Warfare in African History 0521195101, 9780521195102

Citation preview

Warfare in African History

This book examines the role of war in shaping the African state, society and economy. Richard J. Reid helps students to understand different patterns of military organisation through Africa’s history; the evolution of weaponry, tactics and strategy; and the increasing prevalence of warfare and militarism in African political and economic systems. He traces shifts in the culture and practice of war from the irst millennium into the era of the external slave trades, and then into the nineteenth century, when a military revolution unfolded across much of Africa. The repercussions of that revolution, as well as the impact of colonial rule, continue to this day. The frequency of coups d’état and civil war in Africa’s recent past is interpreted in terms of the continent’s deeper past. Richard J. Reid is Reader in the History of Africa, Department of History, School of Oriental and African Studies, at the University of London. He is the author of several books, including Frontiers of Violence in North-East Africa (), War in Pre-Colonial Eastern Africa () and Political Power in PreColonial Buganda ().

New Approaches to African History Series Editor Martin Klein, University of Toronto Editorial Advisors: William Beinart, University of Oxford Mamadou Diouf , Columbia University William Freund, University of KwaZulu-Natal Sandra E. Greene, Cornell University Ray Kea, University of California, Riverside David Newbury, Smith College New Approaches to African History is designed to introduce students to current indings and new ideas in African history. Although each book treats a particular case, and is able to stand alone, the format allows the studies to be used as modules in general courses on African history and world history. The cases represent a wide range of topics. Each volume summarises the state of knowledge on a particular subject for a student who is new to the ield. However, the aim is not simply to present views of the literature; it is also to introduce debates on historiographical or substantive issues and may argue for a particular point of view. The aim of the series is to stimulate debate, to challenge students and general readers. The series is not committed to any particular school of thought. Other books in the series: . . . . .

Africa since , by Frederick Cooper Muslim Societies in African History, by David Robinson Reversing Sail: A History of the African Diaspora, by Michael Gomez The African City: A History, by William Freund Warfare in Independent Africa, by William Reno

Warfare in African History

Richard J. Reid School of Oriental and African Studies University of London

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi, Mexico City Cambridge University Press  Avenue of the Americas, New York, NY -, USA www.cambridge.org Information on this title: www.cambridge.org/ © Richard J. Reid  This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published  Printed in the United States of America A catalog record for this publication is available from the British Library. Library of Congress Cataloging in Publication data Reid, Richard J. (Richard James) Warfare in African history / Richard J. Reid. p. cm. – (New approaches to African history ; ) Includes bibliographical references and index. ISBN ---- (hardback) – ISBN ---- (paperback) . Africa – History, Military. . Military art and science – Africa – History. . Politics and war – Africa – History. . War and society – Africa – History. . War – Economic aspects – Africa – History. . Civil war – Africa – History. . Coups d’état – Africa – History. . Africa – Colonial inluence. I. Title. DT..R  .–dc  ISBN ISBN

---- Hardback ---- Paperback

Cambridge University Press has no responsibility for the persistence or accuracy of URL s for external or third-party Internet Web sites referred to in this publication and does not guarantee that any content on such Web sites is, or will remain, accurate or appropriate.

Contents

LIST OF MAPS PREFACE ACKNOWLEDGEMENTS

 The Contours of Violence: Environment, Economy and Polity in African Warfare  Arms in Africa’s Antiquity: Patterns and Systems of Warfare, to the Early Second Millennium CE  The Military Foundations of State and Society, to circa   Destruction and Construction, circa  to circa 

page viii

ix xiii

   

 Transformations in Violence: Military Revolution and the ‘Long’ Nineteenth Century



 Revolutions Incomplete: The Old and the New in the Modern Era



INDEX



vii

Maps

. Main vegetation zones . Physical Africa . The tsetse ly zone . Arms in antiquity, to c.  . Military states, c. –c.  . Nineteenth-century military revolution . The ‘scramble’ and its aftershocks . Warfare since the mid-twentieth century Maps follow page xiii

viii

Preface

The central role of warfare in human history can hardly be in doubt. Yet the fact remains that the key centres of research into the history of war have in recent decades been clustered, with few exceptions, in Europe and North America, societies in which direct experience of conlict – no matter how fresh in the memory it might be for a dwindling few – is largely lacking. It means that war is something that either happened some time ago, its signiicance hardly questioned but the interpretation of it necessarily abstract, or is happening somewhere else, usually in parts of Africa and (perhaps rather better known, and certainly better reported) central and western Asia. For the peoples of those regions, war is emphatically not abstract: It is something very much here and now, constantly evolving, part and parcel of daily existence, reaching some way into the past and, it would seem, for some distance into the future. This most fundamental global division is no coincidence, for it may well be that the ability to relect historically on organised violence is a luxury – a dividend of peace, perhaps – but it also relects the broad distinction between those who have largely ‘done with’ war, and those who have not. Whereas generations of Europeans have recently grown up only with the celebration or commemoration of conlict, and with a considered narrative of the role of war in their histories close to hand, millions of Africans – the subject of this book – have no such cultural and intellectual equipment at their disposal, yet. The story of war is still unfolding around them, often in the most horriic of ways. This has serious implications. For example, it has meant that from a global perspective, war studies (encompassing the rather less ix

x

PREFACE

fashionable military history) is decidedly Eurocentric as a discipline, and that particular conventions – regulations wired into the narrative, as it were – have been imposed by one part of the world onto everywhere else. The most obvious manifestation of this, perhaps, is in the depiction of particular forms of war as less worthy than others of close examination – including those along the supposedly barbaric, undeveloped frontiers of humanity, many of which, for reasons explored in this book, are believed to be in Africa. It is also true, of course, that for the time being, at least – and it might be suggested that ‘the West’, or ‘the North’, is enjoying only a temporary pax, broken intermittently by terrorist attacks at home – European methods of warfare have proven remarkably successful, globally, and that in war studies, perhaps more than any other discipline, the victors have indeed written the history. Many have been written out of the story – either because their violence is not deemed worthy of classiication as ‘war’, or because they have been ‘defeated’ and consigned to the dimmest corners of public memory. Thus one of the core objectives of the present study is to overturn the misleading imagery which has long been generated around African warfare. Zulu spearmen charging down British guns in southern Africa in  and the ragged members of the Lord’s Resistance Army in northern Uganda in the s may appear to have little in common, but in fact they embody two of the key stereotypes of African warfare, as perpetuated by the Western mainstream. One is the brave, honourable, but doomed ‘primitive’ warrior, the ‘noble savage’, dying heroically but tragically before the inexorable march of modernity; the other is the raddled, bloodthirsty brute, rolling around in the ditches that line the road of progress, his blood-curdling screams reminding everyone else of what happens when order breaks down. The aim of this book is to tell the story of African war beyond these crude but enduring images. Insofar as such a story can be told in such a tight space, the book endeavours to give war its due place in the larger narrative of the continent’s history. The objective is to do so over la longue durée, and to make sense of violence as a force for both construction and destruction – and therefore to better enable readers to place modern conlict in context. Some preliminary points are worth making. The irst concerns the organisation and scope of the book itself. Following an introductory chapter which considers the broad themes germane to the study of African war, the book is structured chronologically, which, after a great

PREFACE

xi

deal of pained deliberation, seemed to offer the best means of exploring African warfare over la longue durée. The divisions of time employed are by no means hard and fast, because no system of periodisation is; but these do indicate distinct episodes comprising particular experiences and processes which need to be understood in turn, in order to fully grasp the evolution of organised violence in Africa and the implications of that evolution for Africa today. Inevitably, within this broad narrative arc, things will have been left out. Further reading will point those interested in the direction of greater detail, but those seeking indepth treatment of a particular topic within the vast canvass presented here may well be disappointed. Thus, while the continent as a whole is the subject of study, there is rather greater focus on Saharan and sub-Saharan Africa than on the northern region, although reference is made to the Mediterranean world where appropriate. Even within sub-Saharan Africa, particular areas inevitably receive more attention than others. In other words, while there is plenty that a volume of this scope and size can do – and it is hoped that this book has – it needs to be acknowledged at the outset that there is rather a lot that it cannot. One of the central theses of this book is that war has more often been constructive than not, and that it has frequently been an expression – if at times an unpalatable one – of political and cultural energy, demonstrating tremendous creativity in social and political affairs. Too often, that energy, that creativity has been misunderstood by outside observers. However we need to be careful here, because sometimes war has emphatically not ‘worked’, and violence has descended into vacuous brutality – both in the deep past, and in more recent times. There is no ‘pristine’ pre-colonial past, in which Africans are always doing the ‘right thing’ and making rational choices: This would make them something more (or, perhaps, less) than human. Mistakes have been made, and central dilemmas have remained unresolved. This needs to be explored, too, particularly in terms of the need to connect warfare with economic development. It is clear, moreover, that the highly racialised interpretations of African war developed by Europeans – who dismissed it as indicative of political, cultural and even biological backwardness – have proved extremely durable. In broad terms, depictions of a savage barbarity which was uniquely African had their beginnings during the era of the Atlantic slave trade, and took on new and ever more paternalistic forms during the nineteenth century, especially during the era of partition from the s onward. Africa was, according to this view, a brutal and inherently violent continent

xii

PREFACE

which needed outside intervention. There are many today who persist in this outlook, if in more subtle ways. It is almost superluous to suggest that this book aims to elucidate the wrongheadedness of the basic interpretation; yet neither is it proposed that we espy instead a ‘merrie Africa’ in the mists of time, a place of innocence and industry and joy in which no violence was done to anyone before sinister foreigners brought trouble to paradise. The best we can say is that Africa has been no more violent than anywhere else, averaged out over the aeons of human history; but it has indeed been an extremely violent place at times, and in many respects, parts of it are so now. The book intends, however, to explain that violence, both at the micro and at the macro levels, and to give it historical depth. Such conlict has often been seminal, but it has been consistently misrepresented. Finally, it is clear enough that this is a contentious topic: Many will disagree with the periodisation and the analysis which follow. But hopefully all will recognise the need to attempt such an analytical and narrative synthesis, even in a preliminary way, in order to more properly understand the role of warfare and the military in African history over la longue durée. In our modern rush to ‘stop’ conlict, perceiving it as uniformly a ‘bad thing’ which must be eradicated, we have too often been guilty of presentism, of historical foreshortening: We have been, in other words, staring down the wrong end of the telescope. We need to begin at the beginning and understand, above all, that most precious of tools in the scholar’s kit, namely historical context.

Acknowledgements

Thanks go to Martin Klein and an anonymous reviewer for their rigorous scrutiny of an earlier draft; Murray Last for numerous conversations on the topic, most of which alerted me to the dangers of attempting this book; Robin Law, for his support and the example provided by his own scholarship; and Robert Bates and James Robinson, who in a different context offered insightful comments on warfare and development. I am grateful to SOAS for granting me the sabbatical year during which this book was completed. Above all, deep gratitude to my wife Anna for holding it all together following the arrival of our daughter May, who appeared in the midst of the violent nineteenth century. The juxtaposition she offered was a matter of both stress and delight. RJR London

xiii

Tropical rainforest High grass–low tree savannah Tall grass savannah Short grass savannah Mediterranean type Desert and sub-desert steppe Montane vegetation Temperate grassland (veldt) 0 0

1000 500

2000 1000

3000 km 1500

MAP  Main vegetation zones.

xiv

2000 miles

Atlas Mts

Tibesti

Hoggar 1 3

1 2

2 6

4

7

3

4

5 5

Rivers 1 2 3 4 5 6 7 8 9 10

Lakes

Senegal Gambia Niger Benue Congo White Nile Blue Nile Zambezi Limpopo Orange

1 2 3 4 5 6

Chad Tana Turkana Victoria Tanganyika Malawi

6

8

9

10

Equatorial forest Land over 1,000 metres 0 0

1000 500

2000 1000

3000 km 1500

2000 miles

MAP  Physical Africa.

xv

Tsetse fly zones

0 0

1000 500

2000 1000

3000 km 1500

MAP  The tsetse ly zone.

xvi

2000 miles

Berber

M

AL

OR

AV

I

DS

ALMOHADS EGYPT ARAB MIGRATIONS Fezzan Garamantes

Sanhaja Berber

Awdaghust

Soninke

KANEMBORNO

GHANA Bambuk

Mande

NUBIA MEROE

Adulis

AXUM Tigray

Bure

Shoa

Zeila Ifat

BANTU DIFFUSION

Fault lines/zones of conflict GHANA Key state systems Soninke Population groups Bure

Key places Early sub-Saharan state formation Towns

0 0

1000 500

2000 1000

3000 km 1500

2000 miles

MAP  Arms in antiquity, to c. .

xvii

MOROCCO

A

ALMOHADS

S ID AV OR LM

MAMLUK EGYPT

1 7

SONGHAY MALI

SOLOMONIC ETHIOPIA

HAUSA STATES

8

OYO IFE

ASANTE

DAHOMEY

SOMALI

NILOTIC MIGRATIONS

OROMO MIGRATIONS

9

BENIN 3

2

4

10

5 4

Fault lines/zones of conflict 1 2 3 4 5 6 7 8 9 10 11

Segu Buganda Bunyoro/Kitara Nkore Rwanda Rozvi Futa Toro Futa Jalon Karamojong Maasai Mutapa Tondibi

1 2 3 4 5 6

KONGO

Sofala Kilwa Mozambique Mombasa Sena Teta

0

1000 500

5 6

GREAT ZIMBABWE

3000 km 1500

2000 miles

MAP  Military states, c. –c. .

xviii

6 11

NGONI

2000 1000

2

OVIMBUNDU LUNDA IMBANGALA

Songhay Mali

0

LUBA

MBUNDU

1

3

Tunis

Algiers 8

Tripoli MOROCCO EGYPT

SANUSI MUSLIM BROTHERHOOD TUAREG MAHDIST 7 BAMBARA SEGU

TE TA RS

LO KO YORUBA TU SAMORI’S STATE

6

5

ASANTE 4

Lagos

Massawa

Khartoum

KANEM

STATE

ETHIOPIA

BORNO

SOKOTO

IBO BUNYORO BUGANDA

DAHOMEY TORO ANKOLE RWANDA TIPPU TIP’S DOMAIN

KONGO 1 2 3 4 5 6 7 8

SOMALI

OROMO

BENIN

Cape Colony (Br.) Boer Republics Portuguese settlement Gold Coast (Br.) Sierra Leone (Br.) Gambia (Br.) Senegal (Fr.) Algeria (Fr.)

Luanda 3

MAASAI NYAMWEZI

Zanzibar

LUBA

MBUNDU

LUNDA IMBANGALA OVIMBUNDU

YAO

BEMBA LOZI SHONA

3

NDEBELE TSWANA

Fault lines/zones of conflict

2 ZULU

Ottoman suzerainty Swahili/Arab movement 1

XHOSA

Major states Diffusion of Ngoni/Zulu military model Egyptian expansion

0

European settlement/control

0

1000 500

2000 1000

3000 km 1500

2000 miles

MAP  Nineteenth-century military revolution.

xix

Fr. Ital.

EGYPT Br. 1880s

ALGERIAN HINTERLAND 1870s–1920s SANUSI 1880s–1930s TUKOLOR 1880s–90s

Fr. ASANTE 1870s, 1900 1

SAMORI 1880s–90s

SOKOTO 1900–03 BENIN 1897

Fr.

Fr.

Br. DAHOMEY 1890s

Eth.

Br.

Ital. Fr. 1890s

TIPPU TIP 1890s

CHILEMBWE REVOLT 1915

Battle of Adwa 1896 Battle of Omdurman 1898 Battle of Isandhlwana 1879 Also uprising, 1900 Also uprising, 1906

OVIMBUNDU 1880s–1900s

Ger.

Br.

Direction of Imperial expansion Wars of conquest, 1870s–90s Anti-colonial insurgency, 1890s–1900s

0

1000 500

2000 1000

3000 km 1500

2000 miles

MAP  The ‘scramble’ and its aftershocks.

xx

YAO 1880s–90s

SHONA/ NDEBELE 1896–7

HERERO 1904–7

Br. 1888–9 Ger. MAJI MAJI 1905–7

HEHE 1890s

Port.

0

SOMALI 1890s

BUNYORO 1890s

1913

2

SOMALI 1890s–1926

OROMO 1890s–90s

1904

Bel.

1

Ital.

MAHDIST SUDAN 1880s–90s

1897–1900

Port.

BOERS 1890s 2 ZULU 1879, 1906

ALGERIA– MOROCCO 1963–64

ALGERIA 1954–62 1990s LIBYA 2011– WESTERN SAHARA 1960s– MAGHREB 2000s–

CHAD 1970s– DARFUR 2003–10

7

ERITREA–ETHIOPIA 1960s–90s 1998–

1

2

SIERRA LEONE, LIBERIA 1990s

SOUTH SUDAN 1960s–2000s

BIAFRA 1967–70

IVORY COAST 2000s

3 8

4 SOMALIA 1990s–

5 CONGO 1990s–2000s

6 KENYA 1950s

KATANGA 1960–64 ANGOLA 1960s–2000s

1 2 3 4 5 6 7 8

Tigray 1970s–90s Oromo 1970s– NAMIBIA 1966–90 Ogaden 1960s– N. Uganda 1980s–2000s W. & C. Uganda 1970s–80s Rwanda–Burundi–E. Congo 1990s–2000s Guinea-Bissau 1960s–90s Ethiopia–Somalia 1963–4, 1977–8, 2006–9 Fault lines/zones of conflict

0 0

1000 500

2000 1000

RHODESIA/ ZIMBABWE 1966–1980

MOZAMBIQUE 1960s–90s

SOUTH AFRICA 1960s–90s

3000 km 1500

2000 miles

MAP  Warfare since the mid-twentieth century.

xxi

CHAPTER 

The Contours of Violence Environment, Economy and Polity in African Warfare

An understanding of warfare in historical perspective – in Africa or anywhere else – can only begin with an appreciation of certain key drivers and dynamics over the longue durée. The roles of environment and economy are central in this respect, and certain social, cultural and political parameters and permutations need to be surveyed before embarking on an exploration of the iner detail. It might be pointed out that some of the broad observations made in what follows represent ‘universal truths’, in that they are by no means peculiar to Africa’s circumstances: So, for example, dense forest and open savannah facilitate markedly distinct forms of combat and military organisation; political authority, in whatever guise, is rarely without a military component, whether covert or overt; all human communities celebrate their own courage and sense of honour on the ield of battle, whatever form the battle itself takes. But it will be clear from the ensuing discussion that there are particular dynamics at work in Africa’s military history – despite the enormous diversity across the continent – which can be, and need to be, highlighted by way of a preliminary overview.

Terrain and Population Particular physical environments and climates have had a direct bearing on forms of organised violence. They have implications for the very nature of militarism – by which is meant the centrality of warfare in society and culture – as well as military structures and strategies 



WARFARE IN AFRICAN HISTORY

of defence and aggression. Two broad features need to be noted from the outset. Firstly, the relative regularity of Africa’s coastline and the infrequency of natural harbours – with the notable exception of the Mediterranean coast – meant the absence, in general, of the armed ocean-going leets so prominent in western European and east Asian military history; only on stretches of navigable rivers, on inland lakes, and along the West African coastal lagoon system was war waged on water. Secondly, the key distinction within the continental landmass itself is between the tropical zone infested with the tsetse ly – very approximately, a band contained between the th degree latitudes on either side of the equator – and the tsetse-free land north and south of it. The tsetse attacks livestock – including cattle, sheep, goats, camels and horses – and spreads trypanosomiasis, which is usually fatal; there is a human variant, known as sleeping sickness. The presence of the ly means that between the Sahara and the Kalahari, horses and other pack animals are almost completely absent, feasible only in isolated zones. Cavalry forces are therefore conined to the Sahel belt and the adjacent Sahara, and more recently the southern tip of the continent; everywhere else, only infantry forces have historically been possible, and war on foot – involving large contingents of porters carrying supplies and women in a range of support roles – predominated across sub-Saharan Africa until the introduction of mechanised transport in the twentieth century. Beyond this fundamental division, several broad physical zones can be identiied. Equatorial forest, covering much of the Atlantic coastline, much of the West African hinterland and the Congo basin, accounts for little more than  per cent of the continent’s geography. Rather more widespread are Africa’s savannahs, grasslands and acacia woodlands, which stretch across the Sudanic belt and into northeast and eastern Africa, while a separate band of woodland covers centralsouthern Africa, encompassing modern-day Angola and Zambia. The continent’s desert expanses segue out of the grasslands – in the north the Sahara, and in the southwest the Namib and Kalahari – and indeed those zones of transition between desert and savannah, especially along the Saharan fringes, have long constituted some of Africa’s most volatile and militarised frontiers. There is also the great highland plateau complex – often referred to as Africa’s ‘spine’ – which stretches from the Ethiopian Highlands in the northeast, through the Great Lakes region of eastern and central Africa, down to the highveld of South Africa. Several riparian systems have shaped Africa’s history in the

THE CONTOURS OF VIOLENCE



most fundamental of ways: The Niger, Benue, Senegal, Congo, Nile and Zambezi rivers are central to the histories of the regions through which they slice. The same is true of the major lacustrine clusters, notably Lake Chad in the western savannah and the lakes of the Great Rift chain, including Turkana, Albert, Victoria, Tanganyika and Malawi. Most of Africa’s enduring centres of population and indeed of state formation lay in close proximity to these sources of water. The central Niger bend, for example, where a rich soil was deposited by the lood waters coming from the Futa Jalon hills to the southwest, supported some of the region’s most extensive military states, stretching between desert and coastal forest. The history of the Nile is well known in this respect, carving a trail of volcanic soil through some of the most arid landscapes on the planet. Interlacustrine eastern Africa, too – especially the cradle land encompassing southern and western Uganda, Rwanda, Burundi and northwest Tanzania – was well watered, in terms of both precipitation and proximity to the lakes themselves, and was a region blessed with a rich earth which supported more people than in the equatorial forests to the west or the grasslands to the south and north. Beyond these zones of fertility, soils were thinner – or foliage denser – population was spread more thinly, and the struggle for water was more intense. Vulnerability to signiicant variations in rainfall, moreover, has long been a key driver of conlict, and correlations can often be traced over the longue durée between environmental shifts and levels of violence – as demonstrated across the Sahel and northeast Africa in the late twentieth century. The deining dynamic in a great deal of African war – in antiquity as in more recent times – has been the quest for proximity to water. This was especially true in areas with fragile ecosystems; in these places, war itself has been a cause of deprivation, removing people, whether agriculturist or pastoralist, from their natural sources of sustenance. Compelled to eke out a living in harsh frontier lands, such displaced communities have then clashed with those already there – as there almost always were. Thus the cycle of militarisation has been perpetuated, whereby war has been made central to social and political organisation, as well as cultural celebration. Elsewhere, war has been one of the drivers of urbanisation, especially in the western half of the continent; in some areas, the roots of urbanisation lie in itinerant royal camps comprising armies and their enormous entourages, whereas in other circumstances, communities have come together in fortiied urban clusters for protection,



WARFARE IN AFRICAN HISTORY

a phenomenon associated particularly with the nineteenth century – among the Yoruba, for example, or the Nyamwezi. It should be noted, however, that violent conlict has just as often stymied fragile urban growth, causing the dispersal of people away from turbulent areas and abandoning nascent settlements to the bush. Dificult and diverse terrain meant both distinctive local variations and innovations, as well as limitations on the scope and range of war in those areas. The open spaces of desert and grassland allowed for relatively rapid movement – particularly outside the tsetse zone, where horses were deployed – and military operations could develop over large areas; the problem facing leaders, however, was control of those operations and the maintenance of logistical support for – and linkage to – the mounted outriders of expansionist projects. Conversely, movement in the equatorial region could be ponderous and dificult, especially during the wet season, and within the tsetse zone more generally, the deployment of men and material without draft animals or wheeled transport meant limitations on the range and the type of military operations. Communities large and small tailored their military structures and their tactics accordingly. In thick forest, where manoeuvrability was restricted, useable battlegrounds were comparatively limited and were probably carefully chosen, becoming well established in the process; at the same time, ambush along forest paths was especially important in tactical terms. In fertile, relatively densely populated regions, military hierarchies made use of vertical chains of recruitment, command and deployment; across more thinly peopled regions, horizontal systems of organisation were often more effective, allowing as they did the mobilisation of – and operation across – larger areas, such as the age-set systems favoured by dispersed pastoral groups and segmentary societies. Throughout Africa’s history, however, logistical challenges, stemming from vast and dificult terrain, have persisted, and what has been described as a ‘skulking way of war’ or ‘raiding war’ has often resulted: This refers to cycles of assaults on the enemy – particularly on their human, animal and material resources – as well as the more familiar (to Western audiences) ‘campaigning war’ involving prolonged, largerscale operations and set-piece battles. Although discrete ‘episodes’ in wars of both types might be relatively brief – a matter of days, for example – it is clear enough that many such episodes – that is, individual raids or particular campaigns – constituted wars of considerable

THE CONTOURS OF VIOLENCE



duration, sometimes over a number of years. In many respects, the highly mobile, skirmishing style of combat has endured into the modern era. Ultimately, then, much organised violence in Africa’s history was concerned with the struggle to domesticate often harsh, and certainly fragile, environments, and this needs to be understood alongside much of the continent’s historical underpopulation. One of the key vectors in Africa’s military past has been the drive to maximise and control productive and reproductive labour; coercion has often been required to control people in a relatively land-rich continent, and thus slavery and polygamy have long been underpinned by cyclical military operations, seen as critical to the growth and indeed very survival of the community. The continual formulation and reinvention of unifying ideologies, moreover, has long been used to justify warfare, as well as in the attempt to prevent the perpetual process of ission which has characterised much of Africa’s history. The armed frontier is continually reproducing itself as groups split from the metropolitan community for economic or political purposes, migrate to adjacent territory, and fuel regional rivalries over local resources. Naturally, this has been especially true in areas where population levels have increased somewhat over time – for example, as a result of inward migration. Relatively land-rich these areas may still have been, but a shift in the land-to-people ratio means that frontiers become ‘closed’ – that is, communities press in on them from all sides – rather than ‘open’, in other words offering opportunities for continued expansion into empty or sparsely peopled land. In these cases, competition over ever more restricted space, and conlict within and across local ‘closed’ frontiers, is increasingly violent. Nonetheless, in general, it remains the case that people have been the focus of much warfare rather than the annexation of territory per se, for land is plentiful and easily devalued, although there are a number of important exceptions in which territorial expansion has itself been a core objective. Again, however, as earlier instances of local population growth – and in particular the population boom in Africa since the mid-twentieth century – have demonstrated, more people does not necessarily mean less war, indeed often quite the opposite: In recent decades, especially, economic failure has rendered population growth disastrous, and a new dynamic has emerged whereby there are too few people with valued and fulilling lives.



WARFARE IN AFRICAN HISTORY

War and Economic Development These geographical and demographic dynamics have had a direct bearing on the continent’s economic systems. Different forms of violence were supported by different economies, which either grew or contracted as a result of war over the long term. The relationship between war and economic development is complex and multi-faceted: A range of factor endowments – arable land, pasture, minerals, water, livestock, people and skills – both drove and placed constraints on warfare in Africa’s history. Centralised military states appeared in areas with fertile soil and reliable rainfall, and were thus able to use agricultural surpluses to maintain armies in the ield for longer, and over a larger area – although, of course, soldiers routinely supplemented their own supplies by feeding off enemy land, and often were wholly reliant on being able to do so. Strong sedentary economies – for example, in the Great Lakes, the Ethiopian Highlands, the West African forest and parts of the western savannah close to the Niger – thus facilitated expansionism. Elsewhere, however, the demands of pastoralism, combined with the lower yields of subsistence agriculture in the grasslands and drier savannahs, placed limits on the range and type of operations. Only short-term mobilisation was possible, and the outcome was the distinctive pastoral militarism – organised around age-sets – whereby young men (among the Fulani, the Oromo, the Maasai, and the Ngoni) were both warriors and herdsmen. In either case, however, terrain and environment remained inhibitive, especially south of the Sahara, again, where most African armies were infantries with no access to pack animals, and the seasons dictated military activity. Movement during particular seasons was extremely dificult, and impossible over large distances: In particular, as noted earlier, the wet season presented problems for soldiers on the move and was usually avoided completely. Some societies were able to take advantage of waterways – rivers and lakes, usually, rather than oceanic coasts – but this was exceptional rather than normative. More generally, military operations were dictated by agricultural calendars and the success of harvests; armies marched as far, and stayed on campaign for as long, as their duties in the ields allowed them, or as their extant food supplies permitted. For these elements, societies could lay plans, but lack of rainfall, the failure of the harvest, or the outbreak of livestock disease could either wreck military ambitions or push communities thus deprived to violence in search of basic resources. Again, peoples were

THE CONTOURS OF VIOLENCE



often forced to move to less fertile districts, which in turn perpetuated economic instability and militarised frontier zones. A great deal of African conlict involved total economic war, the objective of which was the seizure of key resources or their destruction in enemy territory. Thus, ‘raiding war’ could be enormously destructive from an economic and social point of view, involving the capture of livestock, foodstuffs and, above all, people. Starvation and impoverishment often were political processes as well as biological and economic ones. The defeat of a particular province or chiefdom was usually followed by episodic professions of loyalty, underpinned by the regular payment of tribute in goods and people. Such tribute, importantly, might be redistributed among loyal oficers of the state, who might also expect to be appointed to lucrative positions of administrative authority over newly conquered territory. Violence, then, was often critical from an economic point of view, especially where trade was dificult or insuficient. War, again, was used to underpin local systems of slavery; women and children were prime targets. Population growth, and even mere demographic stability, often depended on successful campaigns of people-gathering, and conlict frequently occurred over technical skills and economic specialisation. Wars over people clearly intensiied with the growth of the external slave trades – notably the Atlantic system between the sixteenth and nineteenth centuries, and in eastern Africa during the nineteenth century – which unquestionably involved new levels of organised violence, economic in objective; warfare escalated as the export of people became more lucrative, while expanding states likewise needed slaves to maintain levels of domestic economic growth. Nonetheless, it represented an extension of African economic war, and in particular of the long-standing practice of waging war for the capture and incorporation of people. Even so, the Atlantic system signiied the growing importance of long-distance commerce, the drive for control of which had been a factor in African warfare since antiquity – particularly in the northern half of the continent, where the lucrative trans-Saharan trade network rewarded the ambitious and well-organised military states of the savannah, or where Red Sea trade provided a crucial commercial outlet for the restless Axumite state. But with the expansion of the global economy from the middle of the second millennium, and the relative decline it signiied for Africa, commerce became ever more critical to the latter. Wars were increasingly about exports and imports; opportunities for rapid commercial gain presented themselves – not without



WARFARE IN AFRICAN HISTORY

considerable risk – to armed entrepreneurs, and states and societies had to weigh up the costs of war alongside the commercial beneits that might result. Many miscalculated; the ‘golden goose syndrome’ was much in evidence, as armed polities sought to take advantage of the new opportunities, only to ind that the violence involved in the endeavour destroyed trade itself. This remained a dilemma into the twentieth century. Still, many lourished, especially those for whom this was not solely a matter of responding to external demand for certain commodities, but for whom the construction of political and cultural ediices was just as important. There has certainly long been evidence of both a ‘political’ model – by which we mean that over the long term, African war was the result of endogenous development and stimuli, such as local processes of state formation – and an ‘economic’ model – in other words, that ultimately violence in Africa has been stimulated by exogenous commercial pressures, from the slave trades onwards. States have often resulted from such activity, but just as often they have not, largely because states represented unnecessary encumbrances on the relatively ‘straightforward’ business of raidingand-trading. Either way, there can be little doubt that in more recent times, and certainly since the nineteenth century, a great deal of violence has been entrepreneurial, and indeed developmental in aim if not in outcome, as African communities seek to access some of the wealth they perceive to be moving around a global economy within which they are marginal. One sphere where overseas commerce has been especially critical, of course, is in terms of imported weaponry. There existed almost limitless local variations on the basic technology available to African societies – spears, swords, bows and arrows – and innovation and reinement were continual, particularly with the spread of ironworking. Economic specialisation was central to military success, and the quest for new materials – whether iron ore from which to fashion spearheads, or particular types of wood for weapons or canoes – was often behind military adventures. But some of the most radical innovations came as a result of imports, from horses in the early centuries of the irst millennium to irearms and other equipment from the sixteenth century onwards. Entrepreneurs needed to export in order to acquire not only the commodities the distribution of which secured the continued loyalty of their followers, but the arms and ammunition needed to maintain military success. The pioneers of expanding military states were often not soldiers but traders in distant commercial centres,

THE CONTOURS OF VIOLENCE



negotiating for irearms and bullets; and by the nineteenth century, in some places at least, the efforts of merchants were supplemented by the skills of craftsmen able to repair guns and make ammunition. Since the middle of the twentieth century, the business of arms trading and local adaptation has continued as before, albeit on a greatly enlarged scale. War often has indeed facilitated economic expansion, sometimes dramatically so, and indeed economic growth itself has in the past facilitated ever more ambitious military operations. But again, environmental and demographic constraints have frequently placed limits on the nature of this growth. At the same time, a great deal of violence in Africa’s past may be seen to have been inimical to economic development – and it is another of our ‘universal truths’ that sometimes war is indeed just that. Yet it has rarely been economically irrational, even where it has not achieved the kind of stability and growth witnessed beyond the continent. Indeed, in the modern era, armed conlict is more frequently a symptom, not the cause, of economic instability.

Questions of Culture and Society War is a social and cultural phenomenon, relecting – often in the starkest of terms – a community’s values and sense of self. War and military prowess permitted, in the most unambiguous of ways, the display of courage and the winning of honour – achievements celebrated by all societies, even if they differed in the detail of how these might be realised. Success in war brought social standing and respect; it also often brought material beneit and political advancement. Almost everywhere, however, ideas about participation in war, and thus about access to honour, were deeply gendered, for it was men who usually did the ighting and who designed the military systems and the cultures that celebrated them. In truth, it is dificult to avoid analyses of war becoming oppressively androcentric affairs. Courage was almost always masculine, and cowardice feminine; for young males, war was a means of attaining full manhood. However, women played a myriad of roles: They were logistical support; they maintained economies when not travelling with the soldiers; sometimes, indeed, they too fought, and led. Women certainly provided much of the narrative, and critique, both during the battle and when the dust settled. But women were invariably targets, as well as rewards by which success



WARFARE IN AFRICAN HISTORY

was measured in the aftermath of battle. The sources for before the twentieth century are problematic, being produced by men and for men; but unquestionably further research is needed on women’s roles in other spheres – in local recruitment, for example, and in inluencing military strategies. In certain ‘warrior societies’, or communities which had experienced high levels of male casualties, larger numbers of women than of men awarded polygamy and the control of women a matter of particular urgency; yet such circumstances must have afforded women opportunities for considerable leverage. We know enough already to suggest that they did indeed play such roles, even if the detail is still lacking. Certainly, both men and women invested in the cultural and ethnic identities which were developed around war and violent struggle and sacriice; stories of heroism and martial prowess were woven into national narratives which themselves became – among other things – celebrations of the worthier characteristics of a particular community, and essential guidelines for would-be leaders. Very often, these cultures of militarism emerged in response to the perennial need to maximise human resources and to create wider unity as a defence against hostile neighbours. Militarism was not – or not solely – about bloodthirsty smiting and the thrill of killing: It was a celebration of the community’s achievements, and tapped into the social needs of cohesion, camaraderie, self-reliance, and courage in the face of danger. And war was continually historicised, too: The past was very much present in the organisation of violence, even during the revolutionary upheaval of the nineteenth century and the self-consciously modernist struggles of the mid-twentieth century. States and societies sought historical precedent and heroic forebears, for both inspiration and solace, and constructed both oral and written narratives – although very rarely physical monuments – for the purpose. Such celebrations and traditions have rarely lacked a spiritual dimension, and indeed the role of religion and spirituality is frequently central. Much war was waged with spiritual sanction and according to belief systems which periodically demanded violence, not least to appease ancestral spirits. Ghosts attended the battle, and the dead proffered or withdrew their blessing; diviners were critical in providing sanction and even tactical guidance, while particular gods – specialising in violence – were routinely consulted in advance of the campaign. Ultimately, only spirits could keep men safe and only through spiritual observance could post-war healing take place. There was much local

THE CONTOURS OF VIOLENCE



variation in this respect, but the global faiths were present, too. Across the western savannah, Islam provided a framework for righteous violence from at least the turn of the second millennium onwards, while jihad aimed at the restoration of true Islam was periodically invoked. In northeast Africa, Muslim-Christian relations were likewise intermittently violent, and Christian Ethiopia’s expansionist militarism was at least in part anti-Islamic in provenance. On both sides, but perhaps especially for Solomonic Ethiopia, these wars were part of a grand narrative of spiritual struggle, sanctioned by God. More generally, religious symbolism and ideology were especially important in uncertain and troubled times: witness the advance of Islam during the nineteenth century, and local spirituality in some areas in the late nineteenth and early twentieth centuries. It is not generally dificult to measure, or at least acquire an appreciation of, a particular community’s sense of its own sanctity. It is less straightforward to measure acceptable behaviours in war, and changes in these: Speciically, in dealing with levels of killing, and degrees of ‘cruelty’ and ‘barbarity’, analyses can date rapidly, and are value-laden. Sufice it to remind ourselves here that all war is brutal. In the African context, normative behaviour did exist as a concept, but it was mobile, according to context. It is also the case more generally that distinctions were not drawn between combatants and noncombatants; that was not the nature of African warfare, which, as we have noted, was concerned with assaults on the enemy’s very social and economic fabric, involving attacks on women and children – and their capture – as well as against fellow men-at-arms. This escalated in those regions affected by external trades in slaves. Terrorising tactics were common, too, and that included the mutilation and killing of the most vulnerable members of society. The destruction of crops and livestock – whether this was the primary objective of an attack, or a by-product – frequently led to many deaths from starvation and destitution. This was clear enough during the ‘scramble’ for Africa, but it had long been common in African warfare. As for levels of actual killing, generalisation is impossible, beyond the observation that whereas comparatively bloody wars with large numbers of casualties, by modern standards, did occur, they were probably rather rare before the nineteenth century, and certainly before the mass killing of the twentieth. A great deal of pre-colonial African war did not involve large numbers of deaths – sometimes, indeed, hardly any – and this was especially true in underpopulated regions, although it is worth



WARFARE IN AFRICAN HISTORY

adding the obvious point that whether a casualty rate is ‘high’ or ‘low’ is indeed entirely dependent on population levels. In a small, compact chiefdom, a dozen men killed is more devastating than the loss of a thousand in a sprawling polity with a large pool of manpower. Finally, in terms of a preference for ‘peace’ over ‘war’, the risk of anachronism is again very real. Everywhere, ‘peace’ is a relatively modern invention. The modern ixation with it may be understandable, given the traumas of the twentieth century, but in fact it has tended to obscure our understanding of ‘war’, and certainly a more nuanced appreciation of it; it is not always entirely clear, moreover, what is actually meant by ‘peace’, beyond particular groups ‘not ighting’. The history of African warfare demonstrates that periods of rest, or armistice, or resolution, were never taken for granted – nor were they always particularly welcome, because war was economically, politically and socially important. Nonetheless, all societies had mechanisms for resolution, or a suspension of hostilities; it is in the nature of human affairs to practice diplomacy, even while the battle continues. It may well prove proitable to resurrect these older, organic processes of resolution, diplomacy and moderation; as it is, the current ‘peace industry’ in Africa is manifest in the corporate multi-nationalism of the African Union and its regional subsidiaries, invariably pressured (and funded) by an international community possessed of a self-appointed mandate to bring order to the troubled non-Western world – usually with inancial incentives attached. The motives behind, the meanings attached to and the cost of such ‘peace initiatives’ perhaps need to be examined more closely.

Fortiied Centres and Mobile Frontiers: War and Polity The challenge of understanding war in a political context is formidable, given the enormous diversity of polity across Africa, temporally and spatially. The crudest of analytical dissections involves the separation of ‘state’ from ‘non-state’ or ‘segmentary’ forms: the former characterised by centralised structures, hierarchy and some degree of territoriality, founded on sedentary agriculture; the latter by decentralised political authority which was essentially gerontocratic and collective, wielded via councils of elders whose directives were implemented by junior age-sets, and often – owing to networks of clan alliances and secret associations – over wide areas. These were predominantly,

THE CONTOURS OF VIOLENCE



albeit by no means exclusively, pastoralist economies. As we have noted, these two broad systems waged war very differently, relecting the exigencies of economy and environment; often, indeed, they competed violently with one another at their respective peripheries. But the key point here is that everywhere, political authority incorporated military command in one form or another, and political culture was frequently militarised. In the early chapters of this book, we will see that this was the case from antiquity; but the militarised political culture becomes even more prevalent in the nineteenth and twentieth centuries, the outcome of a revolution in military affairs – a transformation in the organisation and scale of warfare – with which Africa is still grappling. In particular, military culture, and war itself, was frequently the result of the proliferation of armed, mobile frontiers which witnessed the production and reproduction of polities over the long term. This phenomenon needs to be understood in the context of the process of ission alluded to earlier. Communities continually split and reformed in a land-rich and underpopulated environment, and obviously enough levels of violence generally rose where there were more communities in direct competition with one another: migratory exoduses broke away and sought to claim new land, which meant conlict both with societies they encountered and often with those they sought to leave behind. On these frontiers, war represented part of a creative political vortex; organised violence and attendant military cultures were central to the struggle for control of people and access to resources. Wars were thus often intensely local affairs, waged between neighbouring groups within and across frontier zones which became economic and ideological, as well as physical. This has remained one of the core themes of Africa’s history. It brought about a great deal of political and economic creativity – the dynamic military cultures which resulted were aimed at both internal unity and external threats – but in many respects it also meant inhibitions on ‘growth’, for political instability was frequently endemic as a consequence. Larger, more ambitious projects aimed at stability are much in evidence: notably, the armed and mobile frontier has frequently sought to capture ‘the centre’, however the latter is deined – as a political hub, or a cluster of material resources, or a particularly important cultural or spiritual core (or indeed a set of ideas which ‘inhabit’ a physical location). But in the process, new frontiers were created and new (or reinvented) military cultures appeared on them. This has shaped Africa’s political



WARFARE IN AFRICAN HISTORY

systems since early in the second millennium in particular; was a central dynamic in the nineteenth-century military revolution; and can be discerned behind much of Africa’s present-day political volatility. The issiparous and luid nature of African polities and institutions has meant a prominent role for the armed adventurer, or more commonly groups of them, with cultures of militarism frequently developing around charismatic leadership; sometimes these igures have gone on to establish their own sedentary polities, in which political leadership was thus founded in military endeavour and couched in the rhetoric of military greatness. Military leadership thus often morphed into political power, and military cultures underpinned the political order. Armies were rarely ‘apolitical’; politics and military command were closely interlinked, and successful rulers needed to have proved themselves in battle – or, more commonly, to surround themselves with enough people to say they had. This was especially important in older, better established states in which new rulers rarely, if ever, saw actual combat. Success in war, in sum, demonstrated not only manly attributes, but also the successful leadership of other men, and thus loyalty and cohesion, key pillars of the stable polity. Later, this was demonstrably the case during the ‘scramble’ for Africa, and was true of the era of European rule which followed it; certainly, the colonial order was in many ways founded on an alliance of African and European armed adventurers. Even though the political nature of colonial armies only really became clear after decolonisation, this was very much a long-term dynamic in African history. Moreover, the more prosaic struggle for control of resources – productive and reproductive – gave rise to, and doubtless was often driven by, more abstract and intellectual ideas about identity, territory and history, and underpinned by cultures of militarism and violence which framed the righteousness of war in both spatial and temporal terms. Sometimes wars were wrong, usually started by bad rulers; but the good ones were duly celebrated and rendered righteous, and memorialised at the centre of the polity. This was cyclical war which was shaped by relations between shifting frontier and centre, with cultures developed to justify actions and unify communities. But the scale and reach of such war was limited by the mobility of the frontier itself, and entrepreneurs of violence needed, ultimately, to exercise caution: In that sense, arguably, the most ‘successful’ – that is, the most enduring and cohesive – states and societies were those that remained compact, territorially but also culturally and ideologically, for to stretch beyond

THE CONTOURS OF VIOLENCE



natural boundaries was to risk unleashing and ultimately losing control of frontiers, which then threatened to overwhelm the very heart of the polity. Africa’s history is replete, then, with rebels, bandits and irreconcilables, and the launch of armed challenges against established orders. Sometimes, these became the established order in their turn; the analytical category deployed in contemporary sources – whether criminal banditry or righteous insurgency – usually depends on how far the armed frontier has been able to march toward the centre, metaphorically or physically. This is as true for the continent’s deeper precolonial past as it is for the post-colonial era, with its anti-colonial revolts, revolutionary guerrilla insurgencies, and diamond-smuggling warlords. Guerrillas often do become the government, but more often they do not, and remain on their stalled frontiers, brimming with violent frustration. Such militarisation often occurred in the residue of failed state-building exercises, the product of the ebb and low of political tides. The ‘view from the bush’, as it were, is one of the more revealing angles from which to view Africa’s violent past: The symbiotic dynamism between fortiied centralism and the creative, centrifugal frontier is a tension which has driven much construction as well as destruction in Africa’s history, not least in the nineteenth and twentieth centuries. Given the restraints on their power and reach, leaders developed various strategies of military recruitment and control. The age-regiment system instituted by ‘stateless’ societies, which mobilised entire age groups for activities requiring youthful vigour and aggression – war included – was one of the most effective. For sedentary, hierarchical states, the issue was often more complicated, for centralising elites had to trust the task to local administrators, and often had to outsource military service altogether. In the broadest terms, relatively few African societies had standing, professional armies, although a number did, especially by the nineteenth century: in other words, fulltime dedicated soldiers, sustained by either campaign booty or land and funds bestowed by their leaders (or both), and developing selfsustaining martial cultures which celebrated honour, courage, skillsat-war, and esprit de corps. However, through much of Africa’s history, most armies comprised part-time warriors. In centralised states, the general practice was to devolve responsibility for recruitment to provincial chiefs – themselves drawn from a political elite associated with or related to the royal court – who thus supplied men on demand.



WARFARE IN AFRICAN HISTORY

These soldiers would then join a core of troops which remained under the direct command of the ruler himself. In this way did ‘national’ armies relect the strengths – and the population levels – of the various districts, as well as the prowess of local chiefs, whose success was measured not in the extent of the land under their jurisdiction, but in the number of men they could contribute to the cause, whether through persuasion or coercion. This model – with some similarities to the feudal pattern of organisation in Europe – was widespread by the middle of the second millennium, and can be discerned in a range of states from Asante to Buganda to Ethiopia. Such states also made use of mercenaries and slaves, the latter serving as both soldiers and porters; and the practice of recruiting from the frontiers, and harnessing the prowess of ‘hardy’ borderland peoples, was common long before it became standard among European colonial regimes. Again, the vast bulk of sub-Saharan armies were infantry forces; only in the northern half of the continent, with the development of cavalry-based armies, did a costly class of professional mounted warriors emerge which spent most of its time engaged in war, which was serviced by a class of retainers, and which in many cases wielded considerable political power. Control of this wilful rank of warriors was a challenge for all would-be state builders across the savannah and desert edges, and reliance on them both rendered power a precarious business and necessarily meant a strong military element to governance. The warrior horseman may have represented a clearly elevated social status, but even among infantry states further south, successful warrior chiefs often played a similar king-making role, albeit without the expensive skills associated with cavalry. Soldiery may have been a part-time business, but cultures of professionalism still emerged, and considerable social status could still be achieved through military activity. It was, indeed, one of the surest routes to social advancement, as it would be during the colonial era and beyond, too. Successful societies harnessed youthful male aggression to good effect; but social crisis led to young men seeking leaders to follow, expectant of booty and adventure. The professionalisation of war and military organisation – involving the dedication of an ever larger share of socio-economic and political resources to waging war and celebrating it – increased dramatically during the nineteenth century, and was a core element in the transformation of violence across Africa during that era. Skills became more expensive, too, among those who mastered the increasingly

THE CONTOURS OF VIOLENCE



ubiquitous irearm. During the ‘scramble’ for Africa and its aftermath, indeed, colonial recruitment oficers were able to take advantage of heightened levels of military professionalism, as well as of social and political conlict, in building the armies that underpinned the colonial order. Moreover, young men – at least those not forcibly conscripted – espied considerable advantages in joining up. Men-at-arms have been despised and admired in equal measure in African history, but the idea of soldiering – in its loosest sense – as aspirational has persisted into the modern era, in some places at least, however brutal those aspirations may seem to the outsider. But so too have the challenges confronting leaders who would arm the young and harness their aggression for political and economic ends. The African military maintains a powerful presence in the polity; and although there are a number of novel elements to this, the products of the continent’s recent past, it represents the replication of a pattern of some considerable antiquity. Further reading Goody, J. Tradition, Technology and the State in Africa (London: Oxford University Press, ). Iliffe, J. Africans:The History of a Continent (Cambridge: Cambridge University Press, nd ed., ). Kopytoff , I. (ed.) The African Frontier: The Reproduction of Traditional African Societies (Bloomington: Indiana University Press, ). Lamphear, J. (ed.) African Military History (Aldershot: Ashgate, ). Mazrui, A. A. (ed.) The Warrior Tradition in Modern Africa (Leiden: E. J. Brill, ). McCann, J. Green Land, Brown Land, Black Land: An Environmental History of Africa, – (Oxford: James Currey, ). Reid, R. J. A History of Modern Africa:  to the Present (Oxford: WileyBlackwell, ). Uzoigwe, G. N. ‘Pre-Colonial Military Studies in Africa’, Journal of Modern African Studies, : ().

CHAPTER 

Arms in Africa’s Antiquity Patterns and Systems of Warfare, to the Early Second Millennium CE

It is no easy task to assess warfare and military organisation in Africa’s antiquity. A serious lack of sources – documentary, material or oral – across much of the continent means that a comprehensive survey of processes and structures in the deep past is impossible; certainly, little is known about warfare across much of sub-Saharan Africa before the early centuries of the second millennium. However, it is possible to offer some speculative remarks about that vast region in the context of the spread of Bantu languages: The available evidence concerning economic development and socio-political change can be combined with intelligent guesswork to assess the military dimensions of this early period. More concrete reconstructions are possible further north, across the western savannah, where the irst millennium CE witnessed a succession of dynamic and expansionist polities which depended on a combination of agricultural productivity, long-distance trade and armed force. Likewise, the range of sources for the Ethiopian Highlands and the upper Nile valley are comparatively rich and permit a closer examination of ancient African militarism. Of course, both western savannah and northeast riparian and highland polities were the products of their particular physical, climatic and demographic environments, and therefore the extent to which they can be used to speculate more broadly about the nature of early African warfare is limited. They do, however, demonstrate certain themes which are germane to other regions of the continent.



ARMS IN AFRICA ’S ANTIQUITY



Cavalry Empires of the Savannah We have already observed that the presence or the absence of the horse is the irst principle in distinguishing the development and nature of warfare in northern and southern Africa, respectively. Whereas the ancient Egyptians had long used horses in war, their spread southward across the desert took rather longer, and their adoption in the Maghreb is associated more with the expansion of the Roman Empire in the centuries immediately preceding and immediately following the beginning of the Common Era than with any diffusion westward from the Nile valley. Nonetheless, the Berbers (or Gaetuli, in the ancient sources) adopted horses in the course of the irst millennium BCE, and from the Mediterranean coastal zone, the horse drifted gradually southward, transforming political structures and warfare in the course of the irst millennium CE. The Garamantes of the Fezzan were using horses by the irst centuries CE, and to great effect, for they straddled a vast area across the central Sahara, clashing with the Romans to the north – with whom they eventually established a mutually advantageous commercial relationship – and raiding in the area of Lake Chad. Indeed, many of the slaves they supplied to the Romans in exchange for Mediterranean goods (and ideas) were captured in the southern fringes of the central savannah. The Garamantes practised agriculture – they developed complex irrigation systems to support it – but their use of horses gave them an aggressive and proitable mobility which was replicated by a number of peoples across the desert and savannah to the south in the later centuries of the irst millennium CE. By the seventh and eighth centuries, the pastoral polity known as Kanem, a sprawling confederation ruled by the Zaghawa close to Lake Chad, was supplying slaves to the Islamic north; in exchange for slaves they acquired horses in order to expand the reach of their slave-raiding parties. By the twelfth century, Kanem may have had a cavalry force some , strong. Borno, which neighboured and in time succeeded Kanem, likewise traded slaves for mounts. Horses, thus, had reached the southern edge of the Sahara and the northern savannah zone by the end of the irst millennium CE, and they were instrumental in the state-building projects which characterised the western savannah for close to a thousand years, between c.  and c. . Nonetheless, the creation and the maintenance of effective cavalry was neither a straightforward nor a



WARFARE IN AFRICAN HISTORY

uniform process. Horses survived well enough in the northern savannah to make cavalry possible – if enormously expensive, as it was elsewhere – and Saharan nomads made highly effective use of both horses and camels as mounts. However, there were problems with equine health a little further south. The faster and larger mounts had to be imported from the desert, and these did not reproduce well in the southern savannah. Local animals were smaller and weaker, which meant that the superior breeds had to be continually imported if societies were to maintain their cavalries. Horses therefore became associated, as they did in Europe, with elite status, owing chiely to their expense and the skills needed to use them effectively. The military role of horses therefore drove not only state formation, but also social stratiication. Cavalry units tended to represent privilege, drawn from or closely associated with ruling elites. Horses did not always make for successful state building – quite the reverse, sometimes. In many Saharan cultures, to be a man was to be a horseman and a warrior, which made for a potent military ethos; but centralised structures and codes of discipline were frequently absent, and bands of horsemen were rarely more than that – loose associations of men with temporarily aligned interests. Such mobile organisations were rendered volatile and brittle by a iercely individualistic military culture which made large-scale action dificult, if not impossible, and certainly rare. Nomadic horsemen might, at times, unite under messianic leadership, particularly as Islam iltered from the coastal Maghreb southward: the Almoravids in the southwest Sahara in the second half of the eleventh century offer such an instance. But all would-be leaders and state builders across the Sahara and the savannah struggled to impose unity and cohesion over far-lung and highly mobile populations. Centrifugal forces and individualistic military cultures would inhibit the growth of permanent states across the region, as they would elsewhere in Africa. It meant that scattered amongst centralised polities, in the vital frontiers between expansionist and extractive imperialisms, lay a host of smaller-scale communities which were not ‘states’ in the conventional sense, but organised themselves according to generation and collective responsibility, or which lived only as long as individual, charismatic leaders did. These were as much a critical part of the military environment in the savannah as were the great cavalry empires themselves. Meanwhile, across the savannah, relations between agriculturalists and pastoralists were alternately peaceful and violent, involving trade and war; they fought

ARMS IN AFRICA ’S ANTIQUITY



over land, but cattle keepers supplied farmers with milk, meat and hides, for which farmers exchanged grain, salt and iron. The great states arose on the back of both the cyclical conlict between farmers and herdsmen and the vital commerce which existed between them. The pastoral-agricultural dynamic made for highly mobile, volatile frontiers of interaction, and extremely creative ones; it is one of the great overarching themes, of course, in world history. Large states did come; many of them came and then went, but they were remarkable while they lasted, dominating the savannah belt for a millennium. In the tsetse zone of sub-Saharan Africa, the high costs of equine military organisation experienced by Eurasia over several thousand years were generally avoided, as seen in the later discussion; this key dynamic – or the absence of it – may be regarded as absolutely critical in terms of Africa’s long-term development because it meant that the incentive to raise capital to fund high-cost militarism was lacking. After all, in much of Europe and Asia, it was the need to meet those costs which drove state formation over the long term. However, it is clear that in the case of the west African savannah, a similar pattern to Eurasia is discernible in that the states looping around the Niger River also had to raise funds to meet the costs of equine militarism – a situation further complicated by the fact that the good horses, and certainly those most effective for use in cavalry forces, were imported from the desert, thus making for a high-cost long-term investment on the part of political and commercial elites. Cavalry cost money, and between Mauritania and east of Lake Chad, those costs escalated as states raided sedentary communities which in turn armed themselves and developed new methods of defence. But continued slave-raiding campaigns against those communities were vital, as slaves were exported in return for the horses so necessary to maintain regional hegemony; it was a self-sustaining cycle, with slaving wars both the cause of escalating militarism and the product of commercial dynamics. Over the longer term, political and economic elites became tied into a cycle of dependency on expensive foreign imports – neither local resources nor sources of revenue could cover the mounting costs – in the attempt to build sustainable polities. Between the middle of the irst millennium and the middle of the second millennium CE, then, a succession of sprawling empires appeared across the savannah, based on the effective – but costly – organisation of violence. Detail on weaponry and tactics is extremely thin, and often absent altogether; in general terms, however, swords were commonly used by



WARFARE IN AFRICAN HISTORY

both infantry and cavalry for close ighting; and an array of spears – whether for thrusting or for throwing – were likewise probably used by horsemen and foot soldiers alike. Bows were generally an infantry weapon, but later evidence suggests that in some areas – Borno, for example – archers were mounted. Shields were carried by both infantry and cavalry, and the latter also wore protective quilting. As for numbers, these are impossible to come by for this early period, although it seems reasonable to speculate that the enlargement of military scale meant a shift from campaigns involving a few dozen horsemen and a couple of hundred foot soldiers to those comprising several hundred horsemen and perhaps several thousand infantry. The skills and technologies associated with good horsemanship, meanwhile, developed piecemeal: Stirrups were introduced only gradually, whereas saddles and harnesses were unknown in some areas for several centuries. Riders frequently rode bareback, sometimes cutting their mounts’ backs so that the blood would ‘glue’ them on more securely. In ‘normal’ conditions, such horsemen had good control over their animals, but such control was more dificult to achieve in stressful battle conditions. For these reasons, and given the inadequacies of local mounts, many western African cavalries tended to use missile tactics, arming themselves with javelins or bows, rather than using horses in shock assaults. Horsemen skirmished with and harried the enemy in this way, and only when the latter appeared broken might the cavalry charge with sabres. Polities grew up on or around the key river systems of the savannah, chiely the Niger, Benue and Senegal; they were based on diversiied economic systems – farming, ishing, hunting, salt-mining, ironworking – which facilitated the growth of urban centres and of internal trade. These were the basic building blocks of empire; but above all, they depended on long-distance commerce, and in this respect the camel was more valuable than the horse. Camels were occasionally used in war – if not in actual battle, then to carry supplies to it – but they were absolutely essential in opening the desert to commerce. In particular, they carried gold and other trade goods from West Africa to the Mediterranean world, and it was revenue from these trades which was the most important source of funds for state building and warmaking. The Soninke were one of the earliest groups to seek advantage from this combination of factors. They were ironworking farmers whose expanding settlements west of the Niger bend required defence from desert raiders: Pressure from the Sanhaja Berber at the desert’s

ARMS IN AFRICA ’S ANTIQUITY



edge drove militarised state formation among the Soninke, while iron technology was used in the development of new weaponry for the purpose. Militarisation took place as the process of desiccation drove the desert edge southward, and the Soninke organised themselves in the irst instance around the defence of pasture- and farmland against armed and displaced nomads, the latter increasingly motivated by religious zeal which further radicalised what was in essence a struggle over scarce resources. The desert edge, indeed, was – and remains – one of Africa’s most dynamic, and ambiguous, frontier zones. It contained a thriving and distinctive economy – based on salt, grain and livestock – from which various groups drew succour, including the savannah state builders; but the desert edge was also intermittently violent, a volatile place where pastoralists and farmers, again, sought exchange for mutual beneit, but where they also clashed over space and resource. Early Soninke militias probably fought on foot, but increasingly infantry was joined by more elite units on horseback using animals imported from the north – and probably from Berber traders who had to tread carefully, as war-zone merchants always do, along corridors of conlict to deliver the merchandise. The line separating defensive from aggressive militarism can be a thin one indeed, and by the eighth century, the Soninke had crossed it: The resultant kingdom of Ghana was an expansionist state, using cavalry and infantry equipped with iron weaponry both to attack and to patrol, funded by the proceeds of the gold which was transported from Bambuk on the upper Senegal northwards to the desert. Raids were also carried out on neighbours, particularly to the south, in search of captives to be sold as slaves to Saharan merchants, much as Kanem and Borno did. As such commerce expanded, in the ninth and tenth centuries, so did Ghana, which seized the key trading town of Awdaghust in the eleventh century. To awed Muslim merchants, whose accounts are invaluable, Ghana was the mighty and wealthy ‘land of gold’; one report declared, improbably, that the kingdom’s army was some , strong. Proits from trade were critical: Without them, Ghana was deeply vulnerable. As long as commercial dividends looded into central treasuries, good horses could be imported, and the state could defend itself against northern marauders, whether the latter operated as jihadists or otherwise, and cavalry could be deployed to control outlying districts perennially given to secession. Indeed, Ghana was in reality a sprawling, decentralised polity, and increasingly



WARFARE IN AFRICAN HISTORY

so, with little direct central control over the provinces which were obliged, above all, to supply men for military service. It was a fragile system. By the early twelfth century, new goldields had been opened in Bure, and trade routes running east of Ghana, beyond the state’s reach, were increasingly lucrative. This, combined with gradual environmental degradation, explains the decline of the empire in the course of the twelfth and thirteenth centuries. Funds dried up with which to inance cavalry; soil degradation led to a dispersal of farmers and herdsmen from the former heartlands, for the time being at least, which meant shrinkage in available manpower. And war itself had taken its toll, for the recurrent violence on the frontier where desert met savannah, between the Soninke, the Sanhaja Berber and the Almoravids, disrupted the trade routes which had been Ghana’s lifeblood. War was costly, and for all Ghana’s economic resources and political ingenuity, its rulers could not afford to wage it – or at least could not do so successfully – unless a series of interrelated factors were running in their favour. Once that system was disrupted, at any point, severe trauma resulted. As Ghana slowly disintegrated, the frontiers revolted – or, perhaps more accurately in some cases, simply waited until the inal vestiges of Ghanaian authority had evaporated. The most spectacular – and the most enduring – of these insurrections involved the Mande to the south. Here, in the early thirteenth century, one of the savannah’s great warrior-kings, Sunjata, organised the Mande to wage war on their former masters and to take advantage of the new gold routes. Traditions relate the story of a momentous battle, in around , in which Sunjata’s forces crushed those of the Susu king Sumanguru who had subjugated the Mande in the course of asserting independence from Ghana; but just as likely it was prolonged war of attrition which would result in the emergence of Mali, which succeeded – and at length absorbed much of the former territory of – Ghana as the dominant power in the western savannah. Sunjata has been immortalised in story and song since the thirteenth century, his life and deeds told and retold in epic form over the centuries by jails, or bards, in modern Mali and Guinea. Sunjata is recalled as a great warrior, his violence righteous, but he was even more than that: The symbolism surrounding Sunjata in Mande culture and history is extraordinarily powerful, for his story embodies a moral code and recalls a timeless sense of honour and courage; his name evokes a glorious past, a time, indeed, that may come again. He remains one of the most

ARMS IN AFRICA ’S ANTIQUITY



important igures, and one of the most signiicant reference points, in early African history. In the context of warfare, moreover – apparently invariably dominated by the male sex, both the practice and the study of it – the Sunjata epic is striking in terms of the prominence it awards to women. Two women in particular have key roles – Sunjata’s mother, ill but heroic and a seminal inluence on him, and to whom he is devoted; and his sister, beautiful, wily and courageous, who plays a key part in his inal victory over Sumanguru. The epic is a reminder of the relationships which are necessary in order for great goals to be achieved: Sunjata, great man though he was, possessed of wondrous powers, still could not do it alone. At its greatest extent, in the fourteenth century, Mali encompassed territory from the Atlantic coast south of the Senegal to the eastern stretch of the central Niger bend. It was one of the most dramatic instances in the region’s history of the armed periphery marching on the heartland and inheriting its assets – although, as we see in the next chapter, Mali also inherited Ghana’s intrinsic weaknesses. There were political developments elsewhere, including within the province of Songhay to the east, which had its roots in a cluster of ishing settlements positioned along the Niger bend from the eighth century. These were also well positioned to take advantage of trans-Saharan trade networks, and a process of amalgamation in the ninth century represented the seeds of state formation. Songhay thus had a long history of independent development prior to its incorporation into Mali in the s, an independence only rediscovered following Mali’s own demise in the ifteenth century. Notably, Songhay’s riparian culture would also develop a military component in time, as canoes, in addition to their use as commercial transport, would be used to wage war on the Niger and to ferry troops to the theatre of combat. In the meantime, the spread of Islam southwards from the desert, carried by Berber caravans and trade migrations, introduced new codes of brotherhood, honour and chivalry to the warrior castes of the savannah. The rulers of Ghana had welcomed Muslim traders but eschewed outright conversion until the late eleventh century; the conversion of the ruling elite in the s gave rise to the controversy surrounding Ghana’s supposed ‘conquest’ by the Almoravids, to which we turn later. Songhay elites were likewise Muslim by the eleventh century, and Malian kings sought – not always successfully – to balance Islam alongside local systems of belief. They needed to do so because in the early centuries of the second millennium, Islam



WARFARE IN AFRICAN HISTORY

remained a largely urban faith, conined to political, economic and military elites who used it, quite apart from any spiritual ediication, to strengthen commercial linkages both within the savannah zone and between the savannah and the Muslim world to the north. Islam also facilitated new forms of military cooperation and cohesion, as it had in the Sahara. Not all West Africans embraced the idea of the state in this era (or, indeed, in any other era); in fact, for all the might and extent of Kanem, Borno, Ghana and their successor empires, most West Africans probably did not live in a ‘state’, or in anything other than a very imprecise and loose arrangement with one. We do not know how these societies fought, although it is sensible to suggest a network of collaborative horizontal linkages for the purposes of defence distinct from, and probably on a rather smaller scale than, the vertical and hierarchical systems of the large states. Clearly certain activities were dificult, if not impossible, without centralised political structures, and these included the waging of war, whether defensive or offensive, on a large scale and as an extension of political and economic strategy. Hierarchical command, inancial resources and the mobilisation of key sectors of the population (mostly young males) are characteristic of centralised statehood, and the cavalry empires of the savannah were able to organise these critical elements. This was expensive, centralised militarism paid for by long-distance commerce – as it was elsewhere in the world at this time. But individualistic martial cultures, again, rendered fragile the equine militarisms which nonetheless depended on them, while commanders of elite bands of horsemen had a propensity to revolt, and even to challenge the very heart of the system itself. Meanwhile, low population densities and the scattered nature of populations, as well as the enormous distances involved and the great expense incurred by trying to traverse them, meant that these states struggled to overcome the stubborn localism of savannah political culture. The kafu – the small-scale, local community, essentially an autonomous cluster of villages which was the microstate of the region – endured while empires rose and fell. Those empires needed to root themselves in the kafu while also seeking to move beyond it; but it seemed that no amount of military force could impose the state on the kafu for long, and certainly not very far beyond the cosmopolitan centres which strove to meet the costs of ever larger political and economic projects. Yet the visions behind those ambitious projects would persist.

ARMS IN AFRICA ’S ANTIQUITY



Colonisers and Converts in Northern Africa The western savannah was inextricably linked to events in the Sahara and northern Africa, where in the second half of the irst millennium CE, the main dynamic was the Arab conquests and their larger repercussions. Indeed, the envelopment of much of the northern half of Africa by Islam – from Senegambia and Mauritania to Eritrea and Somalia – was a key driver of conlict between the late irst and early second millennia CE, as well as the facilitator of cooperation and new identities over larger areas. Arab forces had moved into northern Africa via Egypt by the middle of the seventh century CE: In the early s, they expelled a deeply unpopular and oppressive Byzantine administration and established a capital at Cairo. From there, Arab armies were both in a position to dominate the upper Nile and poised to advance westwards into the region they would call al-Maghreb – literally ‘the West’. Nonetheless, Nubian resistance prevented southward expansion along the Nile valley, in the same manner as, in earlier centuries, the Nubians had halted attempts by successive Egyptian regimes to impose themselves on the region. Islam did spread slowly into Lower Nubia as a result of largely peaceful trading relations along the Nile, but from the middle of the seventh century, for several centuries, the Arabs themselves were held at a frontier around the irst cataract of the river. Only in the fourteenth century would the southward migration of Arab nomadic clans towards Darfur and Lake Chad transform the genetic and cultural makeup of the savannah lands west of the Nile valley. Islamic armies encountered tough resistance to the west, too, certainly tougher than they had encountered in Egypt. Byzantine forces made good use of their naval advantage, although the main Byzantine opposition had been overcome by the s and s. Then there were the nomadic Berbers on horseback, whose skilful armed resistance had long before conined the Romans to their bases in coastal Libya and Tunisia, and whose undoubted military skills were now turned on the advancing Arab armies. The Arabs were compelled to overcome a multitude of Berber chiefdoms one after the other, which made for exhausting campaigns for the Arabs, but which also represented the key Berber weakness, namely lack of unity and common purpose. By , the invaders had reached the Moroccan Atlantic coastline and were preparing to cross the straits into Spain. However, although Berber resistance had been subdued along the main lines



WARFARE IN AFRICAN HISTORY

of the Arab advance – if not in the deeper hinterland, from where a loose tribute was paid to the coastal settlements – their martial reputation remained intact. Berber tribesmen were increasingly recruited into the Arab armies, and were deployed extensively in the invasion of Spain. Crucially, rates of conversion to Islam among Berber recruits were high – the result, at least partially, of military service itself – and they quickly gained a reputation not only as formidable soldiers but as zealous Muslims. Nonetheless, many Berbers retained a basic hostility to the Arab settlers, and particularly resented their innate sense of ethnic and cultural superiority, along with the abiding notion that they were superior Muslims, too. This was especially a matter of deep offence among those Berbers who themselves had converted to Islam. Therefore, the second key development following from the Arab conquest of northern Africa was the emergence of centrifugal forces in the west, and by the mid-eighth century, the Muslims of the Maghreb were already defying the authority of the Caliphate. Berber converts were drawn to sects which set themselves up in opposition to the Caliph, and the inluence of the Kharijis – with their emphasis on equality of rights across the umma, the Muslim community – was particularly pronounced. Khariji Berber polities were emerging in the central Maghreb in the eighth and ninth centuries. Further west, Shi’a Islam was important. Across North Africa, periodic attempts were made at regional unity, based on the marginality of the Maghreb to the Islamic world itself, the shared culture of the Maghreb, and the call for a puriication of Islam: a return, in other words, to the basic principles of the faith, and the creation of a more just society based on the principles of Islam. Politics in the Maghreb would continue to be characterised by localism and fragmentation, but the ideological framework had been created whereby more radical Islamic movements, achieving wider, if short-lived, unity, periodically appeared across the region. Thus was a pattern of ideological militarism created in north and northwest Africa. The Islamism of the desert, moreover, would have a direct impact on developments further south. The process of Arabisation in northern Africa continued, meanwhile, with the westward migration of the pastoralist nomads known as Bedouin from Arabia. Highly mobile, they migrated between the tenth and the thirteenth centuries in two broad patterns, those hugging the coast known as the Banu Hilal and those moving further inland as the Banu Sulaym. These movements have been associated with a fair amount of

ARMS IN AFRICA ’S ANTIQUITY



wanton violence and destruction, although this may have been exaggerated, and in any case they brought not only war but also Islam and the Arabic language, as well as a large amount of Arabic culture. However, it is worth noting a larger theme in Islamic history at this juncture, for the Bedouin armies that swept out of Arabia were probably not, in the irst instance, especially committed to the Faith; only in the waging of war did these Arabs become devoted Muslims, a pattern also discernible in the case of the Berbers, whose conversion rates, as we have noted, were probably directly linked to their service in conquering Arab armies. Notwithstanding the inherently fragmentary nature of Berber society, Islam provided the Berbers of the Maghreb with a common cause and, in the loosest meaning of the term, a sense of brotherhood, and it facilitated stronger links among traders and between traders and herdsmen. In particular, however, Muslim Berbers along the expanding desert edge had common cause against the inidels to the south, with whom they were engaged in conlict over pasture land. The key conlict was between the Sanhaja Berbers who were in the front line of a struggle against the Soninke, builders of the Ghana kingdom between the Senegal and Niger river valleys, and this desert-edge competition drove militarisation on both sides. For some Sanhaja leaders, this was nothing less than a jihad, and they sought to mobilise their brethren as part of a larger spiritual conlict. But the message frequently fell on deaf ears among those Berbers whose Islam was both less public and less militant, and whose concerns were ultimately very local ones. Indeed it was the failure of so many to respond to the messianic call to arms that prompted renewed vigour on the part of reformist leaders. So dismayed was the Muslim Berber scholar Abdallah ibn Yasin at the almost total lack of ‘genuine’ Islamic faith when he toured the southwest Sahara in the mid-eleventh century, he embarked on a vigorous campaign of teaching and mobilisation, arguing that the failure to truly embrace Islam was the cause of Sanhaja disunity and their dificulties against the Soninke. Initially unable to make much headway among the local population, he withdrew to the Atlantic coast and created his own devoted following at a remote site. These became known as the Almoravids, from the Arabic al-murabitun, or ‘men of the ribat’, a holy fortress or retreat in the defence of Islam. The Almoravids were one of northern Africa’s earliest radical Islamic movements, preaching the rejuvenation of the faith and the waging of jihad against unbelievers. Following Abdallah ibn Yasin’s hijra, the



WARFARE IN AFRICAN HISTORY

message spread among the Sanhaja and was embraced in particular by those locked in conlict with the Soninke. Again, as in other similar scenarios, many Sanhaja probably only became Muslim in the course of waging war in the name of Islam. Across the western Sahara, newly converted zealots built a new polity, and Almoravid leaders harnessed local military culture, mobilising young nomads on horseback to participate in raids against inidels; nonetheless, while a sense of ethnic superiority for a time meant the exclusion of non-Sanhaja from military service, new recruits were not always forthcoming, and so use was made of Christian mercenaries from the north. By the s and s, the restless new state was muscling in on the highly lucrative gold and slave trades lowing across the central western desert. It proved a temporary unity, however. Abdallah’s death led to the state splitting in two, with the northern Almoravids dominating much of Morocco and advancing into southern Spain, and the southern grouping continuing the conlict with Ghana. It is to this period that we can date the long-running debate about whether – as claimed in many of the Arabic sources – the southern Almoravids, under the leadership of Abu Bakr, conquered Ghana and forcibly converted its leaders to Islam. The conquest supposedly happened in , and it is certainly the case that Ghana’s leaders were Muslim by the late eleventh century. But although clashes certainly continued between Ghana and the Almoravids, the conversion of the former was probably peaceful, and there is no real evidence of Ghana’s military subjugation. Abu Bakr died in , and disunity re-emerged, weakening the southern Almoravids and leading to a shift of power back to the north. Ultimately – and notably – the Almoravids did not remain in the desert, but sought more attractive locations beyond it, whether the better-watered grasslands to the south or, perhaps even more alluringly, Morocco to the north. There, the process of withdrawal from the desert was underlined by the attempt to create an army which was less a force of nomadic tribesmen and more of a centralised, ‘imperial’ army, complete with drums and lags – and it was more heterogeneous, too, for it was increasingly composed of Christian captives from Spain and slaves from the south. But the northern Almoravids had their own problems: By the early twelfth century, some detected a certain backsliding among a formerly devout leadership. Islam was no longer strictly observed, and in the s, a century after the Almoravids had irst emerged, a jihad erupted against them in turn. The Almoravids were overthrown by the

ARMS IN AFRICA ’S ANTIQUITY



Almohads, who continued the work of purifying Islam in the region and uniting the Maghreb itself under a single cause. The Almohads were named after a corruption of the Arabic al-muwahhidun, which meant ‘the unitarians’, relecting the particular emphasis placed by the movement on the oneness of God. They also incorporated the Shi’a notion of the Mahdi, or divinely guided messiah. Ibn Tumert, their early spiritual leader, claimed to be that messiah, but it was his successor, Abd al-Mu’min, who was largely responsible for creating the formidable Almohad polity. Its success rested on horsed warriors and infantry, and, as with the Almoravids, both wings contained Christian mercenaries hired in Morocco and even in Spain. The Almohads were no more successful than their Almoravid predecessors in creating permanent political structures or an enduring, uniied identity; by the early thirteenth century, they were already weakened by serious internal rifts. There was a new enemy, too – or the revitalised version of an old one – in the form of Christian Spain, whose forces were advancing steadily by this time. In the course of the thirteenth century, the Maghreb descended into a series of rival polities while Christian forces pushed Islam from Spain, and, in a marked reversal of fortune, by the ifteenth century, the Spanish and Portuguese were themselves establishing outposts on the northwest African coast. Nonetheless, transient though the Almoravid and Almohad states were – their dominance lasted little more than two centuries – their legacy proved rather more enduring. The Berbers were more thoroughly Islamicised, while Arab migrations between the tenth and thirteenth centuries had implanted Arabic culture and language more deeply across the Maghreb. Despite the political fragmentation, commercial networks were stabilised, both within the region and between northern Africa and the western savannah on the one hand and Christian Europe on the other. Precedents had been set, moreover, for the rise of militarised theocracies across and on the edges of the Sahara, and frameworks had been created for religious links across vast regions of northern and western Africa. If one key theme across northern Africa was the armed religious sect, another was the increasing muscularity of the army in political affairs – and both were frequently in response to, or at least intensiied by, foreign threats. In Fatimid Egypt, economic stability from the late tenth century onwards was protected by an increasingly powerful army which was bolstered by the import of Turkish slave soldiers, the Mamluks. The rank and ile, meanwhile, were composed of Sudanese



WARFARE IN AFRICAN HISTORY

slaves brought in from the south, as many Egyptian armies had been throughout the Pharaonic era. The army’s increasingly troublesome independence coincided with the Crusader threat in the mid-twelfth century, and Egypt was saved only by a young Kurdish warrior named Saladin, later the hero of wars against Richard the Lion Heart’s Crusaders. With the death of the last Fatimid ruler, Saladin was declared sultan of Egypt in , and the Abbuyid dynasty was thus founded; over the next century, however, the army continued to accrue wealth and political position. Abbuyid reorganisation involved the systematic import of Mamluks, seized and trained when very young and fed into a corps of professional soldiers who were, by the beginning of the thirteenth century, a highly trained and disciplined military elite. This was especially true of the ‘royal Mamluks’ – those of former sultans as well as of the reigning sultan, and those of the bodyguard and corps of pages – who received extensive training in barracks for the purpose. They were primarily mounted archers – Mamluks were famously ine horsemen – but in addition to the bow, their equipment included the lance, sword and shield. Successful Mamluks gained their freedom and transformed themselves into a landed military aristocracy, owning agricultural estates on the banks of the Nile – many of which had been developed under the Fatimids – and gaining steadily in economic as well as political inluence. In , the last Abbuyid sultan was assassinated, and the Mamluks seized control of government itself. Mamluk Egypt was the military state writ large, and a markedly successful one – although it would be matched in military vigour and violent creativity by events further south.

Violence and Empire in Northeast Africa In the other major conlict zone of Africa’s antiquity – northeast Africa – faith played a key role in warfare and militarism as well (indeed, as we shall see, it has done so ever since). Here, the early story centres on one of the ancient world’s foremost civilisations, Axum, located in the northern Ethiopian Highland plateau – encompassing Tigray and central Eritrea – and reaching to the adjacent Red Sea coast. The nucleus of the state emerged in the irst century CE, and early in the fourth century, Christianity was introduced, in all probability patronised alongside Judaism. Judaic elements survived, but Christianity became fundamental to the growth of the centralising

ARMS IN AFRICA ’S ANTIQUITY



state; between the ifth and seventh centuries, the scriptures were translated into Ge’ez, and a culture of militant monasticism was taking shape across the highlands. By the middle of the irst millennium CE, Axum was a dynamic, prosperous and militarily energetic state, the model of the centralised military polity, territorially and ideologically expansionist, its armies built around loyal provincial governorships and directed from the imperial core. While it lasted, it was arguably an even more successful exemplar of the model than its counterparts in the western savannah. Axum controlled abundant local resources – notably in terms of carefully managed farmland, which was highly productive and could support relatively dense populations as long as rainfall was good – and sat astride a thriving commercial network through which it was linked to the Mediterranean and the Middle East. Although the details of its military organisation are sketchy, it was clearly an expansionist polity whose elites placed great emphasis on martial success. At its height, Axum’s military reach was considerable: Beyond its territorial core, it had tributary provinces and even overseas ‘protectorates’ in the Arabian peninsula, and is notable, in African terms, for its development as a maritime power. A navy extended its inluence across the southern Red Sea into Arabia and enabled it to vie for dominance with Saba, in present-day Yemen. Early king lists – one of the best means of imposing order on particular campaigns and thus charting the process of expansion – are hazy, but even before the reign of King Ezana, the ruler associated with the adoption of Christianity in the early fourth century, a number of inscriptions attest to military expeditions into southern Arabia and the establishment of military garrisons there. Campaigns against the Nubian kingdom of Meroe are also described for this earlier period; as for Ezana himself, there is little doubting his ‘glories’ from the relevant inscriptions, which describe a king regularly on campaign, conquering and subduing. He was – using a moniker adopted by later monarchs – ‘King of Kings’, his inscriptions making clear that he was the overlord in a web of tributary relationships with other rulers in the region. By the fourth century, then, Axumite kings were using inscriptions on stone monuments to emphasise their authority and, through their exertions, the achievements of the state of which they were personiications. These inscriptions were frequently concerned with military conquests, listing campaigns and the booty acquired – including slaves seized to supplement the domestic labour force, women, livestock – and describing the extent of the area under tribute. Although



WARFARE IN AFRICAN HISTORY

we have scant data on how these wars were actually fought, this was clearly a polity with a highly developed culture of militarism, while the inscriptions themselves may be seen as the forerunners of later Ethiopian royal chronicles which were centrally concerned with warfare and struggle – foreign invasions, righteous campaigns of civilising violence – and with the necessity, indeed desirability, of conlict and arms in the achievement of individual and national greatness. This would be one of Axum’s key legacies to the polities which followed it. Another was the transformation of Christianity into a royal cult, its basic tenets deployed to enhance central authority and the personal prestige of rulers. Axum’s decline, in the second half of the irst millennium CE, was a long one. The evidence suggests a prolonged period of low rainfall, which clearly eroded the foundations of Axum’s success – its strong agricultural base. The population drift and demographic decline which resulted from soil erosion, as in Ghana far to the west, fundamentally weakened the state’s ability to wage expensive long-distance, or even local, warfare. This was, after all, a fragile environment. Wars elsewhere, too, damaged Axum. Recurrent conlict between Persia and Byzantium, encompassing the northern Red Sea and the eastern Mediterranean, disrupted the trade routes so important to Axum’s growth; but more important was the rise of Islam on the other side of the Red Sea in the mid-seventh century. Over the longer term, the rise of the early Caliphate, and in particular its effective conquest of the Red Sea world, including Egypt, isolated Axum from the Eurasian heartlands of Christianity itself and severed critical commercial links with the wider world. In sum, Axum’s formerly secure elites found themselves locked into a cycle of decline, unable to maintain control over overseas conquests or even neighbouring tributary provinces, which in turn caused a dramatic reduction in revenues. While Axum probably continued to direct some commerce in the southern Red Sea, such control was increasingly tenuous. The much diminished polity’s isolation was compounded by an armed revolt on one of the state’s most volatile frontiers, the one inhabited by the Beja peoples to the north, formerly under Axumite suzerainty. This caused the centre of political gravity to shrink further southwards, away from the coast and escarpment and deeper into the highland hinterland. There were elements in the process which were distinctive to the region, but it was a pattern of warfare – insurgent frontier pitted against contracting centre – discernible across the continent throughout this era, and beyond it.

ARMS IN AFRICA ’S ANTIQUITY



Axumites were not the only Christians in the region – Egyptians carried Christianity to the Nubian kingdoms in the ifth century – but their relative seclusion had a profound impact on the development of polity and society. With its church cut off from the key Mediterranean and southern European centres of Christianity, including Alexandria, Axum’s faith – and the institutions building around it – took on very distinctive forms. As the centre of the Caliphate moved to Baghdad in the eighth century, the centre of commercial and political gravity in the region shifted to the Persian Gulf and northwest Indian Ocean, leading to the appearance of Islam on the Somali coast. By the tenth century, the old Axumite port of Adulis had withered away and had been replaced by ports such as Zeila, heralding the beginning of an era which would see Islam pushing steadily into the interior of modern Somalia, Djibouti, Eritrea and Ethiopia itself. A note of caution is required here, however: The clash between Christianity and Islam in northeast Africa was by no means predestined, nor was it relentless. Later writers, notably nineteenth-century Europeans, tended to depict it as such in their portrayal of Ethiopia as a semi-civilised outpost of distorted Christian faith, somehow over the ages holding the ferocious Muslim hordes at bay. It is also true that over several centuries, Ethiopian chroniclers themselves projected similar imagery, albeit with a somewhat different aim in mind. Rather more inevitable was the degree of peaceful coexistence between proponents of the two faiths and the commercial cooperation which characterised ChristianMuslim relations across the region. Nonetheless, it is true that violent conlict was recurrent, and at times intensely bloody. It is also true that this conlict had its roots in the nature of the economic competition which escalated in the eastern highlands in particular, and in the gradual radicalisation of both lowland Islam and highland Christianity as these were co-opted into state-building projects on either side of the northern Rift Valley. Much early war was indeed commercial in provenance. As the irst Muslim traders advanced up the Awash valley and into the highlands in the tenth and eleventh centuries, they represented a threat – if a low-level one – to Christian management of the ivory and slave trades; when a series of Muslim trading clans began to amalgamate themselves into compact, dynamic mini-‘sultanates’ in the course of the twelfth and thirteenth centuries, the threat became rather more urgent. An early sultanate at Shoa was later succeeded by another Ifat, which expanded to dominate commerce from its position in the highlands



WARFARE IN AFRICAN HISTORY

south of the Blue Nile. It was this process of expansion which escalated conlict with the post-Axumite Zagwe state in the Christian heartlands. Christian military superiority, irst under the Zagwes and later under the irst Solomonic rulers from the s onwards, kept the Islamic challenge under control, for the time being. For obvious reasons, perhaps, the Ethiopian region’s trials and tribulations following the decline of Axum have been likened to northern Europe’s ‘dark ages’ after the fall of Rome. The evidence is thin, but late in the irst millennium, the area was convulsed by violent insurgencies of various kinds – including, importantly, that of the Agawspeaking population, involving a shadowy female igure. A common theme running through the fragments of source material – traditional as well as documentary – relates to this pivotal female character, a queen or a person of royal blood, in some way drawing the Axumite era to a close and violently ending Axum’s dynastic cycle. In one tradition, she is the Agaw queen Gudit, or ‘Judith’, persecuting Christians and repeatedly savaging the already weakened kingdom. At its most basic, the tradition perhaps describes an Agaw people, who were indeed led by a woman, overthrowing the Axumite state.Yet the Gudit character is almost certainly a composite of several individuals, for her lifespan is improbably long when all the strands of the story are pulled together – although she appears most active at the end of the tenth century. Either way, this is a violent nemesis in female form: This mysterious woman is at certain times a wicked pagan, at others a Jew, or a lapsed Christian; but all traditions agree that she was driven by an awful bloodlust, seeking to destroy by ire and sword churches, monasteries and symbols of Christian and Axumite power. In Amharic, she was known as Esato, derived from the word for ‘ire’; and in this tradition she killed the last emperor of Axum, seizing the throne herself and reigning for forty bloody years. The violence of ‘Judith the monster’, the iery one, was proverbial. In fact, the sources for her are few, and she remains in a half-light at best. But in their broadest sense, these stories certainly describe an era of violent upheaval and the insurgency of the frontier against the weakening centre. This was the march of the peripheral Agaw against the old Axumite imperial heartland, the two trading places so that the core of Axum became the periphery and the Agaw occupied the centre. Nonetheless, what is clear is that the idea of the uniied Christian empire survived successive crises and would continually re-emerge at key junctures, as successive state-building elites sought to realise

ARMS IN AFRICA ’S ANTIQUITY



the political ideal which they had inherited. This was certainly true of the Zagwe, who seized power in around  – or , according to Ethiopic tradition – and who imposed a pax on the troubled northern highlands, presiding over one of the most energetic periods of Christian expansion across the region, as well as over an era of commercial and cultural interaction with Egypt and the Middle East. The problem, for the Zagwe, is that they were Agaw-speakers – associated with the violent destruction of Axum – and that they were thus depicted in subsequent chronicles as illegitimate usurpers, transgressors with no true ‘Israelite’ connection. Yet the Zagwe were effective governors, part of an increasingly important military and political class that had become assimilated into Christian culture in the centuries following Axum’s decline. Their ‘seizure’ of power probably did not represent a signiicant break with the past, but rather was the logical culmination of a process dating back at least  years. This was probably not a military state to quite the same extent that Mamluk Egypt would become, but it was nonetheless a highly militarised one: Army commanders, appointed from among the royal family and or those to it, served as extremely powerful military governors of key territories, while military command itself was distributed among Zagwe courtiers drawn from their home province of Lasta. Zagwe forces were also engaged in subduing the Muslim trading states to the east and in providing the raw material for the lucrative Red Sea slave trade. But the biggest challenge facing the Zagwe – and one which they failed to overcome – was internal issures, and the refusal of a range of groups to recognise their legitimacy. Increasingly hostile and well-armed opposition movements appeared among the Tigrayan and Amhara populations by the early thirteenth century; these were groups who regarded themselves as the natural ethnic enemies of the Agaw-speaking Zagwe and the rightful heirs to the Axumite Christian legacy. An insurrection of especial vigour erupted in Shoa in the late s, involving a Christian community which had ample inancial resources as a result of its involvement in eastbound commerce, and – just as important – the backing of the church. The leader of the Shoan rebellion was Yekuno Amlak, and his forces won crucial engagements with the Zagwe army across Lasta and Begemedir between  and . When the last Zagwe ruler was killed in , Yekuno Amlak proclaimed himself king and declared that he represented the ‘restoration’ of the legitimate Solomonic line. In asserting



WARFARE IN AFRICAN HISTORY

its legitimacy, the new dynasty sponsored the redaction of a carefully contrived and powerful narrative, a story that formed the basis of the holy book known as Kebre Negast, ‘Glory of the Kings’. This was a text which became Christian Ethiopia’s ‘national epic’, on a par with the traditions evolving out of Sunjata’s exploits further west earlier in the century. The ‘Solomonic’ dynasty claimed that it was descended from King Solomon of Israel and Makeda, Queen of Saba (Sheba), and the Kebre Negast – replete with righteous violence in an imagined past – would justify ideological warfare against barbaric and inidel enemies on a greatly enlarged scale as the highland Christian state reinvented itself once again through force of arms.

The Bantu Diffusion and Its Politico-Military Implications South of the Ethiopian Highlands and the vast savannah belt, comparatively little is known about war in antiquity. Certainly, before the early second millennium CE, only the broadest generalisations and the most tentative speculations are possible, made so by the work of archaeologists and historical linguists. It is obvious that the very gradual shift from nomadic hunter-gatherer groups – whose organisation of violence would look very like the kind of ‘primitive war’ beloved of late-nineteenth- and early-twentieth-century ethnographers – to increasingly sedentary societies engaged in agriculture and the keeping of livestock was a fundamental one and denoted the escalation of ‘war’ as conventionally understood; it involved an expansion of scale and a marked increase in the range of issues at stake. Those systems were further transformed by the Bantu-speakers’ expansion, although the nature of the spread of Bantu languages across sub-Saharan Africa is still contested, and therefore the positioning of this remarkable but indistinct phenomenon in the story of African warfare is a problematic one. An early model proposed for the Bantu spread involved a sudden demographic ‘explosion’, a mass migration which moved coherently and uniformly from a homeland on the Nigeria-Cameroon border across the rest of the subcontinent. This is now considered inaccurate and overly simplistic; this was no great military conquest, nor was it a process of violent colonisation, although, to be sure, conlict must have often resulted. It now seems more likely that we are dealing instead with a multitude of groups whose movements ebb and low over two millennia, beginning not earlier than  BCE. The

ARMS IN AFRICA ’S ANTIQUITY



movements themselves were complex: some groups spread eastwards across the edge of the great tropical forest, and thence into the lacustrine and savannah regions to the south; others turned southwards earlier, edging along the Atlantic coastline; yet others drifted through the western forest and down into the Angolan woodlands. Migratory movements in central Africa in the early centuries BCE are best considered as swirling eddies, with pioneer groups peeling off toward the Zambezi valley and Lake Tanganyika and others hitting the northern fringe of the Kalahari Desert. By the early irst millennium CE, Bantu-speaking communities who had gone in different directions several centuries earlier were running into one another across centralsouthern Africa. The Bantu spread is associated with a number of key transformations which are directly relevant for the military history of sub-Saharan Africa. Bantu-speakers brought settled agriculture with them; some groups practised the keeping of livestock. In eastern Africa, for example, the migration of Cushitic-speakers – who had reached northern Kenya by c.  BCE – introduced domesticated cattle to the region, and by the early irst millennium BCE, a new kind of economy based around pastoralism had emerged in central Kenya, inluencing Bantu groups which later drifted into the area. These integrated economic systems, and the associated use of pottery, denoted a superior material and technological culture which became increasingly pervasive across the subcontinent. Often small groups of settlers clung to rivers and lake shores, where they slowly built up relatively concentrated centres of population; and as their populations grew rather quicker than the surrounding ishing and foraging indigenes, the latter were drawn to the settlements of the former and adopted their language and culture. Critically, too, some Bantu-speakers brought iron with them, or built on the ironworking techniques already in existence, although how and when remain unanswered questions. But certainly a cluster of sites across the centre of the continent (e.g. Niger, central and southern Nigeria, Gabon, southern Uganda, northwest Tanzania) can be dated to the middle centuries of the irst millennium BCE, and from this period onwards, ironworking skills spread, if unevenly, across much of sub-Saharan Africa. Across eastern Africa, for example, Bantuspeaking, iron-using farmers transformed the landscape by cutting down woodland, pushing back the tsetse frontier and opening new areas of land for cattle keeping. By the ifth century CE, virtually the whole of sub-Saharan Africa had begun the transition to iron-based



WARFARE IN AFRICAN HISTORY

farming, and by the end of the millennium, hunter-gatherer populations had either been absorbed into the new socio-economic systems or been pushed to the edges of those systems, into areas where the latter could not function, notably the excessively dry southwestern corner of the continent. This is a simpliied picture, no doubt, but this vital process of interaction and exchange laid the foundations of polity and ethnicity and led to new systems of warfare and militarism in the course of the irst millennium CE. The combination of ironworking, farming and (in some areas) cattle keeping transformed society and polity, and with them the systems of violence. Workers in iron themselves, especially across eastern and southern Africa, were often associated with political authority and the creation of centralised statehood, and indeed, sometimes, were seen as possessing arcane magical powers. Indeed, smelting and smithing became as emblematic of masculinity as participation in war. Iron greatly increased agricultural productivity and enhanced communities’ military capacity. With iron-tipped spears and bows, Bantu-speaking Africans could build larger polities, expand into new land and defend it. More productive and intensive farming led to divisions of labour, which freed up sections of the population for more specialised roles, including political leadership and the waging of war. There was more to defend – settled communities, arable land, livestock, material culture – and more people to defend it, although armies in all likelihood remained relatively small throughout this era: part-time infantries called up in time of need. Ideologies were developed to justify the ownership of land and the control of people, often against supernatural and natural enemies alike, and these commonly involved the weaving of military prowess and obligation into political philosophy. Violence was at the heart of these political systems from the beginning, especially in regions characterised by demanding, even hostile physical environments, and which required the stringent control of people in order to lourish or even survive. In many respects these were frontier societies, seeking to maximise populations and develop their factor endowments more broadly, and regularly having to face local enemies who might once have been kin but who had set out to colonise their own frontiers and to compete for those same endowments. As in the case of the western savannah, the outcome was not always centralised statehood, but rather the emergence of a host of differently structured polities according to local environmental needs. Some of these polities were more heterarchical than hierarchical – in

ARMS IN AFRICA ’S ANTIQUITY



other words, networks of theoretical equals engaged in exchange and reciprocity, in communities which were horizontally rather than vertically governed. Again, many of the speciics on how these polities actually waged war are elusive. We do know, however, that in the broadest terms, the scale and regularity of violent conlict in many areas increased with population spread, and thus the proliferation of closely contested frontier zones. This, in turn, meant that the most successful communities were those which drew people to them, and which were thus able to ield more men and impose themselves over a wider area. In the equatorial forest, for example, villages organised themselves across districts for the purposes of defence; alliances based on mutual aid were developed for common security, and in the area of modern Angola, political overlords were military protectors. Similarly, in the Great Lakes region, early linguistic evidence suggests the waging of war through complex networks of alliances. Leaders who developed extensive client relationships were able to mobilise young men as warriors and organise military expeditions. Almost certainly, the relatively sophisticated process of warfare by alliances grew out of – and coexisted alongside – older notions of conlict in the form of interpersonal feuds. Alliances – as with mutual security treaties between villages in the equatorial forest – allowed for more large-scale mobilisation, and the warfare which ensued was probably more intense as population levels became denser. Meanwhile, cattle offered distinctive opportunities for militarism, in large part because their protection demanded extensive military organisation, which in time facilitated peculiar political and cultural forms. In equatorial Africa, a distinction was made between ‘restricted’ and ‘destructive’ war. The former was constrained by a set of regulations which were overseen by elders who intervened when the contest was seen to have gone on long enough. ‘Destructive’ war, by contrast, was unimpeded and usually involved the victor burning the villages of the vanquished and otherwise seeking to destroy the latter. Such clashes probably increased over time, as population levels themselves increased and placed strains on existing resources, and probably involved alliances of villages pitted against one another. From the lack of evidence of early fortiication across much of central Africa it might be supposed that these wars were not quite as destructive as they would become; but from the early second millennium, military innovation was the result of a heightened security problem, as



WARFARE IN AFRICAN HISTORY

communities in favourable locations found themselves vulnerable to attack. In the equatorial basin, older weapons such as the stabbing spear and shield were joined by the scimitar – curved and hooked to rip away the opponent’s shield – although it is unclear whether this was introduced from the north or was a local innovation. In eastern Africa, spears and bows were the standard weapons. In open combat, the spearmen of equatorial Africa fought several ranks deep behind a wall of shields, often in clearings appointed for the purpose; but roadside ambushes were probably also common in dense forest, as was the use of war canoes on rivers. Limited available evidence certainly suggests – for west central Africa at least – a style of war in the early second millennium requiring training and discipline. Ultimately, military success depended on numerical superiority, and thus wider cooperation. Safety in numbers meant larger settlements with greater capacity for armed deployment. Similar developments are discernible in the uplands of central Angola, where there is also early evidence of stone walls and other defences on fortiied hilltops. By the early second millennium, a patchwork of regionally distinct states and societies based on iron-age agriculture and pastoralism were emerging across sub-Saharan Africa. Variously militarised polities – often, but not always, states with centralised kingship (sometimes with divine elements), hierarchical political structures, economic specialisation and social stratiication – appeared in different parts of the subcontinent: in the Shaba / Katanga region of Congo between the eighth and the thirteenth centuries; in Mozambique in the tenth century, and between the Limpopo and Zambezi river valleys soon after; in lacustrine eastern Africa by the fourteenth; and in the lower Congo region in roughly the same period. In the dry open grasslands of modern Botswana, military structures formed around cattle, control of which facilitated the emergence of chieftainship and thus early statehood; here, the ‘Toutswe Tradition’ involved hilltop settlements which combined specialist cattle keeping and some cultivation, and which lourished between the seventh and the thirteenth centuries. Pastoral militarism would be predominant in many parts of southern and eastern Africa in the centuries which followed. In southern Nigeria, both in the forest and the savannah edge, a system of city-states emerged in the eleventh and twelfth centuries – the earliest and most important being Ife, later to be replaced in signiicance by Oyo – which would drive dramatic politico-military developments in the millennium to come. Oyo, located just north of the forest belt, would soon tap into

ARMS IN AFRICA ’S ANTIQUITY



the commercial network to the north and acquire horses for a cavalry army which would dominate southwest Nigeria between the sixteenth and the nineteenth centuries. A little further east, the city-state of Benin, which may also have owed much to Ife, likewise emerged in the eleventh century under centralised kingship; it, too, would shortly become one of the major military powers in the region. In some areas, at least, political and military developments were responses to the need to manage ever larger groups, to control ever larger territories, and to access ever wider pools of resources – or, indeed, to control diminishing resources. The communities and polities which resulted were sometimes more collaborative than coercive, so-called heterarchies unburdened by the expenses of the militaries to the north. It is also the case that pastoral cultures had a range of different political responses, often developing decentralised systems less deined by territorial delimitation, and these diverged from the more compact sedentary model of statehood. But elsewhere hierarchies proliferated. Population settlement and growth led to political and thus military sophistication, but the demographic increase was never so great as to render the need for militarised control of precious human resources redundant; population growth based on ironage agriculture was enough to spur innovation but not suficient to disarm competing militarised frontiers. And while peaceful commerce and exchange of goods and ideas continued apace, conlict grew more sophisticated as well, laying the foundations for more complex societies and, in turn, expanding the scale and signiicance of warfare and attendant military establishments.

Relections: War and the Military in the Deep Past The diversity of Africa’s wars and military establishments in antiquity is clear enough, but it is also clear that certain broad themes apply across the continent. Warfare expanded steadily in scale and sophistication, and violence underpinned many exercises in state creation and the building of economic and cultural communities. This was an expensive business in the north, where the use of horses meant costs which could only be met through commercial entrepreneurialism and extractive and territorially expansive state structures; but these costs were largely absent in the south, in the vast tsetse zone, where the keeping of horses was impossible and where violence was organised on



WARFARE IN AFRICAN HISTORY

foot. Costs were comparatively low, and although the spread of Bantu languages was an extraordinary phenomenon, the range and capacity of individual communities was also comparatively limited. The incentives to expand military operations were similarly lacking, at least for the time being, for there were not many enormous costs to meet. But here, too, the foundations were being laid for military growth: Just as the use of horses in the savannah involved marked social stratiication – cavalrymen had elite status – so, too, in the forest zone did the transformation in farming techniques, driven by innovations in metallurgy, produce stratiication and specialism, including in the waging of war. By the early second millennium, these new communities – forged through the interaction between Bantu-speakers and indigenes – were coalescing into polities which would further hone the organisation of warfare. Whether in the savannah, in the northeast highlands, or in the forest, it is not clear just how ‘professional’ Africa’s armies were in the irst millennium. In most areas, it can reasonably be supposed that the bulk of ighters were part-time, and when not on campaign were engaged in other occupations – as cattle keepers, farmers, craftsmen and so forth. But horsemen can be considered a mounted class of professional warrior, dedicated full-time to the business of war and serviced by retainers, and both Ethiopia under the Zagwe and Mamluk Egypt represented recognisably professional military systems involved in political administration. In many ways, the early second millennium CE was a remarkable era, and in which key foundations were laid. The thirteenth century was particularly spectacular, witnessing two major insurgencies – by Sunjata among the Mande in the West African savannah, and by Yekuno Amlak among the Shoan Amhara in the Ethiopian Highlands – which led to two of the continent’s most remarkable military states and two of the great military traditions. It is noteworthy that two of the richest and oldest African epics were born of violent struggle, the product of militarised frontiers – a pattern which would be replicated elsewhere in the centuries to come. The age of Sunjata and Yekuno Amlak was also the age of the mighty Mamluks in Egypt, one of the earliest and inest instances of the politicised military. Mali, Ethiopia and Egypt would be dominant in the irst half of the second millennium, although Mali proved less enduring as a functioning polity over the longer term. In the centuries to the turn of the second millennium, faith became increasingly important in driving conlict across

ARMS IN AFRICA ’S ANTIQUITY



the northern half of Africa. Whether the political and military elites of the western savannah, or the new armed jihadist movements of the desert, or the clashes between a distinctive Ethiopian Christianity and the pioneer Muslim merchant states of the eastern Rift, arms and militant identities were organised around faith and drove state-building exercises, even if these were not always wholly successful. Many of Africa’s iron-age states were forged in war, then, and these states involved strategies which sought the effective marshalling of factor endowments, exercises in coercion and co-option which were made possible by force of arms and justiied by militarised ideologies. Nonetheless, these large states were faced with obstacles which not only prevented further growth, but actually caused massive shrinkage and even collapse: environmental and thus demographic fragility; the vagaries of long-distance commerce, and a periodic inability to balance commercial and military needs; and the pervasive localism which deied armed centres. But creative responses to these challenges would continue to evolve. Further Reading Bamba Suso and Banna Kanute. Sunjata: Gambian Versions of the Mande Epic (London: Penguin, ). Brooks, M. (ed. & tr.) A Modern Translation of the Kebra Negast (The Glory of the Kings) (Lawrenceville, NJ: Red Sea Press, ). Humphreys, R. S. ‘The Emergence of the Mamluk Army’, Studia Islamica,  &  (). Levtzion, N. Ancient Ghana and Mali (London: Methuen, ). McDougall, E. A. ‘The View from Awdaghust: War, Trade and Social Change in the Southwestern Sahara, from the Eighth to the Fifteenth Century’, Journal of African History, : (). Oliver, R. and B. M. Fagan. Africa in the Iron Age, c.  BC to AD  (Cambridge: Cambridge University Press, ). Phillipson, D. W. Ancient Ethiopia (London: The British Museum Press, ). Schoenbrun, D. A Green Place, a Good Place: Agrarian Change, Gender and Society Identity in the Great Lakes Region (Oxford: James Currey, ). Sergew Habte Selassie. Ancient and Medieval Ethiopian History to  (Addis Ababa: United Printers, ). Vansina, J. Paths in the Rainforests: Toward a History of Political Tradition in Equatorial Africa (London: James Currey, ). How Societies Are Born: Governance in West Central Africa before  (Charlottesville: University of Virginia Press, ).

CHAPTER 

The Military Foundations of State and Society, to circa 

The era between the early second millennium and circa  is signiicant as a discrete unit of study owing to several key developments. The era of the great cavalry empires in the western and central savannah reached its apex in Mali and Songhay, but by the beginning of the seventeenth century, alternative frameworks for military activity would emerge, moving away from the empire-state model which had dominated the region for so long. Another military state with its roots in antiquity, Solomonic Ethiopia, would likewise achieve remarkable success in the fourteenth and ifteenth centuries, before a series of wars in the sixteenth brought it to the point of extinction. Unlike its West African counterparts, however, Ethiopia would survive – just. Further south, the foundations built in the irst millennium, particularly in the wake of the Bantu diffusion and attendant political and economic transformation, would be used to create some of the most dynamic military organisations on the continent by the end of the sixteenth century, if frequently more compact and less concerned with the overt control of territory than some of their counterparts north of the forest zone. A series of state and non-state structures emerged across sub-Saharan Africa, each in different ways placing war at the centre of policy and culture. Some of these would prove remarkably durable. Once again, as in earlier centuries, migratory movements proved seminal. If the Bantu expansion and Arab migrations had been among the crucial dynamics of antiquity, in the irst half of the second millennium there were population movements of no less import, if on a somewhat smaller scale: The Nilotic migrations into lacustrine eastern Africa and the Oromo expansion into the Ethiopian Highlands in particular revolutionised 

THE MILITARY FOUNDATIONS



the nature of warfare and military organisation in their respective arenas, and transformed forever the polity itself. The other key theme of the era concerns the marked increase in encounters between Africans and foreigners, an exchange which had long-term implications for African military development. Most obviously, the appearance of the Portuguese along Africa’s Atlantic coastline was signiicant, not least in terms of the slave trade – which was well underway, if far from its brutal climax, by the end of the sixteenth century; the kingdoms of Kongo and Benin would be early participants, although their experiences were markedly different. The violent opportunities which the escalating trans-Atlantic commerce offered would be seized by new political actors, many of whom were scarcely visible by , although it is clear that certain processes were already in motion. The Portuguese visited new forms of war on the Swahili coast, too, and were involved in the wars of the Ethiopian Highlands. In North Africa, the expansion of the Ottoman Empire brought new military structures and tactics to the region, as conlict between Christian and Muslim intensiied in the Mediterranean zone. Ottoman adventures led to the militarisation of the Red Sea basin, too. We note these developments as much for what they hinted about the future as for their contemporary importance, which was questionable even by . European military intrusions into the interior in the sixteenth century – in Mozambique, for example – usually ended in disaster, after which they conined themselves to coastal bases; their impact, whether military or any other sort, was minimal, and across most of Africa largely invisible. In the meantime, of course, they introduced irearms to sub-Saharan Africa, while Egyptians, Turks and Moroccans carried them across north Africa, into the Sahara, and ultimately into the savannah zone. Portuguese and Turks introduced muskets almost simultaneously into the Ethiopian region. Although guns would remain largely irrelevant for a long time to come in most of these areas, it is clear enough that new drivers of violence, both economic and technological, appeared across the continent in the ifteenth and sixteenth centuries.

Expanding Horizons in North and Northeast Africa Between the thirteenth and the sixteenth centuries, the western and eastern Mediterranean and Red Sea regions became part of a vast



WARFARE IN AFRICAN HISTORY

militarised zone, interconnecting the Maghreb and northern Africa, the Nile valley, the Ethiopian Highlands and the Somali promontory. In many respects the fulcrum of this elongated zone of interaction was Egypt, irst under the Mamluks and then as part of the expanding Ottoman Empire. The Mamluks developed a remarkable military system from the mid-thirteenth century onwards, sustaining themselves by importing slave soldiers from Turkey and southern Russia; but perhaps most noteworthy is the fact that while it was a military autocracy, Mamluk Egypt was also one of the key cultural and intellectual centres of the Islamic world. Mamluk regiments of skilled horsemen succeeded in defeating the Christian Crusaders, the result of Mamluk mobility but also of Crusader over-ambition; and then in , they saw off the Mongols at the battle of Ain Jalut, in which a small Mongol force was ambushed by a Mamluk army under the gifted command of Baibars, who had served with the Mongols and now replicated their speed and audacity. As they won these wars, the Mamluks created a cultural environment of extraordinary vitality and creativity, an aesthetic and spiritual realm guaranteed by military success. Indeed, each fed the other, as the Mamluks extended their control over Mecca and Medina, further underscoring Egypt’s regional importance. And commerce was facilitated by the Mamluk pax, with Egypt positioned at the centre of a trade network linking the Indian Ocean and eastern Africa with Europe. Increasingly, however, the Mamluks were the victims of their own success: The earlier military ethos of rigorous discipline and order was diluted by material success, and in the course of the ifteenth century, the system began to weaken. An oppressive tax regime – necessary to maintain an expensive and well-appointed military aristocracy – undermined economic stability, as did, coincidentally, a series of droughts and resultant famines. Yet ittingly, it was defeat on the battleield which toppled the Mamluks – a pivotal and deeply symbolic engagement which seemed to represent the passing of one age and the advent of another. By , the horsed regiments which had held their own against Christians and Mongols appeared archaic opposite the ranks of Ottoman gunmen, the latter the harbingers of a creeping technological revolution begun a long way from Egypt. When the Ottoman Turks attacked in , the Mamluks were overwhelmed – in a battle which involved the irst large-scale deployment of irearms on African soil – and Egypt became the province of a western Asian empire. Under a viceroy appointed by Istanbul, Egypt now became a mustering station for Ottoman expansion. It was used as a base to push

THE MILITARY FOUNDATIONS



southwards, up the Nile valley and into Nubia, and also into the Red Sea, where the seizure of ports such as Suakin and Massawa and the nourishing of Somali Islam became parts of a much larger imperial project, not least aimed at countering the Portuguese advance into the Indian Ocean. But the Ottomans were also pushing westwards, too, and the Maghreb and the western Mediterranean became a key zone of armed engagement. North and northeast Africa were becoming encompassed, not for the last time, within a burgeoning global conlict, their militarisation the direct result of geopolitical importance, the armed frontier zone writ large. In the course of the sixteenth century, there was heightened conlict over control of the western Mediterranean between Christian Spain and the Ottomans, the former having expelled the last Muslim occupiers of the Iberian Peninsula, the latter expanding energetically under Selim I and then Suleiman the Magniicent. Key ports along the north African coast, including Algiers, Tunis and Tripoli, changed hands repeatedly between Spanish and Ottomans as the struggle for control of shipping escalated. Christian naval forces won a series of engagements in the s and s – most importantly in , when the Holy League, headed by Spain, overcame the Ottoman leet at Lepanto owing in large part to their superior irepower – severely constraining Ottoman activities at sea. Istanbul was more successful on the north African coast itself, inally securing Tripoli, Tunis and Algiers. Yet even here success was qualiied, as the military governments set up in the coastal provinces were soon only nominally under Ottoman suzerainty. They developed as largely autonomous enclaves, reaping the beneits of their position at the termini of trans-Saharan trade routes. Beyond the immediate hinterland, the Berbers repulsed all attempts to bring them more closely under control, just as they had resisted Romans and Arabs. More broadly, northern Africa witnessed heightened interaction between nomadic and sedentary groups – both peaceful and violent – which fundamentally transformed the demographic composition of the region. Ottoman inluence did not extend beyond the area of presentday western Algeria: Here, in the irst half of the sixteenth century, a coalition of Arab nomadic clans was extending its control over the region northwest of the Atlas Mountains, operating within the domains of the old Almohad polity. Moroccan power was formidable by the late sixteenth century, and particularly under Sultan Ahmad alMansur: The Portuguese were expelled from their Atlantic ports and



WARFARE IN AFRICAN HISTORY

recalcitrant clans were brought to heel by a reorganised army in which highly skilled cavalry was bolstered by muskets and Iberian mercenaries. Ahmad al-Mansur’s ambitions stretched far beyond the Atlas Mountains, and in  he dispatched an expedition across the Sahara to attack and subdue the savannah empire of Songhay. We examine this in more detail later; sufice it to note here that this weakened the sultanate considerably, proving expensive in terms of manpower and resources. But despite a cycle of insurgency and dynastic competition, Morocco survived, at least in rump form, the state incubated by the walled cities that stretched out between the mountains and the Atlantic shore. At the other end of the Ottoman world, in the southern Red Sea region, the Turks encountered another remarkably resilient polity, Ethiopia, whose militarism, like that of Morocco, was the product of its particular physical and cultural environment. It also provides us with arguably the most detailed example of a ‘medieval’ African military state – some of whose features are peculiar to Ethiopia, but some of which are more broadly illustrative, not least in terms of the tension between centre and periphery. In the late thirteenth century, the Shoan insurgency – presented by its leaders as the ‘restoration’ of the legitimate ruling line, the ‘Solomonic’ dynasty – gave rise to a new regime which very much deined itself through righteous warfare and in which glorious militarism was celebrated at the heart of highland culture. There was continuity in this context from the earlier Axumite state, which had likewise self-consciously projected the image of military might; in Solomonic Ethiopia, the violence which drove the state was sanctiied by God, for highland Christianity was as militarised as any of the Islamic movements of western Sahara and savannah zones, and arguably even more successful in terms of the translation of such ideological militancy into political action. In Ethiopia, this distinctive manifestation of Christianity was armed, and church and state formed a potent alliance – a partnership of mutual beneit as Ethiopia expanded in the centuries which followed Yekuno Amlak’s coup d’etat in , but one which was also deemed necessary in a dangerous and volatile region, for Ethiopians, God’s chosen people, were surrounded by inidel enemies. It was an idea enshrined in a text, the Kebre Negast, or ‘Glories of the Kings’, which the new rulers sponsored – or at least reordered and reissued – in the early fourteenth century. Here, the story is told of Ethiopia as the product of a liaison between Solomon and the Queen of Saba; of how God turns away from the Israelites

THE MILITARY FOUNDATIONS



and favours Ethiopia, the New Zion, instead; of the Ethiopians’ violent struggles against a host of enemies, but always with God at their backs, for they are possessed of the Solomonic blood line and (just as important) the Ark of the Covenant, the physical manifestation of their contract with God. This was the state’s ideological militarism, its armed Christian mission, writ large. Yekuno Amlak had been a career soldier, and his immediate successors needed to be warriors, too, as they struggled to keep centrifugal forces and armed regionalism under control in a jagged, mountainous landscape where rebel fastnesses were often easier to defend than they were to attack. It was an environment over which it was dificult to establish anything approaching hegemony: Alongside large and often cumbersome corps of foot soldiers, horses were used in the Ethiopian region, and although detailed evidence is scarce, their reach and impact was certainly more limited in the rugged, broken landscape of the highlands than in the western savannah. Amda Tsion reorganised the army in an attempt to meet those challenges, although in the long term the problem of localism and resistance to central command persisted. The system which had prevailed since the Axumite era involved a standing royal regiment of men-at-arms at the centre, under the direct command of the monarch’s close associates, complemented when the need arose by provincial forces under local rulers, whose loyalty (and that of their troops) to central authority was tenuous at best, and highly contingent. This form of military organisation was prevalent, too, across much of the continent south of the Sahara, as we see a little later; Ethiopia – while exceptional in some respects – in fact provides us with an excellent illustration of the military challenges associated with an expansion of political scale. When the system worked, it appears to have worked well. The problem, however, was that regional commanders could accumulate power and armed followings which were essentially autonomous from the royal centre, enabling them to challenge central authority. Yekuno Amlak’s own rebellion against the Zagwe exempliies the dynamic. Amda Tsion strove to strengthen the forces at the disposal of the royal centre by creating new regiments which cut across regional loyalties and which were more pan-territorial than previously; in addition, there were new units comprised of mercenaries and slaves – a system developed by Moroccans and Egyptians, too – and others which comprised recruits from newly conquered territories, an ancient strategy which persisted into the twentieth century in



WARFARE IN AFRICAN HISTORY

Africa. In essence, it involved the centre co-opting and mobilising the frontier, whose manpower was then posted elsewhere and which was paciied in the process. The command of these various new units was distributed among trusted members of the royal court, sometimes relatives and certainly close allies of the king. Over the longer term, a spirit of competition between regiments evolved, with each seeking to win the favour of the king – and such favour could have material and political implications. Moreover, while the royal court itself was itinerant, a more permanent, armed physical presence was instituted across the empire. Military colonies were established at key points, and land was distributed to senior soldiers for the purpose. Known as gult, these estates entitled soldiers to the food produced by tenant farmers as well as to tax in kind, and tied military commanders (and other loyal recipients) to the king, for such land was not hereditary, and therefore the grant itself was conditional on continued good service and loyalty. Beyond the Christian, Semitic core, populations diversiied in terms of culture, language and belief systems; unity was fragile, and ‘Ethiopians’ – especially newly conquered ones – continued to think locally. Gult warlords still accumulated enormous power, and although they were in theory the representatives of central imperial authority, the wealth they could store up, and the size of the military garrison they might develop, meant that they were the linchpins of the Solomonic order, but also that they could mount serious challenges to that order. The material tribute (including horses) which they passed to the central coffers, and the soldiers they contributed to large-scale military campaigns – failure to do either was tantamount to rebellion – meant that they empowered the emperor, the Negus Negast, or ‘king of kings’; but the emperor was dependent on their fealty at least as much as they depended on his favour. It was a potentially fragile system. The success of the centre was in many respects conditional on personality, which indeed was a deining dynamic in much of Africa’s history: Energetic, charismatic monarchs could enforce their authority and command loyalty, whereas a weak, inactive incumbent empowered provincial overlords. Crisis at the centre could lead to a shrinkage of the territorial state, and often dramatically so. Factionalism around royal succession, in particular, episodically destabilised the polity and could lead to civil war fought between warlords championing an array of candidates. Amda Tsion’s ‘glories’ were achieved mainly against the Muslim settlements which were encroaching on the highlands, but this was

THE MILITARY FOUNDATIONS



not simply holy war; it was an economic struggle, too, for in the early fourteenth century, the Solomonic state strove to re-establish control over the trade routes linking the highland interior with the Red Sea and the Gulf of Aden – part of commercial network which had been rejuvenated by the Mamluks in Egypt. The slave trade was important in this era, and there is little doubt that much early Solomonic warfare was aimed at capturing slaves who were then exported eastwards, although the wars themselves were given a supposedly more noble purpose, namely the extension of God’s glory across the region. Violence attended the territorial expansion of the state, too, as a militant monasticism followed the army across the highlands, the frontier pushed forwards by Christian missionaries and soldiers alike. Christians and Muslims were trading partners across the central and eastern highlands, but in the course of the fourteenth century, the religious frontier in northeast Africa became more violent. As the commercial port of Zeila thrived, Muslim trading states fanned out across the eastern and southeast highlands, and these were the target of a great many of Amda Tsion’s campaigns – particularly Ifat, which for a time in the s spearheaded a more uniied Muslim response to renewed Christian aggression. But Muslim disunity and internal rivalry, and Amda Tsion’s well-organised and disciplined forces, put paid to sustained resistance, and in the second half of the fourteenth century, Ifat relocated itself to the Harar plateau, where its rulers reconstituted the polity as Adal, and set about the resuscitation of Muslim power in the region. Under the Walasma dynasty, Adal sought control of the vital trade artery between Zeila and the highlands, and also initiated the more systematic conversion of the Somali pastoralists in the area. As with the Berbers in northern Africa, Somali proselytes also proved to be enthusiastic soldiers in Adal’s reformed army, which held its own against highland Christian armies during the recurrent conlict of the ifteenth century. As in western Africa, trade periodically suffered, but the assertion of identities rooted in violence was as important as any potential commercial gain in a spiritually and ethnically fractured region. In the last decades of the ifteenth century, Adal raided Ethiopia’s eastern frontiers continually. In the early sixteenth century, while Ethiopia struggled with internal dissension, its army less potent than at any time since the Solomonic coup d’etat, Adal was undergoing renewed radicalisation and militarisation under the messianic and charismatic leadership of Ahmad ibn Ibrahim al-Ghazi, who became



WARFARE IN AFRICAN HISTORY

effective leader of Adal in the s. Ahmad ibn Ibrahim – known as ‘Gran’, the ‘left-handed’ – was a senior soldier-turned-kingmaker who deposed and killed the reigning sultan, replacing him with his brother. He titled himself imam and preached the need for renewed Islamic purity – in particular through self-discipline and abstinence – in which he was supported by an array of missionaries and activists arriving from the Arabian peninsula; and he called for jihad, seeking to overcome Somali and Afar clan differences through promises both spiritual and worldly – of afterlife glory for the slain and material and political gains for those who might survive the forthcoming showdown with the Christian enemy. Armed with Turkish irearms, Arabian missionaries, Somali convert soldiers and Islamic zeal, in , Ahmad launched jihad against Christian Ethiopia in what would be one of the deining wars of the region’s history. An early encounter at the battle of Shimbra Kure, not far from present-day Addis Ababa, saw the Adal army overcome Emperor Lebna Dengel’s forces. As Ahmad’s army advanced, it swept along groups hostile to Ethiopia, and its ranks swelled; pushing into Shoa, the Muslim force inlicted a series of heavy defeats on the Ethiopians and steadily occupied the southern and central parts of the kingdom by the mid-s. Lebna Dengel had mobilised a large army, from Tigray, Shoa, Begemedir and Gojjam: Size alone was not the issue. The problem was its cumbersome and rigid nature, and the rivalries and jealousies which paralysed effective command – a recurrent problem for pan-Ethiopian armies. Ahmad’s forces, by contrast, were well armed and led, and organised into highly motivated, lexible units. The Christian state teetered on the brink, its unity shattered, its defences blown apart, and its citizenry either killed or forcibly converted. Muslim governors were appointed over conquered provinces; churches and monasteries were destroyed. Lebna Dengel was reduced to the role of an itinerant leader of a rump army, avoiding battle when he could, engaging when it was either unavoidable or advantageous to do so. Lebna Dengel had managed to send off a plea for assistance to the Portuguese before his death in . They contributed a small force of musketeers, but their intervention was not straightforward: They were defeated by Ahmad’s troops in , and their commander, Cristovao da Gama, was killed. They had guns, but so did the Turks in Ahmad’s army. Nonetheless, they did play a part – how important being a matter of debate – in the defeat of Ahmad’s forces in early . Ahmad

THE MILITARY FOUNDATIONS



himself was killed in the ighting near Lake Tana, and demoralised and exhausted – their zealous unity revealed to be rather more supericial than it had irst appeared – Adal’s overstretched Somali army disintegrated and drifted back eastwards. The Ethiopians, now under Galawdewos, retook Shoa and set about rebuilding their defences. The Solomonic state had survived, but the polity was fundamentally shaken and profoundly weakened. The interconnectedness of local economies meant that commercial interaction was absolutely vital, and coexistence continued; still, Ethiopia was conscious of its anomalous position in a predominantly Muslim region. Indeed, ighting with Adal continued sporadically: Galawdewos himself was killed ighting Ahmad’s successor, Emir Nur, in . In the wake of the devastating war, there were new encroachments on ‘fortress Ethiopia’, this time from far to the south, in the form of the Oromo migrations which had begun by the middle of the sixteenth century. The Cushitic-speaking Oromo, predominantly pastoralists originating in the grasslands northeast of Lake Turkana, took advantage of the devastation and moved into the depopulated and ravaged highlands in search of new pasture. Their appearance in the war-torn Ethiopian region at this moment caused great consternation at the time, and has continued to fuel polemics ever since. Some Oromo groups were armed and hostile, but many settled peacefully in the southern and central highlands, taking up Christianity and agriculture and assimilating into Semitic society. Others adopted Islam, especially those in the eastern highlands and around Harar. In time, they became key actors in both political and military terms, as Oromo recruits swelled the ranks of provincial armies and contributed their skills on horseback and their own distinctive military cultures to the Ethiopian military tradition. But contemporary chroniclers depicted them as barbarous hordes, taking advantage of Ethiopia in its weakened state, and negative imagery would continue to haunt them in the centuries which followed. Similar imagery, likewise generated by Amhara chroniclers, adhered to the Somali. The Oromo were organised around the age-grade system known as gada – as much a philosophy as a military structure – which affected all aspects of society but which was particularly associated with militaristic pastoralism. The gada system deined male activities in eightyear cycles; young warriors could gain honour and inluence through raids on neighbouring societies, with youthful aggression marshalled for the defence of the community. Recent studies of the Oromo in



WARFARE IN AFRICAN HISTORY

history tend to underline the image of the proud, armed pastoralist, given to expressions of controlled, masculine violence; certainly, the gada system made a deep impression on highland Ethiopians, who were prompted to examine their own social and military structures as a result, and who increasingly absorbed Oromo into those structures, even while they feared and despised these cattle-keeping barbarians. The stark truth, as one chronicler, a monk called Bahrey, wrote in , was that the Ethiopians had forgotten how to ight, and could learn much from these ferocious invaders – who, again, were grudgingly admired for their skill on horseback, which was arguably, in time, their most important contribution to highland military organisation. Some moved into the southern and western uplands, where they subverted their pastoral egalitarianism to found agricultural and commercial monarchies which subdued local Sidama and Omotic peoples; and other groups spread into the central highlands, their expansion facilitated by the shrinkage of the Ethiopian state – a deliberate policy of territorial retrenchment undertaken by Emperor Sarsa Dengel (–). Sarsa Dengel is notable for his limited adoption of irearms; but while inevitably the adoption of the new technology might be described as marking something of a ‘new era’, guns were not widely used in Ethiopian armies for another  years. It seems reasonable to speculate that he made limited but effective use of irearms while relying primarily on spear- and sword-wielding infantry – organised according to province – supplemented increasingly by cavalry drawn from the same Oromo groups with whom he was ighting. There was military reorganisation, too, as Sarsa Dengel withdrew many southern garrisons, built up strength in the north and sought the creation of a more readily deployed force. Restructuring was certainly necessary, owing to the fragility of the polity: The loss of so much territory in the course of the sixteenth century had deprived the political centre of the most important resource with which to purchase support, namely land in the form of gult estates. The state had become more predatory, and the armed nobility more restive; increasingly, there was the problem of praetorianism, which meant that the central army asserted its right to intervene in the royal succession, which it did repeatedly in the early seventeenth century when it perceived the gult system to be under threat. At the same time, provincial forces took to raiding peripheral peoples on their own account rather than serving the central cause in order to bolster local revenues. Little wonder that Ethiopians admired,

THE MILITARY FOUNDATIONS



even if grudgingly, the eficient martial prowess of the new arrivals. This was recognised by Emperor Susenyos (–), who owed his succession partly to Oromo support and who thus initiated the risky strategy of integrating Oromo units into the central army, and deploying these forces against other, more hostile Oromo. The regimental system created by Amda Tsion had been thrown into disarray by the jihads, and Solomonic power now needed to be carefully reconstructed and, above all, renegotiated between centre and province. Sarsa Dengel had been a gifted warrior who successfully confronted a number of enemies, but he failed to address the mounting socio-political crisis in Ethiopia which was the result of prolonged warfare. Military magnates might be mobilised in defence of the Solomonic ideal, but they could just as easily take up arms against those at the centre who claimed its inheritance. It was a pattern of military organisation – and of violence – which would be prevalent across much of Africa, and in this sense Ethiopia, however distinctive culturally and ideologically, is thematically illustrative. By the early seventeenth century, the army – or, more appropriately, armies – had become the decisive force for change and/or continuity in the Ethiopian polity. As soldiers increasingly comprised a distinct social class aware of its rights and privileges, that class’s great weight was supported by those who worked the land on estates parcelled out for the purpose; it was a system which at times would come under enormous strain.

Shifting Centres, Advancing Peripheries: Saharan and Savannah Militaries Particular patterns of military state building – involving the mobilisation of resources on a large scale and over long distances – are discernible in the early second millennium in the savannah and southern Saharan zones. Indeed, in many respects, the empire of Mali – arguably one of sub-Saharan Africa’s most remarkable and successful military polities – represented the apex of one particular model of development. Founded through armed insurgency – the uprising of the Mande under the inspired leadership of the legendary Sunjata – Mali emulated Ghana in that it rested ultimately on the effective utilisation of arable land, as well as control of the vital gold trade. The unifying drive at the heart of Sunjata’s empire involved the surrendering of local powers – vested in the village head, the mansa – to the political



WARFARE IN AFRICAN HISTORY

centre: From the outset, Sunjata had proclaimed himself sole Mansa, and successive leaders followed suit. Islam was widely promoted, particularly under Mansa Musa (–), although not at the expense of local spiritual orders – an indication, again, of the fragility of the empire at the level of the kafu. The class of professional merchants which emerged in the course of the fourteenth century was protected by a greatly enlarged military apparatus. The army guaranteed the commercial viability of the state, the vibrant economy based on a cowry-shell currency protected by an army which was supported, in turn, by the proits from long-distance trade, as well as by the tribute paid by the far-lung polities over which Mali exercised suzerainty. The horse was critical across the savannah, because owing to dispersed population and to landscape characterised by dry pastureland, it was dificult to mobilise and move large numbers of infantry over long distances for long periods. Cavalry was essential to centralisation; horsemen represented the offensive and predatory wings of the militarised polity, with bow-using infantry deployed in the rear and, generally, more defensively. Cavalry also implied social stratiication, as the horsed warrior represented an investment in terms of training as well as inancial resources, giving rise to a more professional ethic and in turn to codes of honour and nobility which became an integral part of horsemanship. In some respects Mali represented a more eficient manifestation of the military model developed by the rulers of Ghana; but Malian elites inherited the same problems, too. The power of the kafu had endured, for all the centralising innovation of the empire. Centrifugal forces and continual rebellion in outlying provinces exposed the limited capacity of the army, while the cumulative loss of territory meant an attendant loss of the tax revenue essential to the funding of such large cavalry forces. Moreover, Mali’s army had no monopoly on decent horses: Its enemies had them, too, and just as Ghana had been gradually enervated by prolonged frontier conlict with the Sanhaja Berber, so nomadic groups on horseback, and in particular the Mossi and the Tuareg, kept up a pattern of raiding on Mali’s frontiers which amounted to a low-level war of attrition, and which drained the military and inancial resources of the state. Mali’s inability to protect the great seat of learning, Timbuktu – which it lost to the Tuareg in  – was a clear indication of decline. Indeed the Mossi and the Tuareg had become the nemeses of the imperial centres of the savannah, representing an alternative, more luid and more mobile model of military

THE MILITARY FOUNDATIONS



organisation in the region. By the beginning of the sixteenth century, Mali’s dominions had been reduced to the Mande heartlands, and even the federation which governed the imperial rump had largely dissolved by the end of the seventeenth century. While the dreams of a great Malian national revival persisted until the nineteenth century – and the stories of Sunjata’s military glories echoed further still – Mali’s place was taken by Songhay, long governed by Malian overlords. A process of amalgamation had been underway for some time, and in the ifteenth century it asserted its independence, again through force of arms. Its cavalry was complemented by infantry and a leet of canoes operating along a lengthy stretch of the central Niger, one of the earliest recorded – and indeed one of the very few – examples of waterborne warfare, certainly in sub-Saharan Africa. The military foundations were laid by Sonni Sulayman Dandi, but Songhay’s great warrior-statesman was Sonni Ali, ‘the Great’ to posterity, who reigned from  to  and who effectively ensured that Songhay eclipsed Mali as the region’s dominant power through military conquest. To Arab and Muslim chroniclers, he was a barbaric backslider who displayed a violent disrespect for Islam; to Songhay chroniclers, he was their Sunjata, the conquering hero who imposed a new order on the turbulent savannah. He spent much of his reign ighting the predominantly Muslim Tuareg, from whom he wrested Timbuktu in , and the Mossi, much like his Malian predecessors; just as the Mossi raided farming communities under Songhay’s protection, Songhay horsemen made proitable forays into Mossi territory. Yet Sonni Ali’s exploits also demonstrated the dilemma of the centralising warrior, for in the late ifteenth century his wars caused considerable disruption to the trade on which Songhay, like Ghana and Mali before it, so depended. If Sonni Ali’s imperialism had been largely political and military, that of his successor, Muhammad Ture (–), was a blend of the economic and the spiritual. Notably, he declared a jihad against the Mossi in , which endeared him to Muslim chroniclers – in contrast to his violent predecessor, who had waged war on his co-religionists. For Muhammad Ture, the army was no less important, but the militarism was perhaps less overt. Commerce recovered, and for much of the sixteenth century, Songhay was a trading empire irst and foremost, exporting gold, slaves and a range of agricultural produce in exchange for cloth, salt and horses. Nonetheless, Muhammad Ture’s programme of consolidation had had limited success, for even while



WARFARE IN AFRICAN HISTORY

trade lourished, Songhay was periodically – and cumulatively – weakened by factionalism and eruptions of civil war. Once again, regional overlords and disaffected members of the Sonni dynasty, each basing their strength on a combination of horse and foot, deied central authority and took advantage of porous frontiers. The internal threat to the savannah polities was real enough, whether from provincial rebellion, recalcitrant frontiers or local adversaries; in the case of Songhay, however, an additional menace approached from the north. The Moroccan expedition across the Sahara of  culminated in one of the earliest recorded pitched battles in West African history, that at Tondibi on the northeast bend of the Niger, between a Moroccan force of cavalrymen and musketeers and the horsemen and infantry of Songhay. Interpretations of the clash differ, but it may still safely be regarded as a pivotal moment in the region’s military and political history. From a distance, it appears a fundamentally unbalanced contest: Morocco, its military recently reorganised, equipped with muskets and led by dynamic and ambitious commanders, versus Songhay, weakened by internal issures over several decades, lagging behind the enemy in technological terms, with its only advantage an understanding of local climatic and environmental conditions. But the battle of Tondibi needs to be understood in more nuanced terms, and it is certainly inadequate to perceive it as representing the irst advance of inexorable modernity on a backward periphery – just as it would be teleological to regard the irst Portuguese armed excursions, discussed later, in the same manner. Morocco’s expansionist-minded Sultan Ahmad al-Mansur, driven by a desire to more closely control lucrative trans-Saharan trade networks, as well as by heightened competition for gold supplies, had dispatched a series of diplomatic missions to Songhay after  whose primary objective appears to have been the gathering of intelligence. At the beginning of , a Moroccan force of some , men – the bulk of whom were Iberian mercenaries, including a number of recent Muslim converts – started out across the Sahara under the command of Judar Pasha with the aim of subduing Songhay and breaking its hold on the southern end of the commercial system. They took with them horses, muskets and some light cannon. Several weeks later, the ravaged and depleted army reached the Niger, having lost around , troops from dehydration and starvation, and while the survivors regrouped, Songhay mobilised. The armies met on  March, and the Moroccan force of around , found itself massively outnumbered

THE MILITARY FOUNDATIONS



by the Songhay army, although the sources vary in their estimates of the latter, from , to , horsemen, and from approximately , to , infantry. Seeking to exploit home advantage and the supposed weakened state of the Moroccans, the Songhay army advanced – in what may be assumed to be typical battleield formation – with its infantry in the centre driving a thousand cattle before it and cavalry on the wings, but the Moroccans marched to meet them, likewise pushing their cavalry toward the Songhay lanks. The Songhay tactic of advancing behind cattle backired when the Moroccan musketeers began shooting, startling the animals and causing a backward stampede. The Moroccans then rushed forward, but they encountered a new line of defence, as the Songhay archers – an elite force, according to the sources – held irm for a time, their ankles fastened to their calves, famously, to make retreat dificult if not impossible. But the defence was soon exhausted, while the Moroccans repulsed a further attack by the Songhay cavalry; and ultimately, the Songhay army withdrew, leaving the ield to the Moroccans. The battle illustrates the kind of formation noted earlier: cavalry as an offensive wing (if, in this case, ineffective when confronted by disciplined musketeers), with bowmen situated more defensively. On the face of it, initially at least, total victory had been achieved: The Moroccan army put much of the surrounding area to the torch and then advanced into Gao, the Songhay capital, where it inlicted heavy damage. Moroccan troops also occupied the cultural and spiritual prize of the region, Timbuktu. Certainly, an older standard interpretation held that the Moroccan triumph was unconditional and that it spelled the end of the Songhay empire; according to this view, moreover, the introduction of irearms signiied the beginning of the end of cavalry in the savannah and heralded the beginning of an era deined by superior foreign technology. Each of these positions, however, has been seriously qualiied. The Moroccan victory was in no way absolute; rather, the invasion sparked a prolonged war of attrition which lasted for another twenty years, and the evidence suggests that the Moroccan commander had not anticipated a permanent occupation and was awaiting orders to return. Indeed, Songhay offered an indemnity and an acknowledgement of Moroccan suzerainty in return for which the Moroccans would withdraw, but Sultan Ahmad al-Mansur rejected the proposal and instead, in the course of the s, dispatched a further , troops to the region to inish Songhay off. But the Songhay cavalry overcame their initial awe of the new weaponry – the



WARFARE IN AFRICAN HISTORY

noise and smoke of the muskets had probably been at least as important as their physical impact – and adapted rapidly, becoming ever more successful in their engagements with the Moroccan musketeers, who were soon reduced to the role of roaming marauders. Malaria took its toll on the invaders too, as did the hills and swamps of the area. The goal of controlling the gold trade remained hopelessly out of reach, and confronted with both disease and resistance, the Moroccans were unable to impose themselves beyond the Niger bend. The Sultan sought to establish a military administration of the supposedly conquered territory, but over the next few years the invaders lost contact with Marrakesh, with local commanders creating their own miniature dominions and throwing off any pretence of fealty to Morocco. They and their descendants in the seventeenth century constituted a disruptive new element where they settled, especially in the environs of Timbuktu. As for Morocco itself, the whole project proved damaging and costly, and the sultanate would enter a period of decline. Nonetheless, it is true enough that Songhay’s power had been broken, while the internal crises which had plagued the state through much of the sixteenth century now re-emerged. Much diminished by the Moroccan war, the central state had no further resistance to offer regional leaders with no interest in loyalty to a lost cause, and by the early seventeenth century, Songhay had largely ceased to exist, replaced over the next few decades by a chain of smaller polities. In terms of military technology, the Moroccans’ muskets had clearly been signiicant in the context of the actual battle; but guns would have a limited impact on the savannah beyond Tondibi itself, and cavalry remained the dominant mode of warfare in the region until deep into the nineteenth century. Localised equine military power had proved resilient; the kafu had reasserted itself. There were successor states, and of especial signiicance was the ‘raiding state’ phenomenon which in many respects dated back to the early and mid-irst millennium, but which would come increasingly to the fore. Raiding states were based on ‘war houses’ established across the southern fringes of the Sahara and in the savannah, engaged in slave raiding and the violent production of booty, sustaining warrior elites and the continual reproduction of often short-lived warrior states. These groups, representing the shifting and fragmentary nature of regional politics and warriorhood, would also be mobilised by Islamic revivalism; and by the nineteenth century, raiding states were the restless drivers of much violent change, as we see later. Islam,

THE MILITARY FOUNDATIONS



moreover, would continue to spread across the region. Nonetheless, the savannah zone entered an era of political turbulence and economic stress. Whereas the main trade routes shifted eastward towards the central Sudanic zone, evidence also suggests increasingly unreliable rainfall in the Niger valley and an upsurge in the incidence of famine, dispersing populations away from an erstwhile fertile belt that had formed the backbone of a series of dynamic militaries stretching back over a thousand years. For, although the Niger valley could support dense populations and complex political cultures when the rains were good, this was an environment which remained essentially fragile and prone to crisis. Indeed, through the sixteenth century and beyond, heightened competition between herders and farmers and predatory horsemen in an increasingly dificult landscape destabilised states and thwarted centralising political ambitions while militarising communities and creating violent frontier zones. These were communities and frontiers, however, which would yet be mobilised by, and co-opted into, much larger causes – religious and political.

Founding Fathers and Militarised Mythologies: War, State and Society in Sub-Saharan Africa During the irst half of the second millennium, there was striking diversity across the subcontinent, but broad patterns emerged. Numerous polities represented pioneer political organisms imposing new political and economic order on larger areas hitherto witnessed south of the Sahara; indeed, the celebration of founding fathers creating civilisation out of the wilderness, and subsequently imposing new systems on savage neighbours, was at the heart of the traditions and mythologies which formed the ideological basis of the new states themselves. They were the products of fertile and frequently violent frontiers, from the forested Atlantic coast to the central grasslands, from the great riparian cultures of the Congo and Zambezi valleys to the lacustrine region of eastern Africa. Kings were often spiritual war leaders, or the temporal links to the gods of war; but of particular interest is the association of sacred or semi-sacred kingship with the business of ironworking. Royal regalia commonly included the blacksmith’s tools – and it is probable that the superiority of iron weaponry, monopolised by a few, facilitated the development of military states which thereafter associated leadership with the new technology. Of course, in addition to the



WARFARE IN AFRICAN HISTORY

pragmatic links between military commanders and suppliers of superior weaponry, there also may have been a less worldly association: Both kings and blacksmiths controlled supernatural forces. More broadly, there was a symbiotic relationship between war and economic development. Territorial expansion brought arable land and pasture under control and led to the further growth of productive surpluses; furthermore, effective military organisation enabled the channelling and protection of long-distance trade routes. Both facilitated, if not necessitated, state formation; agricultural surpluses and commercial wealth further enabled the specialisation of labour and supported the creation of more elaborate military systems. Meanwhile, political structures were developed which resembled the spokes of a wheel, whereby the centre delegated power to provincial governors who were frequently military overlords with more or less direct links to the kingship. They were responsible for defensive and offensive war, the patrolling and extension of frontiers and the periodic recruitment of soldiers as and when needed. It is clear that such polities emerged in response to particular environmental and demographic circumstances, although again militaries had broad structures and objectives in common. Wars were fought to secure external resources – including land, water and commercial highways – and for slaves, especially women, who were wanted for labour and reproduction, thus rendering a clear link between external and internal processes of violence. Owing to economic and demographic pressures, most polities did not have full-time permanent armies, but rather part-time forces raised according to need; often they were concerned with their immediate, adjacent frontier zones, through which the discontented escaped to found their own rival polities, and which were also proitable hunting grounds for armed entrepreneurs, whether at the level of the state or of the individual. And owing to the closely integrated nature of society itself, there was little or no separation between the political and the military, but rather the army was highly political, intimately involved in politics, and the kingship was invariably associated (at least at the rhetorical level) with command in war and military prowess. War, meanwhile, may have been a predominantly masculine affair, with the prowess of both service and leadership intimately linked to the iner manly attributes; militarism, indeed, was frequently the clearest manifestation of ideas about masculinity. But although women were clearly among the primary targets of military campaigns – and societies were armed, in the irst instance, to protect childbirth – they

THE MILITARY FOUNDATIONS



were also central to military activity itself, providing essential logistics, cheerleading and critiquing, as well as, in elite circles at least, inluencing war policy. They could be military leaders, as well, as the story of the Ndongo queen Njinga, outlined later, demonstrates. As ever, the substantive evidence is slight, and igures such as Njinga appear exceptional rather than conventional; but it must be assumed that women were indeed at the heart of warfare, not only as victims, but as protagonists, too. A brief survey reveals a number of regional variations on the core themes outlined earlier. In central-southern Africa, for example, the earlier civilisation of Mapungubwe was superseded in the fourteenth century by Great Zimbabwe, whose famous stone walls may have been defensive structures in origin but which became architectural elaborations on the theme of omnipotent, charismatic kingship. Great Zimbabwe’s armies enforced tribute across the plateau and protected one of the most lucrative trade routes in sub-Saharan Africa, the one linking Zimbabwe’s gold mines with Kilwa on the coast. When overcultivation and commercial shifts caused the abandonment of Great Zimbabwe in the middle decades of the ifteenth century, a new state emerged farther north. Mutapa’s ruler bore the title Mwene Mutapa, ‘Conqueror’, and it was a heavily militarised state from the outset. By the late ifteenth century, in addition to establishing control over the Shona, Mutapa’s armies ranged from the central Zambezi valley to the coastal lowlands, funded by proits from the gold trade and drawing tribute from a wide area to supplement the state’s impressive income. Only the appearance of the Portuguese on the Swahili coast threatened their position, although the pattern of militarised hinterland power in uneasy symbiotic relationship with commercial port settlement was already well established by the time they sailed over the horizon. In eastern Africa, by the early centuries of the second millennium, Bantu-speaking iron-wielding farmers were cutting back woodland to make way for cultivation, thus creating an environment in which dynamic militarisms, underpinned by integrated economic systems, could lourish. Between the eleventh and the ifteenth centuries, a series of small chiefdoms emerged, and by the middle of the ifteenth century, a number of these had merged to form the state of Kitara. Relatively little is known about Kitara beyond its earthwork defences and the importance of its ruling dynasty, the Chwezi; but it was soon eclipsed by a series of migrations into the eastern African region from



WARFARE IN AFRICAN HISTORY

southern Sudan and southwest Ethiopia between the ifteenth and the seventeenth centuries. Much of the region’s subsequent military history is shaped by the arrival of the Nilotes, whose migration may be compared in some respects to that of the Oromo in the Ethiopian region. Probably driven south by drought and land pressure, they brought pastoralism and the cultures and economic systems associated with livestock; more importantly for our purposes, their gradual arrival introduced new military systems to the region. The Western Nilotes had entered northern Uganda by the mid-ifteenth century, from whence they dispersed in small groups around Lake Victoria. One group, the Bito clan, appears to have ejected the Chwezi – the details are hazy – and created a new dynasty in a new state, known as Bunyoro. Related groups had an inluence in the establishment of neighbouring Buganda, a small chiefdom at the north of Lake Victoria, in much the same period, whereas still others gave rise to pastoralist groupings in Nkore (where they were known as Hima) and Rwanda and Burundi (Tutsi). In no sense was this a series of ‘conquests’, perhaps with the exception of Bunyoro, although violent conlict was clearly often involved. Generally, however, Nilotes interacted with Bantu-speaking communities – themselves only comparatively recently arrived – to build new polities. These latest immigrants did not introduce the ‘state’ to the region – centralised polities had emerged prior to their arrival, and can probably be associated with the irst Bantu-speaking communities – but they did greatly enhance it and contributed to a considerable enlargement of scale and vision. Moreover, in many places, the immigrants constituted militarised ruling elites whose impact is recorded in the oral traditions of the kingdoms involved. Bunyoro, the dominant power in the sixteenth century, used agricultural and pastoral wealth to maintain a roving army whose main function was to extract resources from the surrounding area, including more cattle. In the course of the sixteenth century, too, Buganda emerged as a small, centralised state whose stability was rooted in the banana plantation, which facilitated a relatively dense population. This, combined with the needs of defence against Bunyoro, facilitated the rise of an armed centre, although it would be some time before Buganda could properly compete with Bunyoro. A chain of new polities was emerging in the lacustrine region in the course of the sixteenth century – including Bunyoro, Buganda, Toro, Nkore, Rwanda and Burundi – which had in common highly hierarchical systems, centralised kingship with important symbolic and

THE MILITARY FOUNDATIONS



ritual functions, and provincial governorships responsible for military mobilisation and resource extraction. They arose in one of the most fertile belts in sub-Saharan Africa, characterised by rich volcanic soil and good rainfall, sustaining a denser population than was possible elsewhere; this combination awarded these polities a level of stability and permanence lacking in other areas. Yet violence was intrinsic, too, manifest in the institutionalisation of resource conlict, and in the periodic civil wars which erupted around royal succession – political contests which came about because monarchs routinely killed or exiled family members who constituted a threat. As a result of both economic and political warfare, ruling dynasties forged militaristic cultures which made much of their superior immigrant stock and their martial prowess, and sponsored traditions which told of righteous civilising war and offered salutary lessons for good citizenship around ideas of heroism and manly courage. These traditions, orally transmitted, emerged in much the same way – and according to many of the same needs – as the epic of Sunjata in Mali and the Kebre Negast in Ethiopia. War and what might be achieved during and as a result of it – both individually and collectively – was at the heart of these states’ self-deinition. Violence was pragmatic, but it was also a profoundly moral dynamic, if employed by good, righteous people in pursuit of legitimate aims – all of which were, of course, entirely subjective. In the meantime, other Nilotic migrations did not lead to territorial or centralised states, but to other forms of social organisation no less militarised. Notably, for example, the Eastern Nilotes moved in an arc encompassing northeast Uganda, western and central Kenya and northern Tanzania, and this region saw the rise of the Karamojong and the Maasai, broadly decentralised societies which privileged cattle-keeping, eschewed amalgamation with Bantu-speakers and – like the Oromo – created systems and cultures of military service around age-sets. We know next to nothing about the nature of warfare itself in eastern Africa during this period, although Rwandan tradition recalls an invasion of Nilotic warriors as ferocious as they were numerous, equipped with iron spears and immensely strong shields which easily delected Rwandan arrows. They committed the barbarous sin of killing cattle for food while on the march, and devastated communities in their path; women and children were enslaved and driven in front of the main army. More generally, it is clear that charismatic and ambitious leaders cultivated client networks through cattle gifts, and territorially expanding relationships of debt and service formed the basis



WARFARE IN AFRICAN HISTORY

of political development in the thirteenth and fourteenth centuries. In the course of the sixteenth century, kings harnessed these networks with a view to consolidating political and military power, using client communities as the manpower for military expeditions which – equipped with bows, spears and sturdy wooden shields – ranged ever further aield and clashed on the increasingly violently contested frontier zones between expanding polities. Yet alongside such centralised structures were political organisations without kings or provincial warlords in the sense of Bunyoro or Buganda, but whose male youth was just as readily mobilised for attack or defence, as the need arose. These ‘non-state’ systems, characterised by pastoral militarism, emerged cheek-by-jowl beside centralised, agricultural polities in the early centuries of the second millennium, and were just as important as models of African militarism as the territorial hierarchies. Maasai and Karamojong age-sets, in which spear-wielding young men tended cattle and undertook raiding expeditions on behalf of the community, had much in common with those of the Oromo further north. In much the same formative period, in the Atlantic zone, a range of polities emerged, some of them broadly similar to the interlacustrine model, others taking the form of compact city-states which often extended inluence over wide areas. In the course of the ifteenth century, the Kongo kingdom emerged near the mouth of the eponymous river, its extensive tributary system and eficient military order sustained by the produce of the fertile heartland. City-states emerged in the coastal forest of Nigeria and Benin in the eleventh and twelfth centuries, where the Yoruba built a string of kingdoms which traced their lineages to a mythical founding father, Oduduwa. Ife was the earliest and most important of these. Then there was Benin, the creation of the Edo-speakers of the Niger Delta, where kingship was emerging in the eleventh century; Benin’s famous bronze sculptures, although actually made of brass, date to the fourteenth and ifteenth centuries and clearly celebrate warrior prowess and the military power of the oba, the king, in a manner just as stunning and articulate as Sunjata’s epic or the Kebre Negast. Further west, the Akan states of modern Ghana – particularly Denkyira, Akwamu and Fante – were emerging in the fourteenth and ifteenth centuries in response to heightened competition for land, gold and trade; several of these would at length coalesce to form the Asante state. The key external dynamic for Atlantic societies was the slow growth, in the course of the sixteenth century, of the long-distance

THE MILITARY FOUNDATIONS



slave trade to the Americas. From the early sixteenth century, with the opening of the trans-Atlantic route to the Americas, the transport of African slaves to the ‘new world’ increased steadily as a result of heightened demand among European traders. The numbers remained fairly constant for the irst century or so: From the s – with the irst trans-Atlantic shipments – until the s, at least , slaves a year, as an average and according to available records, were landed alive in the Americas. In the s and s, the numbers involved increased markedly, as the Portuguese were joined by English, Dutch and French traders. Again, however, it is important to note regional variation: The Senegal-Gambia area, for example, was a key source of slaves throughout the sixteenth century, as was the area of presentday Angola. West-central Africa – the area of the Kongo kingdom – was also heavily involved from an early stage. Nevertheless, although it varied from place to place, the Atlantic slave trade clearly had a major – and cumulative – impact on the military history of this vast region. The general tendency was towards the militarisation of state and society in the Atlantic zone, largely because the supply side of this new economic system – ‘new’ in the sense that the slave trade was driven exclusively by external factors, and more intensively than anything earlier – required militarisation and the regular deployment of armies in slave-gathering for export. In bad neighbourhoods, dominoeffect militarisation occurred, as communities armed themselves to defend against predatory states (as states invariably were); in tandem, moreover, there emerged cultures of militarism which reiied warfare and celebrated the achievements won by men-at-arms – indeed, often going so far as to suggest that war was necessary to the very existence of the community, and that the sacriice of war captives was essential to placate judgemental ancestors. The impact was by no means uniform. In the course of the ifteenth century, Benin, for example, was becoming an extensive urban civilisation with a permanent army and a considerable territorial reach, and when the Portuguese arrived they encountered an expansionist, dynamic kingdom willing to sell the war captives who were a by-product of its wars. The arrival of Portuguese buyers on the shore was economically fortuitous for Benin, but once the process of territorial expansion slowed down, in the sixteenth century, there were no more slaves for sale in Benin, and its rulers did not pursue the activity, offering ivory, cloth and spices instead. As it turned out, Benin was unusual. By contrast, by the beginning of the seventeenth century, Oyo was



WARFARE IN AFRICAN HISTORY

establishing itself as one of the key powers in the West African hinterland, one whose position owed much to the expanding slave trade, as we will see in the next chapter. Systematic predation in response to external opportunities underpinned this new kind of statehood. But it is perhaps Kongo which demonstrates the early impact of the slave trade most powerfully. With the establishment of diplomatic relations with the Portuguese in the s, the Kongo elite sought, among other things, Portuguese weapons and soldiers and espied an opportunity for a useful military alliance. But it proved a fundamentally destructive partnership, for as the Portuguese lent muskets and musketeers to the Kongo army’s slave-raiding expeditions – which in turn enabled Kongo to widen its area of operations – a debate opened up within the royal court over the merits, or otherwise, of involvement with the foreigners. Some favoured it, but others saw in it portents of disaster. The former carried the day, and a new king, the Christian convert Afonso I (–), relied increasingly on Portuguese armed support; the Portuguese, meanwhile, used Afonso to access ever greater numbers of war captives to be sold as slaves, especially from the s onwards with the opening of the trans-Atlantic routes. Armed factionalism increased – as in Ethiopia a few decades later, the Portuguese presence prompted a violent backlash – and the kingship, which had initially used foreign arms to bolster its authority, now saw its base weaken in the eyes of rival groups which associated beleaguered central authority with the foreigners. Kongo’s downfall was caused at least partially by the fact that the Portuguese had been allowed to reside in the capital – something that was not permitted either by Benin or by the later slaving states of Asante or Dahomey – which meant that irearms lowed freely and unregulated not only to the state, but also to rebellious provinces and dissident subjects. By the end of the sixteenth century, Kongo was profoundly unstable, and in the course of the seventeenth it would disintegrate into various rival factions, each ighting the other for access to war captives for export. It was the violence at the heart of the Atlantic commercial system writ large. It was much in evidence elsewhere in central Africa in the same period, as well. Further south, at the burgeoning sixteenth-century slaving station of Luanda, traders purchased slaves from the Ngola, the king of Ndongo, who built up his military capacity and the range of his operations as a result. Violence heightened across the hinterland as war represented new economic opportunities.

THE MILITARY FOUNDATIONS



The Imbangala – highly mobile professional warriors who overran much of the Ovimbundu area and who invaded the coast in the s and s – best demonstrate the phenomenon. Their origins are uncertain: Whereas some traditions linked them to the central Angolan uplands, earlier accounts associated them with what may have been a renegade faction of a ‘regular’ Ovimbundu army. It is certainly clear, in any case, that they were the product of commercial and political upheaval in the late sixteenth century. These were skilled mercenary soldiers – hiring themselves out to whoever was willing to pay, whether European or African – who were thus also entrepreneurs, armed adventurers seeking both martial and material gains, with a reputation for ferocious pillaging, infanticide and (almost inevitably) cannibalism. They were organised according to a complex command structure and sustained their ranks by capturing and integrating young boys into the Imbangala system, which found further cohesion through spiritual sanction. By the middle of the seventeenth century, there were several provinces under the control of Imbangala bands stretching across central Angola. They represented not only an increasingly familiar pattern in sub-Saharan African warfare – the armed mobile band, the military frontier in motion, inding cohesion in a combination of ideological and pragmatic violence – but an early indication of the military revolution more commonly associated with the nineteenth century, namely the newly militarised society shaped by increasingly professional soldiery. Their impact across a swathe of central Africa was considerable, and is perhaps best demonstrated through the story of the remarkable warrior queen Njinga Mbande of Ndongo. Njinga recognised the potential of the Imbangala early on – both as a model of militarism and as hired help. When a combined Portuguese force using Imbangala mercenaries destroyed Ndongo between  and , the ongoing war between Portugal and Ndongo entered a new and bitter phase: Throughout the s, both Njinga and the Portuguese deployed Imbangala mercenaries. Njinga was defeated in , and she withdrew eastwards, unsuccessfully seeking an alliance with Kasanje, the most powerful Imbangala band in the area. Instead, she underwent initiation as an Imbangala herself, and emulated the Imbangala model in rebuilding her army, leading numerous successful campaigns against a range of enemies in the s and s, before her death at a ripe old age in . Contemporary European documentation on the Imbangala notwithstanding, evidence for African tactics and formation is hard to



WARFARE IN AFRICAN HISTORY

come by for this earlier period. For example, little is known about warfare in Atlantic Africa prior to the slave trade. As we noted in the previous chapter, the enlargement of political scale in the equatorial region and in west-central Africa meant more urgent security issues, and this engendered military innovation, in particular greater cooperation across wider areas and the organisation of larger forces, both of which appear to have involved more destructive warfare. What is certainly clear is that a martial ethos predated the violence of the slave trade era, conirmed by linguistic, material and oral evidence of the intensiication of warfare between approximately  and , the organisation of young men into age-sets – notably in western and central Angola – and innovations in weaponry such as the adoption of throwing knives and battle axes, use of body armour and differentiation between throwing and stabbing spears. As in eastern Africa, the irst half of the second millennium witnessed the emergence of kingdoms in west-central and equatorial Africa, polities which aimed to develop more eficient means of conquest, given that enlargement meant the perpetual struggle to control people and outlying communities; wars of secession along regional fault lines were probably common. All the armies under discussion here were infantry forces, but clearly there were signiicant regional variations. In the coastal forest belt of West Africa, the slim evidence suggests that infantry fought in relatively tight formation, especially those (as in the Benin Gap and in the Senegambia region) which regularly encountered cavalry forces. Soldiers using hand-to-hand weapons – short spears, knives, clubs – were supported by archers, and something similar is discernible in central Africa, where armies often sought hand-to-hand combat, with archers deployed as skirmishers. Shields might be carried as defensive arms in both regions; but more generally, in central Africa, looser formations are in evidence. The impact of Europeans on tactics and organisation was fairly minimal, at least in the sixteenth century: Europeans served as mercenaries in some situations, and were admired, as in Angola, for their abilities in hand-to-hand combat; they also supplied irearms, most obviously. But guns were treated as simply another missile weapon by most African armies and did not come close to replacing indigenous weapons such as bows until at least the early eighteenth century. Moreover, many early trade guns available through commerce were largely ineffective in tropical conditions: Powder and matches drenched by seasonal rains were impossible to light, and gun barrels rusted in the damp heat.

THE MILITARY FOUNDATIONS



Broad patterns of warfare can thus be identiied. The enlargement of political scale – as witnessed, for example, in Kongo, Bunyoro, Great Zimbabwe – led to centralised and expansionist polities whose armies comprised provincial forces under local overlords ighting alongside royal regiments under the direct command of kings. It is a model to which Solomonic Ethiopia likewise belongs. Military campaigns were increasingly far-lung, although of course the speed and range of armies depended on local environment, whether thick forest, light woodland, or relatively open savannah, and the limitations imposed by economy and demography. Smaller and shorter-range military expeditions, as were probably prevalent in the West African forest and in much of lacustrine eastern Africa, involved perhaps a few hundred men, lightly armed and highly mobile, looking to make lightning strikes on neighbours; larger campaigning armies, as in Zimbabwe and Ethiopia, probably looked like enormous communities on the move, with several thousand soldiers supported by sizeable (and largely female) logistical corps, travelling considerable distances over many weeks. Yet in both cases, campaigns were increasingly underpinned by larger political and economic visions, by strategies which aimed at extraction and conquest, or at least neutralisation, of restive frontiers. Nevertheless, those same shifting, fertile and brutal frontiers also produced highly militarised societies-at-arms – whether mobile predatory bands, such as the Imbangala, or the more sedentary but no less martial Mutapa state on the Zimbabwe plateau. In both cases, warfare was increasingly professionalised – in other words, prosecuted by men (and even women) who dedicated themselves to it, and who in turn drew material sustenance as well as socio-political status from it. On a smaller scale, many of the chiefdoms of lacustrine eastern Africa were centred on entrepreneurial frontiersmen, who used the distribution of cattle – and other largesse, in time – to ensure a pool of military recruits. The latter were deployed in campaigns which may initially have been relatively limited in objective – the awed tone of Rwandan tradition notwithstanding – but which grew in range and ambition as networks of clientship expanded. Thus were the foundations laid for the territorial monarchies of Buganda, Nkore, Toro, Rwanda; as these jostled for space in the increasingly competitive environment that was the lacustrine frontier, they had much in common with their counterparts in the coastal forest of western Africa, where armed, compact communities were beginning to amalgamate and



WARFARE IN AFRICAN HISTORY

expand. Meanwhile, predominantly pastoral communities – which both fought and traded with their sedentary, agrarian neighbours – produced a distinctive form of militarism involving the organisation of younger age groups into regimental sections tasked with the defence of cattle, as well as their violent accumulation. In each of these ‘types’ of war during the period under examination, an expansion in the scale and aim of violence is discernible, especially markedly in Atlantic Africa as the slave trade intensiied. These, then, were the patterns which were emerging at the same time as Europeans began to appear along sub-Saharan coasts – and it is that novel development with which we conclude our discussion of this epoch. It is important to note that Portuguese attempts to intervene more directly in the affairs of the interior were either severely constrained or completely doomed, defeated by both armed resistance – there was no marked European superiority in this context – and disease, particularly malaria. Portuguese adventures had enjoyed limited success in Kongo, but they were a resounding failure in the Luanda hinterland, and merchants resigned themselves to a coastal presence – which was, in any case, suficient, save for occasional political intrigue, to stoke the ires of interior conlicts. It was a different story on the eastern coast, where the violent irruption of the Portuguese into the Indian Ocean had devastating consequences for the Swahili city-states of the coast. From the early s, Portuguese galleons equipped with cannon rounded the Cape of Good Hope and made their way up the coast, launching attacks on towns which refused to submit to the King of Portugal. In what amounted to a holy Christian war against Islam, they systematically subdued the Swahili coast: The city-states themselves were ill-defended against seaward assault, as they had long been more concerned with attack from the landward side. The submission of towns was often followed by the erection by the conquerors of stone fortresses, as in Kilwa, Sofala and Mozambique. There was more obdurate resistance along the northern coast, however, with Mombasa repeatedly refusing to pay tribute to the Portuguese, provoking ferocious attacks, irst in  and again in . Finally overcome, Mombasa was chosen as the site of the formidable Fort Jesus, completed by , which became the centre of Portuguese dominance in the western Indian Ocean. Swahili commercial culture entered decades of decline – the result of this early encounter with a burgeoning military imperialism originating thousands of miles distant.

THE MILITARY FOUNDATIONS



Nonetheless, as in the Atlantic zone, Portuguese success was largely conined to the coast. In the s, an expedition was dispatched up the Zambezi River which did manage to capture the Swahili trading posts of Sena and Tete; thus were direct commercial relations established with Mutapa on the Zimbabwe plateau. But Portuguese attempts to win a greater share of the gold trade, as well as the failure to convert the Mwene Mutapa to Christianity, culminated in an attempted invasion of Mutapa in . It was a disaster, as was the follow-up assault in . Portuguese weapons were no match for local military structures, and Portuguese bodies were ravaged by disease. It would be some time before Europeans could seriously contemplate anything but the vaguest of inluences over the interior – and even then, they would be unwittingly co-opted into African military projects. In the meantime, at the end of the sixteenth century – just around the time that the Imbangala were sweeping down to the Atlantic coast – stretches of the East African coast were ravaged by the invasions of the strange and fearsome ‘Zimba’, ravaging shoreline settlements as far north as Mombasa. They were gone almost as soon as they had come, but the stories about them had much in common with those told of the Imbangala: their lack of mercy, their cannibalism, their generic brutality. The ‘Zimba’, whoever they were, served as a reminder to foreigners perched perilously on the coast that the savage interior was very close to the shore indeed – an idea which would loom increasingly large in the Western imagination.

Relections: The Roots of Modern Military Establishments Over the irst few centuries of the second millennium, key transformations took place across the continent, processes of change which may be considered, in different ways, as laying the foundations of modern African military structures and dynamics. Distinct patterns are discernible in the western savannah, for example. For close to a thousand years, horsed conquerors sought to extend new forms of authority beyond the kafu, rooting power in concentrations of wealth and population and relying on long-distance commerce and elaborate tax systems to fund the kind of military capacity required to enforce their authority. But although such experiments in armed imperialism were stunning, their effects were too often supericial, as elites struggled to make their presence felt at distance from political centres, whatever the



WARFARE IN AFRICAN HISTORY

resources and weaponry at their disposal. Attempts at long-distance military dominance – and at assertions of legitimacy through the seizure of cultural and religious centres such as Timbuktu – were episodically met with equally ferocious armed responses. Tensions were thus inherent in the process. Cavalry states were invariably fragile owing to their dependency on external sources for good mounts, and as a result became locked into expensive and unstable ‘slave-horse’ cycles, exporting slaves in exchange for horses. Thus successful equine militarisms were essentially dependent on good commercial relations with the north. This was true of most Saharan and savannah polities from Kanem onwards, and remained true of later states, including Oyo further south, just emerging at the end of the period examined in this chapter. Moreover, with military power so rooted in cavalry, leaders of horsemen were well placed to negotiate for a share in political power, and indeed could challenge the state for it directly. Horsemen were kingmakers and brokers, and great equine chiefs episodically deied political centres altogether and created new polities of their own – sometimes of great potency, especially those motivated by ‘new’ visions of Islam, which was itself a simultaneously disruptive and unifying force across the region. Only very slowly would irearms be adopted by these military orders; cavalries largely eschewed them until the nineteenth century. Broadly similar patterns, if within somewhat distinctive parameters, can be detected in northeast Africa, too. Ethiopia’s bloody sixteenth century was characterised by religious war which profoundly weakened the bristling nation-at-arms created by Amda Tsion  years earlier. The violence inherent in that system was now revisited on Ethiopia itself, the outcome of mounting Christian-Muslim rivalries in the region; and in addition to this shift in the balance of power, there was the movement of the Oromo into the highlands, changing forever the usage of land and creating new patterns of farming and herding. Once again, environmental change appears to have driven military change in Africa. But other patterns emerged, too, including an over-powerful military aristocracy in the much-reduced provinces of the Solomonic polity, and a politicised army at the centre of it. As in the western savannah, great war chiefs became – or fancied themselves to be – power brokers and the makers of emperors. Disloyalty was the result of a lack of land with which to reward loyalty. South of the Sahara, there were neither horses nor land grants to keep the costs of warfare high. Nonetheless, across this vast region, an

THE MILITARY FOUNDATIONS



array of military structures appeared in the irst half of the second millennium, from armed pastoralism to sedentary militarised statehood. Flexibility and mobility were the hallmarks of many of these evolving systems, with frontier zones between polities becoming arenas of conlict and creativity, as those polities competed for control of people, resources and trade. Such competition became all the more violent with the intrusion of external dynamics, most obviously in the form of the Atlantic slave trade. Iron technology was used to subdue harsh environments and increase productivity; complex integrated economic systems resulted, with attendant social stratiication; communities were able to build up population levels; trade networks ensued, and along commercial highways lowed ideas and cultures as well as commodities. These were networks which would ultimately become global in their connections. Between  and , a series of state systems emerged across sub-Saharan Africa – in the Great Lakes, in the coastal forests of western Africa and the west-central savannah, in southeast Africa – which essentially represent the political and cultural foundations of the modern continent. These represented the superiority of iron weaponry and iron implements, as shown by the incorporation into royal rituals of references to ironworking and the elevated status of the blacksmith. Long-distance trade and thus enhanced proit-making capacity is an explanatory factor, as is, again, economic surplus and the resulting specialisation of labour. Larger populations resulted from greater food-producing capacity, and larger populations meant political sophistication – a process supplemented, in several areas, by migrations which drove militarism in different ways. For all these reasons, the irst half of the second millennium is one of the critical fertile phases in African history. This is evident everywhere in the celebration of arms and militarism: the sixteenth-century Benin Bronzes, the epic of Sunjata, the Kebre Negast and the oral traditions of the Great Lakes kingdoms. External contact had increased markedly by the beginning of the seventeenth century, and it is tempting, as always, to talk of harbingers and portents. And perhaps justiiably so, but we need to be careful to avoid teleology in this respect. To be sure, commercial dynamics which were European in provenance had begun to drive militarisation in Atlantic Africa – a process which would only intensify in the two centuries which followed. In terms of new technology, irearms occasionally decided the outcome of the engagement; in the hands of the Portuguese in Ethiopia, or the Ottomans in Egypt, or the Moroccans



WARFARE IN AFRICAN HISTORY

in Songhay, they played a decisive role, while guns on Portuguese ships were used to subdue the Swahili city-states. But elsewhere, European activities were constrained by climate and disease, and irearms had a very limited impact, as a rule, across sub-Saharan Africa. Only in the centuries to come would the irearm fundamentally alter African tactics and organisation. But if irearms were here to stay, so too were a number of dynamic military states – from Ethiopia and the lacustrine monarchies to Morocco and the states proliferating along the Atlantic littoral. Some of the great militarisms of antiquity had faded by about  – those of the western savannah, notably – but elsewhere the military foundations of state and society had been laid. Further reading Crummy, D. Land and Society in the Christian Kingdom of Ethiopia: From the Thirteenth to the Twentieth Century (Oxford: James Currey, ). Fisher, H. ‘“He Swalloweth the Ground with Fierceness and Rage”: The Horse in the Central Sudan’, Journal of African History, : () & : (). ‘The Eastern Maghrib and the Central Sudan’, in R.Oliver (ed.), The Cambridge History of Africa: Vol. , from c.  to c.  (Cambridge: Cambridge University Press, ). Huntingford, G. W. B. (tr. & ed.). The Glorious Victories of Amda Seyon, King of Ethiopia (London: Oxford University Press, ). Kaba, L. ‘Archers, Musketeers, and Mosquitoes: The Moroccan Invasion of the Sudan and the Songhay Resistance (–)’, Journal of African History, : (). Levtzion, N. ‘The Western Maghrib and Sudan’, in R. Oliver (ed.), The Cambridge History of Africa: Vol. , from c.  to c.  (Cambridge: Cambridge University Press, ). Mair, L. African Kingdoms (Oxford: Clarendon Press, ). Oliver, R. & A. Atmore. Medieval Africa, – (Cambridge: Cambridge University Press, ). Schoenbrun, D. A Green Place, a Good Place: Agrarian Change, Gender and Society Identity in the Great Lakes Region (Oxford: James Currey, ). Taddesse, Tamrat. Church and State in Ethiopia, – (Oxford: Clarendon Press, ). Vansina, J. How Societies Are Born: Governance in West Central Africa before  (Charlottesville: University of Virginia Press, ).

CHAPTER 

Destruction and Construction, circa  to circa 

Merchant and Military in Atlantic Africa: The Slaving State As the external demand for slaves increased in the seventeenth and eighteenth centuries, vast areas of Atlantic Africa – including both shoreline communities and those across several hundred miles of hinterland – became militarised zones. The scale of warfare increased, and war itself became an extension of polity and economy; militarism became a matter of cultural celebration within many communities from the southern fringes of the Sahara to the northern edge of the Kalahari. Asserting the correlation between the slave trade and organised violence is hardly controversial; what is of interest here, however, is the manner in which warfare changed in this period, and how both construction and destruction resulted, in political and economic terms. Inevitably the impact and intensity of the trade varied from area to area: Whereas Senegambia and Angola had been mainstays of slave supplies for Europeans since the mid-sixteenth century, new regions opened up in the seventeenth century, notably the Slave Coast (southwest Nigeria and Benin) and the neighbouring Gold Coast on the Gulf of Guinea. This stretch of shoreline, some  miles long, became home to a series of forts – more or less permanent European settlements – from the mid-seventeenth century, tapping into the veritable lood of slaves brought about by the rise of such states as Dahomey, Oyo and Asante. It was a circular process, not to say a matter of chicken-and-egg: While such states arose at least partly 



WARFARE IN AFRICAN HISTORY

in response to the slave trade, the presence of Dutch, English, French and Danish traders on the coast – themselves responding to ready supplies – encouraged further slave trading (and raiding) to go on, which in turn encouraged more buyers to cluster at the coast. Thus the militarisation of African society and economy was a cumulative process, and involved the emergence of what has become known as the ‘slaving state’. The Slave Coast – a major source of slaves from the mid-seventeenth to the early nineteenth century, and closely connected to the rise of the Caribbean plantation economy – demonstrates the dynamic very clearly. To begin with, the history of Oyo is indelibly intertwined with the story of the slave trade itself: From the early seventeenth century, the Oyo state imposed itself across the woodland north of the coastal forest belt, using horses imported from the north. Oyo’s reliance on cavalry meant that its military operations were conined to the tsetsefree zone – enemies could stay out of range by taking refuge in the forest – but Oyo was able to gain direct access to the coast thanks to the ‘Benin Gap’, a small area where there is a break in the coastal forest and the savannah reaches down to the shoreline. In the s, an Oyo force attacked and subdued this vital stretch of coast, and with it the kingdom of Allada and the vital port of Whydah, although it was only with the reduction of Dahomey to tributary status in  that Oyo gained secure access to European merchants. The Oyo empire thus became one of the key transit handlers and producers of slaves in the eighteenth century. Its own wars produced a surplus of captives, some of which were retained within Oyo itself; others were imported from the north and passed to the coastal forts. Oyo acquired cloth, metal goods and cowries in exchange – wealth which was used to buy horses – as well as guns, although these were adopted rather more slowly by the Oyo army than by other polities closer to the shore. Oyo cavalry, operating in tandem with infantry forces, struck terror into coastal communities, doubtless largely owing to the fact that forest peoples rarely encountered horses; a brooding Oyo presence looms large in the eighteenth-century European sources, with stories aplenty concerning the extent of this ‘savage’ and ‘warlike nation’ with its ,-strong army. It appears not to have been a standing army in the strict sense, but a highly effective part-time ‘national’ force. Among the key military ofices, beneath the alain himself, that of the basorun was the most important for much of the eighteenth century: Its holder was the commander of the metropolitan army, as well as a

DESTRUCTION AND CONSTRUCTION



senior political chief. The kakamfo, meanwhile, was commander of the provincial army. In the capital itself, there existed no less than seventy principal war chiefs, or eso, divided into sixteen senior and ifty-four junior positions. This oficer corps was under the control of the Oyo Mesi, the central council of chiefs, with each of the seven senior chiefs commanding ten eso. The political importance of the army needs to be noted here, for its role was by no means a subservient one. In the late seventeenth century, it was the perceived tyranny of Alain Karan which prompted a mutiny, after which direct royal control over the army was limited, although the king still had the power to dispatch military expeditions. The autonomy of the Oyo army gave rise to an intermittent political rivalry which escalated in the s, when Alain Abiodun allied himself with the kakamfo against the basorun. In the struggle which ensued, the alain relied heavily on troops from out of town against the metropolitan units. Thereafter, the kakamfo became particularly important politically as well as militarily, which was a new problem for the alain: When Alain Awole sent units against his overly ambitious kakamfo Afonja a few years later, it was Afonja who marched to the Oyo capital and forced the suicide of the king. By the late eighteenth century, the army had become a major player in internal politics – no doubt owing, at least in part, to its role in underpinning the Oyo economy, and its role in the state’s very existence. Similarly, Dahomey had its origins in the early seventeenth century – like Oyo, its development was linked to the expansion of the slave trade – and quickly became one of the most highly centralised, militarised states in the region, or indeed anywhere on the continent. But the militarism which evolved was not solely concerned with the export of captives, whether or not this had been its original driver. Dahomean military success was ultimately an end in itself, and a remarkable cult of kingship had developed by the early eighteenth century, involving annual human sacriices – the victims were normally war captives – to honour royal ancestors, without whose blessing Dahomey’s armies would meet disaster and the kingdom would fall. It was a ruthlessly cyclical militarism. Wars produced tangible outcomes – captives for export, in return for highly valued goods, including irearms – but they also became the focus of cultural celebration which held feats of arms in the highest esteem. It would become a shibboleth of the Dahomean ruling elite that war was the sine qua non of the kingdom. Meanwhile, the practice of human sacriice, zealous militarism and participation in the slave trade rendered Dahomey, in



WARFARE IN AFRICAN HISTORY

the eyes of Europeans, the epitome of African primitive savagery – and accordingly a reference point throughout the eighteenth and nineteenth centuries for those wishing to highlight everything that was ‘wrong’ with the African character. Dahomey expanded at the expense of the less centralised communities which bordered it, selling the resultant captives to middlemen in direct contact with European buyers. Whether local expansionist violence merely coincided, in the irst instance, with the expansion of the slave trade, or whether it was driven by the latter from the outset is perhaps a matter of speculation, but it is clear that the two processes quickly became indelibly interlinked. Between  and , Dahomey conquered a stretch of the coast encompassing Allada and Whydah, which gave it direct access to the trade – although interestingly, the initial intervention in Allada’s affairs appears to have been of a mercenary nature, the brother of the king of Allada paying the Dahomean king Agaja for his assistance. Despite becoming tributary to Oyo after , Dahomey continued to expand and thrive, becoming the main supplier of slaves in the area, its military organisation far in advance of that of most of its neighbours. Its acquisition and effective deployment of guns made it one of the best-armed states in the region. Again, however, Dahomey’s wars were not solely about export: Slaves were critical to the kingdom’s political and social systems and were used extensively in the domestic economy. Certainly, apologists for the slave trade in the late eighteenth century argued vigorously that Dahomey fought wars largely for their own sake – such was the African ‘condition’ – and that only a small proportion of captives were actually sold off to Europeans. King Kpengla, who reigned between  and , supposedly told the governor of the British fort at Whydah that Dahomey’s wars were fought regardless of European commerce, although this was contradicted by King Gezo in the midnineteenth century, who explicitly identiied Dahomean militarism – indeed the very stability of the kingdom – with slave gathering and export. He did so, it should be noted, while under pressure from the British to cease Dahomey’s involvement in the ‘illegal’ slave trade. The truth, no doubt, lies somewhere between these two positions – and, again, the signiicance of war itself in Dahomean political culture was scarcely in doubt. Like Oyo, Dahomey offers a model of intricate military organisation, as well as of the melding of the political and the military in a manner increasingly common across sub-Saharan Africa. In

DESTRUCTION AND CONSTRUCTION



the eighteenth century, a system of regiments – each with its own emblems – was sustained by a programme of rigorous training for young boys, who were assigned for the purpose to regular soldiers. By the late eighteenth century, women appear still to have been conined to ceremonial roles; Dahomey’s famous female regiments would only be properly developed in the nineteenth century. Contemporary evidence makes much of the celebration of killing and plunder, the two very much intertwined, which was at the core of Dahomean regimental culture. The head of the army was the agau, who was assisted by deputies, the zohenu and the fosupo; the army as a whole appears to have been organised into ‘right-hand’ and ‘left-hand’ wings. The various commanders themselves had considerable political inluence, the army being as central a part of the Dahomean political ediice as it was in Oyo. As a result of a remarkable political dynamism, Dahomey was in a position to take full advantage of the power shifts of the late eighteenth and early nineteenth centuries. The kingdom overthrew its long-standing subordination to the ailing Oyo state and took advantage of the chaos which followed Oyo’s collapse to begin operations to the east among the Yoruba, out of bounds as long as Oyo had remained hegemonic. This shift in military gravity was timely indeed, for Dahomey’s traditional hunting grounds to the north and west had experienced serious depopulation by the early nineteenth century. It meant that as the nineteenth century dawned, Dahomey was well placed to dominate the thriving ‘illegal’ slave trade in the area. Further west, too, the Asante kingdom emerged out of a federation of several Akan states in the course of the seventeenth century, again in large part as a response to external commercial opportunities, although the process was also the outcome of local competition for land and goldields. When the military overlord Osei Tutu, the Asantehene, took control from his base at Kumasi in the s, he laid the foundations of a highly effective system whereby tributary chiefdoms supplied soldiers for a large permanent army, which had enormous capacity to drive territorial expansion. By , Asante controlled most of the goldields of the forest in the vicinity; in the course of the eighteenth century the kingdom continued to expand, waging wars which produced plenty of captives, many of whom were duly delivered to European buyers at the coast.Yet the standard view is that Asante never became as dependent on the slave trade as either Oyo or Dahomey, enjoying a more diverse range of factor endowments,



WARFARE IN AFRICAN HISTORY

including gold and some of the inest farmland in coastal West Africa. The Asantehene derived enormous wealth from long-distance trade, both coastbound and northbound, and the Asante army was well equipped with irearms. In the late eighteenth century, further reforms awarded the Asantehene even greater control over the army. To some extent, levels of violence depended on the degree of political control exercised by unitary authority within a given area – although this should not be overstated, for some polities were neither able nor willing to curb the desultory warfare which invariably operated alongside more conventional, state-level slave raiding. At the same time, new military forms emerged out of the rapid change brought about by contact with Atlantic commerce. The emergence of the Wolof kingdoms in Senegal in the seventeenth century was intimately linked to the increasing demand for slaves at the coast, unlike the Jolof state – for whom the northbound Saharan trade had been more important – which had preceded them. The Wolof states were organised around armies of royal slaves (ceddo) whose very existence depended on slaving wars, and which developed distinctive cultures of military honour accordingly. The power of this new commerce to transform military and political structures was clear, too, in the foundation of the Segu kingdom on the banks of the upper Niger. In this case, a charismatic young warrior named Biton Kulibali embodied the violent entrepreneurialism of the age by taking command of his age-set and expanding it into a highly effective force of slave soldiers known as the tonjon. Around , he established Segu, and in the course of the eighteenth century, the state resembled the Wolof ceddo polities in terms of its deep involvement in the slave trade and highly militarised culture, as well as its use of slave labour in the domestic economy. But whereas the Wolof were exporting few slaves by the eighteenth century, Segu had become a major ‘producer’ of slaves on the scale of Dahomey or Oyo. In the seventeenth century, across Atlantic central Africa, the breakdown in order which followed the collapse of states such as Kongo resulted, as we have seen, in the emergence of new military realities, exempliied by the Imbangala – mobile, highly militarised, tradingand-raiding iefdoms for which violence was both cause and effect. Widespread slaving wars meant that the area of modern Angola produced a steady stream of captives for export: By the s, the port of Luanda alone was exporting , slaves a year. Other polities arose even deeper in the interior: The Lunda ‘empire’, for example,

DESTRUCTION AND CONSTRUCTION



was a loosely organised federation of chiefdoms straddling the southern Congo basin, run by the Mwata Yamvo (itself a title with military connotations), which emerged in the second half of the seventeenth century in response to the demand for slaves and ivory from the distant coast. They imported guns – although these were less effectively used than at the coast, and did not yet have a signiicant impact – and developed a rich tradition of militarism which celebrated conquering heroes and martial prowess. The Lunda followed a well-established pattern of issure, migration and reconstitution across sub-Saharan Africa when a faction migrated eastward towards the Luapula River and established a new state, headed by the Mwata Kazembe. The Kazembe state borrowed Lunda military culture and methods of control, straddling a long-distance commercial system which stretched both west and east – to the Atlantic and, rather closer to hand, the Indian Ocean shore. Slaves were only one of its exports, but warfare was key to Kazembe’s success in the eighteenth century. A chain of militarised polities now reached from the Atlantic littoral to the Congo-Zambezi watershed – and indeed beyond, towards the East African coast. The ‘slaving state’, then, proliferated across Atlantic Africa in the seventeenth and eighteenth centuries. Predatory, highly militarised polities such as Dahomey, Oyo, Asante, Segu and Lunda imposed order through violence on the disorder wrought by the slave trade. The costs of militarism were high in these entrepreneurial state systems, but they were met by reinvesting the proits of business with Europeans in military organisation. To a greater or lesser degree, each came to be dominated by military aristocracies, deined by war – and in many cases those aristocracies began as renegades and bandits, who hijacked processes of state building in order to grasp the opportunities offered by Atlantic commerce. There were differences, of course, in how they were organised and how they extended their ‘reach’ into the deeper interior. The armies of Segu and Oyo comprised horse-riding raiders leading the assault, complemented by infantry, and were able to organise several campaigns simultaneously which ranged considerable distances in search of human prey. In more compact Dahomey, military campaigns tended to be organised annually, dispatched into favoured hunting grounds, especially northwards. It is impossible to generalise about the size of armies in this period, not least because they probably varied greatly from year to year even within a particular polity. Field forces might number no more than a few hundred men, whereas the more compact militaries of the Atlantic coast appear



WARFARE IN AFRICAN HISTORY

generally to have comprised anything from , to , troops. The states with the greatest capacity for large-scale mobilisation – such as Oyo or Asante – might ield somewhere between , and , men, although evidence suggesting still greater numbers must be treated sceptically. Yet even the larger states probably regularly deployed much smaller forces for the purposes of ambush – a long-established tactic in the forest – in pursuit of slaves, against both enemy armies and unsuspecting village settlements. What is clearly true is that larger armies might be more devastating when ighting in the immediate vicinity of the homeland, but further aield they were often simply more cumbersome. Most armies were accompanied on the march by support systems in the form of looters and porters – and often women, children and slaves – whose job was to carry food and supplies. Armies also resupplied themselves through foraging, hunting and (more commonly) looting, although rivers might be used to send supplies to campaigning forces. Overall, there were signiicant logistical obstacles to the size of armies and their range of operations. This meant that most slaving states were served by client polities and trading partners which did their own raiding and passed slaves into their systems. By the late eighteenth century, for example, Asante was acquiring most of its slaves as tribute from states which its armies had conquered some way north of Kumasi. The slaving state differed from patterns of war elsewhere – including much of eastern and southern Africa – in the sheer scale of its operations and the eficiency and professionalism of its organisation. These were political systems dedicated to war and underpinned by highly developed cultures of militarism, to a degree only occasionally discernible in eastern and southern Africa in the same period – and this is true even though, as we see later, there was a general trend towards militarisation in those regions, too, during this period. The economies of slaving states were rooted in violence, and the political authority vested in their leaders was based ultimately on their abilities as military commanders and as commercial operators. In virtually all cases, slaving states may have begun as producers of slaves through warfare, but they also became important commercial centres, hosting key regional markets for slaves. And the polities which resulted – suddenly, in some ways, as in Dahomey and Segu – were novel, innovative creations, bristling with violent entrepreneurial energy in ways that many eastern and southern political systems, which were developing more slowly over the longer term, did not, yet.

DESTRUCTION AND CONSTRUCTION



This was an era in which the scale and professionalism – in other words, military activity as a way of life, underpinned by cultures of martial celebration – of armed units appear to have increased markedly. This is certainly borne out by evidence relating to the Slave Coast polities of Oyo and Dahomey, while Asante likewise developed a class of professional soldiery; further south, the Imbangala offer an example of a somewhat different kind of militarism – mobile bands of militia comparatively unencumbered by state apparatus – but also indicate new levels of commitment to war as vocation. These were standing armies, in the simplest deinition of ‘professional’, either continually in service or in anticipation of it; which did not have to produce their own food; and which were sustained materially and politically by the state (or its equivalent). Violence, moreover, was increasingly outsourced: The use of mercenaries – professional gunmen (or spearmen, or bowmen) hiring themselves out to neighbouring polities – is much in evidence across Atlantic Africa, and to some extent may indicate both an attempt to keep levels of local militarisation under control and a lack of military capacity. Allada and Whydah, for example, were making extensive use of mercenary soldiers from the Gold Coast, particularly Akwamu, by the s and s. Meanwhile smaller, decentralised societies in the interior developed other survival strategies in the era of the slaving state: Some sold slaves, or offered slaves as tribute, to protect themselves, and certainly small-scale raiding took place in the shadow (or indeed at the behest) of larger neighbours. The constant threat posed by the latter, more generally, caused profound insecurity across substantial areas and resulted in shifts in living patterns, as villages became fortiied, or as groups retreated further into hilly defensive positions, often some distance from arable land. Conlict along the jagged frontiers of predatory states – and indeed at their very centres – led to ission, as experienced by Segu, for example: In the s, a contingent broke away to form another tonjon state, Kaarta, further north. Yet Segu itself survived the process. In the previous chapter, we noted the collapse of Kongo; in this era, only one major slaving state, Oyo, disintegrated in the face of mounting commercial and military pressures. Oyo’s ruling elites derived a great deal of economic power from slave exports, but in the end it was the huge dependency on these external sources of wealth which proved the kingdom’s greatest weakness; indeed, Oyo exempliies the eighteenthcentury (and later) phenomenon of African military power derived from external economic relationships over which there was ultimately



WARFARE IN AFRICAN HISTORY

little African control, although while it lasted, African elites certainly bent the relationship to their local advantage. As the demand for slaves dipped from the s, alains found their revenues much diminished, leading directly to a sharp decline in military capacity and a series of provincial uprisings in the early years of the nineteenth century. At least one of these – the Muslim uprising at Ilorin in  – was largely coincidental, related rather to the jihad to the north, and this is dealt with in the next chapter. But it was all symptomatic of Oyo’s steady decline. Contemporary reportage from the early nineteenth century suggests that the cavalry which had once terrorised the surrounding area was largely defunct – there were no more funds with which to buy horses from the north – and that Oyo’s much depleted armies were now infantry forces, unable to defend the state against predatory neighbours nor take advantage of the illegal slave trade which would lourish in the nineteenth century. It is clear that during this era, the irearm, Africa’s most signiicant military import since the horse north of the forest, begins to have a major impact. In Europe, the key development was the shift from the matchlock – predominant in sixteenth-century European armies – to the fusil, or lintlock musket, which was standard across Europe by the s. In terms of its impact in Africa, of course, we must be careful. In the savannah lands north of the forest, guns were still eschewed by equine warriors armed with spears and protected by quilted armour. Preferring bows, javelins and swords, for example, Oyo cavalrymen were hampered in the early and mid-eighteenth century by the fact that coastal soldiers’ guns frightened the horses. Deeper in the interior, the simple fact of limited or at least less direct commercial linkages meant that guns were again slow to have much inluence, and old tactics held sway. The Lunda, notably, although they did import guns, persisted with hand-to-hand combat using locally produced weapons – including clubs, throwing knives and thrusting spears – and were successful even against adversaries equipped with irearms. Nevertheless, elsewhere the import of lintlock muskets into Africa from the early eighteenth century onwards had a signiicant impact on many polities and societies, particularly those on or a short distance from the coast. In some places, bows and arrows – most common, for obvious reasons, among grassland communities and in more lightly wooded areas – were eventually almost completely displaced. By the late seventeenth and early eighteenth centuries, matchlocks and lintlocks were in common use along the Slave Coast, for example, and

DESTRUCTION AND CONSTRUCTION



Gold Coast mercenaries were frequently hired at least partly for their experience and skill in using irearms. Dahomey adopted irearms rapidly, and effectively. Perhaps the most dramatic tactical impact in many areas was in terms of ighting formation, for the tight arrangement of troops along much of the West African coast in the seventeenth century frequently gave way to rather looser formations in the eighteenth. With ever larger numbers of troops equipped with guns, soldiers fought in more open conigurations, involving smaller, more mobile units; the evidence suggests that commanders sometimes sought to avoid close combat altogether, instead seeking to overwhelm enemies with gunire from various vantage points. In central Africa, too – in western Congo and in Angola – looser formations than previously are in evidence, although here it is possible that the smaller, specialist warrior units of old were replaced by larger, more unwieldy levies of musketeers, including some Portuguese. Exceptions to this broad trend were found in the forest-savannah zone of interaction, where infantries which regularly engaged with cavalries needed to retain tight formations. In Dahomey, compact phalanxes of musketeers, apparently drilled to maintain a high rate of iring, faced down Oyo cavalry charges in close order, and with some success, it seems. Dahomean tacticians had quickly realised that horses do not normally charge through a compact mass; other West African armies which occasionally campaigned along the frontier between forest and savannah, including those of the Asante, likewise developed close formation tactics, and maintained these even following the adoption of irearms. More broadly, even though much political and military development in Atlantic Africa has been explained in terms of a gun-slave cycle, it is an oversimpliication to do so: Local competitions for resources and identities, which reached into the deeper past, were critical, and local actors co-opted exogenous dynamics into endogenous processes of militarisation and patterns of conlict. Nonetheless, the expansion of the Atlantic economic system did indeed mean an increase in the incidence and the scale of war in many areas. It has been suggested that something approaching ‘total war’ resulted – as irearms were imported in ever larger numbers, and were deployed more effectively, and as levels of destruction, particularly of smaller, stateless communities, heightened – but in truth, there was always a tendency towards this in African warfare. The seizure or the total destruction of the enemy’s human, animal and mineral resources was not something introduced by the Atlantic slave trade: It had been a core objective in much



WARFARE IN AFRICAN HISTORY

African violence for several centuries. That said, there can be little doubt that the capacity to achieve it, and the willingness to inlict it, increased markedly during the seventeenth and eighteenth centuries – and would continue to increase still. Overall, the slave trade meant the creation of what were essentially war economies, with ruling elites commanding armies which were economic assets as much as they were designed to bring about political aggrandisement. Great wealth was generated as a result, when the system ‘worked’. It is fair to say, however, that the wealth which was generated was neither widely distributed – it remained conined to powerful elites, even if Dahomean soldiers, for example, were paid for the captives they brought to the royal court – nor was it constructively reinvested in the long term, however rational the system was in the immediate term. And it was eminently rational: Powerful men sold slaves in order to acquire the goods which would attract followers, and which would enable them to pay for the military systems increasingly necessary to support the economy. This had a deadly circularity about it, and it was a zerosum game: The irony was that as individual communities and polities sought to maximise numbers – as they always had – they were selling off that most valuable of assets, namely people. In a sense, of course, this was a variation on what African states had always done: sustaining political centres and bolstering economic well-being by preying on contested borderlands, and using coercion to control a range of resources. The tragedy, now, was in the scale of the forced migration, both internal and external, and in the irretrievability of the lost assets. Increasingly there was an emphasis on warfare rather than production as generator of wealth, and in many areas this would only intensify into the nineteenth century and beyond. Widening circles of violence, moreover, meant considerable internal displacement, as vulnerable communities shifted to new locations which might be more easily defended but which were less economically viable. It was a process which in turn drove the militarisation of state and society. This was very expensive militarism indeed.

Reform, Consolidation and Crisis: Patterns of War and Polity in Eastern and Southern Africa Some of the iercest struggles for regional hegemony in the seventeenth and eighteenth centuries, meanwhile, were in eastern and

DESTRUCTION AND CONSTRUCTION



southern Africa. Across this vast region, the trend – as in Atlantic Africa, although usually somewhat less dramatic – was towards an escalation of violence as transformative force and the militarisation of the polity. Ethiopia may have deied the general movement towards uniication and consolidation in political and military affairs between  and , but the region does demonstrate the increasing prominence of the military in political and cultural life. From the s, Ethiopia sought stability following a century of violent upheaval – the thriving urban centre of Gondar itself represented something of a commercial and cultural renaissance, the absence of serious fortiication indicating, perhaps, of a greater sense of security. Nonetheless the Turks on the Red Sea coast simultaneously represented a military threat and a commercial opportunity. The captives generated by warfare – especially along the western borderlands – were dispatched northwards, and the slave trade lourished: In the seventeenth century, some , slaves per year were being exported through the port of Massawa. At the same time, one of the key themes of this period is the continued immigration, and integration, of the Oromo. The strategy of co-opting Oromo units into the central imperial army gathered pace under Emperor Iyasu (‘the Great’) in the late seventeenth century. As the eighteenth century progressed, Oromo chiely lineages – many of which were intertwined by marriage with the Amhara political establishment – were positioned at the heart of imperial power, their role as kingmakers and brokers based very largely on their command of much-feared infantry and cavalry forces. The armies which might be raised by Ethiopian leaders were sizeable, especially in the larger provinces, and certainly comparable to Oyo and Asante; ield forces of between , and , soldiers and their attendants were probably not unusual. The trend was towards vast ‘non-combatant’ attachments to armies on the move; these could be half the size again, or even double that, of the army itself. As armies expanded, so too did the logistical corps. At the centre, meanwhile, the ever weakening kingship sought to counterbalance the increasingly unreliable and overtly political imperial guard by creating new Oromo units under the direct command of the emperor. Infantry and cavalry operated in tandem, and guns were having some impact, but this should not be exaggerated. Even though large numbers of irearms were present in northern and central Ethiopia, especially in Tigray, in the second half of the eighteenth century, they were only slowly becoming decisive in



WARFARE IN AFRICAN HISTORY

battleield engagements and important as measurements of politicomilitary power. These forces were frequently deployed against other Oromo groups pressing in on the kingdom’s eastern lanks; but just as often they were deployed against recalcitrant provincial overlords, for centrifugalism remained the great scourge of would-be unifying monarchs, and in particular the ability of powerful men in the regions to raise large numbers of men in opposition to the state. This was a struggle, ultimately, over control of gult land, the estates which sustained military overlords. While the armed nobility reasserted itself at the centre from the early eighteenth century, the provinces pulled away; the emperor himself became little more than a local magnate, and then not even that. By the late eighteenth century, the armed periphery had marched decisively on the weakened centre, and in effect the Gondarine era was over – as indeed, for a time at least, was the era of Ethiopia as a uniied polity. Through the s and s, although the idea of Solomonic unity persisted – much diminished emperors still walked the corridors of the royal palace at Gondar, but with little power beyond it – the Ethiopian region entered a period of prolonged war fought between rival provincial nobles over access to gult estates and the economic and military power control of such land brought. Ethiopia had been ripped apart by centrifugal forces. Meanwhile, elsewhere, the amalgamation of military power and new armed statehood was in evidence. In the Great Lakes region, in the course of the seventeenth century, Bunyoro was increasingly challenged by the rise of Buganda, which started out as a compact cluster of clans on the northern shore of Lake Victoria, dominated by Bunyoro in political and economic terms. The Ganda, indeed, based much of their burgeoning sense of cohesion as well as their military tradition on the fact that they had, against all odds, survived numerous onslaughts from Bunyoro; the Nyoro were in some respects the sine qua non of militant Ganda nationalism. For more than  years, Bunyoro based its regional hegemony on the stability provided by mixed farming and the resultant ability to ield relatively large, mobile armies; but Buganda’s rapid economic growth, driven by well-watered banana plantations which in turn facilitated a relatively dense population, enabled it, too, to ield sizeable, well-fed armies made up of men contributed by provincial overlords with an investment in central authority. Active defence morphed into aggressive adventurism, and by the early eighteenth century, Buganda was poised to overtake Bunyoro as the key

DESTRUCTION AND CONSTRUCTION



military power in the northern lake region. The reign of Kabaka Mawanda, some time between  and , is particularly associated with politico-military reform and successful expansionism: He extended central control over land and the appointments that came with it, and oversaw the further development of a complex hierarchical system which facilitated rapid military mobilisation and greater executive control over the army itself. Buganda bore some resemblance to Ethiopia in its ability to ield ever larger armies, swollen by the presence of non-combatant logistical support which extended the reach of such armies. Nonetheless, terrain and the supply of food and water naturally placed constraints on the range of campaigns, as did the fact that Ganda soldiers were part-time, and therefore expeditions needed to be planned carefully according to agricultural timetables. But war was increasingly serious business, and armies were used either to periodically raid for key commodities – cattle and slaves, chiely – or to occupy and annex particularly desirable provinces, often with considerable bloodshed. Ganda traditions recall, through allegory, the critical switch from wooden to iron-tipped spears, whereas the bow was largely eschewed – although it was common among Buganda’s grassland neighbours – and personal protection came in the form of the enormous, body-length shields characteristic of eastern Africa. The military ethos grew in tandem with the public-works schemes which operated on the same principle: Men (and women) assembled at appointed times to build roads and clear ields just as they did to ight wars on or beyond the kingdom’s borders. But Buganda’s soldier-farmers were no faceless automatons: Ganda military culture celebrated individual glory, especially by young heroes, and oral traditions told of the humble raised to greatness through extraordinary deeds on the ield of battle – and, conversely, of hubris laid low by revelations of cowardice and selishness. Honour and courage were central to the polity’s military ethos, and a culture of arms grew richer and deeper with tangible success – and there was plenty of that in the eighteenth century, as Ganda armies carried the frontiers of the state eastwards to the banks of the Nile and southwards around the shore of Lake Victoria. As the kingdom expanded, it incorporated some of the inest farmland and pasture in the region, as well as sources of iron ore for its spearheads and the wood which in time would be used in the construction of war canoes. Its southward expansion also gave it greater access to the long-distance trade routes which were opening up in the late eighteenth century, linking the interior with the Indian



WARFARE IN AFRICAN HISTORY

Ocean coast; and in that sense Buganda exempliies a pattern much in evidence elsewhere, too, in which the material gains resulting from military success fuelled a culture of militarism at the very heart of state and society. Militarism was also a core component in the state-building exercises underway at the north end of Lake Tanganyika, in Rwanda and Burundi. Here, too, rich farmland sustained healthy and active armies, although it was ownership of cattle which bestowed aristocratic status and facilitated the creation of warrior elites. Immigrant pastoralist Tutsi clans had interacted extensively with the indigenous Hutu, but by the eighteenth century, to be Tutsi was as much about social status as ethnicity, and livestock owners made much of their aristocratic warrior heritage, offering protection to farmers from rival clans in return for food and service. And although the Tutsi had absorbed much in the way of Hutu custom and culture, the former nonetheless emphasised their superior genetic stock to justify their political and economic dominance – mythology critical to the cohesion of the polity, however much it was contradicted by the realities of integration and mutual dependency. Chiefdoms across the area of present-day Rwanda and Burundi were iercely competitive, militarised societies which, like Buganda, positioned the celebration of arms at the very heart of the polity, and which drew armed strength as well as the social power of sanction from unstable and dangerous borderlands. A further variation emerged among the pastoralist Maasai, who, like the Oromo further north, organised themselves into age-sets – the key categories were children, warriors and elders – which offered the most effective system for the defence of cattle and the mobilisation of the community over considerable distances. These were organised into highly mobile units of spearmen – probably numbering on average no more than a few dozen men – capable of mounting rapid attacks on a regular basis, thus applying pressure on their neighbours along contested frontier pastureland. They were warriors and herdsmen: No distinction was made between the two social roles. The Maasai may have eschewed the political hierarchies of Buganda, Bunyoro, Rwanda and Burundi, but they were no less expansionist, clashing with neighbouring Kikuyu and Kamba over land, and justifying the seizure of others’ cattle and pasture by claiming, in effect, that all the cattle in the world belonged to them. Again, it was mythology which could be utilised when needed, whereas the periodic deployment of warrior age-sets – overseen by watchful and often wary elders – represented

DESTRUCTION AND CONSTRUCTION



the attempt to harness the youthful aggression so necessary, and yet so potentially detrimental, to the health of the larger community. War needed to be regulated, and the young carefully managed, and in this sense the Maasai struggled with the same problem as – although their pastoral economies required a different strategy from – the lacustrine monarchies to the west. Across eastern Africa, the tendency was towards an enlargement of the scale of warfare and military organisation in the eighteenth century – even if in some areas, scale remained small until the following century, as among the grassland communities of Tanzania. Here, before the nineteenth century, population densities remained low and settlement remained scattered; people lived in small groups, able to move on when land became infertile or when political disputes arose, such lexibility demonstrating the underlying theme of continual movement, issure and fusion. This would change in the nineteenth century, but the expanded scale was already evident elsewhere – slowly, at irst, in places, but gaining momentum towards the late eighteenth century. For example, in the well-watered southern African lowveld, until about , Ngoni chiefdoms were small, self-reliant communities, clusters of related family homesteads with access to farmland and pasture. Individual chiefdoms tended to grip hilltops, separated from one another by the numerous river valleys that cut through the landscape. Available evidence suggests, however, that during the eighteenth century, Ngoni society was transformed. Competition for resources heightened during this period, in part, it would seem, because the introduction of maize at a time of unusually high rainfall led to a rise in population over several decades. With larger populations, more land was brought under cultivation, and the delicate balance which characterised the pre- lowveld was placed under increasing strain. Livestock herds grew larger – not least because of an improvement in pasture – and men clashed along the grazing frontier, with cattle both the cause and the prize of escalating conlict. A further driver of enlarged warfare – as elsewhere across the continent – was expanding long-distance commerce with the Portuguese on the Indian Ocean coast. In this area, it was the heightened demand for ivory which caused violent clashes among Ngoni over the rich hunting grounds near the coast. The political effect of all this was, again, the enlargement of scale: A number of northern Ngoni chiefdoms in particular expanded and absorbed smaller neighbouring polities, and the age-set system which had been such an integral part of Ngoni society



WARFARE IN AFRICAN HISTORY

was increasingly militarised. The trend in the eighteenth century, as elsewhere, was towards larger military units, and the expansion and militarisation of the age-set among the Ngoni is comparable to the experience of the Maasai and the Oromo, among others. Age-set regiments were deployed as the armed wings of expanding polities, and those of defeated and absorbed societies were integrated into the armies of victors. Ngoni armies were beginning to favour the short stabbing spear, the assegai, as emphasis was increasingly placed on the speed with which one closed on the enemy and engaged in close combat. Ngoni soldiers were still probably mostly part-time in the eighteenth century – as with the Ganda, they balanced military activity with economic duties at home – but the emphasis on speed of manoeuvre and the use of assegais suggests a process of professionalisation involving training and discipline. The outcome of this, by the late eighteenth century, was the emergence of three more or less centralised chiefdoms which now dominated the region: the Ngwane, under the leadership of Sobhuza; the Ndwandwe, under Zwide; and the Mthethwa, under Dingiswayo. Yet in many ways the major clashes were still to come: Around , the era of high rainfall came to a close and in its place came poor rains and drought. Wars of expansion became wars of survival; out of these conlicts would come further revolutionary changes in politics and warfare in the nineteenth century. These changes are largely associated with the Zulu, who at the end of the eighteenth century were only one small part of the Mthethwa federation of chiefdoms. The southeast lowveld was one of Africa’s most creative frontiers of revolutionary change. In other areas, too, notably in the transZambezi region, new heavily armed states arose in what might be considered modern eastern Africa’s ‘irst wave’ of militarisation. In the seventeenth and eighteenth centuries, these were situated in a vast commercial intersection, where the impulses emanating at the Atlantic coast were still felt, and met with those coming from the Indian Ocean – a conjunction of forces especially evident across a zone stretching from Lake Tanganyika to the Zimbabwe plateau. The Maravi state on the southern shores of Lake Malawi, for example, actually had its origins in the early sixteenth century, but evolved into a major military power under the most dynamic ruler of the Kalonga dynasty, Masula, between  and . Masula forged close links with the nearby Portuguese in the Zambezi valley and maintained military control over a large area straddling southern Malawi, northeast

DESTRUCTION AND CONSTRUCTION



Zambia and northern Mozambique. The state’s wealth derived from the ivory trade, but in the second half of the seventeenth century, after Masula’s death, the Maravi disintegrated into several competing chiefdoms. The violent disunity of the area laid it open to Yao slaveraiding predations in the nineteenth century. South of the Zambezi, the Mutapa kingdom was the key military power in the area until its gradual decline – largely the result of succession disputes – in the course of the seventeenth century. It was replaced by a new military polity under the leadership of the Changamire who commanded a famously effective and disciplined army known as the Rozvi, the ‘destroyers’. The state expanded in the eighteenth century, and although foreign trade was important, the basis of the Changamire’s power was careful management of livestock, good farmland and hunting. It is not easy to assess just how much military innovation the Changamire’s army represented: As in the case of Mutapa, this was not a standing army, as mobilisation took place as and when necessary, although evidence suggests structural reform in the creation of distinct military divisions. Firearms were imported in ever greater numbers, but the success of the Rozvi appears to have been based more on discipline and a fearsome reputation rather than any signiicant technological or tactical innovation: Relatively small forces – around , men was the usual strength – could be mobilised rapidly and dispatched over considerable distances in good order, with a view to extracting tribute or punishing wayward chiefs. However, although warfare may still have been relatively limited in its objectives and probably involved few casualties, the Changamire system was a portent of what was to come in this region in the decades which followed, in terms of highly eficient military structure and increasing professionalism. Elsewhere in southern Africa, other equally violent but quite distinctive frontiers were emerging. In the attempt to consolidate their control over the Zambezi valley, the Portuguese established prazos, or land grants, which were supposed to be held by settlers but which, as a result of intermarriage, gradually came to be controlled by Afro-Portuguese overlords. These commanded large armies of slave soldiers, known as Chikunda, whose recruits were supplied by neighbouring chiefdoms – both war captives and the violently uprooted. As direct control from Lisbon withered in the eighteenth century, the prazeiros became local warlords exercising considerable autonomy. The Chikunda were comparable to the armies of Segu and the Wolof states as armed slave retainers, equipped increasingly with muskets



WARFARE IN AFRICAN HISTORY

and used to hunt elephants as well as to muscle in on trade routes and raid neighbouring prazeiros for tribute. The Chikunda slave-soldier model would be replicated in other areas of eastern and central Africa in the nineteenth century. Meanwhile, to the southwest, the creation of a permanent base by the Dutch East India Company at Table Bay in  signalled the beginning of a new era of relations between European and African in the area – relations which were initially peaceable, with the local Khoisan at irst welcoming the commercial opportunity. But with the establishment of independent farms by Boers – demobilised Company soldiers whose task was now to increase crop production for the colony – conlict quickly escalated. The Dutch settlers took to seizing cattle, prompting Khoisan counter-raids, and before too long the Euro-African frontier had become a militarised one. There were conlicts with the Khoisan in –, and again in the mids. The Dutch commander van Riebeeck’s principle of ‘right of conquest’ – meaning that African land was there for the seizing – was an alien concept vis-à-vis Khoisan notions of warfare, but its brutal implications became clear throughout the late seventeenth and early eighteenth centuries. Dutch soldiers and traders, equipped with matchlocks and later with rather more effective lintlocks, penetrated deeper into the interior, exploiting Khoisan divisions and laying claim to Khoisan pasture. Dispossessed of cattle, many Khoisan organised themselves into hunter-raider communities; their systematic attacks on Boer farms led to a further escalation of violence, as Boers formed militia units to hunt down Khoisan raiders, the latter regarded as little better than vermin to be exterminated. Armed with a particularly strident interpretation of the Bible as well as irearms, Boer settlers justiied their unrestrained brutality towards the Khoisan in racial and ideological terms. While violence intensiied along the northern frontier of the Cape Colony, to the east, the Boer line of advance brought them into conlict with the Xhosa – with whom, again, initial relations were often cordial, involving trade and even intermarriage. But in the s and s, as Boers moved into the Zuurveld, a key Xhosa grazing area, a series of frontier wars began, which would endure intermittently for a century. The Boers found the Xhosa a rather greater obstacle than the dispersed Khoisan; at the same time, frontier settlers, perceiving a lack of support and empathy from the Cape government, rebelled against the latter in , and again in , this time against a British

DESTRUCTION AND CONSTRUCTION



administration which had seized the Cape as part of a global strategy aimed at Napoleonic France. Boer uprisings against colonial rule signiied another cycle of conlict which would only intensify in the nineteenth century.

Wars of Faith, Land and Livelihood In the eighteenth century, as we have seen, the forces of commercialism led to the creation of newly militarised polities across West Africa; at the same time, however, Islamic forces were also central to political and military transformation. During this era, a series of movements emerged across the Islamic world at large, which aimed to restore a purer, truer faith and to reorder the community according to the principles laid down by Muhammad himself. Too many nominally Muslim leaders had become corrupt, and the observance of Islam itself had become lax. Such revivalist, fundamentalist traditions were as old as Islam itself, but they gathered renewed strength and reach in the eighteenth century. In the west African savannah and southern Sahara, wars of faith dated to the fourteenth and ifteenth centuries, the era of Mali and Songhay. In the eighteenth century, revivalist dynamics are associated with the Fulani, or Fulbe, pastoralists who following a series of migrations came to inhabit a swathe of territory from the Senegal valley, across the central Niger delta, to the savannah of northern Nigeria. The Fulani had migrated long before they became Muslim, and were central to the regional economy, to which they contributed milk, meat, hides and manure; their hides were particularly valued in the making of harnesses and sheathing. However, the gradual Islamicisation of the Fulani was the outcome of several dynamics. One was prolonged interaction with other Muslim communities – and particularly traders – across the Sahel, notably the Tuareg, encountered as the result of migration in search of superior pasture and (often) political asylum. Another was a sense of persecution and marginality, for in some areas Fulani settlers were confronted with ill-disguised hostility from sedentary populations and ruling elites who sought to restrict their grazing and trading rights. According to this view, Islam offered an alternative form of government with which to confront would-be oppressors and provided some degree of ethnic solidarity and a common purpose around which to arrange a broad set of grievances. Although caution is



WARFARE IN AFRICAN HISTORY

required in the face of overly neat explanations, it is little coincidence, at the very least, that by the early s, Fulani numbered among the leading Muslim scholars and preachers of the Sahel belt, giving rise to the notion of an increasingly politically active ‘Fulani aristocracy’. Like their counterparts in Arabia and India, Fulani religious leaders were ever more vehement in their condemnations of those they identiied as lapsed Muslims, and invoked the concept of jihad against ‘unbelievers’. There was nothing intrinsically novel about this in the savannah zone, witness, for example, the Malian leader Muhammad Ture’s waging of jihadist war against transgressors and inidels two centuries earlier. What was perhaps distinctive about the Fulani-led movements of the eighteenth century was the bottom-up nature of such activism, and the fact that they were often instigated by outsiders, as opposed to earlier top-down exercises in the construction and consolidation of states in the name of Islam. Eighteenth-century spiritual warriors preached the need to combat not only ‘unbelievers’, but also – and altogether more odious – backsliding political and socioeconomic elites whose disregard for shari’a and even the most basic tenets of the faith was considered an outrage. The brewing conlict among the Hausa states – scattered across the area of present-day northern Nigeria, Niger and Burkina Faso – demonstrates the case in point. The Hausa, who emerged out of an intermixing of savannah farming communities and southern Saharan nomads, had been exposed to Islam through commerce from an early date: Certainly, by the fourteenth century, Hausa urban elites were mostly Muslim, if not yet the wider populace. Relations between the Hausa city-states were increasingly violent, and evidence suggests that in the course of the sixteenth and seventeenth centuries, heightened levels of warfare greatly undermined agricultural development, and was ultimately paid for through a system of heavy taxation on farmers themselves. Furthermore, as the Hausa became drawn into the slave trade – they were key suppliers of slaves to states such as Oyo – slaving violence accordingly increased, and many of the victims were themselves Muslims, Islam having spread beyond urban centres into the countryside. The enslavement of Muslims by Muslims, illegal in Islamic law, combined with the economic predations of Hausa warrior elites, meant that by the eighteenth century, immigrant Fulani activists – viewed with deep suspicion as outsiders by Hausa leaders – found a political, social and spiritual environment ripe for insurrection. Sermons concerned with the corruption and virtual apostasy of

DESTRUCTION AND CONSTRUCTION



ruling elites resonated with a Hausa Muslim community ever more ready to use violence in pursuit of a range of objectives both temporal and spiritual. The growing popularity of many Fulani religious leaders prompted violent responses on the part of Hausa elites who sought to crush or drive away emergent Islamic movements. The irst of a series of jihadist wars broke out in the hills south of the River Gambia, in the region of Futa Jalon, in ; a series of campaigns, during which commanders relied particularly on bowmen, lasted in the area until around  and produced a uniied theocratic state which served as a model for later jihads. The Fulani had been leaders and organisers in Futa Jalon, as they were on the Senegal River where a jihad erupted in . This second major uprising led to the creation of the Futa Toro kingdom in . In both cases, tentative beginnings gave way to increasingly conident insurrections which were joined by Fulani resentful of taxation and inspired by early victories; and as we have seen in earlier contexts, many of these ighters became passionate Muslims in the course of actually ighting for Islam. These armed theocracies provided inspiration across the savannah. It was another Fulani, Usman dan Fodio (–), from a scholarly and religious family, who began preaching the need for the proper application of shari’a in the Hausa state of Gobir in the s. In time, he formed a separate community and attracted a signiicant following through the s and s as a result of his condemnations of Gobir’s rulers. His eventual armed revolt would fundamentally reshape West Africa.

Relections: The Local and the Global In the course of the seventeenth and eighteenth centuries, extant patterns of warfare and military organisation were exacerbated, and new forms were developed. Yet even though common threads can be discerned across the continent – sharper conlicts over identity and land, an increase in the scale of warfare, ever more professionalised military systems and ever greater fusion between those systems and the political order – the fact remains that Atlantic Africa was distinguished from the rest of the continent in its engagement with a burgeoning and pervasive global economic system. The trans-Atlantic slave trade reached its apex between  and  and drove forward the militarisation of a multitude of states and societies between Senegal and Angola – a



WARFARE IN AFRICAN HISTORY

process which inevitably involved both destruction and construction. Although many newly militarised polities arose on the back of the international trade, their practices and cultures of warfare were discernibly African, predating the oceanic commerce in human beings and representing forms of armed entrepreneurship which could be traced to the very foundations of sub-Saharan state and society. The slave trade offered new opportunities for the armed entrepreneur, a igure who looms large in Africa’s deeper past, but with whom we can associate much of the growing military professionalism of this era in particular. He, and on occasion she, inspired new forms of creative thinking about the cultural and political signiicance of organised violence – thinking which unquestionably led to the formation of some remarkable militarisms, but which also led to the death, or the enslavement and exile, of millions of Africans. Africans, by and large, remained in control of both the trade itself and of their wars; the instability which gripped the Atlantic region destroyed some communities at the same time as it enhanced others, which were in a better position to take advantage of it. In areas most directly involved, the era of the slave trade initiated many of the long-term processes of violent change which would become evident in the nineteenth and twentieth centuries. The awful human tragedy notwithstanding, the slave trade compelled a remarkable political energy and dynamism which manifested itself in a series of centralised states, characterised by vigorous military cultures. External trade, then, facilitated political creativity – albeit creativity which lourished at the expense of a host of communities on the peripheries of expanding, predatory militarisms. Thus long-extant patterns of political and economic centres preying on contested borderlands were reinforced. In many respects, this represented the beginnings of a military ‘revolution’ – indelibly intertwined with transformations in Africa’s economic relationship with the Atlantic world – which would deepen in the nineteenth century, and which continues to shape the continent today. Developments in Atlantic Africa – notably in terms of the ‘slaving state’ – were characterised by a particularly violent energy. But in eastern and southern Africa, too, large-scale militarisation was taking place, and heightened conlicts were unfolding over access – to people, resources, land. Warfare in these regions was probably not as devastating, nor as professional, in comparison with many Atlantic African societies, but the trend was clearly in this direction. Faith and identity both drove these wars and were called into service to justify

DESTRUCTION AND CONSTRUCTION



them, as they were on the western side of the continent. And as in the vast Atlantic zone, there were winners and losers from the Ethiopian Highlands to the Drakensberg mountains; but everywhere a violent creativity was in evidence, inspired by battles over ideas, goods and, above all, people. It needs to be noted, too, that across this vast region global commerce was also rapidly expanding by the late eighteenth century, and eastern Africa was increasingly linked to a resurgent trade network in both the Red Sea and the Indian Ocean. These external intrusions would intersect with local conlicts, the protagonists of which would use global opportunities to pursue local agendas. This dynamic would become more visible in the century which followed. Across the continent, in practical terms, the intensiication of external contact led to the increasingly widespread adoption of imported weapons – irearms, in other words – and the shifts in tactics and organisation which this invariably involved. The fact that much basic materiel de guerre was widely available meant that it was dificult for any sub-Saharan socio-political system, state-led or otherwise, to establish a monopoly on the capacity to wage war, or to achieve the kind of technical innovation which would provide them with decisive advantages over adversaries. It was in this context that irearms were increasingly important, and in certain areas of particularly intense commercial activity, something approaching an arms race emerged in the seventeenth and, especially, the eighteenth centuries. Of course, the reach of the gun was limited on the eve of the nineteenth century, certainly beyond coastal zones and their immediate hinterlands; it is also the case that mere possession of irearms did not in itself bestow tactical superiority – in fact, sometimes quite the reverse – and that they were not uniformly deployed in an effective manner. Ammunition supply, moreover, continued to be the major problem for many societies seeking to arm themselves with this new technology. It was not uncommon for unloaded irearms, never to be deployed in actual combat, to acquire a prestige value of their own. Nevertheless, the direction of travel was clear enough by the late eighteenth century, and it was toward the import of ever more effective foreign technology, to be adapted to local conditions, integrated into local military structures and deployed in local conlicts. Guns would be a critical part of the rapid developments in organised violence which transformed the continent in the nineteenth century. Ultimately, this was an era in which the intersection between the local and the global became all the more intense. The key dynamics



WARFARE IN AFRICAN HISTORY

and parameters of war and military growth remained African, but war was both driven by internal considerations and waged to achieve the best position from which to take advantage of external opportunities. From the perspective of long-term African development, key sources of inance and equipment now lay beyond Africa itself – the beginning of Africa’s ‘reversal of fortune’ in global economic terms, which was by no means coincidentally followed by the quickening militarisation of the continent. It was an imbalance which led to African warfare being both more total in its nature and more mercantilist – the destruction of rivals being not merely desirable but necessary – thus exacerbating traits which had long been present in African war. Violence also became in some ways more intractable, because competition over external opportunities, inite as these were, left little room for manoeuvre and certainly greatly diminished the chances of demobilisation once escalation had begun. The period between  and , then, reveals much commonality across the continent as a whole. Firstly, the era witnessed the beginnings of signiicant reform and innovation in terms of politicomilitary and ultimately social organisation. The pressures building up, for example, along the Atlantic seaboard in the era of the overseas slave trade were creating movements for change which would be unleashed in the nineteenth century. The continent’s tragic loss in terms of population and global positioning nonetheless forced a violent creativity with which much of Africa is still living. Secondly, much political leadership across the continent – certainly from early in the second millennium – was indelibly intertwined with military endeavour, command in battle, heroism, courage and honour. Thus there were political and cultural mechanisms in place for military innovation, because talented adventurers could quickly establish followings and communities – one of the key elements in the nineteenth-century revolution. Particular patterns in terms of the militarisation of the polity emerge from a continent-wide perspective. Both Oyo and Ethiopia, for example, were facing collapse at the close of the eighteenth century; and although many of the factors in each case differ markedly, both cases demonstrate the power of the army in the polity. Central authority had created, and subsequently sought to control, sophisticated and potent military force; but armies soon became alternative centres of power, while the alain and the negus negast quickly realised that wealthy soldiers and warlords could not be easily controlled. This was a risk run by all the key military states of the era, although some managed the situation

DESTRUCTION AND CONSTRUCTION



better than others. In some respects, ‘compact’ militarisms – such as Dahomey and Buganda – fared rather better, and in the eighteenth century represented emerging powers with capacity to spare. Into the mix was sometimes added deep-seated religious belief, giving rise to charismatic and indeed messianic leadership, as in the revivalist jihads of western Africa in the eighteenth and nineteenth centuries and, as we see later, the zeal of Christian leadership in the Ethiopian Highlands in the nineteenth century. No doubt in part these were both related in some way to shifts in the relationship between Africa and Europe; but in the irst instance, they came from a lexible responsiveness which was characteristic of African political culture, a responsiveness that often, in this era, took violent and innovatory forms. Thirdly, the very mobility of many African communities meant, given the correct socio-economic circumstances and political leadership, military professionalisation was possible, and very abruptly so. It is also the case that the military states of this era were in many respects fundamentally unstable because of the lack of adequate rules of internal succession, which in turn produced heightened levels of militarisation and the fusion of the military and the political. This was already clear in many parts of Africa in the seventeenth and eighteenth centuries; all of these processes would become clearer still in the nineteenth.

Further reading Alpers, E. A. Ivory and Slaves in East Central Africa: Changing Patterns of International Trade to the Later Nineteenth Century (London: Heinemann, ). Birmingham, D. Trade and Conlict in Angola:The Mbundu and Their Neighbours under the Inluence of the Portuguese, – (Oxford: Clarendon Press, ). Kea, R. ‘Firearms and Warfare on the Gold and Slave Coasts from the Sixteenth to the Nineteenth Centuries’, Journal of African History, : (). Klein, M. ‘The Impact of the Atlantic Slave Trade on the Societies of the Western Sudan’, Social Science History,  (). Law, R. ‘A West African Cavalry State: The Kingdom of Oyo’, Journal of African History, : (). ‘Horses, Firearms and Political Power in Pre-Colonial West Africa’, Past and Present,  (). The Oyo Empire, c. –c. : A West African Imperialism in the Era of the Atlantic Slave Trade (Oxford: Clarendon Press, ).



WARFARE IN AFRICAN HISTORY

The Slave Coast of West Africa, – (Oxford: Oxford University Press, ). Lovejoy, P. Transformations in Slavery:A History of Slavery in Africa (Cambridge: Cambridge University Press, ). Oliver, R. ‘Discernible Developments in the Interior, c. –’, in R. Oliver and G. Mathew (eds.), History of East Africa Vol. I (Oxford: Clarendon Press, ). Smith, R. S. Warfare and Diplomacy in Pre-Colonial West Africa (London: James Currey, nd ed., ). Thornton, J. ‘The Art of War in Angola, –’, Comparative Studies in Society and History,  (). Warfare in Atlantic Africa, – (London: UCL Press, ).

CHAPTER 

Transformations in Violence Military Revolution and the ‘Long’ Nineteenth Century

In the early nineteenth century, a series of intertwined processes, combined with a concatenation of unrelated events, instigated a longterm revolution in military and political affairs. In part, this was the outcome of the new era of economic relations with the West, across Atlantic Africa; at the same time, the so-called illegal slave trade perpetuated and intensiied earlier patterns of organised violence, while the slave trade escalated on the eastern side of the continent, leading to new military and political forms. Elsewhere, heightening internal competitions for resources and religious fervency combined to produce novel military processes, or more effective variations of old ones. By mid-century, a range of new states and societies, as well as evolving versions of existing ones, was emerging through violence; violence itself had become one of the key expressions of political and cultural creativity, and war became ever more important as an extension of economic policy. The equipment at Africans’ disposal became more sophisticated and more effectively deployed; the organisation and structure of African forces became ever more eficient and more professionalised. The transformation in military affairs was still evolving in the late nineteenth century with the onset of the European partition, the latter driven in no small part by a wilful or unconscious misreading of events in Africa. Yet the ‘scramble for Africa’ itself – in fact a multitude of invasions and prolonged armed incursions – was made possible by Africa’s own military innovations, not least in terms of the manpower available to a small number of European oficers and Africans’ tactical skill and knowledge of both enemies and terrain. 



WARFARE IN AFRICAN HISTORY

Europe was also able to take advantage of the intense intercommunity and interstate conlict across the continent which was itself one of the clearest manifestations of the military revolution; comparatively small European forces, with frequently superior weaponry and better organisation and drill, were able to insert themselves between states and societies and exercise considerable leverage over them. In a sense, therefore, the era of the partition may be considered an extension of a nineteenth-century African military revolution which has not yet run its course.

Deining ‘Revolution’ The military transformation envisaged here involved the concentrated, institutionalised and increasingly innovative deployment of violence in the pursuit of political and economic goals; this was war as production and construction as well as destruction. The key drivers included the reach for resources at the local level, in some areas almost certainly motivated by environmental and demographic shifts, and the advance of the global commercial frontier in the form of long-distance trade. Rapid economic change, ideological fervour, demographic shifts and political crisis – and occasionally a combination of all of these – gave rise to violent creativity, and a great deal of brutality, across the continent. The key features of the revolution included the professionalisation of the military, new modes of command and organisation and the increasing adoption of irearms, attended by tactical innovations. An estimated  million guns may have looded into Africa in the course of the nineteenth century. The impact of guns was, of course, highly variable. Trade guns were often unreliable and even dangerous to the user; many communities did not deploy them effectively. But where effective deployment and training were present, irearms awarded many states and societies new potency in the course of the nineteenth century: the Yoruba, the Nyoro, the Nyamwezi, the Ovimbundu, the Asante, the Amhara – all accumulated large arsenals of lintlocks and reorganised their armies accordingly. Firearms almost everywhere facilitated greatly expanded military horizons and underpinned these new military forms. Flintlocks – lighter and easier to load than matchlocks – had been increasingly in use from the early eighteenth century, facilitating looser tactical formations involving smaller, more mobile units of soldiers, and allowing a greater emphasis on the concentration

TRANSFORMATIONS IN VIOLENCE



of ire. However, the technological gap between European and African armies widened signiicantly with the development in Europe of breech-loaders in the s. Soldiers were able to feed cartridges into the chambers of breech-loader riles instead of reloading powder and ball into the muzzle – a painstaking and time-consuming business by comparison – thus dramatically increasing the rate of ire and the mobility of infantry units. Many African armies acquired breechloaders in the s and s, but usually only in small numbers, and they remained beyond most Africans’ reach, thus awarding European forces a decisive advantage in irepower. The era was also characterised by an enlargement in the scale and vision of political violence. In political terms, the nineteenth century in many ways witnessed the culmination of community- and state-building exercises in progress since the fourteenth and ifteenth centuries in some areas, more recent and older still in others. New forms of statehood, or at least greatly enhanced systems of identity and organisation, often resulted. The aim across the continent was cohesion and unity – as among the Ganda and the Zulu, for example – and this was sometimes posited as reuniication, for ideological reasons, for example among Mirambo’s Nyamwezi, or in Solomonic Ethiopia. In the second half of the nineteenth century, there was also the external threat, which arrived in different forms and which was interpreted differently according to local circumstances and ideology, but which everywhere prompted innovation. Ultimately, the goal was the maximisation of human resources, usually at the expense of someone else’s – as had always been the case, of course, except that this was now the reinement of total war, and war with larger ends in view. Violence was aimed at the creation of wealth, sometimes succeeding but at other times creating wider and more concentrated pools of poverty, dispossession and marginalisation. Over the longer term, clearly, the revolution involved marginalisation as well as innovation. Across the continent, social and economic systems came to be underpinned by violence. Meanwhile we also see an upsurge in urbanisation across the continent as a result of heightened violence, and the political centralism that it often entailed. Some of the most dramatic instances of wholesale change involve igures who sought socio-economic as well as military change. Muhammad Ali in Egypt perhaps best embodies the phenomenon, and was in many ways the most successful of the nineteenth-century revolutionary state builders: He perceived that sustainable military



WARFARE IN AFRICAN HISTORY

change could only be achieved through the transformation of Egypt’s domestic economy, in terms of both commercial agriculture and industrialisation. Indeed, it was his very success which prompted European intervention to curtail his expansionism in the s. There were others who shared this more holistic vision later in the century, notably Samori in the West African savannah and Menelik in Ethiopia, the latter seeking economic expansion to support an increasingly costly but effective military establishment. But although a host of other states and societies fell some way short of Menelik’s or Muhammad Ali’s achievements, many were clearly moving in the same direction, and were seeking to engender the kind of social and economic change needed to underpin military ambition. The results were mixed, but the vision was much in evidence. Finally, the notion of a ‘revolution in military affairs’ is most closely associated with Europe between c.  and c. , during which time western and central Europe experienced a radical transformation in the organisation and nature of warfare, involving new technologies and cultures – notably portable irearms and professional soldiery – and facilitating (indeed necessitating) the rise of modern states. This was a revolution in military hardware and political systems, then, but it also involved transformations in economic and social organisation. Attempts to make direct comparisons between Europe and Africa in this context are dangerous, but it is important to bear in mind the broad precedent thus established in Europe for the dynamic – and intimate – relationship between military revolution and political, economic and social change. Africa’s nineteenth century, indeed, can be considered as something of a ‘golden age’, during which some of the restrictions and inhibitions described earlier in the book began to loosen, in certain ways, to produce the kinds of changes under discussion. The revolution suggested here did not happen everywhere (nor did it, it might be pointed out, in Europe), but there are numerous instances of dramatic change and innovation, in different ways, in different places and at different moments. During the ‘long’ nineteenth century between c.  and c. , there were new forms of war, new military organisations, the expansion of state power and socio-economic change. Of course some of the elements of the putative nineteenth-century revolution in military affairs were present well before , as we have seen, including, notably, irearms, and in some areas the increasing professionalisation of the military corps. But the ‘long’ nineteenth century witnessed substantive shifts in the deployment and manifestation of

TRANSFORMATIONS IN VIOLENCE



these tools, heralding the crystallisation of a series of processes which arguably had their roots several centuries earlier.

A Revolution in the Practice and Culture of War The global backdrop to change in western Africa was the abolition of the slave trade by several European countries, staggered over several decades, which in some areas of Atlantic Africa led to the adoption of new economic systems based on cash crops. Nonetheless, the ‘illegal’ (in the European view) slave trade persisted through much of the nineteenth century, alongside ‘legitimate’ commerce, and long-term economic shifts did not necessarily involve demilitarisation – indeed, quite the reverse, in some areas. This was in large part because slave labour was needed to generate the agricultural exports which formed the basis of ‘legitimate’ commerce – and warfare continued to be the source of slaves themselves. As the export trade expanded, so did domestic slavery – and so, too, did the scale of violence needed to meet demand. To a very real extent, therefore, long-term commercial shifts actually facilitated an expansion in military scale – an uncomfortable truth for the humanitarian lobby in Europe which had long advocated trade in agricultural produce. That same lobby, meanwhile, had to countenance the resilience of the external slave trade itself for several decades after its supposed abolition. For more than a century, Oyo had imposed itself over the Benin Gap, its imperialism establishing a pax internally and directing its violence outwards; but now, weakened by the decline of the slave trade as well as provincial revolt in Ilorin and the advance of Fulani jihadists from the north, Oyo during the s and s collapsed into civil war. What followed was a half-century of bloody vitality and military reform, carried through by the Yoruba city-states which succeeded the old Oyo imperium – Ijaye, Ibadan, the Egba and the Ijebu, who were among the most assiduous importers of guns into the region and exporters of slaves out of it. These city-states – a demonstration of the fortiied urbanism which was another key feature of the age – were often governed by war leaders in charge of newly professionalised bands of young soldiers equipped with both matchlocks and, more potently, lintlocks. Flintlocks – known locally as ‘Long Danes’, owing to their Danish manufacture – were preponderant in Yoruba warfare until the late s, when some breech-loaders were



WARFARE IN AFRICAN HISTORY

imported. Firearms largely displaced indigenous weapons, if never completely; soldiers became trained in marksmanship, and fought in relatively open order. Often these forces were essentially mercenary armies, attaching themselves to the wealthiest leaders with the most to offer. It was a development which well exempliies the phenomena of the age. As commercial competition increased, and global trade itself reached deeper into communities in terms of both social impact and geographical range, it drove new forms of professional violence. The Yoruba wars resulted in large-scale population movement, broadly north to south, as people led from the open savannah lands to the forest for protection from slaving expeditions; in the forest, too, there was safety from Fulani attacks on horseback. By the same token, urbanisation increased rapidly, and towns were typically protected by ditches, moats, defensive mounds and wooden stockades. Armies developed skill in the use of camoulage and siege tactics accordingly, as would be the case in eastern Africa, too. There was also a dramatic increase in slaving campaigns within Yoruba territory, previously largely exempt from the slave trade owing to the security provided by Oyo whose own slaves for export had largely come from the north. The rise of the Sokoto Caliphate had largely closed off those sources, while Oyo’s collapse had rendered countless communities vulnerable to the violently competitive predations of the city-states which succeeded it. War, fought essentially for access to global trade networks, offered potentially very high returns, and its practitioners became increasingly professional as a result. A major player in these events, too, was Dahomey, now one of the region’s most dynamic ‘illegal’ exporters of slaves, notably during the reign of Gezo (–). Dahomean forces preyed on Yoruba settlements to the west, formerly protected by Oyo but now vulnerable in an age of lux. In particular, Dahomey’s wars with the Egba state escalated from the early s, largely over commercial highways. In the s, the opportunities for slave export available to Dahomey evaporated rapidly, and the state, as with many others along the coast, turned increasingly to palm oil production; but the new economic reality did not denote demilitarisation – indeed, the production of palm oil was attended by a signiicant expansion in the use of domestic slavery, as it was elsewhere in Atlantic Africa. War remained critical to the functioning of Dahomean state and society: Gezo himself had come to power following a coup d’etat against Adandozan, whose reign had been marred by a distinct lack of military glory. Gezo had become

TRANSFORMATIONS IN VIOLENCE



king promising just that, and accordingly was under a great deal of pressure to keep the army active as a result, as he explained, somewhat pleadingly, to a British oficial in . Commercial rivalry only fuelled further conlict, with competition increasing for control of, or access to, key ports in order to acquire the guns and ammunition necessary for economic and political dominance. Yet this was not ‘simply’ about the exclusion of rivals from the marketplace: Warfare was increasingly aimed at the total destruction of competitors, evidenced by the fact that in  and again in , Dahomey launched unsuccessful attempts to seize the Egba capital itself, Abeokuta. Such ferocious assaults, notably, were used by the European missionary community based in Abeokuta as evidence of Dahomey’s savage bloodlust. But in fact, the kingdom’s army was comparatively compact, certainly when set alongside earlier estimates of coastal armies, as well as that of Oyo in the previous century. At the core of Dahomey’s army was a force of some ,–, regular soldiers equipped with lintlocks, although this might rise to approximately , if the need arose; and the kingdom was distinguished by its deployment of female soldiers in the nineteenth century, perhaps up to ,–, of them at any given time, under female commanders. Few nineteenth-century African armies made use of women in quite this way. The kingdom remained the major military power in the region until the eve of the French invasion in the early s, and was one of the most militarised cultures anywhere on the continent. Further west, Asante – another monarchical state system with its roots in an earlier age – sought to more carefully balance commerce and war, under a highly effective ‘bureaucracy’ and, indeed, a skilled diplomatic corps. Here, it is possible to discern, broadly, an internal competition between an ‘imperialist’ party which sought the generation of wealth and security through military aggression, and a ‘peace’ party which sought the development of trade. Diminishing returns from slave exports from the s onwards strengthened the hand of the latter group, which in turn pushed forwards Asante’s participation in ‘legitimate’ trade, and from the s and s successful merchants – and not just great soldiers – were accorded considerable honour and respect within the political establishment. Of course Asante had always enjoyed greater economic diversity than many states along the Atlantic littoral, but war remained important as an extension of ‘national’ policy: Units of men equipped with irearms – the ‘Long Danes’ common among the Yoruba, too – were increasingly necessary for the subjection



WARFARE IN AFRICAN HISTORY

of surrounding areas and rebellious provinces alike. The army itself comprised a core unit – rigorously trained, by all accounts – stationed at the capital, Kumasi, which was complemented by regiments raised in the provinces; it was a potentially fractious arrangement which nonetheless repeatedly proved robust owing to a highly effective command structure and a culture of military prowess rooted in a carefully maintained sense of Asante’s glorious martial past. Conlict with the coastal Fante states, lying on Asante’s southward lank, was ongoing. The Fante regularly stirred up insurrection within the Asante empire, as well as being successful commercial middlemen representing an obstacle to Asante’s own direct trade links with the coast; they also had the British as allies. Periodic Asante attacks on the Fante invariably brought the British to the aid of the latter – most dramatically in –, when the British invaded Asante and inally decided, after decades of irresolution, on the formal annexation of the Fante coast. They withdrew from Asante itself, but now – as with attitudes to Dahomey – the Asante came to be seen by the British as the archetypically barbaric, slave-trading African empire, the fundamental root of instability north of their precarious Gold Coast colony. Perceptions of Asante had a wider resonance, for British military thinking about such polities – inluenced in the irst instance by a group of oficers, including General Wolseley, who had seen in action in the – Asante campaign – moved sharply towards the view that they were, ultimately, obstacles to progress, and that they needed, in the end, to be destroyed. Dahomey, Asante and the Yoruba city-states were the most dramatic manifestations of the highly competitive war economies proliferating across Atlantic Africa, but they were representative of dynamics at the more localised level. In the Niger Delta, for example, a host of smaller-scale communities – trading ‘houses’, such as Bonny, Calabar and Opobo – competed violently for domination of the lucrative palm oil trade. Across this vast region, among small commercial polities and larger territorial states alike, domestic slavery expanded, fed by wars which continued to be fought for the purpose in hinterlands between Senegal and Angola. While such patterns of violence continued to sustain old political and military elites, however, it is clear that such elites often faced what has been described as a ‘crisis of adaptation’ – meaning that while the slave trade was organised by large-scale state monopolies, run by warrior elites, trade in mineral produce and cash crops did not necessarily involve any special advantages for military

TRANSFORMATIONS IN VIOLENCE



leadership. Participants in the new ‘legitimate’ commerce were often smallholder farmers utilising family labour. It meant, in some areas, a challenge to older, slave trade–era military hierarchies; certainly, as irearms looded into Africa in the course of the nineteenth century, they were acquired by small-scale traders as well as by states, and socio-political disruption was often the result. The Niger Delta illustrates the point, for here upstart merchant communities with economic as well as military muscle were able to set up on their own, and commercial centres proliferated as a consequence. The models of both newly created professional military corps and expansionist and innovatory states had their echoes across the continent. Eastern Africa between Lake Victoria and the Limpopo River was transformed by the rapid expansion of long-distance commerce, and in particular the slave and ivory trades, driven by the rejuvenated Sultanate of Zanzibar, between the s and the s. Across this vast region, existing military systems were greatly enhanced, and new patterns of militarisation created, as a result of socio-economic change – change which was then further driven, in symbiotic fashion, by the new military systems themselves. Much has been made of the key differences between west and east Africa in the nineteenth century, with a focus on the decline, however gradual, of the slave trade in the former region and its escalation in the latter. Undoubtedly there were indeed differences, but these should not be exaggerated. In the Atlantic zone, the slave trade remained robust in many areas for much of the period under examination, and the militarism rooted in an earlier era remained very much intact. Furthermore, slaves needed for domestic purposes continued to be procured through warfare, and the rise of the Islamic states in the savannah, examined later, also drove both the slave trade and military mobilisation across a wide area. Likewise, in eastern Africa, warrior elites sought to control the trade which brought them both guns to wage war and goods with which to reward loyal followers. Kingdoms of long standing such as Buganda and Bunyoro had much in common with Dahomey and Asante, in terms of both their deeper roots and their attempts to seize control of approaching commercial and strategic opportunities. During the half-century encompassed within the reigns of Suna (c. –) and Mutesa (–), Buganda proitably exported slaves and ivory for cloth, guns and other goods, and a highly competitive military establishment sustained the state, in political, economic and cultural terms. Armies were deployed to secure commercial interests



WARFARE IN AFRICAN HISTORY

in the corridor west of Lake Victoria, and indeed on the lake itself: The Ganda, frustrated by the limitations of war on land, created one of the continent’s largest canoe leets from the s and s onwards, designed to render Lake Victoria a Ganda sea. Meanwhile, organisational reforms also increasingly militarised the political establishment at home. Yet in many respects, Bunyoro was rather more successful at such internal restructuring; in particular, Kabalega’s creation of new regiments of rilemen, the barusura, in the s was critical in the kingdom’s late resurgence on the eve of colonial rule. These rapacious musketeers were a burden on the community, to be sure, and were in many ways an ill-disciplined bunch; but Kabalega was able to harness their aggressive zeal to Bunyoro’s beneit. In Buganda, by contrast, territorial overreach and a political culture which was increasingly self-possessed and riddled with internal fractures undermined military performance in the late nineteenth century. The Ganda also made less effective use of guns than the Nyoro, as young rilemen were awarded unwarranted privilege in command structures, and the military establishment became bound up with court rivalries. Some older warriors actually considered guns the instruments of cowardice, diminishing honour on the battleield. Still, in both kingdoms, revolutions in military affairs – whether evident in naval strategies, deployment of irearms or the ever more intimate association between political leadership and military command – heightened the ierce competition between the two states, a conlict which spilled out across the lacustrine region and into which the British would be co-opted in the s. Further south, new war bands and state formations more akin to those in Yoruba country were emerging. The boom in the slave and ivory trades across eastern and central Africa throughout the nineteenth century facilitated new forms of organised violence and new levels of militarisation. The Nyamwezi and the Yao were increasingly in the vanguard of political and commercial change, as soldiers and as traders, from the southern shores of Lake Victoria to the Zambezi valley. Across this vast zone of violent change, new militarised leadership arose and communities and identities formed around it, if sometimes only for brief periods. Among the more noteworthy examples are Mirambo, a brilliant Nyamwezi political and military leader whose new state dominated much of the Tanzanian interior in the s and early s; Nyungu-ya-Mawe, who did the same among the Kimbu; and Msiri, ‘the mosquito’, a Nyamwezi commercial adventurer and

TRANSFORMATIONS IN VIOLENCE



warrior who created a substantial private army and a loose hegemony in Katanga, known as Garenganze. Mirambo’s short-lived but extraordinarily dynamic polity exempliies the complex concatenation of dynamics. In many respects his own military revolution was a reaction to burgeoning commercial opportunities, but also to the political inluence exercised by coastal merchants over the neighbouring Nyamwezi chiefdom of Unyanyembe. Here, Arab traders based at the crucial trading town of Tabora had instigated a civil war between the late s and mid-s as part of a strategy of installing a ruler who would protect their economic interests in the area. The deposed ruler, Msabila, organised himself in the bush, and the ierce guerrilla campaign which resulted prompted the militarisation of the communities caught up in it; Mirambo, who inherited a small chiefdom to the west of Unyanyembe, was the most dramatic manifestation of that process. He created a much larger military realm with a view to both seizing the commercial opportunities represented by coastal traders and restricting their political inluence. His new state dominated the region between Tabora, Ujiji and Lake Victoria through the s and early s, with Mirambo usually at the head of a highly mobile, drilled and (on the whole) disciplined army, equipped with lintlocks and specialising in rapid marches resulting in surprise attacks – often, unprecedentedly, at night. He spent much of his reign at war with Unyanyembe; indeed, the central dilemma of his reign was the disruption such conlict caused to trade, on which he depended for supplies of guns and ammunition, for which he traded slaves and ivory. In the end, his politico-military ‘system’ proved ephemeral, too centred as it was on Mirambo himself, and it deteriorated rapidly after his death in . In many respects, rapid military change had outrun the political reform required to create more robust systems. The case of Tippu Tip – the Arab-Nyamwezi trader who carved out a restless, mobile ‘empire’ across western Tanzania and the eastern Congo and traded and raided in equal measure from the s onwards – is similar. He was a slave-dealer – his military power was based on the deployment of slave soldiers, and he used slaves on his own plantations – but his wealth derived mainly from the export of ivory, itself a violent business requiring large numbers of armed men to both hunt elephants and protect bulky valuable goods in transit. Both Mirambo and Tippu Tip exempliied – and were the most successful manifestations of – the privatisation of violence in eastern and central Africa in this period; but such war economies were



WARFARE IN AFRICAN HISTORY

ultimately inimical to long-term development, failing to address the war/commerce conundrum – that is, the degree to which violence actually hindered trade. At the same time, for all their military brilliance, these were porous and intrinsically unstable systems based on the violent extraction of raw materials – including human beings – and were therefore unsustainable over the long term, not least in terms of the slave trade, because of abolitionist forces gathering on the horizon. There are indications that Mirambo, for example, envisaged a rather more diversiied – and by implication less militarised – economy at some future date; but almost everywhere, again, little attention was paid to the creation of more permanent frameworks of governance which might have harnessed and built upon the creativity evident in the military revolution alone. Mirambo and others had been at least partly inluenced in the s and s by the strangers who had arrived east of Lake Tanganyika with new ideas about military organisation and new methods of war. These were Ngoni, offshoots of a series of violent upheavals hundreds of miles to the south which had begun half a century earlier, as we see later. Schooled in the ‘total war’ associated with the Zulu, the Ngoni – many of whom were not actually Ngoni at all, but rather displaced peoples who had absorbed the Ngoni culture of warfare – did indeed introduce new levels of violence to eastern Africa, destroying or incorporating less militarily able polities; they were associated with a particularly unrestrained approach to killing and mutilation, and the evidence suggests that by the s, Nyamwezi, Kimbu and Sukuma soldiers were likewise engaged in the killing of women and children with an intensity which apparently had not existed previously. Above all, in purely military terms, exposure to Ngoni attack – if one survived it – was an instructive experience: Mirambo, for example, emulated their speed of deployment, their close-combat tactics, and their expansionist drive, again revolutionising the scale of warfare vis-à-vis what had gone before in the Tanzanian grasslands. Many others sought to adopt the Ngoni model, reorganising and arming for defence, and then for offence as well, thus sustaining the momentum of regional militarisation. In time, however, irearms replaced assegais, and a more individualistic military culture replaced the strict regimental model characteristic of an earlier era. Armed urbanisation, meanwhile, proceeded apace, and fortiied villages and towns became commonplace across a region formerly characterised by dispersed patterns of settlement.

TRANSFORMATIONS IN VIOLENCE



The region in the second half of the nineteenth century was characterised by new leaders with enormously expanded visions of what was possible in terms of social, political and economic change; they were also entrepreneurs of violence, and their programmes were carried out by slaves armed with irearms and a new generation of aggressive and displaced youth, likewise equipped with muskets and increasingly adept at their use. These were the professional soldiers known as ruga ruga, who bore some resemblance to the Yoruba war boys and the barusura of Bunyoro. To some, the ruga ruga were mere thugs and criminals; it is clear enough, however, that they were part of a self-perpetuating cycle of socio-political reformation, for military states both created the upheaval and displacement which made young manpower available, and then fed off that manpower in pursuit of expanded political visions and commercial proit. Ruga ruga either operated in small, opportunistic bands, raiding the busy commercial highways, or served as rootless mercenaries, or committed themselves to the larger state-building projects proceeding across the region: Notably, they formed the backbone of both Mirambo’s and Nyunguya-Mawe’s states. Among the Ovimbundu and Chokwe of west-central Africa and in the Zambezi valley, there too emerged personal and mercenary armies under leaders trading slaves for guns. Along the Zambezi, private Chikunda armies were commanded by prazeiros, or plantation owners. To the north, around Lake Malawi, the Yao exempliied the creative violence characteristic of the age, with individual warlords commanding private armies of musketeers and becoming ruthless raiders for, and exporters of, slaves in the second half of the nineteenth century. Elsewhere, the militarisation of youth – notably in terms of age-set structures – gathered pace, for example among the Maasai and Turkana, for whom environmental change may have been the motivating dynamic. Again, the emergence of gifted and charismatic leadership, provided by prophets and diviners, was at the heart of this centralising process. Individual leaders now actively encouraged youthful aggression for the material rewards it brought, particularly in terms of the acquisition of cattle, and age-sets among the Maasai became increasingly self-contained military units. In many of these scenarios, older patterns of both leadership and the deployment of force were being supplanted. Across eastern Africa, then, the militarisation of state, society and economy involved new types of armies, new forms of authority, new identities or the arming of old ones and generational conlict. At the same time, the



WARFARE IN AFRICAN HISTORY

proceeds of burgeoning global commerce were unevenly distributed, and economic inequality, more pronounced now than at any time in the measureable past, drove young men to arms in various capacities. Meanwhile, as in western Africa, Europeans espied in eastern Africa bloody insecurity and irrepressible brutality, and were increasingly convinced that intervention would be needed to bring these ‘savage’ peoples to order. They were correct only in the dimmest comprehension that eastern Africa was a violent place in the nineteenth century; they misunderstood, however, what it meant. Much further south, heightened competition over land following a period of population growth in the eighteenth century led to prolonged war between several centralised Ngoni states which made ever more effective use of age regimentation. By the mid-s, a commander named Shaka came to the fore following the death of Dingiswayo, leader of the Mthethwa grouping of chiefdoms, imposing himself and his relatively small clan, the Zulu, on these processes. As a result of what became known as the mfecane – loosely translated as a time of violent upheaval – a Zulu empire-state was created under Shaka based on the carefully managed aggression of male youth who both fought and tended cattle, and at the same time practised sexual abstinence. Their absolute loyalty and esprit de corps, which broke down older group loyalties, was pledged to a novel style of kingship which sat astride a dynamic state system; that state system functioned through internal discipline, and the pursuit of external expansion, through both commerce and conlict. Zulu success was rooted in both tactics and strategy. Tactically, there was the extraordinary speed and mobility of the regiments, the Amabutho, which meant that enemies were routinely surprised and caught unprepared; the highly effective use of the short stabbing spear, preferred over the throwing spear, and the close-order drill which maximised its eficacy; and the famous ‘bull’s horns’ manoeuvre, a pincer movement in which enemies were caught on the lanks by fast-moving Zulu columns. Strategically, the incorporation of conquered districts into the expanding Zulu state, and the conscription of young men into a regimental system, inculcated a new sense of loyalty and identity; in this highly militarised society, ‘off duty’ regiments sustained the Zulu economy by herding cattle and hunting, and regimental towns which represented the military and economic engines of the state were scattered across the territory by the s. Yet the Zulu state was in many ways parasitically unstable. As enemies dispersed, or became armed and organised, the Amabutho found

TRANSFORMATIONS IN VIOLENCE



the early success so essential to the functioning of the polity increasingly dificult to replicate. Moreover, the very roots of the polity itself, characterised by a revolutionary militarism, led to a larger problem, one which had its echo in other war states in the nineteenth century, namely the absence of an adequate system of succession. Shaka, seen increasingly as unhinged and tyrannical, was assassinated in  by a cohort which included his half-brother, Dingane. But Dingane lacked the political and military acumen of his remarkable predecessor, and his reign witnessed signiicant desertion from the Zulu cause and chronic external uncertainty, largely caused by the encroachment of European settlers from the south. Dingane himself was ousted by Mpande in , and although the latter aimed at consolidation in the wake of a debilitating civil war, the lack of effective political infrastructure rendered the Zulu state restive, fragile and externally precarious. The dynamic Cetshwayo reduced Mpande to the role of igurehead between , and upon the latter’s death in , Cetshwayo became king. But although Cetshwayo is regarded as the most gifted ruler since Shaka, even he struggled to cope with the deteriorating economic basis of the Zulu state in the s – rapid population growth, pasture degeneration and the recurrence of drought weakened the polity – to say nothing of the looming British and Boer threats. Cetshwayo’s embracing of diplomacy in the face of these problems is exempliied by his insistence that the British secretary for native affairs in Natal, Sir Theophilus Shepstone, crown him in  in order to guarantee his legitimacy. It was hardly a successful move – the British would invade the kingdom six years later – but it is an indication of the sense of insecurity at the heart of this remarkable polity. Shaka himself swiftly passed into semi-mythology, and in recent years, against the backdrop of dramatic political change in South Africa, some have questioned whether the mfecane really happened at all. The ensuing debate was interesting not only in terms of its speciic historical content, but because it indicated the moral and ideological problems which arose in discussing African violence, even in the deeper past: Perhaps, after all, stories about Shaka and his bloody wars were invented by white land-grabbing racists for whom the depiction of an intrinsically unstable African savagery held obvious advantages. What is clear, nonetheless, is that Shaka’s state – in its professionalised military, its appreciation of violence as a transformative experience and a force for radical change and its economic reorganisation – had its echoes in other parts of the continent in the course of the



WARFARE IN AFRICAN HISTORY

nineteenth century. The Zulu achievement was neither an aberration nor an invention of the fevered white imagination. The mfecane had ramiications beyond the Ngoni homeland, according to the standard interpretation, and had ripple effects across a vast region as migrating groups carried the new tactics and organisation beyond the Limpopo, beyond the Zambezi, and further north still, to the shores of lakes Malawi, Tanganyika and ultimately Victoria. Between the s and the s, refugees from the violence led to the establishment of militarised polities of various kinds across southern Africa. For example, the Ndebele lived an itinerant existence under Mzilikazi, a former lieutenant of Shaka’s, who in the course of the s and s led them from Mozambique, to Transvaal, to eastern Zambia, before inally settling in present-day Zimbabwe in . This highly regimental society differed somewhat from that of Sotho under Moshoeshoe, whose mountain stronghold (modern Lesotho) attracted a range of communities on the run. Moshoeshoe combined military skill with diplomacy to ensure his remarkable new kingdom’s survival. By the time various Ngoni groups reached Zambia, Malawi and Tanzania, a host of societies were emulating their structures and tactics – their regimental system, tight combat formation and short stabbing spear – in large part for defence, as among the Bemba, Hehe, Bena and Sangu, whereas among the Nyamwezi, again, Mirambo’s state replicated many of the military strengths, as well as the political weaknesses, of the ‘Zulu model’. Meanwhile, the Bemba and the Sotho bought guns from Boer traders, and irearms again became important to these rolling military revolutions. Yet southern Africa was being militarised from another direction, too. As Boer migrants, or Voortrekkers, drifted beyond the reach of the British-administered Cape Colony in the s, they opened up new military frontiers in the interior. Small groups of Boers often allied themselves to African communities which welcomed gun-toting partners into local conlicts; the potential power of those irearms was demonstrated to the Zulu in . When Dingane made the panicked miscalculation of wiping out a Boer delegation, the ‘Battle of Blood River’ shortly afterwards witnessed Boer guns mowing down ranks of attacking Zulu warriors in what was one of the most ferocious Afro-European clashes of the nineteenth century prior to the partition. During the s and s, conlict between the Boers of Orange Free State and Moshoeshoe’s Sotho prompted the latter to request British protection. The so-called Boer republics of the interior

TRANSFORMATIONS IN VIOLENCE



remained vulnerable throughout the middle decades of the nineteenth century, but the irepower they possessed promoted a degree of cohabitation between African and European, clashes with Sotho notwithstanding. In most of the contexts already discussed, economic drivers played a prominent role – the competition over pasture and farmland, over access to labour and natural resources, over burgeoning long-distance commerce. In some areas, however, these factors combined with ideological and religious fervour as older traditions of militarism and political centralism sought to reform themselves. The Ethiopian highlands witnessed a state-building process which was based, above all, on a cumulative militarism in political culture and the skilful institutionalisation of violence; but it was driven in large part by the Christian nationalism of Tewodros, Yohannes and Menelik, the sequence of neo-Solomonic rulers responsible for the reinvention of the medieval Ethiopian empire between the s and the s. Again, the professionalisation of the military, the effective deployment of irearms, and the creation of militaristic cultures which channelled expansionist violence outwards were all present. Although the Ethiopian situation has always been seen in particularistic terms, in fact it offers us one of the inest examples of the nineteenth-century military revolution and the violent enlargement of scale, albeit with some unique local lavour. The collapse of the medieval Solomonic state in the late eighteenth century was followed by several decades of intense interstate contest, the constituent parts of the old empire competing over access to increasingly lucrative Red Sea trade, as well as the Solomonic inheritance. It was an era known in the local literature as the zemene mesaint, the ‘era of the princes’, a self-consciously Biblical reference: Between the s and the s, in other words, the Ethiopians were like the Israelites, obeying no law and turning on one another in the most barbarous and godless fashion. It was the culmination of the violent centrifugalism that had deined Ethiopian military history over several centuries, as we have seen. Tigray, Shoa, Gojjam and Wollo were, to all intents and purposes, autonomous kingdoms – with their own political administrations and traditions, economic systems, and, above all, sizeable professional armies, often around ,–, strong – but the concept of a united, pan-Ethiopian empire persisted throughout the violence of the early nineteenth century. Regional overlords aspired to the title of negus negast, king of kings, and to the recreation



WARFARE IN AFRICAN HISTORY

of the Solomonic empire; and as the military struggle escalated, guns became increasingly important. Firearms, of course, had been present in the Ethiopian Highlands since the mid-sixteenth century, but only in the early decades of the nineteenth did armies equipped with lintlocks, and later breech-loaders, become decisive instruments of state power, permitting a much greater concentration, and a faster rate, of ire and much more rapid deployment. But just as important was command of Oromo cavalry, and access to the Oromo manpower and skills-at-arms which connections and alliances with their key chiely lineages opened up. In a long-militarised zone, the transformative power of violence was now being realised in new ways; yet, as in other parts of the continent, the truly creative violence came from the direction of the armed frontier. And it came in the form of a shifta, a bandit or outlaw of minor aristocratic provenance, named Kassa Hailu, later Emperor Tewodros, who operated on the northwest borderlands between Ethiopia and Sudan. From here – he encountered Egyptian armies with lintlocks and quickly recognised the futility of lance and spear against disciplined gunmen – he built up an army and honed his battleield skills, and during the s, Kassa challenged for the imperial inheritance. Ethiopia, indeed, provides one of the relatively few instances in Africa of decisive battleield engagements leading to signiicant regime change, as opposed to more prolonged processes by which the nature of war and polity were transformed: Between the mid-s and the mid-s, Kassa took on and overcame regional rivals in a series of pitched battles in which he made use of intelligence, increasingly effective ields of ire, and cavalry in lanking manoeuvres. In , he was proclaimed Emperor Tewodros, King of Kings of Ethiopia. The reinvention of the Solomonic state was underway – but, above all, this was a military revolution leading to a major enlargement of political scale. Tewodros professionalised the army, creating more regular units, and was successful insofar as he demonstrated the feasibility of a larger, centralised polity based on the more effective use of force. It was a formidable achievement, but it was attended by considerable bloodshed: Much of his reign was spent attempting to subdue a sequence of ferocious provincial insurgencies, and in the end his regime was a horriically violent one, the army reduced to a brutal and traumatised rabble. A major part of his project had been the import of military skills and hardware – he wanted European craftsmen to build him cannon – but, brilliant military campaigner though he had

TRANSFORMATIONS IN VIOLENCE



doubtless once been, he was unable to achieve the monopoly on force essential to his own security. His military success outran any tangible political achievements – and in that sense he had much more in common with leaders such as Mirambo than is normally supposed. Tewodros’s wars were waged against his own recalcitrant subjects as much as the Muslims and Oromo he purportedly despised. He killed himself in  in the face of an advancing British force organised to free some European hostages, but it seems reasonable to suppose that he would not have lasted much longer in any case. The Christian militancy which deined the neo-Solomonic statebuilding project was inherited by his successor, Yohannes (–), who likewise understood the political advantage of a decisively superior military infrastructure. Both men made articulate appeals to a history of violence in the Ethiopian region, adding their own rhetorical appendices to the story of armed struggle contained in Ethiopia’s national narrative, the Kebre Negast. Yohannes recognised the utility of diplomacy, and a lighter touch when it came to bristling provincial rulers, but he was still capable of marshalling considerable armed force and was willing to use it when necessary. Ethiopian armies by the late s were not truly ‘national’ forces in the uniied sense, but unwieldy yet brutally effective conglomerates, comprising contributions to the imperial cause by the key kingdoms. Yohannes died at the head of such an army ighting the Sudanese Mahdist state in , to which we turn later. Into the vacancy slipped Menelik, ruler of Shoa and owner of one of the most potent arsenals of modern irearms, including the inest breech-loaders, anywhere in Africa, acquired through many years’ trading with Europeans on the Somali coast. Menelik, the beneiciary of the bloody pursuits of his two predecessors, was skilled at channelling the restless militarism which had built up in the course of the nineteenth century into a still more professional infrastructure and a ‘national’ cause. It was a militarism which – fortuitously – was given full expression when the Italians invaded in  (we return to this later); but it was also directed at the conquest and exploitation of the vast fertile lands to the south and west in the s and s, as the Solomonic state embarked on its own form of latenineteenth-century imperialism. Amhara soldiers became settlers and landowners; military prowess was rewarded with land and economic opportunity. Modern Ethiopia, then, was founded on armed force and a political culture which emphasised the power of violence and was forged in the heat of a military revolution spanning several decades.



WARFARE IN AFRICAN HISTORY

Meanwhile, militant Islam made advances across Africa throughout the nineteenth century. In the western savannah, in particular, revivalist Islam pushed forwards military and political innovation. The jihads of the eighteenth century in Futa Toro and Futa Jalon drew on a tradition of reformism across the Saharan and savannah zones, and in many ways they presaged the even more remarkable revolution of Uthman dan Fodio among the Hausa states in the early s. The Gobir authorities had attempted to restrict Uthman’s movements, as by the s he was seen as a dangerously subversive character preaching the need for a fundamental reform of Islamic practice in the region; some believed him to be the Mahdi, or divinely appointed saviour, even though he did not claim the title for himself. Emulating the Prophet’s own actions at Mecca and Medina, Uthman withdrew his community from Gobir and declared jihad in . His irst victory, at the battle of Tabkin Kwotto in , may have been fortuitous – the archers’ ire from an apparently impromptu square drove the enemy cavalry and infantry from the ield – but at any rate the tactic was consequently much used. This critical opening triumph was followed by a series of attacks on the Hausa states, most of which had been overcome by . The infantry formations equipped with bows, which were deployed in the irst instance, were combined later with cavalry units; horsemen wearing padded quilts and chainmail and wielding lances and swords were the offensive arm of the expanding state, whereas the highly effective infantry squares were its defensive bedrock. The resultant Sokoto Caliphate was the outcome of a coalescence of dynamics, the motivations for participation in holy war clearly multidimensional: spiritual, for sure, but also ethnic, as indicated by the prominent role of the Fulani; and socio-economic, insofar as the revolt was at least partly a response to the cattle tax imposed by agricultural rulers on pastoral communities. Sokoto represented much that was novel, but it also represented the crystallisation of several lines of long-standing conlict, including the role of Islam in public life, and the tensions inherent in the cohabitation of pastoralists and farmers across the Sahel belt. Sokoto was now the exporter of both ideology and military practice across a wide region, and the Caliphate’s success had a knock-on effect, with Bornu reforming its military in order to check the advance of Hausa and Fulani forces. Guns were increasingly adopted across the savannah, after several centuries during which they were largely eschewed by horsemen, and by the time the Tukolor jihad of al-Hajj

TRANSFORMATIONS IN VIOLENCE



Umar Tal – declared in  on the upper Senegal River – was underway, the holy warriors were equipped with irearms, some of which were manufactured and maintained locally. They were also equipped with four artillery pieces, captured in early skirmishes with the French, and deployed in the capture of Bambara territory, including Segu, to the east. The army itself was prone to internal friction – between an elite cavalry corps of early devotees, the talaba, slave-soldier converts known as sofas, and conscripts, the tuburru – but nonetheless Umar Tal’s jihad encompassed a swathe of territory in present-day Guinea and southern Mali, and conquest was swift, although it was less successful than Sokoto in terms of the political consolidation of military gains. The ability of the Tukolor army – both cavalry and infantry – to manoeuvre in formation explains much of its success, yet it was increasingly confronted by French forces possessed of both greater mobility and greater irepower. A few years later, Samori Ture, the extraordinary Mandinke leader in the tradition of Sunjata, was combining horses and guns in his dogged resistance against the French advance. In the course of the s, this former-slave-soldier-turned-warrior-entrepreneur carved out a swathe of territory between the Niger and present-day Liberia. Some degree of cohesion was provided by Islam, but Samori was not a ‘holy’ leader in the Sokoto or Tukolor mould, and unity was rooted instead in his creation of a military system notable for its solidity and complexity. From the two provinces which he controlled directly he drew oficers who were then appointed to command the army corps of the provinces – eight in total – each of which comprised some , professional soldiers (sofas), supplemented by local conscription. When not actively in service, or in training, sofas were released to engage in agriculture. On campaign, three provincial corps were normally deployed, with a fourth in reserve; cavalry and infantry operated in tandem, although Samori’s armies relied rather less on cavalry elites than either Sokoto or Tukolor – the emphasis on infantry denoting, perhaps, something of a social revolution as much as a military one. Be that as it may, Samori depended on his merchants to carry out the business of exporting both slaves and agricultural produce in order to sustain the military state; in this way he acquired large numbers of breech-loading riles, in addition to older lintlocks, via trade routes running to the Sierra Leone coast. As with the Tukolor state, local gunsmiths were capable of repairing and modifying irearms. His impressive arsenal notwithstanding, however, Samori altered his tactics from



WARFARE IN AFRICAN HISTORY

the early s onwards in recognition of the fact that pitched battle with the French was invariably disastrous. Increasingly, he favoured guerrilla warfare instead, as we see later. The polities of Umar Tal and Samori were among the most dynamic and violent state-building projects anywhere in Africa in the nineteenth century. The increasing sophistication of military recruitment and organisation, the deployment of increasingly modern weaponry and the expansion of political scale exemplify the processes of the age. Yet in many respects, these new states – and here we include Sokoto, too – were confronted with the same challenges as their early-secondmillennium predecessors, in terms of the prevalence of sectional and provincial interests and the struggle to create lasting political superstructures capable of overcoming centrifugal forces. They had a new threat to deal with, too, in the form of European armed aggression. In the meantime, Mediterranean northern Africa represented the key zone of interaction – economic, political, cultural and military – between the continent and an increasingly well-armed, aggressive Europe. Sheer geographical proximity, coupled with the absence of malaria, meant that early European experiments in the exercise of inluence, if not outright control, would be played out along Africa’s northern coast. Ottoman administration was increasingly nominal along the coast in the seventeenth century, and the autonomy of maritime powers such as Algiers was manifest in low-level attacks on European shipping in the western Mediterranean, as well as raids on European coasts as far away as southern England. The ‘pirates’ (or ‘corsairs’, named after French privateers) of the Barbary Coast – as the Europeans called the Maghreb coastline – were particularly active in the irst half of the seventeenth century, gradually becoming less troublesome thereafter, not least because of an increasingly effective European naval presence. Peace treaties were also imposed on Barbary rulers by European governments, the irst indications of the approaching imperial order. In time, political and religious elites along the northern African coast would face a choice in their confrontations with Europe: to borrow, adopt, emulate European military culture in order to more effectively defend themselves against it; or to fall back on long-standing traditions of Islamic reform and revival in tackling the imperialism emanating from Christian Europe. Millions of African Muslims would face similar dilemmas as the nineteenth century progressed. The inidel threat frequently prompted military reorganisation and the militarisation of faith in the irst instance.Three distinct movements

TRANSFORMATIONS IN VIOLENCE



demonstrate this. In the s and s, Abd al-Qadir led resistance to the French conquest of the Algerian coast by organising horsed tribesmen into highly mobile military units under a banner of rejuvenated Islam; it was a military system which drew on Sui notions of deference to authority, and which lent itself, for a time, to very effective guerrilla warfare. Around the same time, Sidi Muhammad ibn Ali al-Sanusi, founder of the Sanusiyya order in the s, was similarly organising dispersed desert communities in Libya; the Sanusiyya urged the renewal of faith as well as resistance to inidel rulers on the coast, and Sanusi merchants were also prodigious traders of slaves northward towards Tripoli. Half a century later, in the s, Suism was again critical in Muhammad ‘Ahmad ibn Abdallah’s uprising in central and eastern Sudan. Here, the region’s subjection to Egyptian rule – representative of an aggressive modernisation programme associated with Europe – provoked a militant Islamic state which was as remarkable in its way as Sokoto, if rather shorter-lived. Muhammad Ahmad had proclaimed himself the Mahdi, although after his death in  the state was rather more secular, and was primarily concerned with the adversaries – European and Ethiopian – gathering on the frontiers. Mahdist soldiers may have proven themselves able ighters, famously crushing the British defence of Khartoum in –, but they would be overcome by the heavily armed return of the British in . It is worth noting that armed theocracies appeared most commonly on the militarised frontiers and peripheries of nineteenthcentury Africa, the fertile seedbeds of violence which in turn generated long-term change. Certainly, these shifting frontiers of violence would continue to play a key role in the evolution of African warfare and militarism into the twentieth century. European imperialism was a driver of other forms of military change across northern Africa, too, where ruling elites sometimes sought to emulate European models of military modernity in order to ward off the threat represented by those same models. Egypt, in particular, was a key battleground of ideas as well as military control. Under Ottoman administration, the Mamluk aristocracy had nonetheless gradually reasserted itself in the course of the seventeenth and eighteenth centuries; by the late eighteenth century, several key Mamluk families, under patriarchs known as beys, virtually ran Egypt, while the pashas, initially Turks appointed directly by Istanbul, themselves established autonomous dynasties in the eighteenth century. Three centuries before, the Mamluks had been overwhelmed by a superior military force of



WARFARE IN AFRICAN HISTORY

the Ottomans; now, a new threat arose – again better equipped and organised – from the direction of Europe. To Napoleon, the Egypt and the Nile Delta represented control of the eastern Mediterranean; the symbolism of resurgent France exercising domain over the glories of antiquity was also powerful. The French invaded in , and the British soon arrived in pursuit. Only through their naval power off the Delta coastline were the British, in alliance with the Ottoman forces, able to expel Napoleon in . This is the context for understanding Muhammad Ali, the Ottoman army oficer who was the autonomous military ruler of Egypt from  until his death in . During those four decades, Muhammad Ali created a military bureaucracy modelled on the European system, imported vast amounts of irearms, and employed European advisors to oversee his remarkable programme of military reform. It would be paid for by a greatly expanded taxation system and a dramatic increase in the production of cotton for export (itself facilitated by new irrigation schemes); Muhammad Ali also sought the development of Egyptian industry, importing European technology with a view to manufacturing both weapons and textiles. It was a veritable socioeconomic revolution – comparable to developments in Europe in the seventeenth and eighteenth centuries, at least in objective – directed towards military expansion. With an army of close to , men, he effectively took control of the Red Sea, expelling the Wahhabis from western Arabia (and thus from the holy cities of Islam, Mecca and Medina) and seizing the ports of Suakin and Massawa. He initiated the Egyptian occupation of Sudan, and in the early s he invaded the Levant. Increasingly perceived in Europe as a dangerous menace, he was forced to withdraw from Syria and Palestine by the early s and was compelled by a European alliance to dismantle his bristling military establishment, his army being reduced to a mere , men. Egyptian ambitions had been effectively thwarted, but successive rulers – Muhammad Said and Ismail – continued to make use of European advisors and money to create a ‘modern’ military state, albeit one under ever less subtle European control. Fittingly, it was another army oficer, Urabi Pasha, who in – led a revolt against a regime which was seen to have exchanged Egyptian independence for the meaningless accoutrements of ‘modernity’. The revolt was crushed and Egypt was seized by the British. From the middle decades of the nineteenth century onwards, then, a host of these leaders – Umar Tal and Samori; Mirambo and

TRANSFORMATIONS IN VIOLENCE



Nyungu-ya-Mawe; Tewodros, Yohannes and Menelik; Mutesa and Kabalega; Moshoeshoe, Mpande and Cetshwayo – perceived the advance of new technological and strategic frontiers from the direction of Europe. Sometimes this advance was clearly a threat, and in some areas rather earlier than others; but very often it represented a model which offered various kinds of inspiration. This was most evidently the case across northern Africa, where a series of polities from Morocco to Egypt sought to ‘modernise’ along the lines of European militaries, and often hired European advisors for the purpose; in Ethiopia, Tewodros, worsted in an encounter with Egyptian irearms in the s, saw the great potential of modern irepower, and spent the next twenty years trying to acquire it. But although the results were often dramatic, in many respects these exercises in what has sometimes been termed ‘defensive modernisation’ were but the latest stage in an ongoing process of military reform. A further stage still was the host of wars waged by an array of states and societies against ‘new’ African armies under European oficers at the outset of the colonial era. There were many ‘scrambles’ for Africa, and a myriad of armed responses and counterresponses; by the late nineteenth century, European imperialism was drawn into a series of turbulent vortices which were simultaneously misunderstood and altered, but only temporarily becalmed.

Conlict and Co-Option in the Age of Imperialism Intermittent European military incursions into Africa stretched back over several centuries – notably Portuguese musketeers in west-central Africa, Ethiopia and the Zambezi Valley in the course of the sixteenth century, and Dutch soldiers and later armed farmers pushing forwards the frontiers of settlement in southern Africa from the mid-seventeenth century onwards. But these were backed by no grand imperial plan, nor were they especially successful: Europeans were often undone by climate and terrain, and they possessed no marked superiority in either arms or tactics over African adversaries. Sometimes, in any case, as with Portuguese gunmen in Angola in the sixteenth and seventeenth centuries, they served in a purely mercenary capacity. Only in the nineteenth century, pushed forwards by industrial expansion, did Europe develop a decisive advantage in terms of military technology over Africa and indeed much of the non-European world. This was the essential foundation for any imperial project; the other key



WARFARE IN AFRICAN HISTORY

element was political will, which was clearly lacking for much of the nineteenth century. To be sure, European military activity on African soil increased in the course of the century: the French in Algeria during the s and s, and indeed beyond; the Boers and the British (not always on the same side) in South Africa, where conlict escalated from the s onwards; French troops along the Senegal, Casamance and Saloum rivers, from the s; the Anglo-Asante wars between the s and the s; and Britain’s bombardment of Lagos in , followed by its annexation a decade later. Worth noting, too, is the British expedition against Ethiopia in –, which was somewhat anomalous. These forays were limited in scale and objective, although they were indeed portentous as demonstrations of irepower and organisational capacity: The symbolism, for a generation of African war leaders to come, of Emperor Tewodros’ suicide in the face of remarkable British logistical genius and massive ordnance is rich indeed. Yet European military adventures in Africa were piecemeal, and indeed before the s, such activity was rarely supported with enthusiasm from governments in London or Paris. It could be expensive, and threatened deeper and even more costly commitment. Thus the British showed no great zeal to expand the limits of their Gold Coast colony at the expense of Asante, ighting the latter only when absolutely necessary; they were reluctant to become embroiled in frontier wars with the Xhosa in South Africa’s eastern Cape. The French vacillated over the size, and the purpose, of their settlement on the banks of the Senegal, although largely because of problems rather closer to home. The groundwork for France’s West African empire was laid by General Louis Faidherbe between the mid-s and the mid-s, but France was compelled to put any grand plans on hold following its defeat by Prussia in –. Only by the beginning of the s were the French ready to expand into the Sudanic interior. Across the continent, then, it was only during the s that there was a massive (and uncoordinated) escalation of violence between Africans and Europeans, and a dramatic expansion in the objectives driving European incursions: This was the ‘scramble’ for Africa, and it lasted, to all intents and purposes, until the s. It is not our purpose here to restate the range of factors behind the European partition of the continent. It is, however, important to note the emergence of one dynamic in particular – signposted earlier – namely the highly racialised perception of Africa in Europe, which hardened as the century wore on. The essentialist interpretation of

TRANSFORMATIONS IN VIOLENCE



African culture focused speciically on its apparently inherently violent nature; European writing on Africa from mid-century onwards invariably had something to say on Africans’ intrinsic savagery and brutality, and on the unending cycles of warfare – and the persistence of the slave trade – which perpetuated the continent’s state of horrible despond. Racial contempt prevented commentators from accrediting most Africans with the capacity for ‘proper war’, which was associated with the appropriate level of civilisation; rather it was Africa’s unceasing and ritualistic violence, along with its bloodthirstiness, which was the clearest manifestation of the profound laws in the continent’s cultural ‘development’, such as it was. Childlike, bickering tribes stood in the way of economic and political modernisation. Thus, although hardened racialism may not in itself have been a direct cause of the partition, it provided a legitimising framework within which the ‘scramble’ proceeded, and certainly legitimised civilising violence – unfortunate but necessary – against recalcitrant natives. To borrow an analogy from the missionaries who provided much of the humanitarian propaganda of the age, some Africans would have to die in order for others to be born anew in Europe’s image. It is one of the great ironies of Africa’s modern history that the era of some of its most remarkable creativity was so chronically misunderstood by betterarmed foreigners with designs on its destiny. Europeans were indeed better armed, but this always needs to be carefully qualiied. Breech-loaders from the s, and machine guns from the s, gave European armies decisive irepower (and of course the telegraph and the steamship made the ‘scramble’ possible), but often armies of imperial expansion were not yet actually equipped with these weapons when they sallied forth against African adversaries. Machine guns, notably, were too few in number in the s and s to decide campaigns. In many respects, rather, Europe’s advantages lay in a capacity for structure, organisation and drill. Of course, Europe’s main advantage was Africa’s own military revolution – but we return to this a little later. For now, we need to be clear that where Europe’s best and most modern weapons were indeed available and deployed, they were devastating: For all the advances and transformations brought about among African militaries since the late eighteenth century, they were rarely a match for European irepower and organisation. Breech-loaders enabled a much more rapid rate of ire than most Africans could muster; machine guns – especially the Maxim gun, deployed increasingly in the s – were devastating



WARFARE IN AFRICAN HISTORY

from a stationary position, notably when African armies obligingly amassed in front of them. African soldiers charging on foot, or on horseback, were mown down by concentrated ire. Of course, machine guns were ineffective when Africans declined frontal attack, and were largely useless in mountainous terrain and jungles. But when conditions were suitable, and African attacks reckless, machine guns were bloodily effective. Even if Africans managed to penetrate the ield of ire and actually reach the enemy, European formations were often as tight in defence as they were in offence: The British ‘square’, for example, was a disciplined, close-order infantry formation which proved resilient against most (although not all) attacks, including cavalry, and usually withstood assaults even by numerically superior forces. In southern Africa, white settler forces favoured the laager, usually impenetrable circles of wagons and other makeshift fortiications. Artillery, meanwhile, was more problematic, and European expeditionary forces largely eschewed heavy artillery pieces; terrain alone usually prevented their transport inland. But between the s and the s, smaller, lighter artillery was deployed increasingly effectively, although the impact on Africans was often more psychological than physical, as in some cases African soldiers quickly realised they could duck under shells, or shift position even slightly to avoid the bombardment. Even so, ield guns were used in assaults on the walled towns of the western savannah, for example, and removed the need for long and exhausting sieges. The tendency of African forces to ight in the open and seek decisive engagement was usually disastrous. Only Ethiopia offers us a stunning exception, as well as a glimpse of an alternative future: At the battle of Adwa, in March , Menelik’s well-equipped, ,-strong army (perhaps four in ive Ethiopian soldiers possessed modern riles, many of which were breech-loaders) destroyed an Italian invasion force a ifth of its size, including Eritrean recruits. It was thus clear what might have happened more often had modern weaponry been more effectively deployed against slightly less competent European opposition. In many ways, of course, Menelik’s triumph was as much a matter of luck as of skill: Had the Italians waited a few more days, the Ethiopian army, on the point of starvation, would have disbanded. But seeking glory for domestic consumption, the Italian force advanced. Poor intelligence and planning – the various columns quickly became separated from one another – combined with a dangerously ill-judged underestimation of Menelik’s military capacity within the Italian command led

TRANSFORMATIONS IN VIOLENCE



to catastrophic defeat. Nonetheless, it was still Ethiopian soldiers who inlicted the defeat itself: Several army corps, positioned on higher ground and guarding key deiles, were ready and waiting, and sheer manpower as well as skilful use of modern irepower – soldiers fought in loose formation and made excellent use of cover – overwhelmed the various Italian columns within a matter of hours. Elsewhere, African armies occasionally inlicted temporary (if bloody) defeats on incautious or ill-prepared European forces – as the Zulu and the Asante did. However, the general rule was that African skill in terms of concentrated ire, advancing in appropriate order and taking advantage of cover lagged behind that of Europe. Although millions of guns were in African hands, they were rarely used as effectively as they were when in other Africans’ hands, namely those under European command. African armies had professionalised, but they were now up against smaller forces which understood professional discipline and ields of ire altogether better. At the same time, by the s, restrictions on the sale of guns to Africans were in place, and African armies could no longer replace ageing stock or replenish precious supplies of ammunition. It meant that Africans faced European armies equipped with a motley array of weapons old and new. Even so, the technical innovation shown by some in making ammunition and repairing guns foreshadowed the kind of creativity characteristic of guerrillas and militias in the later twentieth century. Across the continent, African bodies were strewn along the frontiers of violent European expansion. Victory at Adwa notwithstanding, the brittle structures of centralised states often fell relatively quickly in the face of concentrated assault once European irepower – and, just as crucially, logistics – had been brought to bear. Often, a single decisive engagement was enough for such centralised militaries. The Asante army, equipped with ‘Long Danes’, was destroyed by British irepower – Snider breech-loading riles and -pounder guns – in the main battle of the  campaign. The Asante force was also undone by its decision to attack the British squares directly, but in truth it was British administrative planning which laid the foundations for such a swift triumph: Drinking water was carried up from the coast, bridges were built over the numerous rivers and carriers were hired to supply ammunition and food for the force of , men. Similarly, in , when the French invaded Dahomey, the compact force of some , men – equipped with breech-loaders and artillery pieces, and well supplied with potable water and tinned food – was shadowed by



WARFARE IN AFRICAN HISTORY

the gunboat Topaz on the river Oueme. French irepower decimated the Dahomeans who attacked en masse, while the Topaz was subjected to several futile assaults from the riverbank. Elsewhere, African armies might catch European forces unawares, but such battleield supremacy was usually short-lived. Despite the Zulu success in surprising a British force at Isandhlwana in  – they used ‘bull’s horns’ encircling tactics against an overstretched perimeter, parts of which were defended by ill-equipped and ill-disciplined local recruits – their army was decimated during futile attacks on the British infantry square six months later. An alternative to pitched battle, of course, was prolonged war of attrition; but economic systems were rarely suited to guerrilla insurgency, especially across swathes of eastern and southern Africa, where a rinderpest epidemic devastated pastoral societies and rendered them destitute. Even so, some attempted just such a strategy. Most famously, perhaps, Samori’s war against the French during the s and s involved guerrilla tactics, and his troops became skilled at ambushing French columns. Moreover, Samori’s deliberate destruction of crops – the scorched-earth tactics increasingly common in the era of imperial warfare – meant that French columns were reduced to around , men, the maximum number which could be supported by French logistics. Samori inlicted heavy casualties on the French in several encounters through the s, although the latter were repeatedly undone by their own disorganisation as well as inancial checks imposed by politicians in Paris, who were often at odds with their commanders in the ield. But the French ultimately brought irepower and mobility to bear on the situation, while the scorched-earth approach probably damaged Samori more than it did the French, for his own economic base was being gradually eroded. He was captured by a French column in . Samori’s tactical skill was legendary, but similar patterns of conlict developed elsewhere, if in less spectacular fashion. In Tanzania, the Hehe engaged the Germans in hit-and-run attacks throughout the s, although in many ways this violence was an extension of wars dating to the s. The Hehe organised themselves into lying columns equipped with spears and lintlocks and preyed on vulnerable European outstations. In time, however, the Germans reinforced – deploying both their own troops and local recruits – and superior irepower as well as organisation had subdued the Hehe by decade’s end. In the same period, the Nyoro were waging prolonged guerrilla war against the British in one

TRANSFORMATIONS IN VIOLENCE



of the nastiest campaigns of the era. Kabalega used hit-and-run and ambush tactics – well established in the region – against British-led columns which included neighbouring Ganda troops, but dwindling ammunition stocks greatly diminished the irepower which might have been brought to bear by his barusura. Ultimately, however, it was scorched-earth tactics by the British force which destroyed Kabalega’s kingdom: Tens of thousands of people died of hunger and disease, and Kabalega’s starving troops were no match for an enemy supplied from outside the immediate vicinity. Bunyoro remained underdeveloped for decades to come. Elsewhere, the northern Somali, under the spiritual and military leadership of Muhammad Abdille Hassan, fought the Italians and the British – and rival Somali clans – between  and . Bands of Sanusiyya horsemen across the Libyan desert deied Italian pretensions until the s. Yet such sustained resistance invariably proved more devastating to the insurgent community than to the invader. In some areas, famine, not European military technology, likewise crushed resistance, and new diseases – bacteria was carried by foreign invaders, inadvertently, although this was not how some Africans saw it – similarly and signiicantly undermined societies’ capacity for mobilisation. Again, such violence was usually attended by bitter economic warfare, the scorched-earth tactics – the burning of land, the destruction of stores, the killing or stealing of livestock – which struck at the very heart of African state and society. These tactics were employed by both sides, but Europeans also used scorched earth to provoke Africans into open attack, with a view to crushing African resistance through decisive engagements. Nonetheless, this was a form of total war which was not merely symptomatic of Europe’s imperial age, but which rather had long been a feature of African warfare, namely assault on the enemy’s very economic and environmental foundations. African soldiers followed European orders in this regard, but they were well acquainted with this kind of warfare, and indeed almost certainly often showed some initiative as the ‘scramble’ unfolded. Even so, some areas took a very long time for Europeans to subdue, notably the drawn-out wars of attrition waged by the French in the dense forests of what would become Cote d’Ivoire, or by the British in the Niger Delta, or by the Portuguese in Angola. These wars lasted a generation, fought mile by mile by small communities for whom no great central authority could surrender. Europeans, hampered by thick forest and forced to move in small units – often without



WARFARE IN AFRICAN HISTORY

artillery or machine guns, largely ineffective in such terrain – were continually harried by highly mobile enemy forces which were only gradually worn down by a combination of breech-loading irepower and scorched-earth policy. Other areas still were never really under European control at all, notably swathes of Saharan Morocco, Algeria and Libya. African Muslims, indeed, were often the most problematic adversaries Europeans had to face. They had often been undone by the sectarian rivalries dividing brotherhoods, while the British, the French and the Italians had all sought in different ways to ‘deal’ with, and co-opt, Muslim elites. But Islamic Africa based its resistance to inidel administration on long-established precedent, and the reformism of the eighteenth and nineteenth centuries would continue apace, both in the deserts beyond Europeans’ reach and in the richly creative frontiers between the Sahara and the savannah. An older generation of scholarship espied two distinct waves of ‘resistance’, namely the initial defence of sovereignty in response to European invasion, between the late s and the late s, followed by a series of revolts against colonial rule, more offensive in nature and led by a new generation of men, in the course of the s and s. In reality, the societies involved would scarcely have recognised such a dichotomy; instead, the Afro-European conlicts of the half-century between  and  need to be considered in terms of continuity, while many had their roots in the earlier nineteenth century. New factors had doubtless been introduced in terms of the impositions of the colonial state: tax, in the case of the Bambatha uprising among the Zulu; the aggressive expropriation of land, in that of the southern Rhodesian uprising in . But these wars were fought in much the same way, and with much the same outcome, as those a few years earlier. In the case of the Zulu, during the Bambatha uprising of , the destruction of  was repeated, with the assegai-wielding Zulu mown down in their hundreds by British machine-gun ire. In Rhodesia in –, the Shona and Ndebele achieved initial surprise against unsuspecting European outposts, but attacks by the Ndebele on laagers defended by machine guns quickly proved horrendously futile. One of the most dramatic uprisings, the Maji Maji revolt in southern German East Africa in –, spread rapidly along the Ruiji valley and caught the Germans ill-prepared: The entire territory was policed by only  local troops, under the command of a few German oficers, and only  marines were available by way

TRANSFORMATIONS IN VIOLENCE



of reinforcement. It was enough: The Germans organised three columns which destroyed crops, killed livestock and burnt villages as they marched southwards, initiating a famine which killed upwards of , Africans. Starvation, and the swift realisation that the maji (‘water’) itself – a medicine designed to turn German bullets to water – did not work, undermined the cohesion and purpose of the rebellion. It is the case, naturally enough, that many war leaders of the later s and s represented a new generation, and were sometimes more millenarian, with more spiritual manifestos which invoked ancestral assistance; in many respects they represented deiant militarism, unwilling to accede to shifts in the political and moral economy. But these novel elements frequently merely compounded the local conlicts and stresses – around resources, authority, class – which predated the age of European imperial expansion. Maji Maji, for example, was prompted in the immediate term by colonial demands for labour, but it was also a spasm of older tensions around competing power centres and shifting socio-economic relations. Ultimately, indeed, the European ‘scramble’ for Africa must be understood as an extension of the continent’s ongoing, violent transformation, a process of change which offered Europeans a series of critical opportunities. Heightening levels of conlict and militarisation in the course of the nineteenth century meant that there were plenty of issures between states and societies into which Europeans might insert themselves in order to achieve political leverage. Not only was there a lack of what is sometimes anachronistically termed ‘African unity’ in the late nineteenth century, but there was active hostility between polities, enmity of which European oficers – the ‘men on the spot’ on the expanding frontiers of empire – were able to take full advantage. Thus, some African leaders viewed Europeans as allies and partners in essentially local struggles of some antiquity: Europeans, essentially, were co-opted into African wars, using African troops. Buganda offers the classic example of such ‘sub-imperialism’: In the course of the s, Ganda soldiers, under both a handful of British oficers and their own chiefs, helped to subdue old regional rival Bunyoro and much of the southern half of what would become Uganda; their political elite lost little opportunity to remind the British of their own glorious military past in the region. The French used their Bambara troops to campaign hundreds of miles across the southern Sahara and the savannah, and locally raised levies carved out France’s equatorial



WARFARE IN AFRICAN HISTORY

empire of Gabon, Middle Congo and Ubangui-Shari in the s and s. Britain deployed African troops in their conquest of northern Nigeria. Italy was able to deploy its Eritrean soldiers at Adwa and in Libya. In a few places, the nineteenth-century privatisation of violence took new forms, as in Belgian Congo, where rubber companies raised militia units whose reputation for barbarity and scorched-earth tactics caused consternation in Europe but whose approach followed nineteenth-century practice, well established in the vast forests of the Congo basin. European oficers depended absolutely on African recruits. Simply put, locally raised forces were essential, as no European government could countenance the enormous costs and logistical dificulties which would be involved in deploying its own troops in a multitude of campaigns on African soil. Without Africans, there could be no ‘European’ partition. In this way did Africa’s own military revolution facilitate European imperialism, for African soldiers in the late nineteenth century provided European commanders with high levels of skills, warworthiness and expertise in terms of terrain and climate. Emergent mercenary cultures meant that service in other men’s armies was a well-established route to material and social security, as well as satisfying the entrepreneurial leanings and quest for adventure within many of the societies embroiled in the dramas and traumas of the ‘long’ nineteenth century. Floating populations, displaced and rendered rootless by earlier wars and the expansions and contractions of states old and new, meant a further critical reservoir of manpower for colonial armies, without which there would have been neither the political will nor the military capacity for colonial expansion. They were cheap to maintain, better protected against disease, lived effectively off the land while on campaign and – the occasional mutiny notwithstanding, mostly over pay and conditions – were markedly loyal. Europeans recruited refugees, ex-slaves, mercenaries, fugitives. At the same time, of course, the case was made by imperial strategists that the recruitment of erstwhile savage ‘natives’ was itself part of the civilising mission, for service in European uniform awarded Africans an altogether nobler purpose, as well as some practical lessons in discipline and order. Thus necessity was recast as virtue. We look at the Afro-European armies which resulted in more detail in the inal chapter. The First World War witnessed the expansion of European colonial armies and the last major colonial military campaigns for some

TRANSFORMATIONS IN VIOLENCE



time. The French conducted recruitment campaigns across their West African possessions (mainly for service on the Western Front), which doubled up as punitive expeditions against recalcitrant chiefdoms. The British brought the Darfur Sultanate to heel in . And of course the war between the British and their allies on the one hand and the Germans on the other across eastern and central Africa relied heavily on mass local recruitment of both soldiers and porters, in itself a further extension of imperial power – although those bitter campaigns of scorched earth and violent extraction were, to the Africans involved, merely the latest stage in a series of wars dating to the mid- and late nineteenth century. These years also saw the last signiicant demonstrations of armed force against European administrations for a generation, as in Dahomey and Nyasaland, for example. More broadly, recovery and consolidation was needed on the part of millions of Africans across the continent exhausted by the violence and upheaval of the past century – but respite for a generation was all that was required before the revolution was reprised, with some new elements. By the beginning of the s, the era of the ‘scramble’, in the conventionally deined sense, was over. The regiments recruited for the purpose were either maintained in barracks or engaged in policing – often more concerned with tax evasion than with armed adversaries – while colonial administration was generally demilitarised, the army becoming less overt in the running of the new territories. Certainly internal order, rather than external adventurism, was the main function of colonial forces after the First World War. The pax colonia – the colonial state’s monopoly on violence – had been achieved across a large part of the continent, involving the suppression of both localised violence and large-scale patterns of conlict through the co-option and marshalling of African manpower. As for the pax itself, it was to prove a temporary armistice, although its transience would not have been apparent to the imperial optimist; it was also uneven. Again, much of Algeria, Morocco and Libya remained beyond the reach of European authority, military or otherwise; in the borderlands of Somalia, Sudan and Kenya, and across the western savannah, low-level military activity continued. In these places, African regiments were deployed against ‘bandits’, and frontier zones between colonial territories were frequently militarised in ways unthinkable closer to commercial and political centres. The armed potential of the fringes, again, is one of the abiding dynamics in the



WARFARE IN AFRICAN HISTORY

history of African warfare, and frontiers had long produced some of Africa’s most dynamic and revolutionary militaries. The further marginalisation of those politically and militarily fertile breeding grounds would fuel new senses of armed outrage, especially among young men. In this way were new issures opened up within colonial populations, between the locally recruited representatives of the new order and the unreconciled restive groups on the borderlands between metropolitan authorities. In many respects, it had been always thus.

Relections: Africa’s Violent Transformation Africa in the nineteenth century witnessed episodic eruptions of violent creativity, the unfolding of a veritable military revolution with far-reaching implications for state, society and economy. There was much destruction and suffering, to be sure – but overemphasising these aspects of this transformative era is to neglect, and misapprehend, the remarkable achievements of the epoch by a host of states and peoples across the continent, and by an array of dynamic entrepreneurs-at-arms. Existing states sought to innovate and reform in terms of military systems and weaponry, while new militarised societies arose across the continent, often underpinned by personal or mercenary armies; an enlargement in the scale of violence was the result. Campaigning war became increasingly common, culminating in decisive pitched engagements. Marauding war of attrition still remained the norm, but even this ‘skulking way of war’ was ever more rigorous and sophisticated. Increasingly professionalised soldiery gave rise to regimental organisation and spirit; furthermore, upwards social mobility could be achieved through violence – a theme common across Africa throughout the nineteenth and indeed into the twentieth century. There was enhanced group cohesion and the creation of new collective identities, or the reinforcement of extant ones – identities which frequently found expression through warfare and which might be mobilised as the need arose. Yet it is important to highlight the limitations on the revolution, along with its fragility. A great deal of instability resulted, and many of the new states and societies which emerged were inherently unstable, owing in large part to an absence of adequate rules of political succession, which in turn bred high levels of militarisation: Violence, in other words, was intrinsic to these developing political cultures. Political

TRANSFORMATIONS IN VIOLENCE



reform, particularly in terms of effective administration based on power rather than force, frequently lagged behind military creativity. In terms of the destruction wrought by military upheaval, there was an increasing emphasis on war rather than production – a trend with its roots in the sixteenth and seventeenth centuries – and this pattern would resurface in the mid-twentieth century. In many areas, for all the enlargement of scale, wars remained essentially large raids, extractive and predatory. The net result was the creation of displaced and militarised communities on the peripheries of expanding war economies, and the perennial logistical limitations on African warfare – particularly in terms of terrain and transport – meant that peoples along the sharp frontiers of expanding militarisms, no matter how energetic these were, could not be fully incorporated into the kind of largerscale, inclusive political and economic systems which might otherwise have engendered some degree of pax. Instead, the very existence of such populations destabilised metropoles into the twentieth century and beyond, perpetuating cultures of violence and the politics of militarisation: Inhabiting armed and hostile borderlands, they represented the dark dividend of the nineteenth-century military reformation. Military change was in many areas based on slaving warfare, and again the relationship between economy and violence was a complex one. Military elites needed to balance the export of ‘unwanted’ people – seized along the internal and the external frontiers of expanding polities – with the maximisation of numbers, people who would be drawn to their military and thus their economic power. But the balance was an extremely dificult one to get right, and the violence which proliferated as a result of the slave trade was uncontrollable and inimical to projects of political stabilisation; such violence represented, in economic terms alone, an unforeseen and unmanageable cost. In eastern Africa, for example, armies built for predation became ever more professional, but they were either not quite professional enough or numerous enough to impose themselves on the widening circles of violent instability and social breakdown which their predation generated. Meanwhile, polities which armed themselves in order to dominate trade found themselves confronted with a dilemma, namely that the wars they waged were often self-defeating in commercial terms. It was something of a ‘golden goose’ scenario. Of course, the fact that war undermined trade is an indication that war was not always about trade; but where it clearly was, it is evident that African elites were often



WARFARE IN AFRICAN HISTORY

prepared to accept a certain amount of ‘collateral damage’ in pursuit of larger (or, perhaps, smaller) long-term goals. Moreover, although the military revolution was driven by the opportunities offered by the expansion of long-range commerce, it is nonetheless important to note that these new factors intersected with extant local struggles for material resources and control of people, with environmental and demographic shifts, with long-standing competitions between communities, and with ongoing processes of political reform, often violent in its character, some of which dated to the early second millennium. Thus wars with a local economic rationality – the violent annexation of a district with particularly good supplies of wood, for example, thus disturbing a long-distance commercial highway running through it – might appear illogical in the context of burgeoning global trade. Be that as it may, nineteenth-century elites invariably failed to resolve the basic conundrum, nor could they escape the reality that the generators of commercial opportunity – over which so much violence occurred – lay thousands of miles from their own ields of battle. The revolution in the military affairs, for all its creativity and vitality, could not bridge the economic gap between Africans and their foreign trading partners; all it could do, in the end, was facilitate ever more eficient methods by which to seize the opportunities provided by the latter. Such opportunities most commonly came in the form of irearms. But the import of guns, no less than the export of people, itself produced new layers of insecurity and instability over which polities, however ambitious, had little control. The new hardware of war was all imported, thus rendering the armies most reliant on them vulnerable to the vagaries of both international commerce and international politics, at least in the short term. In the long term, guns would not be in short supply. This was, again, a fragile revolution, in large part because the tools of change were often imported – chiely in the arrival of the irearms and ammunition increasingly important to the extension of political power, and because the sources of inance (namely those driving the lucrative external commercial system) ultimately lay beyond the continent. As the century drew on, in particular areas, these various reformist and revolutionary impulses grew sharper with the appearance of external threats, which manifested themselves in different ways; there is no doubt that in some areas, military change was a reaction to European imperialism. The ‘scramble’ for Africa introduced further new elements into ongoing processes of reform. Nonetheless, the African

TRANSFORMATIONS IN VIOLENCE



revolution spilled into the colonial era of the late nineteenth and early twentieth centuries, and two broad themes are worth reiterating in this respect. The irst is that the violence of the nineteenth century had produced a swell of manpower – dislocated communities, refugees, newly militarised societies – available for the imperial armies of the ‘scramble’ era, and thereafter in colonial armies between the s and the s. Without these men, the era of European control would have been impossible, and unthinkable. Just as Europeans co-opted African elites and manpower into the imperial project, Africans likewise co-opted Europeans into the violent vortex, the armed lux, which was Africa’s nineteenth-century military revolution. However, it was Europe’s misunderstanding of those processes – the hardening of racial ideas, particularly in terms of the supposed natural African proclivity towards violent and ultimately meaningless brutality – which drove forth imperial intervention in the guise of rugged humanitarianism. The late-nineteenth-century European misreading of African warfare reiied a series of misapprehensions which have proven robust. The second theme coming out of the era of partition and subsequent colonial occupation is an appreciation – what in fact was really an affirmation – of the role of force in political affairs on the part of Africans. Colonialism conirmed this in the minds of two generations of Africans – leaders and followers alike. Even though what we commonly regard as a pax colonia endured between the s and the s, and even though sections of an African metropolitan elite invested in the idea of the civilising mission, the revolution was soon reprised, and the armed and mobile frontier reasserted itself. It is a fundamental error to assume that the dramatic military reformation of the nineteenth century was quashed by the onset of colonial rule, its gains, as well as its failures, consigned to memory and rendered irrelevant. The potential of Africans to continue to innovate and develop was enormous, and the brief colonial interlude facilitated a continued process of innovation and development. In fact, colonial rule itself became part of the ongoing revolution in military affairs, although it certainly provided new elements in the ideologies and the armouries of twentieth-century military revolutionaries.

Further reading Ajayi, J. F. Ade and R. Smith. Yoruba Warfare in the Nineteenth Century (Ibadan & Cambridge: Cambridge University Press, ).



WARFARE IN AFRICAN HISTORY

Alpers, E. A. Ivory and Slaves in East Central Africa (London: Heinemann, ). Crowder, M. West African Resistance: The Military Response to Colonial Occupation (London: Hutchinson, ). Echenberg, M. Colonial Conscripts:The tirailleurs senegalais in FrenchWest Africa, – (London: James Currey, ). Guy, J. The Destruction of the Zulu Kingdom:The CivilWar in Zululand, – (London: Longman, ). Hamilton, C. (ed.) The Mfecane Aftermath: Reconstructive Debates in Southern African History (Johannesburg: Witwatersrand University Press, ). Kanya-Forstner, A. S. The Conquest of the Western Sudan (London: Cambridge University Press, ). Lonsdale, J. ‘The European scramble and Conquest in African History’, in R. Oliver and G. N.Sanderson (eds.), The Cambridge History of Africa: Vol. , from  to  (Cambridge: Cambridge University Press, ). Pakenham, T. The Scramble for Africa – (London: Weidenfeld & Nicolson, ). Reid, R. J. Political Power in Pre-Colonial Buganda: Economy, Society and Warfare in the Nineteenth Century (Oxford: James Currey, ). War in Pre-Colonial Eastern Africa: The Patterns and Meanings of State-Level Conlict in the Nineteenth Century (Oxford: James Currey, ). Rubenson, S. The Survival of Ethiopian Independence (London: Heinmann, ). Smaldone, J. P. Warfare in the Sokoto Caliphate: Historical and Sociological Perspectives (Cambridge: Cambridge University Press, ). Smith, R. S. Warfare and Diplomacy in Pre-Colonial West Africa (London: James Currey, nd ed., ). Uzoigwe, G. N. ‘The Warrior and the State in Pre-Colonial Africa’, in A. A Mazrui (ed.), The Warrior Tradition in Modern Africa (Leiden: Brill, ). Vandevort, B. Wars of Imperial Conquest in Africa, – (Bloomington: Indiana University Press, ). Wilks, I. Asante in the Nineteenth Century: The Structure and Evolution of a Political Order (London: Cambridge University Press, ). Wright, J. ‘Turbulent Times: Political Transformations in the North and East, s–s’, in C. Hamilton, B. K. Mbenga and R. Ross (eds.), The Cambridge History of South Africa: Vol. , from Early Times to  (Cambridge: Cambridge University Press, ).

CHAPTER 

Revolutions Incomplete The Old and the New in the Modern Era

Africa has experienced dramatic – and traumatic – change in terms of warfare and military organisation in the last century. Yet the twentieth century has witnessed, in many respects, a continuation of the nineteenth-century revolution in military culture and the practice of war, with a number of novel elements, and that there is little in the continent’s recent history which cannot be better understood by examining its deeper past. In many ways, of course, the period between c.  and c.  witnessed a hijacking of the nineteenth-century revolution. The colonial era, as is well attested, involved a retardation of Africa’s economic development in a number of respects; it was also to some extent a period of military underdevelopment, distorting an essential part of the continent’s socio-economic and political evolution over the long term with dire implications for Africa’s subsequent development. However, this is too easily exaggerated, and a modern historiographical obsession with the colonial impact has frequently obscured long-term patterns. In many respects, in fact, the colonial era represented the co-option of European imperial impulses into extant African processes. The supposed accoutrements of modernity – what would become the nation-state and its Berlin Congress–era boundaries – are either irrelevant or have only created new spaces within which ‘pre-colonial’ conlict can be played out, with new material and ideological equipment. Even within colonial structures, African military innovation is evident, for example. The most damaging colonial legacy, indeed, is not the artiiciality of the African nation-state; rather, it is the enduring concept, although it now goes by different aliases, 



WARFARE IN AFRICAN HISTORY

of the pax colonia, namely the crystallisation of the notion that there was and is no place for habitually ‘warring tribes’ in the non-European world, a world which was now prised open to inexorable modernity. It was then, and continues to be, received wisdom that such violence was inimical to ‘development’, but it was believed in equal measure that it could be diagnosed and that it was ultimately ‘curable’. The African military revolution has continued unabated, with certain key processes reasserting themselves. Thus coups d’etat, liberation struggle, civil war, insurgency and militia activity represented a continuation of the military revolution by other means. Guerrillas, rebels and warlords have operated in fertile frontiers and marginalised zones, both real and abstract; have represented, if sometimes in the vaguest of terms, a range of disenfranchised groups; and have pursued many of the same goals – however unpalatably – as their nineteenthcentury forebears.

The Military in the Polity: National Transitions Afro-European armies already had a long history, and their origins, at least, were not especially noble. Since the seventeenth century, factory governors along the West African coast had assembled raggedy bands of men – ex-slaves, runaways, outcasts – to protect forts and merchandise. The so-called Forty Thieves, or the Oil Rivers Irregulars, were one such ‘regiment’, guarding English commercial interests in the Niger Delta; the British also raised local levies in each of the coastal territories of Gambia, Sierra Leone and the Gold Coast from the early nineteenth century onwards. The French began systematically recruiting men into the Tirailleurs Senegalais from the s. In the later nineteenth century, the pool of potential recruits increased in line with widening circles of conlict, which displaced communities and often drove people to the coast in search of economic opportunity. In the course of the s and s, loosely run militias acquired more permanent and oficial aspects – the Royal West African Frontier Force, the West African Regiment, the King’s African Riles, the Schutztruppe. Fruitful recruiting grounds were found in places with particularly violent recent pasts, notably Sudan and Somalia, where the skills of uprooted young men were harnessed. Actual numbers were often small compared with later national armies, and indeed with many nineteenth-century African armies: The British force occupying

REVOLUTIONS INCOMPLETE



northern Nigeria, for example, numbered no more than , African soldiers in the early s, and at the same moment Britain’s entire armed strength across its sub-Saharan territories was around ,; the French army which campaigned across the western savannah in the s against Samori comprised at most , men. On the eve of the First World War, the King’s African Riles numbered just ,. An exception was the notorious Belgian Force Publique, which was one of the larger forces at some , in the early s, recruited from among the rootless, the displaced and the downright criminal. Increasingly, however, African troops were well drilled, well armed, and enjoyed excellent logistical support, and such initially modest forces were the foundations on which Europe’s African territories were built. As colonial states consolidated, European recruitment strategies focused on supposedly ‘martial’ races. Certain groups were identiied as possessing particularly desirable characteristics for military service: Such ‘tribal traits’ included loyalty and an understanding of obedience, a proud warrior culture and physical hardiness and an uncomplicated view of the world. Europeans looked beyond the coasts and lowlands for such men, as it was believed that those environments bred vice and guile and other unmanly tendencies; instead, mountains and open savannahs would be the recruiters’ hunting grounds of choice, because those places produced masculine, militaristic cultures which could be harnessed for the greater good. Moreover, the elevated status such men enjoyed was indelibly tied to the colonial authority, for in patrolling the streets of colonial capitals they too were strangers, like Europeans, and hailed from distant frontiers and peripheries, having little in common with the people over whom they watched. Thus the Acholi of northern Uganda were favoured in that territory, the Kamba and Kalenjin in Kenya. Soldiers in the West African coastal territories tended to be from the savannah to the north. For example, the Hausa in northern Nigeria and southern Niger were favoured, with Hausa becoming the language of command, even if many non-Hausa comprised the rank and ile. The Bambara, scattered across Mali, Burkina Faso, Senegal and Cote d’Ivoire, illed the ranks of the Tirailleurs Senegalais, and their dialect became the force’s lingua franca. Two further points might be made in this context. The irst is that many of the great military identities of the nineteenth century were not present in colonial armies, or at least were nowhere near as prevalent as might be expected: Thus smaller or more peripheral groups were preferred



WARFARE IN AFRICAN HISTORY

over the Asante, the Ganda and the Maasai, for example. The second is that military service often strengthened nascent or emergent ethnic identity, which drew on common experience and practice: This was true of the Bambara, for example and a Bangala identity in the Belgian Congo was likewise reinforced through military service. First and foremost, then, colonial regimes sought to secure a monopoly on the use of violence, in the classic Weberian sense, and therefore needed to convert the motley armies of conquest at their disposal into something rather more permanent. Therefore, the forces assembled, often on an ad hoc basis, and deployed in the political and military reordering of Africa in the late nineteenth and early twentieth centuries, became regularised armies according to the European model in the course of the s and s. Structured along European regimental lines, they acquired the regimental ‘traditions’ and ethos, symbols and insignia, associated with their European counterparts. Concessions to an essentialised African heritage might include leopard-skin coverings for regimental drums, but otherwise – in uniform, order and command structure – these units incorporated no reference to Africa’s own military history, thus constituting an invented military culture largely severed from the actual pasts of its members. The force of that history would reassert itself in the fullness of time; for now, however, African soldiers had been remade in the image of their imperial overlords, arguably the most visible – and certainly the most visually striking – manifestation of the civilising mission in progress. Members of African ruling elites were likewise invited into the rareied ambience of European martial tradition. Experiments of this kind dated to at least the mid-nineteenth century: The British had sent the young son of Tewodros, the Ethiopian emperor who embodied savage African tyranny and who killed himself in the face of defeat in , to Sandhurst. A rather more successful example – the boy died miserable and alone in  – is the kabaka of Buganda during the Second World War, Edward Mutesa, who joined the Cambridge Oficer Corps while at university there, and then was commissioned as a captain in the Grenadier Guards. The French, too, trained the sons of some chiefs and made them junior oficers. Like their European equivalents, colonial armies had been forged in war; quite unlike European armies, however – which had emerged irst and foremost out of the needs of national defence (and offence) – the primary function of African colonial forces was internal security and control. Following the era of ‘paciication’, African units were in

REVOLUTIONS INCOMPLETE



effect lightly armed gendarmeries, patrolling troublesome borders – not against hostile neighbours, but in search of tax evaders and lowlevel insurgents – and otherwise enforcing law and order across the new territories. During the apex of colonial rule between the s and the s, colonial regimes generally operated through the projection of power rather than the application of force, the imperial pax sustained at least in part through the image of enormous armed might just beyond the native line of vision. It was an illusion that would dissipate in the later stages of colonial rule and during the postcolonial era, when frontiers once again became armed and violence was privatised once more. Owing to the ‘martial race’ theory, African units were increasingly ethnically deined, which enhanced esprit de corps, as did the elevated sense of social status with which soldiers were deliberately imbued by colonial oficers. Regular salaries and access to housing and pensions contributed to the sense of separateness from the broader populace among serving soldiers and their families; colonial administrations also consciously pursued social and political segregation in the attempt to keep soldiers apolitical. They could not be exposed to, and could not be moved by, the ideas leading to unrest across the colonial world – and therefore would be in a position, if required, to put down the unrest itself. The segregated and professional ethos of colonial forces would be inherited by the national armies of independent states, but so too would the sense of ethnic and social superiority, whereas the endeavour to create an apolitical military class was, in the end, a resounding failure. Greatly expanded African regiments saw action during the First World War, which was a harrowing and profound experience for the Africans involved. This was especially true of those caught up – whether as armed recruits, porters or victims of the destruction and pillage – in the prolonged war of attrition between the Allies and the Germans which ranged across eastern and central Africa between  and . Meanwhile, some , troops raised in French West Africa served on the Western Front in France itself, and were subjected to the horrors of trench warfare. In –, the French famously used the Tirailleurs Senegalais to occupy the Rhineland, much to the dismay of the locals; but otherwise, African units would rarely see action beyond their own borders until the early s, excepting the Italian ascari who served in Libya and Ethiopia in the s and s. In the interwar years, most colonial armies were again



WARFARE IN AFRICAN HISTORY

reduced in size, although France, Belgium and Italy maintained conscription, and only with the emergence of the Italian threat in northeast Africa in the mid-s did the British, for example, enlarge the King’s African Riles and the Sudan Defence Force. Indeed, Africa’s war began with Mussolini’s invasion of Ethiopia in the mid-s, a vicious throwback to the overtly militaristic imperialism of the late nineteenth century, complete with extensive deployment of local (mostly Eritrean) recruits. The Second World War proper again saw the dramatic expansion of African colonial forces in the British and French zones through a deft combination of compulsion and persuasion. Approximately half a million Africans had served in the British armed forces by , while recruitment in the Francophone zone expanded signiicantly: Here, indeed, African manpower attached to the Free French cause across western and equatorial Africa played a critical role in securing the vast Allied southern l ank below the Mediterranean. African servicemen saw action within the continent itself – chiely the northeast African campaign in Eritrea and Ethiopia in – – as well as in Burma and Italy. As in the First World War, the early s witnessed a fair amount of aggressive conscription; unlike –, however, this was now matched by an attempt to explain the nature of the war to potential recruits – in a sense, colonial regimes having to justify themselves to their own subjects – and it has been argued that political consciousness resulted, both among servicemen and wider communities. After , African armies were once again reduced in size, with demobilisation representing a headache for colonial administrations. In particular, African ex-servicemen now had a wealth of new experiences and a widened worldview – especially those who had served in Europe and Southeast Asia – and therefore a heightened sense of the frailties of the colonial world as well as anticipation of reward for service. The point is contested whether or not service veterans played a markedly greater role in post-war nationalist agitation than any other social group. The evidence suggests that soldiers were often more politicised, and did sometimes become active in politics after the war, but that many instead returned to villages and farms with no particular interest in challenging the colonial order. It is also the case that where soldiers did challenge their oficers, it was not – or at least not directly – in the name of a larger nationalist cause. When men of the Tirailleurs Senegalais mutinied at Thiaroye in Senegal in late , they did so over pay and conditions, although later Senegalese

REVOLUTIONS INCOMPLETE



nationalists would claim the incident – which ended with the massacre of the mutineers by French troops – as a cornerstone of the struggle. Similarly, ex-servicemen were prominent in the Gold Coast disturbances of , but they were motivated more by lack of employment and rising prices than by territorial nationalism. Nonetheless, in both cases, events involving African soldiers can ultimately be seen as pivotal moments in the rapidly shifting politics of colonial West Africa; and although soldiers themselves were not necessarily central to those processes, they were indeed politically active at the local or micro level. That said, the ambiguity of colonial service is evident in the Kenyan context. Here, the Mau Mau revolt of the early and mid-s involved many embittered and politicised ex-servicemen, perhaps most famously ‘General China’, whose memoirs made much of his wartime experience in Burma; yet equally the King’s African Riles units deployed against them comprised African troops who remained – in the broadest sense of the term – ‘loyal’. As decolonisation approached, oficer corps became Africanised, although in a number of territories European oficers remained in place for much of the s, overseeing the transition from the colonial to the national. Otherwise, African oficers were drawn from educated elites, and not from the ethnic pools which made up most of the rank-and-ile, which meant a perpetuation of the core problem of colonial forces that national armies were commanded by oficers who had little in common with their own men. This was an issue which would continue to ferment; nonetheless, outgoing colonial regimes could feel conident that they had created professional forces which would defend the new socio-political status quo, uphold law and order, and respect inherited boundaries; the illusion had been created of a staunchly apolitical military. Senior African army oficers stood loyally alongside new heads of government at decolonisation ceremonies, the martial baton apparently safely handed over, the ghost of the pre-colonial savage warrior long since exorcised. But problems were close to hand, not least because newly independent regimes saw the armed forces as instruments of power over potential opposition and in pursuit of particular ethnic, regional or sectional interests. In the early years of independence, armies were often greatly expanded and became politicised as governing elites sought to co-opt oficer corps into ideological projects. Even aside from a deeper military heritage, African soldiers were therefore predisposed to intervention in political affairs.



WARFARE IN AFRICAN HISTORY

Nonetheless, early coups d’etat – which had become emblematic of African politics by the early s – were often welcomed by populations disillusioned with ineffective and rapacious civilian politicians. They were frequently bloodless and promised a return to political rectitude and sound economic management: The army supposedly embodied disinterested public service, unsullied by the ilthy business of politics, and surely capable of eficient government, at least as a temporary measure. Yet there was more to this than met the eye. To begin with, these ‘politicians in khaki’ represented the resurgence of a distinctly African military tradition, in which soldiers, far from being apolitical and merely serving to defend ‘the state’ from attacks both internal and external, were in fact deeply involved in politics and highly political. The military coup d’etat was the irst indication that the revolution of the previous century had not yet run its course. The only difference – and in many ways it was an important one – was that few of these new soldiers had seen much action, although they had much experience of a system which had functioned quite successfully through force or the threat of it. More immediately, soldiers had grievances and expectations of their own. With independence, they anticipated better pay and conditions, and were often moved to action when these were not forthcoming. Particular ethnic clusters in service sought to maintain their elevated status and even exert inluence beyond the army itself. In some cases, they would come to embrace complex and radically ideological solutions to social and political problems, imagining themselves beyond the barracks as the vanguard of revolutionary change in pursuit of the nation. The weakness of postcolonial political structures, and in particular the rapid failure of parliamentary systems and the frailty of civil society, created plenty of opportunity for ambitious soldiers. Between the early s and the mid-s, there were more than seventy successful military seizures of power, and many unsuccessful attempts; by the mid-s, a large number of African states were governed by soldiers in one capacity or another, and the trend was especially clear in West Africa, where – as in Dahomey (renamed Benin in ) – soldiers were overthrown by other soldiers with almost revolving-door regularity. Countries prone to coups tended to be those where the army was especially large, absorbing a higher-than-average proportion of government expenditure, and whose oficer corps was particularly factionalised along ethnic lines and over economic beneits. The image of the soldier as saviour, according to which dynamic young

REVOLUTIONS INCOMPLETE



oficers held the nation together in the face of secessionist violence or economic collapse – Mobutu’s Zaire (former Belgian Congo), Gowon’s Nigeria, or Rawlings’ Ghana – was soon displaced by that of the soldier as predator, with igures such as Idi Amin in Uganda or Jean-Bedel Bokassa in the Central African Republic quickly seen as creating rapacious and often tragically ludicrous regimes wrecking economies and seeking revenge against a host of socio-political and ethnic enemies. Some made the transition from one to the other, notably Mobutu, who was initially projected at home and abroad as the military strongman capable of holding together his vast and fractious territory, and who later became synonymous with brutal ineficiency and state-sponsored kleptomania.The reality was that Mobutu, Amin and Bokassa signiied the resurgence of the nineteenth-century (and earlier) phenomenon of the armed entrepreneur, seeking to attract and control people through the distribution of political and material largesse. The immediate context was the dysfunctionality of the postcolonial state; but in fact they represented a pattern of some antiquity. Even so, most military regimes relied on civilian personnel to various degrees, and many ‘civilianised’ themselves over time, as erstwhile coup leaders appeared increasingly commonly out of uniform and sought – ironically enough – the legitimacy which the civilian accoutrements of ofice supposedly bestowed. Occasionally, as in Ghana and Nigeria, the military might even hand power back to civilians, having steered the ship of state through tough times – only, in some cases, to take it back again when it was clear that civilians still could not be trusted to do the job, or so the rhetoric claimed. Meanwhile, the rapid ascent of often relatively lowly oficers signiied in many places a fundamental socio-political shift, combining the re-emergence of the armed adventurer with the renewed march of the angry frontier. A deadly military authoritarianism frequently resulted. The transition encapsulated in Idi Amin’s coup was especially poignant in this respect: the former NCO, the professional soldier, from the tiny (but supposedly ‘martial’) Kakwa group in northwest Uganda, rising to prominence under the patronage of Prime Minister Obote at the expense of the Cambridge-educated, commemoration-plate soldier, Edward Mutesa, head of the formerly mighty Ganda. In Ethiopia, too, the ancient Solomonic dynasty, whose own position had centuries earlier been rooted in a ruthlessly effective deployment of organised violence, was swept aside by a cluster of oficers which included the



WARFARE IN AFRICAN HISTORY

descendant of former southern slaves from a peripheral ethnic group, Colonel Mengistu Haile Mariam, another professional soldier. Postcolonial armies themselves varied greatly in size: Around half of Africa’s armies were, in fact, relatively small in the s and s, with less than , men, lightly armed; this remains the case. Other military states, however, relied on extensive arsenals and greatly swollen pools of manpower. North African states, notably Egypt with its Middle Eastern involvements, generally had large militaries with upwards of , men, as did Ethiopia. Many regimes were buoyed by the external linkages facilitated by the Cold War. The United States, Israel, China and the Soviet Union, as well as former colonial powers, all extended military patronage. A number of regimes in Francophone Africa maintained extremely close military links with France, which also routinely stationed its own troops on the soil of former colonies. Successive French governments intervened to prop up ‘client’ regimes in Rwanda, Cote d’Ivoire and the Central African Republic, using armed force in so doing. In Zaire, Mobutu was favoured by Western powers for a generation, between the mid-s and early s, as their key central African ally against Communism; his vast military complex beneited from injections of cash and hardware as a result. On the other side of the ideological divide, the Soviet Union reinvigorated the military capacity of Ethiopia’s Marxist regime from the late s, supplying tanks, aircraft (and pilots), light and heavy guns and even oficers. As Ethiopia stumbled from one military crisis to the next, both internal and external, in the s and s, the region became awash with small arms in particular; and here, as elsewhere in Africa, the long-term legacy of the Cold War was the sudden availability of weaponry, not just to national armies, but to the guerrilla movements emerging to challenge them. In this way did the AK-, notably, become ubiquitous, the automatic rile of choice for a host of state-run armies, militias and insurgents alike across the continent. Cold War rivalries, in short, represented a military hardware bonanza for warriors of many different hues. Some states, and indeed non-state organisations, created highly effective armies and adapted modern weaponry to local conditions with considerable success. When Chad went to war with Libya over the Aozou Strip in , the Chadian army, assisted by France and the United States, used four-wheel-drive pick-up trucks, mounted with anti-tank weapons, to overcome a Libyan army generously equipped by the Soviet Union. In fact, the Chadians had borrowed the tactics

REVOLUTIONS INCOMPLETE



from the Sahrawi guerrillas of the POLISARIO (Frente Popular de Liberación de Saguía el Hamra y Río de Oro) Front, ighting for the independence of Western Sahara, who had adapted jeeps mounted with machine guns against the Moroccans in the s. The tactics of both the Chadian army and POLISARIO were in effect those of the Saharan nomadic horsemen of earlier centuries – highly mobile, and using local terrain to their advantage. The modiied Toyota four-wheel-drive vehicle would become as iconic as the AK- in some areas, notably Somalia, where – itted with heavy machine guns or anti-aircraft guns – it was known as the ‘technical’ in the s. Meanwhile the South African army, between the s and s, was similarly adapted to local conditions: Well trained, extremely mobile and lightly armed, it was the highly effective military wing of the apartheid state across the wider region. Yet the problem for most postcolonial states, ample supplies of equipment notwithstanding, was the fact that their armies were often profoundly corrupt, notoriously incompetent, and poorly trained. The Zairean and Ethiopian armies were cases in point – as too was Nigeria, where the army had been one of Africa’s better forces at independence. Armies were often fractured along ethnic lines, and units were sometimes more loyal to individual commanders than to the states they were sworn to defend. Oficers themselves frequently demonstrated rather greater concern for personal self-interest – the acquisition of wealth and political advancement – than for the military arts, and in that sense they in fact represented a signiicant distortion of the nineteenth-century revolution, which at root had been characterised by the professionalisation of organised violence. Armies became chronically ineffectual as a result, notwithstanding their capacity for violent brutality, notably against civilian targets. This was one of the reasons, combined with the fact that postcolonial states generally lacked the economic capacity for them, why major interstate wars were relatively rare. The inconclusive Morocco-Algeria war of the mid-s, the Ethiopia-Somalia conlicts of – and – over the status of the Ogaden, and Tanzania’s invasion of Uganda to oust Idi Amin in  are among the very few examples, as are South Africa’s military operations against Mozambique and Angola in the s and s. Proxy war – support for armed groups across frontiers – has always been a much more common form of ‘interstate’ violence. It is true, as we will see later, that in the last twenty years or so the capacity and the will to intervene militarily in neighbours’ affairs has increased somewhat; but in some respects, the



WARFARE IN AFRICAN HISTORY

search for the ‘classic’ pattern of interstate war along eighteenth- and nineteenth-century European lines is a distraction. The fact remains that for the irst few decades of Africa’s postcolonial era, not only was the capacity to wage aggressive external war lacking, but so too was the need, for most states were threatened not from without but from within. Internally, national armies increasingly faced enemies with access to the same weaponry as they themselves had. At the same time, and perhaps even more importantly, their insurgent adversaries occupied ever more moral high ground in wars where popular participation – whether explicit or tacit – was vital. Whatever weaponry and manpower postcolonial armies had at their disposal, it was insuficient to halt the wave of militarisation coming from the peripheries – both physical and political – of their unstable estates. The soldier-politicians of the s and s were but one manifestation of Africa’s military tradition and but one offshoot of its nineteenth-century revolution; another was the re-emergence of the armed frontier, driven by a range of grievances and aspirations, its violence better organised and the tools at its disposal more sophisticated than ever before. This next aspect of Africa’s violent transformation, and the failure of the continent’s professional soldiers-in-ofice to contain the threat, demonstrated just how fragile the colonial state’s monopoly on the use of force had been, after all.

The Armed Frontier, Resurgent and Insurgent The short-lived colonial order was based on a brittle pax, the outcome of a temporary advantage in military technology – and of a series of extraordinarily effective, if again impermanent, coalitions of European and African interests across the continent. It was brittle, and it was patchy: Even beyond the First World War, conventionally regarded as the watershed between the era of imperial ‘paciication’ and one of peaceful stability, violence persisted among the northern Somali until the early s, and among the Sanusiyya in Libya until the early s. Other armed frontiers remained ‘live’ with low-level insurgency – for example, in Algeria beyond the High Plateau, and on the Sudanese borderlands, especially those abutting Ethiopia and Eritrea. Nonetheless, it was not until the early s that the imposed armistice began to unravel signiicantly. In the course of that decade,

REVOLUTIONS INCOMPLETE



there were the beginnings of a renaissance of violence as an expression of political will and as a manifestation of economic discontent. It would become more intense, and more widespread, because a great many postcolonial states contained the seeds of their own chronic instability: They encompassed a range of inequities, political and economic, which would lead to increasingly violent contests – ethnic, regional, religious – in the decades that followed. In many instances, carefully managed processes of decolonisation ensured a peaceful transition and precluded – or postponed – nationalist violence; in others, however, tardy, limited or otherwise compromised transfers of power drove upsurges in anti-colonial violence, and in a number of these situations, patterns of conlict dated to the nineteenth century and indeed earlier. What is clear, in any case, is that the particular forms taken by colonial administration either greatly exacerbated long-running wars or introduced new grievances and patterns of conlict. The presence of sizeable and politically powerful European settler communities certainly did the latter – in Algeria, in Kenya, in southern Africa (Namibia and Rhodesia-Zimbabwe, as well as South Africa itself) – as did especially obstinate metropolitan regimes, as in the case of the Portuguese colonies of Mozambique, Angola and Guinea-Bissau. The success or otherwise of anti-colonial insurgencies clearly hinged on a number of factors: internal cohesion and organisation; available military skills and access to hardware; the strength and resolve of the enemy regime; and external bases of support. The Mau Mau uprising in Kenya, underway by , was in many ways redolent of an earlier era of anti-imperial violence in terms of its relative isolation and the weaponry at the disposal of the insurgents. Small groups of ighters – armed with assorted hunting riles and pistols but mostly with machetes, and bound together through oaths – operated in the hills north of Nairobi, avoiding pitched battle and engaging in ambushes and night raids. The British responded in force, bringing in troop reinforcements, developing counter-insurgency gangs and a loyal ‘Home Guard’ and using combat aircraft. Brutal concentration camps cowed and traumatised the civilian population. By the late s, Mau Mau had been crushed, having failed to reach out beyond Kenya and to articulate a political message to potentially sympathetic governments – or even, in many respects, to other Kenyans. Yet during the same decade, the insurgency in Algeria pointed toward a new typology of nationalist violence, one which would be



WARFARE IN AFRICAN HISTORY

widely replicated. In , the Front Liberation Nationale (FLN) opened hostilities against a recalcitrant white settler regime, backed by a French government which regarded the territory as an extension of metropolitan France. Like Mau Mau, the FLN operated in small units, generally eschewing pitched battle in favour of the ambush, and carrying out raids on settler estates, Algerian loyalists and French military posts; unlike Mau Mau, the FLN had developed, by , a thoroughly professional military organisation as well as pools of sympathy and support from across the Arab world. Some , full-time ighters – equipped with riles, grenades and mines, operating in the mountains of northern Algeria – were supplemented by , auxiliaries who carried supplies, gathered intelligence and organised a network of clandestine cells in towns and villages. FLN forces were divided into battalions, companies and sections; particular zones of combat were jointly administered by political and military committees. Massive French reinforcements – troops, mechanised armour and air support and (as with the British in Kenya) a heavy-handed approach to the civilian population – led to serious setbacks for the FLN by the end of the s, but chronic divisions within France over the Algerian question ultimately led to de Gaulle’s withdrawal in . The FLN had won the political battle, and in the process had demonstrated what might be achieved militarily, given eficient internal structure and leadership, diversiied tactics and outside support, both material and moral. In many respects, the FLN became a model of modern armed resistance for nationalists elsewhere in Africa, while the ferocious and uncompromising nature of the conlict – it was the bloodiest anywhere in Africa until the escalation of war in southern Africa in the s – haunted colonial regimes and moderate African elites everywhere. In southern Africa, similar organisations waged wars of ‘national liberation’ from the early s onwards; the Lusophone territories of Angola and Mozambique best demonstrate the nature of these conlicts. The Movimento Popular de Libertacao de Angola (MPLA) – inclusive but dominated by the Mbundu – was founded in the wake of a popular uprising in  which was driven at least in part by the continuation of forced labour. By the early s, the MPLA numbered about , soldiers, equipped with Kalashnikov riles, light machine guns and mortars supplied from Eastern Europe; it was especially skilled in ambush and mine laying and was organised into large mobile units which were further subdivided into smaller units capable of operating

REVOLUTIONS INCOMPLETE



across a wide area. While the MPLA was the largest and arguably bestorganised of the Angolan movements, its struggle was waged against rival groups – the Frente Nacional de Libertacao de Angola (FNLA), with up to , ighters, and UNITA (Uniao Nacional para a Independencia Total de Angola) with some , primarily Ovimbundu – as well as the Portuguese. Such bitter factionalism hampered the nationalist cause, perhaps as much as the Portuguese response, which included expanding the military garrison to , men and, most importantly, use of air-to-ground strikes on vulnerable columns of guerrillas which were easily detectable in the sparsely populated and lightly wooded terrain of southern and eastern Angola. This was a messy, fractured war in which both sides were confronted by logistical challenges – not least the huge areas of operations to be covered – and the sheer cost, in men and funds, contributed to Portugal’s eventual withdrawal. The subsequent MPLA regime continued to face armed resistance from UNITA through the s and s. In Mozambique, the armed struggle was dominated by FRELIMO (Frente de Libertacao de Mocambique), a Marxist coalition. FRELIMO started out with perhaps a couple of hundred ighters, growing to close to , by the end of the s, operating in mobile raiding units which made devastating use of landmines and which attacked economic targets (notably the Zambezi dam project) as well as military ones. A war of attrition wore down the Portuguese garrison, whose morale, as in Angola, diminished in the face of a growing casualty list as FRELIMO ighters became adept at ambush, harassment, and the use of Chinese and Russian rocket launchers. Again the Portuguese had mastery of the air, although by , the guerrillas had acquired Soviet anti-aircraft missiles. Notably, by the early s, FRELIMO had tapped into a tradition of resistance to the Portuguese in southern Mozambique dating to the nineteenth century – a tradition of which much was made by FRELIMO’s leadership. In Southern Rhodesia (Zimbabwe), too, the Shona-dominated Zimbabwe African National Union (ZANU) – at war with the obdurate white settler regime from  – regarded itself as the successor of the great uprising against white rule in the area, that of –. Unfortunately for Zimbabwean nationalism, ZANU was intermittently at war with the preponderantly Ndebele ZAPU (Zimbabwe African People’s Union). ZANU began inauspiciously: In the late s, small groups of ill-prepared and ill-equipped insurgents were repeatedly worsted at the hands of the highly professional Rhodesian security forces. But ZANU forces grew



WARFARE IN AFRICAN HISTORY

in capacity and conidence, cooperating with FRELIMO across the border, and accumulating an arsenal of assault riles, grenade launchers and mortars from a variety of sources, including China, Algeria, Cuba and the Soviet Union. By the mid-s, there were some , active insurgents, with several thousand more in bush training camps. The Rhodesian army made excellent use of its light mechanised units, a powerful air force and (as with the British in Kenya and the French in Algeria) counter-insurgency gangs; but the Zimbabwean ighters’ rural politicisation and mobilisation campaigns and the territory’s economic downturn in the course of the s drew new recruits to the guerrillas’ ranks – as did, it should be noted, a fair amount of forcible recruitment, as was the case in Angola, too. By the late s, joint ZANU and ZAPU forces amounted to upwards of , ighters within the territory or across its borders, now equipped with tanks and artillery: White Rhodesia had lost the military as well as the political battle. Armed insurgencies were increasingly common in already independent states. In a handful of cases, these took the form of wars of secession, as in Biafra in Nigeria and the Ogaden in Ethiopia. When the Ibo-dominated southeast region of Nigeria – self-styled ‘Biafra’ – declared its independence in , a hastily recruited Biafran army was initially able to repel the attacks of the ill-prepared Nigerian army; but the latter expanded to some ,, and increasingly enjoyed superiority in the air and – through mechanised armour – on the ground. Nonetheless it was the economic blockade of Biafra, as much as the relentless bombing of civilian targets from the air, which ultimately destroyed the short-lived republic: By the time Nigerian forces captured the Biafran capital at the beginning of , at least , Ibos were dead, mostly from disease and starvation. This was a discernibly ‘modern’ war, tactically and technologically, and the same was true of the wars between Ethiopia and Somalia over the status of the Ogaden (Ethiopia’s Somali province) in –, and again in –. In the irst of these wars, newly independent Somalia armed small groups of secessionist guerrillas in the Ogaden, who launched a series of attacks on isolated Ethiopian military garrisons; only when the Ethiopian army was bolstered by the arrival of U.S. equipment – chiely guns and armoured transport – was it able to advance through the Ogaden to the Somali border, lushing out rebels and compelling the Somali government to withdraw its support for them. Somali forces did rather better in , when a mechanised invasion advanced

REVOLUTIONS INCOMPLETE



deep into the Ogaden in conjunction with the locally active Western Somali Liberation Front; this time, however, it was Soviet weaponry, air power and military advisors which reinvigorated Ethiopia’s armed forces. In early , the Ethiopian tanks and mechanised infantry advanced blitzkrieg-style, sweeping the Somali troops back to the border. Crucially, there was no international support for Somalia’s claim to the Ogaden; yet despite Ethiopia’s military success, no political solution had been achieved in the province either. More often, however, wars in newly independent Africa were guerrilla insurgencies, representative of increasingly professional organisation of the armed frontier. In the s and s, the Eritrean People’s Liberation Front (EPLF), in the north of Ethiopia, was the exemplar of that professionalism. It combined guerrilla combat with conventional war, and waged a war for Eritrean independence from a liberated base area in the north of the territory. Organised and led like a regular army, the EPLF extolled the virtues of ‘self-reliance’ (much of its weaponry, including automatic riles and artillery, was captured from the Ethiopians), advocated pragmatic Marxism and sought to mobilise rural communities in the name of ‘social revolution’. It was supported by a network of urban cells and auxiliaries, inspired in part by the experience of the FLN in Algeria. By the mid-s, in addition to small-scale raids against military and economic targets, the EPLF was capable of launching frontal assaults on Ethiopian positions, manned by increasingly demoralised troops. The same was true of a number of movements in neighbouring Ethiopia, seeking the overthrow of the Marxist dictatorship of Mengistu Haile Mariam. The Tigray People’s Liberation Front (TPLF), another Marxist organisation, began as a handful of badly equipped (and initially badly led) ighters and became one of the best-organised guerrilla outits in the region. The TPLF at length linked up with several other ethnically deined movements across Ethiopia, notably the Oromo Liberation Front (OLF), to create a broad coalition of nationalist guerrillas, the Ethiopian People’s Revolutionary Democratic Front (EPRDF). By the time they, and the EPLF, seized power in Ethiopia and Eritrea respectively in , they looked, and fought, like fully professional, mechanised armies, lacking only airpower. In Uganda, too, the National Resistance Army (NRA) imposed itself on the bloody anarchy that characterised the country in the period following Idi Amin’s overthrow in . It did so through skilled and intelligent leadership, the capture of weaponry from the



WARFARE IN AFRICAN HISTORY

Ugandan army, and the reorganisation of its ighters into mobile battalions adept at ambushes and lightning strikes and bolstered by the promise of the revolutionary, democratic regeneration of a future Uganda. As elsewhere, the political struggle – the motivation of ighters and the mobilisation of the broader population – was as important as any clash of arms. Nowhere was it more urgent than in neighbouring Rwanda, where the Rwanda Patriotic Front (RPF), an army of mostly Tutsi exiles, seized control of the Hutu-led country in , in so doing bringing to an end the terrible genocidal slaughter of close to a million Tutsi and ‘moderate’ Hutu which had ripped Rwanda apart between April and July of that year. In various ways and to different degrees, these various conlicts intersected with the Cold War dynamics which were particularly powerful across Africa between the s and the s. The Soviet Union provided support to movements espousing various forms of socialist revolution (or simply expressing anti-Western sentiment). At the same time, the United States and/or its Western European allies backed the guerrilla organisations most aligned with their interests; U.S. counterinsurgency activities in particular escalated from the early s. In the Horn and in Lusophone Africa, Cold War intervention arguably prolonged wars which might have been resolved one way or the other rather sooner. But we need to be careful not to exaggerate African victimhood at the hands of nefarious external interests. As in the late nineteenth century, African military dynamics were the key drivers, and at the height of the Cold War, ultimately, emergent African military elites co-opted external forces and sought to bend them to their own ends. Between the mid-s and the mid-s, then, a large number of African states had been seized by guerrilla organisations in what represented a new wave of militarisation across the continent, notably FRELIMO (Mozambique), MPLA (Angola), ZANU (Zimbabwe), NRA (Uganda), EPRDF (Ethiopia), EPLF (Eritrea) and RPF (Rwanda). Elsewhere, the African National Congress (ANC) seized power in South Africa – electorally, in the end, rather than militarily – and the South West African People’s Organisation (SWAPO) did the same in Namibia. There were many others for whom success, as yet, remained elusive; but more broadly, the power of the armed frontier had proved transformative, as it often had in Africa’s past. These movements had much in common. They were the products of forces both old and new: Many arose along fault lines of conlict dating back

REVOLUTIONS INCOMPLETE



to the nineteenth century or earlier, whereas others represented rather more recent contests, in particular the profound instability of the political territories of mid-twentieth-century Africa. Some claimed descent from pre-colonial struggles, or (as in the case of Zimbabwe) from late-nineteenth- and early-twentieth-century resistance, although most were doggedly modernist, with ambivalent relationships with the deeper past. But all of them belonged to a tradition in African warfare of organising on the frontier and proceeding to march on the centre. The development of broadly ‘nationalist’ manifestos was a common theme, although many insurgencies were more explicitly ethnic or regional in their composition, with their grievances and objectives articulated accordingly. More speciically, each of them adopted, with local variations, the revolutionary programmes and ideologies of the age: With a critical mass forming between the late s and the early s, these were organisations which sought to create popular platforms – even if, as in many cases, this was in reality conined to rhetoric and nomenclature, and if the movements themselves were frequently revolutionary vanguards; they espoused popular, broadly socialist programmes of reform embracing democracy and land rights as core elements in the process of ‘national liberation’. Gender equality was also a central theme, and some movements deployed women combatants, as in Eritrea, Ethiopia, Mozambique and Zimbabwe; in virtually all movements, women were key organisers and activists, although usually with extremely limited implications for their postwar status, as it turned out. Although usually led by members of the urban educated elite, guerrilla armies were based in, and most of their recruits were drawn from, the countryside – hence the emphasis on land reform, and on taking the ‘political struggle’ to the ‘rural masses’ through elaborate administrative systems. Thus the struggle against the state was political as well as military. A generation of guerrillas from across the continent underwent political and military training in China in the s and s, including Zimbabweans and Eritreans – often with clear implications for their attitudes regarding the sanctity of the revolution and perceived opposition to it, as well as inculcating a belief in the principles of armed struggle lowing from the Long March and enshrined in Mao’s writings. An emphasis on endurance and patience; the importance of the long view; the ability to slowly drain the enemy’s capacity to resist; the participation of ‘the people’ – these basic tenets were present in very different geopolitical and environmental settings across



WARFARE IN AFRICAN HISTORY

Africa from the s onwards, as guerrilla forces mounted their challenges against the neo-colonial order. Most movements, therefore, were broadly of the Left, although some advocated free-market economics, or were more generally pragmatic about particular socialist principles. Moreover, several made signiicant political journeys from Left to Centre or Centre Right in the course of advancing from bush to presidential palace, and came to embrace versions of political and economic liberalism. Reformist or revolutionary rhetoric clearly held great appeal, particularly among movements’ own natural ethnic or regional constituencies; even so, recruitment was frequently undertaken forcibly (as in Angola and Zimbabwe), whereas upsurges in voluntary enrolment were often driven by especially adverse economic circumstances or by increased levels of political brutality on the part of the state. It is worth noting, too, that states themselves often sought recruits for national armies from among the same groups of people – and many joined up for broadly the same reasons that had driven enrolment during the colonial period. Militarily, and in addition to the overseas political lobbying so important to many causes, these movements engaged in low-level guerrilla combat, hitting targets representative of the state, and carrying out hit-and-run attacks on security forces. Some combined these tactics with more conventional campaigns, ighting along discernible frontlines and engaging in pitched battle; most sought to create liberated zones, while rear bases might be created across borders ( preferably in territory belonging to friendly or weak regimes), or at least backing onto them. Southern Africa best demonstrates the dynamics of this, as Angola and Mozambique, abandoned by the Portuguese in , created new zones of support for struggles in Zimbabwe, South Africa and Namibia. The insurgencies themselves were ultimately facilitated by the massive availability of weaponry, particularly small arms: Owing in large part to the inlux of military hardware as a result of the Cold War, and to the inability of the postcolonial states to maintain a monopoly on the use of armed force, African insurgents had access to a range of weaponry not seen since the late nineteenth century. While the Kalashnikov assault rile and the rocket-propelled grenade were the most common pieces of the guerrilla’s standard equipment, many movements also acquired heavy artillery, tanks and transport, often by capturing these from the enemy; only in air power could states maintain dominance. Everywhere, guerrillas adapted to local conditions,

REVOLUTIONS INCOMPLETE



from Sahrawi four-wheel-drive vehicles deployed in what amounted to mechanised cavalry units, to the EPLF’s superlative use of the northern mountains of Eritrea in the creation of a defensible rear base, to the diligent use of local spiritual sanction on the part of ZANU ighters in Zimbabwe. Anti-insurgency campaigns on the part of the state drew fewer distinctions between combatant and civilian: The very nature of guerrilla warfare meant that ‘civilians’ were legitimate targets, often harbouring ighters (voluntarily or otherwise), while insurgencies, or the threat of them, provided states with excuses for the wholesale targeting of particular ethnic communities or regions. In Angola and Mozambique, the Portuguese routinely burnt villages and carried out reprisal executions; in Ethiopia, the government likewise criminalised entire communities suspected of providing succour to rebel groups. Often, these were not simply wars of national liberation; they were also complex struggles between various groups – organised according to ethnicity, culture, region – seeking to secure access to political and material resources within the artiicial conines of colonial territory. Rebellion and civil war invariably had ethnic dimensions, or became ethnicised, as insurgencies organised themselves around a particular cultural group, at least in the irst instance, or mobilised ethnicity in pursuit of a stable platform. There is little doubt that many such identities were of considerable antiquity, and certainly predated colonial rule. But it is also clear that putatively ethnic struggle was often a surrogate for socio-economic conlict, a basic contest for access to resources. Such economic competition was as old as the communities involved themselves, even older; it heightened with the encroachment of the global economy via the slave trade and its replacements in the nineteenth and early twentieth centuries. In the twentieth century, however, material inequity reached new levels in the crucible of competition that was the colonial state and its postcolonial successor, and violence escalated as a result – driven, too, by rapid population growth across the continent from the s onwards. These struggles were man-made in many places, most dramatically in territories where white settlement had marginalised Africans within the economy – as in Kenya or Rhodesia. But they were also heightened by natural catastrophe, as in the case of prolonged drought and resultant famine across the Sahel belt in the s and s, which drove much conlict and militarisation from Mauritania to Ethiopia. And war itself clearly caused further economic and environmental disaster, whether



WARFARE IN AFRICAN HISTORY

through deliberate starvation – for example, Biafra in the late s – or by curtailing the rural economy through restricting movement and even internment, as in Kenya in the s and s or Rhodesia in the s. The Ethiopian government’s response to the twin threats of famine and insurgency in the s was the wholesale relocation of communities away from the north as part of a brutal villagisation programme. More generally, conlict drove people away from arable land, dramatically undermining food production, and millions of landmines across the continent effectively rendered fertile farmland worthless. A catastrophic outcome of prolonged war has been an enormous refugee population across the continent, several million Africans deprived of signiicant economic capacity. Deliberate and sustained attacks on the enemy’s economic base had long been an established facet of African warfare; economic ruination was not simply an unfortunate outcome, but one of the purposes of actually waging war. With enhanced technology, this became all the more devastating. It proved particularly devastating for the peoples on the peripheries of national economies, and pastoralists were especially vulnerable. Long held in deep suspicion by ‘sedentary’ regimes which viewed them as untrustworthy members of the national community – their very way of life, nomadic or at least highly mobile, was a reminder of the weakness of the state at its furthest reaches – they found themselves increasingly squeezed between land-grabbing states, forced to move because of war or otherwise competing for survival in increasingly dificult environments. Pastoral militarism had long clashed with the sedentary kind, and indeed had driven much of the conlict in Africa’s deeper past; this continued to be the case in the later twentieth century. Pastoralists armed themselves against other cattle-keepers as much as enemies from outside; violence occurred over water and – owing to bride price inlation in some areas – livestock itself. Although their conlicts became in many ways overshadowed – and frequently sidelined – by the great liberation struggles unfolding around them, their escalating cycles of violence were often barometers of economic and environmental severity at the local level, across southern Ethiopia and Sudan, northern Kenya and Uganda and the southern edges of the West African savannah. A prevailing Western-centrism is evident in the fact that with the ending of the Cold War in the early s came a conviction that conlict elsewhere might be swiftly resolved, and a pax Americana enjoyed; and there is no doubt that, Africa not being immune to the currents

REVOLUTIONS INCOMPLETE



of international politics, across the continent change did result from events in Eastern Europe and the Soviet Union. But in the main, this change involved the overthrow of regimes which had previously received substantial support, now withdrawn, from one or other of the two Cold War blocs, and an escalation of long-suppressed conlict in the course of the s. African dynamics, again, had always been present; now, they were clearly visible, if not yet readily understood.

Vortices of Violence: Established Patterns and New Directions in Contemporary War Neat distinctions between types of conlict are always dangerous; war itself is a messy business. Our relative proximity to violence in Africa since the s makes it appear all the murkier, following the apparently straightforward era of anti-colonial insurgency around which clear lines of analysis could be (and were) drawn. The age of righteous rebellion appeared to give way to that of paramilitary pandemonium; and as the twentieth century drew to a close, this new violence seemed more ragged, less centralised, and more brutal, and the aims of the protagonists so much less coherent, than that of the guerrilla insurgencies a generation earlier. Some might argue, indeed, that this particular form of violence does not at all belong in a book dealing with ‘war’ and ‘the military’; the supposed messiness and savagery of so many ‘tribal’ wars was the cause of much foreign incomprehension, particularly in the West, where such conlicts were incompatible with notions about ‘real war’ (as indeed they had been in the nineteenth century). Barbarity, of course, was omnipresent – as it was in the nineteenth century, and as it is in all war. But the events of the late twentieth century did not represent a collapse into atavistic yearning, a resurgence of the generic African savagery so beloved by lazy Western reportage – the direct result of native ingrates’ rejection of modernity – but a renaissance of many of the core dynamics driving the nineteenth-century revolution in military affairs. As a number of states ‘collapsed’ in the late s and s, the lines between various forces and their objectives, and between combatant and non-combatant, became blurred, or at least highly mobile. Yet in fact this was an extension of a great deal of earlier anti-colonial violence, and the outcome of the laws inherent in the political order imposed by colonial regimes in the irst place. Above all, Africa in the



WARFARE IN AFRICAN HISTORY

late twentieth century resembled Africa in the mid-nineteenth rather more than many cared to admit (or were even aware). This was true not in some despairing humanitarian sense, but in the actual composition, form and modus operandi of many of the armies and militias themselves. In many cases of large-scale insurgency between the s and the s, states had been captured intact, despite the enormous trauma involved in their seizure – Ethiopia, Rwanda, South Africa, Angola, Mozambique and Zimbabwe, for example. New states had been similarly fashioned out of armed revolution, notably in Eritrea and Namibia. As a general rule, this successful transfer owed much to the fact that the movements themselves had essentially been run as highly eficient, hierarchical and bureaucratic states-in-waiting, with clearly deined objectives; post-war pragmatism also helped, not least in absorbing the skills and institutions of the old regime. But a key feature of the ‘new’ violence of the s was the subordination of ‘the state’ to military organisations with their own internal logic, which fed off chronic mismanagement and economic crisis, as in Sierra Leone, Liberia, Congo and Somalia. In other areas, warfare took on new and ever more brutal forms. Again, however, this was a process of militarisation which also had its roots, in many respects, in a pre-colonial era. The kind of ragged, total violence characteristic of the s and s had been signposted in the more recent past, too: Horriically brutal ‘liberation war’ had involved little distinction between combatant and civilian, as we have noted. But such violence became rather more common, and more complex, as the twentieth century drew to a close. The overthrow of governments and the collapse of state authority in Sierra Leone, Liberia and Congo during the s and s resulted in the unleashing of a multitude of ethnic and socio-economic conlicts, long suppressed by a sequence of authoritarian military regimes. There followed the proliferation of militia bands, sometimes composed along ethnic or religious lines, but often a melange of groups; these were irregular forces which were essentially private armies led by ‘warlords’, military entrepreneurs driven by deep personal ambition, and which continually shifted shape according to circumstance. Some started out looking much like the nationalist guerrilla movements of the s and s; but ‘warlordism’ also became big business as states ceased to function in any meaningful sense. Private armies came to occupy the spaces vacated by the state.

REVOLUTIONS INCOMPLETE



This was most clearly demonstrated in Liberia, where from  the bloody regime of Samuel Doe was confronted by armed revolt in the form of the National Patriotic Front of Liberia (NPFL) under Charles Taylor. Its troops were armed with AK-s and rocketpropelled grenades, but were ill-disciplined and led by mostly illiterate oficers. The civil war was exacerbated through the early s by the emergence of various other factions, the result of frequent splits within the NPFL itself. The ,-strong NPFL controlled significant swathes of Liberia, setting up own administration and radio station, but no side possessed much in the way of military skill. This was, rather, a profoundly brutal conlict, in which rape and plunder – by ighters dressed in bizarre costumes, many of whom were children – was more common than evidence of martial proiciency. As the result of a tortuous ‘peace’ process, Taylor became president in , only for the civil war to restart in ; but in the meantime, the violence had spread to neighbouring Sierra Leone, where the NPFL supported the Revolutionary United Front (RUF). The RUF – similarly equipped, lacking in military discipline and prone to treat civilians with stunning brutality – paid for its operations (and those of the NPFL) through the sale of diamonds, and like its Liberian counterpart routinely abducted children to serve as soldiers. Drug abuse among the latter was habitual. The Sierra Leone Army succeeded in recapturing territory, but they acted in much the same manner as the rebels, asset-stripping to fund operations which were as militarily inept, and as savage toward non-combatants, as those of the enemy. Only foreign military intervention, spearheaded by Britain, at length imposed order on this traumatised land; in Liberia, only Taylor’s eventual exile (and his arrest on war crimes charges) initiated a process of reconciliation. A similar scenario unfolded in eastern Congo, which in the mids became caught up in the genocide and subsequent political turmoil in Rwanda and Burundi. The Alliance of Democratic Forces for the Liberation of Congo-Zaire (ADFL) – predominantly Tutsi-led but comprising a range of groups opposed to the Mobutu regime – launched a rebellion which was supported by Uganda and Rwanda. As in West Africa, the ADFL was an ill-disciplined, motley force, possessing some light artillery but mostly equipped with automatic riles and rocket grenade launchers. In time-honoured fashion, its soldiers – frequently starving and raddled with disease – lived off the land as they advanced. Some ,-strong, many of whom were children,



WARFARE IN AFRICAN HISTORY

the ADFL became noted more for uncontrolled plunder and attacks on Hutu civilians than for any particular martial prowess, although it was still able to overcome Mobutu’s demoralised army in spite of the latter’s use of European mercenaries. However, as violence escalated across eastern Congo, and state authority (as much as it had ever been present) receded, armed bands of various persuasions proliferated across eastern Congo, carving out spheres of inluence in the dense rainforest. A variant emerged in northern Uganda, where the Lord’s Resistance Army (LRA), mainly Acholi, developed not out of state collapse, but rather as the result of a ‘new order’ in the form of the NRA’s seizure of power in the late s. The LRA’s predecessor – the Holy Spirit Mobile Forces, a Christian cult under the leadership of Alice Lakwena – had taken advantage of an atmosphere of fear and uncertainty following the appearance of NRA forces in the area. Lakwena preached the need to wage a ‘holy war’ against evil spirits, pagan healers, impure soldiers; claiming to have had visions that Uganda’s new rulers were witches, she subsequently raised an army and attempted to march on Kampala in . Defeated, she led to Kenya, but the cause was taken up by one of her aides, Joseph Kony, who proceeded to devastate a swathe of northern Uganda. The LRA emphasised the purity of the Ten Commandments and developed a jumbled ‘ideology’ which was partly Christian fundamentalism but very largely hysteria underpinned by distorted aspects of local spiritualism. Above all, the movement became notorious for its abuses of the local population: rape, abduction of children, terrible physical punishments. Never more than around , strong, organised into brigades of – ighters using automatic riles supplied by Sudan, the LRA nonetheless succeeded in terrorising a wide area, not least because of its mobility. It moved continually, even shifting across borders into Congo and Sudan. The Ugandan army itself responded almost as viciously, demonstrating again that the violent frontier sanctioned behaviour which would have been unacceptable beyond it. These movements became synonymous with brutality, largely as a result of drug and alcohol abuse, the mutilation of ‘enemies’ (i.e. anyone unfortunate enough to get in their way), the mobilisation of a shady spiritual universe, and of course the use of child soldiers. The subjection of women to sexual assault, meanwhile, was not simply a tragic by-product of war, but central to a strategy of terror and control. A vicious gun and drug culture thrived, and brutalised children were

REVOLUTIONS INCOMPLETE



an intrinsic part of this nightmarish world of fragmented criminalised bands with few or no discernible aims. Above all, however, these were usually struggles for basic resources, which could only be accessed through violence. Warlords sold diamonds in Angola and Sierra Leone, and iron ore and timber in Liberia – although this activity was not conined to militias, as even national armies sold assets for weapons and other goods, as in Burundi, or plundered neighbours, as Ugandan forces did in Congo. War was business, and, for opportunistic elites at least, often good business; entrepreneur warlords – like their nineteenth-century forebears – tapped into global trade, this time in natural resources as well as in arms, and conducted business from the West African bush via the Internet and satellite phones. In the post–Cold War world, they found willing suppliers, in the former Soviet bloc in particular. More broadly, control of imported goods meant that patrons were able to extend their networks of clients, distributing a range of commodities – from food to luxury items – to loyal followers. Young men joined guerrilla movements in search of material wealth, honour, and the respect of their peers; in many cases, militia culture was rooted in explicitly male behaviour – and in some cases at least relected a historical pattern of aspiration to a life in arms. Whereas some militia groups sprang up locally and autonomously, others had shadowy links to factions within governing elites which took advantage of the weak state to inance or otherwise harness armed groups in pursuit of a particular agenda, or simply for personal protection. For example, the Rwandan government harnessed a network of extremist Hutu militias – known as the Interahamwe – who were responsible for much of the genocide against Tutsis in . The Interahamwe later operated among the enormous Hutu refugee populations in eastern Congo. Once in power, Liberian president Charles Taylor spent enormous amounts of money on a range of loyal paramilitary groups for his own personal security. In recent years, the academic debate has centred on ‘greed’ versus ‘grievance’ in analysing ‘new’ forms of warlordism and guerrilla activity. Thus, whereas some espy the politics of plunder at work – groups being motivated by greed, and for whom ideological or political concerns are mere window-dressing – others see genuine political or economic grievances as the key factors, especially in neglected and marginalised regions where violence is merely a form of entrepreneurialism of which one aspect is plunder. It is not easy, in the inal



WARFARE IN AFRICAN HISTORY

analysis, to distinguish between ‘greed’ and ‘grievance’, certainly in terms of individual motivation; the predatory economic systems which result are often much more than the sum of their parts. Whatever the case, it is clear enough that intractable cultures of violence proliferate, with members of those cultures unwilling or unable to conceive of ‘alternatives’. These are the shifting frontier cultures, possessed of a brutal creativity, which have long driven warfare and indeed political and cultural change in Africa. In the meantime, widening circles of violence produced enormous populations of refugees which crossed and recrossed increasingly irrelevant colonial-era borders – with militia groups in close attendance. In Somalia, and across the Maghreb, the situation was a little different, although driven by many of the same factors. Somalia, always possessed of an unsteady unity, was beset by violent clan politics by the late s, with southern warlords in particular running business concerns as much as aspiring to political power in the conventional sense. Perhaps a quarter of a million Somali died in the clan wars of the s alone. Meanwhile, militant Islam emerged, or re-emerged, as a force for violent change, offering (as it had in the past) an alternative framework for action against a backdrop of chronic socio-political division and the abject failure of secularism. The Islamicisation of Somali politics gathered pace in the early s and received a further boost when the Ethiopian army occupied southern Somalia between  and . A new generation of young, angry Somali men was drawn to radical Islam – the broad umbrella groups Hizb al-Islam and al-Shabaab, in particular – which spearheaded the insurgency against what were perceived as foreign-backed, illegitimate Somali governments scarcely capable of functioning beyond a few Mogadishu streets. These battles were fought using the AK- and the ‘technical’, a Toyota truck mounted with a rocket-propelled grenade launcher. Importantly, foreign inluence, in the form of al-Qaeda cells, was ever more signiicant in advancing the cause of militarised faith – but the historical precedent was there in the Horn of Africa, in the nineteenth century and much earlier, when Islam and/or militant Christianity had been readily mobilised. The same was true across the western Saharan world, where ‘al-Qaeda in the Islamic Maghreb’ was developing an ever stronger presence in the s; but here, too, any novelty was supericial, for radical warriors of the faith had long been active across southern Algeria, Mali and Niger, and al-Qaeda was simply tapping into extant networks of Islamic militancy of some

REVOLUTIONS INCOMPLETE



antiquity. The difference, now, was that there was a rather larger pool of disenfranchised young men from which to recruit. Certainly, while Islamism was on the rise in the early twenty-irst century, it often disguised deeper socio-economic problems at the local level. It is clear, then, that many of these contemporary conlicts contain novel elements, or are the products of novel developments. They have arisen on the back of a population explosion which has awarded Africa an extraordinarily young, and underemployed, populace since the s and s. They are the catastrophic failure of the colonial order – and the colonial pax – writ large, a failure greatly exacerbated by postcolonial mismanagement of epic proportions. Poverty and unemployment have driven people to arms; in the contemporary era, civil war has been much more likely in countries with low incomes, poor growth rates and a reliance on the exploitation of natural resources. Violence has also tended to erupt in mineral-rich countries where there is extremely limited access to that mineral wealth. Yet much of this violence has deep roots and constitutes a resurgence of many of the practices and dynamics of the nineteenthcentury military revolution. It is notable, for example, that these movements have generally arisen in historic frontier zones, along fault lines shaped by violent competition, or peripheries characterised by neglect and underdevelopment: eastern Congo and the central African lakes; northern Uganda; the frazzled edges of the Sahara; and the West African coastal forest. Moreover, the seizure of resources and people has long been more signiicant than the winning of pitched battles. The kinds of ragged, mobile, brutal wars fought by the RUF in Sierra Leone, or the LRA in northern Uganda – lack of military skill notwithstanding – were archetypes of a particular form of precolonial African warfare: skirmishing, raiding, cyclical wars in which new identities, or variations of old ones, were created in opposition to the old order, or the hegemonic new order, and which were aimed at terrorising the enemy and destroying their economies. The warlord culture of West Africa has more than mere echoes of the nineteenth-century war bands – the Nyamwezi ruga ruga, or the Yoruba ‘war boys’ – which emerged in response to rapid commercial and political change, most obviously their adoption of modern weaponry and quest for access to imported commodities, in new communities held together by supernatural sanction. Mirambo would recognise at least certain aspects of the LRA’s modus operandi. The celebrated Barusura of late-nineteenth-century Bunyoro, representative of a



WARFARE IN AFRICAN HISTORY

remarkable revolution in military affairs, spent a great deal of time plundering and terrorising their own people – and were allowed to do so by Kabalega – in not dissimilar fashion to recent militias and national armies alike. Child soldiers, abhorrent in our own age, have long been a core element in African armies, and indeed modern West African and northern Ugandan variations are not so distant from the militarised age-sets so vital in the history of pastoral peoples across the savannah belt. Fighters take ‘traditional’ oaths, or modern variants of them, as assertions of loyalty to new leaders, use charms for personal protection, and spirit mediums to secure the blessings of ancestors. Rebel leaders often perceive themselves to be the reincarnation of precolonial war chiefs. The links with the past are everywhere present. The violent convulsions of the past twenty years have been dreadful to behold, and they indubitably relect the awful severity of the human condition – but they are rooted in a long and complex past, and need to be understood in their historical context. One key development since the early s has been a marked increase in external intervention in these conlicts. International intervention from beyond Africa has frequently been co-opted by warring parties: Humanitarian aid was siphoned off, with food in particular a crucial weapon, while well-intended but often wrongheaded peace initiatives only served to encourage the proliferation of the parties themselves, seeking a piece of the post-war pie. Externally imposed peace agreements have all too often not stopped violence, but only prevented victory, or at least undermined the potential for local, organic resolution. African interventionism, meanwhile, has taken various forms. Proxy war – long a tactic of pre-colonial rulers seeking to bolster inluence over neighbours – remains common, especially in the Horn, where Eritrea and Ethiopia continue their war by other means, notably in Somalia and the Ogaden; wars in southern Sudan and Darfur have also attracted actors from across the region seeking leverage. And in Congo in the s, several countries, including Uganda, Rwanda and Angola, became involved in carving out political inluence and economic access under the banner of regional security as local state authority disintegrated. This was central Africa’s ‘great war’ – truly an ‘African solution to an African problem’, or more precisely an attempt at one. In the meantime, continental and regional groupings have emerged with more muscle. The Economic Community of West African States Monitoring Group (ECOMOG) involved the dispatch of West African

REVOLUTIONS INCOMPLETE



peacekeepers into neighbouring Sierra Leone and Liberia, where they assisted, despite early dificulties, in initiating a slow recovery. The ofices of Intergovernmental Authority on Development (IGAD), the northeast African grouping, were central to negotiations aimed at ending the Sudanese civil war. Sometimes, however, regional alignments are undermined by their own members’ foreign adventures: IGAD is essentially paralysed by the Eritrean-Ethiopian standoff; the South African Development Community routinely called for peace in Congo in the late s and early s, while three of its members – Zimbabwe, Angola and Namibia – were actively involved in the conlict there. The African Union itself is willing, if selectively, to take on peacekeeping roles across the continent. Peace deals, meanwhile, have been achieved, even though the stability and legitimacy of many such agreements may be questioned. The OAU/AU and its regional contingents were involved in bringing about the Eritrean-Ethiopian armistice in ; in the Sudanese Comprehensive Peace Agreement of ; in a series of Darfur ceaseires since ; and in a great many peace initiatives for Somalia over the past two decades. The AU, moreover, is in theory intolerant of the once-staple coup d’etat, and to some extent this is manifested in the isolation experienced by countries in which national armies continue to wield extraordinary political inluence – most recently Niger, Guinea, Guinea-Bissau and Cote d’lvoire. In Mauritania, a coup in  managed to gain acceptance, largely owing to the alacrity with which elections legitimised it. Yet there is pragmatic acceptance of those states now run by former guerrilla movements. In government, the fortunes of and the paths chosen by former freedom ighters have varied greatly. Whereas South Africa and Mozambique might be regarded as ‘democratic’ or ‘semi-democratic’, depending on various criteria, others – Angola, Ethiopia, Uganda, Rwanda – have developed ruthlessly effective ‘semi-authoritarian’ regimes with at least some of the trappings of popular participation and accountability. In a rather different class are the authoritarian systems in place in Zimbabwe and – the exemplar of entrenched militarism – Eritrea. Nonetheless, it is generally the case that former liberation movements have dominated the political space after military victory, developing a powerful sense of both their own blood sacriice and themselves as guardians of their peoples’ destinies; they have quickly proved unwilling to share the moral high ground with those who did not (or who they deemed did not) contribute directly to the armed struggle, and have generally showed themselves



WARFARE IN AFRICAN HISTORY

to be highly sensitive to criticism from anyone (compatriots and foreigners alike) beyond the circle of camaraderie. Thus, although in many countries there has been a clear trend toward demilitarisation and demobilisation – whether the result of external incentives and encouragement, peace agreements, war (and military) weariness or strengthening ‘civil society’ and civilian political sectors – the military presence across Africa is still a powerful one, whether covert or overt. War, or the threat of it, persists in Somalia, eastern Congo, Chad and Cote d’lvoire; between Eritrea and Ethiopia; between North and newly independent South Sudan. Meanwhile, at the edges of states – particularly, again, in the western savannah, and across the grasslands of eastern Africa – pastoral groups continue to wage low-level wars for cattle and land. These conlicts are often barely visible, but they are among Africa’s oldest, being played out in the frontier zones between weak, unstable or otherwise distracted states. And they are escalating, in large part because of heightened competition over that increasingly scarce resource – water. And even though caution must be exercised against technological determinism, the availability of the Kalashnikov has unquestionably created a gun culture and a propensity for violent retaliation which at one time might have been curbed by local mechanisms of resolution. These conlicts remind us of the continued inability of states to control their own frontiers; but they are also warnings, for these militarised margins, encapsulating many of the drivers of African warfare over the longue durée, have long been breeding grounds for creative and violent programmes for change, whether ethnic, political or religious. They are ignored at enormous peril. States remain profoundly unstable, and too many Africans are not suficiently integrated into those states – indeed, they are actively excluded from them, and often abused by them. In truth, it remains much too early to discern the formation of patterns over the past decade or so, and to judge whether the general direction of travel is away from the military; the evidence either way is patchy, and the diversity across the continent deies generalisation. It is also unclear whether ‘peace agreements’ – rudimentary and unsatisfactory though they often are – will become more prevalent, or more stable. ‘Peace’ – a relatively novel and in many ways ambiguous concept – is routinely, and no doubt understandably, regarded as a desirable end point; but as the historical experience of other parts of the world has demonstrated, it only ever has meaning after violence, and often a great deal

REVOLUTIONS INCOMPLETE



of it. How Africa’s military revolution will ‘end’, and what in reality it will be seen to have achieved, remain open questions.

Relections: War and Peace, and History The pax colonia, supericial as it was, did not, in the longer term, resolve or eliminate the deeper patterns of African warfare – and indeed in many respects it actually prolonged violence, because the nineteenth-century transformation of polity and military might well have led to an organic pax of much greater permanence, stability and authenticity. Ironically, postcolonial states were in general much less stable and much less legitimate than many of their nineteenth-century antecedents (and some of these had been pretty unstable). The reassertion of some long-term dynamics, discernible in Africa’s deep past, has been evident in the last half-century or so; yet nineteenth-century processes of political construction – misunderstood by outside observers owing to the violence which attended them – were retarded by external intervention, and the unstable elements inherent in those processes were exacerbated. Meanwhile, the very nature of colonial rule reafirmed the idea that violence – whether its transformative or conservative power – worked, or had the potential to. The deeper historical context of this upsurge in conlict needs to be considered for a moment in order to appreciate both continuity and discontinuity, for the second half of the twentieth century witnessed the convergence of two long-term dynamics, namely the reassertion of the soldier in political leadership, discussed earlier, and the reignition of the violent frontier as an agency of construction as well as destruction. The latter phenomenon was caused in part – as it had been throughout Africa’s history – by the ebbing of state power and the lowing of more potent forces from the direction of the periphery. By the mid- and late twentieth century, however, Africa’s military revolution had become a military crisis, compounded by a novel set of dynamics, including economic decline, rapid population growth, and the chronic political instability of the postcolonial state. Additionally, modern weaponry of all kinds became readily available, while the alliance of technology and ideology signiicantly increased the organisational capacity of the marginalised. Violence was once again privatised. Postcolonial states across Africa were thus faced with enormous challenges, and their own porous and unstable militaries were no more capable of imposing



WARFARE IN AFRICAN HISTORY

order on the frontiers than their pre-colonial predecessors had been – and in many respects rather less so, given the volatility of their own political centres as well as their questionable legitimacy. Clearly, moreover, much of Africa’s twentieth-century history of conlict needs to be understood in the context of economic crisis. Of course there was nothing inherently novel about such economic drivers of conlict per se, as we have seen – militaries had long risen and fallen according to economic calamity as well as opportunity. In many cases of postcolonial conlict – which seem at irst glance to be the products of the failure of ‘modernity’ – violent struggle over factor endowments and access to commercial opportunities can in fact be traced to the nineteenth century and earlier. Nonetheless, economic problems now combined with particularly severe levels of political insecurity, as well as with demographic change, to militarise the margins in unprecedented ways. The inequities and imbalances of the global economic system were greatly exacerbated in the twentieth century by the violently competitive arena provided by the postcolonial nation-state. Overall, in both political and economic terms, particular groups and regions were privileged, and others marginalised, by the colonial, and thus the immediate postcolonial, order. And although, again, in some instances these struggles can clearly be dated to earlier periods, there is little doubt that colonial rule bequeathed on postcolonial Africa new parameters and new networks of violent competition. As had long been the case across the continent, these armed struggles were further exacerbated by natural catastrophe. Lest we forget, of course, war is never destiny: It is the product of particular circumstances. Throughout Africa’s history, the same creativity that has produced violence has also led to integrated, (relatively) peaceful and highly productive societies. The problem, however, is that, certainly over the past two centuries, the circumstances which lead to violent conlict have prevailed across much of the continent. The violent phenomena of the postcolonial era, then, contained their own political and economic logic, historically rooted. Africa will not be served by mere prevention and by foreshortening analyses which highlight the dysfunctionality of African modernity, which fail to historicise, and which focus on those contemporary dimensions which most resonate: state collapse, poverty and even, in some quarters, endemic savagery. Throughout the past two centuries or more, the perception of African violence as essentially illegitimate by external (and, of course, many African) observers remains a constant – with the

REVOLUTIONS INCOMPLETE



brief exception of anti-colonial insurgency in the mid-twentieth century, regarded (on the political left, at least) as justiied and righteous. Certainly, it has become a given that African violence is ‘out of step’ with inherent concepts of Western modernity and the uses of force, and indeed warfare more broadly. Rather, deeper historical dynamics must be appreciated and considered more carefully. This is a matter, therefore, of giving war a chance, however unpalatable such a notion might be to the liberal and humanitarian conscience. Intervention to stop war is invariably not the answer because it reinforces all manner of Western institutional misunderstanding and misremembering of African war, and indeed of violence more generally. Throughout the nineteenth and twentieth centuries, African violence has often been a response to global disparity in access to wealth and represents a struggle to gain access; but for the bulk of this era, the West has generally demanded African quiescence on the basis that material aspirations might be managed by forms and institutions which are based on European, rather than local or indigenous, experience. It is furthermore the case that Africa’s military revolution has not necessarily meant state building and state consolidation, but often very different, innovative forms which, in the modern age as in the nineteenth century, have proved incompatible with current orthodoxy surrounding ‘modernisation’ and ‘development’, whether political or economic. In sum, then, violence and militarism in African history – for all their regrettable implications at the human level – have often been expressions of dynamic political creativity. This is particularly true of the nineteenth century, which represents something of a violent ‘golden age’ for African political development. Yet it must also be acknowledged that war has placed limitations on societies’ capacity to effect far-reaching economic change, while, conversely, particular local challenges – economic, demographic, environmental – have prevented war itself from becoming an agency of change and development, to the point where war itself might be considered unnecessary. The persistence of these limitations, in addition to the modern global economic disempowerment of Africa, has created an environment, visible in the postcolonial era, in which a reversion to the creative militarism and violence of Africa’s nineteenth century was not only possible but necessary. War in African history, therefore, does not represent some anomalous and marginal offshoot with little relevance to a mainstream, global history of organised violence, with few of the meanings and signiicances frequently attributed to war in the



WARFARE IN AFRICAN HISTORY

northern hemisphere. Rather, the role of war in Africa’s past reveals an enormous amount about how violence fuels creativity in the most remarkable, and often the most terrible, of ways; it is instructive about the political, economic and cultural implications of militarism as an inventive, as well as a destructive, force. African war, approached over la longue durée, must be understood in a global context not as some primordial, perennial aberration, but as representative, in some of the most challenging environments on earth, of a struggle for development as yet uninished. Further reading Ali, T. M. and R. O. Matthews (eds.) Civil Wars in Africa: Roots and Resolution (Montreal and Kingston: McGill-Queen’s University Press, ). Boas, M. and K. C. Dunn (eds.) African Guerrillas: Raging against the Machine (Boulder: Lynne Rienner). Clapham, C. (ed.) African Guerrillas (Oxford: James Currey, ). Decalo, S. Coups and Army Rule in Africa: Motivations and Constraints (New Haven and London: Yale University Press, ). Echenberg, M. Colonial Conscripts: the Tirailleurs Senegalais in FrenchWest Africa – (London: James Currey, ). Kaarsholm, P. (ed.) Violence, Political Culture and Development in Africa (Oxford: James Currey, ). Killingray, D. Fighting for Britain: African Soldiers in the Second World War (Woodbridge: James Currey, ). Lee, J. M. African Armies and Civil Order (London: Chatto & Windus, ). Parsons, T. The African Rank-and-File: Social Implications of Colonial Military Service in the King’s African Riles, – (Oxford: James Currey, ). Reid, R. J. Frontiers of Violence in Northeast Africa: Genealogies of Conlict since c.  (Oxford: Oxford University Press, ). Reno, W. Warlord Politics and African States (Boulder: Lynne Rienner, ). Richards, P. Fighting for the Rain Forest:War,Youth and Resources in Sierra Leone (Oxford: James Currey, ).

Index

Abd al-Qadir,  Abdallah ibn Yasin,  Acholi, ,  Adal, , ,  Adwa, battle of, – African National Congress (ANC),  African Union, ,  age-set organisation, , , , , , , , , , , ,  Ahmad al-Mansur, , , ,  Ahmad ibn Ibrahim, ,  Ain Jalut, battle of,  Akan, ,  Albert, Lake,  Algeria, , , , , , , ,  and war of liberation, – Alliance of Democratic Forces for the Liberation of Congo-Zaire (ADFL), ,  Almohads,  Almoravids, , , , ,  Amda Tsion, , , , ,  Amhara, , , , , ,  Amin, Idi, , ,  Angola, , , , , , , , , , , , , , , ,  and liberation struggle, –, ,  anti-colonial warfare, – Arab migrations, , ,  armies

age-set,  Asante, ,  Axumite,  Chikunda,  colonial, , , –, , – Dahomean, – of early Bantu-speakers, , – Ethiopian, , , –, –, –, , , – Ganda, ,  guerrilla, –, – Malian,  Mamluks, –,  militia, – Mirambo’s,  Ngoni, ,  nineteenth-century eastern African,  Nyoro,  Oyo, –,  postcolonial, –,  pre-colonial, , , –, , , –, ,  Rhodesian,  Rozvi,  Samori’s,  of slaving states, –,  Sokoto,  Tukolor,  Yoruba, –,  Zulu,  Asante, , , , , –, , , , , –, , , ,  Axum, , , , , , , , , 



 Bambara, , ,  Bambatha uprising,  Bangala,  Bantu expansion, , –, , , , ,  Barbary Coast,  Belgian Congo, , ,  Benin, , , , , ,  kingdom of, ,  Benue River, ,  Berbers, , , , , , , , , , ,  Biafra, ,  Biton Kulibali,  Blood River, battle of,  Boers, –, , ,  Bokassa, Jean-Bedel,  Borno, , , ,  Botswana,  Britain, , , , , , , , , , , , , , , , , , , , , , , , ,  Buganda, , , , , –, , , , , , , ,  Bunyoro, , , , , , , , , , , , ,  Burkina Faso, ,  Burundi, , , , ,  Byzantium, ,  camels, , ,  Cameroon,  canoes, , , , , ,  Cape Colony, ,  cavalry, , , –, , , , , , , , , , , , , , , , , , , , , , , , , , ,  Central African Republic, ,  Cetshwayo, ,  Chad, ,  Lake, , , ,  Chikunda, ,  China, , , ,  Christian-Muslim conlict, , , –, ,  Christianity, , , , , , , , , ,  Cold War, , , , ,  colonial rule, –, , –, 

INDEX

commerce, conlict over, , , , , , , , , , , , , , , , , , , , , – Congo, modern state of, , –, , , ,  Congo River, , , ,  Cote d’Ivoire, , , , ,  coups d’etat, , , , , –,  cultures of militarism, , , –, , , , , , , , , , , ,  Dahomey, , , , –, , , , , , , , , ,  French colony of,  Darfur, , , ,  decolonisation, ,  demography, war and, , , , , , , , , , , , , , , , ,  Dingane, ,  Dingiswayo, ,  economy, war and, , , –, , , –, , , , , , , , –, , , , , , –, –, , , , , , , , , , , , , , –, –, –, , , – Egypt, , , , , , , , , , , , –, ,  Egyptians, ancient, ,  environment, –, – Eritrea, , , , , , , , , , , , , ,  Eritrean People’s Liberation Front (EPLF), , ,  Ethiopia, , , , , , , , , , , , ,  ancient and early modern, –, , –, ,  Gondarine, –,  nineteenth-century, , –, – twentieth-century, , , , , –, , , ,  Ethiopian People’s Revolutionary Democratic Front (EPRDF), , 



INDEX

ethnicity, , , , , , , , , , , , , , , , , , ,  Europe and imperialism, , , – military advantages of, , ,  nineteenth-century perceptions of African war, , , , , , , 

Hausa, –, ,  Hehe, ,  heroism, , ,  Holland, , , ,  honour, , , , , , , , , , , , , ,  horses, , , , , , , , , , , , , , , , , , , ,  Hutu, , , , 

Fante, ,  First World War, , , , ,  Force Publique,  forest, , , , , , , , , , , , , ,  fortiication, , , , , , , ,  France, , , , , , , , , , , , , , , , , , , , ,  Frente de Libertacao de Mocambique (FRELIMO), , ,  Frente Nacional de Libertacao de Angola (FNLA),  Front Liberation Nationale (FLN), ,  frontiers, role in warfare of, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,  Fulani, , –, ,  Futa Jalon, , ,  Futa Toro, , 

Ifat, ,  Ife, ,  Ilorin, ,  Imbangala, , , , ,  Interahamwe,  iron, , , , , , , , , , , , , ,  Islam, , , , –, , , , , , , , , , –, , –, ,  jihad, , , , , , , , , , , , , , ,  Italy, , , , , , , ,  ivory trade, , , , , , , , 

Garamantes,  Garenganze,  Germany, , , , ,  Gezo, ,  Ghana, kingdom of, –, , ,  republic of, ,  Gold Coast, , , , , , ,  Gondar, ,  Gowon, General Yakubu,  Great Zimbabwe, ,  guerrilla warfare, , –, –, – Guinea-Bissau, , 

Kabalega, , , ,  kafu, , , ,  Kalahari, ,  Kalonga dynasty,  Kanem, , , ,  Karamojong, ,  Kazembe,  Kebre Negast, , , , , ,  Kenya, , , , , , ,  Khoisan,  King’s African Riles, , ,  Kitara,  Kongo, , , , , , , ,  Kony, Joseph,  Lagos,  Lebna Dengel,  Liberia, , , , ,  Libya, , , , , , , , ,  Lord’s Resistance Army (LRA), ,  Luanda, , ,  Lunda, , , 

 Maasai, , , , , , ,  Maji Maji uprising, – Malawi,  Lake,  Mali kingdom of, –, , , – modern state of, ,  Mamluks, –, , , ,  Mande, , , ,  Mandinke,  Mapungubwe,  Maravi,  martial race theory, –,  Mau Mau revolt, , ,  Mauritania, , , ,  Mawanda,  Menelik, , , , ,  Mengistu Haile Mariam, ,  mercenaries, , , , , , , , , , , , , , , , ,  mfecane, , ,  military revolution, , , , –, , , –, , , , , ,  Mirambo, , –, , , , , ,  Mobutu Sese Seko, , , ,  Morocco, , , , , , –, , , , ,  Moshoeshoe, ,  Mossi, ,  Movimento Popular de Libertacao de Angola (MPLA), , ,  Mozambique, , , , , , , , , ,  and liberation struggle, , , ,  Msiri,  Mthethwa, ,  Muhammad ‘Ahmad ibn Abdallah,  Muhammad Abdille Hassan,  Muhammad Ali, ,  Muhammad Ture, ,  Mutapa, , , ,  Mutesa, ,  Mzilikazi,  Namibia, , , , ,  National Patriotic Front of Liberia (NPFL),  National Resistance Army (NRA), , , 

INDEX

Ndebele, , ,  Ndongo, , ,  Ndwandwe,  Ngoni, , –, , ,  Ngwane,  Niger, modern state of, , , , ,  Niger River, , , , , , , , , , , , , , ,  Nigeria, , , , , , , , , , , ,  Nile River, , , , , , ,  Nilotic migrations, , – Njinga, ,  Nkore, ,  Nubia, , , ,  Nyamwezi, , , , , , , ,  Nyasaland,  Nyungu-ya-Mawe, , ,  Ogaden, , ,  Oromo, , , , , , , , , , , , , , ,  Oromo Liberation Front (OLF),  Ottoman Empire, , , , , , , , , ,  Ovimbundu, , , ,  Oyo, , , , , –, , , , , , , , , , ,  pastoralism, , , , , , , , , , , , , , , , , , , , , , ,  peace, concept of, ,  peacekeeping intervention, – POLISARIO,  political systems, and the military, , –, , –, , –, , –, , , , , , , , , , , , , , , –, ,  polygamy, ,  Portugal, , , , , , , , , , , , , , , , , , , ,  postcolonial insurgency, –, – prazeiros, ,  al-Qaeda,  Rawlings, Jerry, 

INDEX

Red Sea, , , , , , , , ,  religion, –, ,  Revolutionary United Front (RUF), ,  Rhodesia, , , ,  and liberation struggle, – Roman Empire, , ,  Rozvi,  ruga ruga, ,  Rwanda, , , , , , , , , ,  Rwandan Patriotic Front (RPF),  Sahara, , , , , , , , , ,  Sahrawi, ,  Saladin,  Samori, , , , , ,  Sanusiyya, , ,  Sarsa Dengel, ,  savannah, , , , , , –, , , , , , , , , , , , , , , ,  scramble for Africa, , , , , –,  Second World War, ,  Segu, , , , , ,  Senegal, modern state of, ,  Senegal River, , , , , , ,  Shaka, , ,  Shimbra Kure, battle of,  Shoa, , , , , , , ,  Shona, , ,  Sidi Muhammad ibn Ali al-Sanusi,  Sierra Leone, , , , , , ,  Slave Coast, , , ,  slave-raiding, , , , , , , , , , ,  slave trade, , , , , , , , , , –, , , , –, , , –, , , , , , , –, ,  slavery, , , , , , , , , , , , , , ,  slaving states, , –,  Sobhuza,  Sokoto, , , ,  Somali, , , , , , , , , 

 Somalia, , , , , , , , , , ,  Songhay, , , , –,  Soninke, , , ,  Sonni Ali,  Sotho,  South Africa, , , , , , , , ,  South West African People’s Organisation (SWAPO),  Soviet Union, , , ,  Spain, , , ,  state-formation, war and, , , , , , , , , , , , ,  Sudan, , , , , , , , , ,  Mahdist revolt, ,  Sunjata, , , , , , , , ,  Swahili, , , , ,  Tabkin Kwotto, battle of,  Tanganyika, Lake, ,  Tanzania, , , , , , , ,  Taylor, Charles, ,  Tewodros, , , , ,  Tigray, , , , , ,  Tigray People’s Liberation Front (TPLF),  Timbuktu, , , ,  Tippu Tip,  Tirailleurs Senegalais, , , ,  Tondibi, battle of, – Toro, ,  Tuareg, , ,  Tukolor, ,  Tunisia,  Turkana,  Lake, ,  Tutsi, , , ,  Uganda, , , , , , , , , , , , ,  Umar Tal, al-Hajj, , ,  Uniao Nacional para a Independencia Total de Angola (UNITA),  United States, , ,  Unyanyembe,  urbanisation, role of warfare in, , , , ,  Uthman dan Fodio, 



INDEX

Victoria, Lake, , , , , 

woodland, , 

warlords, , , , , , , , , ,  water-borne warfare, , , ,  weaponry in anti-colonial warfare, , , ,  artillery, , , , , , , , ,  during the scramble for Africa, – irearms, in pre-colonial era, , , , , , , , , , , , , , , , , , , , , , –, , , , , –, , , , , , , , , , , , , , , ,  indigenous, –, –, , , , , , , , , , , , ,  in postcolonial warfare, –, , –, ,  Wolof states, ,  women, , , , , , , , , , , , , , , , , , 

Xhosa, ,  Yao, , ,  Yekuno Amlak, , , ,  Yohannes, , ,  Yoruba, , , , , –, , , ,  Zagwe, , , ,  Zaire, , , ,  Zambezi River, , , , , , , , ,  Zambia, , ,  Zanzibar,  Zimba,  Zimbabwe, , , , , , , , , , ,  Zimbabwe African National Union (ZANU), , , ,  Zimbabwe African People’s Union (ZAPU), ,  Zulu, , , , –, , ,  Zwide, 