)circle at the foot of a mountain, adjacent villages come to each other’s support. In the monyomiji societies, this clustering process frequently polarized into a confrontation between moieties. In the Shilluk case it seems that the number of sections was just too large for a moiety as a whole to get involved in an intersectional conflict. If moieties were mobilized at all, it was after being prompted by royalty in a top-down manner (Schnepel 1990: 115–18). As a rule, according to Hofmayr (1925: 344), sections stopped fighting after a few casualties at a point when the number of victims on both sides was about equal. The king could use the fight as a pretext to raid the party that had started the fighting. The chiefs of the sections concerned therefore had an interest to quickly bring the parties together as a matter of urgency and negotiate peace by determining the bloodwealth for excess victims and overseeing its payment. During the Condominium period, the king participated in such meetings and demanded a share of the bloodwealth. According to Howell, most feuds were between adjacent sections. Fighting between sections was punished by the royal army (bang reth). Instead of sending his own army, the king could request the sections adjacent to the one to be punished to raid a fixed number of cattle on his behalf (the ‘royal levy’), while giving them carte blanche to burn villages of the neighbouring section and rewarding themselves with extra loot. This last practice may explain Howell’s observation that adjacent sections were each other’s worst enemies. It seems to

82  Simon Simonse

be a ­particularly cynical application of the ‘divide and rule’ principle. However, ­Howell adds that sections preferred being punished by their neighbours to a direct attack by the king’s army (Howell 1952: 113), a remark that points at a degree of local autonomy. Apparently, the sections concerned could secretly agree on how to minimize the damage of the royal levy. The sections consisted of a number of villages (myer, sg. pac), each inhabited by a cluster of lineages, - usually distinguished in lineages of original settlers, and immigrant lineages. Though the villages were certainly not havens of peace, their conflicts rarely spilled over into wider society. For the villagers, interference in their problems by higher level authorities such as the section chief and the king usually meant trouble rather than help. In the monyomiji societies, processes of fusion and fission of segments shifted from top to bottom and vice -versa, in response to the level of inclusiveness of the adversary the community had to deal with. In the Shilluk kingdom, bottom-up flexibility in the formation of coalitions has existed only at the level of the section. Sections were free to unite in military expeditions against enemies of the king as long as they gave him a share of the booty. All other processes of fusion and fission were top-down, driven by the dramas played out at the level of the royals. Kings mobilized sections and, occasionally, moieties to suit their own interests. If a society is so prone to division, war becomes the only remaining lever to keep all segments on board. In fact, all the older records confirm the reputation of the Shilluk as a warring nation. The relationship between kwareth and the collo was one of stratification rather than segmentation. The superiority of the royals was irreversible. Nyikang, their divine ancestor, was believed to have unilaterally collected the sections while migrating to Shillukland. The myth of origin that is shared by the related Anuak and Pari, and also by the Lotuho, has the ancestor king being captured from the river by his future subjects. This reverse scenario has been retained by the Shilluk as the beginning sequence of the royal installation ritual. The material from which the ceremonially important effigies of Nyikang and his son Dak are made are solemnly fished from the Nile (see below p.164). Open confrontations between the people and the king as occurred in the monyomiji cluster were unthinkable in the Shilluk context. Browsing the literature on the Shilluk, the only account of a confrontation between the king and the people that I could discover was mythical: ‘Nyikang had to discover that his people hated him. They disrespected and insulted him, making him tired of life. When Nyikang found out that they were looking for an opportunity to kill him, he decided to kill himself. One day, he called everyone together for a banquet. It lasted for four days and uncountable numbers of sheep and oxen were slaughtered.

The Shilluk Reth  83

The praise of the host was on everyone’s lips. On the last day, a suddenly rising whirlwind scattered everybody. Nyikang used the opportunity to take his life; he allowed his face to be tied with a cloth so that breathing was no longer possible.’ (Hofmeyer, 1 1910: 330, author’s translation). The showdown between King Nyikang and his people came to a premature end in an apparent act of ‘passive revenge’ by the king. Is the story telling us that there can only be unilaterally hierarchical relations between the king and his people, – as in Buganda, rocking the boat meaning suicide for the king and for the surviving people a perpetual state of guilt-ridden indebtedness?3 War united king and collo in a common project. The Shilluk were a feared enemy for their neighbours. Their compact mode of settlement must have given them an advantage over their pastoralist neighbours in terms of cultivating an esprit de corps, improving strategic and tactical skills and stimulating inventiveness. Contemporary assessments like this one by the Italian missionary-explorer Giovanni Beltrame, confirm the importance of war: ‘The Shilluk are the most turbulent, the boldest, the most treacherous and the most thievish people in the whole Valley of the White Nile. However, inside the tribe, the king and his section chiefs know how to use force to curtail this delinquent predisposition but when outsiders are targeted, the kings do nothing to stop them. Instead, they encourage their warriors because they receive part of the loot.’ (Beltrame 1881: 83). Before being incorporated in the Ottoman Empire the Shilluk kingdom was a regional superpower. Its navy controlled the Nile banks up to Eleis, 400 kilometres north of Muomo, the northernmost Shilluk section. The raids were ventures of a number of sections led by the king. Hundreds of warriors took part in these waterborne expeditions. Beltrame was impressed by the degree of organization of the Shilluk fleets. As waterborne cattle-raiders attacking from the water, they were feared by the Dinka and Arab pastoralists on both Nile banks. The Shilluk raiders had developed a method of moving cattle home over water, hopping from island to island. By the mid-nineteenth century, Shilluk raids were so frequent that the Nile crossing point for caravans travelling between the Sennar and the Tegali kingdoms was abandoned (Mercer 1971: 410–11).

Agem: Dynastic Rivalry and Regicide The Shilluk belong to a group of kingdoms in which succession used to be the outcome of a violent confrontation between royal pretenders to the throne. Shilluk history is full of kings who came to power after a showdown with a rival. There are several travellers’ reports of the excessive precautions that kings took against potential usurpers (Brun-Rollet 1855: 94; Lejean 1862: 871; Poncet

84  Simon Simonse

1863: 19; Beltrame 1881: 83). The reason given by Seligman for King Fadyet’s sleepiness during his interviews with him became the material for one of the most celebrated passages in Frazer’s The Golden Bough. The king was reported to stay awake during the night out of fear of a potential usurper (Frazer 1913, Pt. III: 22; Seligman 1932: 91). The custom of taking the royal throne by force has been reported for several communities that once were part of the Funj kingdom and for the Anuak, the Shilluk’s closest cultural relatives. Because the Anuak were only brought under colonial administration after 1920, they may also offer a lead as to how succession worked among the Shilluk. Studies of the Anuak political system were carried out by Evans-Pritchard in the 1930s and 1940s, by Lienhardt in the 1950s, by Perner in the late 1970s and early 1980s, and, in Ethiopia, by Kurimoto, in the 1990s. Agem is the name of the institution of the transfer of power. Translated by Evans-Pritchard as ‘revolution’ and by Lienhardt as ‘rebellion’, agem was practised to replace the headmen in the small-scale Anuak polities and the king in the larger Anuak kingdom. The agem was a joint venture of a legitimate pretender to the headmanship and a group of warrior-supporters (luak) of the would-be successor. Agem was also the name of the Shilluk institution that ensured the succession of section chiefs before they became appointees of the king. Evans-Pritchard learnt this in an interview with Alfred Heasty, one of the founders of the Doleib Hill American Mission Station and the author of the first Shilluk-English dictionary (Evans-Pritchard 1971: 161). This information is significant because it confirms that within the living memory of Heasty’s informants, Shilluk sections selected their own chiefs, independent of the king.4 The Anuak agem was a divisive institution. The price the community paid for this open rivalry was high. In a period of just over a year in the early 1940s, Evans-Pritchard witnessed three of these revolutions in small-scale Anuak polities. In the first, there were twelve casualties, in the second, there were also twelve, and in the third, there were six. As a consequence of the number of casualties in the village where the first fight took place, the community fell apart into two separate polities. For the relatives of the victims of these fights, it was impossible to negotiate compensation for the loss of life, because the only judge available was the winner of the fight. ‘Killing in a rebellion is compensated for by killing in a rebellion’ runs the Anuak maxim (Lienhardt 1958: 32). It is this logic of negative reciprocity that explains why succession came to alternate between two – and, in the case of the Shilluk, three – lineages of the royal clan. The alternation is not a selection criterion developed to limit the number of eligible candidates for the throne, as Schnepel (1990: 106–7) suggests, but a corollary of succession by agem. Alternation could only have become a positive rule after the colonial administration banned agem. Hofmayr puts the problem very succinctly: ‘Once the king has been killed the judges dealing with the case will be the perpetrators of the m ­ urder’ (1925: 179).5

The Shilluk Reth  85

Lienhardt also perceived a political benefit for the community in agem: The Anuak speak of agem, rebellion, with great enthusiasm; and it is possible even for a stranger to see eventually that it is by virtue of constant ostracisms of headmen that the Anuak are able to avoid diffused and uncontrolled conflict within the village, and to rule themselves, rejecting any submission to the symbol of their implicit contract which they themselves have created. (1957: 32) Lienhardt’s view echoes Evans-Pritchard’s assessment of Shilluk royal rivalry in his Frazer Lecture, which also points at the intimate link between kingship and consensual antagonism: It must here be remarked that Shilluk rebellions have not been made against the kingship. On the contrary, they were made to preserve the values embodied in the kingship which were being weakened, or it was believed so, by the individual who held office. They were not revolutions but rebellions against the king in the name of kingship. (Evans-Pritchard, 1948: 83) This may be true, but agem only united a part of the population behind the new incumbent, and against another part that was expected to one day take revenge. It made the kingship divisive, not unifying, as in the monyomiji societies. In the latter case, if there was a stand-off between the people and the king, and the issue at stake – usually rain – was not resolved, the opposition to the king would grow, gradually including all sections, age-classes, men and women, and, ultimately, leading to regicide.

The Shilluk King’s Strategies to Consolidate His Power The main threat to the Shilluk king’s life did not come from his people, but from his royal rivals, of whom there were many. For him, the strengthening of his grip on the throne was also a central concern. In the previous chapter, eight strategies were identified that were deployed by the monyomiji kings to tip the balance of power in their favour. Let us see whether and how these and similar strategies worked out in the context of Shilluk kingship. Generate Wealth and Create Dependants Like the Anuak headman (Lienhardt 1958: 35) and the monyomiji kings, the Shilluk king was the beneficiary of his subjects’ agricultural labour. Villages were ordered by the palace to take turns in clearing and weeding the king’s land. According to Tappi (1903: 122), the king was the only beneficiary, while sectional and village chiefs had to work their own fields. The explicitness of the addition

86  Simon Simonse

may suggest that the exclusivity of this privilege was a recent achievement by the king. The king was heir to the largest herd of cattle of the kingdom. The royal herd had been built over many generations. The most precious cattle were those descending from the herd of the dynastic ancestor Nyikang. The royal herd must have allowed the king to marry as many wives as he wished and also to provide wives for any of his sons (nyireth) and for his clients. The marriages of the king were an important political tool. When a wife became pregnant, she returned to her village of birth, where the child – who would be one of the potential heirs to the throne – would grow up and be at the centre of a royalist colony. This strategy of ‘planting princes in the periphery’ was also practised by the Lotuho and the Bari. The section where the prince grew up was destined to become his support base once he reached the stage where he would compete for the royal succession. As a result of this policy, the royal clan (kwareth) had a presence in all sections and was by far the largest in the kingdom. The large size of the royal clan also meant that there was a huge reservoir of potential rivals. If conflicts about princely entitlement escalated, collo could not avoid being part of the fighting. Create a Royal Army (Bang Reth) and Use Firearms As early as the first quarter of the nineteenth century, the king was reported to possess firearms. Most of the soldiers in the king’s army were war captives and debt slaves. Families who owed the king a debt, a fine or did not have the available cattle to reimburse a loan often provided the king with a family member to serve him as the guarantor of the payment. When the king presided over the settlement of a feud, he demanded, in addition to the bloodwealth paid to the family of the victim, one man or the man’s equivalent in heads of cattle to serve in the army. After the closure of the slave market of Khartoum, many slave traders moved to Kaka, which had a market overseen by the Shilluk king and was outside the zone where slaving was prohibited by the Egyptian government. When the market closed, many slaves were absorbed into the king’s army. Create Alliances with Neighbouring Kingdoms There is little information about kings’ alliances with surrounding kingdoms. From Hofmayr’s king-lists, we know that a number of kings married foreign princesses, which provides an indication of political alliances: – Dak married an Anuak princess as the conclusion of a war expedition into Anuak land. After Dak’s marriage, every king is said to have married a wife from Anuak; – Bwoc married a Nuba wife who became the mother of king Tokot;

The Shilluk Reth  87

– Tokot ruled that each king should at least marry one wife from the Nuba and one fromthe Dinka; – Tugo is known to have divorced his Nuba wife because of her poor cooking; – Kudit’s mother could have been a Nuba. When his brother Tyelgut assassinated – all the eligible nyireth, fifty in number, he fled to the Nuba Mountains. His hosts offered to avenge the murder of the fifty princes, but Kudit stopped them (Hofmayr 1925: 62–87). How exactly these marriages served the international relations of the kingdom is not remembered. In the nineteenth century, the most important allies of the kings became the successive governments in Khartoum. The sources on possible alliances between Shilluk princes and the Mahdist rulers in Khartoum are contradictory. Centralize Royal Power by Abolishing the Age-Class System The Shilluk age-system organized the young men at the level of the section (podh). The Shilluk age-system must have been ‘in tune’ with the age-systems of the surrounding peoples. I found no records of age-grades of different sections joining in ceremonies or in warfare, as was common practice among the Nuer and in the monyomiji societies. Age-sets were divided into three age-based subdivisions: ‘head’, ‘neck’ and ‘middle’. As in the monyomiji societies, young men were eager to become warriors and push their predecessors out into retirement. The agegrade of warriors was led by a sectional war leader (bany) who was a member of an age-grade senior to that of the warriors. Participation was compulsory for all young men. The cicatrisation of the forehead, characteristic for Shilluk men, was performed in the context of the promotion ceremony of the age-set. The age-system must have provided an important countervailing power to that of the king for a long time. Hofmayr (1925: 342) briefly mentions ‘our army of unmarried lads divided in age-grades’, adding that their chairmen/choir-leaders were appointed by the sectional chiefs. Seligman (1932: 46) tried to collect information on the age-system, but never got a straight answer to his questions. Were age-classes a sensitive subject in the circles of King Fadyet, who received the Seligmans? Howell mentions that the king considered the age-classes to be a subversive influence. A chiefs’ meeting in 1934 advocated an age limit for the bany (Howell 1952: 107). Age-classes were formally banned in 1941 in a ruling that had the full agreement of both King Papit and the colonial government. Despite the ban, one southern section appointed a new bany in 1947 and set out to parade him in the chief village of the northern moiety. They passed within earshot of the palace in Fashoda singing their songs. A stand-off between the king and the warrior age-set followed.

88  Simon Simonse

In the article in which Howell discusses the case, he takes the side of the bany and his age-set pleading that an aggressive self-presentation should be expected from warriors that are being stopped from fighting (Howell 1941: 56–60; Howell 1952: 108–13). As an episode in a power contest between the king and the warrior age-grade, the incident shows that even under Condominium rule, the king was concerned about his power base. Here we see the Shilluk king winning a political battle against the bottom-up age-system that his counterparts in the monyomiji cluster never managed to win. Centralize the Kingdom by Bringing the Sections under Direct Royal Control The historical evidence suggests that the sections of the kingdom that lost a great deal of independence after the Shilluk were brought under the Egyptian sphere of influence. From being led by an aristocrat who succeeded to the chieftaincy by agem, the section-chief became a jago, a word meaning ‘second-in-command’ in many Lwoo languages. The jago continued to enjoy some autonomy when carrying out raids. He paid a percentage of the value of the booty to the king. Needham (1980), who suggested that the section chiefs exercised ‘political sovereignty’ in a hierarchical system in which the king exercised ‘sacred sovereignty’, completely misses the competitiveness that animated Shilluk kingship. Before the establishment of Condominium rule, when the king still had his army, the king asserted his authority by raiding ‘disorderly’ sections as a punishment and by a policy of divide and rule (see also p. 147). It was also a context in which feuds between sections were rampant, went very deep and were long-­ lasting. Howell examined a feud that lasted from 1905 to 1932. During those years, it resulted in thirteen violent incidents in which forty-six men died and many more were injured. Only in 1932 was the king able to persuade the parties to reach a settlement (Howell 1952: 108). One could ask whether an intersectional feud could have lasted that long if the king had still disposed of his army. Howell, who had earlier served as an administrator in the Nuer and Dinka districts, considered the feuds of the Shilluk ‘bloodier and more lasting than any likely to occur among the Dinka and Nuer’ (ibid.: 106). With the establishment of Condominium rule, the power of the king quickly diminished. From every murder case tried under his auspices, he received one boy for his army in addition to the girl paid to the party of the victim (Seligman and Seligman 1932: 46). When the government prohibited the transfer of persons in the payment of bloodwealth, only cattle were used at the traditional rate of ten head of cattle for a victim. The portion for the king was no longer fully paid and quickly became smaller (Howell 1952: 114). The practice of paying a person to the king also explains the reluctance of lower-level chiefs to submit cases to higher-level jurisdiction (see also p. 147). During Condominium rule, the role of the king became increasingly that of a peacemaker. He often acted alongside the District Commissioner, persuading

The Shilluk Reth  89

the parties to accept the verdict of the court or the peace deal proposed. He presided over the accompanying sacrifices and occasionally, in order to underline the historic importance of an event, provided a cow that was descended from Nyikang’s own herd (Howell 1952: 113). Consolidate the Kingdom’s Cohesion by Regularly Waging War on Enemies Not only was the competition between princes for the throne divisive, but so was the manner in which the king repressed sectional subversion. Warfare with outsiders was the easiest option to keep the segments of the kingdom aligned. Names of kings were identified with the wars they had fought, as Hofmayr’s kinglist shows (1925: 59–136). Diversify the Social Composition of the Kingdom and Introduce Complementarity in the Division of Labour In the literature I consulted, I found no specific traces of complementary role-­ relationships that might have developed in the context of the role that the Shilluk played during the nineteenth century in regional trade and in hosting a community of foreign traders in Kaka. If there were any, they may have remained shortlived. The king’s army, consisting of men in debt-bondage, war-captives, and slaves operating under a top-down discipline different from that of warrior agesets, was part of a general transformation of the kingdom around the unilateral, nonreciprocal relationship between the king and the people, and by extension between kwareth and collo. Trade Trade was an important asset in the power of the king. As early as 1840, the Shilluk king maintained an embassy in Khartoum to promote and control trade with his country. He offered protection to trading caravans for a price. This safe conduct was absolutely reliable according to reports by traders. The protection must have been ensured by the bang reth and not by the warriors, as Mercer suggests (Mercer 1971: 418). In 1843, the king established a market in Kaka supervised by his appointees. For a long time, this was the only place in the kingdom where foreigners were allowed – apparently, a measure to stop traders from dealing with other Shilluk partners. The king established a monopoly on ivory and guns. When Egypt abolished the slave trade in the 1850s, the slave markets moved to the south, outside the territory controlled by the Turco-Egyptian government. Kaka became an international boomtown. Political enemies of the government in Khartoum also flocked to Shilluk-land as refugees. The royal stores in Fashoda were filled to the brim with merchandise (Mercer 1971: 423). This wealth must have served the king to build the bang reth and to

90  Simon Simonse

provide them with firearms. This, in turn, increased his leverage to raise taxes in cattle he traditionally received from his subjects. According to Schnepel (1990: 115), this must have been the time that the balance of power between the section chiefs and the king shifted in favour of the king. Establish and Promote a Cult that Unified the Dynasty and Regulated Succession The generalized discord produced by the relentless competition for power of the princes combined with sectional rivalry cried out for ideas that could unite the people among themselves and with their king. King Tugo, who ruled the Shilluk kingdom from 1690 to 1710, is credited with the establishment of Fashoda as the capital and with the introduction of the cult of Nyikang, the ancestor of the royal clan (Hofmayr 1925: 77). Apparently, before Tugo’s rule, kings were peripatetic, moving from section to section in order to collect tributes, to perform sacrifices and to settle conflicts, like the kings in the monyomiji cluster (Simonse 2017: 258–61). The new cult which is being observed till the present day involves the following: • The belief that each king is and should be a successor of Nyikang. From the moment the king is enthroned, the spirit of Nyikang is believed to inspire him. The king’s actions and statements are fully in line with Nyikang’s intentions. Any action of the king is an expression not just of his own will, but also of the will of Nyikang (Hofmeyer, 1925: 48). • The building and maintenance of shrines in all sections of the country, where Nyikang could be invoked and sacrifices for rain could be made. • An elaborate script for the enthronement of the new king that consisted of a series of ceremonies that resulted in the identification of the new king with the figure of Nyikang. The installation is so organized that a large cross-­ section of the population participates. • A large, coherent body of myths and legends about Nyikang as the dynastic ancestor and the Shilluk cultural hero, which must have replaced earlier, more diverse and localized oral traditions. • A loose collection of beliefs regarding appearances of Nyikang – in wind, in particular animals, in behaviours of people affected by his power, etc. This cult ensured that there could only be one true king at any time. Any person claiming to be king who had not been subjected to the formal installation procedure did not represent Nyikang and had no right to the throne. The scenario of the installation ceremonies was extensive and full of dramatic detail.6 Two separate but intertwined sequences can be distinguished in the proceedings: one focused on the relationship of the new king to Nyikang, and

The Shilluk Reth  91

another, more archaic, sequence focused on the sacrificial role of the king extending from his designation as king to his final enshrinement in his village of birth. The execution of the first sequence is in the hands of the Nyikwom, ‘the men of the four-legged sacred stool’, the ‘enthroners’. This sequence dates from the reign of King Tugo, who is remembered as having created the installation ceremonies. Of the other sequence, a special group of royals called the Ororo was in charge. They managed the king’s sacred being: from the incest he committed during the period of seclusion to his dying and funeral. The Ororo were once members of the royal clan, but had been degraded and no longer qualified for the kingship. There are several stories about why and how they were degraded. It is also obvious that the officials who managed the dying process of the king could not be in the same social category as those competing for the succession. The first sequence starts with the search for Nyikang in the river abode of his mother, who is a half-crocodile. A piece of ambatch wood appears from the river and is crafted into an effigy of Nyikang. Artists from all over the country contribute to the finishing of the effigy. The enthroners, surrounded by a mock army, carry the effigy of Nyikang from its northerly temple to the south, where the king-elect is waiting. At the moiety boundary, the army of the northern moiety (Gol Dhiang) carrying the effigy and that of the southern moiety (Gol Nyikang) clash in a mock battle. In the melee, the enthroners capture the king-elect and establish physical contact between him and Nyikang’s effigy. They then take the king and the effigy to the house of Nyikang in Fashoda. There they put the king on the four-legged throne where Nyikang, as a spiritual being, takes possession of the king. The Shilluk king is captured by the spirit of kingship. This capture scenario is the diametrical opposite of that played out on his Anuak counterpart. The enthronement of the Anuak king is conditional on the aspiring king’s succeeding in capturing of the royal necklaces. Lt. Col. Bacon, the army officer sent to obtain the king’s submission to the Condominium government in Khartoum, witnessed the installation ceremony of King Akwei Cam in 1920. When the Master of Ceremonies had seated the new king on the higher of two four-legged stools, he placed the royal necklaces around the king’s neck and handed him the royal drum and a short drumstick. The king then beat the drum and took the oath of office: ‘I am a true son of the king; I am wearing the royal necklaces and sitting on the throne. If I am not a son of the king, let me die!’ (Bacon 1924: 116).7 If we assume that the Anuak installation ceremony represents an archaic variant of that of the Shilluk, the innovation introduced by King Tugo transformed the enthronement from an act of victorious self-promotion into one in which a humbled prince is captured by a dynastic power greater than himself. This reading is the diametrical opposite of Graeber’s proposition:

92  Simon Simonse

Divine kingship … refers not to the identification of rulers with supernatural beings … but to kings who make themselves the equivalent of gods–arbitrary, all powerful beings beyond human morality–through the use of arbitrary violence. (Graeber 2011: 13) The conduct required from the new Shilluk king during his installation is one of humble respect for the rules of succession. The gist of Tugo’s reform was that the installation was not a divinization of the person of the king, but that it was the royal office that was divine and not the person holding it. The new Nyikang cult was aimed to contain succession rivalry. This matches the way King Tugo is remembered as a particularly peaceful king ‘concerned to bring happiness and prosperity to his people’ (Hofmayr 1925: 76). There is no lack of Shilluk kings who would fit Graeber’s profile of a defiant, amoral, superhuman being: Tyelgut (ca. 1765) and Akoj (ca. 1840) are good examples (Hofmayr 1925: 83, 96). These kings do not represent the spirit of the Nyikang cult. They may have been the type that King Tugo wanted to exclude from power. Their violent style brings them closer to the kings of Buganda.

Safeguarding the King’s Sacrificial Destiny: The Reth’s Dying from an Inter-Nilotic Perspective The old-time religion, before the introduction of the cult of Nyikang, was centred on the king’s sacrificial role and extended from his installation to the enshrinement of his mortal remains in his village of birth. It began with the official announcement (or is condemnation a better term?) of the prince’s designation as king by the royal electoral college. The pronouncement was made by the son of the Chief of Gol Nyikang, the southern moiety. It ran as follows: ‘You are now our Dinka slave. We are going to kill you.’ The next important step in the sacrificial sequence is made after the spirit of Nyikang has taken possession of the new king. The king enters a three-day seclusion in a makeshift shelter (adul) erected for the installation. He is treated as a simple cowboy and regularly humiliated. His body is washed alternately with hot and cold water and adorned with the bracelets, ankle-bands, skins and precious beads that are the insignia of royal office. One of the nights is spent with a half-sister from the Ororo clan, an act classified as akwalo (incest). The public installation following the seclusion takes place under the responsibility of the Nyikwom, while the king’s achievements and failures as a ruler are judged with reference to the cult of Nyikang. Only when the king gets old or sick do the king’s wives belonging to the Ororo clan reappear on the scene. Measures must be taken to prevent the king’s death from occurring naturally. In most cases, a cloth is pushed onto the king’s face so that he is suffocated. While it is clear that the Ororo wives of the king play the

The Shilluk Reth  93

crucial role, unambiguous information on the decision-making process leading to the king’s assisted death is lacking.8 Wailing is prohibited in the whole of the kingdom. The king’s body is sealed up in a house until it has decomposed. In the past, the king would be accompanied into the grave house by a nubile virgin. There are older traditions that speak of kings having been buried alive. When the king’s body has decomposed, the bones are returned to his village of birth. Ororo men dig his grave and build a hut on top of it that will serve as the late king’s shrine. This is followed by several days of continuous dancing.9 The episode in which the king is killed or ‘is assisted to die’, as the case may be, was surrounded with secrecy and mystery. This secrecy may be a side effect of King Tugo’s promotion of the public Nyikang cult that seems to have given the king’s sacrificial death a more private character. We should also be aware that the mystery surrounding Shilluk regicide could very well be an artefact of the anthropological interest in the subject that began with Seligman, who eagerly interpreted the bits and pieces of information he collected on the deaths of Shilluk kings as a mirror image of Frazer’s drama of the Rex Nemorensis. Evans-Pritchard was aware of Seligman’s bias and proposed an interpretation of Shilluk regicide that was based on his understanding of Anuak succession rivalry and especially of agem. He argued that there was no ethnographic evidence of ‘ritual regicide’ (Evans-Pritchard 1948). Subsequently, dozens of anthropologists picked the version of the facts that suited their theoretical preference. It should be noted that Schnepel in his otherwise admirable listing of the episodes of the king’s installation and funeral skips the killing of the king. His summary begins with a scene where the king is already lying in state, for which no background is provided. For Graeber, the regicidal scene is just another piece of evidence showing the decisive role of violence in the emergence of kingship. While I agree with Graeber’s view on the crucial role of violence, the regicidal scene made famous by Frazer and Seligman has a different meaning, as I will now show by making a comparison with the funeral practices of other Nilotic kings. The announcement ‘You are now our slave; you are our Dinka war-captive’ as the opening of the sacrificial trajectory is a declaration of ‘capture’. Capture is also a customary way in which the monyomiji bring the new king’s status home to him. He is captured at an unexpected place and time10 against the wishes of the future king and his family. Upon receiving the news of their nephew’s designation to be king, the maternal relatives of the new Lulubo king immediately run to the monyomiji demanding bloodwealth.11 The Lokoya monyomiji used to abduct their new king as if he were a leopard. They dropped him at the home of the Master of the Hunt, who will domesticate him by making him shoes of giraffe leather and by smearing his body with giraffe fat, metamorphosing the carnivore into an herbivore. When the king dies, he is

94  Simon Simonse

reunited with the animals he was separated from at his installation. These animals are expected to congregate at his grave. When, at the death of their king, the Lopit fear that the king is not returning to his former state, they move a leopard skull in front of his face to remind him of his pre-royal companions. In some old Shilluk practices, there are remote echoes of the same way of proceeding. For instance, royals did not have their lower incisors removed like their ‘cultured’ commoners (Hofmayr 1925: 349; Seligman and Seligman 1932: 72). During the enshrinement dance, men imitate the roaring of lions (Howell and Thomson 1946: 25). Immediately after death, kings in the Lotuho- and Bari-speaking communities had their bodily orifices plugged with a paste-like substance or with leaves or both. If it is not done, informants say, general disorder in nature and society would follow. A predator may jump up from the deathbed. People believe that there is a charge inside the king’s body that is beneficial for the community, but also very dangerous if released irresponsibly. The utmost care must be taken that the blessings are not lost or spoilt by maleficent forces. The same reason is given for the taboo on wailing and making noise at the king’s death. Any sacrifices accompanying the king’s death should also be noiseless, by suffocation or strangling. Communities that plug the king’s orifices with a mixture of leaves and sesame paste remove the plugs before burial. Their removal, their conveyance by old women (interrupted by stops at fixed stations where they devotedly kneel) and their disposal in a riverbed are believed to be among the most sacred duties a person could ever perform. What was dropped in the river was the dangerous, potentially destructive forces clinging to the king’s body with the ultimate aim of safeguarding the integrity of the ‘power of the king’s blessings’. I must make mention of the remarkable interpretation of the same scenario by some west bank Bari-speaking communities. The body of the king was put on a sorghum-drying platform and allowed to bloat until the corpse started releasing liquid. At that time, peace, rain and food are believed to spread over the land ‘like the air from a punctured ball’, as an informant put it. One community in the monyomiji area, the Pari, who migrated from the northeast and spoke a dialect of Northern Lwoo like the Shilluk and Anuak, observes a different custom of disposing of the king’s body. While they are fully integrated into the monyomiji cluster, they have kept many customs from their place of origin, one being the way of dealing with their dying king. A few years ago, during a trip to his research area, the anthropologist Eisei Kurimoto was told that when the Pari King Fidele Ongang was about to die in 1980, his body and face had been covered with the freshly removed skin of a bull especially killed for that purpose. The skin was pressed against the mouth and nose of the king by a close kinsman so that the king died of suffocation. This was done, Kurimoto’s informants (who were both members of the royal clan) told him, ‘to preserve the gweth of the king’. Gweth is the opposite of cien, the posthumous curse left by a

The Shilluk Reth  95

person who dies while in discord with his family or community. It is the dying king’s power of blessing.12 It is obvious that the Pari practice is an archaic version of what the Shilluk do using a cotton cloth. The switch to a piece of cloth could be explained by the greater privacy of the Shilluk king’s death after Tugo’s reform. The fact that women (the Ororo wives of the king) were put in charge of the procedure may also explain why the slaughter of a bull was abandoned. The conclusion that in these three cases we are dealing with the same ritual procedure – the purification, optimization and celebration of the blessings of the dead king – is compelling. The difference is in the body functions used to make the ritual point. The Bari and Lotuho have picked the digestive process as the transformative agent, while the Pari and the Shilluk opted for the respiratory system.13 For the Shilluk and the Pari, but also for the Dinka, breath (wei in Shilluk and Dinka, and jwei in Pari) is the carrier and manifestation of the force of life. At death, this vital force is not lost. In the case of a sacrificial animal, it is the wei of the sacrificial victim that boosts the wellbeing of the community (Lienhardt 1961: 207), and so it does when a Dinka Master of the Fishing Spear dies. In certain sections and clans of the Dinka, the culturally preferred death of the Spearmaster was to be buried alive when reaching an advanced age. This was a momentous religious occasion. The wei released at the moment of death was believed to be the Spearmaster’s ultimate blessing for his people and should at all costs be prevented from accidentally ‘expiring’: What [the Dinka] represent in contriving the death which they give him is the conservation of the ‘life’ which they themselves think they receive from him, and not the conservation of his own personal life. (Lienhardt 1961: 300) In other Dinka sections, the Masters of the Fishing Spear used to be strangled or suffocated when they reached old age. We may assume that the concern that motivated this mode of dying was the same as that of being buried alive: control in the capture of the king’s wei (Titherington 1925: 196; Seligman and Seligman 1932: 196–98; Bedri 1939 and 1948). All this adds substance to the argument that the Shilluk ‘regicide’ is part of the widely spread Nilotic practice of ‘catching life in the spell of the king’s death’ (Simonse 1992; Simonse 2017, Chapter 18). When digestion is the symbolic medium, the management of the release of the blessings can wait until after the king has died. But respiration can only be controlled when it is still occurring. When we compare the death of the Shilluk kings to that of their Equatorial counterparts, their being killed appears to be a ritual technicality, imposed by the choice of body symbolism, a means to an end rather than an end in itself.

96  Simon Simonse

Interestingly, Dinka vocabulary still appears to carry a trace of the symbolic equivalence of respiration and digestion. The Dinka word wei also refers to ‘chyme’, the stomach contents in the process of digestion. While Evans-Pritchard and Lienhardt are both well aware of the purificatory and life-enhancing use of rumen in ritual, they stop short at making the connection (Evans-Pritchard 1956: 214; Lienhardt 1961: 256, 294). If this analysis is correct, the rationale of Shilluk regicide is not the maintenance of the positive, imitative relationship between the king’s health and the country’s wellbeing that Seligman (1934) and Frazer postulated, but the concern that an uncontrolled death of the king might spoil its blessings. In the Bari and Lotuho kingdoms, no successor was installed until the ‘reign of the dead king’ was complete. The installation of a new king would interfere with the flow of blessings from the dead king and backfire. In some areas, the reign of the dead king lasted up to four years. During this period, there was a taboo on all behaviour associated with violence, dancing, drumming, noise, etc. The only indication that the Shilluk may also have respected a ‘reign of the dead king’ is the duration of the interregnum of four to thirteen months reported by Schnepel (1988: 443). Depending on the season during which the king died, this might be the shortest period possible, and the longest necessary for the dead king’s rain to preside over one harvest. Nowadays, for the Shilluk, the period after the king’s death is known as the ‘year of fear’ because of the ongoing succession strife. In this respect, Shilluk succession has more in common with that in Buganda where the interregnum was also a period of violent disorder and fear.

Conclusion In the preceding chapter I argued that it was possible to overcome the dichotomy that African Political Systems postulates between Group A (hierarchical, centralized) and Group B (acephalous, segmentary) societies if, from the start of our analysis, we focus on the power relationship that is constitutive for the political entity under study. It then appears that in a number of African kingdoms, the relationship between the king and his subjects plays out in much the same way as the ‘balanced opposition’ that is characteristic of segmentary societies. I labelled such kingdoms where royal power was permanently in the balance ‘early kingdoms’. In these kingdoms, the oppositional relationship between the king and the people compelled the king to strengthen his position. The ultimate outcome of this royal drive for domination were polities in which the king ended up monopolizing the use of physical force – Max Weber’s primary criterion for a political entity to qualify as a state. Within states, social relations tend to be restructured along unilateral, top-down lines, while the complementarity of social roles gradually replaces reciprocity.

The Shilluk Reth  97

Among the Nilotic kingdoms in South Sudan, the Shilluk kingdom is by far the largest and the most elaborate. It has greater time depth, a larger population than other kingdoms, and its kings have gone a longer way in terms of eliminating or subordinating rival sources of political power, such as the age-class system and the sections and their chiefs. The royals (kwareth) who live spread over the whole extent of Shillukland form a distinct class superior to the commoners, the collo, and have privileged access to the king. Their dispersal over the kingdom contributes to its geographical cohesion. By providing patronage to commoners in their area of residence, they play an integrative role vertically between the royals and the collo. This royal upper stratum of the Shilluk is larger and more clearly marked and institutionalized than in the societies of the monyomiji cluster. One reason for this difference is that the time depth of the monyomiji rain-dynasties is relatively shallow. Among the peoples of the Upper Nile, the Bari who border the monyomiji cluster on the west, come closest to the Shilluk set-up. Like the Shilluk their society was stratified in a royal class (kör), a class of free commoners (böngön) and various unfree groups (‘dupi, tomonok) while their age-class system was patronized by big men. At the peak of its power during the early nineteenth century, the Bari kingdom of the Bekat straddled the Nile over a distance equal to the length of the Shilluk kingdom and had at least as large a population. The Bari kingdom fell apart, while the Shilluk kingdom survived the upheavals of the late nineteenth century. This chapter suggests that the relative stability of the Shilluk kingdom may be due to the Nyikang cult that established a framework for the legitimacy of succession that the Bari lacked and maybe needed more due to their more intense exposure to invading powers that were eager to make use of their internal rivalry. By indiscriminately attacking sections of his kingdom that he suspected of disloyalty and by ordering sections to attack one another in the king’s name, the Shilluk king went a long way towards destroying the capacity of the sections to unite against him and in so doing building a political body that was unilaterally top-down. Yet the sections survived – and so did the clan and lineage-based constituent divisions of the sections. The Shilluk kings did not touch the lineage-based access to land as the kings of Buganda did. Because the power of the king of Buganda trickled all the way down into the capillaries of local communities, in order to remain credible and keep his people together, the king was less dependent on a continued performance in the battlefield, as seems to have been the case for the Shilluk king. Taking into account the scale of the political body and the degree of transformation of its constituent social relations into top-down, unilateral complementarity, the Shilluk case might be halfway between a nascent early kingdom – like the small-scale kingdoms in the monyomiji cluster – and a fully fledged early state like Buganda.

98  Simon Simonse

Simon Simonse is a social anthropologist who trained in Leiden and Paris. He carried out field work on kingship and social cohesion in the societies of the ­monyomiji cluster in South Sudan while lecturing in the University of Juba. His doctoral thesis ‘Kings of Disaster’ shows how models derived from Girard’s mimetic theory account for the dynamism that animates both hierarchical and segmentary social structures in South Sudan. He has also taught in the Democratic Republic of Congo, Uganda, Indonesia, Japan, France and the Netherlands. For the last twenty-five years, he has worked as a conflict transformation expert in the Great Lakes region and the Horn of Africa.

Notes   1. The societies of the monyomiji cluster distinguish themselves by practising a system of ‘dual rule’, by the age-grade of adult men (monyomiji) on one side and by the king on the other. They occupy the mountainous east bank of the Nile between the Bari to their west and the Kidepo river as their eastern boundary. They include the Lulubo, the Lokoya, the Pari, the Horiok, the Lotuho, the Lomiya, the Ngotira, the Dorik, the Tenet and the Ngaboli (the last five collectively known as ‘Lopit’), the Dongotono, the Logir, the Ketebo, the Lokwa, the Lorwama and the, Imotong (the last six collectively known as ‘Lango’). All are speakers of dialects of the Eastern Nilotic language that is spoken by the Lotuho, the largest and best known group, except the Lulubo, Tenet and Pari. The Lulubo (Olu’bo in their own language) are Central Sudanic, the Tenet are Surma speakers and the Pari belong to the Western Nilotic Northern Lwoo group like the Shilluk and the Anuak. I follow the convention to use the term ‘Lwoo’ for the language group and Luo as the name of the two ethnic groups in Kenya and South Sudan. The cluster counts more than thirty kingdoms varying in size from a single village to the fifteen large agglomerations that make up the Lotuho kingdom of the Mayya dynasty. On the monyomiji age-system, see Kurimoto and Simonse (1998: Chapters 2 and 3).   2. Patricia Mercer’s article (1971) on Shilluk trade before the kingdom’s inclusion in the Egyptian Sudan helps to fill this important gap.   3. This story may also be read as a Shilluk reflection on the way in which Dinka Spearmasters ended their lives, some taking the initiative to be buried alive, others being killed by their community (Lienhardt 1961: 310). Nyikang ends up using the typical Shilluk suffocation technique. The story does not say who tied the cloth.   4. Evans-Pritchard only published Heasty’s information in 1971, more than thirty years after collecting it, in a French journal. The article also contains a long list of ethnographic errors made by Seligman in his Shilluk chapter in Pagan Tribes of the Nilotic Sudan (Seligman and Seligman 1932). Evans-Pritchard considers his discovery of the use of the term agem among the Shilluk significant and urges future researchers to follow up on it. Heasty was Seligman’s key informant and Seligman was instrumental in bringing Evans-Pritchard to Sudan.   5. The way in which alternation of succession between royal lineages was practised in the Funj area offers further confirmation of this. On his very first research trip in Sudan in 1924, Evans-Pritchard collected information on regicide in the Funj area. In the community of Jebel Ulu, out of twelve successive kings, four had been killed by enemies. Another four were killed by agnatic brothers or a son (descendants in the same patriclan, but of a different lineage), while three of these killers succeeded their victim (Evans-Pritchard

The Shilluk Reth  99 1932: 14). In Jebel Kele, out of ten kings, five had been killed by an agnatic brother or son and three of these killers succeeded their victims (ibid.: 43). The regicide scenario among the Berta of Fazogli, where the last two kings (Yassin and Asussu) were killed in 1838 and ca. 1872 respectively, seems closer to that of the monyomiji societies (Lepsius 1853: 202; Marno 1874: 68; Evans-Pritchard 1932: 44–46).   6. For a detailed summary, see Schnepel (1988).   7. Akwei Cam was the son of Cam Akwei who rebuffed the British patrol sent in 1912 to conquer his kingdom. At the young age of twelve, he inherited the necklaces from his father. He did not acquire them by agem. Before the ceremony took place, a rival prince across the border in Ethiopia had already manifested his displeasure at Akwei’s enthronement. Elements of the Anuak ceremony of enthronement are recognizable in the more evolved Shilluk ceremony, e.g. the holding of the four-legged stool by prostrate enthroners when the new king is made to sit on it for the first time (Bacon 1921: 164).   8. Schnepel (1991: 42–50) provides a useful summary of the discussion on this topic.   9. The Shilluk-Luo of Bahr-el-Ghazal, closely related to the Shilluk, dealt with the death of their kings in a very similar way to the Shilluk. See Santandrea and de Georgi (1965: 14–16). 10. The current king of Tirangore who took office in 2008 was abducted from his office in Juba.. 11. For a presentation of the ritual career of the monyomiji kings, read ‘“Catching Life in the Spell of Death”: The Ritualisation of the King’s Victimhood’, (Simonse 1992 and 2017, Chapter 18). 12. Personal communication with Eisei Kurimoto. 13. On Nilotic digestive symbolism, see Simonse (1994).

References Bacon, Lt. Col. C.R.K. 1921. ‘Sobat and Pibor District. Kingship amongst the Anuak’, Sudan Notes and Records, 4(3): 162–64. ———. 1924. ‘The Investiture of an Anuak Nyeta or Sultan’, Sudan Notes and Records, 7(2): 114–17. Bedri, I. 1939. ‘Notes on Dinka Beliefs in Their Hereditary Chiefs and Rainmakers’, Sudan Notes and Records, 22(1): 125–31. ———. 1948. ‘More Notes on the Padang Dinka’, Sudan Notes and Records, 29(1): 40–57. Beltrame, Giovanni. 1881. Il Fiume Bianco e i Dinka, Memorie. Verona: G. Civelli. Brun-Rollet, Antoine. 1855. Le Nil Blanc et le Soudan, Etudes sur l’Afrique Centrale, Mœurs et Coutumes des Sauvages. Paris: Librairie de L. Maison. Evans-Pritchard, E.E. 1932. ‘Ethnological Observations in Dar Fung’, Sudan Notes and ­Records, 15(1): 1–61. ———. 1948. The Divine Kingship of the Shilluk of the Nilotic Sudan. Cambridge: University Press. ———. 1956. Nuer Religion. London: Oxford University Press. ———. 1971. ‘Sources, with Particular Reference to the Southern Sudan’, Cahiers d’Études Africaines, 11(41): 129–79. Forde, Daryll (ed.). 1954. African Worlds: Studies in the Cosmological Ideas and Social Values of African Peoples. London: Oxford University Press. Frazer, James G. 1913. The Golden Bough: A Study in Magic and Religion, 13 vols. London: Macmillan.

100  Simon Simonse Graeber, David. 2011. ‘The Divine Kingship of the Shilluk: On Violence, Utopia, and the Human Condition, or, Elements for an Archaeology of Sovereignty’, HAU: Journal of Ethnographic Theory, 1(1): 1–62. Hofmayr, Wilhelm. 1925. Die Schilluk: Geschichte, Religion und Leben eines Niloten-Stammes. Nach P. Banholzer F.S.C. und eigenen Aufzeichnungen dargestellt, Mödling bei Wien: Anthropos Verlag. Hofmeyer, P.W. 1910. ‘Zur Geschichte und sozialen und politischen Gliederung des Stammes der Schillukneger’, Anthropos 5(2): 328–33. Howell, P.P. 1941. ‘The Shilluk Settlement’, Sudan Notes and Records, 24: 47–67. ———. 1952. ‘Observations on the Shilluk of the Upper Nile: The Laws of Homicide and the Legal Functions of the Reth’, Africa: Journal of the International African Institute, 22(2): 97–119. Howell, P.P., and W.P.G. Thomson. 1946. ‘The Death of a Reth of the Shilluk and the Installation of His Successor’, Sudan Notes and Records, 27: 5–85. Kurimoto, Eisei. 1992. ‘Natives and Outsiders: The Historical Experience of the Anywaa of Western Ethiopia’, Journal of Asian and African Studies, 43: 1–45. Kurimoto, Eisei, and Simon Simonse (eds) 1998. Conflict, Age and Power in North East Africa: Age Systems in Transition, Oxford: James Currey. Lejean, Guillaume. 1862. ‘Le Haut-Nil et le Soudan: Souvenirs de Voyage, I, Les Empires Noirs et les nouvelles découvertes du Fleuve-Blanc’, Revue des Deux Mondes, 37(4): 854–82. Lepsius, Karl Richard. 1853. Letters from Egypt, Ethiopia and the Peninsula of the Sinai. ­London: Henry G. Bohn. Lienhardt, Godfrey. 1954. ‘The Shilluk of the Upper Nile’, in Daryll Forde (ed.), African Worlds, Studies in the Cosmological Ideas and Social Values of African Peoples. London: Oxford University Press, pp. 138–64. ———. 1957. ‘Anuak Village Headmen I’, Africa: Journal of the International African Institute, 27(4): 341–55. ———. 1958. ‘Anuak Village Headmen II, Africa: Journal of the International African Institute, 28(1): 23–36. ———. 1961. Divinity and Experience: The Religion of the Dinka. Oxford: Clarendon Press. Marno, Ernst. 1874. Reisen im Gebiete des blauen und weissen Nil; im egyptischen Sudan und den angrenzenden Negerländern, in den Jahren 1869 bis 1873. Vienna: Carl Gerold’s Sohn. Mercer, Patricia. 1971. ‘Shilluk Trade and Politics from the Mid-seventeenth Century to 1861’, Journal of African History, 12(3): 407–26. Needham, Rodney. 1980. Reconnaissances. Toronto: University of Toronto Press. Perner, Conradin. 2015. The Political Body – Power and Authority: The Anyuak, Living on Earth in the Sky, vol. VI. Basel: Schwabe Verlag. Poncet, Jules. 1863. ‘Notice géographique et ethnologique sur la région du Fleuve Blanc et sur ses habitants’, Nouvelles Annales des Voyages, de la Géographie, de l’Histoire et de l’Archéologie, 4: 5–62. Pumphrey, M.E.C. 1937. ‘Shilluk “Royal” Language Conventions’, Sudan Notes and Records, 20(2): 319–21. ———. 1941. ‘The Shilluk Tribe’, Sudan Notes and Records, 24: 1–45. Quigley, Declan (ed.). 2005. The Character of Kingship. Oxford: Berg. Santandrea, Stefano, and Luigi de Georgi. 1965. ‘Morte Violenta per i “Re Divini” Scilluk E Dinka’, Africa: Rivista trimestrale di studi e documentazione dell’Istituto italiano per l’Africa e l’Oriente, 20(1): 15–32. Schnepel, Burkhard. 1988. ‘Shilluk Royal Ceremonies of Death and Installation’, Anthropos, 83(4–6): 433–52.

The Shilluk Reth  101 ———. 1990. ‘Shilluk Power Struggles and the Question of Succession’, Anthropos, 85(1–3): 105–124. ———. 1991. ‘Continuity Despite and through Death: Regicide and Royal Shrines among the Shilluk of Southern Sudan’, Africa: Journal of the International African Institute, 61(1): 40–70. Seligman, Charles, G. 1934. Egypt and Negro Africa: A Study in Divine Kingship. London: Routledge. Seligman, Charles, G., and Brenda Z. Seligman. 1932. Pagan Tribes of the Nilotic Sudan. ­London: Routledge. Simonse, Simon. 1992. Kings of Disaster: Dualism, Centralism and the Scapegoat King in Southeastern Sudan. Leiden: Brill. ———. 1994. ‘The Burst and the Cut Stomach: The Metabolism of Violence and Order in Nilotic Kingship’, Nilo-Ethiopian Studies, 2: 1–13. ———. 2005. ‘Tragedy, Ritual and Power in Nilotic Regicide: The Regicidal Dramas of the Eastern Nilotes’, in Declan Quigley (ed.), The Character of Kingship. Oxford: Berg, pp. 67–100. ———. 2017. Kings of Disaster, etc., revised and illustrated edn. Kampala: Fountain Publishers; also available as an e-book by Michigan State University Press. Stubbs, J.M. 1935. ‘Notes on the Beliefs and Customs of the Malwal Dinka of the Bahr-elGhazal Province’, Sudan Notes and Records, 17(2): 243–54. Tappi, Carlo. 1903. ‘Notes Ethnologiques sur les Chillouks’, Bulletin de la Société Khédiviale de Géographie, 111–41. Titherington, G.W. 1925. ‘Burial Alive among the Dinka of the Bahr-el-Ghazal Province’, Sudan Notes and Records, 8: 196–97. Westermann, Diedrich H. 1912. The Shilluk People, Their Language and Folklore. Berlin: Reimer.

Chapter 5

A New Approach to the Political Anthropology of Africa From African Political Systems and Tribes without Rulers via The Early State Petr Skalník Introduction: Classics The aim of this chapter1 is to suggest a new approach to the political anthropology of Africa, focused not on the past paradigms that concentrated on states, whether precolonial or modern colonial and postcolonial, but on the interplay of original African political forms such as chiefdoms and modern hybrid forms of politics. This cannot be made without acknowledging and assessing the achievements made by the classics of political anthropology. The pioneering role of African Political Systems (APS), Tribes without Rulers (TWR) and The Early State (TES) is undeniable. At the same time, the years after APS have seen several paradigm shifts. Certainly, one of them is the departure from the structural-functional type of analysis. The next has been the temporary popularity of Marxist or rather neo-Marxist precepts. Foucauldian discourse and subsequent postmodernist debates have influenced thinking in political anthropology as well. Not to mention the shift to neo-evolutionism in Russia and the United States among archaeologists and some anthropologists. However, what seems to be the most important shift is the turn towards the active role of Africans who have now emerged as critical researchers. The days when non-African Africanists were hegemons in knowledge production about Africa are gone. What I see as the most significant positive feature of APS is the fact that its two editors came up with an analysis of societies without political centralization, without titular rulers, and showed that they were also political. However, noncentralized societies took up only three of the book’s eight chapters. Chapters on

A New Approach to the Political Anthropology of Africa  103

the Nuer of Southern Sudan, the Logoli and Vugusu of Kavirondo in western Kenya, and the Tallensi in the Northern Territories of the Gold Coast/Ghana dealt with noncentralized societies; the other five dealt with politically centralized societies, even though the term ‘state’ was only used once, in Nadel’s chapter on the Kede ‘riverain state’. Other terms used were ‘kingdom’ (Zulu, Ankole) and ‘political system’ or ‘political organization’ of a particular tribe such as the Ngwato and the Bemba (Fortes and Evans-Pritchard 1940: passim). This may have been influenced by Radcliffe-Brown’s rejection of the ‘power of the State’ as something above the society of concrete actors (Radcliffe-Brown 1940: xxiii) and the official British position that the state and the king can only be located in London (cf. Quigley 2005). As a result of this policy, all other traditional leaders in the African parts of the British Empire could not be called kings, but only paramount chiefs. Another negative aspect of the volume, perhaps, was the latent evolutionist thinking that was evident in these early political anthropological essays. In other words, noncentralized politics of Group A were assumed to change into a centralized way of dealing with public affairs (Group B). What is remarkable in Radcliffe-Brown’s preface is that he emphasized the ‘organized exercise of coercive authority through the use, or the possibility of use, of physical force’ without mentioning that this was a prerogative of the state. He contrasted ‘organized states’ with ‘political systems’ in ‘simple societies’ in which law and war dominate and regulated vengeance stand between the two extremes (ibid.: xiv). The editors of APS, Meyer Fortes and E.E. Evans-Pritchard, were freer in their usage of ‘primitive states’, which they saw as synonymous with both chiefdoms and kingdoms (Fortes and Evans-Pritchard 1940: 5). At any rate, their typological exercise of dividing the societies discussed into Groups A and B was a rudimentary classification. They stated that societies that did not have central government and administration (denoted as ‘stateless’ in APS) would be able to ‘develop into states’ as a result of conquest. They mentioned the cases of the Zulu and the Banyankole. What they also emphasized was the role of the segmentary lineage system in the political systems within stateless societies. By contrast, primitive states had territorial administrative organizations. The political system of the Ngwato in today’s Botswana resembles the modern nation-state, in which political units are territorial, while ‘the plexus of kinship ties serves merely to cement those already established by the membership of the ward, district, and nation’ (ibid.: 6). The editors mentioned a third type, ‘very small societies’, in which ‘political relations are coterminous with kinship relations’ (ibid.: 6–7). Unfortunately, no case studies of this type were included in the volume. It is not clear whether they meant hunter-gatherer societies or some other face-to-face communities. Nevertheless, as a rule, stateless societies smaller in terms of population han the primitive states. The subsequent volume, TWR, appearing eighteen years after APS (Middleton and Tait 1958), included six case studies of segmentary politics. In this

104  Petr Skalník

way, the anthropologists made the number of rulerless cases of African politics even with those of centralized polities. The introduction by the editors heralds a more refined classification that is the result of the essays included in the volume. Dinka, Mandari, Amba, Lugbara, Tiv and Konkomba had, as far as they are analysed as ‘indigenous systems, unaffected by European contact’, no centralized political authority combined with unilineal descent groups. The editors define political relations as ‘those in which persons and groups exercise power or authority for the maintenance of social order within a territory’. Externally, it is uncontrolled power, while internally, cohesion is maintained by a legitimate authority (ibid.: 1). In uncentralized societies, there are also types where legitimate authority is not specialized and not always vested in corporate lineages, but rather in chiefs and headmen. Also, local groups obey holders of authority in age-sets and age-grades. Finally, village councils and associations are holders of political authority. However, in TWR, the authors deal with segmentary lineage systems. This is, of course, a limitation that has had repercussions for further research on African indigenous politics. The bias towards the segmentary lineage structures has meant, on the one hand, a refinement of studies of these structures, while, on the other hand, other social forms within the uncentralized category have tended to be less studied or simply have not entered the typologies constructed by the theorists. The ‘balance of power’ between and among segments and their clusters was the hallmark of the perspective on equilibriums of the structural functionalists. This did not exclude violence, especially feuds and warfare. However, these would as a rule result in a new equilibrium. Still, conflicts of this kind between uncentralized and centralized groups could result in the permanent victory of one over the other. Thus, in today’s northern Ghana, the chiefly centralized Dagbamba vanquished the uncentralized Konkomba, which led the latter to retreat into other uncentralized territories beyond the reach of the centralized chiefdom or to migrate into another chiefdom such as Nanuŋ (Skalník 1989; Talton 2010). In TWR, Buxton’s chapter on the Mandari brings in chiefship of the simplest type, which means that there are many tiny chiefships without any tendency for amalgamation. The Mandari can be considered an example of a segmentary structure of a chiefly kind that shows no signs of further centralization. Yet another, more isolated, example was the proposition to study a ‘segmentary state’ made by Southall, who worked among the Alur of Uganda (Southall 1956). I find this view significant because kingdoms and chiefdoms are located somewhere between leaders without hereditary titles on the one hand and complex nonmodern states with writing, standing armies, bureaucracy, etc. on the other.2

The State as a Teleology? Nevertheless, there was a fashion among historians of Africa, who in the 1950s and 1960s were mostly non-Africans (for exceptions, see for example Bâ and

A New Approach to the Political Anthropology of Africa  105

Daget (1955; Dike (1956)), to bring out evidence that Africans created their original statehood long before colonialism and modern independence. Later authors focusing on African statehood included Marian Małowist (1964), who generalized about the medieval states of Western Sudan, Murray Last (1967), who wrote on the Sokoto caliphate, Joseph Miller (1976), who equated kingdoms in Angola with early states, Michal Tymowski (2009), who loyally worked with ‘states’, and many more mentioned in Skalník (2004). Their work was in contrast to what most anthropologists had produced. One of the first Africanist anthropologists, Melville Herskovits (1938), called Dahomey ‘an ancient West African kingdom’ and many followed. Anthropologists were joined by anthropologizing historians, such as Jan Vansina, who continued writing about African kingdoms and chiefdoms but not states (Vansina 2001),3 or Edward Steinhart (1977), who published a book on conflict and collaboration in the kingdoms of western Uganda. Another line of research inspired by the publication of APS has been the juxtaposition of ‘traditional’ chiefs or chiefdoms and the ‘modern’ state without falling into the evolutionist trap (Busia 1951; Owusu 1970; Otite 1973). This line of research is closer to the approach I am favouring. The point is that the reconstruction of the African past is often very difficult and unless done by very refined oral tradition methods, elaborated namely by Vansina (1961), it can be closer to conjecture than provable facts. On the other hand, comparative study of the parallel existence of modern colonial and postcolonial states and encapsulated neotraditional chiefdoms is an authentic exercise that answers questions posed by really existing situations. Are the chances of democracy in Africa strengthened or weakened by chiefs and chieftaincy? Can modern states in Africa gain from the vetting of their exercise of power provided by chiefs and chiefdoms? Does African statehood, whether in the precolonial period or more recently, require monitoring by chiefs as representatives of the common people? Is the promotion of the state as the greatest achievement of history, whether in Africa or elsewhere in the world, the right path towards the self-management of societies? These and related questions can be answered by political anthropologists working on the relevant topics of our time. The celebration of the longevity of the state is a fairly recent phenomenon. It is a projection by nationalist elites and bandwagoning academia of the present day that harks back to the imaginaries of the past. In fact, except for ancient Rome or China during some specific dynasties, kingship was much more significant than something we would like to designate as the state. I say ‘would like’ because so much in our anthropological writing, and even more so in historiography, is concerned with imposing recently developed categories onto the past. Past conditions and events are thus ‘explained’ by an uncritical projection of the present onto the past, especially the distant past. Unfortunately, many African writers have jumped on the bandwagon of confirming the existence of statehood

106  Petr Skalník

and even empires in the distant past, slavishly following those non-Africans who, by searching for African equivalents of statehood on other continents, did a disservice to Africa because they suppressed the original qualities of chiefdoms and kingdoms. An example is the neo-Marxist study of Futa Djallon by Tirmiziou Diallo (1979). Another case is the Czech collection Social Stratification in Tribal Africa (Holý 1968). There the authors searched for the existence of social strata as synonyms of classes in precolonial tribal Africa that was described based on research carried out during the colonial period. ‘Tribal’ was any society analysed in the book (Rwanda, Zande, Agni, Herero and Bamileke), irrespective of the fact that one was described as a monarchic unitary state, the other a collectivity of states and the rest as displaying various types of chiefdoms. The book’s topic was social stratification regardless of the overall character of the society. It was a mixture of structural functionalism with latent Marxism. In 1973 in the Soviet Union, Olga Tomanovskaya published a short but challenging article on the ‘genesis of the state’ in Africa. Through a critique of APS, she called for a return to an evolutionary approach to the question of the origins of states in Africa. This she saw as a function of a ‘transitional’ period in which material inequality and class formation combined with the formation of political differentiation. She mentioned the striking difference of various African developments from European and Asian ‘socioeconomic formations’. She parted with the dogmatic Marxist schemas and calls for a detailed study of African chiefdoms and kingdoms whose stability she explained by traditions of primitive order and the sacred character of rulers. That is why Tomanovskaya concluded that ‘African kingdoms cannot be called states in the Marxist understanding of this term’ (Tomanovskaya 1973: 281). She lamented the lack of notional tools for the analysis of the preclass and early class processes. She rejected terms such as ‘state formation’ or ‘early state formation’. Then she proceeded to mention ‘barbarian kingdom’ in medieval studies and suggested that the term ‘kingdom’ should be conditionally accepted in African studies. Finally, she wondered when a community can be considered to be organized into a state-like formation or a ‘kingdom’ and remarked that the criteria were thus far absent. I previously wrote about such conflations and here only need to reiterate that the originality of African political systems lies not only in the fact that they were not bureaucratic clockwork systems, but also functioned according to the logic that ran parallel or against the logic of the state, as defined by Western political scientists (Skalník 1996). In effect, they were not real systems. In 1991, I published an article on ‘political systems’ for a French encyclopaedia of anthropology that was also published in Russian and English, in which the term ‘political system’ was submitted to scrutiny. The point was that the term ‘system’ indicated rigidity and boundedness instead of the reality of flexibility and change (Skalník 1991; cf. Skalník 2010: 104). Another problem with the term was that structural-functionalist analyses, until then dominating political anthropology,

A New Approach to the Political Anthropology of Africa  107

were latently evolutionist and did not allow for a more dynamic understanding of politics in both uncentralized (‘stateless’) and centralized polities (chiefdoms, kingdoms, states). Briefly, the rigidity of conceptual tools led scholars into the trap of viewing the state as a historical teleology, if not eschatology. If the state is inevitable, then a particular kind of state might also appear inevitable. Such a way of thinking is only another form of Eurocentrism and colonial dependence. The classical work refuting the inevitability of the state is, of course, Pierre Clastres’ Society against the State (1977). The author describes situations where the imported colonial state is resisted by chiefdoms and uncentralized societies because people already knew what the state meant for the life of the majority.

Chiefdoms, Early States and Carneiro’s Impact The Early State (TES) volume, edited by Claessen and Skalník and published twenty years after TWR (Claessen and Skalník 1978),4 presented twenty-one cases of centralized polities, divided into three types, of which the first type, that of the ‘inchoate’ state, was more or less synonymous with chiefdom. The lack of a theory of chiefdom both in the essays and editors’ introductory and synthetic chapters in that volume was quite significant. In fact, in TES, chiefdom was temporarily expunged from the theory of political anthropology. That happened despite the fact that Sahlins and Service had used the term since the early 1960s following Kalervo Oberg (also one of the contributors to APS). These neo-­evolutionists placed chiefdom as a stage between ‘segmented tribes’ and the state. Service wondered how ‘repressive states’ could suddenly appear out of ‘egalitarian primitive societies’. The answer to the riddle was placing chiefdom as a benign centralization lacking ‘forceful repression’ (Service 1975: 15–16). However, the question remains as to what is then the role of violence, including war, in the formation of both chiefdoms and states. In other words, does the neo-­evolutionist approach hold? For example, how many cases are known and well described where chiefdoms evolve into the state? Service himself introduced the case of the Cherokee Indians, who according to him knew both theocratic chiefdom and the state. But both political centralizations were the result of the contact, whether belligerent or peaceful, that the Cherokee had with the advancing English and French colonizers (ibid.: 140–48). Of the cases published in TES, eight were drawn from Africa (ancient Egypt, Axum, Jimma, Ankole, Kuba, Zande, Yoruba and Volta). Apart from Ancient Egypt and Axum, all six cases discussing tropical centralized polities were polities that lacked writing (Claessen and Skalník 1978). Some of them, such as the Zande and Volta, were clusters of political centralization that rather resembled a plurality of chiefdoms than states. The three types of the early states that were postulated – inchoate, typical and transitional – clearly suggested the neo-evolutionist character of the early state theory. The volume has been referred to by many anthropologists and other writers, but its neo-evolutionist slant has

108  Petr Skalník

not helped its popularity. Elsewhere I have discussed the causes of the rather reluctant acceptance of the concept (Skalník 2009). In 2004, I published an article called ‘Chiefdom: A Universal Political Formation?’, which critically examined the term ‘state’. It was an indirect response to the original Polish edition of Michal Tymowski’s book Państwa Afryki Przedkolonialnej [States of Precolonial Africa] (Tymowski 1999), which came out in English ten years later under the title The Origins and Structures of Political Institutions in Pre-colonial Black Africa (Tymowski 2009). This prominent historian, equally versed in European and African historiography, has been one of the belated promoters of the existence of (early) states in Africa (on his influence in Poland, see below). Although he was aware of the literature on chiefdoms, he rather uncritically presented the state in Africa compared to Europe and other continents. In the essay, I showed that in most of the literature, authors did not use the term ‘state’, but rather ‘kingdom’. The latter term along with ‘chiefdom’ was a much more widely used designation when discussing African cases of political centralization. Aside from that, I showed that chiefdoms were an original African contribution to governance and as such should be considered before conclusions about states and empires are drawn (Skalník 2004). Previous research by historians and some anthropologists (Vansina 1965; Last 1967; Izard 1985) demonstrated that there has never been an automatic development from one form of political centralization to another, especially from chiefdoms into early states and further into mature and modern states. Colonial conquest imposed a new type of state, derived from European models. Original African institutions such as chiefdoms were either vanquished and abolished or incorporated into these imported states in a subordinated nonsovereign position (i.e. indirect rule). When the colonial situation (Balandier 1955) entered the scene, African states (except for Ethiopia) disappeared, only to be ‘rediscovered’ by zealous writers such as Basil Davidson (1959). Even before him, the German linguist Diedrich Westermann published a special book on state formations in Africa south of the Sahara, which was inspired by earlier German conquest theories (Westermann 1952). One of the leading American neo-evolutionists, the late Robert Carneiro, first suggested the circumscription theory of the origin of the state (Carneiro 1970). In Carneiro’s own words: ‘[T]his theory came to be known as the circumscription theory since it pointed to the key role played by tight environmental constriction in giving rise to population pressure, which in turn had brought about recurring warfare, culminating, in certain areas, in the rise of the state’ (Carneiro 2012: 10). What was new in Carneiro was the dichotomic combination of the voluntaristic and the coercive approach. The voluntarist explanation of the origin of the state Carneiro relates to the late nineteenth-century sociologist Leslie Ward, who suggested that the state was the result of the ‘rational faculty’ of ‘a single brain or a few concerting minds’ (Carneiro 2012: 1, quoting Ward 1883: 224). The twentieth-century theorists, including Wittfogel, Fried,

A New Approach to the Political Anthropology of Africa  109

Claessen, ­Trigger and many others, argued that the state is the result of ideologies or ecological pressures alone, without recourse to coercion and violence. In the article, published forty-two years after the original paper in Science, Carneiro admitted that circumscription was a misleading name for his theory. He argued that ‘coercion – warfare, essentially – lies at the very heart of the process … Warfare is fuel – the propellant – that powers political evolution’. Carneiro asserted that he did not find a single case around the world of ‘historically or ethnographically known chiefdom or state having arisen without warfare having played a significant role at some stage of the process’ (ibid.: 14). ‘Chiefdom’ was proposed as a stage in political evolution by Service only in 1962 in Primitive Social Organization (1962). This was not widely accepted, as Morton Fried in his oft-cited The Evolution of Political Society (1967) introduced rank and stratified societies in between egalitarian societies and the state. Fried mentioned ‘chieftainship’ in various places of his book, but ignored Service’s suggestion about chiefdom as a stage. Instead, he discussed at length the usage of ‘tribe’ in the works of Sahlins and Service. TES, as already mentioned, avoids chiefdom as a stage. The collection Origins of the State, published in 1978, does not pay attention to chiefdoms even though Service was one of the two editors (Cohen and Service 1978). However, Carneiro in his contribution to that book discusses ‘competitive exclusion’ and the evolution of warfare (Carneiro 1978: 208–11) as a kind of prelude to his important text in which he designated chiefdom as a precursor to the state (Carneiro 1981). Here Carneiro defined chiefdom as ‘an autonomous political unit, comprising several villages or communities under the permanent control of a paramount chief ’ (Carneiro 1981: 45). But how did chiefdoms emerge? Carneiro argues that this happened by ‘intensive warfare’ (ibid.: 64). He recently corrected his original view about the steady expansion of chiefdoms, stressing that the growth of chiefdoms is also vitally connected to warfare, which ‘is the mechanism par excellence which enabled the chiefdom, and its successor, the state, to emerge’ (Carneiro 2012: 17). The most powerful and active from among the expanding chiefs – called pendragon by Carneiro – becomes the permanent, paramount chief. Although he refers mostly to American cases, his theory seems plausible for other continents, Africa included. This course of events does not disqualify demographic or environmental circumscription, but makes it more dependent on the human factor, not least economic motivation like taking agricultural land of the defeated (ibid.: 18–21). Carneiro proposed the revision of his circumscription theory as follows: A heightened incidence of conquest warfare, due largely to increase in population pressure, gave rise to the formation of successively larger political units, with autonomous villages being followed by chiefdoms, the process culminating in certain areas with the emergence of the state. (Ibid.: 27)

110  Petr Skalník

Carneiro’s 2012 paper was followed by responses by many specialists who agreed that war was important, but was not always a decisive factor. Grinin and Korotayev, the editors of the special issue of Social Evolution and History, stress the evolutionary multilinearity in political processes, but seem to disagree with Claessen’s scepticism as to the role of wars. The point is that only abrupt changes in evolution produce novelties such as chiefdoms or states. And war and the threat of war are prominent producers of such changes (Grinin and Korotayev 2012: 199). In a recent contribution, I follow Carneiro on African data (Skalník 2018).

Russian Neo-evolutionism and Its International Network This brings me to the concerted effort by several Russian archaeologists, ethnologists and historians, who though bred on Marxist historical materialism were decisively inspired by the creative revisionism of Lev Kubbel’s ‘potestal-political ethnography’5 (1988). Their advent symbolically coincided with the demise of the Soviet Union and the Soviet brand of Marxism. Trying to strike a creative neo-evolutionist note, this loose group has published a string of collective volumes over the years (Korotayev and Chubarov 1991; Popov 1993b and 1995; Kradin and Lynsha 1995; Bondarenko and Korotayev 2000; Kradin et al. 2000; Grinin et al. 2004; Popov 2012 and 2013; Carneiro et al. 2017) and in 2002 launched a biannual international journal Social Evolution and History. They found and described alternatives to the early state in various parts of the world, Africa included. Kradin’s summarizing text on chiefdom mentions Kubbel’s ‘potestal power’ criteria which distinguishes between (1) priestly-theocratic, (2) protobureaucratic-aristocratic, (3) military and (4) plutocratic chiefdoms. These subtypes are not exclusive; one can change into another (Kradin 1995: 23). The year 2004 saw the appearance of the collective volume The Early State, Its Alternatives and Analogues (Grinin et al. 2004). This volume contained ­twenty-two chapters, of which twelve were written by Russian authors. This was significant as the idea of alternatives and analogues to the early state stemmed from Russian research reacting to the publication of The Early State and several other volumes, edited primarily by Henri Claessen. In this volume, the Africanist Dmitri Bondarenko wrote on a megacommunity in precolonial Benin, on chiefdoms in precolonial Benin and on hierarchical alternatives in Benin of the second millennium BCE. Bondarenko shows that Benin was a megacommunity without bureaucracy, even though he continues to call Benin a kingdom under the leadership of the Oba. However, he concludes that ‘the Benin megacommunity was not a state … [but rather] a specific type of the complex hierarchical socio-political organization … alternative to the statehood’. At the same time, he asserts that ‘Benin was not less developed than the majority of early states’ (Bondarenko 2004: 353).

A New Approach to the Political Anthropology of Africa  111

In contrast to Bondarenko, Eleonora L’vova discusses personal dependence and the state, and the formation and development of states in the Congo Basin. She writes about the Kuba, Luba and Lunda as states whose first rulers were ‘war chiefs’. These states were ‘based on territorial conquests and their aims were war robbery and a tribute’ (L’vova 2004: 288). The conquered chiefs had to recognize the supreme chiefs titled as Mulohwe, Nyimi and Mwata Yamvo. L’vova never uses terms like ‘king’ or ‘kingdom’. Carneiro actually supports his argument on the role of war in state formation by reference to L’vova’s (2004: 288) finding that the first rulers in the Congo Basin were war chiefs. This stands in contrast to Vansina’s thesis of the origin of Sub-Saharan states from ‘conceptions in the mind’ (Carneiro 2012: 16). Leonid Grinin profiled himself as the leading theorist who looked critically at the concept of the early state. The Volgograd-based scholar became the spiritus movens of the journal Social Evolution and History (SEH) that soon became a representative forum for all scholars interested in human social evolution, including political relations. SEH brought to light the most important texts in the field of neo-evolutionist anthropology and history. The journal carried several special issues: besides the above-mentioned volume that also reproduces several articles from SEH, there was a special issue on thirty years of early state research (vol. 7, no. 1, guest-edited by H.J.M. Claessen, R. Hagesteijn and P. van de Velde in 2008), on chiefdoms (vol. 10, no. 1, edited by L.E. Grinin, A.V. Korotayev, and D.M. Bondarenko in 2011) and a special section on the early state theory in anthropological theory (vol. 9, no. 1, guest-edited by P. Skalník in 2009). Other single issues carry articles on these clusters of problems of theory and practice. The 10th Anniversary Issue was particularly notable, with a leading article by Carneiro (vol. 11, no. 2), followed by a large number of comments by leading American, Russian, Dutch (Claessen included), Chinese, Italian, French and Austrian specialists. Henri Claessen, the main protagonist of early state research, has also published his variations on The Early State theme in SEH (Claessen 2002, 2005, 2008, 2010, 2011, 2012, 2014 and 2016). The substance of his contributions can be summarized as a painstaking search for lacunae in the theory and new topics that emerge as the theory moves forward. Let us return to Leonid Grinin, who does not write specifically on Africa, but generalizes on various aspects of the evolutionary sequence of statehood. In one of his numerous texts (Grinin 2008), I noticed the potential for further theorizing on what some authors, such as Krohn-Hansen and Nustad (2005), call ‘state formation’. Although Grinin still seems to be attracted by typological exercises so characteristic of Marxist historical materialism, he admits that ‘evolution’ is multiform and multidirectional. He accepts the early state as a ‘separate stage of evolution’, but finds the quite tentative category of ‘mature state’ as inadequately analysed. In TES, Claessen and Skalník (1978) did not focus on mature states. Grinin comes with an intermediate category of ‘developed state’, distinguished

112  Petr Skalník

by the existence of preindustrial bureaucratic centralization. These states would exist from Late Antiquity up to the Industrial Era, at which point truly mature modern industrial states took over. The value of this revision is not in positing another stage and dividing it into three substages à la Morgan or Engels, but in that we begin to see that state formation is a continuous process that had not been captured by theory. We have to overcome the typical anthropological obsession with archaic and perhaps also ‘medieval’ states and join forces with other social sciences in forging a new theory of the state that includes its ‘mature’ or ‘national’ forms. If, however, we consider chiefdoms on the one hand and contemporary forms ‘beside the state’ on the other (Bellagamba and Klute 2008), then the state is only one element, however crucial and multiform, in the whole spectrum of political centralization theory. Political anthropology, by its focus, then appears as well equipped to create an overall theory that would not hide behind the cloak of (over)specialization, but instead would generalize the knowledge produced by anthropologists, historians, sociologists and political scientists.

Towards a Paradigmatic Change in Political Anthropology: The Polish Volumes Polish archaeologists and ethnologists have turned their interest to political anthropology and especially the questions of early states through the influence of the historian Michał Tymowski, who has been drawn to the international conferences and volumes on the early state topics convened by Henri Claessen and his colleagues since the late 1970s (starting with the 10th International Congress of Anthropological and Ethnological Sciences, held in New Delhi; see Claessen and Skalník (1981)) and has published mostly on African themes in both Polish and English ever since. Already in 1987, he organized a national conference in Warsaw that echoed early state research. The title was still carefully formulated, due to Marxist orthodoxy, The Genesis and Functioning of Early Forms of Statehood in a Comparative Perspective (Tymowski and Ziółkowski 1992). Tymowski in his two-page preface acknowledges the influence of political anthropology as a ‘recently developed branch of knowledge’ that prompted intensive research on early forms of statehood. This research is comparative in its nature. He asserted that ‘in Polish science’ the study of the beginnings of statehood had been carried out for a long time, but by different disciplines that differed methodologically and organizationally. Thus, the origin of the Polish state was studied by archaeologists, historians, state and law specialists and linguists. Polish scholars also studied ancient Egypt, Mesopotamia and Greece, but also pre-Columbian America and precolonial Africa. The 1992 book (which consisted of 350 copies printed on a mimeograph) introduces two essays on Africa: L. Adamowicz discusses Great Zimbabwe as the rise and fall of the first Mashona state, while M. Tymowski devotes his paper to the theoretical problems of precolonial African states.

A New Approach to the Political Anthropology of Africa  113

Tymowski also wrote the closing chapter, in which he summarized the comparative study of early forms of statehood. In it, he fully accepts ‘the early state’, but cautions that the majority of scholars, especially in the science of state and law, understand the state as a historical phenomenon that emerges and withers. However, he admits that to determine the boundary where the political system becomes an early state system is a difficult task, as it is a prolonged process. Tymowski reminds his readers that Marxist research identifies the formation of social classes as the search for a boundary. But he refers to non-European cases that do not corroborate the economic roots of statehood. In the West, according to Tymowski, Claessen and Skalník speak about ‘emergent classes’, while Soviet scholarship theorizes a transitory phase between the pre-state or tribal level and the state level. Tymowski believes that this transitory phase is identical with the early state. As to class formation, it is part of the social stratification processes, but there is no priority among the economic, political and ideological differentiation of societies. He further characterizes the 200 years of research on the Polish state that was comparative at first only within the Slav supraethnic community, but then also further comparisons were extended to early Middle Ages in Europe, which included not only Germanic states but also polities created in Europe by nomads arriving from Asia, i.e. Avars, Bulgars, Khazars, Pechenegs and Magyars (Gieysztor and Manteuffel 1968). However, the 1992 volume was the first in Poland to undertake a comparison of European early states with non-European ones. Tymowski in his summarizing chapter then lists various aspects of early states as presented by the authors of the chapters. What seems characteristic for the Polish approach is the identification of the tribe with a chiefdom. Several contributors, Tymowski stresses, pinpoint warfare and especially booty as decisive factors. Tymowski mentions one author, Tyszkiewicz, who when discussing the boundaries of ‘early feudal’ Poland argues that ‘early states were usually military, aggressive organizations that conflicted with their neighbours’. Another chapter writer, Wielowiejski, asserts that weaker neighbours were either annexed by the aggressor or their territory ravaged (Tymowski in Tymowski and Ziólkowski 1992: 278). A specific African early state, Great Zimbabwe, is subjected to a profound archaeological analysis by Leonardo Adamowicz (1992). Nevertheless, the author cannot give any conclusive evidence about the existence of the state, except for some data on interregional trade and gold extraction. The voluminous collection based on a conference held in Poznań in 2012 (Banaszkiewicz et al. 2013) contained two chapters on Africa. One was on the emergence of the state of the pharaohs in Egypt and the other was on succession and the legitimation of power in chiefdoms of northern Cameroon. The latter, written by Ryszard Vorbrich, compared succession in four polities of the Kirdi people. Wandala (Mandara) he classifies as a state, although little is known about how it functions. Gudur, situated to the south of Wandala, according to Vorbrich, who relied here on the field research by Catherine Jouaux (1988, 1991), is

114  Petr Skalník

a chiefdom founded by a group of immigrants led by Biya, possessor of the holy stone – mahura nga kuley – giving him control of rain and other forces of nature, and ensuring the prosperity of his realm. Biya was in effect a chief priest called bay. Bay’s authority was limited by the lords of the Earth – masa hwayak – who practised agrarian cults, but also served as kingmakers, whose consent was essential for succession to happen. Gudur appears as a kind of ‘seed’ chiefdom for several smaller chiefdoms, the most important of which is Guiziga-Muturua. This chiefdom was founded at the beginning of the nineteenth century by an escapee from Gudur named Bildengeur, who stole a holy object from Gudur – an instrument of sacred rule. The local inhabitants, the Guiziga, were subdued and the new kinship group assumed the name ‘Muturua’. The autochthons were, among other things, assigned the role of funerary agents of the chief ’s family. This chiefdom, however, was weakened by Muslim Fulbe lamidates and its chiefly traditions moved further south from Muturua, into the chiefdom of Daba-Musgoy. The ruling dynasty was from the Guiziga, while the local people subjected to it are the Daba, who remained the custodians of the Earth cult (Vorbrich 1913: 98–99). The chiefdom Daba-Musgoy eventually created a pact with Modibo Adama, the founder of the Yola Fulbe emirate. The Fulbe would not force Daba-Musgoy to convert to Islam in exchange for an annual supply of slaves. However, the chiefdom started to copy the lamidate, adopted slavery and politically became a ‘calque’ of the lamidate, and eventually many inhabitants of the chiefdom converted to Islam. The ruling elite was ‘fulbeized’ and managed to legitimize itself even vis-à-vis the German and French colonial and also the postcolonial state of Cameroon. Thus, the chiefdom Daba-Musgoy still exists today, while Gudur and Muturua disappeared from the political map of Cameroon (ibid.: 100–1). This case shows how chiefdoms move not only on the map but also over time through their ability to adjust to the more powerful state structures. The Poznań volume also contains a theoretical chapter by Michał Tymowski, who discussed the problems of the application of the early state theory on precolonial states in Africa (in a way, the chapter relates to a similar chapter in Tymowski and Ziółkowski (1992: 253–66)). Tymowski mentioned five reasons for early state theory, namely: (1) universality; (2) atemporality; (3) comparative methodology; (4) interdisciplinarity; and (5) a balance between case studies and generalizing theory. Early state theory was confronted with neo-evolutionism, and multilinear, general and specific evolution. The automatism of evolution was rejected and the transformation of nonstate organizations into early state organizations as well as mature state organizations a possibility, not an unescapable unavoidability. Another confrontation was with Marxist theory, especially the empirical fact that early states emerged before or without the formation of social classes and that they were not a variant of feudalism. Clientelism was too broad and applicable to various societies. According to Tymowski, the theory of the early state was also confronted with ‘state and law theory’, which places the

A New Approach to the Political Anthropology of Africa  115

emergence of states exclusively in Europe in the late Middle Ages and the early modern period. According to this ‘idealistic’ conceptualization of the state, there were no states in early medieval Europe or outside of Europe before the colonial conquest (Tymowski 2013: 76–78). The author further discusses the precolonial states in ‘Black’ Africa (i.e. south of the Sahara). He laments that there was no special treatment of the problem of violence in the literature on early state formation. Tymowski himself paid attention to the problem of violence when he studied the nineteenth-century states of Samori and Kenedougou. But regarding state origins, the late Jan Vansina already wrote back in 1962 that the kingdoms in Africa were enlarged through conquests by groups of immigrants. Tymowski asserts that there are no documented cases of contract that would assign power to the chief (Tymowski 2013: 79). Here I have to note that in the West African savanna, there were quite a few cases of ‘ritual-political equilibrium’ or pacts (see e.g. Fortes 1945: 108) and conquest stood in the background as a threat. Tymowski further discusses the question of territory and boundaries, stressing that boundaries are less important than the centre. This means that the ability of the early state to control diminishes the greater the distance from the centre. The allegiance of the chiefs from the periphery is measured by the regularity and quantity of sent gifts, tributes, supplies of armed men, etc. to the centre. If these stop, it may mean a change of allegiance to another early state and subsequently a military intervention from the losing state. Tymowski also mentions the prolonged coexistence of states and stateless polities in precolonial Africa. He mentions the suggestion of Skalník that chiefdoms were more universal than early states, while the latter were an exceptional type of political organization. Tymowski places Southall’s segmentary state hypothesis in parallel with tribal confederacies or tribal unions of medieval Europe and Grinin’s alternative types of political organization. While Germanic and Slav tribal unions disappeared as a result of European state-forming processes, in Africa the ‘segmentary states’ lasted longer (Tymowski 2013: 80–82). Tymowski also admits that in research on precolonial Africa, the term ‘empire’ is used ‘very often’, even more often than in other continents. Perhaps to soften this improbability (actually impossibility – cf. Skalník 2004: 80), Tymowski proposes calling large multiethnic four-layer polities ‘early empires’ or ‘early imperial formation’. These were fairly fragile vis-à-vis external and internal pressures of segmentation (cf. Tymowski 2011). Finally, Tymowski stresses the fact that in Africa, the co-existence of stateless tribal communities and chiefdoms with the early state structures allowed for oscillation between these political forms. Besides, it is also unclear whether one should describe some contemporary fragile or failed states in Africa with the help of the early state theory (Tymowski 2013: 84–85).

116  Petr Skalník

Conclusion The International Union of Anthropological and Ethnological Sciences (IUAES), a global forum of anthropologists and ethnologists, has also paid attention to early states, state origins, etc. One of the activities organized by its Commission on Anthropological Theory (COTA) was the panel ‘Towards a Universal Paradigm in Political Anthropology’, which took place during the 16th World Congress of Anthropological and Ethnological Sciences held in Manchester in 2013. There, the Russian Africanist Vladimir Popov presented a paper on the correlation between political and ethnogenetic processes. He examined what he calls the modus of potestarity by pointing out that the ‘great ethnoses’ of contemporary Africa such as the Ashanti, Malinke, Yoruba, Baganda, Bakuba, Banyarwanda or Zulu emerged as a result of potestal-political processes of the precolonial period. In other words, these ethnonyms are derived from politonyms, meaning that the name of the polity gave name to the newly-formed ethnic group of that polity. Ethnicity is thus a result of the modus of potestarity. Much more ambitious was another COTA event, namely the ‘World Congress on State Origins and Related Subjects’, which took place in Wigry, Poland in September 2014. Several papers presented there were devoted to African state formation. A selection of revised papers from this major event is currently under preparation for publication. Because chiefdoms and kingdoms co-existed with modern postcolonial states in Africa and other parts of the world such as Oceania, political anthropology began to study the complex relations characterizing the system and process of politics in countries such as Ghana, Nigeria, Benin, South Africa, Cameroon, Uganda, Burkina Faso, Niger, Mali, Côte d’Ivoire and others. Political culture appears to be an especially decisive factor for the character of the oscillation between stability and instability, processes leading towards democracy or away from it. By including data from African political systems, the entire body of political theory concerning the state and its alternatives, analogues are being redefined (cf. Drulák and Moravcová 2013). Africa offers original solutions to the queries concerning optimal governance for today and tomorrow. I believe that it is fair to say that the future of political anthropology does not lie in an antiquarian interest, but in trying to come to grips with the parallelism of modern forms of the state and various nonstate polities, be it chiefdoms or acephalous communities, but also present-day terrorist, ultranationalist, utopian and extremist religious political forms that may or may not call themselves states (Chabal et al. 2017). This novel approach is closely connected with the origin of states and state-formation processes that are currently taking place in the world, but also the attempts of various traditionally noncentralized groups to acquire the attributes of chiefdoms or states. The theory of states will only be good if it overcomes the limitations of the hierarchization characteristic of Eurocentrist approaches and the limits of historical categorization that almost forbids comparison over time and space.

A New Approach to the Political Anthropology of Africa  117

Petr (Peter) Skalník is a political anthropologist who specializes in the state and chiefdom studies. He was educated in Prague, Leningrad and Cape Town, and has taught African studies, social anthropology and political science at universities in Czechoslovakia and the Czech Republic, the Netherlands, South Africa, Switzerland, Poland, Lithuania and France. He has edited and co-edited two dozen books, among them The Early State (Mouton, 1978), The Early Writings of Bronislaw Malinowski (Cambridge, 1993), Postsocialist Europe: Anthropological Perspectives from Home (Berghahn Books, 2009) and Africa: Power and Powerlessness (LIT Verlag, 2011). He was ambassador of Czechoslovakia/Czech Republic to Lebanon and Vice-President of the IUAES (2003–13), and is Chevalier dans l’Ordre des Palmes Académiques (2006).

Notes 1. The author wishes to thank Han Vermeulen for his selfless critical reading. His editorial experience enabled the revision and expansion of the original text. However, responsibility for the views expressed rests with the author. 2. It should be noted that political scientists and theorists of state and law generally do not speak of states before the fifteenth century (i.e. Machiavelli). This may sound preposterous and even dogmatic to an anthropological ear, but all forms that existed previously in the category of political centralization have not been considered machinery identified with the state as we know it. 3. Characteristically, Vansina’s chapter in The Early State volume was called ‘The Kuba State’, thus avoiding both the terms ‘kingdom’ and ‘early state’ (Vansina 1978). This was caused by the stipulation of the editors who insisted on using the state concept. 4. Under the leadership of Henri Claessen, five other collections followed. 5. ‘Potestal’ is derived from Latin potestas, which means ‘power’, ‘might’ and ‘faculty’ depending on the historical and social context. Kubbel’ (1929–88) coined the term ‘potestal-­ political ethnography’ in order to distinguish his creative Marxist scholarship from Western political anthropology (cf. Popov 1993: i). ‘Potestal’ in Kubbel’s understanding was meant to clearly distinguish pre-state politics from state politics. But a problem arises in this context with the early state, as this form of polity straddles pre-state and state politics.

References Adamowicz, L. 1992. ‘Wielkie Zimbabwe. U źródeł powstania i upadku pierwszego państwa Maszonów [‘Great Zimbabwe: Sources of Emergence and Decline of the First State of the Mashona]’, in M. Tymowski and M. Ziółkowski (eds), Geneza i funkcjonowanie wczesnych form państwowości na tle porównawczym [Genesis and Functioning of Early Forms of Statehood in Comparative Perspective]. Warsaw: Wydawnictwa Uniwersytetu Warszawskiego, pp. 215–51. Bâ, A.H., and J. Daget. 1955. L’Empire peul du Macina, vol. 1 (1818/1853). Dakar: IFAN. Balandier, G. 1955. Sociologie actuelle de l’Afrique Noire. Paris: PUF. Banaszkiewicz, J., M. Kara and H. Mamzer (eds). 2013. Instytucja ‘wczesnego państwa’ w perspektywie wielości i różnorodności kultur [Institution of the ‘Early State’ in the Perspective of Quantity and Variety of Cultures]. Poznań: Instytut archeologii i etnologii PAN.

118  Petr Skalník Bellagamba, A., and G. Klute (eds) 2008. Beside the State: Emergent Powers in Contemporary Africa. Cologne: Rüdiger Köppe Verlag. Bondarenko, D.M., 2004. ‘From Local Communities to Megacommunity: Biniland in the 1st Millennium B.C. – 19th Century A.D.’, in L.E. Grinin et al. (eds), The Early State, Its Alternatives and Analogues. Volgograd: Uchitel, pp. 325–63. Bondarenko, D.M., and A.V. Korotayev (eds). 2000. Civilizational Models of Politogenesis. Moscow: Center for Civilizational and Regional Studies, Russian Academy of Sciences. Busia, K.A. 1951. The Position of the Chief in the Modern Political System of Ashanti: A Study of the Influence of Contemporary Social Changes on Ashanti Political Institutions. London: Oxford University Press. Carneiro, R.L. 1970. ‘A Theory of the Origin of the State’, Science, 169: 733–38. ———. 1978. ‚Political Expansion as an Expression of the Principle of Competitive Exclusion’, in R. Cohen and E.R. Service (eds), Origins of the State: The Anthropology of Political Evolution. Philadelphia: ISHI, pp. 205–23. ———. 1981. ‘The Chiefdom: Precursor of the State’, in G.D. Jones and R.R. Kautz (eds), The Transition to Statehood in the New World. Cambridge: Cambridge University Press, pp. 37–97. ———. 2002. ‘Was the Chiefdom a Congellation of Ideas?’, Social Evolution and History, 1(1): 80–100. ———. 2012. ‘The Circumscription Theory: A Clarification, Amplification and Reformulation’, Social Evolution and History, 11(2): 5–31. Carneiro, R.L., A.V. Korotayev and L.E. Grinin (eds). 2017. Chiefdoms: Yesterday and Today. Clinton Corners, NY: Eliot Werner Publications. Chabal, P., G. Feinman and P. Skalník. 2017. ‘Beyond States and Empires: Chiefdoms and Informal Politics Fifteen Years Later’, in A.V. Korotayev, L.E. Grinin, R.L. Carneiro (eds), Chiefdoms: Yesterday and Today. Clinton Corners, NY: Eliot Werner Publications, pp. 309–24. Claessen, H.J.M. 2002. ‘Was the State Inevitable?’, Social Evolution and History, 1(1): 101–17. ———. 2005. ‘Early State Intricacies’, Social Evolution and History 4(2): 251–58. ———. 2008. ‘Before The Early State and after: An Introduction’, Social Evolution and History, 7(1): 4–18. ———. 2010. ‘On Early States: Structure, Development, and Fall’, Social Evolution and ­History, 9(1): 1–51. ———. 2011. ‘On Chiefs and Chiefdoms’, Social Evolution and History, 10(1): 5–26. ———. 2012. ‘Reconsideration of a Reformulation’, Social Evolution and History, 11(2): 36–43. ———. 2014. ‘From Incidental Leader to Paramount Chief ’, Social Evolution and History, 13(1): 1–41. ———. 2016. ‘The Emergence of Pristine States’, Social Evolution and History, 15(1): 3–57. Claessen, H.J.M., R.R. Hagesteijn and P. van de Velde. 2008. ‘The Early State Today’, Social Evolution and History, 7(1): 245–65. Claessen, H.J.M., and P. Skalník. 1978. The Early State. The Hague: Mouton Publishers. ———. 1981. The Study of the State. The Hague: Mouton Publishers. Claessen, H.J.M., and P. van de Velde. 1987. Early State Dynamics. Leiden: Brill. ———. 1991. Early State Economics. New Brunswick, NJ: Transaction Publishers. Clastres, P. 1977 [1974]. Society against the State. New York: Urizen Books. Cohen, R., and E.R. Service (eds). 1968. Origins of the State: The Anthropology of Political Evolution. Philadelphia: ISHI. Davidson, B. 1959. Old Africa Rediscovered. London: Victor Gollancz.

A New Approach to the Political Anthropology of Africa  119 Diallo, T. 1979. ‘Der theokratische Fulbe-Staat “Fuuta-Jaloo”. Beitrag zur Theorie der Staatenbildung im Westafrika des 17. bis 19. Jahrhunderts’, Ph.D. dissertation. Berlin: Freie Universität Berlin. Dike, K.O. 1956. Trade and Politics in the Niger Delta 1830–1890: An Introduction to the Economic and Political History of Niger Delta. Oxford: Clarendon Press. Drulák, P., and Š. Moravcová (eds) 2013. Non-Western Reflection on Politics. Frankfurt: Peter Lang. Fortes, M. 1945. The Dynamics of Clanship among the Tallensi. London: Oxford University Press. Fortes, M., and E.E. Evans-Pritchard (eds). 1940. African Political Systems. London: Oxford University Press. Fried, M.H. 1967. The Evolution of Political Society: An Essay in Political Anthropology. New York: Random House. Gieysztor, T., and T. Manteuffel (eds). 1968. L’Europe aux IXe–XIe siècles. Aux origines des Etats nationaux. Warsaw: Państwowe wydawnictwo naukowe. Grinin, L.E. 2008. ‘Early State, Developed State, Mature State: The Statehood Evolutionary Sequence’, Social Evolution and History, 7(1): 67–81. Grinin, L.E., and A.V. Korotayev. 2012. ‘Emergence of Chiefdoms and States: A Spectrum of Opinions’, Social Evolution and History, 11(2): 191–204. Grinin, L.E. et al. (eds). 2004. The Early State, Its Alternatives and Analogues. Volgograd: Uchitel. Herskovits, M. 1938. Dahomey: An Ancient West African Kingdom, 2 vols. New York: J.J. Augustin. Holý, L. (ed.). 1968. Social Stratification in Tribal Africa. Prague: Academia. Izard, M. 1985. Gens du pouvoir. Gens de la terre. Les institutions politiques de l’ancien royaume du Yatenga (Bassin de la Volta Blanche). Cambridge: Cambridge University Press. Jouaux, C. 1988. ‘Gudur: chefferie ou royaume?’, Cahiers d’Etudes Africaines, 29(114): 259–88. ———. 1991. ‘La chefferie du Gudur et sa politique expansionniste’, in J. Boutrais (ed.), Du politique à l’economique: etudes historiques dans le bassin du lac Tchad. (Actes du IVe colloque Mega/Tchad, Paris, du 14 au 16 septembre 1988). Paris: Orstom, pp. 193–224. Korotayev, A.V., and V.V. Chubarov (eds). 1991. Arkhaicheskoe obshchestvo. Uzlovye problémy sotsiologii razvitiya razvitiya [Archaic Society: Central Problems of the Sociology of Development]. Moscow: Institut istorii AN SSSR. Kradin, N.N. 1995. ‘Vozhdestvo: sovremennoe sostoyanie i problémy’ [‘Chiefdom: Contemporary State and Problems’], in V.A. Popov, (ed.), Rannie formy politicheskoy organizatsii: ot pervobytnosti k gosudarstvennosti [Early Forms of Political Organization: From Primaeval State to Statehood]. Moscow: Vostochnaya literatura, pp. 11–61. Kradin, N.N., and V.A. Lynsha (eds). 1995. Alternative Pathways to Early State. Vladivostok: FEB RAS. Kradin, N.N. et al. (eds). 2000. Alternatives of Social Evolution. Vladivostok: FEB RAS. Krohn-Hansen, C., and K.G. Nustad (eds). 2005. State Formation: Anthropological Perspectives. London: Pluto Press. Kubbel’, L.E. 1988. Ocherki potestarno-politicheskoy etnografii. [Essays on Potestal-Political Ethnography]. Moscow: Nauka. Last, M. 1967. The Sokoto Caliphate. London: Longman. L’vova, E.S. 2004. ‘Formation and Development of States in the Congo Basin’, in L.E. Grinin, R.L. Carneiro, D.M. Bondarenko, N.N. Kradin and A.V. Korotayev (eds), The Early State, Its Alternatives and Analogues. Volgograd: Uchitel, pp. 288–297.

120  Petr Skalník Małowist, M. 1964. Wielkie państwa Sudanu Zachodniego w późnym średniowieczu [Great States of the Western Sudan in Late Middle Ages]. Warsaw: PWN. Middleton, J., and D. Tait (eds). 1958. Tribes without Rulers: Studies in African Segmentary Systems. London: Routledge & Kegan Paul. Miller, J.C. 1976. Kings and Kinsmen: Early Mbundu States in. Angola. Oxford: Clarendon Press. Otite, O. 1973. Autonomy and Dependence: The Urhobo Kingdom of Okpe in Modern Nigeria. London: Hurst. Owusu, M. 1970. Uses and Abuses of Political Power. London: Clarendon Press. Popov, V.A. 1993. ‘Predislovie’ [‘Foreword’], in V.A. Popov (ed.), Rannie formy sotsial’noy stratifikatsii. Moscow: Vostochnaya literatura. ———. (ed.). 1993. Rannie formy sotsial’noy stratifikatsii [Early Forms of Social Stratification]. Moscow: Vostochnaya literatura. ———. (ed.). 1995. Rannie formy politicheskoy organizatsii: ot pervobytnosti k gosudarstvennosti. [Early Forms of Political Organization: From Primaeval State to Statehood]. Moscow: Vostochnaya literatura. ———. (ed.). 2012. Rannie formy politicheskikh system [Early Forms of Political Systems]. St Petersburg: MAE RAN. ———. (ed.). 2013. Rannie formy potestarnykh sistem [Early Forms of Potestal Systems]. St Petersburg: MAE RAN. Quigley, D. (ed.). 2005. ‘Introduction: The Character of Kingship’, in D. Quigley (ed.), The Character of Kingship. Oxford: Berg, pp. 1–23. Radcliffe-Brown, A.R. 1940. ‘Preface’, in M. Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press for the International African Institute, pp. xi–xxiii. Service, E. 1962. Primitive Social Organization: An Evolutionary Perspective. New York: ­Random House. ———. 1975. Origins of the State and Civilization: The Process of Cultural Evolution. New York: Norton. Skalník, P. 1989. ‘Outwitting Ghana: Pluralism of Political Culture in Nanun’, in P. Skalník (ed.), Outwitting the State. New Brunswick, NJ: Transaction Publishers, pp. 145–68. ———. 1991. ‘Système politique’, in P. Bonte and M. Izard (eds), Dictionnaire de l’ethnologie et de l’anthropologie. Paris: Presses Universitaires de France, pp. 583–85. ———. 1996. ‘Authority versus Power: Democracy in Africa Must Include Original African Institutions’, Journal of Legal Pluralism and Unofficial Law, 37–38: 109–21. ———. 2004. ‘Chiefdom: A Universal Political Formation?’, Focaal-European Journal of ­Anthropology, 43: 76–98. ———. 2009. ‘Early State Concept in Anthropological Theory’, Social Evolution and History, 8(1): 5–24. ———. 2010. ‘Political System’, Zeszyty Etnologii Wroclawskiej 1–2(12–13): 101–6. ———. 2018. ‘War as a Decisive Factor in State Formation: War and Peace in Africa – Local Conflict and the Weak State’, in T. Hüsken, A. Solyga and D. Badi (eds), The Multiplicity of Orders and Practices: A Tribute to Georg Klute. Cologne: Rüdiger Köppe Verlag, pp. 155–65. Southall, A.W. 1956. Alur Society: A Study in Processes and Types of Domination. Cambridge: Heffer. Steinhart, E.I. 1977. Conflict and Collaboration: The Kingdoms of Western Uganda, 1890–1907. Princeton: Princeton University Press. Talton, B. 2010. Politics of Social Change in Ghana. Basingstoke: Palgrave Macmillan.

A New Approach to the Political Anthropology of Africa  121 Tomanovskaya, O.S. 1973. ‘Izuchenie problemy genezisa gosudarstva na afrikanskom materiale’ [‘Studying the Problem of the Genesis of the State on African Data’], in Osnovnye problemy afrikanistiki. Etnografia, istoria, filologia. Moscow: Nauka, pp. 273–83. Tymowski, M. 1999. Państwa Afryki Przedkolonialnej [States of Precolonial Africa]. Wroclaw: Fundacja na rzecz nauki polskiej. ———. 2009. The Origins and Structures of Political Institutions in Precolonial Black Africa. Lampeter: Edwin Mellen Press. ———. 2011. ‘Early Imperial Formations in Africa and the Segmentation of Power’, in P.F. Bang and C.A. Bayley (eds), Tributary Empires in Global History. Cambridge: Cambridge University Press, pp. 108–19. ———. 2013. ‘Kilka problemów wynikających z zastosowania teorii Early State do badań przedkolonialnych państw Afryki Czarnej’ [‘Some Problems Arising from the Application of the Theory of the Early State to the Research on Precolonial State of Black Africa’], in J. Banaszkiewicz, M. Kara and H. Mamzer (eds), Instytucja ‘wczesnego państwa’ w perspektywie wielości i różnorodności kultur [Institution of the ‘Early State’ in the Perspective of Quantity and Variety of Cultures]. Poznań: Instytut archeologii i etnologii PAN, pp. 75–89. Tymowski, M., and M. Ziółkowski (eds). 1992. Geneza i funkcjonowanie wczesnych form państwowości na tle porównawczym [The Genesis and Functioning of Early Forms of Statehood in a Comparative Perspective]. Warsaw: Wydawnictwa Uniwersytetu Warszawskiego. Vansina, J. 1961. De la tradition orale. Essai de méthode historique. Tervuren: Musee Royal de l’Afrique Centrale. ———. 1965. Kingdoms of the Savanna. Madison: University of Wisconsin Press. ———. 1978. ‘The Kuba State’, in H.J.M. Claessen and P. Skalník (eds), The Early State. The Hague: Mouton, pp. 359–80. ———. 2001. Le Rwanda ancien. Le royaume nyiginya. Paris: Karthala. Vorbrich, R. 2013. ‘Suksecja substancjalna a legitymizacja władzy w wodzowstwach pólnocnego Kamerunu. Między perspektywą antropologiczną i historyczną’ [‘Substantial Succession and Legitimation of Power in the Chiefdoms of Northern Cameroon: Between Anthropological and Historical Perspectives]’, in J. Banaszkiewicz, M. Kara and H. Mamzer (eds), Instytucja ‘wczesnego państwa’ w perspektywie wielości i różnorodności kultur [Institution of the ‘Early State’ in the Perspective of Quantity and Variety of Cultures]. Poznań: Instytut archeologii i etnologii PAN, pp. 91–103. Ward, L.F. 1883. Dynamic Sociology, Vol. 2. New York: D. Appleton and Co. Westermann, D. 1952. Geschichte Afrikas. Staatenbildungen südlich von Sahara. Cologne: C.H. Beck.

Chapter 6

Retaliation, Mediation and Punishment in Ankole Revisiting the Chapter by Oberg Günther Schlee

The list of contributors to African Political Systems (APS) is replete with famous names, comprising founding fathers of the second generation of social anthropology or people who were to become founding fathers of important schools within it. Along with Günter Wagner, Kalervo Oberg is one of the least well known of these contributors, particularly among Africanists and British social anthropologists. This may have to do with him (a Canadian of Finnish origin) having worked on the Americas before and after his stay in Britain, which was interrupted by two years of field research in Uganda in the mid-1930s. He enjoys quite a standing, though a lesser one than he deserves, in America, and there particularly among Americanists. Not being well versed in the anthropology of the Americas, I can only illustrate that through chance findings. Carneiro, in his polemic against Pauketat, attributes the concept of chiefdom (which he defends and Pauketat contests) to Oberg: Surprisingly, the chiefdom, the concept as well as the term, was not introduced into anthropology until 1955. In that year the ethnologist Kalervo Oberg, in a ground-breaking article in the American Anthropologist, unveiled it as one of several structural forms he identified in the native cultures of Central and South America. (Carneiro 2010b: 173) In the article to which he refers, Oberg (1955) indeed takes up a theme of his contribution to APS. As we shall see below, his account of precolonial Ankole

Retaliation, Mediation and Punishment in Ankole  123

can also be read as the description of a transitional phase between two structural types: egalitarian, lineage-based organization or chiefdom/kingship. Oberg spent much of his life working for government institutions and is regarded as one of the founders of applied anthropology. One of his best-known papers, based on a talk he gave in 1954, deals with the development community itself. It is the one on ‘culture shock’, a concept he borrowed from Cora DuBois and elaborated into a theory, which divides culture shock into different phases. The paper can also be read as practical advice for people working abroad, providing lots of common sense (Oberg 1960). The development of British social anthropology had a strong focus on Africa, in particular East Africa. It owes much to the Azande, Nuer, Bemba and Zulu, and the people who wrote about them, the volume revisited here being a milestone in the course of this development. Oberg shared this British/African focus only temporarily. His journey, both institutionally and in terms of research interest, took him back to the Americas. Another factor might be his ‘innate modesty’, which the obituary in the American Anthropologist attests to him (McComb and Foster 1974), a feature that – no need to tell the younger generation of anthropologists who struggle for ‘visibility’– is harmful to the one who exhibits it. Oberg’s Ankole kingdom1 is classified as belonging to Group A in the introduction to APS (Fortes and Evans-Pritchard 1940: 5), but on a closer look, it straddles the dichotomy between this group, the state-like societies with central authority, and Group B, the societies with segmentary lineage2 systems. There is no monopoly of power. Clans and lineages have to sort out internal affairs themselves and interclan violence needs to be authorized by the king, but the authorized party would have to carry it out by itself. The Banyankole therefore provide a model case for the interpenetration of different logics of action. Using a conventional spatial metaphor, we can call these vertical and horizontal. ‘Vertical’ refers to top-down rule and ‘horizontal’ to give and take, tit for tat, bargaining and negotiations among formally equals. This distinction between different types of action, both in their pure and their mixed versions, has informed much of our own recent work, conducted at the Max Planck Institute for Social Anthropology in cooperation with other institutes.3 We have developed a research programme called ‘Retaliation, Mediation and Punishment’ (REMEP) (see e.g. Turner and Schlee 2017) that deals with the interpenetrating logics of action of these three components. For example, mediation with the aim of compensation payments and reconciliation may be undertaken in order to prevent violent retaliation; both retaliation and mediation among formally equals may be pushed into the background where the logic of punishment administered by a higher authority takes over, etc. In the deliberations that led to the institutionalization of this research programme, Oberg’s chapter in APS played a significant role (see e.g. Schlee and Turner 2008: 54–59). The forms of conflict regulation of the inhabitants of the precolonial kingdom

124  Günther Schlee

of Ankole4 in the western part of what now is Uganda neither correspond to the model in which equals retaliate against each other unless they negotiate other solutions, nor to the model in which justice was administered and punishment was dealt out from above; rather, they show a combination of the logics of retaliation and punishment. I will now try to illustrate this in a selective summary5 of Oberg’s chapter, focusing on these aspects, and will then come back to more recent research inspired by this perspective. In the kingdom of Ankole, punishing homicide or other crimes of the subjects against one another did not belong to the Mugabe’s (i.e. the king’s) duties. He might have passed judgment, but if the enforcement of his judgments did not favour his personal interests, he left it up to people who had an interest in enforcing them. This led to the victims of a violent crime no longer being entitled to exact revenge as a lineage on their own, as they would have been in a stateless setting; they would have been obliged to receive permission from the king.6 The king first had to withdraw his protection from subjects accused of homicide or their lineage before the damaged side or party could exact revenge or demand compensation. All this applies to the dominant stratum of society, the cattle-keeping Hima. The majority of the population were Iru, agriculturalists who were not allowed to keep reproductive cattle. Like many kingdoms in this corridor between Lake Victoria to the east and the Ruwenzori range to the west, Ankole has been described as a conquest kingdom. The best known, for deplorable reasons, of these (former) kingdoms is Rwanda, where the genocide of 1994 made headlines worldwide. In order to give a rough orientation, it may suffice here to say that the equivalent of the Tutsi are the closely related Hima in Ankole and the status equivalent of the Hutu are the Iru. To take revenge against Iru, the consent of the king was not required (Oberg 1940: 134). For a Hima lineage with a justified grudge against another Hima lineage, like a homicide, which required revenge, but without the military clout to carry out the revenge, the power of the king was of little help. The king’s power mainly served to satisfy his own or his subjects’ claims. Conversely, the culprit’s party had a number of options to prevent the revenge killing of one of its members. The sacred drum, spiritually connected to the king, symbolized an area of asylum. Touching it meant the neutralization of a king’s permission for revenge (Karugire 1971: 82, 103f.). In other cases of violations, the Mugabe was well authorized to pronounce a death sentence, although the default punishment for culprits was the confiscation of their cattle for the king’s benefit (ibid.: 53). Limiting violent disputes and avoiding their escalation in society was nevertheless in the king’s interests in order to maintain his own claim to power. A violent act also demanded the restoration of order, which fell within the purview of the king. The expectations he was held to thus comprised more than simply protecting his subjects and their possessions against foreign enemies. But m ­ eeting

Retaliation, Mediation and Punishment in Ankole  125

these expectations cannot strictly be interpreted as fulfilling a duty to others. Showing strength was essential to kingship, and thus to his own identity and survival. Protecting his subjects is the hallmark of a strong ruler and, as such, the king of Ankole did protect his subjects, especially when they requested it and offered him tribute. This did not constitute an abstract duty of protection, but rather a personal relationship of mutual services. The king had no tangible duties, whose dereliction anyone could have lamented, as he was not a public servant in the sense of a government structure. Any notion of him serving or being responsible to anything other than the sacred objects he was spiritually connected to would have seemed very odd to him.7 However, the king not protecting his subjects in conflict situations could be interpreted as a lack of spiritual energy and would reduce him in his capacity of virility and leadership, much like lacking sexual stamina. Subjects owed no loyalty to an unfit ruler and would cast votes with their feet simply by emigrating to a different dominion without anybody being able to stop them (see Karugire 1971: 65, 105). On the one hand, the king was above the law8 and was entitled to any woman or cattle he wanted. On the other hand, he was still dependent on certain people. The king could not go against his mother or his sister. These women had supported him during the murderous succession struggles with their intrigue so that he would be victorious against all his half-brothers, the sons of the previous king (his own father) had had with other women. The king was also bound to his mother’s clan in return for its support. Each clan fought for its nephew (the son of their clan sister) to become the new king in order to make sure the woman it gave to the previous king would be the mother of the new king and that all other sons of the king, whose mothers came from rivalling clans, would be slain or exiled. The king’s patrilineal descent was clear because all clans but one were excluded from kingship. The king would normally be the only remaining son of the previous king. Other clans than the king’s clan could never install a king, but they could provide the woman to be the king’s mother. The uterine descent of the king was thus the stepping stone to the contested second rank in the clan hierarchy.9 The king also depended on his wives. If they let on about any signs of weakness he might display, his days would be numbered. The poison cup saved him from old age. The king’s power, both his siring and violent, destructive power, was analogously transferred to the crops of the land and its people to secure their general wellbeing. This analogous-magical or symbolic-ritual role was perhaps the one element of the kingdom that most closely resembled modern ideas of serving the common good. However, it would be a gross exaggeration to speak of self-sacrifice on the part of the king. Oberg’s text never hints at the king’s consent being a prerequisite for his killing.10 The kingdom of Ankole can be considered a hybrid, halfway between an acephalous (or polycephalous) system, in which each lineage itself was to take revenge or seek compensation for wrongs it had suffered, and a state that claimed

126  Günther Schlee

the purview of law enforcement. Pronouncing sentences was partly the task of the king, but their enforcement remained with the lineages. In order to enforce its claims due to suffered wrongs, any lineage had to rely on its own strength. This strength and the inherent capability to exact revenge also determined the potential to demand compensation. Someone with no right to revenge and no capability to exact it does not need to be conciliated with compensation. Thus, the right to compensation is unimaginable in a state-free space or – as is the case of Ankole – in a state that does not reserve the prosecution of homicide for its own purview, but instead only gives permission to do so themselves to its subjects. It is tempting to generalize and infer that such conditions of budding statehood must mark the beginning of state formation. Can Ankole serve as a representation of an ‘early state’? Historically, we know nothing about the initial genesis of statehood. The oldest written sources already mention the oldest-known state – Sumer – in fully developed form, complete with complex institutions such as the military, bureaucracy and sciences (cosmology, astronomy and astrology) (Winkler 1907: 10). It has thus often been attempted to derive theories on early states11 from current or recent12 examples of real or apparently budding statehood. The classic nineteenth-century unilineal evolutionists came up with the most ambitious development schemata.13 Political philosophers also pondered state genesis – some with and some without recourse to ethnographic or historic examples – and designed social contract theories, whereby independent individuals freely agree to surrender some of their rights to a central authority that is better able to succeed in issues relating to the common good.14 If we consider the kingdom of Ankole as an example of budding statehood, where centralization had only been halfway realized, such an example would contradict the social contract theorists. The king was the strongest son of the strongest man. Every one of his paternal ancestors was the survivor of a succession struggle. The king enacted his power without restraint. His life’s purpose was to amass loot and tribute (which he could share with his brothers-in-arms) and to sire countless children with countless women. The only children of his who would survive were many daughters and the strongest, most ruthless son. It might have been challenging to convince such a man that he was an official tasked with functions for the common good. The king did not serve any other human being or collective than himself. Considering the beliefs surrounding the Banyankole, the king served – at best – the sacred drum, which embodied independent power and was the one insignia of the kingdom that was sought after, hidden and fought over during the succession struggle. Social contract theories might only be applied to the king and his armed retinue. These men apparently formed implicit or explicit agreements to rob women and cattle from other men or force them into services.

Retaliation, Mediation and Punishment in Ankole  127

Privilege and power came to the king at a cost. His entire life was dominated by certain behavioural rules. He had very little freedom of movement, had to adhere to dietary rules, had to be sexually active and would not die of natural causes. It is doubtful whether these conditions can be considered compatible with the idealistic ideas of the Enlightenment’s social contract theorists.15 The king fulfilled no tasks for the common good; instead, he could bestow favours like granting protection. Requesting the king’s protection, one first had to present him with a gift. The king thus had personal relationships that comprised giving and taking, but no official duties. His protection was a gift in exchange for a gift. He could persecute culprits for crimes or not. His rule was powerful and personal. But if he was unable to maintain order in general society – e.g. if he did not punish revenge actions not condoned by him – this would have been seen as demonstrating a reduction in his power.

Comparisons to International Law Drawing this parallel might seem outlandish, but the International Criminal Court (ICC) in The Hague and the kingdom of Ankole have one thing in common. Both merely resolve the legitimacy of claims – they pronounce sentences – but neither enforces its rulings. Both the Ankole kingdom as reconstructed by Oberg and the world (which is the level at which the UN and its institutions like the ICC operate) have preserved a feature of acephalous societies; they do not monopolize executive power. In Ankole, the king’s retinue was far too busy with raiding to have the time to support the relatives of a homicide victim against the culprit. The ICC has no executive force to begin with, as police and military forces belong to states. International law entirely depends on states or ‘coalitions’ for its enforcement. The UN or the ICC might approve and legitimize an action and yet it might never be executed. This would require an executive force to take action, and whether any does naturally depends on the national interests of any potential intervening ­nation-states. Furthermore, these interests may be only faintly – if at all – ­connected to the legal basis for the intervention (typically violations of international law or human rights). Allegorical portrayals of Justitia always depict her as a woman with a tunic and long hair, but more importantly with three central features: the sword, the scales and the blindfold. Which of these attributes can be found in the practice of international law? It has no sword of its own. Calling the Lady Justice toothless would seem impolite; however, international law does lack the sword. I once mentioned this fact during a presentation, to which Silja Vöneky, an expert in international law, replied to the contrary that international law has many swords. In her contribution to our book on retaliation (Schlee and Turner 2008), Vöneky lists the punitive instruments international law can employ: these comprise some

128  Günther Schlee

that are allowed in any case, with or without wrongdoing of the other side, like the extradition of diplomats, withholding development aid or not concluding contracts one would otherwise have concluded. Other measures, which do require justification by prior wrongdoing of the other side, can be noncompliance with existing treaties or restricting traffic, like denying planes the right to land. The use of violence (the ‘sword’ in a narrower sense) is restricted to self-defence, a contested concept, because it is not clear whether self-defence can only be the response to an actual attack or whether anticipatory action to prevent an attack is legitimate. Another contested issue is the issue of ‘humanitarian’ interventions: the use of violence with the aim of helping civilians to survive. However, the impression remains that Justitia may not wield these swords only on her own authority. Vöneky speaks of Selbstdurchsetzung, the necessity for states to enforce their rights themselves (Vöneky 2008: 150). International law legitimizes such enforcement, but as the institutions administering international justice do not have executive force, the states – or other states supporting them – have carry out the enforcement themselves. An economically, politically and militarily weak state, whose rights have been impaired by another state, would wait in vain for the sword of justice. And it would not have any effective sword at its own disposition. In this it would resemble a clan that has been authorized by the king of Ankole to carry out revenge against another clan, but which is not strong enough to fight the matter out with its own resources. The same is true for the Somali, where the injured party must either wield the sword itself or – if it is too weak – find an ally to do so, or abandon the claim altogether (Schlee 2017a). Given the circumstances of the sword, what is the situation with the scales? Ideally, the process is impartial, although the enforcement of a sentence becomes a partial issue as, in some cases, the national interests of the intervening power facilitate enforcement, but not in others. And the blindfold? In itself it does not represent blindly striking out, but the absence of any sympathies for one party or another and the disregard of the interests of the judging party. Do these concepts hold true? We can assume that, in the absence of her own sword, Justitia’s blindfold also loses its effect. Lady Justice will always have to peek out below her blindfold at the interests of those parties who might lend her a sword. The independence of the law – the notion that the judiciary is a power that never bends to other powers – might not be realized in its ideal form anywhere. Law, in all its forms, is always only a part of society and the interdependencies inherent therein – thus being a controlling force and also being controlled. Considering the selective enforcement of international law, we can assert that it is much more power-sensitive than other forms of law.

Retaliation, Mediation and Punishment in Ankole  129

Internal Law and Politics and the Cohesion of the Kingdom Elsewhere, I have discussed the power sensitivity of customary law among Somali (Schlee 2017a). Since Lewis’ classic monograph, A Pastoral Democracy (Lewis 1961), the Somali rival the Nuer, who play that role in APS, as the paradigmatic example of an acephalous, segmentary lineage society. The reason for the power sensitivity of Somali customary law is the same as the reason for the power sensitivity of international law. The Somali do not have a government that takes up the task of jurisdiction and, more importantly, law enforcement upon itself. The same applies to the society of nations: the world. There is no global government willing to enforce the law and capable of doing so. What we have in terms of international institutions does not comprise one government with an effective executive arm. What the Banyankole example shows is that even in a setting where we do find a state capable of centralizing (although not monopolizing) power and organizing violence on a terrifying scale, using this power to administer justice to everyone is simply not a task of the state. Even within the dominant Hima stratum, the rights of the weak were not enforced against the strong. In the case of homicide, the king withdrew his protection from the culprits, but it fell upon the group of the victims to take revenge according to their own capabilities. If the culprits or presumed culprits were Iru, the permission of the king to take revenge was not required. The group of a Hima ‘victim’ could take revenge as they saw fit. So did the state have a purpose for the common good? Did the state identify with the people and their interests? Did the people have a reason to identify with the state and to be loyal to it, apart from blunt coercion? Yes, in a way, all these questions can be answered in the affirmative, although the ways in which this happened may have been quite different from those in a modern nation-state or a functioning democracy. By some magical equivalence, the king stood for the country. His fertility, his sexual prowess and his wealth in cattle were not only the symbol of the wealth and fertility of the country, but somehow magically promoted them. However, there was little conflict of interest. The duty of the king did not differ a great deal from his inclination. While in other political systems, the ruler might face tough moral decisions between the public good and his personal enrichment, the duty of the king of Ankole very much was what a strong, healthy male, who was brought up as a warrior, liked to do in any case: loot; capture territory, cattle and people; take many women; and beget many children. The physiological base of this seems to have been permanent feasting on meat and beer. As the kings of Ankole were neither known as vegans nor for their abstinence from intoxicants, it is quite probable that they did not mind such a diet. All that lasted as long as it did, and then, as we have seen, the king was given a cup of poison.

130  Günther Schlee

Unlike modern constitutional monarchies, the kingdom of Ankole did not claim to be a democracy. But were there elements of democracy in it? Was absolute power moderated? Did the subjects have a say? The answer to all these questions is also yes, but again in a particular way. The kingdom of Ankole was not a monarchy in the strict sense; rather, it was bicephalous – two-headed to some extent. The drum, Bagyendanwa, was not just a ritual object and an insignia that the king needed to appropriate in order to become a king; it was thought of as having a personality of its own and it had its own court with its own elders. The king had an ethnicity. He was a Hima. It is true that he was a protector of the whole country and had to protect the Iru as well (against Hima raiders and conquerors from rival kingdoms), but he was also a symbol of military prowess and Hima dominance. By contrast, Bagyendanwa had no ethnicity. Much more than the king, it was a symbol that stood for the whole country. Even Iru in need turned to it and received help from the guardians of the sanctuary. Bagyendanwa was a moral instance. Through its magical powers, it punished wrongdoers, and people who suffered some misfortune redressed it by making donations to it and seeking reconciliation with their victims. It was also a centre of redistribution. People who had lost cattle through raids or disease or who were in economic distress for other reasons received help there. Special consideration was given to those who were not favourites of the king (APS, p. 154). It thus offered an alternative: it pluralized options; it made the monarchy less monarchical; it united Ankole and distinguished it symbolically from other kingdoms; it belonged to all of Ankole and Ankole only. Other kingdoms had other drums.

On Statehood In many ways, Oberg’s Banyankole provide food for thought about statehood in general. On the one hand, statehood among the Banyankole appears rudimentary. The king and his retinue are warriors and they go on cattle raids. Is the state here much more than a gang of armed men? Apparently yes, because in combination with the drum, kingship is at the core of some magical perception of the wellbeing and fertility of the country, land, people and cattle. On the other hand, the same question can be asked not only about some precolonial states like Ankole, but also about postcolonial states and all kinds of states around the globe. Many of them essentially appear to be gangs of armed men who extract wealth from others. What makes a state above and beyond that is the subject of endless debate. In some cases, there is more to statehood, in other cases not much more. We see purely or mainly predatory states in some settings that appear archaic and in other cases we see them recently emerging on the fringes of global capitalism. We have also mentioned that in the case of the oldest known ­statehood – that of Sumer – we have no information of an ‘early stage’ or how it has ‘emerged’. In the oldest records, a complex state with internal functional ­differentiation already

Retaliation, Mediation and Punishment in Ankole  131

appears in front of our eyes. Unilinear evolutionism is long dead and enumerating these examples that speak against it might be like flogging a dead horse, but its ideology has become so pervasive in our language that it is difficult not to fall back into its trap. We have stated above that Oberg’s ‘account of precolonial Ankole can also be read as the description of a transitional phase between two structural types: egalitarian, lineage-based organization or chiefdom/kingship’. Now we have to hasten to add that these phases are not always found in a strict chronology and that they are not associated with a clear dichotomy with ‘revenge or compensation’ on the stateless side and ‘punishment’ in the case of states. In a recent book called Ehre und Rache (Honour and Revenge), Ruch (2017) describes the legal changes concerning revenge in the transition between the archaic period in Ancient Greece, as reflected by Homer, and the classical period in the fifth century BCE. Ruch laments that legal historians attribute too much originality to Roman law and ignore its Greek predecessors, and that those who are interested in classical Greece attribute the invention of too many institutions to fifth-century Athens and ignore statehood and law in earlier periods. Evolutionism needs primitives to start with and has no use for sophisticated beginnings. It seems to fit unilinear evolutionist assumptions better to have state institutions develop out of tribal structures or even anarchy, rather than having law and order in the earliest records. In the period described in Homer’s epics, there were not only fully fledged little city-states (poleis) with public legislative assemblies, but also an entire sphere of international law had developed between them. This world of miniature states that engaged in trade and war with each other had produced rules even for hostile interaction. Wars were proclaimed and could be averted by conciliatory procedures. There were contractual alliances, an armistice after a battle to collect the bodies, there was arbitration, and treaties about extradition with each other existed (Ruch 2017: 60–66). Had, in this world of states, the role of revenge declined? This is what one would expect along evolutionary ideas. One would expect that revenge is necessary in the absence of the state and that with the development of statehood it is replaced by something better: History as progress, history as improvement. What Ruch reports is that there were changes in the laws about revenge, but not in the sense of weakening it or reducing its importance. In certain cases, where hubris rather than mere damage was involved, it became obligatory. Originally, the right of revenge conceded by the state to the harmed party could be seen as a way to channel or limit violence. The legitimacy of revenge was an incentive to the party of the perpetrator to ‘buy off’ the right of revenge by offering compensation (Ruch 2017: 289). With the concept of honour and infractions against it by hubris acquiring a religious dimension and becoming a concern of the state, intentional killings and other crimes motivated by hubris

132  Günther Schlee

could no longer be dealt with by buying off revenge and paying compensation. Deadly revenge became obligatory and, willingly or unwillingly, the avenger became an agent of the state and lost his own honour (implying the loss of his social status and economic existence) if he did not carry out the revenge. However, at some point, unlike in Ankole, the state might have come in by helping to enforce its own law. If in trouble, the avenger could rely on state support (Ruch 2017: 273). So here revenge is not an indicator of a ‘weak’ state; it exists in a setting in which the state is quite strong, even strong enough to demand revenge where the parties concerned would rather reconcile. The difference to a modern (posthonour) state and society is that the ordinary citizens were also strong in the sense of being entitled to carry out violent action and expected to be capable of doing so. Oberg does not report any obligation for Banyankole to carry out revenge, but not to carry it out would have had consequences. Like in Ancient Greece, it would lead to the loss of one’s status and would be tantamount to giving up any claims about property, since it would invite future transgressions. Ruch (2017: 34) would call this the ‘materiality of honour’. There was a close relationship between the perception of the working of human societies and law. The founding fathers of anthropology were lawyers (Morgan, Bachofen, etc.). Before you could study anthropology, law was as close as you could get to what people thought constituted society. Some lawyers still believe that human behaviour is shaped by law, but the emancipation of the social sciences from their mother discipline, law now let the relationship between law and what people actually do appear much looser than it appeared to our predecessors. Critical law studies, which show the unintended consequences of laws rather than the norms on which they are based and their noble intentions, have contributed to this effect. All this implies that one cannot fully explore statehood or examine the workings of a particular state by looking at its constitution. It also implies that anthropology, including legal anthropology, needs to be more than the study of the ‘customs and traditions’ of a group, which is still a notion very closely associated with (unwritten) law. To capture ‘statehood’ and what precisely a state is, there has been a shift from normative to descriptive notions. Seeing Like a State is the title of a book by James Scott (1998). States perceive things in certain ways: as countable, measurable, ‘legible’, taxable, etc. Sureau describes how, in the most recent modern ‘nation’-state, South Sudan, rudiments of bureaucracy in the form of making lists were essential for emerging statehood. There are state crops like rice or other types of grain that can be standardized and stored (Persoon 1992). Some states may revolve around revenue from the export of raw materials, which trickles down networks of support and loyalty; others are about trade routes and tolls. Ankole statehood seems to have had a focus on cows, women and the fertility of the land, and that seems to be what made life meaningful to Banyankole people in general.

Retaliation, Mediation and Punishment in Ankole  133

Routines, accompanied by a sense of responsibility and a desire for order, are an important element of statehood. In the only African country characterized by a long tradition of statehood on a larger scale, Ethiopia, three days after a revolutionary, violent regime change (from the Derg to the Tigrayan People’s Liberation Front (TPLF) in 1991), a new month started, salaries were due, and all salaries were paid on time. The administration had simply continued (de Waal 2015).16 Where such routines are interrupted and large numbers of people suffer, one speaks of state failure. But state failure always needs to be discussed with a reference parameter: failed for whom? States that have the purpose (mostly in the form of a hidden agenda) to create bottlenecks in the economy and to control these bottlenecks in order to channel wealth into the pockets of an exclusive ‘elite’ cannot be said to have failed when exactly this happens. This leads us to the distinction between public and private. Can one distinguish between the state and the state class – or between the state and the people who man the state? Can one accuse corrupt officeholders of misusing their ‘public’ office to stuff their ‘private’ pockets? In Oberg’s Ankole kingdom, this distinction, which is so important for our thinking about modern states, certainly did not exist. The king ruled as a person and he acquired cows, women and agricultural tribute for himself in all his capacities – as a man, a father, a king and a warrior. The separation of roles had certainly not taken place in the sense in which they are thought of in modern sociology or political science. In recent years, identification has become increasingly important to me as an element of statehood. A smaller or a larger number of people can be identified as ‘the state’ in inclusive or exclusive concepts of statehood, and states can have an idea who should their subjects or citizens be and how these people should be, which likewise boils down to mechanisms of inclusion or exclusion or ranked varieties of citizenship rather than uniform, ‘universal’ citizenship (Schlee 2018). In the kingdom of Ankole, both the king and the drum, Bagyendanwa, were symbols of identification, and Bagyendanwa seemed to be the more inclusive of the two.

Concluding Remarks The dichotomy used in APS between Group A societies (early states) and Group B societies (segmentary lineage systems) has been criticized many times, including in the present volume, as too rigid and not covering the complexities of African political systems, either now or at the different periods of time to which the contributions to APS refer. In a comparative essay that extends beyond Africa, Aidan Southall (1988), has shown that kingdoms can also segment (see the chapters by Lewis, Simonse and Skalník in this volume). However, segmentary kingdoms do not fit the APS dichotomy, since there the adjective ‘segmentary’ invariably comes with the noun

134  Günther Schlee

‘lineage’. Sons of the rulers of core kingdoms can found offshoots that are not under the political control of the core, but remain attached to it by notions of seniority and ritual adherence. So, we have a wider area of shared ritual kingship that contains autonomous political units belonging to individual kings as its ‘segments’. ‘Ritual’ versus ‘political’ may be another simplistic dichotomy. My point is that there are different kinds of offices, social positions and forms of power that may overlap in the same areas and affect the same people. There are not only two forms of power. Bernardi (1985: 19) has pointed out the relative neglect of age-grading systems in APS. Alemayehu (2021) reminds us that age-grading systems can provide a large number of political and ritual roles and are a form of government that combines clans and lineages with something else, namely age-sets that are recruited from all clans and lineages, and demand forms of cooperation and leadership that are not adequately described either by the lineage paradigm or by the kingship paradigm. They are the frame for one or many other kinds of power. Alemayehu regards the re-enactment of complex ritual cooperation between the holders of many different kinds of power as an explanation of the remarkable temporal stability of the Gadaa system of the Borana Oromo. We have learned from Southall that political units need not be coterminous with ritual domains. The various Gadaa systems of groups of the Oromo and the Somaloid in the Kenyan/Ethiopian borderlands take clues for timing and require ritual inputs from each other. There is a metalevel on which Gadaa systems of different ethnic groups, whose interrelationships are even intermittently hostile, interact with each other (Schlee 1998). Oberg’s description of Ankole also comprises different kinds of power. There is the king and there are the lineages that have to rely on their own power to defend their rights. The king authorizes assertive actions, but he does not carry them out for any offended party. And then there is the drum, Bagyendanwa, which does not belong to the king. Bagyendanwa does not need the king, but no one can become king without Bagyendanwa. It does not belong to a lineage either. It is supra-lineage and even supra-ethnic. It clearly incorporates a third kind of power. This makes Oberg’s Ankole not only an example of where two types of logic of action interpenetrate, with kingship/vertical power relations/punishment on the one side and lineage/horizontal interactions/retaliation or bargaining on the other, but where other kinds of (counter)power also come in, which, for lack of more specific terms, we tend to combine under ‘ritual’. Günther Schlee is one of the founding directors of the Max Planck Institute for Social Anthropology in Halle, Germany. Prior to this appointment, he was Professor of Social Anthropology at the University of Bielefeld until 1999. He conducted fieldwork in Kenya, Ethiopia and Sudan, and was a guest lecturer

Retaliation, Mediation and Punishment in Ankole  135

in Padang (Sumatra) and at the École des Hautes Etudes en Science Sociales in Paris. His main publications include Identities on the Move: Clanship and Pastoralism in Northern Kenya (Manchester University Press, 1989) and How Enemies Are Made: Towards a Theory of Ethnic and Religious Conflict (Berghahn Books, 2008). After his retirement from Halle in 2019 he has taken up a professorship for social anthropology at the Arba Minch University, Ethiopia.

Notes   1. Here we take his description as a model case for theorizing, being well aware, as Oberg himself was, that this description refers to a historical moment that he himself did not witness, but only reconstructed.  2. For more on lineages, see Oberg (1938). Unilinear descent plays a greater role for the Hima than for the more cognatic Hutu. Oberg shows how this is reflected in their respective kinship terminologies (two terminologies within the same language!) and discusses the connections between these forms of kin organization and the different forms of livelihood (cattle pastoralism versus agriculture). For a more general perspective on kinship terminologies and their distribution, see Schlee (2017b).  3. The Max Planck Institute for Foreign and International Criminal Law, Freiburg, the Max Planck Institute for the History of Law, Frankfurt, and the universities in the respective locations, including Halle. See Turner 2014; Kohlhagen 2015; Sureau and Auge 2019; and Härter, Hillemanns and Schlee 2020.  4. It was quite common in the ethnographic writings of social anthropologists of the first two generations to present descriptions of the precolonial situation, which, by the time of their field research, was still fresh in people’s memories. This is often implicit and the impact of the colonial ‘pacification’, to which the anthropologists after all owed the possibility to undertake field research, was simply left out. Oberg makes this omission explicit in the last paragraph of his chapter (Oberg 1940: 162). In his other article, Oberg (1938) takes the actual time of his field research as the ethnographic present. He discusses not only the differences in the kinship organization of the Hima and the Iru, but also deals extensively with recent changes brought about by the colonial power, such as labour migration, tax payments, bank accounts, and their impact on family structures and kinship organization.  5. The following paragraphs are an expanded and modified translation of Schlee and Turner (2008: 54–59).   6. Oberg 1940: 133f., 137. Oberg’s text originates from a work from 1940 work and refers to the early colonial era and a reconstruction of precolonial times, around and before 1900. Therefore, we use the past tense. See also Roscoe 1923; Karugire 1971: 115f.; ­Stenning 1972.   7. See Stenning (1972: 262f.) on the Mugabe as the divine king.   8. For examples of legal competencies and duties of a king in African states, see Turner (2005).   9. Similar succession wars with similar importance on uterine descent have also been known outside of precolonial African states, such as in the Ottoman Empire. 10. Oberg 1940: 157. Karigure (1971: 83, 93) extensively discusses the concept of royal suicide, which is a very controversial research topic, and argues that his subjects regard the divine king to be immortal and, as such, his death could only be explained with suicide.

136  Günther Schlee 11. See, for example, Claessen and Skalník (1981). 12. This includes the kingdom of Ankole, whose description above reflects the period between the late nineteenth and early twentieth centuries. 13. On the so-called sacred kingdom, see the classic work by Frazer (1907–15). 14. On more recent social contract theory, see, among others, Buchanan and Tullock (1962); Rawls (1971); Nozick (1974). 15. Other examples of centralized political rule in Africa come closer to this approach or, rather conversely, the interpretation of empirical data was influenced much more strongly by this approach. See e.g. Gluckman (1955) on the Lozi. 16. Elements of persistence across revolutionary regime changes are also discussed as ‘transcontinuities’ (Schlee 2002).

References Alemayehu Debelo Jorgo. 2021. ‘Gadaa Reconsidered: The Structurationist Approach to Understanding the Social System of the Borana Oromo of Southern Ethiopia’, Ph.D. dissertation. Halle: Philosophical Faculty I, Martin Luther University Halle-Wittenberg. Bernardi, Bernardo. 1985. Age Class Systems: Social Institutions and Polities Based on Age. Cambridge: Cambridge University Press. Buchanan, James M., and Gordon Tullock. 1962. The Calculus of Consent: Logical Foundations of Constitutional Democracy. Ann Arbor: University of Michigan Press. Carneiro, Robert L. 2010a. ‘Pauketat’s Chiefdoms and Other Archaeological Delusions: A Challenge to Social Evolution. Social Evolution and History, 9(1): 135–65. ———. 2010b. ‘Critique of Pauketat’s Volume’, Social Evolution and History, 9(1): 172–76. Claessen, Henry J.M., and Peter Skalník (eds). 1981. The Study of the State. The Hague: Mouton. De Waal, Alex. 2015. The Real Politics of the Horn of Africa. Cambridge: Polity Press. Fortes, Meyer, and E.E. Evans-Pritchard. 1940. ‘Introduction’, in Meyer Fortes and E.E. ­Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 1–23. Frazer, James G. 1907–15. The Golden Bough: A Study in Magic and Religion. Oxford: Oxford University Press. Gluckman, Max. 1955. The Judicial Process among the Barotse of Northern Rhodesia. Glencoe, IL: The Free Press. Härter, Karl, Carolin Hillemanns and Günther Schlee (eds). 2020. On Mediation: Historical, Legal, Anthropological and International Perspectives. Oxford: Berghahn Books. Karugire, Samwiri R. 1971. A History of the Kingdom of Nkore in Western Uganda to 1896. Oxford: Clarendon Press. Kohlhagen, Dominik (ed.). 2015. REMEP – Retaliation, Mediation and Punishment: Research Agenda and Projects. Max Planck Institute for Social Anthropology, ‘Integration and Conflict’ Department, Field Notes and Research Projects, vol. XII. Retrieved 12 September 2021 from http://www.eth.mpg.de/pubs/series_fieldnotes/vol0012.html. Lewis, Ioan M. 1961. A Pastoral Democracy: A Study of Pastoralism and Politics among the Northern Somali of the Horn of Africa. Oxford: Oxford University Press. McComb, Marlin R., and George M. Foster. 1974. ‘Kalervo Oberg 1901–1973 (Obituary)’, American Anthropologist, 76: 357–60. Nozick, Robert. 1974. Anarchy, State, and Utopia. Oxford: Basil Blackwell. Oberg, Kalervo. 1938. ‘Kinship Organization of the Banyankole’, Africa, XI(2): 129–59.

Retaliation, Mediation and Punishment in Ankole  137 ———. 1940. ‘The Kingdom of Ankole in Uganda’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press. ———. 1955. ‘Types of Social Structure among the Lowland Tribes of South and Central America’, American Anthropologist, 57: 472–87. ———. 1960. ‘Culture Shock: Adjustment to New Cultural Environments’, Practical Anthropology, 7(4): 177–82. Persoon, Gerard. 1992. ‘From Sago to Rice: Changes in Cultivation in Siberut, Indonesia’, in David Parkin and Elisabeth Croll (eds), Bush Base, Forest Farm: Culture, Environment, and Development. London: Routledge, pp. 187–99. Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press. Roscoe, John. 1923. The Banyankore. Cambridge: Cambridge University Press. Ruch, Philipp. 2017. Ehre und Rache: Eine Gefühlsgeschichte des antiken Rechts. Frankfurt: Campus. Schlee, Günther. 1998. ‘Gada Systems on the Meta-ethnic Level: Gabbra/Boran/Garre Interactions in the Kenyan/Ethiopian Borderlands’, in Eisei Kurimoto and Simon Simonse (eds), Conflict, Age and Power in Northeast Africa. Oxford: James Currey, pp. 121–46. ———. 2002. ‘Regularity in Chaos: The Politics of Difference in the Recent History of Somalia’, in Günther Schlee (ed.), Imagined Differences: Hatred and the Construction of Identity. New Brunswick, NJ: Transaction Publishers, pp. 251–80. ———. 2017a. ‘Customary Law and the Joys of Statelessness: Somali Realities beyond Libertarian Fantasies’, in Bertram Turner and Günther Schlee (eds), On Retaliation: Towards an Interdisciplinary Understanding of a Basic Human Condition. Oxford: Berghahn Books, pp. 208–35. ———. 2017b. ‘Omaha Terminologies: Global Distribution Patterns and How They May Have Come about’, Cross-Cultural Research, 51(2): 117–41. ———. 2018. ‘Identification with the State and Identifications by the State’, in Günther Schlee with Alexander Horstmann (eds), Difference and Sameness as Modes of Integration. Oxford: Berghahn Books, pp. 78–91. Schlee, Günther, and Bertram Turner. 2008. ‘Rache, Wiedergutmachung und Strafe: Ein ­Überblick’, in Günther Schlee and Bertram Turner (eds), Vergeltung: Eine interdisziplinäre Betrachtung der Rechtfertigung und Regulation von Gewalt. Frankfurt: Campus, pp. 49–67. Scott, James. 1998. Seeing Like a State. New Haven: Yale University Press. Southall, Aidan. 1988. ‘The Segmentary State in Africa and Asia’, Comparative Studies in Society and History, 30(1): 52–82. Stenning, Derrick J. 1972. ‘Salvation in Ankole’, in Meyer Fortes and Germaine Dieterlen (eds), African Systems of Thought. Oxford: Oxford University Press, pp. 258–75. Sureau, Timm. 2016. ‘“The Last Bullet”: South Sudan’s State Emergence’, Ph.D. dissertation. Halle: Philosophical Faculty I, Martin Luther University Halle-Wittenberg. Sureau, Timm, and Yelwa Auge (eds). 2019. Understanding Retaliation, Mediation and Punishment: Collected Results. Max Planck Institute for Social Anthropology, ‘Integration and Conflict’ Department, Field Notes and Research Projects, vol. XXV. Retrieved 12 September 2021 from http://www.eth.mpg.de/pubs/series_fieldnotes/vol0025. html. Turner, Bertram. 2005. Asyl und Konflikt. Rechtsethnologische Untersuchungen. Berlin: Reimer. Turner, Bertram. 2014. International Max Planck Research School ‘Retaliation, Mediation, and Punishment’ (IMPRS REMEP), in Max Planck Institute for Social Anthropology: Report 2012–2013: 115–122.

138  Günther Schlee Turner, Bertram, and Günther Schlee (eds). 2017. On Retaliation: Towards an Interdisciplinary Understanding of a Basic Human Condition. Oxford: Berghahn Books. Vöneky, Silja. 2008. ‘Reaktionsformen auf abweichendes Verhalten aus Sicht des Völkerrechts’, in Günther Schlee and Bertram Turner (eds), Vergeltung: Eine interdisziplinäre Betrachtung der Rechtfertigung und Regulation von Gewalt. Frankfurt: Campus, pp. 149–71. Winkler, Hugo. 1907. Die Babylonische Geisteskultur. Leipzig: Verlag von Quelle und Meyer.

Chapter 7

Beyond ‘African Political Systems’? The Relevance of Patrilineal Descent in Moments of Crisis in Northern Somalia Markus V. Hoehne … the political culture of Somali society today is in the midst of profound and fundamental social transformation. —Ali Jimale Ahmed, ‘Introduction’ It [The Nuer] is regarded as the archetypical product of British Social Anthropology, and in particular as the foundation of modern lineage theory. —Meyer Fortes, ‘Preface’

Introduction In this chapter, I differentiate between segmentary lineage theory and segmentary lineage systems as a model. The theory about segmentary lineage societies, advanced by Fortes and Evans-Pritchard (1940), and exemplified most famously by the case of the Nuer (Evans-Pritchard 1940a, 1940b), has attracted harsh criticism and indeed has not proven helpful to exhaustively explain cause and effect regarding social practices in the societies to which it was supposedly applicable. However, I argue that the segmentary lineage system as a model is still useful as a heuristic device.1 This has already been mentioned by Fortes when he argued that ‘[t]he segmentary type of system holds together because it is made up of parts that are homologous with one another in their constitution and are in dynamic, not mechanical balance. But remember, it is a model for inquiry, a heuristic device’ (Fortes 1979: ix). Still, there is confusion in the literature on

140  Markus V. Hoehne

the matter, and in other places, Fortes and others speak of segmentary lineage theory (ibid: vii; Gellner 1969: 42; Kuper 1982: 92). This confusion needs to be sorted out in order to assess the usefulness of the segmentary lineage theory or model today. I argue that at least in the Somali setting, where I have carried out extensive ethnographic field research, the model helps create a clearer understanding of options and variations of behaviour against the background of certain kinship structures in certain situations. This does not deny that Somali society is complex, dynamic and subject to change. There has been a long debate about the merits and flaws of ‘clan-analysis’. The latter is famously related to Lewis (1961),2 who, in his groundbreaking ethnography, represented northern Somali society as a segmentary lineage society. Besteman emphasized that ‘race, heritage, and status invest Somali identities with much greater complexity than simple assumption of one exclusive clan identification suggests’ (1996: 125). Samatar (1992: 627–28) and Kapteijns (2004–10: 3–4) emphasized change through colonial and postcolonial interferences that makes the emphasis of patrilineal descent as the main explaining variable for Somali politics simplistic at best. In my view, the emphasis of other factors shaping everyday life and the multitude of internal contradictions and conflicts enshrined by lineage-based kinship systems based on age, gender, roles, modes of work and exchange (Zitelmann 2018: 67) do not undermine the relevance of segmentary dynamics in moments of escalating conflict. My argument is that in order to capture complex political processes and conflict dynamics in northern Somalia – which was the focus of my research between 2002 and 2018 – the model of segmentary lineage systems is still extremely useful, in combination with a focus on power differences and individual decision-making. This argument touches on the ‘chicken and egg’ question as to whether structure establishes practice or practice establishes structures. In my eyes, there is no solution to this question. One can, however, choose a position. I position myself here on the side of agency or practice, although without trying to get rid of structures completely. The Somali example is – as I will show – an interesting example in which the reformation and transformation of structures through new kinds of practices (e.g. in transnational space) can be followed up. In the following, I will present my preliminary thoughts on this process. Segmentary lineage theory has an intellectual genealogy going back to the social evolutionists of the mid-nineteenth century. Maine, Morgan and others were interested in the formation of corporative groups in ‘primitive’ societies. Kuper stressed that Durkheim, in Division of Labor in Society (orig. 1893) and Rules of Sociological Method (orig. 1894), developed his concept of segmentary societies as ‘societies based on repetitive parts which joined together simply through a sense of mutual resemblance’ (Kuper 1982: 75; see also Sigrist 2004: 4). According to Durkheim’s terminology, segmentary societies were based on ‘mechanical solidarity’.3

Beyond ‘African Political Systems’?  141

In British social anthropology, the theory was developed further by Fortes and Evans-Pritchard in their co-edited volume African Political Systems (1940). The basic tenets of segmentary lineage theory are as follows: there are groups based on unilineal descent, e.g. patrilineal groups. Each group per se exists as an autonomous entity. Among each other, the groups are related in a ‘total genealogy’, e.g. along the father’s line. Within these systems, nearer generations of forefathers are known. The further generations date back, the more fictitious the line of descent becomes. Many northern Somalis, for instance, trace their forefathers back to the family of the Prophet Muhammad, which is clearly an innovation going back to the Islamization of the Somali Peninsula between the seventh and the nineteenth centuries, starting at the coast shortly after the death of the Prophet, but reaching the hinterlands only centuries later. Cassanelli (2010) convincingly argued that the elaboration of the total Somali genealogy is a product of nineteenth-century Islamic reforms in urban centres along the Swahili coast between Mogadishu and Mombasa. Segmentary lineage societies exhibited no centralized political authority, despite the fact that they could consist of up to a million people or more, which fascinated early anthropologists and troubled colonial administrators. Within these societies, relations between political groups could be understood only ­situationally – a person’s sense of belonging changed depending on the level of segmentation that was the most relevant in any given situation. A person could identify as member of a lineage, a clan or a tribe or clan-family. Flexibility was structured through the principle of ‘complementary opposition’. This means that groups regularly located at equivalent levels of segmentation (put simply, primary, secondary or tertiary segments) would cooperate with or confront each other. Cooperation or confrontation could therefore trigger dynamics of fusion or fission. If, for instance, a member of a lineage A1 belonging to clan A was in conflict with a member of a lineage E1 belonging to clan E, tensions flared up potentially between all members of clan A and clan E. If, on the other hand, vital resources like a pasture or water became scarce in an area jointly used by members of clan A, its members would split along descent lines into lineages A1, A2 and so forth, and would compete with each other. A further aspect of this principle of complementary opposition would be that usually, uninvolved ‘third parties’ would be called upon to mediate between groups in conflict. The theory then, in its simplest form, would establish a relationship between unilineal descent and social solidarity. It would state that social solidarity exists primarily between agnates. But due to the principle of complementary opposition, and in a setting in which resources are scarce, groups of agnates would never be in a state of stable relations, but would fight with each other or cooperate against more distanced agnates or others as the situation required. A frequently quoted saying, said to be Arabic, captured the nature of the theory as: ‘Me against my brother; me and my brother against my cousin; me, my brother and my cousin against the world.’ It is

142  Markus V. Hoehne

worth noting that relatively early on, Barth, educated in the structural-­functional tradition of social anthropology, but soon developing his own perspective criticizing the ‘homo sociologus’ model of Evans-Pritchard and others, provides an interesting qualifying discussion to this somewhat simplistic logic: There are limited situations, such as in the allocation of usufruct rights to jointly owned land, when a nesting hierarchy of segments is made tangible; but even then the actual politics is of a more complex nature. In far the most confrontations involving related parties, the opposed units are not unilineal descent segments but factions, built on bilateral and affinal relations, friendship and opportunistic alliances as well as a selection of agnatic relations. As shown most clearly in Barbara Aswad’s recent study, the main schisms of descent groups occur between close collaterals, in her case even full brothers; and these schisms serve as a focus for the alignment of others – not without regard to segmental position, but transmuted by cognatic and affinal ties. (Barth 1973: 13) Therefore, the above-mentioned proverb should instead be: ‘Me against some of my brothers, me and some of my brothers against some of my cousins, me, some of my brothers and some of my cousins against some outsiders’ (ibid.; italics in the original). The segmentary lineage theory is essentially about social cohesion (Gellner 1969: 42). It made it possible to predict (a key feature of a theory) who, in a given situation, would fight against whom in cooperation with whom, and who would be in the position to mediate. Put into colonial practice, a conclusion arrived at by this theory was that one should not try to rule people like the Nuer or the Somali in northeastern Africa through the establishment of fixed hierarchical structures. Rather, one should keep a certain flexibility in colonial rule, punish only the most threatening crimes like revolt or manslaughter under colonial law, leave much leeway to indigenous forms of legal regulation and dealing with conflicts, and establish relations with influential figures at various levels of segmentation (lineage, clan and tribe or clan-family) that could be used, if the need arose, to play off one temporary leader or group against another. In this way, a flexible form of divide and rule could be established to deal with evasive and ­independent-minded people who respected no permanent authority and even frequently lived as nomadic pastoralists – meaning they also did not respect permanent territorial borders. The theory could be complicated by adding the principle of clan-exogamy, which would regularly establish affinal ties between not closely related people and their agnatic groups. This would lead to what Gluckman understood as cross-­ cutting ties (again in its simplest form). In case of conflict, agnates could act in solidarity with in-laws. However, in other situations, group pressure among

Beyond ‘African Political Systems’?  143

a­ gnates could lead to the cutting of affinal ties. Therefore, whether cross-­cutting ties would enhance the peacefulness between groups or, on the contrary, would facilitate conflict escalation remained a matter of debate (Gluckman 1965: 12–20; Schlee 1997: 578). Peters (1967), working among the Bedouin of Cyrenaica, heavily criticized segmentary lineage theory and spoke of it as a ‘folk model’, which is taken for granted within a society, but is actually not in accordance with what is really happening. In his ethnography, he showed how central tenets of the theory, particularly complementary opposition, did not work in practice. A key argument in his case was that competition over local resources would constantly bring neighbouring groups – that were also close agnates – into conflict. Solidarity was therefore not prevalent among these close agnates, and each side sought to receive support from genealogically more distanced groups against each other. Salzman (1978) took on Peters’ critique and emphasized, in a nutshell, that in settings in which mobile groups, like the northern Somali pastoral nomads described by I.M. Lewis, did not lay claim to bounded territories and where resources were frequently not stably localized (due to erratic rainfall, etc.) and essentially had to be shared, the segmentary lineage theory still applied. However, Salzman noted that in practice, solidarity and opposition did not only follow agnatic principles. Demographic imbalances produced by the unforeseeable growth of one group or the decline in membership of another could be supplemented by reference to additional ties of solidarity through, for example, the mother’s patriline (so-called uterine ties). Salzman called this the ‘lineage plus model’; I would call it segmentary lineage plus theory, since it still established causal relations between concepts leading to predictable effects. Yet, Salzman recognizes some validity in Peters’ critique. He admits that Peters issues ‘a fair warning that various factors can enter into political behaviour, and we had better know what they are before we say that political behaviour is lineage based’ (Salzman 1978: 62). Concretely referring to the (northern) Somali, who are one of his main examples for defending segmentary lineage theory against Peters, Salzman stresses that: ‘The Somali, moving about in irregular patterns as a response to erratic resources, cannot establish political relations on the basis of propinquity’ (ibid.). He saves the theory, it seems, by admitting contingencies that ‘are not random and beyond understanding, but patterned in relation to certain circumstances’ (ibid.: 64). In my understanding, he adds a gradual dimension to segmentary lineage theory by integrating the possibility to ask ‘to which extent a particular principle or pattern is present, and to what degree it is modified by this or that factor or circumstance’ (ibid.: 66). In line with this thought, but already before it was expressed, Lewis (1965) had set out to compare several segmentary lineage societies to establish which of these followed the theory most accurately. It does not come as a surprise that in his conclusion, Somali society actually exhibited the central tenets of the theory

144  Markus V. Hoehne

in its purest form. In contrast to Lewis and Salzman, who sought to save the ­theory – and I again insist that it must be understood as theory here – Kuper (1982: 92) asserted in the conclusion of his thorough review of ‘lineage theory’ (sic!) that it had ‘no value for anthropological analysis’. Kuper adds that it was not enough to distinguish ‘folk models’ from ‘decision models’ and that one should not even attempt a guarded defence of lineage theory (ibid.). His main arguments against the theory were: ‘First, the model [sic!] does not represent folk models which actors anywhere have of their own societies. Secondly, there do not appear to be any societies in which vital political or economic activities are organized by a repetitive series of descent groups’ (ibid.) Fifteen years after Kuper, Baştuğ (1998: 95) asserted that ‘claims that it [the segmentary lineage model] has been discredited are seldom substantiated’. She understood it as a model or concept, less than a theory, and claimed that losing this model would deprive us of one way of gaining a better understanding of ‘the cultures of peoples whose system of descent takes the form of segmentary lineage system’ (ibid.). She wished to free the model from elements that, in my eyes, made it a theory, namely explaining the formation of groups in conflict or the maintenance of a stateless political order. She argued that many of these characteristics ‘cannot be, and should never have been, regarded as defining criteria for a type of unilineal descent’. She advocated viewing the segmentary lineage system as part of wider kinship systems and in dialectic relationships ‘with many other (changing and always changeable) social, political, economic, ideological and historical factors’ (ibid.: 96). With this, Baştuğ comes close to what I would understand as segmentary lineage system as a model and as heuristic device. A heuristic device is ‘[a]ny procedure which involves the use of an artificial construct to assist in the exploration of social phenomena. It usually involves assumptions derived from extant empirical research. … [It] proved especially useful in studies of social change, by defining bench-marks, around which variation and differences can then be situated. In this context, a heuristic device is usually employed for analytical clarity, although it can also have explanatory value as a model’.4 In my view, segmentary lineage system as a model can help us – in contexts like northern Somalia, but also elsewhere – to understand options and variations of behaviour against the background of certain kinship structures and dynamic social situations. While in everyday life, descent is only one factor among many, I argue that particularly in situations of escalating conflict, a dynamic and complex social formation can change into a rather rigid descent-based formation. One brief methodological remark should be made here: between 2002 and 2018, I spent some three years in total carrying out research in northern Somalia. I was able to follow up on events I had observed years ago and in some cases could see how certain political strategies were implemented and what the consequences of these were. In this way, I was able to participate in, observe and analyse a series of complex social situations, frequently engaged with people who

Beyond ‘African Political Systems’?  145

Figure 7.1  Map of political divisions, northern Somalia (© Markus V. Hoehne)

over the years sometimes became close acquaintances or friends, became known to many people in the region and thus also was updated – by email or phone – when things happened that people thought could interest me, even while I was not in the field. I have summarized aspects of this engagement in my book Between Somaliland and Puntland: Marginalization, Militarization and Conflicting Political Visions (Hoehne 2015: 9–12).

Ethnographic Approximations The Formation of Puntland as Clan-Administration After years of mounting armed opposition, the Somali government collapsed in January 1991. Subsequently, an ‘all-out’ civil war escalated. Its key drivers were warlords who created factions along lines of patrilineal descent. The ‘clan logic’,

146  Markus V. Hoehne

relating to dynamics of fusion and fission, complemented by complex alliance formations and social hierarchies between ‘noble’ and ‘ignoble’ or majority and minority groups (the latter becoming easy prey for any armed faction), structured important aspects of the formative years of the Somali civil war. Early on, the concrete conflict dynamics followed in principle what Barth (1973: 13) had outlined (see above). Indeed, some close agnates, together with some further-away patrilineal cousins, fought against some genealogically more distant actors (some of whom were allied with non-Somali outsiders). Occasionally, affinal ties were used to facilitate the negotiation of ceasefires or to convince faction leaders to attend peace conferences (which eventually all failed throughout the 1990s). In addition, external interventions (by the United Nations, the United States and neighbouring states) influenced the dynamics of violence dramatically and provided further fuel for the conflict (de Waal 1997), and fostered changing alliance formations and the proliferation of armed factions (given that in such a setting, the chances of Somali clans or lineages gaining a share of power and resources increased if they had their own warlord). Interestingly, attempts to overcome the ensuing collapse of the Somali state through regional peace and state building sometimes also followed the segmentary lineage model.5 The clearest example of this is the formation of the Puntland State of Somalia. It was set up in northeastern Somalia as regional state run by members of the various Harti clans. Harti is a nodal point within the genealogy of the Darood clan-family. While descendants of Darood reside all over the Somali territories, not only inside Somalia but also in eastern Ethiopia and northeast Kenya, the descendants of Harti are mostly concentrated in northeastern Somalia (except for a few groups living in southern Somalia, in the port town of Kismayo). Harti as a nodal point combines the three large clans of the Majeerteen, Dhulbahante and Warsangeli, plus several smaller groups, among them the Deshiishe, Kaskiqabe and Libangaashe. In the early 1990s, the Majeerteen, Dhulbahante and Warsangeli had their own militias, largely defending their own small clan-homelands (Somali sing.: degaan) against intruders. The Majeerteen in particular had in Cabdullahi Yuusuf,6 a former colonel of the Somali military, their own prominent warlord who was actively engaged in Somali politics, competing with the warlords of other factions such as the Hawiye/Habar Gedir leader Maxamed Faarax Caydiid. Yet, after years of fighting and several externally driven peace conferences had not yielded a lasting and comprehensive solution for the Somali governance crisis, political actors in northeastern Somalia called for an interclan conference (Somali sing.: shir) to set up a regional administration that in the absence of a stable Somali government could govern in the northeast and also safeguard the rights and interests of an (imaginary) Somali state. For several months, traditional authorities representing important lineages within the various Harti clans sat together and on 1 August 1998, the Puntland State of Somalia was announced as a

Beyond ‘African Political Systems’?  147

regional administration (Somali sing.: maamul goboleed). The political vision was that Puntland would be the first federal state of a future federal republic of Somalia (which became a reality only years later, when at a peace conference in Kenya in 2004 a federal system for Somalia was agreed upon, which was inscribed in the current – but still in many respects provisional – Somali Constitution of 2012). In practice, power within Puntland was divided between the various Harti clans according to perceived seniority and political ‘weight’. The Majeerteen as the ‘oldest’ son of Harti received the presidency, important ministries, and also the leadership positions within the various military and police forces. The Dhulbahante as the ‘second son’ received the deputy positions (the vice presidency, vice commanders of the military and police units) and the ministry of the interior. The Warsangeli as the ‘third son’ received the position of the speaker of the parliament and several influential ministers. The Deshiishe and other smaller groups received some midrange ministries; the Deshiishe also held the position of major of Bosaso, the largest port town and economic hub of Puntland. In the following years, the government positions were changed several times. Before presidential elections (every four or five years), the lineage heads nominated the members of parliament (sometimes renominating the incumbents, sometimes changing the personnel). The parliament then reconvened to elect the president, the vice president and the speaker of the house. The president then nominated the new cabinet. Interestingly, without any formal regulation, the basic formula for allocating positions within the Puntland state system has not changed for two decades (with changes of power in 2005, 2009, 2014 and 2019). For all these years, it was an ‘unwritten law’ that within the Harti state, power was divided according to considerations of seniority among the collective of agnates. Moreover, within the different Harti clans, various dominant lineages received their share over the years. For instance, while the first president, Cabdullahi Yuusuf (1998–2004), was from the Cumar Maxamuud lineage, the second president, Cadde Muuse (2004–9), was from the Cusman Maxamuud lineage, and the third president, Cabduraxmaan Faroole (2009–14), was from the Ciise Maxamuud lineage. The cycle started again with Cabdiwali Gaas (2014–19), a member of the Cumar Maxamuud lineage, and is currently continued by Siciid Cabdullahi Deni (2019–) who belongs to the Cusman Maxamuud lineage. Together, these three lineages constitute the backbone of the Majeerteen clan. I argue that this informal formula of clan-andlineage power-sharing provided Puntland with considerable stability in the face of the ongoing violence and instability of southern Somalia. Finally, the segmentary logic at work in Puntland also positioned it clearly against Somaliland in the northwest of collapsed Somalia. Somaliland had already seceded unilaterally from Somalia in 1991. It took years to become stable. In contrast to Puntland, Somaliland claims independence based on its separate colonial history (like Eritrea). It was a British Protectorate until 1960, when it united with the Italian Trusteeship territory of Somalia to form the Somali Republic. Somalil-

148  Markus V. Hoehne

and comprises various quite distant descent groups that do not connect at any recognizable nodal point in the patriline (its population comes largely from the Dir, the Isaaq and the Darood clan-families). Thus, its political basis is territorial, not genealogical (as is the case for Puntland). Somaliland clashes with Puntland over controlling the regions of Sool and Sanaag (see figure 7.1 above) inhabited by the Dhulbahante and Warsangeli. The government of Somaliland argues that these regions were part of the British Protectorate. Puntland argues that, as the Dhulbahante and Warsangeli are part of the Harti clan-collective, their clan homelands belong to Puntland. Behind these competing territorial claims, much larger political issues are at stake. I have argued elsewhere (Hoehne 2009) that Puntland seeks to safeguard Somalia’s unity by preventing Somaliland’s independence. The most effective means to prevent Somaliland from breaking away completely is to undermine its claim to clearly demarcated boundaries (from colonial times). The claim to administer all Harti-inhabited areas includes, ‘naturally’, the Dhulbahante and Warsangeli lands. This claim is effective because many Dhulbahante and Warsangeli feel marginalized in Somaliland, where they are dominated by the powerful Isaaq clans. The multiparty democracy introduced in Somaliland in 2001 even further cemented Isaaq domination (Isaaq candidates win all elections, since their clan family is the majority in the region and their patrilineal relatives predominantly vote for them; thus the most important positions remain in Isaaq hands, also under conditions of ‘free and fair’ elections). The example of the Puntland State of Somalia shows that the segmentary lineage model can explain more than conflict dynamics, as it has been the main focus of its ‘inventors’ and their disciples for decades. To push this point even further, segmentary lineage societies are not only confined to ‘ordered anarchy’, as Evans-Pritchard had suggested, but can even, under certain conditions, form states. This, at least, has been illustrated by the Puntland State of Somalia, which was set up in 1998 as a regional state of a clearly defined group of agnates. Within Puntland, at least for two decades after its foundation in 1998, positions of power were divided following a genealogical formula that was unwritten but highly effective. In my analysis, the periodic elections in Puntland served to suggest to outside observers (including donors) that there was some choice, but behind the scenes, the important political actors knew that the choices were limited; at least the patrilineal descent lines of the important office holders were premeditated already.

Taking over Laascaanood (Eastern Somaliland/Western Puntland) Laascaanood is the largest town in the Dhulbahante-inhabited territories in northern Somalia. For many years, neither Puntland nor Somaliland effectively controlled it. The situation changed when Puntland police forces occupied ­Laa­scaa­nood in late December 2003. The man in charge of the operation was Cabdirisaq Af-Guduud. He was about forty years old, belonged to the Majeer­ teen/Cabdiraxiin subclan and was commander of the Ciidamka Poliska (Po-

Beyond ‘African Political Systems’?  149

lice Forces) of Puntland. I was in Laascaanood when the first police technicals (pick-up trucks with heavy machine guns mounted on top) rolled into town. Officially they came to intervene in a bloody conflict between two Dhulbahante lineages in the countryside. Soon, however, it became clear that the forces were there to stay. In the first days of this ‘creeping’ occupation, Cabdirisaq Af-Guduud held many meetings with elders and other influential members of the local community. He established a qaad-chewing circle in town that served his aim of establishing connections. Qaad is the local drug, a stimulant imported in this case from Ethiopia and chewed by men in the afternoon hours; qaad-chewing sessions is where, besides other issues, politics are discussed and made. Cabdirisaq Af-Guduud’s mission was aided by the fact that he had Dhulbahante/Nuur Axmed half-siblings. After having been divorced by his father, his mother (who was also Majeerteen) married a Dhulbahante/Nuur Axmed man. Until he was a teenager, Cabdirisaq Af-Guduud grew up with his half-siblings in Laascaanood. From that time, he still knew a number of local men who by the time he entered the town as police commander of Puntland had become well-established businessmen, nongovernmental organization (NGO) workers or politicians. This cer­tainly eased Cabdirisaq Af-Guduud’s and his forces’ (most of whom were Majeerteen by clan) way into Laascaanood. These events indicate that Harti solidarity (between

Figure 7.2  Map of territories of clan-families and some clans in northern Somalia (© Markus V. Hoehne)

150  Markus V. Hoehne

Dhulbahante and Majeerteen) was not just an ideological discourse or a ‘folk model’, but actually provided a basis for concrete interactions. However, Harti solidarity was not ‘mechanical’. While the Puntland troops were initially welcomed as ‘Harti-brothers’, after a few days, local community leaders urged the government in Garoowe (Puntland) to withdraw its forces again since their presence would provoke a military reaction by Somaliland. Also, the conduct of the Puntland forces in town was rather rough – the soldiers, who were not regularly paid by their own administration, demanded free food and qaad on the local market, and their patrolling technicals ran over several people in the streets, killing some children. However, after four weeks, most people in Laascaanood had adapted to the new situation. One reason for this was, as several older locals emphasized, the special relationship mentioned above that the police commander Cabdirisaq Af-Guduud had and used for his operation – through half-siblings and their agnates from another clan, but related through their common mother. Another factor that contributed to the acceptance of the new forces in town was not related to kinship or segmentary lineage systems as a model at all: it refers to a certain ‘political vision’ of many locals in Laascaanood, who were strong Somali nationalists – and Puntland stood for pursuing the dream of a united and strong Somalia. I argue that many locals in late 2003/early 2004 pragmatically adapted to the fact that a new phase, which might bring about military clashes over the future of Somalia as one state or as divided between Somaliland and the ‘South’, had begun. Numerous locals prepared for war against Somaliland forces (which were positioned some 30 kilometres west of Laascaanood), polishing their guns and checking the technicals (i.e. pick-up trucks with heavy machine guns mounted on top) owned by lineages and parked in some backyards. They had last been used in the civil war in the late 1980s and early 1990s. The political vision of a united Somalia was certainly nurtured by the fact that at the time of the military takeover of the town by Puntland, an international conference for Somalia was underway in Kenya (Schlee 2006). At this conference, and under the auspices of the Intergovernmental Authority for Development (IGAD), a national and unitary Somali government was to be established. Cabdullahi Yuusuf, the founding President of Puntland, was one of the candidates for the Somali presidency (and he eventually was elected Somali President in August 2004). Against this backdrop, ‘Harti hopes’ were high, which certainly also influenced the situation in the contested borderlands between Somaliland and Puntland. In my view, the establishment of a nascent Puntland administration in Laascaanood in 2004 illustrates that the segmentary lineage model was employed consciously by the local decision-makers. The selection of Cabdirisaq Af-Guduud as the person leading the troops into town showed that Harti solidarity was not considered a sufficient precondition for the move. Those sending him, I suspect, had carefully considered his personal ties to important locals, established during

Beyond ‘African Political Systems’?  151

childhood. Thus, cross-cutting ties created through the second husband of the mother, Cabdirisaq’s stepfather, were used to buttress the bold and risky move of the Puntland government in order to make this intrusion acceptable among the locals. In combination with the larger Somali politics underway in those days at the peace conference in Kenya, this became a success story, at least for several years. Somaliland’s forces were kept at bay and until 2007, Laascaanood and the surrounding areas were under Puntland’s control, which not only strengthened Harti politics but also weakened Somaliland’s claim to independence.

Politicizing Local Feuds The following example goes back to events between 2011 and 2015. By chance, I was involved in an incident in February 2015 that crystallized the important problems at stake with regard to local conflict dynamics and peace-making. It also illustrated something I would call ‘anomie in the feud’ (not ‘peace in the feud’; Gluckman (1955)). Anomie refers to ‘social instability caused by erosion of standards and values’.7 This only works as a concept if one can also find some ‘normalcy’ in the feud. Indeed, there are certain ideal rules governing feuding or warfare among Somalis (which have been broken many times, but still, a notion how things ‘should be’ exists). The pretext of my recent stay in the area was that I had been hired as a consultant by the Danish Demining Group (DDG). My task was to conduct a

Figure 7.3  Simplified genealogical sketch of the Dhulbahante clan (© Markus V. Hoehne)

152  Markus V. Hoehne

local conflict analysis in order to establish how one could work in the contested borderlands, particularly in Taleex, and what would be the preferences of local people concerning development projects. Before and during the mission (I was part of a local DDG delegation visiting Taleex and other places), I collected information on an ongoing intergroup conflict that had remained unresolved for several years and complicated all aspects of social life in the area. It was clear that as long as this conflict was ongoing, any DDG project was going to fail. At the local level, in Taleex and its surroundings, the main conflict drivers (at the moment) were two lineages of Nuur Axmed, namely Faarax Cali and Samakaab Cali (see figure 7.3 above). Both had been engaged in a cycle of revenge killings since 2011. The beginning of this conflict was related to a clash between Somaliland and Puntland forces in a place called Mataano near Caroole (southwest of Taleex). On both sides, Dhulbahante fighters were involved. In this clash, the Puntland side captured several technicals of the Somaliland troops and took a number of the Somaliland soldiers as prisoners of war. Among them were injured Faarax Cali fighters. They were imprisoned in Garowe, the capital of Puntland, for several months. At that time, Cabdisamid Cali Shire was Vice President of Puntland. He was an old Somali military officer and was once Vice Commander of the Puntland forces (Ciidanka Darawiishta). By patrilineal descent, he is Samakaab Cali from Taleex. Upon their release, the Faraax Cali soldiers went back to Taleex, ganged up with their close relatives and in late 2011 ambushed and killed two men from Samakaab Cali who were close relatives of Cabdisamid Cali Shire. The attempts to settle this conflict became complicated because the governments of Somaliland and Puntland interfered in the process. One reason for the killings was that Faarax Cali men saw the two technicals lost in the previous clash to Puntland as belonging to their lineage. Also, some of the injured Somaliland soldiers, who happened to be Faarax Cali, had not been treated adequately in Garowe prison. Political/military and family/militia dynamics became mixed up. In mid-2013, Cabdisamid went to Taleex as an ‘elder’ to settle the dispute. Eventually it was agreed that the Faarax Cali lineage should pay 120 camels per killed man from the Samakaab Cali lineage and additionally a US$30,000 fine for ‘killing in a bad way’ (dil xumo).8 Part of the mag/compensation called jiffi (usually one-third of the camels agreed upon) was paid, as well as the fine. But still some camels were missing. It is unclear whether compensation was paid to Faraax Cali for their losses in property (e.g. technicals). After the settlement, Cabdisamid Cali Shire used the positive mood in Taleex to advance his re-election campaign as Vice President of Puntland (elections were due in January 2014). However, this was not well-received by the local population, particularly many members of the Faarax Cali lineage, who were on the side of the Khaatumo State of Somalia. This was an autonomous clan administration of almost all Dhulbahante. It had been established in 2012, after many

Beyond ‘African Political Systems’?  153

Dhulbahante had developed the feeling that Puntland was actually not seriously defending them against Somaliland’s growing military power in the region. Following up on where I left it above: after Puntland had occupied Laascaanood in late 2003, it stayed in control until November 2007. Then, for reasons I have outlined elsewhere (Hoehne 2015: 68–77; see also Mahmood 2019: 6), Somaliland forces moved into town. Despite considerable local resistance, the ‘Isaaq’ troops remained in control and the government of Puntland did not set out to reconquer the town. Desperate about this situation, local activists, supported by the Dhulbahante diaspora, mobilized their people to create their own administration, which was to be autonomous and to keep its distance from Somaliland and Puntland. Between 2012 and 2015, the Khaatumo State of Somalia existed. Its troops fought actively against Somaliland forces in Dhulbahante lands. The troops also opposed Puntland, which created a considerable nuisance for the Harti leadership that was trying to speak with ‘one voice’ in southern Somali power politics (in which Puntland, in contrast to Somaliland, always had a stake) (Hoehne 2015: 99–118). Thus, in 2013 Cabdisamid actually had to leave Taleex without having gained the local support for his political campaign in Puntland. Simultaneously, Maxamed ‘Indhosheel’, then President of the Khaatumo State of Somalia, advanced from Boocaame to Taleex with some fighters to control the situation. On their way, the Khaatumo forces were attacked by Cabdisamid and sections of the Puntland army. The latter consisted of soldiers from various Harti clans, but many of them were Majeerteen. The Puntland forces did not kill any of the Khaatumo militias, who decided to flee to Taleex. Cabdisamid’s forces chased them and entered the town. Indhosheel and the forces close to him fled, but locals in Taleex put up resistance. A shoot-out started in which seven men from Faarax Cali, one from Wacays Cali and one Majerteen soldier were killed. It is worth noting that Cabdisamid considered this attack on Taleex not as part of an intraclan feud, but as an attack of Puntland state forces on rebels (i.e. the Khaatumo militias). But the locals interpreted it differently. For them, Cabdisamid had come as Samakaab Cali leader and had killed locals from other lineages, mainly Faarax Cali. This effectively undid the settlement Cabdisamid had engineered shortly before. Consequently, the incident triggered new revenge killings at the local level. In 2014, an elder of the Samakaab Cali was killed by some Faarax Cali men upon leaving the mosque in Taleex after the morning prayer. In addition, several men from both sides were injured in attacks during 2014. All of these issues had not been settled in early 2015. When the DDG mission arrived, a peace conference (Somali: shirka nabadeynta) was underway that sought to resolve these issues. The conference was presided over by a council, headed by Khader Cali Jaamac, a businessman-cum-religious leader from the Naaleeye Axmed branch of Dhulbahante/ Maxamuud Garaad. The Naaleeye Axmed reside in the vast area stretching from Laascaanood to Ceerigaabo (see figure 7.1 above); they are probably the largest

154  Markus V. Hoehne

Figure 7.4 Elders under the ‘peace tree’; second from left, with white hat: Caaqil Shiine (Samakaab Cali) (© Markus V. Hoehne)

Dhulbahante subclan. Key players in the peace conference were Caaqil Cabdirashid ‘Shiine’, the highest local representative of the Samakaab Cali l­ineage, Wakiil Caaqil Cabdiqaadir Abyan ‘Koore’, the representative of the Faarax Cali lineage leader (the caaqil himself resided in Denmark at that time), a man called Xadeed from Naaleeye Axmed/Bihina Arale, who is a manger of Telesom, the biggest telecommunication company in the region, a man called Maxamuud Cumar Ciise ‘Hiirad’ (from Cumar Wacays lineage), and Cabdirisaq Cumar Cali ‘Ayr’ a businessman from the Wacays Cali lineage. A respected religious leader, Sheekh Siciid Axmed Adare from Jidali near Ceerigaabo (Naaleeye Axmed), was also involved. It is worth nothing that, due to the absence of any overarching traditional or religious authority in Taleex, authorities from other Dhulbahante lineages (particularly Naaleeye Axmed) were called upon to assist with settling the complex conflict. Moreover, it is striking that mostly religious men and businessmen led the peace conference, and not traditional authorities. People in Taleex explained that ‘people go with the money. The businessmen pay the expenses of the meetings’. Certainly, there was also a religious undercurrent to the situation: Islamists from the Muslim Brotherhood and other Salafi-oriented groups sought to enter into lineage politics and thereby to strengthen a reformatory element in Somali society. The idea was that many conflicts in Somalia were related to excessive wrangling between descent groups, and that religious orientations and authority could provide unity and stability. The peace conference was expected to conclude after our visit, on around 18 February 2015; in fact, it concluded later. The

Beyond ‘African Political Systems’?  155

­ amage done to each lineage by the other comprised not only men killed and d injured, but also cars and houses damaged in the exchange of fire (particularly during the capture of Taleex by Puntland forces in late 2013). In monetary terms, the damages ran close to US$500,000. Aside from lineage representatives from Taleex and its surroundings and aside from the Somaliland police car that had escorted us, no presence of any administration was visible in Taleex during our visit. Even the police commander Cabdinasir Jaamac Bulaale emphasized that he acted more as a local clan-­policeman than an official Somaliland representative. Genealogically, he belongs to Wacays Cali and he and his forces were tasked with guarding the peace in town. The men from the Wacays Cali lineage served as a classical ‘third party’ between two other local lineages, safeguarding the peace process. Khaatumo representatives did not appear anywhere, despite the fact that the Khaatumo State of Somalia had been founded in Taleex in January 2012. Puntland representatives were also completely invisible. Due to the involvement of all local/regional ‘administrations’ in the escalation of the conflict (after Puntland and Somaliland had been involved at the beginning, Khaatumo had also taken part), no official ‘showed his head’. At the market, several young boys were carrying Kalashnikovs. They belonged to the two lineages in conflict. When I asked why they were armed, their answer was: ‘We guard ourselves.’ When we visited Caaqil Cabdirashid ‘Shiine’ the next day his house was also guarded by an armed young man. This had to be taken as a very bad sign. If even the caaqil was afraid for his life, the issue under discussion must be extremely painful and complicated. Traditionally, elders, as well as guests, women and delegations, are ‘bir ma geydo’ (‘not to be injured/killed’) in conflict situations. Still, before we left, hopes were high that the very next day the longstanding feud between close agnates (involving political interferences from far-away ‘governments’) could actually be resolved. Thus far, it had become clear that the ongoing conflict in the area was characterized by overlapping layers of fighting – first between Somaliland and Puntland forces that, as usual in the contested borderlands, were mainly staffed by members of local lineages. Their fighting therefore led to the transformation of a ‘national’ conflict (between two ‘governments’) into an intergroup conflict that, in this particular case, was a feud between close agnates residing together in Taleex and its surroundings. When Khaatumo became involved after the formation of this clan administration in 2012, the situation became even more complex. While Khaatumo was supposed to be clan-inclusive (for all Dhulbahante), the reality on the ground was that the feud hindered the stabilization of the clan administration. This provided Puntland with a chance to undermine the rival that, if it succeeded, would put an end to joint Harti politics in the region. On our way back to Laascaanood on 14 February, an incident occurred that illustrated the ‘anomie in the feud’ characteristic of the complex conflict dynamics in the area. Several ‘camel boys’ (Somali sing.: geel jire) from the Faarax Cali

156  Markus V. Hoehne

lineage waylaid us and demanded to search our cars for a young man from the Samakaab Cali lineage. They had been tipped off by someone in Taleex. Indeed, we had a young man with us who was called Ayaanle (which translates as ‘with luck’, which was indeed an appropriate name). Ayaanle was a close patrilineal relative of Cabdisamid Cali Shire and therefore a ‘high-profile target’ in the current conflict. We had consciously invited him after we had heard that the conflict was about to be settled and Muna, a woman belonging to the Samakaab Cali, who was working for DDG and was part of the mission, had demanded that we take a man from her lineage as well, given that I had earlier brought in a close friend of mine from the Faarax Cali to help us with paving the way into Taleex. ‘My man’, called Maxamuud Ilweyn, was indeed the maternal uncle (Somali: abti) of Ayaanle. In Taleex, this balance between the lineages within our delegation had worked out. But on the way back, it became a problem. If those who waylaid us had found Ayaanle, they would have executed him in revenge for some Faarax Cali men killed earlier. In this quite tense situation – we, in our jeeps, facing around ten armed men who wished to kill one of our team m ­ embers – the ‘elders’ of our delegation, Axmed ‘Daakir’, Maxamuud ‘Ilweyn’, Maxamed ‘Qoryooley’ and one of our drivers called Shiine, who all were unarmed, ran towards the camel boys and engaged them in a heated discussion. Two shots were fired in the air. Maxamud Ilweyn, a very famous man in the area and a lineage relative of those who waylaid us, grabbed one man who wanted to walk towards

Figure 7.5  The members of our delegation; third from the left (standing): Ayaanle, the target of the killers; on his left: Axmed ‘Daakir’; on his right: Maxamuud Ilweyn (© Markus V. Hoehne)

Beyond ‘African Political Systems’?  157

our cars and stopped him. The driver, Shiine, also from the Faarax Cali lineage, recognized one close patrilineal cousin among the attackers and slapped him in the face, shouting that our delegation was ‘bir ma geydo’ (sacrosanct) and that the conflict was being settled. Eventually, our men hurried back to us and stressed that we had to escape with the cars before something disastrous could happen. Off we went with the four-wheel drives, another way through the desert. Those who had waylaid us had no car to follow. But for almost two hours, everyone in our car was so nervous that the attackers would call relatives elsewhere in the countryside and motivate them to attack us. Despite the remoteness of the area, virtually everyone had a mobile phone and there was a reasonable net-coverage. During the rough ride, I overheard the analysis of my friends in Somali about what had happened. What I understood was that the patrilineal affiliations and personal histories of most of those who had waylaid us were discussed in detail and their relations to the conflict with the Samakaab Cali were outlined. It turned out that some of the men had lost close relatives in the feud between 2013 and 2014. Regarding the peace process underway in Taleex, it was clear that if the camel boys had succeeded, the negotiations would have collapsed. The incident showed us that the discussions in Taleex had not yet reached the countryside. The conflict was still ongoing (as obviously someone from Taleex must have informed the camel boys about our delegation and its staff). Moreover, the power of the elders, religious leaders and businessmen in Taleex was obviously chiefly confined to the town. In sum, this event shows how political tensions and competition between Somaliland and Puntland influence local conflict dynamics. The latter oscillate between ‘family affairs’ and ‘national political affairs’, and even involve diasporic and transnational actors in the form of Khaatumo supporters, since the Khaatumo administration is strongly backed (also financially) by members of the Dhulbahante clan (mainly) residing abroad. It also shows that from the perspective of local actors, the segmentary lineage model is used to guide action and make sense of incidents.

Conclusion From an early twenty-first-century perspective, it is clear that the 1970s and 1980s critiques of Peters, Kuper and others that ‘there do not appear to be any societies in which vital political or economic activities are organized by a repetitive series of descent groups’ (Kuper 1982: 92) is mistaken. On the contrary, at a more general level, anthropological studies into ‘tribal societies’ in which kinship is an important factor organizing politics has gained renewed attention, particularly on the part of US and European military strategists trying to fight ‘insurgency’ and ‘terrorists’ in Iraq and Afghanistan. They were aided by some anthropologists (mainly from the United States) who were eager to employ, for

158  Markus V. Hoehne

e­ xample, ­segmentary lineage systems as a model to strategize how segments of local societies could be turned into allies (Zitelmann 2018). Moscona et al. (2020) recently engaged in an ambitious study that attempted to establish how segmentary lineage organization and conflicts in Sub-Saharan Africa are correlated. The idea was to gain insights into obstacles for development in the region. Moscona et al.’s economics and political science perspectives draw heavily on political anthropological literature, starting with The Nuer and African Political Systems. Finally, I provided my own case studies in this chapter in order to argue that the segmentary lineage system as a model, combined with other (dynamic) social, political and other factors, has considerable explanatory power. Of course, northern Somalia, where I worked, is seemingly a quite ‘favourable’ ethnographic setting for making such a point. However, there are numerous researchers (Somali and non-Somali) who actually perceive anything referring to ‘clan’ as a divisive (neo-) colonial construct. There is a strong sentiment, particularly among Somalis in the diaspora, to disregard ‘clan’ and ‘clan analysis’ (which my presentation would certainly qualify as) and anything related to this. This is illustrated by Cawo M. Abdi (2015), who in a debate about the future of Somali Studies argues: Most relevant for Somalia, I.M. Lewis’s reductionist analysis of Somali clan structures in great part informs the unworkable clan political dispensation and regional fiefdoms that currently characterize Somalia. The international community, with Somali sectarian warlords and politicians, continues to rely on this outdated and problematic understanding of Somali social structures that was produced by British colonial academic experts. Against this criticism within Somali studies, but also against some of the criticism in social anthropology – most prominently voiced by Kuper – I stick to this model. However, I – and here I follow Baştuğ – do so not in the form of a theory, as establishing causal relations between concepts and allowing to some degree the prediction of social practices. Rather, and in combination with a methodology that privileges long-term field research and dense participation (Spittler 2001), I see the segmentary lineage system as a useful heuristic device to make sense of individual behaviour in a field characterized by changing power structures, situational reference to various forms of kinship, and diasporic and transnational political engagement. It was, in the case I outlined above, also used by local actors themselves to analyse their own situation. This shows that, contrary to Leach (1961) and, following him, Holý (1979), the segmentary lineage model refers to more than ‘anthropologists’ ideas’; indeed, it is related to empirically observable practices. Ibn Khaldun already found in the fourteenth century CE that a pure genealogical tree is useless. Bourdieu (1977: 38, cited in Jama Mohamed 2007:

Beyond ‘African Political Systems’?  159

239), many centuries later, stressed that it was like ‘abandoned roads on an old map’. Jama Mohamed (2007), a Somali historian, outlined the radical difference between genealogy (Arab.: nasab) and group feeling (Arab.: asabiya). Referring to Ibn Khaldun, he argued: ‘What creates such togetherness is “social intercourse, common defense, long association and familiarity, and companionship in childhood and in all other aspects of life”’ (Ibn Khaldun 1999: 327, cited in Mohamed 2007: 239). Part of this togetherness in northern Somalia (and also in the diaspora, as Bjork (2017), shows) can very well be captured by the segmentary lineage system as a model. It needs to be stressed that what one could observe, particularly regarding the situation in and around Taleex (2011–15), was a decrease in complementary relations that would usually play a role alongside unilineal descent. This was part of the conflict escalation in a situation where stable overarching structures of governance were still missing. In a sense, my attempt to make sense of the ethnographic encounter around Taleex by referring to some time-honoured model of British social anthropology that nevertheless, I suggest, is still useful for capturing important aspects of contemporary conflict situations, is seconded by Gluckman, who discovered already some time ago that: As we appreciate more fully that culture is in fact to some extent a hotchpotch, and that customs and values are independent of one another, discrepant, conflicting, contradictory, we shall have to evolve concepts to deal with social life which are less rigid, and which can cope with lack of interdependence as well as the existence of interdependence, with the haphazard as well as the systematic. (Gluckman 1961: 161) I hope I have shown that regarding conflict dynamics in northern Somalia, the segmentary lineage system as a model can do exactly this: cope with the haphazard as well as the systematic. What I also find interesting about the Somali case studies briefly presented above is that one can see how the model is transformed in the course of ­local-global dynamics. The role of traditional leaders, for instance, who – it could be said – put the segmentary lineage model into practice during peace negotiations has changed dramatically (Hoehne 2006). Some became ‘like politicians’, which meant, they attached themselves to the political elites in the towns, were open for corruption and were less accountable to their rural constituencies. Others went abroad, like the Chief Caaqil of the Cali Faraax lineage from Taleex, and left the daily business to some representative who, in this case, lacked authority. These factors contributed to what I called ‘anomie in the feud’. It shows that, even if one can keep the ‘framework’ of the model, its ‘content’ (lineage solidarity, segmentary opposition, role of elders) changes in character over time.

160  Markus V. Hoehne

Markus V. Hoehne is a lecturer at the Institute of Social Anthropology at the University of Leipzig. He received his Ph.D. from the Martin-Luther University Halle-Wittenberg and worked at the Max Planck Institute for Social Anthropology in Halle (Saale). His research interests are identity, state formation, conflict, borderlands and transitional justice. He is author of Between Somaliland and Puntland: Marginalization, Militarization and Conflicting Political Visions (Rift Valley Institute, 2015) and co-editor of Borders and Borderlands as Resources in the Horn of Africa (James Currey, 2010), The State and the Paradox of Customary Law in Africa (Routledge, 2018) and Dynamics of Identification and Conflict: Anthropological Encounters (Berghahn Books, 2022).

Notes 1. A heuristic device is defined as: ‘Any procedure which involves the use of an artificial construct to assist in the exploration of social phenomena. It usually involves assumptions derived from extant empirical research. For example, ideal types have been used as a way of setting out the defining characteristics of a social phenomenon, so that its salient features might be stated as clearly and explicitly as possible. A heuristic device is, then, a form of preliminary analysis. Such devices have proved especially useful in studies of social change, by defining bench-marks, around which variation and differences can then be situated. In this context, a heuristic device is usually employed for analytical clarity, although it can also have explanatory value as a model.’ Retrieved 5 September 2021 from http://www. encyclopedia.com/doc/1O88-heuristicdevice.html. 2. Lewis was a student of social anthropology at Oxford University in the 1950s, where ­Evans-Pritchard was among his teachers. 3. Mechanical solidarity is usually based on kinship ties of familial networks. It involves relatively little interdependence between members of society and is often built on religious rules and collective sanctions. 4. http://www.encyclopedia.com/doc/1O88-heuristicdevice.html (retrieved 5 September 2021) 5. The following section is based on my long-term ethnographic engagement in the region, which is most comprehensively summarized in Hoehne (2009 and 2015). 6. Somali personal and place names follow Somali orthography in this text. C stands for Arabic ‘ayn’ and x stands for the emphatic ‘h’ in Arabic. Double vowels indicate long vowels (like ‘ii’). 7. http://www.thefreedictionary.com/anomy (retrieved 5 September 2021). 8. On the standardization/variation of compensation payments, see Schlee (2017: 350–53).

References Abdi, Cawo M. 2015. ‘The Audacity of “Somali Studies”’. Retrieved 5 September 2021 from https://africasacountry.com/2015/04/the-audacity-of-somali-studies. Ali Jimale Ahmed. 2014. ‘Introduction’, Journal of Somali Studies, 1(1): 5–10. Barth, Frederik. 1973. ‘Descent and Marriage Reconsidered’, in Jack Goody (ed.), The Character of Kinship. Cambridge: Cambridge University Press, pp. 3–19. Baştuğ, Sharon. 1998. ‘The Segmentary Lineage System: A Reappraisal’, in Joseph Ginat and Anatoly M. Khazanov (eds), Changing Nomads in a Changing World. Brighton: Sussex Academic Press, pp. 94–123.

Beyond ‘African Political Systems’?  161 Besteman, Catherine. 1996. ‘Representing Violence and “Othering” Somalia’, Cultural Anthropology, 11(1): 120–33. Bjork, Stephanie. 2017. Somalis Abroad: Clan and Everyday Life in Finland. Urbana: University of Illinois Press. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press. Cassanelli, Lee. 2010. ‘Speculations on the Historical Origins of the “Total Somali Geneaology”’, in Markus V. Hoehne and Virginia Luling (eds), Peace and Milk, Drought and War: Somali Culture, Society and Politics: Essays in Honour of I.M. Lewis. London: Hurst, pp. ­53–66. De Waal, Alex. 1997. Famine Crimes: Politics and the Disaster Relief Industry in Africa. Oxford: James Currey. Evans-Pritchard, E.E. 1940a. The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People. Oxford: Clarendon Press. ———. 1940b. ‘The Nuer of the Southern Sudan’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 272–96. Fortes, Meyer. 1979. ‘Preface’, in Ladislav Holý (ed.), Segmentary Lineage Systems Reconsidered. Belfast: Greystone Press, pp. vii–xii. Fortes, Meyer and Evans-Pritchard, E.E. 1940. ‘Introduction’, in Meyer Fortes and E.E. ­Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 1–23. Gellner, Ernest. 1969: Saints of the Atlas. Chicago: University of Chicago Press. Gluckman, Max. 1955. ‘The Peace in the Feud’, Past & Present, 8(1): 1–14. ———. 1961. ‘Ethnographic Data in British Social Anthropology’, Sociological Review, 9(1): 5–17. ———. 1965. Custom and Conflict in Africa, 2nd edn. Oxford: Basil Blackwell. Hoehne, Markus V. 2006. ‘Traditional Authorities in Northern Somalia: Transformation of Powers and Positions’. Working Paper No. 82. Halle/Saale: Max Planck Institute for Social Anthropology. ———. 2009. ‘Mimesis and Mimicry in Dynamics of State and Identity Formation in Northern Somalia’, Africa, 79(2): 252–81.c ———. 2015. Between Somaliland and Puntland: Marginalization, Militarization and Conflicting Political Visions. Nairobi: RVI. Holý, Ladislav. 1979. ‘The Segmentary Lineage Structure and Its Existential Status’, in Ladislav Holý (ed.), Segmentary Lineage Systems Reconsidered, vol. 4. Belfast: Queen’s University Papers in Social Anthropology, pp. 1–22. Ibn Khaldun. 1999. Al-Muqaddimah. Beirut: Dar al-Kitab al-Lubnani. Kapteijns, Lidwien. 2004–10. ‘I.M. Lewis and Somali Clanship: A Critique’, Northeast African Studies, 11(1): 1–23. Kuper, Adam. 1982. ‘Lineage Theory: A Critical Retrospect’, Annual Review of Anthropology, 11: 71–95. Leach, Edmund. 1961. Pul Eliya: A Village in Ceylon. Cambridge: Cambridge University Press. Lewis, Ioan M. 1961. A Pastoral Democracy. A Study of Pastoralism and Politics among the Northern Somali of the Horn of Africa. Oxford: Oxford University Press. ———. 1965. ‘Problems in the Comparative Study of Unilineal Descent’, in Michael P. ­Banton (ed.), The Relevance of Models for Social Anthropology. London: Travistock ­Publications, pp. 87–112. Mahmood, Omar S.. 2019. ‘Overlapping Claims by Somaliland and Puntland: The Case of Sool and Sanaag’, Institute for Security Studies, East Africa Report 27.

162  Markus V. Hoehne Mohamed, Jama. 2007. ‘Kinship and Contract in Somali Politics’, Africa, 77(2): 226–49. Moscona, Jacob, Nathan Nunn and James A. Robinson. 2020. ‘Segmentary Lineage Organization and Conflict in Sub-Saharan Africa’, Econometrica, 88(5): 1999–2036. Peters, E.L. 1967. ‘Some Structural Aspects of the Feud among the Camel-Herding Bedouin of Cyrenaica’, Africa, 37(3): 261–82. Salzman, Philip Carl. 1978. ‘Does Complementary Opposition Exist?’, American Anthropologist, 80(1): 53–70. Samatar, Abdi I. 1992. ‘Destruction of State and Society in Somalia: Beyond the Tribal Convention’, Journal of Modern African Studies, 30(4): 625–41. Schlee, Günther. 1997. ‘Cross-Cutting Ties and Interethnic Conflict: The Example of Gabbra Oromo and Rendille’, in Katsuyoshi Fukui Eisei Kurimoto and Masayoshi Shigeta (eds), Ethiopia in Broader Perspective: Papers of the XIIIth International Conference on Ethiopian Studies. Kyoto: Shokado Book Sellers, pp. 577–96. ———. 2006. ‘The Somali Peace Process and the Search for a Legal Order’, in Hans-Jörg Albrecht, et al. (eds), Conflicts and Conflict Resolution in Middle Eastern Societies: Between Tradition and Modernity. Berlin: Duncker & Humblot, pp. 117–68. ———. 2017. ‘Customary Law and the Joys of Statelessness: Somali Realities beyond Libertarian Fantasies’, in Bertram Turner and Günther Schlee (eds), On Retaliation: Towards an Interdisciplinary Understanding of a Basic Human Condition. Oxford: Berghahn Books, pp. 319–60. Sigrist, Christian. 2004. Segmentary Societies: The Evolution and Actual Relevance of an Interdisciplinary Conception. Mitteilungen des SFB 586 ‘Differenz und Integration’ 6. Spittler, Gerd. 2001. ‘Teilnehmende Beobachtung als Dichte Teilnahme’, Zeitschrift für Ethnologie, 126(1): 1–25. Zitelmann, Thomas. 2018. ‘Kinship Weaponized: Representations of Kinship and Binary Othering in U.S. Military Anthropology’, in Tatjana Thelen and Erdmute Alber (eds), Reconnecting State and Kinship. Philadelphia: University of Pennsylvania Press, pp. 61–86.

Chapter 8

Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan Munzoul Assal

Introduction Very little is known about the Tuareg in Sudan. Locally known as Kinin, they came to Sudan at the beginning of the twentieth century. Archival records show that they entered Sudan running away from French troops who fought the Tuareg rebels in Mali at the turn of the twentieth century. Defeated by the French, some 10,000 Tuareg rebels reached Darfur and sought refuge and protection of its Sultan, Ali Dinar. Although wary of their militant nature and raiding propensity, Ali Dinar took their arms and allowed them to settle in the southern fringes of present-day El-Fashir. With the defeat of the Sultan and the collapse of the Fur Sultanate in 1916, and with the French victory over the last Tuareg revolt in 1917, most of them left Darfur to Mali (McGregor 2013). However, some decided to stay behind, forming the basis for the present-day Tuareg community in Darfur. They are mainly scattered in villages in Northern Darfur, around the towns of El-Fashir, Tawila and Kutum. They practise farming and animal husbandry, but in recent years some joined the growing urban informal sector in the major towns in Darfur and further afield. Like other ethnic groups in Sudan, the Tuareg are divided into clans, so to speak. Their main clans are Emodas, Igdalan (who are viewed as the descendants of slaves), Enizghaman, Kazkazan, Ekwaddaran and Enizghirsan. In addition to these clans, the Tuareg also have other names that are either related to place or occupation. For example, kailaiair means ‘those of the north’, while kailaghayak means ‘those who farm’, etc. Like other groups in Darfur, the Tuareg were affected by war and the generalized state

164  Munzoul Assal

of insecurity, and they are at the moment displaced in major towns in Darfur. In fact, the village on which this chapter is based was deserted in 2006 following the intensification of war in Darfur and the burning of villages perpetrated by government-allied militias, the Janjaweed. Inhabitants of Maagila village are now displaced persons living in the Abuja camp in El-Fashir. This chapter provides some preliminary notes on the Tuareg of Sudan. Although the analysis presented here is based on information from one village, it sheds light on some general issues relating to the economic and political organization of the Tuareg, their place within the region, and how they access resources in one of the conflict-prone regions of Sudan. While the information on which this chapter is based was gathered some years ago, an attempt is made to provide an update on the current situation, and the Tuareg’s response to the war and polarization in Darfur. The chapter begins with a brief historical background and then discusses their economic and political organization, kinship and marriage. It then moves on to discuss the evolution of the crisis in Darfur. The discussion about the Tuareg cannot be done without analysing or providing some sort of an account about the current situation in Darfur. The final section of the chapter deals with the impact of the crisis in Darfur on the Tuareg, and the type of transformation they went through, how they fare in displacement camps in El-Fashir and what future prospects they have.

Background Information The presence of the Tuareg in Darfur has not received proper scholarly attention, despite dating back to the early years of the twentieth century. Very few ­accounts (Beck 1996; McGregor 2013) mentioned the Sudan Tuareg in connection to raiding activities at the turn of the twentieth century, and some r­ecent media reports in 2013 that alleged the movement of Tuareg rebels from West Africa (Mali) to Darfur. McGregor (2013) compared this to the movement of Tuareg rebels following their conflict with the French troops a century ago. ­Defeated by the French, the Tuareg began to move east to Darfur, where they arrived in 1909. They wanted to seek refuge and the protection of its Sultan, Ali Dinar. Numbering about 10,000, they were given sanctuary by Ali Dinar. Before rebelling against the French, the Tuareg were said to entertain a raiding lifestyle, which they wanted to continue in Darfur. Ali Dinar wanted to expel them, but was asked by the Sanusi of Libya to allow them to stay. Ali Dinar accepted this, but decided to disarm them. In this regard, argues McGregor: a number of Tuareg women were taken off to the royal harem and the Tuareg slaves were impressed into the Sultan’s own army. Most of the remaining Tuareg (known in Darfur as Kinin) were disarmed and resettled near the lead mines of Kutum in Jebel Marra … in 1913, the

Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan  165

Sultan softened his position and allowed the Tuareg to resettle a half-day south of the capital of Elfashir, while small units of Tuareg joined the Sultan’s army, most of the tribesmen did not, nor was any effort made to impress them, the Sultan being as wary of the Tuareg as they were of him. ­(McGregor 2013) Although they portray a peaceful image of themselves, Beck (1996), just like McGregor, argues that even during the period of colonial rule, the Tuareg were involved in raiding and clashes, particularly with the Kababish and the Gura’an groups. The Sultan of Darfur was defeated by the British troops in 1916 and was forced to flee El-Fashir, his capital; by the time British troops entered El-Fashir, the Tuareg had already arrived to begin the looting of the nearly deserted town. By 1917 and after the French defeat of the Kaocen Rebellion, many Tuareg returned to Mali and Niger – but apparently not all of them. The Tuareg (Kinin) of today are the second and third, or probably even the fourth, generation of those who remained in Sudan. The writings on the Tuareg focus on West Africa and there is no scholarly account about those who live in Sudan. This chapter is based on primary fieldwork carried out many years ago (Assal 1997b), with an update on the present situation. The account on Tuareg in Sudan presented here is told by Tuareg in Maagila village, one of the main Kinin villages some 60 kilometres west of El-Fashir, the capital of North Darfur State. Maagila Tuareg recount that there is a legend about the grandmother of Tuareg (the mythical ancestress of Tuareg), whose name was Tehanin, who established the feudal tribe – the Tuareg in southern Algeria. Tehanin descends from the kel Ghali clan. Tehanin established the timnokalat (Sultanate) of al-Ahgar. Her maid Tikmat is said to be the grandmother of the Igdalan clan, which is the clan of slaves. However, there is no account of Tehanin’s husband. The history of Tuareg is full of influential women, such as Tamghart and her maid Ghaida (Rasmussen 1995). Although some accounts (McGregor 2013) link the presence of Tuareg in Sudan to the rebellion in Mali, old Tuareg men in Maagila village argue that their fathers and grandfathers came to Sudan to support the Mahdi, a religious leader who fought the Turks and the British, and liberated Khartoum in 1885 (Holt 1958; Holt and Daly 2000). These accounts also say that not all the Tuareg came at the same time; their migration to Sudan continued up to the 1950s and 1960s in order to join relatives. Some were on a pilgrimage to Mecca, but decided to stay in Sudan. This is actually a practice of many groups of West African origin who settled in Sudan. Related to the West African dimension in Sudan, it should be noted that the different ethnic groups in Darfur also live in neighbouring countries. Members of groups like the Misseiriyya, Mahamid, Taalba, Eringa, Taayaisha, Zaghawa, Bedeyat and many others living in Darfur also live in Chad and the Central African

166  Munzoul Assal

Republic. The region is therefore quite volatile in terms of population mobility, and even more so during times of conflict and violence. When violence erupts, kin groups on the other side of the border cross to fight alongside their relatives. This means that violence in Darfur causes violence in neighbouring countries. In fact, in 1990, the current Chadian President, Idris Déby, marched from Darfur to Ndjamena and overthrew Hussein Habre. The Mahamid are said to have connections to Niger, and although this could not be verified, the Sudanese government has been accused by rebel movements of resettling Mahamid Arabs in the lands of indigenous groups in West Darfur. A distinguishing feature of the Tuareg is the tugulmus (veil), locally known in Darfur as kadmoul. Keenan (1974: 107) described the veil as a noteworthy characteristic of all Tuareg and, although there are differences in the style of the veil between various classes, the veil is the dominant feature of Tuareg identity. To the extent that Tuareg in Sudan are neither politically nor economically autonomous, the veil is withering away; some old men still use it today, but in the context of Darfur, the government banned the veil since rebel groups use it as a disguise (Takana 2016). The Tuareg have their own language, known locally by Sudan Tuareg as Tamagegh. As a matter of fact, they call themselves ‘tamagaghan’. They do not use the written version of Tamagegh, known as tefeinagh.

The Setting: Maagila Village Maagila is a small village some sixty kilometres west of El-Fashir. At the time of the study, Maagila had a population of 530. The village is made up of different clans. Emodas is the main clan with which political power rests. The sheikh of Maagila village descends from this clan. Other clans are Ainadan (who are perceived as blacksmiths), Enizghaman, Zayakan, Igdalan, Kazkazan, Ekwaddaran, and Nizghirsan. In addition to Maagila, there are other villages in North Darfur. In areas close to El-Fashir, there are the villages of Tawila (a small town), Tina, Gailai, Umgaras and Suwelingha. Some Tuareg also live in Kutum in the northern part of North Darfur. Until the 1960s, the Tuareg were nomads who did not engage in aghayak (farming) in significant ways like kel akal (indigenous people). Milk used to be their main source of nutrition, in addition to ainali (millet), from which they make eching (porridge). But from the 1970s, they shifted to cultivation. As a supplementary activity to generate income, they engage in work that includes charcoal and rope-making for men, and mats and basket-making for women. Men from the Ainadan clan make wooden handcrafts, especially grinding utensils – the most famous of which is fundug. It is used to grind millet and other dried items like okra and dried tomatoes. Making wooden utensils is considered to be degrading and is tantamount to blacksmithing. As a matter of fact, members of the Ainadan clan are looked down upon and are called hadahid (blacksmiths).

Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan  167

Members of this clan are rarely accepted as marriage partners for members of other clans. Unlike the case with other major groups in Darfur, like the Fur, Masalit and Rezeigat, the Tuareg have no land or dar that belongs to them or that is written in their name. Their access to land is therefore usufruct. And since land in Darfur is a depot of political power (the tribes with dar are powerful and this is also the case vis-à-vis the central government), the Tuareg are politically marginal. I will discuss this point later on in the chapter. Here, however, it is necessary to state that the Tuareg in Maagila village are under the authority of the late Shartai Mohamed Salih Ahmed Mandi, who is responsible for all villages west and northwest of El-Fashir up to Korma, a major trading centre between El-Fashir and Kutum. Shartai is a local name in Darfur used in the native administration system and is equivalent to paramount chief. While the Shartai is equivalent to the paramount chief, the Tuareg do not deal with him directly. They interact with local sheikhs in matters relating to the allocation of land and minor disputes. Suffice it to note that they have their own sheikh who is recognized by the authorities and by the Shartai, but who has no power or authority to allocate land, except in some insignificant cases. The practice in Darfur is that in each village (or a few adjacent villages), a headman or a sheikh, is responsible for the allocation of land plots for farming purposes to members of the village or villages for which he is responsible. Strangers and individual nonmembers can be allocated land on usufruct terms, and in return they should remit one-tenth of the produce at harvest to the sheikh (Harir 1994). Before the beginning of the current crisis in Darfur, there were no problems in terms of access to land. The reason behind this is twofold: first, the number of people was indeed small; and, second, young people tend to migrate to cities to look for employment there instead of staying at the village and farming the land.

Economic Organization The household is the production unit in both farming and herding. For herding, there is a clear division of labour: males look after animals while females milk them. In farming activities, the division of labour is less clear. Due to the droughts of the 1980s and armed robberies that culminated in 1987 with the killing of one of the Tuareg notables in Maagila village, the Tuareg abandoned herding that required moving away with animals. Those who continue to keep animals do so in combination with farming and by keeping a small number of sheep or goats that are kept in the village. The few camels that exist are used for transport and ploughing purposes. The Tuareg have abandoned raising camels. The main crops harvested include millet (the staple crop), sorghum, okra and tomatoes. These varieties are produced for household or family consumption and are rarely sold. But there is one important cash crop that the Tuareg started to

168  Munzoul Assal

cultivate in recent years: tobacco, locally known as tombak. This is an important cash crop not only for the Tuareg but also for the entire Darfur region, especially North Darfur. The economic organization of the Tuareg is relatively simple. No sophisticated technology is involved. No fertilizers are used and production is small scale. Even with cash crops like tobacco, the areas cultivated are small and, although hired labour is used at different production stages, communal labour is practised for both the staple and cash crops. Moreover, since production is small scale, the chances for the accumulation of wealth are negligible and the Tuareg can generally be seen as poor by local standards (Assal 1997a). There are factors that limit this accumulation. Although generally there is no land scarcity, land with good-quality soil is difficult to find. While millet is cultivated in sandy lands, tobacco requires clay lands along small seasonal streams and riverbeds. Therefore, there is some form of latent competition over fertile lands in riverbeds, with few people having good farming lands. A person receives a farming plot either through his family (fathers allocate farming land plots to their married or grown-up sons and also through inheritance) or from the village sheikh. Depending on a person’s network, he or she may receive a farming plot in an adjacent non-Tuareg village – either from the sheikh of that village or simply from a friend. This is accompanied by the understanding that the owner may claim back their land at any time. But usually such dispensation takes many years. Sometimes it seems to be indefinite. It is rare for unmarried children to have their own farming plots. The practice is that they work on the family farm until they are married. Although a wife usually works on her husband’s farm, she can have her own little garden where she grows okra, tomatoes and other vegetables for family consumption. In other parts of the region, gardening cultivation is geared towards the market (Umbadda and Abdul-Jalil 1985). A farmer may work directly on his farm with his family members, organizing communal labour parties at different stages of the production cycle, hire labourers or, if the farm or farms are big enough, organize other arrangements. One of them is subletting, whereby a farmer brings someone else to do the work and share the produce with them. The most frequent arrangements are bel-nus and bel-tilt (onehalf and one-third), in which the landowner gets either half or one-third of the production, respectively. In these arrangements, the landowner does not provide any input, except the land, where the party to the agreement does all the work and gets either half or two-thirds of the final produce. These arrangements are widely practised in Darfur and other parts of Sudan, and are therefore not particular to the Tuareg. They vary from one season to another and are contingent on a number of factors, including how good or bad the rainy season is, etc. Another thing that is practised by the Tuareg and other communities in Darfur is the shail system. This is a practice that involves someone borrowing either

Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan  169

money or commodities from a rich farmer or merchant at the beginning of the rainy season, with the obligation to pay back in kind immediately after finishing the harvest. Some scholars (O’Brien 1986) described this system in great detail and, again, like other arrangements, this is something that is practised in many parts of Sudan, and the Tuareg are part of wider systems of economic organization. In Maagila, there were two farmers (who were also merchants) who lent people either money or goods in return for produce. The shail has widely been seen as an exploitative system, to the extent that by the time of harvest, crop prices are so low that a farmer may end up giving all that he has produced to the merchant, who will in turn sell the crop or crops later for higher prices. The shail system thus works to the disadvantage of small producers. A further dimension of the economic organization involves one or more village merchants buying the produce from farmers in the villages and selling them at a later stage to other big merchants in local village markets, or even taking the products (especially in the case of tobacco) to the town. At times, agents for bigger merchants visit these villages and buy crops directly from farmers and small merchants. Although simple, the economic organization of the Tuareg is part and parcel of both local and regional processes that involve exploitation, profit-­making and accumulation. The war in Darfur has changed the situation dramatically, as most of the villages where the economic processes described in this chapter were occurring are currently empty. Their inhabitants have fled the conflict and have been reduced to internally displaced persons in major towns in Darfur.

Political Organization In the introduction to African Political Systems, Fortes and Evans-Pritchard (1940: 5) described African political systems as falling into two groups: A and B. Group A consists of those societies that have centralized authority, administrative machinery and judicial institutions – in short, a government – and in which cleavages of wealth, privilege and status correspond to the distribution of power and authority. Group B is made up of those societies that lack administrative structures and judicial institutions. In other words, they lack a government and there are no sharp divisions in terms of rank, status or power. The Tuareg of Darfur certainly fall into one group: Group A. The Tuareg are politically marginal in Darfur and in Sudan at large. One reason that explains their marginality is that they do not have land, which in Sudan is the main source of both economic and political power. As already mentioned, they are part of the traditional system under the reign of the Shartai, which is a form of traditional leadership equivalent to the Nazir, which is used by Arabic-speaking groups or tribes in Darfur and other parts of the country. They are, so to speak, clients to the Shartai, who does not interfere in their day-to-day affairs, while only major problems are reported to him.

170  Munzoul Assal

Villagers choose the sheikh, but this position is mostly hereditary – this is the case in Maagila village. When the Maagila sheikh died in 2011, his son took over. A sheikh must have the necessary personal qualities to be a leader. While in ordinary circumstances the eldest son takes over when the sheikh dies, sometimes a younger son is chosen. This is what happened when the sheikh of Maagila village died in 2011. Asked about the criteria for choosing the younger son, the answer was that ‘he is wiser than his elder brother’. Above the sheikh, there is the Omda, who represents a number of villages. The Omda is normally chosen by village sheikhs and the process is similar to choosing the sheikh. The Omda must also have leadership qualities and the ability to bring people together. Although the functions that these leaders play seem to be minimal, their positions are politically very important, especially when seen within the context of civil war and atrocities in Darfur over the past decade. While ordinarily the sheikh and the Omda solve local disputes, when the war started in Darfur in 2003, they were drawn into larger circles of alliances. Thus, they became part of the conflict, as I will explain later. Here, however, it should be noted that in 2007, a self-­ proclaimed Tuareg Nazir appeared and, although it is not known whether such a proclamation was endorsed or not, no one contested the position – although the Tuareg do not have a dar, which is a necessary condition for having a Nazir. This is a political move consonant with current ethnic and political polarization in Darfur, where traditional leadership is aligned with ruling elites in Khartoum. To carve out a niche, tribes in Darfur are forced to grab land and/or form their own political units that are eventually endorsed by the government. The establishment of the Tuareg Nazirate is part of the political games tribes are playing in Sudan at the present time. The current Tuareg Nazir is a retired police officer. Sudan provides an interesting case study when it comes to the exercise of power by groups such as the Tuareg. The federal system of governance promulgated in 1994 led to the devolution of power and the country was divided into twenty-six states (being reduced to eighteen states after the secession of South Sudan in 2011). In some of these states, some ethnic groups, or even tribes, dominate. For instance, in North Darfur, in which the Tuareg live, the Fur, Arab speaking groups, and the Zaghawa are the dominant groups. Yet, none of these groups could exercise political power independently of the central government in Khartoum. Within this complex set-up, politically speaking, the Tuareg have little room for manoeuvre.

Kinship and Marriage Kinship structures relationships among the Tuareg in significant ways. This comes as no surprise, since marriage among them is endogamous. Tuareg personal names are used most frequently in addressing all descendants and kin of one’s own generation, although cousins frequently address one another by their

Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan  171

respective classificatory kinship terms. Kin of the second ascending generation may be addressed using the classificatory terms inna (mother) and abba (father), and brothers and sisters of parents may be addressed in the same way, although this is variable. Most ascendants, particularly those who are considerably older and on the paternal side, are usually addressed with the respectful term amghar (males) or tamghar (females). The most frequently heard kinship term is aibobaz, denoting cousin, maternal or paternal, and is not confined to direct cousins, since it covers distant cousins too. Tuareg enjoy more relaxed, familiar relationships with their maternal kin, and more reserved, distant relations with the paternal kin. There are joking relationships with cousins and affinal relationships are characterized by extreme reservation. A cousin who is fifteen years old may joke with his cousin who is seventy, and may even make fun of him or her. This resonates with Murphy’s statement that ‘kinship terms are neither mere labels of social status nor simple expressions of basic binary distinctions in the cognitive map of the family’ (1967: 164). The behavior among cross-cousins can be seen as one of humorous or bantering ease. Marriage, known as tadwat, is predominantly endogamous and patrilocal. It is also polygamous. At the time of my first fieldwork in Maagila village, 20% of marriages were polygamous (one man having two or more wives) and only four members of the village were married to non-Tuareg women. The tinisduban (bride) is supposed to provide the furniture of the house. The furniture is quite simple: local kitchen utensils, one or two palm-tree mats, baskets and a locally made bed. Aibiryid (divorce) is not common, since the husband and wife are normally consanguine. The Tuareg prefer cross-cousin marriage, and within this the preference is for the mother’s brother’s daughter. Second in marriage preference is parallel-cousin marriage (the mother’s sister’s daughter). This may relate to the former matrilineal system for which the Tuareg are known. This also resonates with what Murphy (1967) found among the Tuareg of Agades in Niger: a central figure for any young man is the mother’s brother, for one may make extensive material claims upon him and prospects are considered excellent for receiving his daughter in marriage. Intra-group relationships and interactions are intimate, while relationships with other groups, known by Tuareg as etainfinan, are ­peaceful – or at least they were prior to the crisis in Darfur. The question of honour is taken seriously by the Tuareg, especially regarding sexual matters. Women are not supposed to engage in premarital sexual encounters. A woman who is found not to be a virgin when she marries brings shame to her family, and the fact that marriage is endogamous makes things worse. A woman who gets pregnant before marriage risks remaining a single mother for the rest of her life; if she gets married, it is normally to an elderly man or to a man who is already married. A final note on marriage among the Tuareg of Darfur relates to the lack of intermarriage with members of the Ainadan clan. As noted earlier, the Ainadan are

172  Munzoul Assal

considered blacksmiths, although they do not do ironwork, but engage in woodworking, which itself is considered a degrading activity. It should be noted that among other groups in Darfur (e.g. the Zaghawa), blacksmiths are also looked down upon. Members of the Ainadan clan are not accepted as marriage partners outside their clan. The Tuareg are not an isolated group. They are part of the wider social, economic and political systems in Darfur. The processes described above take place within such systems and are affected by them. It is therefore necessary to provide some analysis of the context in Darfur. I will do this at two levels vis-à-vis (1) the evolution of the crisis in Darfur and (2) its impacts on the Tuareg. At both levels, my analysis shall be brief and selective, and will focus on aspects that are relevant to the issues in this chapter. Beyond this, the reader may consult some of the references listed at the end of the chapter, which may provide detailed accounts on Darfur. The most recent book to provide a detailed account on Darfur is that by Yousif Takana (2016).

The Evolution of the Crisis in Darfur Two main narratives are adopted by successive Sudanese governments and their opponents in explaining the crisis in Darfur. The first comes from successive Sudanese governments that have argued that the war in Darfur is caused by ­local-level ethnic conflicts, mainly arising from pressure on a diminishing ­resource base. The second comes from rebel groups in the region that contend that marginalization and underdevelopment represent the main reasons for taking arms and fighting the central authorities (Sørbø 2016). These irreconcilable narratives account for the escalation of the crisis and the failure of the international community to gauge its different dimensions (Assal 2009). Until the early 1980s, Darfur was stable. The 1960s and 1970s witnessed peaceful and symbiotic ethnic interactions between different groups in the ­region. The work of Gunnar Haaland (1969, 1972) on economic determinants in ethnic processes and nomadization as a career among sedentary people in Darfur is illustrative of peace and stability in Darfur at the time. In an influential and much quoted article, Haaland (1969) described and analysed processes of boundary maintenance between two of the major ethnic groups in western Sudan: the Fur and the Baggara. Haaland discussed the nature of ethnic identities and the determinants in charge of identity. The Fur and the Baggara have remained culturally distinct, although they have been in contact for centuries. They differ with regard to general lifestyle, subsistence pattern, overt cultural features like language, housing, weapons and standards for the evaluation of performance. Fur–Baggara contact is regulated by a shared codification of the reciprocal statuses that are appropriate for members of the two groups respectively. Both the Fur and the Baggara are Muslims and thus may interact on ritual

Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan  173

occasions. ­Baggara people camp in the Fur area during the dry season. In the marketplace, they provide complementary goods: the Baggara supply milk and livestock, and the Fur supply agricultural products, of which millet is of major importance to the Baggara. There is a shift of economic activity that is constantly taking place: a Fur may accumulate cattle and pursue a nomadic career, and a Baggara may become impoverished and opt for a sedentary life. Haaland’s analysis reveals that livestock was an attractive investment during the 1960s and that Fur farmers who succeed to build up herds of cattle established themselves as nomads and migrated with the Baggara. Hence, they were assimilated into Baggara groups. Ethnic boundaries at the time were not as rigid and polarized as they are today. However, these processes changed during the 1970s and 1980s. Increased competition for the pasture due to growth in human and animal populations led to changes in ethnic relations. Revisiting his material, Haaland (2009) argued that due to the increase in animal and human populations, pastoralists started to construct enclosures to preserve the dry season pasture and raiding became more frequent. In 1986, the symbiotic relationship between the Baggara Arabs and Fur changed: the Baggara started to keep the nomadized Fur away from Baggara areas and the Fur reacted by preventing Baggara groups from moving into their areas in the dry season. A major escalation took place in 1987 when a congregation of twenty-seven Arab groups formed a coalition. They stated that Arabs make up 70% of Darfur and, as such, they should be duly represented in the different levels of governance in the region. This led to increased tension in Darfur and a year later, in 1988, violent clashes erupted between sedentary Fur and a coalition of cattle and camel-herding Arabs. In 1989, a reconciliation conference for the two fighting groups took place in El-Fashir, but it did not last long. The new military regime was looking for a success story, so it hurriedly organized the reconciliation conference (Harir 1994). Due to these violent conflicts, livelihoods have changed dramatically. There has been a gradual movement southward by camel-herding Arabs. This led not only to conflict between them and settled non-Arab farmers, but also to conflict between them and other cattle-herding Arabs: the Baggara. An escalating factor was the absence of an effective native administration, which had been abolished in 1970 without putting an alternative system into place. And since land is the reservoir of power in Sudan, many groups started grabbing land. According to Takana (2008), many groups found that acquiring land for settlement could be done by political allegiance to the government in Khartoum. Administrative divisions and subdivisions were made and unmade, often at the expense of settled groups whose rights had been recognized for centuries. But such divisions led to violence even between different Arab groups (ICG 2015). When the two main rebel groups – the Justice and Equality Movement (JEM) and the Sudan Liberation Army (SLA) – appeared in 2003, Darfur had

174  Munzoul Assal

already experienced serious intercommunity violence and widespread robbery. These rebel groups made regional and national claims that aimed at transcending ethnic cleavages, with demands for more equitable distribution of wealth and power. Nonetheless, the ethnic base of these movements was mostly non-Arab, particularly Zaghawa, Fur and Masalit. This led to more polarization in the region: Arab groups were seen as government cronies, while non-Arabs were seen as rebels (Assal 2009; Tubiana 2007). War and the generalized violence in Darfur led to polarization and shaped identities in the region. Alex de Waal (2005: 181) examines processes of identity formation in Darfur over the last four centuries. For de Waal, the basic story is of four overlapping processes of identity formation, each of them primarily associated with a different period in the region’s history: namely, the ‘Sudanic identities’ associated with the Dar Fur sultanate, Islamic identities, the administrative tribalism associated with the twentieth-century Sudanese state, and the recent polarization of ‘Arab’ and ‘African’ identities, associated with new forms of ­external intrusion and internal violence. It is a story that emphasizes the much-neglected east-west axis of Sudanese identity, which is arguably as important as the northsouth axis, and redeems the neglect of Darfur as a separate and important locus for state formation in northern Sudan, paralleling and competing with the Nile Valley states. For de Waal, due to the violence in Darfur, negotiable identities, of the sort described by Darfur scholars (like Haaland (1969)), have become fixed. The ‘Arab’ and ‘African’ emerged from violence and militarization and also from the fact that violence in the region became a field for regional, national and international actors and processes. This dichotomy, I have argued elsewhere (Assal 2009), is simplistic and may not reflect realities on the ground. It is also instrumentalized by non-Arab groups in order to demonize all Arabs in the region – even those who were not part of the violence and atrocities. It should be noted that this polarization led to the exclusion of Arabs from the Abuja peace process in 2005–6. The referral of Darfur to the International Criminal Court and the indictment of the Sudanese President further complicated the crisis, which attracted the world’s attention and resulted in massive humanitarian engagement, ­culminating in the arrival of the United Nations African Union Peace Keeping Forces (UNMID). For Sørbø and Ahmed (2013), issues of justice and reconciliation were taken out of local hands. The ambiguous and simplistic international discourse on Darfur led to local voices remaining unheard and Arab groups, even those who had kept away from the spiralling violence, being demonized in the international media (ibid.). Where are the Tuareg in all this madness? How did they react to the crisis? The last section of this chapter deals with these questions.

Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan  175

The Impact of the Crisis on the Tuareg Like everyone else in Darfur, the Tuareg have been adversely affected by war and violence in the region. Even long before the current crisis, the village where I carried out fieldwork many years ago was hit by deadly incidents of armed robbery. In 1987, one of the notables of Maagila was killed while attending his herd. After shooting him dead, the perpetrators took his herd. The next morning, men went after the robbers, but armed with modern rifles, the robbers repelled them and they could not recover the stolen herd. This incident led the inhabitants in the village, and other villages as well, to embark on a process of militarization, whereby they armed themselves with modern weapons. They also organized themselves into some sort of paramilitary units. The leader of such rudimentary military organizations is known as cagiid (lit. the knotter) – someone with the ability to command respect and bring people together. While this is not as significant a militarization process as that of the Arabs and the Fur, it signifies an important transformation of the Tuareg, who have been pacified for a very long time. It also signifies a shift in thinking: protection requires militarization and armament. The cagiid is someone other than the sheikh and normally a young, brave and strong man. During the early 1990s, another shift took place: they started joining the police and the army. The first police officer from Maagila village graduated in 1991. He was killed by a Fur militia in an incident in Jebel Marra in 2002. A few others followed suit and joined the military and the police. The current Nazir of Tuareg is in fact a police officer and a religious leader who is based in El-Fashir. Although the entire society in Darfur was polarized along Arab/African lines, as mentioned in the previous section, the Tuareg managed to avoid aligning themselves explicitly with any of these groups, even though they are accused of being sympathetic with the Arabs (Tubiana et al. 2012). Their villages in fact exist among villages of groups identified as ‘Africans’. They had to form some sort of alliance or agreement with these villages so that they would not be attacked. Such agreements worked well and were strengthened by the fact that they intermarried with the surrounding villages. However, in the face of increasing insecurity, the Tuareg had to be proactive: they left their village before they were attacked. Trust has eroded significantly in Darfur. The most significant change in the lives of the Tuareg of Maagila village, as well as other Tuareg villages across Darfur, happened after 2003 with the intensification of war in Darfur. Before the village was deserted, about four of the village members were killed in different incidents. These killings were blamed on the rebel movements. Even with these killings and the intensification of war in Darfur, the Tuareg in Maagila clung to their village until 2006, when, following the burning of adjacent villages, they decided to leave. They ended up in Abuja, one of the major three displaced persons’ camps in El-Fashir, joined by Tuareg

176  Munzoul Assal

from other affected villages. Abuja camp is an ironic reference to the capital of Nigeria where peace talks between Darfur armed groups and the Sudanese government took place in 2006. Some of the internally displaced persons from other villages ended up in Zamzam, the second-biggest internally displaced persons (IDP) camp in El-Fashir. Displacement has transformed the lives of the Tuareg. Prior to being displaced persons, they combined farming with animal husbandry and a few of them worked on wooden handcrafts. Due to displacement, they have no recourse to their previous employment opportunities. After 2009, a few could go back and work on their farms, leaving their families behind in the IDP camp. Men go back to farm their lands while women and children stay in the camp. When they are done with farming, they return to the camp. And although since 2009, security conditions allow for them to return, it seems that returning to the situation prior to displacement is not going to happen. People have changed and things have also changed. While they lead beleaguered lives in the IDP camps, they are, at least, physically safe. They also have access to education and healthcare services, and a few years ago they even received food relief. One result of what these people went through as a consequence of war in Darfur was that the inhabitants of many scattered villages came together in the IDP camps. On the other hand, even before the crisis in Darfur, some Tuareg were already living in the major cities in Darfur – El-Fashir, Nyala and Kutum. As mentioned above, the Tuareg have long practised migration, especially to Libya. But they also practised internal rural to urban migration (to the major towns in Darfur and also to the national capital, Khartoum), where most Tuareg migrants work in bakeries. Like other migrant communities, they have their own networks of communication and support. They help each other and come together during social occasions.

Concluding Remarks The Tuareg of Sudan not only became part of the local social and ethnic fabric of Darfur, but also part of the country at large, since they are now scattered all over the country. They were also affected by the turbulence in Sudan. They still keep their spoken language, which is spoken by the elderly and at the village level. The young ones who migrate to the towns speak Arabic, which is the lingua franca in Sudan. They do not stand as a distinct groups vis-à-vis other groups in Darfur and although they practise endogamous marriage, many Tuareg men and women are married to spouses from outside their group. Their distinguishing features (riding camels and wearing the tugulmus – the veil) are withering away. Their clothes are no different from average clothes worn by everyone else in the region. The crisis in Darfur disrupted the lives of Tuareg and rendered them internally displaced persons in major towns like El-Fashir. From mixing farming with

Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan  177

limited animal husbandry (sheep and camel), they now depend on food relief and work as unskilled labourers in towns. Their lives have thus been profoundly transformed, as exposure to displacement, although negative at the outset, also made them aware of urban systems of livelihood. They make use of health and education facilities, which has resulted in more children now going to school than during the situation prior to the conflict. In El-Fashir, some of them left the camps and succeeded in buying houses in the town. This means that the likelihood of them going back to their villages or to conditions prevailing before the crisis is quite unlikely. This point actually applies to displaced persons from other groups as well, and therefore the talk of the ‘return of displaced and refugees to their original areas’, as formulated by the United Nations (UN) personnel and some circles within the international community, is not realistic (Assal 2013). Moreover, even if some of the displaced persons in Darfur, the Tuareg included, are willing to return to their original settlements, they cannot do so until peace is fully restored. Darfur is still an active war front and UN/African Union peacekeeping forces are still in the region. The politics of numbers (Harir 1987) renders the Tuareg less visible, politically speaking. Their number is small and they are clients for bigger groups, although a few years ago, one of their notables claimed to be the Nazir of Tuareg in Sudan. This means that unlike some major groups like the Fur or Zaghawa, they will not be influential in the politics of the region, at least not in the near future. In the context of Darfur, and Sudan for that matter, a group that does not have land or arms has no voice. Looking at the Tuareg in terms of numbers, one may be inclined to say that the small size of the groups saved it from being polarized or becoming an active part in the Darfur crisis. Yet the crisis made them aware of the importance of education, and their youth are also joining the army and the police. Finally, judging from the massive displacement and mass killings in Darfur, and from the experience of inhabitants of Maagila village, the damage done by the war to the social fabric is beyond redemption. Trust has eroded in Darfur and people of the same village do not even trust each other, let alone people belonging to different ethnic groups. The policy of ‘divide and rule’ adopted by the central government in Khartoum is responsible for this. Government-supported militias kill and loot with impunity, causing bitterness and a sense of helplessness among local people who, in the eyes of the government, are rebel supporters or sympathizers. Rebel groups, for their part, also perpetrate atrocities, loot properties and levy some form of tax on local people (Assal 2013; Tubiana 2013; Takana 2016). The military and security perspective of the Sudanese government will likely escalate the situation in Darfur and even if the fighting between government troops and rebels ends, security conditions in the region will continue to be precarious for a long time. It seems that the Tuareg will remain in displacement camps and on the periphery of towns for a long time.

178  Munzoul Assal

Munzoul Assal is Professor of Social Anthropology, Dean of Scientific Research and former Director of the Peace Research Institute at the University of Khartoum. His research focuses on refugees, internally displaced persons, humanitarianism and citizenship. His key publications include: Sticky Labels or Rich Ambiguities? Diaspora and Challenges of Homemaking for Somalis and Sudanese in Norway (Bric, 2004); Diaspora within and without Africa: Homogeneity, Heterogeneity, Variation (Nordic Africa Institute, 2006, co-edited with Leif Manger); An Annotated Bibliography of Social Research on Darfur (Bric, 2006); Multi-dimensional Changes in the Sudan 1989–2011: Reshaping Livelihoods, Political Conflicts and Identities (Berghahn Books, 2015, co-edited with Barbara Casciarri and François Ireton) and Past, Present and Future: Fifty Years of Anthropology in Sudan (Chr Michelsens Institute, 2015, co-edited with Musa Abdul-Jalil).

References Assal, Munzoul. 1997a. Some Features of Rural Production in Darfur: Social and Cultural Aspects of Tombak Production. DSRC Discussion Paper No. 107. ———. 1997b. ‘Demystifying Perception and Expressions of Poverty: The Case of Tuareg Pastoralists, Northern Darfur’, Poverty in Sudan: Towards an Agenda for Research, 23–24 August 1997. Khartoum: Sharga Hall. ———. 2009. ‘Locating Responsibilities: National and International Responses to the Crisis in Darfur’, in Salah Hassan and Carina Ray (eds), Darfur and the Crisis of Governance: A Critical Reader. Ithaca: Cornell University Press, pp. 285–96. ———. 2013. ‘The Humanitarian Situation in Darfur and the Future of Internally Displaced Persons and Refugees’, in Abdelwahab Al-Affendi and Sidi Ahmen Salim (eds), Darfur: Ten Years after the Crisis (in Arabic). Doha: Aljazeera Centre for Studies, pp 189–211. Beck, Kurt. 1996. ‘Nomads of Northern Kordofan and the State: From Violence to Pacification’, Nomadic Peoples, 38: 73–98. De Waal, Alex. 2005. ‘Who Are the Darfurians? Arab and African Identities, Violence and External Engagement’, African Affairs, 104(415): 181–205. Fortes, Meyer, and E.E. Evans-Pritchard. 1940. ‘Introduction’, in Meyer Fortes and E.E. ­Evans-Pritchard (eds), African Political Systems, London: Oxford University Press, pp. 1–23. Haaland, Gunnar. 1969. ‘Economic Determinants in Ethnic Processes’, in Fredrik Barth (ed.) Ethnic Groups and Boundaries: The Social Organisation of Culture Difference. London: George Allen & Unwin, pp. 58–73. ———. 1972. ‘Nomadisation as an Economic Career among the Sedentaries of the Sudan Savannah Belt’, in Ian Cunnison and Wendy James (eds), Essays in Sudan Ethnography, London: C. Hurst & Co, pp. 149–72. ———. 2009. ‘Pastoral Peoples in a Globalizing World’, in Henriette Hafsaas Taskos and Alexander Taskos (eds), Connecting South and North: Sudan Studies from Bergen in Honour of Mahmoud Salih. Bergen: Unifob Global and Centre for Middle Eastern and Islamic Studies, pp. 97–114. Harir, Sharif. 1987. ‘The Politics of Numbers: Mediatory Leadership and the Political Process among the Beri “Zaghawa” of the Sudan’, Ph.D. dissertation. Bergen: University of Bergen.

Some Notes on the Tuareg (Kinin) of Northern Darfur, Sudan  179 ———. 1994. ‘“Arab Belt” versus “African Belt”: Ethno-political Conflict in Dar Fur and the Regional Cultural Factors’, in Sharif Harir and Terje Tvedt (eds) Short-Cut to Decay: The Case of the Sudan. Uppsala: Scandinavian Institute of African Studies, pp. 144–85. Holt, Peter M. 1958. The Mahdist State in the Sudan, 1881–1898. Oxford: Clarendon Press. Holt, Peter M., and Martin Daly. 2000. A History of the Sudan from the Coming of Islam to the Present Day. London: Longman. ICG (International Crisis Group). 2015. The Chaos in Darfur. Crisis Group Africa Briefing No. 110. Nairobi and Brussels: International Crisis Group. Keenan, Jeremy H. 1974. ‘The Tuareg Veil’, Revue des mondes musulmans et de la Méditerranée, 17: 107–18. McGregor, Andrew. 2013. ‘The Hunt for Mali’s Missing Islamists: Have Tuareg Rebels ­Returned to Darfur?”, special report, Aberfoyle International Security. (https://www.aber​ foylesecurity.com/?p=180, accessed 8 October 2021) Murphy, Robert. 1967. ‘Tuareg Kinship’, American Anthropologist, 69(2): 163–70. O’Brien, Jay. 1986. ‘Towards a Reconstitution of Ethnicity: Capitalist Expansion and Cultural Dynamics in Sudan’, American Anthropologist 88(4): 898–908. Rasmussen, Susan. 1995. Spirit Possession and Personhood among the Kel Ewey Tuareg. ­Cambridge: Cambridge University Press. Sørbø, Gunnar. 2016. ‘Engaging Anthropology in Sudan’. Unpublished paper. Chr. Michelesens Institute, Bergen. Sørbø, Gunnar, and Abdel Ghaffar M. Ahmed. 2013. ‘Justice by Default? Dealing with Accountability Issues in Sudan’, Nordic Journal of Human Rights, 31(2): 224–47. Takana, Yousif. 2008. ‘The Politics of Local Boundaries and Conflict in Sudan: The South Darfur Case’. Sudan Working Paper No 2: Chr. Michelsen Institute, University of Khartoum, and Ahfad University for Women. ———. 2016. Darfur: Struggle of Power and Resources, 1650–2012: An Institutional Perspective. Bergen: Chr. Michelsen Institute. Tubiana, Jerôme. 2007. ‘Darfur: A War for Land?’, in Alex de Waal (ed.), War in Darfur and the Search for Peace. Cambridge, MA: Harvard University and Justice Africa, pp. 68–91. Tubiana, Jerôme et al. 2012. Traditional Authorities’ Peacemaking Role in Darfur. Washington DC: US Institute of Peace. Tubiana, Jerôme 2013. ‘Darfur after Doha’, in Gunnar Sørbø and Abdel Ghaffar M. Ahmed (eds), Sudan Divided: Continuing Conflicts in a Contested State. New York: Palgrave Macmillan, pp. 161–84. Umbadda, Siddig, and Musa Adam Abdul-Jalil. 1985. ‘Women in Small-Scale Irrigated ­Agriculture: The Case of Wadi Kutum’, Afrika Spectrum, 85(3): 339–51.

Chapter 9

The Nkandla Controversy Insights from African Political Systems Robin Palmer

Introduction In hindsight, the Nkandla controversy was only one of several scandals that ­accompanied the deputy presidency (1999–2005) and later presidency (1909–18) of Jacob Gedleyihlekisa Zuma, but it loomed large in the year when we convened for the workshop to celebrate the seventy-fifth anniversary of African Political Systems, and it remains a significant lens through which to view Zuma’s extraordinary career as a populist leader in light of some of the ideas in that volume. Nkandla is the name and setting of Zuma’s rural homestead, set in former Zululand1 where he was born and partly raised. It is home to his living wives and many children as well, though neither he nor his wives bar the first have lived there permanently. Previously, Zuma’s homestead was little different from that of others in this poor rural area, but when democracy came to South Africa and Zuma took his place in the leadership of the African National Congress (ANC), Nkandla was incrementally upgraded. While the state is obliged to meet all costs associated with the building, maintenance and equipping of official residences, such as those Zuma came to occupy when he became Deputy President and later President, the state is only obliged to meet the costs of security for the private residence of a Deputy President or State President. Not only did the overall costs of the Nkandla project to the Department of Public Works escalate, but these also included features that had nothing to do with security, such as a visitors’ centre, an amphitheatre, a swimming pool, a cattle enclosure, a chicken run and a ‘tuck shop’ run by Zuma’s senior wife, Sizakele Khumalo-Zuma.

The Nkandla Controversy  181

Since the news first broke in 2009,2 it has emerged that Nkandla has cost the South African state R246 million.3 In the aftermath of these revelations, seven complainants instructed the Public Protector,4 Advocate Thuli Madonsela, to look into the matter. Her detailed report of March 20145 concluded with the recommendation that Zuma ‘determine the reasonable cost of the measures implemented … at his private residence that do not relate to security’ and ‘[p]ay a reasonable percentage of the cost of the measures as determined with the assistance of National Treasury’ (Madonsela 2014: 68).6 For the next two years, Zuma and his supporters failed to comply with the ruling while the pace of state capture accelerated, masterminded notoriously by the three immigrant Gupta brothers in collusion, allegedly, with Zuma and his adult children, among others (Martin and Solomon 2016; Pauw 2017). Eventually, on 31 March 2016, the Constitutional Court, the highest court in South Africa, responding to an application brought by the two largest opposition parties, found that both Zuma and Parliament had acted unconstitutionally through their noncompliance with the Public Protector’s recommendation. The verdict was immediately followed by a vote of no confidence in Parliament (overturned, predictably, by the ANC majority), an unconvincing apology from Zuma and a promise to comply with the judgment. Subsequent calls for Zuma’s resignation or recall from both within and outside his party were provoked not only by that judgment and the weak response to it, but also by increasing revelations of highlevel corruption tantamount to state capture7 facilitated by questionable ministerial appointments by the all-powerful president.8 It was only after the Local Government Elections (LGEs) of August 2016 – widely regarded as a proxy for a referendum on the ANC and, in particular, the Zuma presidency9 – that support for a change of leadership began to gather within the ruling party. The ANC’s share of the national vote (at 54%) had declined to below 60% for the first time since 1994 and the party lost its majority in three major metropolitan areas. Fearing further losses in the upcoming national elections, the ANC internally elected a new president, Cyril Ramaphosa (hitherto Deputy President), but it was not until February 2018 that Zuma resigned. Following the national elections of May 2019, Ramaphosa commenced his first term. The ANC’s majority of 57.50% in the LGEs, although still well down from the 62.15% of Zuma’s second term in 2014, showed a marked improvement compared to those three years earlier. ‘Ramaphoria’ was the term coined in the press for the reception of the new president-in-waiting, but it overlooked Ramaphosa’s slender lead in the internal vote of the ANC, and the constraint of a compromise cabinet in which an intense struggle for honest governance and economic recovery would ensue between the Ramaphosa and Zuma factions over the next three years and beyond. The Nkandla controversy was not the first scandal involving Zuma. The first was his involvement in the so-called Arms Deal, from 1999, and he is about

182  Robin Palmer

to face trial for it.10 The second was his trial for rape in 2006 where he was ­acquitted.11 Each of these scandals is revealing of the man and of his party’s appeal to the e­ lectorate during his time as a key leader of the ANC, but I have selected Nkandla and to a lesser extent the rape trial to illustrate certain side-issues of the first two chapters of African Political Systems (APS) (Fortes and E ­ vans-Pritchard 1940a). The focus of APS was eight indigenous African societies that represented a basic division between polities having centralized authority and those with an uncentralized political organization, the fieldwork having been done between the two World Wars. A substantive and conceptual antique such as APS can hardly have direct relevance to the present chapter on a case of power and its abuse in a society as developed and complex as post-apartheid South Africa, yet the book was the apical ancestor of all political anthropology in the British tradition, with its assertion in the introduction of an identity separate from Eurocentric political philosophy (Fortes and Evans-Pritchard 1940a: 4–5). And there are elements of the introduction and first chapter that can provide insights into Zuma’s choices as a traditionalist and his more traditionalist supporters among the electorate, as will be demonstrated below. The anthropologists who contributed to APS did their fieldwork in British colonies that were subject to ‘Indirect Rule’. This system of colonial governance added a controlling layer to traditional political systems12 that left the grassroots (where anthropologists based themselves) relatively undisturbed. Change came gradually, especially to tropical colonies with only recent and limited European settlement. Under the circumstances, there was no point in discussing change if there was so little of it, and for the pioneering anthropologists of the British School, there were more immediate academic concerns, such as a more precise understanding of the properties of social structure (Evans-Pritchard 1940; ­Barrett 1990: 17). The great exception to this characterization of anthropological research settings in Africa even then was South Africa.13 As with India, the Cape was settled by an East India company – Dutch rather than British – from as early as the seventeenth century; also, like India, the Cape colony and the rest of what became South Africa was never settled heavily by E ­ uropeans, even after the colony passed from the Dutch to the British in 1795. As James Belich (2009) has pointed out, British settlers preferred countries that were established by the British, such as the United States, Canada, Australia and New Zealand over, for example, Argentina and South Africa. Afrikaners (native-born descendants of the Dutch and assimilated Germans and French) remained – and still are – a majority among whites, and the white population was always small relative to other settler states and indigenous populations. Although the Dutch had largely eliminated the San and smallpox had reduced the Khoikhoi to a colonized remnant even before the British took over (Elphick 1985: Chapter 9), the more resilient and well-organized Bantu-speaking herders of the

The Nkandla Controversy  183

interior were a different matter. It took ten wars and a hundred years to conquer the Xhosa, and when the British challenged the Zulu Kingdom further up the coast, it cost them their only defeat in Africa, the Battle of Isandlwana in 1879. Colonized earlier and more comprehensively than the rest of Sub-Saharan Africa, South Africa also acquired independence, but hardly majority rule, relatively early on. Besides uniting the colonies and the Boer republics, the Union of South Africa in 1910 made for a self-governing country with some progress towards political inclusion14 that was modernizing its still largely agrarian and extractive economy. Labour supply was critical to these industries, but large-scale black urbanization was both threatening to whites and costly. The local solution was ‘circulating’ migrant labour, which depended on constraining land supply in the Native Reserves (later known as Homelands or Bantustans), restricting black land ownership, segregating black workers in hostels and townships, and imposing curfews (Bundy 1979; Murray 1981). After its election victory in 1948, the Afrikaner Nationalist Party inherited both the system and increasing black opposition to it.15 By giving segregation a unique name – apartheid – and reinforcing it with additional draconian legislation to separate whites from other resident races, while relocating and otherwise oppressing the latter, the National Party (NP) provoked international opprobrium as well as local opposition. External and internal pressure eventually led to the NP’s unbanning of the liberation movements, freeing of political prisoners and acceding to democratic elections. These were held in 1994 and were won by the African National Congress (ANC), with Nelson Mandela as the first President. This atypical African country had a profound effect on the anthropologists who were reared there – or at least the English-speaking anthropologists.16 These anthropologists were inducted into structural-functionalism in the English-­ medium universities and in the course of postgraduate training in the United Kingdom.17 When they returned to undertake fieldwork in South Africa, ­familiarity with their atypical African circumstances encouraged some of them to modify the structural-functionalist paradigm where it failed to describe local reality. For example, Monica Hunter (later Wilson) published the first ever example of a multisited ethnography in an attempt to understand social change in South Africa (Hunter 1936).18 Max Gluckman also opted to carry out his first major fieldwork in South Africa, in deep-rural Zululand between 1936 and 1938.19 Here too, racial domination was inescapable, and in his chapter in APS, Gluckman contrasted the complementary opposition and royal patronage that kept the indigenous Zulu state stable for long periods of time with the altered political situation that he encountered in the field. He characterized the contemporary rule by magistrates in association with local chiefs as ‘not well-balanced, for ultimately it is dominated by the superior force of Government, against which the only reaction of the Zulu is acceptance or passive disobedience’ (Gluckman 1940a: 54). In a single sentence, he dismissed any facile assumption of social equilibrium.

184  Robin Palmer

In that chapter and, even more successfully in his case-driven and reflexive short monograph, ‘Analysis of a Social Situation in Modern Zululand’ (Gluckman 1940b),20 Gluckman demonstrated that social equilibrium was temporary, the outcome of conflict more than the negation of it, and he took that argument even further in Custom and Conflict in Africa (1956). His chapter in APS had a threefold framework: the historical clan before the rise of Shaka, the Shaka era that formed the Zulu into a state and the contemporary situation. It was an analysis that combined Marxism, his training in anthropology and experience of the complex South African society into which he was born in 1913, and this early experience was to make him a major creative force and inspiration for others in the anthropology of modernizing, rapidly changing societies (Epstein 1967; Kuper 1973). Most importantly for the present purposes, Gluckman explained in his chapter in APS how clans had disappeared as territorial units under the regimes of Shaka and Dingane, and extended families (headed by the senior male homestead head) had become the basic political units. Clans retained their cultural significance and through their very dispersal helped to integrate the kingdom; however, territorially and administratively speaking, the homestead or umzi became a crucial unit in the chain of command. Each homestead owed allegiance to a chief but also to the king, with the king’s role more central and crucial than is usual among the Nguni following Shaka’s reorganization of the Zulu into a state. The king appointed chiefs who might be quite closely related to him, controlled a centralized army through conscription, performed ritual functions viewed as crucial on behalf of the nation and became personally involved with homesteads, including loans of cattle for bridewealth. During these reorganizations, the concept of umzi became so potent that it was used to describe larger political units, ultimately the entire kingdom, as a metaphorical umzi (Gluckman 1940a: 9). The notion of a king or paramount chief performing ritual functions regarded as critical for the viability of the polity on behalf of his people was not an invention of Shaka’s, but was pervasive in traditional African polities. Fortes and Evans-Pritchard devote six-and-a-half of the twenty-three pages of their Introduction to such ‘mystical values’: Members of an African society feel their unity and perceive their common interests in symbols, and it is their attachment to these symbols which more than anything else gives their society cohesion and persistence. In the form of myths, fictions, dogmas, ritual, sacred places and persons, these symbols represent the unity and exclusiveness of the groups which respect them. They are regarded, however, not as mere symbols, but as final values in themselves. (Fortes and Evans-Pritchard 1940b: 17)

The Nkandla Controversy  185

In the case of the Zulu, and of Nguni people more broadly, as I shall argue, it is when the mystical values associated with the homestead and those of the leadership come to be closely associated that one gains an understanding of why the particularities as well as the expansion of Nkandla have been so important to him and his supporters.

Contemporary Significance of the Rural Homestead Gluckman was not able to follow up his Zululand study ‘on the ground’, having moved to then Northern Rhodesia (present-day Zambia) and later the United Kingdom and on to other research topics, and he died in 1975 (Gordon 2018: Chapters 5ff.). Nor is there much in the way of relevant Zululand-based anthropology from the apartheid years – it is a region seemingly more inspirational to historians, and these historians were active during the remaining years of racial domination (Bonner 1981; Guy 1981; Wright 1981). Even the contribution of anthropologist Adam Kuper is historical (1993). Among the historians of Natal-Zululand, Jeff Guy was the most interested in the centrality of the homestead and the role of kinship at all levels of the political organization of these societies. He described: commoners’ homesteads consisting of a man (the homestead head, umnumzana, two or three wives, their offspring, cattle and smallstock, grazing and agricultural land. The men worked with the livestock, the women in agriculture, the two fundamental branches of production, and there was a clear sexual division of labour in the many supporting tasks. The wives were ranked and housed separately in the homestead. (Guy 1981: 50) Such homesteads were the basis for a pattern that continued long after Shaka and through the reign of his successor, Cetshwayo: While it was production in the homestead, and the social strength of the lineage within the clan based on this production, which provided the basis for the material strength of the Zulu Kingdom, it was power from above, from the Zulu state, that identified which individuals and which groups would achieve status within the kingdom. Zulu state power was based ultimately on the surplus labour [mainly in the military homesteads in the royal house – the amakhanda] drawn from every homestead in the land. (Guy 1981: 57) More recently, the focus on Zulu homesteads has been updated for contemporary Kwazulu-Natal and investigated in urban as well as rural settings by the contributors to the volume Ekhaya: The Politics of Home in Kwazulu-Natal (Healy-Clancy

186  Robin Palmer

and Hickel 2014). The findings from this volume are highly relevant in the case of the Nkandla controversy, and we will return to an examination of these in some detail below. While Guy tended to emphasize the economic and political significance of the Zulu homestead, McAllister, working among traditionalist Gcalecka Xhosa who were linguistically and culturally close to the Zulu, discussed the homestead’s moral and religious significance as well: A man without a homestead is not only materially poor, he also lacks status in the community and his reciprocal relations with others are limited. Similarly, he will be unable to enter into a full and satisfying relationship with his [ancestral] shades, because the homestead is the chief place of worship and the place where the relationship between a man and his shades is centred. (McAllister 1980: 208) It was thus in their own long-term interests for young men to keep the goal of a homestead of their own in the front of their minds. McAllister has described how younger men were schooled in their primary duty to ‘build the homestead’ with their wages from migrant labour through a ritual led by their elders. This emphasized the importance of resisting modernity and focussing on ‘home’ by remaining within the group of amakhaya (‘home boys’) whilst in town (McAllister 1981, 1991, 2001). This was more possible under the old ‘oscillating’ migrant labour system (Wolpe 1972; Murray 1981; Hay 2014) than the wholesale urbanization that has replaced it. Several other anthropologists undertook fieldwork among the Cape Nguni in the post-Second World War period. These included Philip Mayer (1971, 1980), David Hammond-Tooke (1962, 1975), Chris de Wet (1995), Cecil Wele Manona (1980, 1985, 1991), Leslie Bank (2011), Derick Fay (2002) and me (Palmer 1997a, 1997b; Palmer et al. 2002; Fay and Palmer 2000, 2002; Fay and Palmer 2002; Palmer and Hamer 2011; Khene-Pade et al. 2010). From these contributions, an empirical understanding of changing rural and urban lifeways over the period from the perspective of ordinary Nguni emerged. Exploitative as it may have been and disruptive to rural production, migrant labour provided rural homesteads with a degree of security and predictability. It shored up traditional marriage, secured by the payment of bridewealth21 with migrant earnings. And because labour migrants were legislated as temporary urban dwellers, confined to hostels, the rural home was the only permanent home, the site of the ancestors, a place of sanctuary for the politically active and ultimately a refuge for the sick or the elderly. Under segregation and apartheid, there was ample justification for prioritizing the building of the homestead. At the same time, migrant labour reduced the power of the black patriarch by removing him physically from the homestead for most of the year, and by

The Nkandla Controversy  187

making his migrant sons independent of him financially if they wanted to marry and build their own homesteads before he was ready for them to do so (Beinart 2014). The absence of male labour for much of the year also meant that field cultivation tended to be reduced or disappear altogether in favour of the more proximal gardens that could be more easily cultivated by multitasking women. Mining was dangerous, as were hostels and townships, and women were widowed more frequently, and some were abandoned. While in the past a widow would be incorporated into her husband’s brother’s household (the levirate), these males were also frequently absent; increasingly, women came to head households of their own. With the decline of mining and the repeal of apartheid legislation limiting the access of blacks to the cities, permanent or semi-permanent urbanization for both sexes replaced migrant labour (Reddy 2014). Rural homesteads were depleted not only of male labour as before, but also of younger mature women; many households became ex-agrarian female-headed repositories of older women looking after young children, along with retired men and, increasingly, dependant mature men retrenched from industrial jobs with nowhere else to go (Fay and Palmer 2000, 2002; Fay and Palmer 2002; Hay 2014). After 1992, when black African social pensions were dramatically increased to match those of other races, rural households with qualifying members (i.e. most of them) received a significant increase in their cash incomes, making it possible for them to support their grandchildren and free their urban-based children from the urgent and time-consuming task of job-seeking or, if employed, distant commuting (Palmer 1997a: 43). Since the advent of democracy in 1994, formal social housing has been provided in and around cities and towns as never before, but it has always failed to catch up with urbanization, and shack settlements on the urban periphery have proliferated. Newcomers to the city hoped that formal or informal employment would be easier to come by in town and that their children would be able to join them later, education generally being much better in the cities than in the rural areas; the advent of child grants supported these aspirations (Fay and Palmer 2002). However, due to increasing unemployment and poverty (even since the end of apartheid), not only is it becoming more difficult to support established families, but it is also increasingly beyond the capacity of young blacks to marry and found households of their own, whether in rural or urban areas. There is a crisis of social reproduction in South Africa associated with high and rising unemployment following successive ANC governments’ embrace of neoliberalism exacerbated by cronyism, corruption and state capture (Johnson 2017). Young men simply do not have the means to obtain the traditional wherewithal, ­meaning bridewealth, to get married, and this factor underpins pathologies such as an epidemic of domestic violence and rape, and absent fathers (Steinberg 2008).

188  Robin Palmer

I will argue that it is against the background of this worsening domestic crisis that we must understand the popularity of Jacob Zuma with the masses. But before making these connections, it is necessary to know more about the man himself.

Jacob Zuma Jacob Gedleyihlekisa22 Zuma was born in Nkandla in 1942. Jeremy Gordin23 sets the scene: With its steep gorges and deep ridges, the rainy Nkandla forest has always been a place of mystery, legend and final refuge. King Cetshwayo,24 defeated by Zhibhebhu, spent his last days there; in Nkandla the people once successfully defended themselves against King Shaka; more recently the rebels of Bhambatha25 had taken refuge in its depths. These days the forest is badly denuded. The area around it and the people who live in it are deeply impoverished, and were even more so when Zuma was born … The poverty in those days, however, was also a poverty of spirit – of a once wealthy and powerful kingdom brought to its knees. (Gordin 2011: 22) Zuma’s family were not of Zulu Royalty. They were commoners – peasant ­farmers –, though his father was a policeman He died when Jacob was four and his mother took him and his two younger brothers to her parents’ place for the family support she needed as a widow with such a young family. Later, at the request of the Zumas, she returned Jacob and his brothers to his paternal grandparents at Nkandla and went to work as a domestic servant in the Durban area. There was no school at Nkandla in those days, so Jacob herded his grandfather’s cattle. As reported by Gordin, the widespread belief that he gained his first education when he was in jail on Robben Island is a myth; he received some education secondhand from children who were attending school elsewhere, and he organized an evening class, with other herd boys, conducted by a local woman with some education. But it is true that his ten years at ‘the university of Robben Island’ provided more than political education. Zuma’s politicization began early, not only from listening to the stories of the Bambatha rebellion, which still had living witnesses at Nkandla when he was growing up, but also from the influence of a much older half-brother from his father’s first marriage, who was an ANC member and trade unionist. Later on, his experiences in the townships of Durban when he was in his teens, visiting his mother, also served to ‘conscientize’ him. The electoral victory of the Afrikaner nationalists in 1948 (when Zuma was six years old) was followed by a succession of Acts of Parliament intended to formalize apartheid, and these provoked street

The Nkandla Controversy  189

protests and eventually national protests against the passed laws, culminating in the Sharpeville massacre of 1960.26 This atrocity sparked internal shock and international condemnation, and the NP government responded by declaring a State of Emergency, arresting thousands and leaving the British Commonwealth to form a Republic. These acts would deepen the isolation of the NP and the country as a whole – and intensify the struggle – over the next thirty years. Zuma’s growing political awareness led him to join the ANC in 1959, at the young age of seventeen. His half-brother and another Zuma relative provided access to the South African Congress of Trades Unions (SACTU), the forerunner of the current Confederation of South African Trade Unions (COSATU), which was expanding rapidly in the Durban area. SACTU provided after-hours political education for its younger members, and Zuma attended avidly and regularly (by his own testimony, as recorded by Gordin). As opposition to apartheid intensified, the ANC, a nonviolent organization in the spirit of Gandhi up until that point, declared armed struggle and formed an armed wing, Mkhonto we Sizwe or ‘Spear of the Nation’ (MK), in December 1961. Subsequently, the organization began quietly recruiting activists for military training, and Zuma immediately joined up. In 1963, he was arrested while trying to leave the country in a party of about fifty other members bound for training in Zambia. He was held under legislation that stated anyone could be detained for ninety days on suspicion of political crimes; his arrest followed that of Nelson Mandela and the rest of the ANC leadership in July 1963. Their trial (on charges of treason) followed the exposure of Operation Mayibuye, MK’s plot to overthrow the white government; they received long sentences on Robben Island off Cape Town, historically the place of incarceration for political prisoners. In August 1963, Zuma himself was tried, found to be part of the same plot and sent to Robben Island. While the leadership was placed in single cells, Zuma, as an ordinary cadre, shared a cell with fifty or more other activists, some of whom became friends and allies for life. Days consisted of hard labour on a poor diet, but there was scope for sport – Zuma excelled at football – and the evenings were for recreation, education (which he urgently required) and political debate. Towards the end of 1973, Zuma was released and returned to Nkandla. It must have been quite a homecoming after no visits and no female company on Robben Island for ten years, but he soon resumed underground political activities locally. By early 1974, he was in Durban, having been given a job in a pet shop as cover; the next year he was in Swaziland, ferrying new MK recruits out of South Africa and trained MK soldiers and matériel into South Africa. Meanwhile, he had married his first wife and childhood sweetheart, Sizakele Khumalo. Neither of them knew that he would be away from home for the better part of two decades. After Swaziland, he was sent to Mozambique and then to Zambia, received military training in the Soviet Union, and rose steadily in the ANC in exile. In

190  Robin Palmer

June 1976, Soweto erupted, many schoolchildren protesting against compulsory education in Afrikaans were shot by police and thousands of young activists were on the run – potential recruits for MK if they could evade arrest and leave South Africa. Zuma was a key figure in the facilitation of this process. In recognition of his efforts, he became a member of the ANC’s Executive Committee, was appointed Head of Underground Structures, later Chief of Intelligence and then became a member of the ANC’s Political and Military Council. Not much is known about this secretive period of his life, but as his promotions indicate, he was effective in this role and held in high esteem within the party. In the latter 1980s, Zuma became involved in the encounters with opposition party politicians and businessmen abroad, which eventually led to negotiations between the NP and the ANC. This process of rapprochement eventually led to the unbanning of the liberation organizations by then-President F.W. de Klerk in 1990, and the release from prison (after twenty-seven years) of Nelson Mandela. No doubt Zuma’s considerable charm and negotiation skills contributed to the successful outcome. These talents would continue to be needed, not least in his troubled home province of KwaZulu-Natal (KZN), where a regional civil war that would cost 2,000 lives was raging between the Zulu nationalist party Inkatha (with clandestine support from the NP) and the ANC. In 1975, Chief Mangosuthu Buthelezi had revived this Zulu cultural movement, founded in the 1920s, when he was leader of the KwaZulu Bantustan, seeking to oppose apartheid ‘from within’. But Buthelezi had subsequently fallen out with the ANC and it was only with Zuma’s intervention, as another Zulu traditionalist, that conflict ceased, and eventually ordinary Zulus came over to the ANC cause and Inkatha’s power waned. Meanwhile, Zuma chaired the ANC negotiating team at the internal Conference for a Democratic South Africa (Codesa) that would eventually deliver democratic elections, a Government of National Unity, a new Constitution and a Peace and Reconciliation Commission to deal with the many atrocities committed on all sides in the long-running and complex contestation for power in South Africa. Zuma was also busy on another front: he led the South African mission to make peace in Burundi following the genocide in Rwanda in 1994 that had spread into neighbouring countries. It was a delicate operation that required twenty visits over two years. On the strength of these diplomatic successes, as well as his support for Thabo Mbeki during the years of struggle, Zuma was elected Vice President-in-waiting at the ANC elective conference in 1997, while Mbeki was elected President to follow Nelson Mandela. Both assumed their respective offices following the general elections of 1999. Barring occasional visits that had to be clandestine before 1990, Zuma had been away from Nkandla and his wife, Sizakele, for most of their marriage to date, but he was neither celibate nor monogamous.27 Soon after arriving in Maputo, he married Kate Mantsho, who worked for the Mozambique airline and with whom he would have five children, and a few years later, Nkosazana Dlamini, a doctor

The Nkandla Controversy  191

in Swaziland, who bore him four children.28 Zuma was involved with Nompumelelo Ntuli for some years before he eventually married her in 2008 and she bore him two children. In 2010 he married Thobeka Stacy Mabhija and, in 2012, Gloria Bongekila Ngemqa, with each of whom he had one child. A further child, Edward, who has spent most of his life at Nkandla, was born in around 2000, the result of a liaison. In addition, he has other children born out of wedlock, some of whom are being cared for at Nkandla, making up a total of twenty-two over a wide age range. Although only his first wife, Sizakele, who is childless, stayed at Nkandla in the 1990s, Zuma had to support Kate until her death (by suicide) and Nkosozana until their divorce, as well as all the children. With no savings or assets from his years in exile – on the contrary, he had mounting debts – Zuma could not afford to support his family, let alone build a suitable home base for them at Nkandla on his salary.29 He needed help with his limited and chaotic personal finances. Fortunately for him, this was forthcoming from the brothers Shaik. Moe and Yunis Shaik were well known to Zuma, having long been involved in the struggle under him; another brother, Schabir, was in business, and the fourth, nicknamed Chippy, was in charge of procurement for the Arms Deal (mentioned above) in the Department of Defence. The brothers assured Zuma that they would manage his finances while he concentrated on politics. Their faith in him was not misplaced; as we know, he became Deputy President and later President. The main responsibility for financing Zuma’s household fell on Schabir and his company, Nkobi Holdings. Among other payments he made to Zuma, Schabir solicited R500,000 a year for him from a leading company involved in the Arms Deal, Thomson-CSF (later renamed Thales), a French company of which his Nkobi Holdings was the local ‘black empowerment’ partner. This and other acts of assistance to Zuma became the subject of Schabir’s trial for corruption and eventual fifteen-year sentence in 2004. Zuma was implicated30 and this was the first of two serious charges against him in 2005. In November of that year, Zuma was accused of rape by the thirty-one-yearold daughter of an old friend who was staying over at his Johannesburg home. A sensational trial followed, accompanied by thousands of vocal supporters. In his testimony, in which he admitted to having consensual sex with ‘Khwezi’ (the pseudonym that protected the identity of the plaintiff) but denied the rape charge, Zuma described her coming into to his room in a shift, a ‘khanga’, that indicated to him that she wanted sex with him, and that traditional Zulu male ethics required that a woman who wanted sex should not be denied. He also explained why he went ahead without a condom even though he knew his accuser was HIV positive, taking a shower afterwards to ‘reduce the chances of infection’. (Ever afterwards, Zuma’s nemesis, the cartoonist Zapiro, would portray him with a shower growing out of his head, an image the opposition Economic Freedom Fighters (EFF), led by Julius Malema, his sometime supporter, would use in a

192  Robin Palmer

gesture to mock him.)31 The trial represented a turning point in his prestige, lowering it among the elites but increasing it among the masses, an apparent contradiction that I will examine below. President Thabo Mbeki had already sacked Zuma from the deputy presidency for being implicated in corruption before the rape trial. Hardly a ‘man of the people’ following his years in exile and Sussex degree, and his somewhat distant and authoritarian interpretation of his presidency, Mbeki was losing the support of key ANC constituencies: COSATU, the South African Communist Party and the ANC Youth League (ANCYL). Civil society movements were also questioning Mbeki’s inexplicable denial of the causes of HIV/AIDS at a time when the pandemic was at its height in South Africa. His dismissal of his popular deputy was not a good move. At the 2007 elective conference at which Mbeki hoped for a third term, Zuma was nominated as President instead. With his proletarian background, proclivity for dancing and singing – his trademark was the ‘struggle’ song umshini wam (bring me my machine gun) – and his excellent ANC pedigree, Zuma offered a persona that was quite different from that of Mbeki. Although presidents are not directly elected in South Africa, it was thought that he would boost the ANC’s chances in national elections. In the same year, corruption charges against him were dropped on the grounds of political interference, and Zuma celebrated by marrying MaNtuli at a traditional ceremony for 500 guests at Nkandla in early 2008. Mbeki resigned the same year and was replaced by his new deputy for a few months, until the 2009 General Election. When the election came around, the ANC increased its majority to 66% of the vote – a phenomenon the press dubbed a ‘Zunami’. Interestingly, the party this Zulu traditionalist led did even better in six out of the eight provinces than it did in KZN: what he was and what he stood for resonated with the electorate far beyond his fellow Zulu in his still somewhat politically divided home province (Hunter 2014: 222). Still, he persisted with his winning formula, not by rebranding himself as something he was not, but by having another traditional wedding even more spectacular than his last: in early 2010, he married his fifth wife, MaBhija, with whom he already had two children. As Meghan HealyClancy and Jason Hickel expressed it, ‘in a self-consciously traditional ceremony on his self-consciously traditional homestead … Zuma celebrated the extension of his umuzi32 [homestead, including its members], and thereby the expansion of his political power’ (2014: 1). This time there were 2,000 guests and the ceremony was televised at home and abroad. His followers took to wearing t-shirts emblazoned with his portrait and the legend ‘100% Zulu-boy’, and then-leader of the (ANCYL), Julius Malema, claimed he would ‘kill for Zuma’. There was global interest in South Africa at this time not only for the sake of this colourful new president but also because the country was preparing to host the FIFA World Cup in 2010, which had never been held in Africa before. There

The Nkandla Controversy  193

was much doubt about a successful outcome, but in the end the event exceeded expectations and boosted Zuma. He needed it: in January 2010, the news broke that he had fathered a child with the daughter of another old friend, Irvin Khoza. Although he acknowledged the child as he always did, paid inhlawulo (the customary indemnity to the family) and apologized to the nation, it was a ‘bridge too far’ in the eyes of the press and some of his supporters, and it was the end of his presidential ‘honeymoon’ in the party too, as other evidence of his poor judgement also began to emerge. Malema, the leader of the ANCYL who had been prepared to ‘kill for Zuma’, now showed disrespect and insubordination, so Zuma dismissed him, presumably gambling on the fact that Malema was too young and junior in the ANC for his dismissal to have serious repercussions, but he underestimated the young man and the increasing evidence of his own incapacity to deliver on the ANC’s election promises (Butler 2010). Malema went into vengeful opposition, and along with other malcontents formed the Economic Freedom Fighters (EFF), an opposition party with a radical agenda that won twenty-five parliamentary seats in the 2014 national elections, and continues to go from strength to strength. In August 2012, a ‘wildcat’ miners’ strike at the Lonmin platinum mine escalated violently, culminating in the death of thirty-four miners and the injury of seventy-eight after being shot by police in the worst such incident since the Sharpeville Massacre of 1960 (Reddy 2014). Zuma’s response was unimpressive, and he was implicated for his appointment of an outsider (and a woman to boot) as Commissioner of Police, who faced a storm of criticism. Meanwhile, Malema took the opportunity to visit the area, exposing Zuma’s increasing new proclivity to isolate himself, much as Mbeki had done before him. It was becoming increasingly apparent that Zuma, the singing and dancing populist and sometime candidate of the unions and the communists, lacked leadership capacity, was a party to state capture and was no more of the Left than Mbeki was (Reddy 2014) At the same time, as noted at the beginning of the chapter, the extent and dire implications of state capture and attempts to protect Zuma and favour cronies were being exposed (see e.g. Pauw 2017). When the results of the LGEs of 2016 revealed the political fallout of Zuma’s presidency,33 elements of the ANC leadership that were previously on the back foot received additional support and narrowly succeeded in replacing Zuma with his deputy, Cyril Ramaphosa, as noted. For the last three years, in the face of continuing support for Zuma within the ANC, including its leadership, Ramaphosa has struggled to redress the damage of the previous nine years and continues to do so.

Explaining the Rise of and Support for Jacob Zuma As noted above, Zuma had excellent ‘struggle credentials’ and initial support from the unions and communists, but he also appealed strongly to an ­increasingly

194  Robin Palmer

important constituency in democratic South Africa that has still not been discussed in political terms: the enfranchised but largely unemployed and dependent masses. While it is the case that Zuma drew support on an ethnic basis from fellow Zulus in his home province of KZN, especially since the ANC began to prevail over Inkatha, he also gained popularity within a black lower-class segment beyond any particular ethnicity. I have described above in general terms the worsening domestic crisis of this segment after organized migrant labour declined and rural-based men found themselves without a reliable source of cashflow that, if responsibly managed, could secure marriage – perhaps more than one – and the opportunity to ‘build a homestead’. The free urbanization of individuals and couples that followed was accompanied by higher overheads and higher risks: saving for the traditional goals of bridewealth and a homestead became less likely for uneducated and unskilled men, even if they were now liberated and enfranchised. Accordingly, traditionalist women who shared those goals also had to settle for less. For South Africans such as these, the very marital, extramarital and homestead-making activities of Zuma, which were regarded as excessive by the ‘chattering classes’ both within and beyond his party, were viewed as credentials in the eyes of both the men and women of the large and important constituency of the poor and desperate. Mark Hunter (2014) found himself ‘in the field’ in Mandeni, not far from Nkandla, living in an informal settlement on the periphery of an industrial area and a formal township at two critical stages of Zuma’s career: the rape trial (2006) and the run-up to his presidency (2009). He noticed a tendency among his interlocutors to emphasize the private and domestic performance of leaders. While the broadsheet press ridiculed Zuma’s confession that he took a shower to prevent contracting HIV/AIDS after having sex with his rape accuser, this made perfect sense to Hunter’s equally misinformed interlocutors. When Zuma went on to explain that, in accordance with Zulu traditional morality, he could not leave his accuser unsatisfied once he surmised that she wanted sex and that he had offered to pay ilobolo and marry her, the men and women of Mamdeni were further impressed. They saw him as a ‘stud’, an Isoka, but not an isoka lamanyala – a Casanova (Hunter 2014: 235–37). To them, he was a responsible Isoka, with the inclination as well as the means to support his wives and children, and even to offer to make an ‘honourable’ woman of his rape accuser. Zuma’s rape trial enhanced his prestige in their eyes. When in the 2009 election season Zuma promised to increase the amount of the child grant (on which so many depended) as well as the age range of those eligible, they did not regard this as electioneering, but believed the promise, regarding it as consistent with Zuma’s family values and traditionalist ethics. Even when on another occasion he confessed to homophobia in his youth, this was regarded as further evidence of his respect for the patriarchal family, whose reproduction would be threatened if homosexuality became rife (ibid.)

The Nkandla Controversy  195

In Mamdeni, according to Hunter, the decline of domestic capacity in the lives of commoners began well before the advent of democracy. Increasingly, young men could not raise ilobolo, so young couples could not marry and young women received less support if they became pregnant. In the countryside, the capacity of ordinary men to ‘build the homestead’ was diminished and even in towns, assuming they were lucky enough to get an ‘RDP’ house,34 these were not a patch in terms of size and building quality based on the old four-room brick houses of the apartheid years that could accommodate a family (Hunter 2014: 249). Hunter’s findings accord well with the more general sketch of conditions in the rural areas and small towns culled from anthropologists working in the Eastern Cape and provided earlier. Whilst in the field among rural and smalltown Xhosa speakers who were even more universally ANC supporters than the Zulu among whom Hunter worked, I too witnessed firsthand the deteriorating circumstances of families and their increasing dependence on welfare. I recall especially the desperation of younger men who had been retrenched from the mines and failed to find work in town, and who were now returning to the rural areas and dependent on the pensions of their elders and whatever temporary paid work they could find locally, with little hope of marriage, let alone ‘building a homestead’ (Palmer et al. 2002; Fay and Palmer 2000, 2002; Fay and Palmer 2002; Palmer and Hamer 2011; Khene-Pade et al. 2010). In this general context of worsening deprivation, Zuma’s upholding, embodying and even exceeding the normal expectations of a polygynist and extramarital lover as a super-Isoka, was consistent with his also having a super-homestead, Nkandla. To the thinking of the adulatory but home-starved and culturally nostalgic masses, it was logical that Nkandla should exceed any ordinary rural homestead just as Zuma’s marital and extramarital activities and responsible approach to his marital and family responsibilities and dalliances exceeded anything to which an ordinary black South African man (even an ordinary polygynist) could aspire to or perform, or that any ordinary black woman might benefit from. At this level of society, there was no support for the notion that Zuma was responsible for the overspending at Nkandla or that he owed the state any money.35 Nkandla was in any case bound to be large because of the size of Zuma’s family, including the children born out of wedlock or by absent wives who were incorporated into his homestead. Besides any practical considerations, in addition to being a super-Isoka with a super-homestead, he was a super-chief, the president of a rich country: his position demanded a magnificent homestead, with space to receive guests. The masses, however, were not Zuma’s only reference group, even though they are the main part of the electorate and are mostly loyal to the ANC that liberated them. To Zuma, his fellow ANC leaders, the heads of the state-owned enterprises (SOEs) and the emergent Black Economic Empowerment (BEE)

196  Robin Palmer

business owners were an important reference group. These supporters relied on state tenders and practised conspicuous consumption in terms of accommodation, transport and hospitality (notably weddings). To achieve and maintain standing in their eyes, it was crucial that Zuma and his family match and even exceed them in terms of consumption as he rose in the power structure, and this he was able to achieve, not so much from an expanding salary, but through the patronage he could offer to individuals like Schabir Shaik and later the Gupta brothers. In addition to all this, he headed an administration that was eager to please him when he was Deputy President, and even more so when he became President, and was not averse to directing public funds towards both security and more of his private residence – until the Public Protector became involved. The details of the nonsecurity public expenditure on Nkandla reflect both Zuma’s traditionalism and populism, and his need to need to impress his fellow leaders; the accoutrements of Nkandla are an interesting mix of tradition and ­aspiration. The major cattle kraal and the smaller kraal for rituals are an essential – nay sacred – feature of a traditional homestead, not least a chief ’s, who should perform animal sacrifices for the health of the nation; the chicken run is also a nod to traditional subsistence. A grand entrance and reception area for visitors, and an amphitheatre for gatherings are hardly features of a traditional homestead. The ‘great place’ of a chief or king have always been more elaborate than an ordinary homestead and one would not expect the homestead of a traditionalist State President to neglect the obligations regarding hospitality of a traditional leader in the rural areas. However, the swimming pool is entirely aspirational, and when the accusations of state subsidy of private benefits began to flow, defenders of Zuma were at pains to explain it away as a ‘fire pool’, an essential security measure for a compound with thatched dwellings. A similar combination of attachment to an umzi and social aspiration is exhibited in Zuma’s concern for his ‘home away from home’ on international trips – the upgrading of his official aircraft. The struggle between what his office requires and what the Air Force regards as reasonable and affordable has been ongoing for almost as long the Nkandla scandal.36 Nkandla is also, of course, in its large scale and emphasis on security, a statement of Zuma’s power, a modern take on ‘the mystical values of political office’ as Fortes and Evans-Pritchard put it (1940b: 18). In the traditional pastoral economy of the Nguni, without fences or deeds, land rights were not an issue of legality as much as an assertion to be expanded and defended. One way to assert those rights was to build a homestead at the centre of the domain that was claimed. For chiefs and kings, that domain included the homesteads of kinsmen and others who owed them fealty; for a state president’s rural home, it was more complicated, more abstract, more hedged in with administrative guidelines about where the state’s obligations ended and those of the private individual began. But to a traditionalist such as Zuma with scant formal education, procedural niceties

The Nkandla Controversy  197

have little meaning; his home is something to defend if it is attacked, whatever the merits of the case, with as much determination as any Nguni man would defend his homestead physically and to the last. And that is something that appears to be have been understood by his constituency, at least as long as his legitimacy in other respects was intact. Legitimacy, however, among a majority of ANC supporters who are poor and getting poorer as the economy stagnates and as unemployment increases, was at least as much a function of delivery to this segment as a more abstract respect for Zuma’s embodiment of the traditional and the aspirational as symbolized pre-eminently by Nkandla. It is not difficult for those who have become aware of the inflated cost of Nkandla through the media to calculate how much welfare or how many schools, clinics or RDP houses R256 million would buy; this calculation was made anyway in the media for its readership or viewership with every twist and turn of the Nkandla saga. In an attempt to mitigate the problem and to retain Nkandla, the rural municipality, for the ANC in an impending byelection in 2014, Zuma announced a grandiose R2 billion development for the municipality that the press dubbed ‘Zumaville’.37 However, the Nkandla electorate was not mollified by the promise of Zumaville in 2014 and the ANC lost to its great rival in KZN, the resurgent Inkatha Freedom Party (IFP), in that byelection; in the LGEs of 2016, the IFP returned again with an increased majority.38 As if that irony was not embarrassing enough, Zuma was giving a speech at the official announcement of the election results on 6 August 2016 when four young black women in black evening dresses appeared in front of the stage with placards reading ‘I am 1 in 3’, ‘10 years later’, ‘Khanga’ and ‘Remember Khwezi’. To those who had followed Zuma’s rape trial, these placards were clear references, respectively, to the epidemic of rape in the country, that the trial had not been forgotten, that wearing a Khanga shift was not an invitation to intercourse and that ‘Khwezi’ was a tragic victim forced into exile and now deceased. The women’s silent protest marked the beginning of Women’s Month and highlighted the problem of ‘rape culture’ to which protesters had been drawing attention all year.39 While the traditionalists of Hunter’s Mamdeni and elsewhere and those whom he had helped to make billionaires continued to support Zuma in his decline, the stagnating economy and the rise of intimate partner violence against women were inviting more progressive elements in the black community to revise their opinion of him. That Malema was now opposing Zuma with his own form of rabble-rousing was also having an effect. Taking his cue from Zapiro, the cartoonist who reminded the public about Zuma’s rape trial with the ‘shower head’, Malema and his supporters in the EFF, now a conspicuous presence in Parliament with their red uniforms, symbolic of proletarian occupations, imitated a shower with one hand over their heads to taunt him. With reference to Nkandla, following the Public Protector’s determination, and particularly after the support for it by the Constitutional Court, whenever Zuma tried to speak in Parliament, the EFF

198  Robin Palmer

chanted that he should ‘pay back the money!’ until they were declared out of order and evicted. These stunts have helped to make the EFF the fastest-growing political party in South Africa with younger, urban blacks and seriously undermined whatever modern equivalent of the mystical values of high political office Zuma had managed to spin in his ascendency as a populist Zulu traditionalist. With reference to Nkandla, there is a final irony. Eventually, the Treasury calculated that an appropriate share for the visitors’ centre, amphitheatre, swimming pool, cattle kraal and chicken run would be R7.8 million.40 This may have been only a small fraction of the total cost of Nkandla, but it was a hefty sum for the President and his family and a stiff price for those five items. Zuma was given forty-five days to come up with the money. With his popularity plunging and elections imminent, he declined offers of support from ANC members and former benefactors, and took out a bank loan to repay his Nkandla Bill with the then obscure VBS Bank.41 This bank is no longer obscure; it has been exposed as a Ponzi scheme since its collapse owing the better part of R2 billion, and the heads of the bank are currently on trial, facing lengthy jail terms if found guilty. This is an important test case, which is soon to be followed by that of Zuma himself on renewed charges linked to the Arms Deal. Should any of the accused turn state’s witness to avoid heavy sentences, it would ‘bring down the house’ that corruption in South Africa has built since the beginning of this century and perhaps usher in belatedly the new era that the coming of democracy in 1994 promised.

Conclusion It is self-evident that an eighty-year-old book based on even older anthropological research on indigenous African polities was going to be of scant utility in understanding a postcolonial phenomenon such as Jacob Zuma. Structural functionalism, the paradigm APS helped to launch, was not well adapted to understanding social change, yet change gathered ground rapidly after 1945, even in Africa. Gluckman, the contributor of the first chapter, as a South African, had a better feel for racial domination than other contributors and, as a Marxist, a more dynamic approach to the study of conflict (1956). His focus on Zululand and the centrality of the umzi or homestead in the organization of the Zulu was helpful to our understanding of the choices of a relatively unsophisticated Zulu traditionalist like Zuma. As for the rest, the significant proportion of the introduction to the book dedicated to the role of ‘mystical values’ in the politics of African kingdoms or chiefdoms was also helpful up to a point (Fortes and Evans-Pritchard 1940b: 16–23). While these beliefs and traditions may stabilize a slowly changing, relatively small-scale, preindustrial African society, they are harder to maintain by the leader of a recent regime at the top of a complex nation-state, even when such values are held by a large proportion of the electorate and are readily transferable in their minds from hereditary traditional leader to limited-term president.

The Nkandla Controversy  199

As noted, however, such values readily lose their lustre in the hurly burly of a parliamentary democracy in the age of television and social media. The same media that can rapidly promote a traditionalist on top of his game singing struggle songs, dancing and enjoying lavish traditional weddings in full Zulu regalia can also diminish him when he shows that he cannot count large numbers, is pilloried by opposition parties and ridiculed by cartoonists – and, most importantly, cannot deliver on development promises. As Fortes and Evans-Pritchard pointed out, mystical values are only one essential factor for the functioning of even a pre-capitalist state. They also cite material conditions, the balance of forces and use of force, and the fact that politics does not exist in isolation (1940b) In this connection, Zuma’s capacity to satisfy the masses with improvements to their livelihoods was constrained by a politics of patronage and corruption to the extent of state capture that may still turn South Africa into a ‘failed state’. The SOEs could not perform their functions. Housing and infrastructure contracts favouring tenderpreneurs delivered a substandard result. Municipalities staffed through cadre deployment could not deliver services adequately. The tightening of BEE and the threat of land ­expropriation, as well as continuing deficiencies in the maintenance of law and order, led to a flight of capital and deterred further investment, causing already severe unemployment to rise further. Strikes and protests were more cathartic than useful in this context. Support for the ANC and Zuma were bound to wane under these dire circumstances. Sacrificing a president who was no longer popular besides being the source of some of the problems was an obvious first step to halting the slide and in the hope of bolstering the party’s chances of re-election. In APS, Fortes and Evans-Pritchard and especially Gluckman opened a portal to politics below the level of governance – initially of kinship, family and home, and later of all manner of identities that would not have been considered to have political potential when the editors of APS resisted a Eurocentric and philosophical interpretation of politics as governance and declared an understanding of what happens ‘on the ground’. Few as they were, successive anthropologists of the British tradition in South Africa acknowledged the contributions of the pioneering work of the contributors to APS, but they were also less averse to the incorporation of change into their models and of interdisciplinary transfers of knowledge. Their influence is apparent in local historiography and vice versa. Whether this tradition will survive the postcolonial turn in African anthropology remains to be seen, but the South Africans’ more holistic interpretation of the empirical approach has survived in its essentials into the new millennium. If a new generation of anthropologists adopts the intellectual and ideological independence of mind combined with respect for tradition that those cited in this chapter have exhibited, the discipline will continue to contribute to the study of this extraordinary society at the tip of Africa in interesting ways.

200  Robin Palmer

Robin Palmer is Professor Emeritus of Anthropology at Rhodes University (Grahamstown). He holds a BA (Hons) from the University of Durham and an MA and a Ph.D. from the University of Sussex. His early research was on Italian migrants, based on fieldwork in rural Italy and London. His research projects in South Africa’s former Bantustans have included a survey of black blood donors, ex-donors and nondonors; racial succession in a small town; a focus on gender, households and the environment of two villages; a land claim on two nature reserves and their neighbouring communities on the Wild Coast; and more recently the impact of a digital ‘living lab’ in the same general area.

Notes  1. Since 1994, this former ‘native reserve’ and later ‘homeland’ under apartheid has been fully integrated into the province of Kwazulu-Natal.  2. ‘Zuma’s R65 million Nkandla Splurge’, Mail & Guardian, 3 December 2009; ‘Bunker, Bunker Time: Zuma’s Lavish Nkandla Upgrade’, Mail & Guardian, 11 November 2011.  3. ‘Nkandla: How Not to Spend Your R246m’, Mail & Guardian, 2 August 2015.  4. The Office of the Public Protector is an independent or ‘Chapter 9’ institution created by the Constitution of the Republic of South Africa in 1996.  5. ‘Secure in Comfort: Report on Allegations of Impropriety and Unethical Conduct Relating to the Installation and Implementation of Security Measures by the Department of Public Works and in Respect of the Private Residence of President Jacob Zuma at Nkandla in the KwaZulu-Natal Province’, Public Protector South Africa, Report No. 25 of 2013–14. At that time, R1 was about US$11.50; it is about US$15.50 at the time of writing.  6. ‘Treasury Tells Concourt that Zuma Must Pay Back R7.8 million for Non-security Upgrades to Nkandla Homestead’, Mail & Guardian, 27 June 2016.  7. ‘For the Sake of Your Country, Mr President, Go Now!’, Sunday Times, 3 April 2016.  8. ‘President Zuma Is Stronger than Ever – Analyst’, Times Live, 10 May 2016.  9. ‘How World Sees SA – Local Elections Viewed as Referendum on ANC’s Hegemony’, Financial Times, 1 August 2016. 10. The Arms Deal of 1999, officially the Strategic Defence Package of the South Africa’s Ministry of Defence, sought to modernize the South African National Defence Force (Navy, Army and Air Force) with material at a cost of US$4.8 billion (R30 billion at the time). There have been immediate, and continuing, allegations of corruption (see e.g. Crawford-Browne 2004). 11. In December 2005, Zuma, sixty-four, was charged with rape of the thirty-one-year-old daughter of a family friend who was staying in his Johannesburg flat. He was acquitted in May 2006, but the trial was highly politicized, and is discussed in more detail below. 12. Indirect Rule and its relevance for anthropology has been ably summarized by Kuper (1973: Chapter 4) and critiqued by Asad (1973). 13. The starkly summarized history of the British colonial and subsequent regimes in South Africa that follows is mainly based on Leonard Thompson’s The History of South Africa (1982 [2001]) and www.sahistory.org.za. 14. Until the apartheid regime stopped it, ‘coloured’ people, the descendants of slaves and Khoisan, were represented by a white MP in Parliament. 15. The ANC, the main black opposition movement, was founded as early as 1912.

The Nkandla Controversy  201 16. Afrikaans-speaking anthropologists cleaved more to continental European schools of anthropology (see e.g. Hammond-Tooke 1997). 17. Adam Kuper has written at some length on the challenges of anthropology in South Africa before the coming of democracy in the Introduction to his South Africa and the Anthropologist (1987). 18. Hunter/Wilson included, along with a conventional ethnography of the amaPondo in their rural reserve, an example of an urban community and the conditions of labour on white-owned farms. Monica Hunter Wilson’s remarkable career has been comprehensively discussed by Andrew and Leslie Bank (2013). 19. Robert Gordon (2018: Chapter 3) provides a full account of Gluckman’s fieldwork in Zululand. 20. Robert Gordon (2018: 403) records a story that Gluckman offered the editors of APS ‘The Bridge’ essay, but they turned it down. This would have been consistent as it was even less consonant with structural functionalism than the chapter they accepted. Even so, the Gluckman chapter and, to a lesser extent, Audrey Richard’s chapter were the only ones that dealt with the interaction of indigenous institutions and colonial authorities. 21. Lobola in Xhosa, boghadi in Sotho. 22. Zuma’s unique Zulu forename is a compression of a phrase concocted by his father, which translates as ‘I can’t keep quiet when someone pretends to love me with a deceitful smile’ (Gordin 2010: 22). 23. Biographical material on Zuma is scarce. The main work, Zuma: A Biography by Jeremy Gordin (2010), is out of date and one has to rely on newspaper articles. Gordin, a South African journalist, was not Zuma’s ‘official’ biographer, but he has been criticized for being overly sympathetic to his subject. If not otherwise referenced, this is the source of the following summary of Zuma’s early career. 24. Successor to King Shaka, who sent 24,000 warriors against the British colonial invaders, inflicting the greatest defeat upon the British Empire since the Indian Mutiny at the Battle of Isandlwana in 1879. 25. The Bambatha rebellion of 1901 was a revolt against the payment of Poll Tax; just thirty-six years before Zuma was born, 1,000 men were massacred close to Nkandla by a colonial militia charged with suppressing the rebellion (Thompson 2001: 98). 26. See http://www.sahistory.org.za/topic/sharpeville-massacre-21-march-1960 (retrieved 18 September 2021) for a summary account of the massacre and the events leading up to it. 27. As a Zulu traditionalist and prominent person, polygyny was his preferred mode of marriage as it conferred prestige, but he was also by inclination ‘a Lady’s Man’, as his many additional liaisons attest. 28. Nkosazana divorced him in 1998 and Kate committed suicide in 2000. 29. Initially he worked for the ANC in KZN and later as a Member of the Executive Council (MEC), the equivalent of a minister at regional level. Currently, MECs receive approximately R1.8 million, and Zuma’s salary in the 1990s would have been in a similar relation to the cost of living: it would have supported a monogamous household very well, but it was quite inadequate for Zuma’s circumstances and the high standard of living that the new political elite expected for themselves. Financial assistance for Zuma came from several benefactors besides Schabir Shaik – even Nelson Mandela funded him with R1 million after he was fired as Thabo Mbeki’s deputy. For the full list, see ‘Zuma the Kept Politician’, AmaBhungane Team, 7 December 2012, www.amaBhungane.co.za (retrieved 2 May 2016). 30. It would have been logical to charge both corruptor and corruptee, but while the Minister of Justice admitted there was a ‘prima facie’ case for doing so, he did not think it

202  Robin Palmer

31. 32. 33. 34.

35. 36. 37. 38. 39. 40. 41.

was ­‘winnable’. Actually he did not want to charge a comrade and the Deputy President (Gordin 2010: 118). Nevertheless, the case rolled on for three more years before it was dropped. For a detailed account of the trial, see Robins (2008); Gordin (2010: Chapters 11–12); Thlabi (2017). An alternative rendering of umzi. ‘Election Setback Proves That Zuma’s Scandals Do Matter’, Sunday Times, 7 August 2016; ‘ANC Admits Zuma Scandals Helped Alienate Middle Class’, Sunday Times, 7 August 2016. RDP stands for Reconstruction and Development Policy, which came in with Mandela and persisted as a priority of the ANC in terms of providing free formal housing for households earning below a certain amount. Unfortunately, this worthy and continuing policy has been abused not only by both corrupt crony construction companies that deliver an inferior product, but also by owners who illegally buy, sell and let the houses. ‘ANC Voters Don’t Care about Nkandla’, Sunday Times, 5 July 2015. ‘Zuma’s R500m Backup Boeing: Air Force to Foot Bill for “Unnecessary” Lease’, Sunday Times, 9 November 2016. ‘Zumaville Development Plans a Go’, Mail & Guardian, 4 May 2014. ‘ANC Loses Nkandla to IFP’, Daily Dispatch, 4 August 2016. ‘It Was All Over But the Crying: And Then Came Khwezi’, Daily Maverick, 6 August 2016. ‘How the 7.8 Million That Zuma Must Pay Back for Nkandla Was Calculated’, Mail & Guardian, 28 June 2016. ‘How Zuma Plans to Pay Back the Money’, City Press, 31 July 2016.

References Asad, Talal. 1973. Anthropology and the Colonial Encounter. New York: Ithaca. Bank, Leslie. 2011. Home Spaces, Street Styles: Contesting Power and Identity in a South African City. Johannesburg: Wits University Press. Bank, Andrew, and Leslie Bank. 2013. Inside African Anthropology: Monica Wilson and Her Interpreters. London: International African Institute and Cambridge University Press. Barrett, Stanley. 1990. The Rebirth of Anthropological Theory. Toronto: University of Toronto Press. Beinart, William. 1980. ‘Labour Migrancy and Rural Production: Pondoland c. 1900–1950’, in Philip Mayer (ed.), Black Villagers in an Industrial Society: Anthropological Perspectives on Labour Migration in South Africa. Cape Town: Oxford University Press, pp. 81–108. ———. 2014. ‘A Century of Migrancy from Pondoland’, in Peter Delius, Laura Phillips and Fiona Rankin-Smith (eds), A Long Way Home: Migrant Worker Worlds 1800–2014. Johannesburg: Wits University Press, pp. 59–73. Beinart, William, and Colin Bundy. 1987. Hidden Struggles in Rural South Africa: Politics and Popular Movements in the Transkei and Eastern Cape 1990–1930. Johannesburg: Ravan Press. Belich, James. 2009. Replenishing the Earth: The Settler Revolution and the Rise of the Anglo-World, 1783–1939. Oxford: Oxford University Press. Bonner, Philip. 1981. ‘The Dynamics of Late Eighteenth Century, Early Nineteenth Century Northern Nguni Society: Some Hypotheses’, in J.B. Peires (ed.), Before and after Shaka: Papers in Nguni History. Grahamstown: Rhodes University, pp. 74–81.

The Nkandla Controversy  203 Bundy, Colin. 1979. The Rise and Fall of the South African Peasantry. London: Heinemann. Butler, Anthony. 2010. ‘The ANC under Jacob Zuma’, in Devan Pillay et al. (eds), New South African Review 1: 2010 Development or Decline? Johannesburg: Wits University Press, pp. 164–83. Crawford-Browne, Terry. 2004. ‘The Arms Deal Scandal’, Review of African Political Economy, 31(100): 329–42. De Wet, Chris. 1995. Moving Together, Drifting Apart: Betterment Planning and Villagisation in a South African Homeland. Johannesburg: Witwatersrand University Press. Elphick, Richard. 1985. Khoikhoi and the Founding of White South Africa. Johannesburg: Ravan Press. Epstein, A.L. (ed.) 1967. The Craft of Social Anthropology. London: Tavistock. Evans-Pritchard, E.E. 1940. The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People. Oxford: Oxford University Press. Fay, Derick. 2002. ‘The Trust Is Over! We Want to Plough! Land, Livelihoods and Reverse ­ Resettlement in South Africa’s Transkei’, Ph.D. dissertation. Boston: Boston University. Fay, Derick, and Robin Palmer. 2000. ‘Prospects for Redistribution of Wealth through Land Reform in Dwesa-Cwebe’, in Ben Cousins (ed.), At the Crossroads: Land and Agrarian Reform in South Africa into the 21st Century. Bellville: University of the Western Cape, Programme for Land and Agrarian Studies, pp. 194–210. ———. 2002. ‘Poverty and Differentiation at Dwesa-Cwebe’, in Robin Palmer, Herman Timmermans and Derick Fay (eds), From Confrontation to Negotiation: Nature-Based Development on South Africa’s Wild Coast. Pretoria: Human Sciences Research Council, pp. 146–72. Fortes, Meyer and Evans-Pritchard, E.E. (eds) 1940a. African Political Systems. Oxford: Oxford University Press. ———. 1940b. ‘Introduction’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. Oxford: Oxford University Press. Gluckman, Max 1940a. ‘The Kingdom of the Zulu of South Africa’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems. London: Oxford University Press. ———. 1940b. ‘Analysis of a Social Situation in Modern Zululand’, Bantu Studies, 14(1): 1–30. ———. 1956. Custom and Conflict in Africa. Oxford: Basil Blackwell. Goody, Jack (ed.). 1958. Developmental Cycle in Domestic Groups. Cambridge: Cambridge University Press. Gordin, Jeremy. 2010. Zuma: A Biography. Johannesburg: Jonathan Ball. Gordon, Robert. 2018. The Enigma of Max Gluckman: The Ethnographic Life of a ‘Luckyman’ in Africa. Lincoln, NE: University of Nebraska Press. Guy, Jeff. 1981. ‘The Political Structure of the Zulu Kingdom during the Reign of Cetscwayo KaMpande’, in J.B. Peires (ed.), Before and after Shaka: Papers in Nguni History. Grahamstown: Rhodes University, pp. 49–74. ———. 2014. ‘Colonial Transformations and the Home’, in M. Healey-Clancy and J. Hickel (eds), Ekhaya: The Politics of Home in Kwa-Zulu Natal. Pietermaritzburg: University of Natal Press, pp. 27–30. Hammond-Tooke, David. 1962. Bhaca Society. Cape Town: Oxford University Press. ———. 1975. Command or Consensus: The Development of Transkeian Local Government. Cape Town: David Philip. ———. 1997. Imperfect Interpreters: South Africa’s Anthropologists, 1920–1990. Johannesburg: Wits University Press.

204  Robin Palmer Hay, Michelle. 2014. ‘Surviving Drought: Migrancy and the Homestead Economy’, in Peter Delius, Laura Phillips and Fiona Rankin-Smith (eds), A Long Way Home: Migrant Worker Worlds 1800–2014. Johannesburg: Wits University Press, pp. 144–54. Healy-Clancy, Meghan and Jason Hickel (eds) 2014. Ekhaya: The Politics of Home in Kwa-Zulu Natal. Pietermaritzburg: University of Natal Press. Hunter, Mark. 2014. ‘Beneath the “Zunami”’: Jacob Zuma and the Gendered Politics of Social Reproduction in South Africa’, in M. Healey-Clancy and J. Hickel (eds), Ekhaya: The Politics of Home in Kwa-Zulu Natal. Pietermaritzburg: University of Natal Press, pp. 233–49. Hunter, Monica. 1936. Reaction to Conquest: Effects of Contact with Europeans on the Pondo of South Africa. London: Oxford University Press. Johnson, R.W. 2017. How Long Will South Africa Survive? The Crisis Continues. Johannesburg: Jonathan Ball. ———. 2019. Fighting for the Dream. Johannesburg: Jonathan Ball. Khene-Pade, Caroline, Mitchell Kavai and Robin Palmer. 2010. ‘A Baseline Study of a Dwesa Rural Community for the Siyakhula Information and Communication Technology for Development Project: Understanding the Reality on the Ground’, Information Development, 26(4): 265–88. Kuper, Adam. 1973. Anthropology and Anthropologists: The Modern British School. London: Routledge & Kegan Paul. ———. 1987. South Africa and the Anthropologist. London: Routledge & Kegan Paul. ———. 1993. ‘The “House” and Zulu Political Structure in the Nineteenth Century’, Journal of African History, 34(3): 469–87. Leach, Edmund. 1954. Political Systems of Highland Burma: A Study of Kachin Social Structure. London: Athlone Press. Manona, Cecil Wele. 1980. ‘Marriage, Family Life and Migrancy in a Ciskei Village’, in Philip Mayer (ed.), Black Villagers in an Industrial Society: Anthropological Perspectives on Labour Migration in South Africa. Cape Town: Oxford University Press, pp. 170–204. Martin, Michaela, and Hussein Solomon. 2016. ‘Understanding the Phenomenon of State Capture in South Africa’, Southern African Peace and Security Studies 5(1): 21–34. Mayer, Philip. 1971. Townsmen or Tribesmen? Cape Town: Oxford University Press. ———. (ed.). 1980. Black Villagers in an Industrial Society: Anthropological Perspectives on Labour Migration in South Africa. Cape Town: Oxford University Press. McAllister, Patrick. 1980. ‘Work, Homestead and the Shades: The Ritual Interpretation of Labour Migration among the Gcaleka’, in Philip Mayer (ed.), Black Villagers in an Industrial Society: Anthropological Perspectives on Labour Migration in South Africa. Cape Town: Oxford University Press, pp. 205–54. ———. 1981. Umsindleko: A Gcaleka Ritual of Incorporation. Grahamstown: Rhodes University. ———. 2001. Building the Homestead: Agriculture, Labour and Beer in South Africa’s Transkei. Aldershot: Ashgate. Murdock, J. 1951. ‘British Social Anthropology’, American Anthropologist, 53: 465–73. Murray, Colin 1981. Families Divided: The Impact of Migrant Labour in Lesotho. Cambridge: Cambridge University Press. Palmer, Robin. 1997a. Rural Adaptations in the Eastern Cape, South Africa. Roma: National University of Lesotho. ———. 1997b. ‘The Transformation of the Town of Keiskammahoek’, in Chris de Wet and Michael Whisson (eds), From Reserve to Region: Apartheid and Social Change in the Keiskammahoek District of (Former) Ciskei: 1950–1990. Grahamstown: Rhodes University, pp. 297–321.

The Nkandla Controversy  205 ———. 2010. ‘ICT4D and the Siyakhula Living Lab: An Anthropological Contribution to Digital Development’, Anthropology Southern Africa, 30(1–2): 19–32. Palmer, Robin and Derick Fay. 2002. ‘The Residents’, in Robin Palmer, Herman Timmermans and Derick Fay (eds), From Confrontation to Negotiation: Nature-Based Development on South Africa’s Wild Coast. Pretoria: Human Sciences Research Council, pp. 15–29. Palmer, Robin and Nick Hamer. 2011. ‘Transkei’s Wild Coast: Development and Frustration at Dwesa-Cwebe Reserve’, in Greg Ruiters (ed.), The Fate of the Eastern Cape: History, Politics and Social Policy. Durban: University of Kwazulu-Natal Press, pp. 289–99. Palmer, Robin, Herman Timmermans and Derick Fay (eds). 2002. From Confrontation to ­Negotiation: Nature-Based Development on South Africa’s Wild Coast. Pretoria: Human ­Sciences Research Council. Pauw, Jacques. 2017. The President’s Keepers: Those Keeping Zuma in Power and out of Prison. Cape Town: Tafelberg. Peires, Jeff. 1981a. ‘Chiefs and Commoners in Precolonial Xhosa History’, in J.B. Peires (ed.), Before and after Shaka: Papers in Nguni History. Grahamstown: Rhodes University, pp. 125–44. ———. 1981b. The House of Phalo: A History of the Xhosa People in the Days of Their Independence. Johannesburg: Ravan Press. ———. 1989. The Dead Will Arise: Nongqawuse and the Great Cattle-Killing Movement of 1856–7. Oxford: James Currey. Radcliffe-Brown, Alfred. 1952. Structure and Function in Primitive Society. London: Cohen and West. Reddy, Micah. 2014. ‘Post-apartheid Migrancy and the Life of a Pondo Mineworker’, in Peter Delius, Laura Phillips and Fiona Rankin-Smith (eds), A Long Way Home: Migrant Worker Worlds 1800–2014. Johannesburg: Wits University Press, pp. 254–63. Robins, Steven. 2008. ‘Sexual Politics and the Zuma Rape Trial’, Journal of Southern African Studies, 34(2): 411–27. Southall, Aidan. 1953. Alur Society: A Study in Processes and Types of Domination. Nairobi: Oxford University Press. ———. 1988. ‘The Segmentary State in Africa and Asia’, Comparative Studies in Society and History 30(1): 52–82. Steinberg, Jonny. 2008. Three Letter Plague: A Young Man’s Journey through a Great Epidemic. Johannesburg: Jonathan Ball. Thompson, Leonard. 2001 [1982]. The History of South Africa. New Haven: Yale University Press, Yale Nota Bene. Thlabi, Redi. 2017. Khwezi: The Remarkable Story of Fezekile Ntsukela Kuzwayo. Johannesburg: Jonathan Ball. Webb, Colin, and John Wright (eds). 2001. The James Stewart Archive, Volume 5. Pietermarizburg: University of Natal Press. Wilson, Monica. 1951. Good Company: A Study of Nyakyusa Age Villages. London: Oxford University Press. ———. 1981. ‘Nguni Markers’, in J.B. Peires (ed.), Before and after Shaka: Papers in Nguni History, Grahamstown: Rhodes University, pp. 145–49. Wolpe, Harold, 1972. ‘Capitalism and Cheap Labour Power in South Africa: From Segregation to Apartheid’, Economy and Society, 1(4): 425–56. Wright, John. 1981. ‘Control of Women’s Labour in the Zulu Kingdom’, in J.B. Peires (ed.), Before and after Shaka: Papers in Nguni History. Grahamstown: Rhodes University, pp. 82–99.

Chapter 10

Rethinking Tswana Kingships and Their Incorporation in Modern Botswana State Formation Ørnulf Gulbrandsen

Fortes and Evans-Pritchard introduced African Political Systems (APS) by stating, as a chief aim, to contribute ‘to the discipline of comparative politics,’ assuming that the collection of essays would bring out ‘all the major principles of African political organization’ (1940a: 1). They asserted that a ‘comparative study of political systems has to be on the abstract plane where social processes are stripped of their cultural idiom and are reduced to functional terms’ (ibid.: 3, emphasis added). As is well known to many generations of undergraduate students of anthropology, this approach led Fortes and Evans-Pritchard to identify two major categories of political systems in Africa – centralized and segmentary – with a particular concern with their order, which was mainly seen as a matter of ‘balance of forces’. However, these authors made no effort in the introductory chapter to APS to pursue their comparative ambition further by reflecting upon ‘all the major principles’ identified in a comparative perspective transcending Africa. As I want to show in this chapter, such a broader comparative perspective is useful for rethinking important sociocultural dimensions of African kingships, which are of considerable significance to their incorporation in postcolonial state formations. More specifically, while the relationship between indigenous polities and modern state formation in Africa has, as shown by Bayart (1993) and many others, often been riddled with serious conflicts, causing unstable and weak state governments, the symbolism and sociopolitical institutions of Tswana kingships have, apparently paradoxically, proved mostly conducive – if not ­indispensable – to the formation of a strong and stable republican ­democracy in Botswana.

Rethinking Tswana Kingships and Their Incorporation  207

Nevertheless, as we shall see in the second part of this chapter, the encounter between, on the one hand, value orientations centred in Tswana kingship deeply ingrained in the population and, on the other hand, modern elites’ strong adherence to entrepreneurial practices of liberal individualism has caused major societal contradictions and tensions due to escalating discrepancies of income and wealth, propelled by Botswana’s beef and diamond-driven political economy. I am centrally concerned with how the modern political elites’ appropriation of the symbolic wealth and institutional structures of Tswana kingships has conditioned massive, but often tacit, repressive practices and structures of domination. This is, however, not a matter of ‘resurgence’ of ‘traditional authorities’ to enhance sovereignty of the modern state in Africa (see e.g. Englebert 2002; Sklar 2005; cf. Buur and Kyed 2007), since Tswana kingships were captured into the process of state formation at the outset in this country (Gulbrandsen 2014: 97ff.; cf. Comaroff and Comaroff 2018: 39 for a general discussion of the continuous significance of traditional authorities). As a point of departure for rethinking such an African political system as that of Tswana kingship in a larger comparative context and in relation to modern state formation, I find most helpful an important – but seemingly not much recognized – section in the introductory chapter to APS where Fortes and ­Evans-Pritchard transcend their notion of ‘social processes [that] are stripped of their cultural idiom’, as quoted above. I have in mind the section entitled ‘The Mystical Values Associated with Political Office’ (1940a: 16ff.), which addresses, at length, cultural idioms of great significance to the construction of African kingship. I have found these elaborations inspiring in my efforts to develop an approach suitable for conceptualizing the distinctiveness of kingship in Africa in relation to kingship beyond that continent. Moreover, they are most useful for comprehending the encounter between kingship in Africa and Western modernity in the context of postcolonial state formation. Within the space constraints of this chapter, I shall, as already suggested, approach these large issues via an ethnographic focus upon one of the examples of centralized political systems represented in APS, the Tswana of the present-day Botswana,1 who were discussed in that volume by Isaac Schapera and amongst whom I have myself undertaken extensive anthropological research, albeit at a much later date than Schapera.2 In a critical comment, Dumont (1972: 331, n. 32g) stated in Homo Hierarchicus that ‘one of the tendencies in [the introduction to] the classic African Political Systems is to reduce the religious functions of the king to his political functions’ (emphasis added). However, this complaint is surprising as the introductory chapter contains passages like the following: An African ruler is not to his people merely a person who can enforce his will on them. He is the axis of their political relations, the symbol of

208  Ørnulf Gulbrandsen

their unity and exclusiveness, and the embodiment of their essential values. He is more than a secular ruler … His credentials are mystical and derived from antiquity. (Fortes and Evans Pritchard 1940a: 16, emphasis added) Moreover, Fortes and Evans Pritchard pointed out that ‘sacred symbols, which reflect the social system, endow it with mystical values which evoke acceptance of the social order that goes far above obedience exacted by the secular sanction of force’ (ibid.: 17–18). In their conception, ‘mystical values’ are, to people, ‘outstanding values’ that are ‘dramatized in the great public ceremonies and bound up with their key political institutions’ (ibid.: 21). All this relates to the value of the ‘common interests of the whole society’, which are of ‘general import to the basic elements of existence’ and include ‘fertility, health, prosperity, peace, justice … everything [that] gives life and happiness to a people’ (ibid.: 18). The ‘mystical values’ are ‘their essential values’ because they ‘stand for the greatest common interest of the widest political community to which a member of a particular African society belongs’ (ibid.: 21, emphasis added). In other words, this is a matter of values of the highest existential importance to everybody, the supreme this-worldly guardian of which is the incumbent of the kingship, who, I requote, ‘is more than a secular ruler’. Although these notions make it clear that Fortes and Evans-Pritchard were completely aware of the centrality of the spiritual dimensions – or ‘religious functions’ in Dumont’s parlance – of African kingships, I nevertheless take Dumont’s complaint seriously. Together with Fortes and Evans-Pritchard’s pinpointing of crucial African value orientations, it should challenge us to identify distinctive features of kingship in Africa that are generally neglected in the ethnographic case studies of APS. In this regard, the following statement of Dumont about the chief features of ‘traditional society’ – in his conception, all societies except those of the modern West – reflects an image of Eurasian kingship that is helpful for beginning to think about how ‘religion’ should be conceived in an African kingship such as that of the Tswana. He argues that: As we [modern Westerners] live in an egalitarian society, we tend to think of hierarchy as a scale of commanding powers – as in an army – rather than as a graduation of statuses … [T]he very word hierarchy, and its history, should recall that the graduation of status is rooted in religion: the first rank normally goes, not to power, but to religion, simply because for those [‘traditional’] societies religion represents what Hegel has called the Universal, i.e. absolute truth, in other words because hierarchy integrates the society in relation to its ultimate values. (Dumont 1970: 67, emphasis added)

Rethinking Tswana Kingships and Their Incorporation  209

This can be read as a version of Dumont’s principal rationale for calling for the delineation of the ‘religious functions’ of premodern systems such as that of African kingdoms, which, in his scheme, require a category of ‘religion’ that is distinct from ‘power’ – or ‘politics’. Of course, this is a distinction Dumont applied especially in the case of the Indian caste system regarding the separation of the Brahman from the king, with the former conceived as always superior in rank to the latter. Notwithstanding all the controversies amongst Indian specialists about his insistence on the superiority of ‘religion’, my concern here is Dumont’s claim that there is a separation of ‘politics’ from ‘religion’ that is universally valid for ‘traditional’ societies. This is a claim that I want to challenge, as I argue that in the Tswana case, Dumont’s distinction obscures more than it reveals. Nonetheless, this critical observation invites us to discuss why the conventional Western religion/politics or sacred/secular divide (upon which Dumont relies) is not readily applicable to Africa. As we shall see, this discussion brings out some salient and distinctive features of Tswana kingship that, I presume, pertain to kingship elsewhere in Africa too. Furthermore, Dumont suggests in the quotation above that ‘traditional ­kingship’ – in the singular – prevails everywhere in a societal order centred on a hierarchy of statuses that ‘integrates society in relation to its ultimate values’ of ‘absolute truth’ – a point he tries to sustain by invoking Hegel’s notion of ‘the Universal’. What Dumont has in mind here is obviously Indian caste society, where, in his conception, society’s ‘ultimate values’ – in terms of purity-­impurity – underpin a major division of (caste) statuses in a hierarchical order that is basically kept integrated by virtue of the all-encompassing force of these ‘ultimate values’. However, as I shall demonstrate in the case of Tswana kingship, ‘the absolute truth’ – or ‘the Universal’ in Hegel’s sense – might well be all-encompassing without conforming to Dumont’s conception of hierarchy, that is, to a radical discrimination of categories of people in a status hierarchy like that of the Indian caste system and enforced by the imperative value of purity. Thus, although an African kingship like that of the Tswana is organized, in certain important respects, according to principles of rank, I shall argue that, unlike the case of caste India, the all-encompassing value of social order unifies people culturally rather than providing cultural conditions for discrimination, separation and division. (However, as I shall discuss subsequently, this is not to say that there is no mechanism of exclusion.) This means it is a kingship-centred society that embodies an egalitarian value orientation, though one that does not combine it with individualism as in Western modernity, but instead a holistic order. The all-encompassing holistic value upon which Tswana kingdoms rely seems to have a quality of ‘absolute truth’ akin to Hegel’s notion of ‘the Universal’, which, it should be noted, he did not confine to ‘religion’, but that instead was framed as a much broader conception of ‘spirit [Geist]’ (see, for example, Hegel 1991: §263).

210  Ørnulf Gulbrandsen

In pursuing these issues in Part One that now follows, I am chiefly concerned with the symbolism related to what Fortes and Evans-Pritchard speak of as ‘mystical values’ of ‘greatest common interest’. Subsequently, I discuss the significance of these value orientations in practice, in the context of Tswana kingship sociopolitical realties. I shall be particularly concerned with contradictions intrinsic to Tswana kingship between celebrated holistic values and commonly feared individualistic practices of greed for power and wealth – in other words, as I shall discuss in Part Two, contradictions that have been amplified under the circumstances of the modern state.

Part One: Towards an Understanding of the Distinctiveness of Tswana Kingship First, I shall pursue the issue suggested by this heading by discussing spiritual dimensions in Tswana kingship, with the major aim of explaining how tightly integral ‘religion’ is to what, from a Western point of view, should be classified as ‘politics’ or ‘jurisprudence’. Second, with the point of departure in my comprehension of Tswana ontology with respect to the spiritual dimensions of social order, I challenge the common Western notion that kingship is fundamentally hierarchical in its value orientation, as suggested in the quotation of Dumont above. Third, I shall move from ideals and values of Tswana kingship to ­exercises – including abuses – of power. I shall relate this, on the one hand, to Tswana ideals of the common good and social order in Dumont’s holistic sense, and, on the other hand, to individuals’ greed for power and wealth, which in Tswana imagination is associated with secretive – and hence destructive – exercise of power, above all in the form of occult practices.

On Spiritual Dimensions of Tswana Kingship The Tswana royal kgotla – the king’s (kgosi, pl. dikgosi) residence and the supreme council site – is the imaginary location of the bogosi (kingship), which in Tswana cosmology is anchored in the royal ancestorhood (badimo ya dikgosi), the supreme custodian of Tswana morality, customs and law – mekgwa le melao ya Setswana (‘Tswana custom and law’). The kgosi is the ultimate living custodian of Tswana morality and, by virtue of being installed by the morafe (nation, pl. merafe) as the incumbent of the bogosi, is responsible for ensuring that the population lives in accordance with this moral and normative code. The highly elevated position of the kgosi is distinctly expressed in the Tswana maxim of kgosi ke modingwana, ga e sebjwe (‘the kgosi is a little god, no wrong must be spoken of him’). This suggests a spiritual dimension of Tswana kingship that has often enough been seen as deteriorating due to many of the dikgosi accepting Christianity at an exceptionally early stage (in the larger African context) and allowing evangelizing missionaries to establish church congregations, in which the dikgosi and other

Rethinking Tswana Kingships and Their Incorporation  211

high-ranking men took an active part. Dikgosi often went to the extent of allowing these churches to have a monopoly over evangelization: they virtually became a form of state church (Gulbrandsen 1993a: 44 and 2001: 49ff.; Landau 1995: 51–52). In his contribution to APS, Schapera appears to have taken for granted the fact that the dikgosi’s adherence to Christianity and their abandonment of some ritual ceremonies meant that ‘the people no longer look to the chief [kgosi] for spiritual benefit’ and that the dikgosi received much less reverence ‘as tribal priest and magician’ (Schapera 1940: 76). This understanding of the impact of Christianity is, however, largely premised upon a sacred/secular dichotomy3 that is, as I shall now argue, incongruous in an African context like that of the Tswana, since the Tswana did not abandon their belief that the rule of the dikgosi depended greatly upon their ability to communicate well with the royal ancestors. This validation is, in many aspects, perceived as a matter of spiritual intercourse largely beyond public scrutiny. Similarly, great secrecy normally surrounds dikgosi’s engagement with the other principal source of spiritual power – that is, powerful ‘magicians’, known as dingaka (sing. ngaka). Public confirmation of ancestral support through a kgosi’s demonstration of force, wisdom and knowledge is principally made in what appears, from a Western point of view, to be a typical politico-jural field: the kgotla (pl. dikgotla), which denotes both the council place, the council and those people assigned to a particular kgotla. It is the core institution found in every level, from the royal centre to the elementary sociopolitical unit of the descent group. While the kgotla is, from a Western perspective, readily perceived as distinctively secular, it embodies a discursive field that is, in Tswana comprehension, encompassed by the badimo (ancestral spirits) who are (in fact or imagination) buried in the cattle kraal, adjacent to the kgotla. This discursive field thus connects men, by virtue of the mediation of the kgotla elders, to the spiritual agencies of supreme forces. This means that the apparently ‘secular’ field of kgotla discourse has, in people’s imagination, remained closely connected with the badimo, the major custodians of moral order, as inscribed in mekgwa le melao ya Setswana. More than once, it was suggested to me that ‘in the kgotla we are with badimo and they are with us’, whether in the royal kgotla or in the kgotla at the descent-group level (Gulbrandsen 2014: 173; cf. Gulbrandsen 1993a, 2001). It is true that Tswana dikgosi, like other African kings, used to perform certain rituals in public ceremonies, especially in the pre-Christian contexts, like bogwera/bojale (male/female initiation), rainmaking and go loma ngwaga (‘to bite the new year’ – the first-fruit ceremony). These are the ceremonies and ritual practices that the evangelizing missionaries condemned as pagan and that many dikgosi therefore abandoned or continued to practice secretly,4 as in the case of rainmaking. It is highly significant that the dikgosi could make such changes with relatively limited popular protest because their authority was always supported by people’s spiritual satisfaction with their extensive participation, as I

212  Ørnulf Gulbrandsen

have ­explained, in the apparently ‘secular’ field of kgotla, the principal discursive field for sustaining peace and order by resolving conflicts in accordance with ancestral morality and law. Herein lies a chief condition for the popular celebration, even today, of Tswana kingship’s (bogosi) – and its incumbent, the kgosi’s – great significance as the ultimate source of peace and social order, known as kagiso. Kagiso is, as I shall now explain, perceived as extremely important, even of great existential significance, to everybody, and satisfies most indicators of Ortner’s (1973: 1339) conception of a ‘key symbol’. In Tswana comprehension, kagiso features as the all-encompassing value of the common good, patently resembling Fortes and Evans-Pritchard’s above-mentioned notion of ‘mystical value.’ Kagiso is a matter of a particular orientation that instils in everybody a fundamental fear of social tension and conflict, which, it should be noted, is not restricted to the parties that are directly involved. Everybody located in a social sphere inflicted with conflicts and tension is perceived as vulnerable to disasters caused by the heated (mogote) atmosphere thus generated, like the infertility of human beings and domestic animals, poor crop yields, sickness and bad health, and even death (Gulbrandsen 1995; Gulbrandsen 2014: 174f.). In addition to the avoidance of calamities, the maintenance of kagiso is i­ndispensable for prosperity in all the vital matters of life, as it is the ­fundamental condition for human, crop and domestic animal fertility, prosperity and welfare – in brief, for life. This is a principal Tswana value orientation, expressed metaphorically in maxims like ra, ba bonolo ba sa tshedile ka bonolo jwa bone (the dwellings of fierce men become ruins in ashes, the meek live quietly by reason of their meekness) and kagiso ke go bona mabele (peace gives plenty of corn). The all-embracing character of kagiso is reflected in the fact that everyone, collectively and individually, is responsible for maintaining social order by conducting their lives according to the normative repertoire of megwa le melao ya Setswana (Tswana custom and law), the proper practising of which is indigenously conceived as essential in order to prevent social tension and conflicts from arising. Thus, there is great concern with people’s conduct in relation to each other, which is much a question of whether a person is endowed with a good character, known as botho. A person’s botho depends partly on dispositions profoundly rooted in a person’s inner self, as given by God, ancestors or otherwise inborn (see Alverson 1978: 112–14; cf. Livingston 2008; Klaits 2010: 186; Durham 2011: 145ff.), and partly upon the dispositions inculcated from early childhood through intensive disciplinary interventions to install disposition for civilized behaviour (maitseo). More than once, I witnessed how elderly people took one of their grey hairs and, as a reward, planted it within the hairs of a well-behaving youngster – the grey hair signifying masegò, meaning good luck; a symbolic practice associating botho with fortune and prosperity.

Rethinking Tswana Kingships and Their Incorporation  213

In my experience, botho is, ideally, believed to be ingrained in everyone: embodied as a most important dimension of habitus, as indeed a profound and pervasive precondition for behaving respectfully and inoffensively. I have been struck many times by how kindly, even warmly, the Tswana typically approach each other, seeming eager to create an enjoyable, harmonious atmosphere, characterized much by laughter, affable commensality and expressions of love5 – all certainly felt, more or less consciously, to be conducive to kagiso. Conversely, a person whose behaviour causes bad feelings, tensions and conflicts destructive to kagiso is readily condemned, besides being vulnerable to harm inflicted by elders’ or ancestors’ sadness known as dikgaba (Schapera 1970 [1934]: 298ff; Schapera 1971b [1940]: 228f.; Gulbrandsen 1995: 427; Lambek and Solway 2001; Klaits 2010: 5). Botho is a virtue, above all, because it is perceived as counteracting that which is destructive to kagiso, and hence is conceived as an existentially important disposition that is perceived as ideally embodied, I reiterate, in everyone, to the blessing of oneself, the family and, on an aggregate level, the whole nation. Moreover, the consequent strong commitment to kagiso involves a moral obligation to participate in reconciliation (tetlanyò) between conflicting parties to bring them to mutual understanding6 (kutlwanò). The ultimate source of kagiso is the moral order of the royal ancestry (badimo ya dikgosi) and its custodian – the kgosi. However, this spiritual custodianship is not displayed primarily in terms of practices that, from a Western point of view, appear as a form of ‘religious’ ritual, but in the form of kgotla proceedings of legislation and litigation directed towards conflict resolution and reconciliation in service of the supreme value of kagiso. These counterconflict proceedings are at the core of the Tswana lifeworld and have been an almost everyday activity for adult men at all levels of Tswana society – from the elementary descent-group council (kgotla) up the hierarchy of larger sociopolitical units to the royal kgotla, over which presides the kgosi or one of his deputies. These ubiquitous processes are directed towards creating consensus (dumèla) about rules and judgements of action, and, above all, reducing mogote (‘heat’), in the cause of cultivating kagiso through the reconciliation of conflicting parties.7 The capacity of any Tswana kgosi and kgosana to ensure that the kgotla proceedings they head work harmoniously and have a consensual constructive outcome is continuously assessed by subjects, who often speak of such capacity as a matter of personal character (botho). The proceedings of the discursive field of the kgotla might be conceived in terms of Mauss’ concept of ‘total social fact’ (1990: 78, emphasis in original), which captures ‘phenomena [that] are at the same time juridical, economic, religious, and even aesthetic and morphological, etc.’ (ibid.: 79). The highly elaborate proceedings of the kgotla potentially bring into play the whole register of Tswana morality as well as other normative concerns, which are not restricted to a distinctively ‘religious’ sphere. It is in authorizing laws – often in response to changing societal conditions – and by exercising his authority as the supreme

214  Ørnulf Gulbrandsen

judge that the kgosi impose the cosmologically anchored ancestral morality and normative order, essential to social peace and harmony and hence kagiso. The issues dealt with span all aspects of life, and great importance is placed on the aesthetics proceedings, especially the oratory capacity of the participants. While values of social harmony, peace and order can be traced in a vast number of human societies, especially across the Bantu-speaking peoples of Southern Africa (see Gulbrandsen 2014: 198f.), there is a distinctive sociocultural quality to, at least, the Tswana notion of social order that is of significance for comprehending why kagiso is not simply a matter of rulers imposing a ‘harmony ideology’ (Nader 1990: 291). Instead, kagiso is the existentially critical value immanent in all social relations, at every level, and in any context of a Tswana morafe. In this sense, kagiso has a transcending spiritual value, a quality of absolute truth, apparently akin to Hegel’s conception of ‘the Universal’. Moreover, this means that kagiso, as the existentially critical condition for everybody’s health, prosperity and welfare is equally important for peace and order as the processes captured by Fortes and Evans-Pritchard (1940a: 11) central notion of ‘balance between conflicting tendencies and between divergent interests’. This is a balance conceived in terms of their structure-functionalist-­ oriented, comparative political sociology developed in the West, and then ­applied to premodern Africa. I do not deny the theoretical significance of this notion, but would indicate important additional ways in which ideals of social order are preserved through pervasive everyday practices of consensus-making and reconciliation, as in the Tswana case in the pursuit of kagiso. In other words, I would foreground processes anchored in an African cosmology and working in a sociocultural context patently distinguishable from that of Eurasian kingdoms. In this regard, I am especially concerned with the population’s intimate integration in the structures and processes like that of the Tswana kingdoms, with, as I have explained, clear and strong continuities – in a wide range of ways – between the royal centre and the rank and file on the ground. This means, more generally, that we have here a kind of kingdom anchored in the population as a whole,8 and in a way, and to an extent, that contrasts sharply with the whole range of Eurasian kingship systems, whether feudal ­Europe, caste-oriented India or Southeast Asian kingdoms like the ‘Theatre State’ of Negra in Bali (Geertz 1980). In this broader perspective, it is an intriguing point that kagiso, seen as an all-encompassing value in Dumont’s sense, contrasts sharply with that of the Indian caste system: purity. For whereas the overarching value of purity in the Indian case entails exclusion, division and thus the formation of hierarchical order that separates people into categories of radically different worth, the value of kagiso is a matter of highly inclusive commensality with, as I shall argue below, an important egalitarian dimension. In this respect, this example of African kingship is distinctive in relation not only to the social structures of caste India, but also to those of feudal Europe, and

Rethinking Tswana Kingships and Their Incorporation  215

many of the ‘traditional kingdoms’ (as identified by Dumont) located throughout Eurasia. The way in which the people of the Tswana, located within an African k­ ingship, are integral to processes that work in support of kagiso in kgotla ­proceedings – where the wealth of wisdom, knowledge and experience vested in the population is always in demand – reflects orientations that would probably have pleased Plato. I especially have in mind a passage from The Republic ­(462a–b), in which value is placed upon peace and social harmony, especially through cooperation and friendship among different social groups. Plato also warned against factionalism and other forms of social conflict, hence underscoring the significance of good government to sustaining a highly inclusive societal order as a principal common good. Moreover, the case of the ancient Greek city-state suggests how a ‘traditional’ centralized political system can also be partially principled upon egalitarian values, a point upon I shall now elaborate with ­reference to the Tswana.

Virtues of Rank, Values of Equality Before I pursue issues of equality, let me briefly be more explicit about Tswana values associated with rank and their manifestation in certain hierarchical relationships, which is most evident in the discursive field of the kgotla. As indicated above, this discursive field is found at every level of Tswana kingdoms: all the way down the social ladder from the royal centre to the everyday lived-in-world of the elementary unit of sociopolitical organization, which is a relativity small, agnatically centred descent group. This is the elementary unit of any Tswana morafe and each is located in a distinct part of a village (assigned by the kgosi), which has their kgotla at the centre. A certain number of such descent groups (normally five to seven) make up a ward (see Schapera 1984 [1938]: 19ff.), within which the descent groups are ranked. The headman (kgosana, pl. dikgosana) of the senior descent group is the head of the whole ward. Unsettled disputes within one of the descent groups are appealed to its superior kgosana’s kgotla, whose decision may in turn be appealed up to the kgosi’s kgotla. At each level, the kgotla is under the authority of a kgosana, recruited, in principle, according to agnatic genealogical seniority, although his authority also depends on the kgosi’s delegation. Like the kgosi, the dikgosana, at each level, head a council, located in the discursive field of the kgotla, whose members are internally ranked, again basically according to agnatic seniority. The all-encompassing ranking at the core of the hierarchical structure of the Tswana kingdom is anchored in the royal ancestry and goes from the kgosi down the ladder to the ward kgosana and then the descent group kgosana. The authority of the head of a kgotla is epitomized by him always concluding matters after everyone else has had a chance to express their views, which, according to the discursive principles of the kgotla, is done in an inverse ranked order, so that the person next in rank to the head of the kgotla speaks last, before the head

216  Ørnulf Gulbrandsen

concludes the matter. As this suggests, the symbolism of rank is pervasive, manifesting from the royal centre all the way down to the elementary descent group. The omnipresence of symbolism for conducting respectful behaviour in relation to a senior person is not imposed from the apex of the hierarchical order. The highly sophisticated semiotics of respect are inculcated through routine socialization in the family context, which is evident in such elementary relationships as those between elder and younger brothers. These well-institutionalized practices of expressing symbolically respect towards a person of higher rank are, in Tswana comprehension, crucial conditions for sustaining peace and social order – kagiso – because the principles of ranking are thought of as militating against conflict-generating rivalries, especially during kgotla proceedings. However, the irony is that the deep concern with the value of rank reflects the perception of an ever-present potential for conflict-generating rivalry, as I shall elaborate upon in the following section. This account suggests how tightly the population is integrated within the Tswana politico-jural structures of ranking. As I have discussed elsewhere, the Tswana virtually live their lives in courts (Gulbrandsen 1996b, 2007; Gulbrandsen 2014: 169ff), which is, of course, a radically different situation from that of the feudal kingdoms of Europe and a wide range of premodern Asian states. This contrast is underscored by the all-inclusiveness of the kgotla, which is based upon the maxim of freedom of speech for everyone (mmualebe o a be a bua la gagwe). The distinctive character of an African kingship such as that of the Tswana is, as I have discussed extensively elsewhere (Gulbrandsen 1995), epitomized by the Tswana dictum of kgosi ke kgosi ka morafe (the kgosi is kgosi by virtue of the nation, i.e. the people). This is the basic, constitutive principle of Tswana kingship emphasizing that there are two major entities within its order: on the one hand, the kgosi and, on the other hand, the people (in this context, an undifferentiated collective). By virtue of this principle, sovereignty is vested in the people – to which the kgosi, in the final analysis, is subject,9 meaning that this codification of sovereignty basically relies on a democratic and egalitarian value orientation. Thus, the case of the Tswana supports Graeber and Sahlins’ general claim about premodern kingships: ‘the principle that the sovereignty of the king is delegated by the people, to whom it belongs by origin and by right’ (2017: 7).10 That Tswana sovereignty is in the kgosi’s custodianship only while grounded in the population is asserted by the Tswana dictum of molao sefofu, obile otle oje mong waone (the law is blind, it eats even its owner),11 which means that equality before the law includes even the kgosi. This egalitarian value orientation amongst Tswana is also underscored by another dictum: malao tau, oloma lemokgokgo (the law is a lion, it bites the great man too). Moreover, ‘should any Chief [kgosi] act contrary to the public opinion … the result would be disaster’ (Schapera 1984 [1938]: 83; see also Roberts 1985: 77).12

Rethinking Tswana Kingships and Their Incorporation  217

The expectation that the kgosi should be in service of the common good – and hence equally to all citizens – is perfectly expressed in the Tswana notion that the kgosi is, ideally, not only rich, but also generous, the source of wealth for all (the ideal kgosi is the motswadintle – the one from whom good things come). This observation is further supported by such Tswana maxims as moja morago kgosi (the king eats last) and kgosi ke mosadi wa morafe (the king is the wife of the nation). Thus, Wylie says that: the chief [kgosi] dined daily in the open air, taking pieces of the boiled beef … distributing them in his fingers to each one of the dozen rich men who always accompanied him … The ritual advertised before all villagers the mutual dependency and ideal solidarity … The chief had shown himself to be ‘first among equals’ and a guarantor of feasts. (1990: 32, emphasis added) That sovereignty is ideally located in the population, rather than in the kgosi, indicates a form of interrelationship between the kgosi and the people that is obviously contrary to the Hobbesian conception of the social contract, while being quite akin to that of Rousseau. The Tswana virtue of extensive kgotla debates that should include everyone (‘traditionally’ meaning every household head) seems to resemble some salient features of Rousseau’s notion of ‘the general will’, conceived as a matter of reaching a general agreement amongst all citizens. He states that ‘[a]ll citizens have the right to contribute personally, or through their representatives, to its formation’,13 meaning that each person contributes to the general will of the collective of citizens, the total social body. Furthermore, Rousseau emphasizes that ‘every authentic act of the general will, binds or favours all the citizens equally’.14 All this suggests considerable correspondence between the notions of this Enlightenment thinker and Tswana ideals of highly inclusive discursive practices in the kgotla in service of the common good. However, there is also an instructive difference between Rousseau’s Enlightenment and Tswana virtues that I want to discuss with reference to what Rousseau stated as ‘the fundamental problem’ – in other words: How to find a form of association which will defend the person and goods of each member with the collective force of all, and under which each individual, while uniting himself with others, obeys no one but himself, and remains as free as before.15 Rousseau asserted that ‘the social contract’ holds the solution to this ‘fundamental problem’, which is a matter of reconciling the collective interest in an ‘association which will defend the person and goods of each member’ with the dictum of ‘each individual’ remaining ‘as free as before’. It is beyond the scope of this

218  Ørnulf Gulbrandsen

chapter to discuss how well Rousseau resolved this conundrum, which seems to embody one of the basic contradictions of Western modernity. The point I want to make here is that, in the context of Tswana holism, there was no such ‘fundamental problem’, since individual freedom was clearly subordinated the holistic order of Tswana kingship. However, as I shall discuss in Part Two below, this became a radically different matter under the circumstances of the modern state in Botswana, with massive intrusion of Western egalitarian individualism. Presently, I thus want to underscore that the form of egalitarian value orientation I have identified within Tswana kingdoms is of a radically different kind from that of the Enlightenment, where it is always coupled with individual freedom. By contrast, amongst the Tswana there is a distinctive form of egalitarian value orientation that is always coupled with the overarching value of the common good – above all, peace, harmony and social order as epitomized by kagiso. This value orientation seems akin to what Dumont speaks of in universal terms as ‘holism’, claiming that ‘most societies value, in the first place, order: the conformity of every element to its role in the society – in a word, the society as a whole’ (1977: 4). He contrasts this to societies – especially those of the modern West – that ‘value, in the first place, the individual human being’ (cf. Kapferer 2010: 193ff.). In terms of this conception, we can say that Western virtues of egalitarian individualism are perceived by Tswana as antagonistic to ideals of a holistic order; whereas their form of egalitarian value orientation is constitutive of sovereignty and hence is necessarily compatible with virtues of Tswana holistic order. Therefore, generally speaking, ‘egalitarianism’ – as an anthropological ­concept – should be used with great care, since its significance differs widely from one sociocultural context to another. Finally, I shall briefly highlight a complicating aspect of egalitarianism amongst the Tswana. In the previous section, I placed great emphasis upon the high degree of integration of people within the order of Tswana kingship. There is, however, a very important reservation to be made that is not dissimilar to those often made with respect to democracy in ancient Greek city-states (cf. Robinson 2004) – that is, there are categories of people who are part of the order, but are not on terms equal to those of the majority. In most of the Tswana kingdoms, there are a number of groups located within its realm, and that are hence subject to its social order, who are denied residence in the royal towns and indeed to appear in the kgotla and enjoy the right of freedom of speech. These are largely serfs labouring in agriculture and cattle herding, as well as groups depending on hunting and gathering like San-speaking peoples. Before Botswana’s independence, they were deprived of full citizenship rights by their Tswana overlords and were frequently stigmatized as inferior human beings and, accordingly, harshly treated, a feature of Botswana that is still far from having been eliminated (Solway 1994, 2004, 2009; Sapignoli 2015, 2018; Wilmsen 1989; see also Gulbrandsen 1991; 2014: 211ff.; Schapera 1984: 251f; Werbner 2002a).16

Rethinking Tswana Kingships and Their Incorporation  219

Hence, while egalitarianism is often thought of as a particularly positive value orientation, the Tswana – in the past and even to some extent in the p ­ resent – provide an example in which egalitarianism is valid only within certain limits, and therefore entails humiliating exclusion and even brutal domination too. This possibility is, to be sure, evident in a vast number of contexts, in addition to ancient Greece (combining democracy and slavery), ranging from caste India practising egalitarian principles within castes and harsh discrimination between them (Parry 1974: 110), to Australia, where egalitarian individualism can be seen, argues Kapferer (2012: 183ff. and cf. 377ff.), as ‘a motivating force in interethnic violence’, and to the United States, for which Myrdal argued that ‘the dogma of racial inequality’ is almost ‘the only way out for a people so moralistically egalitarian’ (1944: 89). This brings us to questions about the significance of such value and normative orientations – or ‘mystical values’ – with which I have been concerned so far, in the context of Tswana kingship sociopolitical realities.

Value Orientations in Practice Virtually everywhere, it seems, kingship has been plagued by dynastic competition for power. In the West, celebrated writers like Shakespeare and Machiavelli focused precisely on dynastic intrigues as a way of revealing the tremendous significance of power as a dimension of humanity, and with particular attention to its destructive potential. Thus, Gluckman (1970: 27ff.) opened an important essay on African kingship, ‘The Frailty of Authority’, by invoking the passage in Macbeth where Malcolm – wanting to test Macduff’s intentions in encouraging him to lead an army against the tyrant Macbeth – compares his character with that of the ideal king: The king-becoming graces, As justice, verity, temperance, stableness, Bounty, perseverance, mercy, lowliness, Devotion, patience, courage, fortitude, I have no relish of them17 Gluckman comments that ‘there develops frequently a conflict between the ideals of leadership and the weakness of a leader. This is the frailty in authority. For it is likely that as a leader exhibits his weakness – natural human weaknesses though they be – his subordinates may begin to question his authority, to turn against him’ (1970: 27). A systematic discussion of dynastic conflicts across a range of Southern African kingdoms led Schapera (1963: 157ff.) to conclude that only in a few cases were such conflicts solely a matter of a ruler’s tyranny or misgovernment. More often, conflict within royal families ‘was caused by such events as real or attempted usurpation of the chieftainship, personal jealousies or grievances between a chief and his relatives and attempts to remove real or potential rivals’

220  Ørnulf Gulbrandsen

(ibid.: 165). The general condition of polygamy amongst such chiefs generated many pretenders to high offices, leaving the ground particularly well prepared for dynastic intrigues and rivalries. Amongst the Tswana, such complications of kingship are reflected in a strong concern with the capabilities and character of their kgosi – as well as authority figures further down the hierarchical sociopolitical order (Gulbrandsen 1995: 424f.). I shall address these complications by arguing that the great value the Tswana place upon the holistic value of kagiso reflects their anxiety about certain individual dispositions, especially amongst people of authority. Ideally, as I have explained, the kgosi – as the incumbent of bogosi – should be the superior living agent of kagiso. However, in Tswana experience, he may possibly either be too weak to exercise authority with sufficient force to ensure social order18 or, still worse, he may himself be the source of disorder, as epitomized by engaging in occult practices of sorcery (boloi). While the Tswana are concerned with individuality, especially amongst people of authority, they are certainly not alone on the continent in this respect, as ‘African societies did, in times past, have a place for individuality, personal agency, property, privacy, biography, signature, and authored action upon the world’ (Comaroff and Comaroff 2001: 277; for the Tswana, see Alverson 1978: 110ff.). Yet, it should be noted that this was not of a matter of appreciating individuality in the modern Western sense, which places immense value upon the freedom of individual development (Simmel 1971: 271ff.); it was rather a matter of appreciating certain personal capacities within a restricted, conventional range conducive to social order and thus kagiso. Thus, in the case of the Tswana, there is a popular anxiety that a kgosi or kgosana might transcend this range by acting, from greed for power and wealth, in evil and harmful ways in relation to their people. In extreme cases, Tswana rulers of this kind have been dethroned or even killed by their people, as was the case with Kgosi Motswasele II of Bakwena (ca. 1807–22), who ‘treated his people harshly, robbing them of their cattle and crops, and even of their wives, and was ruthless in imposing the death penalty. Some of the royal headmen led by his younger brother Segokotlo … plotted against him, leading to public assassination in a tribal assembly’ (Schapera 1965: 127, cf. Gulbrandsen 1995: 430ff.). This is, to be sure, an extraordinary case of a person perceived as having an acute lack of the kind of personal character (botho) upon which, as I have explained, the Tswana place immense value. Moreover, the perceived destructive impact on kagiso was certainly considered as unusually threatening to everybody in this case, with possibly wide-ranging damaging ramifications on the entire morafe. I shall return to this issue by addressing issues of greed and selfishness, but first I want to discuss the significance of other dimensions of personal character. A pertinent point of departure is the mythological origin of the ruling line of the Tswana kingship of the Bakwena. The evangelizing missionary (London ­Missionary Society – LMS) Willoughby relates that:

Rethinking Tswana Kingships and Their Incorporation  221

At some unknown date in the past, a weakling who was in direct line of succession to the paramount handed over the chieftainship to an abler but younger brother. The present paramount chief of the Bakwena is head of the house that the younger brother founded, and the Ntloedibe clan is of the superior lineage of the weakling who parted from his birthright. (Willoughby 1928: 228–29) This meant that the elder brother and all his descendants lost the military, political and judicial prerogatives vested in the bogosi. Ever since, the descendants of the younger brother have, in successive order, been in control of the bogosi. However, the head of the genealogically senior descent line has been assigned ‘the title of Kgosi ea Leshótse (“Chief of the Cucurbitaceae [pumpkin]”) – a dignity which is fastidious of birth and disdainful of energy, ability, and wisdom’ (ibid.). I have discussed elsewhere a similar mythological origin of the ruling line in another major Tswana kingdom, the Bangwaketse (see Gulbrandsen 2014: 41–42). These are both high-profile cases that underscore the value of rank by agnatic seniority in Tswana societal order. The Bakwena and the Bangwaketse cases clearly convey the importance of the ruler’s personal capacities as indispensable premises for popular acknowledgement of the leader’s authority (taolò). Amongst the Tswana generally, this is crucially a matter of personal ability (nonohò), wisdom (kitso), knowledge (botlhale), force (patelèlò), and courage (nametshègò), as well as good character (botho) – all highly esteemed virtues, essential for exercising authority in the service of the common good, especially the all-embracing value of kagiso. Hence, a fine-grained set of categories and values is employed to assess a person in a position of authority, above all the kgosi. Where such personal qualities are limited or lacking amongst their leaders, anxiety about the damaging consequences of disorder prevails. Although the ‘general will’ is ideally vested in the population, people are, I re-emphasize, highly aware of their everyday dependency upon the capacities and qualities of the executive powers vested in the hierarchy of authority – reflecting a distinction that was, of course, crucial to Rousseau too.19 However, it is a key point that valuations of individual capacities and qualities perceived as essential to the exercise of authority in the service of social peace and order in no way represent a concession to ‘individualism’ in the sense commonly conceived in Western modernity. In Dumont’s rendering, modern Western societies ‘value, in the first place, the individual human being’ (1977: 4), meaning that the overarching value of kagiso should place the Tswana ‘in the holistic type’ of society where, as Dumont maintains, ‘the requirements of man as such are ignored or subordinated’ to the value of ‘the society as a whole’ (ibid.).20 However, such a conception needs to be tempered by the fact that, from the Tswana point of view, people can never be sure that persons in a position of authority always adhere to the ideals of conduct indicated above. It is intrinsic

222  Ørnulf Gulbrandsen

to the way in which Tswana perceive their lifeworld that there are reasons to fear that even a kgosi – always perceived as a human being with potential weaknesses as well as strengths – may be vested with destructive dispositions. This means that Dumont’s general conceptions – however helpful – should be used with high sensitivity to the particular ethnographic situation in which they are applied. When I discussed this matter with Tswana elders, they regularly invoked a proverb – skgosi thipa, e sego mo tootsi (a king is like a knife that will cut the sharpener), meaning, as I was told, that ‘we give force to the kgosi, you can say that we make him sharp. But we are never entirely sure how he will use it’. Another proverb, bogosi boa raga (kingship is intoxicating), was related to the fear of the kgosi abusing the powers entrusted to him, as epitomized by the ambiguities of the magical dimension of Tswana kingship. On the one hand, the kgosi is entrusted with control over the most powerful magic and the magicians (dingaka, ngaka) so as to obtain the most potent ‘medicine’ (ditlhare) to fortify the royal centre (bogosi) and his own person, family and property. Moreover, customarily the kgosi has distributed protective and productive ditlhare down through the hierarchy of dikgosana and heads of family, to reinforce the ditlhare they, at each level, have obtained from their own dingaka – always in the pursuit of defence against evil forces as well as in support of the fertility and health of people, domestic animals and crops. This is patently a strong symbolic expression of the kgosi’s blessing capacity as a motswadintle and the morafe as an encompassing field of commensality. On the other hand, the dingaka are not only providers of promotive and protective ditlhare; they are also perceived as a source of ‘black magic’, and hence the exercise of sorcery (boloi), a potentiality that is hard to detect because of the highly private, secretive dealings with the dingaka, even to obtain protective and productive ditlhare, because it otherwise might be vulnerable to occult attacks. So, since virtually everybody is believed to have contact with a ngaka, there is a pervasive fear of being interfered with by people who have the evil and destructive intension of committing boloi. The consequent ambiguity also surrounds the kgosi, as well as authority figures down the ladder, which is reflected in two distinctively different kinds of fear amongst subjects. On the one hand, there is a positive notion of fear – tshaba – connoting natural respect (tshisimogo) for authorities who rightfully discipline their subjects for the sake of social order and hence kagiso. On the other hand, there is the notion of ‘fear’ known as boifa, which is radically different: it is connected with the secretive exercise of power epitomized by the damaging practice of boloi. As in Tswana ontology, occult practices are always a secret matter, boifa is ever-present as a highly unpredictable potentiality that underscores the Tswana’s preoccupation with the personal characters of dikgosi and dikgosana.21 During the colonial era, when the dikgosi enjoyed the backing of the British, some of them, like the regent kgosi of the Bangwato, Tshekedi Khama, were

Rethinking Tswana Kingships and Their Incorporation  223

widely feared as dangerous, dictatorial rulers (see Wylie 1990; Gulbrandsen 2014: 87ff.). One significant condition for Tshekedi’s extensive exercise of repressive powers, which provided the basis for people’s fear of him, was certainly the popular imaginaries of the magicality of Tswana kingship. In some important respects, such abuse of power is perceived by Tswana as a matter of the personal greed for power and wealth that is, as suggested, associated with conflict-generating competition for power, position and wealth, and represents a pervasive and dangerous potentiality (Gulbrandsen 2003; Gulbrandsen 2014: Chapter 8). While perceived as dangerous and as typically related to negative personal dispositions, like greed and selfishness, such conflicts are also systemic. In particular, the system of ranking embodies significant incitements for competition and rivalry of this kind. The great popular concern with this source of tension and conflict – always perceived as threatening kagiso – derives from its high frequency in close agnatic relations. Rivalry over positions and property is perceived as pervasive amongst brothers (Comaroff and Comaroff 1992: 103), and fuels processes of conflict resolution and reconciliation in the kgotla. At the same time, the discursive field of the kgotla – the principal home of kagiso – contains incitements to rivalry. This competition surrounds, and is often spurred on by, the significance of cattle wealth, an economic source of power of immense importance, upon which Tswana kingships were founded and expanded from the late eighteenth century; in other words, in the same way as the dikgosi used the returns from their monopolies of the fur and ivory trade to raise huge herds of cattle, which they then exploited to establish vast networks of political retainers and incorporate vast numbers of immigrants into their realm. Thus, the dikgosi became agents of major transformation, and in a way that gave them considerable freedom of action (Gulbrandsen 2014: Chapter 1; see also Gulbrandsen 1993b, 2007). Cattle have a broad political significance, especially because, as the Tswana say, a man with a large herd of cattle is a man who can speak entirely freely in the kgotla – meaning that he will be listened to, not readily challenged, and, indeed, respected. He might even voice opinions against the head of the kgotla, including the kgosi. This is a way in which ambitious people’s efforts to advance their de facto rank challenges customary principles of ranking – such as that of genealogical seniority – which are much esteemed because they are considered to prevent disorder. Such ambitious people might, in certain respects, be admired. But their ways of exercising domination, especially in close agnatic relations, might make them suspected of relying on occult practices. This is especially so if their cattle herd has been growing suspiciously fast and their agricultural fields bring recurrently much higher yields than those of other people. In this sense, there have always been some structural conditions instigating behaviour condemned as governed by individual greed and selfishness. As market forces and related elements of Western modernity gained some momentum

224  Ørnulf Gulbrandsen

after the Second World War, and accelerated following the establishment of the modern state in Botswana, there developed an obsession with the accumulation of cattle herds that gave rise to a category of enterprising people whom Kuper (1970: 54ff.) identified as ‘the New Men’. These were people who progressively departed from Tswana ideals of holistic order, as they were taken by the spirit of egalitarian individualism, and they became the driving force in establishing a political party – the Botswana Democratic Party (BDP) – that has been in power ever since Botswana’s independence in 1966. Intriguingly, at the same time as the establishment of a postcolonial state represented a massive introduction of Western modernity, state formation in this country relied to some extent, as I have discussed extensively elsewhere (Gulbrandsen 2014: Chapters 5 and 6), upon the symbolic wealth and institutional structures of the Tswana kingdoms. The consequent contradiction between holistic and individualistic value orientations is the chief issue of Part Two, to which we will now turn.

Part Two: Virtues of Tswana Kingdoms and the Contradictions of Modernity in Postcolonial Botswana In Europe, kingship has, broadly speaking, either been abolished as a governmental form or absorbed into the modern state in the form of a constitutional monarchy, with sovereignty ideally vested in the population and executed by elected political bodies. The rational orientations of modernity have, with significant effect, worked persistently against imaginaries of the God-given and ‘mysterious’ character of royalty. In this respect, the modern state in Botswana is intriguing because this country forms a contrast both to modern Europe and to most other postcolonial countries in Africa. On the one hand, Botswana is a parliamentarian republic whose state constitution recognizes the Tswana kingdoms to the extent of making them integral, in significant ways, to the modern state. Yet the modern state leadership has never attempted to rationalize away what Fortes and Evans-Pritchard (1940a) would call ‘mystical values’, which I have discussed in Part One above. Rather, the state leadership has contributed significantly to the reproduction of the symbolic wealth of the Tswana bogosi, reflecting its significance to modern state formation in this country. On the other hand, Botswana differs substantially from the quite widespread imagery of indigenous symbolism, practices and institutions of power in Africa as highly counterproductive – even damaging – to the formation of a strong state with a sustainable government. Following Bayart’s (1993) seminal study, The State in Africa, which has the telling subtitle The Politics of the Belly, a vast array of scholarly works, much of an Afro-pessimistic kind, has been centrally concerned with the destructive impacts of Africa’s political heritage, typically conceived as constituted by ‘patrimonial’ or ‘neopatrimonial’ ideas, practices, and structures22 in generating notoriously fragile and weak states.

Rethinking Tswana Kingships and Their Incorporation  225

In what follows, I shall first illuminate why Tswana kingdoms have proved able to provide significant conditions enabling the legitimation of a strong modern state in Botswana with a sustainable parliamentarian democracy. Thereafter, I shall pursue the issue suggested towards the end of the preceding Part One; that is, contradictions between holistic ideals of the common good and liberal individualism, with reference to conflicting conceptions of egalitarianism entailed by modern state formation in this country.

Ambiguities of Tswana Kingship in the Formation of the Modern State in Botswana The efforts of the postcolonial state leadership to anchor the modern state – and itself – in the indigenous symbolic wealth of Tswana kingship is apparent in the officially stated national principles and is above all signified by their embracing of kagisano23 as the overarching value of the nation-state of Botswana. This ­symbolism conspicuously concludes the national anthem: Ka kutlwano le kagisano [Through our unity and harmony] E bopagantshe mmogo [We stand in peace as one] This expression of a strong communitarian value orientation, encompassing all citizens equally, was soon reflected in major governmental policies of welfare, out of which emerged a significant continuity with the paramount ideals of the common good at the core of Tswana kingship. The constitution of the modern state in Botswana was inspired by Enlightenment virtues of equality and individual liberty, and subscribes to the moral value of equal human worth: the legal value of equality before the law (the rule of law), the political value of equal voice in the legislature and the selection of political leaders (democracy), and the social value of equal access to the necessities for a decent life (universal welfare). In view of what I have already explained in this chapter, it should come as no surprise that leaders of the modern state had no difficulty in developing a political rhetoric to ensure its legitimacy by grounding these principles in central values generally held by the population of this country. The tremendous impact of this rhetoric was ensured, above all, when it was spearheaded by the unrivalled national leader, Seretse Khama (1921–80), who was both a very popular heir to the bogosi of the largest Tswana kingdom (Bangwato) and a man of the modern world, being well educated as a barrister at Oxford (see Parsons, Henderson and Tlou 1995). By virtue of his impressive oratory and charismatic performances, Khama communicated very successfully with all sections of the population, especially in the context of the kgotla. As early as the 1950s, he gained the reputation of being strongly dedicated to the common good and general welfare, as patently expressed in his ambition of ensuring everybody access to Western healthcare

226  Ørnulf Gulbrandsen

and education. Conspicuously, as cattle, though always highly unequally distributed, represented an essential asset in most families’ subsistence, he called for ‘a cattle-owning democracy’, asserting that ‘the democratic way to spread wealth to virtually every family in the country was through the intensification and spread of commercial cattle production for export’ (Parsons, Henderson and Tlou 1995: 204). While vast groups of ethnic minorities had been subject to severe discrimination from the Tswana, the postcolonial state government developed a national discourse of development that explicitly rejected such practices. Again, Khama was at the forefront in terms of asserting equality across spectrums of ethnicity and race. Khama – and the people with whom he formed the ruling party – initiated the establishment of a modern state in Botswana by addressing people around the country in their own kgotla, the symbolic core. With great frequency and intensity, political leaders – in perfect command of the discursive symbolism of the kgotla – have up to the present day continued this practice of travelling from village to village in this vast country. Political leaders developed a particular rhetoric by which they quite successfully blended promises of modernity with Tswana virtues of prosperity for everybody. Across Botswana, people assembled in the kgotla were told, again and again, that they had a government that placed high value on social equity and general welfare. The politicians endeavoured to reinforce their message in a context strongly associated with ancestral spirits (badimo) in the discursive field of the kgotla, the significance of which I have elaborated upon elsewhere (Gulbrandsen 2014: 240ff.). Soon, an era of development optimism started to take hold in the population at large, closely associated with Khama’s party, the BDP, which has maintained a majority of the votes ever since. This is also to say that electoral democracy has prevailed since independence in 1966 – a highly exceptional state of continuity in postcolonial Africa. Political leaders in Botswana speak proudly about their parliamentarian democracy as rooted in indigenous virtues of popular participatory politics, centred on the kgotla. The effects of blending of indigenous virtues and modernity were not limited to the field of political rhetoric, but led to the staffing of the central and local governments with a vast and growing technical-bureaucratic staff of ­‘experts’ and ‘volunteers’ from Western welfare states.24 The implementation of ‘­development’ was substantial and progressed impressively fast during the first two decades of independence. Many people around the country soon ­experienced – and greatly appreciated – the delivery of the promised public services like medical care, education and domestic water facilities, according to universal principles of equity. At the same time, the values and institutions of Tswana kingship were efficiently exploited so as to keep the population in the societal structure during a period of radical change that, in particular, involved immense socioeconomic

Rethinking Tswana Kingships and Their Incorporation  227

i­nequality (see below). This ambiguity is especially apparent in the administration of justice, which the state has, to a great extent, left this with the customary courts of the kgotla at all levels, as under colonial and precolonial circumstances. Thus, social tension and conflicts are, at least ideally, quickly identified and brought to popular tribunals close to people’s lifeworlds, with the presidents of the courts (dikgosi and dikgosana) now being state employees. In urban areas, a kgotla has been set up in various quarters. Moreover, the tremendous expansion of education and health services has entailed a massive governmentalization of the state, in Foucault’s (1978) sense, through which vast sections of the population have, more or less unwittingly, been captured into the modern-state formation of this country, impacting their ontological orientations and self-identity under the condition of progressive modernity (Gulbrandsen 2014: Chapter 7; see Gulbrandsen (2018: 287ff.) for my comparison with the Italian island of Sardinia).

Contradictions of the Common Good and Liberal Individualism What might have been a measure to promote social and economic equality under postcolonial circumstances – Khama’s cattle-owning democracy – was never realized. On the contrary, the penetration of market forces gave rise to entrepreneurial engagement, which found its most prominent expression in tremendously large-scale commercial cattle raising, facilitated by one of the major achievements of President Khama’s political leadership: the establishment of an abattoir with a capacity far beyond most slaughterhouses in the southern hemisphere, virtually the only facility in Africa satisfying the strict veterinary standards of the European Union. Obsessed by accumulating cattle, which had by now become an extremely profitable means of production, the economic-cum-political-­cum-­ bureaucratic elites in control of the state (Leith 2005: 55) ensured the development of a number of policies and programmes through which mainly those who had the initial assets required were able to take advantage of very substantial financial and technical state support. Western ‘development aid’ enabled the state-owned abattoir to augment the very good beef prices received from sales in Europe, which in turn accelerated wealthy cattle-owners’ possibilities for investment in livestock, water sources, etc., and thereby enabled a remarkable spiral of wealth accumulation. This trend gained tremendous momentum through very generous foreign ‘development’ aid and, not at least, as massive revenues from the immensely profitable diamond industry started to flow into the treasury around 1975 – a substantial share of which was thus appropriated by those, again, with sufficient assets to take the advantage of the state ‘development’ programs and projects (Gulbrandsen 2014: Chapter 3). The tension between individual accumulation and the common good is epitomized by a major enclosure movement, whereby enormous sections of ‘tribal’ common land have been appropriated by the above-mentioned elites of large-scale cattle-owners (Peters 1994; Gulbrandsen 1996b).

228  Ørnulf Gulbrandsen

That this kind of development had nothing to do with a cattle-owning democracy is amply demonstrated by its escalating inequality, bringing vast wealth to a minority of the population, while almost half the population has persistently remained below the official poverty threshold. Postcolonial Botswana has thus featured amongst the most inequitable countries on this planet (e.g., Good 1993; Good 2008: 67–68; Werbner 2010: 689ff.). At the same time, the situation of massive poverty in the midst of plenty has become progressively pronounced during the postcolonial era (Gulbrandsen 1996a; Gulbrandsen 2014: 282ff.). The ‘African miracle of Botswana’ being ‘one of the most socially unjust countries in the world’ (Samatar 1999: 189) has given rise to major critiques of the ruling elites as greedy and self-aggrandizing (Durham 1999: 200ff.), if not corrupt (Good 1994, 2008). It is apparent that the state leadership has been riding two horses. On the one hand, it has attempted to prevail as legitimate in relation to the rank-andfile sections of the population by dispensing public services and vast handouts, when needed, in the spirit of distributive egalitarianism and universalism. On the other hand, political leaders in Botswana – across the spectrum of political parties – have always broadly agreed on policies premised upon a quite different virtue: egalitarian individualism. In other words, we have here the kind of dualism constitutive to contradictions we know well from modern, capitalist societies elsewhere. In the case of Botswana, these contradictions have gained a special form because of the extraordinary state-centred character of the country’s political economy. One intriguing aspect of this is that Botswana has achieved an international reputation as a country virtually untainted by corruption. This is largely because people of power and wealth – across party political divides – have been tapping the state treasury legally by means of policies, programmes and projects decided upon by the parliament and other democratically elected bodies over which they have had control. As I have discussed elsewhere (Gulbrandsen 2014: Chapter 3), already at the time of independence there was in progress a ‘grand coalition’ of elites – across rural-urban, traditional-modern and even party-political divides. This helps to explain why these policies and programmes have been politically uncontroversial, and actually of great significance to the strength of the state in Botswana and, paradoxically perhaps, to the sustainability of its parliamentarian democracy. Patently, this integration of elites at the apex of the society has made Botswana a deviant case with regard to the image of postcolonial African states as weak and unstable due to factional battles at their political centre. Although one could well speak of ‘politics of the belly’ regarding Botswana, it is obviously of a kind quite different from that conceived by Bayart (1993), which is related to extensive ‘patrimonial’ or ‘neopatrimonial’ political ideas, practices and structures. In other words, his analysis describes vertical political formations of factions external to the order of the state, often destroying vast

Rethinking Tswana Kingships and Their Incorporation  229

resources in violent conflict with each other, and thereby militating intensely against the development of a strong, sustainable modern state. I have addressed these issues more extensively elsewhere (Gulbrandsen 2014: 8–13), explaining why similar state-weakening structures have not evolved in Botswana, despite the importance of Tswana kingship to the formation of the modern state. Hence, in relation to all the tendencies around the continent of Afro-pessimism, Botswana is a case in point to demonstrate that indigenous African political systems might well embody considerable state-building and consolidating potentialities, notwithstanding the highly unequal benefits to their postcolonial citizens in a country like Botswana. The economic-cum-political-cum-bureaucratic cattle-obsessed elites did not only take tremendous advantage of being in control of the postcolonial state; they bluntly justified what, in due course, amounted to a strong dimension of liberal individualism, fashioning themselves as successful entrepreneurs spearheading the development of the nation, to the benefit of the whole country. Soon after its establishment, the postcolonial state accelerated what Macpherson has termed ‘possessive individualism’ – a ‘conception of the individual as essentially the proprietor of his own person and capacities, owing nothing to society for them’ (2011: 3). In fact, as the great beneficiaries of the state-sponsored ‘development’ saw it, ‘society’ had, indeed, to be grateful for their contribution to the national wealth and hence everybody’s welfare. In this vein, entrepreneur-politicians readily invoked the Western economists’ notion of ‘trickle-down effect’ and presented it as a matter of truth in their legitimizing efforts (Gulbrandsen 1996a: 361ff.; Gulbrandsen 2014: Chapter 3). The political establishment thus attempted to gain legitimacy by construing a new kind of holism – the kind that people of the capitalist class everywhere propagate: the welfare of the society at large depends upon the individuals in ownership of productive property. However, in the context of modern Botswana, this notion cannot survive scrutiny. First, the tremendous expansion in livestock production generated an insignificant number of jobs because of its low labour requirements, given the open-range husbandry of ‘tribal’ commonages, where only a few people suffice to herd hundreds of cattle. Second, as argued by political-economist Hillbom, Botswana’s postcolonial economy has remained ‘a case of premodern growth’ because of ‘continued natural resource dependency, failure to build a strong private sector, lack of thorough industrialization, and ongoing reliance on a close connection between the government bureaucracy and the cattle-keeping elite’ (2011: 67). Consequently, she maintains, Botswana cannot be named a ‘developmental state’. In other words, we have at hand a quite distinctive instance of liberal individualism that has taken hold in, and is sustaining, a largely premodern mode of production that, seen from a long-term historical perspective, evolved as an indispensable material – and symbolic – resource base for Tswana kingship (Gulbrandsen 2014: Chapter 1). Under postcolonial circumstances, it has been

230  Ørnulf Gulbrandsen

transformed by Botswana’s highly state-centred political economy in such a way that major contradictions have evolved in the relationship between the privileged elites and the rank-and-file – to a great extent impoverished – sections of the population. With the large majority of political leaders, across the political spectrum, subscribing to major state policies, programmes and projects from which they have benefited tremendously, no leadership has emerged to exploit – for example, in a populist way – the massive dissatisfaction amongst all the rank-and-file sections of the population of the escalating discrepancies in income and wealth during postcolonial times. Those rare instances of people speaking in public about a need for radical change, like a major redistribution of cattle wealth, have been efficiently silenced. Critical voices have hence been relegated to subaltern discourses. However, cultural continuities across the rural–urban divide have remained strong, which means that the subaltern discourses taking hold in urban areas have been significantly inspired by Tswana values of the common good and transparency of political leaders’ exercise of power (Gulbrandsen 2003, 2014: Chapter 8).25 The tensions that have thus built up have been kept under control by means of rather authoritarian, repressive measures – a trend that accelerated under rule of the son of the founding state President Seretse Khama, Lt. General Ian Khama (2008–18). Thus, while his father, Seretse Khama, was celebrated – nationally as well as internationally – as a highly committed democrat, Ian Khama is quite widely reputed to be distinctively autocratic, violating democratic principles. He relied extensively on the intelligence service and ‘an escalation in the militarization and personalization of power’ (Good 2010: 315). Ian Khama’s image is thus, at the present time, not unlike that of his paternal uncle – the previously mentioned regent kgosi Tshekedi Khama – who was perceived by many subjects as causing the kind of destructive anxiety and fear I identified above as boifa, associated with occult practices and other forms of power abuse. Ian Khama’s image as a sometimes uncompromising and brutal hardliner was strongly expressed during a major and longlasting strike in 2011 that spread from towns to villages and culminated in violent confrontations with the police (Gulbrandsen 2014: 306ff.; Makgala and Malila 2014; Werbner 2014: 206ff.; Good 2016).26 The exceptionally – in the Botswanan context – violent character of people’s reactions in this case demonstrated the extent to which popular bitterness and frustration over elites’ appropriation of a lion’s share of the nation’s wealth is being silenced by repressive measures. An earlier (1994) series of such violent and dramatic events was inflamed by the murder of a girl, who was found with her private parts taken away, perceived as a matter of people of power and wealth using them as ‘medicine’ for self-empowerment. This incident was widely perceived as a ‘ritual murder’, committed by people with a greed for ever more power and wealth. Riots ensued and stores were set on fire in the Tswana royal town of Mochudi, where the murder had occurred, and in the capital of Ga-

Rethinking Tswana Kingships and Their Incorporation  231

borone, where many young people had been mobilized in violent confrontations with the police. This case clearly suggests that the kind of individualism brought about by Western modernity into the Tswana context has, to many people, proved highly destructive to kagiso and hence damaging to the welfare of the entire population – as epitomized by a statement I registered during this series of events: ‘now they are even eating our children’ (see Gulbrandsen 2003; Gulbrandsen 2014: 286ff.). Popular reactions to the political leadership have amounted to a fundamental egalitarian critique rooted in the Tswana ontology of power, as discussed in this chapter. However, the exceptional character of these reactions reflects the fact that the potency of ontology in this respect was not, to any significant extent, ‘realized through the ideological engagement of the logic within social and political realities’ (Kapferer 2012: 80), like in Anzac ritual practices of Australia (ibid.: 149ff.). In the Tswana context, popular egalitarian critique, premised upon the perception of highly damaging abuse of power amongst people who have successfully taken tremendous benefits from the practices of liberal individualism, has, I reiterate, been muted by repressive forces and hence has been restricted to subaltern discursive practices with, so far, no enduring mobilizing forces. Hence, Good (1996) speaks of postcolonial Botswana as a paradoxical case of ‘authoritarian liberalism’. Intriguingly, high-ranking governmental officials I knew at the time (in the mid-1990s) were engaged in developing a central-government document known as Vision 2016, which expressed considerable concern about the newly forged characters of people emerging in urban areas, as they were deemed lacking in the kind of politeness, if not submissiveness, they were used to from the lived-in context of Tswana villages. This has remained the idealized sociocultural sphere in which people are inculcated, from early childhood, with strong dispositions towards proper ways of relating to one another: above all, respect for everyone senior to oneself, a moral value still held in high regard within the structures of Tswana kingship. After extensive consultations with people around the country, these officials were highly concerned with the important Tswana notion of good character (botho – see above), which was esteemed as a key value in the Vision 2016 document and in a series of other governmental papers (Livingston 2008: 295; Gulbrandsen 2014: 179f.). Moreover, the value of botho was elevated by the government to the fifth national principle, to be drawn upon, ideologically, in the rhetoric of political leaders. It is true that botho is essentially an existentially critical virtue of considerable significance in people’s everyday lives because of the great value placed upon having an unchallenging and well-mannered character, and behaving kindly and respectfully to other people. Hence, botho is, in Tswana thought, closely associated with the overarching value of kagiso. However, like kagiso, botho was ideologized and captured into a modernist discourse, where it has been – due to

232  Ørnulf Gulbrandsen

its polysemiotic potential and disconnection from its indigenous sociocultural context – readily construed as an individualistic virtue. Thus, according to Vision 2016, a person with botho is a ‘rounded character, who is well-mannered, courteous and disciplined’. Like kagiso, botho has been disconnected from the Tswana customary sociocultural context that is oriented towards the holistic values of the common good. It is instead propagated as an individualized virtue instrumental to sustaining peace and order in a modernized society characterized by huge socioeconomic inequalities and massive and rapid urbanization.

Conclusion The main points of this chapter in relation to the question I posed initially – the comparative distinctiveness of African kingship – are centred on the observation that, in the case of the Tswana, the king is conceived as ‘king by the grace of the people’ rather than as being of divine heritage. According to this ontological orientation, sovereignty is vested in the population, underscoring an egalitarian dimension that marks a strong contrast with Eurasian kingship. Yet, egalitarianism in this context does not combine with individualism as in the Western tradition of Enlightenment, but is, on the contrary, encompassed by the overarching holistic value of social order through peace and harmony, epitomized by the ubiquitously existential significance of the ideal state of kagiso. This observation links closely with Fortes and Evans-Pritchard’s notion of ‘mystical values’ standing for ‘the greatest common interest’; that is, a notion that I have attempted to elaborate upon in this chapter so as to discuss African kingship, such as that of the Tswana, in a comparative context transcending Africa. To sustain social order, and hence kagiso, there is a firmly institutionalized hierarchy of authorities, culminating in the custodian of bogosi, the kgosi and, by extension, the principal source of morality and law, the badimo ya kgosi (the royal ancestry). However, whereas this hierarchical structure is, in Tswana ontology, indispensable to kagiso, it is at the same time a major source of conflict. From the perspective of Tswana ontology, there is a fundamental contradiction embedded in the hierarchical structure, as, on the one hand, it vests people in positions of authority with power to sustain peace and order, while, on the other hand, the practices of ranking, which are intrinsic to the hierarchical structure, constitute a systemic impetus for rivalry and other forms of challenging, conflict-generating behaviour. The tremendous concern with this potentiality is epitomized by the pervasive fear of occult practices and the high value placed on transparency, ideally in all walks of life and particularly in political decision-making. In other words, the Tswana have a form of kingship founded upon democratic, egalitarian virtues of a kind that contrast sharply with those of ­premodern Eurasia, notwithstanding the inherent potential for autocratic rule – which is, after all, conceived as an abuse of power and thus contrary to the ideals of good government.

Rethinking Tswana Kingships and Their Incorporation  233

I have pointed out that the omnipresent concern with kagiso means that the Eurocentric conceptual divide between politics and religion obscures rather than helps our comprehension of Tswana cultural constructions of kingship. Furthermore, I have argued that the strong holistic value orientation amongst the Tswana is antagonistic towards individualistic endeavours, which are readily seen as motivated by personal greed and selfishness because of their perceived ­conflict-generating and destructive character. This is a matter of contradiction amplified and intensified by the postcolonial circumstances of the modern state in Botswana. On the one hand, the postcolonial state leadership was at the outset quick to appropriate the symbolic wealth of Tswana kingship centred on the virtues of the common good. On the other hand, their embracing of Western modernity – bureaucracy, representative political bodies, and commercial beef and diamond production and marketing – has had a massively transformative impact, progressively and seriously challenging Tswana values of the common good and transparency, often to the extent of undermining them. Above all, conflict-generating contradictions have been amplified under the impact of strong elite reorientations towards liberal individualism. One important entailment within the diamond-driven, state-centred political economy is the rapid development of substantial class divisions, which have attained their particular form because of the influence of the symbolic wealth and institutions of Tswana kingship. Ironically, however, the distinctive form of egalitarianism vested in the holistic order of this form of kingship – which is, as we have seen, radically different from egalitarian individualism of Enlightenment and Western modernity – has ultimately lost much of its significance during Botswana’s postcolonial transformations.

Acknowledgements I am indeed grateful for all the helpful comments received during my work on this chapter from Bjørn Enge Bertelsen, Jan-Petter Blom, John Comaroff, Kenneth Good, Elsa Kaarbø, Adam Kuper, Bruce Kapferer, Jacqueline Solway and Harald Tambs-Lyche. During my sabbatical year at the London School of Economics (LSE) in the 1990s, I visited Professor Isaac Schapera, the doyen of Tswana studies, for conversations on Tswana ethnography nearly every fortnight, from which I learned much that I relied on in writing this chapter. In appreciation, I dedicate this chapter to Schap’s memory. Moreover, I want to extend my gratitude to University of Bergen for financial support. Ørnulf Gulbrandsen is Professor Emeritus of Social Anthropology and former Head of the Department of Social Anthropology at the University of Bergen, Norway. He has conducted field research in Botswana since the mid-1970s, in

234  Ørnulf Gulbrandsen

Norway and more recently on the Italian island of Sardinia. He has published on topics including symbolism of political action/structure related to modern state formation, kingship in changing, globalized historical contexts, sorcery as a dimension of subaltern discourse, the political economy of cultural ecology, marriage and sexual relations and on violence related to banditism and vendettas in late modern European state contexts. His most recent book is The State and the Social: State Formation in Botswana and Its Precolonial and Colonial Genealogies (Berghahn Books, 2014).

Notes   1. There are a number of Tswana kingdoms, partly located in eastern Botswana, partly in the North-West Province of South Africa. Like Schapera, my work is on the people of the Tswana kingdoms on the Botswana side of the border, who were formerly included in the Bechuanaland Protectorate, where they enjoyed a much lighter colonial intervention than the Tswana within the successive harsh racist regimes of South Africa, studied by Jean and John Comaroff and others.   2. I have recurrently conducted ethnographic field work in eastern Botswana since 1975.   3. See Gulbrandsen (2001) for extensive discussion of the inadequacy of this dichotomy in the context of Tswana kingship.   4. Such a shift was usually more of an adjustment of customary practices than a discontinuity, because the dikgosi have at all times, in the popular imagination, conducted certain rainmaking rites secretly (Schapera 1971a: 31), often assisted by a hired rainmaker (moroka). Similarly, the dikgosi’s ritual practices and communication with the badimo ya dikgosi have been perceived as partially a secret matter.   5. Thus, Klaits has observed how love can be ‘a nurturing sentiment that builds up other people’s well-being and counteracts the destructive power of resentment and jealousy (lefufa)’ (2011: 214).   6. Social conflicts were thus often related to me as ‘misunderstandings’ (go tlhoka kutlwano, lit. to lack understanding) between the parties, and then not actually in the sense of a conflict of interests and a dispute in the Western sense. So, much of the kgotla proceedings consist of resolving the conundrum of interpersonal intricacies in order to arrive at kutlwano rather than establishing who is right in relation to the law.   7. Brown observed long ago that: ‘The idea of Atonement through sacrifices is well understood by the Bechuana … Atonement strictly means the act of bringing together two who have been alienated; the restoration to a peaceful state of those who have been estranged. It means an act of Reconciliation’ (1926: 144). Cf. Willoughby 1928: 196ff.   8. There is an important reservation to this, as there are always certain subject peoples within the domain of a Tswana kingdom who do not have full citizenship rights, as I shall explain below.   9. The broader significance of this notion is indicated by a quotation provided by Comaroff and Comaroff of an elder’s speech at the occasion of the enthronement of a new kgosi amongst the Tswana people of Barolong, stating that: ‘A chief [kgosi] can only be judged by what he does … If you treat [people] with respect, they will treat you with respect. If you shun them, they will shun you. And if you frighten them, they will run away … We will be watching to see whether you are going to make improvements. Chiefship is not an easy job. A chief never sleeps. A chief does not discriminate. Batswana say that a chief

Rethinking Tswana Kingships and Their Incorporation  235 is chief because of the nation. If we cannot see you in the court [kgotla] we shall draw away from you. And if we do will you still call yourself chief?’ (Comaroff and Comaroff 1997: 131, emphasis added; see also Schapera 1984: 83ff.; Gulbrandsen 1995: 421, 430ff.) 10. However, I have reservations against their bold generalization of ‘stranger-kingdoms’ as ‘the dominant form of premodern state the world around’ (Graeber and Sahlins 2017: 5) 2017: 5), which is in fundamental disagreement with the western Tswana kingdoms whose precolonial development I have examined elsewhere (e.g. Gulbrandsen 1993b, 2014: Chapter 1). 11. This maxim a strikingly similar to that of the Lozi – ‘Even the king is slave of the law (mulao)’ (Gluckman 1955: 164) – whichs suggest an egalitarian dimension amongst the Lozi too (e.g. ibid.: 214). 12. As for the broader context of African kingship, let me refer here to Kapferer’s (2003) discussion of ‘Gluckman’s egalitarian ethos’ and to his observation that ‘Gluckman’s image of the African chief or king was very much at the forefront of his imagination. This was of a person who is a primus inter pares, that is, a person who embodies to the highest degree the qualities (generosity, selflessness) that are the core organizational properties of the group and who is otherwise no different from those whom he commands. Put another way, material difference, the trappings of status, and other forms of social distinction should not be an impediment to social interactions with peers’ (Kapferer 2005: 116 n.10; cf. note 11 above). 13. Quoted in Swenson 2000: 163. 14. Rousseau 1968: Book II, Chapter 4. 15. Ibid.: Book I, Chapter 6. 16. A significant exception, as extensively discussed by Richard Werbner (2004; cf. Werbner 2002b), is manifested by the Kalanga of northeast Botswana, who are ‘[o]f all minorities … the largest, most prominent, politically and economically influential, and culturally perhaps most assertive’ (Werbner 2004: 71). 17. From Act 4, Scene 3. 18. In the case of a kgosi considered by people to be very inactive or too weak to be a blessing to his subjects, it is suggested proverbially that Kgosiya tlhotsa malata ra gagaba (‘when the Kgosi limps, the servants are limping too’). 19. See e.g. Rousseau 1968: Book III. 20. This point corresponds to the Tswana notion of motho ke motho ka batho (‘a person is a person through other people’), which is reflected, for example, in the sincere fear of being left alone that I frequently encountered amongst people moving to urban areas, leaving me with a strong impression of a deeply felt existential need for belonging. I have almost never encountered idealization of individual autonomy in the Western sense. 21. Sorcerous practices by kgosi might be perceived by subjects as legitimate too when seen as for the benefit of the common good, especially when a kgosi is under such severe attack by rivals or external enemies that the stability of social order is brought into danger. In the case of the Zulu, the perception of the king as a potential sorcerer is advanced a step further as ‘the title inkhosi enkhulu (great king) and umtkhakathi omkhulu (great sorcerer) are synonymous’ (see Krige (1950: 243) for elaborations). On the notion of king as sorcerer, see, for example, Kapferer (2002a: 15, 2002b: 110ff.) and Gluckman (1971), and see Bertelsen (2016: 186) for a general discussion of the relationship between sorcery and sovereignty and its significance for the extremely violent turmoil of postcolonial Mozambique. 22. For my recent discussion of this issue, see Gulbrandsen (2014: 5–13). 23. While kagiso means peace and harmony, kagisano is the corresponding verb – staying together in peace and harmony.

236  Ørnulf Gulbrandsen 24. The level of education in Botswana was very low and the government depended heavily upon Americans and Europeans during the first two decades of independence. 25. This also means that despite Tswana virtues of the common good vested in Tswana kingship, popular participation in indigenous institutions – centred on the kgotla – have never resulted in any form of political protest against the ever more progressive inequality that has characterized the postcolonial era (Gulbrandsen 1996a: 355ff.; cf. Gulbrandsen 2014: Chapter 6). While three of the dikgosi have been greatly at odds with the state leadership, they have never attempted to mobilize their respective people in a protest of this kind. As I have discussed at length elsewhere, the most confrontational amongst them – Kgosi Kgafela II of the Bakgatla – instead tried, although unsuccessfully, to enhance his autonomy in relation to the state by mobilizing the Bakgatla by means of a spectacular neotraditionalist endeavour (Gulbrandsen 2014: 148–61). 26. After Ian Khama resigned in 2018, he became seriously at odds with his successor, the present President of Botswana, Mokgweetsi Masisi – so much so that he departed from the party his father had founded, the always-ruling BDP, and joined in the formation of a new party, the Botswana Patriotic Front (BPF). However, at the general elections in October 2019, his party was heavily defeated by Masisi’s BDP, which won a landslide victory of thirty-eight seats (an absolute majority) in parliament, while the BPF won only three seats. Although Ian Khama won the election in his own constituency, this was obviously a most humiliating outcome, given his status as the incumbent of the Bangwato bogosi. His political performance clearly indicated that he had been unable to draw upon the symbolic wealth of the Khama dynasty in order to bring his party to victory at the national level, as his father famously had done following Botswana’s independence.

References Alverson, Hoyt. 1978. Mind in the Heart of Darkness: Value and Self-Identity among the Tswana of Southern Africa. New Haven: Yale University Press. Bayart, Jean-Francios. 1993. The State in Africa: The Politics of Belly. London: Longman. Bertelsen, Bjørn Enge. 2016. Violent Becomings: State Formation, Sociality, and Power in Mozambique. Oxford: Berghahn Books. Brown, J. Tom. 1926. Among Bantu Nomads. London: Seeley, Service & Co. Ltd. Buur, Lars, and Helen M. Kyed. 2007. ‘Traditional Authority in Mozambique: The Legible Space between State and Community’, in L. Buur and H.M. Kyed (eds), State Recognition and Democratization in Sub-Saharan Africa: A New Dawn for Traditional Authorities? New York: Palgrave Macmillan, pp. 105–30. Comaroff, Jean, and John Comaroff. 1992. Ethnography and the Historical Imagination. ­Oxford: Westview Press. ———. 1997. ‘Postcolonial Politics and Discourses of Democracy in Southern Africa: An Anthropological Reflection on African Modernities’, Journal of Anthropological Research 53(2): 123–46. ———. 2001. ‘On Personhood: An Anthropological Perspective from Africa’, Social Identities 7(1): 267–83. Comaroff, John L., and Jean Comaroff. 2018. ‘Introduction’, in J.L. Comaroff and J. Comaroff (eds), The Politics of Custom: Chiefship, Capital, and the State in Contemporary Africa. Chicago: The University of Chicago Press. Dumont, Louis. 1970. Religion/Politics and History in India. Paris: Mouton. ———. 1972. Homo Hierarchicus: The Caste System and Its Implications. London: Paladin.

Rethinking Tswana Kingships and Their Incorporation  237 ———. 1977. From Marx to Mandeville. Chicago: University of Chicago Press. Durham, Deborah. 1999. ‘Civil Lives: Leadership and Accomplishment in Botswana’, in John L. Comaroff and Jean Comaroff (eds), Civil Society and the Political Imagination in Africa. Chicago: University of Chicago Press, pp. 192–218. ———. 2011. ‘Disgust and the Anthropological Imagination’, Ethnos 76(2): 131–56. Englebert, Pierre. 2002. State Legitimacy and Development in Africa. Boulder, CO: Lynne Rienner. Fortes, Meyer, and E.E. Evans-Pritchard. 1940a. ‘Introduction’, in M. Fortes and E.E. ­Evans-Pritchard (eds), African Political Systems. London: Oxford University Press, pp. 1–24. ———. (eds) 1940b. African Political Systems. London: Oxford University Press. Foucault, Michel. 1978. ‘Governmentality’, Ideology and Consciousness 6(6): 5–21. Geertz, Clifford. 1980. Negara: The Theatre State in Nineteenth-Century Bali. Princeton: Princeton University Press. Gluckman, Max. 1970. Custom and Conflict in Africa. Oxford: Blackwell. ———. 1971. Politics, Law and Ritual in Tribal Society. Oxford: Blackwell. Good, Kenneth. 1993. ‘At the Ends of the Ladder: Radical Inequalities in Botswana’, Journal of Modern African Studies 31(2): 203–30. ———. 1994. ‘Corruption and Mismanagement in Botswana’, Journal of Modern African Studies, 32(3): 499–521. ———. 1996. ‘Authoritarian Liberalism: A Defining Characteristic of Botswana’, Journal of Contemporary African Studies 14(1): 29–51. ———. 2008. Diamonds, Dispossession and Democracy in Botswana. Oxford: James Currey. ———. 2010. ‘The Presidency of General Ian Khama: The Militarization of the Botswana “Miracle”’, African Affairs 109(435): 315–24. ———. 2016. ‘Democracy and Development in Botswana. Special Issue on Botswana’s 50th Anniversary’, Journal of Contemporary African Studies 35(1): 113–27. Graeber, David, and Marshall Sahlins. 2017. On Kings. Chicago: HAU Books. Gulbrandsen, Ørnulf. 1991. ‘On the Problem of Egalitarianism: The Kalahari San in Transition’, in Reidar Grønhaug, Gunnar Haaland and Georg Henriksen (eds), The Ecology of Choice and Symbol. Essays in Honour of Fredrik Barth. Bergen: Alma Mater, pp. 81–110. ———. 1993a. ‘Missionaries and Northern Tswana Rulers: Who Used Whom?’, Journal of Religion in Africa 23(1): 44–83. ———. 1993b. ‘The Rise of the North-Western Tswana Kingdoms: On the Socio-cultural Dynamics of Interaction between Internal Relations and External Forces’, Africa 63(4): 550–82. ———. 1995. ‘“The King Is King by the Grace of the People”: Control and Exercise of Power in Subject–Ruler Relations’, Comparative Studies in Society and History, 37(3): 415–455. ———. 1996a. Poverty in the Midst of Plenty: Socio-economic Marginalization, Ecological Deterioration and Political Stability in a Tswana Society. Bergen: Norse Publications. ———. 1996b. ‘Living Their Lives in Courts: The Counter-hegemonic Force of the Tswana Kgotla in the Bechuanaland Protectorate’, in Olivia Harris (ed.), Inside and Outside the Law. London: Routledge, pp. 125–56. ———. 2001. ‘Christianity in an African Context: On the Inadequacy of the Notion of a Sacred/Secular Divide’, Journal of the Royal Institute for Inter-Faith Studies 1: 31–59. ———. 2003. ‘The Discourse of “Ritual Murder”: Popular Reactions to Political Leaders in Botswana’, Social Analysis: Journal of Cultural and Social Practice 46(3): 215–33. ———. 2007. ‘Town-State Formations on the Edge of the Kalahari: Socio-cultural Dynamics of Centralization in Northern Tswana Kingdoms’, Social Analysis: Journal of Cultural and Social Practice 51(3): 55–77.

238  Ørnulf Gulbrandsen ———. 2014. The State and the Social: State Formation in Botswana and Its Precolonial and Colonial Genealogies. Oxford: Berghahn Books. ———. 2018. ‘Inside and outside the State in Italy and Botswana: Historical and Comparative Reflections on State Apparatuses of Capture and Rhizomic Forces’, in Bruce Kapferer (ed.), State, Resistance, Transformation: Anthropological Perspectives on the Dynamics of Power in Contemporary Global Realities. Canon Pyon: Sean Kingston Publishing, pp. 265–94. Hegel, Georg W.F. 1991. Elements of the Philosophy of Right, edited by Allen W. Wood. Cambridge: Cambridge University Press. Hillbom, Ellen. 2011. ‘Botswana: A Development-Oriented Gate-Keeping State’, African Affairs, 111(442): 67–89. Kapferer, Bruce. 2002a. ’Introduction: Outside All Reason: Magic, Sorcery and Epistemology in Anthropology’, Social Analysis: Journal of Cultural and Social Practice 46(3): 1–30. ———. 2002b. ‘Sorcery, Modernity and the Constitutive Imaginary: Hybridising Continuities’, Social Analysis: Journal of Cultural and Social Practice 46(3): 105–28. ———. 2003. ‘Erindring om forgangne tider: Gluckman og Manchester-skolen i social-­ antropologiens historie’, Jordens Folk 38(3): 11–14. ———. 2005. ‘Situations, Crises, and the Anthropology of the Concrete: The Contribution of Max Gluckman’, Social Analysis: Journal of Cultural and Social Practice, 49(3): 85–122. ———. 2010. ‘Louis Dumont and a Holistic Anthropology’, in Ton Otto and Nils Bubandt (eds), Experiments in Holism. Theory and Practice in Contemporary Anthropology. Chichester: Blackwell, pp. 187–208. ———. 2012. Legends of People: Myth of State. Violence, Intolerance, and Political Culture in Sri Lanka and Australia. New York and Oxford: Berghahn Books. Klaits, Frederick. 2010. Death in a Church of Life: Moral Passion during Botswana’s Time of Aids. Berkeley: University of California Press. ———. 2011. ‘Asking as Giving: Apostolic Prayers and the Aesthetics of Well-Being in Botswana’, Journal of Religion in Africa, 41: 206–26. Krige, Eileen Jensen. 1950. The Social System of the Zulus. Pietermaritzburg: Shuter & Shorter. Kuper, Adam. 1970. Kalahari Village Politics: An African Democracy. Cambridge: Cambridge University Press. Lambek Michael and Jacqueline S. Solway. 2001. ‘Just Anger: Scenarios of Indignation in Botswana and Madagascar’. Ethnos 66: 49–72. Landau, Paul S. 1995. The Realm of the World: Language, Gender, and Christianity in a Southern African Kingdom. Oxford: James Currey. Leith, J. Clark. 2005. Why Botswana Prospered. Montreal: McGill-Queen’s University Press. Livingston, Julie. 2008. ‘Disgust, Bodily Aesthetics and the Ethic of Being Human in Botswana’, Africa: Journal of the International African Institute, 78(2): 288–307. Macpherson, C.B. 2011. The Political Theory of Possessive Individualism: Hobbes to Locke. Oxford: Oxford University Press. Makgala, C.J., and Ikanyeng Stonto Malila. 2014. The 2011 BOFEPESU Strike. Cape Town: Centre for Advanced Study of African Society. Mauss, Marcel. 1990. The Gift: The Forms and Reason for Exchange in Archaic Societies. London: Routledge. Myrdal, Gunnar. 1944. An American Dilemma: The Negro Problem and Modern Democracy. New York: Harper & Brothers. Nader, Laura. 1990. Harmony Ideology: Justice and Control in a Zapotec Mountain Village. Stanford: Stanford University Press. Ortner, Sherry B. 1973. ‘On Key Symbols’, American Anthropologist 75(5): 1338–46.

Rethinking Tswana Kingships and Their Incorporation  239 Parry, Jonathan. 1974. ‘Egalitarian Values in a Hierarchical Society’, South Asian Review, 7(2): 95–121. Parsons, Neil, Willie Henderson and Thomas Tlou. 1995. Seretse Khama, 1921–1980. Gaborone: Macmillan. Peters, Pauline E. 1994. Dividing the Commons: Politics, Policy, and Culture in Botswana. Charlottesville: University of Virginia Press. Plato. 1956. The Republic, translated into English by B. Jowett. New York: The Modern Library. Roberts, Simon. 1985. ‘The Tswana Polity and “Tswana Law and Custom” Reconsidered’, Journal of Southern African Studies 12(1): 75–87. Robinson Eric W. 2004. Ancient Greek Democracy: Readings and Sources. Oxford: Blackwell. Rousseau, Jean-Jacques. 1968. The Social Contract. London: Penguin. Samatar, Abdi I. 1999. An African Miracle: State and Class Leadership and Colonial Legacy in Botswana Development. Portsmouth, NH: Heinemann. Sapignoli, Maria. 2015. ‘Dispossession in the Age of Humanity: Human Rights, Citizenship, and Indigeneity in the Central Kalahari’, Anthropological Forum 25(3): 285–305. ———. 2018. Hunting Justice: Displacement, Law, and Activism in the Kalahari. Cambridge: Cambridge University Press. Schapera, Isaac. 1940. ‘The Political Organization of the Ngwato in Bechuanaland Protectorate’, in Meyer Fortes and E.E. Evans-Pritchard (eds), African Political Systems, London: Oxford University Press, pp. 56–82. ———. 1963. Government and Politics in Tribal Societies. London: C.W. Watts & Co. ———. 1965. Praise Poems of Tswana Chiefs. Oxford: Oxford University Press. ———. 1970 [1934]. ‘Oral Sorcery among the Natives of Bechuanaland’, in E.E. ­Evans-Pritchard et al. (eds), Essays Presented to C.G. Seligman, Westport: Negro University Press, pp. 293–305. ———. 1971a. Rainmaking Rites of Tswana Tribes. Leiden: Afrika-Studiecentrum. ———. 1971b [1940]. Married Life in an African Tribe. Harmondsworth: Penguin. ———. 1984 [1938]. A Handbook of Tswana Law and Custom. London: Oxford University Press for the International African Institute. Simmel, Georg. 1971. On Individuality and Social Forms. Chicago: University of Chicago Press. Sklar, Richard L. 2005. ‘The Premise of Mixed Government in African Political Studies’, in Ofuemi Vaughan (ed.), Tradition and Politics: Indigenous Political Structures in Africa, Asmara: African World Press, pp. 13–31. Solway, Jacqueline S. 1994. ‘From Shame to Pride: Politicized Ethnicity in the Kalahari, Botswana’, Canadian Journal of African Studies, 28(2): 254–74. ———. 2004. ‘Reaching the Limits of Universal Citizenship: “Minority” Struggles in Botswana’, in B. Berman, D. Eyoh and W. Kymlicka (eds), Ethnicity and Democracy in Africa. Oxford: James Currey, pp. 129–48. ———. 2009. ‘Human Rights and NGO “Wrongs”: Conflict Diamonds, Culture Wars, and the “Bushman Question”’, Africa 79(3): 329–43. Swenson, James. 2000. On Jean-Jacques Rousseau. Stanford: Stanford University Press. Werbner, Pnina. 2010. ‘Appropriating Social Citizenship: Woman’s Labour, Poverty, and Entrepreneurship in the Manual Workers Union of Botswana’, Journal of Southern African Studies, 36: 693–710. ———. 2014. The Making of an African Working Class: Politics, Law, and Cultural Protest in the Manual Worker’s Union of Botswana. London: Pluto Press. Werbner, Richard. 2002a. ‘Introduction: Challenging Minorities, Difference and Tribal Citizenship in Botswana’, Journal of Southern African Studies, 28(4): 672–84.

240  Ørnulf Gulbrandsen ———. 2002b. ‘Cosmopolitan Ethnicity, Entrepreneurship and the Nation: Minority Elites in Botswana’, Journal of Southern African Studies 28(4): 731–53. ———. 2004. Reasonable Radicals and Citizenship in Botswana: The Public Anthropology of Kalanga Elites. Bloomington: Indiana University Press. Willoughby, W.C. 1928. The Soul of the Bantu: A Sympathetic Study of the Magico-Religious Practices and Beliefs of the Bantu Tribes of Africa. New York: Doubleday. Wilmsen, Edwin N. 1989. Land Filled with Flies: A Political Economy of the Kalahari. Chicago: University of Chicago Press. Wylie, Diana. 1990. A Little God: The Twilight of Patriarchy in a Southern African Chiefdom. Middleton, CT: Wesleyan University Press.

Afterword Bilinda Straight

Reappraisals of classic works provide a critical space to reflect on their enduring importance for the history of theory in the field, consider the accomplishments they made possible and take stock of where the discipline needs to go next. The 1940 publication African Political Systems (APS) made an enormous contribution to the field of anthropology as a whole and to political anthropology as a new field, engendering debates that lasted decades. Although the structural-­ functionalist paradigm of APS yielded to other ways of seeing and doing anthropology, certain elements, such as the distinction between centralized and decentralized societies, persist in the popular, if not the anthropological imagination. The role of history as a theoretical perspective and of a particular historical period in shaping APS has likewise persisted in critical responses to the volume across the longue durée. The authors of APS certainly did acknowledge Africa’s colonization as a political reality. Nevertheless, in most chapters, the constraints of the structural-functionalist paradigm posed a barrier to meaningful consideration of the impact of colonialism on the ground. Given the COVID-19 pandemic that is ongoing as I write this Afterword and worldwide protests in the wake of the death of US citizen George Floyd in police custody, the current volume turns out to be timely in ways that the authors perhaps could not have anticipated when they embarked on their endeavour. In his Preface to APS, A. R. Radcliffe-Brown asserts the role of physical force and coercion in maintaining the social order and the push-pull arrangement between law and war. ‘Within the state’, he notes, ‘the social order, whatever it may be, is maintained by the punishment of those who offend against the laws and by

242  Bilinda Straight

the armed suppression of revolt. Externally the state stands ready to use armed force against other states, either to maintain the existing order or to create a new one.’ In the chapters that followed, African societies served, in their difference, as exemplars through which to understand the role of law and force in maintaining not only African political systems but all political systems. Moreover, they addressed the question Meyer Fortes and E. E. Evans-Pritchard posed towards the end of their introductory chapter: ‘Herein lies a problem of world importance: what is the relation of political structure to the whole structure?’ The present volume revisits this theoretical question and addresses the problem of history, the enduring distinction between centralized and decentralized political systems, and the relationship between law and conflict. The authors provide additional insights as well, which I will discuss in due course. All of the chapters share a commitment to revisit APS in a nuanced manner, correcting inaccurate perceptions and assessing its strengths, weaknesses and lasting importance through close rereading. Bošković both orients us and provides an intellectual history that does right by APS, revealing the depth of the debates that followed in response. He reminds us, for example, that the emphasis in APS on kinship and descent as key elements of African forms of political organization remain important with respect to their outlining in the original chapters and in stimulating ensuing debates. Lewis also grounds us in the intellectual history surrounding APS, taking us on an extended tour of the history of anthropological theory that begins with APS, but that extends well beyond, grounding his chapter in the opposing perspectives by which social structures totally encompass and usurp individual will or, in contrast, are abstractions that only exist by and through the actions of individuals and collectives. Although APS is typically viewed as an extended demonstration of the former, Lewis points out the exceptions within APS that are also key to its legacy. The critical note on which Lewis closes his chapter is one worth contemplating with respect to this present volume as a whole. After carefully evaluating the strengths and weaknesses of APS through his own rereading and in light of the decades of anthropological theory that followed, Lewis views the current state of anthropology critically, for too much emphasis on the perspectives of writers who never lived with anthropologists’ interlocutors and for relying on the single analytical tool of domination. The current volume is responsive to Lewis’ observation, drawing upon APS, to consider a set of African case studies through a variety of perspectives rather than by way of a theoretical monolith. Simonse, Hoehne, Schlee and Skalník each reassess, in different ways, the distinctions in APS between decentralized and centralized political systems. Each of these authors builds critically upon the theoretical contribution of APS to political anthropology, identifies the theoretical weakness of the ‘Group A’/Group B’ dichotomy and raises new questions. Simonse compares and contrasts the ­Baganda state and Mumonyi segmentary lineage and king structure,

Afterword  243

noting the potential to shift between ‘Group A’ and ‘Group B’ systems, even within the same political structure. Schlee revisits Oberg’s chapter in APS and considers Ankole political structure as occupying a space between state (Group A) and segmentary lineage system (Group B). He focuses on the state in particular, noting the deficiencies in evolutionary typologies of the origins of states and identifying the characteristics of both modern and ancient states in relation to the Ankole political system. Hoehne argues that the segmentary lineage system fails to work as theory, but that it can be useful heuristically, as exemplified in his narrativized retelling of events he experienced on the ground in Somalia. Like Schlee, Skalník critically considers the concept of the state, asserting that the long-term tendency to apply Eurocentric systems as models for understanding African political forms is teleological or even eschatological. He provides an overview of classic and contemporary works on the origin and typology of states, concluding that a new paradigm is needed that will consider political forms that are parallel to contemporary states, including terrorist and extremist groups. The chapters by Assal, Palmer and Gulbrandsen represent homages to APS that, like the other chapters, take the opportunity to offer new insights. Assal describes the underexamined Tuareg of Sudan, tracing the historical waves of migration that brought them to Sudan, the dramatic livelihood changes they have undergone, and their fate in the Darfur crisis as they settled in internally displaced persons (IDP) camps. As an homage, Assal’s chapter offers a description of a political system, while as a critical reappraisal, Assal provides an historically dynamic story of a social group that has actively adapted to marginalizing circumstances. Palmer addresses the Nkandla controversy surrounding Jacob Zuma that has developed into a political crisis in South Africa. He credits APS with providing a framework for analysing kinship and family organization as integral to African political systems. Thus, he interprets Zuma’s actions through the symbolically dense concept of homestead, arguing that Zuma’s populism was supported by his many marriages and the ostentatious building up of ‘homestead’ to support wives and children. In spite of this, corruption charges against his regime were made possible by a series of errors of judgement that crossed the line and led to a loss of legitimacy. Gulbrandsen nods approvingly to APS by examining the Tswana polity in its ideal form, but then goes on to describe the system’s postcolonial undoing as crucial cultural elements of the Tswana political system, botho and kagiso, have been reimagined over time in ways that turn them inside out. Concepts by which socially beneficial qualities of the person (botho) contribute to peaceful relations that are fragile and always in need of tending (kagiso), have been turned instead towards an emphasis on individualism that undermines Tswana concepts of the common good. In addition to homages and critical reappraisals of elements of APS that have long featured in anthropological debates, other themes raised in this volume also demand our attention. The editors of APS viewed African political systems as

244  Bilinda Straight

worthy of examination in their own right, but also as rich material through which to ask broad questions about the nature of society and polity. Undeniably Eurocentric in their perspective, the authors of APS nevertheless considered African societies on their own terms and through symbolically dense understandings that accorded meaning and legitimacy to political systems. However, the chapters in the current volume go further, often examining African political systems through their own optic, which is not reducible to Eurocentric viewpoints. Some chapters, such as those by Skalník and Palmer, take what might be explicitly referred to as an Afrocentric perspective. A consideration of the common good also surfaces in different ways. As already noted, Gulbrandsen considers the ways in which postcolonial politics have subverted Tswana concepts, including kagiso, a key symbol that Gulbrandsen considers as a ‘mystical value’ consistent with Evans-Pritchard and Fortes’ discussion in their introduction, and one that, in the Tswana case, enjoins all individuals to take responsibility for maintaining a social order that is ever-­fragile. Here, the common good is understood as a ‘highly inclusive societal order’. Schlee’s chapter examines the common good in quite a different way, which can again pivot on Afrocentric perspectives. Schlee argues that the Ankole king was perceived by the people to benefit the common good through his potency, but the king would not have seen himself as serving the common good. His actions, including raiding, looting, and acquiring cattle and wives, were desirable or aspirational for everyone and yet his motivations were to benefit himself. Palmer identifies a similar pattern in the homestead building actions of Jacob Zuma. What then is meant by the common good and how might we relate it to the central questions in APS of the relationship between political structure and the whole structure? Have colonial and postcolonial political realities so fundamentally changed the ‘mystical value’ by which African social structures legitimate political structures that the line that Jacob Zuma crossed is becoming everywhere always crossed? Let us return to current events unfolding in June 2020 as I write, and in particular the death of George Floyd on 25 May 2020. In the state system governing George Floyd, the police are empowered to take into custody, by physical force if necessary, citizens who break laws and therefore threaten the social order. Is the social order here synonymous with the common good? Deaths in US police custody occur all too frequently, a fact that points analytically to the ability of the state to deprive some of its citizens of life without due process, as part of the cold calculus of maintaining the social order. Yet, this particular death happened after the loss of over 100,000 US citizens to a pandemic disproportionately affecting African Americans, after weeks of stay-at-home orders and massive job losses, and after a series of other deaths and outrages against African Americans. The ensuing protests and riots, which were met with international solidarity, point unequivocally, nakedly, to a mystical value that the authors of

Afterword  245

APS – ­publishing in the early years of the Second World War II and before the civil rights movement and other mid- to late twentieth-century social ­movements – did not yet imagine. That mystical value might be summed up in the United States as ‘equal under the law’, but it also encapsulates the burden of shared history, the ancestral and ongoing guilt of racial dominance, and a new generation’s insistent exclamation that too many lines have been crossed. The current volume is a timely reappraisal of APS; it provides a rich set of analytic tools to better understand contemporary and historical political structures in African societies and to apply those understandings to central questions in anthropology and social systems globally. These analytic tools include a renewed and historically nuanced consideration of authority, violence and the imbalance of power in political systems, and a meditation on the common good in relation to political structures that safeguard the rights of those in power to rule through force. Speaking in the aftermath of ordering military troops to Washington DC and the addition of heavy fortifications to the White House, the 45th President of the United States told Fox News Radio on 3 June 2020: ‘We have to have a dominant force. Maybe it doesn’t sound good to say it, but you have to have a dominant force. We need law and order.’ This edited collection positions us well for a rich set of debates to come, which might begin with an unpacking of that statement. Bilinda Straight is Professor of Anthropology and Gender & Women’s Studies at Western Michigan University, United States. Earlier in her career, she published on the anthropology of experience, gender, aesthetics and material culture. Since 2008, she has focused on the nature of emotion and human experience through the impact of cultural and moral norms on pastoralists in the midst of northern Kenya’s extreme environment and chronic, low-intensity warfare. Most recently, her National Science Foundation-funded projects have examined the intersection between humans and world suggested by epigenetics, utilizing an ethnographic and multidisciplinary approach.

Index

acephalous, 11, 27, 47, 56, 61, 71–72, 96, 116, 125, 127, 129 polities, 1, 56, 61, 71, 72, 96 affinal, 52, 142, 171 affinal ties, 142–43, 146 Afrocentric, 244. See also interpretative perspectives or biases Afro-pessimism, 224, 229 agency, 25, 31, 140, 211, 220 age-class system Shilluk, 52, 87–88, 97, 98n1 agem, 83–85, 88, 93, 98n2, 99n7. See also, rebellion, revolution agnates, 47, 98–99, 141–43, 147–48, 150, 155, 237, 215, 221, 223 antagonism bridging by blacksmiths and woman, 45, 71 consensual, 49, 50, 66, 72, 85 generations, 45 king and people, 53–56, 61, 66, 71 section, 55–56 Alemayehu Debelo Jorgo, 134 alliance, xiv, 10, 57–58, 86–87, 131, 142, 146, 170, 175 formation, 146

ancestor, 23, 48, 62, 75n18, 82, 86, 90, 126, 165, 182, 186, 214–15, 245 ancestors [badimo], 211 discursive field of the kgotla, 211 royal [badimo ya dikgosi], 213, 221, 232, 234n4 anthropological theory, ix–x, xiv, xix, 1, 10, 12nn6–7, 16, 18–20, 26, 27, 29, 33, 35–36, 40n14, 41n20, 41n23, 93, 107, 111–12, 114–17, 123, 126–27, 135, 136n14, 139–44, 158, 241–42 structural-functionalism, structuralfunctionalist, xi, xiv, 12n7, 16, 20, 22, 24, 27, 32, 47, 50, 71, 102, 104, 106, 142, 183, 198, 291n20, 214, 241 anthropology British social anthropology; ix–xiii, 12n3, 15–16, 20–23, 25–26, 35, 41n20, 122–23, 139, 141, 158–59, 182, 199 history of, 1, 10, 39, 105, 242 of the state, xii–xiii, 1–2, 6–7, 10–11, 19–26, 36, 39, 39n3, 40n5, 40n8, 49, 56, 61–62, 64, 67, 72–73, 75n18, 96, 102–17, 117nn2–3, 117n5, 123, 126, 128, 130–33, 144, 148, 207, 224–25, 227, 232, 234, 235n10, 241–43

248  Index army, 57, 60, 87, 91, 153, 164–65, 173, 175, 177, 184, 200n10, 208, 219 Amakuta, 57 Awusa, 57 bang reth, 81–82, 86, 88–89 Buchura, 57 Makatub, 57 authority. See also conflict centralized, xii, 5, 19, 47–48, 123, 126, 136n15, 141, 167, 169, 170, 172, 177, 182, 215, 226, 231, 241–42 frailty of, 219 personal capacities, 221–22 Bachofen, Johann Jakob, 132 Baker, Samuel, Governor Equatoria, 1871–73, 58, 74n12 balancing of power, 49, 53, 54 equilibrium, xiv–xv, xvii, 17–18, 20–21, 24, 27, 29, 32, 40n10, 41n17, 48, 52, 115, 183–84 generations, 51 sections, 48, 71 Bari, ethnic group, 58, 60, 63, 68–69, 74n11, 74n15, 86, 94–97, 98n1 Barth, Fredrik, 23, 30, 31, 35, 37, 41n19, 41n23, 142, 146 Baştuğ, Sharon, 144, 158 Bernardi, Bernardo, 134 bicephalous, 130 bogosi [Tswana kingship], 210, 212, 220–22, 224–25, 232, 236n26 boloi [sorcery], 220, 222 botho [personal character], 212–13, 220–21, 231–32, 243. See also kagiso Botswana, 10, 47, 207, 218, 234nn1–2, 235n16, 236n24 ambiguity of Botswana as an ‘African miracle’, 228–29 ambiguity of modern state – Tswana kingship relations, 103, 225–26, 243 independence, 224, 236n26 parliamentarian republic, 10, 206, 224–25, 233 state-centred political economy, 207, 229–30 (trickle-down effect) Botswana Democratic Party, 224

capitalism, 34, 37, 130, 199, 228–29 Carneiro, Robert L., 107–11, 122 caste, xv, 32, 59, 209, 214, 219 cattle, 54, 57, 61, 67, 81, 83, 86, 88, 90, 124–26, 129–30, 135n2, 173, 180, 184–85, 188, 196, 198, 211, 218, 220, 223–24, 226–30, 244. See also pastoralism commercialization, 226–27 inequality, 226–30 political significance of, 223, 227 centralization. See also political system capital, 60, 62, 64–65, 75, 89–91 centralizing ethos, 69 concentration of ritual powers in the hands of the king, 60 control access to land, 69–70, 75n20 frustrate bottom-up complementary fusion of segments, 80–82 planting princes in outlying areas, 58, 69, 86 turning sectional chief into royal appointees, 88 charisma, 56, 59, 225 Claessen, Henry J.M, 62, 107, 109–11, 113 117n4, 136n11 clientship, 57, 69, 70, 76, 86 cohesion, sociopolitical, 48, 49, 56, 69, 142, 184 commoners, 185, 188, 195 bakopi (Ganda), 65, 69 böngön (Bari), 97 collo (Shilluk), 79, 82, 83, 86, 89, 94, 97 Chaillé-Long, C., 63, 65, 75n17 chiefdom, xiii–xiv, xvi–xviii, 2, 11, 19–21, 25–27, 36, 40n4, 47, 54, 64–67, 76n23, 81–85, 87–88, 90, 92, 97, 102–16, 122, 131, 157, 159, 167, 183–84, 195–96, 198, 211, 216–17, 219–21, 234n9, 235n12 Civil Rights Movement, 37, 39, 41, 192, 200n15, 245 civil war, xvii, xiix, 20, 145–46, 150, 170, 190 clan, x, xii–xiv, xvii, 7, 22–23, 48, 51, 56, 57–58, 60, 69–70, 72, 74n9, 84, 86, 90–92, 94–95, 97–98, 123, 125, 128, 134, 140–42, 145–53, 155, 157–58, 163, 165–67, 171–72, 184–85, 190, 192, 221

Index  249 clan family, 141–42, 146, 148 colonialism, colonization, postcoloniality, ix, xi–xv, xvii–xix, 2–3, 5–6, 8, 10–11, 12n3, 20–21, 25, 26–27, 32, 37–39, 47–48, 54, 61–62, 64, 70, 74n4, 74n6, 81, 84, 87, 102, 105–8, 110, 114–16, 122–23, 130–31, 135n4, 135n6, 135n9, 140–42, 147–48, 158, 165, 182, 198–99, 200n13, 201n20, 201nn24–25, 206–7, 222, 224–31, 233, 234n1, 235n10, 235n21, 236n25, 241, 243–44 common good, 125–27, 129, 210, 212, 215, 217–18, 221, 225, 227, 230, 232–33, 235n21, 236n25, 243, 244–45 Condominium rule, 80, 81, 88, 91 conflict. See also violence, war authority, 245 crisis, 49, 50, 53, 54, 61, 81, 243 looting, 81, 83, 126, 129, 165, 177, 244 raiding, 61, 74n12, 76n23, 81, 83, 88, 127, 130, 155, 163–65, 173, 244 confrontation, xiv, 48, 52–54, 67–68, 71–72, 75n19, 79, 81–83, 114, 141–42, 230–31, 236 Congo, 4, 73, 98, 111 contract theories (of the state), 126–27, 136n14 cooperation, 65, 76n21, 134, 141–42, 215 corporate group, 32, 47, 69, 73 COVID-19 pandemic, 241 Critical Development Studies, 39 cross-cutting ties, 142–43, 151 cultivation for the king, 53, 57, 58 culture shock, 123 curse, 53, 54, 56, 66, 69, 94 customary law, 2, 129 Dahomey, 4, 20, 105 Darfur, 11, 163–77, 243 Darood (clan family), 146, 148 Darwinism, 50 democracy, 10, 98, 105, 116, 129–30, 148, 180, 183, 187, 190, 194–95, 198–99, 201n17, 206, 216, 218–19, 224–28, 230, 232 descent, 9, 20, 22–23, 27, 30, 34, 48, 55–56, 63, 69, 73, 104, 125, 135n2,

135n9, 139–42, 144–45, 147–48, 152, 152, 157, 159, 211, 213, 215–16, 221, 242 diaspora and transnationalism, 39, 157–58 dikgaba [damaging infliction by ancestors and elders), 213 Dinka, ethnic group, 22, 83, 87–88, 92, 94–96, 98n3, 104 ditlhare [productive/protective medicine], 222 divide and rule, 81, 88, 142, 177 Dogon, 7 domination, 8, 20, 23–24, 31, 36, 38, 96, 207, 218–19, 223, 242 ethnic, 25, 148, 170, 183, 185, 198, 218, 226, 235n16 drum (sacred), 124, 126, 130, 133–34 DuBois, Cora, 112–13 dumèla [consensus], 213 Durkheim, Emile, xiii, 15, 16, 19–20, 27, 30, 32, 39n2, 40n10, 41n17, 50, 59, 140 early kingship, 49, 59, 61, 71, 72, 79, 96, 9 fragility, 54, 56, 72 reversal to segmentary configuration, 5 reversibility of power balance, 54 early state, xiii–xiv, 10, 61–62, 72, 79, 97, 102, 105–8, 110–16, 117n3, 117n5, 126, 133 irreversibility of power balance, 54, 61, 66, 68, 80, 82 egalitarian individualism, 219, 224–5, 228, 233 egalitarianism, 208–9, 214–16, 218–19, 231–33, 235n12 elites, economic-cum-political, 227, 233 entrepreneurial, 207 as greedy, 228, 230 land appropriation, 227 state consolidating grand coalition of, 228 ethnic minorities, 218, 226, 235n16 Eurocentric, 107, 116, 182, 199, 233, 243–44. See also interpretative perspectives or biases Evans-Pritchard, Sir E. E., ix–xii, xv, xvii–xix, xixn2, 6–7, 9, 15–23, 39n4,

250  Index 47–49, 52, 55, 59, 71, 73n1, 75n19, 81, 84–85, 93, 96, 98nn4–5, 103, 123, 139, 141–42, 148, 160n2, 169, 182, 184, 196, 198–99, 206–8, 210, 212, 214, 224, 232, 242, 244 evolution, x, xiii, 26, 27, 50, 70, 72, 73n2, 74n3, 103, 105–11, 114, 126, 131, 140, 164, 172, 243 evolutionism, 9–10, 102, 110, 114, 131 existentially important values, 208, 212–12, 214, 231–2, 235n20 family organization, 135n4, 168, 213, 216, 226, 243. See also descent, kinship feud, ix, xvii, 24, 61, 81, 86, 88, 104, 151, 153, 155, 157, 159 firearms, 57, 59, 60, 63, 74, 86, 89, 90 fission, 82, 141, 146 Floyd, George (death), 242, 244 folk model, 143–44, 150 Fortes, Meyer, x–xiii, xv–xiv, xviii–xix, xixn2, 3, 6–7, 9, 15–23, 26–27, 39, 40n7, 47–48, 59, 71, 73n1, 103, 105, 123, 139–41, 169, 182, 184, 196, 198–99, 206–8, 210, 212, 214, 224, 232, 242, 244 Foster, George M., xii, 123 Foucault, Michel, 37–38, 102, 227 Frazer, James, 54, 65, 136n13 functionalism, xi, xv–xvi, 2, 6, 8, 9, 12n7, 16, 17, 20–21, 24, 27, 32, 33, 47, 50, 104, 106, 183, 198, 201n20, 214, 241 funeral of the king, 64–72, 93–95 companions buried with the king, 68, 93, 95, 98n fusion, 48, 82, 141, 146 genealogy, 52, 140–41, 143, 146, 148, 151, 155, 158–59, 215, 223 Girard, René, 49–50, 66 Gordon, Charles, Governor Equatoria, 1873–76, 70, 75n17 Gouldner, Alvin, 67, 75n19 Graeber, David, 91–93, 216, 235n10 Grant, James, explorer, 63, 70, 75n21, 76n23 gweth (Pari: dying king’s power of blessing), 94

harmony ideology, 214 Harti (genealogical nodal point), 146–51, 153, 155 heuristic device, 11, 140, 144, 158, 160n1, 160n4 hierarchy, 68, 73, 83, 88, 96, 98, 110, 116, 125, 142, 146, 207–10, 213–16, 220–22, 232 and authority, 208–9, 213, 221–22, 232 and rank, 208–9, 215–16, 221, 223, 232 ranking as conflict-generating, 216, 223, 232 Hima, 124, 129–30, 135n2, 135n4 Hobbes, Thomas, xix, 4, 40n15, 41n22, 217 Hofmayr (also Hofmeyer), Wilhelm, missionary, 80–81, 84, 86–87, 89 holism, 16, 199, 209–10, 218, 220–21, 224–25, 229, 232–33 Holý, Ladislav, 1, 9, 11, 106, 158 Homer, 131 homicide, 54, 124, 126–27, 129 honour, 54, 131–32, 171, 194 Howell, P.P., Colonial administrator, 81, 87– 88 human rights, 39, 127 humanitarian intervention, 128 humanitarianism, 39, 178 Hutu, 124, 135n2 immigrants, xi, 74n9, 82, 114–15, 181, 223 indirect rule, ix, xi, xiii–xv, xix, 12n3, 21, 108, 182, 200n12 individualism, 36–37, 41n22, 209, 218–19, 224, 232 243 freedom, 220 liberal, 207, 227–33 non-Western, 221 individuality, 40n13, 220 inequality racial, 219 socioeconomic, 106, 227–28, 232, 236n25 installation of the king, 68–69, 82–83, 90–92 Buganda king turned into predator, 69 capture of Lokoya king as leopard and domesticated as giraffe, 69, 93 enthroners (nyikwom), 56, 58, 72, 91 royal incest, 91, 92

Index  251 sacred stool and necklaces, 91, 99 transfer of diseases on new Bari king, 69 triple capture of Shilluk king, 82–91 internally displaced persons (IDPs), 164, 169, 175–78, 243 International Criminal Court (ICC), 127, 174 international law, 127–29, 131 interpretative perspectives or biases, 93, 104 Afrocentric, 244 Eurocentric, 199, 243–44 Iru, 124, 129–30, 135n4 Isaaq (clan family), 148, 153 jago (Shilluk), 88 jwei (Pari), 95–96 kagiso [peace and harmony], 212–16, 218, 220–23, 231–33, 235n23, 243–44. See also kgosi, kgotla as source of prosperity, 212–14 key symbol, 212 Karugire, Sanwiri R., 124–25, 135n6 Kgosi Kgafela II of the Bakgatla, 236n25 kgosi, pl. dikgosi [Tswana kings], 210. See also egalitarianism, kingship, kagiso, sovereignty ambiguity of authority, 220–24 as court president, 213, 215, 227 assassination of, 220 assessment of, 221 by virtue of the people, 216 and Christianity, 210–11 cosmology, 210, 214 enthronement, 234n9 fear of (boifa), 222, 230 kingship (bogosi), 210 as a ‘little god’, 210 and missionaries [evangelizing], 210–11 monopolization of trade, 223 as motswadintle [the one from whom good things come], 217, 222 occultic practices, 222 as priest, 211 rainmaking, 211, 234n4 as sorcerer, 220, 222, 235n21 spiritual power, 212 as ultimate source of prosperity, 212–14 weak, 235n18

kgotla [Tswana popular council/court]. See also kagiso, Kgosi as both secular and spiritual field, 211, 226 as customary court, 216, 227 freedom of speech, 216, 218 ideals of transparency, 230, 232–33 Khama, Ian, Lt. General, former President of Botswana, Kgosi of Bangwato, 230, 236n26 Khama, Seretse, founding President of Botswana, 225–27, 230 Khama, Tshekedi, regent of the Bangwato, 222–23, 230 king blackmailing people, 57 executions, 61–67, 74n15, 75n17 justice, 61, 64, 65 mediator, 53, 81, 88–89 peace of the king (mirembe), 56 peacemaker, 88 police, 62, 65 posthumous curse, 56, 66, 69, 96 punishment, 64, 67, 74n21, 81, 82, 88 rainmaker, 51, 58–60, 74n9 roaming diplomat, 58 royal clan, 51, 57, 60, 70 royal herd, 57, 86, 89 royal levy, 81 sacred power, 92–94 sacrificial destiny, 91–93 share in bloodwealth paid between subjects stealing, extorting rainstones, 53, 60, 78n9 symbolism of king’s sacred power, 94–95 will to power, 56 kingship, xii–xiii, xxii, 1, 3, 4, 10, 19, 20, 23, 39, 47–49, 51–73, 73n2, 74n4, 74n7, 74n9, 74n11, 74n13, 74n15, 74nn16–20, 75nn22–23, 79–98, 98nn1–2, 98n5, 99n5, 99n7, 99n9, 99n11, 103–8, 110–11, 114–16, 117n3, 123–31, 133–34, 135nn7–8, 135n10, 136nn12–13, 183–85, 188, 196, 198, 201n24, 206–10, 212, 214–26, 229, 231–34, 234n1, 234nn3–4, 234n8, 235nn10–12, 235n21, 236n25, 242, 244. See also bogosi, early kingship, kgosi

252  Index African vs. Eurasian, 207–10, 214–15, 216, 224, 232–33 African, xii, 10, 49, 51, 59, 71, 73, 206–10, 214–16 as intoxicating, 129, 220, 222 magicality of, 125, 129–30, 211, 222–23 and modern state, 73, 75n18, 79, 103–5, 108, 133, 206–7, 224–27, 229, 233 mythological origin, 220–21 premodern, 209, 216, 232, 235n10 symbolism of, 206–8, 210, 216, 225–6, 229, 234, 236n26 theatre state, 68, 75n20, 214 Tswana, 10, 206–10, 212, and incorporation in the modern state of Botswana, 224–27 kinship, xiv, 9, 12n5, 16–17, 19–20, 22–23, 25, 29–30, 33–34, 38, 39n4, 40n9, 47, 73n1, 94, 103, 114, 135n2, 135n4, 140, 144, 150, 157, 158, 160n3, 164, 166, 170–71, 185, 196, 199, 242–43 kiwendo, 64, 66, 72. See also mass-executions Kuper, Adam, 9, 12n3, 17, 140, 144, 157–58, 184–85, 200n12, 201n17, 224, 233 Kuper, Hilda, 7, 22 Kurimoto, Eisei, 84, 94, 99n12 kutlwano [mutual understanding], 213, 234n6 Laascaanood (place), 148–51, 153, 155 law, 7, 17–18, 39, 61, 64, 67, 103, 112–14, 117n2, 125–29, 131–32, 142, 147, 189, 199, 210, 212–13, 216, 225, 234n6, 235n11, 241–42, 244–45. See also, customary law, due process equal under the law, 167 Leiris, Michel, 8 Lévi-Strauss, Claude, 7, 9, 40n4, 50 Lewis, Ioan M., 22, 129, 140, 143–44, 158, 160n2 Lienhardt, Godfrey, xiii, 22, 84–85, 95–96, 98n3 Linant de Bellefonds, Ernest de, 70, 75n17 lineage, xii–xiv, 8, 20, 22–23, 26–27, 48, 69, 82, 84, 97–98, 104, 123–26, 131, 134, 135n2, 139–41, 143–44, 146–47, 149–50, 152–56, 159, 185, 221. See also segmentary lineage

Livingstone, David, 5 Lokoya, ethnic group, 51–52, 60, 69, 74n4 Lord Lugard, xiv, 3–4, 12n3 Lotuho, ethnic group, 56, 58–60, 68–69, 72, 74n4 Lulubo, ethnic group, 51–52, 58, 60, 68–69, 74n4 Mahdist rule, 57, 60, 87 masegò [blessing, good luck], 212 mass executions, 64, 68–69, 71, 75n17 Mauss, Marcel, 7, 50, 67, 213 McComb, Marlin R., xii, 123 mechanical solidarity. See solidarity megwa le melao ya Setswana (Tswana custom and law), 210, 212 migration, 26, 81–82, 94, 104, 125, 135, 165, 167, 173, 176, 243 mob lynching, 61 model, xii, xiv–xv, xvii, 2, 9, 11, 16, 20, 22, 24, 26, 29, 33, 34–35, 37, 40n5, 40n11, 41n23, 49, 54–55, 73, 81, 108, 123–24, 135n1, 139–40, 142–44, 146, 148, 150, 157–59, 160n1, 199, 243 modernity, 10–11, 24–25, 27, 36, 73, 102–3, 105, 108, 112, 115–16, 129, 132–33, 186, 207, 209, 218, 220–21, 223–27, 229, 231, 233, 243 mogote [destructive heat], 212–3. See also kagiso monarchy, 39, 106, 130, 224 monyomiji, 49–51 age-class system, 51–53 monyomiji cluster, 98n1 reciprocal victimization of king and people, 53–56, 61, 65, 66, 73, 74n14 Morgan, Lewis Henry, xiii, 26, 112, 132, 140 Muteesa, king of Buganda, 1856–1884, 62–64, 70, 75n17, 76n23 mystical value, 19, 48, 184–85, 196, 198–99, 207–8, 210, 212, 219, 224, 232, 244–45 myth of spear and bead, 56, 58, 72 neoliberalism, 39, 187 ngaka (pl. dingaka) [Tswana ‘doctors’/ magicians], 211, 222 Nigeria, xi, 3–5, 12n3, 21–22, 47, 116, 176

Index  253 Nkandla (rural homestead of Jacob Gedleyihlekisa Zuma, former President of South Africa) acceptance of it by the masses, 195 cost to the taxpayer of non-security upgrades, 181 nature of the scandal, 180–81 Public Protector’s Report, 181 significance as a Nguni rural homestead, 185–87 nnamasole (Ganda queen-mother), 62, 70, 76n23 Nuba, 86, 87 Nuer, ix, xiii, xv, xvii–xviii, 6–7, 9, 20–22, 47, 49, 52, 72, 87–88, 103, 123, 129, 139, 142, 158 Oberg, Kalervo, x–xii, xiv, 3, 10, 39n1, 107, 122–25, 127, 130–134, 135nn1–2, 135n4, 135n6, 135n10, 243 Ottoman Empire, 83, 135n9 patrilineal, 125, 139–41, 143, 145–46, 148, 152, 156–57 patrimonial/neopatrimonial, 224, 228 Pauketat, Timothy, 122 Paulme, Denise, 1, 7–8 peace, 49–50, 55, 66, 71, 81–82, 88–89, 92, 94, 107, 143, 146–47, 151, 153–55, 157, 159, 165, 171–72, 174, 176–77, 190, 208, 212, 214–16, 218, 221, 225, 232, 234n7, 235n23 peaceful relations, 243 personhood, 17, 28, 29–31, 33, 81, 92, 104, 133, 141, 165, 170, 184, 192, 212–13, 220–22, 229, 232, 235n12, 235n20, 243 Peters, E.L., 143, 157 Perner, Conradin, 84 political system (or structure or organization or form). See also authority, due process, kingship, kinship, segmentary lineage system, violence centralized, 27, 53, 56, 58–60, 62, 69–72, 87–88, 96, 102–4, 107–8, 112, 116, 117n2, 129, 184, 185, 198, 206–8 custody, 69, 241, 244 decentralized, 241–42

due process, 244 extremist, 11, 116, 243 (see also terrorism, terrorist) group A, xii–xiii, xxii, 10, 19, 47–48, 55, 59, 61, 71, 96, 103, 123, 133, 169, 243 group B, xii–xiii, 8, 10, 19–20, 47–48, 55, 61, 96, 103, 123, 133, 169, 242–43 imbalance of power, 72, 245 physical force, 18, 55, 61, 66, 72, 96, 103, 241, 244 police, 39n3, 62, 65, 127, 147–50, 155, 170, 175, 177, 190, 193, 230–31, 241, 244 power, x, xii, xvii, 2, 19, 20, 21, 23, 25, 27, 28, 29, 31, 36, 38, 39n3, 40n17, 48–49, 51–54, 56, 59–61, 64–66, 69–73, 75n18, 79–80, 83–85, 87–92, 96–97, 103–4 terrorism, terrorist, xiii, 11, 116, 157, 243 polycephalous, 125 postcolonial state, 105, 114, 116, 130, 206–7, 224, 225–26, 229–33 poststructuralism, 38 poverty, 26, 168, 180, 186–88, 194, 197, 228, 2300 power. See also balancing of power, kgosi abuse of, 23, 48, 75, 182, 210, 222–23, 230–32 destructive, 51, 56, 94, 125, 210, 219, 234n5 greed for power/wealth, 210, 220, 223, 228, 230, 233 magic, 17, 129–30, 211, 222–23 militarization, 145, 174–75, 230 power vs religion (Dumont), 207, 208–10 repressive, 68, 107, 207, 223, 230–31 secretive, 210, 222 spiritual, 211 practice, xiii, xv, xvii, 28, 50, 67, 71, 74n8, 81, 87–88, 93–95, 111, 127, 139–43, 147, 158–59, 165, 167–68, 207, 210–11, 213–14, 216–17, 219–20, 222–24, 226, 228, 230–32, 234n4, 235n21 Puntland, 145–53, 155, 157

254  Index Radcliffe-Brown, Alfred Reginald, ix–xiii, xvi–xix, 2, 9, 12n4, 15–18, 21–23, 27–28, 30, 32–33, 39nn1–3, 40n10, 103, 241 raiding, 61, 74n12, 76n23, 81, 83, 88, 127, 130, 155, 163–65, 173, 244 rank-and-file, 228, 230 rebellion, 19, 24, 26, 40n6, 40n10, 48, 60, 65, 70, 79, 84–85. See also agem rebels, 65–66, 74n12, 153, 163–66, 172–75, 177, 188, 201n25 regicide, political, 64, 71, 73, 81, 85 in Funj, 98 regicide, ritual, 93–96 Dinka Spearmaster Shilluk-Luo king, 99 suffocation, 92, 94, 96–98 reign of the dead king, 94 relativity, 6 religion, 6, 11, 17, 27, 29–30, 41n20n, 50, 72, 79, 92, 95, 116, 131, 153–54, 157, 160n3, 165, 175, 186, 207–10, 213, 233 retaliation, 54, 63, 122–24, 127, 134 revenge, 54, 58, 64, 66, 83, 85, 124–29, 131–32, 152–53, 156 revolution, 20, 84–85. 133, 136n16. See also agem Richards, Audrey, x–xi, xv, xvii–xviii, 3, 15–16, 20–23, 63, 71 ritual, 11, 16, 20, 22–23, 26, 29, 48, 51, 60, 69, 71, 82, 93, 95–96, 99n11, 115, 125, 130, 134, 172, 184, 186, 196, 213, 217, 231, 234n4 abandonment of, 211 murder, 230 rivalry, 56–60, 66, 70, 72, 80, 81, 83–86, 92, 93, 216, 223, 232 Rousseau, Jean-Jacques, xix, 4, 217–18, 221, 235n14, 235n19 Ruch, Philipp, 131–32 sacred, 49, 50, 88, 106, 114, 125, 136n13, 184, 208–9, 211 sacrifice, 50, 54, 89, 90 mass-executions as sacrifice, 66 noiseless, 94 power of sacrifice, 95–96 Said, Edward, 38

Salzman, Philip Carl, 143–44 scapegoating, 54, 68, 74, 75n18 scapegoat mechanism, 49–50 Schnepel, Burkhard, 84, 90, 93, 96, 99n6 secret societies, 3, 7, 12n5, 30 secularism, 26, 30, 39, 72, 208–9, 211–12 segmentary lineage system, xii–xiii, 2, 9, 11, 22–24, 27, 40n11, 103–4, 123, 129, 133, 139– 44, 148, 150, 158–59, 242–43 segmentary model, 22, 144, 146, 148, 150, 157–59 segmentary opposition, 47, 51–52, 55–56, 58, 71–72, 75n19, 79–81, 159 complimentary, 52–56 intersectional conflict, 52–53 moiety, 53, 55, 56, 81–83, 89, 90 non-aggression pact between sections, 52 stick fights, 52 structural homology with early kingship, 55, 56, 79–83 synchronicity in the formation of warrior age-sets, 62 segmentary state, 20, 23, 72, 73, 104, 115 Shilluk king Dak, Shilluk legendary king and divinity, 82 Tokot, ca. 1680, 86–87 slavery, 5, 59, 63, 67, 71, 86, 114, 163–65, 200n14, 219, 235n11 social cohesion, 48, 73, 142, 184 social contract, 126–27, 136n14, 217 social order, 18, 24, 27, 104, 208, 209, 210, 212, 214–16, 218, 220–22, 232, 235n21, 241, 244. See also structure socialism and postsocialism, 39 solidarity, 16–17, 20, 41n17, 59, 140, 141, 142–43, 149–50, 159, 160n3, 217, 244 mechanical, 16, 20, 59, 140, 160 Somali, 22, 128–29, 139–43, 145–47, 150–51, 153–59, 160n6 Somalia, 11, 139–40, 144–50, 152–55, 158–59, 243 Somaliland, 145, 147–48, 150–53, 155, 157 South Africa background, 183 influence on anthropologists, 183–84

Index  255 relevance of ‘mystical values’ concept from African Political Systems, 184–85 Southall, Aidan, 23, 72, 104, 115, 133–34 sovereignty, 54, 61, 66–68, 72–73, 88, 108. See also holism as delegated kgosi by the people, 216–8, 224, 232, 235n2 egalitarianism, 216, 218, 232 enhancement of, 207 sorcery, 235n21 state formation, 49, 50, 61–64, 67, 71–73, 96–98 achieve control of access to land, 69–70 breaking the dynamic of bottom-up segmentary fusion, 80–83 monopoly of the use of physical force, 55, 61, 68, 72, 80, 96 organic solidarity, 59 sovereignty, 54, 61, 67–68, 72, 73, 88 substitution of personal relations based on reciprocity by centrally controlled complementary role-relationships, 67–70, 89 statehood, 2, 72 stratification, 59, 82, 80, 97, 106, 113structuralism, 9, 30, 50, 72–73, 105–6, 110–13, 126, 130–33 structure, ix, xiv, xvi–xvii, 2–3, 6, 17–22, 24–25, 28–32, 34–36, 38, 39n4, 40n14, 52–53, 56, 67, 71, 73, 96, 98, 104, 108, 114–15, 125, 131, 135n4, 140–42, 144, 146, 158–59, 169–70, 182, 190, 196, 207, 214–16, 224, 226, 228–29, 231–32, 234, 242–45 succession, 26, 56, 57, 61, 70–72, 76n23, 83, 84, 86, 90, 91–93, 96, 97, 98n5, 113–14, 125–26, 135n9, 221 alternation between royal lineages, 84, 98n5 interregnum, 65–66, 96 killing of successor’s brothers, 70–71, 76n23, 84, 87, 98n5 nyireth, 79, 86, 87 subalterns and social movements, 39, 230–31 Sudan, xi, xviii–xix, 5, 7, 9, 11, 79, 49, 51, 64, 73, 74n4, 80, 97, 98n1–2, 98nn4–5, 103, 105, 132, 163–66, 168–70, 172–74, 176–77, 243

Sumer, 126, 130 symbolic wealth, 207, 224–25, 233, 236n26 Tanzania, 5, 23 terrorism, terrorist, xiii, 11, 116, 157, 24 tetlanyò [reconciliation], 213–14, 234n7 theory (see anthropological theory) the Universal [Hegel], 208–10, 214 total social fact, 213, 217 trade, xvi, 8, 57–60, 89, 98n2, 113, 131–32, 188, 223 armed escorts, 89 effects on trade of blockage of sudd, 60 firearms, 57, 59, 89 iron trade, 67 ivory trade, 59 Khartoum Embassy of Shilluk Kingdom, 89 slave market in Kaka, 86, 89 transnational, 39, 140, 157–58 Tswana, 10, 47, 103, 206–26, 228–233, 234n1, 234n3, 234nn8–9, 235n10, 235n20, 236n25, 243–44 Tuareg, 11, 163–72, 174–77, 243 Turco-Egyptian rule, 59, 74, 86, 89, 98n2 Turner, Bertram, 123, 127, 135n3, 135n5, 135n8 Tutsi, 124 unilineal, 20, 22, 30, 104, 126, 141–42, 144, 159 United Nations (UN), 127, 146, 174, 177 uterine descent, 125, 135n9 Vinco, Don Angelo, missionary, 57–58 violence, xii, 37, 49–50, 53–54, 63–64, 66–68, 71–73, 75n20, 79, 83, 88, 92–93, 96, 104, 107, 109, 115, 123–25, 128–29, 131–33, 146–47, 166, 173–75, 187, 189, 193, 197, 219, 229, 230–31, 235n21, 245 virility, 125 Vöneky, Silja, 127–28 Wagner, Günter, xi–xv, 5–7, 39n1, 40n5, 122

256  Index war, ix, xi–xiii, xv, xviii, 1–2, 5, 7, 18, 20, 27, 37, 40n11, 41, 49–53, 55, 57, 59, 66, 75, 76n23, 82–84, 86–89, 93, 103–4, 107–11, 113, 129–31, 133, 135n9, 150–52, 163–64, 169–70, 172, 174–77, 182–83, 186, 201n24, 224, 241, 245. See also civil war captive, 57, 89, 93 warlord, 145–46, 158 warrior, 50, 67, 83, 84, 87, 88–89 wei (Dinka, Shilluk), 95–96. See also power of sacrifice Wrigley, Christopher, 64, 69–72, 75n18, 76n23

Zuma, Jacob Gedleyihlekisa decline and fall, 181, 197–98 explaining his grass-roots support, 193–97 finances, and corruption allegations, 191 involvement in transition to democracy, 190 marriages and children, 189–191, 193 origins, early life, and education, 188 politicization and imprisonment, 188–89 presidency, 192–93 rape trial, 191 underground activism, 189–90 Vice-President, 180, 190