African Histoty V

UNESCO International Scientific Committee for the Drafting of a General History of Africa
GENERAL HISTORY
OF AFRICA- A"
V Africa from the Sixteenth
to the Eighteenth Century

EDITOR B.A.OGOT
JAMES CURREY CALIFORNIA UNESCO

General History of Africa V
Africa from the
Sixteenth to the Eighteenth Century
Abridged Edition
UNESCO General History of Africa
Volume I Methodology and African Prehistory

Editor J. Ki-Zerbo
Volume II Ancient Civilizations of Africa

Editor G. Mokhtar
Volume HI Africa from the Seventh to the Eleventh Century

Editor I. Hrbek
Volume IV Africa from the Twelfth to the Sixteenth Century

Editors J. Ki-Zerbo and D. T. Niane
Volume V Africa from the Sixteenth to the Eighteenth Century

Editor B. A. Ogot
Volume VI Africa in the Nineteenth Century until the 1880s

Editor J. F. Ade Ajayi
Volume VII Africa under Foreign Domination 1880-1935

Editor A. Adu Boahen
Unabridged Paperback Edition
Volume VIII Africa since 1935

Editor A. A. Mazrui
Assistant Editor C. Wondji

The abridged edition of
THE UNESCO GENERAL HISTORY OF AFRICA
is published by the following publishers
In Ghana, Sierra Leone,

the Gambia and Cameroon by
F. Reimmer Book Services
P.O. Box CT3499
Cantonments - Accra, Ghana

In Kenya by
EAEP
P.O. Box 45314
Nairobi, Kenya
In Nigeria by
Heinemann Nigeria
P.O. Box 6205
Ibadan, Nigeria
In South Africa, Namibia, Botswana,
Lesotho and Swaziland by
David Philip Publishers
P.O. Box 23408
Claremont 7735, South Africa

In Tanzania by
Tanzania Publishing House
P.O. Box 2138
Dar es Salaam, Tanzania

In Uganda by
Fountain Publishers
P.O. Box 488
Kampala, Uganda
In Zambia by
UNZA Press
P.O. Box 32379
Lusaka, Zambia
In Zimbabwe,
Botswana, Swaziland and Malawi by
Baobab Books
P.OBox 1559
Harare, Zimbabwe
In the United States of America and Canada by

The University of California Press
2120 Berkeley Way
Berkeley, California 94720
And in the United Kingdom, Europe
and the rest ofthe world by
James Currey Publishers ' and
73 Botley Road UNESCO Publishing
Oxford OX2 OBS 7 Place de Fontenoy, 75700 Pans
International Scientific Committee for the Drafting of a General History of Africa (UNESCO)
General History of Africa V
Africa from the
Sixteenth to the Eighteenth Century
EDITOR B. A. OGOT
Abridged Edition
JAMES CURREY CALIFORNIA UNESCO

First published 1999 by the
United Nations Educational, Scientific
and Cultural Organization
7 Place de Fontenoy, 75700 Paris
James Currey Ltd

73 Botley Road
Oxford OX2 OBS
and
University of California Press

2120 Berkeley
California 94720, United States of America
© UNESCO 1992 and 1999
ISBN (UNESCO): 92-3-102497-3

ISBN (UC Press): 0-520-06700-2
British Library Cataloguing in Publication Data

UNESCO general history of Africa. - Abridged ed.
Vol. 5: Africa from the sixteenth to the eighteenth century.
1. Africa - History - To 1884
I. Ogot, Bethwell A. II. UNESCO. III. General history of Africa
960.2'2
ISBN 0-85255-095-2
Library of Congress Cataloging-in-Publication Data

is available on request from the Library of Congress
Typeset in Monotype Bembo 11/11 pt by Long House, Cumbria, UK,

and printed in Britain by Villiers Publications, London N3
Contents
Preface by Amadou-Mahtar M'Bow, former Director-General of vii

UNESCO (1974-87)
Description ofthe Project by B. A. Ogot, President ofthe International xii
Scientific Committee for the Drafting of a General History of Africa (1978
Note on chronology xiv
Members ofthe International Scientific Committee for the Drafting of xv
a General History of Africa
Biographies ofthe authors who contributed to the main edition xvii
1 The struggle for international trade and its implications for Africa 1
2 African political, economic and social structures during this period 13
3 Population movements and the emergence of new socio-politicalforms 25

in Africa
4 Africa in world history: the export slave trade from Africa and the 39
emergence of an Atlantic economic order
5 The African diaspora in the Old and New Worlds 56
6 The Ottoman conquest of Egypt 70
7 The Sudan, 1500-1800 89
8 Morocco 104
9 Algeria, Tunisia and Libya: the Ottomans and their heirs 120
10 Senegambia from the sixteenth to the eighteenth century: 134

evolution ofthe Wolof, Sereer and Tukuloor
vi Contents
11 The end ofthe Songhay empire 153
12 From the Niger to the Volta 166
13 The states and cultures ofthe Upper Guinea coast 187
14 The states and cultures ofthe Lower Guinea coast 204
15 Fon and Yoruba: the Niger delta and Cameroon 221
16 The Hausa states 232
11 Känem-Borno: its relations with the Mediterranean Sea, Bagirmi 248

and other states in the Chad basin
18 From the Cameroon grasslands to the Upper Nile 259
19 Tlie Kongo kingdom and its neighbours 273
20 The political system of the Luba and Lunda: 290

its emergence and expansion
21 The northern ZambeziaLake Malawi region 300
22 Southern Zambezia 315
23 Dependence and interdependence: southern Africa, 1500-1800 335
24 The Horn of Africa 345
25 East Africa: the coast 368
26 The Great Lakes region, 1500-1800 381
21 The interior of East Africa: the peoples of Kenya and Tanzania, 407
1500-1800
28 Madagascar and the islands ofthe Indian Ocean 418
29 The history of African societies, 1500-1800: conclusion 440
Bibliography 447
Glossary 461
Index 480
Preface
AMADOU-MAHTAR M'BOW
Former Director-General of UNESCO (1974-87)
For a long time, all kinds of myths and prejudices concealed the true history of
Africa from the world at large. African societies were looked upon as societies
that could have no history. In spite of important work done by such pioneers as
Leo Frobenius, Maurice Delafosse and Arturo Labriola, as early as the first
decades of this century, a great many non-African experts could not rid them¬
selves of certain preconceptions and argued that the lack of written sources and
documents made it impossible to engage in any scientific study of such
societies.
Although the Iliad and Odyssey were rightly regarded as essential sources for
the history of ancient Greece, African oral tradition, the collective memory of
peoples which holds the thread of many events marking their Uves, was rejected
as worthless. In writing the history of a large part of Africa, the only sources
used were from outside the continent, and the final product gave a picture not
so much of the paths actually taken by the African peoples as of those that the
authors thought they must have taken. Since the European Middle Ages were
often used as a yardstick, modes of production, social relations and political
institutions were visualized only by reference to the European past.
In fact, there was a refusal to see Africans as the creators of original cultures
which flowered and survived over the centuries in patterns of their own making
and which historians are unable to grasp unless they forgo their prejudices and
rethink their approach.
Furthermore, the continent of Africa was hardly ever looked upon as a
historical entity. On the contrary, emphasis was laid on everything likely to
lend credence to the idea that a split had existed, from time immemorial,
between a 'white Africa' and a 'black Africa', each unaware ofthe other's exis¬
tence. The Sahara was often presented as an impenetrable space preventing any
intermingling of ethnic groups and peoples or any exchange of goods, beliefs,
customs and ideas between the societies that had grown up on either side ofthe
desert. Hermetic frontiers were drawn between the civilizations of Ancient
Egypt and Nubia and those ofthe peoples south ofthe Sahara.
It is true that the history of Africa north ofthe Sahara has been more closely
linked with that of the Mediterranean basin than has the history of sub-Saharan
viii Preface
Africa, but it is now widely recognized that the various civilizations of the
African continent, for all their differing languages and cultures, represent, to a
greater or lesser degree, the* historical offshoots of a set of peoples and societies
united by bonds centuries old.
Another phenomenon that did great disservice to the objective study of the
African past was the appearance, with the slave trade and colonization, of racial
stereotypes that bred contempt and lack of understanding and became so deep-
rooted that they distorted even the basic concepts of historiography. From the
time when the notions of 'white' and 'black' were used as generic labels by the
colonialists, who were regarded as superior, the colonized Africans had to
struggle against both economic and psychological enslavement. Africans were
identifiable by the colour of their skin, they had become a kind of merchan¬
dise, they were earmarked for hard labour and eventually in the minds of those
dominating them, they came to symbolize an imaginary and allegedly inferior
'Negro' race. This pattern of spurious identification relegated the history ofthe
African peoples in many minds to the rank of ethno-history, in which
appreciation ofthe historical and cultural facts was bound to be warped.
The situation has changed significantly since the end of the Second World
War and in particular since the African countries became independent and
began to take an active part in the life of the international community and in
the mutual exchanges that are its raison d'être. An increasing number of
historians has endeavoured to tackle the study of Africa with a more rigorous,
objective and open-minded outlook by using with all due precautions
actual African sources. In exercising their right to take the historical initiative.
Africans themselves have felt a deep-seated need to re-establish the historical
authenticity of their societies on solid foundations.
In this context, the importance ofthe eight-volume General History of Africa,
which Unesco is publishing, speaks for itself.
The experts from many countries working on this project began by laying
down the theoretical and methodological basis for the History. They have been
at pains to call in question the over-simplifications arising from a linear and
restrictive conception of world history and to re-establish the true facts
wherever necessary and possible. They have endeavoured to highlight the
historical data that give a clearer picture of the evolution of the different
peoples of Africa in their specific socio-cultural setting.
To tackle this huge task, made all the more complex and difficult by the vast
range of sources and the fact that documents were widely scattered, Unesco has
had to proceed by stages. The first stage, from 1965 to 1969, was devoted to
gathering documentation and planning the work. Operational assignments were
conducted in the field and included campaigns to collect oral traditions, the
creation of regional documentation centres for oral traditions, the collection of
unpublished manuscripts in Arabic and Ajami (African languages written in
Arabic script), the compilation of archival inventories and the preparation ofthe
Guide to the Sources ofthe History of Africa, culled from the archives and libraries
ofthe countries of Europe and later published in eleven volumes. In addition,
Preface ix
meetings were organized to enable experts from Africa and other continents to
discuss questions of methodology and lay down the broad lines for the project
after careful examination ofthe available sources.
The second stage, which lasted from 1969 to 1971, was devoted to shaping
the History and linking its different parts. The purpose of the international
meetings of experts held in Paris in 1969 and Addis Ababa in 1970 was to study
and define the problems involved in drafting and publishing the History:
presentation in eight volumes, the principal edition in English, French and
Arabic, translation into African languages such as Kiswahili, Hausa, Fulfulde,
Yoruba or Lingala, prospective versions in German, Russian, Portuguese,
Spanish and Chinese, as well as abridged editions designed for a wide African
and international public.1
The third stage has involved actual drafting and publication. This began with
the appointment of the 39-member International Scientific Committee, two-
thirds African and one-third non-African, which assumes intellectual responsi¬
bility for the History.
The method used is interdisciplinary and is based on a multifaceted approach
and a wide variety of sources. The first among these is archaeology, which
holds many of the keys to the history of African cultures and civilizations.
Thanks to archaeology, it is now acknowledged that Africa was very probably
the cradle of mankind and the scene in the Neolithic period - of one of the
first technological revolutions in history. Archaeology has also shown that
Egypt was the setting for one of the most brilliant ancient civilizations of the
world. But another very important source is oral tradition, which, after being
long despised, has now emerged as an invaluable instrument for discovering the
history of Africa, making it possible to follow the movements of its different
peoples in both space and time, to understand the African vision of the world
from the inside and to grasp the original features of the values on which the
cultures and institutions ofthe continent are based.
We are indebted to the International Scientific Committee in charge of this

General History of Africa, and to its Rapporteur and the editors and authors of
the various volumes and chapters, for having shed a new light on the African
past in its authentic and all-encompassing form and for having avoided any
dogmatism in the study of essential issues. Among these issues we might cite:
the slave trade, that 'endlessly bleeding wound', which was responsible for one
of the cruellest mass deportations in the history of mankind, which sapped the
African continent of its life-blood while contributing significantly to the
economic and commercial expansion of Europe; colonization, with all the
effects it had on population, economics, psychology and culture; relations
between Africa south of the Sahara and the Arab world; and, finally, the process
1 At the time of going to pre« Volume I has been published in English, French, Arabic, Spanish, Portuguese,
Chinese, Italian and Korean; Volume II in English, French, Arabic, Spanish, Portuguese, Chinese, Italian and
Korean, Volume III in English, French, Arabic and Spanish, Volume IV in English, French, Arabic, Spanish
and Chinese, Volume V in English and Arabic; Volume VI in English, French and Arabic; Volume VII in
English, French, Arabic, Spanish, Portuguese and Chinese; Volume VIII in English and French
x Preface
of decolonization and nation-building which mobilized the intelligence and

passion of people still alive and sometimes still active today. All these issues have
been broached with a concern for honesty and rigour which is not the least of
the History's merits. By taking stock of our knowledge of Africa, putting
forward a variety of viewpoints on African cultures and offering a new reading
of history, the History has the signal advantage of showing up the light and
shade and of openly portraying the differences of opinion that may exist
between scholars.
By demonstrating the inadequacy of the methodological approaches which

have long been used in research on Africa, this History calls for a new and
careful study of the twofold problem areas of historiography and cultural
identity, which are united by links of reciprocity. Like any historical work of
value, the History paves the way for a great deal of further research on a variety
of topics.
It is for this reason that the International Scientific Committee, in close
collaboration with Unesco, decided to embark on additional studies in an
attempt to go deeper into a number of issues that will permit a clearer under¬
standing of certain aspects of the African past. The findings being published in
the series Unesco Studies and Documents General History of Africa2 will prove a
useful supplement to the History, as will the works planned on aspects of
national or subregional history.
The General History sheds light both on the historical unity of Africa and on
its relations with the other continents, particularly the Americas and the
Caribbean. For a long time, the creative manifestations of the descendants of
Africans in the Americas were lumped together by some historians as a hetero¬
geneous collection of 'Africanisms'. Needless to say, this is not the attitude of
the authors of the History, in which the resistance of the slaves shipped to
America, the constant and massive participation of the descendants of Africans
in the struggles for the initial independence of America and in national
liberation movements, are rightly perceived for what they were: vigorous
assertions of identity, which helped forge the universal concept of mankind.
Although the phenomenon may vary in different places, it is now quite clear
that ways of feeling, thinking, dreaming and acting in certain nations of the
western hemisphere have been marked by their African heritage. The cultural
inheritance of Africa is visible every where, from the southern United States to
northern Brazil, across the Caribbean and on the Pacific seaboard. In certain
places it even underpins the cultural identity of some of the most important
elements ofthe population.
2. The following twelve volumes have already been published in this series: Tlie peopling of ancient Egypt and
the deciphenng of Meroitic script, Tlie African slave trade from the fifteenth to the nineteenth century; Historical relations
across the Indian Ocean; Tlie historiography of Southern Africa, Tlie decolonization of Africa, Southern Africa and the
Horn of Africa; African ethnonyms and toponyms; Histoncal and socio-cultural relations between black Africa and the
Arab world from 1935 to the present; Tlie methodology of contemporary African history; Africa and the Second World
War; Tlie educational process and historiography in Africa, Libya Antiqua; Tlie role of African Student Movements in the
political and social evolution of Africa from 1900
Preface xi
The History also clearly brings out Africa's relations with southern Asia across
the Indian Ocean and the African contributions to other civilizations through
mutual exchanges.
I am convinced that the efforts of the peoples of Africa to conquer or
strengthen their independence, secure their development and assert their
cultural characteristics must be rooted in historical awareness renewed, keenly
felt and taken up by each succeeding generation. My own background, the
experience I gained as a teacher and as chairman, from the early days of
independence, of the first commission set up to reform history and geography
curricula in some of the countries of West and Central Africa, taught me how
necessary it was for the education of young people and for the information of
the public at large to have a history book produced by scholars with inside
knowledge of the problems and hopes of Africa and with the ability to
apprehend the continent in its entirety.
For all these reasons, Unesco's goal will be to ensure that this General History
of Africa is widely disseminated in a large number of languages and is used as a
basis for producing children's books, school textbooks and radio and television
programmes. Young people, whether schoolchildren or students, and adults in
Africa and elsewhere will thus be able to form a truer picture of the African
continent's past and the factors that explain it, as well as a fairer understanding
of its cultural heritage and its contribution to the general progress of mankind.
The History should thus contribute to improved international co-operation and
stronger solidarity among peoples in their aspirations to justice, progress and
peace. This is, at least, my most cherished hope.
It remains for me to express my deep gratitude to the members of the Inter¬
national Scientific Committee, the Rapporteur, the different volume editors,
the authors and all those who have collaborated in this tremendous
undertaking. The work they have accomplished and the contribution they have
made plainly go to show how people from different backgrounds, but all
imbued with the same spirit of goodwill and enthusiasm in the service of
universal truth, can, within the international framework provided by Unesco,
bring to fruition a project of considerable scientific and cultural import. My
thanks also go to the organizations and governments whose generosity has
made it possible for Unesco to publish this History in different languages and
thus ensure that it will have the worldwide impact it deserves and thereby serve
the international community as a whole.
Description ofthe Project
B.A. OGOT'
Former President, International Scientific Committee for the
Drafting of a General History of Africa (1978-83)
The General Conference of Unesco at its 16th Session instructed the Director-General
to undertake the drafting of a General History of Africa. The enormous task of imple¬
menting the project was entrusted to an International Scientific Committee which was
established by the Executive Board in 1970. This Committee, under the Statutes
adopted by the Executive Board of Unesco in 1971, is composed of thirty-nine
members (two-thirds of whom are African and one-third non-African) serving in their
personal capacity and appointed by the Director-General of Unesco for the duration of
the Committee's mandate. The first task ofthe Committee was to define the principal
characteristics of the work. These were defined at the first session of the Committee as
follows:
(a) Although aiming at the highest possible scientific level, the History does not seek
to be exhaustive and is a work of synthesis avoiding dogmatism. In many
respects, it is a statement of problems showing the present state of knowledge and
the main trends in research, and it does not hesitate to show divergencies of
views where these exist. In this way, it prepares the ground for future work.
(b) Africa is considered in this work as a totality. The aim is to show the historical
relationships between the various parts of the continent, too frequently
subdivided in works published to date. Africa's historical connections with the
other continents receive due attention, these connections being analysed in terms
of mutual exchanges and multilateral influences, bringing out, in its appropriate
light, Africa's contribution to the history of mankind.
(c) The General History of Africa is, in particular, a history of ideas and civilizations,
societies and institutions. It is based on a wide variety of sources, including oral
tradition and art forms.
(d) The History is viewed essentially from the inside. Although a scholarly work, it is
also, in large measure, a faithful reflection of the way in which African authors
view their own civilization. While prepared in an international framework and
drawing to the full on the present stock of scientific knowledge, it should also be
1 During the Sixth Plenary Session ofthe International Scientific Commitee for the Drafting ofthe General
History of Africa (Brazzaville, August 1983), an election ofthe new Bureau was held and Professor Ogot was
replaced by Professor Albert Adu Boahen
Description ofthe Project
a vitally important element in the recognition of the African heritage and should
bring out the factors making for unity in the continent. This effort to view things
from within is the novel feature of the project and should, in addition to its
scientific quality, give it great topical significance. By showing the true face of
Africa, the History could, in an era absorbed in economic and technical struggles,
offer a particular conception of human values.
The Committee has decided to present the work covering over 3 million years of
African history in eight volumes, each containing about eight hundred pages of text
with illustrations, photographs, maps and line drawings.
A chief editor, assisted if necessary by one or two assistant editors, is responsible for
the preparation of each volume. The editors are elected by the Cornmittee either from
among its members or from outside by a two-thirds majority. They are responsible for
preparing the volumes in accordance with the decisions and plans adopted by the
Committee. On scientific matters, they are accountable to the Committee or, between
two sessions ofthe Committee, to its Bureau for the contents ofthe volumes, the final
version of the texts, the illustrations and, in general, for all scientific and technical
aspects of the History. The Bureau ultimately approves the final manuscript. When it
considers the manuscript ready for publication, it transmits it to the Director-General
of Unesco. Thus the Committee, or the Bureau between Committee sessions, remains
fully in charge ofthe project.
Each volume consists of some 30 chapters. Each chapter is the work of a principal
author assisted, if necessary, by one or two collaborators. The authors are selected by
the Committee on the basis of their curricula vitae. Preference is given to African
authors, provided they have requisite qualifications. Special effort is also made to
ensure, as far as possible, that all regions of the continent, as well as other regions
having historical or cultural ties with Africa, are equitably represented among the
authors.
When the editor of a volume has approved texts of chapters, they are then sent to all
members of the Committee for criticism. In addition, the text of the volume editor is
submitted for examination to a Reading Committee, set up within the International
Scientific Committee on the basis ofthe members' fields of competence. The Reading
Committee analyses the chapters from the standpoint of both substance and form. The
Bureau then gives final approval to the manuscripts.
Such a seemingly long and involved procedure has proved necessary, since it
provides the best possible guarantee ofthe scientific objectivity ofthe General History of
Africa. There have, in fact, been instances when the Bureau has rejected manuscripts or
insisted on major revisions or even reassigned the drafting of a chapter to another
author. Occasionally, specialists in a particular period of history or in a particular
question are consulted to put the finishing touches to a volume.
The work will be published first in a hard-cover edition in English, French and
Arabic, and later in paperback editions in the same languages. An abndged version in
English will serve as a basis for translation into African languages. The Committee has
chosen Kiswahili and Hausa as the first African languages into which the work will be
translated.
Note on Chronology
Also, every effort will be made to ensure publication ofthe General History of Africa
in other languages of wide international currency such as Chinese, Portugese, Russian
German, Italian, Spanish, Japanese.
It is thus evident that this is a gigantic task which constitutes an immense challenge
to African historians and to the scholarly community at large, as well as to Unesco
under whose auspices the -work is being done. For the writing of a continental history
of Africa, covering the last three million years, using the highest canons of scholarship
and involving, as it must do, scholars drawn from diverse countries, cultures, ideologies
and historical traditions, is surely a complex undertaking. It constitutes a continental,
international and interdisciplinary project of great proportions.
In conclusion, I would like to underline the significance of this work for Africa and
for the world. At a time when the peoples of Africa are striving towards unity and
greater co-operation in shaping their individual destinies, a proper understanding of
Africa's past, with an awareness of common ties among Africans and between Africa
and other continents, should not only be a major contribution towards mutual
understanding among the people of the earth, but also a source of knowledge of a
cultural heritage that belongs to all mankind.
Note on Chronology
It has been agreed to adopt the following method for writing dates. With regard to
prehistory, dates may be written in two different ways.
One way is by reference to the present era, that is, dates BP (before present), the
reference year being +1950; all dates are negative in relation to + 1950.
The other way is by reference to the beginning of the Christian era. Dates are
represented in relation to the Christian era by a simple + or - sign before the date.
When referring to centuries, the terms BC and AD are replaced by 'before the Christian
era' and 'ofthe Christian era'.
Some examples are as follows:

(i) 2300 BP = -350
(ii) 2900 Be = -2900
ad 1800 =+1800
(lii) Fifth century BC = Fifth century before the Christian era

Third century AD = Third century ofthe Christian era.
Members of
the International Scientific Committee
for the Drafting
of a General History of Africa
Hie dates cited below refer to dates of membership .
Professor J. F. Ade Ajayi Professor M. Difuila
(Nigeria), from 1971 (Angola), from 1978

Editor Volume VI
The late Professor Cheikh Anta Diop

Professor F. A. Albuquerque Mouräo (Senegal), 1971-86; deceased 1986
(Brazil), from 1975
Professor H. Djait
Professor D. Birmingham (Tunisia), from 1975
(UK), from 1985
The late H. E. M. El Fasi
Professor A. A. Boahen (Morocco), from 1971; deceased 1991
(Ghana), from 1971 Editor Volume III
Editor Volume VII
Professor J. D. Fage
The late H. E. Boubou Hama (UK), 1971-81 (resigned)
(Niger), 1971-8 (resigned 1978);
deceased 1982 The late Professor J. L. Franco
(Cuba), from 1971; deceased 1989
Dr (Mrs) Mutumba M. Bull
(Zambia), from 1971 The late Mr M. H. I. Galaal
(Somalia), 1971-81; deceased 1981

The late Professor D. Chanaiwa
(Zimbabwe), from 1975; Professor Dr V. L. Grottanelh
deceased 1993 (Italy), from 1971
Professor P. D. Curtin The late Professor E. Haberland

(USA), from 1975 (Germany), from 1971; deceased
The late Professor J. Dévisse Dr Aklilu Habte

(France), from 1971; (Ethiopia), from 1971
deceased 1996
Committee for the Drafting of a General History of Africa
The late H. E. A. Hampâté Bi Professor Bethwell A. Ogot

(Mali), 1971-8 (resigned); deceased (Kenya), from 1971
1991 Editor Volume V
Dr 1. S. El-Hareir Professor C. Ravoajanahary

(Libya), from 1978 (Madagascar), from 1971
Dr I. Hrbek The late Professor W. Rodney

(Czech Republic), from 1971; (Guyana), 1979-80; deceased 1980
deceased 1993
Assistant Editor Volume III The late Professor M. Shibeika
(Sudan), 1971-80; deceased 1980
Dr (Mrs) A. Jones
(Liberia), from 1971 Professor Y. A. Talib
(Singapore), from 1975

The late Abbé Alexis Kagame
(Rwanda), 1971-81; deceased 1981 The late Professor A. Teixeira da Mot
(Portugal), 1978-82; deceased 1982

Professor I. N. Kimambo
(Tanzania), from 1971 Mgr T. Tshibangu

(Zaire), from 1971
Professor J. Ki-Zerbo
(Burkina Faso), from 1971 Professor J. Vansina
Editor Volume I (Belgium), from 1971
Mr D. Laya The late Rt Hon. Dr E. Williams
(Niger), from 1979 (Trinidad and Tobago), 1976-8

(resigned 1978); deceased 1980
Dr A. Letnev
(USSR), from 1971 Professor A. A. Mazrui
(Kenya)
The late Dr G. Mokhtar Editor Volume VIII,
(Egypt), from 1971 not a member of the Committee
Editor Volume II; deceased 1998

Professor C. Wondji
Professor P. Mutibwa (Côte d'Ivoire)
(Uganda), from 1975 Assistant Editor Volume VIII,
not a member ofthe Committee
Professor L. D. Ngcongco
(Botswana), from 1971
Secretariat ofthe International
Professor D. T. Niane Scientific Committee
(Guinea), from 1971 C. Wondji
Editor Volume IV Sector for Culture
Director, 'Chargé de mission'

Professor T. Obenga Head of History Projects
(People's Republic ofthe Congo), 1, rue Miollis
from 1975 75015 Paris
Biographies of the authors who contributed
to the main edition
Tlie abridged version was prepared from the texts ofthe main version written by thefollowing authors:
chapter i M. Malowist (Poland), specialist in the economic and social history ofthe Late Middle Ages
and early modern times, author of various publications and articles on the subject, former
Professor of History, University of Warsaw; former member of Clare Hall, Cambridge and
visiting member ofthe Institute for Advanced Study, Princeton, deceased
CHAPTER 2 P Diagne (Senegal). Doctor of Political and Economic Science; economist and linguist; author
of several works on African political power and on modern Wolof grammar; currently
Professor and consultant
CHAPTER 3 J. Vansina (Belgium) specialist in African history, author of numerous works and articles on
pre-colonial history of Africa, Professor of History and Anthropology, University of Wisconsin
CHAPTER 4 JE. Inikori (Nigeria), specialist in economic history, author of various publications and articles
on the subject, Professor and Head of Department, Ahmadu Bello University, Zana.
CHAPTER 5 J. E. Harris (USA) specialist in African diaspora; author of various publications and articles on
the subject, currently Professor at Howard Univer sity, Washington DC.
CHAPTER 6 R. Vesely (Czechoslovakia) specialist in history of Egypt, diplomatics, economic problems of
Arab lands in the Middle Ages, author of several works on history and diplomatic studies ofthe
medieval period of Arab (especially Egyptian) and Islamic countries; lecturer for Arab and
Islamic lands and their history at the Philosophical Faculty of Charles University, Department
of Oriental and African Studies, Prague.
CHAPTER 7 Y. F Hasan (Sudan): specialist in the history of the Sudan; author of various works on the
subject, formerly Professor of History, University of Khartoum and Director of the Sudan
Research Unit, University of Khartoum, at present Vice-Chancellor, University of Khartoum
B A Ogot (Kenya): specialist in African history, pioneer in the techniques of oral history;
author of many publications on East African history; former Director ofthe International Louis
Leakey Memorial Institute, former Deputy Vice-Chancellor, University of Nairobi, Professor
of History, Kenyatta University, Nairobi.
CHAPTER 8 M El Fasi (Morocco) author of a number of works (in Arabic and French) dealing with
linguistic history and literary criticism, former Vice-Chancellor of the KarawîyyTn University,
Fez
CHAPTER 9 M. H Cherif (Tunisia): specialist in North-African social and political history; author of several
articles on North-African history; University Professor and Dean, Faculté des sciences
humaines et sociales, Tunis
CHAPTER 10 B Barry (Senegal). Doctor in History; author of numerous works on Waalo, Senegambia and
Futa Jallon; former Secretary-General ofthe Association des historiens africains, now teacher at
the Faculty of Arts, Université Cheikh Anta Diop, Dakar.
CHAPTER I I M Abitbol (Israel): specialist in Nigerian Sudan; author of works on the history of the Sudan;
Senior researcher at the Hebrew University ofjcrusalem
CHAPTER 12 M. Izard (France): specialist in the pre-colonial history ofthe Mossi kingdoms (Burkina Faso),
anthropologist and historian; author of numerous publications and articles, especially on the
ancient kingdom of Yatenga; Senior Researcher at the Centre national de la recherche
scientifique (Laboratory of Social Anthropology), Pans.
J. Ki-Zerbo (Burkina Faso): specialist in the methodology of African history; author of a
number of works dealing with Black Africa and its history; Professor of History, University of
Dakar.
CHAPTER 13 C. Wondji (Côte d'Ivoire): specialist in the modern and contemporary history of Africa, author
of numerous works on African cultures and the history of Côte d'Ivoire; formerly Head ofthe
Department of History, Faculté des lettres de l'Université nationale de Côte d'Ivoire (Abidjan)
and Deputy Director of the Institut d'art et d'archéologie, University of Abidjan, at present
Councillor at the Permanent Delegation of Côte d'Ivoire at UNESCO, Pans
Biographies of Authors
CHAPTER 14 A. A. Boahen (Ghana): specialist in West African history; author of numerous publications and
articles on African history; former Professor and Head of the Department of History, and now
Emeritus Professor, University of Ghana.
CHAPTER 15 E J. Alagoa (Nigeria): specialist in African history and historiography; author of various studies
of the Ijo, the techniques and methodology of oral tradition and archaeology; Professor of
History, University of Port Harcourt
CHAPTER 1 6 D Laya (Niger): specialist in West African cultures, sociologist, author of works on the subject;
Director of the Centre d'études linguistiques et historiques par tradition orale (CELHTO),
Niamey.
CHAPTER 17 B. Barkindo (Nigeria): specialist in state formation and inter-state relations in the Chad basin;
author of numerous works on the subject; Reader in History, Bayero University, Kano.
CHAPTER 1 8 E. M'Bokolo (Zaire): specialist in the history of Black Africa; author of numerous works on
the subject; Senior Researcher at the Ecole des hautes études en Sciences sociales (EHESS);
lecturer at the Institut d'études politiques (IEP), Paris; producer at Radio France Internationale.
CHAPTER 19 J. Vansina
T. Obenga (People's Republic of Congo), specialist in African languages; author of a number
of articles on African history and of works on Africa in the ancient world, formerly Professor at
the Faculty of Letters of Manen N'Gouabi University, Brazzaville; at present Director-Genera]
ofthe Centre international des civilisations bantu (CICIBA), Libreville.
CHAPTER 20 Ndaywel è Nziem (Zaire): specialist in the history of Central Africa, author of works on the
subject, teaches History at the University of Kinshasa and at the Institut pédagogique national.
Director General ofthe Bibliothèque nationale du Zaire.
CHAPTER 21 K. M. Phin (Malawi): specialist in the history of Malawi, and of Central and Southern Africa;
formerly lecturer and senior lecturer in History, University of Malawi, at present Associate
Professor in History, University of Malawi
O.J. M. Kalmga (Malawi), specialist in the history ofthe Lake Malawi region; author of works
on the subject; Professor of History, National University of Lesotho.
H. H K. Bhila (Zimbabwe): specialist m the history of South-Eastern Africa, author of works
on Southern Africa, former Chairman ofthe Department of History, University of Zimbabwe;
at present Deputy Secretary to Parliament of Zimbabwe.
CHAPTER 22 H H. K Bhila.
CHAPTER 23 D. Denoon (UK), specialist in Southern and East Africa; author of works on the subject;
formerly visiting lecturer at the University of Ibadan; currently Professor at the Australian
National University, Canberra
CHAPTER 24 E Haberland (Germany), specialist in pre-colonial Ethiopia and West Africa, author of works
on Ethiopia, Director ofthe Frobenius Institute Frankfurt; Professor ofthe Chair of Ethnology
and African history, University of Frankfurt; Director, Centre of African Studies in Frankfurt.
CHAPTER 25 A. I Salim (Kenya): specialist in East African history; author of many articles on the Swahili-
speaking peoples; Professor and currently Chairman, Department of History, University of
Nairobi.
CHAPTER 26
J B Webster (Canada): specialist in pre-colonial history, with special attention to climate;
author of works on Acholi and Iteso history as well as Chronology, Migration and Drought m
Interlacustrine Africa, formerly Professor and Head of History at Makerere and the University of
Malawi, now Professor of African history, Dalhousie University
B. A Ogot.
J P Chrétien (France) specialist in the history of Burundi; author of numerous works on the
Great Lakes kingdoms and German East Africa; Senior Researcher at the Centre de recherches
africaines (CRA), University of Pans I.
CHAPTER 27 W R Ochieng' (Kenya), specialist in the history of Kenya; author of numerous works on the
subject; formerly Senior Lecturer in Kenyan history, Kenyatta University College, Nairobi;
currently at Moi University, Eldoret, Kenya
CHAPTER 28 R. K. Kent (USA): specialist in the history of Madagascar, author of numerous works on the
subject; Professor of African History, University of California, Berkeley
CHAPTER 29 B A. Ogot
The struggle for
international trade
and its implications for Africa
Introduction
In 1500 the geo-political map ofthe world revealed the existence of a number
of major, relatively autonomous regions: the Far East, the Middle East, Europe
and finally Africa, with its Mediterranean, Red Sea and Indian Ocean seaboards
which were opening up to growing trade with the orient.
The period from 1500 to 1800 was to witness the establishment of a new
Atlantic-oriented geo-economic system, with its triangular trading pattern Unking
Africa to Europe and the Americas. With the opening up of the Adantic trade,
Europe - particularly Western Europe - gained ascendancy over the Americas
and African societies.
In most European countries, the period from 1450 to 1630 saw considerable
economic, political and cultural expansion. Expansion overseas reached into
vast territories bordering the Atlantic and even as far as the Pacific coastline.
From the beginning the African coast was one such area, but the situation in
North Africa differed from that in the sub-Saharan regions. The Mediterranean
was the scene of bitter rivalry between, on the one hand, Spain, Portugal,
France and Muslim North Africa and, on the other, the Ottoman empire which
was becoming ever more influential. In 1517 the Ottomans conquered Egypt.
Subsequently they subjugated a large part ofthe Arabian peninsula and gradually
established their rule over Tripoli, Tunis and Algiers. In Morocco, however, the
Portuguese managed to dominate a large part of the coast as far as Agadir and
Safi, while the Castilians established bases in Tlemcen and Oran.
These conquests gave the Portuguese control of some long-established
and important gold- and slave-trade routes from Western Sudan through the
Sahara and the Maghrib to the Mediterranean. The Europeans diverted part
of this trade previously sent to the Muslim world to their own benefit, thus
reducing the amount of gold reaching the Maghrib. This might explain the
conquest ofthe Niger bend by the Moroccans in 1591, which enabled the
latter to gain control of a number of gold- and slave-trade routes running
from West Africa to the Maghrib and Egypt. The famous campaign of
Djudar Pasha, himself a renegade of Iberian background, is a typical example
Africa from the Sixteenth to the Eighteenth Century
ATLANTIC
OCEAN
500 1000 km
1.1 Africa: the main points of European trade contact from the sixteenth to the eighteenth
century
The struggle for international trade and its implications for Africa
ofthe great conquests so characteristic ofthe sixteenth century.

Throughout the fifteenth and the early sixteenth centuries the Portuguese
established numerous trading posts on the West African coast and made the
coastal population and its rulers participants in trade with the Europeans. After
1481-2 the fort at El Mina became the busiest trading post on the Gold Coast.
In East Africa the Portuguese used different methods: they crushed the defences
of Sofala, Mombasa and other coastal towns, installed garrisons there and raised
taxes for the king of Portugal. At the same time, they tried to take over the
trade in gold, ivory and metals between the coast, its hinterland and India.
In the early sixteenth century the trade at El Mina, at the mouth of the
River Gambia, in Sierra Leone and in Sofala brought large profits mainly from
the purchase of gold at low prices and, to a lesser extent, from the trade in
slaves supplied from the hinterland. This trade initially brought Portugal great
profit. The considerable inflow of African gold made it possible for John II and
his successor, Manuel, to mint the cruzado a high-value gold coin and, more
importantly, to expand the fleet and strengthen colonial and state administration.
This latter measure opened opportunities for the aristocracy and gentry to obtain
numerous prestigious and profitable offices. Trade with Africa, and
subsequendy with India, accelerated the development of the Portuguese trader
class which had still been relatively weak in the fifteenth century. Thus
Portugal might be thought to have entered upon a path of lasting economic
and political expansion. However, its backward and sluggish socio-economic
structure prevented this from happening. Overseas expansion necessitated large
financial oudays, and the purchase of gold and slaves depended on supplying
Africa with large quantities of iron, brass and copper goods, cheap textiles,
silver, foodstuffs and salt. These goods were not produced in Portugal but had
to be bought from foreign traders or in Bruges and, later, from the major
European trade centres of the time. Furthermore, expansion of the fleet
depended on the import of timber, mainly from the Baltic countries. Portugal
was unable to develop its own production because of its small population and
the severe competition from abroad, notably in industrial goods.
Europe's economic boom triggered a gradual rise in prices from the 1470s.
During the second half of the sixteenth century this rise became enormous,
affecting both agricultural products and industrial goods. The relationship
between the rise in prices and the increase in Portugal's profits from overseas
trade seems not to have benefited Portugal. The high expenditure associated
with overseas expansion was profitable only when it was possible to buy cheap
and sell dear. To do this, however, meant limiting or even excluding European
competitors, and that required a strong fleet. That was a costly enterprise. By
the 1470s Portugal was already engaged in armed conflict with Castile, which it
eliminated from West Africa, thanks to the temporary supremacy of the
Portuguese fleet and to diplomatic manoeuvres. The discovery and conquest of
America turned Castile's attention away from Africa but this did no: happen
with other European countries which were more economically advanced than
Portugal and thus more of a threat.
The Portuguese rulers, John II and Manuel, were compelled to seek support
for their colonial activities among the big financiers of Italy and Germany. The
King of Portugal's accounts, in Bruges and subsequendy in Antwerp, revealed
that the crown was heavily dependent on the major firms such as those ofthe
Frescobaldi, the Affaitati and the Fuggers. In the first stage of expansion, the
Portuguese rulers managed to monopolize the import of gold from Africa and,
to a large extent, the slave trade or at least its indirect profits.
In the mid-1 520s the Portuguese encountered their first difficulties in
acquiring gold, even in the region of El Mina. It seems that they were by then
already unable to supply enough goods in exchange and were facing competi¬
tion from their European rivals who were better placed because they did not
have to import the goods they supplied to the Africans. In addition, France,
England and Holland were not yet burdened with an overdeveloped administra¬
tion regulating overseas trade. Their traders had the wherewithal to buy greater
quantities and sell more cheaply.
The slave trade
Portugal was initially attracted to Black Africa by its gold, which had previously
been exported to the Islamic countries. However, they soon discovered a
second African product attractive to Europeans, namely slaves. Slaves had been
exported to the Arab countries for many years from much of the continent,
particularly Sudan. In the fifteenth and sixteenth centuries the Portuguese
benefited from this tradition and were making regular purchases of slaves from
West Africa, particularly Senegambia, which was a long-established economic
partner of the Maghrib. They successfully applied the trade methods used in
Senegambia, and, realizing that co-operation with local rulers and traders was
the key to success, successfully got them involved in the slave trade.
The Portuguese soon abandoned any moral objections to the slave trade,
believing, as did many Europeans, that it enabled blacks to attain salvation. As
non-Christians remaining in their own countries, they would have been
damned. It should, however, be added that black slaves appeared in Europe at
a time when the trade in white slaves from the Black Sea area was almost at
an end, and that from this time slave and Negro became interchangeable
terms.
Throughout the fifteenth and the beginning of the sixteenth centuries the
main market for 'black merchandise' was Europe, particularly Spain and the
countries subject to Spanish rule, and, to a certain extent, islands in the Atlantic
such as Madeira, the Canaries, the Cape Verde Islands and the island of St
Thomas. The main incentive for the slave trade in these islands was the intro¬
duction of the cultivation of sugar cane and cotton. Slavery could not develop
to any great extent in Europe because there was no economic reason for it.
According to P. D. Curtin, the number of slaves taken from Africa by Euro¬
peans between 1451 and 1600 amounted to 274,900. Of these Europe and the
Adantic islands took about 149,000, Spanish America about 75,000 and Brazil
about 50,000. These figures are characteristic ofthe early period ofthe Adantic
The strugglefor international trade and its implications for Africa
slave trade, before the great development of the plantation system in the New
World, which sparked off a heavy demand for black labour.
The first Africans in America were brought there by the conquistadores who
were their masters. They came mostly from Senegambia and either had first
been brought to Europe or had been born there. They were called ladinos in
America because they knew Spanish and Portuguese and had been at least
pardy influenced by the civilization of the Iberian countries. They were highly
thought of, contrary to the bozales, who had been brought directly from Africa
and were thus carriers of a very different culture.
In the fifteenth century the Portuguese had become increasingly interested
in the slave trade and, in the following century, they began to seek territories
that could supply them with large numbers of slaves. This was the basis for
Portuguese penetration in the Congo which began in the early sixteenth
century, and their subsequent conquest of Angola. The settlers on the island of
St Thomas were also seeking supplies of slaves, not only for their own planta¬
tions, but also to sell to the Spanish colonies in America and also, from the end
of the sixteenth century, to Portuguese Brazil. The size of Brazil's black
population then was only a few thousand but the following century saw an
increase in this number to 400,000 following the development of
sugar-cane plantations.
The conquest of America created great problems for the Castilian crown
which had to supply slaves to the colonists. However, the Royal Treasury
found an abundant source of revenue in the system of licences granted to
traders importing slaves for the colonists, especially as the price of the licence
rose as the demand for slaves grew. Among the first to obtain licences were
aristocrats close to the throne, and also big capitalists, probably because of their
proposals for setdement and mining in Venezuela. Naturally, the potential slave
traders sought to eliminate the costly offices of the Portuguese. For that, they
had to overcome the obstacles set up by the Portuguese in Africa, while in
America they had to resort to contraband as the Castilian crown permitted only
those to whom it had granted licences to import slaves. But the Spanish
colonists in America were permanently short of labour and were therefore
willing to deal with smugglers, who unloaded their cargoes in illegal harbours.
Thus, all seemed to favour increased exports of Africans to America, although
the slave trade only really boomed with the establishment of the great sugar¬
cane plantations. It transpired quite quickly that in Spanish America and
subsequendy in Brazil the Indian population could not stand the steady hard
work on the plantations, whereas Africans had all the required qualities.
Thus from the early 1500s, but more particularly from 1550, Africa played a
major though undesirable role as a supplier of labour and of some gold to a
flourishing world economy.
But the Portuguese position was constantly deteriorating. In Morocco, they
suffered severe blows from the Sa'ädl sharifs' holy war against the infidels. In
1541 they lost Agadir and soon after were compelled through financial
difficulties to give up almost all their Moroccan ports. The year 1560 witnessed
the first bankruptcy of the Portuguese crown. The cost of a colonial empire,
while bringing enormous profits to part of the aristocracy, the bourgeoisie and
some traders, had emptied the Royal Treasury and laid an ever-increasing
burden on much ofthe population.
Arrival of new European powers

From the 1520s the French, and from the 1550s the English, became dangerous
rivals to the Portuguese in Africa; and from the late sixteenth century the
Dutch were even more dangerous. Initially only individual French traders or
trading companies were active. Neither the French king nor the English crown
was direcdy involved in trade with Africa. Francis I even unsuccessfully tried to
stop his subjects from making expeditions to Africa so as not to jeopardize his
relations with Portugal at a time when France was in open conflict with the
Spanish Habsburgs. The French presence was particularly strong in the area of
Cape Verde and Senegal. The French often looted Portuguese ships at sea as
they were returning from Africa with a cargo of gold or from India with goods
from the Orient. By the last quarter ofthe sixteenth century the position ofthe
French in Senegambia was very strong, particularly in such centres as Gorée,
Portudal, Joal and Rufisque (Rio Fresco) in Wolof country. The French
brought textiles from Normandy and Brittany, spirits, metal goods and firearms.
The arms trade seems to have swung the balance in favour of France, as the
Portuguese crown long and stubbornly prohibited the export of firearms to
Africa, despite the eagerness of the local rulers to obtain them. The French
bought gold and ivory, and also malaguetta pepper, hides and palm oil. The
slave trade was not yet of any interest to them. By the mid-sixteenth century the
French were dangerous rivals to the Portuguese on the Pepper and Gold Coasts,
as they exported more goods to Africa than the Portuguese, harming them
particularly in the region of El Mina. Eventually they drove the Portuguese out
ofthe estuary ofthe Senegal and the Gambia, only to be pushed out in turn by
the English at the end ofthe sixteenth century.
English expansion in Africa in the sixteenth century was similar to that of
the French. They began by establishing solid economic relations with
Morocco, after the departure ofthe Portuguese in 1541. Between 1550 and
1565 English trading companies sent several expeditions, reports of which have
survived. They explored the coasts of West Africa as far as the Gulf of Benin
where they bought gold, hides and a few slaves. Portuguese accounts show that
by the end of the sixteenth century the English were in contact with the
population of the Gold Coast, but were little interested in the slave trade. By
the end of the century their position was very strong on the fringes of
Senegambia, from where they had ousted both the Portuguese and the French.
In 1588 the first English company for trading with Guinea was established
mainly by merchants from London and Exeter. There is no evidence that this
company developed to any great extent. Possibly, at the end of the sixteenth
century, the English found that looting Castilian ships on the waters of the
Atlantic was more profitable than trading with Africa.
The struggle for international trade and its implications for Africa
It was at exacdy this time that the Dutch made their appearance on the coast
of the African continent. They were at war with Spain and treated Portugal,
which was then ruled by Philip II, as an enemy. The huge capital amassed by
their traders and the strength of their fleet made it possible for them to expand
in India and Africa. Between 1593 and 1607 some 200 ships set out for Africa
from Holland. In 1594 and 1595 the Dutch came to Gorée, and a few years
later they reached Benin, where they bought cotton goods and cowrie shells
which they traded for gold and silver on the Gold Coast. In 1611 they built
Fort Nassau (Mouri) on the Gold Coast, their first fortified trading post on the
West African coast. They also developed trade with the Accra region. They
exported copper and tin ware, cheap textiles, spirits, weapons, various
adornments, things in everyday use and even spectacles which enjoyed varying
popularity. They also imported sugar from the island of St Thomas and
transported the semi-finished product to their refineries in Amsterdam.
By 1617 the Dutch were so strong in Senegambia that they were a dominant
force on the island of Gorée and had supplanted not only the Portuguese but also
the English and French in Joal, Portudal and Rufisque. They retained this strong
position for over 50 years. At the same time their ships visited the coasts of the
Congo and Angola. The Dutch, like the English and the French, were initially
little interested in the slave trade. But things changed at the beginning of the
sixteenth century when the Dutch began to purchase slaves in El Mina, Accra,
Arda and Benin and also Calabar, Gabon and Cameroon. These slaves were sold
in exchange for sugar to planters on the island of St Thomas, or transported to
Brazil. The conquest of Angola in 1641 was closely associated with the labour
needs of the Dutch in Brazil. Though the Dutch lost north-eastern Brazil and
were driven out of Angola in 1648, the close association ofthe two territories,
based on the slave trade, continued until the nineteenth century.
Throughout this period, East Africa was of little interest to Europeans. The
Portuguese, who held Sofala and politically controlled the other coastal towns,
did not penetrate into the interior. The volume of gold reaching the coast from
the hinterland was already declining in the mid-sixteenth century. This
reduction in supplies of gold may partly explain the decline of such towns as
Kilwa, Mombasa and Malindi, which, prior to the arrival of the Portuguese,
had been active in supplying gold and other goods to traders from India and
Arabia. The coastal inhabitants vainly attempted to persuade the Turks from the
Arab Peninsula to intervene against the Portuguese. The territorial expansion of
the Imâmate of Oman along the East African coast and islands at the end of the
seventeenth century brought about some changes, confining the Portuguese to
Mozambique alone, but it was not till the late eighteenth and the nineteenth
centuries that marked changes occurred.
At the southern tip of Africa the first signs of European expansion
appeared in the seventeenth century, when the Dutch East India Company
encouraged the settlement of Dutch (and German) peasants, who became
known as Boers. The pressure ofthe Boers who reduced the San to slavery or
drove them out of their lands was a bad omen for the indigenous population.
The stmgglefor international trade and its implications for Africa
Africa as a reservoir of labour
P. D. Curtin estimates that between 1541 and 1600, 274,900 blacks were
carried to Europe and America. Over the next few years, the number rose to
1,341,100 to reach 6 million in the eighteenth century, almost 2 million of
them being exported to Brazil. In the eighteenth century the export of black
labour to the English and French West Indies increased enormously and exports
to Cuba also grew. Africa, which had initially been seen by Europeans as a rich
source of gold, now became a reservoir of labour for the many European
estates in America, particularly the rapidly developing sugar-cane plantations.
The import of black labour to north-eastern Brazil, which had been very heavy
at the time of Portuguese rule, increased yet further with the Dutch
occupation. The situation changed only when the Dutch were driven out of
Brazil and began to apply their sugar-refining techniques in the islands in the
Caribbean, whose new plantations became serious rivals ofthe Brazilian ones.
In the English colonies in the Caribbean, slave imports rose from 263,700 in
the seventeenth century to 1,401,300 in the eighteenth century. The situation
was similar in the islands occupied by France, with San Domingo importing
almost 790,000 slaves in the eighteenth century. The cultivation of sugar cane
was also started in Cuba, creating a similar need for labour. Dutch Surinam, the
English and French plantations in Central America and the northern part of
South America also absorbed great numbers of black slaves. In North America
the tobacco plantations in Virginia and the rice plantations in Maryland also
helped to expand the slave trade. In the nineteenth century the development of
cotton plantations would transform the south of the United States into a vast
area whose economy rested on slavery.
The demand for black labour presented Western Europe with an unprece¬
dented task at a time when radical economic and political changes were taking
place. In the second half of the seventeenth century the decline of Spain and
Portugal became increasingly evident. Holland, then at the peak of its power,
began to be ousted by France and England whose economies were developing
rapidly. From the end of the seventeenth century these two countries were
exercising ever-increasing influence in Africa, while the Spaniards and even the
Dutch were playing no more than marginal roles. Meanwhile, Portuguese
success in Angola allowed them to retain an advantageous position in the slave
trade.
In the seventeenth century, Holland, England, and later France and other
countries set up companies for trading with Africa and transporting slaves to
America. The companies obtained from their respective governments monop¬
oly rights for trading with Africa, which enabled them to control prices. In
return, they were obliged to maintain the old forts and build new ones to
protect the European trading posts on the coast of Africa. The number of
European forts increased all through the seventeenth and eighteenth centuries,
particularly on the Gold Coast and nearby.
Meanwhile, whites in Africa were at loggerheads among themselves. The
10 Africa from the Sixteenth to the Eighteenth Century
traders and trading companies sought to secure for themselves control of the
best situated trading posts on the African coast. Each of the different rival
European groups was backed by its government. The great powers could not
be uninterested in Africa, since great stakes were involved. Even less powerful
countries such as Sweden, Denmark and Prussia tried to involve themselves in
African affairs.
The trading companies did not, however, have the success expected of
them. The first two English companies were not particularly active. The Royal
African Company, founded in 1672, controlled a large part of Africa's external
trade passing through the West African coast in the last quarter of the
seventeenth century, but its policy was challenged by American planters and
many English traders. After 1689 its privileges were gradually curbed and it lost
its monopoly.
An era of free trade with Africa had already begun a few years earlier in
England. The Liverpool merchants were the most powerful. Their remarkable
success in the eighteenth century was made possible by the rapid industrial
development ofthe Midlands, particularly of metals in Birmingham and textiles
in Manchester, which enabled them to supply Africa, cheaply, with knives,
weapons and other metal goods, as well as textiles. The English traders
exchanged these goods on the coast of Africa for slaves whom they carried to
America to sell to the planters in the colonies.
During the eighteenth century, England became the leading trading country
on the African coast. Its influence was felt from Senegal to the boundaries of
Cameroon. Though England had to hand over its outposts in Senegal to France
in 1799, its position in the Gambia and Sierra Leone had become stronger. It
also occupied a leading position in the Gold Coast's slave trade which had been
growing since the seventeenth century. But the English were not the only
Europeans involved. Holland and France were increasingly active on the Slave
Coast, in Senegal, in the region of Cameroon and in Loango.
Attempts at setdement began in Angola, where a few Europeans went to
settle. Along the coast from Cape Verde to the Congo, however, European
activity was limited to trade. The trading posts and factories founded by the
Europeans (such as Saint-Louis in 1626) were scattered along the coast, close to
convenient bays and, usually, African towns. El Mina, Accra, Whydah, Porto
Novo, Badagri and both Old and New Calabar - all famous in the eighteenth
century were meeting places where blacks supplied whites with slaves in
return for European goods. African rulers generally reserved for themselves
priority in trading with Europeans, but black traders also played an important
role. Even in Angola the Portuguese captured only a few slaves themselves,
leaving it to local agents to buy or kidnap slaves in the interior.
It is difficult to know the territorial extent of the slave trade in Africa. The
coastal regions were not the only ones to suffer. As early as the sixteenth
century there were reports of the long journeys slaves had had to make from
their places of origin to the ports where they would be embarked. In the
eighteenth century, when the export of Africans became large-scale, the main
The strugglefor international trade and its implicationsfor Africa 11
suppliers were the rulers of powerful countries such as Asante or Dahomey or

the traders of Calabar. The old African states such as Benin or Qyo, were much
less involved. The Congo, which entered a period of complete disintegration
in the eighteenth century, was never an important supplier of slaves.
The Europeans were not yet ready to undertake the conquest of Africa,
mosdy because they found the climate hard to bear and they were helpless in
the face of tropical diseases. The early Europeans, the Brazilians and the North
Americans who first made an appearance on the African continent preferred to
win the friendship of African rulers by supplying them generously with
whatever goods they wanted. A community of interests united those dealing in
slaves, that is African rulers, dignitaries and traders. It should be mentioned that
the movement for the abolition of slavery, started in the second half of the
eighteenth century, was strongly resisted in England not only by West Indian
planters but by merchants in the metropolis. It later became clear that the kings
of Asante and Dahomey were also firmly opposed to the abolition of the slave
trade.
It was only in the eighteenth century that European slave-traders appeared

on the eastern coast of Africa. The great distance from there to the American
markets made the transport of Africans to the colonies in the New World much
more difficult, though not impossible. According to Curtin, the number of
slaves brought to the Americas from central and south-eastern Africa between
1711 and 1810 was about 810,100, or 24 per cent ofthe total number of slaves
imported. In East Africa slave exports to Arabia, carried on by subjects of the
Imam of Oman, increased during the eighteenth century. But it was in the nine¬
teenth century that these regions suffered the incursions of Arab slave dealers.
In North Africa the expedition to the Lower Niger in 1591 seems to have
had no lasting effect on Morocco. The conquerors quickly made themselves
independent of their metropolis and their descendants, the Arma, set up small,
short-lived states. There is no evidence of any important changes having taken
place in trade between Morocco and the Niger bend. The slave trade seems to
have been considerable since, at the end ofthe seventeenth century, the sultans
of Morocco possessed an army composed of slaves which, for a certain period,
also exerted a strong influence on the country's politics.
Conclusion
The following can be said of Africa's external contacts from the sixteenth to the
eighteenth century:
(1) They mosdy occurred on the west coast of Africa and in its hinterland.
(2) Initially, the Europeans were interested in African gold but, from the
second half of the sixteenth century, the slave trade became more impor¬
tant. It helped to ensure the economic development of much of America
and the Caribbean, and also accelerated the accumulation of capital in
Europe (especially in England) and in Africa.
(3) At this time, European expansion in Africa was proto-colonial but largely
commercial. Exchange between the two parties was unequal, since the
whites were selling cheap goods in exchange for huge numbers of slave
labourers. Thus, although they did not try to conquer the continent, they
inflicted great demographic damage on Africa.
(4) Africa's role as a market for many European industrial goods in the six¬
teenth and seventeenth centuries has probably been underestimated.
(5) The advantages accruing to Africa from these contacts were limited to the
introduction of maize and manioc cultivation. That in no way compen¬
sates for the huge demographic losses, let alone the sufferings imposed on
coundess human beings abducted from their country and carried to far-off
lands to toil on plantations.
African political, economic
and social structures
during this period
Largely as a result of internal factors such as demography and ecology or under

the impact of external forces such as the slave trade, Christianity, Islam and
capitalism, African social, economic and political structures were continuously
transformed between 1500 and 1800.
New social structures
Islam and Christianity

In the religious area, European and Middle Eastern philosophies and religions
began to impinge, with Christianity and Islam becoming political forces in areas
where they had previously been unknown.
Any ground gained by Christianity in the coastal areas of East Africa was lost
with the assassination of Father Gonçalo da Silveira, a Portuguese priest who
had been trying to bring the Mwene Mutapa under Christian influence in
1560. Islam, on the other hand, made gains with the conquests of Ahmad Gran
(1531 in Ethiopia and those of the mai of Borno and the askiyas of
Songhay in the Sahara and Western Sudan.
The transition from captive to slave societies

The second important change was the replacement in most of Africa of the
black African jonya system by the European and Middle Eastern slave-owning
system.
The jonya (from the Mande word jon meaning 'captive') was found mainly
in Western Sudan and the NigerChad region. A jon was not transferable; he
owned the bulk of what he produced, was part of a lineage and belonged to a
socio-political category that was part of the ruling class. Jonya played a
considerable role in the states and empires of Ghana, Takrür, Mali, Känem-
Borno, Asante, Yoruba and Mutapa. The élite of the royal slaves belonged to
the ruling class; they exercised some power, made fortunes and could them¬
selves own slaves.
Oriental and Western slavery, on the other hand, in both its ancient and its
13
colonial form, made slaves beings with practically no rights who could be
bought and sold as chattels.
With the setting up of Muslim states or emirates, which progressively took
over the whole of Western Sudan in the course of the eighteenth century,
Muslim law and tradition were established in the region and the slave-owning
system replaced the jonya. The jonya did, however, survive amongst the
traditional aristocracy of Western Sudan and the Niger-Chad region which had
experienced little Muslim influence. Until the colonial conquest, jonya
continued to retain some influence in the Wolof, Serer, Hausa, Kanëmbu and
Yoruba states.
The spread of feudal structures

The third change was the spread of feudal structures in either pure or distorted
forms among the agrarian civilizations of Africa.
Feudalism as a political structure, a mode of production or a socio-economic
system entailed not only ties of fealty and suzerainty but an opportunity to
speculate and to make a profit out of the means of production. Landed
property was a key feature of the European and oriental feudal systems which
took root in the predominantly agrarian civilizations of Africa. Their influence
was to be seen wherever land tenure or control over a territory led to duties,
taxes and rents.
But before the advent of Islamic law or the Western-inspired mailo (land
system), land was not a source of income in black Africa. Those who appro¬
priated or passed on a plot of land or a hunting, fishing or food-gathering area
worked on the basis of simple user rights which entailed neither lucrative
speculation nor the possibility of sale. The sub-Saharan agrarian societies had
given rise to lamana, a land tenure system which precluded land-renting, tenant
farming and share-cropping. The form of economy peculiar to black Africa
entailed above all production for consumption. People produced what they
needed, but did not own the means of production.
Interaction between the different social structures gave rise to hybrid,
heterogeneous societies. From the sixteenth to the eighteenth century, African
society contained marginal areas of debased feudalism mixed with lamana.
Feudalism held sway where there was production for exchange and the lamana
system had been either done away with or modified.
In Egypt, the Turkish beylik system helped to foster the feudal system. The
rural aristocracy was protected, as in the Middle Ages in Europe, by odjaks
(Turkish fortresses and garrisons). Great local chiefs ruled an 'arsh (ethnic area)
and the dwars (tented camps) according to a hierarchical system. In the Sahel
and the Mauritanian Sahara, religious families and djuad (warrior chiefs) took
over and vassalized confederations of smaller communities under the cloak of
religious brotherhood or by right of conquest.
The chief Turkish and native dignitaries in the Maghrib were often granted
large estates. As in Egypt, economic activity was in the hands of the beylik
which monopolized manufacturing industry and controlled the cereal, oil, salt
African political, economic and social structures 15
and textile trades as well as trade routes. The guilds of craftsmen and merchants
were under its supervision.
In the rest of North Africa and northern Ethiopia, the feudal system
developed differendy because of ecological conditions which in some areas
made land ownership less concentrated. The great beylical estates of the Magh¬
rib extended over a wide area, and gave rise to the widespread distribution of
azel (fiefs) worked by the khammäsat (tenant farming) system on a one-fifth basis.
At the regional level, the pattern was still one of milk (small family holdings) and
(arsh (community holdings) but always within the speculative feudal super¬
structure.
In Egypt and the rest of North Africa, several centuries of Graeco-Roman

rule had already prepared the ground for the transition to feudalism, the iktâ'
system of the Mamluks. South of the Sahara, the transition to feudalism took
place under outside influence.
In northern Ethiopia, for instance, there emerged a landed gentry which
created large estates, because the Ethiopian nobility subverted the principles of
ambilineal descent and equal partible inheritance, thereby creating among them¬
selves transgenerational 'families'. Political power was gradually concentrated in
the hands of land owners, and this accumulation of landed property and
political power led to greater class differentiation in northern Ethiopia.
This semi-feudal regime was extended by Christian northern Ethiopians to
the southern areas where the ketema (garrison towns) were established with a
neftenia (landed élite) colonizing the gasha (occupied lands). The gahar (peasants)
who worked the lands for the benefit ofthe landed élite were more or less serfs
or, at least, tributaries obliged to pay the gabir or siso depending on whether
they were share-croppers or tenant farmers.
In the Great Lakes region, especially in the southern area which comprises
much of present-day north-western Tanzania, Burundi, Rwanda and Uvira in
north-eastern Zaire, the institutions of clientship constituted a semi-feudal
bond which regulated relations between pastoralists and agriculturalists. It was a
sort of contract between the donor (pastoralist), who provided catde, and the
recipient (agriculturalist), who put his services and those of his family and
future generations at the disposal of the donor and his heirs. These contracts
varied from society to society and changed over the yean.
The main factors contributing to the adoption of the Ottoman feudal socio¬
economic system were the spread of the Muslim emirate with the advent of the
askiyas in Western Sudan, the expansion of the empire of the Islamized mai of
Borno and the introduction of Kur'anic law as a result of conversion and the
djihäds. In Songhay the askiyas kept part of the traditional socio-economic
structure.
In the emirates the Muslim law of the djihäds was adopted, thus implanting
more firmly semi-feudal socio-economic structures. The almamia of Futa Toro,
Futa Jallon and the Sokoto caliphate were simply copied from Ottoman land
tenure and taxation systems.
The new legal system established under the emirates introduced feudal-type
land speculation. The njoldi (symbolic payments attaching to the master ofthe
land), kawngal (fishing grounds), yial (hunting grounds), hore kosam (grazing
land) and gobbi (mines) were transformed into annual dues payable to those in
power, and collected under state supervision. The office of tax-collector
became negotiable for a fee, as did most of the official posts within the system.
Share-cropping, tenant farming and land rental became the rule. In the
Islamized areas the many landless peasants, dispossessed by conquest or the new
legal system, became serfs and vassals subject to the Ottoman system.
Thus, from the sixteenth century onwards, socio-economic structures from
different backgrounds were therefore combining. The result was the emer¬
gence of a new social order in which feudal structures were superimposed on
the African lamana. In Western Sudan and Northern Nigeria, which were
becoming Islamized, the Muslim institution of the emirate replaced the mansaya
or was superimposed on it. In the Gulf of Guinea and Central and East Africa,
where Christian rulers appeared, the influence of the Christian feudal
monarchy became increasingly apparent.
Architectural and artistic developments

There were some important architectural and artistic achievements. The men
who built the towns in the Nile valley, the Maghrib and the Sudan and on the
coast, and erected Yoruba palaces, the edifice known today as the Zimbabwe
ruins, the houses, palaces and mosques of the East African coast and the tata
(forts) surrounding the Hausa cities were at once architects, masons, decorators
and town planners. The round or pyramidal huts made of stone or clay and the
storey houses of the Jolla were in the same architectural tradition as the
Koutoubia of Marrakesh, the tomb ofthe askiyas in Gao and that ofthe caliphs
in Cairo.
After the sixteenth century new advances in architecture continued to be

made, perhaps mainly in Western Sudan and Nigeria, but the towns of North
Africa and the Nile valley declined. The askiyas, however, who carried on the
tradition in West Africa, were great builders, like their Moroccan con¬
temporary, Abü'l-'Abbäs al-Mansür, whose accession to power coincided with a
vogue for big public works. The architectural traditions of the Sahel and the
Islamic world increasingly spread southwards. The Sudanese architectural style,
of which the mosques at Sankore and Jenne were the prototypes, spread from
the sixteenth century.
In Ethiopia, the Gondar period (c. 1632-c. 1750) witnessed the development
of new architectural styles promoted by the court. In Gondar and in other
towns, the imperial families erected huge and beautiful palaces, casdes, churches
and libraries with elaborate interior decorations. In the Swahili-speaking coastal
regions of East Africa there were major architectural innovations in the period
between 1700 and 1850, both in the patterns and in the design of houses.
These architectural developments were accompanied by great productivity in
related fields such as wood-carving and furniture-making.
New economic structures
Major economic structures developed during this period: the caste system,
which replaced the guild or corporation system; the predatory economy, mainly
in North and East Africa; and the trading-post or entrepôt economy, mainly in
Central and West Africa.
A craft economy and a caste-and-guild society

Medieval urban civilization had contributed to the division of labour through
the development of crafts and manufacturing activities. But in the sixteenth
century this development was uneven, with different tendencies showing
themselves in different social contexts.
In the civilizations of Western Sudan, the Niger-Chad region and the

Sahara, for instance, crafts and manufacturing activities developed on the basis
of more or less closed castes, organized on a lineage basis. Under the growing
influence of the civilizations of Takrür and the Sahara, the caste system tended
to crystallize, notably in southern Senegal, Mande territory and Hausaland. The
Mande nyamankala (caste system) long raised the status of blacksmithing until
the Takrüri occupied the area after djihäds. Metalworkers were held in high
esteem among the Fon and the Yoruba, but there, too, the impact of
immigrants from Takrür and the Sahara was to upset the prevailing order. In
Songhay, the askiyas were already governing a society in which the caste system
had taken root and developed.
The Torodo revolution at the end of the eighteenth century accentuated the
caste system in Takrür by deepening class divisions. The Sebbe peasants, the
Subalbe fishermen and even the Buruure nomadic herdsmen were progressively
reduced in status. They were not identified with the benangatoobe (including
cobblers, blacksmiths or griots) and were segregated from the nangatoobe (higher
caste). The Torodo marabout élite came to despise the defeated aristocraties and
to monopolize high office. The Torodo marabout élite came to despise the
defeated aristocracies and to monopolize high office. In the NegroBerber
societies of the Sahara, religious, ethnic and racial divisions gradually crystal¬
lized into hierarchies of castes.
A major aspect of the organization of industry or crafts at the time was the
degree of state control. In Mediterranean civilizations, there was generally a state
monopoly in a number of activities, such as weaving, shipbuilding, the develop¬
ment of arsenals, refining and foreign trade, but the states in black Africa seldom
exercised such control, even with the expansion ofthe armaments industry.
This period was marked by the contrast between the versatility of country-
dwellers and the marked specialization of townsmen. State industries developed,
with arsenals for weapons and even the building of river- and ocean-going
fleets, both in Western Sudan and on the West Adantic coast and also in the
Mediterranean and Indian Ocean countries.
The multiplicity of wars sometimes lent fresh impetus to metalworking. In

the sixteenth century Sonni 'Ali reorganized the Songhay arsenals, setting yearly
production targets for the workshops. Egypt became skilled in metallurgy and
began to produce Damascus steel. The Mande blacksmiths, organized on a caste
basis, exported their techniques to the new towns that had sprung up as a result
of the Atlantic trade. The precious-metal industry in Egypt and North Africa
continued to be supplied with gold from Nubia, Sofala and Western Sudan.
Craftsmen working in gold and silver stimulated trade in the suks (markets) of
towns in the Maghrib, Egypt and Western Sudan. Berber and Wolof jewellers
were outstanding for their gold and filigree jewellery work. The Swahih also
manufactured beautiful jewellery and other silver and gold items. Ceramics gave
rise to industries, pottery (like basketry) remaining a female preserve. The glass
industry continued to develop, spreading throughout Yoruba country, Nupe,
Hausaland, Egypt and the Maghrib. Among the Shona in the southern Zambezi
region, mining was widespread, and the gold and copper mines sustained the
economies ofthe region up to the eighteenth century.
Leather-working flourished most in Nigeria, where stock-breeding provided
ample raw materials. From the sixteenth to the eighteenth century, basketry
and weaving assumed an important place among the industries of the Niger-
Chad region. The manufacture of paper, which had replaced papyrus,
developed chiefly in Egypt under the influence of Samarkand. The Sudan did
not lag far behind and gradually began to turn out manuscripts, with Kanem
producing Kur'ans that were sold throughout the Muslim world. Trades
associated with the food industry, which had grown up in the Middle Ages in
towns in the north and in Western Sudan, also became established in Nigerian
cities. In the textile area, the growing of cotton and the weaving of cotton
cloth were well established in the Zimbabwean highlands and the Zambezi
valley by the sixteenth century. The Swahili city-states were also famous for
their fabrics. In Central Africa the raphia cloths of Kongo were renowned from
the fifteenth to the nineteenth centuries.
The predatory economy

Prior to the sixteenth century, long-distance trade had played a major role in the
economy of Africa. This had encouraged high productivity and had led to the
rise of urban civilizations and the forging of strong links between town and coun¬
try. Between 1500 and 1800, however, the predatory economy, a consequence of
Spanish and Portuguese expansion, was introduced and became established.
Between 1495 and 1523 Spain and Portugal had gradually extended their
sway over the Maghrib. When the Portuguese occupied Azemmür, in 1513,
the Arab-Berbers and the Sublime Porte decided to oppose European
aggression and the corsairs, sailing under the Ottoman flag, helped to redress
the balance of power. In 1514 one of the Barbarossa brothers, Abu Yusuf,
recaptured Djidjelli and Algiers, and his brother Khâyr al-Dîn consolidated the
reconquest. Tunisia and Algeria came under Ottoman sovereignty once again
and remained so, at least nominally, until the nineteenth century, despite
Charles V of Spain's expeditions. In 1551 Sinan Pasha occupied Tripoli in the
name ofthe Sublime Porte and then Tunis in 1574. In the mid-1 500s, Morocco
asserted its independence after reconquering Agadir, Säfi and Azemmür from
Portugal.
Despite their clashes with the European powers, the North African states
continued to preserve their freedom but their progress was impeded in the
sixteenth century by the breakdown ofthe world economic order. Henceforth,
the ports of the Maghrib and the rest of North Africa lived mainly by piracy
and on tributes and duties, rather than by trade or new industries. The ports of
Salé (Morocco), Algiers, Tunis and Tripoli had the protection ofthe privateer
fleet which, in the seventeenth century, enjoyed its golden age in the
Mediterranean. The corsairs spread terror, and throughout the eighteenth
century, too, there was continual instability in the western Mediterranean. The
Ottoman regencies of Algiers and Tunis were almost constantly at war with
one or other European power and this situation inevitably hurt mercantile
capitalism and the merchant class.
It is against this background that the Moroccan expedition against Songhay
in Western Sudan should be seen. The Moroccan ruler, Abu 'l'Abbäs al-
Mansür, endeavoured after the conquest of Songhay in 1591 to re-open the
gold and slave routes. In 1593 the capture of Timbuktu made it possible to
bring 1,200 slaves across the Sahara, but the expedition, led by Djüdar Pasha,
only hastened the downfall of this trade. That was the end of what had been
the greatest empire in Western Sudan in the sixteenth century.
Tripolitania and Egypt suffered less from the consequences of the decline of
trade in the Sahara and maintained their traditional trans-Saharan routes. The
Sublime Porte, which was established in Egypt and Tripoli, signed an alliance
with Känem-Borno and was thus able to maintain the flow of north-south
trade vital for its own supplies until the nineteenth century.
The predatory economy certainly contributed to the slow-down of
economic and technological growth in the southern Mediterranean, but socio¬
economic and political structures also played a part in the stagnation ofthat area
and its hinterland.
All the countries of the Nile and East Africa, as well as those in the Niger-
Chad region and Western Sudan were affected in varying degrees. In the
western Indian Ocean, for example, the years 1680 to 1720 became known as
'the period ofthe pirates'. Countries in direct contact with the new European
hegemonies were physically disrupted by the predatory economy, although
their decline was also due to their inability to put new life into a socio¬
economic structure increasingly influenced by a backward East. They were also
handicapped by their inability to establish power relationships that would have
enabled them to escape the inequality of the trading system as it existed during
the period.
The decline ofthe countryside:

poverty and insecurity among the peasantry
The predatory economy caused trade between town and country to stagnate
and therefore affected the relationship between them. Previously, their activities
and produce had been complementary. Towns had broken the vicious circle of
subsistence farming, deepened the division of labour and contained the seeds of
the new society. It was urban industries and crafts that had stimulated large-
scale agriculture, stock-breeding, fishing and hunting. They had also been
responsible for the planting of sugar and cotton on an industrial scale and the
growing of dye and perfume plants. Hydraulic installations, roads and stock-
breeding for wool, milk and meat all owed their development to the towns.
The sixteenth century brought disruption and crisis to that world. Urban
depopulation brought about decline in the rural economy, leading to wide¬
spread poverty among the peasants and the reversion to bush of large tracts of
arable land. Increasingly insecure, people in the rural areas sought refuge in the
depths of the forests where, cut off from the consumer culture of the towns,
they withdrew more and more into family or village subsistence farming.
Exchange of produce and barter were the main forms of trade.
The slave trade brought about new upheavals. Depleted of able-bodied men,
the rural economy declined. In some areas, peoples subject to continual
harassment reverted to a nomadic economy of hunting and gathering with
many leaving the savannah region for the forest.
The warrior aristocracy diverted a great deal of peasant manpower to its own
benefit, which had disastrous demographic consequences for the rural areas.
The ruling élites gave up farming and relied instead on raiding using the
services of the freemen and slaves they had captured. Peasants suffered most in
Western Sudan and the Niger-Chad region.
The oppression of the African peasantry by the rural and urban élites
increased with the tightening of the fiscal screw. Under Turkish Muslim law
taxation could be increased in non-Islamized lands. Muslims were made to pay
not only the zakät (the only tax to which a Muslim was Hable) but also the
kharâdj. As the black market in taxation rights in Muslim countries became
widespread, so the burden of taxation grew even heavier.
In such a situation the leaders of the djihäds and the Christian messianic
movements found it easy to enlist the mass support of the peasantry. Religious
leaders promised equality once order was restored. They blamed the traditional
aristocracies and the Europeans for all the disruption and social injustice.
From the seventeenth century, the peasantry's political role increased.
Peasant revolts swept the continent like a religious revolution, thus paving the
way for resistance to colonial conquest. These revolts were not the work of
captives or slaves but ofthe largest and most exploited class, the small peasants.
The countries ofthe Nile and the Indian Ocean
The impact of the predatory economy on the countries of the Nile Valley and
those bordering the Indian Ocean was equally disastrous. The East African
ports had been known for their trading activities since the eleventh century,
and were in direct contact with Arabia, Persia, India, China and the Medi¬
terranean. The Portuguese invasion marked the beginning of the collapse of
this urban commercial civilization. In 1502 Kilwa and Zanzibar were placed
under tribute by Portugal. In 1505 Francesco d'Almeida sacked Kilwa and

Mombasa and then built Fort Santiago at Kilwa. He prohibited all trade between
the towns and the merchants left them for Malindi and the Comoro Islands.
Lamu and Pate were occupied.
With the exceptions of Luanda and Mozambique, none ofthe stations set up
by the Portuguese and later by the Dutch, English and French was as big as an
average town in Western Sudan, or even the Swahili and East African ports of
the tenth to sixteenth centuries.
The entrepôt or trading-post economy

While the predatory economy became widespread in the areas bordering the
Mediterranean, the Nile and the Indian Ocean, the entrepôt or trading-post
economy prevailed along the Atlantic seaboard. The new maritime entrepôt
towns were fortresses before becoming centres of commercial civilization
scenes of violence and spoliation.
On the Guinea and equatorial coasts, the Portuguese, who had established
the entrepôt economy in the sixteenth century, looted more than they bought.
They had litde to offer economically: apart from wine and iron bars, their
goods were imported. They exchanged them for gold, slaves, leather, gum,
ivory, amber, yellow civet, cowries, cotton and salt. In the Gulf of Guinea, the
Portuguese bought goods from the Akan and resold them on the Nigerian
coast or in Kongo or Angola. In Senegambia, they took up residence in the
ports and became successful merchants. In the seventeenth century, when
European industry began to produce textiles and hardware, the economic
decline ofthe towns accelerated.
The trading posts contributed nothing to local prosperity. Before 1800,

Albreda, Cacheu, Santiago de Cape Verde, El Mina, Ketu, Calabar and San
Salvador were the most important trading posts and all had fewer than 5,000
inhabitants. Apart from the Europeans, most of the inhabitants of the trading
posts were laptos (local interpreters). The trading-post economy rested on the
Atlantic slave trade. At the height of this trade none of the trading posts
provided a market for local products or promoted the commercial and
industrial activities of the indigenous population. The slave-trade trading post
was above all an instrument of depopulation.
The direct and indirect contribution of the trading-post economy to world
prosperity, however, was considerable. After the discovery of the American
mines, the trading posts supplied a substantial part of the gold and silver of the
world. Moreover, the bulk of the labour force which developed the American
continent originated from them. In a word, they kept world trade going. They
were the fountainhead of industry, finance, and European and international
capitalism. France's trade, for instance, saw tremendous growth in the course of
the eighteenth century.
By getting a firm foothold in the economic network, the European
navigators interrupted the normal pattern of inter-African trade and established
their own monopoly of middlemen operating from the trading posts. Africans
no longer traded between Saint-Louis and Portendick, Grand Lahou and El

Mina, Angola and Kongo or Sofala and Kilwa.
The Portuguese network was used by other maritime powers from the
sixteenth century onwards. The abundant sources of information on trade in
the entrepôt forts and the seasonal trading stations from the time of the
Portuguese until the arrival of the Dutch, French and English tell a story of
violence and continual conflict. The Atlantic and Indian Ocean trading posts
were destroyed and rebuilt and changed hands several times in the course of the
struggle by the European, Ottoman and Omani maritime powers against the
stubborn resistance of local rulers, who levied duties or curva.
Finally, the technological innovations that revolutionized Europe had little
impact. The African economy suffered most from competition from European
industry and business. By cornering the ports the European traders paralyzed
the links between the coast and the interior. Thus the European states that
annexed coastal areas demarcated Portuguese, Dutch, French and English
spheres of influence even before the colonial conquest and, in so doing,
determined the development and political geography of these areas in the
eighteenth century. From the Moroccan expansionist drive against Songhay to
the shifting fortunes of the internecine wars in Western Sudan, most of the
upheavals on the African political scene originated in the process of disruption
set off by the European powers in the sixteenth century.
New political structures

The African political scene had already reached a degree of balance and stability
in the period between the twelfth and sixteenth centuries. In the sixteenth
century, Mediterranean Africa constituted a sub-system within the Arab-
Ottoman empire, with Morocco, Ifrikiya and Tripolitania making up part of it.
Egypt was an entity on its own. The Nile area, consisting of Nubia and
Ethiopia, stretched south to the state of Bunyoro-Kita in the area ofthe Great
Lakes, the Swahili city-states and the area south of the Zambezi dominated in
the late sixteenth century by the state of Mutapa. Southern Africa as yet had
few state structures. In Central Africa one system was dominated by the Kongo
and Tio kingdoms, another by the Luba kingdom. But the peoples in the
forests were not organized into states. Western Sudan and the Niger-Chad
region adjoined each other, with every-changing borders.
The development of the political map was to reflect external pressures and
their repercussions. New states emerged: either because they were better
armed, such as Känem-Borno, or because they had more oudets to the sea,
such as Kayor in Senegambia, Dahomey in the Gulf of Guinea, Angola in
Central Africa or the kingdom of Changamire south ofthe Zambezi.
The character ofthe African state itself altered. Huge areas without rulers or
sovereigns hitherto peopled by farmers, nomadic herdsmen and hunters were
conquered and turned into states with centralized structures. The village
structures of the Bantu and the Kabyles and Berbers of the Sahara were
replaced by the Maghribi Makhzen. the autocratic mansaya or farinya of black

Africa, the feudalism of the Ottoman beylikates or the system of the Muslim
emirates.
From the sixteenth century onwards, political life centred increasingly on

the coastal areas, the privateering ports and the trading posts. The aristocracy
collected tithes there; African governments equipped themselves with revenue
departments to collect taxes on foreign trade. Many treaties were concluded in
an attempt to codify this taxation system.
Morocco, Tunisia, Algeria and Tripolitania signed many trade agreements and
short-lived treaties of friendship with the Europeans and even the Americans. In
1780 a war between Morocco and Spain was ended by the Treaty of Aranjuez,
which redefined their borders and codified their trade relations. At about the
same time, Algeria, and then Morocco, compelled the United States to pay
ransom to the pirates to secure the freeing of some of their nationals.
It took wars, the destruction of trading posts and prohibitions on trade to
persuade the European powers to pay taxes. But these more or less regular
sources of income were the cause of wan among the aristocracies and ruling
classes throughout the continent.
Political entities were predominantly areas in which a balance had been
achieved and which had developed in relation to their domestic circumstances.
They varied in size and in the stability of their borders and governments. Some
were confederations of states and others unitary states or chiefdoms with limited
jurisdiction. Sometimes they were a clan, or an independent lamana in which
the first occupants lived a completely autonomous existence.
The instability introduced by the predatory and trading-post economies thus
set the pattern, from the sixteenth to the eighteenth century, for states or
economies which could no longer base their development on a consistent and
organized foundation.
Population movements
and the emergence of
new socio-politicalforms
in Africa
The nature ofpopulation mobility

One ofthe main ways in which the history of pre-colonial Africa differs from
the history of Europe and most of Asia is in population mobility, especially the
mobility of farmers. African farmers in tropical and subtropical regions were
more mobile than farmers in Europe or Asia because they practised extensive
agriculture and could not cultivate the same land year after year. They were
thus not tied to definite plots of land. It was the same with herdsmen, although
their techniques of coping with the environment were more comparable with
those of the pastoralists of central Asia and the transhumance practised in
Europe. This population mobility is a basic characteristic to be considered in
any study or reconstruction of Africa's past.
This volume gives evidence of many migrations, from the trekboere at the
Cape to those ofthe Somali and Oromo in the Horn of Africa, the Ngbandi of
Ubangi, the Jaga in Central Africa, the Tuareg in the Niger bend, the Mane in
Sierra Leone, the Fulbe all over West Africa and all the peoples of Madagascar.
These migrations were not any more typical ofthe period between 1500 and
1800 than of any other period; but by the late sixteenth century Portuguese
travellers speculated that these migrations were all related, being for them
characteristic of barbaric tribesmen, a stereotype to be found in African history
right down to recent years. This conception is not only derogatory; it also
makes it impossible to understand what really happened.
The word 'migration' means the movement of a population from one area
to setde in another. Migration is thus a concept expressing a change in the
relationship that exists between people, space and time. In this general sense, it
is better to speak of population movement. Such movements may occur
because the ratio of people to resources alters or because people try to reorgan¬
ize their space and its resources on a relatively large scale. In Africa the main
large-scale spatial organizations were states and trading networks.
Africans stressed the importance of migrations in their oral traditions. Many
traditions that seek to explain the origin of the world refer to founder peoples
or individuals who came from elsewhere - a place of genesis. Other myths
25
26 Africa from the sixteenth to the eighteenth century
assert that people emerged from the soil and were thus the owners of the land.
The first hypothesis was adopted by European scholars who -were confirming
their preconceived ideas. They had an image of constant invasions during
which the peoples of Africa were pushing one another around like billiard balls,
with each conquest setting off a new wave of refugees who would go off to
seek haven elsewhere, or perhaps would themselves drive out other peoples.
Despite appearances, the notion that Africans were forever on the move is a
baseless stereotype. In a crucial article, P. E. H. Hair documented the remark¬
able 'stability' of all the coastal languages from Senegal to Cameroon, and this
observation can be extended to most of the farming societies of Africa after the
sixteenth century.
Like the concept of 'migration', the concept of 'stability' refers to a relation¬
ship among people, time and space; it expresses an absence of change. But both
concepts are relative. Given a large enough area, such as the area stretching east
and north ofthe White Nile, the movement of peoples, such as that of the Jie,
for example, becomes stability, a mere adaptation of people to the soil and the
climate. At the other extreme, the removal of a village 10 kilometres away can
be described as a 'migration'. It is the same with time: only by looking over
long periods of time is it possible to see the 'migrations' of some peoples who
move slowly over several centuries (like that of the Bantu, for example, which
may have gone on for 2 thousand years).
We must first examine the usual movements brought about by the quest for
a livelihood so as better to understand the unusual, abnormal movements and
see how the different population movements actually arise. We can then turn
to the kinds of evidence of these unusual movements and take a look at the
major population movements in Africa between 1500 and 1800.
Mobility and land use

As there are four main ways to exploit the environment for food production
hunting and gathering, herding, agriculture and fishing so there are four main
patterns of usual movement associated with these activities.
Hunters and gatherers roam a territory which is fairly stable as long as the
population density is adapted to the requirements ofthe way of life. They live
in camps and move about frequently inside the territory, like the Ituri pygmies
and the Kung San in Botswana. They have to do this to follow game and find
sufficient quantities of plants on which to feed. Their movements vary accord¬
ing to the season. The mobility is high, yet such populations can sometimes
exploit the same territory for a very long time.
Herders too are mobile. Their animals need water, grass and salt and the
availability of these commodities varies with the seasons. In the Sahara, for
instance, nomads usually live near the edge ofthe desert or major oases during
the dry season and go deep into the desert when the rains come. These move¬
ments are known as transhumance and they often followed the same route year
after year. Such movements could involve enormous distances for camel
herders such as the Rigeibat of the western Sahara. Often, complex reciprocal
Population movements and the emergence of new socio-politicalforms 27
movements occurred between nomads herding different kinds of stock, accord¬

ing to the various requirements of camels, goats, catde and sheep. Moreover,
the nomads also cultivated a few crops and exchanged food with specialized
hunters or gatherers where they existed as did the eighteenth-century
trekboere in South Africa. Thus the natural resources were exploited in various
complementary ways by groups who coexisted, each with its own mobility
over the same area. Herding, however, was more susceptible than hunting and
gathering to climatic variations, especially to short-term ones.
Farmers were also mobile because they practised shifting agriculture, leaving
fallow the land they had farmed the previous year. Villages had to move when
the fields became too far away from people's homes. In recent times this
occurred every ten years on average. The best lands were those that were
fertilized both by irrigation and an annual deposit of silt. On such lands there
could be stable cropping which led to a settled population pattern. But, apart
from Egypt and the belt of wet rice cultivation on the western coast of Guinea,
areas of permanent cultivation were very rare in Africa, for lack of efficient
methods of intensive fertilization.
Most villages moved then, but so long as the population density remained
low, they followed a more or less circular pattern within a given territory. The
mobility of villages was not dependent on the state of the soils alone. In many
regions, for instance throughout the tropical rainforest of Central Africa,
farmers also practised hunting and trapping, with some gathering too. Thus,
among the Nzabi of Gabon, for example, the siting of the village could be
dictated as much by the needs of trappers as of farmers. The greatest menace to
this way of life was climatic variation, especially irregular rainfall patterns.
Famines were possible even close to the equator, despite its being a well-
watered area. To have no rain after planting was a disaster but so was continuous
rain that prevented planting. Droughts were more common in areas close to
deserts, but no region was entirely safe from them.
In short, for farmers as for herders or gatherers there was a maximum
population density and an optimum population density which varied with the
environment the nature of the soil, rainfall, topography, the availability of
supplementary food resources, the state of technology and the arrangements for
the sharing or distribution of resources.
Fishermen were settled people and rarely moved their villages. Yet seasonal
variations in the levels of rivers could force them to undertake long expedi¬
tions. They would then settle in camps on sandbars as on the Zaire or the
Kasai, sometimes hundreds of kilometres from the villages where they had left
their families. Permanent setdement was likeliest among fishermen operating
on the coast or on the shores of large lakes. Because they had boats they had a
cheap means of transport, which turned many into traders acting also as middle¬
men among different groups settled on land.
It goes without saying that peoples who moved regularly could do so for other
than economic reasons. The departure of a group could be precipitated by a rise
in mortality or conflicts with other groups or the need for self-defence. There
was also much individual mobility. Women would often marry into another
village; sons would go to live with their mother's kin. Slaves, pawnbrokers,
traders and pilgrims, reputed medicine men and hunters would often be on the
move. The stereotype that some African ethnic groups had hardly any contact
with the outside world is as false as the opposite one of perpetual wandering.
It should be stressed that the distinction between usual and unusual move¬
ments was much more tenuous in the case of individuals, even though the
resulting population movements could in total be spectacular. The slave trade
after 1660 could be seen as an everyday movement of people. Yet the transport
of slaves to America was by far the largest population movement that Africa has
ever experienced.
Urbanization also represents population movements. By the beginning ofthe
fourteenth century, the population of the city of Zimbabwe could have been
10,000, which means that there must earlier have been migration to the city.
When cultivated land became exhausted, the inhabitants abandoned the city
and moved back to the villages. Their dispersal represents an equally important
population movement. But, urbanization and the slave trade apart, there is litde
information about the movements of individuals and even families. We shall
only know more about them when we know how much population densities
varied decade by decade and area by area. But such individual movements did
exist and they could be enough to have a marked impact on population density.
Typology of unusual population movements

Unusual population movements by communities fall into two major categories
drift and migration. Drift is gradual, slow movement while migration is a
sudden movement. In addition, drift does not necessarily mean abandonment of
the original setdement, although it may eventually, whereas migration ordinarily
means abandoning the territory of origin. These two categories of population
movement can be subdivided into several types. Drift is called expansion when
it enlarges the territory of a group and diaspora when it is discontinuous and
new distinct setdements result. On the other hand, there is mass migration,
band migration (generally a band of pillaging warriors) and élite migration
when it involves tiny groups whose arrival precipitates major changes in the
host society.
The historical importance and scale of a population movement depend on
the number of people involved, the distance travelled, the duration of the
movement, its causes and its consequences. All these faetón need to be taken
into account in each case, but they cannot be used as a basis for classification
because they are too variable in meaning. Thus it is the characteristics of the
movement itself that we shall now examine for each of the five migration types
set out above.
Expansion
Expansion has occurred coundess times because ofthe natural mobility of most
African populations, and its character differs according to the way of life ofthe
society being examined. Shifting agriculture becomes expansion when drift
occurs primarily in one direction rather than in a haphazard or circular pattern.
The expansion may last a very long time and yet the population will always
appear to be setded.
In the forest, Mongo expansion southward from the great bend of the Zaire
began long before the sixteenth century and was still in progress at the end of
the nineteenth century in most of the area between the Zaire in the west and
the Lomami in the east. This expansion caused a flow from nuclei of higher
population density around the equator and Io south, to the river valleys and the
fringes ofthe forest.
Expansion could easily be accelerated either by increasing the distances
travelled during each move or by moving more frequently. Such expansion
then became a matter of precise intentions, often a pull factor. Thus, in just half
a century, the Nzabi of Gabon/Congo moved from their lands in the east of
the great Ogowe bend to the south-west so as to be closer to the trade routes
and be able to exploit new mineral resources. The Fang of the Upper Comoe
reached the Gabon estuary in a mere twenty years and then continued for
another forty years towards the Ogowe delta. They never abandoned their
usual way of life, villages would move one after the other, each one leap¬
frogging past the one that had preceded it.
Expansion by pastoralists followed a different pattern. The most common
was for members of a younger generation to leave the pole of transhumance of
the group and setde with their stock in fresh pastures. The Maasai advanced in
this way in the seventeenth and eighteenth centuries until all suitable lands were
occupied. Another example is that of the trekboere of South Africa. Second-
generation European setders established themselves near the Cape from about
1680 onwards. From the start, the herders complained about overpopulation,
despite the fact that the population was sparse. Taking some of the stock,
members of the younger generation would go off and setde elsewhere. Until
about 1780 this expansion continued into areas occupied mainly by other
herders (Khoi groups) whom the trekboere would replace. But then they came
up against a frontier of much better watered land occupied by Xhosa farmers
and herders.
Even hunter-gatherers could show drift. This may be the explanation for the
presence of Baka hunters in eastern Cameroon. These pygmies speak Ubangian
languages, but live further west than farmers belonging to the same language
group. The reason for this probably lies in a movement from the upper valley
of the Sanga which gradually expanded the area in which the Baka hunted
further west.
Large expansion movements are indicative of new population distribution.

One of the deepest and longest trends in the history of Africa has been the
inexorable advance of people in larger and larger numbers setting out to con¬
quer new areas and mould their environment to their way of life rather than be
determined by it. Thus the expansion of the Igbo east of the lower Niger
resulted in a complete transformation of the forest landscape and a marked popu¬

lation increase. Such expansions into sparsely populated areas had the effect of
raising population density over the continent as a whole as population increased.
Expansion movements must have occurred from an early date in Africa.
Hunters and gatherers were the first to extend the areas over which they
operated, followed by populations that were exploiting their environment in a
more intensive way. In some cases long-term deterioration of resources due to
climatic change led to expansion, the most spectacular case being the desicca¬
tion ofthe Sahara which gradually forced the farmers of Mauritania southward.
A people that was expanding did not necessarily drive out the other peoples
it met with on its way. Often a mixing of populations occurred, with the
indigenous adopting the culture ofthe newcomers, as in the case ofthe Fang; or
a new society and culture might emerge from such a fusion. Thus the Mongo
expansion southward gave rise to various peoples, including the famous Kuba.
When there -was no fusion, the indigenous peoples would gradually abandon
their territory. It appears that it hardly ever happened that a people driven out of
its territory invaded a neighbouring territory. Expansions produced few refugees
given the low population densities involved.
Diaspora
A diaspora is a discontinuous population movement which leads to the creation
of new settlements separated from the parent settlement. All diasporas are
linked to trade or pilgrimage, except perhaps that ofthe Fulbe herders all over
West Africa.
The most typical diasporas grew out of trade. The Phoenician, Greek and
Arab setdements along the coastal areas ofthe Horn and the East African littoral,
the European forts and the colony of Cape Town were all founded by overseas
traders. River trade and fishing played a major role in the history of diasporas.
For instance, between 1750 and 1850 the inhabitants of a large village at the
mouth of the Ubangi founded trading posts and settlements all along the Zaire
as far as the mouth of the Kasai; they mixed with other peoples and founded a
new ethnic group, the Bobangi. In East Africa the Swahili and their culture
spread in a similar way, from the coasts of Somalia and Kenya to the island of
Ibo off Mozambique and the Comoros. Diasporas also followed overland routes.
Mande traders formed Jahanka diasporas from the Upper Niger to the coast of
Senegal and Joola (Dyola) settlements from the Upper Niger to the Akan coast.
Other diasporas were due to the yearly hadj to Mecca. When, for one reason
or another, pilgrims could not continue on their way, they would setde where
they had had to stop. This is how the Takrüri, from West Africa, came to setde
in Sudan in the nineteenth century. Similarly, most of the marabtin bilbaraka in
Barka were descended from pilgrims from North Africa.
Diaspora peoples typically maintained contact with their home countries,
either through trade or because they lived close to pilgrim routes. In some
cases, however, they gradually moved apart. For instance, by the fifteenth
century the inhabitants of Sofala no longer had special links with the Lamu
archipelago or the Bajun islands in the Swahili heartland. Even the European
colonies tended to become removed from their home countries and formed
trading relations with other colonies. Thus the colonists at Cape Town were
closer to the Netherlands East Indies with Batavia as its capital than they were
to the Netherlands.
Diasporas are highly visible population movements. They are a sign of the
existence of long-distance communication networks and flourish with the
development of trade routes. They attest to the success of man's control over
space.
Mass migration
A mass migration occurs when a whole people men, women and children
leave their homeland with all their belongings to travel long distances for a year
or more. The greatest invasion of North Africa was that of the Banü Hiläl and
the Banü Sulaym from 1052 onwards which has been tentatively linked to
recurrent droughts in Arabia. It continued until about 1500 when it reached
Mauritania. It changed the cultural map of all northern Africa which became
fully Arabized.
True mass migrations are dramatic but they are rather rare. In the period
between 1500 and 1800 the only real mass migration was that ofthe Oromo
which precipitated other population movements. In the nineteenth century
there were only the Nguni migrations which convulsed Africa from the Cape
to Nyanza.
Mass migrations were difficult undertakings. The movement called for new
forms of social and political organization, often of a military kind. Migrants
would often resort to raiding and pillaging. Other peoples, disrupted by their
passage, would often join them, gradually increasing their numbers. Migrations
could lead to serious clashes and create refugees and set off secondary migra¬
tions or rapid expansions. Even when the initial migration was fairly rapid, the
population movements that it set off could last for a century and more. Thus
the Oromo migration began perhaps in the 1530s, but stability did not return
to the regions affected until about 1700.
This description applies to the most extreme cases. Earlier, when documen¬
tation is scant, it is often difficult to distinguish between mass migrations and
quick but massive expansions, especially by pastoralists. Thus the expansion of
the Luo is often understood as a mass migration. It involved whole populations
and precipitated numerous secondary movements, and brought turmoil to a
vast area, mainly east of the White Nile. But the movement of the Luo went
on for a very long time, perhaps half a millennium, and what evidence there is
points to drift rather than mass migration, as the Luo usually moved with their
flocks and stopped to raise crops. Parts of the Oromo migration show the same
characteristics. These examples show that a mass migration is sometimes more
like a massive expansion and that drift may also occur at such times. Its dis¬
tinctive features remain the productive capacities and the militarized structure
of a vast people on the move.
Mass migrations are usually attributed to cataclysmic causes, for example a

sudden change in climate with ensuing famine and epidemics. Sometimes, how¬
ever, they are not due to natural disasters. The Vandal migration, for example,
is linked to other migrations and to the fall of the Roman empire. Relative
over-population is advanced as a cause for the migration ofthe Banü Hiläl, the
Oromo and the mfecane but so far there is little proof of this. Population
pressure may well be involved in all mass migrations, but it is inadequate by
itself as a complete explanation.
Band migrations
Migrations by bands, always armed, involves a relatively small number of
people, mostly young men, constituting but a fraction of a population. The
usual process was for bands of warriors to strike out to conquer new lands,
sometimes, but not always, under a single leader. In some cases, such as that of
the Jaga and the Zimba, for instance, the bands finished up disintegrating after
defeat; in others, such as that of the Mane, the Tyokosi and the Imbangala, they
succeeded in establishing chiefdoms or kingdoms.
Bands could cause serious destruction but because their migrations were
smaller in scale and shorter than mass migrations, they caused less turbulence.
In most cases they arose from pull factors, but push factors were also sometimes
important, for example when the expansion of the trekboere drove the Korana
and Gonaqua to migrate to the Orange and Caledon rivers. Band migrations
were sometimes the consequence ofthe formation of a state or an expansion of
trade from which the migrants wished to reap profits, as in the case of the Jaga.
The Imbangala case is an instructive example. Bands formed near the
Kwango river, perhaps following changes within the expanding Lunda state.
These bands mixed with marginal people of Ovimbundu and Mbundu polities.
The Imbangala engaged in pillage for several years as allies of the Portuguese
who were carving out a colony in Angola. By about 1620 they settled near the
Kwango, just out of reach of the Portuguese. Their migration resulted in the
formation of a state, the kingdom of Kasanje, which became the premier entre¬
pot for the processing of slaves exported from inner Africa on the route to
Luanda. The Imbangala case therefore amounts to a reorganization of socio¬
political structures and trade, nothing more.
Elite migration
Elite migration is a favourite subject of oral traditions about the founding of
states. The first king is a foreigner, often a hunter; he comes from elsewhere,
alone or with a few companions; and while the population movement involved
is insignificant, the socio-cultural results are vast.
Some of these stories may well be wholly fictitious and do no more than
reflect the idea that the king had to be a foreigner because he is a being apart,
enveloped in an aura of sacredness and mystery. Others, however, are better
founded. Thus a Kuba tale tells of how an exiled Bushoong, Shyaam a Mbul a
Ngoong, returned home from the west and formed a kingdom out of
antagonistic chiefdoms. It has been shown that linguistic influences from the
west entered the Kuba area and then radiated outwards from the court, which
supports the notion of cultural domination. It is, however, unlikely that such an
effect was obtained from the arrival of one person, and even less from the
return of an exile.
In a study of population movements élite migrations can be ignored, but for

a study ofthe development of socio-cultural formations, especially states, they
can be of considerable importance.
Documenting population movements

The major sources documenting the history of population movements in Africa
are oral, linguistic, archaeological and written. But using them is difficult and
requires great care.
Oral tradition cannot record long-term expansion because daily life is so little
disturbed by it and the movement so slow that the population is hardly aware of
it. No oral tradition can encompass a mass migration because the scale of events
is too large; it can only record episodes in it. Moreover, oral traditions tend to
confuse mass migrations with band migrations which, being more localized but
equally striking, remain in the folk memory. People often remember a diaspora
because the inhabitants of a village know from which village their ancestors
came. Finally, élite migrations are often remembered in detail. The movements
of the ruling family of Mangbetu, for example, are told almost two centuries
before their kingdom was founded.
But if the researcher does not take into account the fact that traditions
express ideologies and cosmologies, he may fall into error. If a people believes
that everything comes from a single place, it will obviously believe too that a
migration brought it from that original place to where it lives now. The
existence of a paradise implies that of an original migration.
Ideology is also responsible for such themes of origin as the foreign hunter
becoming king. This theme is to be found among the Igala of the lower Niger,
the kingdoms of the southern savannah and the Great Lakes region, the
Shambaa kingdom in Tanzania and the Fipa state between lakes Tanganyika
and Rukwa. All these stories reflect not the migration of an élite but the
ideology of a state. But a state may indeed have been founded by a foreigner or
a small group of foreigners. These stories thus need to be checked by other
means, taking account of their precise symbolic meaning and their ideological
worth.
Another error often made in the interpretation of traditions is taking the part
for the whole. The traditions of origin of the Kamba, the Mera and the
Gikuyu in Kenya seem in each case to have elevated the story of one small
portion of the population into that of the whole population.
In general, the more stereotypical themes oral traditions contain (and they
are to be found even in clan, lineage, village or family histories) the more
suspect those traditions are.
Recourse is often had to linguistics, but the facts of language are also difficult
to interpret. The general rule is that, when speakers of two languages mix,
those in the majority end up imposing their language on the minority. That is
why migration or massive expansion generally leads to the spread of the
language of the migrating population. When, conversely, as a result of a dia¬
spora or a band migration, a small number of newcomers is brought into con¬
tact with greater numbers of indigenous people, it is the newcomers' language
that disappears. Most of the apparent exceptions to this rule are not in fact so.
A population retains its language in a diaspora when it is reasonably compact
and retains relations with its original home. Migrant bands also maintain their
language when they do not mix with other peoples: this was the case with the
Tyokosi and Mende of Sierra Leone, as opposed to the Imbangala bands who
abandoned their language because it had disappeared from their home country.
But there are genuine exceptions to the rule of linguistic assimilation by the
more numerous population, which can be traced to the prestige that some
minority languages had. Ndebele and Kololo survived in Zimbabwe and
Zambia less because they were the languages of conquerors than because,
shordy after the conquest, they began to be committed to writing and taught in
school. Otherwise, they would have been absorbed by the majority languages,
Shona and Luyi.
A second basic rule is that the language absorbed always leaves traces in the
language that absorbs it, in the form of loan words, expressions (caiques),
personal and place names or peculiarities of morphology and syntax. Study of
such traces can provide information on the relations among populations.
Linguistics even makes it possible to distinguish the various population
movements that occurred. Diasporas are the easiest to detect because they result
in long-lasting multi-lingualism and sometimes the formation of creóles.
Afrikaans is a créole, as is shown by the major changes in its morphology and
even its syntax, as well as the lexical contributions from Malay, Portuguese,
Bantu and Khoi or San. Bobangi is a creóle from several closely related Bantu
languages. Diaspora languages are characterized by simplified grammar and
multiple-origin vocabulary. Expansions give rise to far fewer linguistic
borrowings. They result in languages being distributed over a wider area than
the region directly affected by the movement, which is fairly congruent with
the supposed kinship ties among the populations. In mass migrations, on the
other hand, this distribution only weakly reflects kinship ties. Mass migrations
often produce mixes between more than two languages. The Langi are unlikely
to have been involved in a mass migration because their language, Luo, has
only been influenced by a single language, Karimojong. When a migrant band
is large enough, it imposes its language on the region where it setdes and leaves
traces in the regions through which it passes, provided that these languages are
not too close to its own.
But the main difficulty with the use of linguistics is that unless loans are
studied in detail using the 'words and things' (Worter und Sache) method, it is
not possible to know whether they ought not to be attributed to other causes
than population movements such as trade relations, the influence of a state

where the language is spoken by the ruling family or the prestige of a religious
language. Such a study would be arduous but might yield valuable results.
Archaeology has often been called to help assert the existence of expansions
or migrations when similar or identical elements of material culture (objects) or
identical customs (such as funerary urns) have been found in different places.
But archaeologists today are moving away from this theory of 'migrationism',
with the recognition that independent invention occurs more frequently than
was previously thought and items could have been diffused by many other
means than migrations.
Nevertheless, in many cases population movements seem to have been the
source of a great deal of diffusion. When it is observed that a new pottery style,
markedly different from previous ones, has spread uniformly in different areas,
it is reasonable to suspect diffusion. Thus archaeology does indeed seem to
confirm the existence of a population movement, called kutama, from south¬
east Africa to Zimbabwe around 1000. There are many other unfortunately
doubtful cases.
Some authors still use ethnographic distributions as evidence of migrations.

This once fashionable approach is now completely discredited. Thus, to argue
that simply because the Fang had crossbows like other peoples in the Ubangi
region, they must have migrated from there, is clearly unreasonable. Ethno¬
graphic characteristics can spread without population movement. Their simi¬
larities may be due to chance or independent invention. They prove nothing if
they are not accompanied by linguistic similarities.
The major population movements Í500

Only two parts of Africa were subject during this period to major redistribu¬
tions of population and the ensuing creation of new societies and cultures. This
was, on the one hand, the Horn area south ofthe Abbay or Upper Blue Nile,
including most of what is now Somalia and northern Kenya, and, on the other,
the area to the east of the White Nile, north of lake Nyanza and south of the
Sobat. There were several population movements in this area. The most
spectacular was the migration ofthe Oromo into Ethiopia in about 1535. By
1700 a large part of Ethiopia was under Oromo control. The Christians and
Muslims had completely lost power in the south-west; the Somali and Oromo
were locked in competition for good lands as far as the Tana and smaller
sedentary populations had been driven out of Shungwaya, a coastal area just
north of the SomaliKenyan border. These groups, ancestors of the Miji-
Kenda, setded in large fortified villages (kaya) behind the major port towns of
Kenya.
Further west, population movements had begun much earlier, perhaps about
1000, with the migration ofthe Luo along the White Nile. A large number of
groups were on the move, notably the so-called Karamojong group, the
Turkana, further east, and the Maasai and southern Nilotes. All these peoples,
except the Luo, were primarily pastoralists, like the Oromo and Somali. All
were seeking 'empty' lands lands with a relatively low population density.
Their movements were heavily dependent on natural conditions. Thus, the
Luo needed well-watered lands, while the camel-rearing nomads, the Somali
and southern Oromo, could live in dry areas. It is rare for a population to
change its way of life. In some cases, as expansions came to an end, armed
conflicts occurred between populations with the same type of economy. Thus,
the Kenya Luo wrested lands from their neighbours in the eighteenth century
and various Maasai groups fought over land in the nineteenth century. Clearly
population pressures were involved here.
These population movements were in effect the story of the colonization of
marginal lands, at least until the eighteenth century. The best-watered lands,
along the Nile, were first occupied by farmers who also kept cattle. They
repelled the newcomers who were seeking to extend their ranges as their
numbers increased. This mobility highlights the stability that prevailed
everywhere else on the continent, where man had mastered space.
It has been suggested that drought played a major role in these population
movements in north-east Africa, but it is no more adequate to explain the
mobility of populations than is the collapse of the powerful Ethiopian empire.
The fundamental causes are to be found rather in the tensions between the
areas where population density was relatively high, as in the Oromo heartland
and perhaps the Ethiopian highlands, and, on the other hand, less populous
areas such as northern Uganda and Kenya and the Rift Valleys in Kenya and
Tanzania. By 1700 these latter regions had been occupied by new groups with
economies and societies that permitted higher densities.
Elsewhere, population movements were occurring on a much more modest
scale. In southernmost Africa the trekboere were colonizing the Karroo veld by
ousting or killing its pastoral occupants, while in Namibia both the Herero and
the Namib were expanding at the expense of the original San and Dama. The
fringes of the Kalahari were being settled as were the arid lands of northern
Uganda and northern Kenya. But in the south-east a major area of dis¬
equilibrium was building up as population growth was outstripping resources.
The first signs of instability, perhaps, were the movements of the Tonga
northwards into south-eastern Zimbabwe and the band migrations of raiders in
Zimbabwe itself.
In the rainforest of Central Africa and in the savannahs of what is now the
Central African Republic, mass population movements were under way, but
slowly, maintaining a reasonable balance between their number and the
available resources. The drift of Mongo groups south led to the formation of
the Kuba kingdom and the powerful chiefdoms north of the lower Kasai.
Between the Zaire and the Ubangi some higher densities built up but in
patches. It was still possible for a people such as the Ngbandi in the Ubangi
valley to drift southward, beyond the river. But by the eighteenth century there
were signs of relative overpopulation and the second half of the century saw
the birth of a new people, the Zande. They expanded rapidly eastward towards
the Nile, creating a series of chiefdoms. In the western part ofthe rainforest of
Central Africa constant migration by drift was taking the populations setded at
the confluence of the Mbam and the Sanaga into less populated areas in the
south and also perhaps to the west.
Most of the small population movements occurred around states being
formed or collapsing. In West Africa the decline of the Mali empire seems to
have caused the Soso (Susu), the Baga and the Nalu to leave Futa Jallon and the
Mane to migrate in bands to Liberia and Sierra Leone. The development ofthe
Mossi states may well have been connected to the northward movement of the
Dogon who setded on the Bandiagara scarp and to the foundation, further
south, of Gonja by Mande bands. The formation of the Asante state led to
expansion in the west by Baule and Agni groups.
In Central Africa fewer such movements are known. The expansion of the
Lunda empire first, then of the Yaka kingdom on the Kwango precipitated the
migration of armed bands. The ones in southern Lunda and the one that
founded the Kazembe kingdom are the best known. In Malawi a spectacular
rise of armed bands occurred as a byproduct ofthe creation ofthe Maravi and
Lunda states in about 1600. In Zimbabwe small movements of expansion and
raiding cannot be confidently associated with the development or decline of
states. Some population movements had more to do with the colonization of
sparsely setded lands in the south. Madagascar is perhaps the prime example of
population movements related to the emergence of kingdoms and chiefdoms:
the Maroserana migrations are a good example of this. By 1 500 the population
ofthe island was still very mobile, and there were still empty lands. By 1800
most ofthe island was organized into states of various sorts. During the pro¬
cess of state formation space had been tamed. On the mainland state formation
was limited to part of central Tanzania, southern Tanzania and northern
Mozambique.
The impressive further development of trade routes south of the Sahara led
to the creation of new diasporas and the strengthening of older ones. From
west to east the Jahanka, Joola, Yarse, Hausa, Bobangi, Vili, Hungaan, Bisa,
Yao and Swahili diasporas are the best known. To these must be added the
European ones: the French on the Senegalese coast, the Portuguese in Luanda
and Mozambique and the Dutch at the Cape gready influenced the African
mainland in this period. Both European and African diasporas played a growing
role in the slave trade as trade routes developed. Peoples living near such major
routes sometimes moved closer to them and sometimes fled from them. Thus
the Itsekiri expanded to the nearby coast in order to gain a foothold on the sea
route to Benin and many inhabitants of Gabon drifted slowly toward the delta
of the Ogowe in order to trade there. But the population that was living
between the Kwongo and the Kasai seems to have fled the traders and raiders
and many in Zimbabwe seem to have moved away from the turbulence created
by the Portuguese feiras. But all these population movements were small-scale.
They represent no more than dynamic readjustments to a reorganization of
space in the light of new developments in trade and socio-political relations.
Conclusion
Space had been tamed in most of Africa long before 1500 and population
density was sufficiently low for processes of expansion to settle the problems
arising from excessive pressure on the land. The main exception was eastern
Africa, from the Horn to the Zambezi, excluding the Great Lakes area but
including Madagascar. The northern half of this area saw mass migration, new
forms of settlement and the rapid expansion of herding groups. In the southern
half and on Madagascar a new social organization, associated with the forma¬
tion of kingdoms and chiefdoms, helped to stabilize settlement and make for
more intensive land use.
Population growth cannot have had serious consequences for Africa during
this period since it was easily compensated for by population movements. Only
here and there do we see the emergence of new technologies making possible
better land use and, in turn, higher population densities. The lower Casamance,
Igbo country, the Cameroon grasslands, the mountains ofthe Great Lakes area,
along the western Rift with their systems of irrigation and intensive cultivation
of bananas, the Kukuya plateau with its novel forms of fertilization or the valley
of the Upper Zambezi were and still are exceptions in western and Central
Africa. Without going into detail here about why population growth was not
greater, we must at least mention that large numbers of Africans left the
continent as a result ofthe trans-Atlantic slave trade.
Outside eastern Africa, population mobility was affected most by the rise and
fall of states and by the extension of trading networks. Despite the collapse of
some of the larger states in the Sahel in West Africa, the percentage of areas
controlled by states was higher in 1800 than it had been in 1500.
An extensive network of trade routes and setdements existed in North and
West Africa long before the sixteenth century. There were also a few trade
routes in Central Africa, but they only developed in the seventeenth and eigh¬
teenth centuries, linking the Atlantic to the Indian Ocean. The spread of
trading networks made it possible to exploit resources on a larger scale just as,
politically, states controlled larger areas than chiefdoms and village confedera¬
tions. By linking Africa to other continents, trade routes drew it into the
hierarchical organization of the world, that is a system which, from 1500
onwards, was increasingly dominated by Europe.
These three centuries must be seen as a portion of a much longer evolution.
Compared with the previous half-millennium, the population of Africa
experienced much more stability and the mastery of space was much greater.
But we still know little about population movements. Historical demo¬
graphy and the history of technology are only in their infancy in Africa. We
need more data and we particularly need to replace vague notions such as
'migration' by more refined analyses. We shall then be better placed to chronicle
a fundamental theme in African history: the slow, gradual colonization of the
continent by its inhabitants.
Africa in world history:
the export slave tradefrom
Africa and the emergence of
an Atlantic economic order
Introduction
While in Graeco-Roman times the slave trade was a known fact in the area
around the Mediterranean, the export of slaves in significant quantities from

black Africa to the outside world dates from the ninth century. This trade
mainly served the area around the Mediterranean (including southern Europe),
the Middle East and parts of Asia, and went on for several centuries, lasting even
into the early twentieth century. However, the numbers exported annually to
this area were never very great. With the opening up of the New World to
European exploitation, following the voyage of Christopher Columbus in 1492,
a much more extensive slave trade from Africa, involving far larger numbers of
people, was added to the older trade: this was the trans-Adantic slave trade
which lasted from the sixteenth century to the mid-nineteenth century. The
place of this trade in world history has hitherto not been properly demonstrated.
It must first be noted that trading in slaves is not something that has only
occurred in Africa. The world has known chattel slavery and large-scale trade in
slaves since the days of the Roman empire. Historical records amply show that
all the races of the world have at one time or another sold their own people into
slavery in distant lands. For many centuries ethnic groups in eastern and central
Europe (particularly the Slavs from whose name the word slave is derived)
supplied slaves to the Middle East and North Africa.
Yet, from the point of view of world history, the export slave trade from
Africa, particularly the trans-Adantic trade, is unique in terms of its sheer size, its
geographical extent and its economics.
The difficulty of determining its exact place in world history is tied up with
the problem of the historical origins of the contemporary world economic
order. The controversy over this issue arises from a number of factors: first, the
tyranny of differing paradigms conditioning the thought patterns of different
scholars; second, the intrusion of political influences into scholarly explanations;
and third, inadequate information at the disposal of many scholars.
In this chapter, an attempt will be made to analyse the consequences of the
slave trade from Africa in the context of the evolution of the world economic
39
o
o
o
e
o
W
c
<
I
u
-a
ru
Si,
^5
"S
TS
s:
«^
Si <*
ru »3
-S §o
». gj
y c
S
£ s
8.3
<S
Africa in world history 41
order from the sixteenth century. An economic order may be defined as a system
of economic relations embracing several countries, simultaneously allocating
functions and distributing rewards among them through the mechanism of a
trading network. The development of such a system of international economic
relations entails the evolution of economic, social and political structures in the
various member countries which make it possible for the operation of the
system to be maintained entirely by the forces of the market. Once it has
reached that point, any important modification ofthe system can only arise from
a deliberate political action, occasioned possibly by a change of regime in one or
more countries -within the system.
The starting-point here is the notion that an economic order linking together
a vast area comprising diverse regions of the world emerged in the Adantic zone
in the nineteenth century. These regions were Western Europe, North America,
Latin America, the Caribbean and Africa. The core of this order was Western
Europe and later North America, while Latin America, the Caribbean and Africa
formed the periphery, with economic, social and political structures to match.
The extension ofthe Adantic economic order to Asia and the rest of Europe in
the nineteenth and twentieth centuries produced the modern world economic
order which has since undergone only minor changes. The regions comprising
the core and peripheral areas of the nineteenth-century Atlantic economic order
have retained their positions ever since, even within the wider order.
This chapter intends to show that the slave trade from Africa was a key factor
in the development of the nineteenth-century Adantic economic order. To
develop this argument, efforts will be made to bring out, on the one hand, the
role of the African slave trade and slavery in America in the capitalist transforma¬
tion of Western Europe and North America and, on the other, the role of the
same faetón in the emergence of dependency structures in Latin America, the
Caribbean and Africa by the mid-nineteenth century.
Methodology
Scholars often use differing conceptual frameworks which make them view the
same social facts differendy. This explains much of the controveny surrounding
underdevelopment and dependency. On the one hand, there is an undifferen¬
tiated view of all social change as leading to economic and social development.
On the other, especially among scholars within the dependency and under¬
development tradition, social change is differentiated: a distinction is made
between two directions of social change, one leading to economic development
and the other to underdevelopment and dependency. Both are change, how¬
ever, and can therefore be studied historically.
Looking at the historical processes leading to the present state of national
economies, the undifferentiated view of social change proves incapable of
offering a satisfactory explanation. Social change has been taking place in all
societies over the centuries. If all social change leads ultimately to economic
development, then surely most economies in the world should by now be
developed. Yet, by any accepted definition of 'economic development', only a

few economies in the world today are developed. The vast majority can only
hope to become so by taking very drastic measures such as those taken by
China. It follows that the social change which led these economies over the
centuries to their present situation is something other than a process of
development. It is a historical process which some analysts describe as a process
of underdevelopment and dependency.
The approach of underdevelopment and dependency theorists needs closer
examination. Social change entails economic, social and political structuring.
Some mixture of economic, social and political structures produces economic
development, but othen create obstacles to development. Analytically, therefore,
three types of economies can be distinguished: undeveloped, developed and
underdeveloped.
A developed economy should be taken to mean an economy with strong
internal structural and sectoral linkages, supported by advanced technology and
social and political structures that make self-sustained growth possible. An
underdeveloped and dependent economy is an economy that is structurally and
sectorally disarticulated by certain internal structures arising from previous inter¬
national relationships which make it extremely difficult for advanced technology
to take root, giving rise to a situation where the economy's expansion or
contraction depends entirely on the external sector. Finally, an undeveloped
economy is one that possesses neither the structures of development nor those of
underdevelopment and hence is still free to move in either of the two directions
depending on the opportunities that arise.
Thus, to undentand the global effects ofthe process through which the inter¬
national economy -was created, we need to examine the kinds of economic,
social and political structures to which that process gave rise in the different
economies it embraced. For that, it will be useful to use a fundamental
hypothesis provided by underdevelopment and dependency theorists, which is
that, during the mercantilist epoch, the capitalist transformation of what became
the core countries of the world economy produced at the same time a
consolidation and further extension of pre-capitalist social formations in areas
that became the periphery. If this was indeed the case, then the development of
the core countries led to both underdevelopment and dependency structures in
the periphery. This chapter is organized around this hypothesis in order to test it
against the historical evidence.
The volume ofthe slave tradefrom Africa

To make a fair assessment of the role of the slave trade from Africa in world
history, it is important to establish as accurate an estimate as possible of the

volume of the trade over the centuries. In this regard, considerable progress has
been made with respect to the most important branch of this trade, the trans-
Adantic slave trade, starting from estimates published by P.D. Curtin in 1969.
Since 1976 other specialists (J. E. Inikori, C. A. Palmer, E. Vila Vilar, L. B. Rout
Jr., D. Eltis and R. Stein) have published the results of detailed research centred
on different portions of these estimates.
Although this research work is not yet complete, all the results go to show
that Curtin's global figures for the Atlantic slave trade were much too low and
require a 40 per cent upward revision. Curtin's global estimate of 11 million
slaves exported thus rises to 15.4 million.
For the trade across the Sahara, the Red Sea and the Indian Ocean, the
existing estimates are less firmly founded. Raymond Mauny computed 10 million
slaves for the period 1400-1900 and Ralph Austen arrived at a total of
6,856,000 for the period 1500-1890, 3,956,000 for the trans-Saharan trade and
2,900,000 for the trade across the Red Sea and the Indian Ocean. Taking
Austen's estimates, which seem more securely based, then some 22 million
people were exported from black Africa to the rest ofthe world between 1500
and 1890.
Tlie capitalist transformation of Western Europe and North America

in the era ofthe slave trade and slavery
When Christopher Columbus arrived in the Caribbean in 1492, the West Euro¬
pean economies were by definition undeveloped. Subsistence agriculture and
self-employed handicraft manufacturing were still the dominant economic
activities of the working population employed. Social and political structures
were such that extra-economic coercion still dominated the distribution of the
social product.
However, some major changes had been occurring in Western Europe in the
previous three or four centuries. The growth of population and its regional redis¬
tribution during the Middle Ages stimulated the development of inter-regional
and international trade. Major innovations had been introduced in the organiza¬
tion of land and labour and these changes were accompanied by some changes in
social structures. The conditions were thus prepared which enabled the econo¬
mies of Western Europe to seize the opportunities offered by the emergence of
the Adantic system.
Of coune, not all the economies of Western Europe underwent changes to
the same degree in the late Middle Ages. In particular, England experienced the
most remarkable changes as a result ofthe combined effects ofthe wool trade and
population expansion. These marked differences from country to country in the
level of change explain the differing rates of capitalist transformation of the
countries of Western Europe between the sixteenth and nineteenth centuries.
In order to analyse the impact of the evolving Atlantic system on the
economies of Western Europe, two periods are distinguishable: 1500-1650 and
1650-1820.
During the fint period, the economies and societies of the Adantic region
were not yet structured in a way that enabled market forces to maintain fully the
operation of a single economic system that could allocate functions and
distribute rewards among member countries. In consequence, Western Europe

used its military superiority to command resources from other economies and
societies in the region.
The most important resources that came to Western Europe from the rest of
the Adantic area during this period were silver and gold. These came mainly
from Spanish America (the Spanish colonies in the New World), the gold trade
from West Africa having declined as the slave trade took hold. From Spain, the
silver and gold from the New World were distributed throughout Western
Europe, accelerating the process of commercialization of economic activities.
The interaction between the rapidly increasing quantities of money in circula¬
tion and population growth produced what is known in European history as the
price revolution of the sixteenth century.
The import of American bullion also gave a fillip to the growth of inter¬
national trade within Europe. In Spain, only Spanish nationals and Spanish
ships were allowed to carry goods to and from Spanish America, and trade was
restricted to only two ports in the whole of Europe, Cadiz and Seville. Yet, all
sorts of underground arrangements enabled merchants from other European
countries to participate in this trade with Spanish America.
Thus, in the sixteenth century, Spain became the centre of a large-scale inter¬
national trade dominated by Holland, France and England. The silver and gold
from Spanish America left Spain within months of their arrival and, injected into
the major economies ofthe region, fuelled their transformation processes. As A.
Christelow wrote: 'Spain kept the cow and the rest of Europe drank the milk',
and this continued all through the seventeenth and into the eighteenth century.
In the second period, 1650 the structuring of the economies and
societies of the Adantic area reached its climax and the process of capitalist trans¬
formation in Western Europe came to depend on the Adantic system. The role
of the latter system in the economic development of Western Europe can only
be fully appreciated when viewed against the background of the general crisis
ofthe seventeenth century in Western Europe.
Imports of American bullion reached their peak in the 1590s and then
declined, and population growth slowed down. The situation was wonened by
the policy of economic nationalism adopted by a number of West European
countries during the seventeenth century. As a result ofthe tariff walls erected by
France, England and other countries to protect home industries, the economic
situation degenerated into a general crisis and inter-regional trade collapsed. The
process of capitalist transformation was stopped in its tracks in some countries,
while a revene process occurred in othen, with Italy being the wont hit.
The nature and origin of the seventeenth-century crisis show clearly that for
the capitalist transformation process to be completed, the region needed much
greater economic opportunities than Europe alone could offer.
The changes that occurred in the structuring of the economies and societies
ofthe non-European areas ofthe Adantic between 1650 and 1820 offered both
immense opportunities and challenges, which would completely alter the
economic situation in Western Europe. In the New World, the production of
precious metals continued to be important, but the key element in the structur¬
ing of the economies and societies of the region was the growth of large-scale
plantation agriculture. In mainland North America, the products were chiefly
tobacco and cotton, but in Latin America and the Caribbean, sugar was king.
The scale of operation of the new economy necessitated a complete repopula¬
tion ofthe New World.
A great volume of trade was organized around the shipping of goods from
Africa and the Americas, slaves from Africa to the Americas and agricultural
produce and precious metals from the Americas to Western Europe. Since the
colony-owning Western European countries restricted the movement of goods,
their control of the distribution of American commodities in Europe became a
major factor in the growth of intra-European trade in the seventeenth and eight¬
eenth centuries. The main beneficiaries of these developments were England,
France and Holland. According to F. Crouzet, a French economic historian,
'the eighteenth century can truly be called the Atlantic stage of European
development'.
The new economic opportunities brought about by the expansion of the
Adantic system increased employment and stimulated population growth all over
Western Europe after the seventeenth-century decline. The growth of domestic
markets in England, France and Holland, combined with the growth of exports,
provided the demand pressures that provoked the inventions and technological
innovations of the eighteenth-century and nineteenth-century industrial revolu¬
tions in Western Europe. In this way, the phenomenal expansion of commodity
production, trade, finance and shipping which occurred in the Adantic area
between 1650 and 1820 enabled the countries of Western Europe to overcome
the crisis of the seventeenth century, break the shackles of traditional economic
and social structures and complete the process of capitalist transformation. The
fint country to succeed in doing this was England.
That part of the New World that in 1783 became the United States of
America, but which in the seventeenth century and until 1776 was made up of
colonies, nevertheless played a considerable role in this expansion even in this
early period. When Columbus arrived in the Americas, these territories were
probably among the most economically undeveloped. Population density was
very low and the territories were dominated by subsistence activities which
continued long after their setdement by European colonists.
The involvement of these territories in the Adantic system in the yean
immediately preceding the Declaration of Independence considerably altered the
situation. The total annual value of the Atlantic trade of British North America
during these yean was ¿£8.4 million. With a total population of 2.2 million in
1770, that represents /J3.80 per head of population. The magnitude of involve¬
ment in the Adantic system stimulated the growth of the domestic market and
production for market exchange, as it encouraged specialization, raised per
capita income and influenced the rate of migration into the area.
As the British North American colonies moved gradually from subsistence
activities to market production under the impact of the Atlantic system, three
economic regions comprising the southern, middle and northern (mainly New
England) colonies became distinguishable. The combination of rich natural
resources and cheap African slave labour encouraged the southern colonies to
expand plantation agriculture, first rice and tobacco, then cotton. The middle
colonies took to foodstuffs production on family farms. The northern colonies
were relatively poor in natural resources for agriculture but were endowed with
deep natural harboun and forest resources for shipbuilding and this encouraged
early specialization in trade and shipping. In this way, the south produced
practically all the plantation commodities that were exported to Europe, while
the north produced the bulk of the invisible exports (shipping, merchants,
services, insurance etc.) and the middle colonies produced foodstuffs for export
and some export services.
These differing forms of involvement in the Adantic system produced
differing economic and social structures. In the middle and northern colonies,
production was based on free white labour, property ownership was widely
spread and incomes were quite evenly distributed. In the southern colonies, the
predominance of plantation agriculture dependent on African slave labour
produced a population with a large proportion of slaves, a high concentration of
property ownenhip and an extremely uneven distribution of income. Of the
697,000 slaves in the United States in 1790, 642,000 were in the southern states,
where they accounted for 36 per cent of the total population. While the
structures in the middle and northern colonies encouraged the growth of a
home market for mass consumer goods, the structures in the south restricted the
growth of that kind of domestic market and encouraged the import of foreign
luxury goods. In this way, while the foundation for self-sustained economic
growth was laid in the middle and northern colonies during the colonial period,
dependency structures were built up in the south.
After the achievement of independence, the economy ofthe southern states
continued to depend on African slave labour, to whom they owed the
phenomenal expansion of their cotton production between 1790 and 1860. In
consequence, the economic and social structures of the colonial period were
maintained.
However, the politically independent United States government adopted

economic measures which gradually turned the south from dependency on
Europe to dependency on the northern states. With government protection, the
ship-ownen and merchants of the north-eastern states took over the shipping of
southern cotton to Europe and the import of European manufactures for
southern planters and their slaves. At the same time, the expansion of cotton
production in the south provided a growing food market which stimulated the
growth of commercial food production in the west and the flow of immigrants.
The creation of a vast domestic market encouraged the growth of import-
substitution industries in the north-east. In this way, the United States took
advantage of its political independence at the right moment to manipulate the
forces operating in the Adantic area to the benefit of its economy, aided by the
favourable structures that developed in the middle and southern colonies
during the colonial period. The dependency structures of the southern states
thus acted as essential conditions for the capitalist transformation of the northern
and western states.
The evolution of underdevelopment structures in Latin America and

the Caribbean
By our definition, the countries of Latin America and the Caribbean were un¬
developed economies when Columbus arrived in the area. Three main factors
were responsible for this general absence of development: population, geog¬
raphy and isolation from the rest ofthe world.
Estimates of the population of all the Americas in 1492 vary enormously,
from 8.5 million to 112 rnillion. The more recent research of the Berkeley
School, however, indicates that a range of between 50 million and 100 million is
more plausible. Relative to the large geographical extent of the Americas, even
the highest figure is small. Moreover, it was broadly concentrated in three areas:
Central America, comprising the kingdoms of the Aztecs and the Mayas; the
Inca empire of ancient Peru; and the Canbbean island of Hispaniola, now made
up of Haiti and the Dominican Republic. The rest ofthe New World was very
sparsely populated.
The low population density had adverse effects on the development of trade
and the division of labour. Areas with high population densities were difficult
of access because of the topography, and this limited intra-American trade. In
addition, the Americas remained isolated from the rest ofthe world until 1492
and, as a result, their rich natural resources had little commercial value and there¬
fore made litde or no contribution to population and trade. This explains why,
by 1492, the ancient civilizations of Central and South America had reached a
high level of cultural development but were not at all developed economically.
The trading opportunities opened up by the arrival of Europeans came under
conditions which led to the evolution of structures of underdevelopment rather
than development. Fint, the West European countries forcefully seized control
of the natural resources of Latin America and the Caribbean. The Indian
population, humiliated and demoralized, then overworked and decimated by

the unfamiliar diseases brought by the Europeans, collapsed all over the region.
Thus, the population of central Mexico, estimated at between 18.8 million and
26.3 million before the European conquest, fell to 6.3 million in 1548 and 1.9
million in 1580. By 1605, it was down to only 1.1 million.
The virtual annihilation ofthe Indian population had two major consequences.
The fint was the massive import of African slave labour, intended to replace it,
and the other was the take-over by European colonists of agricultural land that
was turned into large estates (hacienda or fazendd). Both developments provided
trading opportunities that stimulated the capitalist transformation of Western
Europe and North America, while at the same time producing underdevelop¬
ment and dependency in Latin America and the Caribbean.
oo
a
"S-fe
r c
o
m
?-
i
Ä, <u
"3
tí
r
o
«
«It
c
«i
«i
v. uo
rt
^0
<>
C
"V)
rrt
t¿ ta»
rt
^
^î
ITt
^
o
3
4_»
tt
C/5
¿.
«
J3
p»l
H
-* "
The magnitude of contraband slave imports into Spanish America in the

sixteenth and seventeenth centuries makes it almost impossible to quantify the
contribution of African slave labour to the production of precious metals in this
region during this period. However, a census taken in Spanish America by the
clergy in 1796 shows that there were 679,842 people of African origin in
Mexico and 539,628 in Peru. In Brazil, production of sugar for export in these
centuries depended entirely on African slave labour and, when the gold boom
occurred in the eighteenth century, actual production still practically depended
on their labour. This is borne out by the ethnic composition of the Brazilian
population in the eighteenth and nineteenth centuries. In 1798 people of
African origin made up 61.2 per cent ofthe total population of Brazil and in
1872, 58 per cent. The slave populations were concentrated in the six provinces
that produced gold and agricultural products for export: Bahia, Pernambuco,
Rio de Janeiro, Sào Paulo, Minas Gerais and Rio Grande do Sul.
In the Caribbean, from the second half of the seventeenth century, the large-
scale import of African slaves and the expansion of plantation agriculture
precipitated the demise of subsistence agriculture while production for export
grew rapidly. The ethnic composition was also profoundly altered. Thus, to take
the combined population of Barbados, Jamaica and the Leeward Islands, the
slave population increased from 40.5 per cent in 1660 to 80.2 per cent in 1713.
Similarly, the population of African origin in the French West Indies grew from
about 52 per cent of the total population in the late seventeenth century to
about 88 per cent in 1780.
It was this massive transplantation of African labour into Latin America, the
Caribbean and the southern territories of North America that produced the
phenomenal expansion of commodity production and trade in the Atlantic area
between the sixteenth and nineteenth centuries, but it was far from having the
same effects in these countries as in Western Europe and North America.
Because the population in Latin America and the Caribbean included a high
proportion of slaves, the vast majority of the people earned far too little for a
domestic consumer goods market to develop. Lacking an oudet locally, the
profits from mining and plantation agriculture thus went for the purchase of
manufactured articles imported from Europe or were repatriated to Europe to
finance investment and consumption there. This situation was aggravated by
colonial laws which restricted the establishment of industries in Latin America
and the Caribbean.
The lack of industrial development gave rise to disarticulated economies in

Latin America and the Caribbean, with the mining and agricultural secton being
closely linked to the economies of Western Europe and also, later, to the
economy of the United States. Associated with this development was the
emergence of economic empires closely tied to importing and exporting, the
only activities in which the mining magnates and agrarian oligarchies of Latin
America and the Caribbean saw any interest. The prosperous merchant class also
became enmeshed in importing and exporting. Even after the major countries of
Latin America won political independence in the nineteenth century, govern-
ments continued to favour the production of primary commodities for the

export and import of manufactured goods. In addition, the industrial revolutions
in Western Europe and the United States in the nineteenth century reduced
production costs, which meant that the prices of manufactured goods traded in
the Atlantic area were so low that the newly independent countries of Latin
America were unable to establish competitive national manufacturing industries
for the domestic market. Thus, by the mid-nineteenth century, the economies
and societies of Latin America and the Caribbean had become so structured that
underdevelopment and dependency had become entrenched.
The laying of thefoundations of dependency structures in Africa

The available data show that African societies were going through major pro¬
cesses of transformation when the Europeans arrived in the late fifteenth century.
Recent archaeological findings, such as the remains of Igbo-Ukwu, indicate that
a high level of social and economic transformation had already been attained in a
number of places by that time. However, the process had hardly begun and
economic and social structures still conformed basically to our definition of
undevelopment. The total population relative to the available agricultural land
was still very small and, in addition, was scattered over the huge continent, with
groups separated from one another by long distances and difficult terrain. The
presence of a massive desert between black Africa and the Mediterranean and
the Middle East (the centres of international commerce for many centuries)
limited black Africa's trade with the rest of the world to commodities with very
high value but relatively low transport costs gold and slaves.
The establishment of seaborne commerce between Africa and Western
Europe from the second half of the fifteenth century seemed at first to offer the
kind of possibilities that black Africa needed for rapid social and economic
transformation. The gold trade expanded. Trade in agricultural commodities,
such as pepper, was initiated and some stimulus was even given to African cloth
producen as the Portuguese and the Dutch participated in the distribution of
African cloths to different parts ofthe African coast.
These early developments, however, were short-lived. When the vast resources
ofthe Americas became accessible to Western Europe, the role of Africa in the
Adantic economic system was altered. The population which Africa needed to
build up in order to provide the internal conditions necessary for a complete
structural transformation of its economies and societies was transferred massively
to the Americas. The conditions created over a period of three centuries by this
massive transfer of population discouraged the development of commodity
production in Africa and laid the foundations for dependency structures there.
The fint loss inflicted by this forced migration was the elimination of
population growth and the depopulation of large areas of the continent. It was
estimated above that some 22 million people were exported from black Africa to
the rest ofthe world between 1500 and 1890.
In order to determine how far these exports reduced the reproductive

Africa in world history
capacity of the populations of black Africa, an analysis is needed of the age and
sex composition ofthe exported population, because it is the number of females
of child-bearing age that is important.
In the case of the trade across the Sahara and the Red Sea, young and
attractive females predominated because of the demand for concubines. It is
generally believed that the sex ratio was two females to one male. For the trans-
Adantic trade, recent research now provides data with which to show sex ratios
for 404,705 Africans imported into various New World territories during the
seventeenth, eighteenth and nineteenth centuries, that is, about 3 per cent of
total exports to the Americas. Overall, females made up 32.9 per cent of the
sample.
An examination of data for the trans-Adantic trade shows consistent
variations of the sex ratios according to the regions of Africa from which the
slaves came. The Nigerian area, from the Bight of Benin to the Bight of Biafra,
exported the largest proportion of females, ranging from two-fifths to one-half of
total exports. However, the other major exporting area, the CongoAngola area,
regularly exported a larger proportion of males than the general average. This
regional variation is very important in assessing the demographic impact of slave
exports at the micro-regional level.
For the whole of black Africa, the evidence analysed shows that the number
of females exported annually was of a magnitude that must have drastically
reduced the region's reproductive capacity. Considering the additional popula¬
tion losses caused by exports to the Americas (mortality between the time of
capture and the time of final export) and the 6.9 niillion blacks exported to the
rest of the world, the evidence indicates that the population of black Africa
declined absolutely, at least between 1650 and 1850. As for the areas of origin of
the large numben exported through the Bight of Benin, the Bight of Biafra and
CongoAngola, they must have suffered serious depopulation.
Again, because the enslavement of the bulk of the exported population was
achieved largely through force, the export slave trade had a serious distorting
impact on African social and political structures. Indeed, a number of con¬
temporary observen say as much. In 1730 an officer of the Dutch West India
Company observed that the Gold Coast had become virtually a pure Slave
Coast and that, since the introduction of firearms by the Europeans, the kings
and princes had engaged in terrible wan against one another with the aim of
taking prisoners who were immediately resold as slaves. Later in the eighteenth
century, an African observer, Olaudah Equiano, wrote:
From what I can recollect of these batdes, they appear to have been irruptions of one
litde state or district on the other, to obtain prisoners or booty. Perhaps they were
incited to this by those traders who brought the European goods I mentioned
amongst us. Such a mode of obtaining slaves in Africa is common; and I believe
more are procured this way, and by kidnapping, than any other.
These observations show the strong link that existed between the export slave
trade and the frequency of wan in Africa during the period.
One major distortion was the creation of military aristocracies that became so
politically influential that they determined the direction of state policy in
virtually all major African states of the period. The existence of a large export
market for captives made them see war as a means of acquiring prisonen to be
sold rather than more territories whose natural and human resources could be
integrated and exploited. This had two negative effects on these states: on their
size, which remained limited, and on their internal political stability, which
remained extremely fragile.
In several African societies, the existence of these military aristocracies, com¬
bined with the prevailing economic conditions, also stimulated the growth of
the slave mode of production. Large sections ofthe populations of major African
societies came to be held by individuals who, as either merchants or government
functionaries, were connected directly or indirectly with the export slave trade.
Given the shortage of human resources relative to the available cultivable land,
the structures that had been established remained even after the suppression of
the external demand for slaves.
The overall consequences of these historical processes covering a period of

more than three centuries was to alter the direction of the economic process in
Africa away from development and towards underdevelopment and dependency.
The expansion of the slave mode of production in large areas of Africa had the
effect of limiting the development of domestic markets and market production.
Moreover, the Atlantic slave trade in various ways obstructed the development
of commodity trade with Europe that would have stimulated the growth of
intra-African trade and production for the market. Hence, by the middle of the
nineteenth century, subsistence production of foodstuffs remained dominant in
African economies. This virtually eliminated any capital accumulation in agri¬
culture and, therefore, any increase in productivity in food-crop farming for the
domestic market. The low productivity of African foodstuffs farmers today is the
legacy of three centuries of history, whose negative features were made worse by
the economic impact of colonialism in the twentieth century.
The low level of the division of labour and the limited size of the domestic
markets were unfavourable to the development of manufacturing beyond the

handicraft stage. The development of manufacturing was further hampered by
the uncontrolled import of European and oriental manufactures in exchange for
captives. Thus, with limited domestic markets, non-capitalized agricultural and
industrial secton and small-scale states dominated by merchants and warriors
depending on a slave mode of production, the foundation was firmly laid for
Africa to become dependent on the industrialized economies of the Atlantic
area. And, starting in the late nineteenth century, the full edifice was completed
by colonial rule.
Conclusion
The evidence and analysis can now be summarized. When Christopher

Columbus arrived in the Americas, the economies ofthe Atlantic area were all,
by definition, undeveloped. In Europe as in Africa and the Americas, manu¬

facturing was at the handicraft stage and was part and parcel of agriculture, the
overwhelmingly dominant sector. Pre-capitalist modes of production were
everywhere dominant. By the mid-nineteenth century enormous gaps had
opened up between the economies of the various regions of the Atlantic area.
On the one hand, mechanized industries were concentrated around the fringes
of the Atlantic in north-western Europe and the north-eastern USA. On the
other, the greater part of the Adantic area was dominated by primary com¬
modities: commercial foodstuffs production and plantation agriculture in the
western and southern USA; plantation agriculture in the Caribbean; mining,
large-scale livestock raising and plantation agriculture in Latin America; and
subsistence farming and part-time gathering of wild products for export in Africa
(the export of slaves having been halted). The economies and societies of the
Atlantic area had become structured into a single economic system ruled by
market forces. Barring deliberate political action in one or more member states to
alter these structures radically, the situation was bound to be self-perpetuating.
The facts clearly show that this situation derived ultimately from the export
of slaves from Africa. The industrial revolution in England in the eighteenth and
early nineteenth centuries and in the north-eastern USA in the nineteenth cen¬
tury could not have occurred without the phenomenal expansion of commodity
production and trade in the Adantic area from the sixteenth to the nineteenth
century. There is no doubt whatsoever that it was African slave labour provided
by the slave trade from Africa that made this phenomenal expansion possible.
While this expansion stimulated the growth of free wage labour to become
the dominant form of economic activity in north-western Europe and the
north-eastern USA, in the rest ofthe Atlantic area the slave mode of production
expanded. In Latin America, the Caribbean and the southern USA, its
expansion provided the conditions for unequal development that facilitated the
rapid development of capitalism in north-western Europe and the north-eastern
USA. The historical process that produced capitalism in these areas along the
fringes of the Atlantic produced at the same time a consolidation and an
extension of pre-capitalist modes of production in Africa, Latin America, the
Canbbean and the southern USA.
It is important to note that at the time the Atlantic economic order was being
constructed, Western Europe could not establish with Asia a firm trading link
based on the exchange of what each produced. Europe had litde to offer apart
from the gold and silver of the Americas. By the nineteenth century, however,
Western Europe and North America were able to integrate the economies of
Asia firmly into the Atlantic economic order, as Asians found it difficult to resist
the products of their mechanized industry, notably textiles from English and
North American mills.
In this way, with the help of colonial rule, the Atlantic economic order was
extended to the rest of the world, producing the twentieth-century world eco¬
nomic order, which was originally constructed with the sweat and blood of
Africans.
The African diaspora
in the Old and New Worlds
Europe and the Americas

Lack of documentation makes it difficult to know exactly when Africans first
arrived in Iberia or the rest of Europe. However, it is likely that some Africans
from north and south of the Sahara made their way into the Iberian Peninsula
in ancient times when trans-Saharan trade flourished.
Africans also took part in the Muslim campaign in Iberia in 711, and the
ensuing centuries of Muslim-Christian warfare saw Africans fighting as soldiers
and serving as slaves. Indeed, as early as the thirteenth century, Moorish
merchants were selling slaves from sub-Saharan Africa at fain at Guimarâes in
northern Portugal.
The Portuguese capture of Ceuta in 1415 opened up the era of European
penetration into Africa. Africans were taken to Lisbon from 1441, and this was
the prelude to the slave trade which continued into modern times. Indeed,
between 1450 and 1500 Portugal imported an estimated 700 African slaves
annually. In 1468 the Portuguese crown initiated the famous asiento
(monopoly) over the trade in slaves south ofthe Senegal river.
The papal bulls of Nicholas V (1454) and Calixtus III (1456), which presented
Portuguese expansion into Africa as a crusade to Christianize the continent,
gave a degree ofjustification to the enslavement of Africans. This argument was
further supported by the biblical myth of Ham, the son of Noah, whose
descendants were said to be cuned and condemned to slavery.
In Spain and Portugal black slaves worked in mines, on farms, in construc¬
tion, as soldiers, guards, domestics, messengen, stevedores and factory
labourers; women were concubines. Those who were not slaves were
invariably to be found doing the most menial and arduous tasks.
The sale and use of slaves were primarily urban phenomena, as African labour
largely came in through ports and cities, chiefly Barcelona, Cadiz, Seville and
Valencia in Spain, and Lisbon in Portugal. Urban life provided numerous
opportunities for slaves to escape and, in some cases, to purchase their freedom.
The 'free' blacks congregated and sought to cultivate a community spirit and
defend their interests. Religious brotherhoods were organized in Barcelona in
56
The African diaspora in the Old and New Worlds 57
1455, in Valencia in 1472 and Seville in 1475. These organizations sponsored

recreational activities, raised money to free other slaves and bought burial plots,
as blacks had generally to be buried in separate areas.
Some free blacks achieved distinction in Spanish society. Cristóbal de
Meneses became a prominent Dominican priest; Juan de Pareja and Sebastián
Gómez were both painten; Leonardo Ortez became a lawyer; in 1475 Juan de
Valladolid was appointed supervisor over the blacks in Seville. More remark¬
able still is the case ofJuan Latino, a black scholar, who received two degrees at
the Univenity of Granada and taught there.
There was no official policy for the slave trade to the New World until 1518
when Charles I of Portugal proclaimed the Asiento do Negroes which had the
effect of intensifying competition in the trade in African slaves. Although under
Spanish rule, Portugal established a virtual monopoly ofthe slave trade for itself
by 1600, making contracts with the Spanish, the Dutch and the French. In
1713, following the War of the Spanish Succession, this monopoly went to
England, which then became the biggest slave-dealer in the world.
Yet the number of Africans living in England itself was growing all the time.
In 1556 Elizabeth I observed that there were too many 'blackamoors' in
England and that they should be returned to Africa. From the eighteenth
century, West Indian planters adopted the habit of bringing their African
domestics with them when they came on home visits. Military and naval
officers and captains of slave vessels did the same. Gradually ownenhip of black
slaves came to be seen as a mark of social distinction. Newspapen in London,
Bristol, Liverpool and elsewhere offered slaves for sale and carried advertise¬
ments to secure the return of runaways. In England slave-hunten became past
mästen at capturing Africans who enjoyed no legal protection. Many of them
were identified by the marks that had been burnt into their skins by slave-
ownen. The psychological effects of this control by whites over blacks cannot
be overestimated. The process of dehumanizing the African was well under
way by the eighteenth century.
But the question arose of knowing the status of Africans living in England.
Some English people thought that conversion to Christianity should bring
freedom and the rights of civilized men. Granville Sharp was one of those who
fought for the abolition of slavery. He defended the case of African slaves
before the courts, notably in 1772, that of James Summenet, who had run
away and been recaptured. Lord Mansfield, who ruled on the case, did not
abolish slavery but did rule that a slave-master could not legally compel a slave
to accompany him abroad - a decision that signalled the beginning of the
erosion of slavery in England.
In France, people began to pay attention to the African presence in the
fifteenth century. This was the period when French sailon frequented various
parts of the West African coast, bringing back Africans to France, fint as
evidence of their voyages and later for sale as slaves.
The development of slavery in France was not originally intended. Indeed, a
royal court in 1571 proclaimed that: 'France, mother of freedom, admits of no
So
O
-a
c
So
Ü
Ü
Ü
5*
iQ
^go
"a K<3
Si
P
slave.' However, practices varied from case to case: some Africans were en¬
slaved while othen remained nominally free in a hostile environment. Several
observers noted the African presence in such cities as Angers, Lyon, Orléans,
Nantes and Paris. They -were active as servants, domestics and even as pages in
noble households. Others were soldiers, such as those in the Saxe-Volontaires
Regiment which was made up of people from Guinea, Congo and Madagascar;
they won acclaim in several battles in Europe in the eighteenth century. The
most distinguished black in the French military was General Alexandre Davy
Dumas (1762 born of a French father and an African mother, whose
son and grandson won fame in the world ofthe arts.
From the end of the seventeenth century, Africans were arriving in France
in significant numbers and, during the eighteenth, royal policy permitted
French slave-ownen in the Americas to bring their slaves to France.
Much less has been written about the African presence in other parts of
Europe. It is known, however, that a number of Africans, especially Ethiopian
envoys and pilgrims, visited Europe during the late Middle Ages and other
Africans were living in Venice, the Vatican and neighbouring cities. Venetians
were both slave-traden and slave-owners. Although most slaves seem to have
been of European and Asian origin, some were Africans. Documentation is too
scanty to give a picture of the lives of Africans in Venice and neighbouring
areas. Some reports assert that they became absorbed into local families, which
would explain their virtual disappearance by the late eighteenth century. All
had to be baptized, and that may have helped to alleviate their lot.
The most dramatic effects ofthe African diaspora occurred on the other side
of the Adantic Ocean, in the Americas. For most of the fifteenth century the
slave trade remained limited to the Caribbean and Central and South America.
It was associated with the development of Portuguese plantations in Brazil and
Dutch ones in the Guyanas. The sixteenth-century phase ofthe trade coincided
with African participation in the exploration of the Americas. Two hundred
Africans accompanied Alvarado's expedition to Quito, and othen joined
Pizarro's expedition to Peru. The best known of them was Estevanico, who
played an important part in the Spanish exploration of New Mexico and
Arizona. Africans also participated in the French expeditions to Canada and in
the conquest ofthe Mississippi valley.
In 1619 a Dutch vessel brought 20 'negan' to Jamestown as indentured
servants. This introduction of African labour initiated a demand for black
workers and various practices were introduced to restrict their freedom. This
culminated in the official establishment of slavery in the English North
American colonies in 1660 and, by the end ofthe century, Africans were legally
relegated to the position of slaves, property to be disposed of as their mästen saw
fit without regard to the fact that they were human beings.
Meanwhile, England and Spain were fighting for dominance in the Carib¬
bean. The sugar plantations demanded a big labour force and large numben of
slaves were imported from the Gold Coast, Angola, Congo, Nigeria, Dahomey
and, by the 1690s, Madagascar.
By the last decades of the seventeenth century, the English and the French
had secured their dominance in the Caribbean and slavery developed in
Jamaica and Barbados, providing a model for North America. In addition, the
Caribbean 'seasoning' centres, where African slaves were 'broken in', came into
being. The rebellions of these 'seasoned' slaves also served as models for slave
resistance in North America.
Areas with a high density of black slaves, such as British Guyana, Jamaica,
Brazil and San Domingo (Haiti), were the areas which experienced the most
frequent and most serious conspiracies and revolts. The first was the Maroon
War which broke out in Jamaica in 1725, when bands of slaves fled into the
mountains and established their own communities there. Guyana suffered
several serious revolts during the eighteenth century. In the 1740s black resis¬
tance led the Dutch to sign a treaty of friendship with the Coromante leader
Adoe. The 1760s were marked by the great rebellion of 1763-4 by Africans
and creóles led by Cuffy.
But it was in Brazil that the African armed struggle reached unprecedented
proportions and lasted longest, at least in the seventeenth century. In the state
of Palmares, an autonomous community of Africans estimated to number
20,000, chiefly Bantu from the CongoAngola region, survived from 1605 to
1695. They sought to model their society on that of their homeland and
resisted both the Dutch and the Portuguese before being finally overcome in
1695.
These freedom struggles illustrate the awakening of African nationalism

among the slaves in the Caribbean and Latin America. In these struggles African
religions such as obeah and vodum were powerful organizing tools. Islam played
a similar role, especially in Bahia where it helped to bring the Hausa and
Yoruba together.
In North America too, during the same period, Africans fomented plots and
launched several insurrections. Most plantations were situated far away from
terrain suitable to rebellious activity, such as the mountains of Jamaica or the
jungles of Guyana. But many slaves from the southern colonies of North
America fled to live among the Indians with whom they later raided neigh¬
bouring plantations. Insurrections also occurred in Virginia and Maryland in
the early eighteenth century as soon as it became clear that the indentured
system was going to be replaced by life enslavement.
In 1730 slave conspiracies were uncovered in the three colonies of Virginia,
Carolina and Louisiana. In 1731 a mutiny broke out on board a slave ship lying
off Rhode Island and, four yean later, slaves on board the slaver Dolphin
destroyed themselves and their captors. The most serious revolt was in 1739 in
South Carolina, just as Cudjoe was holding off the British army in the
mountains ofJamaica. It is known as Cato's Rebellion.
Similar troubles occurred in the northern colonies of North America, where
there were fewer than 3,000 Africans among almost six times as many whites
and there were no plantations. In 1712 a group led by a Gold Coast African
attempted to burn down New York City. A similar attempt was made in
Boston in 1723. In 1741 Africans again attempted to burn down New York,
and this incident was widely publicized. By 1772 it was being suggested in the
American colonies that all free blacks be deported to Africa or the West Indies,
they being regarded as the instigators of resistance. Hanging and brutal
treatment were meted out to those caught engaging in revolutionary activities.
Between 1750 and 1775 events that would decide the fate of Africans were
moving to a climax in both North America and the Caribbean. As Bntish

supremacy was becoming more and more firmly established, the anti-slavery
movement began to take off in Britain. In 1772 Lord Mansfield made his
famous decision that it was illegal to hold an individual in slavery within the
British Isles. In the American colonies a movement was developing among
whites in favour of political independence from the British crown.
In 1776 the North American colonies proclaimed their Declaration of
Independence. Literate Africans, whether slave or free, joined Europeans in
claiming the right to freedom. Some fought side by side with whites against the
British. A group of coloured volunteen came from the French colony of San
Domingo to support the colonists who were fighting the British at Savannah in
Georgia. In North America, the struggle for freedom had begun: the Europeans
were seeking freedom from British rule, while the Africans wanted freedom
from both the British and the American colonists.
The Africans in the American diaspora were an integral part of a world

dominated by European powers where powerful economic and intellectual
forces were reshaping political and social structures. Faced with these forces,
Africans were divided: some were convinced that salvation lay in the assimila¬
tion of European values; others sought to assert their Afncanness and were
ready to risk their lives in protest and resistance against European repression.
The Sierra Leone experiment reflected the convictions ofthe former group.
In London there was a sizeable black community and the abolitionists had the
idea of resettling liberated Africans in Africa so that they might found there a
society based on free labour which would spread Christianity, develop a
Western-style economy and contribute to the abolition of slavery. Thus, in
1787, over 400 liberated Africans were sent from England and setded in Sierra
Leone. This was the first attempt to repatriate former African slaves. It inspired
a self-financed effort by Africans organized under the leadership of Paul Cuffee,
in the United States, with the aim of reuniting blacks wanting to return to
Africa. Paul Cuffee's efforts resulted in the repatriation of 38 blacks to Africa in
1814, and his example was to inspire future generations.
While the Europeans in the USA were establishing institutions as expressions
of their culture and their independence, the Africans also established a number
of institutions. Although they had litde on which to base their ethnic identity,
since they were forbidden to speak African languages or practise their religions,
they began calling themselves 'Africans'. In 1787 an African Methodist minister
in the United States, Richard Allen, in protest against the segregation policy in
his own church, broke away and formed the Free Afincan Society, which had
both religious and social objectives. Africans adopted the same approach in
other parts of the United States. At the same time, a Barbadian-born preacher,
Pnnce Hall, who had been initiated as a Mason by some British soldiers during
the American War of Independence, tried to secure the right to found a lodge
for free black men. The Masons of the Scottish Right granted him the charter
to organize an African lodge of which he became the grand master. This was
the first Western-style fraternal organization among black men. The lodges and
religious organizations were to be Afro-Americans' most powerful institutions
in the nineteenth century, since they bound blacks together nationally.
While these organizations identified themselves with African traditions and
culture, they were also carriers of Western values such as thrift, puritan
theology, penonal advancement through work, and education and concern for
the less fortunate. Such was the motivation of George Liele, for example, who
founded Baptist churches in the United States and Jamaica in the late
eighteenth century. These pioneering innovations among American blacks
would later help provide the basis for an evolving community identity in the
United States, the Caribbean and Africa.
Most Afro-American leaders of this period were self-educated. Phyllis
Wheatley, born in Africa around 1753, became a renowned poet; Gustavus
Vassa, born in Benin in 1745 and taken to America and later England, was
active in the anti-slavery movement and wrote his memoin (Tlie Interesting
Narrative ofthe Life of Olaudah Equiano, or Gustavus Vassa). Benjamin Banneker,
nicknamed 'the Ethiopian', became a noted mathematician and astronomer.
Africans living in Europe also contributed to the struggle for freedom and
human dignity. After studying at the Universities of Halle and Wittenberg,
Anton Azmo returned to help his compatriots in the Gold Coast; Ottobah
Cagoano, who was freed by the Mansfield decision, wrote Thoughts and
Sentiments on the Evil and Wicked Traffic ofthe Slavery and Commerce ofthe Human
Species. Ignatius Sancho's posthumously published letters confirm him as an
important spokesman for Africans abroad. In Europe, as in the United States,
African proteston found allies among the whites, such as the Paris-based
Société des Amis des Noirs.
It was in America, however, that the freedom struggle took on a real inter¬
national dimension with the events on the island of San Domingo.
When a revolution broke out in France in 1789 to the cry of 'Liberty!
Equality! Fraternity!', the movement shook the French colony of San Domingo,
covered with prosperous sugar plantations, where half a million slaves and
24,000 free persons of colour lived under the domination of 32,000 French
settlers, known for the cruelty with which they treated their slaves. The free
African population demanded full equality with the whites. Then, in 1791, the
great mass of blacks began to stir under the leadenhip of an illiterate fieldhand,
Boukman, who bound his followers with voodoo ritual and African-style secret
oaths to rise against their masters. The revolutionary government in Paris
dispatched an army to restore order. It was at this point that a quite remarkable
figure appeared on the political scene, a literate, Christian slave, born in the
diaspora of an African father Toussaint, who took the name L'Ouverture.
5.2 Toussaint L'Ouverture of Haiti (The Moorland-Springarn Research Center, Howard

University, Washington DC)
Toussaint called for guerrilla action to support his small army and, within
five yean, had defeated Napoleon's forces, with assistance from yellow fever.
He restored order and prosperity to Haiti and was acclaimed throughout the
world for his military ability, administrative skills, humanity and statesmanship.
Haiti and Toussaint L'Ouverture became symbols that gave inspiration to blacks
aspiring to freedom and even independence in other parts of the Americas and
the Caribbean.
At the beginning of the nineteenth century, a black preacher, Gabriel Presser,

organized a march on Richmond, in Virginia, at the head of a thousand slaves.
They hoped to use this demonstration to secure their freedom, but the
governor called out the militia. Many Africans, including Presser, were arrested
and executed; but the example of Haiti remained very much alive.
With the nineteenth century, a new era dawned for Africans living in
Europe and the Americas. It is true that the various legal systems in both
Europe and America did not grant blacks either equality or real freedom, but
there were instances of greater flexibility. They learned to read and write, even
where the laws forbade it; there were inter-racial marriages, despite the fact that
they were repugnant to most people; and travel facilitated the development of
networks. Nevertheless, Africans in the diaspora realized that their deprived
status stemmed from the fact that they belonged to a different race and a
different culture, whence their efforts to save their racial brothers and their
heritage. This psychological unity continued to prevail, and indeed became a
source of strength among the African peoples culminating in the pan-African
movement ofthe nineteenth and twentieth centuries.
Africans in the diaspora, however, could not escape the influence of the
environment to which they had been transplanted nor several centuries of
inculcation with Euro-American culture. Their language and life-style changed,
and so did their values and goals. Africans in the diaspora in Europe and
America thus became cultural intermediaries between indigenous Africans and
Euro-Americans.
Asia
While the forced migration of Africans to Europe and the Americas is a

relatively recent chapter in world history, the trade in slaves to Asia constitutes
a much older historical phenomenon. The Periplus of the Erythraean Sea, written
in c.50, refers to slaves taken from the Horn of Africa and it seems likely that
trade between peoples on both sides of the Red Sea goes back to prehistoric
times.
From the seventh century, that is with the birth of Islam, a kind of cultural
unity evolved throughout the Indian Ocean and Red Sea area. Several coastal
towns in East Africa became Islamized and Muslims increasingly dominated
trade, including the slave trade.
For many centuries Turkey and neighbouring countries were the major
entrepôts for slaves from Tripoli and Benghazi, and also a transit zone for the
5.3 Africa, Europe and Asia (adaptedfrom map drawn by Dr Dulal C. Goswami, Geology
and Geography Department, Howard University, Washington DC)
inland traffic. The purchase in Constantinople in 1696 of several African boys

for Emperor Peter the Great of Russia deserves mention since one of them was
Abram Petrovich Hannibal, the great-grandfather of Aleksander Pushkin.
Hannibal appears to have been born in Ethiopia where he was captured by
Turks. How many other African slaves arrived in Russia and neighbouring
areas via Turkey is not known, but the numbers were probably small. Since
slavery was abolished in Russia during the first quarter of the eighteenth
century, Africans in that country were probably servants rather than slaves.
Most slaves imported into Asia were children, and the majority were girls.
From the East African ports, slaves were normally taken to the Arabian port of
Mocha on the Red Sea. From there many were sent on to Hodeida, Jeddah,
Mecca and other entrepôts in Arabia. Othen were shipped to Persian Gulf
ports. Indian ports such as Bombay, Goa, Surat, Karikal, Pondicherry and
Calcutta usually received their shipments from Mocha or the Persian Gulf, but
some came direcdy from East Africa.
In Arabia, Oman held the key position in the naval and commercial strategy
ofthe Middle East. Its capital, Muscat, commanded the approach to the Penian
Gulf, through which large numbers of African slaves were transported. Omani
Arabs captured the East African ports of Kilwa and Zanzibar in 1784 and 1785
respectively; from then on they claimed sovereignty over several parts of the
coast and procured slaves to harvest cloves and coconuts on Arab-owned
plantations in the region.
In Yemen and Hadramawt, black people of Ethiopian origin had been living
since ancient times. In Aden, these blacks formed a group comparable with the
caste of 'untouchables' in India. In several other parts of southern Arabia, black
slaves from Africa served in the armies of local sultans; they were also to be
found as concubines and domestic servants, eunuchs, crewmen, stevedores and
agricultural labourers working in the salt marshes and sugar and date
plantations.
Africans were scattered on many Indian Ocean islands. The Dutch procured
slaves in East Africa and Madagascar and took them to Indonesia. The French
and the British setded East Africans as slaves on the Mascarene Islands of
Bourbon and Mauritius. The expansion of the slave trade in the nineteenth
century favoured the development of African communities in those islands,
but, even before this period, there had emerged a community of creóles whose
influence would be felt all through the nineteenth and twentieth centuries.
There were far more African slaves in South Asia than in other parts of Asia,
probably because trade with Africa was older and busier. The Muslims con¬
trolled the west coast of India and Indians had setded in East Africa. There
were many African slaves in India as early as the thirteenth century.
The second half of the fifteenth century witnessed the assertion of an African
presence in one part of northern India, Bengal. Africans had migrated there
from the coastal region of Calcutta, the area of Dakha and several inland regions.
The ruler of Bengal, Rukn-ud-dîn-Barbak (1459-74), used to promote loyal
Africans to responsible military and administrative posts. On the death of
Barbak in 1486, Shähzäda. eunuch commander of the palace guards, seized
power and assumed the tide of Barbak Shah. He was murdered and replaced by
another African, who had remained loyal to the former ruler. Africans remained
in power until 1493 and made their mark during this short period.
In Gujarat Africans served in the army from at least the thirteenth century
and possibly earlier. In 1537 the city of Ahmadäbäd was employing as many as
5,000 Africans in government service. Several of these Africans in Gujarat
distinguished themselves during the sixteenth century. Some were important
military commanders: Ikhtiyar-ul-Mulk. for example, organized an army of
some 20,000 men to challenge the troops of the Mughul Emperor Akbar.
Ikhtiyar was defeated, but he won Akbar's respect.
An African, a former slave, Shaykh Sayyid al-Habshî Sultani, served as a
soldier in Jhujhar Khän's army. When he retired, he purchased some land,
made the pilgrimage to Mecca and then worked his lands, which enabled him
to feed hundreds of poor people daily. He founded a library which attracted
many scholars. In 1573 Sayyid (Sa'ïd) constructed a mosque in Ahmadäbäd
famous for its simple design: a roof of arches and beautiful arched perforated
windows with exquisite tracery and floral patterns.
Another African, Sïdï Bashïr, built a noted mosque in Ahmadäbäd, which is

unique in that it possesses two 'shaking' minarets. When one minaret is shaken,
the vibration is carried to the other.
Not far from Gujarat is Janjira Island which, in ancient times, was part ofthe
prosperous commercial activity of north-western India, including what is today
Bombay, and the Konkan coast. If local traditions are to be believed, the Siddis
ofJanjira descended from Africans who arrived from Gujarat in 1489.
From the 1530s the Portuguese developed political and economic control
over parts of the west coast of India, especially the Konkan coast where many
African slaves were imported. Most of these slaves came from Mozambique and
the Portuguese used them in trading, on farms or as domestics. Some were
trained as priests and teachers for religious schools, especially in Goa, which
became Portugal's headquarters for its East African and Asian colonies.
Throughout this period, Janjira Island retained its autonomy. By the
seventeenth century the Siddis, who comprised the bulk of the Muslims on the
island, had become its principal land owners. A council of elders of the major
Siddi leaders chose a nawab who acted as temporal and spiritual ruler. Having
established their political base, the Janjira Siddis expanded their power over the
north-western coast thanks to their naval power. In 1616 they formed an
alliance with Malik Ámbar, a Habsfñ king ofthe Deccan, in central India. Both
forces fought the Mughuls for many years.
During the seventeenth century, the British East India Company made
repeated efforts to negotiate an alliance with the Siddis, who ruled over the
Konkan coast. The Siddis continued to exercise independent power in the area,
and it was not until 1759 that the British were able to curb their power. Janjira
was only brought under direct British control in the nineteenth century.
The Janjira Siddis played a considerable role in Indian history. There can be
no doubt that their religion (Islam) and their maritime and military skills were
major assets for them. It is of great importance that this small group of African
migrants had such an impact on the policies and actions of Great Britain,
Portugal and the Netherlands, as well as on local Indian states.
Africans also setded along parts of the Malabar coast during the seventeenth
and eighteenth centuries. These were black Jews, descendants of African slaves,
who had left Cochin and Kerala in southern India, to come and settle on that
coast. Most became domestics and intermarried with local inhabitants and other
Jews. The Portuguese were responsible for settling African slaves along parts of
the Malabar coast, especially around Goa, which became one of its head¬
quarters in the sixteenth century.
The Italian scholar and traveller, Pietro della Valle, reported that blacks from
East and West Africa (Guineans and Mozambicans) were sent by sea to the
Portuguese territories. The Portuguese made their black slaves domestics,
bearen and guards for their escorts. The women were often taken as mistresses.
Another part of India, the Deccan, witnessed a dramatic rise to power, that
of the African Malik Ambar, an Ethiopian who had been sold several times as a
slave before reaching India. Ambar's historical significance lies in the fact that
5.4 Malik Ambar, an African king in India in the seventeenth century

(Tlie Ross-Coomaraswamy Collection, courtesy ofthe Museum of Fine Arts, Boston, MA)
he became the commander of a powerful army fighting under the banner of

the king of Ahmadnagär and resisted several attacks by the Mughuls, thus
preventing the Mughul rulers from conquering the Deccan in his lifetime. He
ruled supreme over the area around Ahmadnagär, where several thousand
Africans lived, from 1602 to 1626. He founded towns, constructed canals and
irrigation schemes, stimulated trade with Asia and Europe and attracted scholars
and poets to his court.
Malik Ambar's activities confirm two significant points: first, that individual
Africans played influential roles in Indian history; and second, that Africans
managed to win support and respect from Indians while retaining a sense of
their own identity.
Conclusion
The history of African people has been so much influenced by perceptions

stemming from the slave trade that it is necessary to emphasize that Africans of
their own volition migrated between continents, as merchants, clerics, seamen,
adventuren and so on. This presence of free blacks abroad deserves further
study to provide a more rounded and realistic picture of world civilization.
It was, however, the intercontinental slave trade which, more than anything
else, established a world black presence. It was the nature of this trade and its
consequences which, especially in the Americas and the Caribbean, caused
Africans to organize freedom struggles. This process began at the beginning of
the modern era and took a quite different turn when in around 1800 Toussaint
L'Ouverture emerged as the international symbol of black liberation. The
liberation process continued, despite colonial rule, and may indeed be the
greatest historical consequence ofthe African diaspora.
The Ottoman conquest
of Egypt
At the beginning of the sixteenth century a new Islamic great power entered
the history of Africa and Arabia: the Ottoman empire, which had come into
being in Asia Minor and the Balkan peninsula and went on to subjugate a large
part ofthe Arab countries in West Asia and North Africa.
It was in 1516 that Marnluk sultanates in Syria and Egypt collapsed under
the attack by the Ottoman army. EgyptianOttoman relations had been
strained since the 1480s, but in 1514, following Sultan Selim I's first successful
campaign to put an end to the power of Penia, the rulers of Egypt took a firm
stand against the Ottomans. Selim decided to break the MarnlukSafavid
coalition once and for all.
In August 1516 the Ottoman army routed the Marnluk troops north of
Halab; it owed its victory largely to its technical superiority but it was also
helped by betrayal by a pro-Ottoman faction in the Marnluk camp. Sultan
Selim occupied Damascus without encountering much resistance and took all
of Syria and Palestine as far as the Sinai Desert. But the Sultan's advison and the
Marnluk amirs who had gone over to the Ottoman side persuaded Selim to
crown his campaign with the conquest of Egypt. Selim came within sight of
Cairo without much difficulty and routed the Marnluk troops in a short battle
in January 1517. That marked the end ofthe Marnluk sultanate.
Military superiority was not the only reason for the Ottoman victory over
the Mamluks. The main reason lay in the differences in the economic and
political situations of the two states. In Europe and Asia Minor the Ottoman
empire dominated territories that were economically advanced, and had mines
and raw materials as well as production centres connected to international
trade. The Marnluk states, on the other hand, had almost no mineral resources
and depended virtually exclusively on agriculture and the international transit
trade, whose main branch, the trade in eastern spices, had fallen into the hands
of the Portuguese. Imports of precious metals from Africa were also declining.
For many years the Mamluks had been trying to compensate for this loss of
profit by drawing on their economic reserves, confiscating land and increasing
taxes, measures which had intensified the indigenous inhabitants' hatred of the
foreign caste of Marnluk exploiten.
70
The Ottoman conquest of Egypt 71
Selim's victory had far-reaching consequences for West Asia and North
Africa. The conquest of Egypt in particular impelled the Ottoman empire to
control the Mediterranean sea routes and attempt the conquest of other Arab
countries in North Africa.
The territories conquered from the Mamluks included regions of great

economic, political and strategic importance. Egypt was particularly important
because of its intensive agriculture, its large population and its location on the
Red Sea. In addition, the prestige of the Ottoman sultans was enhanced by the
fact that they became guardians of the two holy cities of Islam, Mecca and
Medina.
Before leaving Egypt in September 1517, Selim placed Khäyr Bey, the
Marnluk defector, at the head of the province of Egypt. He governed Egypt
more as the sultan's vassal than as a provincial administrator and remained in
office until his death in 1522. He retained his Marnluk tide Malik al-umarä
(King of the Commanders) and conducted his court according to Marnluk

ceremonial. He surrounded himself with former Marnluk dignitaries willing to
serve the new regime, entrusting them with administrative tasks and also
political and military assignments. The organization of justice continued to be
based on the system of four Chief Judges, one for each School of Law. Selim
left Khäyr Bey to control the province's income, and the regular presents sent
by Khäyr Bey to Istanbul were expressions of his personal allegiance to the
Sultan rather than obligatory deliveries to the state treasury.
Khäyr Bey worked hard to secure his control ofthe province's resources. He
abolished the ikta1 (extant feudal tenures) and incorporated them into the state
property, their former holden being allotted fixed wages paid from the
provincial treasury. He caused a complete cadastral survey to be made and
revised the land tenure system, which enabled him to discharge the task en¬
trusted to him of supplying the two holy cities of Hidjaz with corn.
Khäyr Bey's powers, however, were not unlimited. Sultan Selim had left a
strong body of troops in Egypt to consolidate Ottoman authority and, if
necessary, deter any attempt by the Mamluks to seize back power. These troops
were made up of two infantry corps (one of Janissaries and one of Azabs) and
two cavalry ones. In order to maintain his authority over these somewhat un¬
disciplined units, Khäyr Bey formed his own cavalry corps, recruited from
among former as well as new Mamluks.
When Khäyr Bey died in 1522, the process of incorporating the city of
Cairo was well under way, while the state of which it had been the centre had
been reduced, after centuries of independence, to the rank of a mere Ottoman
province.
Ottoman administration and conflicts within the ruling class

The beginning of the new period was marked by a revolt against Ottoman
supremacy, launched by the new vice-regent, Ahmad Pasha, in 1524. He pro¬
claimed himself Sultan of Egypt and started minting his own coins. He
6.2 The Janissaries, as portrayed in a sixteenth-century Turkish miniature

(Topkapi Museum, Istanbul, HI524, p. 271A, © Sonia Halliday Photographs)
managed to win the support of a few Arab groups and took the citadel of
Cairo, but was subsequendy driven out of the city after being betrayed by the
Bedouin chiefs who had initially supported him.
In the light of these events, it became obvious that the Istanbul government
had to clarify the precise status of Egypt within the Ottoman empire. This was
done by issuing an edict, the KänTtn Name, which regulated the political,
military, civil and economic life of Egypt. By the terms of this edict, the vice-
regent, the wäll, who had always held the rank oí pasha, was accorded certain
privileges based on customs in use at the court in Istanbul. As his seat he was
given the Cairo citadel, the seat ofthe former sultans, and he was entitled to his
own penonal guard, called the muteferrikas , like the Istanbul sultan's own
personal guard.
The Kânïtn Name adopted a number of Marnluk administrative customs in the
area of financial and civil administration. Egypt was divided into fourteen
districts each headed by a käshif whose task it was to maintain the irrigation
system in good working order and levy taxes from the farmen. The region
south of Asyut was left under the Arab shaykhs ofthe Hawwira group who had
gained control ofthe area after 1517. The Känün Name also enacted the creation
ofthe odjaks, imperial troops introduced into Egypt by Sultan Selim I. In reward
for their loyalty during the revolt, the Janissaries became the main prop of the
Sultan's power. Their commander, the agha, was one of the most prominent
members of the council of state and, in addition, the mint producing Egyptian
currency was situated in the area of their barracks inside the citadel.
All the troops stationed in Egypt received regular pay; the highest digni¬
taries, including the vice-regent, the Chief Judge and 24 other senior figures
with the tide of sandjak bey, also received salaries, säliyäne, paid out of the
Egyptian treasury.
These salaries were paid out of the land tax, taxes and duties levied on
productive activities in the towns and cities and the head tax to which fol-
lowen of religions other than Islam were liable. The land tax was levied by the
administrators of each region and the taxes on urban residents by tax collectors,
käshifs or emins, who were paid a fixed annual salary. The Arab shaykhs in the
region south of Asyut could, in accordance with the iltizam system described
below, raise the tax in their respective areas and retain what it produced,
provided that they guaranteed agricultural work and paid a fixed percentage of
their receipts to the Egyptian treasury.
From 1525 the surplus remaining after covering expenses was sent each year
to Istanbul instead of Hidjäz. The regular remittance of this treasure (the
hazine) was one ofthe vice-regent's most important tasks. Egypt also fulfilled its
obligations to Mecca and Medina.
The Kämm Name also made it legal to levy duties on individual holdings of
tax-farms known as iltizams. At first this system only applied to agricultural
holdings, but in the coune of the sixteenth century it spread more widely, as
the odjaks and their officen grew more powerful, at the expense of the repre¬
sentatives of Ottoman rule. In such conditions, tax collectors found themselves
reduced to being mere officials with no direct influence on tax collection or

any other aspect of fiscal policy.
Towards the end of the sixteenth century two levels of political power
crystallized. One consisted of those officially holding power, appointed by Istan¬
bul, headed by the vice-regent, and the other consisted of the odjaks and their
officen, and the sandjak beys and their Mamluks. Although unofficial, this group
had both economic and military power, and from the late sixteenth century
onwards its significance in Egypt's political life became more and more apparent.
Ottoman influence in north-east Africa

Throughout the sixteenth century Egypt played an important role in the
expansionist policy ofthe Ottoman empire. Because of its geographical position
it formed a natural link between the Mediterranean Sea and the Indian Ocean,
with the trade route bringing eastern goods to Europe passing through the Red
Sea and Egypt itself. By 1517, the trade with the east was in the hands ofthe
Portuguese. Once Egypt had been subjugated, therefore, the Ottomans faced
the task of eliminating the Portuguese from the Indian Ocean or at least
hindering their penetration into the Red Sea. The occasion to intervene was
provided when the ruler of Gujarat solicited Sultan Sulaymän for help against
the Portuguese who had occupied the port of Diu. Sulaymän Pasha, the vice-
regent of Egypt, was entrusted with the campaign. On his way to Diu, in 1538,
he seized Aden, in Yemen, and established Ottoman rule there. Although he
failed to dislodge the Portuguese from Diu, Sulaymän Pasha landed in Upper
Egypt and, after reaching Aswan, continued his march along the Nile as far as
northern Nubia.
It was difficult to maintain authority in this area of Upper Egypt dominated

by the Hawwära shaykhs. In the middle of the sixteenth century Ottoman
supremacy had to be re-established there. An expedition headed by the former
vice-regent of Yemen, Özdemir Pasha, reoccupied Ibrim, Aswan and Say and a
new province, Berberistan, was created and placed under the authority of a
käshif directly subordinate to the vice-regent in Cairo.
In 1557 Özdemir Pasha took Massawa on the Red Sea as well as Zaylä',
opposite Aden, and he conquered part ofthe inland region ruled by the king of
Ethiopia. Massawa became the centre of Habesh, another Ottoman province in
Africa, which played an important role in the Ottoman defence ofthe Red Sea
against the Portuguese. The creation of these Red Sea posts was important to
the continuation of the transit trade in which coffee had taken the place of the
traditional spices from the east. Because ofthe new popularity of coffee, Egypt
reaped considerable profits from this transit trade.
The period from 1517 to the 1580s marked the zenith of Ottoman power in
Egypt. No symptoms of Istanbul's decline had yet appeared; on the contrary, its
power was even further consolidated. In 1583 the province of Upper Egypt
was created, and the Hawwära chiefs were replaced as administraton by an
official sent from Cairo.
MEDITERRANEAN SEA
: Alexandria
Kharga
150 miles
Aswan <
6.3 Middle and Upper Egypt (al-Saïd) (adaptedfrom {Abd al-Rahim (Abd al-Rahman and
Wataru Miki, Village in Ottoman Egypt and Tokugawa Japan A Comparative
Study, Institute for the Study of Languages and Cultures of Asia and Africa, Tokyo, p. (ii)
(map published by kind permission of Professor Wataru Miki, Keio University, Tokyo))
Traditionally, Upper Egypt had occupied a distinctive place within the

political organization of the Nile valley states. It was different from the rest of
Egypt in its geography, its social systems, its politics, and its ethnic and religious
particularities. On numerous occasions it had been the refuge of movements
directed against lawfully constituted authority. It commanded trade routes and
the traffic along the Nile. The Ottomans' interest in this territory was
demonstrated in the special status it was accorded within the province of Egypt.
Thus, the administrator of Upper Egypt, like the administrator of all Egypt, had
the tide of vice-regent and his political importance was emphasized by the
considerable number of troops at his disposal.
Internal military conflicts

The period of peace that followed the suppression of Ahmad Pasha's rebellion
in 1524 lasted some 60 yean. In the 1580s, as a result of its socio-economic
dependence on the Ottoman empire, Egypt began to experience the first

economic difficulties, which sparked fierce conflicts within the ruling military
caste.
Between 1598 and 1609 a number of revolts broke out, led by the spahi
cavalry corps, who were badly hit by inflation. The subsequent years saw the
growth in importance of the 24 sandjak beys who came to form a group in some
ways comparable to the great amirs of the old Marnluk sultanate. Their high
rank entided them to hold the most senior political offices. Until about 1620
they held the office of Serdar, commander of the troops operating within or
beyond Egypt's borden. They had command of the units escorting the
Egyptian hazine sent to Istanbul and of the troops entrusted with protecting the
pilgrim caravans travelling to Mecca. At the beginning of the seventeenth
century one of them was chosen as KäHrn Makam (acting vice-regent).
Their first significant political coup was the removal in 1631 of vice-regent
Müsä Pasha and his replacement by their Kä'im Makam. Despite the protests of
the vice-regent, the Sublime Porte endoned their decision, thereby creating a
precedent that would be repeated several times in the yean to come. The beys
and the odjaks had got rid of Müsa Pasha because he had tried to put an end to
the increasingly common practice of soldien illegally taxing the urban residents
in the form of alleged protection money, himlye. Originally this had been
caution money protecting cities from looting, but it had become a means by
which the soldiers enriched themselves by exploiting the urban populations.
Müsä Pasha was replaced by Ridwän Bey al-Fäkäri who was vice-regent
until his death in 1656. He was the leader of the Fakärlya, political group
which was opposed by the Kâsïmïya. These two groups consisted of beys and
their Marnluk retinues and they sought followers among the nomads and the
urban dwellers. Their struggle also involved the odjaks, the Janissaries
supporting the Fakärlya and the Azabs the Kâsïmïya, with the result that by the
end ofthe seventeenth century Egyptian society was completely divided.
Ridwän Bey secured his position by obtaining from the Sultan a decree
appointing him life commander of the pilgrim caravan while his ally (Alï Bey
was appointed life vice-regent of Upper Egypt. The Fakärlya was thus well-
placed politically, with its two leading figures firmly entrenched at the top of
the hierarchy. After Ridwän Bey's death, however, disagreements within the
Fakärlya led to the collapse ofthe party.
The Kâsïmïya succeeded in gaining, briefly, the offices previously held by the
Fäkäriya beys, but in 1662 the head of the Kâsïmïya was assassinated on the
orders of the vice-regent.
Yet in 1661 the Sublime Porte had managed to suspend all non-military
personnel in the seven odjaks and impose a 20 per cent increase in the
multazims payments from their benefices to cover the treasury deficit caused by
inflation. Despite the discontent, the Istanbul court decided to introduce further
measures. In 1670 the vice-regent, Kara Ibrahim Pasha, was instructed to carry
out fiscal reform in Egypt with the help of the army. After a radical rearrange¬
ment ofthe budget and four years of negotiations with the interested parties, he
doubled the sum ofthe treasury. The principles of this reform became the basis
for solving fiscal problems all through the period from this time until 1798. But
the results achieved were short-lived because the decision to weaken the
influence of the odjaks had come too late. The weakening of the pasha's
position during the fint half of the seventeenth century and the elimination of
the beys' parties in the 1 660s had enabled the odjaks to seize the benefices arising
from exploitation ofthe urban population. The beys, meanwhile, continued to
exploit the rural population.
Between 1670 and 1750, political life became even more complicated. It
was marked not only by dissensions among the odjaks but also by conflicts
generated by party affiliation, with some supporting the Fakârïya and othen the
Käsimiya. The most contentious issues were the two main elements of political
and economic power: the benefices held and their redistribution on the one
hand, and on the other illegal takings. The most strife-ridden odjak was that of
the Janissaries which controlled the most lucrative benefices.
For almost 20 yean the main protagonist in the struggle within the
Janissaries' odjak was the Janissary Bashodabashi Küchük Mehmed. He had been
expelled from Egypt in i 680 and when he returned he won the support ofthe
Agha who introduced him to the Fäkäriya. With the help of the leader of this
party, Küchük Mehmed succeeded in ridding the Janissaries' odjak of its Kasimi
commanders and securing his own domination of it from 1692. Shortly after,
he secured approval by the top officials of all seven odjaks of his decision to
abolish some of the protection charges and all the other illegal charges
demanded by the Janissaries and the Azabs. The main reason for this new
coune was the fear of a repetition ofthe riots of 1678 which had been caused
by inflation and a sharp rise in the price of corn.
During the summer of 1694, tension within the Janissaries' odjak reached a
critical level. At the same time, a low flood ofthe River Nile caused a sudden
rise in corn prices. True to his political principles, Küchük Mehmed moved
against speculators, fixing wheat prices and excluding middlemen from its
distribution. These measures brought the Janissary opposition and the Azabs
who controlled the grain stores together and Küchük Mehmed was assassinated.
Once again the odjaks had access to their lost benefices along with the freedom
to determine food prices. The subsequent rapid escalation in the prices of basic
commodities caused a famine in 1695. It was only the abundant floods of 1697
that put an end to this difficult period.
Since 1688 the state of crisis had been aggravated by the continuous erosion
ofthe value ofthe currency and, by 1692, the situation had reached a critical
point. The provincial administration tried to compensate for the shortfall in the
fiazine by an increase of 4 per cent in deliveries from the iltizams, but Istanbul
demanded that the hazine be delivered in undepreciated currency, a demand
which aroused strong opposition among the odjaks who were deriving huge
profits from the devaluation, as were their merchant clients. It was their
opposition that wrecked the attempt at financial reform and change in the tax
system prepared by the financial expert Yäsif al-Yahüdl in 1697.
In the early 1700s a new wave of price rises occurred, resulting from an
influx of valueless currency from other parts of the Ottoman empire. It was
decided that depreciated coins could only be used for their metal weight and
that silver coins of a higher fineness (fidda diwani) would be minted to replace
them. This decree helped to slow down the decline of the Egyptian currency
but the serious problem of protection charges remained unresolved, with the
odjaks quite unwilling to abandon them. So tension persisted, threatening to
erupt into a new crisis.
The crisis finally broke in about 1710, caused initially by the growth in
coffee exports. These earned the wholesalers and their Janissary protecton
profits that the other odjaks found intolerable. The dispute led to a serious clash
within the Janissaries' odjak in which the beys also participated.
The man at the centre of the affair was Bashodabashi Ifrandj Ahmad, who
was in favour of the Janissaries involving themselves in economic activities. He
was expelled from the odjak in 1707 but was readmitted with the backing of
the Fäkäriya and his position appeared secure. But a decree from Istanbul dealt
the Janissaries a serious blow: the decree put an end to all abuses and the
patronage relations between military protecton and non-military personnel. In
addition, the mint was to be moved from the Janissaries' barracks to a new
building. At the end of 1710 the crisis was further aggravated by a split
between the two leading Fakärlya beys. The delicate balance of power was
disturbed and the crisis degenerated into an armed conflict in which the beys
were openly engaged. In June 1711 the resistance of the Janissaries who were
holding out in the citadel collapsed, the Fäkäriya beys fled Egypt and Ifrandj
Ahmad was executed.
The effect of these events was to weaken the influence of the Janissaries and
the other odjaks, exhausted by 20 years of endless struggle. A new period now
began in which the beys once again took the limelight on the Egyptian political
scene.
Despite the economic and political crises of its final decades, the odjaks'
period of rule had witnessed great economic and social change in Egypt.
Agricultural production had increased as had trade in Egyptian goods, and the
transit trade, especially in coffee, contributed to general prosperity. The healthy
state ofthe economy was accompanied by rapid population growth. There was,
it is true, an ongoing problem of depreciation of the currency, aggravated by
Ottoman defeats in Europe, but Egypt was not overly affected by these wars in
which odjak participation was negligible. Under these conditions, the social
character of the odjaks had gradually changed. The soldien had begun to
integrate with the local people, mosdy through marriage; their discipline
declined and their military ability was no longer what it had been.
After 1711 the struggle for power continued both between political groups
and within each of them. In 1714 leading figures in the ruling Kâsimïya killed
Käytäs Bey and seized supreme authority (riyäsa). But in 1718 the Mamluks of
Abu Shanab formed an opposition group, the Shanabïya, which in co-operation
with the Fäkäriya beys regained power.
The never-ending struggles among the beys gave the vice-regents an oppor¬
tunity, after 1720, to strengthen their authority. The vice-regent, Bakir Pasha,
took advantage of the rivalry among the three leaders of the Fäkäriya to try to
divert both the lawful and the illicit taxes collected by the odjaks.into the state
treasury. To break their resistance, he had all three leaders assassinated. But the
new leadership of the Egyptian military caste, represented by 'Uthman Bey
and the Janissary Kähya Ibrahim Kazdoghlu, took the side of the odjaks. The
opposition was reinforced by representatives ofthe ulamâ, the religious brother¬
hoods.
With Ibrahim Kazdoghlu at the head of the Janissaries' odjak, the Kaz-
dughliya, a strong Fäkäriya faction which had previously remained in the back¬
ground, came to the fore. From 1743, after he had forced 'Uthman Bey to
leave the country, Ibrahim Kähya became the sole leader of the country's
military and theKazdughlïya ruled as the sole party until 1798.
Ibrahim Kähya joined forces with Ridwän, the Kähya of the Azabs. This
duumvirate ruled undisturbed for the years after 1744. Ridwän Kähya indulged
his passion for building, while Ibrahim Kähya turned Egypt into his personal
domain, using every means to enrich himself and building up a strong retinue
of Mamluks. Yet, despite the seeming stability of the duumvirate, symptoms
were developing which heralded the collapse of the socio-economic structures
on which the power ofthe military rested.
The underlying unrest became apparent soon after the death of Ibrahim
Kähya in 1754. Ridwän Kähya perished in a Janissary revolt precipitated by his
decision to impose a new tax on coffee. The odjaks were showing their deter¬
mination to protect their much eroded privileges, but they were also expressing
their hostility towards the Mamluks who, promoted to beys and odjak officers,
were now assuming political and economic power. This shift in power from
the odjaks to the Marnluk beys was reinforced after 1760, and the odjaks
degenerated into mere reserves for Marnluk retinues whose members were
allowed to hold the rank of officer.
Cultural development
The extinction of the sovereign Marnluk sultanate and the transformation of
Egypt into a province of the Ottoman empire had cultural repercussions.
Cultural activity in Ottoman Egypt as it developed bore very distinctly the im¬
print of the new social, economic and national relations. Egypt's new political
position also affected education and the role of Arabic as the medium of
cultural expression.
After the fall ofthe Marnluk sultanate, Cairo ceased to be the spiritual centre
of the Sunnite world, which it had been since 1261. Although Al-Azhar
remained one of the most prestigious schools for students from Arab countries
even after 1517, those who wanted to achieve success in public life found them¬
selves increasingly forced to attend schools in Istanbul.
The incorporation of Egypt and other Arab countries into the Ottoman
empire, in whose public and cultural life Ottoman-Turkish occupied a leading
position, marked a significant step in the decline of Arabic as mediator of Arab-

Islamic culture. The loss of political independence by countries with an Arab
population naturally signified the end of the use of Arabic as the language of
state administration.
As in all the Arab provinces ofthe Ottoman empire, the culture of Egypt fell
under strong Ottoman-Turkish political, social and cultural pressure. It
nevertheless retained a certain character of its own which was to become the
starting-point and regulator of the Arab national awakening in the nineteenth

and twentieth centuries. This special character was retained by Egyptian Arabic
literature of the sixteenth, seventeenth and eighteenth centuries but it contri¬
buted nothing new in the classical genres, although it did bring about a new
orientation of literary production. Writings in the Arabic language were
directed at readers from the middle strata of the urban population, hence the
strong popular character ofthe genres and language ofthat literature.
This trend was particularly pronounced in poetry. As the Dervish orden
became more widespread, mystic siifi poetry became popular through works
marked by the influence of Ibn al-Farid, who sang the praises of the Prophet
Muhammad. Satire was also introduced during this period, a remarkable
example being Hazz al-kuhüf (Shaking the Peasants' Heads) by al-Shirbïnî
(1787), in which the author ridiculed the stupidity of the peasants and the
narrow-mindedness of the itdamä). The language of this work is the Arabic
dialect of Egypt. Another noted poet was Hasan al-Badri al-Hidjazï (died 1719)
who, in addition to a collection of poems on religious themes, wrote a collec¬
tion of satires on contemporary society. Apart from this poetry addressed to a
wide public, formalistic poetry appeared with a tendency to self-conceited
eccentricity, whose representative in Egypt was Abdullah al-Idkâwï (died
1770).
Narrative folk prose was enriched by a novel on the conquest of Egypt by
Selim I, in which the hero is the last Marnluk Sultan, Tümän Bey. Scientific
literature, however, continued to decline. The period from the sixteenth to the
eighteenth century simply continued the previously existing state of affairs,
marked by a lack of originality in the processing of knowledge and sterile
commentaries on the writings of the great figures of the past. Historiography
had better possibilities, but it suffered from a lack of patrons. Apart from the
works of Ibn Iyäs, which although ending in 1522 are considered to be part of
Marnluk historiography, most works were of poor quality, especially in their
methodology. Only a few books from this period are lengthy chronicles
containing well-organized materials. Usually, they are but dry chronological
lists of sultans, vice-regents and ChiefJudges, or short summaries ofthe history
of Egypt. The closing decades of the eighteenth century are described in an
important work written in the nineteenth century, the chronicle by al-Djabarti.
A popular genre was treatises on pilgrim routes to Mecca and Medina, as
well as writings on places of pilgrimage and the tombs of outstanding figures,
especially those of scholan and shaykh saints.
The veneration of saints was associated with the advance of the Dervish
orders and the growing interest in mystics. The most popular order was the
Ahmadlya, founded by Ahmad al-Badawi (died 1276), whose tomb in Tanta
was the centre of a special cult. Among the influential orders, mention should
be made of the Kädirlyya, the Rifaïya and the Nakshbandiya. The Dervish
orders were headed by a representative of the descendants of the Prophet (nakib
al-ashräf), a shaykh from the al-Bakrî family which derived its origins from the
fint Caliph Abu Bakr. These orders were associated with certain social strata
and their solid organization gave their leaden significant political power. The
orders also had a cultural role as they contributed to the education of people by
passing on to them a written literature, mosdy poetry. Mystic poetry was
cultivated by a number of poets, such as the pleiad of members of the al-Bakri
family, 'Abdullah al-Sha'räwi. Ahmad al-Dardïr and othen.
Throughout the eighteenth century Egypt had no printing press, although
from 1729 Turkish and Arabic books were being printed in Istanbul. Copying
by hand continued to be the only way of increasing the number of copies of
literary works.
Cairo and other Muslim cities are still adorned today by numerous
architectural monuments built during the Ottoman period. Most of them are
buildings serving religious or educational purposes which were financed by gifts
from vice-regents, senior odjak officers, beys and even Ottoman sultans or other
memben of the dynasty. These buildings contain a number of specifically
Egyptian elements, both in their architectural conception and decoration. The
imported elements are chiefly represented by the manara (minaret), the faience
wall decoration and the flowers painted on the ceiling and walls. The ample size
and comfort of surviving private houses reflect the well-being of city-dwellen.
They are a testimony to the changes under way in Egyptian society and the
growing importance of traden, the nucleus ofthe future bourgeois class.
Economic decline and attempts to achieve independence

After the deaths of Ibrahim Bey and Ridwän Kähya, fierce fighting broke out
among theKazdughlïya. For a brief period, a number of beys, all of them former
Mamluks of Ibrahim Kähya, followed one another at its head. The previously
rarely used honorific tide of Shaykh al-Balad began to be conferred in 1756. In
1760 'Ali Bey, known as al-Djinn (the demon), became the new Shaykh al-
Balad and soon revealed himself to be an extraordinarily ambitious and
energetic man. He secured his position in Egypt and, through his contacts with
the Sublime Porte, succeeded in persuading Istanbul that he was the only man
capable of putting Egypt to rights. In exchange for permission to retain all the
property confiscated from his adversaries, he pledged to pay Istanbul the debt
accumulated on the hazine over the past ten yean, amounting to 91 million
paras, and to pay the sum of 50 million paras raised by the sale of confiscated
beneficies. The hanhness of his measures against land ownen and the politically
influential triggered the formation of an opposition headed by Sälih Bey who
left for Upper Egypt in 1765 to join forces with Shaykh Humäm, tlie leader of
the Hawwära Arabs, who was already harbouring many of 'All Bey's enemies.
Tlie Ottoman conquest of Egypt 83
Aware of the threat represented by this group which was firmly in control of
Upper Egypt, 'Ali Bey attempted to destroy it through a military campaign but
he failed and was forced to flee to Syria in 1766. But the following year 'All
Bey returned to Egypt with the support of the Sublime Porte; he reconciled
himself with Sälih Bey and rid himself of his opponents who had formed a
duumvirate in his absence.
As soon as 'All Bey had fulfilled his promise to the Sublime Porte to pay off
the deficit, he radically shifted policy. In 1768 he deposed the vice-regent and
appointed himself to the post. By adding the office of titular head of the
Egyptian administration to his political power as Shaykh al-Balad he became the
real ruler of Egypt and his subordination to the sultan was merely formal. 'All
Bey did not go so far as to declare himself an independent sovereign, but he
usurped the right to mint coins of his own and had his name proclaimed during
Friday prayers. He endeavoured to consolidate his economic and political
power by strengthening the army and developing agriculture and trade. He also
opened commercial negotiations with leading European powers.
In 1770 his office enabled him to become involved in the succession dispute
among the Hashemite amirs in Mecca. The setdement of the succession by an
Egyptian intervention and the replacement of the Ottoman vice-regent in
Jeddah by an Egyptian bey marked a gain for 'All Bey at the expense of the
Sultan's supremacy in Hidjäz. Stimulated by this success, 'All Bey began to
cherish the idea of unifying under his rule, but within the framework of the
Ottoman empire, the areas that had once formed the territory of the Marnluk
sultanate: Egypt.Hidjäz and Syria.
'All Bey decided to extend his grip to Syria as soon as favourable
circumstances arose in 1770. The Ottoman navy had just been destroyed by the
Russian fleet and the vice-regent of Damascus was facing problems with the
revolt of Shaykh Zähir 'Umar. 'All Bey sent troops to help the latter and the
two armies, reinforced by further units under the command of Muhammad
Bey Abu'l-Dahab, defeated the government forces and occupied Damascus.
Despite these initial successes, the two beys gave up their adventure and,
assuring the sultan of their loyalty, returned to Cairo with their army at the end
of 1771. In 1772, following desertion by his allies, 'All Bey's power collapsed
and he had to hand over office to Muhammad Bey. With the support of his
allies in Syria and Palestine, 'All Bey made a fruitless attempt to return and died
soon after in captivity.
Although 'Ali Bey was inspired by Egypt's past, some features of his
ambitious career foreshadowed his ultimate failure. He was also ill-served by
the system of Marnluk houses and the fickleness of their allegiance.
While he was in power, symptoms appeared of an economic crisis which
continued, with varying intensity, until the nineteenth century. Prompted by
the need to increase the receipts from the feudal rent, 'All Bey first inflicted his
harsh fiscal policy on the rural areas. It was undoubtedly one ofthe main causes
of the impoverishment of the farmen and their fleeing to the towns, which
however were unable to provide them with the means to make a decent living,
AliHBeí.
ÏVucîui de la ( ' rimes,
6.4 'Ait Bey, vice-regent of Egypt (© The Mansell Collection, London)

as more and more craftsmen and traders there were also subject to high taxes.
This general impoverishment grew worse as the yean passed and was aggra¬
vated by a series of epidemics.
In the person of the new Shaykh al-Balad, the Porte hoped that they had at
last found a man that would be wholly committed to them. In 1775 Muham¬
mad Bey dispatched over 130 million paras as the annual payment and, in the
same year, he sent an expedition against Shaykh Zähir al-'Umar on the Sultan's
orders. But he was killed during the siege of Akka and that was the end of
Egypt's involvement in Syria. The Porte actively exploited the power struggle
which broke out among memben of the Marnluk elite after Muhammad Bey's
death. It was ready to support whoever promised to send the largest portion of
the tax levied from the new benefice holden but had no intention of inter¬
vening directly in Egyptian affain.
In the struggle for power, which lasted almost ten yean, the main
protagonists were three Kazdughllya beys, Ismä'il Bey, Murad Bey and Ibrahim
Bey. The fint phase ofthe struggle saw the latter two succeed in forcing Ismä'il
Bey and his followen to leave Egypt. This was followed by a prolonged fight
between the two victon, as Ibrahim, although officially recognized as Shaykh
al-Balad, was prevented by Murad from being fully master ofthe situation. This
rivalry gave the Porte the chance to manipulate the adversaries with the aim of
increasing to the maximum the sum ofthe annual hazine. But it was not able to
make full use ofthe opportunity as the two men gradually gained control ofthe
whole administration of the province, seizing all the financial resources which
they used for their own purposes.
When the two rivals were reconciled, the Sublime Porte, considering that its
interests were endangered, decided to re-establish direct control over Egypt
through military intervention. In July 1786 an Ottoman expeditionary force
landed at Alexandria under the command ofHasan Pasha. It dispened the
Marnluk troops but Murad Bey and Ibrahim Bey retreated into Upper Egypt
with what remained of their forces. To restore the Sultan's authority, Hasan had
to break the military power of the two usurpen and remould the Marnluk
elements that had remained loyal. He sought to do this by creating a new set of
beys and garrison commanders appointed from among the various Marnluk
households. Given that the main aim of the punitive expedition was to restore
regular delivery of the hazine, Hasan Pasha prepared a number of decrees to
ensure that the province of Egypt complied with its obligations to the Porte
and the holy cities. But before he could implement these laws, he was sent
elsewhere on a military mission.
The military intervention in Egypt did not have the expected results. The
virtual division of Egypt into a lower part, ruled by the Sultan's representative,
and an upper part, dominated by the two rebels, came to be accepted as a
reality. Despite the changes brought about by Hasan Pasha among the holden
of tides and offices, and notably the appointment of Ismä'il Bey as the new
Shaykh al-Balad, the Marnluk regime remained intact and with it all the
conditions for a resurgence ofthe old problems.
The political crisis was seriously aggravated after 1783 by economic dif¬
ficulties such as poor harvests, famines, price increases and currency devalua¬
tions. An unusually severe epidemic broke out in 1791, which claimed the lives
of loyal beys, notably Ismä'il Bey. Murad Bey and Ibrahim Bey were thus able
to return to Cairo in the summer of the same year without encountering any
resistance and re-establish their authority over the whole of Egypt.
The Porte consented to their rule on condition that they fulfilled all their
obligations. In 1792 the two beys signed an agreement stipulating the total sum
to be paid and the method of payment but they kept it reluctantly and only in
part. Their return to power signified the return of the harsh, exploitative
regime that Egypt had experienced before Hasan Pasha's expedition and its
consequences for the economy were even more disastrous. The country had
been disrupted by the political anarchy of the previous ten yean and the never-
ending plundering of its resources and reserves. The critical state of the
economy was a reflection of the political oppression to which the whole of
society was subjected by a small élite made up of the Marnluk beys and their
retinues. But the central government lacked the strength required to eliminate
them and Egyptian society was not yet ready to do so itself. But the ousting of
the Mamluks would not be long in coming. The fint impulse was given by the
totally unexpected arrival in 1798 on the Egyptian coast ofthe French military
expedition under the command of Napoleon Bonaparte.
Conclusion
During the historical period from the fall ofthe Marnluk sultan in 1517 to the
French expedition of 1798, the development of Egyptian society was deter¬
mined by its own inner dynamics. Unlike the European parts of the Ottoman
empire, Egypt remained on the fringes of political developments and its socio¬
economic development did not reach the stage where the bourgeoisie made its
appearance as a new social class. Socially, therefore, it remained at the stage
traditionally known as feudalism.
The incorporation of Egypt into the Ottoman empire was, however, a
significant gain for the empire in that it enhanced its economic potential and
political strength. But relations between the central government and the
province of Egypt were often strained and the gradual decline ofthe Ottoman
empire led to clashes and crises which led to the emergence of decentralizing
forces.
The growth of separatist forces in Egypt in the late eighteenth century was
not an isolated phenomenon in the Arab countries ofthe Ottoman empire. But
in the case of Egypt they were not accompanied by the ambition to get rid of
the last vestiges of subjection to the Ottoman sultan.
Hasan Pasha's expedition in 1787 revealed the military inadequacy of the
Marnluk regime and its vulnerability to an attack by a stronger enemy as
Napoleon's expedition was to demonstrate. Moreover, by trying to destroy
decentralizing forces and reassert ties between the province and the central
6.5 Vie mosque of Muhammad Bey Abu'l-Dahab, 1188/1774 (Institute of Egyptology,

Charles University, Prague)
government, the Ottoman expedition foreshadowed the efforts that the Otto¬
man state would make in the course ofthe nineteenth century.
Egypt's increasing interest in territories beyond its borders and its efforts to
enter into independent commercial relations with other regions show that it
had broken out of its centuries-long isolation. That isolation was ended when
France embarked on its colonial expansion in the eastern Mediterranean. In the
increasingly problematical context of the Eastern Question, Egypt was to
become a country of major importance on the chessboard of world politics.
The Sudan,
1 500-1800
Introduction
The history ofthe Sudan in the period under discussion was characterized by
population movements both into and out of the Sudan. In the north, the slow
penetration of large numben of Muslim Arabs, well under way by the
beginning of our period, led to the integration of the region into the larger
pan-Islamic world. The process of cultural and ethnic assimilation was a two-
way process: it led, on the one hand, to the Arabization and Islamization of
large numbers of Sudanese peoples and, on the other, to the integration of Arab
immigrants.
The influence of Islam and Arabic culture on the Southern Sudan was
negligible. The expansionist energies of the Nilotes from the south succeeded
in arresting the southward march of the Arabs as well as the spread of Islam.
Indeed the Nilotes, especially the Shilluk and the Jieng, posed a serious threat
to the northern Muslim states until the end ofthe period.
This chapter will deal primarily with the establishment and expansion of the
two Muslim savannah states (the Fundj and Für sultanates), their relations with
each other and their interaction with non-Muslim African societies which were
also in the process of formation from diverse linguistic and cultural groups. It
will also attempt to show that the expansion of Islam constitutes an important
factor in the history ofthe Sudan during this period.
The southward process of Arabization and Islamization was halted where the
riven divide (the Bahr al- Ghazäl and the Bahr al-'Arab), thus creating a sort of
cultural frontier between the northern Sudan and the southern Sudan which
has deeply coloured the interpretation of Sudanese history.

From the southern Sudanese viewpoint, relations between the two regions
have been marked essentially by the incunions and economic exploitation by
the Fundj in the Upper Nile and later by the Für in the Bahr al-Ghazâl region.
But from the northern viewpoint political, economic and social changes have
generally been explained by the arrival of Muslim immigrants, the 'wise
strängen' from the north or east. Today, the north is presented as Arab and
Muslim, and the south as African (or negroid) and 'pagan' (or 'animist'). Thus
89
the frontier that separates them becomes increasingly defined in religious and
ethnic terms.
The historical reality was of course more complex than this. First, the frontier
between these two regions was constandy shifting and not always southward or
westward.
Second, the frontier was a vast area which, around the Upper Nile, for
instance, stretched over several hundred kilometres. The same was true for the
western area. Furthermore, within the frontier itself, cultural, ethnic and social
transformations were taking place all the time, with people becoming Arabs,
Für, Fundj, Shilluk, Naath or Jieng according to political and economic
circumstances.
Third, within the northern Muslim sultanates themselves the process of

ethnic change was a complex phenomenon. To begin with, the word Arab,
which had strong cultural connotations, was progressively emptied of its ethnic
significance, as in the case of the Bakkära Arabs who lived along the frontier.
Furthermore, by the eighteenth century, each sultanate was divided into a
number of chiefdoms most of which were plural societies. The social forma¬
tions within these sultanates were further complicated by the factor of slavery.
Many slaves brought from the south formed an important part of the sultans'
army, while others were completely assimilated into their new societies.
Finally, with the exception of the people who had come from the central
and western Bilâd al-Südän (such as the Takrür and the Fulbe) and the
Rashayida from Arabia, no new ethnic groups emerged in the Sudan in the
nineteenth or twentieth centuries, and major population movements were
practically over by about 1 800.
The Abdalläbi state
During the second half of the fifteenth century, a confederacy of Arab kabilas
led by their chief 'Abdallah, nicknamed Djammä' (the Gatherer), succeeded in
conquering the kingdom of "Aiwa'. The conquest of Soba was an indication of
the supremacy ofthe Arabs and marked the beginning of Arab influence in the
eastern Biläd al-Südän. 'Abdallah Djammä' and his descendants, the 'Abdalläbi,
established the seat of their government at Kerri, north of Soba, on the Nile.
Kerri was easy of access for the Butanä Arabs whose support was essential to the
'Abdalläbi. It also made it possible to monitor movements along the Nile valley
and across it on the western bank.
The extent of the new state is not easy to determine. The 'Abdalläbi
hegemony seems to have extended over the Arabs in northern Djazira, the
Butana, east of the Nile, and possibly some of the Bedja groups. The indepen¬
dent 'Abdalläbi state did not last long enough to develop its own institutions.
By the beginning of the sixteenth century it had to face a powerful enemy, the
Fundj, who were cattle nomads migrating down the Blue Nile. The two
migratory forces clashed, probably over possession of grazing land north of the
Djazira. At Arbadjï, in 1504, the Fundj defeated the 'Abdalläbi and reduced
The Sudan, 1500-1800 91
their king to the status of a vassal. The 'Abdalläbi continued to rule the
northern part of the Fundj sultanate as feudatories until the Turco-Egyptian
conquest in 1820.
The Fundj sultanate

The origin ofthe Fundj has given rise to many contradictory hypotheses. Their
kingdom is known in Sudanese traditions as al-Sultana al-Zarka (the Blue
Sultanate). David Reubeni, a Jewish adventurer who went there in 1522-3,
described their monarch as a black Muslim who ruled over a nation of blacks
and whites, referring to the indigenous population and the Arabs respectively.
Local traditions attribute an Arab ancestry to the Fundj, and say that they are
descendants of an Umayyad refugee who came via Ethiopia and who allegedly
married an indigenous princess. He is described as a 'blessed man' who brought
new customs.
It is not clear at what date the Fundj dynasty began to claim an Umayyad
descent but it seems that the Fundj were rapidly Islamized. Like other convert
peoples living on the fringes of Muslim societies, the Fundj wanted to be
regarded as Arabs, hoping thereby to enhance their moral authority over their
Arab subjects.
Having asserted their suzerainty over the 'Abdalläbi territories, the Fundj
ruled their domains from Sennär which became their seat of government. The
'Abdalläbi shaykh became a vassal of the Fundj and bore the title of Mändjil or
Mändjuluk which the Fundj monarchs bestowed on their principal vassals.
In an attempt to shake off Fundj domination, Shaykh Adjïb I, who came to
power in the middle ofthe sixteenth century, challenged the Fundj, defeated
them and drove them into Ethiopia. In the reign of Dakin (1568/9
the Fundj recovered their former supremacy but they had to concede Adjïb I's
right to appoint judges in his domains. Dakin's policy of reorganizing the
kingdom and introducing new regulations appears to have upset the delicate
balance between the two sides and driven Adjlb into open rebellion. In 1611
a Fundj army inflicted a crushing defeat on the 'Abdalläbi in a battle in which
Adjib I was killed. A setdement between the Arabs and the Fundj was
negotiated which restored the status quo. Thus, the descendants of Adjïb, with
the tide of wad lAdjib, continued to rule directly the territories as far north as
Hadjar al-'Asal and indirectly over the ethnic groups living in the Nile valley as
far as the border of Ottoman Nubia. This setdement gave the sultanate a fairly
long period of stability. However, in about the middle of the seventeenth
century, the Shaykiyya chieftaincy revolted against the 'Abdalläbi hegemony
and asserted its independence from the Fundj sultanate.
The extension ofthe Fundj sultanate

The extension of Fundj authority as far as lower Nubia seems to have been
viewed with suspicion by the Ottomans, who had conquered Egypt in 1517. In
the reign of Sulaymän the Magnificent (1520-66) they sent a naval expedition
under Özdemir Pasha into the Indian Ocean against the Portuguese, who were
threatening the Red Sea lanes; as he was returning, Özdemir was instructed to
put an end to the Fundj 'rebellion' in Nubia where two rival factions were
fighting one another. Özdemir captured the strategic fortresses of Ibrim and
Dirr on the border and built a fortress at Say which marked the southern limit
of Ottoman Egypt. The new province, known as Berberistan, was administered
by the Ottomans.
Özdemir Pasha was later entrusted with the task of conquering Habesistän
(Abyssinia). After detailed preparations in Egypt, the expedition started up the
Nile. At Aswan, Özdemir lost control of an undisciplined army and had to call
off the campaign. There it was understood that the expedition had not been
directed against Habesistän but against Fundjistän (the land ofthe Fundj). The
Egyptian frontier was finally consolidated at Hannik in 1622 after some fighting
between the 'Abdalläbi and the Fundj.
Özdemir is also credited with establishing a base at Sawäkin against the
Portuguese and above all the Ethiopians. Sawäkin came under Ottoman
suzerainty and Massawa was annexed in 1557. The coastal strip between the
two ports constituted the province of Habesh. Once the Portuguese threat had
been removed, Sawäkin once again became the main commercial outlet of the
Fundj sultanate. Relations between the Ottoman governor and the Fundj were
at first unfriendly and deteriorated into armed confrontations in 1571.
However, because of the extensive commercial relations that had developed
between the two sides, tension declined and Sawäkin took on an importance
which it retained until the beginning ofthe twentieth century.
By the beginning of the seventeenth century, the Fundj had apparendy con¬
solidated their position at Sennär. The territories they ruled directly extended
from Arbadjï to the south of Fazûghlï. The expansion of their power to the
west into Kordofän had been initiated by Sultan 'Abd al-Kadlr in about 1554.
The increasing pressure of the Nilotic peoples on the White Nile at the
expense of Sennär seems to have led to a confrontation between the Fundj and
the Shilluk. Sultan Bad! II Abu Dikn (1644/5-1718) established a bridgehead
and a Fundj garrison at Alays, on the river. From that strategic position the
Fundj were able to monitor movements across the river and check the Shilluk,
who appear to have entered into an alliance with them.
The Fundj then penetrated the Nüba mountains, one of the principal
sources of slaves. The Islamized kingdom of Takali was reduced to the status of
a tributary. Likewise, the Fundj extended their rule over the northern hills of
al-Däyr and Kordofän. The numerous prisoners captured among the non-
Muslim 'hill-Nüba' were setded in villages around Sennär. They formed a slave
guard to protect the sultan. The establishment of a slave army dependent on
the ruler was viewed with concern by the traditional warriors, the Fundj
aristocracy, and they revolted in the reign of Bad! HI (the Red, 1692-1716).
A new crisis was precipitated during the reign of Bâdï IV Abü ShuUükh
(1721-62), the last effective Fundj monarch. In the fint half of his reign, Bädl
IV left the affain of state to his minister Doka, but when Döka died, Bâdï
The Sudan, 1500-1800 93
banished the ahl al-iisül (members ofthe old lineage and rank) and began to rule
arbitrarily. To rid himself of the Fundj notables who were opposing him, he
sent them on a campaign against the Musabba'ät who had encroached on Fund]
land in Kordofän. After a number of setbacks, the Fundj army, under the
command of Muhammad Abu Likaylik, won a decisive victory in 1747. Abu
Likaylik remained viceroy of Kordofän for 14 years.
The record of relations between the Fundj and the Ethiopians is one of
numerous border disputes, but that did not prevent positive co-operation and
economic interdependence between the two countries. To the Christians of
Ethiopia, Sennär was for many years their sole outlet to the outside world and
trade routes remained open even when relations between Sennär and Gondar
were strained.
Yet the Fund] waged two wars against Ethiopia. The first broke out at the
beginning of the seventeenth century. Relations between the Fundj sultan and
the Ethiopian emperor had detenorated when the latter granted political
asylum to the deposed sultan 'Abd al-Kädir and made him a governor. Frontier
incidents escalated during 1618 and 1619 when large numbers of troops were
involved in clashes. The fact that the two rulers were fighting this war far from
their capitals meant that it did not represent a serious threat to either of them.
It ended in favour of Ethiopia.
In the following century, a second war began with incursions into the
Kaläbät-Dindera region. In March 1744 Iyasu II advanced at the head of a large
Ethiopian army against Sennär. The two armies fought a pitched battle on the
banks ofthe Dindera, in which the Ethiopians were routed. The Fund] victory
was celebrated by Bädl IV and his subjects with religious fervour. This was
echoed as far away as Istanbul and the Ottoman Caliph was said to have been
pleased with the victory of Islam.
The progress of Islam

The Fundj-'Abdalläbi sultanate gave the country a measure of unity and
stability that facilitated the spread of Islam. This was the work of Muslim
scholars whom the rulers encouraged to come and setde in the country. Before
this, Islam had been spread by two groups - traders and nomadic Arabs. Among
the first group, trade and proselytization had always gone hand in hand on the
margin of Islamic societies. The second group, although not well versed in
doctrine, played a major role in the spread of Islam notably through their inter¬
marriage with indigenous Sudanese people. Islamization was the normal
concomitant of Arabization.
The early Fundj period witnessed a rise in the number of Muslim teachers.
Some came from Egypt, Hidjäz, Yemen and the Maghrib but most of them
were born in the country and some had studied in Cairo or other holy places.
On his return from Cairo in the middle of the sixteenth century Mahmud al-
'Araki, the first Sudanese Muslim scholar, established 17 schools on the White
Nile. In about 1570, Ibrahim al-Büläd ibn Djäbir was the first to base his
teaching on two Mäliki textbooks: the Risäla by Ibn Abï Zayd al-Kayrawäni,
and the Mukhtasar by Khalil ibn Ishäk. This teaching helped to establish the
predominance of the Mäliki rite in the region.
The fint trained Muslim scholars were primarily concerned with teaching
Muslim law, the shari'a and its application. Their efforts to transmit orthodox
teaching and raise the level of religious life encountered serious difficulties in a
vast, isolated and backward country. Before orthodox Islam took root there, a
more popular and less exacting type of Islam had been introduced.
Most of the süfi tarlkas (religious orden) came fromHidjäz. The fint, and
probably the most popular, was the Kädiriyya. It was introduced into the Sudan
by Tädj al-Dîn al-Bahäri of Baghdad, who came from Mecca in 1577 and
stayed in the Djazïra for seven years, initiating many prominent Sudanese into
theKädiriyya order. Another order, the Shadhiliyya. was introduced into the
Sudan by a member ofthe Dja'aliyyün clan who had studied in Hidjäz.
When Süßsm began to spread in the Fundj kingdom, it had already lost much
of its importance in the Muslim world and had become tainted with some
unorthodox practices. People believed that baraka (divine blessing) emanated
from holy men and that such men acted as intermediaries between man and
God, hence the emphasis on the cult of saints. Süß teachers were generally
granted land or were exempted from paying taxes and some of them wielded
considerable political influence, such as Idris wad al-Akbär and the Madjadhlb.
Rulen and their subjects came to revere them even more than they did jurists.
The jurists, faced with the gratifying position attained by their rivals, sought
to combine the teaching of law with süß authority. This development was
clearly reflected in local usage in the seventeenth century when the uûefakï (a
corruption oifaklh, a jurist) was applied indiscriminately to jurists and mystics.
Thcfaki established numerous religious centres and gave a degree of stability to
the fluid and heterogenous Fundj society. Their missionary zeal was not con¬
fined to the Fundj kingdom but radiated as far as Kordofän, Därfür and Borno.
Trade and state
Major trade routes crossed the Für and Fundj sultanates linking them to Egypt
and the Red Sea. They played an important role in strengthening cultural and
economic links with the outside world. The Fur and Fundj sultans, like most
rulers in the Sudanic belt, encouraged long-distance trade. Slaves, gold, ostrich
feathers and other African products were exchanged for fine cotton textiles,
weapons, jewellery and other luxury items. Besides the customs duties they
collected on goods passing through, the sultans needed luxury objects to
maintain their prestige and reward their loyal supporten.
External trade ran along two main axes, one west-east and the other south-
north. The first connected Borno-Wadai with Sennär, through Kobbie and
Kordofän. From Sennär it went on to Koz Radjad and Sawäkin either directly
or through Shandi. Through this route, known as the Sudan Road, the Eastern
Sudan was laid open to cultural influences from the Western Sudan and North
Africa. The Sudan Road was taken by indigenous Muslim scholars and pilgrims
who could not afford the cost of crossing the desert through Egypt with the
Tlie Sudan, 1500-1800 95
7.1 Trade routes ofthe Sudan (after Y. F. Hasan)

caravans. Most of these pilgrims travelled on foot and benefited from the protec¬
tion of Muslim peoples. They lived by charity or sold asses, books or other items.
The second route commenced at Sennär, went through Kerri and crossed
the Bayüda desert to Asyüt in Upper Egypt through Dünküla and Salima.
However, the uprising of the Shaykiyya in the eighteenth century caused chaos
along this route and it was abandoned in favour of a route which followed the
eastern bank of the river, crossed the Atbara and then went across the Nubian
desert to Isna in Upper Egypt. At the Salima Oasis, the Nile route joined the
Darb al-(Arba(in (Forty-Day Road), the main artery of Für trade with Egypt. It
started from Kobbie and went to Asyüt by way ofthe Salima and Khardja oases.
Thanks to its wide hinterland and strategic position Sennär became a major
commercial centre. Most of the foreigners who came to Sennär were mer¬
chants and a few were craftsmen. The bulk of the trade seems to have been in
the hands of Sudanese Djallâba (traden). Thanks to their spirit of enterprise, the
Djalläba also played an important role in the development of Fur trade. In
Darfur, long-distance trade was more or less under state control. Through these
trading caravans, the Für sultanate became better known to the outside world.
The commercial ascendancy of Sennär was affected by the rivalry between
Fundj and Für over Kordofän and also by the Shilluk raids which rendered the
Kobbie-Sennär road unsafe. This is why the caravans preferred to take the
northerly route by Shandî. which, in the coune of the eighteenth century,
became a big commercial centre and replaced Sennär as the market for Ethiopian
and Für trade.
In the east, the Hadäriba of Sawäkin were also active in the long-distance
trade between the Red Sea and the Nile. They purchased African products and
slaves from Shandî and exchanged them for Indian goods at Sawäkin, which
was the biggest slave market after Cairo and Massawa.
The movement of slaves northward down the Nile goes back to ancient
times. The Arabs continued the practice when they concluded the bakt (treaty)
with the Nubians. Four hundred slaves a year were sent from Nubia and the
Arab world came to favour them as domestics. They were excellent bowmen
and also formed highly sought-after combat units. Given the increase in the
demand for black slaves, Nubia was unable to meet the needs of the Muslim
world and Arab traders had to tap other sources in south and south-west
Nubia. The slave trade thus took on an importance which it retained until the
last decades of the nineteenth century. At first the slaves were Nubians and
Bëdja, but later they were imported from Kordofän and Darfur and, eventually,
the Bahr al-Ghazâl. Borno-Wadai and other areas in central Biläd al-Südän.
Slaves were acquired through conquest, kidnapping or purchase and the role
of Arab dealen vaned in different periods and different places. It seems,
however, that Arab traders did not procure slaves direcdy themselves (except in
the nineteenth century) but they relied mostly on African supplien and middle¬
men.
The external demand for black slaves for armies shrank in the time of the
Ayyübids (1172-1251) who replaced their black troops with white ones. The
The Sudan, 1500-1800 97
Mamluks (1251 punued the same policy and, so long as they ruled, the
army was almost exclusively made up of white slaves. However, the demand
for black slaves continued in other regions, particularly in the new Fundj and
Für sultanates where such slaves formed the core of the army. In the nineteenth
century Muhammad 'Ali Pasha, viceroy of Egypt, began to recruit black slaves
for a modern army and throughout the nineteenth century the Egyptian army
included a significant number of Sudanese soldiers.
The decline ofthe Fundj sultanate

The commercial decay of Sennär was accompanied by political decline. With
the increase in the repressive measures taken by Bâdï IV, some members of the
Fund] aristocracy, who had accompanied Abu Likaylik to Kordofän, induced the
general to depose the sultan, which was done in 1762. The success ofthe revolt
of Abu Likaylik was an important landmark in the history of the sultanate.
Shaykh Muhammad Abu Likaylik and his successors, the wazirs, became the
effective hereditary rulers or regents of the kingdom until its downfall. The
Fundj sultans were no more than puppets in their hands.
On the death of Abu Likaylik in 1776 the Fundj kings conspired with the
provincial governon, particularly the 'Abdalläbi, to oust their regents. The last
50 years of the Fundj sultanate were punctuated by intrigues, revolts and civil
wars among rival factions.
In the north, the authority of the 'Abdalläbi chiefs over the Nile valley was
also in decline: the Sa'däb of Shandî and the Madjadhîb of al-Damär became
virtually autonomous. The Shaykîyya warriors, who dominated the Dunkula
region, were being constantly harassed by the Mamluks of Egypt. In the west,
the Fundj, harassed by the Für sultanate, were ultimately driven out of
Kordofän. When the Turco-Egyptian forces approached Sennär in 1821, the
sultanate was too feeble to muster any resistance.
The Fur sultanate
The Für sultanate was one of the Islamic kingdoms in the savannah belt of the
eastern Biläd al-Südän. To the west, it was separated from the sultanate of Wadai
by small independent kingdoms whose allegiance was divided between the two
sultanates. To the east, the plains of Kordofän, which stretched between the Für
sultanate and the Fundj kingdom, were the source of dispute between the two
states. The Libyan Desert to the north and the Bahr al-'Arab to the south
formed natural borden to the state. The central region, dominated by Djabal
Marra, the cradle of the Für sultanate, was the meeting place of a number of
trade routes, which were sources of economic prosperity, human migrations
and cultural influences.
The origins of the Für sultanate are obscure but, according to oral tradition,
the Für state was preceded by two indigenous dynasties: the Dädjü and the
Tundjur. The Dädjü kingdom flourished during the thirteenth and fourteenth
centuries, but at the beginning ofthe fifteenth century the Dädjü lost control of
the trade and the Tundjür succeeded them. It was probably during the reign of
the Tundjür that the influence of Islam was fint felt. It resulted from the many
commercial contacts with the Muslim world and the gradual penetration of
Arab immigrants. However, the proper Islamization of the region only began
with the establishment ofthe Für sultanate in the early seventeenth century.
The sultanate owed its name to the Für, a Sudanese people who lived
around Djabal Marra, and who, for reasons unknown, left their mountain
home and setded in the savannah. The Kayra sultanate flourished from about
1640 to 1874. It was restored in 1898 and was finally annexed to the Anglo-
Egyptian Sudan. Sulaymän Solongdungu was probably the founder of the
dynasty (c. 1640 He is remembered as a conqueror who drove out the
Tundjur and annexed the area around Djabal Marra.
The policy of expansion and consolidation started by Sulaymän was
continued by his successors who advanced to the north and north-west. Ahmad
Bukr b. Müsä (c. 1682 annexed the strategic state of Dar Kimr. This
expansion brought the Kayra dynasty into contact with the Zaghäwa whose
territories became a source of conflict between the sultanates of Wadai and
Darfur. Wadai was forced to pay the Kayra sultanate the tribute that it had
formerly paid to the Tundjür kings. When Sultan Ya'küb refused to pay the
tribute and invaded Därfür, Bukr eventually drove him out.
Until the end of the eighteenth century the political history of Därfür was
dominated by the struggle between two factions of the Kayra dynasty: the
sultans who sought to centralize power in their own hands and other princes
who wanted to strengthen the powers ofthe dynasty as a whole. Conflict broke
out when Sultan Bukr sought to ensure that each of his sons should succeed him
in turn. He had a hundred children, five of whom ascended the Für throne. His
fint successor, Muhammad Dawra, began by eliminating or banishing his
brothen and named his son as his successor. When 'Umar Lei came to power,
he was met with hostility from his uncles, the sons of Bukr. Sulaymän ibn
Ahmad Bukr vainly sought the support of fint the Musabba'ät and then the
sultan of Wadai. However, his action provoked 'Umar Lei to invade Wadai,
about the middle ofthe eighteenth century. After heavy fighting, the Sultan of
Wadai, Muhammad Djawda, defeated the Für army and captured the sultan.
The conflict continued during the reign of the sixth sultan, Abü'l-Käsim ben
Ahmad Bukr. To consolidate his position, he fought his brothers and recruited
slave troops rather than the traditional ethnic levies, which offended many title-
holders. Indeed, it produced tensions similar to those that developed in the
Fundj kingdom during the reign of Bâdï IV.
During the battle against the Wadai forces, Abü'l-Käsim was deserted by the
traditional war chiefs and the ethnic levies. He was left alone with his slave
army and lost the battle. The old Für nobility and the traditional warriors
proclaimed a new sultan, Muhammad Tayräb ibn Ahmad Bukr, who followed
the example of his predecessors and created a standing slave army, the Kurkwä
(spearmen). One of them, Muhammad Kurra, a eunuch, became one of Tayräb's
mästen ofthe royal grooms. He later played an even more important role.
The Sudan, 1500-1800 99
Having failed to expand westward against his enemy, the sultan of Wadai,
Tayräb, concluded a treaty with him which remained effective for a hundred
yean. He then turned eastward against the Musabba'ät who were attempting to
establish a strong state in Kordofän. Tayräb may also have had economic
objectives, such as controlling the trade routes and the sources of slaves and
gold in southern Kordofän. His immediate objective was probably to dnve his
brothers and other nobles out of Därfür so as to enable his son to succeed him.
At the head of a large army, Tayräb attacked Häshim, Sultan of Kordofän. The
latter, deserted by his army, took refuge in the Fundj kingdom, while Kordofän
remained in the hands ofthe Für until the Turco-Egyptian conquest.
Tayräb's death was followed by a bitter struggle over the succession. Finally,
Ahmad Bukr's youngest son, 'Abd al-Rahmân, emerged victorious from the civil
war thanks to the support of Muhammad Kurra. The new sultan rewarded
Muhammad Kurra with the position of Ab Shaykh. a rank that gave him great
authority. From 1790 to 1804 Muhammad Kurra was the strongest man in the
state.
'Abd al-Rahmân marked his victory by building a new royal residence

(fishir) at Khor Tandalti, east of Djabal Marra, in 1741/2. Until then the Für
sultans had had no fixed capital. The growth of a permanent capital at al-Fäshir
and the consolidation of the centralizing forces and Islamizing influences made
the reign of 'Abd al-Rahmän the apogee of the Kayra sultanate. There were
increasing contacts with the outside world through trade and the arrival of
religious teachers. 'Abd al-Rahmän exchanged presents with the Ottoman
sultan, who bestowed on him the honorific tide of al-Rasliid (the Just). He also
corresponded with Bonaparte in 1799 during the French occupation of Egypt.
The adoption of Islam in Darfur was probably a much slower process than in
the Fundj kingdom. It speeded up in the eighteenth century, but African
religious rituals and practices coexisted with Islam at the Kayra court for some
time. The propagation of Islam was encouraged by Ahmad Bukr who built
mosques and schools while Muhammad Tayräb obtained religious books from
Egypt and Tunisia. (Abd al-Rahmân al-Rashîd, himself a Muslim scholar,
encouraged jurists and mystics from other countries to come and settle in his
sultanate. Among these was the Tunisian Arab 'Umar al-Tunisi, later followed
by his son Muhammad, whose account of Darfur is one ofthe main sources for
the history of the country. Other scholars came from Egypt, Hidjäz, the Nilotic
Sudan and western Biläd al-Südän. To attract them to his country, the sultan
offered religious teachers lands through the hakura (tax exemption) system, as
was the case in the Fund] kingdom.
By the end ofthe eighteenth century the sultan and his immediate associates,
who had no ethnic affiliations, had gone a long way in developing external
trade and adopting Islamic institutions to govern the country. This state of
affairs helped to alter the ethnic structure of the state and weaken old religious
practices. The emergence of a new class of merchants, jurists and mystics was
instrumental in bringing this change about. However, the Kayra dynasty,
despite its Arab lineage, had its roots among the Für. Although Arabic was used
in diplomacy and trade, Für remained the language ofthe court.

On the death of 'Abd al-Rahmän in 1801/02, his son Muhammad al-Fadl
succeeded him with the help of Muhammad Kurra. But the new sultan soon fell
out with this man whom he had made his chief minister and killed him.
Muhammad al-Fadl reigned for 40 years, during which the state began to
decline.
The Southern Sudan
Despite the existence of serious anthropological works by eminent scholars,

historical research on Southern Sudan is still in its infancy. The same applies to
archaeological investigations and historical linguistic research. From the meagre
evidence, however, certain broad oudines are beginning to emerge. From
linguistic evidence, for example, it is becoming increasingly clear that Nilotic
cultures and probably Central Sudanic ones have their roots in Southern Sudan.
Much of Southern Sudan and northern Uganda was probably occupied by
peoples speaking Central Sudanic languages until the last decades of the last
millennium when the area began to be colonized by Eastern and Western
Nilotic speaken. Indeed, one of the major historical themes during our period
is the progressive Niloticization of the peoples of Southern Sudan. Today the
region is mostly occupied by Nilotes, with only the Moni and the Madi who
live to the west of the White Nile remaining to remind us of the former
presence of Central Sudanic speakers.
Recently, Roland Oliver attempted a synthesis of Iron Age archaeological
evidence for the prehistory of this region which tends to confirm the linguistic
picture. He envisaged two centres of Iron Age cultures, one to the east of the
Nile swamps in the SudanEthiopian borderlands and the other on the
ironstone plateau of the Nile-Congo watenhed. These two cultures, one
Nilotic and the other Bantu, were separated by Central Sudanic cultures. The
Nilotics moved southwards into the Bantu regions and, as a result of that
meeting, the Bantu-speaken began to practise intensive pastoralism.
Regarding the Sudan, Oliver identified two Iron Age periods among the
Nilotes. In Western Equatorial and Bahr al-Ghazäl provinces the fint period is
associated with Iron Age pastoralists, the Luel, who kept humpless cattle and
built artificial mounds in order to protect themselves from flooding during the
rainy season.
The beginning of the Later Iron Age of the Nilotes seems to have coincided
with the appearance of humped catde on the Bahr al-Ghazäl and the full
development of transhumance. Humped cattle probably spread southwards
with the Bakkära Arabs following the fall ofthe kingdom of Dünküla. Thus the
Arab penetration of Northern and Central Sudan in the thirteenth and four¬
teenth centuries not only coincided with but might have triggered the Western
Nilotic migrations.
The rapid expansion ofthe Western and Eastern Nilotes southward and east¬
ward may therefore be linked with the appearance of intensive cattle-herding
The Sudan, 1500-1800 101
combined with cereal agriculture. This combination provided a means of food

production well suited to drier areas which enabled the Nilotes to occupy
many areas previously avoided by Bantu-speakers.
The Shilluk kingdom

In Southern Sudan, the Shilluk constitute the largest component of the
Northern Luo group, the others being the Luo of Bahr al-Ghazäl and the
Anywa who straddle the SudanEthiopia border. The Shilluk initially settled
near Malakal in the early sixteenth century, after driving out the Fundj who
lived in the area at the confluence of the Nile and the Sobat. This region
controls access to Lake No, giving it great strategic importance. This small
group of Luo-speakers incorporated Fundj, Nuba and other elements. All these
groups, with different cultures and economic traditions, combined in the
crucible of history to form what became the Shilluk nation from the second
half of the seventeenth century.
During this formative century, the Shilluk established a mixed economy
based on catde-herding and cereal agriculture. They were essentially sedentary
and did not practise transhumance; they lived in a string of villages along the
western bank of the Nile between Muomo and Tonga.
Frontiers and socio-economic interdependence

By the beginning of the seventeenth century, the west bank of the Nile was
already overpopulated and the Shilluk began to expand into two frontier areas.
Northwards, they attempted to take possession of the White Nile valley
between Muomo and Alays. The area was unsuitable for agriculture but pro¬
vided abundant game, fish and honey. From the reign of Odak Ocollo
(c. 1600-35) to 1861, the Shilluk dominated this region, which the Muslim
peoples called 'Bahr Scheluk'.
The second frontier area was the region between the Nile and the Nuba
Mountains. Tradition dwells so much on Shilluk-Nüba activities in this area
that it must have been important for both peoples before it was occupied by
the Bakkära Arabs in the second half of the nineteenth century.
We learn from Shilluk traditions that, during the reign of Odak Ocollo the
Shilluk supported Darfur in its struggle with the Fundj sultanate for the control
of the White Nile trade. With the capitulation of Takali, the Fundj and the
Shilluk were left confronting each other along the White Nile, exhausted after
three decades of warfare.
From about 1630 the Jieng, another Jii-speaking people, invaded southern
Fundj. Throughout the seventeenth and eighteenth centuries, they expanded
into southern Djazira, which gradually altered the balance of power in the
region. The presence of the Jieng represented such a threat that the Shilluk and
the Fundj joined forces against them. They succeeded in preventing them from
expanding northwards and westwards and even forced them eastwards towards
the Ethiopian border.
The alliance with the Fundj marked the beginning of a socio-economic
102 Africafrom the Sixteenth to the Eighteenth Century
interdependence that the Shilluk established with different groups at different

times (Fundj, Djalläba, Bakkära Arabs, Arab freebooten, European traders and
Mahdists), usually to exploit the Jieng. The fact that the Shilluk combined time
and again with these various groups to exploit a related Jii-speaking people
shows that such socio-economic alliances were untainted by any racial or
ethnic ideology. Racial ideology in Southern Sudan was a post-Mahdiyya-
period phenomenon.
The development of political and social institutions

During the second half of the seventeenth century, political developments
occurred which significandy affected the socio-economic development of
Shillukland. Despite its divenity, the population acquired a sense of national
unity and a more centralized administration was established under the king
(reth). A royal monopoly of economic resources as well as of local and long¬
distance trade was gradually established.
Three names are associated with these developments in Shilluk history:
Abudhok, queen and daughter of Reth Bwoc, and one ofthe many powerful
women in Shilluk history; her half-brother Dhokoth; and Tugo, the latter's
son. Reth Dhokoth (c. 1670 is famous for having made raids down the
Nile and westwards into the Nuba Mountains. The great famine of 1684 (Umm
Laham in Arabic) was probably responsible for making many Shilluk move
north to raid or setde in 'Bahr Scheluk'. Dhokoth's raids were extremely
successful and brought back many captives.
Reth Dhokoth's military and economic successes were largely responsible
for the centralization of royal power which was completed by his son Togo
(c. 1690-1710). Togo founded the village of Fashoda which became the
permanent residence of the reth. He also instituted the elaborate installation
ceremony for a Shilluk reth. His fame spread rapidly both within and outside
the kingdom.
By the beginning of the eighteenth century, the Fundj empire was dis¬
integrating and, as the power of the Fundj declined, that of" the Shilluk
increased. The end of the century saw the total collapse of the Fundj sultanate,
which enabled the Shilluk to strengthen their control of the White Nile. The
English traveller, W. G. Browne, who stayed in Därfür for four yean (1793-6),
reported that the Shilluk were in complete control of the river at Alays where
they provided a ferry service enabling caravan traden moving between Sennär
and al-'Obeyd to cross the Nile.
The decline of Shilluk power

During the reign of Reth Nyakwaa (c. 1780-1820) the Jieng migrated en masse
across the River Sobat and the Shilluk, who for a century and a half had been
in complete control of the White Nile area from Lake No to Alays, had to
accept the presence of other Jii-speaken. But more was to come. A year after
the death of Nyakwaa, the armies of Muhammad 'Ali Pasha, Viceroy of Egypt,
invaded Sudan, put an end to the Fundj administration and installed Turco-
The Sudan, 1500-1800 103
Egyptian rule. A clash with the Shilluk was inevitable. From 1821, despite
sustained Shilluk resistance, the Shilluk frontier steadily retreated southwards
before the gradual encroachment ofthe Arabs and the Turco-Egyptian regime.
In order to dominate the White Nile, the Shilluk had to control the
navigable waterways. They possessed a large number of vessels and were fine
oarsmen. For their raids or military expeditions groups of 30 or 40 canoes
travelled together and constituted a considerable military force in the region.
But a Turco-Egyptian government shipyard was established at Mandjara in
1826 and that put an end to Shilluk naval supremacy on the White Nile.
The peoples ofthe Bahr al-Ghazäl

The country south of the Bahr al-Ghazäl and west of a line from Meshra al-
Rek to the point where the boundaries of modern Sudan, Zaire and Uganda
meet, was occupied by peoples belonging to two major language families: the
Ubangian (Niger-Congo) and the Central Sudanic (Sahara-Nile). Central
Sudanic speakers had been setded there since very early times; they settled
along waterways and lived by farming cereal crops, keeping cattle and hunting.
Ubangian speakers arrived from the west (Central African Republic) and settled
mostly to the west of the Central Sudanic speaking communities or in
between, which is understandable given the very light settlement of peoples in
the area. Ubangians were farmers who first based their economy on yams but
later shifted to producing cereal crops or bananas, depending on the local
environment. They did not keep catde, which implies substantially different
attitudes towards wealth, and especially bride-wealth, from those ofthe Central
Sudanic people.
We know very little about the life of these people before 1800 because
nineteenth-century developments led to the break-up of their communities.
Furthermore, no oral tradition could be found about what happened before
1800. The only genealogies that go back further (to between 1650 and 1705)
are those ofthe ruling families of three Bongo clans. They merely tell us that the
people were already settled in the Wau-Tonj area in the eighteenth century. By
1 800 slave-traden from Darfur were already active in Dar Fertlt and Dar Banda
while the Zande were beginning to conquer the areas further south.
All that can be said for the period covered by this chapter is that Darfur
began to exercise informal control over Dar Fertït well before 1800 and
demanded tribute from the inhabitants of the Hofrat-en-Nahas region, while
Arab or Für families established themselves as leaders over small groups in the
area of Raga. More research is needed throughout this area where it may still
be possible to gather data on ways of life, trade and possible population move¬
ments before 1 800.
Morocco
In Volume IV it was shown that the fifteenth century was marked by Spanish
and Portuguese offensives against North Africa, particularly against Morocco.
Since 1415 the year in which the Portuguese took Ceuta they had seized
many coastal places on the Adantic littoral, making them bases for raids into the
Moroccan hinterland. All these events aroused firm resistance. The zäwiya
Shaykhs and religious brotherhoods encouraged this spirit, using it to prepare
the population to fight the invaden.
Some of the shßrifs of Dar'a, led by Abu 'Abd Allah, known as al-Kä'im bi-
'Amr Allah (he who rises at the command of God) appointed themselves to
fight the infidels and drive them from the forts they were occupying in the
country. Al-Kä'im bi-'Amr Allah's proclamation in 1511 marked the beginning
of the Sa'âdî dynasty. Their struggle lasted some 40 years, and was directed
partly against the Portuguese and partly against the Wattasid kings.
In southern Morocco, the Portuguese invaden were before long safe only in
their forts and, consequently, Portuguese colonization declined at an increasing
pace. The Portuguese were also threatened from the north by the fighters of
Salé who continually harried them and drove them out of Ma'murä, at the
mouth ofthe Sebü river.
During this period, the struggle between the new dynasty of the Sa'âdî and
the old Wattäsid-Marinid dynasty did terrible damage to Morocco. Fortunately,
after an indecisive battle between the Wattäsid sultan Ahmad and the shärif
Ahmad al-A'radj at Tadla in 1527, a treaty was concluded which gave the Sa'âdî
the Süs and Marlakesh and left the rest of the country in the hands of the
sultan, with Fez remaining the capital. These provisions gave Morocco 12 yean
of peace, during which the Sa'âdî organized their forces and concentrated on
the struggle against the Portuguese.
The fint major event was the capture of Agadir, which the Sa'âdî sultan, with
the help of a powerful artillery, succeeded in taking after a six-month siege. The
capture of Agadir led to the immediate evacuation of the towns of Sâfî and
Azemmür by the Portuguese in 1542. These successes made Muhammad al-
Shaykh the hero of national liberation and brought him considerable prestige
throughout the country. He could now resume his struggle against the
104
Morocco 105
Mountain ranges
Rivers
. Towns (e.g. Azemmür)
0 Sa'âdî and 'Alawite capitals
DUKKALA Names of provinces
8.1 Morocco in the sixteenth and seventeenth centuries (after M. El Fasi)

Wattäsids and attempt to reconquer the north of Morocco which they were still
governing under the terms of the Treaty of Tadla.
After defeating his brother and rival, Ahmad al-A'radj Muhammad al-
Shaykh. now the sole leader of the Sa'âdî, decided to deal with the Wattäsids
once and for all and set out to occupy Fez, possession of which would give him
supreme power in Morocco. The struggle between the old and the new
dynasties lasted some ten years and ended with Muhammad al-Shaykh's entry
into Fez on 13 September 1554. But the sultan was concerned about the threat
to his country from the Turks in Algien. In his view, the only real danger to
Morocco came from the Ottoman empire, which had subjugated all the eastern
and western Arab countries. To safeguard his country from a Turkish invasion,
he therefore decided to force the Ottomans out of Africa Two sons of the
Sa'âdî sultan, 'Abdallah and 'Abd al-Rahmän, had already taken Tlemcen in
1550, but the Turks had retaken it in 1552.
To carry through his grand design of conquering Algeria, Muhammad al-
Shaykh sought an alliance with Spain. Talks were arranged with the Spanish
governor of Oran and the two sides agreed to mount an expedition against
Algiers. Two thousand Spanish cavalry paid by the sultan were to join this
Moroccan-led operation. The Turks got wind of these preparations and the
Sublime Porte sent reinforcements to attack Oran, then occupied by Spain.
However, the Turkish fleet which was to have blockaded the Spanish fleet by
sea was recalled to fight the Christian fleet of Andrea Doria that was
threatening the Bosphorus. The sultan could now turn to the conquest of
Algeria. He began by laying siege to Tlemcen and succeeded in capturing it.
In 1557 Sultan Muhammad al-Shaykh was assassinated by an officer of his
own guard who was in the pay ofthe Ottoman government. But this made no
difference to the determination of the Sa'âdî to press on with clearing Morocco
of all foreign occupants and defending it against all incursions, even a Muslim
one. The new Caliph, Abu Muhammad 'Abdallah, who was proclaimed sultan
after the murder of his father, followed the policy laid down by his predecessor.
He was known as al-Ghälib Billäh (victor through the strength of God). He set
out to rebuild the army, obtain the most modern weapons and organize a vast
psychological campaign among the people with the help of the zäwiyas and the
leaden ofthe brotherhoods. When al-Ghälib Billäh felt himself strong enough,
he laid seige to Mazagän with a strong artillary force. Although his attempt
failed, it showed the Portuguese that they had to reckon with a genuine new
power.
The sultan felt that the Portuguese threat to the interior of Morocco was now
averted and he turned to developing his kingdom's prosperity. He encouraged
trade with European states, especially France. Al-Ghälib Billäh was one of the
greatest builden of the Sa'âdî dynasty. He constructed a great mosque in
Marrakesh and restored the Almohad mosque in Agadir Casbah. His reign was
on the whole calm and prosperous and he died peacefully in 1574.
His succession proved more difficult. Transgressing the long-established rule
whereby the family's eldest male had a prior right to succeed over the eldest
Morocco 107
son of the late ruler, Muhammad, the son of al-Ghälib Billäh, was proclaimed
sultan. The new sultan took the tide of al-Mutawakkil, but is still known by
the nickname of al-Maslukha. Mawläy 'Abd al-Mälik, the eldest of the Sa'âdî
princes and uncle to al-Maslukha considered that the throne was rightfully his.
When al-Ghälib Billáh's son was proclaimed sultan of Fez, he took refuge first
in Sidjilmäsa, then in Algiers and eventually in Constantinople. He entered the
Ottoman army and took part in the reconquest of Tunis, where he displayed
great courage. On his return to Constantinople, he found the Caliph disposed
to help him regain the throne of his ancestors. With the small army made avail¬
able to him, Mawläy 'Abd al-Mälik arrived in Morocco where he met with no
opposition, since the people were all on his side. Al-Mutawakkil was aban¬
doned by his army and forced to flee, and Mawläy 'Abd al-Mälik entered Fez
on 31 March 1576, where he was enthusiastically welcomed by the population.
Al-Mutawakkil escaped towards the south and succeeded in retaking
Marrakesh. He was again defeated but managed to escape northwards and asked
the governor of Vêlez de la Gomera to grant him refuge in his town. His
request was granted, with the permission of King Philip II, on condition that
he be accompanied by only a few memben of his family.
Mawläy 'Abd al-Mälik, once freed at least for the time being from his
nephew, began to reorganize the state and rebuild his army, and he made his
brother Mawläy Ahmad Caliph in Fez. To replenish the treasury, he did not
want to levy fresh taxes and sought other methods. He expanded the navy, had
new ships built and repaired the older ones. This policy stimulated many
different crafts and increased trade with the rest of the world, especially with
Europe, and had beneficial effects in all fields.
The battle ofthe three kings

In the sixteenth century, Portugal had a powerful empire which embraced vast
territories in America, Asia and Africa. At the time when the dethroned Sa'ädi
Sultan al-Mutawakkil took refuge at Vêlez de la Gomera, this empire was ruled
by the young King Don Sebastian, who dreamed of conquering Morocco, and
from there the whole of the Maghrib, and then pushing on eastward to deliver
Christ's tomb from Muslim hands. In his search for help to regain the Sa'ädi
throne, al-Mutawakkil had turned in vain to the king of Spain; he now
contacted Don Sebastian, who jumped at this opportunity to embark on the
expedition of which he had always dreamed and for which he had Portuguese
public opinion on his side.
When Mawläy 'Abd al-Mälik was informed of the preparations being made
by the young king, he endeavoured to dissuade him from his rash adventure.
He wrote Don Sebastián a letter which is unique in the history of diplomacy,
not only for its substance but for its sincerity, to convince him to give up his
expedition. The French archives possess the Italian translation of this document
which has been published in Count Henry de Castries' Sources inédites de
l'histoire du Maroc. In that letter Sultan Mawläy 'Abd al-Mälik also displayed
diplomatic skill, making it clear that the real aggressor in the Mediterranean
region was in fact the king of Portugal.
However, Don Sebastián continued to prepare for war and landed at Arzila
(Asila). Mawläy 'Abd al-Mälik, who understood the king of Portugal's sense of
honour, wrote to him again asking him not to attack defenceless people and to
wait for him where he was to join him in batde. Memben of Don Sebastian's
entourage, especially Muhammad al-Maslukh, advised him not to comply
because it was a trap. On the contrary, he should make haste to attack and
occupy Larache and al-Kasr al-Kablr. But Don Sebastian decided not to move
from Arzila, where he remained for 19 days until the eve ofthe batde.
The Sa'ädi forces set out from Marrakesh northwards and were continually
joined by large numbers of volunteen. In addition, the sultan had requested his
brother, the Caliph at Fez, to get to al-Kasr al-Kablr before him with the
contingents from Fez and its region, in particular the elite corps of young
archen (the Rima). When the Sa^dl army arrived at al-Kasr al-Kablr, Mawläy
Ahmad was already there and Mawläy 'Abd al-Mälik charged the archen
once Don Sebastián had crossed the Wädi al-Makhäzin to destroy the bridge
across the river so as to prevent the Portuguese from making their way to the
sea once they had been defeated.
The 'batde ofthe three kings', known among Arab historians as the batde of
Wädi al-Makhäzin and among Spanish and Portuguese historians as the batde
of Alcazarquivir, was fought on 4 August 1578.
The day of the battle

On the Monday morning, at the end ofthe month of Djumada II in the year
986 ofthe Hegira (4 August 1578), the opposing camps prepared for batde. It
was an historic day, the culmination of long centuries of conflicts between
Muslims and Christians which had begun in Spain and then spread to
Morocco.
The opposing forces prepared to attack, each employing its own tactics. The
Christian army was drawn up on the field of batde in a square, with a corps of
legionaries in front; the German troops held the right flank and the Spanish and
Italian ones the left. The cavalry advanced on each wing. There was also a
body of supporters of Muhammad 'the Flayed'. The Moroccan army was
deployed in the form of a crescent, from within which Sultan Mawläy 'Abd al-
Mälik, who was ill, directed the batde from a Utter. The cavalry manned the
tips ofthe crescent, with the infantry and artillery in the centre.
After the dawn prayer, the order was given to open fire. The mercenaries
flung themselves on the Muslims' left flank, taking them by surprise. Sultan
'Abd al-Malik then rose from his litter to urge his men on. He succumbed from
his efforts but, before he died, he gave orden for his death to be concealed,
since the batde had barely begun. As the batde raged, many men fell on both
sides. The Christians fell back and then fled towards the bridge which they
found destroyed. Many then tried to swim across the river, but were carried
away by the current and drowned.
Morocco 109
Don Sebastián stood firm, displaying enormous courage, until he fell on the
field of batde. Muhammad 'the Flayed' drowned in the river as he attempted to
flee. His body was fished out and brought to Ahmad al-Mansür, who was
proclaimed king after the victory (which gave him his name, al-Mansür, the
Victorious).
The losses on both sides were heavy and tradition has it that there were
nearly 14,000 Christian prisonen. Negotiations were entered into for the
ransom of Don Sebastian's body, but al-Mansür was too proud to accept any
offer, especially as the ransoms for the prisonen had yielded him sums which
some historians gave as the origin of his nickname of al-Dhahabî (the Golden).
He therefore returned the king's body without any payment of any kind.
The consequences ofthe battle ofthe three kings

This decisive batde had major consequences which for many centuries set their
mark not only on Islam and Morocco, but also on Portugal and Europe. The
victory imparted fresh vitality to the Sa'âdî dynasty and brought Morocco peace
and prosperity.
Morocco experienced an economic boom, especially in the sugar industry. In
the arts, encouraged by the Sa'ädi dynasty and the newly enriched bourgeoisie,
architecture developed and achieved considerable refinement. The Moroccans'
newly recovered stability and their feelings of pride produced a wealth of art,
poetry and literature.
The batde also had major repercussions internationally. There was a
substantial decline in Portuguese influence in the Muslim world, particularly in
the Gulf countries. The batde finally put an end to the threat from crusaden to
the countries of the East. The Sa'âdî victory also strengthened the power of the
Ottoman empire, and Europeans are right to consider this Moroccan victory as
a victory of all Muslims.
But this victory also resulted in the Ottomans abandoning any idea of
conquering Morocco, which remained the only Arab country outside Turkish
influence. The Arabic language was thus able to retain its purity and authen¬
ticity in Morocco.
Sa'âdî expansion in the sixteenth century

Al-Mansür reigned for a quarter of a century and devoted himself to consoli¬
dating the power of the Sa'âdî dynasty and developing his kingdom in every
possible way.
After the victory of Wâdî al-Makhäzin, he set out to liberate the towns that
were still in the hands of Europeans: Arzila (Asila), Tangier, Ceuta and
Mazagän. He succeeded in driving the occupien out of Arzila (1589), but
Tangier was not liberated until the following century. Ceuta passed into the
hands of King Philip II of Spain, who had inherited the kingdom of Portugal.
Along with Melilla and three other presidios it remains a Spanish dependency to
this day.
8.2 77ie empire of Ahmad Al-Mansür, 'the Golden ' (1578- 1 603) (after M. El Fasi)
Morocco 111
Al-Mansür's success encouraged him to look beyond Morocco's traditional

boundaries with a view to restoring the situation as it was at the time of the
Almoravids and unifying Islam. In the case ofthe expedition to the Sudan, the
real motive was most likely the expansionism common to all great powen.
Before embarking on this campaign Al-Mansür convened a council of war at
which he set out his plans. In his Histoire du Maroc, H. Terrasse writes that
'Almost everyone was against the venture, because it -was too risky and above all
because it would mean waging war against other Muslims'. The (ulamä\ the
true representatives ofthe people, also took this view. But Al-Mansür overrode
this advice,' and decided on his own to undertake the Sudanese expedition.
There is no need to describe this expedition in detail here, since all works on
the history of Morocco deal with it.
Al-Mansür's death in 1603 was followed by a period of instability as a power
struggle broke out among his successors who engaged in endless internecine
strife which drenched Morocco in blood for half a century. In all parts of the
country, adherents of religious brotherhoods rose in the name of patriotism to
restore peace and fight against the Portuguese and the Spaniards who had taken
advantage of the weakness of the last rulers of the Sa'ädi dynasty to blockade
the coasts of Morocco.
The most famous of these war leaden was certainly Abu 'Abdallah
Muhammad al-'Ayyäshi of Salé, who fought successfully against the Portuguese
at Mazagän. He managed temporarily to pacify the areas under his control, in
particular the town of Fez. Despite all his successes he never laid claim to
supreme power, since he was a true saint. He was assassinated on the orders of
his enemies, the Moriscos of Rabat, in 1641.
In 1636-7, for the fint time in the history ofthe Middle Adas, the grandson
of a shaykh univenally renowned for his learning and holiness declared himself
sultan of the whole of Morocco. This pretender's name was Muhammad ibn
M'Hammad ibn Abu Bakr al-Daläl. His grandfather had founded a zäwiya in
the Middle Adas which had become almost as prestigious a centre of Islamic
studies as Karäwiyyln University in Fez. The success of this univenity-cum
monastery encouraged in him ambitions for power, all the more so as the
weakness ofthe last memben ofthe Sa'âdî dynasty left the way open for him.
Following al-Mansür's death, ten of his descendants competed for power for
a little over half a century. Two of them managed to reign for more than 20
yean, with interruptions when their brothen or cousins took power only to
lose it subsequendy - to say nothing of the claimants from among the heads of
the zäwiyas and other adventuren. All these claimants to the throne of
Morocco made themselves out to be savioun of the unity of the country and
champions of stable government.
Mention should be made of the revolt of the Shbanäte who established a
short-lived period of rule in Marrakesh after killing the last Sa'âdî sultan. The
Shbanäte were a faction of the A'räb (Arab nomads) who had been driven out
of Cairo by the Fätimids and invaded the Maghrib in successive waves, destroy¬
ing everything in their path. In the twelfth century they had been brought into
Morocco by the Almohad Ya'küb al-Mansür. They ultimately setded on the

fertile plains of the Adantic coast, and it was they who were responsible for the
real Arabization of Morocco. In 1658-9 a caid ofthe Shbanäte by the name of
Abd al-Karim ibn Abu Bakr al-Shbäni and known as Karrüm al-Hadjdj rose up
in Marrakesh against Sultan Ahmad, known as Mawläy al-'Abbäs. He put him
to death and declared himself sultan and occupied the royal palace. This reign
lasted some ten yean and was marked by Karrum al-Hadjdj's extortion,
injustice and cruelty. Even his followen and supporters sought to get rid of hirn
and he was killed by one of those close to him. His son Abu Bakr ibn 'Abd al-
Karim was proclaimed sultan and reigned for a time until his death, 40 days
before the 'Alawite sultan Mawläy al-Rashîd reached Marrakesh.
The early sultans ofthe Alawite dynasty

The honour of setting up a strong, lasting and beneficent government was to
fall to the Sharifian dynasty of the 'Alawites, who have reigned over Morocco
for three and a half centuries. The 'Alawites are a family whose descent from
the Prophet has been scientifically established by leading Moroccan scholan.
The descendants of this family, created by al-Hasan ad-Dakhil, formed a
religious aristocracy which acquired great prestige among all the peoples of the
Tafilälet. Al-Hasan's great grandson, Mawläy al-Shärif. became very famous. He
had nine sons, one of them being Mawläy al-Shärif whom the Filali proclaimed
sultan. At this time Morocco was divided among the Dilawiyya of the Middle
Adas, who occupied Fez, Ibn Hassün, who reigned over the Süs and the High
Adas, and al-Khadr Ghaylän. who was ravaging the north-west.
In the Tafilalet itself, a family was occupying a fortress, Tabousamt, in
opposition to the 'Alawite shßrifs. There was fighting between the two sides
and, after various incidents, Mawläy al-Shärif in favour of his son Mawläy
M'hammad, who was enthroned as king of Morocco in 1640. This was the
beginning of the 'Alawite dynasty.
Mawlly al-Rashîd was not willing to acknowledge his brother Mawläy
M'hammad and left the Tafilälet. He began by wandering in the towns and the
kabllas in the south and the north, he went to the Todgha valley, to Demnat, to
the zäwiya of Dila and to Fez. Mawläy M'hammad raised an army and marched
on the Dar'a, which was occupied by Abu Hassün. He fought the latter and
drove him out of the area. He then marched on the zäwiya of Dila and gave
batde against the Dilawiyya, but was defeated. He then went and laid siege to
Fez which he succeeded in occupying in 1650. But he failed to establish
himself there and headed for eastern Morocco, occupied the town of Udjda
and advanced towards Tlemcen. The whole of western Algeria was invaded by
his troops. After signing an agreement with the Turks who governed Algeria,
Mawläy M'hammad returned to Udjda.
Mawläy al-Rashîd, punuing his plan to ascend the throne of Morocco, came
and established himself at Täzä. He now had the money and the men to punue
his plans and take power. Having conquered the whole of the Täzä area, he
Morocco 113
declared himself sultan. When his brother Mawläy M hammad learned about
these happenings he came to meet him to put a stop to his doings but he was
killed at the beginning ofthe batde. This was in 1664, a year which saw the
beginning ofthe conquest of Morocco.
Mawläy al-Rashîd then went to the Tafilälet, the cradle of the family, where
one of his deceased brother's sons had risen up against him. He fled on Mawläy
al-Rashîd's arrival and the new sultan was able to enter his native town peace¬
fully. After two vain attempts to occupy Fez, he finally entered the town in
triumph in 1666. The taking of this capital, without which no government can
maintain itself in Morocco, marked the definitive establishment of the 'Alawite
dynasty.
Mawläy al-Rashîd first set about organizing the administration. He
distributed money to the {ulamd and made them his privy councillon. He
appointed the scholar Hamdün al-Mazwär al Kädi of the capital. All these
actions won him the devotion of the Fâsîs. He spent two yean pacifying the
whole of northern Morocco, and, at the end of 1668, he prepared to reduce
the zäwiya of Dila which still constituted the only authority in the Middle Adas
area He met the troops of Muhammad al-Hadjdj al-Dilä'i in the plain of Fäsls
and inflicted an overwhelming defeat on them. He then entered the zäwiya
with his troops, treated the occupants humanely and did not shed one drop of
blood. This happened at the beginning of 1668.
After this victory, it only remained for Mawläy al-Rashîd to crush the last
pockets of resistance in southern Morocco. He fint liberated Marrakesh from
the Shbanäte. and then went on to attack the Samläliyün who had established
their power in the Süs; he took Tarüdant and the fortress of Ighll. By 1670 the
whole of Morocco was pacified and 'Alawite power finally established. A year
and a half later, Mawläy al-Rashîd met his death in a riding accident.
The story of the last rulen of the Sa'âdî dynasty is largely one of events, as
the strife that then prevailed and the instability of government did not allow
arts and letten to flourish It was only under Mawläy al-Rashîd that Morocco
resumed its cultural traditions and its social and economic achievements.
Mawläy al-Rashld built the biggest madrasa (college) in Fez, and another in
Marrakesh. He was also responsible for the construction of the bridge over the
Wädi Sebü 15 kilometres east of Fez.
On the economic side, he lent traders considerable sums to develop their

businesses and thus create prosperity for all. He took an interest in the water
problem, especially in the desert areas, and had many wells dug in the eastern
deserts, particularly in the Dar'a, through which caravans of traden and pilgrim
caravans passed on their way to Mecca.
By and large, the reign of Mawläy al-Rashîd was marked by remarkable
progress in all domains and by peace and prosperity.
The reign of Mawläy IsmäHl

Mawläy Ismä'il completed the task begun by his two brothers, that of unifying
Morocco by putting it under a single throne. He strengthened the basis of the

state and saw to it that Muslim law was extended to all parts of Morocco, in
order to give the country religious as well as political unity.
He was extremely interested in matten relating to religion and proselytiza-
tion on its behalf, and he even wrote to Louis XIV of France and James II of
England inviting them to embrace Islam. He stricdy observed the precepts of
Muslim law and led an austere fife. Christian historians have described him as
cruel, violent and despotic. This false image of Mawläy Ismä'il derives from the
stories told by European prisoners who had naturally been treated m
accordance with the fashion ofthe time that is with retaliatory violence.
Ambassadon from the sultan were sent to all European capitals and, in
neturn, delegations from the European states came to seek his friendship. He
was thoroughly conversant with what was happening in France and England,
through his intelligence agents, and showed perspicacity in politics.
He came to the throne on 16 April 1672. The notables of the towns and
kabllas came to swear allegiance to him, except for those from Marrakesh,
where his nephew Ahmad ibn Muhriz had hastened to have himself proclaimed
sultan. Mawläy Ismä'il was obliged to march against his nephew and fight him
and the kabllas in the Hauz who had joined cause with him. He defeated them,
entered Marrakesh and forgave the inhabitants for their lack of haste in
swearing allegiance to him. His nephew who had fled after his defeat started
plotting again, urging the people of Fez to revolt.
Mawläy Ismä'il marched against him and forced him to flee a second time
into the Sahara. He laid siege to Fez and secured its surrender, but later decided
to make Meknes his capital. He gave orders for palaces, houses, walls, stables,
warehouses and other large buildings to be put up there. He had gardens and
ponds laid out, to such good effect that Meknes came to rival Venailles.
Ahmad ibn Muhriz tried a third time to revolt against his uncle. He
succeeded in occupying Marrakesh in 1674 and fortified himself there. His
uncle laid siege to the town for nearly two years, at the end of which time ibn
Muhriz fled to the Süs. Mawläy Ismä'il then made his entry into Marrakesh and
this time he gave orden for those who had supported ibn Muhriz to be
punished.
One ofthe domestic political events of Mawläy Ismä'il's reign was the revolt
by three of his brothen, Mawläy al Harran, Mawläy Häshäm and Mawläy
Ahmad, which broke out at the end of Ramadan in 1678-9. The rebels were
supported by the A'it 'Atta kablla. The sultan marched against them at the head
of an imposing army and the three defeated brothen fled into the Sahara.
Meanwhile, after suffering various up and downs, ibn Muhriz joined forces
with his uncle al-Harran. When Mawläy Ismä'il learned that they were
occupying the town of Tarüdant in 1684-5, he laid siege to the town; during
the siege ibn Muhriz was killed. The royal army took Tarüdant by storm and
al-Harran fled into the Sahara. From that date, 1687, on, there was no further
challenge to Mawläy Ismä'il's rule.
Morocco 115
^k ^B ml p II
i uvm ^p?
v
~WL
8.3 Sultan Mawläy Isma'il (MiJiammad V Mausoleum, Rabat, courtesy ofthe Curator)
Mawläy Ismä'il's campaigns to recover the Moroccan towns occupied by

the Europeans
After consolidating the achievements of his two predecessors, Mawläy Ismä'il
completed the unification of Morocco by liquidating Christian colonization.
Tlie retaking of al-Mahdiyya

Al-Mahdiyya, which was then called al-Ma'murä, was one ofthe biggest ports
in Morocco. Pirates of various nations attempted to occupy it. Taking
advantage of the Moroccans' weakness as a result of the quarrels among al-
Mansür's sons, the Spaniards occupied al-Mahdiyya in 1614 and held it until
Mawläy Ismä'il decided to drive them out. He marched on the town, laid siege
to it, cut off its water supply and occupied it in 1681, taking all the Spaniards
there prisoner.
The retaking of Tangier

Tangier had come under English rule after one of their kings had married a
Portuguese princess. Mawläy al-Rashîd had attempted to return this town to
Morocco but died without having been able to do so. Mawläy Ismä'il deputed
one of his greatest generals, 'Alï ibn 'Abdallah al-Rîo to lay siege to it.
Historians agree that the English evacuated Tangier without a fight after
demolishing the towen and fortifications but opinions are divided on the
reasons for that evacuation. Moroccan historians say that the English did so
because of the severity of the siege that they were suffering. European historians
attribute it to reasons of domestic politics. Nevertheless, they acknowledge the
difficulty the English were having in repelling Moroccan attacks. Be that as it
may, the Moroccan army entered Tangier in February 1684.
Tlie retaking ofLarache

The town ofLarache had not been forcibly occupied by the Spanish enemy but
had been ceded to King Philip III of Spain by Muhammad Shaykh in return for
help in regaining the throne of Morocco which his two brothers were fighting
him for. The bargain was struck in 1610 and Larache remained under Spanish
rule for 80 years. Mawläy Ismâ'îl sent a powerful army against Larache, and
besieged the Spaniards there for over five months. The retaking of Larache,
which took place in November 1689, gave the Moroccans enormous joy, on a
par with the grief they had felt at its loss.
Tlie retaking of Arzila

The port of Arzila had fallen into the hands ofthe Portuguese at the beginning
ofthe reign ofthe Banü Wattäs. After being recovered by the early Sa'âdî kings
it was retaken by the Portuguese and eventually came into the possession of the
Spaniards. When General ibn Haddu had finished with Larache he received
orden to go and lay siege to Arzila. When the besieged inhabitants, worn out,
sued for their lives, the request was granted subject to the sultan's approval but,
Morocco 117
fearing the worst, they fled by night in their ships. The Moroccans entered
Arzila in 1691.
The armies of Mawläy Ismi'il
The Wadäya militia

The Moroccan kings raised their armies either from thekabllas of their own clan
or from allied kabllas. This was so until the advent of the Sa'ädi dynasty who
recruited their soldien from the kabllas of Beduin Arabs introduced into
Morocco in the time of al-Mansür the Almohad who had settled in the south
of the country. The Sa'ädi dynasty raised a militia known as the 'Wädaya
militia', which was dispened on their decline. During his stay in Marrakesh,
Mawläy Ismä'il had the idea of reassembling them and making them into
soldiers. These new recruits were taken to Meknes, the capital, where they
were joined with Shbanäte and Zirära. Their numben having grown, Mawläy
Ismâ'îî divided them into two groups: one was sent to Fez, and the other stayed
at Riyäd in Meknes.
Tlie Bawäkhir militia
Mawläy Ismä'il was fully aware that the strength of a nation depends on its
military power, but he feared the acquisition of excessive authority by soldiers
and their leaders. That is why he decided to establish a militia composed of
slaves.
One of the secretaries of the Makhzen. named Muhammad ibn al-Käsim

'Alîlîsh, whose father had been secretary to the Sa'adï ruler al-Mansür, told
Mawläy Ismä'il that the king once had a militia of black slaves, adding that he
possessed the register in which their names were recorded. Mawläy Ismä'il
entrusted 'Alîlîsh with the task of tracing these men, of whom there were still
many in the Marrakesh area, and enrolling them, whether they were slave or
free. Public opinion was shocked, especially the {ulamä) repositories of Islamic
law, which forbids the exploitation of free men. This issue gave rise to a long
controveny between the king and the \ilamct of Fez and to a voluminous
correspondence, part of which still exists. These are the famous letters from
Mawläy Ismä'il to Shaykh al-Isläm al-Fâsï.
Mawläy Ismä'il was convinced that he was acting in accordance with Muslim
law in this matter which he regarded as best for Morocco and Islam and he
relied on the assertions of 'Alîlîsh that these men, or at any rate their fathers,
had already been slaves in the time of the Sa'âdî. The earliest letter from the
correspondence between Mawläy Ismä'il and Sidi M'hammad ibn 'Abd al Kädir
al-Fâsï that has survived is dated July 1693. In this letter, the ruler asks Sîdl
M'hammad to study 'Alîlîsh's arguments and to state whether the reduction of
free men to slavery was in accordance with the law. The scholar no doubt
answered that the law did not allow free men to be reduced to slavery, as the
sultan's letten show growing irritation.
Firmly decided to organize the army of the Bawäkhir. the ruler wrote to the
Shaykh al-Isläm in December 1698 that he had freed all the slaves that he had
enrolled in his militia and that he held them in mortmain to defend the
territory of Islam. We do not know what Sîdî M riammad's reply was; he died
in 1703 and Mawläy Ismä'il continued to ask the iulamâ) of Fez for their
agreement. The affair had all sorts of ups and downs until 1708 when the
king finally forced them to approve the Dlwän al-Abld (Register of Slaves).
That is how the Bawäkhir militia came to be formed, and it greatly helped
in maintaining peace and security in the unified country. Mawläy Ismä'il had
established forts and citadels (kasabas) in all parts of Morocco, and they were
garrisoned by these soldien. Thanks to this powerful militia, Mawläy Ismä'il
was able to restore Morocco's strength and prestige in the eyes of the great
nations ofthe day.
Mawläy Ismä'il appoints his sons viceroy in the various parts of Morocco
This selection caused Mawläy Ismä'il serious problems as he had 500 sons and as
many daughters. Hence he could not satisfy them all. In these circumstances,
he would have done better to have adopted from the outset the solution he
finally arrived at.
In 1699 he divided the provinces among his sons as follows: Mawläy
Ahmad (nicknamed al-Dhahabi) was sent to Tadla with 300 black soldiers:
Mawläy 'Abd al-Mälik was sent to the Dar'a at the head of 1,000 horsemen;
Muhammad al-'Alem to the Süs with 3,000 horsemen; and Mawläy al-Ma'mün
al-Käbir to Sidjilmäsa. The last-named set up his base at Tizïmi, but died two
yean later and was replaced by Mawläy Yüsuf in 1701-2. Mawläy Zaydän was
sent to eastern Morocco, where he launched expeditions against the Turks,
once even sacking the amir 'Uthman Bey's palace in Mascara. His father
removed him from office, in view of the pact between himself and the
Ottoman Caliph, and replaced him with Mawläy Häfid.
Those of Mawläy Ismä'il's elder sons who did not receive viceroyships felt
slighted. Some of them tried to occupy provinces by force, such as Mawläy
Abü Nasr, who attacked his brother Mawläy 'Abd al-Mälik and seized the
Dar'a. The sultan had to send his son Mawläy Sharif to recover the province
and then assigned it to him. Mawläy Muhammad al-'Alem rose up in the Süs,
had himself proclaimed sultan and marched on Marrakesh, which he occupied.
Mawläy Ismä'il sent his son Mawläy Zaydän against him, and he fought the
rebel for two yean. Having seen the unfortunate consequences of this
experiment and the squabbles amongst his sons, Mawläy Ismä'il decided to send
to the Tafilälet all his sons who had attained the age of puberty. He set each of
them up in a house, gave them a number of palm trees and a plot to farm along
with slaves to help them in their work. It was a wise solution for the ruler had
too many sons for them all to lead a princely life in Meknes or other towns in
Morocco. In 1717 he removed all his sons from office except Mawläy
Ahmad al-Dhahabi. governor of Tadla, who had succeeded in his task since, in
20 years, there had not been a single uprising in his province.
Following this measure the country had peace and quiet and the Moroccans
Morocco 119
contributed to increasing the country's wealth by developing trade and agri¬

culture. Mawläy Ismä'il had a very long reign. He died on 21 March 1727, after
being on the throne for 57 yean.
Mawläy Ismä'Ü's successors

After Mawläy Ismä'il's death his many sons, who already during his lifetime had
fought for power in the regions, rose up to win supreme power. For decades
none ofthe claimants was successful in maintaining a strong and lasting govern¬
ment. The fint of them, Mawläy 'Abdallah, was enthroned and deposed several
times.
Towards the end ofthe eighteenth century a great king, Sïdï Muhammad ibn
'Abdallah or Muhammad III, came to the throne of Morocco. He restored
order, strengthened the power of the monarchy and made Morocco a country
respected by all nations. His primary interest was in the development of trade
and he saw to the modernization of the ports, in particular the port of
Mogador, which has since that time been called Essaouira. He also took an
interest in the Karâwiyyïn of Fez for which he drew up a whole programme of
reforms.
As far as religion is concerned, he believed in the original purity of Islam,

which rejects maraboutism or the veneration of saints. Nevertheless, as he was
on excellent terms with the shärlf of Mecca, Sourour, he did not push too far in
the direction of religious reforms that coincided too closely with the ideology
of the Wahhäbis, enemies of the shärlfs of Mecca. Nevertheless, the power of
the brotherhoods declined considerably during his reign and that of his son,
Mawläy Sulaymän.
As regards foreign relations, he concluded numerous agreements with
foreign countries. He recognized the independence of the United States of
America, suggested to Louis XV that he abolish slavery and supported the
Ottoman empire in its conflict with the Russian empire. In 1767 he expelled
the Portuguese from Mazagân, but died suddenly while preparing to lay siege
to Ceuta.
In conclusion, the reign of Muhammad III can be said to have had a major
stabilizing effect on the state and the power of the 'Alawite dynasty. He loved
peace and settled issues of domestic and foreign policy through negotiation and
dialogue. He only went to war to liberate Mazagän. His wise and realistic
policy was beneficial to the Moroccan people who, during the second half of
the eighteenth century, enjoyed security and prosperity.
Algeria, Tunisia and Libya:
the Ottomans and
their heirs
The early sixteenth century was marked by a profound crisis which put an end
to the old state structures in the Maghrib and the former balances implicit in
them. Thanks to Ottoman intervention in Algien, Tunis and Tripoli, this time
of troubles was overcome and gave way to a new order which eventually
brought some stability to the Maghrib until the early nineteenth century when
a new structural crisis erupted, heralding colonial domination.
What was the underlying significance of this crisis? To what extent did the
Ottomans help in re-establishing stability in the sixteenth century and how was
it done? It is clear that the situation of Algiers was different from that of Tunis
and that of Tunis from that of Tripoli. Similarly, the seventeenth century, a
time of hesitations and experiments, was different from the eighteenth century,
a period of relative stability.
The sixteenth-century crisis and the Ottoman solution in the Maghrib

The crisis which affected the entire Arab world at the time was a many-faceted
one: it was an economic crisis, due in part to the abandonment of this part of
the world as a major trade route, a social and political crisis consequent upon the
weakening of the unifying forces in society, and a cultural crisis arising from an
excessive attachment to the past. It was particularly dangerous because Europe
was awakening economically, politically and culturally, as the Renaissance
brought about new ideas, new ways of doing things and new technologies.
Only the Osmanli Turks, on the fringe of the Muslim world, managed to
make some adaptation to the times. Limited though it was, this Ottoman
reaction was nonetheless a solution for societies and states in full decline who
were also exposed to outside threats as the Maghrib states were at the begin¬
ning ofthe sixteenth century.
The crisis in the Maghrib

By the late fifteenth century the Maghrib was in a state of severe crisis, marked
by a dwindling population, the dislocation of the economy and society and
incurable political weakness. The reasons for this state of affain there was the
120
Algeria, Tunisia and Libya 121
alarming proximity of the desert, the juxtaposition of diverse societies that did
not always get along with one another, the weakness of the unifying elements
of society and the lack of technological and cultural progress.
The Hilälian nomads have often been blamed for the decadence of the
Maghrib. The accusation lacks subdety, but it is true that through their activities
and organization, their way of life and their warrior ethic, the kabciil (nomadic
Arabic groups or groups assimilated to the Arabs) represented an element of
weakness for the Maghrib as a whole.
In tbe course ofthe fourteenth and fifteenth centuries epidemics and famines
had reduced the total population of the Maghrib to between 3 million and 6
million. This fall in population had led to a decrease in production, already
threatened by the advance ofthe desert. In addition, the treasures of America,
which were pouring into Seville from 1503 onwards, seriously devalued the
Old World's monetary reserves and helped establish the power of those who
owned them, first the Iberians and later those who seized them in the service of
the New World economy, particularly the Dutch, English and French.
The gradual decline which had been undermining the towns of the Maghrib
since the eleventh and twelfth centuries became a rapid decay in the late
fifteenth century and during the first decades of the following century. Leo
Africanus invariably attributed this poverty which he saw everywhere, both in
the towns and in the areas under sedentary cultivation, to oppressive taxation
and the depredations ofthe nomads.
The nomads were indeed freeing themselves from the yoke of the state and
extending their control and their pastoral way of life to the greater part of the
country. They were taking advantage of the vacuum created by depopulation
and the low level of farming and were using violence in order simply to subsist.
The areas cultivated by sedentary populations shrank as they advanced and there
would be tiny fields in which cereals were grown protected by high walls.
It was in such circumstances that the centuries-old states of the Zayyänids (or
Banü 'Abd al-Wadids) in Tlemcen and the Hafsids in Tunis began to founder.
Their resources dwindled as a result of the disorganization of trade and the
decrease in taxable commodies. In addition, there were the ikta{ (concessions) of
land or taxes granted by inpecunious rulen to powerful warrior groups. Before
long the towns far removed from the centre of power would become autono¬
mous (Tripoli, Bougie and Constantine in the case of the Hafsid kingdom) and
the great confederations would reign supreme over the areas that they domi¬
nated. The Biläd al-Makhzen (country under state authority) was gradually reduced
to a narrow area around the sultan's residence and a few pockets further away.
The domination ofthe Zayyänids in Tlemcen and the Hafsids in Tunis was
already deeply undermined when the Spaniards and the Ottomans set foot on
their possessions.
Foreign intervention in central and eastern Maghrib

The designs ofthe conquerors must be seen in the fight ofthe religious passions
of the time: the crusading spirit of the Spaniards who had only just completed
the reconquista of their land; and the defence of Dâr al-Istâm, combined with the
ideal ofghäzl (conquest) of the Ottomans. Moreover, the coastal strongholds of
the Maghrib were of undeniable strategic interest to the two protagonists.
Finally, the Spanish offensive in the Maghrib came shortly after the fint arrivals
of precious metals from America which provided the Spaniards with the means
of punuing a vigorous expansionist policy.
Tlie Spaniards in the central and eastern Maghrib

From 1505 to 1574 the kings of Spain repeatedly tried to secure firm footholds
on the coasts of the Maghrib. One has only to recall the great expeditions of
Pedro Navarro in 1505 Charles V in 1535 and finally those of Don
John of Austria, who recaptured Tunis from the Turks in 1573. The results
were limited, however, as the conquest of the interior of the Maghrib and the
convenion of the indigenous inhabitants rapidly proved impossible. The
Spaniards contented themselves with occupying a few presidios (towns) such as
Oran and Tripoli and building a number of fortresses such as the Peñón at the
entrance to Algiers or La Goulette in the outer harbour of Tunis.
This policy of limited occupation had to be supported by a constant quest
for local allies. The most famous were the Hafsid sultans who, from 1535
onwards, played a subtle balancing game between the Spaniards and the
Ottomans. Similarly, the last Zayyänids of Tlemcen made on-and-off alliances
with the Spaniards until the fall of their capital to the Turks and their final dis¬
appearance in 1551-4. In general, however, the religious and cultural barrien
were such that no durable rapprochement between the Spaniards and the local
rulers was possible and the Muslim Turks took advantage ofthe situation.
Tlie Ottomans 'fight against the Spaniards

The fint Turks to wage war on the Christians in the Maghrib were the Ra'ls
corsairs. They acted first on their own account although in agreement with the
local rulen and aided by the local inhabitants. Such was the case of the
Barbarossa brothers, 'Arrudj and Khayruddïn, between about 1500 and 1519.
After the defeat and death of 'Arrudj and Khayruddïn appealed to the Ottoman
sultan, whose suzerainity he acknowledged, and so the Maghribi iyäla
(Regencies) came into being.
Thanks to the Janissaries and the weapons provided by Istanbul, Algeria was
gradually subjugated by Khayruddïn and his successon, in particular Sälah Ra'is
(1552-6) and the all-powerful beylerbey ofthe 'odjaks ofthe west', Killidj or 'Ildj
'Alï (1568-87). For 40 yean (1534-74), East Ifrikiya was the scene of a long
drawn-out struggle between the Spaniards and the Turks in which Hafsid rulen
and local leaden intervened actively, although not always to their advantage.
The principal stages in the Turkish conquest of the eastern Maghrib were the
taking of Tripoli in 1551, Kayrawän in 1557, Djerba in 1558 and, lasdy, Tunis in
1569 and again in 1574. Thereafter the situation stabilized in favour ofthe Turks
who remained mästen ofthe central and eastern Maghrib, with the exception of
the enclave of Oran, Mana al-Kabïr and the small island of Tabarka.
Hie organization ofthe Ottoman Regencies

The 'odjaks ofthe west', as the new African provinces were called, were never
completely integrated into the Ottoman political and military system. There
was no regular tribute and no direct administration by Istanbul. Initially under
the authority of a single warlord, the beylerbey of Algiers, the three Maghribi
Regencies were separated after the death ofthe all-powerful Killidj.
Each province was entrusted to a pasha appointed by Istanbul and assisted by
a dlwän (council of senior Turkish officen). The militia of the Janissaries,
numbering several thousand, was responsible for the defence of the country,
maintained a minimum of order and played an active role in the raising of taxes
and the running ofthe 'administration'. From the very start it was the mainstay
of the new regime. The conain, who came from the Greek archipelago,
Albania or neighbouring European countries, had all converted to Islam and
were regarded as Turks; they were assimilated into the caste of conqueren and
shared their privileged status.
The Makhzen (administration) fulfilled various simple functions: the raising
of taxes to meet the cost of waging war and maintaining the corps of con¬
querors in the country, maintaining order and dispensing justice; and providing
the postal service. A small staff of khudja (scribes) and accountants, a few
highranking political and religious figures and, above all, soldien of all ranks
carried out these tasks. With time, the Makhzen had to recruit civilian and
military auxilliaries within the country. In other respects, local institutions con¬
tinued to govern society with the new masters intervening now and again.
Although slight and rudimentary, the new Ottoman organization gave the
Maghrib what it needed most - the means of adapting to the modem world. It
introduced an army that was disciplined and used firearms and a relatively cen¬
tralized administration and by privateering it secured some of the money being
circulated in the Mediterranean. The new regime was generally well received by
the cities and also by the clergy, particularly the iulamâ) who favoured a strong
central power. Opposed to the Turks were all those who had benefited from
the freedom of the early eleventh century, the rural population which had its
own social and war-making organizations and found in its new masters only
hanh rule and burdensome exploitation. The Turks had to wage long wars to
subjugate the interior ofthe country and impose a minimum of order there.
The Maghrib in the seventeenth century: in search of equilibrium

In the seventeenth century, Maghribi society, litde changed by the Ottoman
episode, gradually recovered from the crisis of the previous century. Its coastal
cities and perhaps too its sedentary population even experienced a degree of
development, but it remained susceptible to grave crises (epidemics, famines,
civil wan) which shook it periodically. The Turkish ruling class underwent
some internal changes as a result of the entry of new elements and also by its
becoming to some extent rooted to the Maghrib. A number of problems arose
concerning relations with Istanbul, and Barbary privateering, which enjoyed its
golden age between 1600 and 1650. The Regencies grew further apart as their
political development followed different lines.
Seventeenth-century Ottoman Maghribi society

Ottoman North Africa comprised a variety of socio-economic groups. The
rural people were made up of sedentary or semi-sedentary farmen, nomadic
shepherds, nomadic desert camel-herden and sedentary oasis dwellen. The
urban population included principally the clergy and merchants, craftsmen, the
kulughli (descendants of locally married Ottoman immigrants) and the military
and civilian Ottoman ruling class.
The rural people belonged to various kabllas which allied to form confedera¬
tions (sajf, pi. sufuf). These decided, on a case by case basis, whether to rebel or
cooperate with the existing regime. Great blocks, subject to or allied to the
Makhzen. stood in opposition to the slba (dissident blocks).
The seventeenth century stood out from previous centuries by a greater
degree of spatial stability and a consequent reduction in turbulence, at least in
the well-watered parts ofthe Maghrib.
While most of the countryside was Arabic-speaking, substantial Berber-
speaking communities continued to survive that had taken refuge in the
mountains, from the Djabal Nafüsa in Tripolitania to the Aurès and Grand
Kabyila in the central Maghrib and the Atlas and Rif mountains further west.
They were distinguished from Arab communities by their elaborate defence
system which enabled them to resist the Turks suceesfully and preserve various
forms of Berber culture. They mosdy preserved their autonomy, refusing, for
example, to pay taxes.
The village areas, the southern oases, the wheat-producing regions under the
control of the mästen of the towns and the flat country of the towns were
occupied by a society very different from that of the mountain areas in the
hinterland. This society led a sedentary fife, was open to the market and
possessed a better defined milk (land ownership). It was influenced by the
economy and culture of the towns, and values such as submission to authority
spread more easily than in the mountain areas. Patrilineal kinship links there
coexisted with hierarchical relationships, such as those linking the owner of the
means of production to the khammä (one-fifth share-cropper) in the large
cereal-producing areas. All these traits must have been accentuated during the
seventeenth century with the advances in security, the consolidation of urban
society and the establishment of relations with European trading states which
stimulated the extension of export crops.
Thanks to their remoteness from the sea and the central authorities, and
also to the continued existence of the caravan trade with black Africa and the
East, the oases were home to better integrated societies and produced well-
entrenched oligarchies or local dynasties, such as that of the Fâ"sï in the
Fezzan.
The cities, for their part, made their presence felt throughout the Maghrib,
whether it be the coastal capitals such as Algien, Tunis and Tripoli or old
centres such as Kayrawän, Constantine and Tlemcen. Trade and handicrafts

took on renewed vigour from the first half of the seventeenth century. {Ilm
(religious knowledge) slowly revived after the crisis of the previous century,
stimulated by the relative material prosperity of the towns and the consequent
increase in the number oiwakfs (religious foundations). The Turkish authorities
encouraged this revival, for reasons perhaps more secular than spiritual.
At the top of the political and social hierarchy was the ruling class, made up,
in theory, of memben of the Turkish army but open, in fact, to other
categories. These were originally Christians who had converted to Islam and
were called Turks despite their French, English, Conican or Sardinian names.
They introduced modern European technology, especially in military and
maritime matters, and played a very active role not only in these areas but also
in various administrative and political posts. In Tunisia and Tripolitania the
Turks quickly gave their children by indigenous women (the knhtghli) the
responsibilities and privileges reserved for themselves. The Turks in Algiers
were more exclusive and refused to do likewise, thus provoking a serious
kulughli insurrection which ended in the defeat of the latter and their total
exclusion, at least between about 1630 and 1680.
In short, in the seventeenth century the ruling class began to associate with
the local élites in Tunis and Tripoli, while that in Algien remained adamantly
aben. The difference stemmed essentially from the strength or weakness of
those élites: they were powerful in Tunis by virtue of their historic traditions
and their activities and they were strong in Tripoli through large-scale trade,
but in Algien, a new town practically created by the Turks in the sixteenth
century, they were but poorly established.
The political regimes in the Regencies in the seventeenth century

The three Regencies, in principle Ottoman provinces, became largely indepen¬
dent of Istanbul from the beginning of the seventeenth century but, apart from
this common feature, they developed differently.
Developments were most rapid in Tunis. Soon stripped of any real power by
senior militia officers in the dlwan, the Tunisian pasha was nevertheless retained
as a symbol of Ottoman allegiance. However, in 1591, these semor officen had
to yield their places in the dlwân to the representatives of the Janissaries. The
military democracy thus installed did not last long: in 1598 a single Turkish
leader, the dey, seized all powers and established an autocratic regime. In about
1630 a new authority emerged in the country, the bey or commander ofthe
land troops. This office, held by a mamlük (freed man) of Genoese origin,
Murad, enabled its holder to conquer the hinterland at the expense of the large
confederations, hitherto independent. The bey consolidated his position, con¬
centrated power in his family (the Muradite dynasty) and adopted a veritable
monarchical policy in keeping with the country's traditions. He won out over
his rival, the dey, particularly during an armed conflict in 1673. But the
Muradite triumph was short-lived. The crisis of the late seventeenth century,
the weakening of the Turkish militia, the internecine strife among rival beys
and the intrigues and interventions by Algien brought this first experiment in
semi-national monarchy to an end in 1702.
A similar development occurred in Tripoli, although a little later. A dey was
appointed in 1603 and a policy analogous to that ofthe Muradites in Tunis was
adopted between about 1630 and 1672, with the bey pre-eminent, especially in
the time of Muräd al-Maltî (1679-86) and the kulughli bey Ahmad Käramänll in
1711. However, this evolution in the direction of a monarchical regime better
integrated into the country was thwarted in Tripoli by a number of factors:
interventions by Istanbul which was trying to regain real power in the country,
the omnipotence ofthe Beduin confederations, the desire for autonomy on the
part ofthe eastern province, Benghazi and the southern province, Fezzän, and,
finally, the meagreness of the country's resources. Power ultimately depended
on a fragile balance between the Turks and urban notables, maritime and land
interests, the towns and the great kabllas, and the centre and the provinces.
The regime in Algiers preserved for longest its original character as a Turkish
military province. There the pasha kept some of his prerogatives until 1659,
when the dlwän and the ägha (senior officers of the militia) seized power. This
was not for long, as a dey, representing first corsair captains then the Janissaries,
succeeded in seizing power in 1671. But the deys power remained precarious
and most of them were deposed and killed in revolts by the Janissaries. The
Algiers regime was thus a military regime which became more democratic
during the seventeenth century but only to the benefit ofthe Turkish caste, in
the absence of a strong group of indigenous notables.
In general, whatever its attitude to local élites, the regime in the Regencies
remained close to Istanbul. Their policy towards rural communities was very
harsh and sought to put the maximum pressure on their subjects. From the
seventeenth century, however, the ruling class began to use alliances with
certain local forces to impose their domination at least in Tunis. But, for most
ofthe time, force continued to prevail over any other policy.
External revenue: privateering and trade

Some colonial historiography improperly reduced the modern history of the
Magnb to the history of privateering equated with piracy. But privateering,
quite different from piracy, involved only a tiny minority in Maghribi society
and, during the latter part of the seventeenth century, had to contend with the
mercantile interests ofthe great European states and their local allies.
Maghribi privateering in the seventeenth century

A legacy of the great struggles fought by the Ottomans against their Christian
enemies in the sixteenth century, privateering became the prerogative of the
Turkish Regencies in the Maghrib once Turkey had made peace with the
Spaniards in the Mediterranean and the Ottoman provinces in the west had
acquired freedom of action. It was practised primarily by Turks, Albanians and
converted Christians, for their own account, and remained a monopoly of a
fraction ofthe Ottoman ruling class. Its aims and purposes were many: as a holy
war par excellence, it helped to justify Turkish conquest and power in the
Regencies and it brought in enormous profits from the ransoming of slaves and
the capture of Christian ships. It was an extremely lucrative activity for the
privateen and corsairs who engaged in it, for the states that took a not insig¬
nificant share of the profits and indirecdy for the whole population of the
corsair ports.
Its importance varied gready from one period to another. Beginning its
history in the Regencies in the 1580s, it reached its zenith as a result ofthe
European wars of the first half of the seventeenth century. The Regencies were
able to build up substantial fleets, based principally at Algiers and Tunis. The
second half of the century saw an irrevenible decline of Maghribi privateering
consequent upon the increasing fire-power of the European fleets and the
mercantile advances of the big Christian powen. From the 1680s onwards
France and England compelled the Regencies to respect their shipping and
their trade and privateering continued only against nationals of the small
Christian powen. From then on it was only in exceptional circumstances, such
as European wars (particularly those of the late eighteenth and early nineteenth
centuries), that Maghribi privateering could really regain its freedom of action,
and even then only temporarily.
Progress in trade between Europe and the Maghrib in the seventeenth century
Privateering never completely eclipsed peaceful dealings and it began its decline
and yielded to trade in the second half of the seventeenth century. This change
would seem to be due, in the fint pjace, to the influence ofthe major European
states that eventually imposed their views on the rulers of the Maghrib. These
latter were split between a military faction and a civilian one, favourable to trade
and prevailing over its rival, mainly in Tunis. Such were the circumstances in
which trade finks with Europe were strengthened on new bases.
The traditional transit trade was in decline, except for the trans-Sabaran links
through Tripolitania and pilgrim caravans between southern Morocco and
Mecca which crossed the Algerian and Tunisian oases, Tripolitania and Egypt.
What now became predominant, in both quantity and value, was the maritime
trade imposed by the European states.
The products exported to Europe came from the Maghribi countryside.
There was an increasing demand for cereals, in particular, in the southern
provinces of western Europe. To meet this demand and improve their finances,
the Maghribi states intervened strongly in this export trade. Exports of craft
products were of secondary importance and were mainly to other Muslim
countries, such as the chechia (red woollen bonnets) from Tunis, luxury textile
products and worked leather.
Besides the luxury products for the use of a small élite, imports included the
means of controlling the country such as weapons, coins, paper and a number
of products for the use of local craftsmen such as wools and dye-stuffs.
Without doubt the main beneficiaries of these trading relationships were the
European states and their merchants and carriers. The main aim of the great
9.2 View ofthe city and port of Algiers, by the eighteenth-century French engraver
F. A. Aveline (e Bibliothèque nationale, Paris)
9.3 View ofthe city and port of Tripoli, by the eighteenth-century French engraver
F. A. Aveline (o Bibliothèque nationale, Paris)
naval expeditions ofthe 1670s and 1680s was to reduce privateering and put
European trade on a convenient, reliable and profitable basis. This was the
beginning ofthe policy of unequal treaties.
Paradoxically, these treaties were mosdy accepted by the Maghribi authori¬
ties, not only out of fear of European fire-power but also out of self-interest, as
they derived substantial profits from sea-borne trade and could use them to
acquire European products and weapons. Moreover, some social groups
associated with European trade also stood to gain by it, as did olive-growen in
the Sahel and the ownen of the large estates where cereals were grown.
The eighteenth century: achievement or respite?

Historians traditionally see the Ottoman eighteenth century as a period of crisis
and accelerating decline. How far is this true for the Regencies or 'sandjaks of
the west'?
Maghribi societies in the eighteenth century

Unlike during the previous century, the Maghrib did not undergo any
fundamental changes. Yet some progress can be seen, such as the spread of
cereal-growing on large estates and the extension of the valuable prickly pear
well beyond the areas of Andalusian setdement where it was fint introduced.
But the means of production remained unchanged and social structures
retained their distinctive features, whether inherited from a remote past or a
recent one. The only noteworthy changes were those brought about by public
service or the growth of trade with Europe. Thus the power of certain families
of local notables at the head of rural communities, in certain religious offices
and in administrative or venal offices was consolidated. There are well-known
examples of the rise of certain kulughli families in the Algerian beyliks and of
Algerian Jewish families who played a leading role m relations between Algiers
and Europe in the late eighteenth and early nineteenth centuries. In Tunis,
large-scale lizma (farming of provinces) enabled some families to acquire wealth
and power.
The Tunisian state in the eighteenth century

In the eighteenth century it was the best-consolidated and best-integrated state.
It was still a province of the Ottoman empire, ruled by a veli (governor)
appointed by Istanbul; the Turkish caste continued to dominate both politically
and socially but in reality, Turkish suzerainty was becoming increasingly
notional. The bey of Tunis enoyed complete autonomy, the Turkish ruling
class, which had completely absorbed the kulughli, became more and more
open to local notables, and the Janissary militia was restricted to a strictly
military role. Lasdy the effects of the policy of force were attenuated by the
beys' practice of making alliances with local notables, especially religious leaden,
won over to the dynasty by a whole host of advantages.
Taking advantage of the invasion of the Tunisian countryside by Turkish
troops from Algien in 1705, a kulughli officer, Husayn b. 'Ali, had himself
proclaimed a bey, led the resistance against the invader, and took advantage of
his victory to found a lasting semi-monarchical dynasty. Restricting the Turks
to purely military roles and their representatives, the pasha and the dey, to an
honorary role, he relied on the kulughli, the Andalusians and the local notables
and managed to achieve a relatively high degree of centralization. But contra¬
dictions accumulated between the attempts at centralization and the mosdy
segmentary nature of rural society, between the removal of the Turks from
political life and the Regency's status as an Ottoman province and between the
subsistence economy and the large-scale trade in which the beylik was involved.
In 1728 the bey's nephew, 'Ali Bashä, revolted. The country split immediately
between the supporten of the revolt (the bäshiya or pashists) and those loyal to
the reigning bey (the husayniya or husseinists). The crisis dragged on until 1762
with numerous twists and turns.
The Regency then once again experienced peace until just after 1815. This
was the time of 'All Bey (1759-82) and Hammüda Pasha (1782-1814), a rela¬
tively prosperous period despite the terrible plague of 1785 and the famines of
1777-8 and 1804. Revenue from external sources rose considerably as a result
of the strengthening of trade links with Europe, the strong European demand
for food products during the revolutionary and Napoleonic Wars (1792-1814)
and lasdy the resumption of conair activity taking advantage of those wan.
This enabled the state to lighten the tax burden or at least not increase it, thus
lessening political tensions.
In addition, the bey of Tunis triumphed in wan against foreigners at Venice
(1784 Tripoli (1793 and above all Algiers whose domination was
ended in 1807.
This period of equilibrium and success which Tunis enjoyed for over half a
century ended just after 1815, with the resumption of European expansion in
new circumstances heralding colonial imperialism.
The Regency of Algiers in the eighteenth century

Of the three regencies in the Maghrib, Algiers retained an alien military ruling
class longest. Yet it still underwent some changes.
Although privateering had gready declined, it was still practised selectively
against certain Christian countries such as Spain. To safeguard their merchant
ships, the Scandinavian countries and a number of Italian states, such as Venice,
agreed to pay tribute to Algiers. At the same time trade developed with the
major European trading states of France and England which mosdy bought
cereals.
Warfare against neighbouring states paid the ruling class in Algien hand¬
somely. Interventions in Tunis in favour of pretenden in 1735 and 1756 resulted
in considerable booty and a disguised tribute paid by Tunis from 1756 to 1807.
Domestically, taxes continued to be raised through the maHalla (armed
expedition) for the benefit of the Turkish caste, which excluded the kulughli
from the militia and higher state offices. This policy aroused a great deal of
opposition: numerous Kabyle insurrections, popular uprisings in Oran led by

religious brotherhoods, and revolts in Constantine at the beginning of the
nineteenth century led by local feudal leaders.
Since the late seventeenth century there had been an increasing trend for a
single leader, the dey, to monopolize power. He was increasingly supported by
a small body of Turkish dignitaries from amongst whom he was chosen. As a
result, the Algerian regime became more stable and effective. Between 1710
and 1798 the deys remained longer in office, with Muhammad b. 'Uthmän
ruling from 1766 to 1791.
In the provinces the changes were even more marked as the beys of
Constantine, Titteri and the west, with only a small number ofJanissaries, were
forced to rely on local notables and leaden. There were even kulughli beys who
had blood ties to the leading indigenous families.
These changes were felt, although belatedly, in the capital itself. With the
support of the kulughli and the Zwawa, Dey 'Ali Khödja got rid of what
remaiined of the Janissary militia, as had the beys of Tunis and Tripoli a century
earlier. A nationalization of the Algerian regime could then have been enacted,
but the French conquest put an end to a trend which had possibly begun too
late.
The Regency of Tripoli in the eighteenth century

As in Tunis at the beginning ofthe century, in 1711 a kulughli officer, Ahmad
Käramänll, seized power in Tripoli and founded a dynasty of beys which was to
rule until 1835. The success of this family is due to several faetón: fint, there
was the length of the reigns of Ahmad, 'Ali and Yüsuf; second, there was the
existence of multiple alliances between the kulughli and the leading families of
Tripolitania; third, there was the size of the beys' revenues from external
sources. There were the direct and indirect revenues from privateering revived
after 1711 and again between 1794 and 1805, and there were also the revenues
from the large-scale trans-Saharan trade across the Fezzän and the Medi¬
terranean trade with Leghorn and the Levant.
Yet the Regency of Tripoli experienced serious problems. Fint, there were
natural disasters such as the famine of 1767-8 and the plague of 1785. Next,
there was the fact that the country's population was split into two great
confederations whose rivalry was the source of revolts and civil wars, like that
of 1791 which pitted different members ofthe Käramänll family against one
another. Finally, Istanbul did not give up its attempts to regain real power in
Tripoli. In 1793 a Turkish officer, 'Ali Burghül, entered Tripoli and drove the
Käramänlis out. When he extended his action to Djerba, in Tunisian territory,
he was driven out by the bey of Tunis, who restored Yüsuf Käramänll. But in
1835 the Ottomans landed at Tripoli and the Porte resumed direct control over
Libya.
The final source of difficulties for the Regency, and not the least of them,
lay in Christian attempts to reduce privateering. Just starting out on the
international stage, the United States made war on Tripoli from 1801 to 1805.
The war ended with a compromise peace. After 1815, it was the Europeans
who eliminated privateering and opened the country to their trade on terms
that suited them. They repeatedly demanded compensation ofthe Tripolitanian
regime for anything and everything. With its financial resources exhausted, the
Tripolitanian state found itself paralysed, and, to make matten wone, under¬
mined by revolts that it was quite unable to put down. It was easy prey for the
Ottoman empire which re-established itself there in 1835 to rule for a long
time.
Conclusion
In the sixeenth century the Maghrib expenenced a serious crisis, the fundamen¬
tal cause of which was its failure to adapt to the age of firearms, centralizing
monarchies and the treasures of America. The Ottomans provided the
countries of the central and eastern Maghrib with a solution by setting up
modern military and administrative systems there, capable of ensuring external
defence and the minimum of order necessary for common survival. But, at the
same time, they imposed an iron rule along with hanh exploitation of
resources, which contributed to the stagnation ofthe indigenous societies.
Rapidly becoming independent of Istanbul (without ever renouncing official
allegiance to the mother city), the 'odjaks ofthe west' gradually separated into
individual states which were inclined to be mutually antagonistic. There were
no fewer than ten wan between the regimes of Tunis and Algien between
1600 and 1800. These states developed along somewhat different fines: whereas
the Ottoman ruling class gradually opened up to the kulughli and local notables
in Tunis and Tripoli, in Algien it was uncompromisingly exclusionary. The
result was the emergence of semi-national monarchies in the eighteenth century
in the fint two countries and the continuance in Algien of a regime strongly
influenced by its conquering foreign origins. This separate and different
development of the three Regencies (later accentuated by the divenity of their
colonial situations) was to determine the partition of the Maghrib into separate
states down to the present day.
The history of the Ottoman Maghrib was also determined by its relations
with Christian Europe. The latter supplied many of the instruments of
modernity which enabled the states and ruling classes to exercise their hege¬
mony over the local populations. They also supplied men who brought new
technologies and ideas. Relations with Europe were thus vital to the ruling
classes ofthe Maghrib. First came relations of war, with privateering, then rela¬
tions of peace, mainly commercial. The chief beneficiary of these commercial
relations was European capitalism, but the states of the Maghrib and their local
allies also benefited, at least so long as they were able to defend their interests
against their European protagonists until about 1815. That date clearly marks
the end of one period and the beginning of another that of exclusive
domination by Europe.
Senegambia from the sixteenth
to the eighteenth century:
o evolution ofthe Wolof,

Sereer and Tukuloor
Introduction
Senegambia, which comprises the basins ofthe Senegal and Gambia rivers, lies
between the Sahara and the forest. Until the fifteenth century it remained a
dependency of the states of the Sudan and the Sahara but its opening up on its
Atlantic side with the arrival of the Portuguese gave it its full geopolitical
importance as an avenue for the penetration of Europe's economic and political
domination and as an oudet for the products ofthe Western Sudan.
From the fifteenth century the Portuguese trade in gold, ivory, leather and
slaves diverted the trade routes from the interior towards the coast and, in the
coune of the sixteenth century, precipitated the break-up of the Jolof
confederacy and the rise of the Denyanke kingdom in the Senegal valley and
the kingdom of Kaabu in the Southern Rivers.
During the seventeenth century the division ofthe coast into Dutch, French,
English and Portuguese spheres of influence coincided with the development of
the slave trade which remained throughout the eighteenth century the corner¬
stone of the Atlantic trade. The Atlantic slave trade brought violence and
strengthened the warlike and arbitrary character of the ceddo (warlord) regimes,
exemplified by the reigns ofthe damel-teen of Kayor and Bawol and the satigi of
Futa Toro. In the face of ceddo violence, Islam formed the sole bulwark against
the arbitrary rule of the aristocracy. At the end of the seventeenth century the
adherents of Islam took up arms in the Marabout War. Their defeat was
followed by the success ofthe three glorious revolutions in Bundu, Futa Jallon
and Futa Toro during the eighteenth century. The opposition between the
Muslim theocracies and the ceddo regimes thus forms the background to the
history of Senegambia suffering the consequences ofthe Adantic trade.
The Portuguese monopoly and re-drawing of the political map in the

sixteenth century
Senegambia is the sector of the African coast most open to the west. Long a
dependency of the Sudan and the Sahara, in the sixteenth century it was
134
135
Senegambia
T Porlendick
Krffa
ISLANDS
O 100 miles
I h
10.1 Senegambia from the sixteenth to the eighteenth century

subjected to the influence of the Atlantic with the arrival of the Portuguese,
whose trade gave rise to profound changes and, in particular, the re-drawing of
the political map of Senegambia as early as the middle of the sixteenth century.
The Portuguese trade

Senegambia, a dependency of Mali, was soon dominated along the River
Gambia by the Mande Juula (Dyula) who, through a series of stages, linked the
Niger bend to the kola, iron and indigo trade from the forest areas. The Mande
conqueren thus founded the kingdom of Kaabu south of the River Gambia
which, in the name of Mali, came to dominate all southern and part of
northern Senegambia.
The succession crisis that followed the mansa Sulaymän's death in 1360
facilitated the creation of the Jolof confederacy whose ruler, Njaajan Njaay,
extended his rule over all of northern Senegambia between the Senegal and
Gambia rivers. Jolof s hegemony was soon undermined, however, and finally
broke up in the middle of the sixteenth century following the invasion led by
Koly Tengella who left the Malian Sahel with many Fulbe to setde in the high¬
lands of Futa Jallon. After 1490 the many companions of Koly or his son headed
north and founded the Denyanke dynasty of Futa Toro in the Senegal valley.
The appearance of the Denyanke kingdom coincided with the arrival of the
Portuguese who were the fint Europeans to explore the African coast. They
established themselves at Arguin around 1445 with the aim of diverting
towards the Atlantic the trade of the Sudan and Senegambia which, tradition¬
ally, had been directed northward across the Sahara.
The Portuguese, working out of the Cape Verde Islands, participated in the
trade along the coast and at the mouths ofthe Senegal and Gambia rivers in the
form of cabotage. They established themselves firmly in the Southern Riven
and in the Gambia because of the commercial importance of this region. The
Portuguese trade in gold, ivory, hides, spices and also slaves was grafted on to
the old inter-regional trade in kola nuts, salt, cotton goods, iron and indigo.
The Portuguese were also soon heavily engaged in the major trading centre of
Wuli at the starting-point of the caravans that finked the Gambia northward
towards the upper Senegal and eastward towards the Niger bend, thus obliging
Mali, then in decline, to turn increasingly towards the Adantic to market its
gold.
The Sudanese gold trade, which now overflowed in the fain of Wuli and
Kantora, was a major factor in detaching the Bambuk and Bure area from its
links with the Niger bend and the Sahara and making it an integral part of
Senegambia. Leather was Senegambia's second export item after gold. In
addition there were ivory, wax and above all slaves for which Senegambia was
the fint and main source of exports by sea to Europe in the sixteenth century
and, later, to the New World.
Moreover, from the start, on the basis of the slave trade, the Portuguese
developed a plantation economy in the Cape Verde Islands which enabled
them to produce sugar, cotton and indigo. The lançados or tangomàos (the Afro-
Senegambia 137
Portuguese) gradually asserted themselves as the indispensable middlemen

between European trade and Senegambia which by the second half of the six¬
teenth century was undergoing profound changes.
The re-drawing ofthe political map ofthe states of Senegambia in the

second half of the sixteenth century
Despite its relatively minor importance, in the sixteenth century Portuguese
trade brought about an economic and political crisis in the Southern Riven
and accelerated the break-up of the Jolof confederacy in northern Senegambia.
The crisis occurred earliest in the region between the River Casamance and
the Rio Cacheu where numerous Cape-Verdean traden were operating. Here
the Bainuk and Kasanga, who were highly skilled at weaving and dyeing,
bought cotton from the Cape Verde Islands. The Bainuk also became pros¬
perous through their agricultural produce, which was needed both by the
European residents and for the provisioning of slave cargoes. The favourable
position of the Bainuk provoked conflict with the lançados who, in 1570,
exploiting the rivalry between the Bainuk and the Kasanga, sought help from
the mansa Tamba, king ofthe Kasanga. The conflict only ended in about 1590,
with the death ofthe mansa Tamba.
From the beginning, the slave trade began to upset the situation in the
Southern Rivers in a more enduring way. The Mande, who specialized in
large-scale slave-hunting, consolidated the power of Kaabu which thus came to
control all the land between the River Gambia and Futa Jallon. Similarly, the
inhabitants of the Bijago Islands organized themselves so as to participate
actively in the hunt for slaves on the mainland. While the women devoted
themselves to farming, fishing and house-building, the men made almadies, the
famous boats that made up a veritable war fleet and enabled them to spread
terror in the Southern Riven. Thus the mosaic of peoples in this area were the
fint victims ofthe oveneas slave trade. However, it was the minority groups in
the Tenda area, the Bassari, the Koniagui and the Badyaranke, living between
the highlands of Futa Jallon and the Southern Rivers, who were the main
victims ofthe slave trade.
The political evolution of the Southern Rivers lineage states was blocked by
the pressure of neighbouring peoples and by the violence engendered by the
hunt for slaves which induced defensive and isolationist reactions. This
blockage continued practically until the nineteenth century and the colonial
conquest.
Kaabu became the dominant power in the region after the final decline of
the empire of Mali. It controlled the Bainuk and Beafada trading network to its
own advantage and also seized the Mande principalities along the River
Gambia. The Maane and the Saana, who formed the ruling Naanco dynasty in
Kansala, strengthened their warlike character and from the beginning symbol¬
ized the rule of the ceddos, who dominated political life in Senegambia during
the era ofthe oveneas slave trade.
The sixteenth and seventeenth centuries marked the apogee of the kingdom
of Kaabu which replaced Mali throughout the Southern Riven until the
triumph of the theocratic revolution in Futa Jallon at the beginning of the
eighteenth century. The expansion of Kaabu coincided with the development
of the oveneas slave trade and with the takeover by Europeans of trade in the
Southern Riven. The initiative in this inter-regional trade, so vital for the
whole of Senegambia, passed from the indigenous peoples to the Portuguese
who turned the Beafada, Bainuk and above all the Mande and the Bijago into
mere slave-hunten or broken in the Atlantic trade towards the interior.
Northern Senegambia between the Gambia and Senegal rivers also under¬
went a far-reaching and lasting re-drawing of its political map in the course of
the second half of the sixteenth century.
The Jolof confederacy, already shaken by the massive invasion led by Koly
Tengella, broke up completely under the impact ofthe Portuguese trade which
favoured the coastal provinces. Thus, Amari Ngoone, the darnel of Kayor, after
his victory over the Buurba Jolof, declared the independence of the Kayor and
became its first darnel. His example was followed by the provinces of Waalo, at
the mouth of the Senegal river, and Bawol in the south-west. Jolof was thus
much reduced in size, deprived of direct relations with the Adantic trade and
cut off too from the trans-Saharan trade by the powerful Denyanke kingdom of
Futa Toro.
The ceddo monarchies that emerged from the break-up of Jolof introduced
violence not only into relations among the states of Senegambia but also into
political and social relations within each state. The same process also gave birth
to the kingdoms of Siin and Salum, both of which finally freed themselves from
the Jolof empire during the sixteenth century.
At one point Amari Ngoone tried to impose his hegemony by annexing
Bawol and part of Waalo. He took the title of damel-teen, thus inaugurating a
long series of short-lived unions between the kingdom of Kayor and the
kingdom of Bawol. But the duel between Kayor and Bawol soon prevented
the darnel from achieving his ambition of unifying the former provinces of the
Jolof confederacy under his rule.
This setback, however, favoured the rise to power of the Denyanke king¬
dom of Futa Toro. The satigi took advantage ofthe break-up ofJolof to extend
his domination over most of northern Senegambia. The Denyanke dynasty thus
reached its apogee at the beginning ofthe seventeenth century during the reign
of Samba Lamu. Futa Toro now controlled both the trade between the Sudan
and the Sahara and the European sea-borne trade.
The partition ofthe Senegambian coast and Muslim reaction in the

seventeenth century
At the beginning of the seventeenth century the Portuguese monopoly on the
coast of Africa was successfully challenged by the arrival one after the other of
the Dutch, the English and the French. The European powen engaged in sharp
Senegambia 139
competition, creating spheres of influence in Senegambia jealously protected by

fortified trading posts along the coast.
These trading posts served above all as entrepôts for the slave trade which,
by the second half of the seventeenth century, had become the main activity of
the European powers. The slave trade, with its corollary of manhunts,
engendered violence in relations between states and the militarization of power
and promoted the advance of militant Islam. Thus, by the end of the seven¬
teenth century, a widespread marabout-\ed movement had developed which
sought to unify the states of the Senegal valley to fight the noxious effects of
the overseas slave trade.
The trading posts and the partition ofthe coast

The re-drawing of the political map went hand-in-hand with the recrud¬
escence of violence between states and the development of the slave trade. It
also coincided with the arrival ofthe Dutch, the French and the English, whose
presence on the Senegambian coast was consolidated during the second half of
the seventeenth century.
In 1621 the Dutch established themselves on Gorée, followed in 1659 by
the French at Saint-Louis at the mouth ofthe River Senegal, while the English
built Fort St James on the estuary ofthe Gambia in 1651. The Portuguese were
thus progressively eliminated from northern Senegambia but retained, in
addition to their permanent base in the Cape Verde Islands, the trading posts at
Cacheu and Bissau. But they were already being forced to share the rich
market ofthe Southern Riven with the new European powen.
The fort at Saint-Louis enabled the French to control all trade along the
River Senegal between February and May, and the building of Fort St Joseph
in Gajaaga at the end ofthe seventeenth century ensured a French monopoly as
far as the upper Senegal at the gateway to the Sudan trade.
Gorée Island, which held the monopoly of trade along the petite côte, was
fint occupied by the Dutch, then taken by the Portuguese in 1629 and 1645,
re-taken by the English in 1667 and finally by the French in 1677. Thus, from
Gorée, the French traded with Kayor at the factory at Rufisque, with Bawol at
the factory at Portudal and with Siin at the factory at Joal. Further south, they
encountered competition from the Portuguese and above all from the English.
Fort St James controlled all the trade along the Gambia as far as the Barakunda
Falls. Each Mande principality along this river, which is navigable all year
round, was a potential port for the dominant English trade.
Senegambia was thus enclosed all along its seaboard by a series of fortified
trading posts whose main function was to divert the region's trade towards the
ocean. These permanent structures were designed to protect each sphere of
influence against competition from other European powen and prevent the
states of Senegambia from uniting against the monopoly of European trade.
The massive presence of the Dutch, the French and the English, following
that ofthe Portuguese in Senegambia, was closely linked with the emergence
of colonial mercantilism. In their fever to accumulate money wealth, the
Adantic powen threw themselves into the conquest of the markets of Africa,
Asia and the New- World.
Each of the European powers had companies which had been granted a
monopoly of seaborne trade: the Dutch West India Company, set up in 1625,
the French West India Company, set up in 1664, and the Royal African
Company, set up by the English in 1672. These companies symbolized the rise
of nation-states in Europe and reflected the competition among these powen
for the conquest of markets.
The Dutch, who had been the fint to upset the Portuguese monopoly, were
eliminated from the Senegambian coast by the English and the French in 1677,
except for Arguin and Portendick, where they remained until the first half of
the eighteenth century. Senegambia then remained divided between the
French sphere of influence from Saint-Louis to Gorée and the English sphere
of influence in the Gambia, where the French and the English vied for the
Southern Riven with the Portuguese. Possession of the Gambia was of vital
importance because of its proximity to Europe at a time when the development
ofthe plantations was accelerating and tripling the demand for slaves bound for
the French and English West Indies between 1651 and 1700.
Saint-Louis, Gorée, St James, Cacheu and Bissau were transformed into
entrepôts where slaves brought from the interior down to the coast awaited
slave ships to take them to the New World. Senegambia was important in the
early days of the slave trade, in the sixteenth and seventeenth centuries, as the
great markets ofthe Gulf of Guinea and Angola were only opened up later.
It is not possible to give a quantitative estimate of exports from Senegambia
but it is certain that the Adantic slave trade engendered a profound political and
social crisis. This crisis in turn provoked a widespread marabout-led movement
in the Senegal valley a few years after the building ofthe fort at Saint-Louis.
The Marabout War
After the reorganization of states during the sixteenth century under the impact
of Portuguese trade, Senegambia, in the fint half of the seventeenth century,
experienced an acute crisis, which was particularly manifest in northern Sene¬
gambia.
The marabout movement, led in the name of Islam by the Moorish marabout
Näsir al-Din, was well aware of the effects of the European presence in
Senegambia, for it started in the south of present-day Mauritania where Berber
society was suffering an acute economic crisis as a result of the decline of the
trans-Saharan trade, which had accelerated since the establishment ofthe French
at Saint-Louis in 1659.
The island of Saint-Louis, by virtue of its strategic position at the mouth of

the Senegal, was drawing the valley's trade towards the Adantic, thus breaking
up the age-old complementary relationship between the nomadic Berbén of
Chamana and the sedentary agriculturalists living along the river. The trading
monopoly of Saint-Louis thus deprived the Moon not only of the slave labour
Senegambia 141
that had been used for centuries, but also of the cereals from the valley which
acted as a granary for the countries ofthe Sahel north ofthe Senegal river. This
economic crisis exacerbated the political and social antagonism between the
Hasaniyya warrion of Arab origin, the Banü Maghfar, and the Sanhädja
marabouts of Berber origin.
Berber society was caught in a vice between the southward movement of
the Hasaniyya Arab warriors and the monopoly of Saint-Louis which was
diverting the trade of the Senegal valley to its own benefit. Näsir al-Dïn raised
a religious movement, based on puritanical Islam, to save Berber society from
disintegration. This movement attempted to regulate political and social fife
according to the teachings of the sharfa (Islamic law) in its purest orthodox
form, by putting an end to the arbitrary power of the Hasiniyya warriors and
establishing a Muslim theocracy.
The proclamation of a djiltäd in the kingdoms of the river valley was
motivated by both economic and religious considerations, to reconquer the
trade in grain and slaves and to convert the peoples and purify the practice of
Islam. In 1677 the success ofthe holy war in Waalo, Futa Toro, Kayor and
Jolof was gready assisted by the state of crisis in northern Senegambia as a result
ofthe slave trade.
Nâsir al-Dïn's movement was initially opposed to the continuance of the

oveneas slave trade and strongly condemned the tyranny of kings participating
in the hunt for slaves. This opposition in no way signified a desire to suppress
domestic slavery or the small-scale trade in slaves, an age-old practice which, in
the framework ofthe trans-Saharan trade, had never precipitated such a crisis as
had the Adantic slave trade.
Islam, which served as an excuse for the marabout movement, was changing
its character. From being the religion of a minority caste of merchants and
courtiers in the royal courts, it was becoming a popular resistance movement
against the arbitrary power of the ruling aristocracies and against the noxious
effects of the Adantic trade. With the connivance of indigenous Muslims and
popular support, it swept away one after the other the ruling aristocracies in
Futa Toro, Waalo, Kayor and Jolof without meeting any great resistance. After
the defeat of the four kingdoms, Nâsir al-Dïn replaced the fallen aristocracies
with religious leaden who espoused his cause. This marked the triumph in all
of northern Senegambia of Muslim theocracies, with specific features in each of
the kingdoms.
In Futa Toro the victory of the marabouts was swift and marked by massive
and violent popular participation in the overthrow of the rule of the satigi. In
Waalo the brak, Fara Kumba Mbodji, put up strong resistance but died in battle
and the marabout party installed a puppet brak, Yerim Kode, a member of the
royal family who accepted the conditions of the theocratic system set up by
Nâsir al-Dïn.
In Kayor the marabout movement was gready assisted by the political crisis
within the aristocracy. Oral tradition supplies many details about how the
movement, led by the xaadi, espoused the cause ofthe linger, Yaasin Bubu, who
had been removed from office by the new darnel. Yaasin Bubu converted to
Islam, bringing with her part of the garmi (ruling class) and its clients. She
married the marabout Njaay Sail who, after various twists and turns, proclaimed
himself viceroy.
But the death of Nâsir al-Dîn in 1674, killed in battle against the Hasaniyya
warriors in Mauritania, precipitated the decline of the marabout movement.
This decline enabled the French at Saint-Louis to intervene direcdy to give
their support to the fallen aristocracies in Futa Toro, Waalo, Kayor and Jolof.
The French at Saint-Louis wanted to prevent the consolidation of a vast
political grouping which, under cover of Islam, might be able to dictate to
them their terms of trade. They also wanted to resume the trade in slaves
which was so vital for the plantations in the New World.
In the framework of their own interests, the French put their support behind
the brak of Waalo, Yerim Kode, who immediately deserted the marabout
movement and played a key role in the annihilation of the movement in Futa
Toro, Jolof and Kayor. Thus, by 1677 the marabout movement had been almost
wiped out and the old aristocracies had recovered their former prerogatives.
The failure of this first popular resistance to the overseas slave trade and the
arbitrary power of established authority had lasting consequences for the
development ofthe kingdoms of Senegambia.
In Mauritania, the movement's original home, the defeat of the Berber
marabouts ensured the survival of the political power of the Hasaniyya warriors
who founded the emirates of Trarza and Brakna. In the eighteenth century,
when the development ofthe gum trade enabled them to participate profitably
in the Atlantic trade, these emirates kept up constant military pressure on the
states on the left bank of the Senegal. The Marabout War known as the
Tubenan movement in European sources and Shurbuba in the Berber
chronicles had more lasting consequences than the Almoravid movement
which had started in the same area in the eleventh century.
The Chamama Berbén, attracted by the Atlantic trade towards the Senegal
valley, participated increasingly in the political, economic and religious history
of Senegambia. While the emirates of Trarza and Brakna participated in the
widespread violence in relations among states in the Senegal valley, the zwawiya
marabouts continued to forge close links with the established marabout parties in
the kingdoms of Senegambia, thus participating in the Islamic opposition to the
military powers.
This long Marabout War gave rise to a series of famines all over the region
and above all to the repression of the Muslims. Their defeat, with the
complicity of the aristocrats who were the only Africans to benefit from the
Atlantic trade, ensured the continued commercial expansion of Saint-Louis.
Political disintegration proceeded apace as a result of internal crises and wars
among the various kingdoms over who should supply the slave trade. The use
of firearms became widespread, and autocratic and military governments came
to power in all the kingdoms. Islam now formed the chief focus of opposition
to the ruling regimes throughout Senegambia.
Senegambia 143
From this time many marabout families began to leave the coastal areas and
the Senegal valley and take refuge inland, notably in Bundu and Futa Jallon,
where they endeavoured to consolidate the autonomy of the Muslim com¬
munities. The Muslim revolutions at the beginning of the eighteenth century
marked the triumph of militant Islam.
The impact ofthe overseas slave trade: ceddo rule and Muslim
revolutions in the eighteenth century
The Adantic trade, and particularly the slave trade, accentuated the crisis in the
states of Senegambia throughout the eighteenth century. In an atmosphere of
widespread violence, the ceddo aristocracies strengthened their warlike character
and imposed centralized monarchical rule. In reaction against the arbitrary rule
of the aristocracy, Muslim communities formed enclaves within the states or
organized revolutions as in Bundu, Futa Jallon and Futa Toro.
The strengthening oí ceddo rule and political crises

The Wolof kingdoms of Kayor, Bawol and Waalo and the Sereer kingdoms of
Siin and Salum experienced the same strengthening of monarchical power.
Kayor in the reign of Lat Sukaabe Fall (1695-1720) was the supreme example
of the evolution of the ceddo regime towards autocracy. Lat Sukaabe Fall took
advantage of the unrest in Kayor to bring together the two crowns of Bawol
and Kayor under the tide of damel-teen. He enforced a royal monopoly over the
sale of slaves and the purchase of firearms and strengthened monarchical rule by
eliminating the Dorobe and Gelowar royal branches to the benefit of his own
meen (maternal family), the Geej.
Aware ofthe danger that the marabout movement represented, he introduced
institutional reform to ensure integration of the marabouts into the political
system. He appointed serin lamb (holden of new tides) who became agents ofthe
central government responsible for the defence of the frontien and soon adopted
the military ways ofthe ruling ceddo party. Convenely, the serin jakk rejected any
compromise with the rulen and, while devoting themselves to religious
activities and teaching, continued to focus the discontent ofthe peasant masses.
Lat Sukaabe Fall's reign marks the final advent of the warlords in the Wolof
and Sereer kingdoms. Lat Sukaabe Fall imposed himself through his skill at
settling internal political disputes to his own advantage and also because of the
royal monopoly on arms imports. As ruler of both Kayor and Bawol he could
now dictate his terms to the trading posts at Gorée and the factories at
Rufisque and Portudal, which considerably harmed French commercial
interests. On the death of Lat Sukaabe Fall, the French took advantage of the
succession crisis to prevent the reunification of Kayor and Bawol under the
authority of a single ruler.
In Waalo too, the French at Saint-Louis intervened in the succession crisis
among the three royal families - the Tejek, the Loggar and the Joos. This inter-
ventionism came at a time when the economic situation was changing owing
to the growing importance ofthe gum trade. The French wanted to attract this
trade to factories on the Senegal river as they were facing competition from the
Dutch and the English in the factories at Arguin and Portendick on the
Mauritanian coast. The fint gum war between 1717 and 1727 had lasting
consequences, particularly on the development ofthe kingdom of Waalo.
After the failure of de la Rigaudière's expedition in 1723 to recapture the
Mauritanian factories from the Dutch, Brué, the officer in charge at Saint-
Louis, sought an alliance with the beech, Malixuri, kangam (provincial chief) of
Roos beecio. The aim of this alliance was to induce Alichandora, the amir of
Trarza, to give back the fort at Arguin to the French and also to act as a
counterweight to the enmity shown by the brak of Waalo and the darnel of
Kayor. In 1724, supported by Saint-Louis, Malixuri rebelled against Yerim
Mbanik, brak of Waalo. But, after the failure of mediation between Saint-Louis
and Alichandora, he lost the company's support and was defeated by the brak
Yerim Mbanik. By 1734 the latter, who had a good army, had become one of
the most powerful kings in the region. This enabled his successon, his two
brothen, Njaag Aram Bakar (1733 and above all Naatago (1756-66), to
exercise hegemony over the neighbouring kingdoms, particularly Kayor, which
had been undermined by famine and seven years of civil war.
The policy of hegemony punued by Waalo was ended by the English who
occupied Saint Louis from 1758 onwards. The brak, Naatago Aram, who now
controlled access to trade on the Senegal repeatedly demanded an increase in
the dues and the price of slaves. In 1764 he twice blockaded Saint-Louis's
trade. The English reacted by giving assistance to the darnel of Kayor, Makoddu
Kumba Jaaring, who succeeded in recovering most of the territory annexed
from him by Waalo.
The English governor, O'Hara, took advantage of Naatago Aram's death
finally to break the power of Waalo. His chief concern was to get as many
slaves as possible out of the region for his own plantations in the West Indies
and he supplied arms to the Moors who overran all the Senegal river area. In
1775, in the kingdom of Waalo alone, the English took more than 8,000 slaves
in less than six months. The over-supply of slaves on the market was such that
in the streets of Saint-Louis a slave was being sold for one piece of cloth.
This tremendous drain of manpower coincided with the beginning of a long
civil war lasting almost 29 years, during which two royal families, the Loggar
and the Joos, tried to recover the power which the Tejek family had
monopolized since the early eighteenth century. But Waalo was feeling
pressure from the Trarza Moors who were intervening regularly in the
succession disputes and completely destroying the power of the kingdom,
which was henceforth incapable of punuing an independent policy.
The Sereer kingdoms of Siin and Salum underwent a development similar to
that of the Wolof kingdoms. But whereas Siin, being small, tried to close itself
off from external influences, Salum, being larger, with a cosmopolitan popula¬
tion and an advantageous commercial position on the River Salum, partici¬
pated in the slave trade and expanded toward the Gambia.
Senegambia 145
Futa Toro, under Denyanke rule since Koly Tengella, experienced a

situation identical to that of the Wolof kingdoms. The lack of detailed rules for
succession to the office of satigi encouraged war among various claimants. In
1716 Bubakar Sire called on the Moroccans for help, thus giving them the
opportunity to interfere in the affain of Futa Toro which from then on was
forced to pay the muudal horma (tax payable in grain). In addition, Futa Toro
took part in the struggle between Alichandora, amir of Trarza, and Brakna.
In 1720 Alichandora, driven out by his powerful neighboun, the Ulad
Dellim, sought the help of the sultan of Morocco. But the Moroccan troops,
the Ormans, subsequendy acted quite independendy and put all the countries
along the Senegal river to fire and sword. They ended up splitting into two
factions, one of which allied with Trarza and the other with Brakna.
Alichandora, defeated in 1722 by the faction allied with Brakna, took refuge
with the beecio, Malixuri, in Waalo. From this time onwards the Ormans inter¬
vened actively in the numerous succession crises in Futa Toro during the first
half of the eighteenth century. From 1721 to 1724 there was total confusion,
until the legendary Samba Gelaajo Jeegi seized power in 1725 with the help of
the Ormans of Gaidy and the commander of Fort St Joseph.
Samba Gelaajo Jeegi (ruled 1725-31) is the prototype ofthe warlord whose
exploits still fuel the legendary tales ofthe griots of Futa Toro. He was the ceddo
leader par excellence, who, with his army of sebbe equipped with firearms,
waged 45 battles during his reign to the sound of bawdi peyya yiyan (blood
drums) and dadde yiyan (war songs/blood songs).
The saga of Samba Gelaajo Jeegi is magnificendy evoked in the two versions
published by Amadou Ly and Amadou Abel Sy. The ceddo epic is still
marvellously sung by the sebbe in their war songs, gumbala and lenngi. The
gumbala, a hymn to dash and courage, is the epic song of death in which the
ceddo assumes his destiny as a warrior, his faithfulness to his anceston and to the
ethic of his caste. The lenngi, sung solely by sebbe women at weddings or
circumcisions, are heroic songs summoning up contempt for death and the
protection of honour. But the saga of Samba Gelaajo Jeegi is unfortunately
evoked out of its true historical context which was dominated by the violence
ofthe overseas slave trade.
Two factors were responsible for this state of permanent violence. First,
Morocco, with its army of Omans, intended to control the Moorish emirates
tied into the Adantic trade circuit. Second, the French trading post at Saint-
Louis was active throughout the Senegal valley, with the aim of extracting as
many slaves as possible from the region. Samba Gelaajo Jeegi fint allied with
the Moors but later sought a rapprochement with the French at Saint-Louis to
shake off the Moroccan yoke. In 1725 he asked the company for firearms and
also to build a port in his capital at Jowol. He even tried to protect French
interests against pillaging by the Moon. It was perhaps this alliance that enabled
him to reign uninterruptedly from 1725 to 1731 despite the claims of his two
rivals, Bubu Musa and Konko Bubu Musa.
However, Konko Bubu Musa seized power with the help of the tunka of
Gajaaga. After an exile in Bundu, Samba reconquered power between 1738

and 1741 but he remained a prisoner of his Orman and Moorish allies and died
in obscure circumstances. According to tradition, he died at the treacherous
hand of his wife suborned by his enemies during his second exile in Bundu.
In 1752 the new satigi Sube Njaay, was in turn driven out by Yaye Hoola
and his warrion who ravaged Bundu with the help of Xaaso and the Ormans.
Futa Toro seemed to be at its nadir, as the satigis followed one another in
bewildering succession to the sole profit of the Moors who dominated the
country. The conditions were now ripe for the success of the Muslim revolu¬
tion in 1776.
Gajaaga, on the upper river, had also been integrated into the Adantic trade
circuit since the end of the seventeenth century. But despite the dynamism of
Soninke traden who were the main supplien of the Niger bend in salt and
European goods and of western Senegambia in cotton goods, Gajaaga suffered
the same political and social crises linked to the slave trade and the invasion by
the Ormans. The political crisis began in about 1700 with the struggle between
the tunka Naame of Maxanna and his cousin, Maxan of Tamboukane, and was
extended in 1 730 by the war between Gey and Kammera. Tensions multiplied
and led to a series of civil wars between 1744 and 1745 which finally destroyed
the unity ofthe Soninke confederacy.
Litde is known about the development of the kingdom of Kaabu which
dominated southern Senegambia until the victory of the Muslim revolution in
Futa Jallon. But the power of Kaabu was based on trade in slaves. In about
1738 the mansa of Kaabu was delivering 600 slaves a year to the Portuguese
alone, while the Southern Rivers, under Kaabunke control, were exporting
thousands of slaves. Kaabu strengthened its hold over the coastal provinces
while also raiding its inland neighboun, the Bajaranke, Fulakunda, Koniagui
and Bassari. With Biram Mansa, who died in about 1705, Kaabu seemed to be
at the height of its power under the leadenhip of the Naneo aristocracy. But
here too, dissension among the three royal lineages, Sama, Pacana and Jimara,
was the source of numerous civil wars. This situation explains the success ofthe
holy wan later led from Futa Jallon and Bundu and the internal Muslim
revolutions against the Soninke state of Kaabu.
Muslim revolutions in the eighteenth century

The military defeat of the marabout movement led by Nâsir al-Dïn in the
second half of the seventeenth century was followed by the spread of under¬
ground action by Islam against ceddo rule and the disastrous effects of the
overseas slave trade.
Within the states controlled by the powerful military aristocracies, the

Muslim communities strengthened themselves under the leadenhip of
influential marabout families. Linked to one another through a long chain of
religious, political and economic ties, they embarked either on creating new
states or seizing power where they were through violence and the proclamation
of a holy war.
Senegambia 147
Thus, at the end of the seventeenth century, Maalik Sy founded the Muslim
theocracy of Bundu which was followed, at the beginning of the eighteenth
century, by the Muslim revolution in Futa Jallon under the leadenhip of
Karamokho Alfa. The second half of the eighteenth century saw the triumph of
the Torodo marabout party led by Sulaymän Baal in Futa Toro, the stronghold
of the Denyanke regime. These three successes demonstrate the continuity and
solidarity ofthe marabout movement throughout Senegambia.
Tlie Muslim revolution in Bundu
The repression directed against the marabout after the defeat of Nâsir al-Dïn led
to the mass departure of Muslims from Futa Toro to Bundu where Maalik Sy
founded the first Muslim theocracy on the borders of Senegambia in about
1690.
Maalik Sy was one of that group of Muslims who had received their
religious education in Pir or Kokki in Kayor, which had close connections -with
the Berber zäwiyas. After travelling through Senegambia he finally settled on
the borden of Gajaaga with the permission ofthe tunka of Ciaabu who granted
him a piece of land. Maalik Sy, settled in this cosmopolitan area, took
advantage of the weakness of Gajaaga to declare a holy war. His religious
prestige, and the military organization he forged with the help of Muslims,
most of whom had come from Futa Toro, enabled him to create the theocratic
state of Bundu.
Maalik Sy supported the Jaxanke marabouts whose commercial interests were

constandy threatened by the pillaging of the Gajaaga aristocracy. He took the
tide of almamy, which is the Fulfulde version of al-Imäm, which had earlier
been adopted by Nâsir al-Dîn. Although he did not take a direct part in the
Marabout War, he achieved some ofthe political and religious objectives ofthe
marabout party.
Tlie Muslim revolution in Futa Jallon

The success ofthe Muslim revolution in Bundu was followed a few yean later
by a revolution in Futa Jallon. The fate of the mountainous massif of Futa
Jallon, which over the centuries had become a place of refuge for the Jallonke,
the Soso and the Fulbe, was transformed by Koly Tengella's invasion and,
above all, the development ofthe Adantic trade. This major crossing-point was
economically revitalized by the existence of large herds of cattle belonging to
Fulbe herdsmen who had been flooding into the area since the fifteenth
century, drawn by the abundant grazing areas in the highlands, and had become
the richest and most powerful group.
The Muslim revolution -was not at all, as some have suggested, an ethnic war
between Fulbe herdsmen and Jallonke setded farmen. Tradition clearly demon¬
strates the multi-ethnic character of this revolution, initially led by twelve
Fulbe marabouts and ten Mande marabouts, who were certainly of Jaxanke
origin. Conversely, the movement was confronted with opposition from the
leaden of Kafu, who were Jallonke, and also from non-Muslim Fulbe, who
lived in the bush with their cattie. The Muslim Fulbe, who wanted to abolish
catde taxes, formed an alliance with the Mande Juula or Jaxanke, whose trading
way of life had always been associated with the practice of Islam, to create a
vast political grouping capable of ensuring the security of the population
threatened by the overseas slave trade. The marabout party was in fact consoli¬
dated by the participation of numerous Fulbe who could put their catde wealth
to profitable use in the framework of gradual sedentarization. The trade in
cattle and hides gave them economic power while Islam gave them the
ideology necessary for the construction of a new political and social order.
Thus, after the victory of the marabout party in the holy war against the
various ruling Janonke aristocracies, the Muslim leaden created the Confedera¬
tion of Futa Jallon under the leadenhip of Ibrahima Sambegu, known as
Karamokho Alfa, the head of the Sediyanke lineage of the Barry family of
Timbo, who took the tide of almamy. The confederation was divided into nine
diwe (sing, diwal) (provinces) whose heads bore the tide of alfa and were
appointed from among those who had contributed to the liberation of the
country. From the beginning the power of the almamy, with his seat at Timbo,
was limited by the wide autonomy granted to the heads of the provinces, and
also by a council of elden acting as a parliament at Fugumba, the religious
capital.
The Muslim theocracy in Futa Jallon was thus the outcome of a series of
military campaigns between the marabout party and the leaders ofJallonke Kafu.
After the famous batde of Talansan, which sealed the victory of the marabout
party, the holy war was continued by the attempt to convert to Islam the non-
Muslim populations of the Futa Jallon massif. The Muslims met opposition
from the Fulbe herdsmen, who had been leading a nomadic way of life in the
area for centuries and were hostile to Islam, which they saw as synonymous
with sedentarization and political and economic control. All this explains why
hostilities lasted so long and why the theocratic regime was so slow to
consolidate itself during the fint half of the eighteenth century.
On the death of Karamokho Alfa in about 1751, the almamyship devolved
on Ibrahima Sory, known as Sory Mawdo (Sory the Great). The religious
leader of the ¿Jihad was thus followed by the commander of the army who
involved Futa Jallon in an aggressive policy against neighbouring countries. As
with the kingdom of Dahomey or the Asante confederacy, the evolution of
Futa Jallon is incomprehensible outside the context of the oveneas slave trade.
These kingdoms, originally formed in reaction against the noxious
consequences of the hunt for slaves, ended up participating in that trade. Islam
then became just one ideology among many to maintain and consolidate the
power of the ruling aristocracy.
Ibrahima Sory Mawdo, assisted by the leader of the Jallonke kingdom of
Solimana, thus engaged in a series of wan against his neighboun to procure
slaves and booty. But the alliance was defeated in 1762 by Konde Burama, king
of Sankaran, who occupied Timbo. It took an outbunt of national energy to
stop Konde Burama's army at the gates of Fugumba and it was only in 1776
Senegambia 149
that Ibrahima Sory finally defeated the king of Sankaran. His victory consoli¬
dated his power and he imposed the authority of the military faction over the
religious faction until his death in 1791.
The death of Sory Mawdo gave rise to much political confusion. It is from
this time that the system of alternating rule dates between the two families of
the Alfaya, descendants of Karamokho Alfa, and the Soriya, descendants of
Ibrahima Sory Mawdo. This duality considerably weakened the central
government, but despite the weakness inherent in the political system, the
kingdom of Futa Jallon was able to preserve its independence until the colonial
conquest and even to extend its frontiers. But the new regime lost its
revolutionary character as the marabout party transformed itself into a religious
and military aristocracy actively participating in the overseas slave trade. As
elsewhere, the trade in slaves became a monopoly of the state which supervised
the trade routes and organized caravans to the coast. Islam became a mere
pretext for the hunt for slaves among the infidels on Futa Jallon's frontien.
Non-Muslims, captured in vast numben, were either sold on the coast or
simply kept in runde (slave villages). Situated between the Bambara states and
the coast, Futa Jallon participated in raids or bought slaves for its domestic use
and sold the surplus to obtain European goods and the salt needed for its
pastoral economy. In the eighteenth and nineteenth centuries this trade intro¬
duced a large number of slaves from various areas into Futa Jallon: Bambara,
Kisi, Jallonke, Fulbe, Bassari and Koniagui.
The internal history of Futa Jallon was marked by the formation of a
hierarchical society based on Islam as the ideology of power. The Muslims had
all the rights of free men while non-Muslims had servile status within the new
society governed by the ¿haría. The predominance of the Fulfulde language
and culture should not obscure the true dynamism of internal development
which was marked by the existence of distinct social classes.
Beyond the basic distinction between rimbe (sing, dimo), free men, and
maccube, slaves, within the dominant society of free men there was a hierar¬
chical ranking reflecting relations of inequality and exploitation. Among the
rimbe there were at the top the las li (the aristocracy ofthe sword and the lance
and the aristocracy ofthe book and the pen), descendants ofthe great marabout
families who had launched the holy war and monopolized power. Next there
was the great mass of free men whose condition derived from their position in
relation to the ruling class. Classed among lowest free men, the bush Fulbe,
slow to convert to Islam, had their cattle as their only wealth and were subject
to endless taxes and labour dues. But it was the development of domestic
slavery, closely correlated with the Adantic slave trade, that was the major
feature of the evolution of Senegambian societies during the eighteenth
century.
The practice of domestic slavery doubtless lay behind the cultural revolution
in Futa Jallon where the marabout and political class, freed from agricultural
work, could devote itself to teaching. The new regime set up many Ku'ranic
schools throughout the country. A strong political and social organization
henceforth based on the sitaría and the prohibition of the sale of Muslims
enabled Futa Jallon to avoid anarchy and depopulation. The theocratic state
thus enjoyed a degree of stability.
The marabouts soon translated the Ku'rän (Qoran) into Fulfulde to facilitate
the religious instruction of the masses. The outcome was not only a plentiful
and rich literature in Fulfulde but also a deeper Islamization of the masses.
Thus, the Islam of cities such as Timbuktu and Jenne became, through the
Muslim revolution in Futa Jallon, a popular Islam which subsequently inspired
the formation of a string of theocratic states throughout West Africa.
77ie Muslim revolution in Futa Toro
Islam triumphed in Futa Toro during the second half of the eighteenth century
under the leadenhip of the Torodo marabout party led by Sulaymän Baal and
'Abd al- Kädir. Here, even more so than in Bundu and Futa Jallon, there is a
clear line between the Torodo movement and Nâsir al-Dïn's movement in the
late seventeenth century. At the same time, however, the Torodo movement,
working closely with the Moorish zäwiyas, took much of its inspiration from
the success ofthe djihad in Bundu and Futa Janon.
The succession crisis initiated in about 1716 by Bubakar Sire continued
thoughout the eighteenth century, plunging Futa Toro into insecurity and civil
wars. The situation worsened when the new English governor, O'Hara,
facilitated the occupation of Futa Toro by the Brakna and Trarza Moors. In
these circumstances the Torodo revolution was directed not only against the
Denyanke regime but also against the domination by Brakna and the sale of
Muslims as slaves. The Torodo party, led by Sulaymän Baal, was able at once to
win a military victory against the Ulad 'Abdallah at Mboya and thus put an end
to the muudul horma. After establishing its authority in central Futa Toro, the
Torodo party put an end to several centuries of Denyanke domination and, in
July 1776, forbade all English trade with Galam in reaction to the ravages
perpetrated by O'Hara to procure slaves.
The Torodo victo coincided with the death of its famous leader, Sulaymän
Baal. His successor, 'Abd al-Kädir, once elected almamy, borrowed many
ceremonial practices from Futa Jallon while maintaining some of the traditions
of the Denyanke regime. He carried out a redistribution of bayti (vacant lands)
while confirming the rights of the powerful Torodo families over most of
central Futa Toro. The three 'Abe' families the Bosseyabe, the Yirlabe and
the Hebbyabe - provided most ofthe great jaggorde (the Council of Electon).
Nevertheless, 'Abd al-Kadir consolidated the new regime and extended its
religious influence beyond the frontien of Futa Toro. His success gave rise to
high hopes of change and increased the tension between the Muslim reformers
and the ceddo ruling classes of Waalo, Jolof, Kayor and Bawol.
In 1786 the Torodo regime embarked on the conquest of Trarza where 'Abd
al-Kadir wanted to impose his authority and the payment of tribute, as he had
on Brakna. He defeated Trarza, whose amir, Ely Kowri, was killed in battle.
After this victory, 'Abd al-Kädir righdy considered himself the Commander of
Senegambia 151
the Faithful, the legitimate heir of Nâsir al-Dïn. His ambition was thus to
impose Islamic law on the rulers of Waalo, Jolof and Kayor and extend his rule
over the upper river.
But in 1790 the new darnel of Kayor, Amari Ngoone Ndeela, renounced the
submission made by his predecesson to Futa Jallon and hanhly suppressed all
attempts at independence on the part of the reformers in the Muslim enclaves
in the province of Njambur. 'Abd al-Kadir then organized a great military
expedition with the purpose of colonizing Kayor. This expedition ended in the
disaster of Bungoy where the great army was beaten through the scorched-
earth tactic organized by Amari Ngoone Ndeela. Numerous Futanke were sold
to the slaven and 'Abd al-Kädir was held prisoner in Kayor and later sent back
to Futa Toro.
With the defeat at Bungoy, 'Abd al-Kädir's authority began to wane. He was
now challenged in Futa Toro by two influential memben of the jaggorde. The
powerful Torode family of Thierno Molle, hostile to the religious strictness of
the Muslim leader, obliged the almamy to leave his capital for Kobbilo, on his
own lands, while the new unlettered princes, 'Ali Sîdi and 'Ali Dundu, imposed
themselves as sole intermediaries between the central governent and the
western and eastern provinces of Futa Toro.
The internal power struggle coincided with the development of hostilities
between Futa Toro and the trading post at Saint-Louis, whose trade on the
river was interrupted between 1801 and 1803. But in 1806 a new agreement
confirming that of 1785 was signed between the two parties, as the stoppage of
trade was hurting both parties.
'Abd al-Kadir then embarked on an expedition to the upper river to put
down the ravages ofthe almamy Sega, at the expense ofthe marabouts of Bundu.
He had Sega executed and appointed his own candidate, Hammâdî Pate, thus
precipitating an alliance between Hammâdî Aissata, the unsuccessful but popular
claimant of Bundu, and the king of Kaarta. Deposed by the jaggorde, 'Abd al-
Kädir allied with Gajaaga and Xaaso but was killed in 1807 by the combined
forces of Bundu and Kaarta with the connivance of the second generation
Torodo party. His death opened the way for the triumph of the jaggorde who
could henceforth impose an almamy devoted to their cause and retain a wide
degree of autonomy in their respective chiefdoms. As in Bundu and Futa
Jallon, the marabout party, initially made up of learned men, gave way to a
political system in the hands of a warrior aristocracy with no religious learning.
'Abd al-Kädir's failure to impose Islam as the state ideology in the Wolof
kingdoms was largely offset by the considerable progress made by indigenous
marabout parties. Growing numben of Muslims tried to challenge ceddo violence
from within. In Kayor in particular, the defeat at Bungoy led to the departure
of large numbers of Muslims from the province of Njambur for the Cape Verde
peninsula where they helped to found a theocracy under the leadership of Jal
Joop. From there they encouraged the Lebu opposition and Kayor separatist
movement. After several yean of resistance the marabout party won its
independence, the first territorial breach in the kingdom of Kayor.
Conclusion
The evolution of Senegambia from the sixteenth to the eighteenth century was
profoundly influenced by the impact of the Adantic trade which made black
Africa dependent on the dominating power of Europe. During this period the
Jolof confederacy disintegrated and gave way to the kingdoms of Waalo, Kayor,
Bawol, Siin and Salum, while the Denyanke kingdom became dominant in the
Senegal valley. In the Southern Rivers, Portuguese trade ruined the inter¬
regional trade and facilitated the rise of the military power of Kaabu which
took over from the declining empire of Mali.
The dominance of the slave trade in overseas exchanges resulted in the
seventeenth century in the partition of the coast into spheres of influence and
the building of fortified trading posts. The slave trade strengthened the violent
character of the ceddo regimes which in turn gave rise to a widespread marabout
movement hostile to the military aristocracies. After the failure of Nâsir al-Dln's
movement (1673 the adherents of militant Islam organized themselves in
Bundu, Futa Jallon and Futa Toro.
At the end of the eighteenth century, however, the theocratic states
gradually lost their revolutionary character just when Europe was thinking of
abolishing the slave trade whose role in the accumulation of finance capital was
diminishing. Europe then tried to integrate Senegambia into the capitalist
system as a direct periphery of the European centre for the supply of raw
materials for industry. Senegambia, already ravaged by the profound crises of
the era of the slaven, stood no chance of resisting the military conquest
embarked on by Europe in the second half of the nineteenth century.
The end
ofthe Songhay empire
The collapse ofthe Songhay empire
Reasons for the Moroccan invasion
When Mawläy Ahmad al-Mansür acceded to the Sa'adi throne after the batde
of Wädi al-Makhäzin. his victory over Portugal placed him among the great
defenden of Islam, and the money paid him to ransom his Christian captives
made him one ofthe most influential figures on the world scene.
As Caliph, Imam and Prince ofthe Faithful, his ambition was to bring all the
Muslim peoples together in a 'single same way of thought' and to revive the
tradition of the djihâd. Thus the profits he planned to derive from the Saharan
salt deposits at Taghäza would be devoted to replenishing the Bayt al-Mal
(Treasury), and slaves won in the conquest of Songhay would be set to work in
the navy that would go and attack the infidels.
But these lofty motives did not exclude far more tangible ones the gold
and slaves of the Sudan. Slaves were needed in the sugar industry in southern
Morocco. As for Sudanese gold, supplies had dwindled rapidly since the rise of
the Songhay empire in the Niger bend. In addition, trade between Morocco
and the Sudan was still threatened: by the Portuguese, who in 1565 had tried
to reach Timbuktu via Senegal, and by the Turks, some of whose moves
suggested that they were seeking to extend their supply lines towards the
southern Maghrib. Finally, the Sa'adis' hopes focused on Taghäza faded as the
Songhays developed the new salt deposits at Taghäza al-Ghizlän (Tauodéni).
In 1582 al-Mansür seized the oases of Tüwät and Gurära. This occupation of
the oases was officially presented as a measure to restore order, but was really a
fint step towards the conquest ofthe Sudan. In 1583 the king of Borno, May
Idris Alaöma, gave al-Mansür an unhoped-for opportunity to realize his
ambitions. Fearing a Turkish advance into his territories from the Fezzän, he
asked for al-Mansür help to fight the non-Muslim groups 'on the borden ofthe
Sudan'. The Moroccan ruler agreed after extracting from the king of Borno a
bay'a (an act of allegiance) in due form. The following year a Moroccan
expeditionary force entered the Adantic Sahel in the direction of Senegal, but
153
was obliged to turn back in circumstances which are not clear.

The order to attack the Songhay empire was almost given in 1586, but in
view of the difficulties presented by the operation, al-Mansür gave himself a
few years to equip his army, collect all possible intelligence on the Askiya's
empire and finally convince his own leading citizens of the soundness of his
plan.
Tondibi and the causes of Songhay's collapse

On 30 October 1590 an impressive Moroccan column set out from Marrakesh
under the command of Djüdar Pash. It crossed the High Adas and went down
the Dar'a valley to Ktawa, where it entered the Sahara. On 1 March 1591, after
a forced march of 60 days, the Moroccan army reached the banks ofthe Niger,
and eleven days later arrived at Tondibi, some 50 kilometres from Gao, the
Songhay capital.
The askiya Ishäk II belatedly mobilized his troops. He neverthless faced the
invader with a considerable force but, against the Moroccans' guns, the
Songhay troops were crushed after a day of heroic resistance. And so collapsed
the last great Sahel empire whose rulen, absorbed in their domestic quarrels,
had failed to recognize the extent of the Moroccan threat.
Since the fall ofthe great askiya Muhammad al-Hadjdj in 1529, the court at
Gao had been the scene of relendess struggles among various pretenders to the
throne. Five yean before the Moroccan invasion the Songhay empire was
practically divided into two as a result of the rebellion of the balama al-Sadduk,
whose headquarten were in Timbuktu. Ishäk II put down the revolt, but did
not have sufficient time to reunite the country.
Economically Songhay had for several decades been suffering from the
advene effects of Portuguese trade on the coast. In addition, the military
reverses in Dendi, Borgu and Mossi country, together with the loss of Taghäza
in 1585, exacerbated the economic and social difficulties.
Finally, the empire built up by Sonni 'Alî and the askiya Muhammad
stretched over a vast area but it lacked an ethnic and socio-cultural framework
which would give it genuine unity. Paradoxically, the empire's centre of gravity
was neither in Gao, the political capital, nor in the Songhay hinterland, but in
conquered territory, at Timbuktu and Jenne.
The setting up ofthe Moroccan pashalik

In pursuit ofthe remains ofthe Songhay army, Djüdar Pasha marched on Gao,
which had been deserted by in inhabitants. Ishäk II did not make any attempt
at a counter-offensive and preferred to negotiate with Djüdar the conditions of
his return to Morocco. The Moroccan pasha, disappointed at Gao's lamentable
state and aware that his own men were in poor physical shape, was prepared to
accept the askiya's suggestions, but al-Mansür was not. He recalled Djüdar and
replaced him with the second-highest ranking officer in the Moroccan army,
Mahmud b. Zarkun, who was given the task of conquering the whole of the
Sudan.
The end ofthe Songhay empire 155
03
"n
S"
*3
S)
i1-
«
a;
«
IJ
;n
S
-^
u
a
t>
U
-r)
a
«
C
t-l
«J
J3
t:
o
Z
G
«J
«j
>
0
o
r/5
C
Ü
TS .3
r.
m
r^
4J -Ö
¡5 ^
>n
H
<1)
-s*
G
^
<s
oq -c
*2
ín ru
->
«» P
e* O
P
t
:*-,
U
-s* *-u
ex» ^
rJ ¿*,
o
</3 ûi
^ ©
r^
"i
ri oo
Mahmud Pasha set about destroying Songhay political power. He seized the
traditional capital of Kukya, drove Ishäk II from the country, laid a fatal trap for
his appointed successor, Muhammad-Gao, and then endeavoured to wipe out
the last pockets of Songhay resistance in Dendi (1592-4). Zarkun then returned
to Timbuktu to destroy the educated classes as a political force: dozens of
iulamä) were slain or sent into exile in Morocco.
Mahmud b. Zarkan was killed in 1594 in an ambush laid by Songhay resis¬

tance, and the task of completing the occupation of the area was again
entrusted to Djüdar. This quickly proved impossible in the face of the fierce
opposition from Fulbe, Bambara and Manden grouped around the mansa
Mahmud. A modus vivendi was established between the Moroccans installed in
Jenne and the main peoples of the area who accepted, although 'only in
words', the Moroccan occupation.
Al-Mansür's soldiers were forced to limit themselves to the effective occupa¬
tion of a few river ports, where they installed kasabas (permanent garrisons).
Most of them were on the river line from Jenne to Timbuktu, while on both
banks ofthe Niger lay vast areas scarcely touched by Moroccan influence.
The pashas did not try to change the local administration. Whenever an
indigenous chief was appointed he had to be approved by the pasha who also
granted investiture to the kadis and the imams in the big towns as well as the
Fulbe ardos and the Tuareg amenokals. The Moroccan representatives, like the
askiyas before them, only rarely interfered in the choice of candidates.
The Moroccan garrisons were not closed on themselves: they were not
living in enclaves or fortified camps. Al-Mansür believed the fruits of conquest
would be short-lived if pacification were not followed by colonization of the
country and its setdement by people from the Maghrib who might take
permanent root there. Thus the Sudan saw the arrival ofthe Guish people from
Süs and the Haha along with Ma'kil and Djusham elements whom al-Mansür
was anxious to get rid of. From 1599 onwards the legionaries of Christian
origin who had come with Djüdar were sent back to Morocco.
Sudanese politics to the end ofthe eighteenth century
The post-imperial experience ofthe peoples ofthe Niger bend:

general features
In the climate of uncertainty and insecurity resulting from the fall of the
Songhay empire, political power was within the reach of any who could
effectively defend and protect their peoples. The imposing states of the past
were to be replaced by a mosaic of kingdoms and princedoms limited to an
ethnic group, a clan, a town or even a string of large villages.
The political leader was fint and foremost a warrior, influenced more by
local tradition than the univene values transmitted by Muslim scholars. Islam,
which had made such a brilliant contribution to the building of the Sudanese
The end ofthe Songhay empire 157
empires, now temporarily ceased to play an important political role, but it

continued to be carried ever farther afield by Jula traders who travelled the
trade routes from the Sahel to the forest.
Timbuktu, Jenne and the Arma

After 1618 Morocco stopped appointing the chief officials of the pashalik or
sending military reinforcements and the last survivon of al-Mansür's army and
the descendants, the Arma, were left to their fate. They remained legitimate
masters of the Timbaktu area until the beginning of the nineteenth century.
Although its military power was much reduced, the pasha state proved
remarkably durable and retained its structures virtually intact until the advent of
Shaykhü Ahmadu. And yet, before the foundation of the Fulbe empire of
Macina, there were forces in the Niger bend capable of wiping out the
remnants of the old Moroccan colony. There were the Bambara of Segu, the
powerful Tuareg confederations of the Kel-Awlhmiden and above all the Kel-
Tadmekket who, despite their overwhelming victory over the Arma in 1737,
never thought of seizing power in Timbuktu. Similarly, the Kunta only entered
Timbuktu on the eve of its conquest by the Fulbe in 1825
At first there seemed to have been no special rules governing appointment to
the various offices in the pashalik, including that of pasha itself. But from the
middle of the seventeenth century, when the first generation of locally-born
Arma came to power, rules for the transmission of authority began to emerge,
based on the rotation of the chief offices among the three great Divisions to
which all the Arma belonged.
This system led to frequent interregnums and short 'reigns'. In addition, as
the same individuals would be called to power more than once, a number of
great families or lineages emerged, distinguished from the rest of Arma society
by their political power and economic influence. There thus came into being a
'ruling class', described in the chronicles as a 'class of chiefs'.
Between 1646 and 1825, 145 pashas were appointed in Timbuktu. Most of
them belonged to the three lineages of the Tazarkini Mubarak al-Dar'i and al-
Zar'i. It was to this last lineage that the pasha Mansur b. Mas'üd al-Zar'i belonged:
he seized power by force in 1716 and plunged Timbuktu into a reign of terror.
In 1719 the people rose up and drove the pasha and his leghas (henchmen) from
the town and restored the old Arma political system with its chronic instability.
In 1766 the Arma elected as pasha Ba-Haddü b. Abü-Bakr al-Dar'i, who
performed the unusual feat of remaining in office for more than eight years,
but, after his death, the town remained without a pasha for 18 years. In 1794,
when the great Divisions agreed on a pasha, their choice, al-Mustafä al-
Tazarkini, restored all the former offices of the pashalik. Under his successor,
Abu Bakr b. Ahmad al-Dar'i, the office of the pasha became hereditary and his
two sons, Muhammad and 'Uthmän, succeeded him. The Ka'ld Uthmän was
the last pasha of Timbuktu.
By the end of the seventeenth century the pashalik had begun to break down
into a number of more or less autonomous units grouped around the great
kasabas of Gao, Bamba, Timbuktu and Jenne. While continuing to recognize the
authority of Timbuktu, each garrison elected its own leaden quite independendy.
The garrison at Jenne was as independent as the other kasabas, except when a
particularly enterprising pasha came to power in Timbuktu, as happened in
1767, when Ba-Haddü Pasha himself appointed a new governor of Jenne. The
commercial links and political relations between Timbuktu and Jenne were
never interrupted and the two kasabas repeatedly rendered services to each other.
But in 1796 the Scot Mungo Park was told in Segu that Jenne officially
belonged to the Bambara kingdom though in fact it was governed by the
'Moon'. Should this be taken to indicate that the Bambara exercised a
'protectorate' over Jenne? Local sources tend rather to confirm René Caillie's
assertion that Jenne lived 'alone and independent' until it was conquered by the
Fulbe of Macina in 1818-19.
The Songhay of Dendi

After being driven out of Gao and losing two kings and dozens of members of
the imperial family within a few months, the Songhay, led by the askiya Nuh,
managed to halt the Moroccan advance in Dendi. Resorting to ambushes, they
held off Zarkun's troops and he was killed in Bandiagara. His successor, Mansür
b. 'Abd al-Rahmän Pasha (1595-6) made them pay dearly for their victory.
Nuh, defeated and forced to leave some of people in the hands of the
Moroccans, withdrew to Dendi while the Moroccans appointed an askiya for
the Songhay who remained under their rule.
The Songhay of Dendi broke down into several kingdoms, although they did
succeed in maintaining their unity untd the middle of the seventeenth century.
In 1630 they signed a peace treaty with the Moroccans who subsequently began
to interfere in their internal affain and arbitrate in their succession disputes. But
they retained their freedom until the beginning of the nineteenth century, des¬
pite strong pressure from the Fulbe and Tuareg nomads from Liptako and Aïr.
The Bambara kingdoms of Segu and Kaarta

After the disintegration of the western marches of Songhay, the Bambara peasants
along the Niger were dangerously exposed to Fulbe and Arma incunions.
Divided into several kafus whose Marka and Muslim leaden maintained reason¬
ably amicable relations with Jenne and Timbuktu, they appealed for protection
to the traditional hunter confraternities or to clans specialized in warfare.
In the middle of the seventeenth century the Bambara revolted against their
Marka leaden and it is probably at this time that the Kulibali clan made their
mark; it would later found the kingdoms of Segu and Kaarta.
In Segu, Biton Kulibali (c. 1710c. 1755) imposed his authority with the help
of his ton-dyon (captives or former captives), but no sooner had he established
his position than, around 1739, his kingdom was attacked by the Joola of
Kong, who remained in the country until 1745. Biton Kulibali, now
weakened, spent the rest of his life consolidating his kingdom by getting rid of
competition from his Masa-si cousins who were setded to the north-west of
The end of the Songhay empire 159
Segu in the region of Murdia. In 1753-4 he carried the war into their own
territory, defeated them and captured their leader. Following this defeat the
Masa-si, led by Sebamana, moved into Kaarta where they established their rule.
After Biton Kulibali's death, the kingdom of Segu went through a long
period of anarchy which ended only in 1766 when a new dynasty rose to
power founded by Ngolo Diarra. The excellent relations between him and the
Azawad Kunta and their leader Shaykh al-Mukhtär. seems to have made him
treat Timbuktu with circumspection.
Ngolo was succeeded by his son Monzón (cl 789 1808), who was the real
organizer of the kingdom of Segu. Like Biton Kulibali, he had to deal with
rivalry from the Masa-si, who seized Nyamina on the Niger, thus cutting one
of Segu's main supply routes. Monzon's reply was terrible: after liberating
Nyamina he moved on to Kaarta, sacked the capital, Guemu, and forced the
Masa-si king to flee to Guidimakha. Monzón then attacked the Awläd M'Bark
Moon in the Nioro region who had refused to help him in his war against
Kaarta.
Monzón died in 1808, followed three yean later by Shaykh al-Mukhtär.

After the demise of these two leaden, the Niger bend experienced a time of
troubles until the advent of Shaykhü Ahmadu, troubles for which the Fulbe
and Tuareg forces were largely responsible.
Fulbe and Tuareg

One ofthe chief consequences ofthe collapse ofthe Songhay empire was social
disorganization.
From the beginning of the seventeenth century the Saharan nomads were
advancing towards the Niger valley and the lake area south of Timbuktu. In
Macina the Fulbe achieved complete preponderance. The Fulbe withstood
Moroccan attacks from Timbuktu and Jenne, preserving their independence,
and stepping up their migration towards Futa Jallon in the west and Liptako
and Hausa country in the east.
Also contemporary with the Moroccan occupation was the expansion of the
Tuareg of Adrar the Kel-Tadmekket and the Kel-Awllimiden. While the
former remained in the background until the end of the seventeenth century,
the latter soon brought their weight to bear on the eastern part of the Niger
valley, especially between Gao and Dendi. Interfering in conflicts among Arma
chiefs and pillaging the farming villages along the river, they became a scourge
for Timbuktu. The eighteenth century was full of confrontations between the
Tuareg and the Arma.
In May 1737 the amenokal Oghmor ag Alad overwhelmed the Arma forces
at Toya, inflicting considerable losses. In 1770 the Tadmekket laid siege to
Timbuktu, reducing the population to starvation, after a group of Arma had
murdered their amenokal Habatlt. The town was only saved by the intervention
of the shaykh of the Kunta, who managed to reconcile Bä-Haddü Pasha
Habatlt's successor, amenokal Hammiyuk. An agreement was reached between
the two parties in August 1771, but the Tuareg broke it and the shaykh
withdrew his moral support from Hammiyuk and set up a rival, thus causing
the Tadmekket to break up into two branches. At the same time he won the
confidence of the Awllimiden who took advantage of the disunity among the
Tadmekket to try and extend their rule over the Saharan nomads of the
Timbuktu region.
The Awilimiden thus became the chief support of the Kunta who, as the
Arma grew weaker, succeeded in filling the political vacuum left in the region
and limiting the effects of the resulting anarchy. But until the founding of the
Fulbe empire of Macina they took care not to assert any political rights in
Timbuktu or to drive out the last vestiges of Arma power. Thus Kawa ag
Amma, the powerful amenokal ofthe Kel-Awllimiden, followed the old custom
and, in July 1796, went to Timbuktu to receive investiture at the hands ofthe
pasha, Abu Bakr.
Western Sudan and the outside world
Despite Morocco's gradual withdrawal after the death of al-Mansür, the pashas
of Timbuktu remained loyal to the last sultans ofthe Sa'adl dynasty. The Friday
khutba (sermon) was recited every week in the name of the rulen in Marrakesh
who took care to announce their accession to the throne to the pashas of
Timbuktu and the heads of the garrisons in Gao and Jenne. When the 'Alawite
dynasty seized power the Arma renewed their allegiance to the Moroccan
throne, fint to Mawläy al-Rashld and then to Mawläy Ismä'il.
But unlike the Sa'adis, the 'Alawites paid little attention to the Sahel regions
along the Niger. Their policy was focused on Mauritania rather than the
Sudan. Mawläy Ismä'il supported the amir of Traza and occasionally sent more
or less regular troops, the Ormans, to the Senegal valley, where they terrorized
the inhabitants until the 1720s. Once masters of Futa Toro, they brought strong
pressure to bear on the inhabitants of the upper Senegal valley and the captives
they took there went to swell the ranks of the famous black army formed by
the sultan of Morocco.
With the coming to power of Sultan Sîdî Muhammad (1757-90), Morocco's

policy in the Sudan made a fresh start, with the revival of trade across the
Sahara. Like the last Sa'adi rulen, the 'Alawite sultan referred to himself as
'sovereign of Gao and Guinea' in his correspondence with European govern¬
ments. This was not mere boastfulness, according to contemporary evidence.
The fact is that the political evolution of Timbuktu in relation to Morocco is
comparable in many respects to the evolution of the Barbary regencies in the
seventeenth and eighteenth centuries, when Constantinople's sovereignty,
though now nominal, continued to be significant.
Economic and social evolution
Natural calamities and human environment
By the end of the eighteenth century Western Sudan was no longer the
dazzling and prosperous country of which al-Mansür had spoken in 1591.
The beginning of the seventeenth century brought a string of calamities

(drought, food shortages, epidemics and famines) which destroyed harvests and
decimated populations, exacerbating tensions between nomads and sedentary
peoples. In 1639 there was famine in the Jenne region, the granary ofthe Niger
Bend. From the central delta it spread all over the bend for four consecutive
yean. The resulting distress was probably the origin of the social movement
that preceded the rise ofthe Bambara kingdom of Segu,
There were few yean without crises in the eighteenth century. After a
seven-year food shortage from 1711 to 1718, Western Sudan experienced one
of its wont famines ever starting in 1738. It affected the whole ofthe Sahel and
much of the Maghrib. In 1741 the scourge was combined with a plague
epidemic. The effects of the famine began to fade after 1744 but plague
remained endemic, breaking out now and again until 1796.
The land between Timbuktu, the Great Lakes and the Niger was the
meeting-place of many ethnic groups, and for much of the year became a
mosaic of peoples and a contact zone between two ways of life pastoral and
rural sedentary. But because good grazing land was also the best arable land, this
contact often produced conflicts, all the more serious because there was no
political force in the area strong enough to limit the incunions of herden at the
expense of farmen' fields, and vice vena. The pressure was greatest at the end
of the hot season, when the first rains brought the peasants out into the fields:
any delay or disorder in the withdrawal of the nomads was enough to produce
a serious threat to the harvest.
Local production and regional trade

In general, agriculture receded in the areas bordering the desert. But in some
places occupation of land for agricultural purposes intensified, as is shown by
the Bambara migrations in Bara, the beginnings of sedentarization among the
Fulbe of Macina and the permanent setdement of some Tuareg groups around
Timbuktu.
Furthermore, thanks to the incomparable advantages ofthe Niger as a means

of communication and the extension of traditional trade networks, inter¬
regional trade flourished.
In Jenne, food production was supplemented by contributions from the sur¬
rounding region of San: groundnuts, flour from the baobab tree, shea, honey,
dried onions, pimentos, beans and fonio. There were also cotton, indigo, senna
for making soap, wool, wax, iron from Bendugu and, of course, the two main
imports from the forest area, kola nuts and gold.
In the Timbuktu region the areas of cultivation were near Lakes Tele, Oro
and Fati and along the backwaten. Wheat-growing spread after the arrival of
the Moroccans in the Sudan but the Arma and the Maghribi merchants in the
big towns were the only people who ate bread. Among the industrial crops
produced were tobacco and gum arabic, collected in large quantities around
Gundam and Lake Faguiban.
The commercial life of Timbuktu was underpinned by numerous craftsmen:
salt-processon, weavers, tailors, tannen, goldsmiths and makers of agricultural

implements and weapons. Each craft was the monopoly of an ethno-professional
corporation: all dyers, for example, came from Sansanding while the Arma had
the exclusive privilege of making shoes.
The eastern part of the Niger valley, suffering from a wonening climate and
isolated from trade networks as a result ofthe decline of Gao, was the area that
had suffered most since the days ofthe Songhay.
Trans-Saharan trade and the Atlantic trade
Despite political changes, the commercial organization of the Niger bend

remained established on firm geographical economic and social foundations
which survived both the Moroccan conquest and the intensification of
European trade on the coast. Overcoming numerous political and military
difficulties, trans-Saharan trade remained the main economic activity of vast
areas ofthe Maghrib and the Sahel. What changed after the end ofthe fifteenth
century was not the volume of trade but its economic significance in relation to
the development possibilities ofthe countries involved.
Routes
After 1591 the route from Dar'a to Taghäza became the royal road for trade
between Sa'adl Morocco and Western Sudan. But this was the fint route to be
affected by the disturbances which followed the death of al-Mansür. Merchants
gradually abandoned Dar'a and Taghäza for Süs and Tafilälet, where the local
religious leaders wielded a beneficial influence. Thus, less than half a century
after the Moroccan conquest the road network had almost reverted to what it
had been before. The most lasting changes were those affecting the Sahel and
Sudanese part ofthe network.
77ie Atlantic route The revival of this route was begun in the 1620s and
continued by the 'Alawites. The Tadjukant were masters of all caravan traffic
between southern Morocco and Mauritanian Adrär. Furthermore, the dissemina¬
tion in the western Sahara of religious doctrines such as those of the Kädiriyya
and Tîdjânïyya brotherhoods led to a considerable movement of people and
ideas between the religious centres of southern Morocco and those of the
southern Sahara.
Caravans from Shinguetti or Waddän travelled to Saint-Louis in Senegal via

Trarza, Galam via Brakna country and Hödh along the Dhär track. From
Hödh, several tracks led towards the Bambura kingdoms of Kaarta and Segu.
Moorish caravans travelled as far as Segu, Banamba and even Nyamina, but
most Bambara trade was transacted at Sansanding, the commercial centre ofthe
kingdom of Segu.
Hie eastern route This group of roads started in Tafilälet and crossed Tüwät,
through which the great caravans of Moroccan pilgrims passed. At Timimun
the road from Tüwät to the Sudan connected with the tracks from central
Maghrib. Further south, at In-Saläh, was the junction with the track from
Ghadämes. disputed by Tunis and Tripoli throughout the eighteenth century,

which was the channel through which Sudanese goods flowed to both capitals.
After Tidikelt the caravans skirted the western border of Ahnet and then, after
the arduous crossing of Tanezruft, reached Azawad where there were Kunta
encampments. The Kunta provided services, knowledge and baraka and were
also the most effective protecton ofthe roads to Timbuktu.
On arrival in Timbuktu, the Maghribi merchants were looked after by
diatiguis, who provided them with hospitality, transport and, if necessary, an
armed guard. For many of them, the journey continued as goods from North
Africa had to be taken on to Jenne.
TJieJula network Jenne was linked by two main routes to the forest regions and
the countries that produced gold and kola nuts. The fint, to the south-west,
went to Bure or Wurodugu, the second, to the south-east, to Kong and Asante.
Kong supplied Jenne with kola nuts and gold from Lobi and the Gold Coast.
Both were carried by the same Jula traden who sold slabs of salt from Taoudéni
to Buna The rest of their load would be made up of cotton goods and
European products which were marketed around Kumasi.
Parallel to this route, the Mossi Yane had established direct links between
Timbuktu and Asante country, skirting Jenne and the inner delta.
Trade goods
Salt played a very small part in trade between the people of the Maghrib and
those of the Sahel and the Sudan. From the beginning of the seventeenth cen¬
tury, towns along the Niger had access to the Saharan salt deposits by means of
separate and independent routes: the Moorish and Tuareg azalai (large caravans).
The Sudan traditionally exported gold, slaves and ivory, but it also needed to
dispose of a number of manufactured goods, such as cloth and jewellery, the
proportion of which grew as the export of gold and slaves declined.
Although the amount of gold exported never again equalled the huge
amounts taken out immediately after the Moroccan conquest, Sudanese gold
formed part of the returning freight of all caravans coming from the Sudan. At
the end of the seventeenth century there was an appreciable increase in the
export of slaves after Mawläy Isma'îl formed his 'Abid army and, in the second
half of the eighteenth century, gum arabic played an important role in exports
to Morocco. The opening ofthe port of Mogador provided Sudanese caravans
with a new maritime oudet where, in addition to the articles already
mentioned, they also sold large quantities of ivory and ostrich feathers.
Moroccan exports to Western Sudan were extremely varied, whether they
were local products (cereals, Meknes tobacco, silk garments and religious
works) or goods from the eastern Mediterranean (spices and silk) or from
Europe (textiles, sugar, coffee, glassware and firearms).
In the central Maghrib, Sudanese trade was the chief source of activity for
the provinces of Tüwät, Mzäb, Süf and Djarid. It covered a wide range of
goods, from indigo cloth and turkedi from Kano to kola nuts from Asante.
This trade, both varied and continuous, was very lucrative. In addition, the
trans-Saharan trade was transporting not only merchandise, but also the ideas
and values of a whole civilization.
Cultural and religious development
Islam in the Sudan on the eve ofthe djihäds ofthe nineteenth century
The period being examined is often described as one of decay and cultural
stagnation, which is untrue if by that is meant a decline or regression of Islamic
culture.
Through the continuous commerce across the Sahara, through the zâwiyas
(sanctuaries), the brotherhoods and the marabout groups and through the highly
organized networks of Jula traders, the influence of Islam continued to spread
in varying degrees through all the peoples ofthe Niger valley.
René Caillié observed at the beginning ofthe nineteenth century that all the
inhabitants of Timbuktu and Jenne were able to read and write Arabic.
Furthermore, the ta'rikhs. the famous chronicles of Timbuktu, were all written
between the seventeenth and nineteenth centuries.
Islam was no longer a purely urban phenomenon as it had been in the

previous period. Propagated by Berber or Sudanese marabout groups, it now
spread into the countryside and affected both Bambara peasants and Fulbe
herders.
The emergence of marabout groups (zuwaya or insilimen) is one of the most

striking aspects of the ethno-social stratification which occurred on the
southern edge of the Sahara during the second half of the seventeenth century.
Nomad society, from Senegal to Aïr, split into two distinct strata: 'warrior'
groups and clerical groups. The latter enjoyed great religious prestige, and their
zâwiyas attracted students as well as travelling merchants seeking protection.
Many of these sanctuaries, established along trade routes, later became caravan
stages. The shaykhs of the desert were to eclipse the urban 'ulamâ' and become
the spiritual mästen of most of the instigaton of the djihäds of the eighteenth
and nineteenth centuries.
The Islam of the marabouts was to a certain extent a continuation of the
Almoravid tradition, and its militancy contrasted sharply with the tolerant
syncretism which characterized the 'Black Islam' ofthe Sudanese towns and the
Jula centres. Also, by quickly associating itself with brotherhoods and religious
orders as universal as the Kädiriyya, the Islam of the marabouts offered its
followers a framework which went beyond the traditional categories of
identification, such as clans or ethnic groups.
It was a rallying point, a force for political and social emancipation, and as
such it attracted all the Tukuloor of Futa Toro who were opposed to the
Denyanke dynasty; it also attracted the Fulbe of Macina, who were shaking off
the Bambara yoke, as well as the Fulbe and the farmen of Hausa country, who
were struggling against the hegemony of various royal powers.
From the Niger
to the Volta
The fall ofthe Gao empire in 1591 following the Moroccan invasion created a
major political vacuum which other powers gradually began to fill. In the
Niger bend and the upper reaches of the Voltas, four poles emerged during the
seventeenth and eighteenth centuries: the Bambara kingdoms of Segu and
Kaarta, the Mossi kingdoms, the kingdoms of Kong and Gwiriko and the
Gulmanceba kingdoms. This took place against a backcloth of numerous ethnic
groups with non-centralized authority. It has been said that power was
'tribalized' during this period. But this is inaccurate, as the traditions of Mali
and Gao continued but in a context increasingly dominated by external factors
and internal conflicts that shaped new political structures. In addition, certain
socio-economic and religious factors, that were in no way 'tribal', influenced
the reorganization processes and, at the end of the eighteenth century, precipi¬
tated the first cracks that heralded the upheavals ofthe nineteenth.
Peoples, chieftaincies and kingdoms
The Bambara kingdoms of Segu and Kaarta from the sixteenth to the
nineteenth century
Political evolution
The origins of the Bambara kingdoms, while not ancient, are obscured by the
divenity of the oral traditions and the chronicles. The order of succession of
rulers is not always the same, nor is the length of reigns. This study, however, is
more concerned with the organization and relations of force that moved these
peoples from the sixteenth to the nineteenth century than with events.
A highly disputed issue relates to the peoples responsible for taking the
historical initiative in the Bambara kingdoms. L. Tauxier is categorical on this
point: 'In the final analysis', he writes, 'the Bambara were never capable by
themselves of founding kingdoms: the kings of Segu and Kaarta were of Fulbe
origin, those of Sikasso and Kong of Manden-Jula origin.' Although the
166
From the Niger to the Volta 167
\ /
\
/. y -- .
/v "- \ _
¡ >í ~ "*>
/>' \
Jj
J \
-*~ "v>
«¡5/ '*\ _-* ^
1 /
|->...' Q = O .- Si"
' i .,-
' « "
O
ruz
§> . o
..
% ;
1 (0
, <n
~ ...
/"
/"" -
13_to»S
a | j| a/
'
S-=cd
«e
_.
^
/ . * :> . }I . .So. >- ro
* I ' 1 ff
1
f ° i
1
\
* ^.IS'*^! \ X j&f
I _X Ï O & 'A V*S-^ /
> s? a U % . / se S 2 x_i->
E V
\< cj J 1 J 'o s / a 1
Z ( * 2 = -SO- 3 ? ^>
»^1
P 1 J «*Ntí </> X 9 * m -s:
VX * ^ \ « / «<¿-^^- a
"^
-S
O1 O^V -v % \ f/
íu
'**t ^
'n_
o
a
TD
r> I ^-^~~~-~^ r~//^M~^-^^ S *
t- . \ ? l-\ ru
A \ <" i \ |
O -a
2
«
Ë _,8
E '
i
* V «= -! 1 l S
8 " m i
o
'¡o
O
O
o
m
. *~ y
1 1
So
2 '\ X ^
<u
^ \ \ »
» \ p
O - - ° i
r-»
?; xnanva y \ __ , *ó. Vfc 1-H
/' /* N .' ' ""
r-i
Bambara's origins are certainly lost, contemporary movements are reflected in

traditions that make it possible to assert that they have long been indigenous to
the Niger bend. The Bozo were almost certainly there before them and they
also mixed with the Soninke, Soso, Malinke, Fulbe and others.
The story of the migration of the eponymous ancestors of the Bambara
dynasties contains the theme of the river crossing found in so many legends of
origin in Africa. Two brothers, Baramangolo and Niangolo, pursued by an
enemy and having no canoe, crossed the river thanks to an enormous m'polio
fish (catfish) which saved them at the last minute.
The striking aspect of this story of origin is the rapid association that came
about with the Soninke (Marka). The Soninke were traden who had been
setded for centuries, sometimes as slave-dealen, and had little interest in
soldiering although the Bambara newcomers made it almost a profession. One
of the brothers, Baramangolo, sought asylum on the right bank of the Niger
with the Buare of Segu. The Segu Buare, who were Soninke, gave him a
district which, although separate, allowed the newly arrived Bambara to defend
the entire Segu kingdom. Meanwhile, the other brother, Niangolo, built a tata
(fort) at Ba'iko. After failing to take the fort by storm, the Marka eventually
settled down to peaceful relations with the newcomen. Hence the Bambara's
(no doubt gradual) movement from the status of refugees to that of protectors
of their hosts and then to that of mästen. This process stands out clearly in most
accounts ofthe origins of Bambara power.
Another key aspect of this history was the strategic importance of the Niger
where the bend of the river reaches farthest into the desert and towards the
routes to North Africa. By unifying the two banks, the middle Niger had
polarized political power for four centuries. It was difficult for power to be
shared, which is why the two sister kingdoms of Segu and Kaarta were con¬
tinually attacking one another.
It is difficult to relate the two legendary anceston of the Bambara dynasties
to the fint leaders known historically. Regarding Segu, we are told that one of
Baramangolo's descendants was Kaladian Kulibali. Kaladian, who does not
appear in the oral traditions, is represented as having almost restored the great¬
ness of Mali in the seventeenth century, notably by reconquering Timbuktu
from the Moroccans. Meanwhile in Kaarta, on the left bank of the Niger,
according to Nioro legend, Sunsan (c. 1635), the son of Niangolo, is said to
have founded Sunsana, near Mourdia. His son, Massa (c. 1666), was a famous
farmer and a prolific father who systematically married off his daughten not to
princes he could not win over but to poor men on condition that they come
and live with him and espouse his cause. Massa's son, Benefali (c. 1710-45),
considerably extended the power of the Massasi using methods that combined
the patriarchal farming fife with the brutality of military raids.
Fulakoro, who succeeded his brother Benefali, was unable to avoid the fint
serious clash between the Massasi and the kings of Segu, who were more
powerful because their territory was richer and its different peoples more com¬
plementary. The person hie had to confront was Mamari Kulibali (1712
great-grandson of Baramangolo, the founder of the kingdom. Mamari, who

was endowed with extraordinary strength, rapidly became the leader of a ton
(an association of boys circumcised at the same time), hence his name, Biton or
Tiguiton, and then a social and political leader outside the gérontocratie and
religious institutions ofthe country.
He seized the chance to consolidate his domestic achievements with external
successes, one against Kong and the other against the Massasi.
Mamari was only able to repel the fint intervention by Kong through his
coalition with the Macina Fulbe. The second attack was driven off with the aid
ofthe Tyero Somono, using swarms of bees against the Kong cavalry.
Fulakoro was besieging Mourdia, which sent an appeal for help to Mamari.
Mamari, who was only waiting for such a signal to intervene, defeated
Fulakoro and took him prisoner, and once again the Massasi had to move
westward. Mamari Kulibali conquered the whole of Bambara territory,
including Beledugu. Macina and Jenne were placed under his authority and he
had palaces built by a Jenne architect. Finally, in 1751, Mamari easily con¬
quered Niani, the capital of declining Mali, whose ruler, Massa Maghan Keita,
paid him tribute. The pashas of Timbuktu had to do likewise after Biton's fleet
and cavalry had cleared the Niger bend of Tuareg exactions.
Biton's successors were rulen of no great stature and his son Denkoro
(1755-57), an arrogant bloodthinty man, who had seized the golden axe (the
symbol of royal power), was executed by the war chiefs. The average lengtb of
reigns was less than three years. Only Ngolo Diarra (1766 reigned long
enough to develop a serious policy. He put an end to the agitation by the ton-
dyon chiefs who had become mercenaries, and restored a true monarchy. He
abandoned Segu-Koro to setde up-river at Segu-Si-Koro. He strengthened the
Bambara hold on both Macina and Timbuktu where civilian officials descended
from the Moroccans, the Arma, were now to be watched over by Bambara war
chiefs. His campaign against the Fulbe chief Sua Makan Yero took him even
into Dogon country. When he was very old he went on an expedition against
Yatenga which failed, and he died on the return journey, having created a new
dynasty at Segu.
After Ngolo, internal conflict erupted again, and only Monson Diarra
(1792-1808) made the power of Segu felt from San to Timbuktu and from the
land of the Dogon to Kaarta.
But Kaarta was not yet finished. After Fulakoro's death in captivity, it had
recovered through raids at the expense of Beledugu, Bambuk and Khassonke
country. After Deniba Bo (1758-61), a great warrior - Sira Bo Kulibali
(1761-80) - had established the royal residence at Guemu. He had seized half
of Bakhunu while Ngolo Diarra was occupied in the north and east. He
pillaged Kita, a large Malinke town in Fuladugu, and, taking advantage of a
dispute between two Diawara clans, the Sagone and the Dabora, he drove the
latter out while the Sagone were exempted from paying tribute but were
obliged to provide a contingent of warrion in the event of war.
The socio-political organization ofthe Bambara kingdoms

The most striking feature of the Bambara's move towards a state form is the
novel formula they evolved for reconciling Bambara practices with the impera¬
tives inherent in the life of large-scale multi-ethnic kingdoms. Starting from the
working of the ordinary basic association of the ton, they gradually forged the
machinery of a state, passing, in the space of a few decades, from 'clans to
empires'.
In the kingdom of Segu The starting-point was the fla-n-ton (peer association) led
by Mamari Kulibali. A fla-n-ton was an association of young men who had
undergone the rite of circumcision together.
The ton-den (association members) elected a ton-tigui (chief) who might
equally well be a house captive as a son of a notable. The prime characteristic
of the fla-n-ton was the absolute egalitarianism among people of all social back¬
grounds, which contrasts with the prevailing clan or gérontocratie hierarchy.
The ton's aims were mutual assistance and the sharing of resources for
participation in social life.
Mamari Kulibali had to strengthen his group against the elden who opposed
his ton plan, using first their role as faaya (fathers) and the demands of farm
work. By killing the elders of Donkouna and Banankoro, Mamari at once
broke with gérontocratie authority and replaced it zsfaa (father) ofthe ton-den.
But, as the leader's successes mounted, many other association memben had
joined the original group - insolvent debtors, pardoned condemned men,
prisonen who had broken parole and young adventuren. Such new recruits
were admitted with the status of ton-dyon (captives ofthe ton).
When the memben of the association - which basically coincided with the
Bambara state's army - poured in, it was no longer a matter offla-n-ton but of
foroba-dyon (captives of the big common field) or furuba-dyon (captives of the
Great Union), the first ton-dyon of which formed the aristocracy. This structure
was strengthened by the involvement of the ton-dyon in cults of which Mamari
was the grand master.
The end-result of this process of the penonalization and accumulation of
power was the attribution of the tide faama (lord holding supreme power) to
the leader of this politico-müitary machine. The confusion of this status with
that of patriarch of a family community was one of the system's serious
contradictions. Originally, property accumulated through contributions, raids
and war-takings had remained in the foroba (public treasury) and profited the
whole community. But eventually the faama had discretionary control of state
property. The ruler was indeed assisted by a consultative senate of 40 memben
including warrion and holy men, but these notables had sworn fidelity to him
at secret rituals on an island in the River Niger and had sworn allegiance after
the investiture.
Another serious contradiction was the attempt by Mamari's successors to

settle the foroba-dyon as state serfs tied to farming, a move which precipitated a
serious resistance crisis among the warriors.
Except during the reigns of strong figures such as Mamari Kulibali and
Ngolo Diarra, however, the egalitarian and almost anarchical spirit of the fla-n-
ton never completely disappeared. It was this spirit that impelled the assembly of
ton-dyon to eliminate Denkoro, Mamari's son, 'to change the seed'. However,
the council of military leaden gradually supplanted the general meeting, itself
becoming weakened as some of its members moved away from Segu as the
kingdom expanded. To rekindle his troops' devotion to him, the faama still
convened a general meeting at least once a year for the re-swearing of loyalty
oaths, the performance of religious rites and the incorporation of young
recruits.
SEGU's ARMY At the time of the election, the royal insignia comprised the bow,
the quiver and the golden axe, unequivocal symbols ofthe main activity ofthe
state. Oral tradition records the feats of the faama of Segu's army, notably in the
siege of towns. When the army was arrayed in batde order, the main mass the
sofa (grooms) -was in the centre, called disi (chest). This main body of troops
was flanked right and left by bob (arms) which were made up of foroba-dyon
officered by ton-dyon. Behind the disi was a reserve corps of experienced ton-
dyon called ton-koro-bolo.
Arms consisted of axes, arrows, lances and guns, the fint of which, of
blunderbuss type, seem to have come from Kong. The drums played an impor¬
tant role, either the great tabala drum associated with each ruler, which
announced war, or the dunuba drum, which transmitted messages over consider¬
able distances. The man responsible for inspiring this whole body was the griot,
whose job it was to ensure that the men had the heart for batde.
The army was the main arena for people of different nationalities to rub
shoulders. It contained men of every social, ethnic and geographical back¬
ground. Young people were thus attracted, being assured if they survived of
rapidly making their fortunes.
The overall organization of the territory, as in most African kingdoms, was
concentric - from the original core along the River Niger, from the capital,
Segu-Koro, to Kirango, to the territories annexed over the yean, which
stretched along the river from Kanbaga to Timbuktu. This main area of state
activity by the kings of Segu was given the evocative name too-daga (millet
paste pot). As a result ofthe increasing confusion of state property and penonal
wealth, the eldest sons of kings were generously provided for and their
residences dotted the coune of the Niger. In the oudying areas, it was almost a
matter of delegated power, either to the indigenous chiefs or to governon
appointed by Segu.
Particular use was made in the organization of the army and the kingdom of
two communities - the Somono and the Fulbe. The River Niger was the
kingdom's main artery, vital for the supply of fish, for civil transport and
military logistics. The Somono fishermen were required to provide transport
and supply fish, while the Fulbe were expected to raise the public herds. These
two groups were backed by foroba-dyon, but there were also many foroba-dyon
established as peasants on agricultural lands for the needs ofthe rulen of Segu.
In the kingdom of Kaarta The socio-political organization ofthe Massasi kingdom

of Kaarta was similar to that of the kingdom of Segu. The royal succession did
not pose problems among the Massasi, as they had firm control over the
country through provincial governon and heads of army corps. Moreover, the
sofa corps direcdy bound to the king was highly organized and all men in
authority, all the corps of troops and the allies of the clan were memben of the
brotherhood that venerated the guardian spirit ofthe clan.
In short, the Massasi clan seems to have been more homogeneous than that
of Segu because it was not initially locked into a pre-existing indigenous
institution. It was formed and developed freely under the initiative of the
Massasi family which always jealously kept its leadenhip for itself.
On the Central Volta Plateau: Mossi kingdoms from the sixteenth to the
nineteenth century
Origins
The most important development on the Volta plateau during this period is
undeniably the formation of the Mossi kingdoms, the earliest of which dates
back to the twelfth century. Assuming, as most authors do, that the Na Gbewa
and Na Nedega of Mossi traditions are the same penon, through Naaba Rawa
and Naaba Zûngrana, the sons of Naaba Wedraogo, the Mossi dynasties are
linked to the Mamprusi, Nanumba and Dagomba kingdoms. With Naaba
Wubri, whose advent at the head ofthe future kingdom of Ouagadougou dates
to about 1495, and Naaba Yadega, Naaba Wubri's agnatic grandson who
founded the kingdom of Yatenga in about 1540, there began the process of
organizing the political structures ofthe Mossi kingdoms.
Of the three kingdoms (Mamprusi, Nanumba, Dagomba) founded by Na
Gbewa's descendants, only Dagomba was to play a major role, from the time of
Na Nyaghse (1460 Little is known about succeeding reigns from the
early sixteenth to the mid-seventeenth century. Dynastic strife seems to have
been the main feature. But in the second half of the sixteenth century a
Manden chief (Naaba) established the kingdom of Gonja.
In the second half of the seventeenth century, the Gonja kings launched
expeditions from their capital, Yagbum. One of these wan, against Dagomba,
was successful and Gonja took Daboya, the centre of a salt-producing area. But
in 1713, during the reign of Na Zangina, a convert to Islam, Dagomba held off
a new invasion by Gonja. The peace that followed put an end to acts of
aggression by Gonja which also harried the Nanumba kingdom.
In about 1740, however, a war of succession broke out between the newly
designated Na Garba and a prince who had been an unsuccessful candidate for
the throne. Taking advantage ofthe situation, in 1744 the Asante ruler, Opoku
Ware, mounted an expedition against Yendi in support of the faction opposed
to the new king. Na Garba was taken prisoner and only freed after undertaking
to send to Kumasi an annual tribute of 2,000 prisonen. Dagomba's future was
dogged by the aftermath of this treaty, which compelled it to engage in man-
hunting and to be ever finding new sources of prisoners. By the end of the
eighteenth century, Dagomba was no more than a channel for Asante influence
which reached as far as the borden of Mogho (the name ofthe Mossi country).
Hie kingdom of Ouagadougou

Twenty-three successon to Naaba Wubri are known in central and southern
Mogho between 1500 and 1800. Naaba Kûmdumye, the son of Naaba
Nyingnemdo who had established Mossi government over the old settlement
of Ouagadougou, began his reign with a dynastic dispute which was to have
serious consequences. Naaba Kûmdumye had competed for the kingship with
another candidate, Naaba Yadega, who, after he had been rejected, stole Naaba
Wubri's regalia. With the insignia of royal legitimacy, Naaba Yadega made for
Guni and conquered the region which subsequently bore his name, Yatenga
(from Yatega-tenga: the land of Yadega).
Naaba Kûmdumye put his sons to rule in Yako and Tema, which were both
later to become independent chiefdoms. By finding a place for princes whose
loyalty was not above suspicion, the king was killing two birds with one stone:
he was occupying the ground politically, and also keeping happy potential
candidates for the throne. But it was through this process that several peripheral
chiefdoms were transformed into kingdoms or independent principalities:
initially loyal dynasties became less so as they became genealogically and
territorially more distant from the royal line.
At the end of the sixteenth century, Naaba Kuda moved the royal residence
to Sapone, which shows that the state apparatus was still poorly developed. He
also placed his sons in such areas as Laale, Zorgo and Riziam.
Almost nothing is known of Ouagadougou until the reign of a Fulbe
usurper, Naaba Moatiba (1729 a warlike man who is said to have died by
poisoning. His name was erased from the official genealogy ofthe court.
The reign of Naaba Warga (1737 was short but full of new moves to
strengthen the monarchy. He was a great legislator. According to Yamba
Tiendrebeogo, he established the ritual announcement of zabyuya at the cere¬
monial investiture of chiefs. He organized the royal court, establishing two
corps of servants, one made up of Mossi and the other of prisonen. He also
institutionalized the royal pogsyure (napogsyure) system a system of capitalization
and distribution of women to enable royal servants to have children. He is
said to have codified customary law to the state in which it existed in the pre-
colonial period.
In short, the end of the seventeenth century and the first half of the
eighteenth century saw the final establishment of the Mossi political system
both in Yatenga and in central and southern Mogho. However, after Naaba
Zombre's long and peaceful reign (c. 1744-84) and that of Naaba Koom (c.
1784-91), who was the fint to spread Muslim influence, the reign of Naaba
Saaga (1791 was marked by internal conflicts which heralded the upheavals
ofthe nineteenth century.
The small kingdoms of Mogho

The small kingdoms and chiefdoms of southern Mogho are as yet little known.
In the north and north-east, however, three kingdoms stand out: Bulsa, Busuma
and Ta tenga.
The principality of Bulsa probably dates back to the beginning of the
sixteenth century with the installation of Naaba Namende, son and kurita of
Naaba Wubri. The kingdom of Busuma absorbed one by one the chiefdoms
that were still scattered in this area: Naamtenga (in the Luda region) at the
beginning of the eighteenth century, then the chiefdom of Pisila, and finally
Salmatenga (in the Kaya region). Tbe neighbouring state of Tatenga, founded
by one of Naaba Kuda's sons, dominated the region in the eighteenth century
when a formidable and ambitious warrior, Naaba Manzi, came to power.
However, Naaba Ruubo of Busuma succeeded in eliminating the terrible
Manzi in battle. His son Wema had no choice but to take refuge in the farthest
parts of the Riziam massif, which was later to lose more land to Yatenga during
the nineteenth century.
Yatenga
The successon of Naaba Yadega, the founder ofthe kingdom, held sway over a
limited area, gradually moving their royal residences northwards. The north,
however, was occupied by the kingdom of Zandoma under the sway of Naaba
Rawa's descendants. Naaba Lambwega's political and military activity consisted
precisely in dismantling Naaba Rawa's inheritance, while in the east he
incorporated the former Kurumba chieftaincies of Lurum.
In the late seventeenth and early eighteenth centuries, Naaba Lambwega's
descendants tried to saturate the considerably enlarged area of the kingdom by
establishing Mossi chieftaincies there, containing the Macina and Jelgoji Fulbe,
and also by maintaining the internal territorial status quo in Mogho.
Naaba Kango
No sooner had Naaba Kango come to power in 1754 than he was faced with
violent opposition from princes led by Naaba Wobgo and he was obliged to
abdicate and go into exile with a handful of trusty followen. He reached Kong
and then moved on to Segu, where he won the support of the Kulibali to
regain his throne. In 1757 he returned to Yatenga at the head of an army of
mercenaries and defeated Naaba Wogbo, who went into exile.
Naaba Kango began his reign with a spectacular gesture which marked a
decisive break with the past: he refused to make the ringu (enthronement)
journey, which alone gave the ruler of Yatenga the rank of rima (king). In
about 1780 he founded the capital at Waiguyo, not far from the former royal
seat at Busigi. It was a new town around a vast palace built in the Malian style
and bringing together his old comrades in exile. For the fint time the
inhabitants of a capital consisted exclusively of representatives of the state

apparatus, the world of politics and members of the royal household, both
prisonen and free. The name ofthe new capital, Waiguyo, is itself indicative of
a whole programme: 'come and bow down and make your submission'. This
demand was addressed fint and foremost to all members of the royal lineage,
royal nakombse (princes of the blood) who had supported the usurper Wogbo
and who now had to be reduced to subjection. But, up to the time of coloniza¬
tion, the princes refused to acknowledge defeat.
To restore order, which had been seriously disturbed during the years of
instability, Naaba Kango put down banditry mercilessly. The indiscipline of
chiefs was also severely punished, as was all ostentatious munificence on their
part.
Did the massacre of Bambara warriors precipitate an expedition by Segu
against Yatenga? There is no reference to such an expedition in Mossi tradition,
whereas there is in Bambara traditions. In any event, Naaba Kango, who
intended to give royal prisoners a special place in the system of government,
needed many more of them. Some were taken from among prisoners of war,
but most were captured in raids.
When Naaba Kango died in 1787, he was buried in his capital and not in
the royal cemetery. His death must have come as a relief to many, for he ruled
without pity. The princes' resentment was so great that they had the only
daughter of the dead king stifled to death; but the court aristocracy which had
remained loyal were able to have one of his lifelong companions, Naaba Saagha
(1797-1803), nominated to succeed him.
Trie socio-political structures of Mogho

By the Mossi kingdoms is meant not states comprising a homogeneous society
the Mossi ethnic group but composite socio-political formations resulting
from conquest, by warriors called Mossi, of the White Volta basin. However,
the process of intermarriage and slow infiltration by settlement carried out by
Mossi peasants was certainly much more decisive than military conquest. Each
time that an area was taken over, it was organized on the Mossi socio-political
model.
In the case of Yatenga, at the end ofthe nineteenth century, the population
comprised three distinct societies: Mossi society, Silmiga (Fulbe) society and
Silmi-Mossi society. The fint alone was subject to the authority of the king,
the Yatenga-Naaba. The Fulbe or Silmiise had, as it were, the status of guests, on
the basis of settlement contracts under which a broad strip of territory in the
northern part of the country was reserved for them. The Fulbe had been settled
in Yatenga since the seventeenth century and had established permanent
villages there from which the transhumance of cattle was organized. It is
perhaps misleading to speak of Silmi-Mossi society, but the sedentary small
stock-breeden, quite numerous in the south-east of Yatenga, certainly fall into
a separate category. Their compound name is a reminder that they were the
offspring of mixed marriages (which anyway were forbidden) between Fulbe
12.2 Mossi statuette commemorating a female ancestor, made of wood with a natural patina.
Height: 47cm (H. Dubois, Brussels)
men and Mossi women. Arriving from the chiefdom of Tema in the eighteenth
century, they did not come under Mossi authority and eventually passed under
the somewhat distant protection of the Fulbe politico-religious centre at
Todiam in the east ofthe kingdom.
But what of Mossi society itself? Its internal cleavages were linked to the
distinction made by the Mossi between the naaba (chief), the naabiise (chiefs'
sons) and the nakombse (sons or descendants of sons of princes who had not
become chiefs). Only chiefs' sons could become chiefs, and so, stricdy
speaking, except for the tiny minority of chiefs and chiefs' sons, all other Mossi
who claimed to be descendants of Wedraogo could be regarded as nakombse.
The nakombse belonging to the royal lineage were themselves divided into five
branches, each consisting of one generation of princes. The royal lineage was
apparently very early seen as having been of a constant generational depth (five
generations), the entry of a new generation into the royal lineage entailing the
exclusion of the oldest generation and its transfer to the talse group
(commoners).
To the extent that they held village chieftaincies, the nakombse constituted a
category of power-holders below the chiefs. There were two othen: the
tasobnamba (war chiefs), local notables of ancient stock, and the memben ofthe
king's household. In the case of court offices, the structure that prevailed at the
top of the political hierarchy around the king was the same as that around each
local chief. Three dignitaries assisted the chief, to which the royal court added a
fourth, the Bin-Naaba or Rasam-Naaba, the chief of the royal prisoners.
Thus memben of the royal lineage too deeply involved in the struggle for
the throne could not serve the king. Lasdy, the institutional holders of power
were actually effectively divided into two sub-groups with opposing interests:
the war chiefs and royal servants, on whom the king relied to govern, and the
nakombse, against whom he governed. This division went deep among the
Mossi as these men of lowly origin who held the great offices of state were at
the same time the great electors who chose a new Mogho-Naaba. Internal
conflicts have always punctuated the history of all the Mossi kingdoms and they
are always about the transmission of power. Generally, there was rivalry
between the ruling monarch's younger brothen and his sons.
In contrast to the world of government, to which the royal prisonen
belonged, lay the world of the land. The people or sons of the land were, in
theory, the descendants of the indigenous peoples, blacksmiths excepted.
Deprived of all political power, they were responsible for the earth rituals,
which concerned both the harvests and the social order and the perpetuation of
the local group. The figure of the naaba was contrasted to that of the tengsoba
(earth-priest), the holder of sacred power. This dualism was reflected cosmo-
gonicaÜy in the divine couple of Naaba Wende, the King-God, and Napaga
Tenga, the Queen-Land.
The definition of the category of people of the land changed substantially
down the centuries because the true Mossi became absorbed into the category
of indigenous peoples and used this status to become earth-priests. Alongside
the earth-priest there was the bugo (fertility priest), an office that seems to have
been of Dogon origin.
The political world and the world of the land were integrated into a system
with the king as its central focus through the medium of great yearly rituals.
The Yatenga Mossi, like the Kurumba, had an annual solar calendar divided
into lunar months, the discrepancy between the lunar year and the solar year
being compensated for by an intercalary month every three yean. After the
new year filiiga (thanksgiving festival) came the ceremonies of napusum
(greeting the king) at which, in three separate ceremonies, the royal servants,
the war chiefs and the nakombse paid homage to the king. The following
period, from the second to the sixth month, was taken up with a great
ceremonial cycle called bega, which involved the king and all the dignitaries of
the land in sacrifices intended to ask for a good harvest. The feasts of the bega
ended at the beginning of the rainy season. Ceremonial activity was resumed
during the harvest season, with two festivals of first-fruits.
This highly complex religious-cum-political system gave cohesion to a state
society which was made up of groups of varied origin, which had usually
retained cultural features from their pre-state past, starting with their own
socio-political stratification. Thus the division of society as a whole into four
large functional groups - power-holders, people of the land, blacksmiths and
craftsmen, and traden - turns up, with certain differences, among the Kurumba
or Fulse.
All the ethnic groups were divided in two ways: into descent groups and
local groups.
In this composite society, patrilineal and patrilocal, the Mossi term budu
denotes any descent group that operates as the exogamous reference unit. With
a history of its own, distinguished by the name of a founder and a place of
foundation, a budu defined its identity by the existence of a bud-kamsa (head)
and a kiims'rogo (ancestral shrine) along with having one or more sonda (collec¬
tive names or mottoes).
The patrilineage existed territorially only at its first level of segmentation, the
zaka (section). The section was divided in turn into yiiya (households), which
were production and consumption units. The sections belonging to one lineage
were generally scattered and hence came under several village units. In other
words, a Mossi village contained parts of several lineages, and a lineage was
dispened in several villages.
Gulma and Borgu

For a long time, litde was known ofthe history ofthe Gulma (or Gurma) bank
of the Niger - the right bank downstream from the bend, but thanks to
research carried out by the historian Georges Madiega we now have
incomplete but scientifically sound information about northern Gulma.
The Dogon constitute one of the oldest populations of north Gulma. After
them came the group called today the Tindamba (people of the earth) and the
Woba from the south. North Gulma was also peopled by the Kurumba. In
south Gulma the oldest inhabitants included the Tankamba, the Berba and, as
in the north, Tindamba and Woba. The latter certainly occupied a huge area
before being partially assimilated by the Gulmanceba and other peoples who set
up states there.
On this substructure of early populations a foreign power superimposed
itself, the Bemba or Buricimba, who later gave rise to the Gulmanceba states.
Madiega has two major hypotheses: the first is that the anceston of the
Mamprusi passed through Gulma before the arrival of the Buricimba; the
second is that fint Gulmanceba dynasties were contemporary with the fint
Mossi dynasties. This would suggest that the beginning of Gulma's history as a
state was probably in the fifteenth century. The figure of a historical-mythical
ancestor called Jaba is associated with the birth of the Gulmanceba states. He
does not seem to have been a great warrior, like Naaba Wedraogo in Mossi
history, as the powen attributed to him have more to do with magic than with
generalship. What seems certain is that the links suggested by the Mossi
between Na Bawa's descendants and Jaba's are simply late reconstructions
thought up at the court of the Mogho-naaba to justify the virtual amalgamation
of Gulmanceba power and Mossi power during the colonial period. All the
evidence suggests that, from the point of view of dynastic origins, the Mossi
and Gulmanceba hegemonies must be treated separately.
Where did the Buricimba come from? There is at present no way of knowing
which warlike migrations gave rise to the Gulmanceba empire. We know that
the earliest Gulmanceba political centre was Lompotangu or Sangbantangu,
south-east of Nungu. The Buricimba later moved to Kujuabongu, south of
Pama. They were not the only founden of kingdoms in this area. The
Jakpangu dynasty is of Berba origin, and the Gobnangu one of Hausa origin.
There are also dynasties whose founden came from Yanga.
Buricimba expansion continued throughout the sixteenth and seventeenth
centuries, and the apogee of Gulmanceba power can be put at about the middle
of the eighteenth century. At that time the Gulmanceba held sway over an
enormous, though no doubt sparsely populated, territory, bounded to the south
by the Mamprusi kingdom and Borgu, to the east by Torodi and the last
vestiges of the Songhay empire, to the north by the Sahel areas inhabited by
Kurumba, Songhay and Fulbe, and to the west by the Mossi chiefdoms of
Tuguri, Bulsa, Kupela and Tankudgo. In the middle of the eighteenth century
two large kingdoms, Bilanga and Kuala, shared most of the territory of north
Gulma between them, plus three minor states, Piala, Bongandini and Con.
The history of central and southern Gulma remains obscure. Among the
eleven kingdoms in the south, the kingdom of Nungu was founded by
Yendabri, a descendant of Jaba's, in the mid-eighteenth century. That was the
time when southern Gulma was overrun by the Tyokosi, Manden mercenaries
belonging to the Wattara group who had originally been in the service of the
Mamprusi kings. Yendabri, the ruler of Nungu, took on the leadenhip of a
coalition of kingdoms to drive the Tyokosi out of Gulma. Hard pressed by
their enemies, the Tyokosi withdrew to their capital, Sansane-Mango, to which
the allies laid siege. It is questionable whether the victorious Yendabri then
took advantage of his success to strengthen his authority over the league he had
formed. What is certain is that, in the eighteenth century, the Nunbado (the
ruler of Nungu) became the leading ruler in Gulma. His royal seat became the
main town of the land, largely because of the economic role of the Hausa
traden there. It is difficult to interpret the relations Nunbado had with the
Gulmanceba rulen. There does not seem to have been a Gulmanceba
confederation, and the Nunbado's actual authority was no doubt just as

territorially limited as that ofthe other kings, but he enjoyed exceptional moral
and ritual prestige in that he came belatedly to be regarded as the direct
heir of Lumpo and hence of his 'father', Jaba.
Borgu, south-east of Gulma, has a somewhat confused history, with several
state-type formations, the earliest having apparendy been Busa which reached
its apogee in the sixteenth century and then declined. The other kingdoms,
such as Nikki, are more or less direct offshoots of Busa.
The peoples with non-centralized authority

These peoples are described in this way for lack of a better term. They had
come into being at a very early date and were settled in the upper Volta basin.
Although these peoples did not dominate the political stage, their contribution
to the history of Africa must not be overlooked, as they form the underlying
human population on to which external contributions were biologically grafted.
The very term Moaga (pi. Mossi) means mixed. We cannot insist too
strongly on the role of the Nioniosse, who existed at a very early date. The
peoples of Yatenga themselves were not untouched by the contributions from
the Manden peoples of the Niger bend. On the cultural and economic levels,
the contribution of the indigenous peoples was a considerable one. We have
seen the contribution of the Dogon in the area of political and religious
organization and ofthe Kurumba to the system of dynastic rule in Yatenga.
The indigenous peoples, whether of Manden or Voltaic culture, were open
to exchanges but refused to accept domination a strategy that was highly
successful until the colonial period.
Despite the absence of a centralized state, the Gurunsi and the Bwaba
developed a marked individual penonality. The Gurunsi, living withdrawn into
family groups in remarkably designed huts, were fiercely opposed to any form
of complex political hierarchy. In the east, however, there did exist a more
organized structure under a canton chief with a court and a religious advisor
responsible for the cult of his kwara (magical symbol). The Bwaba appear to
have emerged as a specific entity between the tenth and fifteenth centuries.
They did not recognize any political authority above the village level. The
religion of Do formed a bond between those initiated in one village and those
in neighbouring villages.
Near Yatenga were the northern Samo, whose concentrated setdements are
marked by enormous grain stores. Their political organization consisted of
reasonably stable confederations of dozens of villages around a few political
centres corresponding to the kafu (Manden canton). Their system rested on both
clan alliances and territorial coalitions. The Sana (Samo is the Manden name),
hardened warrion and peasants fiercely attached to their freedom, never gave in
to the pressures from Yatenga, which tried several times to overcome them.
As for the Bisa, who were related to the Sana and separated from them after
a clan dispute, they were based in south-eastern Mogho. They, too, put up
remarkable resistance despite mutual cultural exchanges and a tribute in
captives at the end ofthe eighteenth century.
Kong and Gwiriko

Since the time ofthe Sudanic empires from the tenth to the fifteenth centuries,
Jula traders, called Wangara, had been travelling the routes to the forest where
gold and ivory were to be found. But, from the sixteenth century, a new factor
arose along the Gulf of Guinea: the trade in slaves and firearms. This single
event is enough to explain why groups of Jula now traders, now soldiers,
now Muslim missionaries -went further and further south into the savannah
areas where the goods most demanded by the new dispensation were
increasingly traded. It was Jula who helped to establish the great trading centre
of Begho. The Begho route was soon linked to the one crossing present-day
Côte d'Ivoire from Assinie to Bobo and Bamako through Yakasso. At the end
ofthe sixteenth century the Gonja kingdom came into being and, at the end of
the seventeenth century, after the destruction of Begho, Jula refugees fell back
into the Abron kingdom and founded Bonduku.
A number of important circumstances gave a sharp push to the activities of
the Jula from the seventeenth century onwards. Fint there was the fall of the
empire of Gao and, second, the creation of the Asante empire -which
constituted a major centre for the supply of gold, arms, salt and manufactured
goods. Third, there were the Voltaic savannah societies, which were quite
densely populated and most of which had no centralized political authority, and
therefore could supply the slaves and also the catde and gold that the coastal
areas wanted. It is thus easy to undentand why the Dagomba should have
organized the Kulango kingdom of Buna in Lorhon country on the same
model as their own.
It was in this context that a new diaspora of setdements that were at once
commercial, political, military and religious came into being on the initiative of
the Jula. In the middle of the eighteenth century, armed bands that had come
down from Segu the Diarrasouba brushed the Senufo aside and set up a
Manden kingdom at Odienne.
The Senufo belong to the Voltaic group of language-speaken. They were
setded in clans around Korogho, Seguela, Odienne and Kong. The fall of Mali
seems to have opened up to them possibilities of territorial expansion northwards
as far as Sikasso and Boungouni and southwards into the region of Bouaké,
where they would be absorbed into the Baule block. In the east they gave rise to
isolated groups before falling under Abron control. The Senufo were above all a
peasant élite. They were egalitarian and independent-minded and they had only
one large-scale social institution, the Poron, which had a religious character and
also helped to regulate the social hierarchy. They had consummate artists who,
since early times, had been producing some of the masterpieces of the
NegroAfrican symbolist style. It was only in about the nineteenth century that
the Senufo set themselves to building a few centralized kingdoms.
At the beginning of the eighteenth century the Manden added the famous
centre of Kong to the chain of Jula settlements. It was there that Keita and
Kulibali Mande, who had become Wattara, subdued the indigenous people.
One of them, Sekou Wattara, eliminated the other Jula groups and reigned at
Nafana and Kong, thanks to a powerful army. This military force enabled him
to conquer part of western upper Volta as far as Dafina.
Subsequently, the Kong forces subdued Turka country and Folona, laid
waste the Sikasso region and part of Minianka and Macina, and even got as far
as Sofara, opposite Jenne on the River Bani. We have seen how this expedition
was finally driven back by Biton Kulibali. After Sekou Wattara's death in 1740,
the empire was profoundly disrupted. There was a split between the caste of
non-Muslim Joola and Senufo warriors, the Sohondji, and the Salama Jula,
who were traden and Muslims. Kong had indeed become a great centre of
Islamic learning, but attempts to control Jenne show that the Wattara's grand
design was above all economic: to control the trade routes linking the forest to
the Niger bend over the greatest possible distance. After the failure of this
ambitious project, the Jula groups fell back on more limited undertakings.
One of the boldest was the creation of the kingdom of Wiriko, founded by
Famaghan Wattara, in a gold-rich area around the watenhed ofthe Banafin (a
tributary of the Niger), Comoe and Black Volta rivers. Famaghan Wattara
seized Tiefo, Dafin and Bwamu (the land of the Bwaba) more or less
permanently. His successon only managed to contain the revolts by peoples
reduced to Jula domination by repeated repression. This domination was above
all economic, even when it was made to appear as proselytism.
Economic life from the Niger to the basin ofthe Voltas

J B. Kietegha dates the period of high gold production from the Black Volta at
the mid-eighteenth century, when the newcomers monopolized the extraction
ofthe gold of Pura and introduced technological improvements.
Trade was busiest in the western and northern half of the area between the
River Niger and the basin of the Voltas. But during the seventeenth and
eighteenth centuries even in the Mossi kingdoms and among the peoples
with non-centralized authority, above the subsistence economy base there
gradually came into being a trading network which came to involve more and
more professional traders.
In Mossi country, an area of near monoculture in millet on exhausted soils
with irregular rainfall, shortages, even famines, were not uncommon. In
addition to millet, the staple food, crops included maize, groundnuts, cowpeas,
beans, sesame and some potherbs. Cotton, the main plant used in crafts, seems
to have been grown for a long time. The Muslim Yane have been associated
with weaving since the beginning of Mossi history and strips of cotton cloth
formed their cargo goods in the north-south caravan trade. The Marase
(Songhay), who specialized in dyeing, used indigo. The main wild plants
gathered were nere and shea nut.
In Lurum's pre-Mossi era (up to the fifteenth and sixteenth centuries) the
Kurumba had among them Marase who engaged in the Saharan salt trade. After
the formation of the northern Mossi states, including Yatenga, the Marase were
supplanted by the Yane, who in addition to the salt trade took up trade in kola
nuts, which they bought on the northern fringes of Asante. The large markets
of the kingdom, such as Yuba and Gursi, were the termini of the salt and kola
caravans. The Yarse of Yatenga were a dynamic group of traders found not
only at Timbuktu, but also in the valley ofthe Bani and throughout Macina. In
Mossi country they lived in symbiosis and compromise with the leaden: in
return for certain privileges, they undertook not to encourage any subvenion,
to sell to the Naaba as a matter of priority, to support him through appropriate
rituals and even, on occasion, to act as intelligence agents.
While the accounting unit for the traden was the cowrie, other units were
used in the caravan trade and there were various systems of equivalence between
the cowrie, the cubit of cotton strip, the block of salt, slaves and hones.
Yatenga was one of the great metal-working areas of West Africa, and the
blacksmiths combined metal-working with the export trade to central Mogho,
which was poor in iron ore.
The slave trade, going on on every part ofthe coast of black Africa during
this period, introduced into the mechanism of trade and socio-political relations
a terrible logic from the moment firearms, slaves and political power were
made part of the same equation. The greatest kings of the time participated in
this slaving activity, fuelled by the host of local chiefs who, willy-nilly,
embarked on this business to enter the political game.
We have seen how, in the Bambara system, the status of captive had been
ingeniously institutionalized by turning to account the already existing institu¬
tion of the ton-den (association member) to create the ton-dyon (captives of the
community). A further step was taken with the furuba-dyon (crown slaves).
These were prisonen of war grouped into contingents, each of which was
defined by the faama who had formed it. After suffering the fate of any trade
goods sold on the market, they saw their situation improve when they were
purchased by a community. A woman, as soon as she produced a child,
acquired the status of woloso (born in the house) and a man could acquire the
same status as soon as his master had sufficient confidence in him. As soon as he
reached the status of woloso, he became a caste member. He could no longer be

put on sale, and had a right to property and to transmit his estate to his
children.
It is true that the status of woloso was transferred to his descendants, even if,
as happened at royal courts, a few caste men might rise to the highest positions.
The mark penisted ineradicably in the collective mind. Thus, the great king
Ngolo Diarra's family was unable to eradicate the memory of the servile status
of their ancestor, which provoked taunts from the Massasi of Kaarta.
Islam and traditional African religions

It used to be a commonplace to say that the Bambara and the Mossi had consti¬
tuted strong barriers to the advance of Islam in West Africa. This is a simplistic
assertion that needs revision. Islam, at work since the eighth century if not
before, increased its influence in the great empires from the empire of Ghana to
the empire of Gao. Naturally, there were obstacles and resistance. Even as late
as the eighteenth century the Jelgoobe and Feroobe Fulbe were often followen
of traditional religions. For a millennium, however, from the eighth to the nine¬
teenth century, Islam continued to penetrate the whole of this region through
numerous channels including traders, prayer-leaders and even violence.
The Bambara initially looked upon Islam syncretistically: its presence im¬
pelled them to stress once again the Supreme God, Maa Ngala. Having
assimilated Islam in this way, the kings of Segu and Kaarta saw no obstacle to
consulting the ministen of this great God through the persons of the marabouts,
while remaining faithful to their own cults of which they were practically ex
officio the high priests and to the magician.
Although they practised traditional rites, the Kulibali took part in Muslim
religious festivals, at least by their presence, and offered lavish gifts to the
Sarakolle marabouts. These syncretisms, which at the same time were develop¬
ing with Christianity in Afro-American worship in Brazil, Haiti and Cuba,
were one of the characteristics of the Negro-African religious spirit. Thus, if
the appeals to the spirits failed to provide a clear and satisfactory response, the
intervention of the marabout would then be sought and welcomed piously by
the various Bambara kings.
What is certain is that Islam presented itself to the Bambara with institutions
such as polygamy, divorce, repudiation and slavery that did not systematically
challenge their own. Moreover, the marabouts did nothing that might put off
Bambara neophytes. Thus at Dia, probably one ofthe oldest centres of Islam in
the Western Sudan, teaching generally adapted itself to indigenous customs and
gave plenty of attention to magic. The most numerous marabouts among the
Bambara were the Marka, the Bozo and the Somono, who claimed to be in the
line of Moorish or Fulbe Islam. Some ethnic groups, such as the Soninke, the
Marka, the Fulbe and the Torobe, Islamized long before and enjoying religious
toleration, acted as models and catalysts.
In short, this mutual tolerance was of great benefit to both parties: it
associated the Muslim leaden with government and it used endogenous
channels to win these peoples to the faith of the Kur'ân. It also contributed to
the strengthening of the power of monarchs who, faced with the ethnic and
social heterogeneity of their states, could not afford the luxury of religious dis¬
cord. Such at least was the state of affairs before the dissimilar djihäds of Sekou
Ahmadu Barry and al-Hädj dj 'Umar.
In Mossi country, the beginning of Muslim influence at the court of

Ouagadougou probably dates from the reign of Naaba Kom, son of Naaba
Zombre (c. 1744 Islam spread into the Volta basin in the wake of Jula
traden and warriors, along a north-south line in the west of the Volta area,
along the Black Volta valley. The town of Bobo-Dioulasso -was one ofthe main
centres of Muslim proselytism.
The founden of Gonja were Muslim Manden. From Gonja, Islam moved
into Dagomba country. Trading centres in these areas soon became religious
centres, distinguished by the presence of madams (Muslim scholars). The first
Dagomba king to be converted was Na Zangina, who reigned at the very
beginning of the eighteenth century.
Islam penetrated Mamprusi country at roughly the same time. Gambaga, the
main market, quickly became a Muslim centre. Thus it was from the south that
Islam came to Mogho at the end of the eighteenth century, during the reign of
Mogho-Naaba Zombre, who showed himself sympathetic to the Muslims. It
was in the reign of Naaba Dulugu (c. 1796 that the decisive step was
taken. He formally converted and built the fint mosque in Ouagadougou. This
early Mossi Islam, like that which then prevailed in the Mamprusi and
Dagomba kingdoms, was an aristocratic Islam, and for the Mogho-Naaba and
the other chiefs and dignitaries its practice was in no way inconsistent with
observance of the traditional religion. Far from trying to impose the new
religion on his subjects, Naaba Dulugu seems to have been afraid that Islam
might make too rapid progress in his kingdom.
Thus, in the Bambara kingdom as in Mogho, although there was a decided
tolerance towards Islam, the traditional religion remained on its guard and
sought compromise formulae. For the other two political centres, the empire of
Kong and Gwiriko on the one hand, and the Gulmanceba kingdoms on the
other, the general attitude was decidedly different. At Kong, as among the
Bobo-Jula, the Islamic religion was both a reason to live and a way of life, such
that their policy, imbued with a militant faith, foreshadowed the djihäds of the
nineteenth century. Conversely, the Gulmanceba were resolute upholders of
the traditional religion.
Conclusion
For the countries ofthe Niger bend and the upper Volta basin, the period from
the sixteenth to the eighteenth century was unquestionably a state-building
phase.
In accordance with an ancient tradition in the region, the Jula, Mossi,
Gulmanceba and Bambara states were all multi-ethnic units. Even if the domi¬
nant ethnic group claimed certain privileges, it was itself the outcome and
source of inter-ethnic mixing. The sometimes highly sophisticated political
machinery was nevertheless vulnerable, mainly because of the almost total lack
of a written form of administration.
Moreover, the states were undermined by internal contradictions. In the fint

place, the devolution of power often gave rise to serious difficulties. When
Naaba Kûmdumye of Ouagadougou found positions for the unsuccessful
candidates for the chieftaincy, little did he suspect that his successor, Naaba
Warga, would have to wage war on the descendants of those chiefs. This
political difficulty was compounded by a social problem, for the princes
excluded from power turned their animosity against the peasants, who were
thus exploited mercilessly. Second, there was a formidable opposition between
the royal entourage, made up of commonen, and the king's relatives, who
were held in check by a series of measures the most striking of which was the
policy of Naaba Kango of Waiguyo.
Furthermore, the question of religion must not be played down. It is true
that, until the end of the eighteenth century, the Mossi and Bambara states
succeeded in maintaining a modus vivendi based on tolerance, syncretism and an
exchange of services. But there was a noteworthy difference between the Mossi
and Bambara states, on the one hand, and the empire of Kong and the
Gulmanceba kingdoms, on the other.
State-building in the region, however, fitted into an overall economic
context which spelt ultimate doom for such political experiments. All these
countries lay between the southern edge of the Sahara and the Adantic coast,
which was increasingly controlled by the Europeans, whose commercial policy
reshaped commercial circuits and the structure and terms of trade to their
advantage. Supply and demand were gradually disrupted. In the economic
chain that each ruler attempted to control, the relative importance of the
commodities used to win and maintain power (horses, arms and captives)
constantly increased, with the consequent spectre ofthe slave trade.
By comparison with the African rulers along the coast, who were direcdy
confronted by the Europeans and had no other choice but submission or war,
these countries of the interior certainly enjoyed a respite and seemed to have
their fate in their own hands. But that fate was in fact caught up in a process
that was increasingly dependent on the outside world. The hegemonies of the
region did not have enough time to establish a de jure state providing the
stability and order mentioned by chroniclers in reference to the previous
empires. Nevertheless, it has been demonstrated that, in difficult circumstances,
people proved capable of building a state with their own African means.
The states and cultures
ofthe Upper Guinea coast
Introduction
From the Casamance to Côte d'Ivoire inclusively, there stretches an enormous

area of coasts and forests, inhabited by many and various peoples. This area is
far bigger than the one historians normally call Upper Guinea. This chapter
aims at sketching the outline of its development between 1500 and 1800.
The societies
By comparison with the big ethnic masses of the Sudan, where state-type
societies predominate, this area is characterized by many small socio-cultural
units organized on the basis of lineages, clans and villages. From the Casamance
to the Tanoe, between the northern savannah and the southern coastline on the
one hand, and between the mountain ranges of Futa Jallon and the Guinea
Spine and the western and south-eastern coastline on the other, there are more
than a hundred ethnic groups and sub-groups.
This multiplicity of human groups explains the many linguistic differences
that characterize the cultural landscape. There are sometimes dialectal variants
within a single language, which restricts mutual linguistic intelligibility within a
single ethnic group,
Despite this divenity, there are wider linguistic entities. Three big language
families share the area between the Casamance and the Tanoe. Within the
family of Manden languages, the Southern Manden sub-group is predominant

Northern Manden only appearing in the form of the Manden spoken in the
Gambia, Casamance, Guinea-Bissau, Sierra Leone and Liberia. South of the
Manden languages, along the coasdine from the Casamance to Liberia, the so-
called West Atlantic languages are also divided into northern and southern
groups and are extremely varied. Lasdy, to the east and south-east, the so-called
Kwa languages comprise the Kru-Bete and Akan languages, which show the
same heterogeneity as the West Atlantic languages.
The difficulty of producing an historical synthesis

Tracing the development ofthe countries ofthe coast of West Africa from the
187
Casamance to Côte d'Ivoire between the fifteenth and nineteenth centuries is a

difficult task for the historian.
The difficulties relate fint to the sources themselves. After the fifteenth
century, written European sources become more and more numerous and
accurate as trade intensifies, but they are unevenly distributed by periods and
regions: they are plentiful for Senegambia, the Rivers Coast and the sector
from the Gold Coast to the Niger delta, but are less so for Liberia and Côte
d'Ivoire. Although they reflect the prejudices of Europeans, they nevertheless
give a good picture of the coastal area, including the geographical setting,
economic activities, systems of government and habits and customs.
When these sources are inadequate or non-existent, they need to be
supplemented by archaeology, oral tradition and other sources. While the
archaeology of the dry Sudan-Sahel area is developing, that of the humid
Guinea area is still in its infancy despite excavations in Casamance, Guinea-
Conakry, Sierra Leone and Côte d'Ivoire. Research on oral tradition has made
crucial progress in the Manden and Fulbe sectors but has not yet given its full
measure elsewhere. Moreover, oral tradition rarely goes back beyond the
seventeenth century. Lasdy, apart from the relatively homogeneous Manden-
speaking groups, historical linguistics in this area of amazing linguistic com¬
plexity is still at the planning stage.
There are also difficulties relating to the socio-political fragmentation typical
of this region. Historians of the states that came into being as a result of the
sharing out of colonies often give more space to the peoples of the Sudanic
parts than to those of the coast. Moreover, no overall historical study of these
coastal and forest peoples is possible without precise knowledge of each of their
histories.
Our study will keep to historical reality as experienced by the peoples

themselves, without losing sight ofthe flow of world history. The West African
coast acted as the main link between the historical pressures of the Sudan and
those of European trade; and down the centuries different parts of it became
active, came to life again and settled down.
Population movements
The Manden movements
In the area between the Gambia and the rivers of Guinea and Sierra Leone,
signs of the fint Manden (Malinke) advance towards the Adantic go back to the
twelfth century. Oral traditions collected in the Gambia, Casamance and
Guinea-Bissau mention Manden agricultural setdement in the late twelfth
century. This peaceful peasant migration was followed by the conquering
migration ofthe soldien of Tirmaghan Traore, one of Sundiata's generals. The
conquest of the Adantic coastal areas gave Mali wide access to the sea by the
late fifteenth century and it dominated the area between the Gambia and Sierra
Leone.
The states and cultures ofthe Upper Guinea coast 189
However, it is misleading to see these movements purely as military con¬

quests. A deeper analysis of this westward expansion suggests it may have been
the consequence of certain factors inherent in Manden society. The practice of
collateral succession made it difficult for sons of rulen to succeed their fathers.
Many therefore set out westwards with their families to found their own states.
Other memben of Manden society, intent on exploiting trade with the West
and finding areas for setdement, were actively involved in this westward
expansion.
The peoples ofthe coastal belt and the hinterland in 1500

The peopling of the coasts and forest before the European voyages of the
fifteenth and sixteenth centuries is one of the most obscure areas in West
African history. Only a few -written sources dealing with the Sudanese empires
and the gold trade can give us some information on the peoples of Senegambia,
the Southern Rivers and the Gold Coast. For the areas in between we can only
guess the history on the strength of hasty notes by Portuguese travellen and the
oral traditions of some peoples of the hinterland who claim 'to have always
been there'.
Tlie Southern Rivers between the Gambia and Cape Mount
Southern Senegambia Casamance and Guinea-Bissau Three main groups occupied

the estuaries and lower rivers when the Portuguese arrived: the Bainuk or
Banhun, a people of Manden origin who had come from the east in the four¬
teenth century; the Joola, who had come from the south before the fifteenth
century; and the Balante from the east, who had imposed themselves on the
Bainuk before the fifteenth century. Behind these coastal groups there were the
Manden who had arrived from the east in the twelfth and thirteenth centuries
driving the other peoples westward towards the sea. After the Manden, Fulbe
from Macina came in the fifteenth century.
Guinea-Conakry and Sierra Leone Archaeological discoveries at Yengema in

Kono country, confirmed by linguistic research, have prompted C. Fyfe to
suggest that the Windward Coast had long been inhabited, since the third
millennium before the Christian era. The earliest inhabitants were probably the
Limba of Sierra Leone. Next came the Temne and Baga, with their related
languages. Lastly came the Kisi and Bullom, two related groups who split up,
the former remaining in the interior and the latter going on to the sea.
At the end of the twelfth century, movements following the fall of Ghana
and Sosso brought the Soso to Futa Jallon, where the Dyallonke, Baga, Nalu,
Landuma and Tyapi were already. There followed a mingling of peoples which
ended in some migratory movements to the west. The Nalu, Baga and
LandumaTyapi settled on the coast and in the immediate hinterland. At the
end of the fifteenth century the pressure of the Fulbe on the SosoDyallonke
Baga and Temne set off new migratory movements which enabled the Baga to
spread southwards along the coast, thus bringing them into contact with the
Bullom.
Cape Mount to the River Bandama: Liberia and western Côte d'Ivoire
This area was inhabited by the Kru, Kwa-speaking peoples, and this section of
the coast was the area of the trade in malaguetta or fool's pepper. The
hinterland was dominated by the high forest and the eastern heights of the
Guinea Spine, which allowed only limited links with the world ofthe Sudanic
savannahs.
The European voyagers of the late fifteenth century met peoples along this
coast from Cape Mesurado to Cape Lahou. Historians think that these were
part of a long-established Kru settlement. Archaeology and research in oral
tradition also show that this region has been populated for a very long time.
We must therefore suppose that the savannah north of the forest was occupied
by Southern Manden groups. In the fifteenth century they had, no doubt,
begun to thrust south into the forest under pressure from the Northern
Manden from the upper Niger. This north-south drive must have brought
many litde groups of Kru to the seaward fringe of the forest.
Population movements from Sierra Leone to Côte d'Ivoire

In the sixteenth century new peoples came from the interior to join those
already there. They were mainly Manden, driven by Mali's economic and
political difficulties to move southward. This expansion to the forest fringes
was the work not only of warriors but also of merchants.
The fint of the Manden to move were the Kono-Vai, who came from the
upper Niger and migrated towards the coast between 1500 and 1550. By the
beginning of the seventeenth century, the Vai occupied the mouths of the
Mano and Moa rivers, while their Kono relatives were in the immediate
hinterland.
In the mid-sixteenth century, the Mane-Sumba, a group of Manden invaders

from the upper Niger, also arrived on the coasts of Sierra Leone and Liberia.
The Mane reached Cape Mount in 1545 and then moved northward, crushing
the indigenous peoples and making an incursion into Futa Jallon by way ofthe
Scarcies. They were driven back by the Soso, the allies ofthe Fulbe, and had to
return to the coast, which they occupied from Liberia to the islands of Los,
merging with the original inhabitants.
The precise composition of the groups that were involved in this migratory
flow is not yet clear. What is certain is that this invasion had important
consequences for all the peoples in this area - Southern Manden, Kru and West
Atlantic - starting with the destruction ofthe famous kingdom ofthe Sapes and
continuing with the building of a new political society under Manden rule.
The peoples of this coast were reorganized into four kingdoms: Bullom, Loko,
Bure and Sherbro. These peoples were then integrated socio-culturally. In the
coune of their conquest the Mane embarked on a process of miscegenation
which led to the implantation of Manden influence in this region. Finally, the
Mane introduced military innovations which enhanced the fighting prowess of

the coastal peoples. By the end of the sixteenth century the invaders seem to
have become naturalized, and, by the beginning ofthe seventeenth century the
original Mane had been completely absorbed.
In the sixteenth century a Southern Manden group, the Manden (Malinke)
of the Cámara clan from the upper Niger, settled in Liberia and western Côte
d'Ivoire, causing not only the movement ofthe Kono and the Vai towards the
coasts of Sierra Leone and Liberia but also that ofthe Toma, Guerze and related
groups towards the forests of what are now the republics of Guinea and Liberia.
At the same time the Diomande, brothers of the Cámara, were moving
eastward, settling at Touba (Côte d'Ivoire) and driving out the Dan-Toura.
Advancing still further east, the Manden (Malinke) entered Worodugu and
displaced the Guro south-eastward, towards the forest.
This thrust by the Southern Manden put pressure on the Kru, who moved
deeper into the forest in the direction of the coast. By the middle of the
sixteenth century the northsouth movement of the Kru was probably already
completed. In the Côte d'Ivoire sector the movements seem to have taken
place later.
Population movements in the seventeenth and eighteenth centuries

The population movements ofthe seventeenth and eighteenth centuries affected
the whole coastal and sub-coastal area and took the form of multi-directional
migratory flows. The sector most affected was the area between Liberia and the
River Tanoe. Warren L. d'Azevedo has suggested that these population move¬
ments were largely stimulated by the desire to control the slave trade which in
turn led to conflicts and the formation of military and commercial alliances
between and within ethnic groups.
From the Gambia to Sierra Leone
Casamance and Guinea-Bissau In the seventeenth and eighteenth centuries no

new peoples appeared in this sector; but with the boom in the slave trade there
was a mingling of peoples due in part to the devastating raids of the Manden,
backed by the expansion of Kaabu, then at its peak. The Manden and their
Kasanga vassals continually harassed the Bijago, Joola, Balante and Bainuk. To
escape the pressure of the slave-traden, many peoples sought refuge in the
inaccessible coastal swamps, abandoning to Manden control the riven leading
into the interior.
Guinea-Conakry and Sierra Leone The growing influence ofthe Adantic trade on
the peoples of the Sudanese zone and the expansion of Islam marked by the
Fulbe revolution in Futa Jallon in the eighteenth century formed the backdrop
to the population movements in this sector in the seventeenth and eighteenth
centuries.
In the first third of the eighteenth century the Fulbe of the holy war set off
the movement of the peoples of the Tyapi group towards the coasts of what is
now the republic of Guinea. Rejecting the authority ofthe masters ofthe new
religion, and having lost several hard-fought batdes, such as that at Talansam
fought by the Baga, the last groups left Futa Jallon for the coast.
Another major movement was the one that brought powerful waves of
Soso-Dyallonke to the coast. Driven from Futa Jallon, the Soso moved slowly
towards the coast which was already occupied by the Nalu and Baga. But the
big wave started in 1760 when the Soso, under the leadenhip of Souba
Toumane, invaded the Baga and set up the chieftaincy of Toumania in the
immediate hinterland ofthe Kaloum peninsula. They welcomed their Dyallonke
relatives, who had also been driven out ofthe mountains by the Fulbe, and, by
the end of the eighteenth century, dominated the coast between the Rio
Pongo and Sierra Leone.
However, the peoples of the Tenda group, on the north-western spurs of
the Futa Jallon mountains, resisted the Muslim Fulbe. So did the Koniagui and
Bassari who had escaped domination by the invaders since the fifteenth century.
Other prime movers of this population movement into Guinea and Sierra
Leone were the Northern Manden, who caused the migrations of the Toma,
Guerze and Manon into the forest zone of Guinea, the Mende into Sierra
Leone and the Quoja into Sierra Leone and Liberia. A distinction must be
made here between movements connected with the already long-established
migratory movement of the Manden southward towards the sea and
movements which stemmed from the expansion of Islam.
With the advent ofthe Muslim revolution in Futa Jallon, the Manden joined
with the Fulbe to convert or subjugate the Soso, Baga, Bullom and Temne.
Thus they occupied the Mellacorie in the eighteenth century, organized
Muslim Manden chieftaincies and were represented all along the coast from the
Island of Matacong to the Saint Paul river. At the end of the eighteenth
century, resistance to Islamization broke out among the Soso and the West
Atlantic peoples when the almamy of Futa Jallon sought to impose his authority
over the whole area. Nevertheless, the society of the Guinea and Sierra Leone
riven was not to escape the influence of Futa Jallon, nor the political and
cultural influence ofthe Manden trader-marabouts.
Liberia to the River Bandama
The migrations that helped to establish the peoples of this sector in the
seventeenth and eighteenth centuries were primarily a continuation of the
north-south movements that took place in the fifteenth and sixteenth centuries.
They were also due to the upheavals in the Akan world during the seventeenth
and eighteenth centuries, and to the various attractions and disadvantages ofthe
situation on the coast at that time. The results were migratory flows in various
directions which, in the forest, were sometimes decidedly circular in nature.
77te Mande and the We The Quoja-Karou invasion ofthe coasts of Sierra Leone
and Liberia shows that the Manden thrust was still pressurizing the forest and
coastal peoples in the eighteenth century. The Northern Manden had continued
their advance to the south and south-east and reached the Bandama River.
Their movement intensified that of the Southern Manden; the Dan continued
to move south-eastward and south-westward while the Guro and Gagu were
heading towards the Bandama river and beyond.
This Southern Manden thrust affected the We, in particular the We of Toule-
plu in Côte d'Ivoire, who reached the forest in the mid-seventeenth century
and the River Cavally later.
The Magwe-Kru C. Behrens thinks that the Kru moved into Liberia along the
coast from east to west during the sixteenth century. After splitting into various
groups, they occupied what is now Grand-Bassa County in the middle of the
seventeenth century. As regards the Kru settlement of Côte d'Ivoire, it seems
that a westeast movement followed the northsouth one. Thus the Kru of
Grand-Béréby came to Côte d'Ivoire from Liberia following the fragmentation

ofthe Guerre groups (the We) from the north.
This great migratory wave led large groups of the Magwe from the Cavally
basin in the west to the Sassandra basin and beyond to the Bandama River. It
supplied population for the settlement of the area between the Sassandra and
Bandama rivers until the end ofthe eighteenth century.
Political mutations and socio-economic changes
The establishment of sea routes: late fifteenth to early seventeenth

centuries
Once open to traffic and trade, between the late fifteenth century and the early
seventeenth century the coastal area became a magnet to the coastal and sub-
coastal peoples and the Sudanese traden from the interior. The establishment of
sea routes had begun before the arrival of the Portuguese caravels when
Manden expansion linked the Central Sudan to the Atlantic coast, from Sene¬
gambia to the Gold Coast.
Internal actors leading to the establishment of sea routes

In the fourteenth and fifteenth centuries, with its expansion westward, Mali
had created the conditions for the commercial expansion of the Manden into
the coastal area. The Gambia and Casamance, having become western pro¬
vinces of Mali, exchanged their agricultural produce and craft products for iron
and other metals from the interior. The trading towns of Kantor ran this trade
which used the waterways, particularly the Gambia and Casamance rivers,
whose mouths were linked by a busy coastal trade. A different trade flowed
from the River Niger to the forest, where the Joola exchanged salt, copper,
cotton goods and fish for kola nuts and, sometimes, palm oil. The Joola
brought goods from the Niger bend to El Mina, as they are reported at Begho
in the first half of the fifteenth century.
13.1 Sixteenth-century carved ivory hunting-horn from Sherbro Island (Bullom), Sierra
Leone. Height: 43 cm (© Luigi Pigorini Museum, Rome, photo: Rossini)
The trading sphere ofthe Manden relied on a network of routes and market
staging-posts with which the forest fringes were amply supplied and which the
river valleys extended down to the coast.
External circumstances: the Portuguese explorations

The Portuguese were the fint Europeans to explore the West African coast in
the fifteenth century, once the development ofthe caravel had enabled them to
round Cape Bojador in 1434. They landed on the island of Arguin in 1443 and
established trading posts at various places along the coast, from Senegal to the
Gold Coast, and explored the whole of the Gulf of Guinea. In search of gold
and spices, they joined their emergent maritime empire to the Manden trading
sphere, nurtured by the expansion of Mali.
Portuguese documents of the time provide lists of the goods which formed
the cargo of returning caravels. They include rice, copal, civet, palm mats and
sacks and carved ivory objects mainly spoons, pedestal bowls and hunting-
horns. Some of these items have been preserved and are to be found in the
museums of three continents. The finest of them, which are the work of the
Bullom of Sherbro Island, bear witness to the high artistic level reached by the
coastal peoples before the arrival of the Europeans.
Throughout the sixteenth century, the Portuguese maintained maritime and
commercial supremacy on the West African coast from Arguin to Angola. In
Upper Guinea they travelled by river in search ofthe gold ofthe Sudan. In this
area, which was then Mali's ocean gateway, they helped to improve the links
between the Rivers Coast on the one hand and the upper Gambia, upper
Senegal and upper Niger on the other.
In the sixteenth century, Joola inland trade routes led to coastal positions
occupied by the Portuguese. The gold of the Sudan therefore reached Lisbon
together with malaguetta pepper. In this way the European and Manden
trading spheres became intertwined and the economic takeover of the Saharan
traffic was under way.
European voyages and early trading

Portuguese supremacy, however, had to contend with French and English
pirates operating on the Guinea coast from 1530 onwards. Later their difficul¬
ties were deepened by Dutch competition. Dutch ships began to ply to West
Africa in the 1590s and a decade later the Netherlands was doing more trade
with West Africa than most European nations put together. Dutch supremacy
replaced Portuguese supremacy.
Nevertheless, by the end ofthe sixteenth century the Portuguese had com¬
pleted the opening up of the Atlantic coast of West Africa to trade and
inaugurated the strategy that was to prevail during the following centuries.
Early socialformations and the 'Kingdom ofthe Sapes'

The West Adantic civilization, which stretched from the Joola of Casamance to
the Temne of Sierra Leone, was characterized by its adaptation to the
ecological environment of the lowlands and stagnant waters. It consisted of

swamp agriculture, the extraction of salt, paddy rice-growing and the use of
waterways for transport and communications.
Organized into village societies without slaves or castes but with age-groups
and work associations, the Joola had already reached the Iron Age when they
came into contact with the Manden. They did not have any state structure but
did have a unified civilization and culture. All authors accept that the caste
system, patrilinear succession and state structures are institutions of foreign
origin. Thus, under Manden influence brought about by the expansion of the
Mali empire, the various peoples of the coast were organized into chieftaincies
and principalities subject to the suzerainty of the mandi-mansa, the emperor of
Mali.
In the fifteenth century the Portuguese observed the existence of these

'western provinces' but reported the presence to the south ofthe 'Kingdom of
the Sapes'. Was this a large confederation comprising the Baga, Nalu, Landuma,
Limba and Temne peoples, or was it a kingdom formed between Cape Verga
and Cape Saint-Anne following the infiltration southward of the Tyapi-
Landuma-Baga-Nalu under the leadership of Manden chiefs? Historians have
divergent hypotheses on this question.
Kaabu: from emancipation to hegemony

Mali's expansion to the Atlantic along the Gambia basin had brought about the
establishment, in the fifteenth century, of politico-military structures such as the
little kingdoms of the Gambia, Casamance and Guinea-Bissau which formed
the western marches of the Manden empire. Their rulers paid tribute to the
mansa of Mali through his farins (governon). But by the sixteenth century it
was the farin of Kaabu who was receiving the tribute from the Manden chiefs
throughout the Gambia.
Lying between the middle Gambia, the Rio Grande and Futa Jallon, the
Malian province of Kaabu controlled the trade of Kantor. In the sixteenth
century Kaabu annexed the province of Sankola, south of the upper
Casamance, and became a powerful province, although still a dependency of
Mali. But the old Mali empire was weakening under attacks from Songhay, and
most of its western provinces freed themselves from its sway, with the farins
setting themselves up as independent kings; Kaabu did so in 1537.
After its emancipation, Kaabu unified all the countries between the Gambia,
the upper Casamance and the upper Geba, and supplanted the Mansa of Mali in
this region. It went on to extend its sway over the Kasanga, Balante, Joola and
Beafada peoples. Its farin wielded authority over all southern Senegambia and
controlled the trade of the region.
Atlantic trade and the peoples ofthe Southern Rivers

In Sierra Leone the Bullom, Temne and neighbouring groups played an active
part in trade with the Portuguese. The Bullom, who traded sea salt for gold
from Mali, now sold this gold to the Portuguese for cotton goods, copper bells
and various utensils. At the end of the sixteenth century the Temne, keen to
profit from the maritime trade, reached the estuary of the Sierra Leone, thus
splitting the Bullom into two.
By this time, not only had the Portuguese set up many trading setdements
along the coasts of the Southern Rivers, but they had established numerous
points of contact with the Manden in the middle and upper basins ofthe riven.
Links with the hinterland were thus intensified, as were northsouth ones. The
Cape-Verdeans traded directly with the region of the Sierra Leone rivers,
selling their cotton on the way to the Bainuk and Kasanga, and handicraft
articles from the coast reached as far as the Scarcies.
The coastal areas as a new West African trading front in the seventeenth
and eighteenth centuries
By opening the West African coast to world trade, the great European explora¬
tions committed the countries of Guinea to an unprecedented process of
historical development, with the expansion of trade, population movements
from the interior to the coast and the appearance of new societies. These
changes brought about the gradual marginalization ofthe Sudanese zone which,
to survive, had to reactivate the routes leading to the forest and the sea.
Tlie establishment and consolidation of European trade

The development of European trade on the West African coast was character¬
ized by three features: the consolidation of this trade through the systematic
organization of commercial exchanges, the increasing importance of the slave
trade and the bitter competition among European nations.
In the seventeenth century, the Dutch perfected the organization of the
world mercantilist system by creating chartered companies. The West India
Company, established in 1621 and given a 24-year monopoly over trade with
America and Africa, had a military and naval guarantee from the state and could
punue its commercial and colonial expansion in complete security. The Dutch
redistributed in Europe the products that they imported from Asia, Africa and
America, in particular a great quantity of gold and ivory, and thus made huge
profits. They supplied slaves to the American colonies of the other European
nations (Spain, Portugal, France and England) which, in 1641, acknowledged
their trading monopoly over the whole ofthe Guinea coast.
However, after 1660, France and England embarked on a merciless struggle
against the Dutch monopoly, equipping themselves with the same tools of
power as their rival. The English created the Company of Royal Adventurers
(1660) and later the Royal African Company (1664), while the French created
the Compagnie française des Indes occidentales (1664) and later the
Compagnie de Guinée (1685). Between 1640 and 1750, many forts and trading
stations were built on the coasts of Africa but changed ownenhip continually
according to the vicissitudes of war between the slave-trading nations.
From the shelter of these forts and around the trading stations, the Euro¬
peans punued the commercial exploitation of the Atlantic coastal area. From
Gorée to Sherbro Island, in Sierra Leone, trade flourished but no French or
English company could monopolize trade because of the many coastal islets
and offshore islands where any ship that wanted to could anchor. This part of
the coast was thus a happy hunting ground for slaven of all nations who were
not associated with the big companies. The chartered companies did not
bother with the Grain Coast or Côte d'Ivoire in the seventeenth and eigh¬
teenth centuries because malaguetta pepper was no longer prized on world
markets, ivory was in decline and anchoring difficulties kept ships away from
the coast. But by the beginning of the eighteenth century the growing
demand for slaves in America was encouraging individual traders to take an
interest in this sector.
Changes in the Sudan

In the Sudan, and particularly in the Manden world, the consequences of the
Adantic trade, dominated by the slave as merchandise, were various. At the
socio-economic level, relations with the coast became more important and the
slave trade was accompanied by the spread of firearms by Joola merchants. At
the socio-cultural level, Islam having lost its sway with the fall of the great
empires, the ruling class split into two rival factions. One, traditional and
political, set itself up as a military aristocracy, while the other, made up of
marabouts and merchants, advanced Islam and inspired politico-religious
revolutions. Both sought to resolve the social crisis born ofthe break-up ofthe
empires and the spread ofthe slave trade.
After the political fragmentation that followed the decline of the great
empires, surrogate hegemonies tried to reconstruct the various parts of the
hinterland. In the north and north-west the Grand Foul empire arose in the
seventeenth century on the ruins of Songhay but later, in the second half of the
eighteenth century, gave way to the empire of Kaarta. Along the Adantic, the
countries that had emerged from the dismemberment of Mali were unified by
Kaabu in the seventeenth and eighteenth centuries, and then by Futa Jallon in
the eighteenth and nineteenth centuries. Along the Niger, the recovery took
place at the beginning of the eighteenth century under the aegis of the
Bambara of Segu, under Biton Kulibali. But in the south, from the upper Niger
to the Bandama and eastward into the Volta-Senufo area, the Joola increased
their influence. In the eighteenth century they organized the empire of Kong
to ensure movement along the trade routes from the middle Niger to the forest
and the Gold Coast and they founded market towns.
Just as trans-Saharan trade had helped bring about Sudanese political hege¬
monies along the Sahel strip between the seventh and sixteenth centuries, so
the intensification of European trade encouraged the emergence of political
hegemonies on the coast and in the hinterland in the seventeenth and eigh¬
teenth centuries.
The sub-coastal hegemonies ofthe north-west: from Kaabu to Futa Jallon

In the seventeenth century Kaabu established itself as the great power in the
Southern Riven sector. Organized into twelve confederated Manden pro¬
vinces, it was led by a supreme mansa supported by an aristocracy of soldier-
officials and a standing cavalry corps. This kingdom, which had become an
empire, controlled the trade of the Gambia valley and the Guinea rivers on
which the European trading stations were situated.
In the second half of the eighteenth century the Fulbe Muslim confederation
established itself, emerging out of the Muslim revolution that shook the
Guinea-Sudan massif in about 1725-30. Situated at the confluence ofthe riven
that run down to the sea from the upper Niger area, Futa Jallon was to set itself
up as a serious competitor to the Kaabu Manden and fight them until they
collapsed in the nineteenth century.
The coastal hegemony ofthe Karon in Sierra Leone and Liberia

The history of the Karou has been written by O. Dapper. First there was the
Quoja kingdom, around Cape Mount, which was conquered by the Karou and
Folgia, peoples from the interior. The Karou leaden were essentially war-lords.
Absolute monarchs, defenders of the kingdom, they succeeded one another
from father to son. They put down revolts in the conquered provinces, repelled
the attacks of neighbouring peoples and constandy defended the integrity ofthe
empire. Yet they still remained allied to the Folgia and Manou of the interior,
with whom they had relations of vassalage. A rising chain of vassalage led from
the coast inland to the Mandi-Mani (the Lord of the Manou), the greatest
suzerain in the whole region.
The function of Karou hegemony was to regulate trade between the coast
and the interior. Their main economic base was trade with the Europeans, from
which they made enormous profits as middlemen. The trading area that they
dominated seems to have been indispensable to the operation of the mercantilist
slave-trading system in the region in the seventeenth and eighteenth centuries.
European trade and the comprador peoples ofthe coast

The development of trade in the seventeenth and eighteenth centuries brought
about the formation among the coastal peoples of merchant groups acting as
middlemen between the European ships and the African societies of the
interior. Such were the mestizos (half-castes) and Creoles of the riven in Casa¬
mance, Guinea and Sierra Leone.
From the Portuguese lançados to the mestizos and creóles ofthe rivers in Guinea and
Sierra Leone
Whether Afro-Portuguese descended from Portuguese setders of the fifteenth

and sixteenth centuries or Anglo-Africans descended from British traders in the
seventeenth and eighteenth centuries, the mulatto groups made up a very
special people whose influence on the historical development of the coast
between the Gambia and Sierra Leone was decisive.
Lançados and Afro-Portuguese Many Portuguese expatriates populated the

Southern Rivers and the Adantic islands (especially Cape Verde) from the fif¬
teenth century onwards. They included merchant adventuren and agents ofthe
trading companies and were called lançados. Most of them were Portuguese,
with a sprinkling of Greeks, Spaniards and even Indians. They recruited grumetes
(African auxiliaries) and plied the valleys of the Senegal, the Gambia, the
Casamance and the small riven of Guinea. Many setded in the Cacheu and
Geba sectors in what is now Guinea-Bissau.
The lançados soon laid the foundations of an Afro-Portuguese community:

setded in villages and living close to black people, they took African wives or
mistresses. But the flow of migrants slowed in the seventeenth century and the
number of pure Portuguese fell below that of the mulattos. These mestizos
became darker and darker until they could hardly be distinguished from pure
Africans. By the end of the seventeenth century the Afro-Portuguese
community thus comprised few whites and many blacks, but it was run by the
mestizos who gave it its main socio-cultural features.
The lançados had produced a new socio-cultural group on the coasts of
Africa. Speaking Portuguese, dressing in European style, living in houses with
whitewashed walls, they nevertheless adopted such African customs as tattooing
and ethnic markings. They practised a rather peculiar Catholicism in which the
celebration ofthe Christian saints did not inhibit ancestor wonhip.
The lançados initially established ties of friendship with the peoples of the
coast. Later, some of them became bound by kinship ties into the social,
political and cultural life of the African peoples. Thus José Lopez de Moura,
grandson of a Mane king of Sierra Leone, often took part in the political fife of
this sector. Others were simply Europeanized Africans.
This Afro-Portuguese group drew its strength from its function as a
comprador class. Its members did no productive work but supplied goods for the
European ships and carried on a seasonal trade related to the agricultural
activities of the indigenous Africans. Some were even rich enough to do
without the European ships. Among the great mulatto families, the Bibiana Vaz
family became prominent in the Cacheu sector in the seventeenth century. José
Lopez de Moura was the richest man in Sierra Leone in the first half of the
eighteenth century.
There were many disagreements between these wealthy mulatto merchants,
Supporten of unrestricted trade, and the trading companies anxious to preserve
their monopoly. The Afro-Portuguese merchants fiercely defended their free¬
dom to trade along the coast. Thus, when the Royal African Company
attempted to break the alliance between the African rulers and the mulattos and
put an end to the latter's role as middlemen, there was a sharp reaction. Led by
Lopez de Moura, the mulattos destroyed the English company's factory and it
was never able to resume its activities in Sierra Leone.
From English merchant adventurers to the Anglo-Africans in Sierra Leone

Another mulatto group was the Anglo-Africans of Sierra Leone. By the end of
Tlie states and cultures ofthe Upper Guinea coast 203
the eighteenth century they formed a group of nearly 12,000 people, among
whom the Tücken, Rogers, Corkers and Clevelands were the leading families.
The Tücken, Rogen and Caulken (later Corkers) came from England in the
seventeenth century and were initially involved in the trade of the English
companies. They married African wives and penetrated indigenous society,
where they were associated with the ruling class. As with the Afro-Portuguese,
a cultural hybrid social group came into being in which a man such as James
Cleveland, whose mother was a Kisi woman, could occupy a prominent
position in the secret society ofthe Poro.
Thus, in the seventeenth and eighteenth centuries the Afro-Portuguese and
the Anglo-Africans constituted a social group with specific economic functions,
and provided a cultural milieu in which Europeans and Africans came together.
But, conscious of their interests, they exploited the Africans from whom they
extracted maximum profits and, although they rebelled against the rigid
monopoly of the trading companies, they were none the less agents in the
service of European mercantile capitalism.
Trade and peoples from the Grain Coast to the Tooth Coast
From Cape Mesurado to Cape Lahou, the development of European trade in
the seventeenth and eighteenth centuries did not give rise to any dynamic
merchant groups among the Kru peoples. Although poorly linked to the
Sudanese hinterland, this coastal sector nevertheless boasted many villages built
at the mouths of rivers and abounded in a great variety of products. The
English and Dutch ships took on supplies of malaguetta pepper at the River
Sestos, Cape Palmas and Cavally and slaves at Bassa, Drewin and Saint Andrew:
ivory they bought everywhere.
Conclusion
Despite the difficulties inherent in drawing up an historical synthesis, it can be

seen that between 1500 and 1800 the peoples and countries of this region
broadly experienced the same development. From relative isolation they were
gradually integrated into the circuits of a world market built up by the
Europeans' Adantic voyages since the great explorations. This integration was
accompanied by population movements from the interior to the coast and by
sharp social and political changes wherever the peoples tried to take advantage
ofthe opportunities European trade offered them.
The states and cultures
of the Lower Guinea coast
For the peoples of the Lower Guinea coast living in the area between Côte
d'Ivoire and modern Benin, or between the Bandama and the Mono rivers, the
period between 1500 and 1800 was probably one of the most revolutionary
periods in their history. It witnessed, fint and foremost, the completion of the
migrations of these peoples and the formation of most of the ethno-linguistic
groups that exist today. Second, it saw the intensification of the trading and
cultural links between these peoples and those of western Sudan, the Sahara
and the Maghrib and the new possibilities of links across the Adantic fint
with Europe and then with the Americas. Third, it saw the gradual centraliza¬
tion of states and the evolution of larger political entities. Finally, the period
was one of social and cultural transformation. ,
Migrations and theformation of new ethno-linguistic groups

In 1500, most ofthe peoples were still living in their various places of origin
and only a few of them had penetrated into some of the areas in which they
live today. Their migratory movement occurred between 1500 and 1800,
particularly in the seventeenth century and the early decades of the eighteenth.
For economic, social and above all political reasons, the Akan, Ga-Adangbe
and Ewe peoples began to scatter in small clan and lineage groups.
In the sixteenth century the Akan migrated first north and east into the
present Kumasi, Mampong and Akyem areas, and then south and south-west
into the Wassa, Igwira, Sanwi and Assini areas. In the seventeenth century,
other Akan moved in various matrilineal clan groups north and north-west, and
also south-west into Wassa and Sefwi and the Lagoon areas of Côte d'Ivoire.
Between 1680 and 1730 there was a particularly large migration of Akan into
the Nzima, Aowin, Sefwi, Ahafo and northern Bono areas and the Anyi and
Baule areas of Côte d'Ivoire. By the mid-eighteenth century the Akan flow
seems to have dried up.
The Ga continued their dispenion towards the coast, attracted by the presence
of Europeans and the consequent economic opportunities that opened up. The
Ga-Mashi, the Nungua and the Tema were the fint to move in the sixteenth
204
The states and cultures ofthe Lower Guinea coast 205
century, followed in the seventeenth by the Osu and Teshi peoples. Their
kinspeople, the Adangbe, also began to spread out southwards and northwards
into the Accra plain during the sixteenth and seventeenth centuries.
The most dramatic migrations, however, were those involving the Ewe. In
the late sixteenth or early seventeenth century, to escape the tyrannical rule of
their king, Agokoli, the Ewe left Notsie in two main groups, the southern Ewe
and the northern or interior Ewe. The southern group moved towards the
coast while the northern groups occupied the central and northern part of what
is now the southern Volta Region. Sub-groups soon moved off to found other
settlements, and this process of fission and diffusion went on throughout the
seventeenth and eighteenth centuries.
These peoples did not only dispene and migrate; they also underwent ethno-
linguistic changes. The Ewe and Ga-Adangbe, who moved into very spanely
populated areas, were able to maintain their ethnic and linguistic purity. Many
of the Akan, however, migrated into areas quite densely populated by peoples
speaking different languages. The mixture of these peoples and the incoming
Akan gave rise to, among othen, the Mbalo, Alladian, Ebrie, Eotile, Avikam,
collectively known as the Lagunaires. Similarly, it was the mixture of the later
Akan migrants and some ofthe pre-existing peoples that gave birth to the Anyi,
Baule and Sefwi. The Akan therefore became divided into the two broad groups
of today, the Eastern Akan and the Western Akan. Each of these groups became
further divided: the Eastern Akan, divided into the Asante, Akuapem, Akyem,
Akwamu, Bron, Wassa, Kwahu, Fante, Assin, Denkyira and Gomua, all speak
the same language, Twi, while the Western Akan, divided into the Anyi, Baule,
Nzima, Ahanta, Sanwi, Aowin and Sefwi speak mutually intelligible Akan dialects.
In the sixteenth century, the socio-political institutions of these groups in
their new homes showed similarities as well as differences. The Ga-Adangbe
and the Ewe lived in scattered independent settlements, in patrilineage and clan
groups. In each group, each lineage had its own senior god whose priests were
its leaders. However, by the end of the fifteenth century the Ga-Mashi had
provided themselves with a king, who kept his court at Ayawaso.
The Akan lived in towns ruled by kings and queens and villages ruled by
chiefs. Each town or village was made up of families belonging to the eight
matrilineal clans into which they were all divided. Each family had an
abusuapanin (head), and so did each clan. It appean that each family or clan had
its own god or gods, as did each village, and that their priests wielded consider¬
able power. In other words, society had already become divided into three
broad classes: a ruling aristocracy consisting of priests and kings, ordinary
subjects and relatively small numben of domestic slaves. This social set-up was
gready altered by the political and economic developments of the period
between 1500 and 1800.
Economic changes in Lower Guinea

Far-reaching economic changes, both internal and external, took place in Lower
Guinea during this period. Internally, already-established economic activities

continued and expanded. These included gathering, farming, livestock-raising,
hunting, fishing, salt-making and gold-mining. The main item that was
gathered was kola nuts in the forest areas ofthe Gold Coast. These nuts formed
the mainstay of trade between the Akan and northern and western Sudan.
The main crops were plantain, bananas, yams and rice. In the sixteenth and
seventeenth centuries new crops introduced from Asia and the Americas by
Europeans were added to these traditional ones, including maize, cassava, various
types of yam, tomatoes, onions, aubergines, avocado pean, sweet potatoes, citrus
fruits and groundnuts. The rapid adoption of these crops divenified the agricul¬
tural economy ofthe Guinea coast and thereby accelerated population growth.
Livestock-raising involved poultry, sheep, goats and pigs. Fishing was the
main economic activity of the Ewe, Ga and other coastal peoples. Hunting was
a universal punuit.
Salt-making and gold- mining became increasingly important between 1500
and 1800. Nearly all salt was produced by the coastal inhabitants while gold-
mining was the exclusive preserve of the inland forest peoples, especially the
Akan. It was obtained either by panning the alluvial soils or by deep-mining
(nkoron). Gold-mining was rigidly controlled by the state and formed the main
source of income of the rulers. It reached its peak in the late seventeenth
century and remained the exclusive monopoly ofthe Akan, despite European -
particularly Dutch - efforts to participate in it direcdy.
Alongside these activities, trade developed, with slaves playing a growing
role in it. A distinction needs to be made between internal and external trade.
The Ewe, the Ga and the Akan had long been trading among themselves. Fish
and salt produced by the Ewe, the Ga and the Fante and Adangbe pottery were
exchanged for gold, chewing sticks (tweapea), ivory and iron and other metal
goods produced by the Akan of the interior. Two branches of external trade
developed: the older one with the savannah and western Sudan, and the other
across the Adantic with Europe and the Americas. The Ga, the Ewe and the
Akan had already established trading links with the Guan and the Gur to the
north, with the Hausa of northern Nigeria to the north-east and with the
Manden of the Niger bend to the north-west in gold, kola nuts, textiles, slaves
and salt. With the rise and expansion of Songhay in the fifteenth and sixteenth
centuries, trade between the Akan and the Mande prospered and it was to
control this trade effectively that the Joola established a number of trading posts
or staging posts such as Bobo-Dioulasso, Kong, Bonduku, Buna and Begho
between the Niger bend and the gold-producing countries in the south. The
towns of Jenne and Begho, the main entrepôts ofthe trade to the north-west,
became extremely wealthy during the sixteenth century, as is evident from
both written and archaeological sources. In the north-western markets, the
Akan exchanged mainly gold, kola nuts and, later, European imports for cloth,
blankets, Turkish carpets, striped and blue silks, leather goods, ironware, brass-
ware and salt.
The overthrow of Songhay in the late sixteenth century and the subsequent
insecurity were once thought to have destroyed trade along the north-west
routes. But it has now been convincingly shown that trade did not decline and
even increased in volume during the eighteenth century.
Trade between the Akan and Hausaland and Borno also continued. It was
already of some importance by the beginning of the sixteenth century, as

evidenced by the eye-witness account of Leo Africanus, who visited Songhay
and Hausaland at that time. It grew in intensity in the seventeenth and
eighteenth centuries with the growth ofthe Mole-Dagbane states of Mamprusi,
Dagomba, Nanumba and Mossi and of the Hausa states. Towns such as Kano
and Katsina, in Hausaland, and above all Salaga to the north ofthe Gold Coast
were extremely prosperous.
The export of gold by the Akan to Hausaland seems to have ceased in the
eighteenth century, probably because of strong European competition, but Akan
kola continued to be the mainstay of trade with the north-east. Kola nuts were
re-exported from Hausaland and Borno into the Sahara and to the Barbary states.
A far more important trading relationship, however, was also developing
elsewhere the trans-Atlantic trade between the Lower Guinea coast and
Europe and the Americas. It started when the Portuguese dropped anchor off
the coast of modern Ghana in 1471. They were followed in the sixteenth
century by the French, the English and the Dutch and in the eighteenth by the
Danes. Direct trade with the West Indies began in 1518 when the first cargo of
slaves was sent there in a Spanish ship. Trade with the North American main¬
land began in 1619 when a Dutch frigate discharged the first cargo of West
African slaves at Jamestown in the state of Virginia. Thus, by the midddle of
the seventeenth century, the triangular trade linking the Guinea coast, Europe
and the Americas was fully established.
The commodities sold to the Europeans varied from region to region and
century to century. Until the late seventeenth century the Akan between the
Bandama and the Ankobra rivers sold mainly ivory to the Europeans, while the
Akan and the Ga living between the Ankobra and the Volta sold mosdy gold.
Throughout the sixteenth century there seems to have been little trade
between the Ewe and the Europeans in the area between the Volta and the
Mono rivers. It was not until the mid-seventeenth century that the Dutch and
the Danes embarked on trade there, and the main commodity was slaves. It was
as a result of this specialization that European traden divided the Lower Guinea
coast into the Ivory Coast, the Gold Coast and the Slave Coast.
Until the end of the seventeenth century, the gold trade was by far the most
important. It impelled every European nation to gain a footing on this coast,
with the result that a huge number of forts and casdes was built along the coast
between the Ankobra and the Volta riven.
Between 1650 and 1800, however, the export trade between the Ga and the
Eastern Akan and Europe underwent a revolutionary change. Starting in the
mid-seventeenth century, the Ga and the Akan exported slaves in such increasing
numben that by 1710 the trade in slaves exceeded that in gold. The number of
slaves exported from the Gold Coast continued to grow throughout the first half
ofthe eighteenth century. According to K. Y. Daaku, 5,000-6,000 slaves were

being exported annually by the late seventeenth century, and this figure rose to
6,000-7,000 in the eighteenth century. According to P. D. Curtin, a total of
474,000 slaves was exported from the Gold Coast in the eighteenth century, but
it is widely believed that Curtin underestimated the number of slaves exported
from West Africa by 7.3 per cent. The Ewe areas also exported large
numben of slaves, especially after 1730 when the Akwamu invaded the region.
Why did the trade in slaves supenede the trade in gold on the Gold Coast?
There are three main answen to this question. The first is the great increase in
the demand for slaves following the introduction of sugar-cane plantations in
the West Indies and on the American mainland from the 1640s onwards. The
second is the enormous rise in the number of war-captives. The third reason is
that after the rise of the Denkyira, Akwamu and Asante empires, the vassal
states paid their tribute in the form of slave deliveries. It should be added that
the supply of slaves to the Europeans on the Gold Coast was the exclusive
preserve ofthe Africans themselves.
Imports into Lower Guinea also underwent revolutionary changes in both
volume and quality. In the late fifteenth century and throughout the sixteenth
centuries, the principal imports were articles of clothing mostly manufactured
not in Europe but in the Barbary states, Benin and Côte d'Ivoire. Fabrics from
Benin and Quaqua cloths from Côte d'Ivoire were very popular and they were
certainly still being imported in the seventeenth century, but by then the list of
imports had gready lengthened. Pieter de Marees, writing in the early
seventeenth century, gives a detailed list of the goods brought by the Dutch
and concludes that the most sought-after wares were linen cloth, brass and
copper objects, basins, pots, knives and corals.
This list does not mention firearms, since the import of guns and gunpowder
only began in earnest in the 1640s, when the English and the interlopen started
selling arms on the coast. In 1 660 the Dutch lifted their firearms export ban and
started exporting them in large quantities. The volume of firearms exports
continued to rise throughout the eighteenth century. They remained the item
most in demand on the Gold Coast and ended up becoming the mainstay of
English trade with West Africa.
The effects of economic developments

One ofthe first effects ofthe economic changes was the appearance of a com¬
plex network of major and minor trade routes. These fint linked the peoples of
the area to one another, then linked them to the Mole-Dagbane and Hausa to
the north-east, the Manden to the north-west and the Barbary states and the
Muslim world across the Sahara. Finally the network spread across the Atlantic,
fint to Europe and then to the Americas. At the centre of this network lay the
town of Kumasi.
A number of urban centres developed along these routes. These included

Kong, Bobo-Dioulasso, Buna, Begho and Bonduku to the north-west, Salaga,
Yendi and Sansanne-Mango in the north-east and Tiassalé, Sakasso, Yakasso,

Krinjabo, Kumasi, Kete-Krachi and Akwamufie to the south. These centres
served as entrepôts, market centres and termini. Furthermore, the European
presence accelerated the growth ofthe towns on the coast at the expense ofthe
former capital towns immediately inland.
The effect of the new network established with Europe was the integration
of the Lower Guinea coast into the larger economic area formed by the
countries of western Europe and America on the one hand and the Manden-
Hausa-Muslim world on the other. However, this latter link gradually
weakened while the one with Europe and the Americas grew ever stronger.
The effect of this integration was the destruction or atrophy of economic
and industrial development in Lower Guinea. The slave trade the most
inhuman of all trades steadily replaced trade in natural products, depriving the
region of needed manpower as well as of many of its artisans and artists. In
addition, Europe exported cheap mass consumer goods, thereby killing most
existing industries or seriously retarding their growth. In short, between 1500
and 1800 there was economic growth in Lower Guinea but without economic
development. Moreover, since Europe controlled both imports and exports, it
kept most of the profits. Herein lie the roots of the process of underdevelop¬
ment, aggravated in the nineteenth century by the abolition of the slave trade
and the establishment of colonialism in Africa.
Political change on the Lower Guinea coast

Political change between 1500 and 1800 was even more radical than economic
change. By 1500 the process of state formation had already begun and a
number of states had emerged both on the coast and in the savannah regions.
The Portuguese arrived to find the coastal states of Ahanta, Shama, Aguafo,
Fetu, Asebu, Agona and Accra, and there were also the Mole-Dagbane states
and the Akan state of Bono in the savannah belt. The process continued to
gather momentum when the first Akan kingdom, Bono, emerged in the second
half of the fifteenth century, and reached its peak through the lucrative trade
between the Manden and the Akan.
During the sixteenth century, the Ga-Mashi, the Nungua and the Tema were
joined on the Accra plain by the Labadi and the Osu, all of whom founded
numerous setdements. Until the fourteenth century the Ga had priests as chiefs,
but by the end of the following century secular kingship had evolved, probably
borrowed from their Akan and Adangbe neighboun. At the beginning of the
seventeenth century some Ga moved to the coast, attracted by the European
presence, and they were followed by the La and the Osu. All the coastal towns
that they founded recognized the suzerainty of the Ga mantse (king) and Ayawaso
remained the kingdom's capital until 1680, when it was moved to the coast.
What seems to have happened among the Akan in the sixteenth century is
the formation in the Pra-Ofin basin of a number of small communities and
city-states or chiefdoms bound together not by allegiance to a common
sovereign but by kinship, agnatic and clan ties. However, it appears from the
oral traditions of Adansi that, in about 1550, there was a move towards
centralization which led to the formation of the Adansi confederation of states
and also precipitated the migration of some Akan peoples northwards and
southwards where they founded chiefdoms, city-states and principalities.
This process of state formation seems to have accelerated bet-ween 1580 and
1630, as is demonstrated by a map of the states of the southern Gold Coast
between the Tano and Volta rivers drawn by a Dutch cartographer in 1629.
This map describes 38 states and kingdoms. All except Great Incassa and Incassa
Igwira have been identified and still exist in the same areas.
The states founded by the Ga and the Akan who had migrated were
probably small but were all similarly structured, with a single leader, chief or
king with his queen. In the Adansi confederation, power rotated among the
royal families of the various states. In the single states, the king was selected
from the first family or clan to arrive in the area. He was advised by a council
composed of family and clan heads. Each state had its own gods, for example
riven, lakes or rocks, and their priests wielded considerable influence in society.
Between 1630 and about 1670, two main political developments took place
in the Ga and Adangbe areas: the consolidation ofthe states shown on the 1629
map and the emergence of new states. Oral sources show that kingdoms like
Akwamu, Denkyira, Accra or Ga, Fante, Wassa and Adorn gready expanded
their frontien, largely through peaceful means. The Ga kingdom reached its
greatest extent and the peak of its power under the king Okai Akwei (c. 1640
During this period the Aduana state of Akwamu also became a strong
kingdom while Denkyira broke away from the Adansi Confederation after a
series of wars and firmly established itself at the confluence ofthe Ofin and the
Pra. On the coast the Fante also extended their territory inland and European
records frequendy refer to wan between the Fante and the Etsi to the north.
The DenkyiraAdansi wan (1650-70) and the Bono wars had accelerated
migration southwards and westwards into the forest, where older peoples such
as the Adisi, the Ewotre, the Agwa, the Kompa and the Lagunaires lived. These
new groups founded not only Aowin, near present-day Wassa Amanfi, but also
the three Sefwi states in the south and states such as Assini, Abripi quern and
Ankobra and a host of coastal towns. Situated in one ofthe richest gold-bearing
areas and on the main trade route linking the northern markets with the coast,
by the 1 670s Aowin had grown into a rich and powerful kingdom.
The Denkyira-Adansi migrants who moved northwards founded the Oyoko
states of Kumasi, Kokofu, Dwaben, Nsuta and, later, Bekwai and the Bretuo
states of Mampong and Afigyaase, all within a 50-kilometre radius of Kumasi.
Further north, othen founded the Aduana state of Gyaaman or Abron.
The new states seem to have been organized on the same basis as the old ones.
The Sefwi and Aowin states, for example, superimposed the sophisticated Akan
clan system and the institutions of matrilineally elected kingship on the existing
socio-political structure based on asago (warrion) grouped around living-quarten.
However, in the last 30 yean of the seventeenth century a political
revolution took place marked not by the emergence of new states but by the
increasing centralization of existing ones. Aowin and Denkyira seem to have
led the way in this shift. Aowin conquered the Sefwi states to the north and
west and expanded far west to embrace such towns as Keteso, Yawu, Brako
and Sikasso in what is now Côte d'Ivoire.
In the 1670s and 1680s the Agona rulen of Denkyira conquered all the
Adansi states together with the pre-Asante states near Kumasi to the north and
Assin and Twifo to the south. Between 1686 and 1690, Denkyira defeated
Aowin, the Sefwi and Wassa states to the south-west and the coastal kingdoms
of Adorn and Fetu. By 1690, Denkyira dominated the south-western Gold
Coast and parts of Côte d'Ivoire.
In the south-eastern regions, the Akwamu leaders were also embarking on
expansionist wan. They first attacked the Ga kingdom, which they conquered
in 1681. They then conquered the coastal kingdom of Agona to the west in
1689. In their final campaigns (1702-10), they conquered the Adangbe states to
the east and Kwahu to the north and, crossing the Volta, subdued the Ewe
states of Peki, Ho and Kpandu.
The Denkyira and Akwamu evolved similar machinery to administer their
empires. Each was divided into two areas, the metropolitan area and the
provincial area. The metropolitan area was centred on the capital, the seat of
the omanhene (king), who reigned over the whole empire. Under him there was
a series of officials or kings who performed certain functions at court. For
politico-military purposes, each state was divided into wings, three in Denkyira
and five in Akwamu. Each wing was headed by the king of a town or state in
the metropolitan area or in the capital itself. He exercised political control over
this wing in peacetime and became its osafohene (war-leader) in wartime. All the
wing chiefs were members of the council that advised the omanhene. The
provincial area consisted of all the conquered states now part of the empire.
Each continued to be ruled by its own king but was placed direcdy under the
control of the omanhene or a wing chief. Each state was expected to pay annual
tribute and to fight in battle under its wing chief.
Many historians have seen the political history of the Gold Coast in the
eighteenth century almost exclusively in terms ofthe rise ofthe Asante empire,
but the coune of political events was much more complex. First, other states
came into being outside the Asante empire. Second, there was the revival,
albeit short-lived, of Aowin and the overthrow of Denkyira and Akwamu by
Akyem and Asante respectively. Third, there was the expansion of the Fante
kingdom to its widest extent and, finally, the brilliant rise ofthe Asante empire.
The wholly new states that came into being in the eighteenth century were
Nzima, the Aowin or Anyi states of Sanwi, Ndenye, Diabe, Moronou and
Bettie and many Baule principalities. After being defeated by the Denkyira in
the 1680s, the Aowin, or Anyi, left Anwianwia and, crossing the River Tano,
recreated their kingdom around Enchi. They were still consolidating when in
about 1715 the Asante attacked. Some then migrated westward where they
founded the kingdom of Sanwi. Having established their capital at Krinjabo
they quickly seized Assini, thus securing control of trade between the hinter¬
land and the Aby lagoon.
North of Sanwi emerged the other Anyi states, Ndenye and Diabe, founded by
the group of refugees from Anwianwia that moved north-westward. Until about
1715, the Anyi of Ndenye owed allegiance to Aowin, but a section of them
refused to accept this and moved away to found the kingdom of Bettie. Some
Anyi, the Monfwe, crossed the River Comoe to establish the kingdom of
Moronou.
North-west ofthe Anyi, between the Comoe and Bandama rivers, there also
arose, in the first half of the eighteenth century, a host of Baule principalities or
chiefdoms. The oral traditions of the founders show that they arrived from the
Gold Coast in two separate waves. The first brought the Alanguira Baule in
about 1700. They had come from Denkyira after its defeat by the Asante, and
setded in the area ofthe present-day canton of Agba. From there some moved
west to live among the Guro and the Koro.
The second, much larger wave, the Assabu, came from Kumasi following a
disputed succession that broke out after the death of Osei Tutu in 1717.
According to their oral tradition, they were led by their queen Abla Poku, who
sacrificed her son to the spirit of the River Comoe before they could cross it,
whence their name Baule (the infant is dead). The group divided, but the bulk,
still led by Abla Poku, crossed the Bandama and went and settled near Bouaké.
Akwa Boni, who succeed Abla Poku, was able to impose his authority over the
Baule, the Malinke and the Manden, who were occupying the Warebo region.
After his death, however, the kingdom broke up into independent chiefdoms.
Both waves of invaden soon mixed with the Guro, the Malinke, the Senufo
and the Goli to form the modern Baule people.
The Nzima kingdom, which also emerged at this time, was the creation of
three brothers who amalgamated under their rule the three existing states of
Jomoro, Abripiquem and Ankobra through the fortune that they had acquired
from trading with the Europeans. Known in the European records as
Apollonia, this new state dominated the south-western corner of the Gold
Coast throughout the eighteenth century.
Meanwhile, some of the older states, such as Bron, Sefwi-Wiawso, Aowin
and Fante were extending their power and influence. Aowin, for example,
appears to have regained its independence from Denkyira in the 1690s. Greatly
strengthened by refugees fleeing the Asante-Denkyira wars, it came to dominate
the Sefwi states and embarked on the conquest of the gold- and ivory-
producing areas to the north. Aowin became so powerful that its warriors
invaded Asante in 1718-19, sacked the capital, Kumasi, and returned home with
booty and captives, including some members ofthe Asante royal family.
Also at the beginning of the eighteenth century, the state of Bron consider¬
ably enlarged its territory through conquest. The new Bron state was unique,
being made up of Akan, Kulango, Nafana and Manden elements. The Akan
invaden, who dominated the state politically, retained their clans, their matri¬
lineal system of inheritance, their judicial system and aspects of their traditional
religion. But they left untouched Kulango and Nafana social structures and
political organization at the rural and village level. Finally, some adopted Islam,
introduced by the Manden-Jula, who also greatly influenced the economic
system ofthe kingdom.
It was also between 1700 and 1710 that the Akwamu extended their
territorial empire across the River Volta. Partly in response to this Akwamu
expansion and later that of Asante and Akyem, the Fante conquered their
neighbouring coastal states, thereby gaining control of the coast from the
mouth ofthe Pra to the Ga kingdom.
The political changes that occurred in the central forest regions between the
Comoe and Volta rivers were even more dramatic. First there was the fall of
the Denkyira empire in a series of wan between 1699 and 1701 waged by the
young Asante confederation of states. This defeat was followed by the conquest
of all Denkyira's former vassal states: the Sefwi states, Twifo and Wassa, then
Aowin, Nzima, the Anyi state of Ndenye, Wenchi and Bono, and finally the
Bron state and Gonja. By 1730 the whole ofthe area between the Comoe and
Volta rivers was under Asante control.
Then there was the defeat of Akwamu by Akyem Abuakwa and its allies, the
Ga, Kotoku and Agona. The Akwamu rulen were driven out of their lands
across the Volta where they founded their present capital, Akwamufie. Thus,
by 1733, the whole area between the Comoe and Volta rivers was partitioned
among the Asante, the Akyem and the Fante.
The process of centralization was completed between 1731 and 1750 when
the Asante conquered not only Akyem Kotoku and Akyem Abuakwa and the
Ga state to the south, but also eastern Gonja and the Dagomba state to the north
of the Volta river and the Krakye and Bassa states to the north-east. The Fante
alone maintained their sovereign existence throughout the eighteenth century,
partly through their own diplomatic skills and partly through the support of the
British who wanted to stop the Asante gaining control over the whole coast.
The structure and government of the Asante empire were similar to those of
the Akwamu and Denkyira empires. It was divided into two parts: metro¬
politan Asante and provincial Asante. Metropolitan Asante consisted of all the
former states within a 50-kilometre radius of Kumasi. All recognized the ohene
ofthe Kumasi state as their asantehene (paramount king). They were represented
by their own king on the Asanteman Council, the governing body of the
confederation and the entire empire. For politico-military reasons, metro¬
politan Asante was divided, like Akwamu, into five wings.
Provincial or Greater Asante consisted of all the states conquered and
reduced to vassal status. These states had no direct representative on the
Asanteman Council nor any direct access to the asantehene. Instead each state
served the asantehene through an intermediary, who was one of the kings or
one ofthe member states ofthe confederation or one ofthe wing chiefs ofthe
Kumasi state. This system of provincial administration (adamfo) was liberal and
the states were largely left alone so long as they paid their annual tribute and
participated in any wan being fought by the Asante.
Thus a political revolution had taken place in the forest and coastal regions
of Lower Guinea between 1670 and 1750. From the 38 states ofthe 1629 map
had emerged the three great empires of Aowin, Denkyira and Akwamu, which
by 1750 had merged into the single empire of Asante.
If these empires were able to come into being, it is because they had the
means to buy arms and ammunition and also the necessary motivation to
expand. Aowin, Denkyira, Akwamu and Asante were the main gold-producing
states and they acted as commercial middlemen between the savannah and the
coastal regions. By expanding northwards and southwards these states would
gain control of the major trade routes. In addition, the presence of the Euro¬
peans on the coast provided a strong motivation, since trade with the Europeans
had become much more lucrative than trade with the savannah. It is thus in no
way surprising that so many should have tried to fight their way to the coast.
The states that made up metropolitan Asante also had a political motivation.
They were eager to rid themselves of their tyrannical Denkyira conquerors and
were only waiting for the right chiefs to lead them. They found them in Osei
Tutu and Opoku Ware, the founden ofthe Asante empire.
A further reason for the rise of these empires is that, when they embarked
on their expansion, they were free from interference in their internal affairs.
The Akyem states, though not lacking either the means or the motivation,
were unable to develop into great empires because they did not have that
freedom. Sandwiched between expanding Denkyira, and, later, Asante and
Akwamu, they were constantly on the defensive. Similarly, the coastal states
were hampered by constant interference by rival European nations. The inland
states of Aowin, Denkyira, Akwamu and Asante were at the outset free from
such interference and could sufficiently consolidate their power before coming
into contact with the Europeans.
There are two other factors which should be taken into account in
explaining the rise of these empires: the adoption of a new technology and the
high quality of their leadership. The rulers of these states had the means to buy
weapons and ammunition from the Europeans, and that gave them a decisive
edge over those who did not have them. The old way of fighting was replaced
by the organization of the army into fighting units and this new military
formation was superimposed on the traditional political structure. The rulers'
ability to adapt to change thus became of vital importance.
At this stage little is known about the Aowin kings, but both oral and
European documentary sources acclaim the courage and ability ofthe rulers of
Denkyira, Akwamu and Asante. The Denkyira kings included Werempe
Ampem, Boadu Akafu Brempon and the most illustrious, Boa Amponsem;
Akwamu's included Ansa Sasraku, Basua and Akwono and Asante's Osei
Tutu and Opoku Ware. Many historians believe that these empires were
products ofthe slave trade. What may be true of other parts of West Africa is
not for this area. On the coast of Ghana the slave trade only became
significant economically from 1700-10, while expansionist activities had
begun in the 1670s and 1680s. On the Gold Coast the slave trade was the
consequence, not the cause, ofthe state-building process.

In the Akan areas west of the River Tano and the Ewe areas east of the
River Volta, ancestral home ofthe Lagunaires, the Anyi and the Baule, no such
dramatic political change took place at this time. In 1800 they were organized
much as they had been 150 yean earlier, with small kingdoms or chiefdoms,
each made up of loosely united family groups from clans of common ancestry.
In 1800, Ewe territory was still split into many independent territorial units
or dukowo of varying sizes. Each dukowo was ruled by z fia (king) elected patri-
lineally from one or two lineages of the founding families and assisted by a
Council of Elders.
There are several reasons why the Western Akan of the Tano-Bandama basin
and the Ewe of the Volta-Mono basin did not undergo political change. First,
migrations into the two areas continued throughout the seventeenth and
eighteenth centuries. Consequendy, by 1800, the Ewe and Western Akan were
not sufficiendy settled to embark on wan of expansion. Second, as these
migrants were themselves escaping the centralizing activities and tyrannical
behaviour of their former rulen, they had no desire to recreate the same socio¬
political structures in their new homes. Third, in the case of the Ewe, the
economic motivation for state-building was lacking. Without gold, ivory or
kola nuts, the Ewe could not participate in the lucrative northsouth Adantic
trade. During the seventeenth and eighteenth centuries the region's main
export was slaves, an activity which jeopardized peace and stability and thus did
not stimulate political expansion.
The Western Akan did have gold, ivory and kola nuts, but because they
were not in control of the trade routes to the north, they did not really have
the means to embark on a large-scale expansionist venture.
For the Western Akan, as for the Ewe, there was also an ecological con¬
sideration in the shape ofthe lagoons which formed a barrier between the coast
and the hinterland and discouraged commercial exchanges.
Finally, their more powerful northern neighboun' constant interference kept
them constandy on the defensive and concerned above all with maintaining
their own independence.
Social and cultural changes in Lower Guinea

The most obvious socio-cultural change in Lower Guinea between 1500 and
1800 was demographic. With the introduction of so many food crops from the
New World and Asia, the population gready increased. During the eighteenth
century, however, largely as a result ofthe slave trade, the population remained
stagnant or even decreased.
Second, society had become more sophisticated. In 1500 it had been com¬
posed of a ruling aristocracy consisting of a religious élite and a political élite,
ordinary subjects and domestic slaves. By 1800, the political élite had super¬
seded the religious, except among the Ga-Adangbe and the Ewe. The
tremendous increase in economic activities and the wars of expansion had led
to an increase in the number of domestic slaves. By 1800 most slaves had

become integrated into the society in which they lived and it had become a
sacred law, especially among the Akan, not to divulge their origins.
Moreover, as a result of trading activities and the European presence, three
new classes unknown to traditional society had emerged a wage-earning class,
an independent class of wealthy traders and merchant princes and a mulatto
group. The first class consisted of people employed by the Europeans as masons,
carpenters, interpreten, writen or secretaries, gold brokers, civil servants, ambas-
sadon and public-relations officen.
The second class consisted of Africans who, through their own efforts in
farming or trade, became exceedingly wealthy and more powerful than the
traditional rulen. On the coast of present-day Côte d'Ivoire, for example, the
Kosehirange had become extremely powerful and were playing a key role in
the choice of lineage heads.
The mulatto group was the product of unions between European traden and
African women. Such people could be found all along the Lower Guinea coast.
Some, such as the descendants of Richard Brew, played important roles in the
commercial and political life ofthe region.
Another important change was the introduction of Christianity and Western
education by the Europeans, and of Islam by the Manden and Hausa traders.
The Dutch and the English established elementary schools at Cape Coast, El
Mina and Accra, and in the 1750s the Society for the Propagation ofthe Gospel
sent missionaries to Cape Coast. Moreover, some children of mulattos and
traditional rulen were sent abroad to be educated, returning home as teachen
and missionaries. Thus, by 1800 a small educated élite and a few Christian con¬
verts could be found in some coastal towns.
Even earlier than Christianity, Islam and Muslim culture spread along the
northern trade routes, fint into northern Ghana by the fourteenth century and
then into Asante and Baule areas by the mid-eighteenth century. By 1800,
Kumasi had a thriving Muslim quarter with a Kur'änic school. However, neither
Christianity nor Islam had much impact on the peoples of Lower Guinea,
although reading and writing in both European languages and Arabic had been
firmly established.
As regards arts and crafts, the peoples of the Guinea coast had, by 1800,
become highly skilled in pottery-making, were already making pottery, carving,
weaving, metalworking and casting brass and gold objects.
Pottery-making in Ghana goes back to the Later Stone Age (after 3000
before the Christian era) and had gready developed by the early eighteenth
century, particularly among the Adangbe. According to J. Anquandah, the
Asante produced some of the best-quality polymorphic pottery, such as the
abttsua kuruwa, clan pot, and the mogyemogye, a wine jar used for pouring
libations on the Golden Stool.
Carving and sculpting in wood, ivory and clay was also developed during
this period, particularly among the Akan. Seventeenth- and eighteenth-century
European visitón to the coast of Ghana, such as W. Bosman, were impressed
by the beauty ofthe ivory side-blown trumpets. Among typical objects of this
art, the best-known are the akuaba (fertility dolls) in wood and clay and the
sculptured clay portraits of deceased kings and queens.
Weaving became widespread throughout the region in the sixteenth and
seventeenth centuries and, according to the 1629 map, Nsoko (Insoco), present-
day Begho, was an important weaving centre. But it was in the eighteenth cen¬
tury that weaving among the Akan and the Ewe attained its maturity, as demon¬
strated by the famous colourful kente cloths of the Akan and the rich adanudo
cloths ofthe Ewe. Adinkra cloth, with Akan traditional motifs and symbols stamped
on it, became famous in the Brong region and was later copied by the Asante.
However, it was in metalworking and casting, especially in gold and brass,
that the peoples of Lower Guinea and the Akan in particular excelled. Using
mainly the cire-perdue (lost-wax) process, the smiths produced exquisite gold and
silver objects, including sword handles, rings, bangles, chains and headgear. The
Akan smiths also produced thousands of geometric and figurative gold or brass
weights, now found in all the world's great art museums.
With the rise of the Asante empire in the eighteenth century, the arts and
crafts of Lower Guinea reached their peak. The Asante kings did everything in
their power to promote them. After the defeat of Denkyira, Tekyiman and
Akyem, they collected the best craftsmen and goldsmiths of those states and
sent them to Kumasi. A number of specialized centres was established in the
neighbourhood of the capital. The asantehene established the Apagyafie, a group
of goldsmiths and other craftsmen brought from Denkyira whose duty was to
make the king's regalia. He also introduced the golden stool (asomfo), whose
fint occupant, Nana Tabiri, was the son of a Denkyira chief.
This blending of different artistic experiences, expertise and traditions
enabled the Asante kings in the eighteenth century to raise the cultural
development of the Akan people to its highest peak of excellence and to ensure
that their power was reflected in works of art ofthe highest quality.
Conclusion
The period between 1500 and 1800 was indeed a period of radical change for
the states and peoples of Lower Guinea. Politically, the process towards
centralization was completed. Economically, the trade in gold and ivory was
eclipsed by the slave trade; the commercial and economic centres moved from
inland to the coast; strong commercial links were forged with the Americas and
Europe, thereby integrating the region's economy with the world economy.
Above all, this was a period of social change, marked by the emergence of new
social classes although on a limited scale and confined mainly to the coast, the
beginnings of literacy, the introduction of Western education and Christianity,
the spread of Islam and the flowering of indigenous cultures, expressed
especially in weaving and metalwork. These were indeed dynamic centuries for
the peoples of the Lower Guinea Coast, all the more so because at the end of
the period they were still in full control of their destiny.
Fon and Yoruba:
the Niger delta

and Cameroon
This chapter coven the region extending from the valley of the River Volta in
the west to the River Cameroon in the east. Most of the area is tropical forest
bordered by savannah with shrub forest to the north. The western part, from
the border of Nigeria to the River Volta, is also savannah. The peoples living
within this part of the Guinea forest and the surrounding savannah comprise
the Fon or Aja ofthe modern Republic of Benin, the Yoruba, the Ijo ofthe
Niger delta, the Igbo to the north-east of the delta, the Ibibio and various
peoples in southern Cameroon.
All the languages ofthe area belong to the Niger-Congo family, the majority
of them being within the Kwa sub-family. The Efik/Ibibio and the other
languages of the Nigeria-Cameroon border area of Nigeria and the languages
of Cameroon itself are closely related to the Bantu languages of central, eastern
and southern Africa. The peoples and cultures in this border region form a
unifying link between West Africa and Bantu Africa. Among the Kwa language
groups, the Yoruba and the Igbo are the largest in terms of the numbers of
speakers (between 8 and 12 million) and geographical spread. The Edo group is
also large, with a spread of related peoples such as the Isoko and the Urhobo,
the Ishan (Esan) and others. Among the Kwa languages, Ijo in the Niger delta
is the most divergent from its neighboun, Igbo, Edo and Yoruba.
The long continuities in this region's history should be borne in mind when
evaluating the changes supposedly induced by the arrival of Europeans on the coast
at the end ofthe fifteenth century. The Portuguese reached Benin in 1486 and
established relations with the oba of Benin. They established commercial relations at
various points along the coast which became the major origin of stimuli to change.
The slave trade was obviously the focus of European activity between the six¬
teenth and the eighteenth centuries. The region covered in this study was one of
the main markets for slaves on the West African coast. Some states, such as the
kingdom of Dahomey, derived great impetus for their formation and growth
from the trade. The development of othen, from the Niger delta to Cameroon,
was influenced by the profits to be derived from it. Other communities,
especially those organized in non-state forms, tended to be the victims of the
overseas slave trade.
221
Fon and Yoruba: The Niger delta and Cameroon 223
The increase in the local slave trade also led to social and political changes
within the communities. Oral traditions show the effects of the trade on the
fortunes of lineages, groups or dynasties.
The Fon kingdom of Dahomey

The political development of the kingdom of Dahomey and of the neighbour¬
ing states of Aliada, Whydah, Popo and Jakin was largely related to the
activities of the European slave-traders on the coast and to the influence of the
Yoruba kingdom of Oyo to the north-east.
The traditional institutions of the small communities and states of the area
were weakened by the introduction of the slave trade and, by the end of the
seventeenth century, a political vacuum had been created. It was at this point
that the founden of the kingdom of Dahomey re-established order by
fashioning a new form of political organization out of various groups of Aja
peoples comprising the Gun, Fon, Arada and other peoples from the southern
part of the modern Republic of Benin. By 1700, Dahomey had become a
major power in the area. Between 1724 and 1727, its ruler, Agaja, embarked
on the conquest of the small but older states surrounding Abomey. This
provoked intervention by Oyo, but not to establish a system of its own. After
1730 Dahomey accepted the political authority of Oyo and agreed to live
within the economy ofthe slave trade.
The crisis that began about 1767 culminated in the fall ofthe Agaja dynasty
in 1818 and the rise of the new dynasty of Gezo. The new dynasty adapted
fully to the slave-trade economy and made it the basis of its strength. It thereby
benefited from the breakdown of the Oyo kingdom and the Yoruba wars in
The inter-relationship between the Aja communities and the Yoruba to the
east and north-east was deep and long-standing. There are indeed states of
Yoruba origin and culture within this area, notably the Ketu, a migrant
community of Yoruba from Ile-Ife.
The states in the area were largely organized in systems similar to the Yoruba
pattern, the 'father' kingdom being Aliada, which had been founded about
1575. Oyo intervention, by restricting Dahomey as a military power, enabled
the young state to strengthen its internal administration. Some aspects of Oyo
organization were taken over, such as the ilari system introduced by Tegbesu,
the last ruler ofthe founding dynasty.
The effects of the slave trade were so significant because at the point when
the trade started in the sixteenth century, the states were still weak. Economic
activity was undermined and the rules of society destroyed. Moreover, the
rivalries arising out of the competition for trading posts destroyed the tradi¬
tional relationships among states. Early in the seventeenth century, the Dutch
established agents at Assim, the capital of Aliada. The French set up a rival
trading station at Whydah in 1671. That destroyed the already weakened ties
between the two states. It was in this atmosphere of political and commercial
competition that migrants from Aliada created the kingdom of Dahomey.

They setded in Abomey, out of reach ofthe Europeans, in about 1625, and
tried to set up a new political system. They likened the state to a perforated pot,
symbolized by the king. For the perforated pot to hold water, each citizen had
to place a finger over a hole, thus sinking his identity in an absolute state. It was
the dogged development of this idea of a strong centralized state with a king
demanding total loyalty that marked out Dahomey from other states. It was thus
that it was able gradually to bring all the neighbouring states under its control.
The predominant role attributed to the Atlantic slave trade in the history of
Dahomey is currently being revised by recent research. W. Peukert directly
challenges two central concepts: that the history of Dahomey was determined
by the slave trade and that its economy was archaic. The historianis new
hypotheses call for a revision of old ideas and further research.
The Yoruba kingdoms

The history of the Yoruba communities of south-western Nigeria can be sum¬
marized around three points. The first concerns the traditions which make Ile-
Ife the centre of origin ofthe Yoruba kingdoms, an assertion which needs to be
reconciled with the archaeological evidence. The second is the rise of the
military and political power of Oyo, thought by some to have coexisted with
the spiritual authority ofthe oni of Ife over the whole of Yoruba country. The
third point is the nature of the political, cultural and social development in
other Yoruba communities and kingdoms.
The primacy of Ife in Yoruba history is based on many factors. According to
tradition, its founder, Oduduwa, was supposed to have come from heaven or
Mecca and his sons or grandsons to have founded all the other Yoruba states.
Recent interpretations of these traditions suggest that Oduduwa, or a group of
migrants, came to the area about 1,000 yean ago but that the land was already
occupied, possibly by the Igbo of Ife tradition.
The famous Ife bronzes have played a part in confirming the traditions ofthe
relationship between Ife or Yorubaland and the Edo kingdom of Benin. But
the bronzes also relate Ife to Nupe and Benin and regions around the Niger as
clear similarities have been found between the large bronzes found in Nupe
and those of Ife. Likewise, it has become clear that the 'Mecca' of oral
traditions did not refer to any place in the Middle East, Egypt or Meroe but to
regions just across the River Niger to the north of present-day Yoruba
territory. In addition, the art of Ife has been compared to the Nok terracotta art
in central Nigeria, despite the time gap between the two cultures.
Qyo not only became pre-eminent among the Yoruba kingdoms, but it also
developed special features. Some of these derived from its location close to
Nupe and Borgu. For example, Oyo relied more heavily on officials of slave
origin in its military and social organization than did other Yoruba states. In the
armed forces, Oyo's supremacy was probably the result of its use of cavalry and
archen, derived from its early contact with the trans-Saharan trade and the
Fon and Yoruba: the Niger Delta and the Cameroon 225
15.2 Fifteenth- or sixteenth-century commemorative headfrom Benin, Nigeria, cast in

bronze and inlaid with iron. Height 22 cm (National Museum of African Art, Eliot Elisofon
Archives, Smithsonian Institution, Washington DC, photo: B. Fleischer)
northern states. These external contacts as well as its location in the savannah
enabled Oyo to make innovations in the Yoruba concepts it shared with the
other kingdoms.
The foundation of Oyo is related to Ife and Benin through its legendary
founder, Oranyan (Oranmiyan), who is stated to have reigned at both Ife and
Benin before moving to Qyo. But other states already existed in this area, and
Oyo made several of them vassal states, including Owu in the south and Ede in
the south-east. The expansion of Oyo was finally stopped by the Ijesha, as the
Oyo cavalry could not operate in wooded regions. The Ijebu and the hill
country ofthe Ekiti also escaped direct Oyo control. The kingdom of Benin to
the east proved another barrier to Qyo expansion. Qyo opened up a trade
route to the coast through the territory of the Egba and Egbado and it was
through this route that Qyo power expanded to Dahomey.
Oyo power developed out of advenity. In the fifteenth century its rulers had
been driven out of Old Oyo (Oyo He or Katunga) and had taken refuge in
Kusu among the Borgu and later at Igboho. Undaunted, Oyo had reorganized
its army and adopted a new policy of militarism. By the beginning of the
sixteenth century Oyo had reconquered its territory and pushed the Nupe back.
Oyo remained outside the area of direct European influence until the
nineteenth century, developing its major institutions and starting its expansion
independently. Its adventure into Dahomey may have been connected with
participation in the coastal trade but other oral traditions say that Oyo kept out
ofthe slave trade and avoided any contact with the Europeans.
The areas of Yoruba country beyond Qyo expansion, to the east and north,
consisted of small states in the Ekiti area and othen, such as the Igala state, that
were more oriented towards the Niger-Benue valley. Other states, such as
Owo and Ijebu, appear to have had much more to do with the Edo kingdom
of Benin than with Oyo. Works of art recovered from excavations at Owo
show that by the fifteenth century the sculptural forms of this region were
already related to both Ife and Benin styles.
The Niger delta

The history of the Niger delta is finked with the history of parts of the coastal
region east to Cameroon and west to the lagoons of Lagos and beyond. Until
the nineteenth century, trade routes crossed the region from east to west and
north to south. Cloth made in the Ijebu area of Yorubaland was sold in the
western delta and appean to have been resold as far east as Nembe in the
eastern delta. The Itsekiri (western delta) were related to the Ijebu, as shown by
the similarity of language, but they had derived some of their cultural values
from the Ijo and traded in pottery, salt and cloth with the states of the eastern
delta, especially Nembe.
In the western Niger delta, Ode Itsekiri, the capital ofthe kingdom, was the
focus of political power. When the Portuguese came to this area, the Itsekiri
became their major delta contacts. In the eighteenth century some Itsekiri left
the capital to establish setdements in the estuary of the Benin river at points of
vantage for the trans-Adantic trade. It was mainly from these new centres that
the Itsekiri served as the agents of the Benin kingdom and as middlemen on
their own account for the export ofthe produce ofthe hinterland peoples.
The Ijo of the western Niger delta were mostly organized in loose stateless
communities and tended to participate in the oveneas trade by preying on it as
pirates. The bronzes found among these Ijo groups, and also among the Mein
and Kabowei, may be indications of affluence derived from participation in
internal and external trade, and also from contact with Benin and other hinter¬
land centres.
The Ijo of the central delta formed the heartland of the group. Oral tradi¬
tions suggest migrations from this area to the eastern and western delta and on
to the delta peripheries. The present communities in the eastern delta are
estimated to have settled there a thousand yean ago. Radio-carbon dating of
finds from sites at Ke and Saikiripogu (Ewoama) confirms that people were
living there before the ninth century.
Excavations in the eastern Niger delta have opened new vistas for Niger
delta history. They show that the earliest setders knew how to exploit the shell¬
fish of the delta and also kept some animals. The abundance of pottery at such
sites as Onyoma suggests a complex economy, based partly on agriculture,
partly on trade. The excavations have also revealed evidence of iron-working.
This suggests contacts with the hinterland for raw materials and finished goods.
Art in the Niger delta consists principally of sculptures in wood depicting
water spirits, anceston or masks for dancing. Excavations have recovered a
small number of masks in terracotta from Ke and human figures from Onyoma,
which are unique among Nigerian terracotta but reminiscent of Nok and Ife
terracotta.
The Atlantic slave trade and, prior to it, internal long-distance trade, played
a large part in the formation of the states of the eastern Niger delta - Bonny,
Elem Kalahari (New Calabar), Okrika and Nembe. From the oral traditions it
would appear that their founders came from the fresh-water central delta and
adapted to life in the salt-water eastern delta. Social and political adaptations
were made and kingship institutions were set up from about the thirteenth
century. The Adantic slave trade accelerated the pace of change and the form
of state established by the seventeenth century has been named the city-state or
trading state. The slave trade provided wealth as a base of power for the
amanyamabo (king) and ruling élite.
Igboland
A number of Stone Age sites in the Igbo heartland suggest a long history of
human habitation. A rock shelter at Afikpo produced stone tools and pottery
5,000 yean old, and sites with similar evidence have been found in the Nsukka
area. It seems clear that people farmed there at least 3,000 yean ago, with yams
as their main crop. Other crops of local origin include oil palms, okro, egusi and
15.3 Yoruba female statuette, dedicated to the worship ofthe orisha of creativity, Obatala.
She is usually clothed with a white cloth ¡ike the priests and devotees of that god, an iron
bracelet and, here, a necklace of white pearls with a pendant. The bowl that she is holding is
intended to receive an offering ofthe white blood ofthe snail, a symbol ofpeace and calm. It is
thefemale element ofthe orisha, the male element being represented by a statuette holding a
fan and a fly-whisk. This bowl-bearer should not be confused with the olumeye, or vesselfor
kola nuts, represented by a much larger kneeling maternalfigure holding a bowl which often
has a lid. Height: 49 cm (H. Dubois, Brussels)
Fon and Yomba: The Niger delta and Cameroon 229
some varieties of kola nut. Cassava, rice, bananas, plantain and other crops were
imported from the Americas.
Iron-working was established early and the bronze art of this area has
become world-famous through the excavations at Igbo-Ukwu. The bronzes
from this area derive from a different tradition from those of Ife and Benin but
equal them in beauty and quality. They are associated with the Nri divine
kingship and ritual centre. The Nri priests exercised authority over wide areas
of Igboland with power to install ozo and eze title-holden and wash away
abomination. It could have been the income brought back by travelling Nri
priests which provided the wealth that sustained this bronze art.
The Nri priesthood performed a vital role in Igboland where communities
were organized on the basis of the title system. But some Igbo groups west of
the Niger and on its east bank adopted the kingship institutions of the
communities with which they came into contact. Thus the Aboh, Onitsha and
Oguta kingdoms apparently derived their obi from the oba (king) of Benin.
These states were created by migrants from areas under Benin influence in the
sixteenth and seventeenth centuries.
Other states on the River Niger, such as Osomari, claim ancestors from the
Igala kingdom of Idah to the north. But Igala influence was probably more
pronounced among the northern Igbo in the Anambra valley and in the
Nsukka region. The Igbo states along the Niger became the first participants in
the overseas slave trade and later in the palm-oil trade ofthe nineteenth century
in collaboration with the Niger delta states.
The Igbo political and social system had mechanisms for -wider control than
the village or town. One was religious sanction by an oracle. Oracles have
operated in Igboland at different times and places, but the best known were
Agbala of Akwa and Ibini Okpube (Longjuju) of Arochukwu. Thus, the Awka
blacksmiths, who worked at markets and settled all over Igboland and the
Niger delta, spread the influence of the Agbala oracle. But the Aro Chukwu
oracle became even more successful thanks to the Aro traden, slavedealen
who spread word of him far and wide. The Aro trade network developed in
the port of Calabar and later also came to deal with the delta states. The Aro
traden used the utterances of the oracle to procure slaves but were also willing
to use violence and enlisted the help of the warrion of Abam, Edda, Ohafia,
Abiriba and othen -with whom they shared their loot.
The areas of Igboland without strong centralized government probably
suffered most from the effects of slave-raiding. In the eighteenth century, the
eastern Niger delta ports were the main centres for the export of slaves from
West Africa, and most of them came from the Igbo hinterland.
The Cross river valley and Cameroon

The peoples in this region show affinities of language and also of historical
origins. Most of the languages belong to the family of Bantu languages. The
largest ethnic group in the Cross river valley, the Ibibio, has long been
established in the region. In the northern parts of the valley, the Ogoja area is
occupied by a wide variety of peoples with oral traditions of migration from
the Benue valley, further north, or from Cameroon. Some communities in the
Ibibio group or closely related to them, such as the Andoni and the Ibeno, also
claim that they came from Cameroon. Likewise, some communities in Came¬
roon, such as the Isangele, are of Ibibio origin.
The communities in this region were largely organized in non-centralized soci¬
eties of great complexity. Age organizations as well as secret societies such as the
Ekpo and the Mgbe (leopard society) provided effective social and political control.
The Efik are closely related to the Ibibio, their proximate home having been
Uruan Ibibio on the west bank of the Cross river. Some oral traditions state
that they had previously lived at Ibom, near Arochukwu in Igboland. Their
final settlement at Ikot Etunko (Creek Town), Obutong (Old Town) and
Atakpa (Duke Town) by the beginning of the seventeenth century made these
places the major centres for the oveneas slave trade in the region. The state the
Efik founded on the lower Cross river now known as Calabar exported
slaves from the Igbo hinterland from the centre at Arochukwu.
The slave trade was pardy responsible for the reshaping of the Ibibio Ekpo
and the Ekoi Mgbe secret societies into the Ekpe at Calabar. The Ekpe became a
class-structured society uniting the free aristocratic elements and keeping in
control the slaves and the poor. It also enforced political and social rules,
collected debts and kept order. Because of the common influences of the slave
trade and trans-Atlantic contacts, Calabar developed lineage-based organizations
similar to the Houses ofthe eastern delta states.
The most important groups on the Cameroon coast were the north-west
Bantu groups of the Kpe-Mboko, the Duala, the Limba and the Tanga-Yasa.
There were numerous groups of fisherfolk, farmers and hunten. Most were
organized in small village units but by the eighteenth century the Bubi, the
Duala and the Isubu had organized larger political units. They were either con¬
nected with the slave trade or took advantage of it. The Jengu secret society, a
major focus of social and political control, became the most prestigious one in
the coastal region of Cameroon.
Conclusion
The major external influence in the history of this region of coastal swamp and
rainforest from the fifteenth to the eighteenth century was the slave trade. It
had an enormous impact on the peoples of the area but it is hard to assess its
exact role in relation to internal factors for change that were at work long
before the advent ofthat trade.
In many oral traditions, the slave trade features as an activity that enriched
people and increased the population. This was the case for the coastal com¬
munities which took part as middlemen, not themselves engaging in hunting
for slaves but buying them to re-sell to the slave traders. The Niger delta states
and the Efik state fall into this category.
The Fon kingdom of Dahomey exemplifies a second type of participation in

the slave trade by active procurement of slaves for the slave traders. In Igbo¬
land the Aro and their mercenary allies played a similar role as slave hunters.
The third category of community affected by the trade were the victims.
These were the communities from whom most slaves were taken for sale on
the coast and included parts of Yorubaland, Igboland and Ibibioland. Raids,
kidnapping, wars and the general cheapening of human life precipitated the
collapse of the economic and social systems of these communities. Villages
were destroyed or dispened, farms were abandoned and people lived in terror.
It is sometimes argued that the slave trade brought these tropical regions of
Africa out of isolation and integrated them into the international economy. In
addition, the introduction of food crops such as maize, cassava and rice could
be seen as redressing the demographic balance by encouraging population
growth. These are speculative hypotheses. For the African peoples of the
region, the slave-trade era is seen as a bad dream best cast into the dark recesses
of their memory.
The Hausa states
The region discussed in this chapter is readily associated with the idea of
wealth. This wealth arises from the complementarity between neighbouring
regions and Hausaland as a producer and a consumer and from the integrated
aspect of the West African economy, with the Wangara, Hausa and Kanuri in
the savannah, and the Fante, Bini, Ijaw and Arochukwu in the forest.
The documentation available on the region's development between 1500
and 1800 is of very uneven quality. Source materials on the Hausa states are
extremely varied. They include eye-witness reports, surveys and reference
works, which are all easily obtainable publications, theses mainly written by
researchen from the region - and the proceedings of seminan covering history
and archaeology. Lastly, the point of view ofthe peoples directly concerned is
set out in a number of documents published in English or Hausa, such as the
Kano Chronicle.
The uneven pattern of research means that the various political entities
cannot all be treated with the same degree of thoroughness. The relatively
peripheral position of Borgu, Kwararafa and Nupe during the period under
consideration is also pardy responsible for this situation.
The Hausa states
Political development
Whenever post-1500 central Sudan is being considered, it has become cus¬
tomary to point to the specific influence of western Sudan - Mali and Songhay
- and Känem-Borno. The century opened with the askiya Muhammad's cam¬
paign in Azbin. There was also a feeling of nostalgia for the sarakunan noma
(masters of the crops) emperon, the mai (king) of Känem and the mansa
(emperor) of Mali. Since western and eastern Sudan were better known, little
attention had been paid until that time to the changes taking place in central
Sudan. However, it was soon to be integrated into the commercial and
ideological system uniting the societies of western Sudan with the Muslim
world. Its political development came to be viewed in terms of its relationships
232
The Hausa states 233
with the neighbouring states, the conflicts between Kano and Katsina and
interludes involving Kebbi, Zamfara and Gobir.
The chronological framework adopted here distinguishes between three
periods:
(1) 1500-1620, when the states gained in strength but there were bitter
conflicts between Kano and Katsina, and Kebbi remained the dominant
power in the western part of Hausaland.
(2) 1620 when Kano was on its way to becoming a caliphate, at a time
when Zamfara and Gobir were consolidating their positions following the
decline of Kebbi and Kwararafa was further increasing its capacity for
attack.
(3) 1730 a period marked by the collapse of Zamfara and Gobir's rise to
the peak of its power.
Relations with the neighbouring regions

Although Hausaland's relations with its neighbours are being studied in in¬
creasing depth, there still remain some areas of controveny.
With Songhay, undue importance has been attached to the authority of
askiya Muhammad. The main outcome of his campaign in Azbin was the
emergence of Kebbi as an independent state in 1516. Songhay was subse¬
quendy to attempt to regain control of Kebbi. Askiya Muhammad II Benkan
Kiriai (1531 led an expedition in about 1533 but was defeated. Another
expedition was organized in 1552 in the reign of askiya Däwüd, the conflict
being brought to an end by the conclusion of a peace treaty in 1553.
Azbin provided the pretext for the clash between Kebbi and Borno. As
Kebbi continued its repeated forays, Azbin sought the help of Borno in 1561.
After initially winning the day at Surame, the 100,000 man Borno army was
defeated at Nguru by the Kanta. Then, at the end of the century, a dynastic
crisis flared up in Azbin. Yüsuf, after being deposed by Muhammad ben al-
Mubärak (c. 1601) appealed to Kanta Däwüd for support. The latter, who
wished to maintain his influence, helped him defeat his rival, who was backed
by Borno. This success checked the mai's designs on Azbin, which was at its
peak under the reign of Muhammad al-Mubärak (c. 1653 In 1674, taking
advantage of the continual conflicts between Gobir and the eastern part of
Hausaland and between Zamfara and Kebbi, Muhammad dispatched an
expedition, led by his son Agabba, who conquered Adar and thereby hastened
Kebbi's decline. From then onwards, it was left to Zamfara and Gobir to
temper the influence of Azbin in the Kwanni regions. In 1675, the conflict
between Azbin and Zamfara gave rise to massacres and batdes. Gobir, which
was pillaging the same area, was attacked and defeated by Azbin in about 1689.
Finally, Agabba, who had become sultan, marched on Surame and killed Kanta
Ahmadu. In 1722, the Kebbi court -withdrew towards the western areas.
However, Azbin was entering a period of dynastic crises and disasten, and this
left Zamfara and Gobir to occupy the leading role.
Before 1561, Borno was in a position of strength. Sarkin Kano 'Abdullahi

(1499 was attacked by the mai and adopted a subservient attitude which
prompted the aggressor to depart. However, his successor, Muhammad Kisoki
(1509-65), made for the city of Nguru in Borno, where he gave orden that
only horses and clothing were to be seized. Caught by surprise, the mai took
the initiative the following year and attacked Kano, but was again forced to
withdraw. The humiliation inflicted by Kisoki on the mai was a sign of the
might of Kano at a time when Borno was suffering from internal strife and
several yean of famine.
It was not long before Kano was subjected to repeated attacks from
Kwararafa: between 1582 and 1618, its population was forced to seek refuge at
Daura. The sovereign of Kano was again driven out in 1653, and its inhabitants
once more sought refuge in Daura in 1671. According to C. A. Palmer, the
peace treaty signed between Kano and Katsina (c. 1649-51) was inspired by fear
of Kwararafa, which was eventually to be defeated by Borno in 1680.
These conflicts point to an unstable equilibrium between powerful neighboun.
The struggle for hegemony

The GwanjaKanoBorno route had been opened up between 1438 and 1452.
Moreover, in the fifteenth century, Agades had replaced Takedda as a caravan
centre, thereby making Katsina a terminus of the trans-Saharan caravan route,
and a trading centre for all Hausaland. The military conflicts between Kano and
Katsina may have been due to competition between the two states for control
ofthe trans-Saharan terminus, but nothing is known ofthe nature and scale of
these wars.
The first conflict broke out during the reign of Rumfa (c. 1463 an
extremely wealthy ruler who was the first to be escorted by fully accoutred
hones in the war against Katsina. That war lasted eleven years, with no decisive
outcome.
The second conflict took place at the time of Abu Bakr Kado (c.156573),
when Ibrahim Badankarî was reigning in Katsina. The Katsinawa advanced to
the very gates of Kano, were victorious and returned home.
To avenge this defeat, Muhammad Shashîri (c. 157382) organized an expedi¬
tion against Katsina, where Muhammad Wari (c. 1575-87) was ruling. The
battle took place at Kankiya, Kano -was defeated and its sovereign was deposed.
The next conflict broke out during the reign of Muhammad Zaki (c. 1582
1618) of Kano. In Katsina Muhammad Wari was succeeded by Sulaymän
(c. 1587-1600) and 'Uthman Nayinawa (c. 1600-18). There is some confusion
as to what happened next, and there may have been a change of dynasty.
Katsina was so powerful that Kano was afraid it might be attacked. However,
Kwararafa invaded Kano, defeating it and sapping its strength. Shortly after¬
wards, Muhammad Zaki of Kano consulted the (ulamä) and obtained a talisman
at a very high price. Suitably protected, Kano then attacked Katsina's military
encampments and emerged victorious.
As soon as he assumed power in Kano, Muhammad Nazaki (c. 1618-23)
made peace proposals which were rebuffed by Katsina, which was again
defeated. In the reign of Kutumbi (c. 1623-48) Kano kept up the pressure on
Katsina and there was a series of incidents. In one expedition, Kutumbi's army
was routed and he himself killed. One of his successon, Shekarau (c. 1649-51),
managed to make peace after negotiations conducted by the iulamä). Katsina's
power was further consolidated without the peace treaty with Kano being
broken.
There was accordingly a long political struggle for control of eastern Hausa-
land.
Interludes
Until the end of the sixteenth century, Kebbi feared none of its enemies, not
even Morocco. Among the specific faetón instrumental in consolidating this
state, one author has suggested the mixed origins ofthe population, the polariza¬
tion of protests against Songhay domination and the determination of its
military leaders to maintain the country's independence. After clashes with
Zamfara, Kebbi defeated Borno in 1561 and rose to occupy the leading place in
the political life of Hausaland.
The capital of Gobir had been transferred from Azbin to Hausaland, at
Birnin Lalle, in the centre ofthe densely populated and well-watered region of
Gulbin Tarka in about 1450. The rulers remained there until about 1600, when
the Tuareg launched an attack. The Gobirawa were forced to take up their
wanderings again and set out southward. They then struck out towards the
north-west and founded the new Gwaran Rama in about 1685 From this
city, Uban Doro launched attacks against Kebbi and even against Yorubaland
and Gurma; his successor, Soba, attacked Adar, Kebbi and Maradi. However,
he established friendly relations with Zamfara, which opened up Alkalawa to
the farmen and traden of Gobir. The old military aristocracy was gradually
eclipsed by a new aristocracy based on money. The decline of Gobir began just
as Zamfara was clearly gaining strength.
In the mid-seventeenth century Zamfara was governed by strong sovereigns
who drew their support from Islam. Aliyu, the fint Muslim ruler of Zamfara,
established good relations with Katsina and built mosques there. Zamfara
stopped making its sporadic raids and instead concentrated its attacks on Kebbi's
major towns. In 1674, Sulaymän put the army of Kebbi to flight. In the same
year, Kebbi lost Adar, which was wrested from it by prince Agabba. It was a
series of separate defeats rather than a concerted action by Azbin, Gobir and
Zamfara. After this victory, Zamfara became the leading power in the area, as
was revealed by its victory over the army of Kano during the reign of
Muhammad Sharif (c. 1703-31).
Zamfara had thus recovered from the defeat it suffered at the hands of
Azbin, but the military strength of Gobir was also growing: In the reign of
Kumbari (c. 1731 there was a fierce war between Kano and Gobir, whose
sarkin's name was Soba. Defeats and acts of revenge followed one another year
in year out, and the massacres only came to an end at the death of Kabi. Gobir
was soon to feel the weight of various restrictions imposed upon it by the sarkin
Zamfara, who was worried by his restive neighbour. At the outset, Gobir was
content with harassing its enemy, but then, taking advantage of a dynastic crisis,
it destroyed Birnin Zamfara in about 1762.
This development had repercussions on the political situation in the peri¬
pheral areas. In the east, the sultanate of Damagaram was founded towards the
beginning of the eighteenth century, while the Tsotsebaki states were consoli¬
dating their position before subsequently splitting up. This area, which marked
the transition between Borno and Hausaland, was extremely sensitive to
political and cultural movements.
In the north-west, the history of Adar is better known, but it is not clear
what its links were with Kurfay, which is sometimes associated with Adar,
sometimes with Borno.
Zarma, the easternmost part of the Songhay empire, first established links
with Hausaland when Gobir was part of Azbin. At that time Kebbi, Zamfara
and Gobir were all fighting over the area. Before it was eliminated in 1722,
Kebbi played a decisive political role in Zarmatarey, where its name was
associated with cavalry protected by lifidi (quilted armour) which spread terror.
Finally, in the west, on the gurma (right) bank of the river Niger, certain
Gulmanceba dynasties claimed to originate from central Sudan, Borno or
Hausaland a claim seemingly substantiated by archaeological excavations, at
least for the areas along the hausa or left bank.
The situation, however, stabilized considerably after Zamfara's defeat in
1762. Katsina, despite a domestic crisis, was able to defeat Gobir, while in
Kano, Babba Zaki (c. 1768 terrorized his collaborators.
Political and administrative organization

Thus, despite the military conflicts, this host of states remained in existence.
After their severe defeats, the sovereigns of Kebbi and Zamfara retreated to a
smaller territory to safeguard their power. The explanation for this develop¬
ment may lie in the process whereby the sarauta (kingship) system was intro¬
duced into and developed in central Sudan.
The sarki
At the head ofthe state, which was first and foremost a kasa (territory), there
was the sarki, whose ancestor had seized political power. In Kano, Katsina and
Zamfara, it had been wrested from the hands of a high priest, while in Kebbi it
was a magaji (warrior) who rose to the rank oí sarki.
The appointment of the successor to the throne was the responsibility of an
electoral college. In Katsina, this consisted of four members. It is difficult to say
whether such a college existed in Kebbi at this time, but it certainly appeared
there at a later stage. In Zamfara, Gobir and Kano, it bore the name tara (the
nine), followed by the name ofthe state. The tides and functions ofthe various
electoral colleges differ, but one incumbent can be identified as holding an
ancient office, the basase, in Zamfara, and there were, besides, governors of
cities and important regions and high-ranking public servants. They also
included representatives of ousted and reigning dynasties.
The successor had to be chosen unanimously by the grand electors. Once he
had been appointed, the enthronement and installation ritual took place. The
electoral college also formed the council ofthe sarki, and it was not uncommon
for the council to clash with the sarki or for the latter to debar one of the
council members.
The government
The sarki exercised his authority through three groups of officials: members of
the dynasty, public servants, governors ofthe towns and regions.
The memben ofthe dynasty were assigned important duties. The sovereign's
sister played a leading political role, although she did so through the traditional
form of worship. The history of Kano is studded with the names of famous
princesses. If the madaki (the queen mother) Auwa had not intervened,
Abdullahi (1499 would most certainly have been driven out by a
rebellion. His son, Kisoki, governed the city with the support of the same
madaki Auwa, his grandmother, Iya Lamis, his mother, and Gulli, Auwa's
brother.
In what might be called the central government, there were several categories
of officials:
(1) The court dignitaries managed the affairs of the palace and the city. They
were in a position to act as middlemen between the sarki and the regional
governments.
(2) The representatives of the guilds were appointed from among the skilled
craftworken, such as blacksmiths, weavers, dyers, tanners, masons, butchen
and hunten. They were responsible for relations with the different trades
and occupations and, in particular, for collecting the state dues.
(3) The autochthonous groups had their own representatives. The village of
Sarkin Naya and the region of Sarkin Mazum, for example, kept their tides
even after Gobir had established its capital at Hisatau.
(4) The numerous immigrants were also allowed to have their own repre¬
sentatives. In Gobir, the sarkin Azbin handled relations with the Tuareg
living in the territory, and the sarkin Fullani did likewise for the Fulbe.
(5) The Islamic community was present everywhere with its mallamai (scholars).
In terms of regional government or, more specifically, territorial control, a

distinction has to be made between several categories:
(1) The authority of the governors of certain cities was independent of the
dynasty. Their relations with the sovereign could be difficult, as they were
eventually reduced to the status of vassals.
(2) The governors of other cities or regions were either nobles whose ancestors
had become allies of the dynasty while preserving their own authority, or
public servants. Zamfara provides an illustration of the fint case with the
danan, who owed his tide to the town in which he resided, an important
commercial centre from where he kept watch over the roads leading south
and west of Kebbi. In Kebbi, the innamme kept watch over the western
frontier. In Katsina, the marisa (destroyer) resided in Gwiwa from where he
supervised the eastern frontier, while the gatari (axe) of Ruma kept an eye
on the north-western frontier.
(3) The other governors became mere cogs in the governmental machinery.
The immigrants retained a local hierarchy.
(4) Finally, certain titles were the outcome of historical developments. In
Kebbi, the kokani was responsible for relations with the population after the
conquest of Kwanni; the office of saburu was created in about 1650 to
provide security on the roads that led to Kwanni and Azbin.
All these dignitaries, nobles, commoners or slaves performed duties of a non-

military nature except that, when the occasion demanded, they became valiant
warriors with their own armed groups.
Military organization
The foundation of Kebbi signified the increasingly important role played by
military leaden in the affairs of the state. The kanta's fint companions were
from the country's leading families but, subsequently, the elements that had
participated in the struggle for independence were granted two representatives,
one of whom, the kunduda, was to be Kebbi's military leader.
Given the need to seek protection against its neighbours and to provide for
security on the borders and the roads, the number of military leaders increased,
and a ranking order was instituted. The highest-ranking title varied from state
to state: it was kunduda in Kebbi, kaura in Katsina, ubandawaki in Zamfara and
Gobir, while in Kano it appears to have been the galadima.
As weapons improved, so strategies and tactics changed. In the event of
mobilization, the services of different trades, such as hunters and blacksmiths,
were called upon. The army consisted of two main corps the infantry,
subdivided into archers and lancers, and the cavalry. Kebbi also had a river
flotilla.
The hones mainly originated from Azbin and Borno and were treated with
special care. The sovereign imported them and set up stables: the kanta of
Kebbi, for example, had stables in three places. The role played by the horse
can be explained by such innovations as the acquisition of sulke (coats of mail)
and the manufacture of lifidi (quilted armour). The sarkin lifidi (general in the
heavy cavalry division) and the lifidi (commander-in-chief of the heavy cavalry
division) were among the highest-ranking officen.
Muskets had been introduced to Kano by a Borno prince in the reign of
Dauda (1421-38), but it was not until three centuries later that they were
imported from Nupe. Babba Zaki (c. 1768-76) was the first sovereign to set up
a corps of fusiliers to serve as his personal bodyguard. Songhay, which had been
able to gauge their deadly effectiveness at its own cost, did not bother to
retrieve those abandoned by the Moroccan soldiers half a century later
(between 1591 and 1640). Hausaland seems to have deliberately neglected a
weapon that it could have procured for itself with its wealth.
The war tactics used consisted of surprise attacks, ambushes, pitched batdes
and sieges. The stepped-up construction of fortifications around the cities was
justified by the frequent sieges and conflagrations. In this region, in which the
pace of economic development had increased, improvements in the art of war
led to the growth of pillaging. It is not easy to distinguish between wan waged
for purposes of conquest or consolidation, the repression of rebellions, wars of
intimidation and simple raids. Gobir was continually obliged to fight for its
survival as a state. The political, administrative and military aristocracy indulged
in pillaging, as the Kano Chronicle clearly shows: it grew rich and presented gifts
to the sovereigns and scholars. In Kano, in the space of two centuries
between 1573 and 1768 it became so powerful that it urged the state to wage
war, participated in plots and gave the sovereign cause for concern. The total
number of dignitaries varied from one state to another. Gobir had 22 (13
nobles and 9 commoners) and Katsina 46 (16 nobles and 30 commoners).
Resources
The Hausa state developed an ingenious system for controlling the territory and
levying taxes. Three major sources may be singled out.
Taxes and duties
Taxes and duties were the most regular sources. They included the following:
kudin kasa (land tax), paid by farmers. The unit taken into account seems to
have been the gandum gida (family field).
kudin sana1 a (professional tax), paid by craftsmen and traden.
kudin hito (customs duty), payable on certain products entering the territory. In
addition, salt shipped to Zarmaterey and Arewa was subject to tax.
jangali (livestock tax), paid by the stock-breeden, particularly the Fulbe. In
Kano, it was levied for the first time by Kutumbi (c. 1623-48). In Kebbi,
the nomadic Fulbe paid kudin haki (grazing duty) while a dignitary, called
the nono, collected the milk and butter intended for the sovereign.
Gifts
Governors, dignitaries and other figures sent the sovereign a gaisuwa (gift made
to a superior). This was a political act whereby the interested party paid tribute
to the sarki, hoping in return to enjoy his favoun. Those who were appointed
to high office also presented gifts to the sarki. One common method of obtaining
the wherewithal to send a gaisuwa was, in fact, pillaging.
Spoils
Pillaging supplied slaves, hones, catde and a variety of goods. The Kano Chronicle
stresses the value attached to hones between 1582 and 1623. The slaves were
sold or divided among the great royal properties.
Other resources
There were coundess measures whereby the sarki could fill the coffen of the
state. For example, when the sovereign pardoned an offender, the latter had to
pay kudin laifi (forgiveness tax). Thanks to the Kano Chronicle, it is possible to
follow the process whereby the taxes and other duties were created for the
benefit ofthe state. Sharif (c. 1703-31) instituted seven taxes that were regarded
as oppressive, including a duty payable by girls on marriage. His successor,
Kumbari (c. 1731-43), made the scholan pay a duty, as a result of which the
Arabs left and went to Katsina.
Overview
The development of the sarauta system evolved in such a way that commonen
and slaves could occupy the highest offices if they were considered highly
trustworthy. Thus it was that royal slaves, particularly eunuchs, came to
constitute an essential cog in the state machinery throughout the region.
This resulted in opposition between the mai sarauta (the ruler) and the
talakawa (the ruled). To a lesser degree, the governors ofthe cities and regions
were potential opponents. The frequent uprisings fomented by the governon of
Gaya and Dutse illustrate the recurring friction between sovereign and vassals in
Kano's history.
The political, administrative and military aristocracy represented a uniform
group which grew rich by various methods of exploitation. It adopted a way of
life commensurate with its resources, and decked itself out splendidly to show
its prestige, while at the same time becoming difficult to keep under control
because ofthe bribery and corruption in which it indulged.
Economic relations
Agriculture and stock-raising

From 1500 to 1800, Hausaland remained primarily a region of farmen. The
manoma (peasant) made judicious use of the agricultural potential through a
variety of techniques, including fertilizers and crop rotation and association.
Their tools were as numerous as elsewhere in Africa, with a wide range of hoes
adapted to the nature of the soil. The bulk of the labour force came from the
gida (extended family) and the gayya (mutual aid system).
Besides millet, sorghum, rice, maize, peanuts and beans, farmen grew
cotton, indigo, henna, tobacco and onions. Shea, tamarind and nere were also
cultivated and honey was collected. They also engaged in fishing and hunting.
Jibda (musk) was removed from the civet-cat and used for making perfume.
There was a substantial amount of stock-raising for domestic purposes. Goats
were sacrificed by having their throats cut for certain ceremonies, while donkeys
were used as a means of transport, particularly by traden. The Fulbe, with their
sheep and cattle, setded in several areas of Katsina, Kebbi and Zamfara.
Agriculture and stock-raising became closely associated and pockets of eco¬
nomic fife developed based on a combination of the two, such as in the region
of Ingawa in Katsina.
Crafts
The divenity and high technical standards attained by craftworken were already
remarkable at the time of Leo Africanus: people worked with iron, wood and
leather, and also engaged in basket-making and pottery, while there was
spectacular growth in weaving and dyeing. In addition, big demand stimulated
high-quality crafts.
A trend towards group specialization can also be observed. In Kebbi, weaving
and dyeing were in the hands of the Kebbawa; Zamfara attracted weaven and
dyers from Kano. In both Kebbi and Zamfara, the Zoromawa specialized in
silver jewellery and pottery; in Kano, pottery was made by the Bambadawa.
A variety of articles was manufactured. Leather, sandals, harnesses and saddles
were exported. Jewellery was purchased by the wealthy. Clothes, such as tunics
and cloths, were famed for their quality. Hausaland was also among the areas
producing excellent-quality woven and dyed goods.
Trade
The stereotype of the bahaushe (trader) became widely recognized and it is

impossible to overstress his integration into the trading networks of western
Africa and the international trader class in the savannah.
The area of influence of the markets differed considerably. Some were

important locally and retained their social and economic character: goods were
traded, but social life on market day was considerably enlivened. At a higher
level, the regional market was a centre to which local products were brought
and where imported articles in daily use were distributed. In Zamfara itself, the
northern and north-eastern markets supplied cotton, indigo, tobacco, onions
and cattle, while the southern ones received abundant supplies of cereals.
Upper Kebbi sent yarn, fabrics and slaves to Lower Kebbi, which supplied nets,
harpoons, hides and abara (large dugout canoes).
Foreign trade remained in the hands of the Hausa, although some Azbinawa,
Arabs, Kanuri and Wangara also took part. Caravans linked Kawar to Gao,
Azbin to Hausaland and Kano to Gwanja. Each sovereign took security pre¬
cautions to protect the caravans in his own territory as they paid taxes to the
states through which they passed.
There were several international markets within the region and on its
periphery. In the north, Agades and Bilma served as links with northern Africa;
in the centre, Katsina and Kano were staging-posts in north-south and east-
west relations; in the south, Zaria, Birnin Gwari and Birnin Yawuri enabled the
networks to be extended towards Yoruba, Nupe, Borgu and Gwanja. A
considerable volume of merchandise circulated in Hausaland.

The Sahara and Azbin sent Arab and European goods to these markets,
including mirrors and paper, but particularly hones, camels, dates, henna, salt,
swords and other articles. A certain proportion of the salt and swords was in
transit to the south. In return, Hausaland supplied them with slaves, clothes,
fabrics, millet, hides, iron, gold dust and kola nuts from Gwanja.
Borno had horses, natron and salt to offer and in exchange received metal
articles, gold dust and kola nuts. Hausaland exported salt, swords, condiments,
hides, clothes and fabrics, slaves and hones to Gwanja, Borgu, Nupe and
Yoruba and received in return various European goods, local iron, antimony,
slaves and eunuchs, muskets from Nupe (for Kano) and kola nuts from Gwanja
for domestic consumption.
Social relations
Three basic groups in society can be distinguished.
Producers of material goods

Farmen formed the largest category. The more intensive use of the land and
the improvements in farming practices were to bring about far-reaching
changes. Until that time, the main source of labour had been the extended
family. It is clear that some population groups had been reduced to a state of
bondage: the Mazumawa, for example, lost their independence as soon as Gobir
reached the Birnin Lalle area. The Maguzawa of Fanfui were subordinate as a
group and thus tributaries. The last source of labour was slave labour.
Among craftworken there was intense specialization, and many slaves engaged
in craft activities at their master's behest and for his benefit, at least to begin with.
A large number of herden adopted a sedentary life-style. They employed slaves,
who rapidly became integrated into the political community.
Whether the producen were free men, tributaries or slaves, they were mem¬
bers of a political community and their relations with the state were codified.
Distinctions based on wealth and on proximity to the machinery of state were
gradually introduced.
Traders
Traden were divided into several categories, ranging from wholesalers engaging
in fatauci (trade over medium and long distances) to retailers who organized the
kasuwanci (retail trade). A group of attajirai (rich traders) was formed which
included merchants dealing in slaves, hones, kola nuts and clothing. Traden
and scholan seem always to have been closely linked. The traders were united
by common interests and they did not hesitate to emigrate when they felt they
were being taxed excessively,
Between 1500 and 1800, the various Wangara and Hausa sub-groups
succeeded in uniting the trade networks throughout western Africa and consti¬
tuted the middle class in the main towns. They continued to consolidate their
position as a class and spread the Islamic faith to their advantage.
Tlie aristocracy
The ruling class included the masu sarauta (all those with any political authority).
It was subdivided into several levels.
The sarki was at the head of the nobility, which was composed of the
princely families, families governing the semi-autonomous cities and vassal
provinces, and representatives of the various nationalities, particularly the
Azbinawa, the Shuwa Arabs and the Fulbe. The nobility performed various
duties within the machinery of state, and tended to become a homogeneous
group whose cohesion was strengthened by marriage ties.
All the dignitaries appointed by the sarki to run the state formed the second
level. They were of commoner or serf origin but their position gave them
access to wealth and esteem through the gifts they received, but particularly
through pillaging. In some instances, they held key military office. It was they
who urged Muhammad Nazaki to reject Katsina's authority and opposed the
attack against Katsina planned by Zaudai. Kebbi's decline began when the
dignitaries holding military office had grown so rich that they lost interest in
affain of state.
The aristocracy, primarily the princes and royal slaves, confiscated the
property of the talakawa particularly when the sovereign showed signs of
weakness.
The question of social relations seems to have been dominated by two

factors: the sale of Africans by certain sovereigns and the participation of Kawar
and Zawila in the slave trade. In Känem, the first indicators of the export of
slaves and their use within it appear as early as the eleventh century. In the
mid-fifteenth century galadima Dawuda went and waged war and sent sarkin
Kano a thousand slaves every two months. By the end ofthe campaign, he had
accumulated 21,000 slaves scattered among 21 villages and all named Indabo.
The term indabo (village of slaves) is reminiscent of the Soninke debe, the
Fulfulde debeere and the Songhay dabey. In these villages, the productive
activities were crafts and particularly agriculture. In Katsina, the city of Tsagero
was a royal property where large numben of slaves were kept, and where even
princes were sent. In the estates in the Gozaki area, cotton-growing eventually
came to depend on slaves imported from Zazzau, and part of the crop was
exported to Kano.
Whether the slave was a commodity, a servant, a high-ranking official or an
independent producer in a subordinate position, he had a role to play in the
development of the economy and the state. However, the export of slaves has
to be examined in relation to the sources of supply in Europe and the Orient,
even before the Atlantic trade began to claim its share. The contribution of
slavery to the prosperity of the region will become clearer when it has been
singled out from all other forms of subordination.
The masu sarauta (aristocrats) can be regarded as opposing the talakawa, who
were free producen but had no political power. As the aristocracy, the scholan
and the traders grew rich, the distinction shifted to an economic level between
the attajirai, the wealthy, and the talakawa, the poor. The bawan sarki (royal
slave) ceased to be a talaka, from both the political and the economic stand¬
points. The situation was thereby clarified, because the ethnic and religious
differences had been relegated to the background, and this left those in power
and the second-class citizens confronting one another.
Culture and religion

Political and economic developments gave rise to many cultural changes. On a
material level, for instance, architecture was improved and, even today, the
cities still compete with one another in their styles. In music, some instruments,
such as the kakaki and the algaita, became widespread and orchestras were
brought into the royal ceremonial. The main feature of Hausaland in this
respect was its richness and cultural unity the outcome of a homogenizing
process that made Guber a lingua franca in Africa under the name Hausa.
Islamization at the beginning of the sixteenth century was confined to
traders and the political élite. However, the end of the period was to be
marked by an intensification of the open struggle between the aristocracy and
the scholan. The Kano Chronicle and the Wangara Chronicle show the develop¬
ment of their relations from the inside.
Kano witnessed a massive and regular influx of scholars. In Kisoki's reign

(1500-65), several scholars arrived. The first, Shaihu Ba-Tunashe, brought with
him the ashafa book; the second, Dan Gwarandume, was to read this book; and
the third, Shaihu Abdussalami, introduced three books. Then a group of three
brothers from Borno arrived, one of whom accepted the post of kadi. In the
reign of Abu Bakr Kado (c. 1565-73), a second group of scholan, from
Bagirmi, immigrated to Kano. They had stayed some time in Katsina, and
Muhammad Zaki (c. 1582 married one ofthe daughters of Tama, their
leader. It was on account ofthe scholars that the Katsinawa did not sack Kano,
and it was the scholars who negotiated peace between Kano and Katsina in the
period between 1648 and 1651.
In Birnin Katsina, a distinction could be made between several groups
descending from Wâlï Abu 'Abdullahi b. Masani, Malam Buhäri and Malam
'Uthmän. Many cultivated people from differing backgrounds, ethnic groups
and places of origin lived in other cities. This intelligentsia, which was scattered
over a territory whose population and leaders considered themselves Islamized,
kept their distance from the ruling class.
Zamfara differed in that Islam did not take root there until a late stage,
probably because most of its urban centres were far from the caravan routes of
central Sudan. Nevertheless, Kanuri scholan may have participated in con¬
verting the sovereign to Islam. The imam of Anka had nothing short of a fief,
and his residence was a place of refuge for all those who incurred the wrath of
the rulers. Some posts were reserved for scholars: the limanin ciki was
responsible for the education ofthe royal family, while the dan kodo and the dan
dubal were advisors on religious affain, custodians ofthe history of Zamfara and
responsible for praying for the army's victory.
For their income, the scholars relied on the generosity of the aristocracy.
However, it is difficult to fault the objectivity ofthe scholan, even when, as in

Kano, they were granted fiefs and estates. Their critical attitude was reflected in
the Kano Chronicle.
The regular travelling of the scholan and the introduction of books partly
explain the area's intellectual awakening. However, there must be added to that
the existence of centres of learning and, in particular, the use ofthe ajami script
(transcription in Arabic) in Kano and Katsina in the sixteenth century. The
scholan wrote in Arabic, Fulfulde and Hausa.
Praying for victory in war and consulted on legal matters, the scholan had a
decisive influence on social life, particularly through the wa'azi (sermons). In
Zamfara, among the most celebrated, mention has been made of Ramadan b.
Ahmadu (from Fezzan), Hashimu Bazanfane, Maman Tukur dan Binta and al-
Mustafa Gwani, a Kanuri who confronted 'Uthman while he was touring
Zamfara on the simultaneous presence of men and women during the sermons.
Two names stand out in Katsina: Abu 'Abullähi b. Masanï b. Muhammad al-
Barnâwï al-Kashinawï (c. 1595-1667) and Muhammad al-Sabbägh al-Kashinawï,
better known as Dan Marina, who flourished in about 1650 and composed a
poem in praise ofthe mai 'Ali, who defeated Kwararafa in about 1680.
However, there can be no doubt that it was Malam Djibril dan 'Umaru who
stood out most strikingly from the group. This scholar, with his encyclopaedic
knowledge, was born in Adar and died there, after having been on several
pilgrimages. His main subject of preoccupation lay in the reform of Islam in
Sudan. After his failure among the Tuareg in Adar, he approached the Hausa
princes, whose hostility he aroused. 'Uthmän was his disciple. The second half
of the eighteenth century was marked by a wealth of intellectual activity and
lively debate among scholars, a simple form of opposition to the established
order represented by the masu sarauta.
Conclusion
Just before the djihâd, there was a marked decline in the turbulent character of
the relations between states. Kebbi and Zamfara had been defeated, but the
othen were faced with difficult problems. In Kano, peace and stability were
punctuated by two serious disagreements between the sarki and the dignitaries.
In Gobir, Bawa (c. 1777-89) for the fint time introduced a tax on a certain
variety of maize, and stock-raisers had to pay the jangali on several occasions
during the same year. Two dynastic crises shook Katsina in 1767 and 1796. The
risk of instability was noted almost throughout the region, due to resistance on
the part of the dignitaries, the increased weight of tyranny and the latent
opposition of the talakawa. The political authorities had attained a threshold of
oppression which no longer spared the scholars, but they were the only ones
who ventured to combat the established order openly.
Economically the apparent opposition to progress that seemed to
characterize the years between 1600 and 1790 has been oventressed. It is true
that neither the wheel nor the windmill was used, and even the musket was
Tlie Hausa states 247
scorned. However, despite a host of natural calamities, the area developed in a

remarkable manner, owing not only to the growth of trade, thanks to state inter¬
vention in various forms, but also to progress in production and processing.
Politically, despite the armed conflicts, none ofthe states disappeared during
the period under consideration. Elsewhere, Ghana, Mali and Songhay disinte¬
grated, while Känem-Borno enjoyed remarkable longevity. The economic
success of Hausaland has overshadowed the great stability of the state as an
institution. Its bureaucratic and oppressive nature was highlighted by the
measures introduced by Sharif in Kano. In fact, the machinery of state was
efficient and few alterations were made to it by the regimes that followed.
Känem-Borno: its relations
with the Mediterranean Sea,

Bagirmi and other states
in the Chad basin
In the thirteenth century the Muslim state of Känem was one of the most
extensive in the Biläd al-Südän. The Sëfuwa mais (kings) controlled the vast area
from the eastern shores of Lake Chad in the south to the oases of Fezzän in the
north, but their expansion had not been accompanied by appropriate political
and economic development. The emergence of various more or less autono¬
mous ethnic groups distinct from the Sëfuwa ruling dynasty had prevented the
development of a centralized political system. In addition, Känem, mostly
desert and semi-desert, lacked the primary resources needed to make such a
large system viable.
A serious crisis finally led to the collapse of the Känem state in the latter part
ofthe fourteenth century. The mai 'Umar b. Idris (1382-87), at the head ofthe
Magumi group and its supporters, left for Borno to the west of Lake Chad,
where the necessary resources existed and where the Sëfuwa had already setded
their vassals. A large number of other immigrants from Kanem had also already
setded there.
The Sëfuwa's main objective on their arrival in Borno appears to have been
the building of a strong regional economy to support a well-organized, Sëfuwa-
dominated political structure. However, during their first century there, they
faced many problems which at times posed threats to their very survival:
constant civil strife, dynastic crises, periodic attacks from Búlala and overmighty
title-holders.
Mai 'Alt Gaji and the foundation ofthe Borno califate c. 1465-97
'Ali b. 'Dunäma, commonly known as 'Ali Gaji, was able to consolidate power
in his own branch of the ruling dynasty. When, in about 1471, the Búlala
launched new incunions into Borno, he defeated them but did not punue
them further. In about 1472, he built the fortress of Birnin Gazargamu which
eventually developed into the capital of the Sëfuwa and remained so through¬
out their rule in Borno.
'Ali Gaji is remembered as a reviver of Islam. He attempted to found a
248
Känem-Borno 249
proper Islamic government, surrounding himself with 'ulamâ' whose advice he

sought. Many of the overmighty title-holders had their powen reduced,
especially the kaigama and the yerima (governors ofthe northern provinces).
During his pilgrimage to Mecca in about 1484, 'AlT Gaji was said to have
been invested as the khalifa of Takrur. From this time onwards the rulers of
Borno regarded themselves as khalifas, a claim that came to be accepted by
many scholan and rulen in the Bilâd al-Sûdân.
Problems and opportunities in expanding the Borno califate:

c. 1497-1 564
'Ali Gaji's policies were continued by his son and successor Idrîs b. 'Ali,
surnamed Katakarmabe (c. 1497-1519), who consolidated his gains and tried to
expand the state. But for the next 50 yean or more Katakarmabe and his
successors faced many challenges which diverted their attention from their
goals.
The Sëfuwa's fint problem was the resumption of Búlala attacks.
Katakarmabe had to face a Búlala attack as soon as he became mai. Not only did
he defeat them but he victoriously re-entered Ndjimi, the former Sëfuwa
capital which, however, was never reoccupied. After concluding a treaty with
the Búlala, Idris returned to Borno, but the truce was short-lived and inter¬
mittent hostilities continued into the reign of Idrîs Alawoma (1564-96).
Another problem for the Sëfuwa kings was the emergence of many states in
the Chad basin and elsewhere in the Sudan, which forced the Sëfuwa to setde
their problems with each state.
By the early 1500s numerous petty kingdoms had emerged along the
southern border of Borno. These included Bagirmi, Mandara, the Kotoko
states, Northern Bolewa of Daniski, Yamta and Margi. Some of these states
were forced to recognize some form of Sëfuwa hegemony. But more often the
Sëfuwa tried to enter into peaceful relations with these nascent states,
encouraging them to develop their own economies and establish regular trade
with Borno. Through this association the emerging states absorbed much of
Borno culture, which may have contributed to their rapid growth.
In the same period, many states were also emerging in Hausaland. Katsina
and Kano soon developed as the termini of the trans-Saharan route as well as
entrepôts of the east-west route by which Akan gold and kola were taken to
Borno. Further north, there was a slight shift of the trans-Saharan route when
Agades supplanted Takedda as the main entrepôt.
Borno had to react to these developments. The new trade route between
Borno and Hausaland soon became unsafe and efforts must have been made to
protect it. The problems concerning this trade route and the struggle for
control of the setdements that developed along it may have contributed to the
conflicts between the mai Idris Katakarmabe and two kings of Kano, 'Abdullahi"
(c. 1499-1509) and Muhammad Kisoki (c. 1509-65).
Borno must also have wanted to control the new entrepôt of Agades which
the then very powerful Songhay sought in 1501 and 1515 to bring within its
sphere of influence. However, the emergence of Kebbi in the reign of
Muhammadu Kanta (c. 1516 frustrated the efforts of both Songhay and
Borno. The problems of Agades continued to bedevil the Sëfuwa kings almost
throughout their rule.
The period from about 1480 to 1520 was a time of active Islamization in the
Biläd al-Südän. In the east, the Fundj sultanate, which was founded in about
1504, soon adopted Islam. In the west, many scholars from Mali, North Africa,
Egypt and the Saharan oases visited Hausaland and contributed to its Islamiza¬
tion. Further west, the Songhay empire reached its zenith with the askiya
Muhammad (c. 1493-1528), whose reign coincided with a bunt of Islamic
fervour. During his pilgrimage to Mecca (c. 1496-8), this ruler was invested
with the tide of khalifa of Takrür, as the mai 'Ali Gaji had been ten years earlier.
The Sëfuwa mais had been Muslims since the eleventh century and they
traditionally surrounded themselves with (ulamä\ which gave them an edge
over other rulers. The decline and fall of the Songhay empire in the late
sixteenth century finally ensured their hegemony.
Borno became a great intellectual centre visited by scholars from the Biläd al-
Südän and other parts of the Muslim world and its cultural influence was felt in
many states.
From the reign of Idrîs Katakarmabe, records exist of diplomatic and com¬
mercial relations between the Sëfuwa and the various authorities controlling the
North African coast. In about 1512 Katakarmabe sent a diplomatic and trade
mission to the Spaniards, who had recendy occupied Tripoli, to renew com¬
mercial ties with the area. When the Ottomans were established in the Maghrib,
Dunäma b. Muhammad in about 1555 sent an embassy which established with
Tighürt Pasha a treaty of 'friendship and trade' which was renewed by their
respective successon. Strong ties were probably also established with Egypt.
Katakarmabe's descendants continued his policy in dealing with the Búlala
incunions and the emergence of new states. In about 1561 there was a war
between Kebbi and Borno over control of Agades, which Borno seems to have
lost.
Establishment of a strong regional economy and a centralized political

system and the emergence ofthe Kanuri, c. 1564-77
Most scholars agree that Borno reached its apogee during the rule of Idris b.
'Ali, posthumously surnamed Alawoma (1564-96). He was seen as a military
and administrative innovator as well as an Islamizer and a skilled diplomat.
His main objective on his accession was to equip his country with a strong
economic and political system. He was the fint mai to attempt to bring the whole
metropolitan province firmly under control. The cavalry was reorganized. There
was also a corps of Turkish gunmen which had been developing under his
Känem-Borno 251
predecesson. Larger boats were built for easier crossing of riven. It was with such
a force that Alawoma launched his conquests.
Most of the hostile groups which refused to be persuaded or coerced into
submission by the mai were continually attacked. These included the Ngafata,
the Talata, the Dugurti, the Maya, the Ngizim and the Bedde. Some, such as
the Mukhulum, were allowed to retain their homes in return for a promise of
tribute in wheat. Others, considered implacable, were ejected from the metro¬
politan province.
In place of these groups expelled from their country, others, mostly from
Kanem, came and setded in Borno. These large-scale demographic changes,
together with extensive intermarriage with female slaves captured during the
conflicts or exchanged with satellite states, contributed to the development of
the Kanuri group within the metropolitan province.
In southern Borno ribäts (frontier fortresses) were built to contain dissident
groups such as the Gamergu, and agreements were entered into with satellite-
state rulers for joint military expeditions against the dissidents.
The final solution to the problems of Känem

After the pacification and integration of most of the metropolitan province,
Idris Alawoma turned his attention to Känem to put an end once and for all to
the threat it posed to Borno. He had three objectives: to destroy completely
the military strength and power base of the Búlala; to destroy the economic
base ofthe Känem state; and to transfer as many groups as possible to Borno.
Alawoma undertook several expeditions against the Búlala and other groups
who supported them. Three ofthe most fertile valleys of Känem were ravaged
and some of the most important towns, such as Ikima, Aghafi and Ago, were
destroyed and their population removed to Borno. The groups forced to settle
in Borno included the Tubu, the Koyam, the Kulu and the Shuwa Arabs. The
last two groups, who were catde-owners, were settled along the southern
shores of Lake Chad and to the west of the metropolitan province. The Tubu
were settled on the desert margin, both for strategic reasons and to participate
in the production of salt.
By the 1580s, Alawoma had achieved most of his objectives and Känem state
was, by common agreement, partitioned between Borno and the Búlala, the
latter accepting some loose form of control by Borno.
Internal reforms
Alawoma seems to have made a pilgrimage to Mecca in about 1571. On his

return he attempted to introduce a number of reforms to bring the country
into line with other Islamic states. He tried, apparendy without success, to
separate the functions of the judiciary from those of the executive. He did,
however, succeed in establishing a court of appeal where 'The learned men and
Imams held disputations before the Amir 'Ali concerning doubtful points of law
and dogma' (H. R. Palmer, 1936, pp. 33-6). The court attracted numerous
North African and other scholars, which gave it a cosmopolitan character.
The rise of a strong regional economy

The conquests of and demographic changes under Idris Alawoma, as well as his
other reforms, led to the emergence of a strong regional economy based on the
metropolitan province, an area of some 20,000 square kilometres.
In the east there was large-scale fishing along the shores of Lake Chad by the
Buduma and Känembu, while the good grazing lands attracted a large number
of Känembu, Shuwa and Fulbe cattle-raising nomads. Natron and salt were
produced by the sedentary Känembu and Buduma. Important urban centres
sprang up, including Mungono, Kauwa, Burwa and Ngurno.
In the west of the country there were also big natron deposits, which were
worked by the Manga and the Tubu. In the south was a large concentration of
farming peoples including the Ngizim and the Bedde. This area received a
great variety of immigrants and the many urban centres which developed there,
including Nguru Ngilewa, Mashina Kabshari and Maja Kawuri, were more
cosmopolitan than their eastern counterparts.
Birnin Gazargamu and its environs formed the third economic centre.
Situated on the trade route that crossed the country from east to west, Birnin
Gazargamu was also the terminus for the BornoKawärTripoli caravan route
as well as for the main route connecting the area with Hausaland. Thus the city
was the nerve-centre of the country's economy and its large market attracted
many foreign merchants. In the fertile, irrigated Yobe valley there was a heavy
concentration of the emerging Kanuri group which had settled in many urban
centres in addition to Birnin Gazargamu.
The development of basic industries led to the establishment of supporting
ones such as pottery, weaving, leather work, dyeing and transportation.
The natron and salt trade extended to the Atlantic coast, Hausaland, the
Volta basin and northwards to Azbin and Adrär. Dried fish was traded within
the country and with Mandara, Hausaland and the Saharan oases. Slaves,
eunuchs, hides, ivory, perfumes, leather goods and gold were sent across the
Sahara to North Africa and Egypt. In exchange, hones, horse-trappings,
armour, copper, bronze and other goods came from Europe and North Africa
to be distributed in Borno and other parts ofthe Biläd al-Sudân.
From the reign of Alawoma, various peoples around the Chad basin were
apparently encouraged by the Sëfuwa to engage in commercial activities. These
peoples were all referred to as Bornoans and must have helped to spread the
emerging Bornoan culture in Hausaland and other areas. It was through its
regional economy, its control of trade and its primacy in Islam that Borno came
to dominate the affairs ofthe Biläd al-Sudän during this period.
The mais interfered little with the trade itself, limiting their role to pro¬
tecting the trade routes and entering into agreements with Saharan and North
African governments to ensure the free flow of trade. On the other hand, they
were active in the organization of markets to encourage traders and to obtain
revenue. In the capital and the other major centres, the wasili (North African
traden) were recognized as guests ofthe government. A quarter was assigned to
Känem-Borno 253
them and a titled official, the zanna atjinama, looked after their affairs. In the
markets, the mala kasuube supervised sales and took pains to ensure honesty in
major commercial transactions.
The diplomatic relations of Idris Alawoma with the sultans of Turkey

and Morocco
Ottoman-Borno relations started with the arrival ofthe former in the Maghrib.
From about 1549 when the Ottomans took control of Fezzän to about
1570, these appear to have been cordial. Things changed in about 1571, when
the Ottoman authorities suddenly raised the annual tribute of Fezzän. This
decision coincided with a period of famine, and many people were forced to
migrate to Hausaland and Borno. The Ottoman officials in Fezzän forced those
who remained, many of them old people, to make up the difference. Even the
pilgrims from Hausaland and Borno were stopped and forced to pay taxes.
Alawoma was able to obtain first-hand information about the situation when
he went on a pilgrimage to Mecca in about 1571. In about 1574, anxious to
ensure the security of the route both for pilgrimage and for trade, Alawoma
dispatched an embassy to Istanbul with three specific requests: a guarantee of
safe passage and security of property for all travellen from Borno crossing
Ottoman territory; proper management or, failing that, the ceding to Borno of
all the recendy acquired fortresses south of Fezzan, including Guran; and co¬
operation between the two powers in dealing with the troublesome Tuareg and
any other people disturbing the peace in the area. There appears to have been
no mention or solicitation of arms.
In May 1577, Sultan Murad III agreed to all these requests except the ceding
of the Guran fortress, which he promised would henceforth be properly
managed. From then on, cordial commercial and diplomatic relations
continued between the two powen.
The Moroccan victory at Kasr al-Kablr in 1578 was generally hailed as an
ideological victory for Islam over Christendom. Borno, like the leading Muslim
powen, sent a delegation to congratulate the sultan of Morocco. However,
Alawoma was fearful of proposed joint Ottoman-Sa'ädi expeditions against the
Saharan oases and even the Sudan and sought to frustrate the possibility by
suggesting a joint Borno-Moroccan expedition to the Saharan oases, once
again in a state of some insecurity. The outcome of this embassy, according to
Moroccan sources, was the acceptance of Borno's suggestions on condition that
Alawoma acknowledge al-Mansür as the khalif of the time. For Morocco,
securing recognition of his califate by one of the leading powen in the Sudan
would constitute an ideological victory and strengthen his position in relation
to Songhay. For Alawoma, the price was worth paying if it forestalled the
threat of joint Moroccan-Ottoman action against Borno or any other area
where its interests lay.
Towards the end of his reign, Alawoma was probably preoccupied by the
rapid development of the Mandara and Bagirmi states. Alawoma was forced to
lead several expeditions against Mandara, most of them unsuccessful. In

Bagirmi, Abdullah b. Lubetko (c. 1561 despite having probably gained
the throne with Bornoan assistance, started making trouble. It was on a
campaign against 'Abdullah that Alawoma is said to have been assassinated by a
Gamergu.
The death of Idris Alawoma, contrary to the views of some writers, did not
lead to the collapse of the Borno empire. It was in fact during the reigns of his
four immediate successors (c. 1596 that his conquests were consolidated
and the administrative machinery assumed its final form. It was also at this time
that the Kanuri finally emerged as an ethnic group with a distinct culture.
The social and material culture ofthe Kanuri
The term Kanuri probably came into use in the early seventeenth century. It
refers to the dominant ethnic group of Borno upon whom the power of the
Sëfuwa mais was based. The Kanuri were the product of intermarriage of the
Magumi from Känem and the indigenous Chadian speakers of Borno, a
development accelerated by the population policy of Idris Alawoma.
Most Kanuri lived in villages in compounds containing several round huts
with conical grass roofs. Many compounds were surrounded by siigedi (grass
matting). In the larger towns, most huts and mosques were mud-built, and the
houses of the rich were protected by high mud walls. Most towns and villages
were in the shape of a U, in the middle of which the dandal (main street),
running eastwest, led to the palace of the local governor, next to the main
mosque.
At Birnin Gazargamu and in some of the larger towns, the houses were
usually rectangular with thick mud walls and flat roofs. Royal palaces at Birnin
and Gambaru, their main mosques and the residences of important dignitaries
were surrounded by impressive walls of fired bricks, one of the distinguishing
traits of Kanuri culture from the mid-sixteenth to the early eighteenth century.
A typical Kanuri was distinguished by language and physical appearance.
Both men and women had distinguishing vertical marks on each cheek. An
unmarried girl wore her hair in a kela yasku (a special style) while a married
woman wore her hair in a jurungul (crown). Women ate kola nuts and stained
their teeth with tobacco flowers. They usually wore gimaje (long dyed cotton
dresses) and women of distinction wore a kalaram (turban). Men wore tobe
(large open cotton gowns either plain or dyed blue). Men of the upper class
wore a multiplicity of gowns of expensive imported cloth, and in the late-
eighteenth century they wore outsize turbans, a custom that appean to have
been borrowed from seventeenth-century Ottoman practice.
Marriage in Kanuri culture had certain distinctive characteristics involving a
mixture of Islamic rites and Magumi and other pre-Islamic practices.
Kanuri society was highly stratified. It was broadly divided into two classes,
the kontuowa (ruling class or nobility) and the tala'a (commoners), and both of
these had several subdivisions. Differences of speech, dress and housing dis¬
tinguished the classes and their subdivisions. Upward mobility was achieved
Känem-Borno 255
through amassing more wealth, associating with the ruling class or obtaining a
high-ranking office.
Kanuri socio-political culture laid especial stress on inferior-superior relations.
A social inferior was always required to show his social superior respect in
public. In addition, he was also required to pay respectful visits to the house of
his 'social father', who could ask him to do any favour. In return, the superior
took care of his inferior's basic needs and advanced his interests in society.
Borno by this time was a centre of learning which attracted large numbers of
students and visiting scholars from the Biläd al-Südän, North Africa and the
Middle East. Borno specialized in tafsir (commentary on the Kur'än) in the
Kanembu language, and the custom of writing the Borno language in Arabic
letten seems to have developed in the seventeenth century.
There were two types of scholar. The fint held official religious offices. The
imams, the kadis, the mainin kinendi (Islamic and legal adviser to the mai), the
talba (head of the police and the magistrates), the khazin (treasurer) and the
wazir were the great {ulamä) who helped the mai govern according to Islamic
precepts. Their positions were hereditary and restricted to a few leading
scholarly families. The institutionalized character of the offices they held
associated them with the temporal power and increasingly alienated them from
the non-court scholars.
The more austere scholan lived in the rural areas and propagated their
religion. During the time of the Sëfuwa, most of them enjoyed the protection
ofthe kings, who granted them mahrams (privileges). However, it was also they
who produced the fint intensive criticism of Sëfuwa rule.
Political organization
The mai was the head of the royal family, the supreme head of state and the
larde kangema (nominal owner ofthe land). He was the symbol ofthe unity and
continuity of the state. As the amir al-milminlm (commander of the faithful) he
was the leader of all the Muslims and the final arbiter in both public and private
matters ofjustice.
Many of the sacred attributes of the pre-Islamic kings seem to have survived
into the Islamic period. The mai was still largely secluded, appearing in public
only in a fanadir (cage), and people could not speak to him direcdy. Muslim
scholan appear to have enhanced the sacred attributes ofthe king.
The mai was selected from among the maina (eligible princes), but the
succession was not always smooth. From the time of the mai Idris b. Ali
(c. 1677-96), or earlier, the kings seem to have resorted to eliminating rival
candidates a habit they probably learnt from the Ottomans.
The other members of the royal family consisted of the king's four titled
wives, the concubines, the princes and princesses, the magira (queen-mother)
and the magram (the king's official sister). The gumsu (the king's head wife) was
responsible for administering the palace; the magira was the holder of the largest
number of fiefs in the kingdom and had the right to grant sanctuary; the
magram usually supervised the cooking ofthe king's food.
Most princes resided outside the palace and their activities were strictly
controlled, although some were given fiefs. The most important of these were
the chiroma (heir apparent) and the yerima (governor of the northern provinces)
who was always a maidugu (grandson of a previous king) which meant he could
have no claim to the throne.
The highest council of state was the majlis, which was normally presided
over by the king and was composed of military and religious notables. Most of
the Muslim advisen already mentioned as well as military commanders were
memben. It also included powerful slaves of the mai, some of whom were
eunuchs. The majlis examined all great affairs of state. The noguna (mai' s court),
which consisted of all the notables in the capital, met daily.
The towns, villages and ethnic units were grouped into chidi (fiefs) and all major
officials of state were chima (fief-holden). The chima were responsible for the
maintenance of order in their fiefs, for the collection of taxes and for the raising
of troops for the army. The mai could confiscate, reduce or redistribute fiefs
entirely as he saw fit. At the local level, the people were ruled by mbarma or búlala.
The Sëfuwa's income included the zakat (compulsory alms), dibalram (road
tolls), kultingo (tribute) and war booty.
The golden age in Borno

The seventeenth century not the sixteenth as traditionally claimed should
be regarded as Borno's golden age.
It was during the rule of 'Umar b. Idris (c. 1619-39) that the offices ofthe
galadima of Nguru and the alifa of Mao in Känem were established as semi-
autonomous vassals to take care of the western and eastern marches of the state
respectively. Buffer states were established along the desert fringe at Muniyo
and Mashina as barrien to Tuareg attacks on the metropolitan province. This
was the period when 'Abd al-Karim ben Jame (c. 1611-55) founded Wadai -
no doubt with Borno's blessing, since Wadai checked for some time the expan¬
sionism of Bagirmi. As with the other neighbouring states, Wadai fell within
the political, cultural and commercial orbit of Borno.
Diplomatic and commercial relations with the Mediterranean littoral entered
one of their briskest periods. Relations with Egypt must also have been good,
through trade and because Egypt lay on the pilgrimage route to Mecca.
By the seventeenth century, particularly by the reign of Ali b. 'Umar
(c. 1 639-77) , Borno had become the dominant power in the Biläd al-Sudän.
Not only was it a centre of learning and culture but it also controlled all the
easily accessible sources of mineral salts at Muniyo, Bilma and around Lake
Chad. Its close trade connections with the Mediterranean coast led to its
domination of the redistribution of imported goods to the states of central
Sudan. The mais were recognized as the undisputed leaden of the Islamic states
ofthe region. Most of these states appear to have placed themselves voluntarily
under the protection ofthe khalif in order to benefit from his baraka (blessing)
and to obtain the goods they needed. It was the general erosion of this system
that led to the decline and final collapse of the Sëfuwa.
Kanem-Borno 257
Crises of the late eighteenth century

With the reign of 'Ali b. Dunäma (c. 1742-92), Borno entered a period of crisis
which climaxed in the djiltâd of the nineteenth century.
In western Borno, the Bedde were intensifying their raids on the trade
routes to Hausaland, while in the north the Tuareg of Agades were stepping up
their incursions. All this precipitated the southward shift ofthe Manga groups.
In Bagirmi, Muhammad al-Amin (c. 1751 threw off Borno's overlord-
ship and launched a series of attacks on Borno's eastern frontiers, while Wadai
under Jawda (c. 1747 continued its expansionist policy towards Bahr al-
Ghazäl. These attacks set in motion the migrations ofthe Tubu, the Känembu
and the Shuwa Arabs into Känem and the metropolitan province, thereby
causing conflicts over grazing lands.
In about 1791, Mandara revolted, soon followed by the Sosebalki states and,
in about 1785, by Gobir.
Faced with insecurity, famines and impoverished grazing lands, many Fulbe
nomads abandoned metropolitan Borno and moved to Hausaland, Mandara and
Fombina. Nowhere, however, did they find the security they were looking for,
and this may pardy explain their dominant role in the djihâd.
The non-court scholars and the ordinary Muslims blamed the crisis on the
increasing corruption of the Sëfuwa rulen and the inability of the mais to
prevent the growing syncretism among the people. When the djiliäd broke out,
the client rulers in Hausaland called upon the mai to fulfil his obligations as
khalif and come to their aid. That he could not and that he could not even
prevent his own expulsion from his capital meant that the khalif had lost his
primacy.
However, although the djihâd did contribute to the loss of the vassal states
and to the end of the Sëfuwa rule, the system that the mais had built,
particularly in metropolitan Borno, survived, though modified, well into the
twentieth century.
From the Cameroon grasslands
to the Upper Nile
In this part of Africa, historical information for the period 1500-1800 is less
plentiful, coherent and reliable than for other regions. In areas where political
organizations took the form of states, these were mostly formed very late, in
the eighteenth and nineteenth centuries. Oral traditions appear uninformative
and rarely go back more than a few generations. Those which do exist often
record the individual history of a particular lineage and raise the usual problems
of interpretation. Other sources are very sparse. A few travellers' accounts exist
from the sixteenth century onwards for the areas in contact with foreigners.
Finally, archaeology, historical linguistics and ethnolinguistics are only just
beginning in the area.
There is, moreover, no single factor uniting the different parts of this region.
The forest is certainly its dominant physical feature, but many parts of the
region have long been savannah. The low population density appears to be
another common characteristic, but there were areas of relatively high density,
notably in the savannah and along the waterways. As for the peoples, their
languages, ways of life and forms of organization were extremely diverse. This
divenity and the inevitable uneven distribution of information necessitate a
regional approach to the area.
The northern border area
The northern part of this region, from the central Cameroon highlands in the
west to the River Nile in the east, is the area for which historical information is
most lacking. Moreover, the tendency of ethnological and historical works to
restrict themselves to the limits of ethnic groups and, more recently, to the
limits of contemporary states, makes any regional synthesis difficult. It seems
wiser therefore, to try to define micro-regions which have some degree of
cultural homogeneity or, at least, some degree of common destiny.
The central Cameroon highlands

These relatively fertile grasslands were probably setded very early. There are at
least two indications of this early settlement. The first is vegetation, which is
259
today wooded savannah but was once dense forest. This savannahization (the
chronology of which is not yet established) might equally well be attributed to
the slow and gradual action of compact human communities, to the sudden and
brutal action of a small number of immigrants, or to climatic influences. Second,
the presence of numerous stone tools testifies to settled human occupation
which probably occurred in the second half of the fint millennium. But it
remains impossible to identify these first inhabitants. They may have been
pygmy hunten, a few of whose descendants still survive on the banks of the
Mbam, or proto-Bantu language speaken who had not yet discovered metal-
working; or a fluid mixture of several distinct peoples.
It is only from the sixteenth century onwards that facts begin to crystallize.
The population movements recorded in the traditions involved many different
sorts of people. All the evidence suggests that the number of migrants was small
for, although they dominated the peoples they found already in situ, every¬
where they abandoned their own language and adopted that of the country
where they settled. The traditions also reveal the countless reasons for these
movements, which included internal disputes, the search for iron ore and wood
to work it, the quest for new land, the salt trade and the pressure of invaders
from the north. The northern factor, particularly the Fulbe raids, which is well
known for the nineteenth century, was relevant much earlier, from the mid-
eighteenth century or even before. The vocabulary and syntax of the Mbum
language, spoken in the central Cameroon highlands, include many borrowings
from Fulfulde. It thus seems that the Fulbe drove the Mbum southwards and
that they in turn forced the Voute to move southwards, and that this led to the
migration ofthe Fang and Beti.
This constant intermingling of populations, spread over several centuries, is a
main factor in the striking cultural homogeneity of the peoples of central
Cameroon. The peoples there all speak Bantu or semi-Bantu languages. There
are numerous similarities in the way in which power is distributed in their
political and social institutions. Males belonged to countless associations with
ritual or police functions. Technologies were often comparable, too, especially
in the working of iron and copper. In all these societies the smith was believed
to have magical powers and is prominent in myths and historical traditions.
This homogeneity, well established by about 1800, had been gradually formed
during the previous centuries.
Three of these numerous peoples deserve special attention. The first are the
Tikar, who are important because many other peoples in the Cameroon
highlands either claim to derive from them or have been strongly influenced by
them. The Tikar are said to descend from a trader from Borno who settled and
had offspring among the Mbum. Their various traditions locate their origin in
north-east Cameroon. Tikar was initially the nickname of the fint group of
Mbum to leave their homeland for the highlands, but in time it came to be
applied to all Mbum emigrants and the peoples they had conquered. The major
part of this migration probably took place in the seventeenth century. When
they reached the Bamenda region, the Tikar met the Tumu whom they first
From the Cameroon grasslands to the Upper Nile 261
allied with and then dominated. But the price of victory was their adoption of
the language of the conquered and their institutions, particularly the political
titles and secret societies.
The foundation of the Bamum kingdom was the work of a group of

emigrants related to the royal dynasty of the Tikar of Nsaw. In the History and
Customs of the Bamum, ten kings are fisted between the founder, Nchare, and
Kuotu, the ruler immediately before Mbu embue, whose reign is dated to the
fint half of the nineteenth century. Nchare was a conqueror who crushed some
18 rulen. He founded the capital of the Bamum, Foumban, then called
Mfomben. Nchare endowed the kingdom with political institutions mostly
taken from those of the Tikar. He set up a body of numerous title-holders,
including the kom ngu (counsellon of the kingdom) with whom he divided up
the available lands; he formed two secret societies, one of which, ngiri, was for
princes only and the other, mitngu, was open to the rest of the population
irrespective of social status. However, the Mbum gave up Tikar and adopted
Mben, the language of the conquered. The nine kings who followed Nchare
get short shrift from tradition. They were not conquerors, for territorial
expansion began only at the beginning of the nineteenth century during the
reign of Mbuembue. The kingdom even faced the serious threat of Chamba
and Fulbe invaden from the north, particularly during the eighteenth century.
However, some significant developments can be detected during this period.
The growth in large-scale royal polygamy led to a vast increase in the number
of pnncely lineages and the palace nobility also grew substantially. The king
recruited most of his retainers from twins and the sons of princesses. Material
and social developments remain litde known. At the end of the eighteenth
century, the kingdom had perhaps 10,000-12,000 inhabitants, among whom
there was a small number of slaves. The economy, heavily agricultural, was
open to external trade. The Bamum were importing salt, iron, beads, cotton
goods and copper objects from a very early date.
As for the Bamileke, their ancient history is closely linked to that of the two
previous groups. All came from the north, from the region today occupied by
the Tikar. Their migration probably began in the seventeenth century and took
place in successive waves. The migrants occupied present-day Bamum country,
where they founded several villages before being driven out of most by the
Bamum: those who remained were assimilated. Among the various Bamileke
groups, the Baleng were the first to cross the river Nun, almost certainly at the
beginning of the eighteenth century, and were soon followed by the Bandeng,
Bapi and Bafusam. The various chiefdoms had features common to the political
societies of the Cameroon highlands. The peoples were highly mobile, a
characteristic which appears to be linked to their fairly high population growth
and the inheritance system that transferred an estate to a single heir, thus
obliging the other sons to secure land from the chief or settle elsewhere.
The areas around the Uele and the Ubangi rivers

Knowledge of the areas around the Uele and the Ubangi riven has made little
progress since the works of the first anthropologists. The most vexed question
is that ofthe settlement of these lands. All the research recognizes three types of
population: neolithic, Bantu and Sudanic. The difficulties appear in detailing
the vagaries and patterns of settlement and the successive forms of relations
among these peoples.
Traditionally, two sub-regions were distinguished, the Uele and the Ubangi.
The Ubangi area, today inhabited mostly by Sudanic peoples, was probably
occupied until the seventeenth century by Bantu, who themselves had taken
possession of the region from little-known populations whose neolithic
implements have survived. The Sudanic peoples probably began to move into
the region from Darfur and Kordofän in the seventeenth or early eighteenth
century. Those who were to form the Ngbandi group probably arrived first,
followed by the Banda and the Ngbaka.
The succession of groups in the Uele area, today dominated by the Zande,
Zande-ized people and Mangbetu, seems to have been different. Until the
sixteenth or early seventeenth century the Uele, Mbomu and Aruwimi basins
were probably occupied by neolithic populations, the present-day survivors of
whom are said to be the Momvu, the Logo and the Makere. Two population
groups, traditionally portrayed as invaders, grafted themselves on to this
underlying population. There were first the Sudanic peoples, from whom came
the Mangbetu, Ngbandi and Zande peoples and then the Bantu, who followed
the rivers northwards. It was during this period the seventeenth and early
eighteenth centuries that the ethno-political groups were formed.
Some researchers have proposed a synthesis covering all Ubangi and Uele
areas. Their argument rests on linguistic and ethnolinguistic data and dis¬
tinguishes three broad stages of settlement. Fint, in the first millennium before
the Christian era, there was a large-scale west-east movement bringing the
proto-Ubangians to the River Nile. In the eleventh century there was a Nilotic
drive which broke up the Ngbaka-Sere-Ngbandi group and drove the Ngbandi
southwards into the area between the Ubangi and Zaire rivers where they met
the pygmies and the Bantu. This Nilotic drive also pushed the Zande-Nzakara
populations westward into the lands watered by the Uele and Mbomu rivers.
The third stage, which began in the eighteenth century, corresponds to the
drive by the Bantu from the south, fleeing the effects ofthe slave trade.
This synthesis is attractive, but calls for two basic reservations. Fint, it seems
difficult to speak of migrations when, in best-known cases, those of the Zande
and the Mangbetu, the invaders were élites, content to divide the local popula¬
tions into chiefdoms and kingdoms. Second, all the traditions, particularly the
genealogies of the major clans, show that it was only during the seventeenth
century that 'conglomerations of peoples' (E. de Dampierre) came into being as
a result of various waves of migration. Later developments, probably dating to
the eighteenth century, enabled two clans, the Vungara and the Bandia, to
assert their dominance at the expense of the others. The Vungara went on to
conquer, assimilate or make allies of the most divene populations. It was above
all their skill in politically organizing these populations that cemented Zande
society. As for the Bandia, the high point of their expansion came during the
eighteenth century thanks to Ngubenge who carried through the conquest of
Nzakara country and helped to drive the Vungara eastward.
Little is known of the material civilization of these peoples. It seems likely,
however, that the conquering aristocracies had long engaged in hunting, which
remained a highly valued activity. The Zande learned their farming skills from
the conquered populations. Iron-working was held in high esteem and among
some groups, such as the Ngbandi, smiths were grouped in strictly closed
professional associations. The major river network formed by the Ubangi, the
Uele, the Mbomu, the Aruwimi and their numerous tributaries was the basis
for a busy trade. In the markets, craft products, foodstuffs and animals, together
with slaves, were the main items traded. This trade grew rapidly as the trade
carried on by Europeans on the Adantic coast penetrated inland.
The coastal societies and the Atlantic trade
For the forest lands of central Africa, the period from 1500 to 1800 also saw the
establishment and development of the coastal trade, soon to be reduced to the
slave trade. The coastal fringe was the last link of a long trading chain which
reached ever deeper into the interior. By 1800, this chain had reached the
upper Mungo, the grasslands of Cameroon, the upper Ogowe and, through the
River Congo (Zaire) the River Ubangi.
Peoples
The state in 1500 ofthe populations and societies in the coastal zone remains
obscure. The writings of European navigators and traders show that the whole
region was occupied, but we do not know by whom.
For the coast itself, O. Dapper's Description de l'Afrique, which contains a list
of numbers used by the peoples on the coast, makes it possible to assert that in
about 1660 they were definitely Bantu-speaking. These same written sources
make it possible to date the reigns of several rulers named in oral traditions.
Thus, the height ofthe reign of Mulobe, the third historical king ofthe Duala,
can be put at about 1650. That makes it possible to date to the sixteenth
century the exploits of his grandfather Mbedi, and to beyond the sixteenth
century those of the legendary Mbongo, to whom the Duala and related
peoples refer.
While it is thus possible to accept a Bantu peopling of the coast by the
beginning of the sixteenth century, one is still reduced to conjecture for the
neighbouring regions ofthe interior. For the Minlaaba region and Beti country,
six levels of populations have been established with no chronological data, the
oldest of which was that of the pygmies, followed by hunter-gatherers, then
various Bantu groups whose migrations continued into the nineteenth century.
Further south, in modern Gabon, the picture is clearer. At the end of the
fifteenth century, the Portuguese found Mpongwe populations in the estuary
region. The lands along the lower Ogowe (Cape López and Fernán-Vaz) did
not yet have any Bantu populations; the peoples there were probably pre¬
dominantly Negrillo. The only organized state was in the Gabon estuary.
According to the later description left by Dapper in the seventeenth century,
the king bore the title o£mani pongo.
In the far south, the kingdom of Loango was well established by about 1500,
since all the voyagen' accounts of the early sixteenth century refer to Mani
Loango and Mbanza Loango by name. Some traditions make Loango a state
derived from the kingdom of Kongo; for othen the foundation ofthe kingdom
of Loango resulted from processes internal to Vili society. It is possible that
Loango may have maintained close, perhaps vassalage, relations with Kongo
until the late fifteenth century. As in many states in Central Africa, the king's
main functions were ritual ones. The institution of royalty, founded by a family
of smiths, was linked to the fire cult. During the coronation ceremony, each
maloango lit the ntufia (sacred fire) which was to burn until his death. Another
ritual concerned rain, for which the king was responsible. The most striking
feature of this society was its complex economy. A dynamic food-crop
agriculture blended with diversified craft production which included the
making of palm-cloth for weaving, the working of ivory, copper and redwood
and salt-making. These products were sold on the local markets, notably the
one in Buali, the capital, and also reached markets outside the kingdom. The
Atlantic trade thus did not create trade in this region: it merely gave it a new
intensity and a new dimension.
The Atlantic trade
The local history ofthe Adantic trade in this part of Africa is still little known.
The account books of European traden, many of which have not survived, are
often fragmentary. But some facts have been established.
The trade was initially the monopoly of Portugal, whose navigators Fernäo
do Po and Lope Gonçalves reached the island of Formosa and the Gabon in
1472 and 1473 respectively. Trade only began to increase after 1500 when
Portuguese setded on the island of Sao Tomé, which needed slaves for its own
sugar-cane plantations while also serving as an entrepôt for cargoes destined for
the Americas. At the end of the sixteenth century, slave rebellions, the most
important of which took place on Sao Tomé in 1574 and 1586, ruined
Portuguese trade in the region. The Dutch arrived after 1598 and developed
the new trading centres of Douala and Cape López, and eliminated the
Portuguese from Loango. After 1660, the slave trade underwent a massive
expansion which continued until about 1850. France and Great Britain became
the main trading powen in the region, but the activity of the Portuguese,
Dutch and Danes succeeded in maintaining a competitive situation.
The exact scale of this trade, which varied from place to place along the
coast, is hard to determine. The position of these lands at the bottom of the
Gulf of Guinea constituted a hindrance. The slaving vessels often followed the
'small route', stopping at the various slaving stations along the coast from
Senegambia to Calabar, and by the time they reached Cameroon, Gabon and
Loango, they had already placed much of their cargo. The 'great route' allowed
boats to avoid these intermediate stops and arrive direcdy south of Cape López
and begin trading at Loango or in neighbouring ports.
On the Cameroon coast, trade entered its busiest phase in about 1750, with
the Wouri estuary and, more precisely, Douala, as its main centres, especially
for Dutch ships. Douala got most of its supplies of slaves from the north and
north-east. On the River Gabon, the main trading centre was the estuary
region. Cape López only began to take an active part in the Atlantic trade in
the reign of Reombi-Mpolo (c. 1790-1810). Any estimate of the numbers
involved would be risky, but the slave trade was quite extensive, to judge from
the seriousness of the conflict between the Portuguese and the Dutch over
possession of Gabon at the beginning of the eighteenth century and from the
number of ships sent to this region by the French port of Honfleur. In addition
to slaves, there were large quantities of ivory, wax and sandalwood supplied by
the Kele and Fang hunten and gatheren ofthe neighbouring region.
The Loango trade is the best known. Its trade in slaves lasted a short time
compared with that of the Slave Coast or Angola. It began on a large scale in
about 1670 but by 1780 was almost over. Until the mid-seventeenth century
Loango supplied mainly ivory, dyewoods and palm-cloth which was widely
sought after in Central Africa as a currency. The number of slaves exported by
the Loango coast - 100 a year in about 1639 - rose to 12,500 a year in the
period 1762-78 and to 13,000-14,000 a year between 1780 and 1790. Slaving
captains' account books have left valuable indications of the ethnic origins of
slaves bought by Europeans at Loango. The presence of Quibangues (Bubangi
and 'river people' from the Congo and Ubangi riven) shows that, at its peak,
the Adantic trade had reached the heart ofthe continent.
Complex dynamics
Over such a small region, an analysis ofthe effects of this system of intense and
prolonged trade must be conducted with rigour and care. We shall not enter
into the debate over the destructive or positive effects of the Adantic trade, a
debate which remains open. We shall confine ourselves here to stressing, in
relation to those exchanges, the dynamic processes of all sorts that were occur¬
ring, bringing long-term changes that affected social organization, political
structures, population and attitudes.
One of the fint changes was the introduction, at the beginning of the
seventeenth century, of food crops from the Americas: maize, cassava, ground¬
nuts, beans and tobacco. The spread of these crops did not take place only in
the coastal region: they moved rapidly inland, sometimes along trade routes.
Cassava had great success, as its yield was distincdy higher than that of yams. It
had many uses: its leaves could be eaten, while the root lent itself to several
techniques of processing and preserving. Doubdess more slowly, maize, whose
presence on the coast is recorded before 1600, replaced sorghum. The other
crops were gradually incorporated into the various systems of crop rotation.
These changes testify to the great capacity for technological innovation of the
peasant communities of Central Africa. Did they contribute, through a more

secure and more divenified food supply, to a greater physical resistance among
the populations and a higher rate of population growth? Nothing could be less
certain, for the nutritional value of cassava is limited. Socially, this agricultural
revolution contributed, along with the trade itself, to the establishment of a
new division of labour. Numerous agricultural tasks were gradually abandoned
by men, who preferred to devote themselves to trade, which was more lucra¬
tive, and entrusted to women and slaves. The development of domestic slavery
and other forms of dependence was a direct consequence of these agricultural
changes.
In the regions nearest the coast, in direct touch with oveneas trade, the
transformations were highly complex. In the seventeenth and eighteenth
centuries, the population stabilized and a process of redistribution began which
was to be completed in the nineteenth century. The seventeenth century saw
the occupation of the Gabon estuary by new Mpongwe clans and the eight¬
eenth century the arrival of the Orungu at Cape López and the Nkomi at
Fernán-Vaz. As regards political structures, the coastal trade had contradictory
effects. In only one case, among the Orungu of Cape López, did it lead to the
formation of a strong, centralized state, whose royal clan Abolia or Alombe
held the exclusive monopoly over trade with the Europeans, at the time when
the slave trade was at its height. Everywhere else on the coast, the opposite
mechanism was at work a breakdown of centres of authority and a splintering
of political power. In this case, the organization of trade, on the African side,
was not a state monopoly. Competition benefited a few individuals, princes,
commonen and even some ex-slaves, who formed a new aristocracy keen to
secure political power. In Douala, rivalry crystallized between the Bela family
('King Bell') and the Ngando family ('King Akwa') and ended in a final break
in the nineteenth century. At Loango it was the new men - broken, traders,
caravan leaders and other middlemen who came out best as they had the
means to purchase land from the king and build up an entourage of many free
or servile dependants. The maloango was reduced to selling senior positions to
these nouveaux riches.
Meanwhile the condition of the ordinary people was deteriorating. The

eighteenth century saw the appearance or development of new magical
practices and new conceptions of sickness centred on the individual instead of
on the family group, with the main aim being to ensure protection against
attacks from all sides. This development is no doubt related to a greater degree
of insecurity.
The Congo river axis and large-scale Congolese trade

Until colonization, the River Congo (Zaire) acted as a powerful factor bring¬
ing together the various peoples ofthe vast Congo basin. This role did not only
involve the river itself along its navigable portion above Malebo Pool, but also
its many tributaries, particularly the Alima, the Likwala, the Sangha and the
18.1 Fang mask worn by a member ofthe Ngil society, an association of men organized on
the basis of specific initiations, whose members maintain order in the community and protect it
from evil spells. Height: 10 cm (G Collection Musée de l'Homme, Paris, photo: D. Destable)
Ubangi on the right bank and the Kwa, the Ruka, the Ikelemba and the
Lulonga on the left bank. Since early times a remarkable trading network had
been organized which had moulded most of the human societies living along
the river and its affluents.
An ancient exchange economy

There is much evidence of these ancient trading networks. It remains difficult to
reconstruct the chronological stages and the precise lines of the setdement of
this region, but two main movements can be distinguished. The fint brought
these Bantu-speaking peoples into this region from their original home. The
strong similarities between these languages and certain Nigero-Cameroonian
languages suggests a direct migration from the Nigero-Cameroonian homeland
towards their present location, following the waterways, which form the most
convenient routes in this manhy and heavily forested area. Besides this first
movement, there were also smaller-scale movements which probably occurred
later and were in an eastwest direction, and which gradually led to the
stabilization and separating out of some groups.
These latter movements helped to create the conditions for an exchange
economy by distributing the population over complementary ecological zones
including waterfronts, flood lands and dry lands. Fishing communities could
easily trade their fish, which -was highly valued, for foodstuffs or craft products.
All known layen of ZaireanCongolese archaeology are to be found around
Malebo Pool, thus bearing witness to a highly unusual degree of continuity in
both density of settlement and the variety of human activities. Two sites are
particularly revealing on the subject of trade. In the first, at Kingabwa, on the
right bank of Malebo Pool, several types of pottery have been uncovered, one
of which has also been found at other minor sites in the Kinshasa region and
along the river. This pottery, dated 1450 is evidence of very old-
established commercial relations along the Congo and its tributaries. At the
other site, at Mafamba, many clay pipes have been discovered, most of which
have not been used, which suggests that Mafamba was a production centre or
an entrepôt redistributing the pipes in the region.
Two products seem to have dominated the long-distance trade: camwood,
for cosmetic purposes, and copper mined from the deposits on the lower
Congo. Local trade involved products such as salt, cloth, mats and baskets.
The grafting ofthe Atlantic trade and its consequences

This grafting took place in several stages. In the first, from the late fifteenth
century to about 1560, the kingdom of Kongo played the leading role.
Documents of Portuguese origin dating from 1529 indicate that the Malebo
Pool region was one ofthe main supphen of slaves, through the routes linking
it to Mbanza Kongo (Sào Salvador) and Mpinda. The period 1560-1750 wit¬
nessed a decline in Kongo's trade, due to the growing competition of Angola
and internal problems caused particularly by the Jaga. During the third stage,
from about 1750 to about 1850, the basin of the Congo and its tributaries
became the principal suppliers of slaves to the Americas. A dense network of

caravan routes linked it to a long chain of ports on both sides of its mouth,
Loango, Cabinda, Boma, Ambrizette, Ambriz and Luanda.
Throughout these three stages, Malebo Pool was the meeting-point between
long-distance Congolese trade and the Atlantic trade. The latter's galvanizing
effect on the regional economy was based on mechanisms that are easy to
undentand. The export trade involved high-value goods and this made long¬
distance transport on the Congo and its tributaries profitable. Trade in locally
used goods, therefore, even if transported over long distances, also became
profitable, and this in turn stimulated regional specialization.
The geographical area encompassed by these various levels of trade grew
continuously from about 1500. By about 1690, Malebo Pool was in permanent
trading relations with the lower Kasai and the lands around the River Alima. A
century later, the River Ubangi was fully a part of this trading pattern, as were
all the riven in between, together with their tributaries. Only in the nineteenth
century does the trade frontier appear to have reached the confluence of the
Ubangi and the Uele. Numerous markets existed along these waterways, the
largest being naturally at the biggest junction, Malebo Pool. Each of the four
markets, at Ntamo and Kinshasa on the left bank and Mpila and Mfwa on the
right bank, had 3,000 permanent residents to whom must be added
people passing through, who were especially numerous during the trading
season, which coincided with the dry season, from April or May to September.
The trading system directed towards the outside world rested essentially on
two items slaves and ivory. Slaves, trade in whom really boomed after 1750,
came mainly from four regions: the Lulonga basin; the lands around the River
Alima; Boma country between the confluence of the Congo and the Kwa and
Lake Mai-Ndombe; and the Ubangi basin. Around the main market, Basan-
kusu, in the Lulonga valley, there were numerous villages where the slaves
cultivated the fields until they could be sold. Some of the slaves sold followed
the course of the river to be exported; othen went up the River Ubangi to be
sold to the Loi in exchange for ivory destined for the markets at the Pool and
on the Atlantic coast.
Among the Tio, domestic slavery was so developed that the word mboma
(man of the Boma ethnic group) came to mean a stupid person or anyone in a
lowly occupation. With the development of the slave trade, the number of
mechanisms of enslavement grew. Among the Bobangi, the main trading
people in the Congo basin, a distinction was made between a montonge
(captured slave) and a montamba (slave sold by his kin), which indicates how far
social values had been overturned by the slave trade.
The second export stem, ivory, appean in sixteenth-century Portuguese
texts as one of the most profitable lines. Ivory had long been used locally for
bracelets and hair-pins, but its trade value in the region seems to have remained
low. There were numerous herds of elephants in the forest zone, and the
pygmies almost monopolized elephant-hunting, but the network of rights over
slaughtered animals was so complex that, once the chiefs had taken their share,
litde profit was left for the hunter. Until the late 1700s, ivory took the same
routes and went through the same hands as the slaves who, by value, provided
the bulk of the trade. The relationship between the two items of trade was
reversed only after 1830.
The range of goods received in exchange from the coast, at fint limited to
salt, nzimbu (shellfish) from Luanda, cloth and beads, was widened in the mid-
1600s by the introduction of metal objects, particularly knives, and mirrors. The
eighteenth century saw a dramatic increase in muskets and gunpowder, copper
and tin products and alcoholic beverages. How these items circulated in the
Congo basin is poorly understood and their speed of circulation has been the
subject of speculation since Stanley estimated, in about 1880, that it took an
average of five yean for a European article to move from the coast to the
Ubangi.
The trade in locally-used goods covered the same area, perhaps slighdy
larger, as that in export goods. People often engaged in both trades at once.
This local trade involved mainly foodstuffs and salt. Foodstuffs had the advan¬
tage of an enormous market because, from the eighteenth century onwards, an
ever-rising level of demand had to be met. Cassava, cultivated and processed by
women, was one of the most sought-after items. The oil-palm provided oil for
domestic use and palm wine. Salt was another vitally important item, whether
rock salt from Mbosi country or vegetable salt produced mosdy by peoples
along waterways. These peoples supplied many species offish, fresh or smoked
according to well-developed techniques. Tobacco and local alcohols were also
sold.
Handicrafts also contributed to trade. These were the work of specialists:

among agriculturalists, the women farmed, leaving handicrafts to men; among
fishing communities, however, catching fish was the men's job and handicrafts
the women's. The most sought-after products were: mats; camwood powder,
produced by women; palm-cloth, for which the Tio were famous; iron imple¬
ments; pottery; and finally, canoes of all sizes, 3-20 metres long, or even
longer.
There were thus considerable differences between the two types of trade.
The export trade was based on an economy of destruction and profited only a
tiny minority. The trade in local goods was based on an economy of produc¬
tion and the beneficiaries were the ordinary people, men and women, each
according to how much he produced.
Diverse societies
This internal trade promoted the progressive unification of the various peoples
ofthe Congo basin, despite the obstacles that encouraged variety. Among these
was the way trade was organized. It was a segmented trade and remained so
until the nineteenth century. The river and its tributaries were divided into
spheres of influence, each of which was controlled by one ethnic group. It was
only in the mid-eighteenth century that the Bobangi began to control the
whole Congo-Ubangi axis. Mistrust long ruled relations among the various
partnen and pillaging was not uncommon. Conflicts of interest often led to
full-scale wars, such as that between the Bobangi and the Tio, who reached a
lasting agreement only in the early 1800s.
Among the many faetón making for unification, most important were the
trading techniques themselves. Thus, from the eighteenth century, the whole
region began to adopt the same currencies for transactions palm-cloth,
nzimbu, cowries and copper rods. The widespread practice of credit created
long chains of solidarity throughout the trading area. The profitable conduct of
trade also led to strategies of extended alliances through marriage and, above all,
blood brotherhoods that created numerous obligations between partnen. The
river trade thus served as a crucible, bringing different peoples into contact,
spreading the same customs and beliefs and facilitating the hegemony of Bobangi
as a trade language. This shared civilization, the formation of which speeded up
in the eighteenth century, reached its apogee between 1850 and 1880.
However, this growing unity went hand in hand with a degree of diversity,
linked pardy to the variety of ways in which peoples adapted themselves to
large-scale Congolese trade. Not all engaged in trade. In this region, where
water was plentiful, fishing, one of the oldest occupations of the Bantu peoples
of the Congo basin, remained the main activity of many groups. Fishing
communities were characterized by high mobility, which was determined by
variations in the water level and the regular movements of fishing banks.
Depending on the season, people would live in fixed villages or in fishing
camps built during low water and abandoned as the level of the river rose.
Socially and politically there was a distinct lack of congruence between political
and economic units. The basic political unit was still the village: the chiefs
authority, essentially ritual and religious, operated solely within this limited
framework. The activities of teams of fishermen thus escaped his control. But
the basic economic unit was precisely the loosely structured fishing team based
on the initiative of a renowned fisherman and including some of his kin and
other volunteers. In this way, the family cells on which the villages were based
were dislocated to meet the requirements of fishing.
Those who specialized in trading activities evolved differently, as is shown
by the contrasting evolution of the Bobangi and the Tio. The details of the
Bobangi's long migration from the middle or lower Ubangi to Malebo Pool are
hard to reconstruct. What is certain is that what was originally a society of
fishermen rapidly switched to trade, which became the sole foundation of all
social life. Henceforth, there was complete convergence between the political
and the economic unit. Bobangi society was based on trading firms belonging
to wealthy traden. It was an open society, with much social mobility: anyone
active and successful in trade could acquire wealth, which was the means to
increase the numbers of his dependants, to gain power and to lay claim to
political tides.
The Tio had the good fortune to live at the biggest crossroads and main
break-of-bulk point in the Congolese trading network. Trade had not, appar-
endy, upset the political structures ofthe kingdom. It was widely decentralized.
The underlying structure consisted of relatively autonomous chiefdoms. The

chief had major ritual prerogatives but limited political power. The same
situation occurred at the higher level. The king, ôkoo (deformed by the Euro¬
peans into makoko), had above all a spiritual role. Temporal power belonged to
a few high notables such as the ngaailiino, responsible for collecting tribute and
heading the administration, and the muidzu, the country's chiefjustice. Unlike
the chiefship, the dignity oî ôkoo was not hereditary: the king was elected from
among the leading chiefs in the kingdom by his peers. This old structure
retained its essential features until the advent of colonization. It -was at the social
level that the most lasting changes occurred. There was fint an increasing
dissociation between the riverine peoples, who monopolized trading activities,
and the Tio in the interior, who specialized in agricultural production (cassava
and tobacco) for the river market. Among the riverine peoples, those who
profited most from trade were the chiefs, who exploited to the limit their right
to sell human beings and ivory. Convenely, the chiefs and tide holden estab¬
lished away from the Pool, such as the king who resided at Mbe, participated
only fitfully in trade by sending their dependants there. They held political
power but their economic power was less than that ofthe riverine peoples.
These Bobangi and Tio trading economies did not become capitalist, despite
the large profits made. It was impossible to buy land or a labour force, except
in the form of slaves. Money, accepted in some dealings, was not a univenal
equivalent which could purchase all goods. Some goods and some social and
religious services, such as fines and dowry payments, never came within the
money economy.
The picture thus drawn of the peoples and societies of Central Africa clearly
demonstrates their divenity as well as their aptitude for innovation, even in
constraining situations. It also shows, through the gaps in knowledge, the
directions in which researchen should move: large-scale research with ample
resources, directed principally towards collecting new materials through
archaeology and oral traditions, and detailed theoretical modelling in the study
of social structures, cultural evolution and the history of mentalities.
The Kongo kingdom
and its neighbours
The western part of Central Africa, to the south of the equatorial forests, is
inhabited by peoples who speak dialects of the Kongo language and of other
closely related languages. That unity is strengthened by a profound cultural
unity. This ethnolinguistic group occupies a territory stretching from southern
Gabon to the Benguela plateau and from the Atlantic Ocean to well beyond
the River Kwango.
The history of this region is well documented from 1500 onwards. Over the
past hundred yean, texts and guidebooks have been published in many editions,
and a historiographical school has been developing since the seventeenth
century.
The historical tendency in these areas in the sixteenth and seventeenth
centuries was very different from that of later times. Territories of great size
were organized, using political structures, to form states. After about 1665, how¬
ever, these territories were reorganized on an even greater scale, the organizing
principle now being derived from the imperatives of an economic structure
which was the product of an intensive slave trade. In our approach to the early
centuries, therefore, we shall give priority to the history ofthe kingdoms; and
then, as the dynamics of trade brought the kingdoms low, trade will be the
focus of attention.
The landscape of these regions is determined by the relief and the rainfall.
The mountainous nature of the terrain explains why a population in search of
better habitats was unevenly distributed. The most favourable region lay to the
north of the Zaire/Congo river, from the coast to the area called Mayombe.
Here there were also worthwhile mineral deposits. This was where the two
largest states ofthe coast, the kingdoms of Kongo and Loango, came into being.
Since 400 yean before the Christian era at least, farmen speaking western
Bantu languages had been setded to the north and south of the lower Zaire,
where they produced yams, vegetables and oil palms. From the second to the
fifth century, this population was augmented by the arrival from the east of
people speaking eastern Bantu languages. These people grew grain and, where
the tsetse fly allowed, kept herds of catde. Lasdy, perhaps during the sixth
century, banana-growing was introduced to round out the production pattern.
273
Thereafter, the socio-political organizations became more complex, and

chiefdoms were formed between the ocean and the Zaire river upstream from
Malebo Pool. It was in the region called Mayombe that the regional division of
labour developed most effectively. By about 1500, the coastal dwellen were
supplying salt and fish and had transformed the coastal plain of Loango, towards
the estuary of the Zaire, into a vast palm-grove producing palm-oil. Further
north, near the edges of the great forest, raffia palm was cultivated and used to
make fabric. Lasdy, in the same area and further into the forest, red dyewood
was exchanged for products of the savannah. This was the birthplace of the
Kongo civilization.
The Kongo kingdom began in the Vungu chiefdom north of the Zaire river.
Nimi Lukeni, the founder of the kingdom of Kongo, founded the town of
Mbanza Kongo where Sâo Salvador now stands. His kingdom was constituted
partly by alliance with the local chief, the kabunga, and with a king who ruled
the Mbata in the Inkisi valley, and pardy by the conquest of other lands
towards the sea and the lower Inkisi valley. The date of the kingdom's
founding is unknown, but it was possibly some time between 1300 and 1400.
The Tio kingdom was probably founded earlier. The kingdom of Loango,
which developed not far from Vungu to the north of the river, is thought by
some scholars to have reached its zenith in the sixteenth century, but this
hypothesis is disputed. To the south of Kongo, the state of Ndongo, whose
king bore the tide ngola, which was deformed into Angola, was taking shape
about 1500.
A hegemony: Kongo
In the fifteenth and sixteenth centuries a single state, Kongo, held sway over
the entire region between the Benguela plateau and the Bateke plateaux and
from the sea to beyond the River Kwango. Around 1500, the borden ofthe
state proper followed the River Zaire from its mouth upstream to the
confluence with the Inkisi, in places extending northward beyond the river. It
included the Inkisi basin and all the lands to the south as far as the Loje. It had
a population of between 2 million and 5 million we do not know the precise
figure. The areas in Kongo's sphere of influence at least equalled the kingdom
in population. The population of the Tio kingdom must have been very spane
except around Malebo Pool.
The kingdom was divided between a large town, the capital Mbanza Kongo,
and the countryside. Three well-defined social strata co-existed: the nobility, the
villagers and the slaves. The nobility formed the structural framework of the
kingdom, and lived in towns except when required to serve as provincial gover-
non. The upper nobility was composed of relatives of the king or one of his
predecesson. In relation to the villages, the nobility made up a group. Matri-
hneality determined access to the land, residence and succession to the headship
of the village. The king appointed his close relatives to key posts. Royalty was
elective; the royal council was made up of 12 memben, including four women
The Kongo kingdom and its neighbours 275
19.1 Kongo and its neighbours in the sixteenth century (afterJ. Vansina)
who, according to A. Hilton, represented the clans of the king's grandparents.

Struggles for succession were commonplace. After a struggle between the lords
of the north and those of the capital and south, Nzinga Mbemba, otherwise
known as Afonso I, succeeded his father Nzinga Nkuwu in 1506.
The kings, much given to polygamy, had many children, and their house
quickly reached huge proportions. Kongo was highly centralized. The king
appointed the territorial governon except the governor of Mbata and, after
1491, the governor of Soyo. He dismissed governon and other officials as he
saw fit. The nobility had no hereditary functions and at every generation its
hierarchy was redefined in relation to the new king. Centralization was
favoured by the existence of a currency controlled by the king. It consisted of
shells called nzimbu, taken from the fishing-grounds of the island of Luanda,
which was part of the kingdom. There was also military centralization. In the
late sixteenth century, the royal guard, composed of slaves, was 16,000
strong and was the sole standing armed force of the kingdom. In time of war
abroad, the peasants were called up in territorial units.
The ideology of royalty, called nkisi, derived from religious conceptions in
which three important cults played a role. These were ancestor worship, for
which the holy place was the royal cemetery grove, the wonhip of the spirits
of the land and the worship of royal charms. The nkisi concept was
fundamental, and the Christians were to adopt the term, using it to mean 'that
which is sacred', for example calling the church the house ofthe nkisi. Diseases
and misfortunes were attributed to the ndoki (sorcerer) whom nganga ngombo
(diviners) were able to identify, sometimes by ordeal. Royalty was sacred. The
king was addressed as Nzambi mptingu (Supreme Creator). By his tumba
(blessing) the king expressed his protection of royal officials and, in general,
guaranteed fertility, principally through his control over rainfall. The king
himself was, in a real and profound sense, the nkisi ofthe country.
Structures in the neighbouring kingdoms were different, but the ideology
was almost identical. These kingdoms were less centralized. In Loango, the
noble houses had not supplanted matrilineal groupings, and this class was much
less cohesive than in Kongo. Ndongo had no territorial structure above the
petty chiefdom level, and the military system was intended to make good the
deficiency.
Afonso I's victory ushered in the longest reign in Kongo's history, from 1506
to 1543. The role played by this king was crucial. He opened up the country to
Portugal, thus setting in train a vast economic and political reorganization and a
deliberate assimilation of features of Christianity which was to prove perma¬
nent. On becoming king, this factional chief rapidly made the Catholic Church
the state religion under the leadenhip, from 1518 to 1536, of his son Henrique,
who was consecrated bishop in Rome. The bishopric was later to fall under the
control of Portugal. The slave trade developed from 1514 onwards. The king
fint tried to control it by establishing royal monopolies and then, in 1526, to
abolish it. This attempt failed, and even the royal monopolies were constandy
breached by the Afro-Portuguese of Sao Tomé, by the kingdom's neighbours
and even in Luanda, which was an integral part ofthe kingdom. The king used
his revenue from the slave trade, the ivory trade and the trade in raffia fabric to
attract Portuguese technicians and, above all, missionaries. In his reign, social
and political life was transformed. The gulf between the nobility and the
ordinary people widened: the former became literate, espoused Christianity and
took part in the slave trade, while the ordinary people were harshly exploited.
All subsequent kings were descendants of Afonso through one or other of
his three principal daughten. The ever-increasing number of claimants to the
throne led to a scission ofthe royal house into enemy factions and, after 1665,
to a civil war. The presence of the Portuguese in the town introduced a new
political dimension. Connected by marriage to various noble houses, they were
split up into Afro-Portuguese and people from the mother country, and they
led the opposing parties at court until 1 665 and took part in all the struggles for
succession.
International trade, which had been very minor before 1506, developed after
it started dealing in slaves. Exports totalled some 4,000-5,000 slaves yearly until
about 1540, and 6,000 thereafter. Imports from Malebo Pool must have
been considerable, for many slaves were recruited at Mbanza Kongo not only
for the royal guard but also for the food plantations which fed the capital, and
other slaves were assigned to the port of Mpinda and the retinue of royal
nobles. In addition to this supervised trade, a clandestine slave trade carried on
by the inhabitants of Sao Tomé developed on the northern and southern
borden ofthe kingdom, especially after 1526, when exports from the kingdom
of Benin came to an end.
Portugal took a keen interest in the mineral products of the country and
wanted to control them. The king of Kongo, however, was determined to
keep the mining of Bembe copper and the working of Mbanza Kongo iron
under tight control. The Portuguese sovereign, for his part, never allowed the
sale of ships to Kongo and opposed its attempts to lay on its own transport to
Sao Tomé or Europe. The Portuguese kept control of trade and made it an
unequal exchange. The Kongo court used a substantial part of the revenue to
import fabrics, wine and luxury items which quickly became an ostentatious
necessity.
Under Afonso I's successors, Kongo's relative power diminished, especially
in relation to Ndongo, which was gaining strength mainly through the clandes¬
tine traffic with Sao Tomé. In 1561, Kongo cut itself off from Portugal almost
completely, but the successive deaths of two kings during a war with the Tio of
Malebo Pool in 1566 and 1567 led to a disarray which, with the invasion of
warriors called Jaga from the east, turned into a catastrophe.
Three states: 1575-1640
The Jaga routed the royal forces and the court had to take refuge on an island
in the lower Zaire. Numerous refugees were sold as slaves by the Jaga to the
people of Sao Tomé. The king had to appeal to Portugal, which sent an
expeditionary force, and the country was reconquered between 1571 and 1573,
although the troops were not withdrawn before 1575 or 1576. The hegemony
of Kongo over the region was destroyed, for the colony of Angola was founded
in 1575 and the Portuguese, in large numbers, came to Loango to trade from
that year onwards.
The identity of those who invaded Kongo has never been established. The
name Jaga is used in the sources as a synonym for barbarian. The invasion must
be associated with the KongoTio wars and probably with the cessation of the
slave trade in 1561. It is thought that the ranks ofthe invaders were swollen by
a great many villagen who were weary of being exploited by the nobility.
Paulo Dias de Novaes, after a stay at the court of Ndongo, worked to such
purpose that he obtained a contract for conquest and colonization from the
Portuguese court. The colony was to be named Angola, from the title of the
king of Ndongo. Dias founded Luanda in 1576 and began trading in slaves,
endeavouring to supplant a community of Afro-Portuguese from Sao Tomé
who were plying the trade there. This community yielded the market to him
and moved to the ngola's court. In 1579, however, pressure from the mother
country forced Dias to terminate his contract. When the king of Ndongo was
told of this, he had all the Portuguese at his court massacred as a preventive
measure, thus furnishing the pretext for a war which was to last for almost a
century, until 1671. At fint, the fortunes of war fluctuated, and the Portuguese
managed - with difficulty - to erect a few fortifications inland. In 1612,
however, the Portuguese entered into an alliance with some Mbangala (called
Jaga) and, with their aid, and especially that of the Jaga Kasanje, they occupied
a large part ofthe kingdom from 1617 to 1621. The king fled to the east ofthe
country and the allies sacked the region. A peace treaty was drawn up in
1622 the ngola being represented by his sister Nzinga Mbande, who was
baptized and named Anna at Luanda. Nzinga became the leader of an anti-
Portuguese party; at the beginning of 1624 the king died and Nzinga became
regent, then queen in 1626. War broke out again against her, but also against
Kasanje who was still occupying the best part of Ndongo. In the course of
operations, Kasanje found a base in the Kwango valley, where he built the
Imbangala state during the 1630s, while Nzinga conquered the kingdom of
Matamba, where she established a formidable base of opposition to the Portu¬
guese regime. When the Dutch occupied Luanda, she was still campaigning
and she formed an alliance with them.
The colony of Angola, ruled by a governor appointed by Lisbon, was

divided into the town of Luanda and a few presidios. The administrative
structure was feudal, its legal basis being contracts of vassalage which linked the
soba (petty chiefs) to the Portuguese court. The revenues of the state and of its
representatives were derived from feudal tributes, all of which, whether due in
goods or in labour, were payable in slaves. This system was not replaced until
the eighteenth century, with the introduction of a poll-tax and the reduction of
the ties of vassalage to a legal instrument justifying the sovereign rights of
Portugal.
¿3ÊS-
19.2 Court ofthe King of Loango, 1668 (O. Dapper, Description de l'Afrique,
Wolfgang, Waesberge, Boom and van Someren, Amsterdam, 1686, © and photo: Dapper
Foundation, Paris)
Meanwhile Kongo had recovered. Despite its loss of hegemony and the
gradual changes taking place in the structure of the nobility, the kingdom
remained sound and even spread eastwards. Kongo successfully resisted other
Portuguese attempts at control by forming a diplomatic alliance with the Vati¬
can and by pitting Portugal against Spain and even both against the Nether¬
lands. The most serious loss was that ofthe maritime province of Soyo, with
the port of Mpinda. After 1636, this province became independent.
Loango had become a great power even before 1600. Its territory extended
from the Fernán Vaz lagoon to the south of Pointe-Noire. Its cultural
influence, spread by trade, extended even further afield; but we know almost
nothing of its political development before 1700.
Loango traded in ivory, hides, red dyewood and raffia fabrics, but exported
relatively few slaves. From 1600 onwards, the Dutch frequented Loango and
Soyo. Dutch ships had better goods to offer at lower prices. They bought
copper and, above all, had guns and gunpowder for sale. The upshot was that
the traders on the Loango coast, the Vili, who organized the caravan traffic to
the interior, extended their networks at an astonishing rate.
At this period, the Vili network was in competition with the existing net¬
works, not only for the sale of ivory or fabrics but also for the slave trade. The
Kongo and Angola network was disrupted by the founding of Luanda, which,
from the start, exported more slaves than Mpinda; the first exports varied
between 12,000 and 13,000 slaves a year, mainly prisoners of war. After the
fint few years, more and more slaves came from purchases inland, either in
Malebo Pool or in the market of Ocanga on the River Kwango. In the 1630s,
slaves began arriving in Luanda from Matamba and Cassange, the capital of
Kasanje.
These developments affected Kongo adversely, for its capital was no longer
the mandatory entrepôt for goods and slaves. Furthermore, its currency greatly
depreciated, the Portuguese having imported other shells. By 1619 the nzimbu
had lost two-thirds of its value and the king two-thirds of his revenue. The
king, however, managed to redress the situation and the currency gradually
regained its former value. The revenues of the king and the nobility must, in
fact, have suffered from the diversion of trade to Luanda and Loango.
This period saw the introduction first of maize cultivation (between 1548
and 1583) and, soon after 1600, that of manioc, tobacco and probably beans,
groundnuts and other New World plants. Pigs are mentioned from 1583
onwards, and these were probably imported animals. They were to spread
mainly in Kongo and central Angola. The new crops transformed agriculture in
the seventeenth century, especially after 1650, increasing yields and providing,
with manioc, a diet better suited to occasional drought and war conditions.
Nutritional standards generally improved. As early as 1560, however, reference
is made to smallpox, an epidemic introduced from Europe, which was to
become a recurrent scourge.
The fint half of the seventeenth century thus witnessed changes in nutrition
and health conditions and the introduction of the slave trade. All this
inevitably set up new population trends, the details of which are unknown. In
Angola in particular, however, the population must have declined.
Towards a new order: 1641
In 1641, the Dutch captured Luanda and occupied a vast part ofthe colony of
Angola before a fleet fitted out in Brazil drove them out in 1648. From then
on, the Brazilians dominated Angola's trade, totally until 1730, and partially
thereafter. Also in 1641, Garcia II became king of Kongo and, like Nzinga,
formed an alliance with the Dutch. The Restoration left them face-to-face
with a great Portuguese army with which they had to come to terms. Kasanje,
which had stayed out of the wars, signed a treaty of friendship and became an
ally of Portugal for more than a century. Disagreements over the treaty
ultimately led the Angolans to invade Kongo. Antonio I of Kongo assembled
all his nobles and was defeated in 1665 at the batde of Mbwila. He perished in
the batde together with many nobles. But a second Portuguese army was so
disastrously beaten at Soyo in 1670 that Angola gave up its attempts to conquer
Kongo. In 1671, the last vestige of Ndongo was conquered, and by about
1680, peace had been imposed on Matamba (Nzinga), Kasanje and the chiefs to
the south ofthe middle Cuanza. The conquest was complete.
The kingdom of Kongo did not recover from Mbwila. A disputed royal
succession led to a civil war between noble houses, which resulted in Mbanza
Kongo being destroyed in 1666 and totally abandoned in 1678. Since the town
was the pivot of the political system, everything collapsed. When a single king
at last returned to Mbanza Kongo in 1709, his kingdom was no more than a
collection of principalities often still rent by internal struggles between claim¬
ants. Everything had fallen apart except Soyo, and even Soyo had the greatest
difficulty in remaining united. The nobility, having deserted Mbanza Kongo,
had to adapt itself to life in the countryside and contended there for the posts of
command. A sizeable number of nobles crossed the river northward to found
small principalities in Mayombe and between the riven Zaire and Niari.
The upheaval was total and demolished the very foundations of society, so
much so that prophetesses began to appear. In 1704, Dona Beatrice Kimpa Vita
began preaching a refurbished Christianity called Antonianism. She rejected the
missionaries and the whites, but above all she exhorted rival claimants to the
throne to give up the struggle, restore a king and repopulate Mbanza Kongo.
She had a large popular following and persuaded a claimant to settle in Mbanza
Kongo, where she crowned him with the black Antonian crown. But in 1706,
she was captured and burnt as a heretic. Pedro II restored the kingdom and
repopulated Mbanza Kongo, but a 40-year tide of political evolution could not
be turned back.
With Kongo, the organization of a vast area disappeared. The structural

framework of this area, like that of Angola, was henceforth to be economic,
with a framework of trade routes articulated by places of transshipment.
Angola, which had expanded towards 1680, was also weakened. Neither the
governon nor the Cámara of Luanda could maintain their ascendancy over the
quimbares and ovimbali (the Afro-Portuguese), or over Brazilian traders.
In Loango, the state still appeared to be strong, but when the king died in
1701, it took more than a year to elect a new king. Royalty, for all its religious
prestige, would begin to be redefined with regard to its authority over trade.
On the periphery of the slave-trading areas, states strengthened themselves
through a growing trade which they controlled, or at least remained sound for
such time as they controlled it. Kasanje had become a great power on the
Kwango river because it controlled an ever-increasing flow of slaves after 1648.
In the north, the Tio kingdom held its ground, apparendy without great
changes.
The Dutch conquest and the Restoration had substantial economic
consequences. Currencies - raffia cloth and nzimbu - depreciated. In Angola,
the slave trade was reorganized by the Brazilians, who supplied capital, ships
and European goods and acted in concert with the organizers of caravans and
the Afro-Portuguese slaven.
On the Loango coast, the French and English, with an insatiable demand for
slaves making itself felt in their West Indian colonies, entered into hard-driving
competition with the Dutch after 1660-65. Here, companies financed in
Europe carried on the triangular trade. So began the large-scale slave trade
which was to reach its full development in the eighteenth century.
Social and cultural transformations: the sixteenth and
seventeenth centuries
We are best informed about Kongo. The division of society into three social
strata, the mwisikongo (nobles), the babuta (sing, mubata) (rustics) and the babika
(sing, mubika) (slaves) was to persist until some time between 1666 and 1678,
the shift occurring fint among the nobility and then in the countryside.
The term ekanda (root: kanda) today means 'matri-clan' or 'matrilineage'. At
that time it meant family matrilineal, to be sure but also any community. The
people of a village made up an ekanda, as did the Christian community. The
village was thought of as belonging to a matrilineal kinship group, descended
from its founder, who was represented by the village leader, the nkuluntti (the
old one). In fact, since women followed their husbands and their sons did not
always go to live in their uncle's village, the village group was bound together
by a territorial tie. The village held the rights to the land, and the spirit that
dwelt in that land was perpetuated by the village kitomi (former lord of the
land). Ideologically, these rights belonged to the matrilineage of the founder,
but in practice the village acted as a body. This can be seen in particular in the
existence of associations such as khimba, nzo longo or kimpasi, all of which were
initiation cults for boys or healing cults which differed from one area to the
next.
In the course of time, the power of the matrilineage and the village
diminished. The villagers were already being hanhly exploited by 1525. They
may have revolted with the Jaga in 1568, and rebellions took place during the
reign of Garcia II. At that time the kimpasi, a cult whose aim was to banish
mpasi (suffering, poverty, calamity or affliction) was very active during periods
of oppression or natural disasten. The villagen' lot became more and more
precarious as the nobles increasingly intervened to demand concubines and set
the sons they bore them to keep the villages under surveillance.
The noble houses, taking the royal house as their model, grew in size and
importance. They absorbed a great many slaves. Their heads had large harems
of wives from other noble or rural families. They paid high bride-prices for
their principal wives, thus securing for themselves the ownership ofthe future
children. The bride-price went to the wife; in this way the noble ladies grew in
economic and political importance. The senior woman of a house by bilateral
descent was its standard-bearer. Yet the position of patrilineal descendants
sons grew in importance. Moveable property acquired by trade belonged to
the sons, who, after about 1550, no longer took their father's name but a
Portuguese surname symbolizing their house. The inheritances, however, were
less substantial than might be thought, as everything acquired through the
performance of public functions reverted to the king after the death or dismissal
ofthe incumbent.
After 1666, the houses underwent two different series of changes. In Soyo
the reigning house became increasingly patrilineal, neglecting nephews in
favour of sons even sons by female slaves. Eventually, in the nineteenth
century, the whole of Soyo adopted the patrilineal ideology. In Kongo, on the
other hand, the nobles, having lost their town, had to make a new place for
themselves in the countryside. The great houses melted away like snow in
summer, to be replaced around 1700 by the formation of large matrilineal
groups, the mvila (clans). Government had been decentralized and ruralized.
Slaves initially war captives did not remain a single stratum for long. A
distinction came into being between saleable slaves and domestic slaves who
could no longer be sold. The latter were attached to noble houses, and some
groups such as the royal slaves and mission-station slaves became
independent, locally powerful groups, and so remained up to the nineteenth
century. Since the saleable slaves were exported, they never formed cohesive
groups. In time, Kongo's three strata thus became compressed into two the
nobility and their exploited subjects.
In Angola, the indigenous social structure comprised the same social strata as
in Kongo, with a Portuguese stratum at the top. But the situation was compli¬
cated by the emergence of an Afro-Portuguese stratum made up of mulatto
traden, who were, culturally speaking, as much Angolan or Kongolese as
Portuguese. This group first formed in Sao Tomé through miscegenation with
Kongo nobles, then hived off to the Kongo capital and to Luanda. By about
1680, there were two large groups in existence, one around the fortress of
Ambaka, the other founding Cacunda Velha. In the eighteenth century, the
Afro-Portuguese were to spread most widely on the Benguela plateau. They
represented a floating population of caravaneen and traden, who worked in
concert with African chiefs and married into the local nobility.
The history of religions and ideologies is marked on the surface by con¬
version to Catholicism, which, in Kongo, spread fint among the urban nobility
and in the chief towns of the provinces. The ecclesiastical structure remained
chiefly Portuguese until 1645, when the Italian Capuchin missionaries began an
intense campaign of convenion to Christianity which continued in Kongo
until about 1700. In Angola, the Catholicism imposed by the conquerors did
not progress beyond the colony except in Matamba. It did not spread in
Loango, despite the conversion of a king in 1663.
The development of religious ideas and practices in Kongo shows that the
Christian doctrine influenced the old religion and coexisted with it. The
Christian terminology stems from the domain of the nkisi, the ndoki and the
nganga, the nkisi becoming 'that which is sacred' or 'grace'. The missionaries
fought against the rites of the kitomi and of witchcraft, ancestor wonhip and
healing associations, but tolerated the medicine practised by the nganga. The
Capuchins used processions and prayers to replace fertility rites, and practised
exorcism.
From the sixteenth century onwards there can be said to have been a single
religion in which, at least among the nobles, features of Christianity and features
of the old religion had merged. The main spread of this religion took place in
the seventeenth century. Garcia II was the defender of both Catholicism and the
kitomi. He was even nicknamed 'the sorcerer'. This new religion was the
source of Haitian voodoo.
Various religious movements have been recorded since the 1630s, when the
fint attempt was made to establish an indigenous church. In 1704, Dona
Beatrice preached a far-reaching reform which cannot be explained solely in
terms ofthe political situation. She claimed to be in direct contact with heaven
and was regarded as a mnnaki (prophetess). She called for thorough Africaniza¬
tion. The Holy Family was black and hailed from Mbanza Kongo, and the
symbols she used were evocative local symbols associated with water, the soil
and vegetation. But she recognized the Pope, despite her desire to drive out
the white missionaries who falsified revelation. In short, her vision was even
more syncretic than the religion practised in her time.
It has been said that after 1700 Kongo lost its Catholicism. This is false, but
the Christian missionaries of the late nineteenth century, steeped in the atmos¬
phere of colonialism, failed to recognize the indigenous Catholicism which had
become an integral part of Kongo religion.
The arts express culture. While nothing survives in Kongo of the per¬
forming arts, such as music, dance and oral art, visual art objects have survived,
together with many texts on the subject. These objects, created to symbolize
class differences (the art of costume for example), political concepts (emblems
and ceremonial objects) and religious concepts (Christian art, objects used in
divination, statues of ancestors and spirits and masks) were commissioned by
the court, the Catholic Church, the leaders of villages and of village cults and
even heads of household.
What is striking in Kongo art is the continuity of stylistic elements, allied to

a dynamism of form that assimilates the many European influences. Examples
include geometrical decorations criss-crossing at acute angles on a royal
emblem (an ivory horn) from before 1533, fabrics from about 1650 to 1800
and basketwork, tattoos and mats from the nineteenth and even the twentieth
century. European influence was strong and was reflected in the importation of
European stone architecture (churches and palaces), symbols of authority
(swords, crowns, flags and vestments) and, above all, religious objects. Almost
all characteristics of Kongo art, especially statuary and the graphic arts, have
been attributed to Christian influence. But the manifestations of ancient graphic
art rock drawings surviving from different periods have hitherto been
ignored. In the seventeenth century, we find alongside stylized drawings a series
of pictograms which are clearly derived from writing and an expression of
popular culture. This use of pictograms was to continue and merge with the
production of geometrical decoration. Alongside these drawings, a more varied
figurative tradition was still practised: witness the narrative bas-reliefs and wall
paintings ofthe eighteenth and nineteenth centuries.
Sculpture is represented fint by many Christian objects. They contain

elements of the European canon of proportions but the stylization allied to
Kongo realism was there from the beginning.
Artists used the most varied materials - copper, ivory, wood, stone, ceramics,
fibres and fabrics and worked on the most diverse objects, including even
cooking utensils. The very fint chronicles extol the merits of Kongo raffia
fabrics, embroidered or worked in velvet. Even the clergy used them to make
priesdy robes to supplement those which they imported, from Italy in
particular. The sculpting of cult objects used in the old religion -was forbidden
and works were destroyed at various times, especially under Afonso I and
Garcia II, and by Kimpa Vita. However, the demand for these objects was
constandy renewed.
The influence of Kongo art extended north and east during the whole of
this period. The spread ofthe so-called 'white' masks from Gabon derives from
northern Kongo masks and probably accompanied the expansion of Loango,
especially in the seventeenth century. The technique of making raffia velvet
spread along the Okango route to the east and was the forerunner of the
velvets of Kasai, and especially of Kuba.
The eighteenth century: the northern areas

In the eighteenth century, the whole of Central Africa was reorganized.
Trading networks not states became the dominant factor. A merchant class
replaced or complemented the old ruling classes. There were two slaving net¬
works: one, Portuguese, traded from Angola to Brazil and the other, in the
north, was tied in with the other European powers; but the overland trade
remained firmly in African hands.
This trade in the northern areas, that of the Loango coast, was fuelled by
companies which financed the entire triangle of trade. Competition among the
different European firms was an even greater factor in driving up prices than
the increased demand for slaves in the West Indies and North America.
Demand increased steadily from about 1665 to 1755, attaining its peak between
1755 and about 1797, when the great European wars brought it to a temporary
halt.
The ports used were fint Loango and then Malemba, Cabinda and, after
1800, Boma. The European traders rented warehouse buildings in the ports,
where they exchanged their merchandise for slaves through the agency of
brokers, the mercadores. Broken were indispensable, for monetary systems were
different and it was necessary to agree on exchange values. To that end, a unit
of goods - the parcel - was established, corresponding to a unit value of slaves.
The parcel was made up of three components: guns and gunpowder, fabrics
and miscellaneous goods, especially metal goods. In addition to the purchase
price, the European merchant had to pay trade duties and make gifts to the
local king and the notable in charge of European trade. The broker, for his
part, earned a large commission on the selling price ofthe slaves.
About 1 million slaves and small quantities of ivory, copper and wax were
exported during the eighteenth century. Imports consisted mainly of fabrics and
weapons. It is estimated that 50,000 guns a year were imported during the
second half of the century.
The caravans were led by guides who were skilled in negotiating rights of
passage and had an expert knowledge ofthe big inland markets. They sometimes
travelled very long distances, taking with them not only European products but
also salt, salted and dried fish and locally manufactured items, especially jewellery
from Loango. On the way they were able to buy other local products which
they sold elsewhere.
The principal route linked the ports to Malebo Pool, where the slaves
shipped along the river and its tributaries arrived in large numbers. The Bobangi
gave up fishing for that commerce, and their trade network stretched from the
Ubangi river to the Kwa. These waterborne carrien also conducted an extensive
trade in other products pottery, fish, mats, food, beer, red dyewood and other
commodities bought in one place and sold in another; this encouraged different
areas to specialize in different goods. The slave-trading zone had reached the
Ubangi, the lower Tshuapa and the areas around Lake Mayi Ndombe well
before 1800 and continued to expand as the demand for slaves grew. Other
caravan routes travened Kongo, reaching Matamba and Cassange. After about
1750, the capital ofthe new Yaka state ofthe Kwango river became a slave-
trading centre into which thronged the slaves taken in the constant military
campaigns waged by the Yaka beyond the Kwango.
The northern slave trade was so lucrative compared with that of Angola that
it pushed the latter's field of operations towards Luanda and Cassange. This led
to Portuguese military action against the port of Mosul (later Ambriz) and to
the occupation of Encoje in central Kongo in 1759 and subsequently, in 1783,
of Cabinda. The Portuguese were driven out by a French squadron supported
by local forces.
The demographic effects of the slave trade are still largely unknown. The
population fell in Kongo and Mayombe, although we do not know by how
much. Elsewhere, the main result was to slow down population growth.
We know more about the economic effects. To the north of the Zaire river
a vast area organized itself into an aggregate of regions, each with its own
speciality: raffia, tobacco, cane-sugar wine, ivory, foodstuffs, ironware, pottery,
boats and so on. This led to technological impoverishment in each individual
region, as some activities were given up, but the economic vitality of the
aggregate increased steadily, owing mainly to the development of waterborne
transport.
The social and political effects were far-reaching. The brokers and notables
in charge of the slave trade steadily gained in importance at the courts of
Loango, Ngoyo and Kakongo, at the expense of the old nobility linked to the
royal house. Royal authority suffered accordingly. In Loango, the royal council
witnessed the replacement of territorial nobles of the royal house by these new
nobles, the titles being sold by the king to the highest bidder. In or before
1750, the royal line became extinct and six factions contended for the throne,
but the royal kitomi eventually succeeded in imposing a neutral king at the cost
of gravely weakening his power. Moreover, distant provinces such as Mayumba
began seceding from Loango from 1750 on.
The Tio kingdom was also in difficulties. The regional great lords trading on
their own account had grown rich and adopted a new ideology legitimizing
their secession. This power was spiritual, centred upon a talisman, the nkobi,
and could dispense with the support of territorial spirits. From about 1780
onwards the Tio state was in the grip of civil war and did not recover until
about 1830.
In the middle Zaire basin, the expansion of the slave trade led to renewed
small-scale population movements and hostilities, both in the Mayi Ndombe
area and further north, while large centres of population began to take shape
beside the river at points where it could be crossed. These centres became
market towns in the nineteenth century.
In Kongo itself, Christian ideology, and especially the Order of Christ to
which the rulers belonged, remained the basis ofthe ruling ideology. Its centres
were the king, who was now no more than a supreme nkisi, and the former
mission stations, run by former mission slaves. Territorial units broke up again
and again, sometimes even at village level. At the same time, however, large
mvila (clans) appeared, structured in lineages that had put down deep roots in
the land. This network of clans became the cement that bound the
independent villages together.
These transformations were accompanied by significant cultural changes.
The role of the great ancestors, linked with the clans, increased. Cemeteries
and churches, particularly those of Mbanza Kongo, grew bigger. The notion of
zombi, a kind of corpse-slave, took hold, as did the confusion of saints with
great anceston. After about 1700 prominent people began to be buried with
ever greater ostentation. The fabric wrapping around the corpse was now
replaced by an enormous funerary bundle. The wooden heads or busts which
were positioned on the bundles in the eighteenth century became wooden or
stone statues or ceramic monuments to adorn the tombs on which an in¬
creasing number of broken objects were laid. The Boma area and Mayombe
were at the centre of these developments.
The group-healing cults prospered everywhere. The lemba, an already old
example of these nkisi in Loango, changed in character and became a nkisi of
protection for members of the élites who could pay the high enrolment fees.
When an important man fell ill, the lemba priest came to treat him.
Christianity held its ground but evolved. The scarcity of foreign or local
priests left the initiative to the missionaries' former helpen, and at court, to the
king's entourage. The reliefs of Ambriz show crucifixion scenes in which the
spear has become a python, reminiscent of Mbumba or Bomba, associated with
the slave trade and wealth, while the scene is accompanied by drummers.
Crucifixes were retained as emblems of authority and, in what was essentially a
judicial role, as emblems of nkangi (the saviour). Other crosses developed, such
as santu, or charms associated with divination by hunting and used to discover

the cause of collective misfortunes. They were the two sides of the same official
religion.
Angola in the eighteenth century

The Angolan slave trade was conducted on a significant scale long before that
of the north developed. The system, which was focused mainly on Brazil, was
highly fragmented by comparison with that ofthe north. Before 1730, it had
the following components: the export merchant in Portugal; brokers in Brazil;
shippers; slave-dealen in Luanda or Benguela; Afro-Portuguese caravaneers;
and African lords and traders offering slaves for sale at the markets. A single
firm could supply several components but rarely all of them together. Everyone
tried to avoid risks and to maximize profits. The greatest risk was that slaves
would die of malnutrition, untreated diseases and ill-treatment. J. C. Miller
quotes an estimate to the effect that half the slaves bought died between the
market of purchase and the port of embarkation and another 40 per cent at
Luanda while awaiting embarkation. Losses at sea varied according to the state
of health ofthe slaves shipped and the degree of overcrowding on board. They
worked out at 10 per cent ofthe total number embarked.
The parcel, called hanzo in Angola, was supplied on credit to the caravaneer,
binding him to his sleeping partner. The parcel consisted mainly of fabrics from
Goa or Europe, brandy from Brazil, local salt, beads and sometimes a few
weapons. The caravaneers' expenses were high, as were their risks, and they
would soon fall into debt and become completely dependent on their creditor.
But from about 1760 onwards, the caravaneers began selling their slaves to any
trader in Luanda or Benguela in an effort to redress their situation. The
sleeping partner was left with a write-off.
After 1730, metropolitan Portuguese firms came back to Luanda to supply
European goods, and the Brazilians withdrew to Benguela. The ships also
brought numbers of poor immigrants, common convicts and adventuren seek¬
ing to grow rich on the slave trade, especially as caravaneers. An unrelenting
struggle began between the quimbares and the new arrivals. The latter, as
traders, had the full support ofthe high authorities of Luanda, but the quimbares
enjoyed the backing ofthe provincial authorities and, in general, ofthe African
chiefs in the inland markets. The Brazilians and quimbares not only succeeded in
maintaining their position in Benguela but also increased their exports of slaves.
But the Portuguese firms, by virtue of their capital and the administrative
backing they enjoyed, succeeded in shifting the slave risk. They bought, not
slaves, but ivory or bills to be collected in Brazil.
At the turn ofthe century, the governors supported the Brazilians. Later, they
tended to favour the big Lisbon firms. The Jesuits, the allies of the Brazilians,
were expelled in 1660. Francisco Innocencio de Sousa Coutinho (1664
made a vigorous effort to re-establish the mother country's control in Angola,
but neither he nor his successors achieved this. The Afro-Portuguese emigrated
once again, moving mainly to the Benguela plateau where a new bridgehead
for trade with the interior developed. The slaves were taken to Benguela,
avoiding Luanda. But Luanda continued to attract the slave trade through the
kingdom of Kasanje. The caravans took the route from Cassange to Mussumba,
the capital ofthe Lunda empire.
The Lunda empire took shape during the eighteenth century. Constant mili¬
tary campaigns yielded large numben of slaves, while tributes in slaves flowed
into the capital from the subject regions. After 1750, the empire succeeded in
founding the kingdom of Kazembe on the Luapula river and in establishing
communications with Tete and Zumbo in Mozambique. Before then, west¬
ward expansion had reached the River Kwango. The Yaka kingdom, founded
between 1740 and 1760, occupied a large part ofthe Kwango valley down¬
stream of Kasanje and extended its conquests eastward in the direction of the
Kwilu. Large numben of captives were brought back from there to be sold to
the Zombo, Soso and Vili, but also to the Imbangala and to caravans passing
through Nkoje.
In Angola, the slave trade had such an ascendancy that, despite the efforts of
de Sousa Coutinho and othen, the country failed to develop a diversified
economy. The colony remained economically dependent on Brazil; around
1800, 88 per cent of its revenues still came from the slave trade with Brazil and
a litde less than 5 per cent from the ivory sent to Portugal.
The political system of
the Luba and Lunda:
its emergence and expansion
The history of the Luba is the story of the development of a political system
which fint emerged in Shaba, in the present-day Republic of Zaire, and
subsequendy spread into a large part of the southern grasslands, an area
stretching almost from the River Kwango to the River Zambezi.
The connections and the differences between the Luba and Lunda states are
implicit in their two names, which reflect not so much ethnic distinctions as
two political and cultural systems within which a whole range of separate
ethnic referents sprang up. The history ofthe Luba covers matters affecting the
present-day Luba of both Shaba and Kasai as well as matten relating to the
Songye, Kanyok, Kete, Sala Mpasu, Bindji and Lulua. The history ofthe Lunda
deals both with the Rund groups (the Lunda in the narrow sense) and the Lozi,
Ndembo, Lwena, Imbangala and other groups (the Lunda in the broad sense).
The linguistic classification given to the Luba and Lunda languages is indicative
of this intertwined relationship.
There is general agreement that the archaeological finds of the Upemba
Depression at Sanga and elsewhere are to be attributed to the anceston of the
Luba. Emblems of rule, later common among the Luba, appear by the
thirteenth century at the latest, along with evidence at Katota and Sanga of the
formation of two chiefdoms. Yet Katota and Sanga were not the direct
forerunners of the Luba kingdom. All we know is that the main Luba kingdom
came into being and began to expand before the Rund state, fountainhead of
the Lunda, was developed. Oral tradition mentions the Rund state by 1680 but
does not allow us to date its founding.
The emergence of the Luba and Lunda states

Shaba and the adjoining parts of Zambia and Angola are covered by wood¬
lands, the soils are poor and the dry season is very long. The best lands lie
mosdy in the river valleys and the worst in eastern Angola. But the southern,
half-barren country, especially the south-east, contains mineral treasure
copper, iron and salt.
It is not surprising, therefore, that the population lived in pockets of fertile
290
The political system ofthe Luba and Lunda
land, with vast expanses of almost uninhabited areas in between, used mainly
for hunting. This population distribution explains the location of the earliest
territorial organizations, which were small chiefdoms, each corresponding to
one setdement. Thus the Sanga string of settlements becomes the later chief¬
dom of Kikondja, while Katoka upstream, at the other end ofthe Lualaba lake
system, formed another. In both these areas, people were fishermen and
farmen. There were many other chiefdoms, all located in major river valleys.
These population clusters were linked by trade and, presumably, inter¬
marriage. From the north came raffia and palm oil, from the Lualaba fish, from
the south copper and salt, and from the central southern area mbafu oil. The
trade was important enough for currencies to develop from at least 1000
onwards. In the Upemba Depression around 1300, the copper cross became the
currency. After 1700 crosses disappeared and, by the nineteenth century,
imported beads were becoming the new currency.
In this context developed the political ideology that later would be central to
the main Luba kingdom. This consisted of two interlocked principles, that of
the sacred character of kingship (bulopwe) and that of rule through a closed
association.
According to some traditions, the forerunner of the Luba kingdom was the
small Kalundwe kingdom. Its capital was at Cifinda. One of its kings, Kongolo
(Rainbow), moved to found a capital on the plains at Mwibele, not far from
Lake Boya, in what would be the heartland of the future Luba state. Other
accounts, however, have Kongolo coming from elsewhere. The epic narrative
of foundation tells how Kongolo was defeated and slain by Kalala Ilunga
(Ilunga the hunter), a foreigner from the east who moved the capital to Munza.
Kongolo may well not be an authentic figure, but Kalala, as mwine Munza
(master of Munza), represents a founding-father figure that depicts what an
ideal Luba king should be. The Luba kingdom dominated, perhaps from the
outset, the Kalundwe to the west and the Kikondja to the south and controlled
the major northsouth trading routes. Nonetheless, until the eve of the
eighteenth century it would remain a fairly small kingdom.
Meanwhile, further west, other political units were being formed. The
foremost among them, Nkalany, developed in the Mbuji Mayi valley from the
fusion of several small areas headed by tubung (masters of the land). Nothing
really predestined the Rund lands to become an empire. Yet at one point the
northern tubung were united under one of the Rund lands. The northern
tubung traditions explain why an Nkond chief left his office to his daughter,
Rweej or Rueji, who married Cibinda Ilunga, the hunter from the Luba court,
and then abandoned office in his favour.
According to Carvalho Dias, after the conquest stage, power went to Rweej's
fint son, Yav, who became mwant yav (lord Yav). However, the venions that
have been collected more recendy claim that Rweej was sterile. In order not to
jeopardize the succession, she gave her husband a second wife, Kamonga, who
became the mother of his successon. This situation accounts for the institu¬
tionalized existence of two female dignitaries at the court of the mwant yav: the
swan murund or swana mulunda, the symbolic mother of the society, who
symbolized the perpetuation ofthe role played by Rweej, and the rukonkesh or
lukonkeshia, the queen mother, who occupied the role played by Kamonga.
This testimony explains the different institutions that evolved after the
appearance ofthe Luba hunter: first, the two female aristocracies, one symbol¬
izing social fertility and the other biological fertility; and second, the royal tide
based on the chiefly tide mwant, to which the name of the first king was added.
The Luba: internal organization and development to 1800

With Kalala as mwine Munza, a state appeared on the plains incorporating the
Kikondja kingdom to the south and the Kalundwe kingdom to the west. It
further expanded to the south-east as far as the lower reaches of the Luvua river.
The foundation and first consolidation of the kingdom led to some distur¬
bances in the region and a number of emigrants left the lower Luvua to found
the Shila state on the western shore of Lake Mweru and along the lower
Luapula. The most important emigration was towards the north. Groups of
Luba Kasai left Shaba and spread northwards along the Lubilash river, where
The political system ofthe Luba and Lunda 293
'
WW
Id'
À
3CT"
m
^ t 1
*
*
*
i
4
20.2 Hemba, Zaire: monoxyloid caryatid seat, the bottom of which is worn away. Note the
prominent scarifications and the quality ofthe bracelets. Height: 35 cm (G. Berjonneau, ART
135, Boulogne-Billancourt, © and photo)
20.3 Liba, Southern Zaire: the knob of a ceremonial cane, in theform of a human head
decorated with a very tall headdress surmounted by two figures holding each other by the waist.
Overall length: 164 cm (G. Berjonneau, ART 135, Boulogne-Billancourt, © and photo)
they found better watered lands. Famine is given as a major motive for move¬
ment in many traditions of related groups in Kasai.
Although, to the south, in Shila and Kanyok country, a variant of the
Shaban Luba political system was set up, in Kasai this did not occur. Here, the
political organization was either limited to the village and its land, or higher
office was given for life or for a limited period to the highest bidder, a
system also found among some Songye. The Luba Kasai also adopted some
emblems and some political practices from the Kanyok. Most Songye were,
however, ruled differently. By the 1880s, and probably some centuries earlier,
the Songye were living in huge farming settlements which were virtually
towns, ruled by the Bukishi association and dominated by chiefs with a few
tided notables.
The organization of the Luba state corresponded to a pyramid of pyramids.

At village level, households were linked by patrilineage. Relations between
villages were also thought of as relations between lineages, although each
district had a tided chief. Above this, the kingdom itself was ruled from the
capital, whose very lay-out reflected this structure with royal quarters at its
centre and quarten for tided officials of both sexes, separated according to their
military or civilian function. A closed association, the bambudye, helped the king
to rule. The king was not supposed to have a lineage or clan, although the office
was normally handed down from father to brother or son. The king was at the
same time above the political fray and yet linked by kinship to many of the
district heads. He stood at the apex ofthe pyramid of pyramids of kinship. The
tide mulopwe signifies the indivisibility of power that could not be shared.
The mulopwe was surrounded by a household and tided officials. There was
fint his harem, reflecting the real political alliances that bound the kingdom
together. His officials saw to it that tribute was paid in mingilu (unpaid labour),
in milambu (taxes payable in food and local produce) and in gifts offered at the
investiture of dignitaries. The territorial administration was in the hands of the
bilolo (sing, kilolo), each responsible for a region. The central administration
supervised the collection of tribute, organized the army and advised the king
through the tshidie (general council) and the tshihangu (king's court).
We do not know exacdy how the bambudye association functioned, as its
secrets have not been revealed. But it had numerous lodges and exerted both
ntual and secular control over the country and even the king.
Local courts were patterned after the organization of the capital and local
chiefs had some sacred paraphernalia and links to local territorial spirit cults,
often for their own anceston. The coercive power of the king's warriors was
the practical force that kept the country together. But there was no standing
army, so that the ideological force constituted by the bambudye was of great
importance. Heavy tribute could only be enforced when central districts were
willing to back the king against internal or external forces. It is not surprising,
therefore, that the kingdom did not expand rapidly.
There is little information about territorial history until about 1700 and the
reign of the king Kadilo. He embarked on unsuccessful campaigns to the north
20.4 The Liba kingdom in the eighteenth and nineteenth centuries (afterJ. Vansina)
against the adjacent Songye towns, but the real expansion of the kingdom
occurred in the valley of the Lualaba, along the Luvua (Kiambi) and south of
the Upemba Depression. Such expansion faltered as a major succession crisis
developed from about 1780 to 1810. During this crisis, the Kanyok freed
themselves from Luba rule, and Songye political and cultural influence reached
a high point. Later, expansion continued towards the east, with the conquest of
the lands between the River Lualaba and Lake Tanganyika and a push towards
the north along the Lualaba river as far as Buli.
The Lunda: internal organization and development to 1800

The Rund kingdom became the Lunda empire in the eighteenth century. Its
earlier organization is not fully known, except that its military character was
much more developed than in the Luba kingdom. A little is known, however,
about the principles of social and political organization.
The succession ofthe tubung had been matrilineal, but succession to the royal
title and at court was bilateral. The matrix of society was based on positional
succession and perpetual kinship. This meant that every incumbent of a
position or tide was supposed to become his predecessor. The incumbent took
the name of his predecessor, and his wives, children and other relatives. As a
consequence, kinship was perpetual. If, for example, the first holder of a title
was the king's grandson, five generations later the holder of the title was still
the king's grandson.
It was possible, therefore, to conceive ofthe state as an organization run by a
single family, headed by the mwant yav (emperor). New chiefs could be
incorporated by giving them a kinship link (as son-in-law, for instance, after a
marriage alliance) and providing them with a stable position in the empire. The
organization of the empire on a family model automatically regulated relations
between officials, based on a strict notion of generation.
So, to the Rund, the state was a family writ large very large indeed as the
empire came to stretch from the River Kwango to beyond the River Luapula.
But it was a family of warriors and a family that would thrive on slavery. By
about 1700, Lunda warriors had subdued the populations on the empire's
periphery, incorporated them and then moved further afield. From 1750 to
1760, strong subsidiary kingdoms took shape from Yaka country on the
Kwango river to Kazembe on the River Luapula, along an east-west axis that
encompassed the mineral resources of Shaba and access to the Portuguese
emporia on the River Zambezi as well as the trading routes from the Rund
country to Imbangala territory on the Kwango.
The imperial administration was loosb. The empire consisted of a core, plus
peripheral kingdoms whose rulen merely paid occasional tribute to the court.
The king had created special officials who roamed the countryside with a
military detachment to force the bilolo to pay tribute. Apart from these forces,
there was a major military organization headed by a kazembe (general). The
kazembe wielded all power in his operational area, but once the area was
incorporated, he either lost his status or became a ruler in his own right, a kilolo
ofthe emperor.
The capital was laid out like an army camp, with a front and a vanguard,
wings and centre. Its hub was the royal palace, where the emperor, the swan
murund, the rukonkesh and dignitaries resided. Power was vested in the emperor;
his tide, mvant yav (lord ofthe viper), referred to the distance between the king
and other mortals, but also to his ambiguity, as he stood for both peace and
war, prosperity and destruction.
Research has shown that the expansion of the Lunda empire started before
1700. The conquest was towards the south-east, towards the salt-pans of the
Lualaba river, near present-day Kolwezi. From there, in about 1700, a break¬
away group went south into present-day Zambia and set up a state in the
Ndembu area after partly assimilating the indigenous Mbwela and pushing
othen southwards.
From the salt-pans of the Lualaba river, Lunda generals moved eastwards on
a new campaign and reached the rich Luapula valley. The Lunda army beat off
a Luba attack against them and subjugated the chiefs of Shaba and the Shila
state of Luapula. This state grew and by the second half of the eighteenth
century Kazembe ofthe Luapula was lord of a large tighdy-knit kingdom. Soon
after their setdement, the Lunda of Kazembe made contact with the Portuguese
of Sena and Tete. A mission led by F. de Lacerda e Almeida visited the new
kingdom from 1798 to 1799. It was impressed by the military power of
Kazembe and with the severity of measures used against the inhabitants. This
Portuguese visit inaugurated a period of intense trade relations between
Kazembe and the lower Zambezi. This was to strengthen the autonomy of
Kazembe from the capital, so that the regular tribute payments ofthe 1790s had
become more an equal exchange of gifts by the 1830s.
Towards the north and west, other Lunda expansion was closely linked to
the development ofthe slave trade in Angola. Contacts with the Angolan slave-
trading system date from the 1670s at the latest and developed subsequendy.
The Imbangala developed a system of trading by caravans which brought
European goods, such as cloth, beads and crockery, to the Lunda capital, where
they served as luxury goods for the local aristocracy. The Lunda would not
accept guns, preferring to remain faithful to their sword, the mpok. In return,
they exported slaves captured initially in the areas north and east of the Rund
core. Later, Mbwela captives from northern Zambia and eastern Angola were
also added to those exports. At the same time, demand for slaves grew within
the Lunda empire itself. They were used to cultivate large fields to grow
cassava; they also worked as ferrymen and porters in the caravans.
Shordy after 1700, military expeditions were sent westwards, and by about
1750 Lunda chiefs had reached the Kwango. Eastern Angola, the land of the
Lwena, was their fint goal. The Lunda leaders superimposed themselves on
fairly large-scale societies organized around matrilineages of great depth. West
of the Lwena country various chiefdoms were also created by the Lunda by
superimposing themselves on existing societies. Among these peoples, the
Chokwe were later to prove the most important. Even in the eighteenth
century, they were valuable to the empire as metallurgists, carven and hunten.
In the Kwango valley, the Lunda invaden found Yaka, Suku and perhaps at
Okango yet other chiefdoms structured after the Kongo pattern. These they
overran. Many inhabitants fled, but the structures were not destroyed. As a
result, a complex system came into being in which even the ideology of the
state acknowledged the existence of indigenous people and Lunda invaden. But
the leader of the Yaka, the kiamfu, could not control all his own men. In the
north, a semi-independent state, Pelende, arose, whilst a group called the
Sonde-Luwa emigrated east of the middle Kwango and founded chiefdoms of
their own.
Once established, the Yaka kings started raiding the lands towards the Kwilu
to gather slaves for sale to Angolan, Kongo and Vili traden. This provoked
major upheavals in the region. The Suku kingdom put up a successful resistance,
but elsewhere people fled north-eastwards, provoking high population densities.
The final expedition went from west ofthe Kasai river northwards, probably
along the valley of the Tshikapa. The Lunda founded two small states: one at
the main falls ofthe Kasai with its capital at Mai Munene, which soon became
a major market for caravans from the River Kwango, and the other, called
Mwaka Kumbana, on the River Loange. In both areas the Lunda mixed with
other recent immigrants, the Pende. The resulting political systems were
marked by Pende influence.
Conclusion
By 1800, Luba and Lunda expansion had structured all the grassland area of
central Africa east of the River Kwango. The peoples of this area shared a
common culture, with common rituals and common emblems and symbols.
The spread of this culture was facilitated by the existence of the trade routes and
even by raids. Population mobility, especially through marriage alliances, was
astonishingly high, as evidenced by the spread of clan names inherited through
the mother. The same names are found from the Kwango to the Kasai and
from the Kasai to the Luapula rivers.
Lunda expansion, however, also led to enormous devastation. The militaristic
nature of the Lunda states should not be underestimated, nor the scale of their
slave-raiding with its consequences for the populations. This, and the presence
of a common political culture from the Kwango to the Luapula, still constitutes
for the inhabitants of this huge area the legacy of those centuries.
The Northern Zambezia-
Lake Malawi region
The region discussed in this chapter is bounded on the south by the Zambezi,
on the north by the Songwe and Ruvuma riven, on the west by the Luangwa,
and on the east by the Indian Ocean. The southern part of this region was
dominated by the Chewa-speakers with their sub-groups. To the west of the
Chewa were the Nsenga and to the east the Lolo-Makua-Lomwe speakers and
the Yao. The northern zone stretched on the western side of Lake Malawi
from the Chewa-Tumbuka marginal zone in the south to the Songwe river in
the north. Three language families occupy this zone: the Tumbuka, the
Ngonde-Nyakyusa and the Sukwa-Lambya-Nyiha. In 1500, all the people of
the southern zone and the Tumbuka of the north belonged to the 'matrilineal
cluster of Central Bantu speaken' which stretched from southern Zaire in the
west to the Indian Ocean in the east. In the centuries that followed, however,
the Tumbuka changed to patrilineal descent. The Ngonde-Nyakyusa and
Sukwa-Lambya-Nyiha had always been patrilineal since the dawn of their
historical tradition. The region today comprises eastern Zambia, all of Malawi,
and northern Mozambique.
In the period under discussion, the southern zone was dominated in the
fifteenth century by the arrival of the Maravi and the rise of their states, whose
expansion continued in the sixteenth and seventeenth centuries. In the north,
in the sixteenth century, the people were organized in clusters of autonomous
clans. Late in the same century, however, an immigrant group the Ngulube
founded the states of Lambya, Ngonde, Chifungwe, Sukwa and a number of
Nyakyusa chiefdoms. In the same period, the expansion of the Maravi into the
Tumbuka-Chewa marginal zone led to the establishment of new Chewa
chiefdoms which imposed their sway over the Tumbuka peoples, creating in
the main the Tonga people and language. In both zones the trade in ivory was
a major factor in the eighteenth century. The southern zone was then
characterized by the decline of the Maravi state system, the rise of successor
states and the outward spread of the Yao, initially as traden and latterly as state
founders. A group of immigrants from Nyamwezi country the Balowoka
setded among the Tumbuka and created economic spheres of influence which
eventually emerged as political entities. The older established states south ofthe
300
The Northern ZambeziaLake Malawi region 301
Songwe, however, remained aloof from these new commercial developments

until well into the nineteenth century.
Prior to the sixteenth century, the region experienced a succession of Iron-
Age immigrations, including those which brought the fint ancestors of the
modern Bantu-speaking inhabitants. These new immigrants appear to have
arrived from the north after 1200 and they displaced or assimilated the Bantu
and pre-Bantu agriculturists and hunter-gatheren already there.
In northern Malawi, the ancestors of the modern Bantu-speaking population
were a stateless and pre-dynastic population. In eastern Zambia, east of the
Luangwa river, the earliest of the modern Bantu-speaking inhabitants are
thought to have been the Katanga. Further east, in central Malawi, they com¬
prised a number of clans to whom the term 'proto-Chewa' has been applied.
In oral tradition, the proto-Chewa are sometimes referred to as Kalimanjira
(path-maken) , who cleared the country of its earlier nomadic inhabitants, the
Batwa and Kafula.
Litde is known of the origins and early composition of the Lolo-Makua-

Lomwe and Yao communities of eastern Mozambique. There is no tradition of
how they occupied their present homelands. This is probably because they
occupied their country many centuries ago and have therefore long since lost
the memory of their exact origins.
In the northern zone, the earliest modern Bantu-speaking inhabitants for
whom there are records setded in the area between the Songwe and the
southern Rukuru. Their traditions claim that they came from the north-eastern
side of Lake Malawi and may have been related to the abilema. Not long after¬
wards, more powerful groups of immigrants began to infiltrate the area. These
included the Simbowe, who occupied the Karonga plain, and the Mbale, who
invaded the Phoka area. The Mbale migration into the Phoka highlands
probably occurred in the fourteenth century. The new immigrants were noted
iron-smelten and they easily established good relations with the indigenous
people, the Mzembe and Chiluba, who supplied them with agricultural
produce and coal and received in return iron agricultural tools and weapons.
Thus it would appear that there was a general migration of peoples southwards
using the corridor between Lakes Tanganyika and Malawi as a main passage.
Between the Phoka and Chewa areas in the north and south respectively,
there lived various Tumbuka-speaking clans. Some of these groups seem to
have spread as far east as the lakeshore and as far west as the Luangwa valley in
the modern Lundazi district of Zambia. The Nsenga, the present-day inhabi¬
tants of the latter region, speak a language that is akin to Tumbuka and share
the same clan-names with the inhabitants of Tumbukaland.
Thus, prior to about 1500, the political structure ofthe entire region from
the Songwe river in the north to the Zambezi in the south, with the exception
of a few pockets, was characterized by the prevalence of stateless small-scale
polities. The different ethnic groups formed territorial chieftaincies. The local
chief presided over a group of closely related lineage-based villages of which his
own was the most senior genealogically. To the community of villages
The Northern Zambezia-Lake Malawi region 303
surrounding him he rendered religious, judicial and military services, and was
in turn entided to the allegiance of all his followers.
Though politically and sometimes physically divided into ethnic and clan
spheres of influence, the Lake Malawi region enjoyed a certain degree of social
and religious cohesion. Over the greater part of this region, religious action was
expressed at two levels. At a local level, religion fulfilled the function of ensuring
the moral and material well-being ofthe population, while at the territorial level
it also fostered cultural and ecological co-operation. Religious practice for most
people involved ancestor veneration, spirit possession, rain-making and control
of witchcraft. The chief manifestation of a territorial religious experience were
the rain cults, dedicated to a god or spirit acknowledged over a wide area.
It has been shown that at least within the matrilineal belt extending from the
Tumbuka-Chewa marginal zone in the north to the Zambezi valley in the
south, there was a considerable degree of interaction among the various terri¬
torial religious cults. The deity was represented in the same way throughout.
Among the Tumbuka and Chewa, for example, he took the physical form of a
snake, was thought to be a male force, and had several spirit wives devoted to
his service. The functionaries dedicated to the deity were possessed and set
apart for their special role. Thus religion provided an important means of
communication among peoples who were otherwise divided politically.
Among the Chewa and related peoples of the southern zone, the loose
political structure which prevailed before 1500 was greatly transformed by the
emergence of the Maravi state system at the beginning of the sixteenth century
and its later expansion. The Maravi thus dominate the sixteenth- and
seventeenth-century political history of the northern Zambezia region just as
the builden of the Mutapa state dominate that of southern Zambezia in the
fifteenth and sixteenth centuries.
It is now generally agreed that the Maravi were immigrants who originated
in the Luba area of south-eastern Zaire, and that they entered central Malawi
from the west. On reaching the southern end of Lake Malawi, they rapidly
established themselves as rulen over the earlier proto-Chewa inhabitants, con¬
solidated their rule, and then embarked on a campaign of territorial expansion
which took them beyond central and southern Malawi. By the early seven¬
teenth century, their confederation of states encompassed the greater part of
eastern Zambia, central and southern Malawi, and northern Mozambique. In
this way, the Maravi state system came to have a political, military and
economic impact over a vast area.
However, the process which led to the emergence of the Maravi state system
is not yet clearly undentood by historians. Some historians hold that the Maravi
arrived in the country fully equipped with the symbols of chiefly power, with
whose help they established themselves as a ruling class over what must have
been a stateless indigenous population. Others would stress the role of demo¬
graphic factors and argue that a growing population engaged in surplus produc¬
tion, albeit modest, would have served as a suitable basis for the accumulation
of dynastic power. Another hypothesis takes into account the trade factor,
maintaining that the formation of a state at the southern end of Lake Malawi in
the fifteenth or sixteenth century arose from the need for an authority to
control the development of the trade in ivory between the southern shores of
the lake and the east coast of Africa via the Shire-Zambezi water system.
Oral tradition records little ofthe early history ofthe area or exactly how the
Maravi state was founded. It simply maintains that under the leadenhip of the
kalonga (the tide of the ruler) Chidzonzi, who headed the most senior Phiri
royal lineage, the Maravi established their fint kingdom around Mankhamba
and Manthimba on the south-western side of Lake Malawi. According to
Portuguese written sources, by 1624, Manthimba was an important commercial
and political centre, and its inhabitants were trading with Portuguese traders
from Tete on the Zambezi.
Oral tradition, however, provides the only indication of the institutional

mechanisms through which the kalonga's state was held together. One of these
was the mlira cult. Once every year, the heads of the different Phiri royal line¬
ages were invited to Manthimba for the ritual veneration of Mlira, the spirit of
the great kalonga Chinkhole, who had led their anceston into the country at the
time of the migration. This ritual, we believe, was a surface manifestation of a
royal cult which had a significant integrative function within the Maravi state.
Equally important to the integration of the kingdom was the manner in which
appointive offices of state were allocated. The kalonga is said to have appointed
some heads of proto-Chewa clans to important leadenhip positions in his state.
The Banda were caretaken of the kalonga's shrine at Mankhamba while the
Mwale led the kalonga's warrion. Traditions clearly undencore the extent to
which non-Phiri or non-Maravi clan leaden were involved in the making of
decisions pertaining to land tenure, the distribution of wealth and the making of
war.
The extent to which the kalonga was able penonally to direct the affairs of
his kingdom may have begun to decline towards the last quarter of the
sixteenth century when the state began to expand. The heads of the junior
Phiri lineages left the ManthimbaMankhamba zone to occupy adjacent
territories. Taken together, the territories controlled by the kalonga and these
subordinate chiefs formed the heartland ofthe Maravi state system as it evolved
in the seventeenth century.
Another dimension of the Maravi expansion involved the migration of the
kalonga's senior kinsfolk into distant lands to the south and south-west of the
heardand. This may have been a calculated move on the part of the early
kalongas as they sought to distance possible pretenden to the throne at
Manthimba. Kaphwiti and Lundu migrated into and colonized the lower Shire
valley. Kaphwiti initially controlled the whole valley before he lost much of his
authority to the Lundu who, after imposing his ascendancy in the lower Shire
valley, spearheaded Maravi expansion eastward into Lolo and Makua country.
The process of Maravi expansion in this part of northern Zambezia started in
the mid-sixteenth century.
The Maravi invasion and conquest of Lolo-Makua country first owed much
to the fanaticism of the lundu' s warrion, whom some historians of eastern
Africa have identified with the notorious Zimba marauders ofthe late sixteenth
century whom the lundu subsequendy employed. The success of this conquest
was also due to the statelessness of the Lolo and Makua societies. The Lolo
political system was segmentary; the functions of government were performed

by village headmen assisted by councils of elden. The Makua also had a
segmentary political system, but among them lineages belonging to the same
clan sometimes, in moments of crisis, combined under the leadenhip of the
most powerful lineage. Thus, as a result of pressure from the Portuguese, strong
chieftaincies emerged among the coastal Makua in the late sixteenth century.
The Lolo and Makua conquered by warrion ofthe lundu were incorporated
into a tributary state which the Portuguese residents of the lower Zambezi
termed 'Bororo'. As a result, by the beginning ofthe seventeenth century, the
lundu controlled a vast territory north of the Zambezi, from the lower Shire in
the west almost to the Indian Ocean in the east. Not only was he recognized as
'the second most powerful penon in the empire of Maravi (the fint being
Kalonga)' but his warrion periodically descended on the Makua part of the
coast to raid Portuguese setdements.
The lundu held his vast kingdom together with the help of loyal military
generals and vassals drawn from junior lineages within his clan. Furthermore,
his state had an economic underpinning in that the different peoples whom it
encompassed depended on the salt, iron and machila cotton cloths that were
produced by the the lundu's Manganja subjects in the lower Shire valley.
Manganja religion was another unifying factor in the state. Centred on the
Mbona cult, whose central shrine was at Khulubvi in the lower Shire valley,
the theology of Manganja religion encompassed many elements of commoner
protest against the misuse of royal power and prerogative. Consequendy, the
official cult of the lundu paramountcy found many adherents over a large
territory, stretching from the seat of the lundu in the lower Shire valley to the
Zambezi delta on the east coast.
The immense prestige enjoyed by the lundu in the eastern part ofthe Maravi
confederation did not go unchallenged. The kalonga, who himself controlled a
vast territory on the western and eastern sides of Lake Malawi as well as along
the upper Shire, became extremely anxious about the lundu's growing power.
In the 1620s and 1630s, therefore, Maravi expansion eastward was stalled by
the vicious internal rivalry which developed between the kalonga and the lundu.
Thereafter, Maravi expansion shifted from the east to the south-west. In this
latter direction, the undi then established his own kingdom which he later
expanded.
According to one venion of the traditions relating to the establishment of
the undi's kingdom, the founder was sent out by the kalonga to occupy the
sandy plains near Nsengaland. Another, more reliable venion, however,
suggests that the undi's departure for the south-west was occasioned by a
conflict within the ruling Phiri clan at Manthimba.
In the west, the undi and his followen successfully colonized the Kapoche
river area, a tributary ofthe Zambezi. Thereafter, they extended the boundaries
of their nascent kingdom toward the confluence of the Zambezi with the
Luangwa. There, the undi's warrion came into conflict with the Tawara along
the Zambezi and the Nsenga along the eastern bank ofthe lower Luangwa. But
historians differ on the extent to which the Maravi then came to dominate the
Nsenga. However, what can be concluded from this debate is that during its
expansive phases, the undi's state encompassed a number of Nsenga chief¬
taincies, but not all. The south-western Nsenga near Feira in Zambia, for
example, have no traditions of having been a part ofthe undi's 'empire'.
There is still the question of how the undi may have established a certain
degree of political control over the Nsenga. The traditional argument is that
the undi and his subordinates used force and created perpetual kinship relation¬
ships according to which the conquered Nsenga chiefs were recognized as
'sons' or 'nephews' ofthe Maravi chiefs. Of late, however, a stronger argument
has been presented. This is that trade and famine may have played a crucial role
in the expansion of the undi's influence over neighbouring peoples. The
Nsenga chiefdoms along the Luangwa valley suffered from endemic famine,
and their inhabitants often sought relief in the undi's territory, which was more
fertile than their own. The undi also enjoyed immense influence with Portu¬
guese traden at Tete, and this fact enabled him to exercise some control over
traden from neighbouring communities. There is also every indication that the
undi and his kinsmen allowed the Nsenga chieftains who came under their
influence to continue to administer the affairs of their clans and areas, as long as
they sent tribute in ivory and slaves to the undi's headquarters.
The Maravi and the Chewa also expanded northwards into Tumbuka country.
Among the princely leaden of Maravi origin who migrated in that direction
were Chulu, Kaluluma, Kanyenda and Kabunduli.
Chulu and Kaluluma moved into the southern Tumbuka area known as
Chimaliro and the surrounding plains. Here, they encountered a Tumbuka

people who were self-sufficient economically but loosely organized politically.
The manner in which Chulu and Kaluluma took power is subject to con¬
flicting interpretations in Chewa and Tumbuka oral traditions. It would seem,
however, that after a period of peaceful cohabitation, the Tumbuka tried to
resist and were crushed by force.
Kanyenda and Kabunduli migrated into what later became Tongaland. Both
originally came from the heartland of the Maravi empire and their migration
northward brought them to Khota Khota, where Kabunduli split from
Kanyenda and migrated further north into Tongaland. There Kabunduli soon
earned a reputation for reorganizing the once scattered Tonga groups whom he
apparently defeated and brought under his own control. As his reputation
increased, he was followed by a number of commoner families from the Chewa
heartland.
This intermingling of the Maravi with the original inhabitants of Tongaland

accounts for the peculiarities of Tonga society as we know it today. Chitonga,
the language of the area, is a bridge between the Tumbuka and Chewa
languages. The Tonga are also the only group in northern Malawi to have
remained matrilineal in descent and inheritance customs to this day.
For the Chewa-Maravi peoples, the seventeenth century was a golden age,
when, as a result of their territorial expansion, they emerged as the most
renowned power-brokers north of the Zambezi. Even the Portuguese,
ensconced in their setdements along the Zambezi, were forced to co-operate
with them. This was because the Maravi did not confine their interest to the
area north ofthe Zambezi. From the 1590s through to the 1630s, they actively
interfered in what was the Portuguese sphere of influence south of the
Zambezi. For two generations at the beginning of the seventeenth century, the
Portuguese, from Angoche on the east coast to Tete on the Zambezi, sought to
contain the Maravi by collaborating with them. They recruited large numben
of Maravi warriors to help them in their campaigns against the Karanga or
Shona south of the Zambezi as well as against enemies in the immediate
vicinity of their Zambezi setdements.
The Maravi expansion had several effects. For example, the creation of the
lundu's state east of the lower Shire valley made it possible for the Mbona cult
among the Manganja to spread its influence more widely. The economic and
cultural effects were perhaps as significant. Within territory encompassed by the

Maravi confederation in the seventeenth century, trade in ivory flourished. As a
result, northern Zambezia had considerable interaction with the mercantile
capitalist economy. The Maravi 'empire' may also have fostered cultural
uniformity among the different matrilineal peoples of the region, since they
share the same clan names and have a common descent system.
While the Maravi were expanding over the southern zone in the sixteenth
and seventeenth centuries, the north was being penetrated by the Ngulube
immigrants from the north-east, a major event in the history of this zone.
In the vanguard of the Ngulube migration, the mwaulambya crossed the
Songwe nver south of the Misuku hills into what came to be called Ulambya.
There he found the Sikwese and Chilima clans. The mwaulambya ensured
political control through more peaceful means than the kyungu who was shortly
to invade the Karonga plain. This was partly because the numben accom¬
panying the former were smaller than those accompanying the latter. The
major officials of the mwaulambya were drawn from local families, who shared
prestige and political power with the new rulen. Similarly, the Msukwa
assumed political power over the Simwayi and Silumbu clans without the use
of excessive force. Both Ulambya and Misuku were founded on compromise.
The kyungu, Kameme and party migrated to the Karonga plain via Unyiha,
Uiwa, Unamwanga and then Ulambya and Misuku in the west. Within a short
time of their arrival in Karonga, Kameme returned and settled in the area
immediately west of Ulambya. Here he established political authonty over a
people who were mainly of Nyiha stock, although in later yean many Mambwe
and Namwanga migrated into the Kameme chiefdom. How Kameme estab¬
lished his state is not clear. It was clearly a smaller polity than that of his brother,
the kyungu, who founded his kingdom on an already established state ruled by
the Simbowe clan.
The Simbowe clan came from Unyiha in modern Tanzania and had estab¬
lished itself at Mbande hill. The basis of Simbowe 's power is not clear but it
would appear that he was a trader and part of a trading network extending to
the east coast. Simbowe's relations with the indigenous people do not appear to
have been easy; after his arrival, the kyungu formed an alliance with the
Mwenekisindile who were the custodians of an important religious shrine
associated with a snake cult, and it seems that the Mwenekisindile assisted him in
launching an attack upon Simbowe. The kyungu assumed power at Mbande by
force and gradually was able to establish a new order. Once setded, the kyungus
maintained their supremacy over their new subjects by establishing a ritual
centre devoted to the cult of their royal anceston. The kyungu and his officials
further established themselves in the area by marrying into local families.
Before the arrival of the kyungu clan, the people seem to have wonhipped
the high god through intermediary spirits such as the divine snake. The kyungus
employed the anceston as mediums for the wonhip of their own high god,
Ngulube. As the people slowly came to accept ancestor wonhip and aban¬
doned the snake cult, they also accepted the divinity ofthe kyungu. He became
priest-king and communicator with the royal anceston. To the Ngonde, the
kyungu became the living representative of god. His health determined their
welfare and prosperity and to safeguard this he was restricted to his residence.
Should the kyungu show even minor signs of illness he was immediately
smothered by his councillors, the makambala.
These developments occurred during the reigns of the fint four kyungus,
roughly between 1600 and 1720. Soon after the death of the third kyungu,
Mwakalosi, his son and successor, renounced the throne because he feared that,
should he fall ill, the makambala would kill him. He also refused to have all but
one of his children killed, a practice designed to avoid competition over the
succession. The crisis caused by the refusal of Mwakalosi to accede to the
throne ended when the makambala installed his brother Magemo as kyungu.
The Kameme, Mwaulambya and Msukwa chiefdoms also promoted royal
ancestral veneration. Yet they became more Nyiha in their philosophy as in
their language, and smothering the monarch, restricting his movements and
killing royal children never featured in their history. However, all these chief¬
doms represent the introduction into the area of a new form of political organiza¬
tion which brought together religion and politics under prestigious leaders and
where relations among the citizenry were based more upon political than upon
kinship ties.
A new succession crisis occurred in the Ngonde kingdom upon the death of
the kyungu Magemo, whose son, Mwangonde, was too young to succeed him.
His sister's son, Kasyombe, became kyungu. With his accession to power the
makambala began to lose some of their powers to the kyungu. Kasyombe, whose
father was a Ndali, had been brought up at Ngana in the northern part of the
kingdom. He ended the practice of killing the male children ofthe kyungu, and
the number of princes began to rise, leading to a further increase in the powen
ofthe monarch, especially after some of his successors had assigned portions of
land to some princes, thereby ensuring that the kyungus influence was felt
throughout the region. There can be no doubt, however, that by 1800, the
Ngonde kingdom was confident enough to begin asserting itself in the politics
ofthe wider region south ofthe Songwe.
Thus between the mid-thirteenth and mid-eighteenth centuries the area
south ofthe Songwe river had developed through a number of stages, from an
early network of religious shrines associated with a snake cult, to a system of
dominant clans and, finally, to the foundation of a number of states: in the
north Ulambya, Kameme, Mikusu, Ungonde and Mwaphoka Mwambele, and
in the south Kanyenda, Kabunduli, Kaluluma and Chulu. In the north, Kyan-
gonde had been introduced as a new language by the Ngonde and Nyakyusa
founders, while in the south Chitonga was evolving out of a fusion of the
Tumbuka and Chewa peoples and languages. The central part of the region
was dominated by the Tumbuka, who had no centralized government and
whose early history is extremely difficult to chart.
The kinds of religious change which occurred following the establishment of
the Ngonde state were also discernible in the Maravi zone of influence. The
Maravi state system encompassed a network of pre-Maravi or proto-Chewa

rain-shrines centred on Msinja in what is western Lilongwe today. Each of the
Maravi state-builden tried to exercise control over the shrine that fell within his
jurisdiction. He would offer assistance and protection, and usually make annual
pilgrimages to the shrine. Both the kalonga and the undi tried to gain control of
the Msinja shrine, while in the lower Shire valley the lundu managed to establish
a considerable degree of control over the Mbona shrine at Nsanje. With this
level of dynastic participation in the affain of the rain-cults, their theology
gradually became syncretic. It began to accommodate the veneration of the
royal ancestral spirits, while purely Chewa creation myths were suppressed.
However, the most notable religious change of the period occurred where the
dynasts failed to gain control over the rain-cults. They then tended to establish
their own cults, and one of their functions was to deify the royal family.
Towards the middle of the eighteenth century, the Maravi state system
began to decline. This decline took the form of secessions from or revolts
against the kalonga, lundu and undi paramountcies. However, the faetón
accounting for the decline of these Maravi states are not clear. In the case of the
undi, the Portuguese who invaded his kingdom during the gold rush of the
1740s and 1750s played only a minor role in undermining his authority. It has
also been argued that the authority of both the undi and the kalonga was
weakened by the growing reluctance of Makewana, who controlled the central
Chewa religious shrine at Msinja, to use her ritual power in the interest ofthe
ruling dynasties. The early Makewanas had been key links in the religious unity
ofthe kalonga and undi kingdoms but the Makewanas ofthe eighteenth century
developed secular ambitions and carved out their own sphere of influence.
Their resistance to the the kalonga and the undi was probably indicative of an
attempt by the proto-Chewa clans, to which they belonged, to reassert their
autonomy.
Finally, a factor which proved crucial to the decline of the lundu's influence
east of the Shire river was the emergence of Makua-Lolo nationalism in the
eighteenth century. Their nationalism was primarily directed against the Portu¬
guese and their African allies, against whom they waged a bitter war of attrition
from 1749 to the end ofthe century. The challenge which the coastal Makua
posed for Portuguese imperialism on the Mozambique coast must have inspired
the Makua and Lolo further inland to launch similar campaigns against foreign
domination. In the inland areas, it was the Maravi rather than the Portuguese
who became the object of Makua-Lolo hostility. Maravi supremacy then began
to be shaken from Quelimane in the east to Mount Murambala on the Shire in
the -west.
The peoples of northern Zambezia in the period under consideration were

also gready affected by their growing economic contacts with the outside
world. Such contacts go back to the period of Arab-Swahili commercial domi¬
nance on the east coast and along the Zambezi before 1500, but were then
rather limited. However, after the setdement of the Portuguese along the
Zambezi in the mid-sixteenth century, Portuguese traders regularly sailed up
The Northern Zambezia Lake Malawi region 311
the Shire river from Sena to purchase ironware, machila cloth, salt and ivory
from the Manganja, with imported cloth, beads and brassware. Further up the
Zambezi, the Portuguese settled at Tete established trading connections with
the undi's kingdom and the heartland of the Maravi state system as far as the
upper Shire and the southern shore of Lake Malawi. With the founding of a
feira at Zumbo in 1716, the Portuguese gained access to the ivory market in
several societies as far west as southern and central Zambia. But it was the rise
and spread of the Zambezia prazos which played a decisive role in the
expansion of Portuguese commercial influence north ofthe Zambezi.
Prazos came into existence when a number of Portuguese or Goanese
colonists assumed the status of political chiefs over land that initially belonged
to the indigenous African peoples. The process began around Sena at the close
of the sixteenth century and gradually spread to other parts of the lower
Zambezi valley in the seventeenth. The crown in Lisbon saw the granting of
prazos to Portuguese setden as a way of encouraging private initiative in the
colonization of the Zambezi valley, and also hoped that the prospect of
acquiring land would attract immigrants to settle the region.
In the eighteenth century, prazos dominated the whole lower Zambezi
region from the Luabo or Zambezi delta in the east to Chicoa in the west. The
big prazos were more than private estates. They amounted to areas of juris¬
diction in which the estate-owners, known as prazeros, 'had absolute power of
justice, waged war, imposed tribute, and were often guilty of great barbarities'.
Of the many issues about these prazos which are of interest to the historian,
two stand out. One is that of the complex relationships which they maintained
with neighbouring African states; the other is that of the economic impact
which they had on the rest ofthe northern Zambezia region.
Most of the prazeros entered into diplomatic and military relations with the
surrounding African states with the explicit aim of exploiting their human
resources. For example, the Chikunda, who served the prazeros as retainen, had
access to labour among the Makua, Manganja, Sena, Kalanga, Tonga, Tawara,
Nsenga, Tumbuka and othen. Through their interest in the gold and ivory
trades, the prazeros played an important role in the development of commerce
north ofthe Zambezi. Using some of their Chikunda retainen as middlemen,
they established a network of trade links with different African peoples in the
region: Manganja, Chewa, Nsenga, Lenje and Southern Lunda. These inland
peoples supplied copper, ivory, wax and slaves in exchange for imported cloth,
beads, brassware, alcoholic beverages and salt. From about 1740, the prazeros
were also involved in the mining and exchange of gold north of the Zambezi,
particularly in the undi's kingdom. Approximately 100 kilograms of gold were
extracted annually before the supply began to dwindle in the 1780s.
This gold rush led to the loss of the ivory trade by the prazeros and the
Portuguese community on the Zambezi; that trade fell increasingly into the
hands of other competiton, particularly the Yao.
The Yao emerged as a dominant trading nation in the late seventeenth and
early eighteenth centuries and they participated in long-distance trade through-
out East-Central Africa. According to Yao oral tradition, it was the Chisi iron-
smiths who pioneered trade to the coast. The Chisi depended on their iron-
working skills and local trade in ironware for their livelihood. It is thus possible
that once Yaoland was adequately supplied with iron goods, the Chisi took
their merchandise further afield and established contact with the coast. The
further development of Yao commercial activities was facilitated by the
creation of a stable market for ivory at Kilwa between 1635 and 1698. When
the Kilwa market for ivory fell after 1698 as a result of Omani-Portuguese con¬
flicts, the Yao found an alternative market at Mossuril opposite Mozambique
Island. Trade, it would seem, had by that point become indispensable to the
Yao way of life. The Yao would carry their ivory to either Mozambique or
Kilwa, as dictated by circumstances. To the west of their own homeland, the
Yao had a rich source of ivory in Maravi country. Furthermore, from about
1750, they began to receive more ivory from the Bisa of north-eastern Zambia
for onward transmission to Kilwa.
For the Bisa, trade probably offered compensation for their political impo¬
tence at home. They -were trapped between the expansionist forces ofthe Lunda
and Bemba kingdoms, and trade and travel were important avenues to oppor¬
tunities further afield. The Bisa opened many new routes between the Luapula
region in north-eastern Zambia and the east coast. Between 1790 and 1830,
they also tried to develop commerce between the Lunda kingdom and the
Portuguese on the Zambezi. They procured ivory, copper and slaves from dif¬
ferent places in the interior of Central Africa and took them to the settlements
on the coast or along the Zambezi to be exchanged for cloth, beads, guns and
gunpowder.
One consequence of their trade with the coast was that their own settle¬
ments began to drift eastward into the Luangwa valley and beyond. Several Bisa
villages were established in the valley by the 1760s, and by the end ofthe eigh¬
teenth century there were Bisa villages in the western part of central Malawi.
For the Portuguese, the failure ofthe gold-diggings and the loss ofthe ivory
trade to the Yao and Bisa were factors in their active participation in the slave
trade. In the late eighteenth century there was a complete change in the nature
of trade among the various regions north of the Zambezi. Slaves became the
main trade commodity that could be sold to the outside world. This change
was pronounced enough to warrant the identification ofthe eighteenth century
with the ivory trade and the nineteenth with the slave trade. Portuguese
participation in the slave trade was remarkable enough in the 1 780s when slaves
from northern Zambezia were exported to the French island colonies of the
Mascarene Islands in the Indian Ocean.
As the slave trade became more lucrative than the trade in ivory, the
Portuguese were joined by the Yao and Bisa. The Kilwa market began to be
dominated by the slave trade from the 1770s onwards and the Yao probably
then began to replace ivory with slaves in their caravans from the interior.
The entire northern zone was also drawn into the long-distance trading
network. Until the early eighteenth century, no part of the region had been
involved in the long-distance trade, although a vigorous local trade had existed
near the shores of Lake Malawi and from Tumbuka country into Nsengaland
and the northern Chewa area. During the first four decades of the eighteenth
century the Tumbuka region was linked to a trading network which stretched
from Katanga in the north-west to Kilwa in the east. The people responsible
for this change were heads of families known as balowoka (those who crossed
the lake).
The most famous of these new men was Kakalala Musawila Gondwe, a
Nyamwezi who for some time had been involved in the trade to the east coast.
Searching for ivory, Gondwe crossed the lake at Chilumba and settled in the
Nkamanga plain near the Luangwa valley, which at that time was heavily
populated with elephants. He established contact with the head of the Luhanga
clan and later married into the Luhanga and other influential families, thereby
firmly establishing himself in Tumbuka society.
Katumbi Mulindafwa Chabinga also came from Nyamwezi territory but
after crossing the lake at Chilumba he went into the mountainous area which
adjoins the Misuku hills and the Nyika plateau and setded at Chigoma in
Nthalire. From Chigoma, Katumbi extended his trading contacts westwards to
Malambo in the ivory-rich Luangwa valley.
Another trader, Katong'ongo Mhenga, whose original home was in the
Ubena-Uhehe region, crossed the lake shortly after Gondwe and Katumbi had
setded. Katong'ongo was looking for land where he and his family could setde,
and this he found in the Henga valley. A man of many skills, he knew how to
produce salt, which enabled him to control the high-quality salt-pans at
Kanembe. He also married into local families and gradually became influential.
He came to be known as Mwahenga, owner ofthe Henga area.
The three immigrants began to dominate the trade of most ofthe area south
of the Songwe and east of the Luangwa valley, and they ended up clashing with
one another. Gondwe felt it was important for him to have free use of- if not
control of the Chilumba ferry in order to transport his ivory to the east coast.
Thus he became involved in the succession crisis following the death of
Katumbi and succeeded in getting a weak candidate accepted for the chief¬
taincy. He then worked out an arrangement with Mwahenga whereby the
latter agreed not to impede his trade. Gondwe made similar arrangements with
the rulen east ofthe Henga valley. Thus by 1800 Gondwe was able to claim a
trade monopoly of the area between the Luangwa and the western shore of
Lake Malawi. In the Nkamanga plain he had achieved political authority and
was thus able to extend his trading empire.
There were smaller polities established in the eighteenth century by families
from the eastern side of the lake, and many of their founders were, like
Gondwe and Katumbi, ivory-traders. Just south of Ungonde, the Mwafulirwa
family founded the Fulirwa state in the area that had been dominated by the
Mkandawire clan. A relation of the Mwafulirwa later went west and finally
setded at Zibang'ombe, where they began to forge the Wenya state, which by
the end of the eighteenth century had extended northward to the southern
borden of Ulambya. At about the same time the area south-west of Wenya fell
under the rule of a Mlowoka family, the Mughogho. This area, Uyombe, was
rich in elephants and, a few decades later, attracted many hunters.
Families of hunten crossed the lake further south and setded not far from the
present Nkhata-Bay boma. They displaced the Phiri rulers and after a generation
became the dominant families in the area.
In the eighteenth century, therefore, the Tumbuka area was dominated fint
by ivory-hunten, then by traders whose control of wealth led to the assumption
of political authority. Political decentralization gave way to the power of
trader-chiefs. Such political influence was not possible in the northern states,
which remained outside the network of long-distance trade for almost a
century after the arrival of the Tumbuka. In the eighteenth century the Maravi
empire was disintegrating and the states of the Tumbuka-Chewa marginal zone
in the south were being left in total control of their own affain. After 1800, all
states and peoples would become deeply involved in the commercial age of
ivory and later of slaves, with the well-known disastrous results.
The beginning date for this volume does not coincide with historical
periodization in this region. An historical age opens in the southern zone
c. 1400 with the coming ofthe Maravi, and in the northern zone c. 1600 with
the migration of the Ngulube. However, the closing date is apt because new
themes dominate the nineteenth century: the changeover in trade from ivory
to slaves, the involvement of all parts of the region in externally manipulated
commerce, the development of Yao chiefdoms in areas formerly under the
Maravi and, finally, the arrival of new peoples the Nguni, Swahili, Kololo
and Europeans who would ultimately seek to wield political authority,
regardless of their disparate motives for coming.
Southern Zambezia
The period between 1500 and 1800 is characterized in southern Zambezia by a

scarcity of records. We have to rely largely on the accounts of the Portuguese
travellen and traders who penetrated the region during the first decade of the
sixteenth century. This chapter will therefore concentrate on three polities: the
Mutapa and Rozvi empires and the eastern Shona states. The Mutapa and
Rozvi polities evolved from the Great Zimbabwe culture (1200 see
Unesco General History of Africa, Volume IV, Chapter 21).
Great Zimbabwe gradually declined from the mid-fifteenth century on¬
wards, as gold production decreased. Great Zimbabwe's decline gave rise first
to an obscure state known as Torwa or 'Butwa', according to Portuguese
sources. Its first capital, situated at Khami and probably founded in the mid-
1400s, was destroyed by fire during the second half of the seventeenth century.
The second capital, at Danangombe, was much smaller than the fint. The
archaeological remains suggest that almost all the prestige stone buildings in the
Torwa state were constructed before 1650.
The second important political development after the fall of Great Zimbabwe
was the emergence of the Mutapa empire at the beginning of the sixteenth
century. The empire comprised heterogeneous Karanga-speaking populations.
In theory it extended from the southern margins of the Zambezi to the Indian
Ocean, but in practice the Mutapa rulers exercised a much more limited
authority outside the plateau country. There is evidence that the kingdoms of
Manyika, Uteve, Barwe and Danda hived away from the empire during the
sixteenth century, although they continued to perform their ritual and tributary
obligations until the rise of Dombo Changamire in the late seventeenth century.
It has been suggested that the Changamires themselves occasionally paid tribute
to the Mutapa emperon during the eighteenth century. Similar developments
occurred on the lower Zambezi, where the Tonga and Sena rulen successfully
resisted all efforts by the Mutapa rulen to impose their political hegemony over
them. These developments reduced the empire to its core, the region spanned
by the Dande and Chidima territories.
The practice of prestige stone-building in the Mutapa empire probably
continued until the sixteenth century. Some of the earliest ruins, particularly
315
Chikova
[
^^^^ ~ZdaÑdT^chidima VTete «1

HT / ^^ f f m Bedza ..
- ^n .Sena -
>±^BÊ \ Quelimane VP ;
R5 . 33fBSE
IAUNGV, | 1 BARWE ^*\ Luabo ^
¡ST *=» - ,
^^^mw«* ' ^
3WMMriTir\M »v
^^2Ei; "-'tiiinawniTf»
WnSS *
«^ ^T ¡ÊÊê Masskesa )
\ /^LyTEVE v y
j SÊr Sofala J
23111 v SD J INDIAN OCEAN
/ MADANDA j\
7\ ^X^
¡Tx^ ^ ^-^s*" "\
V5^S5^
H Land over 1 000m \
0 too 200 miles |

i N
1 i i
0 150 300 km
Inhambane IÄ
22 A Central and south-eastern Africa during the period of the Mustapa and Torwa states
(adapted from S. I. Mudenge, A Political History of Munhumutapa c. 1400-1902,
Zimbabwe Publishing House, Harare, 1988; and Zimbabwe Epic, published by the
National Archives of Zimbabwe, Harare, p. 41)
the Zvongombe complex, represented the early capitals of the Mutapa rulers.
Later capitals were in the form of stockades several metres high.
In the fint half of the eighteenth century, the mutapas lost direct control of
the plateau, and the core ofthe empire shifted to the southern lowlands ofthe
Zambezi bounded by Zumbo and Tete. This loss of political control was the
culmination of a long process of fragmentation which began in 1629, when the
Portuguese defeated Mamvura mutapa, and ended in 1917, when the Mutapa
polity finally disappeared.
Administratively, the empire was controlled at three levels: the capital, the
province and the village. The mutapas delegated authority to chiefs at village
and provincial levels. In the empire's early days, only relatives of the mutapas
were entrusted with these positions. For example, it was customary for a prince
or heir presumptive to govern the Dande lands. Those who, though not related
to the Mutapas, had aided in the process of conquest, were also appointed to
Southern Zambezia 317
positions of authority. As the Mutapas came to feel more confident, in the

seventeenth century, they allowed the villages and provinces to elect their own
leaden. In the capital the Mutapas were assisted by high-ranking officials who,
in exchange for their services, were granted land.
Beliefs and methods of administrative control

The Mutapas manipulated several mechanisms to control the empire. A typical
example was the practice whereby the territorial chiefs had to come and
rekindle their fires annually from the original royal fire, thereby reaffirming
their loyalty to the central authority. This ritual of fidelity was repeated at the
enthronement of a new Mutapa ruler. Failure to perform this ritual was
regarded as an act of rebellion and severely punished.
This was done through an efficient army which has been estimated at 30,000
regulars during the sixteenth century and 3,000 men during the eighteenth
century. These figures suggest that, before the decline of the empire, the
Mutapa rulen were in a position to mobilize many peasants to join the army,
but were much less able to do so when the empire fell into decay during the
eighteenth century.
The Mutapas also manipulated religion to control their subjects through the
close relationship between the monarchy and the spirit world either the
spirits ofthe anceston ofthe emperors themselves or mediums representing the
original owners of the land. The emperor was expected to maintain a close
contact with the powerful dead on behalf of the nation. The system of
observing the cult of royal graves also enhanced the emperor's prestige and
strengthened his control over his people.
The owners ofthe soil, like Dzivaguru, were rain-makers and ritual officen
at the Mutapa royal court. The participation of the emperor in rain-making
rituals was seen as crucial to the economic prosperity of the empire. Religion
was an important mechanism of social control, in an economy dominated by
agriculture.
The most common method of political control was the tribute levied by the
Mutapa emperors in the form of agricultural produce, lion and leopard skins,
ostrich feathen, small and large stock, the breast of every animal killed and, for
an elephant, the tusk on which it fell when it died.
Tribute also took the form of labour rent. According to Joào de Barros, all
the officers and servants of the king's court and the captains of the soldiers,
each with his men, had to help him cultivate his fields or do other work
seven days in every 30. The practice of labour rent was also adopted by vassal
rulers.
When opening a new mine, the Mutapa rulen usually sent their trusted
agents to collect the tribute. The gold-miners erected a shelter to house the
tribute-collectors as well as the tribute itself. The nature of the tribute system
does not seem to have changed until the demise of the Mutapa empire in the
early twentieth century.
These various mechanisms of control partially succeeded in maintaining a

centralized empire at a time when vast distances rendered close supervision of
the territorial chiefs impossible. The army's inability to deal effectively with
rebellions in the far-flung parts ofthe empire, internal political intrigue exploited
by Portuguese traden and civil wan all account for the gradual decline and fall of
the Mutapa empire.
The Portuguese factor

Although the Portuguese arrived at Sofala in 1506, it was only from 1550 to
1630 that they made serious attempts to gain control ofthe Mutapa empire.
Until 1540, trade between the Portuguese and the Shona was conducted on a
non-official basis. In 1540, commercial relations were regularized through the
establishment of a diplomatic-cum-trade mission at the Mutapa royal court,
headed by an officer called the 'captain of the gates'. His main duties were to
transmit presents, requests or complaints from the Portuguese traders to the
Mutapa emperors and vice versa. The Portuguese paid the Mutapa rulers a
tribute, called curva.
At the assumption of his office, each new captain of the Portuguese fortress
of Mozambique was under an obligation to pay to the monomotapa the equiva¬
lent of 3,000 cruzados in cloth and beads for the three years of his office. In
exchange for this tribute, the Mutapa emperors granted freedom of passage to
the traders to travel throughout the empire. If the tribute was not paid within
the proper time, the emperor would order the seizure without compensation of
all the merchandise in his empire. This actually happened in 1610, when one
Mutapa ruler Gatsi Rusere declared a mupeto (forcible confiscation) of the
Portuguese traden' merchandise, which resulted in the raiding and killing of
Portuguese traden.
This system of tributary relations penisted with little change until the second
half of the sixteenth century, when the Portuguese began to win control over
the Mutapa rulers as a result of the aggressive wars they waged in southern
Zambezia between 1569 and 1575, when they succeeded in entrenching them¬
selves in the eastern kingdoms of Uteve and Manyika. In 1575, the Portuguese
concluded a treaty with the king of Uteve whereby the captain of Sofala was
obliged to pay an annual tribute of 200 pieces of cloth to the Teve rulers. In
exchange, the Portuguese traden secured freedom of passage to the kingdom of
Manyika where, they believed, there was an abundance of gold. In addition,
the inhabitants of the Portuguese fortress at Sofala were allowed freedom of
passage into the interior along the Sofala river to buy provisions. A similar
treaty was concluded with the king of Manyika in 1573.
This limited success tempted the Portuguese to make further encroachments
into the Mutapa empire. The rebellions which broke out there between 1590
and 1607 provided the Portuguese with an opportunity to delve into the
complexity of Mutapa politics. Gatsi Rusere, concluded a treaty with them for
military assistance, in exchange for which he promised to cede all his gold,
copper, iron, lead and tin mines to them. In fact, the Portuguese lacked both
the expertise and the manpower to exploit the metals. Moreover, the treaty
turned out to be of limited value, because the Portuguese left Gatsi Rusere to
deal on his own with the civil wan which continued to rage until he died in
1624. Gatsi Rusere was succeeded by his son Nyambu Kapararidze, the
legitimacy of whose claim was challenged by his uncle Mamvura. In the war
that ensued between these two pretenden, Mamvura sought and, in 1629,
obtained Portuguese military assistance. The treaty which he concluded with
the Portuguese after his accession required Mamvura to allow the Portuguese
traden freedom of passage throughout the empire, to expel the Swahili-Arab
traden and to allow Dominican missionaries to preach their religion. It also
required Mamvura to stop the curva, and instead to pay tribute to the
Portuguese thenceforth. After this treaty, the number of Portuguese traders and
adventuren within the Mutapa empire increased. The adventuren seized land,
subsequendy recognized by the Portuguese crown as prazos (crown land). From
being the guests of the African rulen, the Portuguese traders now almost
became the masters of the land, but their individualism and lawlessness led to a
period of chaos and anarchy. Portuguese penetration of the Mutapa empire,
therefore, led to its disintegration and gave rise to the prazo system, as well as
the birth of a new polity, the Rozvi empire.
The prazo system

The acquisition of land by individual Portuguese adventurers proceeded almost
unchecked, with the result that many prazos da coroa were created. They were
owned by Portuguese officials, traden, religious orders and frontienmen.
Th° prazo system was a synthesis of two socio-economic systems. The first
was that of the Shona, which consisted of a ruling oligarchy and peasant
producers; the second, which was superimposed on the Shona system, consisted
ofthe prazeros, ruling as a dominant class over the achikunda (armies of slaves).
An African chief continued to perform his traditional duties but he 'no longer
had absolute authority' (A. F. Isaacman, 1969), as the prazero assumed the status
of overlord.
The relations of production on the prazos were essentially feudal. The

African chief paid the prazero a tribute in kind: butter, locally woven cloth,
honey, tobacco, sugar, large and small stock, ivory and gold dust. In his
position as a de facto chief, the prazero adopted African social and religious
practices and married into chiefly families to overcome his major weakness, his
lack of legitimacy in African traditional politics.
A. Isaacman has argued that the prazo as a Portuguese land-tenure system
was so transformed and adapted to the African situation that it became com¬
pletely African. However, this total Africanization theory is difficult to maintain
when it is realized that the prazo system was always more or less linked to
mercantile capitalism, with the prazeros serving as middlemen in the trade
between Mozambique and India and, later, Europe. Moreover, the internal
organization of the prazos had features which could hardly be called African, a
notable example being the coexistence of the colono (setder) and the chikunda.
In reality, an opposite process, that of the de-Africanization of the traditional
African societies on the prazos, was taking place. As the prazo system responded
to the internal demand for slaves, the prazeros enslaved and sold their African
neighbours and victimized their colonos and slaves alike.
Isaacman has argued that 'in terms of the day-to-day governing of the
colonos, the mambo's position remained virtually unaltered'. It is true that the
mambo enjoyed prestige and obedience from his people by virtue of his
connection with the founding lineages. But ultimately the prazero had the last
word in deciding judicial cases or the appointment of an African chief himself.
Even the right to distribute land had been usurped by the prazero. All that
gready modified the traditional political system. The prazos represented the first
phase of the Portuguese colonization of Mozambique: they were pockets of
political and economic exploitation achieved only through a modification of
African political and social institutions.
The geographical spread ofthe prazos

The prazos were dotted on both sides of the Zambezi river. There were,
however, significant differences between the prazos to the south of the river,
which were Portuguese crown property, and those to the north, which were
not. In theory, the grant of a terra da coroa was restricted to three generations,
the land could be inherited only through a female, it could not be more than 3
leagues long, and the holder was subject to an annual quit rent. The prazo-
holden in the north were free from all these conditions and acquired land as a
result of a direct arrangement with the local chief.
The prazo system was sustained by the work of various categories of slaves.
The top slave was the chuanga, appointed by the prazero because of his loyalty.
Every village had its chuanga, whose primary function was to spy on the tradi¬
tional leaders and to collect taxes and ivory. He also had the job of collecting
information about the African population on the prazo whose size was of
crucial importance to the prazero. For the enforcement of his authority, the
prazero depended on a chikunda, whose main function was to police the local
population, ensure that the laws ofthe prazero were obeyed and deal also with
acts of rebellion. A chikunda army usually ranged between 20 and 30 men on
small prazos, while on large ones it could comprise thousands.
In his report of 1766 concerning the coast of Africa, Antonio Pinto de
Miranda described the main methods by which slaves were obtained. Accord¬
ing to him, slave-hunten stole children and sold them to Portuguese traders,
Swahili-Arab traden and African agents. Some people were sold as slaves or
pawned in times of famine; and convicted criminals, prisoners of war and credit
defaulters suffered the same fate. The picture which emerges from other
Portuguese sources is one of voluntary enslavement.
The decline ofthe prazo system

The prazo system declined during the second half of the eighteenth century for
several reasons. The fint was that the prazero's authority was ill-defined com¬
pared with that of the traditional African ruler. Often this tempted the prazero
to abuse his authority by compelling the African peasants to sell their
agricultural produce to him exclusively and at artificially low prices. He would
also impose heavy taxes on the peasants and commit atrocities with impunity.
The colonos reacted by revolting against the prazeros and even by migrating.
These revolts and migrations led to a decline in agricultural production,
drought and famine. The prazero and his slave armies relied for their food on
what the peasants produced; the chikunda used force to get foodstuffs from the
peasants, and the colonos, again, reacted by migrating from the prazeros to look
for food and security elsewhere.
The slave trade was another factor that led to the decline ofthe prazo system.
The oveneas slave trade increased in the 1640s as a result of the Dutch
occupation of Angola, which forced Lisbon to turn to south-east Africa as a

source. The establishment of a plantation economy in the French Mascarene
Islands in the 1730s created an additional demand for slaves. But the numbers
of slaves exported from southern Zambezia by 1752 were still relatively small:
300 from Zambezia and 200 from Sofala. Towards the end of the eighteenth
century, however, the slave trade gathered such momentum that the prazeros in
the Zambezi valley saw it as an opportunity for wealth. They organized slave-
raids in the Chewa, Nsenga and Manganja territories. They also began to sell
the slaves on their estates, thus undermining the very foundation of the prazo
system. However, beyond the Zambezi valley, the rest of Zambezia was little
affected by the slave trade. The decline of the prazo system south of the
Zambezi was accelerated by the rise of a new polity in the second half of the
seventeenth century the Rozvi Changamire dynasty.
The Rozvi empire

The origins of the Rozvi empire are still obscure. The issue has been further
complicated by the name changamire, which is associated with the founder of
the Rozvi empire in the second half of the seventeenth century, but is already
mentioned in the fifteenth century. However, there seems some consensus on
four points: fint, that the Rozvi were an integral part ofthe historic Karanga of
the Mutapa empire until they broke away in the seventeenth century; second,
that 'the designation Rozvi was first applied to a section of those historic
Karanga who were associated with the rise to power between 1684 and 1695 of
Changamire Dombo I' (S. I. Mudenge, 1974); third, that the creation ofthe
Rozvi empire was the work of Changamire Dombo I; and fourth, that the
appellation Rozvi, which was derived from the Shona verb kurozva (to
destroy), was either assumed by the followers of Dombo out of vanity or was
given to them as a nickname by their victims.
Between 1684 and 1695, Dombo Changamire gradually rose to power in
the north-east. How he did so is still unclear. He was a herdsman of the
Mutapa emperors who rebelled, taking himself and his followen first to the
Core of Rozvi empire
Approximate extent of Rozvi empire
Modern boundary of
Zimbabwe
22.2 Tlie Rozvi empire (adapted from D. N. Beach, Zimbabwe before 1900, 1984,
p. 38, Harare, Mambo Press)
Torwa state in the south-west, then to the kingdoms of Manyika and Uteve in
the east. The Portuguese traders were expelled in 1684 from the kingdom of
Maungwe, from the plateau country in 1693 and from the kingdom of
Manyika in 1695. They fled to Zumbo, where they established a new setdement
in 1710, then returned to their former setdement of Masekasa in Manyika. In
every case, however, there was a tacit understanding that they were under the
authority of the Rozvi Changamire overlords. In this way, the Rozvi rulers
were able to create an empire that stretched as far as Buhera, Bocha, Duma
and the south-eastern highlands in the south, to the sodic soils in the north
and to the lowveld in the south-west. The core of the empire probably
spanned the same area as that of the Ndebele, who succeeded them in the
1840s.
Our idea of the Rozvi capital comes from a nineteenth-century Portuguese

description, although it had probably changed litde since the 1600s and 1700s.
The royal court, according to this account, consisted of three large stone
houses, surrounded by walls of elephant tusks. The Rozvi Changamire had
many rifles which he obtained from Portuguese traders at Sena.
The Rozvi maintained their identity as a foreign élite amidst people who
shared the same moyo (heart) totem, having been their allies during the wars of
conquest ofthe Karanga-speaking people as well as those ofthe Torwa state.
Provincial administration
The Rozvi rulen used methods similar to those of the Mutapas to maintain
their empire. The mechanisms involved the allocation of land, the payment of
tribute, religion and a well-trained army.
The Rozvi rulen distributed land to newly instated chiefs and levied tribute
from them. Their tribute system was organized in a hierarchical pyramid, with
the village as the base and the court as the apex. The principal officer and
commander of the army supervised the collection of tribute throughout the
empire, sending out teams of special tribute-collecton to the various provinces
and villages.
The Rozvi developed a religious system altogether different from that of the
mondhoro prevalent among the northern and eastern Shona. The Rozvi cult of
the oracular deity, Mwari, was based on a belief in a high God who expressed
himself through natural phenomena such as lightning or earthquakes. The
principal officen of the Mwari oracle were the Mouth, the Ear and the Eye,
whose main purpose was to collect information. The Mwari cult was manipu¬
lated by the Rozvi rulen for political ends.
The proverbial efficiency of the Rozvi army calls for some clarification.
Portuguese observers, during the eighteenth century, never ceased to marvel at
both the discipline and the efficiency of the Rozvi armies, which were
generally well equipped with bows and arrows, daggen, assegais, battle-axes
and cudgels. Like most armies in pre-colonial African societies, the Rozvi
armies performed certain rituals which instilled military prowess and self-
confidence before they went into batde. The Portuguese came to believe that
the Rozvi emperor had 'magic oil with which he could kill anyone simply by
touching the person with it' (ibid.), and there is no doubt that this belief played
an important role in Rozvi military victories, especially Dombo Changamire's
victory over the Portuguese at the battle of Maungwe in 1684.
The prowess of the Rozvi armies can also be explained by the thorough
training the soldien underwent. The Rozvi armies were organized and trained
much after the fashion ofthe early nineteenth-century Zulu, Ndebele and Sotho
armies.
Another aspect of the Rozvi army that has been exaggerated is its invinci¬
bility. The defeat of the Portuguese at the end of the seventeenth century, for
example, must be seen in the context of the Portuguese presence in East and
Central Africa during this period. The Portuguese were fighting for their
survival along the East African coast from the 1650s to 1729, when their
fortress at Mombasa was destroyed by the sultan of Oman. In other words,
what the changamires encountered in southern Zambezia was not the full
Portuguese military effort.
The Portuguese military effort in southern Zambezia was in fact based on
the prazeros, themselves badly disunited by rather conflicting economic

interests. If these faetón are not taken into account, it is easy to exaggerate the
efficiency ofthe Rozvi military machine.
The economy
The dominant branches of production among the Shona societies were agri¬
culture, animal husbandry, hunting, gathering, fishing and metal-working,
along with long-distance and inter-regional trade.
Agriculture
The Shona cultivated three main cereals: finger-millet, drought-resistant
bulrush millet and various sorghums. The unit of production was the house¬
hold, and there is no solid evidence that the division of labour was rigidly based
on sex. What little evidence there is suggests that the situation differed from
society to society. In some, the men performed the heavier tasks and the
women the lighter, more in the spirit of co-operation than in a strict division of
labour. In othen, however, such as that ofthe Hlengwe ofthe lowveld, there
was some division of labour, with the adult males devoting most time to
trapping and hunting game while the juveniles herded cattle and the women
cultivated the fields.
The ploughing season stretched from September to November, with the hoe
as the basic implement. Shifting cultivation was practised, whereby the old land
was abandoned whenever it became exhausted and new land opened up. The
opening of new land involved clearing the bush, burning it and allowing the
ash to fertilize it. Shona agriculture, like most pre-colonial agrarian economies,
was adapted to local conditions.
The period 1500 witnessed the piecemeal introduction into southern
Zambezia of new crops from Europe and Asia, mainly along the coast and in
regions where the Portuguese had built their settlements. In Sofala and on the
lower Zambezi, the African peasants cultivated rice and yams, although these
never became their staple diet. They also cultivated sugar cane to eat rather
than to make sugar, as they lacked both the expertise and the appliances. The
African peasants also planted many fruit trees from India and Portugal, such as
figs, pineapples, guavas, pawpaws and oranges; they also grew melons, cucum¬
bers, sweet potatoes and lemons. The cultivation of exotic fruits was stimulated
by demand from foreign traders who exported lemons, in particular, to India.
There is evidence that wheat was being cultivated in the kingdom of
Manyika during April and May and that by 1778 peanuts were being cultivated
there. Two grain legumes were grown everywhere, bambara groundnuts and
cowpeas.
None of the crops introduced from Europe and Asia ever constituted the
staple diet ofthe Shona. The case was different with maize, which was intro¬
duced into southern Zambezia during the eighteenth century. This gradually
caught up with millet and sorghum to constitute one of the staple foodstuffs
ofthe Shona in the twentieth century, particularly among urban populations.
The Nyanga and the lowveld

The agrarian economy of the Nyanga and lowveld areas was in many ways
different from that ofthe plateau. The Nyanga highlands were characterized by
poor soils and steep slopes. The greater part of northern Nyanga was terraced
with dry stone walls about 1 metre high. These terraces were made on the
hillsides and their purpose was to conserve the soil and dispose neady of the
stones. There is some evidence that a limited amount of irrigation was prac¬
tised, but only in a supplementary role, as the Nyanga area usually enjoys a fair
amount of annual rainfall. It is likely that the dry-season irrigation of vegetable
gardens, maize and bananas, which is still being practised in this region, is an
agricultural tradition of eighteenth-century origin.
The lowveld economy

The lowveld economy was different. Here a local people, the Hlengwe, prac¬
tised mainly hunting, gathering and fishing and agriculture only to a limited
extent. The lowveld has very low annual rainfall, so that gathering played an
important role in the lowveld economy, as it does even at present. The fruits
gathered were mainly nkanyi and mahanga, the sap of which was used to make
an intoxicating wine, kwaka, and the baobab fruit. These fruits were gathered
by women and children while the men hunted big game. The women and
children killed mice and gathered large silvery caterpillan and locusts.
The Hlengwe were, as they still are today, excellent hunten and there were
many species of game in the lowveld. Fishing was also an important occupa¬
tion. The lowveld riven, then as now, were rich in fish. The Nyanga region is
especially famous for its trout.
Hunting and fishing were not a monopoly of the Hlengwe of the lowveld.
In various parts of the Mutapa empire, men hunted coundess species of
animals, including lions, tigers, leopards, rhinoceroses, elephants, buffaloes,
wild cows, elands and wildebeests, to cite only a few. Guinea-fowl were found
in their hundreds in the bush close to the fields. The professional hunters used
traps and traditional weapons (spears, clubs, and bows and arrows). With the
introduction of firearms by the Portuguese in the early sixteenth century, the
hunters' efficiency improved.
The Hlengwe were not solely hunters. That they were able to differentiate
between various kinds of soil demonstrates that they paid close attention to
agriculture. The Hlengwe differentiated soils mainly on their ability to retain
moisture. They called the sandy soils where they built their homesteads nthlava.
This type of soil was good for growing cucumben, gourds, cowpeas, bambara
beans and groundnuts. They also grew millet and sorghum there, but in or near
the valleys. They called the basalt soils tsovolo: these were fertile soils that retain
moisture for a long period, on which they grew maize and a variety of
vegetables. It would seem that the Hlengwe never lacked land and that they
were constrained only by the erratic rainfall.
Portuguese observen expressed contradictory views on the efficiency of the

agrarian economies of the Shona societies between 1500 and 1800, while
historians have tended to over-generalize about famines both in pre-colonial
and colonial southern Zambezia. David Beach has characterized Shona pre-
colonial agriculture as 'useful but dangerously static'. This, according to him,
was because 'no selection of crops and soil and no expertise in predicting
shangwa (famine) could ward off these disasters, whether caused by climate,
locusts or other blights'. But the Shona societies were not 'static'. They had
contacts among themselves as well as with other neighbouring African societies
the Swahili-Arabs from the tenth century or earlier and the Portuguese from
the beginning of the sixteenth century. For the Shona, these contacts repre¬
sented efforts to modify their material conditions through barter and exchange.
Furthermore, the Shona peasants were adopting new crops throughout the
period from 1500 to 1800. All this makes it difficult to describe the pre-
colonial agrarian economy as static. The agricultural history in the pre-colonial
period is a story of innovation rather than stagnation.
Moreover, the idea that the Shona peasants were helpless victims of droughts
and famines has become dated. There were several methods of fighting famine,
such as bartering salt, meat, fish, tusks, mats, pots, baskets, spears, arrows and
jewels for grain. The Shona also obtained grain during famine through a system
called mukomondera whereby one borrowed grain on the understanding that it
would be refunded without interest in the following good season. Finally,
hunting, fishing and gathering did not disappear after the introduction of new
crops; they constituted important means of combating famine.
Richard Mutetwa has shown that the problem of storage was not serious,
because the three Shona staple crops, millet, bulrush millet and sorghum, could
store very well for more than three or four years. The Shona peasants also took
precautions to preserve their grain from humidity, carefully plastering their
grain bins to make them airtight. These bins were placed on rocks or poles
high enough to escape the termites. Parallels with what happened elsewhere in
Africa concerning the efficiency of pre-colonial agriculture suggest that there is
little justification for suggesting that Shona agriculture was unable to satisfy the
needs ofthe peasants.
Stock-raising
Stock-raising was an important branch of production in both the Mutapa and
Rozvi empires. Stock-raising included the raising of sheep, goats and catde.
These animals provided meat, milk and manure, which farmers either used or
sold. Possession of cattle also performed a social function in that it bestowed a
social status upon an individual. Portuguese documents stress the central role of
cattle in the economy of both the Mutapa and Rozvi empires. Large herds of
catde flourished, particularly in the highveld, where they were not threatened
by the tsetse fly. There is every likelihood that the Mutapa and Rozvi
Changamire rulers practised transhumance, moving cattle from the highveld to
the lowveld during the dry season when water and salt were scarce and pasture
tended to be poor on the highveld. Transhumance also enabled pastoralists and

cultivators to meet and exchange their respective products: animal products for
grain. Conflicts occasionally erupted between the pastoralists and the cultivaton
whenever cattle destroyed the cultivators' crops. This in turn led to a shift in
the orbit of transhumance or migrations to entirely new areas which is how
pastoralists became colonists.
Neither archaeology nor the Portuguese documents tell us much about
cattle management and their distribution from the sixteenth century to the end
of the eighteenth century. Portuguese sources continually point to the fact that
southern Zambezia was rich in catde and other animais. In the early 1500s,
Diogo de Alcacova recorded that 4,000 hornless cows were paid as tribute to a
Mutapa ruler by a ruler of Butwa in the south-west. In 1569, Father Monclaro
likened the large size of catde in Butwa to the large oxen of France.
Besides catde, the African peasants reared goats, sheep and chickens for food
and exchange.
Mining of metals: iron and copper

A large amount of iron, copper and lead was mined throughout the Mutapa
and Rozvi empires, although the available literature leads one to believe that
African peasants were interested only in mining gold. The African peasants
extracted iron from which they made objects such as hoes, assegais and axes.
The Njanja of Wedza acquired a reputation as hoe manufacturen, selling their
hoes as far afield as Manyika, Bocha, Burbera and Ndau country. They also
mined copper from which they made bangles which both men and women
wore round their arms and ankles. A geological survey carried out in 1952
showed that four-fifths of total copper production in Zimbabwe came from
Duma. A considerable quantity, however, also came from Urungwe in the
north-west.
Salt
The salt industry was very important for the Shona economy between 1500
and 1800, especially in regions that did not enjoy sufficient rainfall to grow
crops and raise cattle, such as along the middle Save. This region was also rich
in clay, and the local inhabitants took advantage of this resource to specialize in
pottery manufacture. Salt and pots were exchanged for grain, particularly
during famine yean.
Cloth
Textile production was also an important activity among the Shona peasants of
southern Zambezia. Cotton-growing and weaving flourished mainly on the
eastern bank of the Zambezi river. The Shona wove a cloth from this cotton
and also from the bark of the baobab tree. The cotton-growing lowveld area
was complemented by the highveld region which abounded in cattle, and this
largely determined the pattern of inter-regional trade.
Gold and silver

The existence and extraction of silver from the sixteenth to the eighteenth
century was a matter of great speculation among the Portuguese authorities,
but throughout the period under review, and even later, no silver mines were
located. Silver was in fact obtained as a by-product of gold-mining, and there
was plenty of gold on the highveld, in Butwa in the south-west, and in
Manyika and Uteve in the east.
The amount of gold present in southern Zambezia was gready exaggerated
by the Portuguese during the sixteenth century. It was claimed, for example,
that the whole of 'Mucaranga' was nothing but a vast gold mine and one had
only to dig to find gold. Nevertheless, there was indeed plenty of gold in the
Mutapa empire and, whatever they said, the Portuguese saw their hopes largely
fulfilled until the second half of the seventeenth century.
Historical references suggest that mining was not a year-round activity for
the Shona, except in the kingdom of Manyika in the east, where the extraction
of gold was carried out year-round because the rivers there are perennial.
Shona mining activities were concentrated in the months of August, September
and October, the months when land was cleared. Another reason for mining
during those months was that minen could sink shafts fairly deep as a result of
the seasonal fall in the water-table.
Mining operations followed a pattern prevalent in much of southern Africa.

The basic tools used to break out the ore were hammerstones and iron gads as
well as a type of crowbar consisting of a piece of iron inserted into the end of a
heavy stick. The method for gold-reef mining was fire-setting as can be
deduced from the large quantities of charcoal which Roger Summers found in
many of the ancient mines. In one he also came across troughs containing
charcoal standing against the face of the reef. According to him, they formed a
hearth concentrating the heat against the hard rock face. By cooling the heated
rock rapidly with water, the Shona were able to crack the rock face. The pieces
thus collected were stacked amongst piles of firewood and roasted, after which
the burnt quartz was crushed and the gold washed out of it.
With alluvial gold, the recovery process was village-based. During the
mining season, groups of 400 or more miners, with their women and children,
assembled in a designated place under the command of the village head. They
washed the alluvium in wooden bowls, and the gold that was recovered was
packed into hollow reeds or quills ready for exchange. The peasants preferred
gold-panning to reef-mining as the workings were shallow and they were not
expensive. Nor did they expose minen to the hazards of tunnels through soft
soils that regularly caved in.
The Mutapa and Rozvi rulers exercised tight control over the production of
gold within their empires. It has been estimated that almost 50 per cent of gold
production within the Mutapa empire was appropriated direcdy by the ruling
élite. If a person stumbled upon a gold deposit, he was required to cover it up
and report it immediately to the local chief. Failure to comply with this law
was punishable by death. The Mutapa and Rozvi rulers did not want the
Portuguese to know the location of the gold mines, as this would tempt them
to conquest: moreover, control of the mines enabled them to dictate better
terms of trade.
The decline of gold production during the seventeenth century

The dangerous nature of gold-mining explains the low production and also its
decline during the seventeenth century. Moreover, once the water-table had
been reached, it was not possible to mine deeper without the aid of a new
technology capable of pumping out both the water and the mud. Finally, the
price the Portuguese traders paid for gold was too low to induce the peasants to
risk their lives.
One aspect that has not been examined in detail was the impact of war on
gold production. The first half of the seventeenth century witnessed the height
of Portuguese aggression against the Mutapa empire. Portuguese intervention
in internal Shona politics resulted in wars among the Mutapa rulers themselves
and also between the Mutapa rulen and their vassals. These wars resulted in
both temporary and permanent migrations and famine, even in areas rich in
resources and with favourable weather patterns. The second half of the seven¬
teenth century was characterized by the even more devastating changamire wars
which resulted in the reorganization of populations into several states. Warfare,
therefore, could be blamed for the disruption of gold production in Shonaland
during the seventeenth century in the same way that a natural calamity disrupts
agriculture by reducing the amount of land available.
In addition, the methods adopted by the Portuguese traders alienated them
from the peasants largely responsible for producing gold. In the early 1500s,
African peasants used to take their produce direcdy to the Portuguese feiras in
the interior. Later in the century, however, the Portuguese sent middlemen
into the interior who took merchandise to the peasants' doon. They sold
merchandise on credit to the peasants, who often defaulted, with the result that
the defaulter became the trader's slave. Relations between the Portuguese and
the African peasants deteriorated to the point where they threatened village life.
In the mid-1 600s, the king of Uteve ordered his people not to extract any
more gold but 'to till the land and grow food in order to become rich and have
more peace and quiet'. Social factors, therefore, were probably significant in
the decline of gold production during the seventeenth century.
Trade
Trade records also provide historical evidence ofthe diversity of Shona peasant
economies. The peasants' main aim was to produce consumables rather than
exchange goods. But the production of commodities inevitably resulted in the
exchange of products among the peasants themselves and between peasants and
craftsmen, and this led to the emergence of regional trade.
The Swahili traders established several bazaars in the Mutapa empire. The
fint bazaars in Sofala and its immediate neighbourhood were probably estab¬
lished during the tenth century when the Swahili-Arab traders started trading
with African peasants in the Mutapa empire. The bazaan were operated on
Mondays, with the African peasants exchanging their agricultural produce for
beads, cloth and other exotic items. The demand for foodstuffs must have
increased when the Portuguese established a fortress at Sofala in 1506, which
became an entrepôt where African peasants came to sell their produce to
Portuguese traden.
The establishment of this fortress must be seen in the context of Portuguese
commercial activities from 1498. Portugal's main aim at this time was to
occupy all strategic points along the Indian Ocean. The fortresses of Sofala and
Kilwa were built in 1506 to protect the gold trade, while three othen were
built to control the pepper trade. On arrival at Sofala in 1506, the Portuguese
found the Swahili-Arab traden well established, and rivalry between the two
groups, characterized by several military encounten, consequendy developed.
The Portuguese emerged victorious and the Swahili-Arabs fled north to various
strategic points on the Zambezi river, from where they continued to under¬
mine Portuguese commercial activities. The Portuguese reacted swiftly by
ousting them from most of these places and, in 1531, establishing Sena and
Tete as their main trade centres. Portuguese trading activity accordingly shifted
northwards, thus reducing the importance ofthe Sofala region.
The defeat of the Swahili in 1512 and the occupation of Sena and Tete
brought to an end the independent trade of the Swahili-Arabs. But as the Portu¬
guese had neither the skill nor the manpower required, and as the Swahili-
Arabs were anxious to continue as traden in the region, a natural though
uneasy trading alliance developed between them. Throughout the seventeenth
century, the Swahili-Arabs were the main agents of Portuguese trade in the
interior. African vashambadzi (traden) had been acting as middlemen long
before the advent of the Portuguese, and their role was formally institutional¬
ized within this alliance.
The defeat of the Swahili-Arabs left the Portuguese traders without major
competition along the Zambezi trade route and in the interior. Initially the aim
of the Portuguese crown was to monopolize the entire trade in Sofala and the
interior, but this proved impossible, as greed drove traden inland to make
independent agreements with African rulen.
As with the Swahili-Arab traders, the African peasants exchanged their
produce for a variety of exotic items. Peter Garlake's archaeological excavations
of the feiras of Luanze, Dambarare and Rimuka indicate that beads were the
most popular commodity during the early sixteenth century. The most sought-
after beads in all kingdoms were the red and black ones known as Cambay
beads. There was also a variety of imported beads, including those of 'coral,
crystal, pewter, jet, amber and blue Venetian glass' but these were unpopular.
Convenely, a locally made variety known as caracoes (small stone beads) flooded
the market between 1516 and 1518. Cloth was another important trade item,
particularly the brightly coloured cloth which was measured by arm-lengths.
In exchange for these articles, the Shona brought to the fortress's market or
to the feiras sorghum, maize, bambara groundnuts, cowpeas, baskets, mats, pots,
chickens, eggs, honey, trapped animals and birds, and a wide range of vegetables
and wild fruits. Most smiths, other craft-worken and minen exchanged their
goods either among themselves or with the Portuguese traden. The farmen
found a ready market for their catde, goats, sheep and pigs. The Sofala fortress
and the feiras in the interior struck deep roots into Shona society and became
an essential part of its economic and social framework.
Feiras in Shona country

The period 1575-1684 witnessed a change in the pattern of trade between
African peasants and Portuguese traders. The Portuguese traders consolidated
their commercial and military victory over the Swahili-Arabs by adapting and
transforming the bazaars into Portuguese feiras. In time, the feiras became the
main focal points of Afro-Portuguese intercourse. Each feira had a fortress, a
garrison of 10 soldien, in theory a church with a priest, and a capitao-mor.
Some feiras were administered by the government ofthe Rivers of Sena, while
othen were privately owned. The capitäes-mores bore some resemblance to the
guards of the medieval Portuguese feiras. The position of capitäo-mor in
Manyika, Butwa and Karanga country was comparable with that of the capitao-
mor dos Banianes on the island of Mozambique. Their payment was regarded
only as a subsidy and it was the policy of the Portuguese administrators to
appoint only wealthy individuals from the Portuguese setdement of Sena to
such posts. The duties of the capitäes-mores involved taxation, price control,
arbitrating between Portuguese and African traders, granting licences, pro¬
tecting caravans, raising soldien and enforcing laws relating to weights and
measures. These powers and duties were gready modified as capitäes-mores
interacted with various African rulers.
The mam. feiras

The feira of Dambarare was acclaimed the best of all the feiras in the Rivers of
Sena, where almost all the rich merchants of Sena traded. It was three days'
journey from that of Angwa, where there was much gold but few residents, as
it was so far from both Sena and Tete.
In Uteve, the Portuguese held an annual feira in Bandire to purchase articles

from the interior. This feira lasted from the sixteenth century to the early
eighteenth century, when the Teve authorities stopped it. The methods used by
the Teve kings to control Portuguese trade were somewhat different from
those of other Shona societies. In Manyika, for example, feira trade was con¬
trolled penonally by both the kings and the prince or princess of the area
where the feira was held. But in Uteve the feira was operated largely by an
inyamasango (village head) under whose jurisdiction the administration of
Bandire fell.
The Portuguese do not seem to have created any feiras north ofthe Zambezi
until the early eighteenth century, when Zumbo and Michonga were estab¬
lished. Two reasons may account for the late development of these last two
feiras: the discovery of bares (mines) north ofthe Zambezi during the eighteenth
Zambezi.
Land over 1200-2400 metres
Land over 900-1200 metres
Land below 900 metres
22.3 The main feiras in the sixteenth and seventeenth centuries (adaptedfrom map 3 of
H. H. K. Bhila, Trade and Politics in a Shona Kingdom. The Manyika and their
African and Portuguese Neighbours, 1575-1902, London, Longman. Adapted by kind
permission of Longman Group UK Ltd)
century, and the need for the Portuguese to look elsewhere for trade after their
expulsion from Shona country by the Rozvi emperor, Dombo Changamire,
between 1693 and 1695.
The expulsion of the Portuguese from their feiras by the changamire forced
the Portuguese to revert to the trading methods they had used during the
sixteenth century. They were, however, able to revive the feira of Masekesa in
Manyika, in 1719, but not the one in Uteve. The system of having one officer
to co-ordinate Portuguese relations with African rulers was no longer possible
in the case of the Mutapa emperon as there was no longer any effective central
authority. It is therefore necessary to examine briefly relations between the
Portuguese and some important African rulen during the eighteenth century.
In the east, the king of Uteve did not allow the Portuguese to mine gold in
the bares in his kingdom. His subjects, however, could buy cloth and beads
from the Portuguese settlen at Sena. Even in this trade, the Portuguese traders
expended much of their profit in trying to buy safety. Teve policy during the
eighteenth century was decidedly anti-Portuguese. In the kingdom of Manyika,
Portuguese traders were allowed freedom of passage throughout the land, but
their trading activities were strictly controlled by the rulers, to whom the
capitäo-mor and the Portuguese merchants paid a regular tribute.
Similarly, to keep the feira of Zumbo and the trade route to Butwa safe, the
Portuguese traden paid a regular tribute to the changamire. He used to send his
envoys to the feira of Zumbo, asking for additional gifts over and above the
usual tribute. Despite all this expense, the trade route between Zumbo and
Butwa was not completely safe. The changamire's famous raid of 1756 in
which he confiscated property worth 1,000 pastas (800,000 cruzados), was a case
in point. The Zumbo trade was also compromised by recurring civil war and
famine during the 1760s. Yet the changamires valued their trade with the
Portuguese. The Rozvi rulers sent expeditionary forces on three occasions in
1743, 1772 and 1781 to protect the feira of Zumbo from attacks by the
neighbouring African rulers.
The nature of relations between the Mutapa and the Portuguese appears to
have changed litde during the eighteenth century. The Portuguese continued to
maintain a garrison of 20 soldiers, a capitào-mor, a lieutenant, a captain-general
and a Dominican friar at the Mutapa royal court. Despite these arrangements,
trade between the Portuguese and the Mutapa rulers continued to decline
during the eighteenth century. This decline has been explained in the context
of developments in southern Zambezia during the eighteenth century. First,
there was a gold rush north of the Zambezi with individual Portuguese
adventuren discovering bares and starting to mine gold themselves. Second,
gold production virtually stopped in the eighteenth century. Third, both the
African and Portuguese traden on either side of the Zambezi river took to
elephant-hunting for ivory.
The ivory trade played an important role in the Mutapa empire. A
sixteenth-century Portuguese account tells us that elephants used to 'go about
in droves almost like a herd of cows' and that 4,000 or 5,000 of them died
every year. Elephant-hunting was less subject than gold to control by African
rulen, as the hunting usually took place in remote areas. However, litde is
known about its organization; there are some indications that much eighteenth-
century elephant-hunting took place in Buhera and Rimuka.
Conclusion
Portuguese penetration of southern Zambezia eroded the power of the indige¬

nous ruling class and facilitated direct forms of peasant exploitation by Portu¬
guese mercantile capital and, later, British industrial capital.
The Portuguese merchants were largely concerned with trade in gold and
ivory and, consequendy, the main commodities of regional trade iron, salt and
copper were relegated to second position. But long-distance trade and regional
trade stimulated each other, and this led to the emergence of an African
merchant class, the vashambadzi. For long-distance trade to flourish, there had
to be gold and ivory and porten, food to feed these carriers, hoes to cultivate
the crops and iron to make the hoes. The vashambadzi transported not only
ivory and gold but also salt and iron hoes, which they exchanged for food on
the way. The African cultivaton who lived near the trade routes were gradually
transformed as they grew more and more surplus agricultural produce for sale
to the traders and their porten. Needless to say, the peasants, who were mostly
ignorant of the international value of the commodities they sold, were
ruthlessly exploited by the Portuguese merchants.
Dependence
and interdependence:
southern Africa,
í 500-1 800
Between 1500 and 1800, much of southern Africa was transformed. New com¬
munities established themselves in the region; many pre-existing communities
changed their way of life or their location, and the relationships which grew up
within and between communities were radically different from any before.
Many of these drastic changes were because of changes in southern Africa's
external linkages. When the fint European, Vasco da Gama, rounded the Cape
of Good Hope in 1497, southern Africa had only the most tenuous links with
the rest ofthe world, but by 1800 the region was firmly enmeshed in world¬
wide patterns of trade and strategy. It is worth considering the changing inter¬
national context itself before trying to evaluate its effect on southern Africa.
In 1500, neither sub-Saharan Africa nor the Americas had been drawn into
regular relationships with the rest of the world. International trade was par¬
ticularly concerned with Europe and Asia, and it was conducted almost entirely
overland, at very great expense. To the Portuguese, who fint developed the sea
route round the Cape, and to the Dutch, English and French, who followed
them round the Cape, southern Africa was mainly a dangerous navigational
hazard. The mapping of its coastline gradually reduced its isolation, but the
consistent lack of interest from Arab and European traders is worth looking at.
During the late Iron Age, the temperate regions ofthe southern hemisphere
were not very attractive to international trade. With the low population density
and the rudimentary tools in use, it was difficult to produce a regular surplus of
food, even in regions which permitted cultivation. The Arabs, who conducted
a flourishing trade along the east coast of Africa, were not tempted further
south because the gold, ivory and slaves which bulked large in their cargoes
were much more readily available on the tropical coasts. Similarly, Europeans
travelled round the Cape for 150 years before taking any real interest in
southern Africa.
If seaborne contacts were negligible, land contacts were not much more
important. Many communities in the south had historical and linguistic ties with
the north, but these had litde significance in daily life. The Khoi, who herded
animals, caught fish and gathered food along the southern coastal belt, had
practically no contacts with the north. The same was true of the San, who lived
335
in the hinterland. The speaken ofthe Nguni languages, mainly east ofthe moun¬
tain chain that divides the region, had litde regular contact with their northern
neighboun. In the far west (in today's Namibia) the Herero and Ovambo had
close linguistic ties with each other and with their northern neighboun, while
the Tswana and Sotho, who lived in the centre, traded to the north from time
to time. As late as 1500, these societies were effectively independent ofthe rest
ofthe world: their external links were sporadic and marginal.
When Europeans did begin to interest themselves in the region, they acted
vigorously. In the mid-1 600s, a wholly new community was planted at the
Cape by the Dutch East India Company, which continued to see southern
Africa simply as a way-station to the East. The regional consequences of
establishing the Cape setdement were much more profound than the company
planners intended. The region as a whole was securely linked to Europe and
Asia, although it was not yet resolved whether the Dutch (who held the Cape
from 1652 to 1795) or the British (who seized the Cape in 1795) would be the
main agents. As the new settlement expanded into its hinterland, it established
new relationships, marked by dominance and dependence. This marked the
advent of capitalist relations of production, together with colonialism and
imperialism, which would transform southern Africa more abruptly and more
thoroughly than any other part of sub-Saharan Africa.
It is possible to reconstruct the conditions and events of the sixteenth century,
but it is difficult to do so objectively. Wherever hunting and gathering commu¬
nities have come into conflict with cultivaton during the past two centuries, they
have been destroyed. Cultivaton have survived in sufficient numbers to force
themselves upon the attention of historians: nomads have been less fortunate.
Some very interesting relationships prevailed in southern Africa in the
sixteenth century that defy all notions of historical inevitability. Environmental
differences permitted a degree of specialization between communities. Hunten
were not evolving into herders, nor herders into cultivaton, but rather co¬
existing, despite the hostilities which sometimes marked their interaction.
The western half of southern Africa, including present-day Botswana and
Namibia, enjoys less than 40 centimetres of rainfall each year. The only exception
is the hinterland of the Cape peninsula, which enjoys good winter rain quite
regularly. In the fifteenth century, without dam-building and food-storage
techniques, the inhabitants of this area could not have practised agriculture.
Their commitment to hunting and gathering or herding is evidence of sensible
adaptation.
The origins and history of the San hunter-gatherer groups are shrouded in
myth and misunderstanding. The great divenity of languages they spoke in the
sixteenth century is evidence of many centuries of adaptation within the western
half of southern Africa. The San had reliable means of subsistence as long as
human population density did not outgrow environmental resources. Hunting
and gathering made them independent of other societies, but they interacted
with other people. Sometimes a group of hunters would attach themselves as
clients to herding patrons, obtaining dairy products in exchange for game or
Dependence and interdependence: southern Africa, 1500-1800 337
"*-. > i
150 300 miles
A °
\ 1 l 1
\
( *
\ 0 150 500 km \
3 #r^
v
HERERO
and
OVAMBO
f i \*Delagoa Bay ^i
i
/
y SOTHO
*->. !~\ * 0 fi C 1 ^
ATLANTIC
OCEAN KORA \
Orange
INDIAN
j S A N TEMBU
OCEAN
Cape Town V t\ n 0 /
Cape of Good Hope ^"T*?* -** Aigoc Bay
23 .1 Southern Africa in the sixteenth to eighteenth centuries
advice on animal movements. This form of livelihood relied absolutely on

mobility, which imposed restraints on population growth and made it
impossible to accumulate property, as it could only hamper the band's speed.
Possession of catde meant that the Khoi had to cling to well-watered areas
rather than roam the arid regions. Since oxen could carry baggage, the Khoi
could build tents and transport them whenever they moved. The presence of
immobile old or young people was no longer the handicap it had been. Private
property could also be accumulated, and within these pastoral societies a
measure of stratification developed, differentiating them from the more egali¬
tarian and property-less San. The need for social control did not require the
Khoi to develop a formal structure of political authority; but individuals did
exercise some authority among their clans.
Herding may have offered a higher level of consumption than hunting
during good times, but herds were vulnerable to drought and epidemics, as well
as theft and confiscation in war. When such disasten occurred, the Khoi often
had to hunt for a living, at least until they could build up fresh herds of stock.
Near the Cape peninsula, where hunting opportunities were limited, dis-
possessed herders took to the seashore, making a living by fishing. Life may
have been unstable, but they were not destitute.
There could be no clear ecological boundary between Khoi and Xhosa, as
both setded farming and herding were possible in the areas where they met.
Indeed, interpénétration seems to have been common. Individual Xhosa
abandoned setded farming for herding and sometimes hunting. Whenever a
crisis struck the Khoi, they could become clients of established Xhosa nearby,
because agricultural production recovers more rapidly from famine conditions
than stock-farming. Robert Ross demonstrates how the Gonaqua,
Gqunukhwebe and Ntinde were former Khoi who became mixed farmers and
were incorporated en masse into Xhosa society. It is only with hindsight that
Khoi pastoralism seems a transitional and obsolete means of livelihood. For
most of the period under review, the Khoi interacted intensively with other
specialized societies but without sacrificing their own life-style.
The late Iron Age was well established among the Nguni and Sotho peoples
by 1500. Iron was commonly employed and quite widely processed, although
not all tools and weapons -were of metal. Hunting and herding societies had litde
need of iron and could rely entirely on stone, bone, wood and fibres. It was also
possible for mixed farmen to manage without iron. In southern Africa, some
communities practised both agriculture and pastoralism with very few iron tools,
for almost 1,000 yean. However, the abundance of iron in the late Iron Age
must have made existing processes much easier and more productive. Farmen are
much more efficient when iron tools are available for clearing, and they have
more time available for domestic crafts and for more extensive pastoralism.
Archaeological excavations suggest that domesticated animals and pottery became
more numerous during the late Iron Age, and that implies a more comfortable
standard of living and a heightened division of labour within each community.
Travellers' accounts of the life of Nguni societies east of the mountain range
between 1500 and 1800 are uneven, and a reconstruction of those years relies
on the study of material remains supplemented by such oral accounts as have
survived. The material basis of fife was grain and livestock. The traditional
reliance upon sorghum was modified in the eighteenth century by the adoption
of maize, which gave higher yields. Stock-farming could be conducted best by
the use of a combination of fixed pastures and seasonal transhumance to take
advantage of the differing qualities of the veld. A general decline in rainfall in
the first half of the eighteenth century and a serious drought at the end of the
century would have intensified competition for those pastures which were
usable even in dry years.
Human population densities appear to have been much lower than we are
now accustomed to. J. B. Peires estimates that there were 100,000 Xhosa-
speaking people by 1800, despite the integration of ex-Khoi. There cannot
have been serious population pressure on land at any time between 1500 and
1800. This is important in any consideration ofthe political systems recorded
for the Nguni. Their essential constituent, and the irreducible unit of viability,
was a household of stock-owning and land-using people. The household could
behave as an independent unit, trading and exchanging commodities and wives

with other households, almost at will. In practice, each household was linked
through patrilineal kinship to a lineage, and hence to a clan. The leader ofthe
royal clan, however, had few sanctions with which to enforce his authority
over other Xhosa-speaken, except the manipulation of marriage links and gifts.
The ability to bestow wives, gifts and land depended ultimately on the royal
household's ability to out-produce other households and clans. But there was
no particular barrier to a commoner becoming wealthy in stock and wives, so
that the margin favouring the chief was not always wide. And against the
centralizing ambitions of the chiefs, commoners also had ambitions. The
incessant wrangles within dominant Xhosa clans and, presumably, among other
Nguni societies were undoubtedly encouraged by thoughtful commonen who
sought to paralyse the central authority. These constraints on the evolution of a
despotic type of administration were not broken until the end of the 1700s.
Until then, daily life seems to have been relaxed and usually comfortable, if we
are to believe the reports of the occasional white castaways who found refuge
among the Nguni.
Even less is known about the Sotho and Tswana peoples west of the
Drakensberg, where lighter rainfall encouraged a greater emphasis on herding.
Sotho and Tswana must have interacted with Khoi, San, Nguni and Shona, as
there was small-scale trade in all these directions.
Like the San and the Khoi, the southern Bantu communities interacted with
other specialized communities but were not transformed by that interaction.
The different African groups were, it seems, capable of long-term coexistence
without one imposing its mode of production on another.
The historiography of southern Africa is dominated by accounts of the his¬
tory and expansion of the Dutch settlement at Cape Town. In fact, the
Portuguese had been the fint Europeans to establish a permanent setdement in
the region, and it is easier to grasp the revolutionary quality of the Dutch
setdement ifit is contrasted with the earlier Portuguese setdement.
Portugal in the sixteenth century was feudal, small and poor. Its oveneas
expansion was organized by the monarchy as a trading monopoly to India, and
was buttressed by the Church, which had a pervasive influence on the
administration. By 1510, the centre of overseas rule was Goa, and the sea route
was protected by forts commanding the Mozambique channel from Delagoa
Bay to present-day Tanzania. With superior ships and firearms, Portugal had no
difficulty in conquering the Arab and Swahili ports, but more was required
than military superiority to govern the region. This was a maritime empire,
tentative in its territorial control. It was also an essentially feudal empire in
which important and lucrative offices were sold by the crown for an annual fee.
It was populated by Portuguese criminals, non-Portuguese traden and
adventuren. Profit was to be had through the gold and ivory trades, and later
the slave trade, recognizing the authority of existing African rulen and pro¬
voking in them the most anti-social behaviour. By the end of the sixteenth
century the Portuguese empire in the Indian Ocean had been disrupted by
"'!! i H"! ". " ' I ! ;'"»;"!l"j'r; T»1»" .,iu.....i.i.i.i.i.i. . ... . i i i i i n
' ±m^~ , ,
iliiIilJ.ii.lJ !.. i 'Àiû i ...i ij ilUliC ! » . . ._...i. . i.u
23.2 Khoi Khoi farmers threshing grain (O The Mansell Collection, London)
competition from rival European trading powen, and the Mozambique channel
lost its strategic significance.
The exploitation of Mozambique was inefficient. Instead of introducing new
forms of production, the Portuguese and Indian traden strove to become
middlemen. The land was parcelled up into prazos and leased to prazeros, and
the estates had the quality of feudal and subsistence-based manors rather than
commodity-producing farms. The supply of the export staples ivory, gold and
slaves depended on indigenous collectors. Merchant capitalism without capital,
and colonization without colonists, could certainly disrupt south-eastern Africa:
what it could not encompass was a lasting transformation of existing societies.
During the long struggle against the Portuguese trading monopoly, the
Dutch East India Company was formed in 1602, with royal support. The
company was independent of the reformed churches, and the 'Heren XVII'
who made up the governing body could afford to be single-minded in the
punuit of trading profits. The company could afford considerable outlays
which would only be recouped in the long term. The garrison settlement at
the Cape was precisely such an outlay. It -was a strategic point premised upon
continuing Asian trade, a place where fresh meat and vegetables could be
bought to provision the Dutch fleets sailing to and from Batavia. When neither
the Khoi nor company employees could produce victuals on a sufficient scale,
the company took refuge in a modified form of capitalist production whereby
company employees were released as free burghers to engage in production on
their own account, the company itself retaining control over marketing.
Like other European powen which expanded into temperate grasslands with
scant populations, the Dutch found that the dynamics of their new settlement
were quite different from those of their tropical dependencies. For over a
century, until the encounter with settled mixed farmers, their experience was
similar to that ofthe Spaniards in the grasslands ofthe South American pampa
and ofthe British in the prairies of North America and south-eastern Australia.
Perhaps the fundamental determinant of the Cape settlement was the
painfully slow development of agriculture. For the fint 40 years of its history,
the setdement imported much of its staple food, despite the vast store of agri¬
cultural knowledge of the settlers. The second great determinant of the settle¬
ment was the absence of a coercible, indigenous labour force. Whereas
Indonesians could be directed into commodity production in large numbers by
the manipulation of existing social structures, it was impossible to turn Khoi
into export producers without the total destruction of Khoi society. The few
Khoi who survived and remained within the boundaries of the settlement
merged with other elements to form a new community - the Khoi Khoi
(Khoisan), later the nucleus ofthe Cape Coloured. The company could not
rely upon the Khoi for labour and had to bring in slaves from other parts of
Africa and the East Indies.
The hallmark of the new society was its persistent expansionism. At the
Cape, labour was expensive and scarce, while land was cheap and plentiful, so
the obvious manner of achieving the expansion of capitalism was not through
the more intensive use of land but through territorial expansion. The
geographical extension of the settlement replicated the spane population levels
which had characterized Khoi society. Indeed, commercial pastoralism meant
litde more than grazing much the same cattle and sheep on the same land as
before, sometimes with Khoi employed as herdsmen. It was not land usage so
much as social relations which were changed by the new regime. On the
frontier of the settlement it was particularly difficult for the Dutch settlen to
control their labourers, as it was easy for disgruntled worken to flee eastwards
or northwards. Slavery constituted only a partial solution, as slaves were cosdy
and they, too, could escape.
During the eighteenth century, the imperatives of extensive commercial
pastoralism expressed themselves in a series of social relationships which have
more in common with those of the South American pampa and nineteenth-
century Australia than with the rest of Africa. Cape Town was the entrepôt, the
focus of all trading relationships and meeting point of international cultures and
ideas. Company officials not only administered the setdement but also often
traded on their own account, consorted with the more prosperous land ownen,
traden and inn-keepen of the town, and lorded it over the large and stricdy
disciplined slave population. They felt superior to the pastoralists who came to
town occasionally to sell meat, hides and tallow. Speaking a crude form of Dutch,
barely literate and wearing unfashionable but functional clothes, the pastoralists
obviously lacked urban sophistication. But they were the mainstay of the
colonial economy. The slave population was constandy being increased by impor¬
tation, but the pastoral population increased just as fast, both by immigration and
natural increase. One hundred and fifty yean after its foundation, the setdement
comprised more than 20,000 free burghers and more than 25,000 slaves.
The territorial expansion ofthe setdement tended to disperse the Khoi and
San from the western interior. At the same time, the strict hierarchy and
discipline within the new society tended to extrude the more independent-
minded of the slaves and the ex-Khoi. An occasional white setder also fled to
the frontier, either to escape the law, or in search of quick wealth. These
refugees were hardy and violent. They could seek and obtain employment by
the state as frontier law-enforcen or try to live a setded life among the white
frontienmen, either as wage-labouren or as independent pastoralists. Although
this last was an attractive prospect, it was difficult, especially for the Khoi, to
register tide to land in their own names. There remained, finally, a third
option: to stay one step ahead ofthe frontier of white pastoral expansion.
The wide, dry region to the north of the company setdement saw the
evolution of new societies, such as the Kora, Griqua and Nama societies, which
from 1730 to about 1780 were seldom troubled by company authority. To
make a living, these little societies captured, bred and sold catde, relying on
firearms acquired through the setdement. Leadership fell on men ofthe frontier
who were able to speak the language of the whites - white oudaws and the
descendants of slaves. Many, like the Griqua, remembered their descent from
Khoi bands. However, they were not merely refugees from the company
setdement but also its cutting edge. They aspired to the status of independent
commercial pastoralists, but as they were always obliged to stay ahead of their
white rivals, they could not setde anywhere for long. When the frontier was at
length 'closed' in the nineteenth century, by a combination of environmental
conditions and reinforced colonial administration, the day ofthe Griqua came
to an end.
Similar developments also occurred on the eastern frontier, but there the
coune of events was swifter, and the climax in the 1790s more dramatic, as it
involved not only the white pastoralists, the Cape Town authorities and the
refugee communities, but also the southernmost Bantu.
A degree of control over most Xhosa-speakers may have been exercised by
the ruler Phalo, but his death in 1775 unleashed the rivalry of his sons,
Rharhabe and Gealeka. Gealeka's death divided the forces of his followen, and
the death of Rharhabe in 1782 similarly split his followers between the regent,
Ndlamba, and the young heir, Ngqika. Each of these factions aspired to
dominance over the entire Xhosa-speaking community and the neighbouring
Khoi who were enlisted as allies. From the perspectives of both factions, the
white pastoralists seemed much like the coloured pastoralists - potential allies
and possible clients of wealthy Xhosa. It took a generation for the Xhosa to
grasp the unique quality of commercial pastoralism, with its steadily increasing
appetite for new grazing land. Meanwhile, they behaved as if the threat was
marginal.
The white pastoralists gradually came to provide for their own protection
and expansion. The company paid for a resident magistrate - a landdrost - at
major urban centres, but he alone could not maintain control. In his judicial
functions, the landdrost was assisted by burghen - heemraden - who, in military
affairs, took matten more and more into their own hands. Every adult male was,
at least potentially, the patriarch of a small community comprising his family, his
slaves and his employees; and each patriarchy conducted its affairs largely out¬
side the formal judicial framework. When fighting was imminent, therefore,
the patriarchs would elect an ad hoc leader, and the unit would be disbanded
once the spoils had been divided. As white numben increased late in the eigh¬
teenth century, the burghers became impatient with the company. In 1795,
they formally renounced its authority, and declared two independent republics
of white patriarchs, the short-lived republics of Swellendam and Graaf Reinet.
Meanwhile, a much more serious rebellion was brewing. A necessary
consequence of white pastoral expansion had been the dispossession of Khoi
communities. To the whites it seemed quite appropriate that the Khoi should
be farm labourers rather than independent farmers and to the Xhosa faction-
leaden it seemed appropriate that they should be content to become clients of
wealthy Xhosa. But the Khoi rejected this situation and in 1799 many of them
rebelled, drawing in reinforcements from the armed Khoi in the service of the
new British administration. It was an incipient revolution.
During the French revolutionary wars, the British navy seized a number of
garrisons around the world to consolidate their naval supremacy. One of these
was the garrison at Cape Town, captured after a short tussle in 1795. The
burgher republics were quickly suppressed by the British, but the burghen'
control over their land was confirmed. The Khoi revolt, however, could not be
put down so easily. With insufficient troops for a protracted frontier campaign,
General Dundas was content to calm matters. Some land was granted to
individual Khoi; others were encouraged to return to their employment or to
rejoin the Hottentot Corps; and the troubles were left to simmer for a few
years until they broke out again early in the nineteenth century.
The Cape, which the British acquired at the end ofthe 1700s, already had
many of the characteristics which would distinguish it in the nineteenth
century. Cape Town, with a mere 15,000 inhabitants, was still mainly an
entrepôt for Asian and European trade; but it had also acquired its role as oudet
for an export economy based on the hinterland. In the colony as a whole there
were some 22,000 'Christians', most of whom were white, with a few liberated
blacks. The Khoi and San were enumerated at about 14,000, and slaves at
25,000. Only 'Christians' could hold land, although the Khoi had not been
entirely squeezed out; and only the slaves were entirely without civil rights.
Increasingly, the Khoi were cast in an intermediate situation between the
settlement on one side and the Xhosa, Sotho and Tswana on the other.
Because it was Khoi - as Khoi Khoi, Griquas, Nama and Kora - who bore the
brunt of the pastoral expansion from the western Cape, most of the mixed-
farming communities of the whole region were at this point only marginally
affected. Among the northern Nguni, the emergence ofthe strong confederacies
which led ultimately to the formation of the Zulu state was entirely indepen¬
dent of events at the Cape. Nevertheless, in the long run it was the trans¬
forming force of commercial pastoralism, itself provoked by the capitalist
transformation of Europe and Asia, which would prove the most durable social
force ofthe nineteenth century.
The Horn of Africa
The sixteenth and seventeenth centuries are the most dramatic in the history of
north-east Africa. The mighty political and military power and the outstanding
cultural development of the Christian Ethiopian empire disintegrated. Enemy
invasions not only decimated large sections ofthe Christian population, but also
tore whole provinces away from the empire for long periods. For a time the
Holy Empire was but a shadow of its former self. The external tribulations
were aggravated by religious disorders when emperor Susenyos, who converted
to Catholicism in 1630, raised it to the status ofthe official religion ofthe state.
Violent civil wan swept over the enfeebled empire until eventually the foreign
faith and its supporters were finally driven out. There now followed a period of
consolidation of empire and cultural revival. Then, from 1700, a period of
fragmentation of empire began. The steadily increasing anarchy only came to
an end in 1855, when Tewodros (Theodore) II ushered in the age ofthe great
emperors whom the Ethiopian empire had to thank for its restoration and its
survival through the colonial era.
North-east Africa, however, does not comprise only the Christian Ethiopian
empire. Around 1500, Islamic culture, too, reached its peak in the east and
centre of the sub-continent, and warlike Islam was soon to achieve great
triumphs. Fired by the spirit of djihâd (Holy War), Ahmad ibn Ibrahim al-
Ghazi. nicknamed Gran (the left-handed) in the Christian Ethiopian chronicles,
won victory after victory. Year after year, in the dry season, his hosts overran
Amhara and Tigré, forcing whole provinces to embrace Islam. But once the
leader had disappeared, the forces he had inspired collapsed. The Christian
empire started to counter-attack. Finally the migrations of the great Oromo
(Galla) people destroyed the once flourishing communities and cultures of the
Islamic peoples in central Ethiopia, leaving hardly a vestige.
The Oromo and the Somali were to play a decisive role in the history of
north-eastern Africa. Their evolution followed completely different paths. The
Oromo spread out in all directions, in a series of migrations. They advanced
into areas depopulated through the religious wan; they subjugated and
assimilated other peoples but adapted themselves to their cultures. They did not
develop a true national culture of their own.
345
75 150 miles
24.1 The Ethiopian empire and its dependencies, c. 1550 (after E. Haberland)
The Horn ofAfrica 347
Not so the Somali. The participation of large groups in Gran's Holy War and
the influence of Islam were powerful factors in the development of a strong
national consciousness. Although divided into many ethnic groups, the Somali
considered themselves one people with one culture and one religion.
The political and cultural events in northern, central and eastern north-east
Africa were historically fairly well substantiated, but the history of the west and
south-west remained rather obscure. We know today that this period saw the
beginnings of a very dynamic cultural development. Direct influence from the
Christian empire to the north led to the growth of large states and sophisticated
cultures extending south to the natural highland boundary. Of particular
importance were the foundation of the states of Inariya, Wolayta, Dauro and
the small states ofthe Gamo highlands.
The Christian Ethiopian empire around 1500

By about 1500, the Christian Ethiopian empire had politically and culturally
reached heights that it was not to approach again for centuries. It enjoyed
undisputed hegemony in north-east Africa. Sudan was not yet Islamic. The
Islamic towns along the shores of the Red Sea in what is now Eritrea were
politically insignificant. Even the coast around Massawa, itself Islamic-Arabic,
was part of the Ethiopian empire. The power of the many small Islamic
dominions in eastern and central Ethiopia had been cut back by the crushing
blows of the Ethiopian emperon of the preceding century, in particular the
emperor Zera-Yakob (1434-68).
The influence of the Christian state was even stronger towards the south and
south-west. New states with a Christian ruling class emerged; formerly inde¬
pendent peoples and states were made to recognize the political and cultural
supremacy ofthe northern empire. The Kebra Nagast (Glory ofthe Kings), the
charter of the Ethiopian empire, provided the moral justification of this
imperial expansion.
This work was probably compiled about 1300, shortly after the restoration
of the dynasty that from then until 1974 was called the Solomonic dynasty.
Folk legends, biblical, Talmudic and Ku'ranic traditions were brought together
and harmonized with a divine mission of salvation. The Kebra Nagast records
how the queen of Sheba here equated with Ethiopia undertook a journey
to Jerusalem to learn the wisdom of Solomon. By him she had a son, Menelik,
who was born after her return to Ethiopia and was later to become the fint
Ethiopian king. Later, on a visit to his father in Jerusalem, Menelik seized the
Ark of the Covenant and carried it off to Ethiopia. It was preserved thereafter
in the famous cathedral of Axum where the Ethiopian emperon were
enthroned. The book ends with the announcement ofthe spiritual division of
the world between the two great Holy Empires of Rome and Ethiopia. The
story makes Christian Ethiopians the chosen people of both the Old and the
New Testaments for, unlike the Jews, they have accepted the Gospel. The fact
that Ethiopia was able to withstand so many storms is due above all to the
geographical position of this mountainous land. But the feeling of being the
new chosen people must have given spiritual conviction to the empire's deter¬
mination to extend dominion over most of north-east Africa.
Internally, the empire enjoyed the utmost tranquillity at that time. This is
evident from the reports of Francesco Alvares who, from 1520 to 1526,
accompanied the Portuguese mission from Massawa to Shoa as chaplain and
travelled extensively throughout the country. Order and security reigned; the
governor's instructions were obeyed and the emperor's word carried absolute
authority throughout an empire that measured at least 1,000 kilometres from
north to south. The concept of the Holy Christian Empire, the undisputed
claim to power of the House of Solomon and, finally, the Christian faith and
common culture of the Amhara and Tigrai who formed the state, constituted
strong bonds. The culture and language of these two peoples exerted a
powerful assimilating effect on other ethnic groups which came into contact
with them. This is particularly true of the different groups of the Agaw, the
indigenous inhabitants of central and northern Ethiopia, who were almost all
assimilated by the Amhara and Tigrai. In the south, too, in the parts of Shoa
province now inhabited mainly by Oromo, not only Christianity but also the
Amharic language and culture must have spread at the expense of other ethnic
groups living there.
The attempt by the emperor Zera-Yakob to centralize the empire's political
organization failed, but the word of the emperor still carried absolute authority,
even after 1500. From the chronicles, we can infer that, as regards the auton¬
omy of individual provinces and the question of land law, conditions were very
similar to those still obtaining in the nineteenth and twentieth centuries. Most
land was the residual property of ethnic and family groups who decided how it
should be divided among their members. The Church was also an important
land owner. Finally, there were the imperial lands, granted by the emperor on a
short- or long-term basis to deserving people or for specific purposes. This
feudal land was known as gult, a word also used to denote the right conferred
by the emperor to the grantee of an office to levy tnbute and services.
The extraordinary dynamism of Ethiopian society, with its ideal of the tellek
saw (great man), who had constandy to prove himself anew and in whom
qualities and achievements alone counted, was not conducive to the nse of a
nobility. In theory, any free man belonging to a respectable family could be
elected to office by his district people's assembly. The emperor could confirm
or reject the election. Thus monarchical power and democratic election
complemented one another, preventing the formation of a hereditary nobility
with political privileges which could have disturbed the unity of the empire.
This happened after 1700, with the increasing enfeeblement of imperial
power.
Roads were safe. Imperial rest-houses abounded and also great weekly
markets. There was general well-being. Taxes in kind flowed from the provinces
and the tributary states to the emperor's court, where they were redistributed.
Although neither minted coins nor commercial correspondence were known,
The Horn of Africa 349
there was nevertheless extensive trade over long distances and consumer goods
were imported from India and the Near East.
Ethiopia was closely linked with lands as far away as Europe. Ethiopian monks
the main purveyon of education, art and science received their training in
Egypt and took part in both Coptic and Islamic cultural life. Although literary
production was limited to the fields of religion and history, it flourished until at
least 1650. The graphic arts of which, above all, book illuminations have
survived were stimulated by contact with the Christian east and with Europe.
The existence of still little-known ruins of churches and palaces is further proof
ofthe rich complexity and artistic creativity of this period.
Islamic states and cultures: the djihâd, the Somali and the Hadiyya
By 1300, if not before, Christianity had developed a vigorous missionary
activity in central and southern Ethiopia. As the official religion of the
Ethiopian empire, Christianity could always rely on state support if necessary
backed by fire and sword. The spread of Islam in north-east Africa, however,
depended on the spread of Arab culture and on trade and communications.
Arab and Islamic influences on the north-east African coast are age-old. Trade
between the Arabian peninsula, the Ethiopian coast and the Somali coast led to
the foundation of a number of towns with mixed populations but with the
Arab element predominating. The most important centres, starting from the
north, were: Massawa, the Dahlak Islands, Assab, Oboek, Tadjurä, Djibouti,
Zayla', Berbera, Obbia, Mogadishu, Merka and Brava.
The map of the spread of Islam during the fifteenth and sixteenth centuries
clearly shows that it was concentrated on both sides of the Rift Valley along
the great trade route into central Ethiopia. From the coast inland, there were a
number of states: Adal, the largest, Dara, Dawâro, Bâlî, Arababnï (or Ar'en),
Sarha and Hadiyya. Islamic influence probably extended westwards across the
Rift Valley. To the north of the Rift Valley on the southern slopes of the Shoa
highlands lay the twin Islamic states of Ifat and Fatagar (Fatadjar). Many people
in these states must have spoken Semitic languages which spread alongside and
overlaid the Kushitic languages originally spoken there. The only ones remain¬
ing are the languages ofthe Harari, East Gurage and Argobba.
These states were all marked by the influence of Arab-Islamic culture from
the coast and many elements ofthat culture still penist today. The most striking
example of the town settlements that once dominated the area is Harar, the
only genuinely ancient city on Ethiopian soil. The stone-built single-storey or
two-storey town dwellings with flat roofs and beautiful interior furnishings
betray Arab influences, as do the large number of walled holy sepulchres. Arab
influence also left its traces in intensive horticulture, still practised today, and a
multitude of food crops unknown in Ethiopia such as oranges, lemons, peaches,
vines, bananas and sugar cane. Also traceable to Arab influence is cotton-
growing and the consumption of coffee and khat or cat (celastrus). Widespread
in central Ethiopia are numerous religious concepts, rites and words that stem
from the Arabic, even though the Muslims there later renounced Islam.
The spread of Islam and its culture from east to west brought it into conflict
with the Christian empire. This happened mainly in central Ethiopia, in what is
now the province of Shoa, and to the east of it. This was also the scene of
bloody conflicts between the two powen that went on for several centuries and
remained indecisive until the second half of the fifteenth. Although the
Christian empire remained the undisputed dominant power in north-east
Africa, it was unable to eliminate the Islamic states, although they were
insignificant in relation to its own size and population.
We are not certain what changed the war between the Islamic trading states
and the Christian empire into a Holy War or djihâd, nor why the political
leadenhip of the sultans of Adal, the Walasma, was transferred to the religious
leaden (imams). Nor do we know what gave the Islamic armies the cohesion
and fanaticism to bring the mighty Christian empire to the brink of defeat. One
explanation no doubt lies in the brilliant Ahmad ibn Ibrahim al-Ghazi (Ahmad
Gran) who became the charismatic leader of the djihâd. There is also no doubt
that the rest of the Islamic world, which at that time held northern Somalia
tighdy in its grip, provided its armies in Ethiopia with a recruiting potential
several times the size ofthe comparatively small population ofthe Islamic states.
A war with the state of Adal, provoked by the Christians, and initially
successful wars in the Chercher area during the reign of the weak emperor
Lebna Dengel (1508-40) seemed to herald nothing new in the conflict
between the two advenaries. But the fortunes of war suddenly changed. In the
decisive batde of Sembera Kure (1529), on the western edge of Chercher, the
Ethiopian empire lost not only a whole army but also a considerable part of its
ruling élite. The consequences were to be fearful. Until 1543, Islamic armies
swept over the Ethiopian Highlands, in the south and the north, each dry
season, systematically defeating and subjugating one province after another. In
southern and eastern Ethiopia, in Hadiyya and Bali provinces, the population
went over to the Muslims, for they often shared their religion and greeted
them as liberaton. The same did not happen in the old Christian provinces,
especially in the regions inhabited by the Amhara and Tigrai. There, those who
did not accept Islam were put to the sword so that whole districts converted - if
only nominally to Islam. Churches and monasteries were plundered and
sacked and their treasures given to the Islamic armies. Treasures of Ethiopian
literature and painting (miniatures in books or murals on church walls) were
destroyed. In 1531, Dawaro and Shoa, two large, densely populated provinces in
the south-east and in the heart of the Ethiopian Highlands, were subjugated.
Amhara and Lasta, two of the empire's central provinces, followed in 1533.
The same year, Ethiopia's most important holy place, Axum, was razed to the
ground. (The Tablets of the Law were saved and later brought back to Axum
where they are still to be found today.)
But even in the most desperate times, the steadfastness of the Christian
people and the strength of the concept of the Holy Ethiopian Empire were
apparent. Those forcibly converted to Islam abandoned their new faith the
moment the Islamic armies withdrew. The conquered provinces soon rose
against their new mästen. Even the weak emperor Lebna Dengel, continually on
the run, refused to give his daughter in marriage to Ahmad Gran, who offered
him a pledge of friendship.
Lebna Dengel died a hounded fugitive, in 1540, in the impregnable monastery-
citadel of Dabra Damo in Tigré. Within a few years there had been a complete
change in the political situation. The accession to the throne of the young
emperor Galawdewos (Claudius), one of the most outstanding figures in
Ethiopian history, was followed by the rapid restoration of the Ethiopian
empire, the arrival of the Portuguese expeditionary corps and the decisive
defeat ofthe Muslims.
By the sixteenth century, Portugal had reached the peak of its imperialist
expansion. The Indian Ocean and its inlets had become a Portuguese sea.
Portuguese fortresses, which were never completely subjugated, sprang up on
the coasts at Hormuz, Oman, Socotra, Mombasa and so on. It was an historical
accident that, in 1541, brought a Portuguese fleet into Massawa which, at that
time, still belonged to the Ethiopian empire. The year before, a Portuguese
fleet had been sent from Goa in India to the Red Sea to combat the expansion
ofthe Turks who, after the conquest of Egypt, were gaining ground in Arabia
and even in Sudan. Part ofthe fleet anchored at Massawa in the spring of 1541.
Soon after, the bâliar nagâs (sea-king), governor ofthe northernmost province of
the Ethiopian empire, appeared bearing letters requesting support for the
Christian empire. By July 1541, a volunteer corps of 400 Portuguese, under the
leadership of one of the younger sons of the famous Vasco da Gama, had
marched into the Ethiopian Highlands. Their combat tactics and their superior
firearms made them formidable opponents for the Muslim occupation troops in
northern Ethiopia. A growing number of Ethiopian volunteers accompanied
the Portuguese, who inflicted two severe defeats on the invincible army ofthe
imâm Ahmad Gran in two battles. The imâm succeeded in obtaining assistance
in the form of new types of artillery and sharpshooten from the Turkish com¬
manders in Yemen. Before the Portuguese could link up with the emperor's
army, they were forced to engage in a third batde and were defeated. Fortune
nonetheless favoured the Christians. The imperial army and the remaining
Portuguese joined forces and, in February 1543, the Muslim army was des¬
troyed to the east of Lake Tana and the imam was killed. Whether it was
Portuguese assistance that tipped the balance in this long-drawn-out war is
uncertain. In the final battles, their guns had restored a certain strategic
equilibrium. What was to be of greater consequence for Ethiopia, however, was
the Catholic mission that followed the Portuguese soldien (see below). The
Muslim forces were no doubt strong enough to conquer Ethiopia militarily
initially, but too weak to keep the two great peoples, the Tigrai and the
Amhara, under control permanendy. After 1543, the Christians emerged
victorious from the struggle and, during the next few decades, the empire
succeeded in winning back lost territory. The might of Islam had been broken
for centuries to come.
The Oromo (Galla)

Since the middle of the sixteenth century, memben of the great Oromo people
hitherto known as the Galla had been penetrating central, eastern and
western Ethiopia in ever increasing numben. Their arrival, their setdement and
their quarrels with other groups already there had far-reaching effects, both
political and cultural. The Ethiopian empire suffered more and longer through
them than through the Muslims, and lost many provinces forever.
Although it is still not possible to specify the reasons for the great Oromo
migrations, we know that they were not undertaken to escape the pressure of
other peoples. Oromo traditions say nothing about the matter except that they
started out to seek new grazing lands for a growing population.
Notwithstanding the fanciful accounts of Amharic and European authon as
to their origins, the Oromo are a true Ethiopian people. The highlands sur¬
rounding what is today the province of Bali were their home, whence they
moved out in large numbers into the heart of the Ethiopian Highlands and
south, to present-day Kenya and the Indian Ocean. Equally erroneous is all that
was written about their culture. They were described as savages and primitive
cattle-raisers. The Oromo were and are a people with a developed, sophisti¬
cated culture admired by their neighboun. Their spread was partly the result of
a sort of chain reaction: in the face of their military superiority, and because of
their much-admired social organization, other ethnic groups joined them. This
was particularly true of the Hadiyya, who lived between Harar and Gurage,
almost all of whom went over to the Oromo.
Oromo culture is marked by several distinctive features, particularly the close

relationship with their catde. Cattle were not only the basis of their livelihood
but the object of emotional and ritual concern. This cattle complex, with its
moral overtones, permeated their whole culture.
The Oromo consisted of numerous genealogically connected groups and
clans which, as the population increased, tended to break away and form new,
politically independent groups. Borana and Barentu (or Baraytu) were two
mythical ancestors and founders of the two halves of the original stock that
bore their names. After the migrations began, these groups split. The Barentu
made their home mainly in the east, while the Borana settled in the south,
centre and west. Towards the end of the last century, when the Oromo were
defeated by the emperor Menelik and incorporated into the Ethiopian empire,
they split into more than 80 groups of various sizes.
The gada system, classification by age groups based on an abstract numerical
principle, was the basis of Oromo social and political organization. It was an
institution that ruled their whole life. Nothing was not subject to its rules:
birth, baptism, marriage, circumcision, emancipation from paternal authority,
permission to bear and bring up children, conscription for war and for hunting,
obligation to make sacrifices, civil death by exclusion from the system, burial
rites and a thousand details of everyday life. It represented the sum total of the
laws by which the life of the Oromo was governed. Forty years after his father
24.2 Oromo migrations in the sixteenth century (after E. Haberland)

had done so, each Oromo male, regardless of his actual age, had to enter the
children's class. Ideally he had to pass through ten classes, each lasting eight
yean, during his lifetime. All those entering a class together formed a gada
group, which remained an insoluble warrior community for life. The fifth and
sixth classes constituted the leadenhip and warrior classes. As it was a
community of free men with equal rights, there was no hereditary hierarchy.
The leader of the ethnic group, who usually took the name of abba gada, was
elected from among the members of the fifth class. He was supported by the
hayu (judges), also elected from this class, and by other dignitaries. Of decisive
significance for the military aggressiveness of the Oromo was the gada rule that
at least once during the eight yean ofthe ruling gada class a warlike expedition
had to be undertaken. Only one who had distinguished himself as a brave
warrior and hunter could occupy a position of eminence in the community.
The ideal of the hero as successful killer and hunter was common to all the
peoples of north-east Africa, but nowhere was the killer complex so integrated
into the social system as among the Oromo. Every eight yean the Oromo
warrion would burst upon their neighboun who, until they came to under¬
stand the gada system, remained perplexed in the face of these cyclical episodes.
The Oromo religion was one of remarkable solemnity. Belief in a god who
was creator and heavenly father took Old Testament forms. They possessed a
rich treasure of oral literature, both sacred and profane, including prayen and
invocations to the deity but also lyrical love poetry.
In their early migrations at least, the Oromo had one advantage other than
their military prowess: they were advancing into regions virtually depopulated
after yean of fierce warfare between Muslims and Christians. This was par¬
ticularly true of central Ethiopia, today the province of Shoa, through which
the Islamic armies had marched every year on their way north. Between 1530
and 1554, groups of Oromo warriors advanced northward, and in 1554
destroyed the greater part of the Islamic state of Adal. At this time the Oromo
became acquainted with the horse and soon became renowned horsemen.
Catastrophic for Christians and Muslims alike were the attacks of the
Harmufa group of Oromo (1560 who were invading the provinces at the
heart of the Ethiopian empire, which were just beginning to recover from the
devastation of the djihâd: Angot, Amhara and Bagemder. At the same time,
other Oromo groups again attacked Adal, where there was famine and plague,
the amir Nur himself falling victim to it. When the emperor Minas died, a third
ofthe Ethiopian empire was already permanendy setded by the Oromo.
But this was not the end of Oromo expansion. Even before 1500, other
groups (Gugi, Boran and Orma) had already moved out from the highlands
around Bali into the savannah to the south in their constant search for new
grazing lands for their catde, finally reaching the Indian Ocean at the mouth of
the River Tana. In the centre the Arusi established themselves in the vicinity of
the former state of Bali and pushed their borden further and further westwards.
The Barentu spread over the area of the old Islamic states of Fatagar, Dawäro
and Adal. The fortified city of Harar and the famous pilgrimage centre of Sek
Hussen survived as Islamic islands. Only in the hot, easily defended lowlands,
in the gorges ofthe great rivers and on a few tablelands did pockets of Amhara
hold out. The Wollo and Yeggu Oromo spread over the fertile provinces of
Angot and Amhara, once inhabited by Christians. Although they soon adopted
the local Amharic language, because of their conversion to Islam the Oromo
remained sworn enemies of the Christians. Other Oromo occupied the fertile
lowlands to the east ofthe Ethiopian highlands as cattle-raisen.
At the beginning ofthe seventeenth century, a last wave of Oromo, consist¬
ing of several groups belonging to the large Macea family, pushed westwards into
what are today the provinces of Wallaga and Ilubabor, stopping only when they
reached the western slopes of the Ethiopian Highlands. In the east and south of
the region they came upon well-organized, densely populated kingdoms in close
contact with the Ethiopian empire to the north and whose populations were
partly Christianized. Despite the resistance of these kingdoms, Oromo expan¬
sion continued ever more vigorously, especially as communications between
the empire and its dependencies in the south-west weakened in parallel with
the decline in the strength of the empire. Thus Bosa and Guman disappeared
from the political map and became the Oromo dominions of Gimma and
Gumma. However, the indigenous substratum remained so strong here that the
Oromo took over much of the culture of the conquered country, in particular
the political organization and the monarchy, while the gada system gradually
disappeared. Inariya, famous for its treasures, was the last of these states to fall,
defeated by the Limmu Oromo. After 1700, its king fled to the south, where
the other Gonga kingdoms and the many Ometo kingdoms held their ground
against the Oromo, thanks to the natural strongholds provided by their terrain.
The Somali
During the same period, if not earlier, another great people - the Somali -
began an expansion which had considerable significance for large secton ofthe
eastern part ofthe Horn of Africa.
The Somali country of origin is thought to be the area north of where they
live today. From there they spread - probably from the eleventh century on¬
wards - south and west. As early as the thirteenth century, Arab geographen tell
of Somali settlements in the area of Merka, to the south ofthe city known today
as Mogadishu. As cattle-rearing nomads, the Somali were driven by the pressure
of increasing population to seek new grazing lands in the direction of the
Ethiopian Highlands, and they probably reached the eastern edge of the Harar
plateau in about 1500, coming into contact with the Islamic states there. Names
and genealogical data left by the chronicler Sihäb al-Din suggest that the Somali
played an important role as supporters ofthe imam Ahmad Grañ's djihâd.
The stimulating and unifying effect ofthe Arab-Islamic influence was greater
on the Somali than on any other north-east African people. Islam became an
integral part of Somali culture, and their faith was constandy strengthened by
the missionaries who came from Arabia. These holy men were the founding
fathers of the Somali people, such as the famous Ismä'il Gabarti from whom
many Somali are descended. This Arab-Islamic cultural influence also gave the
Somali a sense of cultural superiority and strengthened their capacity to assimi-
liate other related groups in the huge area between Ogaden and Lake Turkana.
It is extraordinarily difficult to pinpoint individual stages in the expansion of
the Somali because many groups which attached themselves to the Somali
nation deliberately suppressed all reference to a different origin and assumed a
common Somali genealogy.
Under the influence of Islam, the Somali culture changed, especially in the
south. The Somali thus differentiated themselves more than any other group
from the large family of north-east African peoples, linked by many common
historical, cultural and geographical factors. The influence exerted by the
geographical environment is by no means to be underestimated, since, apart
from the Afar who live in the desolate wastes of their homeland, the Somali are
the only inhabitants of north-east Africa whose territory consists exclusively of
dry savannah and semi-desert. This ecological environment caused them to
develop a completely different type of culture from that of other north-east
African peoples, who are mostly highland-dwellers.
The Ethiopian empire's strugglefir survival, Í529-Í632:

the Portuguese episode and the restoration ofthe empire after Í632
Between 1529 (the start ofthe djihâd) and 1632 (the beginning ofthe Restora¬
tion) the Ethiopian empire was struggling for survival. It was not only a dramatic
period politically and militarily but it was also a time of intense intellectual and
cultural conflict, in which Ethiopia was assailed by ideas from another continent.
The Christian empire showed astonishing strength and capacity for survival in
the face of pressure from external and internal enemies. It was during the time
ofthe emperor Sartsa Dengel (1563-97), when the empire was more a fiction
than a political reality, that the spiritual strength of the concept of the Holy
Empire was forcefully demonstrated. The Turks were consolidating their power
on the shores of the Red Sea, seizing all Ethiopian ports and penetrating deep
into the Highlands to Tigré; in the central provinces of Bagemder and Samen, a
bitter civil war was being waged against the Ethiopians ofJewish faith; and the
Oromo invasion had assumed worrying proportions. Real state power may
have been reduced to the area the emperor occupied with his troops, but the
immortality ofthe 'evangelical empire' was apparently never in doubt.
The empire did, of coune, suffer considerable losses of territory and popula¬
tion in this struggle for survival. From the Restoration of 1632, the great
southward thrust, which had been the whole imperial policy of the preceding
centuries, was scarcely pursued. As the imperial power steadily declined, efforts
were concentrated on maintaining stability at the centre. The transfer of the
emperor's residence to Gondar in 1636 is a sign of this withdrawal. Before this
the emperors - in a country where towns were unknown - had moved their
357
The Horn of Africa
24.3 The castle at Gondar, the Ethiopian capital built by Emperor Fasiladas
(© Werner Forman Archive, London)
encampments to a different place every year or half year. This had been a hard-
and-fast rule, followed even when there was no political or military necessity
for undertaking a journey or change of residence. It was therefore not just a
break with tradition but also a symbolic renunciation of the dynamic imperial
policy of his predecessors when Fasiladas built a fixed residence in Gondar in
1636 after the last great civil wan of religion had ended. Gondar was to remain
the centre of what was left of Christian Ethiopia until 1855.
Thus between 1529, the start ofthe djiliâd, and 1597, the death ofthe emperor
Sartsa Dengel and the beginning of the civil wan, imperial policy was concen¬
trated on defence. Emperor Galawdewos carried on the struggle against the Mus¬
lims, in particular those in the state of Adal-Harar, with great success. Former
contacts with the south-west were renewed, the influence of Christian-
Amharic culture was restored there and the Oromo were driven back, at least
temporarily. Under the leadenhip of the amir Nur ibn Mugahid, the brave and
politically gifted successor to Ahmad Gran, the Adal-Harar Muslims once more
attempted to confront the Christian empire. Although the poorly prepared
Ethiopian army was beaten back in an attack on Adal in 1559 and the emperor
Galawdewos fell on the field of battle, this marked the end of Islamic power.
Within a few decades, the Islamic states and Islamic culture had disappeared
from this region. There remained only the fortified town of Harar and the small
sultanate of Awsa, in the Hawas river-delta oasis. Emperor Minas (1559 had
mosdy to fight the Turks who, from their bases in the Red Sea ports, tried
repeatedly to establish a foothold in the Ethiopian Highlands. Sartsa Dengel
(1563 Minas's son, who acceded to the throne while still a child, had to
wage uninterrupted campaigns in all directions, as the Oromo allowed the
empire no respite.
Wan of frightful violence were waged to annihilate the Ethiopians ofJewish
faith in the northern Highlands, mainly in Samen, Waggara, Dambiya, Belasa,
Sallamt and Sagade, where formerly they had enjoyed a certain autonomy.
Their forebears had probably been converted to Judaism by Jewish missionaries
in pre-Christian times. Unlike Jews in other parts of the world, they were
completely integrated into the culture in which they lived, so much so that
their holy books were not written in Hebrew but in Ge'ez, the Ethiopian
literary and liturgical language.
It is not known why this religious war broke out at the very time when the
empire was already hard-pressed by external enemies. As with the extirpation
of the Christians by the Muslim armies of the djihâd a generation earlier, this
war is one of the darkest chapters in Ethiopian history. The Jews were faced
with the choice of embracing Christianity or being exterminated. The Jewish
population was practically wiped out, as the majority refused to be converted.
The few survivon, stripped of their civil rights and dispossessed of their land,
were forced to become artisans, and the word Falasha, used to describe
Ethiopians ofJewish faith, came to be synonymous with smith, potter, carpenter
or weaver, indeed with any type of skilled labourer. Despite the importance of
these occupations, a stigma was attached to them in a society whose ideal was
the free and independent man who never worked for a wage. Thus social
degradation went hand-in-hand with religious persecution.
It is astonishing how the Ethiopian empire found the strength, despite
almost three generations of unrelenting struggle, not only to confront the
Turks in the north, the Adal Muslims in the east and the Oromo in the south
and centre but also, in massive campaigns in the south-west, to force the large
states of Inariya, Bosa and Kaffa (Gomar) to become once more closely linked
to the Ethiopian church and northern Ethiopian culture. The compiler of the
History ofthe Galla reports how the emperor Sartsa Dengel offered to halve the
Inariya tribute on condition that all became Christians. Under pressure from
the emperor, but also mindful of the support to be expected against the
Oromo, the king Badanco of Inariya decided to be baptized. The ceremony
was performed with great solemnity for the king and all his people together.
Shordy thereafter, the king of the Bosa, neighbours of the Inariya, was in turn
baptized. These events, which took place shortly before 1600, represent the
swan-song of imperial expansion. We can conclude from oral traditions still
surviving among south-west Ethiopian peoples that the contacts thus made
during the time of Sartsa Dengel had a lasting and stimulating effect on
southern Ethiopia and its cultural development.
Of far greater consequence for the empire, however, was the almost
century-long ideological conflict with the Catholic Church from 1542 to 1632.
With the creation ofthe Society of Jesus (the Jesuits) in 1540, the Catholic
Church acquired a very effective instrument not only for carrying on the
Counter-Reformation in Europe, but also for converting the 'heathen' and
restoring the 'heretical' Christian churches to the authority of Rome. Even
before 1540, letters had been exchanged between the Portuguese kings and the
Ethiopian emperon and these contacts were now seized upon. The decisive
factor was, of course, the presence in Ethiopia of members of the Portuguese
expeditionary force who had remained behind and their families. In 1557,
Andrea da Oviedo was consecrated bishop and sent to Ethiopia with some
other Portuguese Jesuits to prepare the work of reinstatement - that is, the
union of the Orthodox Church with Rome. His discussions with the emperor
Galawdewos provoked a brilliant exposé of the Ethiopian religion through the
mouth of the emperor, which has gone down in history as Galawdewos'
profession of faith.
Here a voice is raised for the fint time in Africa pointing out that Chris¬
tianity is a religion without limitations of time or place, about which no people
can claim that its own interpretation is the only true one. Galawdewos shows
that certain customs and practices of the Ethiopian Church are neither
'heathen' nor 'Jewish', but have their origins in Ethiopian folk culture, just as
European culture has found expression in European Christendom.
Under emperor Sartsa Dengel the Jesuits were given Fremona, near Adwa in
Tigré, as their centre. There they were allowed to five in complete freedom
and to engage in missionary work. In 1603, Pedro Paez, a Spaniard, took over
as head ofthe mission. He was undoubtedly the most outstanding missionary of
the time, and also an unusually tolerant human being. Hence the Ethiopians
engaged in the most stimulating theological discussions. To reach the people,
the Church went so far as to conduct and record the religious discussions in
Amharic. This innovation ceased after the Restoration of 1632 with the re-
introduction ofthe official Church language, Ge'ez, which was as little under¬
stood by the mass of the Ethiopian population as was Latin by the European
peasants of the time. The fruitful effect of this contact, however, was short¬
lived and the constructive debate soon turned into open enmity between
Orthodox and Catholic factions.
Literary production in Ethiopia experienced a remarkable upsurge at this

time, and the works written, mainly theological and historical, are among the
most important in Ethiopian literature. Theological works that may be men¬
tioned are the Fekkare Malakot (Explanation of the Divinity), the Mazgaba
Haymanots (Treasury of Faith) and the all-embracing Haymanota Abaw (Faith of
the Fathers), translated from Coptic Arabic. For the reinstatement of those
Christians who had temporarily embraced Islam, the Mashafa Keder (Book of
Impurity) was written in connection with the atonement rites. Enbakom
(Habakuk) , an Arab Muslim converted to Christianity who rose to be leader of
the Ethiopian clergy, wrote the remarkable book Ankasa Amin (Gates of the
Faith), a defence of Christianity against Islam. Furthermore, with the translation
of Barla'am and Josaphat, Enbakom made an important work of Eastern
literature accessible to the Ethiopian people.
The three works or annals concerning the reigns of the emperors Galawde¬
wos, Sartsa Dengel and Susenyos and the short History ofthe Galla by the monk
Bahrey are also numbered among the important literary productions of this
period.
The accession of Susenyos (1607) marks the beginning ofthe decisive phase
of the confrontation between the Ethiopian Orthodox Church and the
Catholic Missionary Church. The head of the Jesuit Mission, Pedro Paez,
theologian, preacher, teacher and architect, had had access to the imperial court
and, under the influence of his strong personality, the emperor began to incline
increasingly towards Catholicism. His influential brother, Ras Se'ela Krestos,
officially turned Catholic in 1612 and set up a Catholic mission in his province
of Gojjam. Despite the remonstrances of the abuna (the metropolitan of the
Ethiopian Church), in 1632 the emperor himself also embraced the Catholic
faith. Pedro Paez died shortly afterwards.
The confrontation between the two doctrines and civilizations, hitherto
carried on with intellectual weapons and mutual respect, now turned into open
war. Unlike his predecessor, Alphonso Méndez, the new Spanish bishop sent
by the Pope, was bigoted and arrogant. Under the emperor's protection,
Méndez began to set the Ethiopian Church on what he saw as the right path.
All Ethiopian priests had to be re-ordained and all churches re-consecrated.
The calendar was Europeanized; circumcision was forbidden; and all Ethiopians
had to be re-baptized.
Revolt after revolt broke out against the hated new Church, being particularly
The Horn of Africa
fierce in the central provinces of Bagemder, Lasta and Wag. The emperor, once
so loved, who had brought peace to the land, now had to subdue his subjects
in bloody wan during which the fanatical zealotry of the Jesuits increased
tension still further. In 1632, there was a great batde from which the emperor
emerged victorious, but the emperor was a broken man. He abdicated,
restoring his subjects to the faith of their anceston and asking the former clergy
to return to their own liturgy.
Susenyos died shortly afterwards, and the Catholic mission came to an end.
The people, insofar as they had adopted the new religion, returned in a body to
the traditional faith. The new emperor, Fasiladas, had the Jesuits deported. The
gradually dwindling number of Ethiopians who persisted in the Catholic faith
were persecuted and their most prominent leaders, amongst them Ras Se'ela
Krestos, were executed. Fasiladas went so far as to conclude an agreement with
the Turkish governor of Massawa in which the governor undertook to kill any
European missionaries landing at Massawa in exchange for head money. Thus
ended, through the fault of the Europeans, a contact between Africa and Europe
that had lasted almost a hundred yean, at fint rewarding, but later only negative.
The Ethiopian empire now began a period of deliberate isolation from the
rest of the world and pursued a policy of stabilization that, by the end of cen¬
tury, turned into stagnation. The establishment of a permanent imperial
residence at Gondar, in the militarily secure centre of the empire, was part of
this process. It was no longer possible to punue an aggressive and dynamic
policy from Gondar. There is therefore litde political activity to report in
Christian Ethiopia during the second half of the seventeenth century. Under
the emperor Yohannes (1667 the few remaining Catholic Ethiopians were
given the choice of accepting the Orthodox faith or leaving the country. The
Muslims still in Christian Ethiopia mainly traders were allowed to practise
their religion freely but made to live in separate setdements.
Iyasu I (1682 was the last ofthe great emperon. He undertook one last,
unsuccessful attempt to re-establish contact with south-western Ethiopia and
even visited Inariya. There were no internal or external enemies to endanger the
empire. The important port of Massawa, although nominally belonging to the
Turkish empire and under the control of a Bedja governor, remained open for
trade with Ethiopia. Shortly after Iyasu's death, however, the disintegration ofthe
empire began, degenerating into complete anarchy after about 1755.
During this period, cultural life flourished for the last time under the patron¬
age of the imperial court at Gondar. To a greater extent than ever, art and
science proved to be the sphere of a small élite. It is not possible here to give a
complete account of the different aspects of the intellectual life of the period.
We shall therefore restrict our remarks to literature, architecture and painting.
In contrast to the previous century, literature lacked originality and took on an
edifying and courtly character, with translations from Coptic Arabic predomi¬
nating. Worthy of mention are Faws Manfasawi (Spiritual Medicine), a devotional
book of penance, and the Fetha Nagast (Jurisdiction of the Monarchs), a
collection of laws and instructions intended for the use of Coptic communities
in Egypt. This period also saw the creation of a great many sacred hymns.
The graphic arts are marked by this relationship with court life. The famous
rock churches, which are among the most original creations of Christian
Ethiopian civilization, were still being hewn out of the rock at the end of the
fifteenth century, but this type of construction ceased completely after 1500,
presumably because ofthe religious wars. What appeared later, in Gondar and a
few other places, was something completely different huge castles, libraries,
chapels and annexes to the palaces of the emperors and members of the
imperial family. The form of these structures often indicates foreign models.
Most were probably erected by Indo-Portuguese masons who came from Goa
with the Portuguese mission. The gradual decline of the empire, the destruc¬
tion of the city of Gondar by Tewodros and the dervishes, and centuries of
neglect have completely robbed these buildings of their architectonic finery and
their cosdy interior decoration. All that remains today is ruins or walls which
give litde idea ofthe level of civilization at the time.
Ethiopian painting expressed itself mainly in two media: church wall paint¬
ings and illuminations of religious books. While only little remains of the church
murals before the seventeenth century, many manuscripts date back to the
fourteenth century and clearly show the development of this sort of painting.
In about 1500, book illumination reached a peak that corresponded to the
peak of Ethiopian Christian civilization. Independendy of the text they
illustrated, these paintings were intended to move onlooken through the force
of their representation and strengthen them in their belief. The paintings ofthe
second half of the sixteenth century and the early seventeenth are quite
different in character. They reflect the multiplicity of foreign influences now
penetrating Ethiopia. An illuminating example is the form of the crucifixion
adopted by several Ethiopian painters - a copy of the famous woodcut, from
the Small Passion by the German painter and illustrator Albrecht Dürer. All
these pictures bear witness to the remarkable capacity of the Ethiopians for
integrating foreign models into their own culture. They are also an expression
of a turbulent age, characterized by conflict with foreign forms that did not
easily fit the style of traditional Ethiopian art.
The pictures ofthe Restoration, the Gondar Period, finally stand out above all
the othen for their courtly elegance and graceful form. They reflect the slowly
vanishing splendour of the empire and are painted for a society of courtien who
found delight in their elegant and increasingly conventional forms. In the so-
called Second Gondar Period, from about 1700, the thrust ofthe artistic creation
steadily diminishes and finally degenerates into a canon of repetitive forms.
South-western Ethiopia: the emergence of a new culture and new states

We are reasonably informed about contacts between south-west Ethiopia and the
Christian empire to the north between 1500 and 1700 from Christian and Islamic
written sources, but they do not enable us to determine the influence Christian
Ethiopia actually had beyond political domination and exaction of tribute.
The oral traditions surviving in the Ethiopian south today provide a wealth
of historical information which has unfortunately still not been fully recorded.
Family trees, analyses of cultures and archaeological findings also give valuable
clues and prove that northern influence on the south was so far-reaching as to
cause the emergence of a new culture.
The cultural situation in southern Ethiopia before the fifteenth century,
when influence began to be exerted by the north, can only be deduced today
from the study of the contemporary culture of ethnic groups who remained
untouched by this influence until recently. Such groups are to be found among
the Gimira, Are or Dizi peoples, who still today lack state structures and
sophisticated political institutions.
Before the steadily growing stimulus provided by the Christian empire to
the north was able to bring about decisive changes, a great number of small clan
groups must have lived side-by-side in the south, held together by nothing more
than the sharing of a common language and culture, recognition of a common
lineage and, perhaps, veneration of a high priest or elder, who embodied the
religious affinities of the group. He was surrounded by a certain religious aura
by virtue of his descent from the founder of the group and he was credited with
supernatural powen, particularly over rain, crops and harvests. Compared with
their overriding religious significance, the political functions of these dignitaries
seem to have been almost non-existent. The real holden of political power
were the clans and their leaders as autonomous, originally self-contained units.
The conquerors from the north with their rigid concepts of state and king¬
ship swooped down upon this unsophisticated world of simple peasants.
Conquest and assimilation often proceeded comparatively peacefully, the
people of the south recognizing the cultural superiority of the north. Thus the
conquest of what was later to become the important state of Wolayta is supposed
to have been due to the political shrewdness of the invaden. In the batdes,
which involved small groups, instead of throwing spean at their opponents, the
northernen threw fabrics, strings of beads or pieces of meat. 'If you are so rich
and powerful that you can throw away such treasures', the Wolayta cried, 'then
shall you also rule over us!' Thus the invaden took possession of a country,
which under their influence, soon changed into a dynamic, aggressive state.
It was often accepted without question that monarchically-led states, which
were better armed and possessed cavalry, originally unknown in the south,
were superior to the democratically organized communities. There are there¬
fore reports of groups voluntarily surrendering to the invaders and accepting
the leaden as their new rulen.
In this way a whole series of new states came into being, states that were
never to forget their links with the north, whose ruling classes faithfully
preserved and developed the traditions and institutions ofthe north.
In addition to the purely external, formal structure of state organization and
court life, the myth ofthe Ethiopian empire also made its mark. It claimed that
they were the chosen people and theirs the true kingdom, and intended to
subjugate and assimilate all neighbouring peoples. The history of southern
Ethiopia, as it is known from 400 yean of oral tradition, is the story of the
expansion of these newly created states which widened their spheres of
domination at the expense of independent smaller groups and often contributed
to the founding of new dynasties among their neighbours. This process worked
in a chain reaction. Thus, according to oral tradition, the fint Gonga state,
Inariya, was founded before 1500 by an immigrant from Tigré. In the sixteenth
and seventeenth centuries, memben of the Busaso dynasty of Inariya founded
further dominions: Bosa, which was absorbed into the Oromo kingdom of
Gimma in the nineteenth century, Kaffa, which is reputed to have been the
most powerful state in all south-eastern Ethiopia for generations, and Sekko
(Mocea) in the virgin forest on the western edge ofthe Ethiopian Highlands.
Oral tradition has it that the kingdom of Gangero was originally founded by
Islamic migrants from Arabia. Later an originally Christian dynasty from Gondar
assumed rule. Similar accounts survive concerning the area occupied by the
Ometo: in Dawäro another group of migrants priding themselves on their
Gondar origin succeeded, after years of struggle, in uniting many small groups
of indigenous inhabitants into a powerful state. In about 1600, Wolayta was
conquered by the descendants of one Sum Tamben from Tigré. Other states,
both large and small, have similar traditions. Members of the Gosanaa dynasty,
probably also of northern Ethiopian origin, founded a total often dominions.
In southern Ethiopia, the social structure was originally characterized by the
coexistence and mutual acceptance of free and equal people. Only religious
leaden and warrior heroes enjoyed a few privileges. But in the newly-created
states under the influence of the hierarchical concepts of the Christian north,
things were quite different. Equality was replaced by a complicated hierarchical
system of ranks. At the summit stood the sacred king, inaccessible and often
invisible to his subjects. His hallowed family enjoyed special privileges, often
occupying the most important state offices. The exaggerated pomp and
ceremony of the royal households was often in flagrant disproportion to the
small size of the country and population. As the kingdoms expanded, so did the
royal households, developing into independent, powerful institutions with
hundreds of members. They were in sharp contrast to the former high priests
or clan leaden who had lived by their own labour and that of their families and
voluntary gifts from others.
The scale a royal court could attain is illustrated by Wolayta to the north of
Lake Abbaya. Here, the royal court was not only the centre of political life but
also a sacral district, protected by a multitude of ritual and ceremonial laws. If
the king appeared in public, it was only at a proper distance from the crowd or
surrounded by his followen. The royal court was protected by three ramparts
and three gates, each one triply secured. Only the king could slaughter animals
on the soil of his hallowed residence, or indeed exercise any kind of sexual
activity. Therefore the pages who took direct care of the king's person were
sexually innocent, forbidden just as stricdy as the pages ofthe Christian Ethiopian
emperor to leave the royal court and come into contact with other people. The
ground covered by the royal court was so hallowed that sexual intercourse
between a man other than the king and a woman would have resulted in the
death ofthe offenden and the transfer ofthe palace to a new site.
Under northern influence, the royal insignia were transformed. These had
formerly been symbolic objects, simple in material and form, such as sacred
spears, two-pronged forks or grass rope. Their place was now taken by rings,
mostly finger or arm rings, fashioned in the two precious metals, gold and
silver, which were the mark of royalty in High Ethiopia.
This High Ethiopian influence also had far-reaching effects on the whole
social structure. Although most southern Ethiopians were members of a single
homogeneous class, a sophisticated system of ranks developed after the High
Ethiopian pattern, which found its most marked expression in the worth
complex that extended over all Ethiopia. Worth was not inheritable like riches
or a respected name, but had to be earned anew by every man so that he could
find his publicly recognized place in society. The central criteria were success in
war and big-game hunting, and the warrior's obligation to kill. Only a penon
who earned worth could aspire to one of the innumerable elected offices, the
holding of which was the decisive criterion for a man's status. Name and rank
were often designated by the same word, and to lack a name was synonymous
with social failure. Here again the influence of the north is obvious, for the
names ofthe ranks are borrowings from the Amharic or Tigrai languages.
The free peasants were followed in the hierarchy by the artisans, who were
memben of specific castes and who occupied a comparatively lowly position.
Despite their great economic importance, they were nevertheless looked down
upon. They could not own land, and were subjected to demeaning regulations,
and physical contact with them was considered to be defiling for people of higher
birth.
In the lowest position came the slaves, who, in many countries, made up
more than a third of the total population. Most lived as serfs on their masters'
fields, but could own property. It was their labour and their productivity that
made possible the formation of a rich dominant class which, in Wolayta for
example, was the driving force in the great wars of conquest.
All these states had well-educated administrators, elected by the people and
confirmed in their positions by the king. The states were divided into com¬
munities, districts and provinces. They had public markets, supervised by the
state, and a good road infrastructure. They were protected against external
enemies by great ramparts and walls with carefully guarded gates.
The material advances brought from the north to the south since the four¬
teenth century are many. Curiously enough, the plough, the most important
agricultural implement in the north, never gained acceptance in the south.
Agriculture is, however, indebted to the north for a considerable number of
imported crops. It was probably not until after the fourteenth or fifteenth
century that peas, broad beans, chickpeas, onions and garlic reached the south.
Later came lentils and various oilseeds, including linseed, sesame, safflower and
nogo, all from the Near East. Later came the plants of the New World brought
from America to Africa in post-Columbian times, probably by the Portuguese.
Tlie Horn ofAfrica 367
Of these, red pepper, maize and tobacco are of such significance today that it is
hard to imagine life in southern Ethiopia without them. Pumpkin, sweet
potato, potato and tomato spread to a lesser extent.
Even today, the techniques used by the descendants of the artisans from the
north still bear witness to the impetus they gave this sector: treadle-loom
weaving, goldsmithing and silversmithing with sophisticated tools, leatherwork
with needles and fine woodwork for the production of door panels, wooden
platten and large bowls. Finally, the introduction to the south of the horse
should be mentioned, although it was without economic significance, being
used only in batde and as a prestige symbol.
Let us now turn to the influence of Christianity: was it strong enough to
have a permanent effect on the south and provide a lasting drive? Even today it
is clear that the cultural influence was very strong, its vestiges still being visible
as far as the borden of Kenya and Sudan. It is more difficult to assess the impact
of Christianity on the moral system of individual folk cultures, on customs and
usage and spiritual life in general.
The most striking Christian monuments in this area, by no means rich in
durable man-made works, are the many sacred groves, mainly on hills and
mountain tops, marking the sites of former Christian churches as their names
indicate: Kitosa (Christ), Mairamo (Mary), or Gergisa (George). The materials
used to build them were perishable and all have disappeared. However, their
sites are still regarded as sacred and the descendants of the former Christian
priests, who have become memben of non-Christian clans, still perform
ceremonies that are variants on the sacrifice of the Mass.
How strongly Christianity as a religious entity, not just as a collection of

rites, influenced the essence of southern Ethiopian religious ethics has not yet
been thoroughly investigated. The exceptionally important role of Mary, a role
that relegates Jesus Christ to the background, also needs investigation. The
function of Mary as a helpful and forgiving deity, particularly well disposed
towards women, can be seen in both north and south Ethiopia.
As regards external forms, an astonishing number of Christian laws and
ceremonies survived in the south until recently, when the Ethiopian Orthodox
Church resumed missionary activity. Fasting on specific days survived, as did
observance of the Sabbath, when all work in the fields was prohibited, at least
for the successon of former Christian priests. The seven-day Christian week,
regulating the frequency of markets, continued to be observed. Christian
Sundays have now become public holidays on which the kesigas (successon of
the Christian priests) gather their congregations together and in truncated ritual
invoke God, Christ and Mary, make the sign of the cross and burn grain
instead of incense. Among the great Christian holidays, the Feast of the Holy
Cross (27 September) became the New Year feast, with the exorcising of devils
and a great folk festival. Christmas and Epiphany - of such importance in the
north - were celebrated only by a few groups in the south, while Easter
became the day of the great ritual hunt.
East Africa: the coast
The sixteenth, seventeenth and eighteenth centuries opened with the arrival of
the Portuguese in East African waters and closed with the attempts by the
Omani Arabs to establish some kind of overlordship on the littoral. In those
three centuries, the coastal towns and peoples experienced important economic,
social and political changes.
The coast at the beginning ofthe sixteenth century

In 1500, the coast was still enjoying its golden age. The surviving ruins and
material culture of the fifteenth and sixteenth centuries conjure up a general
picture of prosperity. A closer look reveals differing degrees of economic and
material well-being.
Kilwa, which was growing in the fifteenth century thanks to the control of
the Sofalan gold trade which it had held since the early fourteenth century, was
experiencing the fint signs of decline on the eve of Portuguese intervention.
By the end of the fifteenth century, Kilwa had lost control over Sofala, and its
political influence was declining on the coast to the north, an area of com¬
mercial importance to it. Internally, the rapid succession of rulers 13 between
1442 and 1498 - was symptomatic of a political malaise that sapped its vitality,
whilst its bad relations with Zanzibar reflected a recurrent weakness of Swahih
towns internal dissension encouraging intervention by outside forces.

By 1454, Zanzibar was attempting to impose its nominee on the throne of
Kilwa. At this time, Zanzibar was not, however, a leading coastal town
politically or economically. It was itself divided into at least five settlements,
each with its own ruler. Mombasa, however, ranked with Kilwa and Malindi as
a leading city-state. Its merchants played an important role in the commerce of
the southern coast as far as Angoche. The island town had also established trade
links with its hinterland in honey, wax and ivory, which may explain
Mombasa's sudden rise until, by the time the Portuguese came, it had become
'the most powerful city-state on the coast' (E. A. Alpers, 1975).
Malindi, Mombasa's arch-rival, was also prosperous by 1500. Its trade was
based largely on the export of ivory and, secondarily, of such goods as beeswax,
368
East Africa: the coast 369
250 500 miles
25.1 The East African coast (after A. I. Salim)

ambergris and gum copal. Unlike most other setdements in the region,
Malindi, even in the sixteenth century, had large plantations of millet and rice
worked by slaves.
Further north, petty sultanates were to be found in the Lamu archipelago
which were in a state of virtually permanent conflict and so unable to present a
united front against the Portuguese invaden. Lamu town existed on its present
site by the mid-1 300s, but it was to flourish only after 1650.
In between these major commercial setdements were numerous smaller ones
which often fell within the sphere of influence of one or other of the big city-
states. The larger towns looked more towards international maritime trade than
the smaller ones, which depended on agriculture and fishing. The number of
settlements proliferated between 1200 and 1500, the coast's golden age.
It is difficult to estimate the size of these setdements in terms of area and
population. Malindi is said to have had a population of approximately 3,500 in

1498, based on the occupation of some 1,000 stone houses, but this figure
obviously does not include the plantation labouren or the poor, who probably
lived in mud-and-thatch huts.
The political fragmentation of the coast was much compensated for by

religious and cultural homogeneity, which was well-established by 1500.
African, Arab and Shirazi ethnic groups intermingled to varying degrees to
form a new cultural group known later as the Swahili (people of the coast).
Swahili is a generic term which the coastal people applied to themselves only
later. In the 1500s, they merely formed urban groups whose élite and ruling
families, for reasons of prestige, often claimed dubious, Arabian or Shirazi
origins, even if they were ethnically mixed.
While the coast was politically fragmented and its people came from
different groups, a number of features brought them together. The most
important were the common African denominator and the Kiswahili language
that was soon to evolve into the lingua franca of the coast. Other important
influences were Islam and, to a lesser degree, Arab influence, although the latter
became dominant only in the nineteenth century.
In fact, Swahili culture is a regional example of Islamic culture. It is regional
because it was influenced by the indigenous East African culture which, to a
significant degree, was incorporated into it. Although Kiswahili adopted a large
number of Arabic words over the yean, it is the most important single
manifestation of the African contribution to Swahili culture. Othen, by no
means insignificant, include rituals related to birth and marriage, funerary and
investiture ceremonies for chiefs, a belief in spirits and traditional dances.
The material prosperity of some coastal towns in 1500 was very impressive.
The rulen lived in palaces and the élite in stone dwellings, with elaborately
carved wooden doors. The townspeople's high standard of living is reflected in
the importation of such luxury items as damasks, silks, satins, copper objects,
Chinese porcelain and Middle Eastern glass vessels and glass beads. Only a
minority lived in luxury, however, while the poor and enslaved majority lived
in mud-and-thatch huts.
The populations of these setdements were predominantly African, with a

minority of Arabs whose numben increased substantially in the eighteenth and
nineteenth centuries. Rulen may have carried legitimate Arab dynastic names,
but they -were invariably of mixed descent and often black. Colour was of no
social or political consequence: it was the ukoo (lineage) that distinguished
someone and indicated his place in society and his degree of belonging.
The Portuguese intrusion

The Portuguese, like their Spanish neighboun in the Iberian peninsula, em¬
barked upon their voyages of exploration with the din of battle against the
Muslims still ringing in their ean. The motives behind the Portuguese expedi¬
tions were to outflank Islam commercially, politically, militarily and religiously
to break the Marnluk (later Ottoman) monopoly of the trade routes to Asia
and China and to link up with other Christians to eradicate Muslim rule
wherever it was found.
Prince Henry, Grand Master of the Order of Christ, was the man behind
Portuguese strategy, and a series of expeditions was embarked upon to put his
plan into effect. In January 1498, Vasco da Gama's fleet reached the southern¬
most fringes of the Swahili coast at the mouth of the Quelimane river, where
the Portuguese found an Afro-Arab settlement that was in touch with the
sought-after lands of the East, namely, the island and town of Mozambique,
which had fallen under the influence of Kilwa. The shaykh there welcomed
them until he discovered they were not Muslims, after which mutual suspicion
led to the fint skirmishes between coastal Muslims and Portuguese. Vasco da
Gama's reception in Mombasa was hostile because news of the Mozambique
skirmishes had preceded him. In contrast, the shaykh of Malindi greeted him
warmly. The two sides formed an alliance and the shaykh of Malindi acquired
Mombasa with Portuguese assistance at the close ofthe sixteenth century.
Vasco da Gama's fint voyage was successful in its objective of reaching India.
It had also yielded valuable intelligence concerning Muslim trade on the
Swahili coast, and the wealth of Swahili towns like Mozambique, Mombasa
and Malindi had impressed the Portuguese. This largely exploratory voyage was
followed by othen. One, in 1501, gave the Portuguese a glimpse of Kilwa' s
size but its shaykh, Ibrahim, would not be penuaded to co-operate with them
in establishing a trading agency in Sofala.
They therefore decided to use force, and the following year, to the accom¬
paniment of the sound of guns, Vasco da Gama's vessels sailed into Kilwa
harbour and, under the threat ofthe destruction of his town, Ibrahim agreed to
the payment of an annual tribute and to become a vassal of the Portuguese
king. In 1503, Ruy Lourenço Ravasco indulged in plunder and piracy around
Zanzibar, before landing there and using force to impose tribute. In 1505, on
the pretext that the town was behind in the payment of tribute and had not
hoisted the Portuguese flag as a sign of allegiance, Kilwa was attacked by Dom
Francesco de Almeida, later viceroy of India. Exploiting the factional rivalries,
he installed a pro-Portuguese elder, Muhammad Ankoni, as ruler and, at Kilwa's

expense, work was started on building Portugal's first fort on the Swahili coast.
Mombasa had shown hostility towards Vasco da Gama in 1498, and in
August 1505 greeted de Almeida's fleet with fire from a cannon recovered from
a sunken Portuguese ship. The king of Mombasa and his people put up a
determined resistance to the Portuguese, but eventually the king was forced to
surrender. The town was looted and burnt before the fleet departed. It seems
that the Portuguese aim was to break the power of Mombasa in order to
enhance that of Kilwa, which was to be their stronghold.
As an indication that Portuguese actions were blatandy punitive, in 1528
Mombasa was attacked again and occupied for some four months. This short
occupation, which ended in the town being razed, gained the Portuguese
nothing but the loss of many lives through fighting and disease.
Sixteen years earlier, in 1512, the Portuguese had been compelled to vacate
their fort in Kilwa. By then the Portuguese had been made to realize that there
were limits to how much tribute the town could pay and that that sum barely
covered the cost of maintaining a garrison. They had hoped to monopolize the
Sofalan gold trade, but had only succeeded in disrupting it with their ill-
considered interventions and commercial controls. By 1506, the garrison in
Sofala, stricken by disease, was no longer functioning, and in 1512, the Kilwa
garrison -was transferred to India.
With the departure of the Portuguese, Kilwa's trade improved significandy,
although the ruler remained nominally a vassal of the Portuguese crown. The
town's commercial pattern underwent an interesting change. Its traden now
avoided Portuguese-controlled Sofala and sought greater contacts with the
Angoche coast south of Mozambique which had long-established commercial
ties with the Zambezi hinterland. Their aim was to undermine Portuguese
control of Sofala, and it was to forestall this that the Portuguese ventured inland
and established themselves at Sena and Tete.
Kilwa had been heavily dependent on the trade in gold and ivory that its
traden brought from Zimbabwe, as it had little to export itself. Once Sofala
was lost, Kilwa had to develop trade links with the interior, and ivory became
its most important export. This reorientation of Kilwa trade coincided with the
expansion ofthe Maravi and the Yao.
In conclusion, then, the Portuguese failed to devise a clear-cut policy of
occupation or administration on the Swahili coast. They had only a mercenary
desire to monopolize all commercial traffic. Even in this they were only
partially successful. Because Portuguese occupation was so brief, the Swahili
towns managed to continue their commercial activities.
Many say that, from the outset, Portuguese interests really lay beyond in the
East, the Swahili coast being considered as only an appendage of the Estado da
India (state of India), headquartered at Goa where the Portuguese viceroy
resided. However, the Portuguese appointed a 'Captain ofthe Sea of Malindi'
with a few small ships at his disposal to patrol the East African coast, and he also
had the job of issuing passes to vessels and running the Portuguese trading
factory at Malindi. This factory imported cotton and beads from India which it
exchanged for local goods such as gum-copal, ambergris, ivory and coir: it is
proof of Portuguese interest in the African trade both on the coast and inland.
Until the advent ofthe Turkish expeditions in the later 1500s, sources give
only the scantiest information about the Swahili coast. It seems that a politico-
economic compromise had been imposed by circumstances: Swahili towns
remained independent so long as no conflict of interest arose with the Portu¬
guese. Some towns suffered more than othen. Mombasa recovered well from
the destruction wrought by the Portuguese in the guise of reprisals, and in 1569
was described by Father Monclaro as 'large and populous'; but he found Kilwa
almost deserted, though still trading in ivory with the Comoros and the
interior. By this time, a factor whose impact is even harder to gauge than that
of the Portuguese had entered the picture: population movements in the
hinterland and their effect on the Swahili towns.
From the moment they had entered the Indian Ocean, the Portuguese had
aroused the hostility of not only the local Muslim rulen but also the Mamluks
of Egypt. With the defeat ofthe Mamluks by the Ottomans in 1517, leadership
in the struggle against the Portuguese intruder passed to the Turkish sultan in
Constantinople. Several battles took place on the west coast of India and in the
Persian Gulf. Then, in 1570 an insurrection broke out in the Portuguese
stronghold of Hormuz, whose inhabitants had called for Ottoman help. The
uprising was suppressed, but its example was followed by other towns.
The king of Portugal had advised the viceroy at Goa to put an end to the
misconduct of Portuguese officials on the Swahili coast about which the
Swahili shaykhs had sent complaints. The king had feared that these leaders
would seek Turkish help to free themselves from the Portuguese yoke. Events
were to demonstrate that these fears were well-founded. When Mir 'All Bey
appeared in 1585, with a single galley, he was warmly welcomed by all Swahili
towns between Mogadishu and Kilwa except Malindi. He left with a promise
to return with a stronger force, after collecting booty and taking prisoner some
50 Portuguese soldien in Lamu. The Portuguese responded with a punitive
expedition directed particularly at Faza and Mombasa. True to his promise, in
1588 Mir 'Ali Bey returned with five ships and, once again, won the support of
most towns, except Malindi, which put up only token resistance. When 'All
Bey began to implement his plan to make Mombasa a Turkish base, the
Portuguese responded by sending a bigger fleet in January 1589.
The threat of Portuguese retribution coincided with the arrival of the
cannibalistic Zimba from the interior, who threatened Mombasa. The towns¬
people and their Turkish deliveren found themselves caught between two
enemies. The destruction of the Turkish fleet by the Portuguese allowed the
Zimba access to the island, where they went on a rampage of destruction. The
Zimba then proceeded northwards and it was only the fact that the Segeju
stemmed their advance that saved Malindi from a similar fate. Meanwhile, the
Portuguese expedition sailed north and wreaked vengeance on Lamu, which
paid dearly for supporting the Turks. The shaykh and several other notables
were taken to Pate and beheaded in the presence of the sultans of Pate, Faza
and Siyu.
To what extent the destruction wreaked by the Zimba, like that of the
Oromo later, contributed to the depopulation and decline of some Swahili
towns is difficult to gauge. Mombasa seems to have been sufficiendy weakened
for her erstwhile enemy, Malindi, to take the town with the help ofthe Segeju.
Thus ended the reign of the Mombasa Shirazi dynasty that had put up
continued resistance to Portuguese overlordship. It was replaced by the sultan
Ahmad, the ruler of Malindi, who was thus rewarded for his loyalty to the
Portuguese. But the long-term effect of the transfer to Mombasa of the captain
of the Portuguese garrison, together with Malindi's royal residence, was the
gradual decline of Malindi itself, which continued until the second half of the
nineteenth century.
Kilifi, a town between Mombasa and Malindi, also played a role at this time.
The rulen of Mombasa and Kilifi are said to have been related. Soon after the
Zimba onslaught, Kilifi seems to have fought Malindi for Mombasa, possibly
on the basis of the dynastic links between the two dynasties, possibly through
mere ambition. Malindi complained of Kilifi's raids and decided to put an end
both to Kilifi's provocations and also to its claims to Mombasa. After this
defeat, Kilifi seems to have suffered an irreversible decline.
The two Turkish expeditions had revealed the fragility of the Portuguese
hold on the Swahili coast. To strengthen it, the Portuguese decided to erect a
fortress in Mombasa and place a garrison there. Fort Jesus was built in 1593
and a garrison of 100 men was placed there under a captain whose jurisdiction
stretched from Brava in the north to Cape Delgado in the south. But the
strengthening of the Portuguese presence in Mombasa does not appear to have
led to any significant change in the administration of the rest of the Swahili
coast. Elsewhere, their presence remained minimal and payment of tribute was
all they expected from the town rulers. The Mombasa garrison made it easier
for the Portuguese to react quickly to any attempts at resistance and revolts
which might arise. In 1603, the ruler of Pate rose up in arms for which he was
condemned and executed. But this punishment did not prevent subsequent
uprisings by Pate and other towns.
Human dispersal and resettlement along the coast

Several important developments took place during the seventeenth century.
One was the entry ofthe Dutch and British into the Indian Ocean to challenge
the Portuguese. Another was the deterioration of relations between the
Portuguese and the new rulers of Mombasa, their traditional allies. A third was
the human dispersal and resetdement along the coast, which led to the
emergence of new groups.
Increased insecurity, arising largely from the pressure of the Oromo, led to
the movement of Swahili groups from the north to the southern towns. Pemba
and Mombasa both acquired immigrants, and it was during this period that the
anceston of the nine nations (Miji-Kenda) of Mombasa settled on the island.
Vumba Kuu developed further and became an independent state in the seven¬
teenth century.
Equally important was the dispersal south from Shungwaya (or Singwaya) of
groups which later developed into the Miji-Kenda peoples and the Pokomo.
According to T. T Spear, different groups moved south from Shungwaya and
setded in fortified villages (makaya; sing, kaya) on the hills overlooking the
Swahili towns between Malindi and just south of Mombasa. In time, the
makaya became more than mere havens of security from the Oromo and, later,
the Maasai. They came to occupy a major place in the socio-religious life of
the Miji-Kenda which outlasted the days ofthe fortified villages.
The conflict between Portugal and Mombasa

The amicable relations between the Portuguese and Sultan Ahmad lasted only
three or so yean after the transfer of the seat of government to Mombasa. The
captain of Mombasa was not honouring the administrative and fiscal agreements
made with his ally. The sultan complained of obstacles placed on Mombasa's
commerce and shipping. He also asked for recognition as ruler of Pemba, which
he seems to have conquered with his own forces. Finally, Sultan Ahmad was
given Pemba, but only on lease, his annual payment being 300-500 sacks of rice.
After Sultan Ahmad had died in 1609, his son, al-Hasan, found himself at
loggerheads with the new Portuguese captain, Manuel de Mello Pereira.
Relations between them became so bad that the sultan sought refuge in Arabaja,
rather than submit to a decision to be tried at Goa. At fint the Musungolos
(mainland Africans) gave him refuge, but later treacherously killed him in
return for pieces of cloth offered by the Portuguese.
Lisbon was not party to this treachery and, to make amends, insisted that al-
Hasan's son Yüsuf be made ruler after a period of education in Goa. There,
Yüsuf was converted to Christianity and given a Portuguese-Asian wife. When
he returned to Mombasa in 1630, neither the Mombasans nor the Portuguese
would accept him. In August 1631, the Portuguese captain, Pedro Leitào de
Gamboa, decided that Yüsuf s habit of praying at his murdered father's tomb in
the Muslim fashion amounted to renunciation ofthe Christian faith verging on
treason, and planned to ship him back to Goa for trial. Yüsuf decided to fight.
On 15 August 1631, during the Portuguese celebration ofthe Feast ofthe
Assumption, he made a surprise entry into Fort Jesus with his followen, and all
but a handful of the Portuguese were massacred. Yüsuf saw his dramatic return
to Islam as the beginning of a djiliâd to oust the Portuguese from the whole
coast. But his call was not heeded and no other town would support him.
Disheartened by his isolation, he left for Arabia - probably to enlist Turkish
help.
When this was not forthcoming, Yüsuf returned to the coast, where he
provoked minor uprisings until his death in 1637. The most rebellious region
was the Lamu archipelago, and the Portuguese had to undertake a punitive
expedition in 1636-7 to subdue Faza, Lamu, Manda and Pate.
Anglo-Dutch intruders
The appearance of the Dutch and English in the area had already contributed
to the decline of Portuguese supremacy by the time of Yüsufs revolt. The
Dutch had long been involved in the Eastern trade as agents, shipping to other
parts of Europe colonial goods brought to Lisbon. But when Spain and
Portugal were united in 1580, the Spanish king sought to cut out the Dutch,
who had been fighting since 1566 to liberate themselves from Spanish over-
lordship. This determined the Dutch to reach the East independently and, by
the end ofthe sixteenth century, their ships were challenging the Portuguese in
the Indian Ocean.
The English, meanwhile, had embarked upon piratical raids on Spanish ships
since 1580, when unity between the two Iberian powen made Portuguese
interests in the Indian Ocean legitimate targets. Before the turn of the century,
English ships had rounded the Cape of Good Hope. In 1591, one landed in
Zanzibar before proceeding to India, and in 1608, another dropped anchor at
Pemba. Thenceforth, the Portuguese had to fight these European invaders in
the Persian Gulf, along the Malabar coast of India and around Sri Lanka.
Neither the Dutch nor the English sought to replace the Portuguese by
permanently occupying any East African town. Stop-oven in Zanzibar and
Pemba were for provisions and water supplies only. Nevertheless, Anglo-Dutch
harassment stretched Portuguese resources, and when the English took to
helping the local inhabitants against their overlords, the situation deteriorated.
Although officially at peace with Spain and Portugal, in 1622 several English
ships helped the Shah of Persia drive the Portuguese from their strategic
entrepôt of Hormuz. The Portuguese fell back on Muscat, one of the Omani
coastal towns they had dominated since the early sixteenth century. As at
Mombasa, soon after the Turkish raids in 1588, the Portuguese had built a fort
there as an entrepôt for the IndianPersian Gulf trade.
The end of Portuguese rule in East Africa

Soon after the fall of Hormuz to the Persians, a new dynasty, the Ya'rubi
dynasty, came to power in Oman, determined to liberate the country from the
Portuguese yoke. Between 1640 and 1650, the Portuguese were forced by
military defeats to demolish their fortresses and evacuate Muscat. With control
over their own coastline, the rulen of Oman used their seasoned seafaring
inhabitants to build a strong naval force that started challenging the Portuguese
beyond the Gulf.
With the recapture of Mombasa after Yüsufs revolt, the Portuguese decided
to stay and rule the town directly. Pate seems to have entered a period of
prosperity during this time which led the Portuguese to establish a custom¬
house there in 1633. But the town's trade was making Pate too powerful; the
Portuguese decided to intervene, and relations between the two deteriorated.
Meanwhile, complaints against the Portuguese from other towns, such as Siyu,
Pemba and Otondo, reached not only Lisbon but also the new rulen of Oman
who -were asked for help.
What followed was a prolonged period of struggle during the second half of
the seventeenth century between the Portuguese and the Omani Arabs in East
African waten. In 1652, an Omani expedition attacked the Portuguese in
Zanzibar. The Portuguese retaliated by driving away the queen of Zanzibar and
her son, the ruler of Otondo, for encouraging the Omani Arabs. In 1660, a
combined fleet from Oman and Pate landed in Mombasa and sacked the
Portuguese quarter seemingly without any serious opposition. Other Omani

raids were made as far south as Mozambique (1669). In August 1678, the
viceroy himself led a major expedition against 'Pate the swaggering'. He was
joined by the ruler of neighbouring Faza, which was involved in its local
squabbles with Pate. The expedition was forced to withdraw on the appearance
of Omani ships in January 1679, but the allies had had the time to seize the
rulers of Pate, Siyu, Lamu and Manda, who were then beheaded, together with
several other notables. Pate seems not to have admitted defeat, as other
expeditions were sent against it, in 1687 for example.
An attempt was made to reach a compromise agreement whereby the shaykh
of Pate would retain his throne in return for accepting Portuguese overlord-
ship. But this provisional agreement reached in Goa was repudiated by Lisbon.
By then, Pate had fallen to the Omanis and, on Christmas Day, the captive
ruler and his councillon were killed in an attempt to escape from captivity in
Panjim.
For some years, civil strife at home interrupted the Omani raids on Portu¬
guese positions in East Africa. When they were resumed, their target was
Pemba, which had revolted against the Portuguese. In 1694, however, the
Portuguese managed to suppress the revolt and expel the Omani Arabs. But
the following year came the greatest challenge from Oman. In March 1696,
seven Omani ships, carrying 3,000 men and the shaykh of Lamu, landed in
Mombasa. They easily occupied the town and the island before laying siege to
Fort Jesus until December 1698, when the fortress surrendered. When the
Portuguese expedition from Goa arrived in Mombasa in November 1698,
with orders to engage the enemy, the red flag of Oman was already flying over
the fort.
From 1728 to 1729 the Portuguese managed a brief return to Mombasa,

taking advantage of the scanty Omani presence on the coast and of Swahili
disillusionment in some towns with the Arabs. There is no doubt that the ruler
of Pate was at odds with the Arab garrison in his town. When a faction rose
against him, supported by the Omanis, the ruler finally sought support from
Goa late in 1727. From Pate the Portuguese expedition moved against
Mombasa, where the Omani garrison in the fort had called for help from the
king of Pate against other Arab garrisons guarding smaller forts on the island.
Mombasa and its fort surrendered to the joint Portuguese-Patean forces.
But the return of the Portuguese and the Portuguese-Patean alliance was
short-lived. The ruler of Pate had been so anxious for help against his rivals and
their Arab supporters that he had promised Portugal more than he could
deliver, such as the payment of tribute and the monopoly of the ivory trade.
Within six months, he was in conflict with them. In June 1729, an armed clash
finally convinced the Portuguese that they should withdraw from Pate.
By then, trouble was brewing for the Portuguese in Mombasa. A combined
force of Mombasans and Musungulos attacked Portuguese positions in the
town and laid siege to the small garrison in the fort. In November 1729, the
garrison surrendered and was allowed to sail to Mozambique. Other towns,
including Zanzibar, Pemba and Mafia, had followed Mombasa's lead, murder¬
ing or driving out the Portuguese.
The Swahili towns themselves, therefore, were responsible for the final
expulsion of the Portuguese. But soon after they had left, Pate and Mombasa
were once again occupied by the Omanis, and the Swahili coast entered a new
era in its history.
The Portuguese defeat has been credited to a variety of faetón: their weak,
confused colonial system; the ineptitude and greed of officials; the harsh
vagaries of the climate and disease among an already small Portuguese popula¬
tion; and the local factional struggles, now working for the Portugese now
against them. Towards the end of the period, Portuguese resources were
overstretched and an expeditionary force hard to muster.
These military activities took place amidst significant political, economic and
cultural developments on the littoral. In the southern part of the coast there
developed an important trade between the Yao and Kilwa, which by the end of
the seventeenth century was dominated by a thriving trade in ivory. This trade
suffered a temporary setback in the early 1700s when the Portuguese lost
Mombasa to the Omanis. As a result, the Yao diverted their ivory from Kilwa
to Mozambique. But the advent of the Büsa'idi dynasty in Oman in the mid-
1700s ushered in another period of revival for Kilwa.
Further north, Vumba Kuu had consolidated its position and its ruler had
taken the tide of diwani. Its Shânfite ruling family, the Bâ-'Alawï, had adopted a
number of local Bantu customs which it combined with Islamic rites. The
Segeju and one of the Miji-Kenda groups, the Digo, who had settled nearby,
entered into close relations with Vumba Kuu. The religious influence of the
Shânfite families who ruled Vumba Kuu helped convert them to Islam, but
they retained aspects of their religious beliefs, including the rain-making
ceremonies.
In the Lamu archipelago, Pate was at its zenith in the seventeenth century
and into the eighteenth. Its prosperity was based on maritime trade, which it
exploited at a time when other towns such as Malindi, Mombasa and Kilwa
had recendy suffered attacks by inland peoples such as the Zimba. Pate
established profitable relations with the Oromo, who provided the town with
hides for export, and it also traded in ivory from the mainland. The historical
sources do not adequately explain its seemingly rapid decline during the second
half of the eighteenth century.
Tlie Omani factor in East Africa

The decline of Pate forms the subject of one ofthe greatest tenzi (epic poems)
in Kiswahili, Al-Inkishaft. Its author, Sayyid 'Abdallah bin 'Ali bin Nizir, was a
member of the Shârîfite élite of Pate. The shañfi who had setded in Pate became
indigenized, and with the shañfi of Lamu came to play a key role in the develop¬
ment of Swahili literary and religious tradition. The methods of teaching Islam,
the manuals used, the cult ofthe saints, the veneration of shârïfs and the form
and content of poetic vene constitute the cultural legacy of these descendants of
the Prophet Muhammad, who settled not only in the Lamu archipelago but also
elsewhere on the coast and formed the main part of its religious intelligentsia.
The Omanis contributed significantly towards the economic resurgence of
Kilwa, as did the slave trade. Together with the ivory trade, the slave trade was
to dominate the economic history of the east coast for much of the nineteenth
century. The trade developed initially to supply the French with the slaves they
needed, after 1735, for their plantations in the Mascarene Islands: Ile-de-France
(Mauritius) and He Bourbon (Réunion). Initially, the French purchased their
slaves at Portuguese-controlled Mozambique and Kerimba Islands. But
Mozambique's domination began to decline in the mid-1700s, partly because
of the challenge from Swahili and Arab merchants who had increased their
activities along the coast after the Büsa'idi dynasty had strengthened its grip on
the throne of Oman. Kilwa once again became a major trading centre to which
mainland African slave and ivory traders were drawn.
By 1785, Mozambique's domination ofthe ivory trade in Central and East
Africa had almost ended, although it continued to thrive on the slave trade,
owing to the increasing demand of the French Indian Ocean colonies. Yao
traden sold more and more slaves and ivory to Kilwa, which became a great
emporium for sought-after Omani or Indian imported goods, such as Surat
cloth. By the 1780s, Swahili or Swahilized traders were making pioneering
trading ventures into the interior beyond Lake Malawi.
Politics and trade were often two separate things: the Omani Arabs traded
with Kilwa, although its sultan jealously guarded his independence from Oman.
From 1784, however, the new imam of Oman, Sa'id bin Ahmad, embarked
upon a military campaign to impose his will on a number of coastal towns,
including Kilwa. The ageing sultan of Kilwa, Hasan Ibrahim, could put up only
token resistance. He was forced to acknowledge the imam's overlordship and
accept the presence of an Omani governor and garrison in Kilwa. But these
political developments only helped to promote the commercial revival of the
Kilwa coast by creating greater stability and encouraging more Indian
investment. No doubt Muscat gained even more from this arrangement.
By the turn of the century, the economic value of the few points at which
they were represented was becoming evident to the Büsa'idi. This and other
political and strategic considerations led Sayyid Sa'id bin Sultan (1804-56) to
extend his dynastic authority on the coast and, eventually, to establish his capital
at Zanzibar.
Conclusion
Between 1500 and 1800, there were important changes on the east coast of
Africa. Hitherto independent Swahili towns endured Portuguese invasion,
sometimes facilitated by their own intrigues and rivalries.
The Portuguese occupation was ruthless. The Portuguese were determined
to strike at Muslims everywhere, take over their lands and trade and even
convert them and the followers of traditional African religion to Christianity.
But the Muslims were impervious to Christianity, and by the end of the
Portuguese occupation no trace of Christianity remained.
As for trade and commercial traffic in the Indian Ocean, the Portuguese
only half-succeeded. The Swahili towns were able to retain a significant degree
of commercial activity despite the regulations introduced by the Portuguese.
Greed and incompetence amongst local officials were responsible for such
developments as the decline of the Sofalan gold trade. However, the coast
enriched Portuguese officials, if not the Portuguese crown.
Swahili towns experienced varying fortunes. Mombasa's resilience enabled it
to recover several times from Portuguese reprisals. Kilwa took the blows less
well, although towards the end of the eighteenth century it recovered, thanks
to the slave trade. Other coastal towns disappeared or reverted to villages. Their
decline was not the work of the Portuguese; the activities of the Oromo and
the Zimba, combined with environmental changes, are more likely explana¬
tions. Pate is an exception, and it enjoyed its greatest prosperity ever in the
seventeenth and eighteenth centuries; its decline began only at the end of the
eighteenth century. More research is needed to answer many remaining
questions about the Swahili towns, the factors responsible for their growth and
development and their eventual decline.
The Great Lakes region,
Í500-Í800
This period can be divided into three parts. The fint stretches from approxi¬
mately 1500 to 1580; the second, from 1580 to 1680, is the era of droughts and
famines; the third, from around 1680 to 1800, is the age of state formation and
expansion. Before 1580, the main preoccupation of the southern and central
regions was the consolidation of the states which had succeeded the Bachwezi
empire, while in the north, the fint intrusion occurred of Eastern Nilotic-
speaking peoples, with all its attendant linguistic consequences.
The second period was dominated by droughts and famines. It also saw the
region's greatest movement of populations for the past thousand years. There
were large-scale migrations out of Baar, the Luo people scattered, while
massive invasions of Banyoro reached areas as far away as Rwanda and
Usukuma. In the east, this drought was associated with the appearance of the
Central Eastern Nilotes who joined up with the local people, as well as a
substantial population drift southward.
The third period was concerned with the proliferation and expansion of
state structures. It began with a dynastic crisis in Bunyoro which heralded the
decline of that empire and brought about the creation of numerous states. The
Mporo state was founded then fragmented. The period drew to a close with
the rise and expansion of Buganda in the central area and Rwanda in the south,
as the two super-powers of the Great Lakes region, and in the north with the
broad outlines of the modern configurations of Central Eastern Nilotic ethnic
groups such as the Iteso, the Jie, the Eastern Luo, the Lango Omiro and the
Kumam.
The Northern and Central areas
Formation of new groups and societies

Before 1000 of the Christian era, the Central Sudanese were probably the
predominant peoples over the entire Great Lakes region. 'Muru' was the Luo
term for all these indigenous peoples, and will be used when the ethnic identity
of a Central Sudanese people is not known and particularly to distinguish the
381
early peoples of the region from the later Madi immigrants, who were also
from Central Sudan. Most of the Central Sudanese peoples were organized
along segmentary and gérontocratie lines. They were farmers. They wor¬
shipped an earth god and used rain stones or a mixture of oil and water to bring
on the rain. They had a complex totemic system and buried their dead in
recessed graves. They were known by later immigrants as experts in ironwork.
By 1500 they had been assimilated into Bantu culture, south ofthe NileKyoga
divide, but still dominated the north from the Alur Highlands to the mountains
of Karamoja.
Around 1000 the Nilotic birthplace of Dog Nam broke up. One group
moved to Tekidi and a splinter group from Tekidi migrated to and setded in
Pakwac-Pawir. From Dog Nam another group settled in Wipac (Rumbek). In
all three Luo enclaves a form of hereditary command associated with the
bushbuck totem had been established well before 1400.
The period from 1400 to 1580 was devoted to the internal consolidation of
the states which succeeded the empire and the diplomatic manoeuvring of the
two great powers, the Bahinda and the Babito. The new rulen recognized
certain indigenous chiefs, notably those of Bwera and Buyaga, and there were
few arguments over succession.
North of the NileKyoga divide, the year 1500 is important because it
possibly coincides with the first major appearance of the northern section of
Eastern Nilotes or their Bari cluster. It has been argued that this group left its
homeland in three waves occurring some time between 1490 and 1571. Each
group in turn attacked Tekidi until this Luo setdement was broken up by the
third invasion, that ofthe Lotuho.
Available data suggest that the Nilotes introduced to the peoples of Central
Sudan their organization by age group, their fire ceremony, a sky god, the spear
and shield, the sacred spear for bringing on rain, the long straight hoe and
elaborate headdresses. In many areas their interaction with the Sudanese
peoples probably developed chieftaincies and their language became a new
lingua franca.
The second Eastern Nilotic invasions may have been affected by the
Oromo, who left their homeland north of Lake Turkana between 1517 and
1544 and attacked Tekidi around the same time as they began their invasions of
southern Ethiopia. Anceston of the modern Kakwa possibly took part in this
migration.
The land of Baar became an area of intense racial intermixing between the
Eastern Nilotes and the Madi people. It appeared that new societies of the
Pajulu, Kakwa and Kuku peoples adopted an EasternNilotic language and
combined the Nilotic sky god and spear with the Sudanese earth god and bow
and arrow. They apparendy abandoned the Nilotic organization by age group
and adopted Madi dress, burial customs and rain stones. The secular chiefs were
generally Eastern Nilotes whereas the ritual experts were Madi. The Pajulu,
Kakwa and Kuku were culturally Sudanic and linguistically Eastern Nilotic.
The Bari fusion was very similar, with a somewhat more marked Nilotic
The Great Lakes region, 1500-1800 383
influence. One ofthe main features of Bari and Kuku society was that the dupi
(servile class) were sometimes physically distinct from the freeborn. In Kakwa
and Pajulu societies the dupi were clients rather than serfs. Since they were
often used as assistants in rain-making rituals and were renowned minen and
ironworkers, they may well have been of Muru rather than Madi origin.
The final invasion of the northern group of Eastern Nilotes was associated,
in the oral Luo Tekidi tradition, with the Lotuho. The Lotuho overcame the
Luo settlement of Tekidi, whose king, Owiny Rac Koma (c. 1544 fled
with his subjects to the Luo setdement in Pakwac-Pawir. It would appear that a
Central Sudanese group, the Okarowok, adopted the Eastern Nilotic language
due to the influence ofthe Lotuho.
Apparently the Lotuho assimilated the small Okarowok clans into their own
four larger clans and adopted the Okarowok totemic prohibitions. They intro¬
duced the fire-making ceremony, organization by age group, and, probably
because of their unsettled situation, built tighdy-packed, well-populated villages
in an area previously distinguished by its scattered settlements. This was the
furthest eastward spread ofthe totemic clans and rain stones indicating, possibly,
the eastern limit of Central Sudanese peoples.
While a group of proto-Luo was forced by the incunions of the Eastern
Nilotes to move to the Tekidi area, another group moved northward to
Wipaco Dwong' in Rumbek. This group subsequently broke up between 1382
and 1418 and went westward, towards the Nile, to setde on the river banks.
Later, this cluster re-divided, with one group the proto-Shilluk and its allies
going north and settling somewhere between 1490 and 1517. The Luo forays
into Baar apparently drove the Madi-led Panyimur, Atyak and Koc-Pagak to
migrate south to the Mount Kilak area. There ensued a Madi-Luo struggle for
power in Pakwac-Pawir. In the chiefdom of Atyak it was prophesied that the
sons of Princess Nyilak (c. 1517-62) would assassinate her father, the king, and
seize the throne. Consequendy, the king locked up his daughter, but a Luo
traveller gave Nyilak a son, who, as prophesied, killed the Madi king. Out of
this, two states probably emerged; Attyak (later known as Okoro) led by the
Luo, and Acholi, by the Madi. The story of Nyilak is an allegory (whose
characters vary according to the version) intended to explain the overthrow of
Sudanese rule by either Luo in the north or Bantu in the south.
Around 1560 the Lotuho overran Tekidi causing most of its inhabitants,
along with their king, to flee to Pakwac-Pawir. The main chiefdom to emerge
from the Lotuho was Puranga. This Luo group, much influenced by Eastern
Nilotes, can be called the Eastern Luo. They formed a predominant element of
the population in East Acholi, Labwor, Nyakwai and Lango and along with the
Eastern Nilotes had a decisive influence on the dialects spoken in these areas.
The period of droughts and famines

No area of the Great Lakes region, nor indeed of east-central Africa, escaped
the effects of the climatic disaster of droughts and their resulting famines which
occurred in the late sixteenth century and beginning of the seventeenth
o o
S Y [/)
1 (T >
LU _ U O
m
CÛ 2
5 s O
Û
(1 Y y 1 LL
m m m m
T
8« m n
<J
< I
< =>
03
C)
<N
t>
¿
r^
-s
£
c
a
C
ni
Xi
S
n
>-
Z
«S
century. During this period, rainfall in the northern and central areas was only
average for two yean. The worst years of drought were probably 1588
1601-02, 1613 and 1617-21. In 1621 the summer Nile reached its lowest level
since 622 when records were first kept.
In northern Uganda the entire period is known as Nyarubanga, a Luo term
meaning 'sent by god'. It came to an end with the Great Famine of 1617
which was accompanied by a disease which wiped out catde herds. The spectre
of mass famine is conjured up in all the traditions. In Egypt, accounts that were
handed down are appalling. Cannibalism became commonplace and almost half
the population died of starvation. What was true of Egypt was probably also
true ofthe Great Lakes region, especially the northern part.
If half the population died, most ofthe survivors were forced to migrate and
flee, gathering near riven, lakes, along the banks ofthe Nile, of Lake Nyanza
(Victoria) and the chain of lakes stretching southward into central Africa.
Twelve major accounts of migration in this period have been collected.
The Nyarubanga reshaped the ethnic and political geography of much of the
Great Lakes region and perhaps well beyond. It probably dealt a tremendous
blow to leadership built on the concept of control over supernatural forces and
especially to those who claimed rain-making powen. It bred a new type of
leader, based on those who headed the migrations and whose skills as com¬
manders, warriors and hunters assured the survival of large groups of people.
Once the shock had passed, the leaden of the survivon created new political
units in new regions which were probably spanely populated and had no
political structures of their own.
It is not therefore surprising that traditional sources give the impression that
from around 1600 there was a new beginning. In the majority of areas, there
was, once again, ethnic intermixing, new habitats and new rulers. If accounts
prior to 1600 are skeletal and vague it is because the Nyarubanga destroyed oral
traditions, wiping out their means of transmission along with most ofthe elders
who were their carriers.
The migrations out of Baar gready accelerated during the Nyarubanga.

Pakwac-Pawir became a place of refuge for huge crowds of starving and
desperate peoples of different languages and cultures. The influx of these
people apparendy destroyed the position of the Paluo north of the Nile, and
made the question of who would control the Luo and other groups there even
more pressing. This led to a conflict between Cwa Nyabongo and Nyipir. This
dispute is recounted within the spear-and-bead tradition. Nyipir and Tifool,
whose followen were largely Owiny and Omolo Luo, moved west ofthe Nile
out of the land of Cwa. According to tradition, Nyipir drove an axe into the
Nile's dry river-bed as a sign of eternal separation. As leader of Attyak, Nyipir
founded the Alur kingdom of Okoro whilst Tifool created a state which broke
up into two parts, Nyiganda and Angal. The main body of the Owiny
apparendy headed south-eastwards, passing north of Lake Kyoga and joined up
with the Budola encampments in eastern Busoga. According to Jonam
traditions, the remaining Luo, led by Daca, overthrew Abok and the Koc state
split into three; Koc Ragem to the west of the Nile, Koc Labongo to the east
and Koc Paluo led by the Madi.
The Luo founder of the Terego chiefdom of the Lugbara may have emerged
out of the dispenal from Pakwac-Pawir. Banale and his nephew Raoule
reached Madi country in a state of total destitution. Banale was accompanied
by a leprous Madi woman. When their clandestine relationship was discovered,
a council of Madi elders decided to accept him as chief. He founded Terego,
the biggest chiefdom among the Lugbara or southern Madi. At the same time,
Jaki and his ambitious sons were extending their rule over the neighbouring
Madi country and it is possible that the council of elden preferred to have
Banale as their leader rather than be absorbed into the Eastern-Nilotic speaking
Jaki group. It was this decision which allowed them to preserve their language
and remain essentially Madi in culture.
The Nyarubanga triggered three Luo migrations to the south-eastern bank of
Lake Nyanza. A significant body of the Adhola clans left Pakwac-Pawir and
joined up with the Owiny in Kaberamaido. One of their leaders was Amor,
and the Amor clan later appeared at Padhola with their drum, sacred spear and
bushbuck totem. The Adhola and Owiny groups rapidly moved south to setde
in camps at Budola in eastern Busoga. Another group of clans who were part of
the Omolo cluster, originally of Luo-Sudanese origin, left Pakwac-Pawir to go
north to Tekidi and south towards Mount Elgon. Thus there was a gradual
build-up of Luo settlers in eastern Uganda, who then combined, probably with
the arrival oí Nyarubanga refugees, to form two separate bodies.
By 1600, the original setden had joined together with the immigrant clans
to form at least four groups: the Omolo group in the Banda camps, the
Owiny-Adhola in the Budola camps, two future Adhola clans (the Ramogi and
the Lakwar) in the forests of west Budama and a group of future Busoga north
ofthe Mpologoma river. Between around 1598 and 1650, both the Budola and
Banda camps broke up, probably because ofthe terrible drought in the 1620s
which marked the end ofthe Nyarubanga. The Owiny headed off towards the
Alego region in the Siaya District in western Kenya. A section of the Omolo
followed and both groups came into conflict. The Adhola clans were the last to
leave their encampments and join their brothers in west Budama.
Among those who retained their Luo language in west Kenya, some created
chiefdoms. The majority, however, were like the Padhola, who settled in an
unoccupied area and were content with an acephalous political system (i.e. one
with no head) which maintained peace between the 31 clans. Ethnic unity was
furthered by the myth that all clans were descended from the sons and
descendants of Adhola, the chief who had led the great migration out of
Kaberamaido to west Budama. The Padhola gradually gained land to the south
and east of their original settlement. After about 1650 they also fought the
Maasai in the Tororo area.
Eight clans belonging to the Owiny and Omolo clusten moved out of their
Budola camps into Busoga. They were all members of the bushbuck totem,
with hunting and rearing traditions, although the Owiny group also had a
farming tradition. All of the Owiny clans, along with the Wakoli of Omolo
origin, became ruling clans in Busoga. The Omolo, who were more pastoral,
were reluctant to settle, and only one of their clans became dominant.
Soon after there followed the Pakoyo migrations from Pawir south of Lake
Kyoga, culminating in their founding ofthe chiefdoms of Bagweri and Bugaya.
The Luo migrations gathered many retainen along the way through inter¬
marriages and also because they took over the principal shrines, which then
became centres of ritual and royal pilgrimage. Wherever the Luo went, they
rapidly accommodated existing local rituals and religious structures; their ability
to adapt spiritually was one of their greatest political weapons.
By 1750 nine states had been founded. By the end ofthe nineteenth century
these had fragmented into nearly 30 states, with the Ngobi or bushbuck clan
dominant in almost 20 of them. Those states which maintained a father-to-son
succession and used administraton who were commoners, experienced no

secessions and few civil wars. But those which practised brother-to-brother
succession and used royal princes as office-holders suffered repeated disputes
over the throne and numerous civil wars. Those dynasties which held on to
their original Luo principles were more stable than those which adopted the
succession rights of their Bantu subjects.
The Nyarubanga and the associated spear-and-bead movement was of great
historical importance for the Luo, as it determined their modern geographic
distribution. The Nyarubanga also set off a massive southern Banyoro invasion,
headed by the Bahima-Batutsi, and led to the rural classes' attempt to over¬
throw the supremacy of the herdsmen. The first stage of the invasion was led
by disorganized and ravenous hordes which swept over the whole of the
southern region, some ultimately settling in Buzinza, in Burundi and beyond.
They in turn sparked off other migrations which followed the chain of lakes
into Central Africa.
As the drought penisted, cattle were struck by a disease which wiped out the
royal herds of Kitara. This prompted Cwa II into driving his armies southward
to replenish the herds and reinforce royal authority. The herdsmen were the
obvious target ofthe Banyoro and the agriculturalist Cwa's natural allies.
The Nyarubanga also introduced the central group of Eastern Nilotes on to
the scene ofthe Great Lakes. Historically, this group, made up essentially ofthe
Karimojong-Teso cluster, was divided into two: the Isera, agriculturalists who
reared some catde, and the Koten, catde-rearers who went in for a little
agriculture. While both Isera and cattle-rearing groups are found in all peoples
of the central group of the Eastern Nilotes, the ethnic groups to emerge
primarily from the former are the Iteso, Toposa and Dodos, and from the latter
the Karimojong, Jie and Turkana.
The Okarowok (Ikarebwok) form a major clan among the central group of
the Eastern Nilotes. They seem to have first appeared in the Agoro region
when the Lotuho imposed their language and customs upon the Lukoya, a
Muru population. Agoro was an important centre of Isera-Omiro dispenal.
The major migrations to the south-east of the Luo, linked with the Owiny
and Omolo groups, were also very much mixed with the Isera, especially in the
case of the Omolo who went southward down the AgoroKaramoja corridor.
The Owiny left behind pockets of Luo speaken in what later became Lango
Omiro country in Amac on the northern banks of Lake Kyoga. A significant
part of the population eventually identified as Omolo were perhaps originally
Eastern Nilotes. Those who reached western Kenya were known to the Luo as
Omia, the Lomia being one of the four major Lotuho clans. The last wave of
migrants to arrive in Siaya also included groups coming from eastern Uganda,
such as the Owila, the Matar and the Bayuma. They were probably all Isera-
Omiro descendants ofthe Bako, who, in turn, were of Ethiopian origin.
In addition to the Central Sudanese peoples and the Northern group of
Eastern Nilotes, the Isera included clans of Luo and Ethiopian origin. These
groups were reinforced by immigrants undoubtedly from Anywa, or Pari, from
Baar and Pakwac-Pawir.
In the course of the seventeenth and eighteenth centuries these migrant

groups setded together and formed clusten which were defined by their place
of residence and, in some cases, by a certain degree of self-identification. The
largest of these bodies were the Bako in central Jie, the Okii (or Miro) in the
Mount Moroto-Omanimani River area of central Karamoja, the Abwor in the
Labwor Hills, the Kumam in eastern Teso as well as the Iworopo-Iteso (unique
because the least coherent of these groups) in central and southern Karamoja.
All of these groups included both Luo and Eastern Nilotes in quite differing
proportions and bilingualism was common by the eighteenth century.
Pastoralists and farmers
Historians and anthropologists have tended to divide the societies of the Great
Lakes region into two watertight compartments; those who cultivate the land
and those who raise animals. But in reality the situation is much more complex,
dynamic and changing. Population movements and other social upheavals
brought together groups whose ideas and customs were often very diverse.
Throughout the period with which we are concerned here, linguistically
different groups of herdsmen and crop farmers were fusing together to form
new and relatively well-integrated societies. Here are a few examples which
well illustrate the nature of this social transformation.
Amongst the Central-Eastern Nilotes, a majority of the animal-tending

groups setded in the Mount Koten-Magos Hills area of north-eastern Karamoja.
These Koten-Magos peoples apparently already had a pastoral orientation before
they came south. Hence whilst they arrived in Karamoja with a meagre herd,
they quickly developed a mixed economy based on rearing, agriculture,
hunting and gathering, which was better adapted to the region's climate than
the economies of more agricultural groups. The results of this evolution
became evident between 1680 and 1830 when the Koten-Magos groups
moved into their present homelands and the agricultural groups either merged
with them, adopting their economy, or else retreated to the west.
This process started between 1680 and 1750 with the expansion of the
pastoral groups in four directions: the Ngimonia spread along the River Tarash
and formed the core ofthe Turkana; the Korwakol moved west to the Longiro
river, and became the most important element of the Jie; a third group, made
up of some future Karimojong and Dodos, headed south to the Apule river;
and the fourth, the Ngikora, emigrated north and settled in Dodoth.
The situation then stabilized for a time, but the demands put on the land in
areas controlled by the pastoralists, along with the droughts of 1780 and 1830
sparked off new migration movements. The majority of the populations that
moved into Nyakwai, Labwor, eastern Acholi and Lango were at least bi¬
lingual: they spoke a Luo dialect and as they were all joined by Luo-speaken
from further west, they adopted the Luo language. Those, however, who went
into Teso included few Luo-speaken and thus the Iteso adopted the Eastern
Nilotic tongue.
This western flow of agriculturalist refugees was accelerated by a renewed
expansion ofthe pastoralists. In the 1780s the proto-Dodo groups on the Apule
went north towards Dodoth, forcing the existing Toposa community to move
north to their present territories. Both the Dodo and Toposa herdsmen assimi¬
lated a number of Isera groups and thus adopted an economy that was more
agriculturally based than that of their southern brethren. Then, in the 1820s,
the Korwakol and the Jie Rengen combined to destroy the huge agricultural
community of the Poet who lived on the Kapoeta river. Finally, during the
same period, the proto-Karimojong on the Apule pushed south into the home¬
lands of the last remaining Okii and Iteso-Iworopom, either absorbing them
into their society or forcing them to leave. These developments, along with the
1830s drought, precipitated the last great migrations from this area to the west.
The case of the Bunyoro-Kitara even better illustrates the interrelations
between pastoralists and farmers. Between 1760 and 1783 Bunyoro lost
valuable grazing land to Buganda. Also, the growing autonomy of the animal-
tending Nkore, Buhweju and Buzimba societies further restricted the Bunyoro
state to largely agricultural lands. Pastoralists settled amongst agriculturalists and
together formed new communities where land, rather than cattle, was the basis
for political authority. The different groups, united by wide-scale inter¬
marrying, provided leaders and land owners who collected tributes from the
subjects for the king. Thus a new Nyoro state gradually came into being in the
eighteenth century.
Social and political institutions

It was not only language and economy which distinguished the animal-tending
groups from the Isera groups, and both from the Luo-speaken. The herdsmen
perfected their age-organizations as instruments of gérontocratie control, and
these became the key social and political institutions in their societies.
Organizations by age-group also existed within the Isera and the societies they
influenced, but in a truncated form. The main institutions in these societies
were the ritual groupings of several clans, which were used for settling disputes
and performing religious ceremonies.
In the ethno-cultural mix north ofthe NileKyoga-Bisina divide, hereditary

chieftaincy was a Luo idea. This type of hereditary leadenhip was opposed by
many Eastern Luo and refused by their elders. Nevertheless it had a certain
impact on the political institutions of not only many of the Luo-speaking
groups, but also on certain Isera groups. Among the Luo-speaken, those that
followed chiefs tended to move into eastern Acholi, where their ideas were
reinforced when some Paluo groups came to join them somewhere between
1680 and 1760. In Labwor, Lango, Nyakwac and Kumam, the populations, on
the whole, rejected the idea of a hereditary chieftaincy in favour of géronto¬
cratie ritual rule.
Despite this, these groups sometimes entrusted individual leaders with a

good deal of secular authority. The eighteenth and nineteenth centuries wit¬
nessed an increase in warfare amongst the Eastern Luo and the Central-Eastern
Nilotes. This in turn increased the influence of the military leaden and their
councils, as well as of the younger generation, at the expense of the
gerontocracy. Thus, in Lango, powerful war leaden formed extensive military
confederations in order to conquer their country and raise large mercenary
armies which fought in Bunyoro, Toro and Buganda. Similar confederations
were formed in northern Teso, Sebei (Mount Elgon), Kumam and Padhola,
but in these regions the chiefs were emuron (seers). Because of their decisive
military role, these men gained considerable political muscle but none was able
to secure a chiefly status for his family. The institution of emuron created a new
centre of power which challenged both the war councils and the gerontocracy
but did not combine easily with chieftaincy.
The Iteso who settled in the 'thigh ofthe cow' (the unusually fertile area in
Ngora and Kumi as well as lands colonized from there) closely resembled the
original Isera. They had no age-group organizations, no etem (ritual clan
groupings) and no emuron. They were dependent upon three great assimilating
clans, the Atekok, the Ikarebwok and Irarak who ensured social cohesion.
Being more mixed originally, the Northern Iteso also possessed all the above
institutions. Many Iteso filtered into Luo and Bantu territory to the south, as
land pioneers or mercenaries. They were made particularly welcome in certain
Busoga states where they counterbalanced military and political pressure from
Buganda.
In the central area, Buganda contrasted sharply with Bunyoro. Without a
royal clan, but with a king who could appoint officials with no regard to
custom, tradition or heredity, Buganda had, by the beginning ofthe nineteenth
century, few institutionalized classes, unusual vertical flexibility and a degree of
social inequality which was greater than Bunyoro but less than Rwanda.
Politically and socially, Bunyoro-Kitara stood somewhere in the middle. The
royal Babito clan was originally made up of agriculturalists who also reared
some stock, but the dynasty increasingly acquired a pastoral orientation.
However, the non-royal branches of the Bito clan were to be found at all
economic levels. Furthermore, representatives of other social classes and allied
groups were appointed to various state and court offices, so that a privileged
Tlie Great Lakes region, 1500-1800 391
class was created whose fortunes were linked with those ofthe Bito dynasty.
The mukama of Bunyoro had far less wealth and power than the kabaka of
Buganda, since in Bunyoro the royal court carried out redistribution of wealth
and surplus. Power was less concentrated here, with most positions remaining
hereditary and the mukama merely confirming the chosen candidate; conse-
quendy, few peasants ever attained high office. Furthermore, while in Buganda
the 30 clans were theoretically equal, in Bunyoro there was a clear distinction
between high- and low-status clans.
Among the Nilotic Luo, there were those who were organized in states and
those who were non-centralized. State structures were less elaborate than in
Bunyoro or Buganda and Luo kings exercised a certain influence but little real
power. Although some royal lines go back over several centuries, most of the
Luo states only came into existence belatedly. Furthermore, recurrent droughts
had forced so many migrations and new regroupings of societies that many
states, such as they exist today, date from about 1680 or even later.
None of the Luo states enjoyed the fertile agricultural base of Buganda,
Rwanda or the Bahima states. Neither did they possess the rich iron and salt
deposits of the Bunyoro or Busongora areas. Prior to the Nyurabanga, it seems
that the Luo owned substantial herds of cattle but by around 1800 these had
been wiped out. Consequendy, the Luo state did not exhibit the same
disparities of wealth nor the class consciousness which characterized the south¬
western states. In all the Luo states, the Eastern Nilotes were very influential,
with their presence increasing from west to east and making up the majority of
the population of eastern Acholi. They were extremely egalitarian in their
political and social concepts and it must have been difficult to convert them to
ideas of chieftaincy and hereditary classes. In the eastern regions, the Luo kings
appeared increasingly like presidents and spokesmen for the councils of elders.
In the Luo states, theory and practice regarding classes differ. Theoretically,
the Luo societies are divided into two classes, the royals and the common
people. However, by around 1830 a third distinction, based on Luo or non-Luo
origin, became as important. Very often, royal has come to be identified with
Luo and non-Luo with commoner. In popular thought, non-Luo commonen
are often graded according to how long ago their anceston integrated into Luo
society. Nonetheless, a member of the royal clan separated by ten generations
from the chiefly line is treated little differently from an outsider whose
ancestors integrated into Luo society ten generations ago. In the Acholi states
the royal clan is exogamous and outsiders are quickly assimilated.
The Luo class system can be looked at in a more realistic way. In the coune
of political compromises made over centuries, many non-Luo clans came to
hold office or perform ritual functions which gave the lineages of those office¬
holden great prestige. If there was a superior class, it was that of the nuclear
families of the king, his councillon, the landlords and the ritual experts,
including the rain-makers and the jagos (territorial sub-heads).
The Luo and the Central-Eastern Nilotes, who had no state structures, had
no class system. There were no chiefs and no hereditary offices. The
gerontocracies were governed by the elden whose decisions, it was believed,

were sanctioned by the anceston. These gerontocracies often formed huge
military confederations in which a warrior chief and his subordinates were
recognized and supported by an emuron. There was tension between the
governing elden and the younger warrion. The elden maintained strict and
even oppressive control over the younger men, delaying their possibilities of
marriage and independence by monopolizing both women and catde.
Sometimes the young men emigrated to escape such frustrations.
The acephalous Luo and Eastern Nilotic societies often showed great
economic disparity, manifested in the number of head of catde owned. Wealth
brought respect and aroused expectations of lavish hospitality but it did not
create social inequality. Actually, catde were widely redistributed through the
bridewealth system.
The change in the balance of power; the decline of Bunyoro and the rise
and expansion of Buganda
After the Nyarubanga, the most significant episodes in the history of the
northern and central areas of the Great Lakes region were the crisis in
Bunyoro, which led to its decline, and the internal development in Buganda,
which led to its rise and expansion. These two factors led to a dramatic change
in the balance of power. The Bunyoro crisis began with the death of Cwa at
the time ofthe Nyarubanga-led invasion ofthe south. Cwa's only son, Winyi II,
was captured and Kitara was governed by a regent, Cwa's sister, Mashamba.
Cwa had no sons by a Luo wife, who could have succeeded him according to
the rules of the royal family. Winyi II murdered Mashamba and seized the
throne. Thereafter, the old tradition whereby the heir had to be of a Luo or
Paluo mother, became the exception rather than the rule. Candidature for the
throne was thrown open to all the king's sons and succession disputes became
more frequent and more fierce.
Under the fint dynasty, the chiefdom of Pawir had held a prestigious
position within the imperial structure. A succession of what the Paluo called
'Bantu kings' sought to make use ofthe unrest at Pawir to encourage separatist
tendencies; this they did with such success that the sub-chiefdom of Pawir,
which was unified in 1650, had split itself into six small chieflets by 1750.
Under the reign of Isana (c. 1733-60), these chieflets were placed under the
authority of a country chief by which Paluoland lost its semi-autonomous status
and was integrated into Bunyoro. The Paluo regularly rejected the Bantu
candidates to the throne; some emigrated to the north and east because of the
closing of their political horizons, their loss of status and sheer penecution. The
frequency of succession disputes, combined with Paluo dissidence, undermined
the central power of Kitara.
The most immediate result of the crisis in Bunyoro was the exodus of the
Paluo-Pakoyo into Acholi, northern Busoga, Alur and even into Padhola and
western Kenya. Whilst the Paluo were spreading their language in the north,
the Pakoyo integrated linguistically into Basoga society. In the north, the major
form of political organization was gerontocracies or small chiefdoms. The Paluo
popularized the royal drum, enhanced the dignity ofthe chief and incorporated
small units into larger states, allowing them to retain their hereditary leadenhip.
They founded new states such as Lira Paluo and Paimol and enlarged and
restructured othen, such as Padibe, Patongo, Alero and Koc. Puranga and
Payera were inspired by Paluo theories and assimilated a number of subordinate
units in the process of expansion. Integration into Luo society proceeded either
by assimilation or by fragmentation and proliferation of political unities. Lira
Paluo is an example ofthe former, Alur ofthe latter.
Omukama Isansa played as essential a role in the decline of Bunyoro as did
Kabaka Mawanda in the rise of Bugunda. Isansa's rule is noted for its peculiar
mix of military success and political failure. The campaign that he led in the
south was intended to demonstrate once again Bunyoro's military might. In
reality, Isansa was overturning a policy of administering oudying territories which
had worked acceptably for three centuries. His biggest political miscalculation
was his attack on the palace of Wamara, head of the Cwezi cult in Bwera.
Wamara presumably opposed the second Babito dynasty because it had violated
the traditional rules of succession. Isansa decided on military intervention. So
great was the sacrilege that the trees 'bled' and Wamara cursed Isansa,
prophesying that the Bugunda would swallow Bunyoro. The prophecy came
true; not only did Buganda take possession of a large part of Bunyoro but the
small southern states could no longer remain safe. They became ideal prey for
the imperial ambitions of Buganda, Nkore and Mpororo.
The rise of Buganda

While the Bunyoro kings were busy dismanding a political system which had
ensured their domination for three centuries, the Buganda monarchs were
correcting many of the political evils which had paralysed Buganda ever since
its foundation. The first arose from the growing strength of the monarchy and
its bureaucracy pitted against the clan heads or Bataka leaden. The second was
linked to the multiplicity of ambitious royal princes looking for an opportunity
to manipulate clan political factions in order to seize the throne. The third
problem concerned the bloody succession disputes which followed the deaths
of most kings. These disputes were even more damaging in Buganda than in
Kitara. In Kitara the struggle for the throne could be fierce but once a king was
chosen and crowned he was rarely assassinated or overthrown, whereas in
Buganda a monarch could be murdered or dethroned at any point in his reign.
Buganda was more favourably situated climatically and geographically than
the other major Great Lakes states. There is no mention of drought in its
lengthy and detailed traditions. It had a secure and diverse agricultural base
which freed its male population for war and politics. Furthermore, being
located on the banks of Lake Nyanza, it conducted waterborne trade. It
produced cloth from bark fibre which was greatly prized by neighbouring
peoples. In its expansion, Buganda secured control of iron-ore deposits, a
resource it originally lacked. With the possible exception ofthe southern Basoga
states, no other state in the Great Lakes region possessed such a favourable
geographical and economic framework.
But unlike the other Great Lakes states, Buganda had no royal clan. Each
prince belonged to his mother's clan, whereas the rest ofthe population followed
patrilineal rules. Any clan therefore had a chance to provide the monarch. This
system gave all the people a feeling of belonging to the monarchy but, equally,
encouraged each clan to give a wife to the kabaka, which, of coune, led to a
proliferation of potential royal hein.
In the sixteenth and seventeenth centuries, Buganda was merely a tiny state,
like so many others in Busoga, which tolerated fraternal succession and relied
upon royal administraton. It was unstable and wrecked by interminable civil
wan. However, Baganda nationalism was strong enough to prevent secessions
and a fragmentation ofthe state.
The kabaka Tebandeke (c. 1 644-74) strengthened royal power by a successful
attack upon the ritual religious officers, whose extortionist practices lay heavy
upon the monarchy and the people. The fact that he had succeeded where
Isansa had failed has to do with the different religious organization in the two
kingdoms. In Kitara the Cwezi cult was a centralized institution whereas in
Buganda each ritual officer was apparently independent ofthe next.
Buganda's territorial expansion is associated with three outstanding kings of
the eighteenth century. Mawanda (c. 1674 seized Singo, invaded
Kyaggwe and took over Bulamogi from Bunyoro. Given the immense gain
which this conquest represented, Mawanda wisely abandoned the old system of
indirect rule and appointed his favourites to administer the new territories.
Mawanda was not only a military genius but also a statesman of imagination
and is righdy credited as the father of the modern Kiganda system of
government. Both Junju and Kamanya further extended the limits of the
kingdom between 1734 and 1794. Junju annexed Buddu and forced Kooki
into a tributary relationship. Kamanya seized Buwekula from Bunyoro. By
1800 Buganda had conclusively established its supremacy over Bunyoro.
Semakokiro, in the mid-eighteenth century, turned his attention to the
problem of the royal princes. He began the practice of executing unsuccessful
rivals, even, if necessary, royal sons. He purged princes from the administration
to prevent them from using their positions to plot against the throne. This gave
immense power to the king and paved the way for the absolute monarchy
which emerged in the nineteenth century. Given the turbulence of the early
political years of Buganda, it is tempting to overestimate the great kabaka ofthe
eighteenth century, who acted as centralizen. But it is well to recall that
despotism could only be achieved at the expense of lengthy rebellions, severe
opposition to the kabaka, considerable violence and growing numben of exiles
to neighbouring states.
For centuries, Buganda had existed in the shadow of Bunyoro imperialism,
but, at the end of the eighteenth century, it was ready to challenge this.
Buganda had developed an intense nationalism which allowed its kings to
gather power and develop an efficient administrative and military machine. By

1800 it had a highly competitive society which was more upwardly mobile and
secular than any other in the Great Lakes region.
The agro-pastoral kingdoms ofthe South

The term 'interlacustrine culture', employed for almost a century by Africanist
ethnologists, was based on observations made in the southern half of the region,
in particular in the Rwandan monarchy which has been singled out as an ideal
example by several authon. The formation of this cultural complex has been
generally dated to a relatively recent period from the sixteenth to the eighteenth
century. The work done by historians over the last 20 yean or so has de¬
mystified this image and set in their true historical dimensions peoples with a
past deeply rooted in an ancient Iron Age, going back at least to the beginning
ofthe Christian era. The sixteenth century is a turning point in their history.
A geo-cultural area
The region's natural features and cultural heritage give it a physical and human
landscape full of contrasts. In the west, a gigantic fault line scarp, over 2,000
metres high, the Kabira range, extended north by the Virunga volcanoes,
dominates the Lake Tanganyika Depression and the Lake Kivu basin. To the
east, the high hills of Burundi and Rwanda gradually slope down towards the
terraced plateaux leading down to the shores of Lake Victoria. At over 1,000
kilometres from the Indian Ocean, which gives the region its climate, this
varied relief gives rise to very marked variations in rainfall. The highlands of
the west and the shores of Lake Victoria have an annual rainfall of more than
1,500 mm, while the narrow valley of the Kagera receives less than 1,000
milHmetres.
The pattern of language distribution was similarly rather modey. The regions
concerned were divided into two zones in which Bantu languages were
spoken. People were able to undentand each other without much difficulty,
from the Virunga volcanoes to lower Malagarazi and from Lake Lutanzige
(formerly Lake Albert) to the south of Lake Victoria. In the former area, kings
were known as mwami and in the latter mukama or mugabe. These long-
established geographical and cultural divides were, however, to become less
clear-cut from the sixteenth century onwards, as the kingdoms took shape.
The traces left by Ruhinda and the Bachwezi

The states ofthe south may have grown out ofthe disintegration of an ancient
monarchy, founded between the sixteenth and eighteenth centuries by a
conqueror called Ruhinda; this, at least, was the prevailing belief until the
1950s. The 'empire of Ruhinda' was considered the southern extension ofthe
'empire of Bachwezi', brought about by a new wave of expansion by Bahima
herdsmen. It is now thought that the event, ifit actually took place, occurred
instead at an earlier period, about the fifteenth century. As for the 'Ruhinda
legend', different traditions have handed down widely diverging venions. In

Bunyoro and Nkore, the hero Ruhinda is described as the illegitimate son of
Wamara, the last Cwezi 'king', and a woman servant, Njunaki, while in Haya
and Zinza country, his father is generally said to have been Igaba or Bugaba,
one of the local names for the supreme deity. The Hinda dynasties of Nkore,
Karagwe, Kyamutwara, Ihaniro and Buzinza took as their respective founders a
brother, a son or a grandson of Ruhinda, who is said to have given each of
them a country and a drum.
All the evidence suggests that the reference to Ruhinda reflected the
meeting of two waves of traditions, connected with two major clan groups,
Bahinda and Bayango, which had adopted the same taboo, that ofthe monkey
nkende. As for the founding ofthe Hinda dynasties, the traditions relate that this
occurred only one or two generations after that of the legendary conqueror.
Probably the most interesting historical fact is the continued existence of the
Ruhinda tradition, a source of political legitimacy, well after the sixteenth
century. Ruhinda was invoked in the eighteenth century in Kyamutwara
against usurpen from the Hima clan of the Bankangos; his name was associated
in the nineteenth century with the memory of past greatness (in Karagwe) and
of shattered unity (in Buzinza). It was also used in the nineteenth century to
rally the Bunyoro and Haya people against the imperialism of the Baganda.
The traditions concerning Ruhinda drew their strength from the existing
links between this legendary figure and the exploits of the Bachwezis.
Whatever may have been the reality, the extent and the duration of the 'Kitara
empire', as a socio-religious model it very clearly had its origins in the plateaux
of western Uganda. The south, however, developed its own particular
mythology, where the principal roles were played by Wamara, Mugasha and
Ryangombe. These deities are merely the most important in a pantheon of 30
or more gods, continually increased according to circumstances. Each god is
associated with an area of activity: Wamara with the dead, Mugasha with water,
rain and lakes, hence with fishing and agriculture, Irungu with the bush, hence
with hunting and travel, Kagoro with lightning and Ryangombe with cattle
and hunting.
While Wamara, whose great sanctuary was in Masaka in Bwera (Uganda), is
a recurrent name in the traditions of Kitara, Nkore, and Karagwe, the influence
of Mugasha seems to have spread out from the Sese Islands, on Lake Victoria.
Accounts of Ryangombe are to be found mostly in the west from Rwanda to
Lake Tanganyika. Far from being merely monarchical institutions, these
religions appear to have very ancient origins which predate the founding of the
modern dynasties.
The Cwezi religion (or religion of the Imandwa) was independent in terms
both of wonhip and mythology. In the east (in Haya and Zinza country) the
same families maintained hereditary functions as mediums and were responsible
for the shrines. In the west (Rwanda, Burundi, Buha) initiation to the
kubandwa was widespread. Symbolism and vocabulary seemed to associate the
religion with royalty in Haya country and Nkore, while the acts of wonhip
Ue Great Lakes region, 1500-1800 397
performed in the shade of the erythrina, the sacred tree of Ryangombe and
Kiranga, seemed to be of a more popular nature. But in all cases, these were
practices of initiatory divination and healing, offering protection against threats
coming from near ancestors or abuses of power.
'Clans' or 'castes'?
Faced with a written tradition which gave great importance to the theory of
'caste' and indeed 'racial' opposition between 'Hima and Tutsi lords' and 'Iru
and Hutu serfs', recent historiography has devoted more attention to the 'clan'
structure. Despite its univenal presence in oral culture, the clan does not have
the organic simplicity of a group of kinsfolk, even ifit is sometimes experienced
as such. Lineages, in the strict sense of the term, are classified in units
characterized by a collective name, by one or two taboos, and sometimes by
traditions relating to their origins or by the protection of a god of the Cwezi
pantheon, but they have no territorial unity. Some clans appear in the long
history of several of the kingdoms of the Great Lakes region, examples being
the Bayangos of Nkore in Buzinza, the Bakimbiri of Nkore in South Buha and
the Basita of Bunyoro in Bukerebe. Taboos sometimes concern even larger
groups: the toad is respected in Rwanda by the Bega, the Bakono and the Bahu
(three large clans which provided the queen mothen). Amongst instances of
the spread ofthe nkende taboo, the most surprising example is that ofthe Kiziba
where the taboo is respected by both the dynastic clan of the Babito and the
Bahinda.
The historical background of the clans is clear in Rwanda. The structure

here is exceptionally limited to 18 great clans which all comprise Bahutu,
Batutsi and Batwa memben. Elsewhere, in Burundi, Buha, Karagwe and the
lands bordering on Lake Victoria, there were hundreds of clans but each had
different forms of association, either exogamic units, or associations based on a
joking relationship or the taking of an oath. Identity was often closely linked
with the exercise of political or religious duties of greater or lesser standing.
The fact of belonging to a given clan defined the individual's social status. The
oldest traditions bear witness to the large number of local functions for which
one or other of these clans was responsible, without this necessarily implying
that each of these principalities comprised a homogenous group.
In Rwanda, there are many pre-Nyiginya political entities and several of
these lineages have kept ritual functions in modern Rwanda. In Burundi, the
custody ofthe drums and the religious duties retained by certain clans probably
reflect ancient ruling powers. All of these clan principalities were gradually
absorbed between the fifteenth and the seventeenth centuries (sometimes later)
by the new dynasties, as we shall see. Despite these regional changes and
influences, small sub-regional entities retained their individuality right until the
twentieth century.
The distinction between pastoralists and farmen is not so clearly obvious in
this context as to lend credence to the prevailing hypothesis of invasions and
conflicts. It is not only in Rwanda that clans are of mixed composition; this is
« KITARA to Sese Is.

NKORE \ l)IIAHM S j
MPORORO \ . \ / \ /A ;
'^ -ISINGOROy \f t\ *'
WA ,\ \ 1 ~sA,\9^\ JäUGABO
/ \. BUGARA ^^-^&^yy
X KYAMUTWARA
W>- I KARAGWEJ KYANJAJ I IS VA M IA I
buHOMA. / \f
BUHOMA
l MUBARlY, ,*J (
BÙSHIRU
^BUSHIRU^- '^ AvJ, lit
/ %\-
CYINGOGOI I
.BAHAVU \ "/
/( ^--" J
t'/w I _J /NDUGA,
25
- Kingdom boundaries -f
80 km
Laier regional devisions (1700-1650)
- Main thrusl of expansion (17th-18th century)
Initial political centres (15th- 16th century)
26.6 The southern Great lakes region at the end ofthe seventeenth century (after
J. P. Chretien)
Ue Great Lakes region, 1500-1800 399
also the case for 20 per cent ofthe clans in Haya country, over 10 per cent of
Bahutu, over 50 per cent of Batutsi and 90 per cent of Batwa clans in Burundi.
Other clans are in an intermediate or indeterminate situation, described as bairu
(ennobled) or bahutu (from a good family). The Basita are in some cases
described as pastoralists (in Rwanda for instance), and in others as farming-
blacksmiths.
While the hypotheses concerning a meeting in the distant past between

Bantu-speaking groups and groups speaking south Kushitic or central Sudanese
languages are interesting, they refer to too remote a period (the first
millennium) to throw light on the situation in the sixteenth century, given the
cultural fusion of these populations. Ethnic differentiations occurred along
regional, clan or political lines which reflected the Hima/Tutsi and Iru/Hutu
categories, although it would be risky to attempt to define exactly what these
categories represented four centuries ago. One single fact seems to stand out
the extent of pastoral activities of the middle plateaux from Nkore to Buha,
passing through Karagwe, Gisaka and Bugesera. The 'Hamitic' conquest, so
often mentioned in connection with this period, appears to be no more than
another way of presenting the regional and politico-economic pattern of the
relations between the predominantiy pastoral and predominantly agricultural
secton between the sixteenth and eighteenth centuries.
The formation ofthe modern kingdoms: a geo-political study
Problems of chronology
The wealth of oral sources contributes to the complex problem of reconstruc¬
ting the chronology of the region, since dynastic lists and the genealogies of
princes present many variants, especially before the seventeenth century. Two
solar eclipses often evoked in the traditions provide absolute references but
their identification has been questioned. The 'invasions' organized towards the
south by the Bito sovereigns of Bunyoro also appear as a binding influence
between the different kingdoms, but the traditions compiled in each of them
ascribe the event to different periods. In fact, the Bunyoro must have mounted
numerous cattle raids. None the less, three large expeditionary waves can be
distinguished.
(1) In the first half of the sixteenth century, after the victory in 1520 in
Nkore, the Banyoro are said to have twice invaded Rwanda, under the
leadenhip of Prince Cwa, the son of Nyabongo, Olimi's successor.
(2) During the seventeenth century, the small Bito kingdom of Kiziba seems
to have been subject to raids by its protector Bunyoro, especially during
the reign of Magembe Kitonkire. Karagwe was also affected during this
same period.
(3) During the first half of the eighteenth century the most famous raid
occurred, which is attributed to a mukama known as either Cwa or
Kyebambe. After having for years devastated all the Haya lands, this
sovereign is said to have been defeated and killed by the king Nyarubamba
Kicumbu in Ihangiro. At the same time another Banyoro group was
defeated by a mugabe from Nkore, known as Kitabanyoro (Banyoro
killer). The kingdoms ofthe south were thus finally rid ofthe activities of
the Babito, whose influence was undoubtedly significant especially in the
military organization of the kingdoms.
Tlie Kagera plateau states

This region of grassy plateau has always represented an important geo-political
axis offering an easy route from the valley of the Katonga to that of the Mala-
garazi. Unfortunately, its political history is the wont recorded in the region,
because of the difficulties which have beset its different kingdoms since the
nineteenth century, resulting in the rapid erosion of their traditions. Only the
northern part, made up of Mpororo, Nkore and Karagwe, is relatively well
known.
Despite its prestigious origins (the Sita dynasty), Karagwe began to assert itself
chiefly from the seventeenth century onwards, taking advantage ofthe fint wave
of Nyoro influence. Little is known about the Ntare and the Ruhinda who
came after them, apart from the fact that they were more powerful than their
neighboun to the east. They maintained better relationships with Gisaka and
Rwanda, even including marriage alliances. The crisis in the eighteenth century,
triggered by the Nyoro invasion, led to a fresh, rapid expansion, embodied in
the penon of Ntare Kitabanyoro, and later consolidated by the establishment of
trade relations with Buganda and the Nyamwezi lands.
Mpororo, or Ndorma, inherited a rich tradition of government by clans.
The Bakimbiri are said to have been governed by Ryangombe himself, and the
Baishekatwa by Queen Kitami. Overall government was taken on by a Hima
dynasty from the Bashambo clan who took advantage ofthe vacuum created by
the Banyoro defeat in Rwanda. The Bashambo contracted marriage alliances
with the Bahinda of Nkore, who were then less powerful than the Bashambo.
In the mid-seventeenth century, King Gahaya, son of Ishemurari, attacked
Gisaka and worried Rwanda. The culminating point was reached at the
beginning of the eighteenth century, but, 50 yean later, when Gahaya
Rutindangyesi died, the kingdom split into two rival principalities and the
Murorwa drum was hidden near Lake Bwinyoni in the western mountains.
The Gisaka dynasty claimed to have two different clan origins; the Bagesera
and the nkende totemic group. At the end of the fifteenth century, King
Kimenyi threatened to annex the tiny territory of the Rwandan mwami,
Ruganzu Bwimba. In the mid-sixteenth century, Kimenyi Shumbusho took
advantage of the Nyoro attack to occupy the heart of old Rwanda, Buganza
and Bwanacyambwe. Rwanda was only to regain this lost territory a century
later.
Bugesera could also be seen as a precunor of the political powers which

were later to establish themselves in the western mountains. The Bahondogo
dynasty was closely linked to that of the Banyiginya of early Rwanda and
ne Great Lakes region, 1500-1800 401
marriages and military alliances marked the history of the two countries from
the sixteenth to the eighteenth centuries. Until the end of the seventeenth
century, Bugesera covered a very large territory extending from the Kanyaru to
the Ruvubu riven (the entire northern third of present-day Burundi), and with
its watering-places and rituals, represented a typical pastoral habitat. Difficulties
arose with the growth ofthe Baganwa dynasty in Burundi which tended to side
with Gisaka. The nineteenth century began with the division ofthe country.
Further to the south, several kingdoms were established in the Malagarazi
basin. From the sixteenth century, a 'north Buha' is said to have consisted of
Buyungu, Muhambwe, Ruguru, Buyogoma and even Bushubi. The Bahumbi
dynasty remained very powerful until the eighteenth century. Early on,
however, to the south ofthe river, Ruguru became the centre of another state,
made up of Heru and Bushingo and governed by the Bakimbiri dynasty,
which, with the principality of the Bajiji in Nkarinzi, turned their attention
towards the mountains overlooking the eastern shore of Lake Tanganyika. It
was only in the mid-nineteenth century that Buha began to break up.
Tlie states bordering Lake Victoria

These countries, which were more populated, extended over grassy highlands,
fertile humid plains and valleys with waters teeming with fish. This geographic
divenity left its mark on the communities and the states and accounted in part
for the splitting up of Kyamutwara and Buzinza in the nineteenth century.
Kiziba, wedged in between Lake Victoria and River Kagera, was torn
between its traditional links with Kitara-Bunyoro and the Sese Islands. Further¬
more, it was constantly in conflict with its neighboun in the south-west,
Kyamutwara and Karagwe, which turned conflicts of succession to their
advantage. The Nyoro threat was followed at the end ofthe eighteenth century
by attacks by the Baganda.
In Kyamutwara, Hinda supremacy was apparendy established once and for
all in the seventeenth century by King Karenara. His insane cruelty is said to
have led him to have his son Mukanbya blinded. A new dynasty was founded
by Rugomora Mahe who was regarded as a civilizing hero. He and his
descendants exploited the weaknesses of Kiziba and Ihangiro, until at least the
end ofthe eighteenth century when Kyamutwara was, in its turn, weakened by
internal dissension. Moreover, two Hima lineages of the Bankango clan,
originating from Buzinza in the reign of Kahigi Kasita, gained increasing
military and political influence in the coune of the eighteenth century. At the
beginning of the nineteenth century they established an independent govern¬
ment in Bugabo and a small-scale 'Kyamutwara', implanted on the shores of
Lake Victoria where in 1890 the Germans were to found Bukoba.
Ihangiro can be clearly divided into the flat lake region and an inland plateau
called Mugongo. Succession conflicts at the end of the seventeenth century
gave the kings of Kyamutwara, Rugomora and Kahigi the opportunity for
intervention in the plateau region. Justification for this intervention could be
provided by the dynastic relationship. At the close of the eighteenth century,
Nyarubamba II appealed to the Baganda to help him recover the plateau region
from a rebel prince. From then on, all along the western shore of Lake Victoria
there developed what could be termed Ganda imperialism.
With Buzinza, we take on another cultural and ethnic region where the
influence of the Baha, the Basukuma and even the Tatoga Nilotes inter¬
mingled. It was also the main home of the Barongo blacksmiths. After the
reigns surrounded by the legend of Ntare Muganganzara and the first
Kabambo, there was a kind of second founding under Kabambo Kinwa. In the
eighteenth century the mugabe Kakaraza had to fight off attacks from the Baha
in the west and the Tatoga in the east. His son, Mwihahabi, was the last
sovereign of a unified Buzinza: the war of succession led to the separation of
Rusubi under Ntare Muhire and of all the coastal regions under Ruhinda
Mugangakyaro at the start ofthe nineteenth century.
The history of Bukerebe is one of complex regional influences. From the
cultural standpoint, this country, made up of an island and a peninsula, forms
part of the Zinza group. In the nineteenth century, however, the Silanga
dynasty claimed to be of Ruhindan (even Bunyoro) origin, whereas its founder,
Katobaha, is said to have come from Ihangiro in the seventeenth century. The
kings (bakama) gradually gained the upper hand in the eighteenth century over
the Sita and Kula clans, but the Tatoga menace forced them to establish their
capital in the island. The important trade in ivory was considerably developed
by this principality at the start of the nineteenth century.
The states in the western mountains
Whereas in the east, the dynamics of kingdoms led to fragmentation, in the

west we see instead a concentration to the advantage of two powers Rwanda
and Burundi each of which counted a population of over a million by the
end ofthe nineteenth century.
The tiny Rwandese principality which came into being in the fourteenth
century south of Lake Mohazi (in Buganza) succeeded in firmly establishing
itself along the River Nyabarongo, thanks to the long reign of Cyirima Rugwe
who annexed the lands of the Bongera. But the serious crises of the sixteenth
century were to affect the stability of this kingdom. The two Nyoro invasions
which made King Kigeri Mukobanya and King Mibambwe Mutabazi flee
westwards, culminated in the shifting of the centre of political power to
Nduga, while the original territory was captured by Gisaka. After the warlike
reign of Yuhi Gahima, the succession conflict between his sons was exploited
by the Bashi, the Bahavu and the Bugara people in the west. The king was
killed, the drum ofthe Rwoga dynasty was seized. Despite the tradition which
tends to present an unbroken continuity from the country's origins, it is very
likely that Ruganzu Ndori founded a new dynasty of the Kalinga drum at the
beginning of the seventeenth century. He reoccupied Nduga, and from there,
he and his successon extended the kingdom as far as Lake Kivu and the
Kanyaru river, recapturing Bwanacyambwe in the east at the same time. The
growth of a new dynasty in Burundi held up expansion at the end of the
ne Great Lakes region, 1500-1800 403
century, but it took off again in the eighteenth century as a result of the
exceptional military organization set up by Cyirima Rujugira and Kigeri
Ndabarasa. Burundi and Gisaka were forced to retreat, Ndorwa broke up,
Mubari was occupied and setdements on the shores of Lake Kivu were
extended up to Kinyaga. At the end ofthe eighteenth century, Bugesera, an old
ally, was divided up between Rwanda and Burundi. This expansion continued
in the nineteenth century, especially in the north-west and south-east, but the
old Hutu and Tutsi powers preserved their independence for a long time under
a kind of protectorate ofthe Banyiginya.
There is far less information concerning the history of Burundi during this
period. At the start of the sixteenth century an initial dynasty created by Ntare
Karemera seems to have been established in the mountains of the north-west,
on the border of the powerful Bugesera. Then, at the end of the seventeenth
century, Ntare Rushatsi or Rufuku founded the Baganwa dynasty in Nkoma
which unified the south, the centre and the north (the fusion of clan powen of
the old kingdom) and rose up against Bugesera, whose king, Nsoro Nyabarega,
had to flee to Rwanda. It was not until the long reign of Ntare Rugamba,
during the fint half of the nineteenth century, that the kingdom extended west
to reach the Rusizi river and east as far as the Malagarazi basin.
Further west, around Lake Kivu, the Bashi on the one hand, and the
Bafuriru and the Bahavu on the other, refer to a common origin in the Nyindu
region (in Lwindi). Traditions also testify to the very long history of relations
between these small kingdoms and Burundi and Rwanda.
States, economies and societies
This history of politics and warfare should not overlook the question of
population movements, the evolution of landscapes and crop production, and
institutional and even ideological changes.
Development ofthe relationship between herding andfarming

Geographical factors determined from the outset two areas, each with their
own activity: pastoralism on the Kagera plateaus, and agriculture on the shores
of Lake Victoria and the western mountains. During the period considered here,
however, both activities were increasingly carried out in conjunction through¬
out the region. In fact, they had never been in opposition to one another: the
association of the earliest cereals, eleusine and sorghum, with cattle-raising,
would appear to be a very ancient practice, both in terms of production and
consumption (meal and milk). Ancient Karagwe is famous, according to tradi¬
tion, not only for its cattle but also for its farming and its sorghum beer. Early
accounts mention the use of manure, whose importance in intensive farming
with two harvests a year in the wettest regions, has been sadly overlooked.
But towards the west in particular, factors such as droughts and famines
probably led to the increasing influence of the herdsmen. Such calamities
multiplied in the first half of the seventeenth century and the second half of the
eighteenth century, forcing farmen to seek the help of the herdsmen, who,
through the practice of transhumance, had been able to ensure the survival of
their cattle. In the seventeenth century the shifting of the centres of political
power from the area of the Kagera and Malagarazi rivers towards the wooded
heights of Kiriba overlooking Lakes Kivu and Tanganyika, cannot be explained
simply by expeditions or dynastic changes, but was also due to the growth of a
system of tribute which was favourable to pastoral groups.
In Rwanda or Burundi, however, it can be seen that those who benefited
from the regimes established 300 yean ago represented only a part of the herd-
owners, the rich Batutsi and the ruling circles linked to royalty. They were, in
fact, those who had succeeded, through the actual use or symbolic use of the
cow, in acquiring political control over agricultural production. Agriculture
was of little interest to ethnologists and yet its importance is evident in the
rituals and even the ideology of royalty. In Burundi, for example, the annual
muhanuro festival, during which royal authority was renewed, celebrated the
sowing of sorghum, and determined the most favourable date for this, in a
country with a long rainy season. Furthermore, with regard to food crops, the
introduction of plants of American origin (sweet potatoes, maize and beans)
offered farmen new possibilities of expansion, providing them with the
possibility of two harvests a year and supplying plant protein.
The relationship between tillen and herders, then, does not have the
immutable and univenal character that the socio-biological stereotypes would
lead us to believe. The economic, political and territorial changes which took
place between the seventeenth and the nineteenth centuries brought about a
development of Hima-Iru or Tutsi-Hutu relations from one of local trade to a
more comprehensive hierarchical relationship, whose flexibility depended on
the states involved or the particular moment in time.
Consolidation of monarchical powers in the seventeenth and eighteenth centuries

Before colonization, each state had a system of tribute which varied according
to the ecological situation, the balance between the different forces of produc¬
tion, the clan patterns and the institutional forms in existence. Everywhere, the
sovereign directly controlled agricultural areas close to his residences and
assigned chiefs elsewhere. They were often princes of royal blood, assisted by
delegates who were usually from the most influential local lineages. Tribute was
paid to these courts either in the form of labour or in kind. This allowed the
ruling aristocracy to extend its influence by redistribution, as luxury was scant.
Royal power was consolidated, particularly from the eighteenth century on¬
wards, in the following four ways.
Exploitation of clientship The relationships which were known as ubugabire or

ubuhaka were increasingly removed from their private context and used for
political purposes: protection was guaranteed to a family in exchange for
increased obligations. In Rwanda, the buhake was more especially used to
subjugate influential Hutu lineages in the peripheral regions conquered by the
Banyiginya, particularly in the last years ofthe eighteenth century. It was at this
same time that the gikingi lurid system became established, which granted
exclusive grazing rights to the most important herd-owners, with
administrative rights over families living in these areas. This network of
privileges, sometimes described as feudal, was accompanied in Rwanda by a
remarkable increase in the size of herds resulting from the conquests of the
eighteenth century.
Military organization Rwanda developed, especially from the time ofthe reign of
Cyirima Rujugira, a system of hereditary standing armies, consisting of young
men of certain lineages stationed in camps located on borders under threat.
This militia also looked after herds owned or protected by the king. In the
nineteenth century, the role of this institution was mainly reduced to tax-
gathering in the service of important chiefs. The importance of the military
factor was equally marked in the other kingdoms at the end ofthe period.
Trade possibilities Until a recent date, institutionalized trade and local barter
played a role which was more important in commerce than in specialized
trading. None the less, regional products such as salt, articles made of iron, or
raffia bracelets were the staple wares of the earliest pedlars. Articles from the
coast of the Indian Ocean, including glass beads or ornamental shells, would
appear to have spread through the region from the seventeenth century.
Copper is thought to have been around from the eighteenth century onwards
in Burundi, Karagwe and Buganda. But it was only in the nineteenth century
that rulers in Rusubi, Karagwe and Rwanda made attempts, as they had in
Buganda, to control this trade in luxury goods.
Ideological control The changes brought about in society as a result of the agrarian
crises, the wars of conquest ofthe seventeenth and eighteenth centuries and the
movements of population, furthered the break-up of lineages. The success of
the Cwezi religion in the seventeenth century is understandable for it offered a
kind of refuge in the face of this instability. To a great extent the legitimacy of
the new monarchies was rooted in this religious movement, as testified by
myth and ritual. Initiates were, however, also capable of triggering subversive
movements among the people or blocking royal action.
Ultimately the monarchies set about controlling this double-edged religion
and turning it to their advantage. In Kiziba, the cult ofthe dead kings counter¬
balanced the power of Cwezi spirits; in Kyamutwara, the monarchy, from the
time of Rugomora Mahe onwards, was based on the cult of Mugasha; in
Rwanda, a 'king ofthe Imandwa' was installed at the court from the reign of
Cyirima Rujugira. During the same period the Rwandese monarchy had an
official body of panegyric poetry and historical narratives compiled, which was
circulated by means of the army.
Rwandese oral sources reflect the exceptional nature of the country's
political centralization. In other states, the different strata of society retained a
greater degree of independence right up to the eve of colonization.
Conclusion
By the end of the eighteenth century the modern ethnographic and linguistic
configurations in the Great Lakes region had taken shape. The major
population movements had mosdy come to an end and the last ethnic groups
to be formed in the region (the Bakiga, the Iteso and the Lango Omiro) were
being defined, the process being completed by about 1830. Also, with very few
exceptions, the era of state formation came to an end with Rwanda emerging
as the most powerful state in the region. In the central area, Bunyoro con¬
tinued to decline until the secession of Toro and Paluoland in 1830. By the
beginning of the nineteenth century, Buganda had emerged as the pre-eminent
state in this area. Henceforth, the main concerns ofthe historian shift from the
growth of central power to the efforts of the chiefs to limit the power of the
monarch. In the coune ofthe nineteenth century a multiplicity of new themes
was to dominate Great Lakes history.
The interior of East Africa:
the peoples of Kenya
and Tanzania, 1500
The period stretching from 1500 to 1800 saw the emergence of societies and
social and economic systems that are still characteristic ofthe interiors of Kenya
and Tanzania. The diversity of experience is perhaps the key feature of the
history of this region at this time. Centre stage was taken up by the Maasai,
Chagga, Pare, Shambala, Gogo and Hehe. On the eastern edge lived the
Kikuyu, Kamba, Miji-Kenda, Zigua and Zaramo. To the north-west were the
Abaluyia, Kalenjin, Luo, Abagusii and Abakuria, whilst in the south-west the
Tanzanian communities such as the Sukuma, Iramba, Nyamwezi, Zinza and
Kimbu were to be found.
All these peoples, with the exception of the coastal societies, were still
isolated from the coast. There is no documentary evidence of Arab or Swahili
penetration of the interior before 1700 and no significant collection of
imported objects has yet been found at any interior site north of the Zambezi
dating to the period before 1600. From the mid-1 600s, however, it is possible
to begin to see the emergence of chieftaincies and structured political organiza¬
tions, as well as changes tending towards a general tributary mode of produc¬
tion. Oral traditions portray this development as one of conquest and
assimilation by the migrant populations who were more powerful than the
local communities. However, it can also be reasonably supposed that the local
communities gradually subdued and stabilized the hitherto nomadic or
unsettled populations who infiltrated them.
All history is transition from one stage to another. During the centuries
following 1500, the distinct ethnic groups that we know today, with their own
peculiar linguistic and cultural characteristics, were evolving within the
societies of the interior of Kenya and Tanzania. The predominant economic
activity was, by far, agriculture. In all the settled agricultural communities,
people endeavoured to find techniques for adapting to their local environment
and dealing with it in a rational way. In some areas, advanced methods, such as
terracing, crop rotation, green manuring, mixed farming and farming of
drained swamp-land were used.
While most East Africans were agriculturalists, the Maasai, Pokot and
Turkana were largely herdsmen who drove their cattle to pasture across the
407
125 250 miles
I H
SUDAN 200 400 km
ETHIOPIA
V
V-
"V
''S,.
'"-,---% _,.--
\S L Turkana
J SOMALI /
BLU0L0 BORAN
C, a \jURKANA^
LAtorSL^ UGANDA j
SAMBURU
/ KARAPOKOT KENYA
! TORO
''luyia
MARAKWET
! BAGANDA
I * . LUO * mi Kenya SOMALI E9ualor
i
NANDI KIKUYU
*-UU KIPSIGIS
GUSII \
KAMBA
SOMALI \
OROMO Pal<^
\ MAASAI
Lamu
'N..
*N». SWAHILI s^S M
SUKUMA CHAGGA N taita Malindi
/ Gedi .,
N. TAVETA
MAASAI
SUMBWA IRAMBA
MRE MÙI-KENDAW
HA
NYATURA
SHAMBALA
LUGULU
Tanga Pemba I.
NYAMWEZI
ZIGUA
TONGWE GOG°
KIMBU
\L Tanganyika
BENDE TANZANIA
Dar es Salaam
PIU8WE
ZARAMO
FIPA LUGURU
UHEHE
\. Kilwa
BENA
Kisiwan
ZAMBIA
NGINDO
,L Malawi
K NGUNI , f~
MOZAMBIQUE
27.1 Ethnic groups of Kenya and Tanzania (adaptedfrom map drawn by M. Kivuva, Moi
University, Kenya, after W.R. Ochicng')
Trze interior of East Africa 409
plains of central Kenya and Tanzania. However, the agriculturalists and the
herdsmen at no time sought to pursue an exclusive economic activity. The
agriculturalists, like the Luo and the Abagusii, also kept large herds of livestock,
while catdefolk, such as the Samburu and the Arusha Maasai, also practised
some agriculture. The Baraguyu, the Kalenjin and the Akamba were half and
half.
The Sanye, Okiek, Sandawe and Hazapi remained hunters and gatherers, but
even among these peoples a growing spectrum of subsistence patterns could be
observed. For example, the Okiek traded honey for agricultural products, and
the Dorobi and Athi hunten became involved in the long-distance ivory trade.
During most of the seventeenth and eighteenth centuries, both agriculturalists
and herdsmen quarrelled over the richer and better-watered grasslands and
plateaus. At the same time, they invaded the domains of the hunters and
gatheren, which led to the absorption, decimation or isolation ofthe latter.
Economic evolution
By 1500, the two major economic punuits in the Kenya-Tanzania interior

were agriculture and catde-rearing. Famine represented, therefore, a veritable
distaster. It resulted from poor soils and the unreliable rainfall which affected
both crops and pasture. Not even the most favoured regions were spared from
the scourge of famine. Banana trees might have thrived in Usambara and yet
the traditions speak of a famine at least once every 15 yean. When the rains
failed, people set off into the bush in search of food, or else they exchanged
their cattle for grain from more fortunate groups.
However, the first line of defence against famine was the cultivator's skill.
From 1500 there was constant experimentation with the crops introduced
much earlier by the Bantu and Nilotes. In forested areas with heavy rainfall,
sorghum, eleusine and millet gave way to banana trees and tubers. After 1500, a
number of European and American crops, such as maize, groundnuts, sweet
potatoes and cassava, were introduced by the Portuguese. Various other foods,
including numerous varieties of peas, beans and nuts, were also cultivated and
are still grown today.
The East African cultivator experimented with all available crops and
adapted to his environment. When the Abagusii, who came from the plains
bordering the Winam Gulf, setded in the highlands of south-west Kenya in the
mid-1700s, instant crop failure led to famine which decimated their population.
They were forced to reduce their production of several varieties of sorghum
and other lowland crops and increase production of finger-millet and root
crops that flourished in their new environment. By the end of the eighteenth
century the interior regions of Kenya and Tanzania had many different agri¬
cultural systems. Nevertheless, one common practice over the whole territory
was the burning of the bush to clear the ground for cultivation. When, after a
few seasons, the soil was exhausted, the farmer would move on to allow the
soil time to recover. It appean that all agricultural systems throughout the
Kenya and Tanzania interion relied on the labour of the homesteads equipped
with the same simple tools axes, hoes, pangas (machetes) and digging sticks.
The raising of cattle, poultry, sheep and goats continued to be a vital
economic and cultural activity in the interior of East Africa. Animal manure was
used in agriculture. Livestock provided clothing and food, weapons and utensils.
Among the herdsmen such as the Maasai and Turkana, livestock governed the
daily routine and relations of kinship, as family prosperity and individual security
were measured in terms of ownership. In many primarily agricultural societies,
such as the Kikuyu and Abagusii, the acquisition and ownenhip of catde was an
important indication of wealth and prestige. Among the Turkana, Luo, Kalenjin
and Maasai, it was the custom to freely lend and borrow catde and to use them
for providing bridewealth. In this way, each family herd came to be widely
dispened among friends and relatives often living far away, which was to
benefit both individuals and the society as a whole.
Apart from the wealth brought by cattle, the predominantly pastoral people
were also rich in land. Groups such as the Oromo, Somali and Maasai
expanded their pastoral activity over wide areas of East Africa. Thus, because of
its social status, wealth, its range of territory and political and military power,
the pastoral way of life was both predominant and envied.
Hunting was a complementary activity to agriculture and cattle-raising, both
as a source of food and for protecting crops. Oral accounts, particularly those of
the Luo, Shambala, Pare, and Abakuria, frequently explain the movement of
people by relating how the pursuit of an animal led them to a location where
they decided to setde. The myth of the origin of the Shambala kingdom tells
the story of Mbegha, a Ngulu hunter, who killed wild pigs that were
destroying their crops, and freely distnbuted their meat. 'The Shambala, in
admiration and gratitude, gave him wives and made him king of all Usambara.'
Hunting, therefore, was an important and respectable economic activity.
The importance assigned by tradition to farming, livestock-raising and hunt¬
ing contrasts with the lack of significance given to fishing, apart from amongst
the populations living along the banks of Lakes Turkana, Victoria, Baringo,
Eyasi and on the coast of the Indian Ocean. Various types of fish were caught
by hook and line, basket-nets and fence traps. Dried fish were, and still are, sold
to people far from fishing areas.
The traditions of most of the East African peoples indicate an ancient
knowledge of iron-working. Archaeologists and linguists have also established
that techniques of iron technology were introduced into East Africa by the
Nilo-Saharans and Bantu, probably six centuries before the Christian era. The
earliest Iron Age sites are around Lake Victoria. These sites were characterized
by tall, cylindrical smelting furnaces and a distinctive style of pottery. Iron-work
was usually the exclusive occupation of a few groups and carried much prestige
and, occasionally, wealth. Several Ugweno traditions report that towards the
beginning of the sixteenth century, certain families specialized in iron-smelting
and forging. The major iron-working lineage, the Washana, held political
power in the country. They were later overthrown by the Wasuya who trans-
The interior of East Africa 411
formed what were, hitherto, clan initiation rites into an elaborate state institu¬
tion with unlimited coercive powen. In -western Kenya, the Abagusii claim to
have worked iron since the sixteenth century. Among the Luo, to the north of
the Winam Gulf, the major blacksmiths were the Walowa of Yimbo. They
manufactured hoes, arrowheads, ornaments, needles, axes, spean, knives and
razon. Other important regional activities were salt-mining and crafts such as
drum-making, canoe-building and pottery. Traditionally, these activities were
the guarded specialities of particular clans. Crafts such as basketry and house¬
building, however, were not specialist activities, but were undertaken by any¬
one who had time for them.
To sum up, it must be emphasized that, between 1500 and 1800, food
production was a crucial factor in the survival and expansion of society. As
most of East Africa was put under extensive cultivation and grazing, and as
production rose above the level of bare necessity, people could begin to use
their time for purposes other than subsistence. It became possible to exempt
certain individuals from the task of food production so that they could devote
themselves to other objectives such as the distribution of merchandise, war-
making, state duty, art, religion, medicine or philosophy and the improvement
of technology.
Trade
Until the late 1700s, most societies in the Kenyan and Tanzanian interion
developed independendy of external forces. They were independent politically,
economically and socially.
Trading, the exchange of goods for mutual benefit, is a universal habit found
among even the simplest societies. At no time have all groups and all areas been
endowed with similar means and resources, hence the necessity to procure
essential commodities from one's neighbours. And so, pastoralists and agri¬
culturalists exchanged their products.
Much has been written about the trading connections that, for centuries,
existed between East Africa and the Orient, but it must be emphasized that sig¬
nificant trade relations in East Africa were not limited to trade with the oveneas
markets; they also included important economic links between different internal
regions. Internal African trade had a different purpose and dynamic to that of
Arab or European trade. It was, consequendy, irregular and sporadic. It was also
a means of accumulating wealth in the form of cattle and foodstuffs.
Let us look at several examples of this trade. Traditions from Kikuyuland talk
of ancient trading links between the Kikuyu and the neighbouring Akamba and
Maasai. The Kikuyu economy was divene. While they were predominantly
farmers, certain sections, such as the Kikuyu of Tetu and Mathira in Nyeri,
evolved a semi-pastoral economy and lived almost like their Maasai neighbours.
Convenely, the Athi Kikuyu specialized in hunting and the exploitation of
forest products, such as beeswax and honey. So, the Kikuyu offered foodstuffs
along with manufactured products to the Maasai, who, in turn, paid in live-
stock, milk, skins, leather cloaks and magic. Kikuyuland also provided the best
markets for the Akamba after 1760. From the Kikuyu communities of Muranga
and Nyeri, the Akamba traden sought staple crops such as beans, yams, maize
and ndulu, a green vegetable of which they were particularly fond. The
Akamba paid the Kikuyu in animal skins and uki, a kind of long-fermented
beer. Sometimes, they also traded their labour at harvest-time for an equivalent
share ofthe food crops.
Further west, on the eastern shore of Lake Victoria, there was a widespread
trading network linking the various communities. The main trading partners
were the Abagusii and the Luo. The Abagusii sold to the Luo agricultural
products and iron items such as axes, spean, razon and arrowheads, as well as
soapstone and leopard and baboon skins. In return, the Luo offered mainly
catde, but also catde-salt, skins, ghee, milk, fish, pots and poison.
Further south, in Tanzania, traditions from western Unyamwezi and Uvinza
speak of pre-1800 immigrant groups from the north exchanging grain crops for
pots with the earlier inhabitants of the region. Among the Nyamwezi, forest
products were also traded between villages and were in particular demand in
the northern countries where the Iramba, Sumbwa and Sukuma lived. In
Unyakyusa, most women made pottery, but those in volcanic areas lacked the
necessary clay and so depended on specialists such as the Kisi women of the
Nyasa lakeshore, who bartered their pots in Nyakyusa villages, or the people of
Ngaseni who traded their huge beer pots along the road which wound its way
around the upper slopes of Kilimanjaro.
The two most important items of early trade in central Tanzania were prob¬
ably iron and salt. The main area of iron-working was in the north, among the
Ha and the Zinza. The scarcity of iron stimulated its trade. Throughout Tan¬
zania, iron was used in a great variety of forms: as knife-blades, blades for hoes
and for axes, spear lances and arrowheads for hunting, fishing and fighting. Many
traden from the north and the south travelled to the land of the Ha and the
Zinza to buy these iron items which they would then resell back home. In the
north-east, the Chagga and the Maasai obtained iron products from Pare smelten,
while in the late 1 700s the Mamba chiefdom became the iron-working centre for
much of the Kilimanjaro region. In the south, Fipa blacksmiths exchanged their
produce for cloth woven in the Rukwa valley, while the Nyakyusa bartered food
for the products of the Kinga furnaces. Iron was a scarce and precious
commodity: only the wealthy possessed iron hoes. Those used on Kilimanjaro
towards the end of the eighteenth century were only a few centimetres broad
and the Sandawe used them until they were worn down to the hilt.
Salt was a product of vital necessity to people who lived chiefly on vegetable
foods. The major salt-pans were located in Ugogo and at Ivuna, Kanyene, Lake
Balangida, Singida and Lake Eyasi, as well as at Bukune, Buha and Uvinza. The
most important brine springs were in Buha and Uvinza. The Uvinza springs, in
particular, seem to have been exploited since the eleventh century. This Vinza
salt industry was later boosted by the foundation of the original Vinza chiefdom
in 1800.
ne interior of East Africa 413
Trade between the Miji-Kenda of south-east Kenya and north-eastern

Tanzania only really began to develop by the beginning of the eighteenth
century. This was the period when most Miji-Kenda, who had previously
isolated themselves in fenced-in villages, began to expand outwards. Later, with
their population gready increased, they embarked on secondary migrations
settling in the less fertile, lower parts of their hilly country, where they
frequendy clashed with the Omoro and Maasai.
The Miji-Kenda were largely cultivators of millet, rice and fruits. Through¬
out the eighteenth century they were staunch allies of the Mazrui, supplying
the many coastal setdements with ivory, gum copal, honey, beeswax, tobacco,
grain, foodstuffs and timber for building dhows. In return, the Miji-Kenda
traden obtained salt, beads, cloth, iron hoes and other goods. In the trade that
developed between the coast and the interior, they sent their caravans both
northwards, to the countries ofthe Oromo and Borana, and towards the north¬
west, to the Akamba and Chagga countries. But their control of the interior
trade was short-lived, and by the mid-1700s they had been replaced by the
Akamba.
Trade in the interior regions of Kenya and Tanzania may well, by 1700 or
before, have involved exchanges over considerable distances; however, it was
only from about 1800 that the interior of East Africa began to participate in
long-distance trade and thus to be linked to outside economies.
Social and political activity

Social and political institutions and organizations played an important role in
the cohesion of society and the protection of property and trade. Between 1500
and 1800, society in the East African interior was far from having taken its
definitive shape. There were still significant internal migrations into sparsely
setded or uninhabited areas, sometimes bringing together peoples of different
languages or dialects, and with different political and economic practices. Internal
conflicts multiplied as clans expanded. In some areas, immigrant groups or
families imposed their political leadenhip over established populations. In other
areas, the immigrants were absorbed into the local societies. In both cases internal
migrations set into motion processes of cultural and political integration which
continued into the colonial period.
The period from 1500 to 1800 was, therefore, characterized by a movement
towards political centralization and the evolution of larger and larger linguistic
and social groups. Thus a variety of societies emerged in the East Africa
interior, but the two that contrasted most conspicuously were the pastoralists
and the farmen.
Pastoralists, like hunters, depended on herbivores. They lived a wandering

life, travelling comparatively large distances in search of pasture for their
animals. They followed a more or less regular migrational pattern according to
the season. Pastoralists had to protect their herds from predators, whether animal
or human. Such a life required organized leadenhip.
The political history of the Kenya-Tanzania interior was partly linked to

population growth which was made possible by, on the one hand, the seden¬
tary life of farmers, and on the other, the disciplined politico-military organiza¬
tion required by pastoralism. The balance tipped in favour of one or other of
the protagonists, depending on the fluctuations in social organization and
cohesion, and on the development of technology. By 1800, the pastoralists
were beginning to lose both economic and military power to the farmers who
were rapidly perfecting their political institutions. The pastoral Maasai, in
particular, suffered military and economic decline throughout the nineteenth
century due to cattle disease, epidemics and civil wars.
By 1700, two main types of socio-political organizations existed in the
Kenya-Tanzania interior. On the one hand there were societies such as the
Kikuyu, Miji-Kenda, Kamba and Maasai, who lived in scattered, independent
setdements with no form of centralized, traditional bureaucracy. Decentraliza¬
tion, however, did not mean disorganization or lack of political and social
cohesion. These decentralized societies had family, village and district councils
made up of elders. Members of each family, clan and district were united by
relationships which defined and governed the actions of individuals and
established mutual obligations and rights.
There were also the centralized (or centralizing) societies, such as the
Shambala, Pare, Sukuma, Nyamwezi and Wanga, whose chiefs maintained
rudimentary administrations that provided a socio-political framework. By the
end ofthe eighteenth century, some of these groups, such as the Shambala and
Pare, were ruled by strong despotic kings or supreme chiefs.
Let us look at certain examples of these developments. Among the Pare, the
fint stage of political development was reached when each clan established itself
in its territory and accepted one ritual leader whose descendants could succeed
him. Thus the religious needs of the community were at the heart of their
political evolution. After settling in an area, each group established its sacred
shrine (mpungi) which finked it to the founding anceston and where,
periodically, all members gathered to wonhip. As the population increased, it
became necessary to bring together people of different clans. This formative
period of unification is shrouded in myth, but, according to Isaria Kimambo,
about 'sixteen generations ago', many Ugweno clans recognized as their chief a
Washana iron-smith. The Washana were later overthrown by the Wasuya who
founded a centralized state. This coup d'état was the work of Angovi, but it was
his son, Mranga, who consolidated the Ugweno state, transforming what were
hitherto clan initiation rites into a complex state institution with unlimited
coercive powers. At the peak of its power, the Ugweno state had at its head a
mangi mrwe (supreme chief) who was assisted by a council of ministers and the
wamagi (district chiefs).
Another area of Tanzania where chiefly institutions were strong by 1600 was
Bukoba, where several ruling lineages had evolved among the agricultural
communities. Their political system, which was more centralized than in other
Tanzanian states, involved control of land and cattle, in such a way that the
tribute system was more exploitative. These states belonged to the Great Lakes
region. Further south, many chiefdoms had sprouted, resembling one another
in many ways. Each consisted of a small group of villages and neighbourhoods
ruled by a single chief whom the villagers appointed from the ruling lineage.
He was the holder of special regalia or symbols of power, such as sacred spears.
He maintained the royal fire from which all fires of his chiefdom were
supposed to be kindled.
Andrew Roberts tells us that numerous Nyamwezi chiefdoms are probably
not the result of any single process of migration or diffusion, but originated from
the Nyamwezi environment. 'From an early stage', he writes, 'certain men were
respected as rainmaken and magicians, as pioneen in opening up woodland and
as arbitraton in dispute.' It was from these roots that the Nyamwezi chiefdoms
emerged. But unlike in Upare, where development tended towards centraliza¬
tion, the small Nyamwezi chiefdoms tended to increase in number. Only in the
nineteenth century was there rapid centralization in Unyamwezi due to the
expansion of trade and the emergence of ruthless traders and organisers such as
Mirambo.
Among the Fipa of south-western Tanzania the earliest myths of origin are
tied up with the origin of the Fipa kingdom (Milansi). According to one
version, recorded by Roy Willis, 'the first man in the world, who was called
Ntatakwa, fell from the sky when the world began and founded the Milansi
line of chiefs'. The early settlen of Ufipa were farmen who lived in tightly-
packed villages; this encouraged them to organize collective defence and build
fortifications. Memben of the ruling family seem to have derived their power
from their iron-smithing skills.
Later, in the eighteenth century, the Fipa political regime changed, follow¬
ing an invasion by pastoralists from the north who introduced political
principles no longer based on relationships between 'father chiefs' and 'son
chiefs', but on a penonal bond of loyalty between the sovereign and his
partisans. They were nominated by the king to ruling posts on the periphery of
the kingdom, but had no kinship relation with the king. There was probably a
coup which overthrew the traditional Milansi dynasty and brought the Twa
dynasty to power. The Twa reign was, however, far from being peaceful, as
Ufipa was again invaded towards the end of the eighteenth century, this time
by the Nyiha who burnt down Milansi villages. By 1800, civil war between the
two rivals for the Twa throne was still raging.
A significant number of chiefdoms of varying sizes and centralization appeared
in Tanzania between the sixteenth and the eighteenth centuries, in response to
human, political and economic need. They were shaped by the physical and
human environment. Centralization and economic expansion was the objective
of most Tanzanian states but the process of nation-building often proved difficult.
In western Kenya, the Luo seem to have founded themselves on the kinship
system, ancestral cults and hereditary leadenhip. When they arrived in Nyanza
they created socio-political entities that were more centralized and more strati¬
fied than the earlier communities. And while the Luo clans and lineages were
equal in most respects, the existence of groups of chiefs, priests and commoners
indicated a degree of inequality rarely found in Kenyan societies.
The best analysis of the way the Luo socio-political system worked at this
time is probably that by Peter C. Oloo Aringo. He agrees with B. A. Ogot
that, at the highest level of political organization, the Luo were divided into 12
or 13 ogendini (ethnic sub-groups) of varying sizes. Membenhip ofthe ogenda
was through the gweng', a semi-autonomous political and territorial unit. In
principle, the gweng' was occupied by members ofthe same clan but often parts
were leased to jodak (outside clans).
The highest political assembly in any gweng' was the council of elden. The
gweng' council was the custodian of clan territory. It carried out ritual cere¬
monies and declared war or negociated peace with other gweng' . Its decisions
had force of law on all subjects. Those who infringed its regulations or defied
its decisions could be cursed.
Above the gweng' council was the oganda council which debated major
political, judicial and economic matters, such as murder, catde theft, boundary
disputes, famine, epidemics, invasion, defence, trade and inter-clan and inter-
gweng' conflicts.
The Kalenjin, who have lived for centuries in the western highlands of
Kenya, formed a typically decentralized society. Authority was monopolized by
the elden and influential specialists. Eloquence, the ability to voice acceptable
views at councils, and a sound knowledge of precedence and custom were
considered important qualities for any political leader. But above all, a man had
to distinguish himself as happily married with a good social reputation within
the community. Among the Nandi, a branch ofthe Kalenjin, there is evidence
to suggest that between the eighteenth and nineteenth centuries, a new
dimension was introduced into the governing of society with the adoption of a
spiritual leader called the orkoiyot. H. A. Mwanzi asserts that while the socio¬
political structure of Nandi society remained largely unchanged, with the clan
and neighbourhood councils maintaining their role, by the end of the nine¬
teenth century, the orkoiyot had transformed the Nandi into a theocracy with
himself as sovereign.
In time, the orkoiyot was able to demand a share ofthe spoils of war and to
impose his authority over councils through having penonal representatives as
delegates to each council. Thus a special class of officials was created among the
Nandi to maintain communication between these councils and the orkoiyot.
These officials accompanied the war leaders when the orkoiyot's permission was
sought for a projected raid. They also acted as secret agents for the orkoiyot.
These arrangements gradually centralized Nandi society in the second half of
Conclusion
By 1800, a bewildering number of Nilotic, Kushitic and Bantu-speaking

communities were scattered all over Kenya and Tanzania. Only in the Great
Lakes region of East Africa had large states and kingdoms developed. In the
Kenya-Tanzania interior, the typical socio-political unity was small and clan-
oriented. Most ofthe Bantu-speaken were farmers, though wherever possible,
they kept their own livestock. Herdsmen, like the Turkana, Oromo and
Maasai, lived in more aggressive societies and controlled substantial areas of
territory between the agricultural Bantu lands. Along the coast were the
Swahili city-states, dominated by local Arab oligarchies. The populations of
these coastal cities were mixed, the dominant language being Swahili, and the
tempo of life considerably different from that ofthe interior.
The end of the eighteenth century marked the end of independent East
African growth. Soon after 1800, the independent people of Kenya and
Tanzania were faced with Arab and European invasions. Much of their
traditional civilization was seriously damaged or submerged in a rising flood of
violence.
East Africa was, at this time, badly placed to withstand this assault from
outside forces. Africa as a whole had fallen far behind the enormous powers of
the rest ofthe world in its ability to produce. Whereas Europe, after 1500, had
entered a great era of mechanical and scientific discovery and development, the
African interior had continued with the steady but slow development of its
own civilization. This Iron Age civilization had many achievements. There was
much advancement and invention in the arts of community life, the adoption
of new crops, the spread of metal-working skills, the growth of trade and, more
important, the methods of self-rule and peace-keeping. These were certainly
important gains but they did not match up to the growing power of the strong
industrial nations of Europe. By 1800, the technical capacity of Europeans far
outweighed that of Africans, and by 1900, the gap had become enormous. It is
this imbalance which goes a long way to explain the crisis which began around
1800.
Madagascar
and the islands
ofthe Indian Ocean
Madagascar
The shape of modern Madagascar was fashioned in the course of the nineteenth
century when the Merina people of its heartland established their political
supremacy, thus becoming virtually the only Malagasy in the eyes of outsiders.
Yet, beyond the impact of relatively recent events, there is a different and
farther-ranging past in which the period between 1500 and 1800 stands apart in
the overall history of Madagascar. It is almost certain that most of the
inhabitants of Madagascar came together during these three centuries to form
the social, economic, religious, cultural and political groups that make up the
Malagasy people today: the Antankara, Antandroy, Antambahoaka, Antanosy,
Antemoro, Antesaka, Antefasy, Bara, Betsimisaraka, Betsileo, Bezanozano,
Merina, Mahafaly, Sakalava, Sihanaka, Tanala and Tsimihety.
By the mid-sixteenth century, the arrival of immigrants into the country
ceased but some of the inhabitants of Madagascar had already come into
contact with Europeans, mainly Portuguese, and at least one of the more
influential royal families, the Maroserana, had started to form. At the beginning
of the seventeenth century, Madagascar was a patchwork of small and mainly
self-supporting kingdoms. Before the end of the century much of western
Madagascar regrouped under a Sakalava empire while several kingdoms
emerged in the highlands. In the coune of this same century, contacts with
Europeans became less confined and more frequent at times as the Dutch,
English and French East India companies turned their interests to the great
African island. The slave trade, which hitherto had exported to East Africa and
Arabia, shifted towards the Cape, the Mascarenes and the New World,
embarking from several points along the length of Madagascar's vast coasts.
Firearms began to spread not without certain repercussions on the political
front. The Sakalava empire and in particular the Iboina kingdom, its northern
component, reached its peak in the eighteenth century, while a good part of
the eastern coast united for the first time into the Betsimisaraka Confederation.
Before the end of the century, the northern Sakalava and the Betsimisaraka saw
their power decline irrevocably; in contrast a weak and disunited Imerina
418
Madagascar and the islands ofthe Indian Ocean 419
C dAmbrt
100
-I-
150 300 km
Bay of
Maha/amba '
Boina : Jr f .
fia-V 7^,fMajunga . \s-

Jks Mandritsan C Masoata
Mañanara
/ Sainte-Marte
SIHANAKA
Vk LAIaolrafy Foulpointe
^J 5 \^ / / A/ *- /Tamatave
I I \ .1? «*
Antananarivo
Vatomandry
ANTAMBAHOAKA
t
ANTEMORO
Fianarantsoa/
ANTEFASY :
C. of St V,ncent\ FIHERENANA
Farafangana
Vangaindrano
ANTAVARATRA
Fort-Dauphin
C. Salnte-Marie
28.1 Ethnic groups of Madagascar (adaptedfrom map of Madagascarfrom

Raymond K. Kent, Early Kingdoms in Madagascar, 1500-1700 (© 1970 by
Holt, Reinhard and Winston, Inc.) Map adapted by kind permission ofthe
publisher.)
succeeded in augmenting its food production and increasing its population,

thus undergoing a political rebirth which gave the Merina a solid base for their
future expansion.
Newcomers and coalitions
Before the nineteenth century the Antemoro had become famous as the only
group able to write in Malagasy (using the Arabic script) and also for their
special abilities in the realm of magic and religion. There is no complete
agreement on when and where the first anceston of the Antemoro landed in
Madagascar, nor where they came from, the extent of their Islamization and
their impact within the island. Gustave Julien was the first scholar to suggest an
Eastern African origin, a thesis which has been supported in the past decade.
One aspect should remain beyond dispute: the Antemoro arrived on the banks
of the Matitana river in the south-east, and it is only after they intermixed with
the tompon-tany (mästen of the soil, original inhabitants of the island) that they
formed a society and a state. It is possible, however, to date with some
precision their settling on the banks of the Matitana thanks to the reports of
three Portuguese sea-captains who visited the area between 1507 and 1514.
Only the third who was in the Matitana valley in 1514 speaks of a 'town
inhabited by many Moors'. As the ancestors of the Antemoro are the only
candidates for the 'Moors', the Matitana colony must have developed between
1509 and 1513.
Written Antemoro traditions record the encounters of the Antemoro with
those who inhabited the south-eastern coast before their arrival: they tell us
about the conflict of ideas between the newcomen with their patrilineal
concepts and the matrilineal concepts ofthe original peoples, and they describe
the many conflicts amongst the newcomers themselves. The Antemoro linked
together into four dominant aristocratic clans and four sacerdotal clans. Their
internal history well into the 1800s can be viewed as a series of conflicts
between the aristocratic Anteoni, Antemahazo, Anteisambo and Zafikasinambo
clans. The andrianoni, or supreme ruler of all the Antemoro, could come from
any of the four clans. The sacerdotal clans vied for supremacy in religious and
cultural functions. The Tsimeto, Zafimbolazi, Anakara and Anterotri sought,
for example, to be the guardians ofthe sacred Antemoro manuscripts (sorabe),
to become the high priests of the realm, or to have the closest proximity to an
andrianoni and his clan. These divisions would blur in the nineteenth century
when the sacerdotal Anakara clans monopolized both religious and political
power.
When the Antemoro established themselves on the eastern coast of Mada¬
gascar, they found, apart from several tompon-tany groups, other communities
who had been there before them, like the Zafindraminia, descendants of
Raminia. It is generally agreed that Raminia and his companions arrived in
Madagascar at least before the fifteenth century, but there is no consensus on
their origins. Wherever they came from, Eastern Africa provided an important
port of call en route to Madagascar.
By the time the Antemoro arrived, Zafindraminia enjoyed, on the whole,

privileged positions among the tompon-tany, the custom (sombili) being at that
time, that only a descendant of Raminia should slaughter domestic animals.
The Antemoro ancestors sought to intermarry with Zafindraminia women and
there were even rivalries over them. In the end, the conflicts were between the
two successive waves of newcomen. This created considerable insecurity which
was echoed in certain oral traditions of the tompon-tany as a fight of two
'giants'. The peak of this general conflict coincides with the emergence ofthe
aristocratic Zafikasimambo as the new and powerful rulen of the region around
1550. Not without some difficulties, the genealogical Antemoro texts try to
demonstrate that the Zafikasimambo descended from the Anteoni, so as to
accord them local and aristocratic origins. Nevertheless, it is known from an
independent source ofthe mid- 1600s that the Zafikasimambo were the last of
the overseas immigrants to settle on the banks of the Matitana and that their
name came through a tompon-tany wife. They were considered as ombiasa
(priests) and scribes sent from Mecca to instruct the local inhabitants. The
Zafikasimambo influenced the coune of Antemoro society in some decisive
ways from 1560 onwards. They took over the Zafindraminia ritualistic
monopoly of the sombili and used it as an economic and political tool. They
reduced the freedoms of the common people and made religion the centre of
political life. They also settled the conflicts between Antemoro and Zafin¬
draminia by putting to death as many male descendants of Raminia as possible
and by putting their women and children into confinement. It could be said
that the Zafikasimambo created the fint strong Antemoro kingdom at Matitana
and gave it two really capable andrianoni, namely Rabesirana (c. 1580-1615) and
Andriapanolaha (c. 1630
The turmoil that the Antemoro country experienced had repercussions
beyond its own frontiers. It produced, for example, a tendency towards
expatriation which became a permanent feature of this society. As the
sacerdotal clans fathered more of the specialists than could be absorbed into the
subdivisions of the Antemoro kingdom, it became common practice for the
ombiasa to spread out among other populations and offer their skills to those in
need. Concurrently, they worked towards a significant process of political
change the transition from small, isolated and inward-looking chiefdoms to
wider political unions. Constant unrest along the banks ofthe Matitana also led
to the exile of occasional memben of the aristocracy and the pogrom carried
out by the Zafikasimambo against the Zafindraminia must have driven a fair
number of involuntary migrants towards the interior of Madagascar.
According to Etienne de Flacourt, the Zafindraminia settled among the
Antanosy at the turn of the sixteenth century but they did not manage to
impose their authority. They introduced the sombili into the area, excelled in
the construction of wooden huts, amassed cattle, their main source of wealth,
and had a sense of kinship and a will for power. They thus attained a position
of privilege in the lands of Antanosy. But their acquisitive impulses, internal
rivalries and inability to find symbols that would unite all of the Antanosy
militated against the formation of a single state ruled by the Zafindraminia

kings. Instead, two parallel societies developed, with the Zafindraminia copying
the tompon-tany hierarchy. When the French founded Fort-Dauphin in
Antanosy in 1643, the two societies had interpenetrated not only as a result of
exogamy but also because the upper echelons on both sides were shifting
towards political centralization. However, during its 30 years on local soil
(1643-74), Fort-Dauphin completely arrested this process and political unity
continued to be elusive.
The Zafindraminia were not kingdom-builden and the Antemoro represent

a rare instance of newcomen who created their own society and state, but the
Maroserana stand out as the most important pan-dynastic family in the political
history of Madagascar. They were to give rulers to the Mahafaly, the Sakalava
of Menabe and Boina, to a part of the Antandroy and of the Bara. As with
other newcomen, their origin and formative period remain matters of
controveny. The Mahafaly and Sakalava oral traditions disagree about who the
original ancestor was and about where the proto-Maroserana hailed from. Oral
texts collected when the Sakalava were still independent simultaneously suggest
two very different origins: Andrianalimbe from the interior of Madagascar and
Andriamandazoala from overseas.
These contradictions help to explain the claims of certain authors that the
Maroserana originated, for example, in south-eastern Madagascar, that they
were Indians who had landed near present-day Fort-Dauphin around 1300 and
gone north to found the Zarabehava, the royal family of the Antesaka, thence
continuing westwards. These hypotheses have been corrected in recent years,
on the basis of primary sources. As the Antesaka royal traditions indicate, the
Zarabehava rulen were a branch ofthe west coast Sakalava who crossed Mada¬
gascar at some point between 1620 and 1650. Thus there was no movement
from east to west of the Maroserana dynasty: they formed in the south-west of
Madagascar. Furthermore, there is no evidence to confirm the idea of Indian
origin; no Indian linguistic vestiges have ever been found in the vocabulary of
the Maroserana rulen.
The two divergent Sakalava traditions are by no means contradictory.

Newcomers originating from the interior, represented by Great-Lord-'Ali"
(Andrianalimbe), and those from overseas, represented by Andriamandazoala,
drew closer together by a process of coalitions and conflicts, following a classic
pattern for newcomers in Madagascar. It is highly probable that some of the
Maroserana anceston arrived in south-western Madagascar by a sea route. The
ruling Maroserana among the Sakalava was known as the Zafivolamena (sons of
gold) while the tombs of the Maroserana rulers in Mahafaly are still called
volamena (red silver, i.e., gold). Given that no gold was found in the entire
southern half of Madagascar, traditions which record that it was imported
cannot be wrong in substance. Equally, the fact that volamena became a sacred
metal is a reflection of the pan-dynastic vocation of those who brought it:
when one considen the proximity of large Zimbabwean gold mines, with men
capable of bringing it to Madagascar in a single crossing, along with the
presence of several words and customs common to both Madagascar and

Zimbabwe, one can hardly avoid the probability of links between Maroserana
and Mwene Mutapa.
Some extremely interesting traditions, collected and published by a Sakalava
at the beginning ofthe twentieth century shed fight on the birth of Maroserana
history. They point to the initial failures of newcomers to the south-west to
become politically established and show that political successes were not
possible until links of kinship were established with the families of local chiefs.
They also reveal that the capacity for innovation was an important asset to early
Maroserana. They had a sense of territorial expansion; began to introduce
fortifications and build on elevated sites, used irregular soldiers, settled disputes,
knew how to attract followers by redistributing food and cattle and began to
consult existing diviners for affairs of state. They followed the same path among
the Mahafaly where alliances with the chiefly families, the Andriantsileliky, led
to their political supremacy.
It would appear that a combination of the high birth rate among the
Maroserana and attempts to apply the principle of primogeniture in royal
succession produced endless collateral branches and princes with no right to
rule. It was probably in response to this problem that they began to seek new
lands in order to give non-ruling members of the royal family villages to rule
over. Nonetheless, this problem still existed into the nineteenth century,
resulting in the voluntary exile of the collaterals, alliances with opponents of
the Maroserana and even to the formation of rival dynasties.
Outsiders and their impact

The accounts of Europeans who visited Madagascar prior to the colonial period
have been of inestimable value in their contribution to the study ofthe island's
past. The European presence up to 1800 had, by contrast, much less commend¬
able effects. In a general way, the Portuguese had a negative influence on the
Muslim-dominated trade and its oudets in the western Indian Ocean. The
north-western trading posts were located on small islets in the Bays of

Mahajamba, Boina and Bombetock. They were controlled and inhabited by
communities of Swahili-speaken who exported mainly rice and slaves to
Eastern Africa and Arabia. The north-eastern trading post, called Iharana, was
different in several respects. The Iharanians lived on the mainland; their long¬
standing culture was of local origin, centred on stone polishing and building
crafts. There is litde doubt that the decline of Iharana began at the point when
Europe interfered with Muslim commerce in the Indian Ocean. From 1700
onwards, Iharana was but a ghost of its former self and its culture disappeared
with hardly a tangible trace. In 1506, the Portuguese attacked the main Swahili
trading post in the north-west, in the Mahajamba Bay. They attempted to
destroy Boina but spared Bombetock as its shaykh proved to be friendly. The
Portuguese also came across a dense colony of Africans further north, in the
Bay of Anorontsanga. Some 2,000 strong, these Africans prepared for combat,
then changed their minds and disappeared inland. Their huts were burned and
it was as if 'the whole mountain was ablaze'. Yet this particular colony was not
part of the Muslim trade network; the Africans in question were escaped slaves
from Malindi, Mombasa and Mogadishu.
After a warlike beginning, however, the Portuguese in Mozambique changed
their tune and began to send a ship annually to north-western Madagascar to
take back cattle, ambergris, raffia cloth and slaves. In adding to the demand for
exports, the Portuguese were a factor in the emergence of several minor
political unions in the north-west of the island, not far from the Swahili trading
posts. One of these rapidly grew in importance under a ruler whose title was
tingimaro, who was 'continually at war with his neighboun'. In June 1614, the
tingimaro was visited in his capital of Cuala by Father Luis Mariano ofthe Order
ofJesus in Mozambique, who hoped to secure a religious foothold in the state,
but he refused to sign a treaty with the Portuguese and would not allow any
missionary activity within the island.
In the late 1580s, the Moors of Mazalagem refused to trade with the
Portuguese and a Dominican priest seeking local converts was killed. Only
royal orden from Lisbon prevented war but not before some reprisals against
Moors outside Madagascar. In 1590 north-western Madagascar was placed in
the trading zone ofthe Mozambique Island. Between 1614 and 1620 several
attempts were made to establish a church in north-western Madagascar.
The Portuguese signed several treaties with rulers along the western coast in
1613 and a Jesuit mission was sent to the Kingdom of Sadia near the
Manambolo river in 1616-17. This mission triggered a civil war which had
been brewing for some time and which would have important consequences in
the foundation ofthe Sakalava kingdom of Menabe. In 1641, the Portuguese
officially annexed western Madagascar in a Luso-Dutch treaty, which had no
practical effect. Indeed, around 1700, north-western Madagascar was under
Sakalava-Maroserana control and, a century later, the Malagasy would raid not
only Portuguese vessels in the Mozambique Channel but also their bases on the
Querimba and Ibo islands.
It would appear that in the early 1600s the Portuguese took fewer slaves
from Madagascar than their African and Arabian competitors. At the same time,
the Comoro Islands became the centre where slaves and merchandise from
Madagascar were collected for shipment to Malindi, Mombasa, Mogadishu and
Arabia. In 1614, Anjouan was reported to be teeming with male and female
slaves from Madagascar waiting to be taken to Arabia and exchanged for Indian
cotton and opium. Dhows from Shihir in the Gulf of Aden made direct
voyages to Madagascar to buy rice, millet, and young men and women. Clearly,
the Portuguese were losing their foothold in Madagascar to traditional buyen
and after about 1620 they seem to have lost interest in the island altogether.
As the seventeenth century edged towards its second half, new buyen
appeared in Madagascar who would not limit their activities to the north-west
of the island. At least eight ships are known to have supplied the Dutch at
Mauritius with rice and slaves from Madagascar, especially from the Bay of
Antongil, between 1639 and 1647. The Dutch had frequent conflicts with the
eastern Malagasy until 1655, when they began to make journeys to the island
from the Cape: they also participated in local conflicts. But the eastern coast
was disunited and the Dutch presence had no lasting impact. In the eighteenth
century, however, the Dutch, along with other Europeans, exerted consider¬
able commercial influence in several parts of Madagascar, and more especially
along its western coast.
English vessels came to Madagascar hundreds of times between 1600 and
1800, mosdy to its western shores and especially to the Bay of St Augustine.
Pidgin English, the language of trade, established itself in the area, and certain
local residents adopted English names and tides. In 1645, English Puritans were
sent to the Bay of St Augustine to establish a colony, and five years later
another colonial venture was started in the north-western islet of Nosy Be.
Both attempts failed with considerable loss of life. Unlike the Portuguese and
the Dutch, the English did not create any bases around Madagascar but they
nonetheless became the most active exporters of Malagasy slaves to the New
World, especially to Barbados and Jamaica. A census taken at Barbados at the
end ofthe seventeenth century showed 32,473 slaves, half of whom were from
Madagascar. Pirates took part in local wan, and contributed to the constant
supply of slaves for export. It was in the course of this period of pirates that two
marked changes took place in terms of trade and merchandise. First, as the
Europeans started to compete, the coastal rulers took advantage of the situation
to raise their prices. Secondly, and perhaps as a consequence, Europeans began
to trade discarded and faulty firearms in exchange for the much coveted slaves.
These firearms often exploded in the hands of those who used them.
Amongst the European nations seeking to establish a lasting foothold in
Madagascar, France was the only one to achieve a modest success with Fort-
Dauphin, a fortified setdement in south-eastern Madagascar, which lasted just
over three decades, from 1643 to 1674. Many of the men at Fort-Dauphin
married local women. Pronis, Fort-Dauphin's fint governor, married a
Zafindraminia rohandrian and the soldier nicknamed 'la Case' married an
heiress-apparent in Antanosy land and became a local military hero. The best
educated and longest standing of the Fort-Dauphin governon, Etienne de
Flacourt (1648-58), author of two fundamental works on Madagascar, became
a local potentate, forced to fend for himself as he received litde support from
France.
The impact of Fort-Dauphin has generally been underestimated: it was locally

sensitive but also had a wider dimension. The Antanosy were, for example,
about to attain their elusive political union with the Zafindraminia under the
leadenhip of Dian Ramack when Fort-Dauphin became an obstacle. At a time
when ships no longer came to it (1650-54), a bloody batde for supremacy broke
out with, on one side, the rohandrian newcomen and on the other, French
soldien. By 1653, Dian Ramack and many rohandrian were dead, and a number
of chiefs made allegiances (mifaly) to de Flacourt who had no interest in building
a local state. The situation wonened after his departure but the second French
East India Company, nonetheless, made further colonial attempts in Madagascar.
Outposts were set up at Antongil, on the Matitana and at Sainte-Marie. In

1667, some 2,000 colonists and French soldien landed at Fort-Dauphin. This
prompted a series of mifaly as rulen in Antanosy gave up the struggle for good.
Outside Antanosy the French presence at Fort-Dauphin accelerated the
foundation of Menabe by the Maroserana ruler, Andriandahifotsy, who sought
an alliance with Fort-Dauphin. Fort-Dauphin was abandoned in 1674 and its
last residents were evacuated to Mozambique, India and the Bourbon Isle (now
Réunion).
Plantation economies of the neighbouring Mascarene Islands of Bourbon
and Ile-de-France (Mauritius, abandoned by the Dutch in 1710), had a
disruptive impact on Madagascar, particularly on its eastern coast, in tbe course
of the eighteenth century. It is generally agreed that the majority of the slaves
imported into the Mascarene Islands between 1664 and 1766 came from
Madagascar but that the royal French administration, which bought these
islands from the French East India Company for over ,£7.5 million, gradually
substituted the Malagasy with slaves from Africa. It should not be forgotten,
however, that imports of slaves doubled between 1766 and 1788 and that the
Mascarenes, even after 1766, constandy interfered in the north-eastern part of
Madagascar.
The northern and western states: Mahafaly, Sakalava, Antankara,

Tsimihety
After the early Maroseranas' political success in the south-west, a new society
emerged between the Menarandra and Onilahy riven. Its name, Mahafaly (to
make sacred) is linked with social differentiation and rituals of royalty.
According to tradition, the fint Maroserana went into total seclusion, advised
and protected by the ruler's ombiasa. No longer seen, he became sacred (faly),
and by extension, rendered the land and its inhabitants maha-faly. After the
Masoserana came to power, Mahafaly society consisted of the privileged
(renilemy), the common people (yalohazomanga) , and those who came from else¬
where (folohazomanga). The descendants ofthe most powerful chief at the time
of Maroserana's formation (Tsileliki, the invincible), his former subjects, all
those who arrived with the proto-Maroserana, as well as those who were
occasionally rewarded, became the new élite.
Mahafaly history before 1800 is marked by conflicts with its neighbours, by
a split into four independent kingdoms and by the annexation of lands
belonging to the westernmost branch of the Antandroy. Sometime in the
1650s the first two breakaway kingdoms, the Menarandra and the Sakatovo,
were formed. A litde later, Menarandra spawned two more kingdoms, the
Linta (c. 1670) and the Onilahy (c. 1750). It was one ofthe Menarandra kings
who conquered the western Karimbola-Antandroy lands in the first half of the
eighteenth century. An Antemoro influence can be detected in the Onilahy
kingdom, whose first ruler took the Antemoro tide of andrianoni as his name.
There were six rulen in Sakatovo before 1800, seven in Linta, three in
Menarandra and two in Onilahy. These divisions reflect the numerous internal
dissensions amongst the Maroserana.
While the Maroserana had their fint political success among the tompon-tany
living south of the Onilahy, the largest waterway of the south-west, their
dynasty did not acquire political importance until some of its members went
north into the valley of the Fiherenana. This is where the Maroserana-
Volamena were born, and from where they crossed the Mangoky to form the
dynasty of the Sakalava states, Menabe and Iboina. There is no doubt that
Andriandahifotsy founded the Menabe dynasty and probably the Volamena
branch as well. The foundings of the Volamena and Menabe dynasties have
religious and secular aspects. A royal priest associated with crocodiles, Ndriam-
boay (noble crocodile) is credited with introducing the cult of the anceston
(dady) to the Maroserana, so that the rulen became the ampagnito-be (great royal
anceston, fathers of the people). According to other traditions, Ndriamboay
sacrificed Andriandahifotsy's wife, from whose blood came both the names
Volamena (silver which became red) and Menabe (great red). A more credible
tradition holds that Andriandahifotsy's wife managed to bring the first firearms
to him during a crucial batde. The ten weapons became part of the royal
treasure and her sons became the Volamena.
When the Maroserana came into contact with peoples of Menabe, they
were no longer in the lands of the south-western pastoralists. The coastal belt
which extends roughly between Majunga and Morondava, for approximately
650 kilometres, was inhabited by fishermen and farmen raising few cattle. With
the exception of the densely-populated Sadia, most of the inhabitants lived in
small communities, and had no weapons. This stretch of the coast was known
as Bambala, and its occupants spoke not Malagasy but Bantu idioms. They were
Cafres not Buques. The Sakalava came from Sadia and connected with the
Maroserana near the Mangoky river, probably at one of its smaller affluents
called the Sakalava. All the traditions agree that the Sakalava were the out¬
standing warrion of their time. They were the spearhead of the political
authority of the Volamena rulers but the dady cult gave this authority a religious
base which survived the formidable warriors and their direct descendants. The
formula which gave the kings posthumous names with the prefix andria (lord,
noble) and the suffix arivo (thousand, thousands) was introduced into Menabe
from the outside but probably not direcdy by the Maroserana. This formula
reflects the political idea that a true king must have many subjects.
Southern Menabe was being ruled by Andriandahifotsy in the early 1670s
when he was visited by a French cattle-trader from Fort-Dauphin who saw a
disciplined army of some 12,000 men and received 50 young bulls as a gift for
the fort's governor. Andriandahifotsy died around 1685. Conflicts over the
succession left one of his sons, Trimongarivo (who became Andriamanet-
riarivo) in charge of Menabe, which he expanded and populated with new
subjects recruited from the south-western pastoralists. His younger brother,
Tsimanatona $>y fitahina Andrianmandisoarivo), headed north to found Iboina.
It is difficult to tell whether Menabe had four or six rulen between 1720 and
1800 but it remained stable during this period apart from one case of regicide
reported in the 1730s. In the following decade, an important alliance with the
Andrevola rulers of the Fiherenana valley was concluded by a pact of royal
blood, which made the southern boundaries of Menabe more secure. In
Iboina, protected from the south by its sister-state, Menabe, the Muslim trading
posts and traden were forcibly brought under the protection of the Volamena
ruler. Majunga grew into the capital of Iboina, while its kings and their court at
Marovoay attained a splendour unequalled anywhere in Madagascar. At the
time of Andrianinevenarivo (known as Andrianbaga, died in 1752), Iboina was
at its height. Migrants from the north, moving south with their catde herds,
gradually absorbed most of the remaining Bambala population. This meant not
only the disappearance of the Bantu-speaking tompon-tany but also an advene
economic change at a time when the Sakalava states needed more farmers. As it
became harder to obtain agricultural labour from other parts of Madagascar, the
Sakalava began to raid the Comoro Islands and even South-eastern Africa.
During the last third ofthe eighteenth century, Queen Ravahiny (c. 1767-
1808) secured in Iboina a stable government and economic prosperity.
Nonetheless, Iboina started to decline during her rule. The Iboina Muslims who
had joined with the Sakalava constituted the country's most dynamic element
and sought to seize the throne. Another weakness came, this time indirecdy,
from the east coast, where a Polish-Hungarian noble, Count Benyowsky, had
penuaded a number of Betsimisaraka rulen to stop paying tributes to Iboina.
Warriors sent in 1776 to punish the former tributaries and eliminate Benyowsky
failed in their mission, revealing to the people ofthe interior that Iboina was no
longer omnipotent. Finally, Ravahiny made an error in foreign policy by
deciding to support the ruler of central Imerina, Andrianampoinimerina, against
neighbouring rivals.
The Sakalava kings were both despotic and lavish. Their daily fives were
regulated by the royal moasy (priests) and nothing could be undertaken without
their advice. There was a Royal Council composed of six elden who lived at
court. The fint minister, manantany, dealt direcdy with the many royal
councillors. His aide, the fáltatelo, was usually selected for his knowledge of
clans and lineages as well as Sakalava customs. Each village had its own royal
official who made sure that the people worked four days a week in the royal
rice fields, and that the royal herd was constandy renewed through gifts, which
were in fact fixed quotas. No monarch could rule without being in possession
of the royal dady (ancestral Volamena relics), especially since a new king
automatically became the sole intermediary between the living and the royal
anceston.
The Tsimihety grew out of groups of leaderless refugees from the east coast,
who, had fled the slave wars and setded with their cattle in the great
Mandritsara plain. In the far north, in the rocky part of the island, lived the
Antankara (rock people). Neither of these peoples had developed their own
state and both came to accept rule by Maroserana collaterals, the Volafotsy,
who, unable to rule among the Sakalava, had migrated into the northern
interior in search of a kingdom. The Tsimihety, however, quickly rejected the

Volafotsy (or the Volamena) and deliberately adopted their collective name (tsi¬
mihety) to indicate their non-submission to Maroserana authonty. Since they
failed to form a centralized kingdom, their eighteenth-century history remains
practically unknown.
The Antankara reactions were quite different. Some came to submit to
Iboina's founder and part of them were to be under Sakalava control, at one
time or another, well into the nineteenth century. Nonetheless, a royal line
going back to a Volafotsy named Kazobe emerged from them. A fully
centralized Antankara kingdom did not evolve, however, until the time of his
great-nephew, Andriantsirotso, who ruled, according to the texts, either from
c. 1697 to 1710, or from 1710 to c. 1750. He built the royal necropolis at
Ambatosahana and introduced the unifying symbol of Antankara royalty, the
saina, the flag with crescent and star. He associated his power with a formidable
moasy (priest) who made royal amulets. He became advisor to two of his
successors and lived to be over 100 years old. His son and successor, Lamboina,
had one ofthe longest rules in Madagascar, noted for its absence of wan. The
Antankara borrowed their political organization from the Iboina-Sakalava but
did not adopt the dady cult.
States ofthe interior: Betsileo, Merina, Sihanaka, Bezanozano, Tanala

and Bara
The Betsileo derive their name from a ruler named Besilau who is known to
have opposed the Sakalava expansion of Menabe into the interior highlands
during the 1670s. Yet the populations that later became known as the Betsileo
only came under a single government through external force in the 1800s. A
mixture of tompon-tany and diverse newcomers, the Betsileo were subdivided
into four main states: Arindrano in the south, merging with Ibara; Isandra in
the centre, facing west and the Sakalava; Lalangina in east-central Betsileo with
the Tanala as neighbours; and Manandriana in the north. It is unlikely that the
Betsileo had developed any kingdoms before the 1650s. Rasahamanarivo, the
founder of Lalangina, was forced to abdicate in favour of his brother because he
had leprosy, but around 1680 he went to Anndrano and was accepted there by
the southern Betsileo as their ruler. Struggles for succession and wars with
neighboun arrested the growth of Lalangina which, after a case of regicide,
broke up into four provinces. Nonetheless, the usurping branch of Lalangina's
royal family produced at least three strong rulers in the eighteenth century;
Roanimanalina, who reunited the kingdom and introduced a national militia;
Andrianonindranarivo, who carried out economic and social reforms; and
Ramaharo, who transformed the militia into an economic adjunct of state
intended to increase rice production. Indeed, dunng the eighteenth century,
the Betsileo became Madagascar's principal rice producers.
According to conflicting traditions, the ancestor of Isandra's royal family, the
Zafimanarivo, was either an Antemoro female or a Maroserana prince in exile.
Geography certainly favours a Maroserana antecedent. An early Zafimanarivo is

associated with the term volamena. One could not call Isandra a notable Betsileo
state until the 1750s when it came under the reign of a great monarch,
Andriamanalina I. Sometimes called the jester king' because of his ability to
-win political arguments through humour, Andriamanalina I was the first
Betsileo to conceive of a unified, single state. When diplomacy failed,
Andriamanalina I used military force. He expanded his kingdom westward,
placing his Betsileo on the Midongy massif and moving southward into Ibara,
where he obtained an important vassal, the Bara ruler Andriamanely II.
Retaining the traditional structures of each region, Andriamanalina I placed his
sons at the head of four new provinces where most ofthe Betsileo lived. At his
death in 1790, the Betsileo -were at their political zenith. As for Manandriana, it
had only a very short span as an independent state, having formed after 1750,
with its second ruler placing himself willingly under the Merina around 1800.
The Merina history begins, strictly speaking, with the reign of Ralambo,
variously estimated between 1610 and 1640. Hitherto, Imerina was not an
organized state; it consisted of numerous village chiefdoms peopled either by
tompon-tony, the Vazimba, or by the Hova migrants. The first member of the
Andriana dynasty was Andriamanelo, Ralambo's father. It was he who created
the fint fortifications at Alosava and who started using iron-tipped, rather than
clay-tipped, spears in war.
With Ralambo came the fint of the twelve sampy, amulet-guardians of
Imerina, the royal practices of circumcision and incest, the fandraona or annual
ceremony of the royal bath, divinization of departed monarchs, the noble
classes, silvenmiths and ironsmiths in the service of the state, head tax,
consumption of beef and a small but standing army. His son and successor,
Andrianjaka, took certain internal political measures which inspired fear in his
subjects but nevertheless he founded Antananarivo and exploited the marshes
for rich rice harvests, thus making the Merina the most numerous people
among the Malagasy.
The population growth in Merina increased the risk of famine and com¬
pelled its rulen to continually expand the irrigated areas. While Andrianjaka's
three successors distinguished themselves mainly by lengthening their royal
names, his great-grandson, Andriamasinavalona, was, in his turn, a great
monarch. He punued a policy of aggressive expansion and allowed little
independence to the many chiefs who came under his authority. He increased
the number of noble clans from four to six and rewarded the most deserving
nobles with village fiefs and the tide of tompon-menakely (masters of the fief).
Wishing to settle the question of his succession while he was still in power, he
divided Imerina into four provinces and placed a son to rule each one. Very
soon, he found himself at the head of four independent kingdoms and what is
more, he suffered the humiliation of being imprisoned by one of his sons. The
old king died around 1750 and Imerina became an arena for internal civil wan
in which even outsiders took part. By 1770, Ambohimanga had conquered
Ambohitrabiby and appeared strong under its ruler Andrianbelomasina. His
nephew, Ramboasalama, who became ruler at Ambohimanga around 1777

came to be regarded as the most important of Imerina's kings.
Assuming the name of Andrianampoinimerina, he first managed to secure
peace with his royal brothen and rivals, and then fortified the borderlands of
his kingdom. Antatanarivo fell under his domination in 1797, Ambohidratrimo
shortly afterwards, and several less defined areas of Imerina before 1800. He did
not content himself with unifying Imerina and began sending emissaries to
rulers in other parts of Madagascar with offers to become his vassal while
retaining autonomy, or else, if they refused, to see their territory conquered by
the Merina. This form of diplomacy succeeded at times, for example, among
the Manandriana Betsileo, the Andrantsay of Betafo, and in western Imamo.
Although neither the Sihanaka nor the Bezanozano developed into powerful
state structures, their history remains of considerable interest. The Sihanaka
group evolved around Lake Alaotra while the Bezanozano setded in an area
between the tropical rainforests of the eastern coast and the slopes of the
plateau. Both were ideally situated on the main commercial route linking
Imerina with the eastern coast and became suppliers to the Mascarene slave-
traden. Indeed, both the Sihanaka and the Bezanozano possessed so many
slaves that in 1768 they had to seek help from Europeans to suppress internal
slave rebellions. In 1667, when François Martin penetrated into Sihanaka
territory he found them inhabiting fortified villages. They had the only bridge
seen in Madagascar until that time by Europeans and they were also extremely
well armed with bows and arrows. Martin also confirmed the Sihanaka as one
ofthe most important trading groups in the islands, all of which would imply
considerable political organization. Yet there is no evidence that they ever had
central authority. In fact, the Sihanaka paid tribute to Iboina in the eighteenth
century but not without at least one serious attempt to free themselves of it
through a major attack which ended in failure. The Bezanozano, whose land
was originally called Ankay, were also ruled by local chiefs and lived in fortified
villages. Unlike the Sihanaka, however, they had at least one unifying
institution, in the form of amulets representing 11 protector-deities said to have
come from Sakalava land. Towards the end of the eighteenth century the
Bezanozano seem to have accepted the authority of Randrianjomoina, their
first king, but he was not to rule for very long as the Merina demanded and
secured his submission.
Wedged between the Betsileo highlanders and the coastal Antemoro, the
Tanala took their name from their natural surroundings, 'people ofthe forest'.
The Tanala soon developed into a highly mixed population since as many as 23
of their clans have claimed Betsileo descent. They never formed a state. Some
of them, however, do stand out in local history, such as the Tanala of Ikongo, a
huge rock accessible only by artificial passages. While completely isolated from
the Betsileo highlands, the Tanalo-Ikongo area was the natural hinterland of
the coastal region, accessible by such waterways as the Sandrananta, the
Faraony and the Matitana river of the Antemoro. It is therefore not surprising
that the Ikongo-Tanala accepted migrants from the eastern coast and that one
28.2 Bara ethnic group of Ambusitra, south-eastern Madagascar: memorial statue called
alualu, erected when a person dies without male issue or when the corpse does not rest in the
family grave, ne statue was surrounded by an enclosure ofposts bearing the horns of zebu
cows sacrificed at thefuneral ceremonies. Tie rule was that it was a femalefigure that becomes
the replacementfor the dead male and symbolizes the element destined to ensure issue. Made
from the hard wood ofthe camphor treee. Height: 107 cm (G. Berjonneau, ART 135,
Boulogne-Billancourt, © and photo)
of them, Rambo, said to have been an Anteono-Antemoro, fathered the local

dynasty ofthe Zafirambo, first mentioned by outsiden in 1668. According to
Tanala traditions, Rambo became king of Ikongo through his knowledge of
curative amulets and of sorabe. Rambo is also remembered as the 'law-giver'.
The greater part of the eighteenth century was marked by conflicts between
the Ikongo-Tanala and the Antemoro leading to a devastating Tanala invasion
of the lowlands towards the end of the century. At the beginning of the
nineteenth century, one of the Zafirambo, Andriamamohotra, allied himself
with Andrianampoinimerina.
The Bara, the major pastoralists of Madagascar, who have often been given
an African origin, seem to have emerged from somewhere in the southern
reaches of Arindrano country, along the Ihosy river. To the three great Bara
clans, the Bara-be, the Bara-Ianstsantsa and the Bara-Imamono, were added
two smaller clans, the Vinda of south-western Ibara and the Antivondro. There
were two distinct dynastic periods in the southern interior regions (of which
Ibara is the largest): one which coincides with the proto-Maroserana and is
largely unknown, and the more recent one, the period of the Zafimaneli,
which started around 1650. A total absence of succession rules and the ease
with which dissenters could migrate with their herds to better grazing lands
undoubtedly facilitated the disintegration of authority from 1640 onwards,
when the Mahafaly ruler, Dian Manhelle, invaded the area and began to instal
members of his family as local rulers. After his death in 1653, his descendants,
the Zafimanely, gradually assumed positions of authority in Ibara, but not
necessarily to its benefit. Between 1650 and 1680, Ibara's history is dominated
by competition among Zafimanely kinglets, not so much political as economic
in nature. With no system of regular tribute, the Zafimanely could only main¬
tain their power by securing wealth through legal decisions and more especially
by raiding their neighbours' cattle. This cycle of tension was broken only after
1800 by Raikitroka, a king who deserves to be studied in greater depth.
The eastern and southern states: Betsimisaraka, Antemoro,

Antambahoaka, Antefasy, Antesaka, Antanosy and Antandroy
The Betismisaraka, often divided into 'Northernen' (Antavaratra) and 'Southern¬
ers' (Antatsimo), and the Betanimena, established between the two, lived along
the eastern coast of Madagascar, roughly between the Bay of Antongil and
Vatomandry. According to numerous sources the tompon-tany of this region
were strangers to the larger political unions up until the beginning of the
eighteenth century and remained under chiefs (filohany) who rarely controlled
more than one or two villages. The Antavaratra were favoured by nature since
they had all the ports, while the Betanimena coast had none. As the Mascarenes
came to depend increasingly on eastern Madagascar for their imports, the
control of these ports made the filohany's fortune. This was also the part of
Madagascar which between 1680 and 1720 was most heavily colonized by
European, especially English, and American pirates. Many of them intermarried
with the Antavaratra, and fathered mixed-race children. From one such union,
around 1694, a certain Ratsimilahoe was born and a political change among the
Betanimena served as a springboard to launch him on to the political scene.
Around 1710, the Betanimena (known at that time as Sicoua) elected the chief
of Vatomandry as their supreme ruler, who would lead them into the northern
ports. It was by standing firm against an invasion of these ports that
Ratsimilahoe managed, despite his youth, to unite most ofthe Antavaratra.
Ratsimilahoe recaptured Fénérive in 1712 and the hasty retreat ofthe Sicoua
across fields of heavy, red clay earned them the name Be-tani-mena (the many of
red earth). Ratsimilahoe coined the name Betsimisaraka (the indivisible many)
to underline the lasting nature of the political union and in contrast to the
Betanimena. He secured peace with them by leaving to their king the port of
Tamatave, but eventually he was himself elected king of all the Betsimisaraka
and renamed Ramaroumanompou (the lord of many subjects) at his capital of
Foulpointe. But this peace lasted only six months and Ratsimilahoe retook
Tamatave, forcing the Betanimena king into an inaccessible area further south.
He then entered into alliances with the Antatismo and the Bezanozano.
Around 1730, Ratsimilahoe was one ofthe most powerful kings in Madagascar,
holding together a confederation of clans and families of rather independent
habits. He achieved this by allowing the traditional chiefs to keep their
autonomy and pay as much tribute as each saw fit, in slaves, cattle and rice. He
compensated the families of filohany who had fallen in battles against the
Betanimena and he allowed the Betanimena to bail out their relatives taken as
prisoners during hostilities. He dealt high justice, permitting any Betsimisaraka

to apply directly to him. Keenly aware of the strong centrifugal tendencies
which threatened his kingdom, Ratsimilahoe married the king of Iboina's only
daughter, in order to bind himself to the Volamena and the Sakalava empire.
He died in 1754, after a 40-year reign. Ofthe three kings who succeeded him
between 1755 and 1803, the second, Iavy, achieved fame only as the biggest
slave-trader in Madagascar. In the second half of the eighteenth century,
Betsimisaraka-land came to resemble Angola at the height of its slaving
tragedies in the early part ofthe seventeenth century.
There is no longer any doubt that both the Antesaka and the Antefasy of
south-eastern Madagascar were founded by migrants from the west. The
Antesaka, whose name indicates Sakalava origins, had left Menabe and ended
up in Nosipandra (later Vangaindrano), which became their capital. It would
appear that their migration was over by the 1650s. As for the Antefasy, their
traditions hold that the ultimate ancestor, Ndretsileo, came from the African
mainland to the River Menarandra and settled in Ibara. Difficulties with the
Zafimanely had wonened by the time of his grandson, Ndrembolanony. He
married the daughter of the ruler of Antevatobe and one of their three sons,
Marofela, coined the name Antefasy (sand people), meaning, by the sand
analogy, that his kingdom would be densely populated.
The most important Antefasy ruler prior to the 1800s was undoubtedly Ifara,
who resided at Ambaky (the old name for Farafangana) . Ifara gained a
Madagascar and the islands of the Indian Ocean 435
monopoly of trade with the fint European vessels to visit his coast and became
so powerful that he was, at one time, regarded as the sole proprietor of the
Manampatra river. Another Zarabehava ruler who is particularly remembered is
Maseba, who rescued the Antefasy from a brief Antemoro domination.
According to Antemoro sources, conflicts between the Antemoro and the
Antefasy can be dated to the 1680s. They continued into the 1700s but the
wars were not conclusive.
The Antesaka were more inward-looking and they had many difficulties
over successions. Tradition recalls that one of their Zarabehava kings, Raton-
galaza, 'either killed or expelled all his brothen'. His grandson, Lengoabo, was
the last eighteenth-century king and he extended the Antesaka boundaries to
their greatest limits. The history of the Antambahoaka and the Antemoro
during the eighteenth-century remains unknown, despite the fact that Matitana
is cited by European accounts as one of the main exporters of rice and slaves,
especially after 1724.
The European presence within Antanosy country was sustained not only by
links with Fort-Dauphin but also by the development, after the 1720s, of a
plantation economy by the Mascarenes, with an insatiable demand for slaves,
rice and cattle. The Count of Modave, governor of the second French
establishment at Fort-Dauphin (1767-70), discovered no less than 35 rulen
between the Ambolo valley and the Mandrare river. They battled with each
other over slaves and catde, causing frequent migrations of their own subjects.
Modave, who opposed the slave trade and came to replace it with legitimate
trade, was abandoned by France and ended up becoming a slaver to pay his
debts and resupply his own plantations on Ile de France (Mauritius).
It should be recalled that Fort-Dauphin was built at a time when Antanosy-
land was already one of the most densely populated in Madagascar, with a
surplus agricultural production and serious possibilities of becoming a unified
state. Shortly after the collapse of Modave's mission a French traveller described
Antanosy-land as 'the poorest and the saddest' in the island, thinly populated
and virtually without resources.
Antandroy, the southernmost state of Madagascar, divides into two very
distinct sections. In western Antandroy, the Karimbola have no memory of
migration. Eastern Antandroy was once occupied mainly by the Mahandrovato,
who were later submerged by a great influx of migrants from Ibara and
Antanosy seeking refuge in the arid south. It would appear that the ruling
dynasty of the Zafimanara emerged from among the Mahandrovato and that
their rule was gradually accepted by the Karimbola as well. The collective
name, Antandroy, was given by the Zafimanara to all the peoples living
between the Menarandra and the Mandare rivers, and it reflects a sense of
political unity and implies the existence of a Zafimanara confederation. By the
1790s, the Zafimanara, driven out by floods and unable to cope with the
changes in their former domain, had to take refuge on the Manombo plateau.
As for many other populations in Madagascar, the end of the eighteenth
century did not augur well for the Antandroy.
The Comoro Islands: Grande Comore, Anjouan, Moheli and

Mayotte
The Comoro Islands are situated roughly halfway between Madagascar and
Africa. The Grande Comore is nearest to the African mainland, Mayotte faces
Madagascar, while Anjouan and Moheli are in the middle of the archipelago.
There is general agreement that the Comoros were settled by Bantu-speaking
mainlanders who had no discernible links with Islam, that this great religion
was imported by the Afro-Shîrâzî and that the lasting supremacy of Sunnite
Islam among the Comoreans must be attributed to the political successes of
more recent Muslim arrivals from Africa. In the traditional accounts collected
by Sa'id Ahmad 'All, the pre-Islamic and Shirazi periods are associated with the
ancient chiefs (bëdja) and their fani successon, whose daughters (jumbe) would
come to marry the Sunnite newcomers. A chronology ofthe bedja and fani has
not yet been possible in the absence of detailed archaeological, linguistic and
historical studies. Such studies could go a long way towards a new under¬
standing of three monumental events; the migrations of the Bantu-speaking
peoples, Indonesian migrations to Africa and Madagascar and the advent of the
Swahili culture itself.
Along with the arrival of the last Muslims clearly appeared a political
concept. For the first time, the four Comoros were perceived as a single unity
where attempts were made to bring them under one monarch (sultan) residing
in Anjouan. The sultanic period may have started as late as 1506 or even before
1400. One or more ofthe Comoros might well have been under the control of
traders from Mozambique and Kilwa before the arrival on Grande Comore of
'Muslim Arabs, belonging to the same tnbe that founded Malindi'. Whenever
Mount Kartala erupted, the Grande Comore could always be seen from the
African mainland.
The fint Comorean dynasty is said to have originated in Anjouan, despite

the fact that the first ancestor migrated into Grande Comore. The second
ancestor, Hassan/Hassani, the real founder of the dynasty, is depicted as a
prolific builder of mosques on Anjouan. He is also considered the main propa¬
gator ofthe Shäfi'ite rite and his reign seems firmly linked with the adoption of
Sunnite Islam by the people of Anjouan. His sons had two names, one Bantu,
the other Arabic, and they extended the Hassanite dynasty to the other
Comoros, probably introducing the new religious faith at the same time. But
while Sunnite Islam took hold across the Comoros, bringing with it a new
system of justice with its kädi na(ib (village judges) and its madjelisse (body of
jurists), political rivalry became both a constant and a dynamic feature of
Comorean society.
European sources as well as local traditions reveal that Comorean history is
dominated not by Grande Comore and Mayotte but by Anjouan and Moheli.
Right at the start of the seventeenth century, independent visitors from
different parts of Europe confirm that Anjouan did indeed have a monarch - a
queen regarded as the supreme head of all four Comoros. According to the
same sources, each island had its own ruler who in turn revered the queen of
Anjouan as dynastic head. In 1602, for example, Moheli was ruled by an
independent and imposing Muslim king, who stunned European visitón with
his astonishing competence in navigational matters, his first-hand knowledge of
Africa and Arabia and his command of the Portuguese language. In 1614,
English voyagers found that Grande Comore was divided among ten 'lords',
that the queen of Anjouan (the 'old Sultana') had 'delegated' power over
Moheli to her two sons, and that Moheli's principal port was under Fombony's
governor, a man of great influence. In the same year, the old fani of M'samudu
proclaimed himself an independent sultan and master of northern Anjouan, and
the venerable queen no longer controlled her island. By 1626, there were two
contenders for the Mohelian Sultanate, a descendant of the fani and an 'Arab'
who had both married daughters ofthe 'last Sultan'.
With few exceptions, Comorean rulers sought good relations with the
Europeans who came to their islands. Nearly all the English ships bound for
Mocha, Persia and Surat, put into port at Anjouan, thus forging a long-lasting
Anglo-Anjouan link, a friendship on which the local sultans would call from
time to time. The Comorean traders had, for a long time, been middlemen
between Madagascar, Africa and Arabia, but this does not mean that the four
islands had nothing to export themselves. Wars which raged within an island,
such as the one on Grande Comore in 1620, and frequent sea-raids from one
island to another, often had as their main objective the capturing of slaves for
export. The wont period was that of the pirates in the western Indian Ocean
(1680-1720) when the sacking and destruction of Comorean towns became
commonplace. It was precisely between 1700 and 1720 that British naval
squadrons brought active aid to Anjouan and its sultan. This military assistance
reflected the desire to extend Anjouan's effective control over the other
Comoros and to obtain, in exchange, the sultan's refusal to shelter the pirates.
The pirate supremacy had ended by 1720 but not the British armed inter¬
ventions on behalf of Anjouan.
In 1736, according to Sa'ïd Ahmad 'Ali, Anjouan had a new monarch, Sultan
Ahmad, who believed that all the Comoros should be under a supreme ruler.
His reign was long but marked by an attempted coup d'état in 1743, a full-scale
political war with Mayotte a few years later, and above all, a major internal
revolt. The original inhabitants of Anjouan (the Wamatsaha), led by a
charismatic commoner named Tumpa, rose up in 1775 against the ruling Arab
element, demanding full equality. The Hassanites, who had taken on a platoon
of British marines, had no difficulty in shooting Tumpa down. His death ended
a movement that could have had far-reaching consequences for Anjouan and
the other Comoros. Yet, as the eighteenth century drew to a close, the real
problems were just beginning for the Comoros. Canoe fleets from Madagascar,
manned by the Sakalava and the Betsimisaraka, began to raid the four islands
for slaves, terrorizing the local inhabitants up to 1825. Moreover, upheavals
caused by the Merina expansion in Madagascar had an unexpected consequence.
Following the massive arrivals of Sakalava fleeing from Iboina, Mayotte became
Malagasy-speaking and the island acquired Malagasy sultans. Indeed, the
situation for the Comoros, halfway between Africa and Madagascar, was not at
all easy.
The Mascarene Islands: Réunion and Mauritius
In contrast to Madagascar and the Comoros, the Mascarene Islands east of

Madagascar remained uninhabited until the arrival of Europeans in the Indian
Ocean. Réunion (Bourbon) and Mauritius (He de France) became naval bases
controlled by large maritime companies. The development of a plantation
economy on these islands affected not only the eastern part of Madagascar but
also eastern and south-eastern Africa and even the coast of India. After 1638,
the Dutch concentrated their efforts on Mauritius, exploiting its timber and
developing a slave-trade with Madagascar, but in 1710 they abandoned the
island. Réunion received its fint setders from Fort-Dauphin in 1646 and 1654;
they were French and Malagasy rebels against the fint French East India
Company, which, nonetheless, managed to make Bourbon its preserve from
1664 to 1719. The French had colonized Mauritius from Réunion by 1721 and
both islands came under the control of the second East India Company until
1767, when the Mascarenes were turned over to the royal administration.
Inspired by the French Revolution in 1790, the planters of Réunion and
Mauritius began to demand a form of local nationalism which stood against
such metropolitan measures as the abolition of slavery in 1794. At the end of
the Napoleonic Wan, Mauritius passed to British control for good, and the
supply of slaves was temporarily halted.
Around 1710, Réunion and Mauritius together had some 2,000 inhabitants,
a third of whom were slaves. By mid-century the population had increased by
300 per cent but slaves made up more than two-thirds of this increase. Towards
the end of the eighteenth century the two islands had 120,000 inhabitants,
which included 94,000 slaves. From the beginning of the eighteenth century
five social groups could be seen: whites descended from first-generation
colonists, white creóles, mulatto creóles, resident white foreigners and slaves.
There were few emancipated slaves before 1797. Nonetheless, an important
change occurred within the slave population when Mascarene buyers gradually
turned their interests away from Madagascar and towards Africa and India in
search of new labourers. This change can be explained mainly by the belief that
the Malagasy were particularly prone to form fugitive slave colonies. In fact,
marronnage (fugitive slaves) in Réunion and Mauritius was, like everywhere
else, a response to slavery, made possible by the environment of mountains and
forests.
Although sugar cane had been cultivated on Mauritius since 1639, the fint
sugar-mill only came into being during the administration of Mahé de
Labourdonnais (1735-46). Mauritius did not export sugar until after the advent
of the royal French government which discarded the restrictive measures
imposed by the French East India Company and opened the islands to all
French nationals at the end of the 1760s. Coffee, introduced to Réunion in
1715 to become its main crop, attained its last record harvest in the early 1740s.
By 1767, it was in such a state of decline, as a result of competition from the
Antilles, that Reunion's planten had to fall back on spices. During the next 30
yean, as commercial interest shifted from the Adantic to the Indian Ocean, the
Mascarene Islands became the focal point of a veritable 'route des lies'.
Mauritius, which, in the Company days used to be visited by some 30 ships a
year, saw a tenfold increase in incoming vessels by 1803. Around 1800, the
privileged minorities in the Mascarene Islands had a standard of living which
equalled, or was superior to, that of all the other colonial territories.
The history of African societies,
Í500-Í800:
conclusion
Various labels have been applied to the period stretching from 1500 to 1800 in
African history. Many history books present it as the era of the slave trade,
overestimating the importance of the phenomenon in the continent's history.
They overlook the fact that, in West and Central Africa, the slave trade lasted
until about 1850 and in East Africa, it developed during the nineteenth
century, if we exclude an earlier slave trade to the Muslim world. The label
also takes no account ofthe fact that some parts ofthe continent, such as South
Africa, knew practically no slave trade at all. Certain historians, particularly the
neo-marxists, see the gradual integration of Africa into the world capitalist
economy dominated by Europe as the main feature of the period, thereby
giving more weight to Africa's external relations than to its internal evolution.
They portray the African peoples as hapless victims of world forces which they
can neither understand nor control. Other historians single out the movement
of populations and the definitive settling of the continent as the distinguishing
characteristic of the period. In fact, with the exception of a few cases, one has
to note that there were hardly any mass migrations in Africa after 1500. Still
other researchers would wish to present the history of these three centuries as a
series of ecological catastrophes and consider droughts and famines to be the
determining factors ofthe period.
Each of these labels has some element of truth, but none of them takes
sufficient account of the complexity and the dynamism of these three centuries
of African history. In this final chapter, we shall endeavour to extract the main
themes ofthe historical evolution of African societies during this period, basing
our discussions on the various chapters in this volume.
Let us start by taking the fundamental question of population movements.
Most of Africa was already colonized by 1500 and what oral traditions refer to
as migrations, were, in effect, population expansion and drift. Only north-east
Africa experienced mass movement of population during this period, among
the Oromo, the Somali, the Luo, the Karamojong, the Kalenjin, the Turkana
and the Maasai (see Chapter 3).
What is of much greater significance than the migrations is the fact that the
period from 1500 to 1800 was crucial for the great socio-political formations on
440
ne history ofAfrican societies, 1500-1800: conclusion 441
the continent. It was during these three centuries that most of the inhabitants of
the different regions of Africa coalesced into the wider social, economic, reli¬
gious, cultural and political units that make up the present-day African peoples.
As African societies evolved into distinct ethnic groups with their own
peculiar linguistic and cultural characteristics, much of Africa was changing as a
result of developments in its external relations. While in 1500 most African
societies were independent ofthe rest ofthe world, by 1800 much of Africa
had become integrated into the circuits of the world market which securely
linked it to Europe, America and Asia. This process of integration was
facilitated by the establishment of new communities on the continent, such as
the Dutch setders in southern Africa, the Portuguese in Angola and on the east
coast and the Ottomans in Egypt and the Maghrib. Many African societies had,
gradually, to change their way of life or move, or both. Soon, radically dif¬
ferent relationships grew up within and between societies, characterized by
dominance and dependence, both internally and in a world system in which
Europe had become the dominant power.
Pathe Diagne has identified the main economic structures which developed
during this period (see Chapter 2). The predatory economy of plundering, for
example, was the outcome of Spanish and Portuguese expansionism. It
completely disrupted the sub-systems of the Mediterranean and the Indian
Ocean and impoverished the countryside which was later further plunged into
crisis by the slave trade. A military aristocracy that lived by piracy and raiding,
using the services of freemen and slaves, emerged.
Like the predatory economy, the trading-post economy concerned itself
litde with innovation. The new maritime trading posts were fortresses rather
than commercial or industrial centres. On the coasts of Guinea and equatorial
Africa, in the Kongo, Angola and Senegambia, the Portuguese looted more
than they bought. From 1650 to 1800 the trading-post economy relied on the
transadantic slave trade. The societies affected by the trading-post economy
were gradually transformed in the seventeenth and eighteenth centuries. This
was a complex process involving major reorganizations, the main feature of
which, especially in West-Central Africa, was the predominance of the trading
networks over the states (see Chapter 18).
As the states declined, so did the old ruling élites, with a merchant class arising to
replace or complement them. The Portuguese penetration of Southern Zambezia,
for example, seriously weakened the power of the indigenous ruling class and
facilitated the institution of direct forms of peasant exploitation by Portuguese
capitalists (see Chapter 22). Similar developments brought about the mestizos and
creóles of Casamance, Guinea and Sierra Leone. The Afro-Portuguese and the
Anglo-Africans of Sierra Leone were merchant groups which acted as
middlemen between the European ships and the African societies ofthe interior.
Even in the case of the Fundj and Für sultanates, where the sultans appear
to have patronized and protected the long-distance trade linking them with
Egypt and the Red Sea, the bulk ofthe trade was in the hands ofthe Sudanese
djallâba (traders) who acted as the middlemen and financiers of foreign trade.
Thus from relative isolation in 1500, the various peoples of Africa gradually
became integrated into the world economy. In most cases, this integration was
accompanied by sharp social and political changes.
Despite the collapse of large states in western Sudan and north-east Africa at
the beginning of our period, the total surface area of territories controlled by
the states increased over the course of these three centuries: the period is also
characterized by numerous examples of political consolidation due to the
expansion and centralization of political institutions.
In north-east Africa, for example, although the sixteenth and seventeenth
centuries witnessed the collapse of the Christian Ethiopian empire, there was
expansion of this empire southwards, which stimulated the growth of new
states such as Bosa, Kaffa, Sëkko, Wolayta and Dauro.
In the case of Madagascar, we note the opposite process of centralization. By
the beginning of the seventeenth century, Madagascar still consisted essentially
of small, self-contained chiefdoms, but by the end ofthat century, the Sakalava
empire had established itself in the western part, and several kingdoms appeared
on the island. The Sakalava empire reached its peak in the eighteenth century
and the eastern coast became politically united for the fint time into the
Betsimisaraka confederation. By 1800, these two states were in decline and the
state of Imerina, formerly weak and disunited, gained power.
A similar process of unification and centralization occurred on the coast of
Lower Guinea. According to a map dated 1629, there were 38 states and
kingdoms then in the region, founded by the Ga and Akan peoples. From these
states emerged the three large empires of Aowin, Denkyira and Akwamu,
which by 1750 had merged to form the Asante empire.
A more common political phenomenon during these three centuries was for
a state in decline or collapse to be succeeded by several localized states or by
economic systems. Thus, in Central Africa, the Kongo, the Tio, Loango and
Ndongo (future Angola) kingdoms, which existed in the sixteenth century,
from 1665 began to decline and territories were reorganized on a larger scale
on the basis of economic imperatives dictated by the slave trade.
On the Upper Guinea coast, other states succeeded the empires of Songhay
and Mali. The Grand Foul empire arose in the seventeenth century on the
ruins of Songhay, but gave way to the empire of Kaarta during the second half
of the eighteenth century. All along the Adantic coast, the countries that had
grown up as a result ofthe dismemberment of Mali were unified by the Kaabu
(Gabu) in the seventeenth and eighteenth centuries and then by the Futa Jallon
in the eighteenth and nineteenth centuries. In the centre, the recovery took
place under the aegis ofthe Bambara of Segu and, in the south, the Jula (Dyula)
organized the empire of Kong in the eighteenth century.
A similar process can be seen in Southern Zambezia. The decline of the
Great Zimbabwe gave rise fint to the Torwa state and later to the Mutapa
empire at the beginning of the sixteenth century. The fragmentation and
decline ofthe Mutapa state began in 1629 but the Mutapa political entity did
not finally disappear until 1917.
ne history of African societies, 1500-1800: conclusion 443
In most ofthe new states, original systems of government and administration

were developed. Dahomey, for example, represented a new idea of the state.
The traditional view, which considered the state as a larger venion of the
family, was succeeded by the notion of a strong centralized state ruled by an
absolute monarch demanding unreserved loyalty from all his subjects.
In most of these states, various social classes emerged: aristocrats, military
groups, commonen and slaves. Thus, the Kanuri society in Borno was highly
layered. In Hausaland, as the aristocrats and traders grew rich, the distinction
shifted to an economic level with the attajirai (the wealthy) and the talakawa
(the poor).
The aristocracy (administrative and military) grew rich using various means
of exploitation. In Egypt, for instance, as the Ottoman empire declined,
internal struggles between different social strata developed. The critical state of
the economy was a reflection of a policy of oppression whereby the majority of
the people were controlled and exploited by a small élite made up of Marnluk
beys and their retinues. This oppressive situation led to the rise of popular
literature in Arabic, dealing with the exploitation ofthe peasants.
These oppressive regimes led to peasant revolts in many African regions.
There is an urgent need for a thorough study of these peasant uprisings in the
seventeenth and eighteenth centuries.
The slave trade
What importance should be given to the slave trade in African history? In other
words, what is our interest, as African historians, in the slave trade? All races
have been enslaved in the past: the word itself comes from 'Slavs', which means
East Europeans. But all other groups have found a way of eliminating the idea
from their historical consciousness. The Africans overestimate, to such an
extent the importance of slavery in their history that the term 'slave' is almost
equated with 'African'. As D.B. Davis has clearly shown in his books slavery is
a major phenomenon in the ideology of the modern Western world which we
must endeavour to undentand.
In Chapter 4, J. E. Inikori attempts to measure the role of black slavery in

the economic progress of the Western world. This is another fundamental
question which must be studied frankly and in depth. Marx and Engels argued
that though slavery is and always has been immoral, it has nevertheless been
essential for economic and therefore social progress. Without slavery, they
argue, there would have been no Greek or Roman civilization. Inikori argues
that African slavery was essential for the development of the Adantic-oriented
economic and geo-political system and the industrialisation of Western Europe.
But, Fernand Braudel has presented, in his three-volume Civilisation and
Capitalism, a different picture of the rise of European capitalism to world
power. Starting with the traditional economy of peasant subsistence through
the market place, he explains how, finally, a handful of banken and merchants
managed, by monopolizing trade and maximizing profit, to extend the growing
power of European capitalism. The role of African slavery in this debate needs
a radical reassessment.
In Chapter 5, J.E. Harris raises another important question related to slavery.

From all the available evidence, it is clear that the dehumanizing ofthe Africans
intensified between 1500 and 1800 as a result ofthe increase in slavery from
1619. It is, however, the intercontinental slave trade, more than anything else,
which established a black presence all over the world. It led to a major African
diaspora especially in the Americas and the Canbbean. This is where the
struggle for the liberation of blacks began, which was the foundation of the
pan-African Movement ofthe nineteenth and twentieth centuries.
Introduction of newfood crops

The introduction of new food crops from the Americas has been seen by many
authors as a positive aspect of transadantic trade. Some have asserted that these
crops, which led to a change in diet, improved the health of Africans, thus
contributing to population increase. But, as E. M'Bokolo has observed, the
effects of these agricultural changes on the African population are difficult to
interpret. He points out, for instance, that cassava has limited nutritional value
and the people who relied on it most, such as the Tio and the Mbosi, suffered
from serious malnutrition.
Population trends during the seventeenth and eighteenth centuries are

difficult to discern because this period, during which the slave trade reached its
peak, was also characterized by the introduction into Africa of new diseases,
such as smallpox, that were to become recurrent scourges.
What is certain, however, is that the new crops from the Americas and Asia,
introduced by Europeans to various African regions between 1500 and 1800,
especially maize, cassava, groundnuts, vanous types of yam, sweet potatoes,
citrus fruit, tomatoes, onions and tobacco, diversified the agricultural economy
of the continent. The present-day heavy reliance of Africans on maize and
cassava as the main staple diet dates back to this period.
Socially, according to M'Bokolo, this agricultural revolution contributed,
along with the development of trade, to the establishment of a new division of
labour. Men gradually left the many agricultural tasks to the women and slaves,
preferring instead to devote themselves to trade, which was much more
lucrative. Thus the development of domestic slavery and other forms of depen¬
dence was a direct consequence of these agricultural changes (see Chapter 18).
Ecological catastrophes
Attempts have been made to demonstrate a meaningful correlation between
periods of aridity, such as the one in question here, and major historical events.
These analyses tend to give the physical environment such importance that
Africans are reduced to powerless victims of nature. Nowhere do historians
discuss measures Africans have taken to counteract the effect of aridity, nor
The history of African societies, 1500-1800: conclusion 445
how droughts affect and contribute to changes in systems of production and

distribution, the enhanced value of food crops and the development of methods
of food preservation.
The question of droughts and famines also relates to the larger issue of the
efficiency of African agriculture in the pre-colonial period. The period 1500 to
1800 was characterized, as pointed out by Jan Vansina in Chapter 3, by the
emergence of intensive techniques for land use, which in turn allowed higher
densities of population. In many areas of Africa, advanced agricultural methods,
such as terracing, crop rotation, green manuring, mixed farming and regulated
swamp farming were used. These topics deserve more attention from historians
than the usual generalizations about famine and drought in Africa.
Christianity and Islam

The period from 1500 to 1800 saw a decline in Christianity in Africa, especially
in Ethiopia, on the east coast and, to a lesser extent, in the Kongo. Islam, on
the other hand, gained ground in West Africa, Sudan, Ethiopia and on the east
coast.
On the Lower Guinea coast, for example, Christianity was introduced into
the region by the Dutch and the British. They began by establishing elemen¬
tary schools in their castles at Cape Coast, El Mina and Accra. Later, in the
1750s, missionaries were sent to Cape Coast by the Society for the Propagation
of the Gospel. Also, some of the new converts were sent abroad for further
education and many returned home as teachen and missionaries. Thus the
foundation for the Christian revolution that was to take place in West Africa in
the nineteenth century was laid during this period.
In the Kongo, Christianity was introduced by the Portuguese during the
reign of Alfonso I (1506-43). Under the direction of his son Henrique, who
was consecrated bishop in Rome, he made the Catholic Church the state
religion. It was, to a very large extent, the religion of the urban nobility and
the ecclesiastical hierarchy remained chiefly Portuguese. But from 1645 to
1700 the Italian missionaries set out to systematically convert the population,
particularly in the rural areas (see Chapter 19).
Both Christianity and Islam, during this period, were syncretic. In the
Kongo, for example, Christianity coexisted with traditional religion. However,
attempts were made to organize independent churches. In the Kongo, early
attempts to establish an indigenous church started in the 1630s and reached a
peak in 1704, when Dona Beatrice Kimpa Vita began to preach a reformed
Christianity called Antonianism which rejected the missionaries and the whites.
Henceforth, indigenous Catholicism prevailed in the Kongo.
Turning now to Islam, it is evident that its expansion in Africa is an
important theme ofthe period from 1500 to 1800. In the Lower Guinea coast,
for example, the Manden and Hausa traders introduced Islam. It spread along
the northern trade routes, reaching Asante and Baule in the 1750s. By 1800,
Kamusi had a thriving Muslim quarter with aKur'anic school. In the Upper
Guinea coast, the Fulbe and the Manden were responsible for the spread of
Islam. They formed a FulbeMalinke religious alliance, not only to convert the
peoples of the region but also in order to subjugate them. Thus, the spread of
Islam was associated with political domination in many regions of Africa, as the
case of Senegambia clearly illustrates; here, the opposition between the Muslim
theocracies and the ceddo (warlord) regimes form the background to the history
ofthe region. In the Sudan, Islamization ofthe north ofthe country created an
ideological frontier between northern and southern Sudan which is still very
marked.
This period saw the establishment and expansion of two Muslim savannah
states the Fundj and Für sultanates. In the seventeenth and eighteenth
centuries, Islam continued to spread in the Bambara, Mossi, Kong and Gwirika
kingdoms, through traden and religious leaden, and even through violence.
The same process of Islamic expansion can be seen in Hausaland and in Borno.
Apart from the simple geographical spread of Islam in Africa at this time,
Muslim fundamentalism was an important factor in many regions. There was,
for example, the Nâsir-al-Dîn movement, which started in Mauritania and
then spread southwards. Its motives were partly economic (to control the trade
in grains and slaves) and partly religious (to purify and reform Islam by
replacing arbitrary rule with Muslim theocracy). The same reform tendency
can be seen among the Muslims in Hausaland, especially during the eighteenth
century. A community of Muslim scholan, with similar political, economic and
religious backgrounds, developed in various centres and became critical of the
established order represented by the aristocracy. As the rule of the aristocrats
became increasingly oppressive, the scholars openly attacked the established
order. This is the background to the djihâd ofthe nineteenth century.
As a result of the oppression of the African peasantry by the rural and urban
élites, especially in western Sudan, the Niger-Chad region, Egypt, the Sahara,
the Maghrib, Ethiopia and the Kongo, the Muslim leaders and the Christian
messianic movements found it easy to enlist the mass support ofthe peasantry.
Finally, it should be noted that the Africans viewed both Christianity and
Islam syncretistically. They accepted Islam while at the same time remaining
faithful to their traditional religions. These syncretisms later developed with
Christianity in Africa and in Afro-American wonhip in Brazil, Haiti and Cuba
(see Chapter 12).
Selected
bibliography
The publishers wish to point out that, although every effort has been
made to ensure that the details in this Bibliography arc correct, some
errors may occur as a result ofthe complexity and the international nature
ofthe work
Abbebe, B. (1971) Evolution de la propriété foncière au Choa (Ethiopia) (Paris: Imprimerie National
et Librarie Onentaliste, P. Geuthner).
Abdallah, Y. B. (1919) Tlie Yaos (arranged, ed. and tr. by M. Sanderson, Zomba; 2nd edn,
London: Frank Cass, 1973).
Abitbol, M. (1979) Tambouctou et les Arma de la conquête marocaine du Soudan nigérien en 1591 à
l'hégémonie de l'Empire du Macina en 1833 (Paris: Maisonneuve & Larose), 297 pp.
Abraham, D. P. (1969) 'The roles of Chaminuka and the Mhondoro cults in Shona political
history', in F. Stokes and R. Brown (eds), pp. 28, 46.
Abun-Nasr, J. M. (1975) A History ofthe Maghnb (2nd edn, Cambndge: CUP), 422 pp.
Adams W. Y. (1977) Nubia - Corridor to Africa (London: Allen Lane), 797 pp.
Ajayi, J. F. A. (ed.) (1989) General History of Africa, Vol. VI (Pans/Oxford/Berkeley:
UNESCO/ Heinemann/University of California Press), 861 pp.
Ajayi, J. F. A. and Crowder, M. (eds) (1971) History of West Africa (London: Longman; 2nd
edn 1976).
Ajayi, J. F. A. and Espie, J. (eds) (1965) A Tlwusand Years of West African History (Ibadan/
London: IUP/Nelson), 543 pp.
Akinjogbin, I. A. (1967) Dahomey and its Neighbours, 1708-1818 (Cambndge: CUP), 234 pp.
Alexandre, P. and Binet, J. (1958) Le groupe dit pahouin (Fang-Boubu-Beti) (London/Paris: IAI/
PUF), 152 pp.
Alkali, M. B. (1969) 'A Hausa community in crisis: Kebbi in the nineteenth century' (MA
thesis, Ahmadu Bello University, Zaria).
Allan, J. (1965) The African Husbandsman (London: Oliver and Boyd).
Alldridge, T.J. (1901) Tlie Sherbro and its Hinterland (London: Macmillan), 356 pp.
Alpers, E. A. (1975) Ivory and Slaves in East Central Africa to the later Nineteenth Century (London:
Heinemann).
Anquandah, J. (1982) Rediscovering Ghana's Past (London: Longman), 161 pp.
Anttila, R. (1972) An Introduction to Historical and Comparative Linguistics (New York:
Macmillan), 438 pp.
Aptheker, H. (1944) American Negro Slave Revolts (2nd printing, New York: Columbia
University Press), 409 pp.
Arhin, K. (1979) West African Traders in Ghana in the Nineteenth and Twentieth Centuries (London:
Longman), 146 pp.
Arhin, K. and Goody, J. (1965) Ashanti and the Northwest (Legon: Institute of African Studies,
University of Ghana).
Atkins, G. (ed.) (1972) Manding: Focus on an African Civilization (London: SOAS).
447
448 Selected bibliography
Austen, R. A. (1968) Northwest Tanzania under Gemían and British Rule, 1889-1939 (New
Haven: YUP).
Avaro, J. A. (1981) Un peuple gabonais à l'aube de la colonisation: Le Bas Ogowe au XIXe siècle
(Paris: Karthala/CRA), 290 pp.
Axelson, E. (1940) South-East Africa, 1488-1530 (London: Longman), 306 pp.
Axelson, E. (1973) Tlie Portuguese in South-East Africa, 1488-1600 (Johannesburg: WUP), 276
PP-
Barber, W. J. (1964) 77ie Economy of British Central Africa- a Case Study of Economic Development in
a Dualistic Society (London: OUP), 271 pp.
Barry, B. (1972) Le royaume du Waalo: Le Sénégal avant la conquête (Paris: Maspero), 393 pp.
Bastide, R. (1971) Les Amériques noires: les civilisations africaines dans le Nouveau Monde (Paris:
Payot), trans, as African Civilizations in the New World (London: Hunt), 232 pp.
Bauer, P. T. (1981) Equality, the Tliird World and Economic Delusion (London: George Weidenfeld
& Nicolson).
Behrens, C. (1974) Les Kroumen de la côte occidentale d'Afrique (Talcnce: CNRS, Centre d'études
de géographie tropicale), 243 pp.
Bernstein, H. (ed.) (1973) Underdevelopment and Development: The Third World Today (New
York: Penguin Books).
Berque, J. (1982) Ulémas, fondateurs, insurgés du Maghreb (Pans: La bibliothèque arabe, Sindbad).
Betoto, C. (ed.) (1950) Histoire de la royauté sakalava (Paris: Ecole nationale de la France
d'Outre-mer, typescript), pp. 1-32.
Bhila, H. H. K. (1982) Trade and Politics in a Shona Kingdom. Tlie Manyika and their Portuguese
and African Neighbours, 1515-1902 (London: Longman).
Binger, L. G. (1982) Du Niger au golfe de Guinée par le pays de Kong et ¡e Mossi 1887-1889 (2
vols, Pans: Hachette).
Biobaku, S. O. (ed.) (1973) Sources of Yoruba History (Oxford: Clarendon Press).
Birmingham, D. (1981) Central Africa to 1870. Zambezia, Zaïre and the South Atlantic
(Cambndge: CUP).
Birmingham, D. and Gray, R. (eds) (1970) Pre-colonial African Trade: Essays on Trade in Central
and Eastern Africa before 1900 (London: OUP), 308 pp.
Birmingham, D. and Martin, P. (eds) (1983) History of Central Africa, vol. I (London: Institute
of Commonwealth Studies), 314 pp.
Blackburn, R. H. (1982) Kenya's Peoples: Okiek (London- Evans).
Blake, J. W. (1942) Europeans in West Africa, 1450-1560 (tr. and ed. by J. W. Blake, London: for
the Hakluyt Society, 2nd edn, 2 vols).
Boahen, A. A. (1964) Britain, the Sahara and the Western Sudan, 1788-1861 (Oxford: Clarendon
Press, Oxford Studies in African Affairs) 268 pp.
Boahen, A. A. (1966) Topics in West African History (London: Longman).
Bosman, W. (1967) A New and Accurate Description ofthe Coast of Guinea (London: Frank Cass,
reprint of 1705 1st edn, with Introduction by J. R. Willis).
Boulègue, J. (1968) 'La Sénégambie du milieu du XVe siècle au début du XVIIe siècle' (PhD
thesis, 3rd cycle, University of Pans).
Bouquiaux, L. and Hyman, L. (eds) (1980) L'expansion bantoue (3 vols, Paris: SELAF).
Bovill, E. W. (1958) Tlie Golden Trade ofthe Moors (London: OUP, 281 pp; 2nd rcv. ed. 1968,
293 pp; London: OUP).
Boxer, C. R. (1963) Race Relations in the Portuguese Colonial Empire, 1415-1825 (Oxford:
Clarendon Press), 136 pp.
Boxer, C. R. (1969) 77ie Portuguese Seaborne Empire (London: Hutchinson).
Braudel, F. (1981) 77ie Structure of Everyday Life: the Limits ofthe Possible (trans, and rev. by S.
Selected bibliography 449
Reynolds, London).
Braudel, F. (1984) Civilization and Capitalism (3 vols, New York: Harper & Row).
Burnham, P. (1980) Opportunity and Constraint in a Savanna Society (London: Academic Press),
324 pp.
Bynon, T. (1977) Historical Linguistics (Cambridge: CUP), 301 pp.
Cà da Mosto, A da (1937) Tlie Voyages of Cadamosto and Other Documents on Western Africa in the
Second Half of the Fifteenth Century (trans, and ed. by G. R. Crone, London: Hakluyt Society),
159 pp.
Caillié, R. ("[828) Journal d'un voyage à Tombouctou et à Jenne (Paris: Imprimerie royale).
Cairns, T. (1971) Barbarians, Christians and Muslims (Cambridge: CUP).
Calonne-Beaufaict, A. de (1921) Azande; introduction à une ethnographie générale des bassins de
l'Ubangi-Uek et de l'Aruwimi (Brussels: M. Lamertin)
Capron, J. (1973) Communautés villageoises bwa (Mali, Haute-Volta) (Pans: Institut d'ethnologie,
Mémoire No. 9), 379 pp.
Carrette, E. (1844) Recherches sur la géographie et le commerce de l'Algérie méridionale (Paris: Sciences
historiques et géographiques).
Castries, H. de (1905-36) Les sources inédites de l'histoire du Maroc de 1530 à 1845 (18 vols, Pans:
Paul Geuthncr).
Centre de Civilisation burundaise (cd.) (1981) La Civilisation ancienne des peuples des Grands
Lacs (Paris: Karthala), 495 pp.
Chaunu, H. and Chaunu, P. (1955) Seville et l'Atlantique, 1504-1650 (11 vols, Pans: Ecole des
Hautes études).
Chittick, H. N. and Rotberg, R. I. (eds) (1975) East Africa and the Orient: Cultural Synthesis
in the Pre-Colonial Times (New York: Africana Publishing Company), 343 pp.
Chrétien, J. P. (1977) 'Les deux visages de Cham: points de vue français du XIXe siècle sur les
races africaines d'après l'exemple de l'Afrique orientale', in P. Guiral and E. Temime (eds),
L'idée de race dans la pensée politique française contemporaine (Pans: Editions du CNRS), pp.
171-99.
Chrétien, J. P. (ed.) (1983) Histoire rurale de l'Afrique des Grands Lacs (Pans: AFERA-Karthala),
285 pp.
Cissoko, S. M. (1968) Histoire de l'Afrique occidentale (Pans: PA), 233 pp.
Clérici, M. A. (1962) Histoire de la Côte d'Ivoire (Abidjan: CEDA).
Cohen, R. (1970) 'Incorporation in Bornu', in R. Cohen andj. Middleton (eds), From Tribe to
Nation in Africa (Scranton: Chandler Int. Co.), pp. 150
Commissariat, M. S. (1957) A History of Gujarat (Calcutta).
Cook, S. F. and Borah, W. (1971-74) Essays in population history, Mexico and the Caribbean (2
vols, Berkeley: UCP).
Cornevin, R. (1964) Note sur l'histoire de Sansanné Mango (London, Survey on African
Chronology).
Coupez, A. and Kamanzi, T. (1962) Récits historiques du Rwanda (Tervuren: MRAC), 327 pp.
Courtois, C. (1955) Les Vandales et l'Afrique (Algiers: Arts et Métiers Graphiques), 441 pp.
Crosby, C. A. (1980) Historical Dictionary of Malawi (New York: Scarecrow Press), 169 pp.
Cruickshank, B. (1853) Eighteen years on the Gold Coast of Africa (2 vols, London: Hurst and
Blackett).
Cuoq, J. M. (1975) Recueil des sources arabes concernant l'Afrique occidentale du Ville au XVIe siècle
(Biläd al-Südän) (Pans: editions du CNRS), 490 pp.
Curtin P. D. (ed.) (1967) Africa Remembered: Narratives by West Africans From the Era ofthe Slave
Trade (Madison: UWP), 363 pp.
Curtin, P. D. (1969) 77ie Atlantic Slave Trade: A Census (Madison: UWP), 338 pp.
Curtin, P. D., Feierman, S., Thompson, L. and Vansina, J. (1978) African History (Boston/
Toronto: Little Brown Company), 612 pp.
Cuvelier, J. and Jadin, L. (1954) L'ancien royaume du Congo d'après les archives romaines,
1518-1640 (Brussels: ARSOM).
Dampierre, E. de (1968) Un ancien royaume Bandia du Haut-Oubangui (Paris: Pion, Recherches
en sciences humaines, 24), 601 pp.
Dantzig, A. van (1978) Tlie Dutch and the Guinea Coast, 1674 A Collection of Documents
from the General State Archives at Tlie Hague (Accra: GASS), 375 pp.
Dapper, O. (1686) Naukeurige Beschrijvinge der Afrikaensche gewesten van Egypten, Barbaryen, Libyen
Blledulgerid... (Amsterdam: Van Meurs); 1670 Engl, trans, and adaptation, J. Ogilby, Africa:
Being an Accurate Description ofthe Regions ofAegypt, Barbary, Libya, etc. (London; 1670 German
trans., Besclireibung von Afrika... (Amsterdam: Van Meurs); 1686 French trans., Description de
l'Afrique... (Amsterdam: Wolfgang, Waesberge, et al).
D'Arianoff, A. (1952) Histoire des Bagesera, souverains du Gisaka (Brussels: Institut royal colonial
beige), 138 pp.
Davidson, B. (1965) Tlie Growth of African Civilisation: A History of West Africa, 1000-1800
(London: Longmans), 320 pp.
Davis, D. B. (1984) Slavery and Human Progress (New York: OUP), 374 pp.
Debourou, D. M. (1979) Commerçants et chefs de l'ancien Borgou, des origines à 1936 (Paris:
Université de Paris I, CRA).
Debrunner, H. W. (1965) TJie Church in Togo: A Church Between Colonial Powers (London:
Lutterworth), 368 pp.
Delafosse, M. (1912) Le Haut-Sénégal-Niger (Soudan français) (3 vols, Paris: Larose).
Delcourt, A. (1952) La France et les établissements français du Sénégal entre 1113 et 1763 (Dakar:
Mémoires de l'IFAN No. 17), 432 pp.
Denoon, D. (1983) Settler Capitalism: The Dynamics of Dependent Development in the Southern
Hemisphere (Oxford).
Deschamps, H. (1949/1972) Les Pirates à Madagascar aux XVlIe et XVHIe siècles (2nd edn,
1972, Paris: Berger-Levrault), 1st edn, 244 pp.
Deschamps, H. (1961/1965) Histoire de Madagascar (2nd and 3rd edns, Pans: Berger-Levrault,
Monde d'Outre-mer, Série Histoire), 348 pp.
Deschamps. H. and Vianès, S. (1959) les Malgaches du sud-est (Paris: PUF), 118 pp.
Diagne, P. (1967) Pouvoir politique traditionnel en Afrique occidentale: essai sur les institutions politiques
précoloniales (Paris: PA), 249 pp.
Diallo, T. (1972) Les institutions politiques du Fouta Djalon au XIXe siècle (Dakar: IFAN, Initiations
et études africaines), 276 pp.
Dickson, K. B. (1969) A Historical Geography of Ghana (Cambridge: CUP), 379 pp.
Dike, K. O. (1956) Trade and Politics in the Niger Delta, 1830-1885: An Introduction to the
Economie and Political History of Nigeria (Oxford: Clarendon Press, Oxford Studies in African
Affain), 250 pp.
Diop, A. B. (1981) La Société wolof, tradition et changement: les systems d'inégalité et de domination
(Pans: Karthala), 355 pp.
Diouf, M. (1980) 'Le Kajoor au XIXe siècle et la conquête coloniale' (PhD thesis, Paris,
University of Paris I).
Dramani-Issifou, Z. (1982) L'Afrique noire dans les relations internationales au XVIe siècle: analyse
de la crise entre le Maroc et le Sonrhai (Paris: karthala), 257 pp.
Dumestre, G. (1974/1980) Lageste de Sêgou (Paris: A Colin) (2nd edn, 1980); 1st edn, 579 pp.
Dunn, R. S. (1972) Sugar and Slaves: Tlie Rise of the Planter Class in the English West Indies
Selected bibliography 45 1
1624-1713 (New York: Chapel Hill, UNCP).

Dupire, M. (1962) Peuls nomades (Paris: Institut d'ethnologie), 336 pp.
Dupré, G. (1982) Un ordre et sa destruction (Pans: ORSTOM), 446 pp.
Ehret, C. (1971) Southern Nilotic History: Linguistic Approaches to the Study ofthe Past (Evanston:
NUP), 200 pp.
Ehret, C. and Posnansky, M. (eds) (1982) Tlie Archaeological and Linguistic Reconstruction of
Aßcan History (Berkeley: UCP), 299 pp.
Evans-Pritchard, E. E. (1940) T?ie Nuer (Oxford: OUP), 271 pp.
Evans-Pritchard E. E. (1971) Tlie Azande: History and Political Institutions (Oxford: Clarendon
Press), 444 pp.
Eyongetah, T. and Brain, R. (1974) A History ofthe Cameroon (London: Longman), 192 pp.
Equiano, Olaudah (1967) Equiano's Travels, edited by P. Edwards (London: Heinemann African
Wnters Senes).
Fage, J. D. (1969) A History of West Africa: An Introductory Survey (4th edn Cambridge: CUP),
239 pp. (Earlier edn publ. as An Introduction to the History of West Africa.)
Fage, J. D. and Oliver, R. A. (eds) (1977) Tlie Cambridge History of Africa, vol. 3 (Cambndge:
CUP), 803 pp.
Farb, P. (1969) Man's Rise to Civilisation (New York: Datton), 332 pp.
Feierman, S. (1974) 77ie Shambaa Kingdom: A History (Madison: UWP), 235 pp.
File, N. and Power, C. (1981) Black Settlers in Britain, 1555-1958 (London: Heinemann
Educational Books).
Fisher, A. G. B. and Fisher, H. G. (1970) Slavery and Muslim Society in Africa (London: C.
Hurst), 182 pp.
al-Fishtäll, 'Abd al-Azïz (1964) Manahil al-Safia fi Akhbar al-Muluk al-Shurafa (Rabat: Gunum
Allan).
Flint, J. E. (ed.) (1976) Tlie Cambridge History ofAfrica, vol. 5 (Cambridge: CUP).
Forde C. D. and Jones, G. I. (1950/1962) Tlie Ibo and Ibibio-speaking Peoples of South-Eastem
Nigeria (London: IAI, Ethnographic Survey of Africa, Part III), 94 pp.
Forde, C. D. and Kaberry, P. M. (eds) (1967) West African Kingdoms in the Nineteenth Century
(London: IAI/OUP), 289 pp.
Forde, D. (ed.) (1956) Efik Traders of Old Calabar (London: OUP).
Fortes, M. (1940) 'The political system ofthe Tallensi ofthe Northern Territories ofthe Gold
Coast' in M. Fortes and E.E. Evans-Pritchard (eds), African Political Systems (London- OUP)
pp. 239-71.
Franklin, J. H. (1967) From Slavery to Freedom: A History of American Negroes (fint published
1956) (New York: Knopf), 686 pp.
Freeman-Grenville, G. S. P. (1962a) Tlie Medieval History of the Tanganyika Coast (Oxford:
Clarendon Press), 238 pp. (2nd edn 1966, 314 pp.).
Freeman-Grenville, G. S. P. (1962b) Tlie East African Coast: Select Documents from the First to
Earlier Nineteenth Century (Oxford: Clarendon Press), 314 pp.
Freeman-Grenville, G. S. P. (1963) 'The coast, 1498-1840', in R. Oliver and G. Mathew
(eds), pp. 129-68.
Freeman-Grenville, G. S. P. (1965) Tlie French at Kilwa Island: An Episode in Eighteenth-Century
East African History (Oxford, Clarendon Press), 243 pp.
Freeman-Grenville, G. S. P. (ed.) (1980) Tlie Mombasa Rising against the Portuguese, 1631, from
Sworn Evidence (London: OUP), 166 pp.
Frobenius, L. (1911-13) Und Africa Spracli (3 vols, Berlin, Charlottenburg: Vita, Deutsches
Verlagshaus, S. D.).
Fyfe, C. (1964) Siéra Leone Inheritance (London: OUP), 352 pp.
Fyfe, C. (1965) 'Peoples ofthe Windward Coast, A.D. 1000-1800', in J. F. A. Ajayi andj. Espie
(eds), pp. 149-65.
Fyfe, C. and McMaster, D. (eds) (1977 and 1981) African Historical Demography vol. 1, 1977,
473 pp; vol 2, 1981 (Edinburgh: EUP).
Gado, B. (1972) Gazetters ofthe Northern Province of Nigeria, vol. 1 (London: Frank Cass).
Garlake, P. S. (1973) Great Zimbabwe (London: Thames and Hudson), 224 pp.
Garrard, T. F. (1980) Akan Weights and the Gold Trade (London: Longman), 393 pp.
Godard, L. (1859) Le Maroc, notes d'un voyageur (Algiers).
Goody, J. (1956) Social Organization ofthe Lowili (London: IAI/OUP), 119 pp.
Gorju, J. (1938) Face au royaume hamite du Ruanda, le royaume frère de l'Urundi (Brussels: Vromant)
118 pp.
Gourou, P. (1953) La densité de la population du Ruanda-Umndi: Esquisse d'une étude géographique
(Brussels: Institut royal colonial belge), 239 pp.
Grandidier, G. (1942) Histoire politique et coloniale (de Madagascar), vol. 5 (Paris).
Grandidier, A., Charles-Roux, J., Delherbe, C, Froidevaux, H. and Grandidier, G.
(eds) (1903-20) Collections des ouvrages anciens concernant Madagascar (9 vols, Pans, Comité de
Madagascar) .
Grandidier, G. and Decary, R. (1958) Histoire politique et coloniale (de Madagascar) vol. 5, part 3
(Antanananvo).
Gray.J. M. (1962) History of Zanzibarfrom the Middle Ages to 1856 (London: OUP), 314 pp
Gray, R. (1961) A History ofthe Southern Sudan (London: OUP), 219 pp.
Gray, R. (ed.) (1975) The Cambridge History of Africa, vol. 4 (Cambridge, CUP), 738 pp.
Greenberg, J. H. (1980) 'Classification of African Languages', General History of Africa, vol. 1,
ch. 12 (Pans/London/Berkeley: UNESCO/Heinemann/University of California Press), pp.
292-308.
Grigg, D. B. (1980) Population Growth and Agrarian Change: An Historical Perspective (Cambridge:
CUP).
Guillot, B. (1973) La Terre Enkou (Pans: Mouton), 126 pp.
Gutman, H. G. (1976) 77ie Black Family in Slavery and Freedom (New York: Pantheon Books),
664 pp.
Guy, J. (1979) 77ie Destruction of the Zulu Kingdom: Tlie Civil War in Zululand, 1879-1884
(London Longman), 273 pp.
Hacquard, A. (1900) Monographie de Tombouctou (Pans. Société des éditions coloniales et
maritimes), 119 pp.
Hagenbucher Sacripanti, F. (1973) Les fondements spirituels du pouvoir au royaume de Loando, R.
P. du Congo (Pans: ORSTOM), 214 pp.
Haig, Sir W. (ed.) (1937) T7ie Cambridge History of India, vol. 3 (London: CUP).
Hair, P. E. H. (1974) 'From language to culture: some problems in the systematic analysis ofthe
ethnohistoncal records ofthe Sierra Leone region', in R. P. Moss and R.J. A. R. Rathbone,
77ie Population Factor in African Studies (London: ULP), pp, 71-83.
Hama, B. (1967a) Histoire du Gobir de Sokoto (Pans: PA), 167 pp
Hama, B. (1967b) Histoire traditionnelle d'un peuple, les Zamia-Songhay (Pans: PA), 273 pp
Hama, B. (1968) Contribution à la connaissance de l'histoire des Peuls (Pans: PA), 362 pp.
Hama, B. (1969) Histoire traditionnelle des Peuls du Dallo! Boboye (Niamey: CRDTO), 160 pp.
Hamani, D. M. (1989) Au carrefour du Soudan et de la Berbérie: le sultanat touareg de l'Ayar (2 vols,
Pans: Université de Pans 1 Panthéon-Sorbonne), 521 pp.
Hanke. L. (ed.) (1969) History of Latin American Civilization: Sources and Interpretation (2 vols,
London: Methuen).
Harms, R. (1981) RiVer of Wealth, River of Sorrow: Tlie Central Zaire Basin in the Era ofthe Slave
and Ivory Trade, 1500-1891 (New Haven and London: YUP), 277 pp.
Harris, J. E. (1971) Tlie African Presence in Asia: Consequences of the East African Slave Trade,
(Evanston, 111.: NUP), 156 pp.
Harris, J. E. (1982) Global Dimensions of the African Diaspora (Washington DC: HUP).
Hartwig, G. W. (1976) 77ie Art of Survival in East Africa: Tlie Kerebe and Long-distance Trade,
1800-1895 (New York/London: Africana Publishing), 253 pp.
Hasan, Y. F. (1967) Tile Arabs and the Sudan (Edinburgh: EUP), 298 pp.
Herskovits, M.J. and Harwitz, M. (eds) (1964) Economic Transactions in Africa (London: Rout-
ledge and Kegan Paul), 444 pp.
D'Hertefelt, M. (1971) Les clans du Rwanda ancien: éléments d'ethnosociologie et d'ethnohistoire
(Tervuren MRAC), 85 pp.
Heusch, L. de (1972) Le roi ivre ou l'origine de l'Etat (Pans. Gallimard), 331 pp.
Hill, A. and Kilson, M. (eds) (1969) Apropos of Africa (London: Frank Cass), 390 pp.
Hilton, A. (1985) Family and Kingship (Oxford. OUP).
Holt, P. M. (ed.) (1968) Political and Social Change in Modem Egypt: Historical Studies from the
Ottoman Conquest to the United Arab Republic (London: OUP).
Hopkins, A. G. (1973/1975/1983) An Economic History of West Africa (London: Longman). (1st
edn 1973, 337 pp.)
Hopkins, J. F. P. and Levtzion, N. (1981) Corpus of Early Arabic Sources for West African History
(Cambridge: CUP), 492 pp.
Ingham. K. (ed.) (1974) Foreign Relations of African States (London: Butterworth, Colston Paper,
No. 75), 344 pp.
Inikori, J. E. (1979) 'The slave trade and the Atlantic economies, 1451-1870', in Tlie African
Slave Trade from the 15th to the 19th Century: Reports and Papers of the Meetings of Experts
Organized by UNESCO, Port-au-Prince, Haiti, 31 January-4 February 1978, 77ie General
History of Africa - Studies and Documents, No. 2 (Pans: UNESCO), pp. 56-87.
Inikori, J. E. (cd) (1982) Forced Migration: Tlie Impact ofthe Export Slave Trade on African Societies
(London: Hutchinson University Library for Africa: New York: Africana Publishing Company),
349 pp.
Isaacman, A. F. (1969) 'The prazos da Coroa, 1752-1832: A functional analysis of the
Portuguese system', Studia, 26, pp. 149-78.
Izard, M. (1981) Histoire du Yatenga des origines à 1895 (Ouagadougou).
Jackson-Haight, M. V. (1967) European Powers and South-East Africa, 1796-1856 (London:
Routledge and Kegan Paul).
Jadin, L. and Dicorato, M. (1974) Correspondence de Dom Afonso, roi du Congo, 1506-1543
(Brussels: ARSOM 41-3), 245 pp.
James, C. L. R. (1963) The Black Jacobins: Toussaint L'Ouverture and the San Domingo Revolu¬
tion (New York: Vintage).
Johnston, H. H. (1902) Tlie Uganda Protectorate (2 vols, London: Hutchinson).
Jones, A. H. M. and Monroe, E. (1978) A History of Ethiopia (Oxford: OUP), 196 pp.
Julien, C. A. (1951-56) Histoire de l'Afrique du Nord: Tunisie, Algérie, Maroc. De la conquête arabe
à 1830, 2nd edn (2 vols, Paris: Payot).
July, R. W. (1974) A History ofthe African People (London: Faber), 650 pp.
Kagame, A. (1954) Les organisations socio-familiales de l'ancien Rwanda (Brussels: Académie royale
des sciences coloniales), 355 pp.
Kagame, A. (1963) Les milices du Rwanda précolonial (Brussels: Académie royale des sciences
coloniales), 196 pp.
Kagame, A. (1972) Un abrégé de l'ethno-histoire du Rwanda (Butare: Editions universitaires du
Rwanda), 286 pp.
Kaira, O. Y. (1970-71) 'A survey ofthe history ofthe Kaira Clan of Northern Malawi' (Student
seminar paper, Chancellor College, University of Malawi).
Kalinga, O.J. M. (1985) A History ofthe Ngonde Kingdom (Berlin/New York: Mouton).
Kaniki, M. H. Y. (ed.) (1979) Tanzania Under Colonial Rule (London: Longman), 391 pp.
Keen, B. and Wasserman, M. (1980) A Short History of Latin America (Boston: Houghton
Mifflin), 574 pp.
Kent, R. K. (1970) Early Kingdoms in Madagascar, 1500-1100 (New York: Holt, Reinhart and
Winston).
Kesteloot, L. (1972) Da Monzón de Sêgou, epopee bambara (4 vols, Pans: F. Nathan).
Kesteloot, L. (1983) Biton Koulibaly, fondateur de l'empire de Sêgou (Dakar: NEA), 96 pp.
Kietegha, J. B. (1983) L'or de la Volta Noire (Paris: Karthala, CREA), 247 pp.
Kimambo, I. N. and Temu, A. (eds) (1969) A History of Tanzania (Nairobi: EAPH), 276 pp.
Kimble, D. (1963) A Political History of Ghana: Tlie Rise of Gold Coast Nationalism, 1850-1928
(Oxford: Clarendon Press), 587 pp.
Kirkman, J. S. (1964) Men and Monuments on the East African Coast (London: Lutterworth
Press), 224 pp.
Ki-Zerbo, J. (1978) Histoire de l'Afrique noire (Paris: Hatier), 702 pp.
Klein, H. S. (1967) Slavery in the Americas: A Comparative Study of Virginia and Cuba (Chicago:
University of Chicago Press), 270 pp.
Klein, M. A. (1968) Islam and Imperialism in Senegal Sine-Saloum, 1841-1914 (Edinburgh: EUP),
285 pp.
Kouanda, A. (1984) 'Les Yarse: fonction commerciale, religieuse et légitimité culturelle dans le
pays moaga (Evolution historique)' (PhD thesis, University of Pans I).
Kup, A. P. (1961) A History of Sierra Leone, 1400-1181 (London: CUP), 212 pp.
Labarthe, P. (1803) Voyage à la Côte de Guinée (Pans: Bossange, Masson & Besson), 310 pp.
Laburthe-Tolra, P. (1981) Les seigneurs de la forêt (Paris: Publications de la Sorbonne), 490 pp.
Lamb, V. (1975) West African Weaving (London: Duckworth).
Lamphear, J. E. (1976) 77ie Traditional History of the Jie of Uganda (Oxford: Clarendon Press,
Oxford Studies in African Affairs), 281 pp.
Lange, D. (1977) Chronologie et histoire d'un royaume africain (Wiesbaden: Franz Steiner).
Lara, O. D. (1979) 'Negro resistance to slavery and the Atlantic slave trade from Africa to Black
Amencas' (working paper, Meeting of Experts organized by UNESCO on 77ie African Slave
Trade from the 15th to the 19th Century, Port-au-Prince, Haiti, 31 January-4 February 1978,
Tlie General History ofAfrica - Studies and Documents, 2 (Paris: UNESCO), pp. 101-14.
Laroui, A. (1970) L'histoire du Maghreb: un essai de synthèse (Pans: Maspero), 390 pp.
Le Barbier, C. (1916-17) 'Notes sur le pays des Bara-Imamono, région d'Ankazoabo', BAM,
new series, 2, pp. 63-162.
Lee, R. B. (1979) 77ie Kung San: Men, Women and Work in a Foraging Society (Cambndge: CUP),
526 pp.
Leo Africanus [Jean Léon l'Africain] (1956) Description de l'Afrique, tr. by A. Epaulard, with
notes by E. Epaulard, T. Monod, H. Lhote and R. Mauny (2 vols, Paris: Maisonneuve).
Lévi-Provinçial, E. (1922) Les Historiens des Chorfa (Paris: Larose).
Levtzion, N. (1968) Muslims and Chiefs in West Africa: A Study of Islam in the Middle Volta Basin
in the Precolonial Period (Oxford: Clarendon Press, Oxford Studies in African Affain), 228 pp.
Lewis, I. M. (ed.) (1966) Islam in Tropical Africa (London: OUP, for the IAI), 470 pp.
Lewis, I. M. (ed.) (1968) Islam in Tropical Africa (London: OUP, for the IAI), 470 pp.
Lewis, W. A. (1978) 77ie Evolution ofthe International Economic Order (Princeton: PUP).
Little, K. (1951) Trie Mende of Sierra Leone (London: Routledge and Kegan Paul), 307 pp.
Lombard, J. (1965) Structures de type 'fiodal' en Afrique noire: étude des dynamismes internes et des
relations sociales diez les Bariba du Dahomey (Pans/ The Hague: Mouton), 544 pp.
Lopez, R. S. (1976) Tlie Commerdal Revolution ofthe Middle Ages, 950-1350 (Cambndge: CUP).
Lougnon, A. and Toussaint, A. (eds) (1937) Mémoire des Iles de France et de Bourbon (Paris:
Libraire Ernest Leroux), 203 pp.
Lovejoy, P. E. (1983) Transformations in Slavery: A History of Slavery in Africa (Cambndge: CUP),
349 pp.
Lovejoy, P. E. (1985) Salt of the Desert Sun: A History of Salt Production and Trade in Central
Sudan (Cambridge: CUP), 351 pp.
Ly, A. (1958) La compagnie du Sénégal de 1613 à 1686 (Pans: PA), 316 pp.
MacMichael, H. A. (1922) A History ofthe Arabs in the Sudan (2 vols, Cambndge: CUP).
Madiega, G. Y. (1978) 'Le Nord-Gulma précolonial (Haute-Volta) origine des dynasties,
approche de la société' (PhD thesis, Univemty of Paris I).
Madiega, G. Y. (1982) Contribution à l'histoire précoloniale du Gurma (Haute-Volta) (Wiesbaden:
Franz Steiner).
Mage, E. (1868) Voyage dans le Soudan oaidental (1863-1866) (Paris: Hachette). (New edn by
Karthala, 1980, Frobenius Institute).
Malzac, V. (1921/1930) Histoire du royaume Hova depuis ses origines jusqu'à sa fin (Antananarivo:
Impnmene catholique). (Repnnted 1930).
Maquet, J. J. (1954) Le Système des relations sociales dans le Ruanda ancien (Tervuren: MRAC),
221 pp.
Marees, P. de (1602/1605/1905) A Description and Historical Declaration ofthe Golden Kingdom of
Guinea, abridged English trans, of 1602 Dutch onginal, published in S. Purchas (1613/1905),
vol 6, pp. 247-396; French trans. 1605, Description et récit historique du riche royaume d'or de
Guinée (Amsterdam: Claessen).
Martin, B. G. (1971) 'Notes on some memben of the learned classes of Zanzibar and East
Africa in the nineteenth century', AHS, 4, 3, pp. 525-45.
Martin, G. (1948) Histoire de l'esclavage dans les coloniesfrançaises (Paris: Geuthner).
Martin, P. M. (1972) 77ie External Trade ofthe Loango Coast, 1516-1810: Tlie Effects of Changing
Commerdal Relations on the Vili Kingdom of Loango (Oxford: Clarendon Press, Oxford Studies
in African Affain), 193 pp.
Marty, P. (1927) Les Chroniques de Ovalata et de Nema (Pans: Genthner).
Mauny, R. (1970) 'Les navigations anciennes et les grandes découvertes', in H. Deschamps (ed.).
Histoire générale de l'Afrique noire (Pans: PUF), vol. 1, Part II, pp. 203-18.
Mauny, R. (1971) Les siècles obscurs de l'Afrique noire (Paris: Fayard), 314 pp.
Mauny, R., Thomas, L. V. and Vansina, J. (eds) (1974) 77ie Historian in Tropical Africa
(London: OUP for the IAI), 428 pp.
Mauro, F. (1960) Le Portugal et l'Atlantique au XVlle siècle, 1510-1670 (Paris: SEVPEN).
M'Bokolo, E. (1981) Noirs et blancs en Afrique êquatoriale: les sociétés côtières et la pénétration française
(Pans/The Hague: Mouton, Civilisations et Sociétés, No. 69), 302 pp.
M'Bokolo, E. (1983) 'Histoire des maladies, histoire et maladie: l'Afrique', in M. Auge and S.
Herzlich (eds) Le sens du mal (Paris: Editions des Archives contemporaines).
McCulloch, M., Littlewood, M. and Dugast, I. (1954) Peoples of the Central Cameroons
(London: IAI, Ethnographie Survey of Africa, Western Africa, Part IX), 172 pp.
McLeod, M. D. (1981) Vie Asante (London: British Museum Publications).
Meek, R. L. (1976) Social Science and the Ignoble Savage (Cambridge: CUP, Cambridge Studies in
the History and Theory of Politics), 249 pp.
Meillassoux, C. (ed.) (1975) L'esclavage en Afrique prêcoloniale (Paris: F. Maspero), 582 pp.
Mellafe, R. (1975) Negro Slavery in Latin America (Berkeley, CA: UCP), 172 pp.
Merrick, T. W. and Graham, D. H. (1979) Population and Economic Development in Brazil,
1800 to the present (Baltimore: The Johns Hopkins Univenity Press).

Mettas, J. (1978) Répertoire des expéditions nêgnères françaises au XVIIIe siècle, ed. by S. Daget
(Nantes).
Miers, S. and Kopytoff, I. (eds) (1977) Slavery in Africa: Historical and Anthropological Perspectives
(Madison: UWP),474pp.
Miller, J. C. (1973) 'Requiem for the Jaga', CEA, 49, pp. 121-49. (See also CEA, 69 and 70).
Miller, J. C. (ed.) (1980) 77ie African Past Speaks: Essays on Oral Tradition and History (London:
Hamden, Dawson & Archon), 284 pp.
Minchinton, W. E. (ed.) (1969) 77ie Growth of English Overseas Trade in the 11th and 18th
Centuries (London: Methuen).
Mokhtar, G. (ed.) (1981) General History of Africa. Vol. II: Ancient Civilizations of Africa (Pans/
London/Berkeley: UNESCO/Heinemann/Univenity of California Press).
Monteil, C. (1924) Les Bambara de Sêgou et du Kaarta (Paris: Larose), 404 pp; 2nd edn, 1977
(Pans: Maisonneuve), 441 pp.
Monteil, C. (1932) Une cité soudanaise, Djennê, métropole du delta central du Niger (Pans: Société
d'études géographiques, maritimes et coloniales); 2nd edn, 1971 (Paris: Anthropos).
Morris, A. (1981) Latin America: Economie Development and Regional Differentiation (London:
Hutchinson), 256 pp.
Mudenge, S. I. (1974) 'An identification ofthe Rozvi and its implications for the history ofthe
Karanga', RH, 5, pp. 19-31.
Mutumba, M. (1973) Bulozi under the Luyana Kings. Political Evolution and State-Formation in Pre-
Colonial Zambia (London: Longman), 278 pp.
Mworoha, E. (1977) Peuples et rois de l'Afrique des lacs (Dakar. NEA), 352 pp.
Mworoha, E. (1981) 'Redevances et prestations dans les domaines royaux du Burundi
précolonial', Mélanges R. Mauny Le sol, la parole et l'écrit: 2000 ans d'histoire africaine (Pans:
SFHOM), pp. 751-68.
Nacanabo, D. (1982) 'Le Royaume maagha de Yoko' (PhD thesis, Univenity of Paris I).
Nachtigal, G. (1876) Le Voyage de Nachtigal au Ouadai: Traduction de Van Vollenlioven (Paris).
Ndoricimpa, L. (ed.) (1984) L'arbre mémoire. Traditions orales du Burundi (Pans/Bujumbura:
Editions Karthala/Centre de Civilisation Burundaise).
Niane, D. T. (1975a) Recherches sur l'empire du Mali au Moyen-Age (Pans: PA).
Niane, D. T. (1975b) Le Soudan occidental au temps des grands empires, Xle-XVle siècles (Pans: PA).
Niane, D. T. (ed.) (1984) General History of Africa. Vol IV: Africa from the Twelfth to the Sixteenth
Century (Paris/London/Berkeley: UNESCO/Heincmann/ Univenity of California Press).
Niane, D. T. and Wondji, C. Enquêtes orales faites en Côte d'Ivoire (1913-1980) et en Guinée
(1966-1910).
Nicolas, G. (1975) Dynamique sociale et appréhension du monde au sein d'une société hausa (Paris:
Institut d'ethnologie, Travaux et mémoires de l'Institut d'ethnologie, No. 78), 661 pp.
North, D. C. (1981) Strudure and Change in Economie History (New York: Norton).
North, D. C. and Thomas, R. P. (1973) 77¡e Rise of the Western World: A New Economic
History (Cambridge: CUP).
Northrup, D. (1978) Trade Without Rulers: Precolomal Economic Development in South-Eastern
Nigeria (Oxford. Clarendon Press), 269 pp.
Nsanze, A. (1980) Un domaine royal Burundi: Mbuye (erw. 1850-1945) (Pans: SFHOM), 93 pp.
Nziziwe, I. (1972) Studies in Ibo Political Systems: Chieftaincy and Politics in Four Niger States
(London: Frank Cass), 287 pp.
Obenga, T. (1976) La cuvette congolaise: Les hommes et les structures (Paris: PA).
Ochieng*. W. R. (1974a) A Pre-Colonial History of the Gusii of Western Kenya 1500-1914
(Nairobi: EALB), 257 pp.
Selected bibliography ASI
Ochieng', W. R. (1975a) A History ofthe Kadimo Chiefdom of Yimbo in Western Kenya (Nairobi:
EALB), 78 pp.
Ochieng', W. R. (1975b) 'Undercivilization in Black Africa', in W. R. Ochieng' (ed.) (1975),
77ie First Word (Nairobi: EALB), pp. 1-20.
O'Fahey, R. S. (1980) 'State and Society in Dar Fur' (London: Hunt).
O'Fahey, R. S. and Spaulding, J. L. (1974) Tlie Kingdoms ofthe Sudan (London: Methuen),
235 pp.
Ogot, B. A. (1967) A History of the Southern Luo, Vol. I: Migration and Settlement 1500-1900
(Nairobi: EAPH), 250 pp.
Ogot B. A. (ed.) (1976a) History and Social Change in East Africa (Nairobi: EALB), 235 pp.
Ogot B. A. (ed.) (1976b) Kenya Before 1900 (Nairobi: EAPH), 291 pp.
Oliver, R. (ed.) (1977) Tlie Cambridge History of Africa, Vol. 3; From c. 1050 to c. 1600 (Cambndge:
CUP), 803 pp.
Oliver, R. and Mathew, G. (eds) (1963) A History of East Africa, vol. I (Oxford: Clarendon
Press), 500 pp.
Oloo, P. C. (1969) 'History of settlement: the example of Luo clans of Alego: 1500-1918'
(undergraduate dissertation, Nairobi Univenity).
Ott, T. O. (1973) 77ie Haitian Revolution (Knoxville: UTP), 232 pp.
Pachai, B. (ed ) (1972) 77ie Early History of Malawi (London: Longman), 454 pp.
Pageard, R. (1963) Civilisation mossi et Egypte ancienne (Geneva: Institut africain).
Pages, A. (1933) Au Rwanda sur les bords du lac Kivu: Un royaume hamite au centre de l'Afrique
(Brussels: Institut royal colonial belge), 703 pp.
Palmer C. A. (1976) Slaves ofthe Wlûte God: Blacks in Mexico 1510-1650 (Cambridge, Mass:
HUP) 234 pp.
Palmer, H. R. (1936) 77ie Bornu, Sahara and Sudan (New York: Negro Univenity Press), 296 pp.
Palmer, R. and Parsons, N. (eds) (1977) The Roots of Rural Poverty in Central and Southern
Africa (London: Hememann).
Patterson, K. D. (1975) 77ie Northern Gabon Coast to 1815 (Oxford: Clarendon Press), 167 pp.
Pélissier, P. (1966) Les paysans du Sénégal: les civilisations agraires du Cayor à la Casamance (Saint-
Yrieix, Haute-Vienne: Imprimerie Fabrègue), 941 pp.
Perlman, M. L. (1970) 'The traditional systems of stratification among the Ganda and the
Nyoro of Uganda', in A. Tuden and Plotnicov (eds), Social Stratification in Africa (London:
Macmillan), pp. 125-62.
Perrot, C.H. (1982) Les Anyi-Ndenye et le pouvoir politique aux XVIIIe et XIXe siècles (Abidjan/
Paris: CEDA).
Person, Y. (1964) 'En quête d'une chronologie ivoirienne', in J. Vansina, R. Mauny and L. V.
Thomas (eds), 77ie Historian in Tropical Africa (Oxford: OUP), pp. 322-38.
Person, Y. (1970) 'Le Soudan nigérien et la Guinée occidentale', in H. Deschamps (cd ), Histoire
générale de l'Afrique noire, vol. 1 (Paris: PUF), pp. 271-304.
Person, Y. (1972) The Dyula arid the Manding World (London: SOAS, Conference on Manding
Studies, duplicated).
Phillipson, D. W. (1977) 77ie Later Prehistory of Eastern and Southern Africa (London- Heine-
mann), 323 pp.
Piault, M. H. (1970) Histoire Mawn: introduction à l'étude des processus constitutifs de l'Etat
(Paris: Editions du CNRS), 206 pp.
Plancquaert, M. (1971) Les Yaka: essai d'histoire (Tervuren: MRAC, Annales sciences humaines,
vol. 71).
Pollock, N. C. and Agnew, S. (1963) A Historical Geography of South Africa (London:
Longman), 242 pp.
Porter, D. H. (1970) Tlie Abolition ofthe Slave Trade in England 1184-1801 (New York:
Archon).
Price, R. (1976) Tlie Guiana Maroons: Historical and Bibliographical Introduction (Baltimore: JHUP)
184 pp.
Priestley, M. A. (1969) West African Trade and Coast Society (London: OUP), 207 pp.
Quarles, B. (1961) Tlie Negro in the American Revolution (Chapel Hill: UNCP), 231 pp.
Randies, W. G. L. (1968) L'ancien royaume du Congo des origines à la fin du XIXe siècle (Paris/The
Hague: Mouton), 275 pp.
Ranger, T. O. (ed.) (1968) Aspects of Central African History (London/Nairobi: Heinemann/
EAPH), 291 pp.
Ranger, T. O. (1979) Revolt in Southern Rhodesia 1896-1891: A Study in African Resistance, 1st
paperback edn (London: Heinemann), 403 pp.
Ranger, T. O. and Kimambo, I. N. (eds) (1972) Trie Historical Study of African Religion
(London: Heinemann), 307 pp.
Rau, V. (1966) 'Les marchands-banquien étrangère au Portugal sous le règne de Joao III
(1521-1557)', in Les aspects internationaux de la découverte océanique aux XVe et XVIe siècles
(Paris).
Rinchon, R. F. D. (1964) Pierre Ignace Liévin van Alstein, capitaine négrier: Gand 1133 - Nantes
1193 (Dakar: IFAN, Mémoire No. 71), 452 pp.
Roberts, A. D. (1976) A History of Zambia (London: Heinemann), 288 pp.
Roche, C. (1976) Conquête et resistance en Casamance (Dakar: NEA).
Rodney, W. (1970) A History ofthe Upper Guinea Coast 1545-1800 (Oxford: Clarendon Press),
283 pp.
Rodney, W. (1972) How Europe Underdeveloped Africa (Dar es Salaam/London: TPH and Bogle
L'Ouverture), 316 pp.
Rodney, W. (1975) 'Africa in Europe and the Amencas', inj. D. Fage and R. Oliver (eds), 77ie
Cambridge History of Africa, vol. 4 (Cambndge: CUP), pp. 578-651.
Rotberg, R. and Mazrui, A. (eds) Protest and Power in Black Africa (New York: OUP), 1274
pp.
Rout, L. B. Jr (1976) 77ie African Experience in Spanish America, from 1502 to the Present Day
(Cambndge: CUP), 404 pp.
Ryder, A. F. C. (1969) Benin and the Europeans, 1485-1891 (London: Longmans), 372 pp.
Salim, A. I. (1980) 'Kenya, Muslims' in Encyclopedia of Islam, new edn (Leiden/Paris:
Bnll/Maisonneuve et Larose).
Sautter, G. (1966) De l'Alantique au fleuve Congo, une géographie du sous-peuplement (Paris:
Imprimerie nationale), 1102 pp.
Schove, D. J. (1973) 'African droughts and the spectrum of time', in D. Dalby and R. J. H.
Church (eds) Droughts in Africa (London: IAI), pp. 38-53; 2nd edn, 1977.
Séré de Rivières, E. (1965) Histoire du Niger (Paris: Berger- Levrault), 311 pp.
Shaw, T. (1978) Nigeria: ¡ts Ardiaeology and Early History (London: Thames & Hudson).
Shepherd, J. F. and Walton, G. M. (1972) Shipping, Maritime Trade, and the Economic
Development of Colonial North America (Cambridge: CUP).
Shiroya, O. J. E. (unpublished) 'The Lugbara: at the nexus of three worlds', in J. B. Webster
(ed.), Uganda before 1900, vol. 1.
Skinner, E. P. (1964) 77ie Mossi of Upper Volta: Tlie Political Development of a Sudanese People
(Stanford: SUP), 236 pp.
Smaldone, J. P. (1977) Warfare in the Sokoto Caliphate (Cambridge: CUP), 228 pp.
Southhall, A. W. (1970) 'Rank and stratification among the Alur and other Nilotic peoples', in
A. Tuden and Plotmcov (eds), Soria/ Stratification in Africa (London: Macmillan), pp. 31
Sow, A. I. (1971) Le filon du bonheur éternel par Mouhammadou Samba Mambeya (Pans: Colin).
Steinhart, E. I. (1981) 'Herden and farmen', in C. C. Steward and D. Crummey (eds), Modes
of Production in Africa (London: Sage Publications), pp. 115-56.
Stokes, E. and Brown, R. (eds) (1966) 77ie Zambezian Past: Studies in Central African History
(Manchester: MUP), 427 pp.
Suret-Canale, J. (1970) La République de Guinée (Pans: Editions sociales), 431 pp.
Swartz, A. (1971) Tradition et changements dans la société guère (Paris- ORSTOM, Mémoires No.
52), 259 pp.
Tamrat, T. (1977) 'Ethiopia, the Red Sea and the Horn', Cambridge History of Africa, vol. 3
(Cambridge: CUP), pp. 98-182.
Tardits, C. (1980) Le royaume bamoum (Paris: Armand Colin), 1078 pp.
Tardits, C. (ed.) (1981) Contribution de la recherche ethnologique à l'histoire des civilisations du
Cameroun (2 vols, Paris: Editions du CNRS), 597 pp.
Taylor, J. G. (1979) From Modernization to Modes of Production: A Critique of the Soríologies of
Development and Underdevelopment (London: Macmillan).
Thomas, J. M. C. (1979) 'Emprunt ou parentés', in S. Bahuchet, Pygmêes de Centrafrique (Pans),
pp. 141-69.
Thompson, R. F. and Cornet, J. (1982) T?ie Four Moments ofthe Sun: Kongo Art in Two Worlds
(Washington: National Art Gallery).
Thornton, J. K. (1983) 77ie Kingdom of Kongo in the Era ofthe Civil Wars, 1641-1118 (Madison:
UWP), 193 pp.
Tosh, J. (1978) Clan Leaders and Colonial Chiefs in Lango: the Political History of an East African
Stateless Sodety, 1800-1939 (Oxford: Clarendon Press, Oxford Studies in African Affain),
293 pp.
Toure, M. (1974) Mande Influences in the Gyaman Kingdom, Colloquium, Bonduku, January.
Toussaint, A. (1971) Histoire des l'Ile Maurice (Pans: PUF), 128 pp.
Toussaint, A. (1972) Histoire des Iles Mascareigries (Paris: Berger- Levrault), 351 pp.
Trimingham, J. S. (1970) A History of Islam in West Africa (Oxford: OUP), 262 pp.
Turnbull, C. M. (1966) Wayward Servants: Tlie Two Worlds of the African Pygmies (London: Eyre
and Spottiswood), 390 pp.
UNESCO (1963) Nomades et nomadisme au Sahara (Pans), 195 pp.
UNESCO (1979) 77ie African Slave Trade from the Fifteenth to the Nineteenth Century, General
History of Africa, Studies and Documents, No. 2 (Paris: UNESCO), 330 pp.
UNESCO (1980) Historical Relations across the Indian Ocean, General History of Africa, Studies and
Documents No. 3 (Pans: UNESCO), 198 pp.
Valensi, L. (1969) Le Maghreb avant la prise d'Alger, 1190-1830 (Paris: Flammarion), 141 pp.
Valensi, L. (1977) Fellahs tunisiens: l'économie rurale et la vie des campagnes aux 18e et 19e siècles
(Paris/The Hague: Mouton), 421 pp.
Vansina, J. (1961) De la tradition orale; essai de méthode historique (Tervuren: MRAC, Mémoire
No. 36).
Vansina, J. (1966) Kingdoms of the Savanna: A History of Central African States until the European
Occupation, trans, of Lei anciens royaumes de la savane (Madison: UWP), 364 pp.
Vansina, J. (1972) La légende du passé: Traditions orales du Burundi (Tervuren: MRAC), 257 pp.
Vansina, J. (1973) 77ie Tio Kingdom ofthe Middle Congo, 1880-1892 (London: OUP), 586 pp.
Vansina, J. (1978) T7ie Children ofWoot: A History ofthe Kuba Peoples (Madison: UWP), 394 pp.
Van Thiel, H. (1911) 'Buzinza unter der Dynastie der Bahinda', Anthropos, pp. 497-520.
Vérin, P. (1975) Les êdielles andennes du commerce sur les côtes nord de Madagascar (2 vols, Lille:
Univenity of Lille).
Wallerstein, I. (1976) 'The three stages of African involvement in the world economy', in P. C.
W. Gutkind and I. Wallentein (eds), Tlie Political Economy of Contemporary Africa (London-
Sage Publications), pp. 30-57.
Walvin, J. (1972) 77ie Blank Presence: A Documentary History ofthe Negro in England (New York:
Schocken Books).
Walvin, J. (1973) Black and Wliite: the Negro and English Society, 1555-1945 (London: Allen Lane
and Penguin Press), 273 pp.
Warren, R. (1980) Imperialism: Pioneer of Capitalism (London: NLB and Veno).
Webster, J. B. (ed.) (1979) Chronology, Migration and Drought in Interlacustrine Africa (London.
Longman and Dalhousie Univenity Press), 345 pp.
Weiskel, T. C. (1980) French Colonial Rule and the Baule Peoples, 1889-191 1 (Oxford: Clarendon
Press) 323 pp.
Wilbur, C. M. (1967) Slavery in China during the Former Han Dynasty (New York: Russell and
Russell).
Wilks I. G. (1962) 77ie Tradition of Islamic Learning in Ghana (Lcgon).
Wilks, I. G. (1975) Asante in the Nineteenth Century: Tlie Structure and Evolution of a Political Order
(Cambridge: CUP), 800 pp.
Williams, E. (1970) From Columbus to Castro: Tlie History ofthe Caribbean, 1492-1969 (London:
André Deutsch), 576 pp.
Williamson, K. (1971) 'The Benue-Congo languages and Ijo', inj. Berry and J. H. Greenberg
(eds), Linguistics in Sub-Saharan Africa (Mouton), pp. 245-306.
Wilson, M. and Thompson, L. M. (eds) (1969, 1971) 77ie Oxford History of South Africa (2
vols, Oxford. Clarendon Press).
Womersley, H. (1984) 'Legends and the history ofthe Luba', in T. Q. Reefe (ed.), Legends and
History ofthe Luba (Los Angeles: Cross Roads Press).
Wright, J. (1977) 'San history and non-San historians', in 77ie Societies of Southern Africa in the
19th and 20th Centuries, vol. 8 (London).
Zahan, D. (1960) Sociétés d'initiation bambara: le N'domo; le Kore (Paris/The Hague. Mouton),
438 pp.
Glossary
abara, the (large) dugout canoes of Hausa, Kebbi and Nupe.

abba gada: in Oromo society (Ethiopia), the leader of an ethnic group, elected from among the
memben of the gada,
'abid: (sing, 'abd): is the ordinary word for 'slaves' in Arabic, more particularly for 'male slaves',
'female slaves' being irria' (sing. ama),
abilema: the earliest inhabitants of Unyakyusa (Lake Malawi region)
Ab Shaykh (title) (Arabic): a rank that was second only to that of the Sultan; a post that
commanded great authority.
abuna (title): the Patriarch ofthe Abyssinian Church; the Metropolitan ofthe Ethiopian Church.
abusuapanin. head, leader (of a (Akan) family, etc.).
achikunda: in the prazo system, armies of slaves, i e. the lower class, as opposed to the dominant
class or prazeros.
adamfo ('client state'): in greater Asante, a system of provincial administration in which each of
the states composing the confederacy served the Asantehene through one of the kings or
member-states of the confederacy or one of the wing-chiefs of the Kumasi state, usually
resident in Kumasi.
adanudo: a rich Ewe cloth (in Ghana)

adinkra: broad cloth with Akan traditional motifs and symbols stamped on it.
agha (title): commander of the Janissanes (in Egypt).
aid al-usul (Arabic): memben ofthe old lineage and rank.
ajami- transcnption in Arabic of African languages (e.g. the Hausa ajami manuscript).
akuaba. small wooden or terracotta maternity dolls. Still to be found in Ghana.
alfa: title borne by the chiefs ofthe provinces (diwe) in the confederation of Futa Jallon.
alifa: title of an officer (in Kanem).
almadies: boats (m Senegambia).
almamia land tenure and taxation systems (in Futa Jallon, Futa Toro and the Sokoto caliphate).
almamy: (title in Bundu, Futa Jallon and Futa Toro): a Fulfulde venion ofthe title imâm.
amenokal (title): The Tuareg equivalent ofthe Muslim imäm and kädi.
amir (Arabic): title given to generals, commanden, provincial rulen and sometimes to the
sovereigns of small countnes (spelt 'emir' in the West).
amir al-mu'minm (Arabic): caliphal title meaning 'prince, Commander ofthe Faithful'.
andrianoni: Antemoro title carried by the fint ruler ofthe Onilahy kingdom in Madagascar.
ant-ankara: 'people ofthe rocks', i.e. the Antankara of Madagascar's far north.
apagyafie: a group of goldsmiths and craftsmen from Denkyira.
ardo or rugga: the Fulbe equivalent ofthe Muslim imäm and kädi.
461
462 Glossary
'arsh (Arabic): ethnic area; community or group holdings; communal land.

asago: an Akan warrior.
asantehene: title ofthe paramount king of Asante.
asiento or assiento (Spanish asiento, seat, meeting place of a tnbunal, treaty, contract, from asentar,
to seat, make an agreement): a contract or convention between Spain and another power or
company or individual for furnishing African slaves for the Spanish dominions in America.
askiya: royal title adopted by the Songhay sovereigns to distinguish themselves from the
preceding Sonni dynasty.
asomfo: a golden stool ofthe Asante.
attajirai (or masu arziki): rich traden; the wealthy (in Hausa society).
azalai or azalay: a term for the great caravans made up of several thousand camels (or to be more
precise, dromedaries), which in the spnng and autumn carry salt from the salt deposits ofthe
Southern Sahara to the tropical regions of the Sahel and the Sudan. The salt deposits of
Taoudéni have replaced those of Taghäza, a source of wealth ofthe kings of Mali and of Gao
(fourteenth-fifteenth centuries).
azel: fiefs.
babika or bampika (sing, mubika): the slaves, one ofthe strata in Kongo society.
babuta (sing, mubata, from vata or evata, 'village'): the rustics, one ofthe strata in Kongo society.
baganwa: title ofthe princes of royal blood (in Burundi).
bailar nagaí ('sea-king'): the governor ofthe northernmost province ofthe Ethiopian empire.
bahaushe: trader (in Hausaland)
bahutu: see muhutu.
bairu: ennobled (clans).

bakama: see mugaba.
bakungu: chiefs (in Burundi and Rwanda).
(the) Bambudye: the closed association which helped the Luba king to rule.
bami: see mwami.
bantamahene (title): one ofthe wing-chiefs ofthe Kumasi state, vassals ofthe Asantehene.
baraka (Arabic): 'divine favour', 'gift of working miracles', 'charisma'; a blessing dispensed by
marabouts and especially by descendants ofthe Prophet Muhammad (shurafi').
barbariores barbari: savages (Ethiopia).
bares: mines in Northern Zambezia region.
Basace (title in Zamfara, Hausaland): an elder holding an ancient, long-forgotten office, member
of a Tara.
(the) bashiya (Arabic) or pashists: the followen of 'Air Pasha.

bawa. captive (in Hausaland).
bawan Sarki: royal slave (in Hausaland).
bawdi peyya yiyan: blood drums (in Futa Toro).
Bayt al-mäl: the (public) Treasury, in Arabic.
beecio: a title (in Waalo).
bega (Mossi): a great ceremonial cycle which involved the king and all the dignitaries ofthe land
in sacnfices intended to ensure a good harvest.
benangatoobe: in the caste system in TakrOr, the class involving the sakkeebe, wayilbe znd gawlo.
bey (gentleman, chief, prince in Turkish): 1) the governor of a distnct or minor province in the
Ottoman Empire; 2) the sovereign vassal of the sultan (e.g. the bey of Tunis); 3) a high-
ranking official, senior officer, in the Ottoman empire; 4) the bearer of a courtesy title,
formerly used in Turkey and Egypt.
beylerbey (pi. beylerbeyi): the governor-general of a province, head of a sandjak-beylik.
beylik: the term beylik denotes both the title and post (or function) of a bey and the territory
Glossary 463
(domain, jurisdiction) under the rule of a bey. Later, by extension, it came to mean also 'state,
government' and, at the same time, a political and administrative entity sometimes possessing
a certain autonomy.
bid 'a (Arabic): innovations, i.e. traditional religious practices grafted on to Islam.
biläd al-Makhzen: in the Maghreb, a country under state authonty.
Biläd al-Südän: literally 'the country ofthe blacks' in Arabic; this expression is nearly always used
by Arab authon in reference to the blacks of West Africa, a very extensive zone comprising
not only the Senegal, Niger and Chad basins but also the more southern countries of savanna
and forest. The term 'Sudan' now attaches to a state in the Nile valley never so designated in
ancient writings.
bilbalse (Mossi): adult servants.
bilolo: see kilolo.
birr, silver, in Amharic (Ethiopia).

biton: Bambara title meaning 'the Commander'.
bolo ('arm', in Bambara): a division ofthe infantry in Samori Ture's army, formed of several sen.
boma (Swahili): an enclosure or stockade used for herding beasts and for defensive purposes.
bozales (Spanish): unlike the ladinos, the bozales were slaves brought directly from Africa and who
had had no previous contact with Europe.
brak: title in Waalo.
buch jodong gweng': council ofthe elden (among the Luo).

buch piny: council of Luo sub-ethnic groups.
bud-kasma: (Mossi) head of a budu.
budu: m the patrilineal and patnlocal Mossi society, this term denotes any descent group, from
the broadest and the oldest to the narrowest, that operates as the exogamous reference unit.
bugo (pi. buguba): (Mossi) fertility pnest.
buhake: a type of clientele link (in Rwanda).
bulopwe: the Luba concept of power (Luba royalty was founded on the principle of bulopwe or
'sacredness') .
Burrjullit ('Great Prayer Masters'): religious leaden, in Senegambia.
capitäo-mor (pi. capitaos-mores (Portuguese): in Shona country, a representative of merchants in a

feira, appointed by the Portuguese administration.
caracoes (shells, shellfish, cowries in Portuguese): small stone beads found in the Zambezia region.
ceddo (pi. sebbe): warlord, war chief (in Senegambia).
changamire: ruler, member of a dynasty which rose to power in Butua.
diät: see kät.
chediia: red woollen bonnet.
chidi: fiefs (in Bomo).

chikunda: in the prazo system, a slave army under the authority of a sadiikunda.
chikwange: a well-known bread among the Kongo, obtained from the cassava root.
chimas: fief-holden (in Borno).
chiroma (a Bomo title): heir apparent.
chuanga: in the prazo system, top slave.
ore perdue: lost wax.
colono (Portuguese): colonist, settler (in prazos).
comprador (from Portuguese comprar, to purchase, buy): 1) Formerly, the name of a native servant
employed by Europeans, in India and the Far East, to purchase necessaries and keep the
household accounts; a house-steward. 2) Now, in China, the name of the principal native
servant, employed in European establishments, and especially in houses of business, both as
head of the staff of native employees, and as intermediary between the house and its native
464 Glossary
customen. 3) One held to be an agent of foreign domination or exploitation.

creóle (from Spanish criollo, 'native to the locality, country'): a penon of European ancestry born in
the former colonies (West Indies, Guyana, Réunion); Creole: a language, born ofthe black
slave trade (sixteenth-nineteenth centuries), that has become the mother tongue of those
slaves' descendants (in the West Indies, Guyanas, Indian Ocean Islands, etc.). There exist
creóles based on French, English, Portuguese, etc.: the creóle of Guadeloupe, the English
creóles ofthe Caribbean, etc.
cruzado (or crusado) (pi. cruzados or cruzadoes) (from Portuguese cruzado, literally 'marked with a
cross'): an old gold coin of Portugal, originally issued by Alfonso V (1438-81) about the year
1457, having a cross on the revene in commemoration ofthe king's crusading struggle against
the Muslims of North Africa; also: a similar Portuguese coin in silver fint issued by John IV
(1640-56).
curva: known as 'customs' in lands under Portuguese influence, 'duty' in English; a tribute paid
by the Portuguese to the Mutapa rulers.
dabey: in Songhay, a village of slaves.

dadde yiyan: war songs; blood songs (in Futa Toro).
dady: in Sakalava country (Madagascar), the cult ofthe anceston; ancestral Volamena relics
darnel: title of a ruler (in Kayor).
damel-teen: title of a ruler (in Kayor).
danau (title in Zamfara, Hausaland), a governor who owed his title to the name ofthe town in
which he resided, an important trading centre from where he kept watch over the roads
leading south and west of Kebbi.
dandal ('the opening ofthe "U" '): the settlement's main street of most Kanuri U-shaped towns
and villages .
dan dubal (a title in Hausaland): in Zamfara, an advisor on religious affain, the custodian of
Zamfara 's history and responsible for praying for the army's victory.
dan kodo (a title in Hausaland): see dan dubal.
(the) Darb al-'Arba'ui ('40-Days' Road' in Arabic): the great trans-Saharan route that linked the
Darfur region to Egypt in about 40 days.
debe: a village of slaves (in Soninke).
debeere: a village of slaves (in Fulfulde).
dey (from Turkish dâyï, 'maternal uncle') (hononfic title): 1) a ruling official of the Ottoman
empire in northern Africa, especially the head ofthe Algien Regency from 1671 to 1830, i.e.
before the French conquest in 1830; 2) also designated a lower rank in the Janissary militia;
towards the end ofthe tenth/sixteenth century in Tunis, the name was born by the heads of
the 40 sections ofthe militia.
al-Dhahabi (from dhahab, pure gold in Arabic): 'the Golden': nickname of Ahmad al-Mansur, the
sixth sovereign ofthe Moroccan dynasty ofthe Sa'âdî.
diatigui: a title (in Timbuktu).
dimo (pi. rimbe): free man (in Futa Jallon).
disi ('chest', in Bambara): the centre of Segu's army when arrayed in battle order.
diwal (pi. diwe): a province (in Futa Jallon).
dlwan (Arabic): central state coffen or Treasury, Council of State; military pension roll; collection
of poetry; large hall or chamber; ministenal departments; ruler.
diwe: see diwal.
djalläb (Arabic): slave merchants, slave trafficken, called 'importers'. They acted as the middlemen
and the financien ofthe long-distance trade.
Djammä1 ('the gatherer' in Arabic): the nickname of 'Abdallah, leader ofthe 'Abdalläbi state.
djiliäd: literally 'effort' in Arabic; djihâd akbar. 'supreme effort'; 1) the struggle against one's
Glossary 465
passions; the greatest effort of which one is capable; an effort to attain a specific goal; 2)
penonal effort of the believer to serve the Islamic faith, particularly the struggle to defend
Islam, its land and the Muslims against the impious enemy, or to enlarge Islamic terntory (Dar
al-lsläm), by extension, 'holy war'.
al-Djinn ('the devil' in Arabic): the nickname ofthe Egyptian bey 'All Bey.
djuad (Arabic): wamor chiefs.
Dongo (divinity): in Hausaland, a Borgu hunter whose powen were identical to those of Shango
in the Yoruba culture.
doomi Buur. see garmi.

dukowo: territorial divisions (in Eweland).
dupi: the servile class (in Western Nilotic societies).
dwar. tented camp .
egusi. a local crop (in Igboland).

ekanda: in Kongo society, matnclan, matrilineage, but also ethnic group, republic: in short, any
community.
Ekpe: the Ekpe society of Calabar was an adaptation of a cult ofthe leopard (Mgbe) present among
many forest groups in the Cross nver valley and in parts ofthe Cameroon.
Ekpo: a secret society (among the Ibibio).
emin (from Arabic amhi, faithful, trustworthy): an Ottoman administrative title usually translated
'intendant' or 'commissioner', tax collector. His function or office was called emänet.
emuron: a seer (in Teso).
etem (pi. itemwan): in Teso and Kumam, multi-clan ritual grouping.
eze (Igbo title): king.
eze Nri (title meaning 'ruler', 'the king of Nri', among the Igbo): priest-king with ritual and
mystic, but not military power.
faa (Bambara): father of a biological community.

faama (Bambara title): lord holding supreme power, man of power and authority.
faaya: fathen (in Bambara).
fáltatelo: in Iboina (Madagascar), fint minister; royal official.
fañadir: cage (in Borno).
fandroana: annual royal-bath ceremony (in Merina, Madagascar).
fani: in the Comoro Islands, the fint Islamic chiefs who had originally succeeded the Bedja of pre
Islamic times.
farin arfaren: rulen of provinces, governon.

farinya (comes from Fan and pharaoh): 'ruler' in Soninke, Mande, etc. Farinya denotes a
monocracy as opposed to the controlled monarchical oligarchy ofthe original mansaya
jashir (Arabic): royal residence.
fataua: in Hausaland, trade over medium and long distances; wholesale trade.
fazenda: large estates in Brazil.
feiras: fain (in Portuguese).
fia: the king of a dukowo (in Eweland).
fidda diwänl (Arabic): fine silver coins.
filohany: chief (in Malagasy).
fitahina: a linguistic taboo (among the Sakalava of Madagascar).
fia-n-ton (or ton): peer association; the fla-n-ton brought together the memben of three successive
promotions of those circumcised.
folahazomanga (in Malagasy, 'those who came to Mahafaly from elsewhere'): a component of
Mahafaly society, Madagascar.
466 Glossary
foroba-dyon ('captives ofthe big common field') or furuba-dyon ('captives ofthe Great Union'):
crown slaves.
gabar. in Ethiopia, a peasant who worked the lands for the benefit of the landed élite and was,
like the fallait, similar to a serf, or at least a tributary or client obliged to pay the gabir,
depending on whether he was a share-cropper or a tenant farmer.
gabir or siso: a tax paid by the gabar.
gada or luba: a classification ofthe Oromo society of Ethiopia by age groups based on an abstract
numencal principle. It is an egalitarian system which involves rule by rotation age-groups.
gaisuwa or tsare: payment of regular gifts to superiors (in Hausaland).
galadima (a Borno title): a kind of prime minister or grand vizier in whom the conduct of all
affain of state was vested. The title was borrowed from Borno, but there it designated the
governor of the western provinces, that is, those nearest to Hausaland; in Katsina, an
important official who deputized for the sarki; in Kano, a military leader; in Sokoto, the
caliph's advisor, acting as the caliph's link with the emirates.
garmi or doomi Buur. the ruling class (in Kayor).
^afari ('axe'): in Katsina (Hausaland), title of a governor who watched over the north-western
frontier.
gayya: in Hausaland, mutual aid system; communal work.

gida (pi. gidaje): extended family; groups of families (in Hausaland).
gikingi: a land system granting exclusive grazing rights to the owner (in Rwanda).
gimaje: long dyed cotton dresses (in Kanun).
gobbi: an annual tax on mines.
grumetes: African auxiliaries recruited by the lançados.
gulma: title of a high-ranking dignitary among the Songhay.
gumbala: in Futa Toro, the war song ofthe sebbe, which consist of hymns to courage. It is the epic
song of death in which the ceddo assumes his destiny as a warrior, his faithfulness to his
anceston and to the ethic of his caste.
gumsu: the king's head wife (in Bomo).

gurma (Hausa): right (e.g. the furnia bank ofthe river, as opposed to hausa, the left bank).
gweng' (pi. gwenge): a Luo semi-autonomous political and terntorial unit.
Habshi: one of the terms used in India to denote those African communities whose anceston
originally came to the country as slaves, in most cases from the Horn of Africa, although some
doubtless sprang from the slave troops ofthe neighbouring countries. The majority, at least in
earlier times, may well have been Ethiopian, but the name was applied indiscriminately to all
Africans, and in the days of the Portuguese slave trade with India many such Habshi were in
fact ofthe Nilotic and Bantu peoples.
hadenda: large estates in Latin America.
hakura: tax exemption.
hanzo (from nso, 'house' in Kikongo): parcel (in Angola).
hayu: judges (among the Oromo of Ethiopia).
hazne or hazine: treasury.
heemraden: burghen (in Southern Africa).
himäye (Arabic): protection charges; caution money protecting the conquered townsfolk from
looting.
hore kosan: grazing land.
¡ko: power (in Hausaland).

iktâ' (Arabic): 1) the fiscal allocation of an emir, always strictly monitored and updated, ofthe
Glossary 467
revenue of one or more localities, according to status and the number of men in his service;
2) delegation of tax-collecting powen granted by the pnnce to a military or civilian officer in
regard to a fiscal district, as remuneration for a service rendered to the state; this concession
was revocable; 3) distribution of fiscal concessions for maintenance of the military class; 4)
military fief system.
'Urn (Arabic): religious knowledge.
iltizam (Arabic): according to the iltizam system which superceded the mukatcfat system by about
1658 lands of every village or group of villages were offered for public auction and the highest
bidden (the multazims) were given the right to collect taxes from the peasants and such lands
became their iltizam.
imäm (Arabic): honorary title awarded to the eminent legal experts who, between the
second/eighth and the third/ninth centuries, codified the whole body of Muslim law in
various intellectual centres of the Muslim world, particularly Medina and Baghdad; a title
given to the founden of law schools and to major theologians; chiefs, supreme leader ofthe
Muslim community; among the Shî'ites. equivalent of a caliph (must be a descendant of 'All).
imamate: the leadenhip ofthe Muslim community; caliphate or inherited power; office or rank of
an imâm; function ofthe imam as head ofthe Muslim community; the region or country ruled
over by an imäm (e.g. the ¡mámate of Yemen).
indabo: a village of slaves (in Hausaland).
innamme: in Kebbi (Hausaland), title of a governor who kept watch over the western frontier.
insilimen: see zuwaya.
inyamasango: a village head in Shona country.
iyäla (eyälet, in Turkish) ('management, administration, exercise of power'): in the Ottoman
empire the largest administrative division under a beylerbey or governor-general. In this sense
it was officially used after 1 000/1591.
jaggorde: the council of electon (in Futa Toro).

jagos: territonal sub-heads (among the Luo).
jangali: the livestock tax paid by stockbreeden (in Hausaland).
Jengu: the most prestigious secret society among the Duala, Isuwu and other neighbouring
groups, based on the veneration of water spints.
jibda: musk (in Hausaland).
jon: captive (in Mande).
jonya (from the Mande word jon meaning 'captive'): black African captive.
jumbe: in the Comoro Islands, this term denoted the daughten of the fani, who came to marry
the Sunnite newcomen.
jurungul ('crown'): a special coiffure worn by a Kanuri married woman.
kabaka: ruler, king (in Buganda).

kablla (Arabic pi. kabä'il) a large agnatic Arab or Berber group the members of which claim to be
descended from one common ancestor and who may jointly own an area of grazing land'.
kabunga: chief (in Kongo).
kädi (Arabic): among Muslims, a magistrate concurrently discharging civil, judicial and religious
functions in accordance with Muslim law {shañ'a).
kafu or jamaría (in Mande): provinces; small state-like territorial units (among the Malinke and the
Bambara).
kâ'id (Arabic): commander, army chief; provincial governor; kablla chieftain.
kaigama (Bomo title): governor ofthe northern provinces (in Kano).
kakaki: a long trumpet (in Hausaland).
kalaram: turban (in Kanun).
468 Glossary
kalonga: ruler in the Maravi state.

kangam: a provincial chief in Senegambia.
känün or ¡känün (from the Greek kanôn, rule): the corpus of legal regulations (canon law).
Känün Name: a decree which sought to regulate political, military, civil and economic life in
Egypt under the Ottoman administration.
kasa: territory (in Hausaland).
kasaba or gasba (Arabic) (casbah/kasbah in English): capital; (small) town; citadel, castle or fortress
(in North Africa).
käshif (Arabic): head of district (in Egypt) whose task it was to maintain the írngation system and
levy taxes from the farmers; tax collector.
kaslda: an Arabic or Penian panegyric, elegiac or satirical poem or ode, usually having a tripartite
structure. The term is derived from the root kasada, 'to aim at', for the pnmitive kaslda was
intended to eulogize the kabila of the poet and denigrate the opposing kabllas. Later it was
concerned with the eulogy of patrons.
kasuwanci: retail trade (in Hausaland).
kät: (in Arabic) and chat (in Amharic): a shrub (Catha edulis F.) native to Arabia, with mildly
stimulating properties, grown in Ethiopia and Yemen and exported under the name of kat or
chat; also the narcotic drug obtained from the leaves of this plant.
kaura (title): a military leader (in Katsina, Hausaland).
kaya (pi. makaya)- a fortified village; an enclosed or fortified settlement (among the Miji-Kenda
people of East Africa).
kazembe: a general (among the Lunda).
kente (also Kente): colourful kente clothes (ofthe Akan); in Ghana, a banded material; also, a long
garment made from this material, loosely draped on or worn around the shoulden and waist.
ketema: garrison towns.
khalif or khalifa (a title) (Arabic): caliph, successor of the Prophet, sovereign responsible for
ensuring observance of Islamic law on earth.
khammâsat (Arabic): fifth share-cropping ofthe land, tenant farming.
kharädj (Arabic): a land tax, sometimes paid in kind (and in addition to the cash poll tax or djizya)
on land belonging to the dhimml (non-Muslims living in Muslim territory under the status of
'protégés of Islam'); by extension designates all land taxes (see also raid),
khazin (Arabic): treasurer.
klnmba (an association in Kongo): an initiation cult for boys.
khutba (Arabic): address delivered by the khatlb (preacher), from the top ofthe minbar (mosque
chair) ofthe Great Mosque, during the noontime Friday prayer, in which God's favour was
invoked on the recognized caliph in the city and, where applicable, the prince from whom
the governor of the town held his delegated power.
kiims'rogo: a Mossi ancestral shrine.
kilolo (pi. bilolo): territorial administrators, chiefs (among the Luba/Lunda).
kimpasi (a healing association in Kongo): a cult whose aim was to banish mpasi.
kitomi: in Kongo, former lords ofthe land.
Kokani (a title in Kebbi, Hausaland): the official responsible for relations with the population.
kom ngu: the counscllon ofthe kingdom (in Bamum).
kubandwa: a practice of initiatory rites in Rwanda, Burundi and Buha.
kudin haki: grazing duty paid by the nomadic Fulbe (in Kebbi, Hausaland).
kudin hito: customs duty (in Hausaland).
kudin kasa: a land tax paid by the farmen (in Hausaland).
kudin iaifv forgiveness tax paid by an offender when pardoned by the sovereign (in Hausaland).
kudin sana'a: a professional tax (in Hausaland).
kultingo: a tribute (in Borno).
Glossary 469
Kulughli (from Turkish kuloghlu, 'son of slave'): in the period of Turkish domination in Algeria
and Tunisia, this word denoted those elements of the population resulting from marriages of
Turks with local women.
Kunduda (title in Hausaland): a Kebbi's military leader.

kurita: a Mossi word which means 'reigning dead man', is constructed by analogy with narita,
'reigning chief. The kurita is the representative among the living of a dead chief, and is
generally chosen from among the sons ofthe dead chief, he has no power by virtue of his title
and is debarred from the succession, but he can become a chief in an area outside his family's
command; if a kurita becomes a chief, he retains the nom-de-guerre (zab yuré) oí naaba kurita.
(the) Kurkwâ corps (spearmen in Für): a standing slave-army created by Muhammad Tayräb ibn
Ahmad Bukr.
kutama: migration movement (from south-east Africa to Zimbabwe around iooo).

kwaka tree or monkey orange: 1) either of two deciduous African shrubs or small trees (Strychnos
inocua and Strychnos spinosa) having a hard globose fruit with edible pulp; 2) the fruit of a
monkey orange.
kwara a magical symbol (in Mossi).
kyungu: ruler in the Ngonde kingdom.
ladinos (Spanish) the first Africans in America, brought there from Europe by the
Conquistadores. They were mainly from Senegambia and had either been previously brought
to Europe, or were born there. They were called ladinos in America because they knew
Spanish or Portuguese, and were at least partly influenced by the civilization of the Ibenan
countries.
lamana (from the Sereer 'master ofthe land'): denotes a system of land tenure and a political
system in which the masters of the land also exercised power. The lamana of the masters of
the land developed into the Mansaya of purely political chiefs.
laucados (from Portuguese lançar, to launch out an adventure) or tangomäos ('people who had
adopted local customs'): emigrants who had settled on the (African) continent with the
agreement of the African sovereigns, intermarried and set themselves up as commercial
middlemen. They formed part ofthe many expatriates who populated the southern riven and
the Atlantic islands (especially Cape Verde). Most of them were Portuguese, with a sprinkling
of Greeks, Spaniards and even Indians. They also came to include more and more halfbreed
children, the filhos da terra (children ofthe land).
landdrost: a resident magistrate (in southern Africa).
¡apto (literally, 'translator'): a native interpreter.
¡arde kangema: the nominal owner ofthe land (in Borno).
las li: the descendants ofthe great marabout families who constituted 'the aristocracy ofthe sword
and the lance and the book and the pen' at the top ofthe rimbe (in Futa Jallon).
legha (Arabic): henchmen.
Lenngi: in Futa Toro, the Lenngi, sung solely by sebbe women at marriages or circumcisions, are
heroic songs summoning up contempt for death and the protection of honour.
lifidi (Hausa): a quilted armour; horse trappings; the quilted protection for war-hones.
Lifidi (title in Hausaland): Commander-in-chief of the heavy cavalry division.
Limanin Ciki (tide m Hausaland): a Kanuri scholar responsible for the education of the royal
family.
Linger, a title (in Kayor).
lizma (Arabic): in the Maghreb, farming out ofthe provinces, customs, etc.; tax-farming.
lundu: powerful ruler in the Maravi confederation in the seventeenth century.
maccube: slaves, captives (in Fulfulde).

470 Glossary
machila: cotton clothes produced by the Lundu's Mang'anja subjects in the Lower Shire valley.
(the) Madjelisse (Arabic): a body ofjurists.
madrasa (Arabic): a primary and secondary Kur'anic school: madrasa designates more particularly
an establishment of higher religious education (as a rule attached to a mosque) for the training
ofthe Sunnite clergy, the 'ulamä*.
magaji: a warrior; a successor (in Hausaland).
magira: the queen-mother (in Känem-Bomo).
magram: the king's official sister (in Känem-Borno).
(the) Maguzawa: there are non-Muslim groups in both Nigena and Niger who speak only Hausa
and who share the Hausa culture, but who refuse to be called Hausa people. In Nigeria these
people call themselves, and are called by other Hausa, Maguzawa (or Bamaguje), whereas in
Niger they are known by the name Azna (or Arna), the Hausa word for 'pagan'. Since the
name Maguzawa is probably derived from the Arabic madjüs (originally 'fire-wonhippen',
then 'pagans' generally) it is possible that the polanzation between Hausa and Maguzawa/
Azna began only with the spread of Islam among the common people in Hausaland, after the
seventeenth and eighteen centuries.
mah'alla (Arabic): in the Maghnb, an armed expedition; armed camps.
mahanga (llala palm Hyphaene natalenis): a tree found in the lowveld areas of Southern Zambezia
the sap of which was used to make an intoxicating wine called njemani or chemwa (in
Hlengwe).
mdtiram (Arabic): the granting- by chiefs - of privileges to families or religious notables.
mai: king, ruler (in Känembu).
maidugu: grandson of a previous king (in Borno).
(the) mailo system: the system of land tenure introduced into East Africa under British coloniza¬
tion; it combines chiefdom with ownenhip and the nght to speculate in land.
mairia: eligible pnnces (in Borno).
mainin kinendi (title in Borno): the Islamic and legal advisor to the mai.
mai sarauta: ruler (in Hausaland) .
majlis: in Känem, the highest council of state, normally presided over by the mai and composed
of military and religious notables.
makambala: the councillon ofthe Kyungu (a priest-king wonhipped by the Ngonde).
makaya: see kaya.
makhzen (Arabic): privileged peoples from whom Moroccan state officials are recruited.
(the) Makhzen: originally meant 'treasury', but came to be used for the official system of govern¬
ment in Morocco, and more broadly the political and religious élite ofthe country.
makoko: king, among the Tio of Kongo.
mala kasuube (a title in Borno): the official who supervised sales in the markets and attempted to
ensure justice and fair play during major commercial transactions.
madams: Muslim clerics and scholan.
mallamai: scholan (in Hausaland).

maloango: king (in Central Africa: Angola, etc.).
mambo: a title found in south-east/central Africa.
mamlük: a freed man, a former slave of Christian ongin who has been converted and suitably
trained to serve at court or in the army.
(the) Mamluks: a dynasty that reigned over Egypt and Syna (1250-1517), whose sultans were
chosen from among the militias of slave soldien (mamlüks).
Manantany: the fint minister, in Sakalava kingdoms (Madagascar).
mandi-mani ('the Lord of the Manou'): the title of the greatest suzerain in the Sierra Leone-
Libena region.
mandi-mansa: the title ofthe Emperor of Mali.
Glossary 47 1
mandjd or mandjuluk: the title which Fundj monarchs bestowed on their principal vassals.
Mangi Mrwe: the paramount chief of the Ugweno clans of Tanzania.
mani: in the Kongo kingdom, a high-ranking noble; a landlord; a governor.
mani kongo: the ruler ofthe Kongo.
manoma: farmen in (Hausaland).
mansa (in Maninka): the king, the holder ofthe most important political power (in Mali).
mansaya (in Maninka): 1) royalty; political territorial groups headed by a mansa; 2) a socio¬
political system, whose dominant ruling class was a polyarchy of an élite of laymen or pnests,
freemen or slaves, caste or guild memben, or noblemen or commoncn. It was financed by
the taxes which those controlling the machinery of government levied on trade and produce.
It was not a landed aristocracy or propnetor class.
mantse: king, among the Ga.
marabout: the word does not have the same meaning in the Maghreb as in black Africa. In the
former it applies both to a holy pcnon who has founded a brotherhood and to his tomb; in
sub-Saharan Africa it designates any penon with some knowledge of the Kur'an and other
sacred wntmgs who uses that knowledge to act as interceder between the believer and God,
while drawing upon traditional divinatory sources and the use of talismans. In the eyes ofthe
public he is a scholar in the religious sense ofthe word, a magician, a soothsayer and a healer.
marabtin bilbaraka: in Barka, the offspring of pilgnms, usually North Africans.
marisa (title meaning 'the destroyer'): in Katsina (Hausaland), the title of a governor who kept
watch over the eastern border.
maroserana ('many paths' in Malagasy): reflects the custom of the fint Maroserana kings in
Mahafaly of placing their residences in the middle of habitations with paths radiating in all
directions to villages around them.
morula or nkanyi: the marulu (Sclerocarya caffra) is a tree of the family Anacardiaceae, found in
central and southern Africa, and beanng an oval yellow fruit about 2 inches long that is used
locally for making an intoxicating beverage; also marula plum: the fruit of this tree.
masu sarauta: in Hausaland, all those with any political authority; aristocrats.
mbafu: red dyewood, in Luba (Zaire).
mbarma or búlala: local or ethnic leaden (in Känem).
mboma (a man ofthe Boma ethnic group): among the Tio, the word designates a stupid penon or
anyone in a lowly occupation.
mbua: animal skin (in Kikuyuland).
meen: matemal family (in Kayor).
mercadores (Portuguese): broken; merchants, tradesmen.
mestizo: a Spanish or Portuguese half-caste (Afro- Portuguese); also applied to other penons of
mixed blood, such as Afro-Asians.
(the) Mfecane (crushing), in Nguni languages and Lifaqanel Difaqane (hammenng) in Sotho-
Tswana: a social and political revolution that took place in Bantu-speaking southern Africa
and beyond, during the fint decades ofthe nineteenth century.
Mgbe: leopard secret society (ofthe northern Cross river valley and the Cameroon).
mhondoro ('lion'): in the Mutapa empire, spirit mediums whose function was to advise the
emperor in all matten of state national ancestor spints.
mifaly: submission (in Malagasy).
milambu: taxes payable (to the Luba mulopwe) in food and local produce.
milk: small family holdings; property; possession, ownenhip.
mingilu: unpaid labour (a tribute paid to the Luba mulopwe by his officials).
mithkal (of gold, etc.) (Arabic): the Sudanese mithkal weighs approximately 4.25 grams.
mitngu: a secret society opened to the rest of the population irrespective of social status (in
Bamum).
472 Glossary
mlira (the mlira cult): the ritual veneration of Mlira, the spirit ofthe Great Kalonga Chinkhole, an
ancestor ofthe Phiri royal lineages (see chapter 21).
moasy: a pnest (in Madagascar).
mogho-naaba (a title): the Naaba ofthe Mossi country.
mogyemogye ('ceremonial jaw-bone'): an Asante wine jar used for pouring libations on the Golden
Stool.
Monomotapa: see mwene mutapa.

montamba: among the Bobangi ofthe Congo basin, a slave sold by his kin.
montonge: a captured slave (among the Bobangi).
moyo: heart, m the Rozvi empire; soul, spint in Kikongo.
mpasi (Kikongo): suffenng, poverty, need, calamity or affliction.
mphande: shells, among the Tumbuka of northern Zambezia.
mpok: a Lunda broadsword.
m'polio: a catfish (in Bambara).
mpungi: the sacred shrine ofthe Pare of Tanzania.
mubata: see babuta.
mubika: see babika.
mugabe ('the milker') or mukama (pi. bakama): king (in Bunyoro and Buganda).
muhanuro: an annual festival in Burundi, during which royal authonty and the drums symbolizing
it were renewed, and which celebrated the sowing of sorghum, and determined the most
favourable date for this in a country with a long rainy season.
muhutu (pi. bahutu): ethnic group in Burundi, Rwanda and several other states in eastern and
central Africa.
muidzu: the country's chicfjustice (in Kongo).

mukama: see mugabe.
mukomondera: a cultivation system through which the Shona used to obtain grain dunng famine.
mulatto (Spanish mulato, from mulo, mule): one who is the offspring of a European and a black;
also used loosely for anyone of mixed race resembling a mulatto,
mulopwe: the title ofthe Luba king which signified the indivisibility of power that could not be
shared.
multazims (Arabic): the bidden who collected taxes from the peasants.
munaki: a prophetess (in Kongo).
mupeto: in the Mutapa empire, forcible confiscation.
muteferrikas: the penonal guard ofthe vice-regent of Egypt; the penonal guard of a sultan.
muud al-hiirum or muudul honna: in the Senegal Futa, an annual tribute payable in grain (grain tax)
to the Moon.
mvila: clans (in Kikongo).

mwami (pi. bami): the royal title ofthe former kings of Rwanda and Burundi.
mwant or mwaant: the chiefly title to which the name of the fint Luba king (Yav or Yaav) was
added to become the imperial title.
mwant yav: tbe highest Luba/Lunda title meaning 'Lord Yav', 'Lord ofthe Viper', referring to the
distance between king and mortals: emperor.
mwene mutapa or monomatapa ('lord of metals', 'master of pillage'): a title borne by a line of kings
who ruled over a country rich in gold, copper and iron, whence the title 'lord of metals'.
This country lay between the Zambezi and Limpopo riven, in what is now Zimbabwe and
Mozambique, from the fourteenth to the seventeenth century. It was often called the empire
of the Mwene Mutapa, and is associated with the site known as Great Zimbabwe in south¬
western Zimbabwe.
mwine Munza: a Luba title meaning 'master of Munza'.

mwisikongo: the nobles, one ofthe strata m Kongo society.
Glossary 473
naaba (a Mossi title): chief.

naabiiga (pi. naabiise): Mossi chiefs son.
naabiise: see naabiiga.
nälb (literally 'substitute, delegate, deputy'): the term applied generally to any penon appointed
as deputy of another in an official position, and more especially, in the Marnluk and Dihli
sultanates, to designate: 1) the deputy or lieutenant ofthe sultan and 2) the governon ofthe
chief provinces. In its most common form, in Penian and Turkish as well as later Arabic, rúñb
signified a judge substitute, or delegate of the kädi in the administration of law.
nakombga: see nakombse.
nakombse (sing, nakombga) (Mossi): pnnces ofthe blood; sons or descendants of sons of pnnces
who had not become chiefs, the royal nakombse = the royal lineage.
nangatoobe: the upper caste (in Takrür).
napusum: Mossi ceremonies of greeting the king.
nawab: king.
ndoki (pi. bandoki): worker of spells, sorcerer (in Kongo).
ndulu: a Kikuyu green vegetable.
neftenia- landed élite, landed nobility.
nere: an African tree whose roots and seeds are used in traditional medicine.
ngaadtino (Tio): in Kongo, the official responsible for collecting tribute and heading the
administration.
nganga: in Bantu languages, healer; sorcerer; medicine man; in Kongo, the recognized term for a
religious expert, especially in nkisi.
nganga ngombo: diviners (in Kongo).
ngiri: a secret society for pnnces only (in Bamum).
ngola: a title born by the king of Ndongo, a state to the south of Kongo.
njoldi: symbolic payments attaching to the master ofthe land; annual ground rent.
nkangi ('the saviour'): in Kongo, a crucifix which was the emblem of the judicial power.
nkende: a grey cercopithecus (a genus of long-tailed African monkeys compnsing the guenons
and related forms) used as a taboo by the kings of Bujiji.
nkisi: the term, which means 'initiation; magic power; mystic powen; spint; ancestral force;
sacred medicine; idol; fetish', designates, in Kongo, the ideology of royalty derived from
religious conceptions in general, in which three important cults played a role: ancestor
wonhip, the wonhip of terntory spints and the wonhip of royal charms.
nkobi: a talisman (in Kikongo).
nkoron: deep-mining (in Akan).
nkuluntu ('old one'): the elder; hereditary village chiefs (in Kongo).
noguna: the mats court (in Borno).
nono (a title): in Kebbi (Hausaland), the dignitary who collected the milk and butter intended for
the sovereign.
nthlava: the sandy soils where the Hlengwe used to build their homesteads.
ntufia: the sacred fire lit at his coronation by each king (maloango) of Loango, and which was to
bum until his death.
nyamankala: the caste system (in Mande).

Nyarubanga (the Luo word for 'sent by god'): in northern Uganda, the period of droughts and
famines that ended with the great famine of 1617-21, accompanied by a disease which wiped
out the cattle herds.
nzambi mpungu ('supreme creator', 'superior spirit'): the way the king of Kongo was addressed.
nzimbu (shells of Olivancilaria nana): shells used as currency in the kingdom of Kongo.
nzo longo (association in Kongo): an initiation cult for boys.
474 Glossary
oba (in Edo): the title ofthe founder ofthe ancient Kingdom of Benin, holder of religious and
political powen.
obeah (or obi): an African religion, probably of Asante origin, characterized by the use of sorcery
and magic ntual; also: a charm or fetish used in obeah and the influence of obeah (e.g. put obeah
on a penon).
obi: an Igbo term, of probable Yoruba origin, for king or chief. The obi was appointed by the oba
of Benin.
odjak: a corps of impenal troops introduced in Egypt by Sultan Selim I; also Turkish fortresses
and garrisons.
ohene: the king ofthe Kumasi state.
ókoo (makoko): king (among the Tio).
okro or okra: a tall annual plant (Hibiseus or Abelmosdius esculentus) indigenous to Africa, and its
green seed-pods, used for soup, salad and pickles (also called gumbo),
omanhene: the king ofthe Akwamu and Denkyira empires.
ombiasa: often translated as 'sacerdotal penon'; pnest; medicine man; doctor, etc.; ombiasa encom¬
pass a number of functions and have a number of categories into which they are subdivided
among the Malagasy.
omukama: king.
oni: king (e.g. the oni of Ife).
oranmiyan: title of Oranyan, a son of Oduduwa (the founder and fint oni of Ife) and the legendary
founder ofthe Yoruba kingdom of Oyo, who is stated to have reigned at both Ife and Benin
before moving to Qyo; now the title of a god.
orkoiyot (pi. orkoik): a spiritual leader (among the Nandi); also, a traditional leader (in Kenya).
osafohene: war leader in the Akwamu and Denkyira states.
ovimbali or quimbares: the Afro-Portuguese (in Kongo).
ozo: a high-ranking penon (in the Igbo hierarchy).
panga (Swahili): a large broad-bladed knife used in Africa for heavy cutting (as of brush or
bananas) and also as a weapon (machete).
pasha (Turkish): honorary title attaching to senior office, particularly of military commander and
provincial governor, in the Ottoman empire.
pastas (Portuguese): 1,000 pastas = 800,000 cruzados (c. 1756-57) - see cruzado,
piny ( pi. pinje): the territory of a Luo sub-ethnic group,
(the royal) pogsyure (napogsyure) system: a system of capitalizing and distnbuting women (among
the Mossi).
pombe (Swahili): in Central and East Africa, a (possibly intoxicating) drink made by fermenting
many kinds of grain and some fruits.
pombeiros (in Portuguese): agents from pombo; barefoot mulatto merchants.
prazeros: in the prazo system, a dominant class, holder ofthe crown lands (prazos); estate-ownen.
prazos (Portuguese): crown lands; agricultural estates; the prazo was also a Portuguese land-tenure
system.
prazos da coroa (Portuguese): crown land estates.
presidios (Portuguese): towns.
quimbares: see ovimbali.
raia (from Arabic ra'iyah, flock, herd, subjects, peasants): a non-Muslim subject of the Sultan of
Turkey, subject to payment of such taxes as the djizya (poll tax) and the kharädj (land tax).
(See the dhimml of Vol. III).
ra'is (or reis): a Muslim chief or ruler, a Muslim ship's captain.
Glossary 475
rakire: joking relationship (in Mossi).

Ramadan: the ninth month of the Muslim lunar calendar, dunng which Muslims observe a fast
(saum).
ras (Amhanc ras, head, chief, from Arabic ra'ls) (a title): an Ethiopian king, prince, or feudal lord
also, the ruler of an Ethiopian province.
al-Rashld (Arabic): an honorific title meaning 'the Just',
(the) Reconquista: the process of Chnstian resistance to Muslim domination and the wan to
eliminate Islam from the peninsula, covenng the penod between 722 (the battle of
Covadonga) and 1492 (the fall of Granada).
renilemy: the privileged, in Mahafaly society (Madagascar).
reth: king, in Shillukland.
rima (a Mossi title - from ri, 'to absorb or eat food that is not chewed'): king.
rimbe: see dimo.
ringu (from ri - see above, under rima): enthronement.

riyäsa (Arabic): leadenhip.
rohandrian: the name ofthe highest Zafindraminia estate in Madagascar.
rukonkesh or lukonkeshia (Luba/Lunda title): the queen-mother and head ofthe logistics for the
court.
runde: slave villages (in Futa Jallon).
saburu (a title in Hausaland): the official responsible for security on the roads.
sqff(p\. sufuf) (Arabic): a confederation formed by alliances between kabilas.
säliyäne (Arabic): annual salaries.
sampy: the royal divinity guaranteeing the well-being ofthe Imerina state (Madagascar).
sandjak (from Turkish sanjäk, literally, flag, standard): in the former Turkish empire, one ofthe
administrative districts into which an eyalet or vilayet (province) was divided.
sandjak bey (a title, in Egypt): the governor of a sandjak; the highest military and administrative
penonnel of a district.
sandjak-beylik: the most important and fundamental military and administrative unit of the
Ottoman empire grouped, regionally, under the authority of a beylerbey.
san'dyon: trade slaves (in Bambara).
sango: the thunder cult (in the Qyo religious system).
santu (Kikongo): a type of cross; a crucifix, but also a saint.
sarakunan noma: a title meaning 'master ofthe crops' (in Hausaland).
sarauta: kingship (in Hausaland).
sarki (pi. sarakuna) (a title): head of state; chief; king (in Hausaland).
sarkin (a title): chief; king (in Hausaland).
sarkin lifidi (a title in Hausaland): a general in the heavy cavalry division.
satigi (a title in Futa Toro, meaning 'the great ful'): ruler.
sebbe: see ceddo.
serdar. the commander ofthe military operating within or beyond Egypt's borden.
serin jakk (a title in Kayor): the serin jakk devoted themselves to religious activities and teaching.
serin lamb: agents of central government (in Kayor) responsible for the defence ofthe frontien.
shangwa: famine (in Shona).
sharfa (literally 'way', 'good road' in Arabic): detailed code of conduct; the sharl'a comprises the
precepts governing the ritual of wonhip, standards of behaviour and rules of living. It consists
of laws that prescnbe, permit and distinguish between true and false. The Kur'anic
prescriptions it coven are complemented by prohibitions and explanations contained in the
law (fikh). The sources ofthe Islamic sharî'azre the Kur'än and the hadith.
sliärif (Arabic pi. ¡Jiurafi'): literally 'noble'; an honorary title given to all the descendants of 'Ali
476 Glossary
and of Fätima
shaykh (Arabic pi. mashäyikh): old man; chief of an Arab kablla; spiritual master; title given to the
founden of mystical brotherhoods, to great scholan and to teachen.
Shaykh al-Balad (Arabic): an unofficial honorary title, only indicating a senior among the
Egyptian beys (senior grandee).
slba (Arabic): dissident blocks
siddi one ofthe terms used to denote blacks or African slaves in Asia.
soba: petty chief, leaders (in Angola).

sombili: (ongoing) custom; prerogative (practised by the Zafirambo in Madagascar) of slaughtering
domesticated animals.
sonda (sing, sondre) (Mossi): collective names or mottoes

sorabe (or volan'Onjatsy) (in Malagasy): Arabic script used to transcribe the Malagasy language;
Arabic Malagasy characters; manuscnpt in the Antemoro language written in Arabic
charactcn. These are katibo's (scribes specializing in the writing and interpretation of the
sorabe) traditions.
süfi. literally 'dressed in wool', from the Arabic root, süf, meaning 'wool', to denote 'the practice
of wearing the woollen robe' (labs a-süf), hence the act of devoting oneself to the mystic life
on becoming what is called in Islam a süfi: an adept of Muslim mysticism (Süfism or tasäwwuf)
süfi tañkas: see tañka.
Sufism (tasawwuf in Arabic), ascetic Islamic mysticism originating in the eighth century and
developing especially in Penia into a system of elaborate symbolism of which the goal is
communion with the deity through contemplation and ecstasy.
sugedi: grass matting (in Kanuri).
suk. an Arab market or market-place.
sük: an Arab market or market-place, a souk, a bazaar; shops grouped by corporations.
sulke: coats of mail (in Hausaland).
sultan the sovereign or chief ruler of a Muslim country; especially (formerly) the sovereign of
Turkey; also formerly, a prince or king's son, a high officer.
swan mulopwe: heir-presumptive and commander of the Lunda army.
swan murund or swana mulunda (a Luba/Lunda title): the symbolic mother ofthe society; symbolic
queen-mother named Rweej.
tafslr (Arabic): commentary ofthe Kur'an; exegesis.

talakawa (sing talaka): the ruled; in Hausa society, serfs, free commoners; the poor; poverty-
stncken farmen and herdsmen of Hausaland and the Niger-Chad area
talba: head of police and magistrate (in Borno).
talse: Mossi commoners,
(ara (a Hausa title meaning 'the Council of nine'): in Zamfara, Gobir and Kano, designated an
electoral college responsible for the appointment of the successor to the throne from among
the princes.
tarlka (Arabic pi. turuk): literally 'way'; association or brotherhood (each tañka bean the name of
its founder); congregation, Süfi religious brotherhood; local centre of a religious brotherhood;
lodge of brotherhoods.
tárlkh (Arabic): history in general, annals, chronicles; usually synonymous with 'historical
account'. It is the title of a great many historical works, like Ta'rikh al-Sûdân (the history of
the Sudan, or Negroes of West Africa - see Biläd al-Sûdân); Ta'rikh al-Andalus (the history of
Andalusia), etc.
tasobnamba ('mästen of war' in Mossi): war chiefs.
tata: a fort (in Bambara).
Teen: a title (in Kayor).
Glossary 477
tengsoba ('the master ofthe land' in Mossi): earth priest.

tenzi: epic poems (in Kiswahili).
terras da coroa: in the prazo system, a Portuguese crown property.
tingimaro: title of a ruler (in Madagascar).
tobe (Kanuri): large open cotton gowns either plain or dyed blue.
tompon-menakely (a Malagasy title meaning 'masten-of-the-fief): chiefs of village fiefs.
tompon-tany ('mästen ofthe land'): original inhabitants (in Madagascar).
ton (Bambara): an association of boys circumcised at the same time, political association.
ton-den (Bambara): association memben.
ton-dyon ('captives ofthe ton , in Bambara) mostly captives or former captives taken in war.
tonjon: crown slaves (at the mansa's court).
too: millet paste that forms the staple diet among the Bambara.
too-daga: too pot (in Bambara).
Torodo: the Torodo revolution of 1776 in Futa Toro was begun by the Torodo marabout
movement which took much of its inspiration from the success of the djihâd in Bandu and
Futa Jallon at the beginning of the eighteenth century. It was a revolt of the small peasants
against both the muudul honna, imposed by the Moorish kabllas, and the oppressive eastern
Muslim tax system introduced by the Islamized Denyanke aristocracy.
Trek (the Great Trek - 1834-39): an emigration movement ofthe Cape Bocn towards the Vaal
and the Orange following the British thrust in South Africa.
trekboer: the trekboers were the Boer immigrant farmen who went across the Orange river in
search of land before the Great Trek. They settled mostly in the south of what later became
the Orange Free State.
tshidie: a general council (at the court ofthe Luba king), tshihangu: the court ofthe Luba king.
tsi-tmhety: the collective name adopted by the Tsimibety (Madagascar) to indicate, by refusing to
cut their hair, non-submission to Maroserana authority.
tsovolo: name given by the Hlengwe to the basalt soils.
tubung ('mastcn ofthe land'): heads of small political units (among the Luba/Lunda).
tunka (a Soninke title): king.
turaki: a protocol officer in Katsina (Hausaland).
tweapea: (Akan) chewing-sticks.
ubandawaki (a title): a military leader (in Zamfara and Gobir, Hausaland).

ubuhake: in Rwanda, a type of clientele link (see buhake).
ubugabite: a type of clientele link (in Burundi).
uki: a type of beer made by the Akamba; it is fermented for longer than that of neighbouring
communities.
ukoo: lineage, descent.

'ulamä' (sing, 'â/i'm) (Arabic): Muslim scholan, erudite penons, docton of law or theologians.
undi: powerful ruler in the Maravi confederation in the eighteenth century.
vaha: a medium.
valohazomanga: the commoners in Mahafaly society (Madagascar).

vashambadzi: traden; African merchant class, in the Zambezi region.
veli: governor of a province (sec wall).
vodun: an African religion.
(the) Volafotsy ('white silver')- in Madagascar, the collaterals ofthe ruling Maroserana, descendants
of white silver sons of non-Maroserana women and Maroserana royals.
volamena ('red silver', i.e. gold): in Mahafaly (Madagascar), the tombs ofthe Zafivolamena (sons-
of-gold), the ruling Maroserana branch among the Sakalava.
478 Glossary
wa'azi: sermons (in Hausaland).

wakf (Arabic pi. aukâf): 1) an Islamic endowment of property to be held in trust and used for a
charitable or religious purpose; 2) a Muslim religious or charitable foundation created by an
endowed trust fund; 3) a juridical-religious measure taken by the owner of land or other real
estate to vest its ownenhip in a religious institution (a mosque) or some public or social
facility (a madrasa, hospital, etc.) and/or his or her descendants.
wall (pi. wulät) (from Turkish väli): governor or vice-governor of a province (wiläya).
wamagi: district chiefs, in the Ugweno state (Tanzania).
Wambai: a Kano title.
wasili: North African traders (in Borno).

wasiliram: in Bomo, a special quarter assigned to wasili.
wazir (Arabic) (title of the successon of Shaykh Muhammad Abu Likaylik): in the Ottoman
empire. Title of high-ranking state officials or ministen and ofthe highest dignitanes; holden
of positions analogous to that of Muslim viziers; vice-regents, viceroys.
woloso ('born in the house', in Mande): the status acquired by a woman purchased by a
community as soon as she produced a child. A man could also acquire the same status as soon
as his'master had sufficient confidence in him; house-born slave.
xaadi: a title (in Kayor).
yav: see mwant yav.

Yadega-tenga: the land of Yadega (Yatenga), in Mossi country.
Yatenga-naaba (a Mossi title): the king of Yatenga.
yerima: the governor ofthe northern provinces (in Borno).
yiiri (pi. yiiya): the household, the second level in the Mossi patnlineal and patnlocal society.
zabyuya: a ritual announcement at the ceremonial investiture of Mossi chiefs.

zaka (pi. zakse): the smaller unit in the Mossi patrilineal and patnlocal society.
zakât (Arabic): compulsory alms-giving that, for any Muslim enjoying a certain income, consists
of distnbuting a proportion ranging from 2.5 per cent to 10 per cent - to the poor and to a
specific category ofthe needy. The zakât is the fourth pillar of Islam.
zakse: see zaka.
Zanna Arjinama: a titled official (in Bomo).

zäwiya (pi. zawâya) (Arabic): religious brotherhood; brotherhood seat (and funerary sanctuary of
the founding saint); cultural centre; when fortified and manned by defenden ofthe faith, it is
called ribät.
zombi or zombie (of Niger-Congo origin; akin to Kongo and Kimbundu and Tshiluba nzambi,
god, Kongo zumbi, good-luck fetish, image): 1) the deity of the python in West African
voodoo cults; the snake-deity of the voodoo rite in Haiti and the southern USA; the
supernatural power or essence that according to voodoo belief may enter into and reanimate a
dead body, a will-less and speechless human in the West Indies capable only of automatic
movement held to have died and been reanimated but often believed to have been drugged
into a catalepsy for the houn of interment; 2) a penon thought to resemble the so-called
walking dead; a penon markedly strange or abnormal in mentality, appearance, or behaviour.
zuwaya or insilimen: Berber or Sudanese marabout groups.
Zwayiya (the Zwayiya marabouts): a branch ofthe marabout movement.
Index
Abagusii, 4-7, 409, 410, 411. 412 states, 442-3 impact of Europe Agadir, 1,5, 19, 104, 106, 110
'Abd al-Kädir, 93, 150-1 on, 22 impact of slave trade Agaw, 348
'Abd al-Mïhk, Mawläy, 107-8 on, 52-3, 148, 149, 152,223- age-groups, 230, 382, 389-90
'Abd al-Rahmän, 99, 100, 106 4,227,229,230,231,286, Agona, 210, 212
'Abdullahi, King of Kano 440, 443-4 industries, 17-18 Agoro, 387, 388
(c.1499-1509), 249 languages, 26, 29, 34, 103, agriculture, 12, 15, 16, 20, 53,
abolitionist movement, 1 1 , 57, 125, 149, 150,187,206,221, 101, 161, 196,241-2,265-6,
61,62 229, 268, 307 new economic 280, 324-6, 334, 366-7, 407,
Abomey, 223, 224 structures, 17-22, 441 new 409,414, 444, 445 in Great
Abu 'Abd Allah see al-Kä'im bi- political structures, 22-4, 442- Lakes region, 388-9, 399,
'Amr Allah 3 predatory economy, 18-19, 403-4 and population
Abu 'Abdallah Muhammad al- 20, 21, 441 rural-urban mobility, 25, 27, 29 in
'Ayyâshl, 111 divisions, 17, 18, 19-20 social southern Africa, 336, 338, 341
Abü'l-'Abbäs al-Mansûr, 16, 19 changes, 13-16,441-2,444 Ahmad, Mawläy, 107, 108
Aba Likaylik, 93, 97' stereotypes of, 25, 26, 28 as Ahmad al-A'radj, 104, 106
Abudhok, Queen, 102 supplier of labour, 5, 7, 9-11 Ahmad al-Dhahabi. Mawläy, 118
Acholi, 383, 390, 391,392 trading-post economy, 21-2, Ahmad al-Mansür, 109, 111, 117,
Accra, 7, 10,2-6,210,211,218 24, 441 wars in, 24, 52 and 153, 154. 156, 253
Adal, 349, 350, 354, 358, 359 world economy, 55, 122, 198, Ahmad Gran, 13, 345, 347, 350,
Adangbe, 204, 206, 207, 210, 335, 440, 441-2 seealso 351,355
211,217,218 Central Africa; East Africa; Ahmad Pasha, 72, 74
Adansi, 211. 212 North Africa; North-east Ahmadabad, 66-7
Adar, 236, 237, 246 Africa; Southern Africa; West Aja see Fon
Aden, 76, 424 Africa Akamba, 409, 411, 412,413
Adhola, 386 African diaspora, 28, 30-1, 37, Akan, 21, 192,204,206,207,
Afonso I, 276-7, 285 444 in Americas, 59-64 in 208,210-11,217,218,219,
Africa, architecture and art, 16 Asia, 64-9 in Europe, 56-9, 220
centralization, 22, 214, 442 62, 64 and Freemasonry, 62 in Akwamu, 206, 209, 212, 214, 215
change from captive to slave India, 66-9 in United States, Akyem, 204, 206, 212, 214, 215
society, 13-14 dependency 61-2 Algeria, 18,24, 106, 112, 123,
structures in, 50-3, 441, 444 African religions, 60, 283, 284, 125, 130
diaspora, 28, 56-69, 444 303,306,307,308-9,310, Algiers, 1, 18, 19, 106, 107, 120,
diseases, 1 1 , 444 domestic 317, 393, 394, 396, 397, 405, 123, 124, 126, 127, 128, 129,
slavery, 4, 7, 13, 14,91.92, 445, 446 131-2, 133
96-7, 149, 206, 217-18, 224, African traders, 157, 243, 313, 'Ali Bey, 82-3
230, 243, 244, 266, 269, 282, 331-4 and slave trade, 10-11, 'Ali Gaji, 248-9, 250
283, 444 droughts and 200 Aliada, 223, 224
famines, 27, 32, 36, 102, 131, Afro-Portuguese, 201 , 202, 276- Amari Ngoone Ndeela, 138, 151
132,142,161,326,354,383- 7,278,281,283,288-9,331, Americas, 1, 9, 47-50, 59-64, 69,
8, 440, 444-5 and European 441 198,207,208,209,264, 444
trade, 2, 3, 7, 10, 12 feudal Agades, 105, 233, 235, 249, 250, see also Atlantic slave trade;
structures, 14-16 growth of 258 Caribbean; Spanish America;
479
480 Index
United States Aro, 229, 231 Bahima, 387, 391,395

Amhara, 345, 348, 350, 351 , 354, arts, 16, 109,220,227,284-5, Bahinda, 382, 396, 397, 400
355, 358 363 Bahr al-Ghazäl. 89, 96, 100, 101,
Ampasindavo, 419 Aruwimi, 262, 263 '103,253
Anakara, 420 Arzila, 108, 109, 116-17 Bahu, 397
Andalusians, 131 Asante, 11, 13,37, 164,206,209, Bahumbi dynasty, 401
Andriamanalina I, 430 212,213,214-15,218,220, Bahutu, 397, 399
Anglo-Africans, 201, 202-3, 441 445 Bainuk, 137, 138, 189, 191, 198
Angoche, 368, 372 Asia, 1, 25, 39, 41, 55, 64-9, 70, Bajaranke, 137, 146
Angola, 5, 7, 9, 10,22,52,268, 336 Baka, 29
274,278,280-1,290,321 askiyas, 13, 15, 16 Bakiga, 406
religion, 283-4 slave trade, Assabu,213 Bakimbin, 400, 401
285-6, 288-9, 298 social Atlantic economic order and Bakkära Arabs, 90, 100, 101, 102
structure, 283 see also Kongo African dependency, 50-3 and Bako, 388
kingdom African slaves, 47, 49 al-Bakrt, 82
animal rearing, 20, 241-2, 280, beginnings of, 1, 39-41, 43-7 Balante, 189, 191, 196
326-7,338,388-9,409,410 extension to Asia, 55 see also Bali, 349, 350, 352, 354
see also pastoralists Atlantic slave trade Bambara, 149, 156, 157, 158-9,
Anjouan, 436-7 Atlantic slave trade, 9, 10-11, 21, 161, 163, 165,200
Antambahoaka, 418, 435 38,57,59-61,69, 134, 136, Bamileke, 261
Antandroy, 418, 422, 426, 435 140, 141, 143-6,221,285-6, Bamum kingdom, 261
Antankara, 418, 428, 429 288-9, 440 and African Banda, 262, 305
Antanosy, 418, 421-2, 425, 435 depopulation, 50, 52 and Bandama nver, 192, 193, 213
Antefasy, 418, 434-5 Cameroon, 264-6, 268-9, Bantu, 22, 230, 302, 382, 383,
Antemoro, 418, 420-2, 426, 433, 221, 264-6, 268-9 and Lower 387, 392 in Cameroon, 262,
435 Guinea coast, 208-9, 215, 217 263, 264, 268 languages, 34,
Anteoni, 420, 421 and prazo system, 321 social 221,229,260,263,273,300,
Antesaka, 418, 422, 434, 435 impact on Africa, 52-3, 148, 395,399,416,417,427,436
Antoniamsm, 281, 445 149, 152,223-4,227,229, m Southern Africa, 339, 343
Anyi see Aowin 230,231,265-6,286,443-4 in Southern Sudan, 100
Aowin (Anyi), 204, 206, 21 1, and Upper Guinea coast, 198, Banü Hiläl, 31,32
212-13,214,215,217 200 volume of, 42-3, 50, 52, Banyiginya, 400, 404
Arabia, 1,7, 11,31, 65-6,90 209 and world economy, 39- Banyoro, 387, 399, 400
Arabic language, 81-2, 99-100, 41, 43-7ff. Bara, 161,418,422, 432,433
165,246,255 Awsa sultanate, 358 Barentu, 352, 353, 354
Arabs, 4, 74, 81, 89, 90, 93, 101. Axum, 347, 350 Ban, 382-3
102, 103,242,244,335,349, Azabs, 72, 77, 80 Barry family of Timbo, 148
417, 436 on East African Azbin, 232, 234, 236, 237, 238, Barwe, 315
coast, 370-1, 377-9 and 239, 242, 243, 252 Bashambo, 400
Madagascar, 423-4 in AzemmOr, 104 Bashi, 402, 403
Morocco, 111 and 16th Basita, 397, 399
century crisis, 120-4 slave Baar, 383, 385, 388 Basoga, 393, 394
traden, 11, 96-7 see also Islam, Babba Zaki (c.1768-76), 239 Bassari, 137, 146, 149, 192
Swahili-Arab traders Babito, 382, 390, 393, 397, 400 Batutsi, 387, 397, 399, 404
architecture, 16, 82, 109, 245, Bachwezi, 381,396 Batwa, 302, 397, 399
284, 363 Badagri, 10 Baule, 37, 204, 206, 212, 213,
Are, 364 Bädl IV (1721-1762), 92-3, 97, 217,218,445
Argobba language, 349 98 Bawäkhir militia, 117-18
Arguin island, 195 Baga, 37, 189-90, 192, 196 Bawol, 134, 138, 139, 143, 150,
aristocracies, 15, 20, 21, 52, 134, Baganda, 394, 396, 401 152
143-6, 151,217,244-5,263, Bagemder, 354, 356, 362 Beafada, 137, 138, 196
282, 443, 446 female, 292 in Bagirmi, 249, 253, 254, 256, 258 Bedde, 251,252,258
Loango, 286-7 Baha, 402 Bedouins, 74, 127
Arma, 11, 157, 159, 160, 161, Bahavu, 402, 403 Begho, 193, 207, 209
163 Bahía, 60 Bemba, 312
Index 481
Benghazi, 127 Bron, 206, 213,214 cassava, 207, 231, 265, 266, 270,
Benguela, 273, 283, 288, 289 Budama, 386 409, 444
Benin, 7, 11,37,209,221,223, Buganda, 381, 389, 390, 391, caste system, 17, 18
224, 226, 227, 229, 277 393-5, 405 Castile, 1,3, 5, 6
Berbenstan, 76, 92 Bugesera, 400, 401,403 Catholic Church, 276, 283, 284,
Berbers, 17, 18,22,125,140-1, Buha, 397, 399, 401 345,351,359-62,445
165 Bukerebe, 402 cattle, 90, 100-1, 147-8, 242, 273,
Beti, 260, 263 Bukoba, 414 327, 337, 342, 352, 387, 389,
Betsileo, 418, 429-30,431 Búlala, 248, 249, 250, 251 403, 404, 405, 409, 410
Betsimisaraka, 418, 433-4, 437 Bullom, 189, 190, 192, 195, 196, Central Africa, 16, 18, 27, 264,
Bettie kingdom, 213 198 272,273,285,312,323,387
Bezanozano, 418, 431 Bundu, 134, 143,146,147,151 political changes, 22, 442
Bibiana Vaz family, 202 Bunyoro, 381,389, 390, 391, population movements, 36-7
Bight of Benin, 52 396, 399 decline of, 392-3, slave trade, 265, 440 sources,
Bight of Biafra, 52 394, 406 259, 273
Bijago, 137, 138, 191 Bure, 190 Central Sudan, 232, 237-9, 241,
Biläd al-Makhzen, 122 Burundi, 387, 397, 399, 401, 245, 256, 382
Birnm Gazargamu, 248, 252, 254 402-3, 404, 405 Central Sudanic peoples, 100,
Bisa, 37, 312 Busoga, 385, 386, 387, 390, 392 103,381-2,383,388,399
Bissau, 139, 140 Butanä Arabs, 90 cereals, 128, 130, see also maize
Bito clan, 390-1,399 Butwa see Torwa Ceuta, 56, 104, 109, 119
Biton Kulibali (c.171 0-c.1755), Buzimba (Kigulu), 389 Chad, lake, 251,252, 256, 257
158-9,200 Buzinza, 396, 397, 401, 402 Chad basin, 249, 256
blacksmiths, 17, 18, 220, 229, Chagga, 407, 412, 413
238,260,263,402,411,412 Cabinda, 285, 286 changamires, 22, 315, 321, 329,
Boa Amponsem, 215 Cacheu, 21, 139, 140,202 333
Bobangi, 30, 34, 37, 269, 270, Cairo, 16, 70, 72, 74, 76, 80, 82, Chewa, 300, 301, 303, 305, 307,
271,272,296 83,86,93, 111 309,310,311,313,314,321
Boers, 7, 27, 29, 37 Calabar, 7, 10, 11,21,229,230, Chidima, 315,332
Boma, 285, 287 264 Chisi,312
Bongo clans, 103 Cámara clan, 191 Chitonga, 309
Bono, 204, 210, 211,214 Cameroon, 7, 10, 29, 38, 221, Chokwe, 298
books, 245, 246, 363 226, 230 Atlantic trade, 264- Christianity, 13, 15, 16, 61, 93,
Borana, 352, 353, 413 6, 268-70 central highlands, 218,276,277,281,282,283,
Borgu, 154, 224, 226, 232, 242, 259-61 coastal peoples, 263-4 284, 285, 287-8, 375, 380,
243 exchange economy, 268 local 445, 446 in Ethiopia, 345,
Borno, 13, 94, 96, 153, 208, 234, trade, 269-71, 272 society, 347-9,350-1,355,356-63,
235, 236, 237, 239, 243, 443 271 Uele and Ubangi rivers, 367
centralized political system, 261-3, 268, 269, 286 Christians, 106, 108-9, 116, 126,
250-1 decline of, 258 Cape Coast, 218 128, 132, 133, 154, 156,355,
economy, 252-3 expansion of, Cape Lopez, 265, 266 371
249-50 foundation of califate, Cape of Good Hope, 29, 37, 335, Chulu, 307, 309
248-9 Islam, 15,251,252, 336,337-8,341-3,418 clans, 33, 187,204,206,287,
255 learning and culture, 251, Cape Town, 31,342, 344 309, 364, 389-90, 397, 399,
255, 256 seventeenth century Cape Verde Islands, 4, 6, 136, 414,420
golden age, 256 state, 255-6 137, 151, 198,202 class divisions, 17, 149
see also Känem-Borno capitalism, 11, 19, 41, 42, 43-7, cloth, 50, 209, 242, 327, 330 see
Bola, 355, 359, 365, 442 53, 152,272,319,334,336, also textiles
Bourbon Isle (Réunion), 66, 312, 341,344,443-4 coffee, 76, 79, 349
379, 426, 438, 439 Canbbean, 9, 11, 41, 45, 47-50, Comoro Islands, 21, 373, 424,
Brakna emirate, 142, 144, 145, 53,59,60,61, 62, 64,69, 428, 436-8
150 208, 425 Congo, 5, 7, 11, 52, 266-72 see
Brazil, 4, 5, 7, 9, 49, 60, 280, Casamance, 38, 187, 189, 191, also Cameroon
281,288,289,446 193, 195, 196,201,202 Constantine, 122, 126, 132

British East India Company, 67 Cassange, 280, 286, 289 copper, 3, 7, 18, 260, 264, 268,
482 Index
277,291,327,334 323, 333 problems, 77, 78-9 expansion

corsairs, 18, 19, 123, 124, 127 Don Sebastian, 107, 108-9 of, 75-7, 88 feudalism, 14, 15
Côte d'Ivoire, 187, 188, 190-1, Dona Beatrice Kimpa Vita, 281, industries, 18 Ottomans in,
193,200,204,208,209,212, 284, 285, 445 70-88 political conflicts, 76-
218 Douala, 263, 264, 265, 266 80, 82-6 social changes, 79
cotton, 4, 9, 18,20,45,46,327, Dunkula, 96, 97, 100 taxes. 74-5, 77, 78
349 Dutch, 4, 6, 7, 9, 37, 44, 45, 50, Ekiti, 226
crafts, 17-18, 20, 128, 163, 218- 57, 60, 195, 198, 207, 208, El Mina, 3, 4, 6, 7, 21,22, 193,
20, 242, 243, 244, 264, 270, 209, 264, 265, 278, 279, 376 218
367,411 in Kongo kingdom, 278, 279, elephants, 313, 314, 333-4

Creoles, 34, 60, 66, 201 280, 281, 282 in Senegambia, Emirates, feudalism in, 15-16, 24
Cross river valley, 229-30 138, 139, 140, 144 in England, 4, 6, 43, 44, 45, 55, 57,
Cuba, 9, 446 Southern Africa, 336, 339, 59,60, 114, 116, 128,131,
Cuffee, Paul, 60, 61 341-3 376, 437 and Lower Guinea,
Cwa, 392, 399-400 Dutch East India Company, 7, 208, 209 and Madagascar, 425
Cwezi cult, 393, 394, 396, 397, 336,341,342-3,418,424-5, and Senegambia, 138, 139,
405 438 140, 144 and slave trade, 9,
Cyirima Rujugira, 403, 405 Dutch West India Company, 140, 10, 11, 282 and Upper
198 Guinea, 198, 203
Dädjü kingdom, 97-8 Dyallonke, 189 English East India Company, 418
Dahomey, 11, 22, 221, 223-4, Ethiopia, 13, 35, 36, 76, 91, 92,
226,231,443 East Africa, 3, 13, 16, 64, 66, 305, 93, 101, 388 architecture, 16,
Dakïn (1568/9-1585/6), 91 312, 323 early European 363 art, 363 Christian empire,
Damagaram sultanate, 237 contacts, 7 end of 15,345,347-9,350-1,355,
Dan, 193 independence, 417 iron- 356-63, 442 and Europe, 362
Danangombe, 315 working, 410-11, 412 and feudalism, 15 Islam, 345, 349-
Dandc,315, 316 Madagascar, 420 pastoralists 51,355-6,358-9,360
Danes, 10, 208, 264 and farmers, 407, 409- 10, literature, 360, 362-3 south¬
Darfur, 94, 96, 98, 99, 101, 102, 413-14, 417 population western states, 363-7
103
movements, 30, 38 predatory Europe, 25, 37, 38, 39, 114, 116,
Dauda (1421-38), 239 economy in, 19, 20-1 slave 120, 285, 319, 362, 417, 437,
Dauro, 347, 442 trade, 11, 440 socio-political 440, 441 Africans in, 56-9,
Dawäro, 349, 350, 354, 365 organizations, 413- 16, 417 62, 64 arrival in Africa, 6-7,
Dendi, 154, 156, 158, 159 trade networks, 312-13, 411- 18, 56 and Atlantic system, 1,
Denkyira kingdom, 206, 209, 13 see also East African coast 39-41 economic boom, 3-4,
211,212,213,214,215 East African coast arrival of 21, 43 effect on African trade,
Denyanke kingdom, 134, 136, Portuguese, 368, 371-4, 380 21-2 and Lower Guinea
138, 145, 147, 150, 152, 165 golden age, 368-71 Islam, Coast, 207-10 and
dependency, 41 , 42, 50-3, 441, 370, 371,375, 379 and Madagascar, 418, 423-6, 435
444 Oman, 376-9 and Ottomans, and Maghrib, 128-30, 131,
depopulation, 122 373-4, 376 Portuguese 133 and Senegambia, 138-40,
Dervish orden, 81-2 decline, 376-8 resettlement, 152 and slave trade, 4-6, 9-10
Dhokoth (c. 1670-90), 102 374-5 shârïfs, 379 slave trade, and Southern Africa, 336,
Diabe, 213 379, 380 339, 341 trade with Africa, 2,
Dizi, 364 East Gurage language, 349 3, 7, 10, 12 and Upper
Djabal Marra, 97, 98 economic development, theories Guinea Coast, 195-203
Djallâba, 96, 102 of, 41-2 European settlers, 31, 201-2, 218,
Djerba, 123, 132 Edo, 224 language, 221 314,341-2,343,344
djihäds, 15,20,141, 146-51, 153, Efik, 230 Ewe, 206, 207, 208, 209, 212,
165,258,345,350-1,356, Egba, 226 217,220
358, 375, 446 Egypt, 1, 19,22,92,93,94,97,
Djüdar Pasha, 1,3, 19, 154, 156 99, 103, 128,349,371,373, Fäkäriya, 77, 78, 79, 80
Dodos, 387, 389 385, 443 architecture, 82 and famines, 27, 32, 36, 78, 102, 131,
Dogon, 37 Borno, 250, 252, 256 cultural 132, 144, 161,258,326,354,
Dombo Changamire, 315, 321-2, development, 80-2 economic 383-7, 403, 409, 440, 445
Index 483
Fang, 29, 30, 35, 260, 267 100, 103,441,446 Great Lakes region, 15, 33, 38,
Fante, 206, 207, 211,212, 213, Futa Jallon, 15, 37, 134, 136, 137, 415, 416-17 Central Sudanese
214, 232 138, 143, 146, 147-50, 151, groups, 381-2, 383, 388, 399
farming, 14, 15, 16,20,25,27, 159, 189, 190, 191, 192,200, clans, 389-90, 397, 399, 404
30,241,243,388-9,403-4, 201,442 droughts and famines, 381 ,
417 see also agriculture Futa Toro, 15, 134, 136, 138, 383-8, 389, 403 Eastern
Fashoda, 102 141, 142, 143, 145-6, 150-1, Nilotes, 381 , 382-3, 387, 389,
Fasilidas, emperor, 357, 358 160, 165 390, 391-2 migrations, 381,
Fatagar, 349, 353, 354 385-8 military organization,
Faza, 373, 374, 375, 377 Ga, 204, 207, 208, 210, 211, 212, 405 pastoralists and farmers,
feiras, 329-34 214,217 388-9, 397, 399, 403-4,
feudalism, 14-16, 24, 86, 339, Gabon, 7, 27, 29, 37, 263, 264, Ruhinda tradition, 395-6, 400
341,405 265, 273, 285 socio-political institutions,
Fez, 104, 106, 107, 108, 111, Gagu, 193 389-92, 404-5 southern
112, 113, 114, 117, 118, 119 Gajaaga, 146, 147,151 kingdoms, 395-405 state
Fezzän, 125, 127, 132,248,253 Galawdewos, emperor, 351, 358, formation, 399-403 trade, 405
Fipa, 33, 412,415 359, 360 see also Buganda; Bunyoro;
firearms, 6, 17, 52, 133, 143, 145, Gama, Vasco de, 335, 351, 372 Luo
164, 209, 239-40, 270, 342, Ga-Mashi, 204, 206, 210 Great Zimbabwe, 315
418 Gambia, the, 187, 188, 193, 196 Griqua, 342-3, 344
fishing, 17, 20, 27, 30, 207, 268, Gambia river, 6, 134, 136 guilds, 15, 17,238
270,271,291,325,326,338, Gao, 16, 154, 158, 159, 160, 163, Guinea, 6, 21 , 27, 441 see also
410 242
Lower Guinea Coast; Upper
Fon, 17,221,223-4,231 Garcia 11,280,282,284,285 Guinea Coast
food crops, 27, 207, 217, 231, Gatsi Rusere, 318-19 Guinea-Bissau, 187, 188, 189,
241,264,265-6,280,324, gerontocracies, 389, 390, 392, 191, 196,202
325, 326, 338, 349, 366-7, 393 Gujurat, 66, 67, 76
404, 409, 444 al-Ghälib Billäh, 106 Gulf ofBenin, 6
food production, 18, 26-8 Ghana, 13,215,218,247 Gulf of Guinea, 16, 21, 22, 264
forests, 20, 22, 27, 29, 30, 37, gifts, 240-1 Gulmanceba dynasties, 179, 237
207,214,221,232,259,409 Gimira, 364 Guman, 355
Fort Jesus, 374, 375 Gimma, 355, 365 Guran, 253
Fort St James, 139, 140 Gisaka dynasty, 400, 401, 402, Guro, 191, 193,213
France, 1, 4, 6, 10, 21, 37, 44, 45, 403 Guyana, 59, 60
62,88, 106, 114, 198,264, Gobir, 234, 236-7, 238, 239, 240, Gwanja, 242, 243
282 and Madagascar, 425-6, 243, 246, 258
435 and Maghrib, 128, 131 Gold Coast, 3, 6, 7, 9, 10, 52, 62, Hadiyya, 349, 350, 352
and Senegambia, 138, 139, 189, 195,200,207,208-9 Hafsids, 122, 123
140, 142, 143-4 and slave state formation, 210-17 Haiti, 60, 62, 64, 284, 446
trade, 9, 57, 59 gold mining, 207, 328-9, 333, Hannibal, Abram Petrovich, 65
free blacks, 56-7, 69 422 Harar, 349, 354, 355, 358
French East India Company, 418, gold trade, 1,3,4,6,7, 18, 19, al Harran, Mawläy, 1 1 4
422, 425-6, 427, 435, 438, 21,44,50,94, 136, 153, 164, Hasan Pasha, 85, 86
439 195,208,252,311,334,368, Hasaniyya warriors, 141, 142
French West India Company, 1 40 380 Hausaland, 14, 16, 17, 18, 159,
Fulbe, 25, 30, 90, 136, 147, 148, Gondar, 16, 93, 356-8, 361, 362, 165,207,208, 209,249,252,
149,156, 157, 158, 159, 161, 363
253, 258, 443, 446 anstocracy,
165,189, 190, 191, 192,201, Gonga, 355, 365 244-5 crafts, 242 culture, 245-6
238, 240, 242, 244, 252, 258, Gonja, 214 economy, 241-3, 246-7
260,261,446 Gorée, 7, 139, 140,200 government and administra¬
Fulfulde language, 149, 150, 246 Graaf Reinet, 343 tion, 237-9, 241, 244 Islam,
Fulirwa state, 313 Grand Foul empire, 200, 442 236, 243, 245, 246, 445
Fundj sultanate, 89, 90, 91-4, 97, Grande Comore, 436, 437 military organization, 239-40
101, 102,250,441,446 Great Britain, 61 , 67, 264, 336, and neighbounng regions,
Fur sultanate, 89, 90, 94, 96, 97- 343-4, 438 see also England 234-7 political development,
484 Index
232-4 resources, 240-1 role in Imerina, 430-1 Jesuits, 359-60

West African economy, 232, Inariya, 347, 355, 359, 362, 365 jewellery, 18, 242
235 scholars, role of, 241 , India, 3, 6, 7, 65, 66-9, 319, 324, Jews, 67, 130,356,358
243, 244, 245-6 social 339,351,371,372,373,375, Jie, 26, 381,387, 388, 389
relations, 243-5 sources for, 376, 379, 426, 438 Jieng, 89, 101, 102
232 state m, 247 trade, 242-3 Indian Ocean, 19,20,21,22,43, Jn language, 101, 102
Hawwära shaykhs. 74, 75, 82 64, 66, 75, 76, 92, 300, 312, Joal, 6, 7, 139
Haya, 396, 399 330,339,341,351,354,369, Jola, 16
Hemba, 291,293, 296 372-3, 376, 423-4, 437 Jolof confederacy, 134, 136, 137,
herdsmen see pastoralists industrial goods, 3, 10, 12, 17-18, 138, 141, 142, 150, 151, 152
Herero, 336, 337 20,53 jtmya system, 13, 14
Hilaban nomads, 122 Iramba, 407, 412 Joola, 30, 37, 136, 158, 189, 191,
Hima, 397, 399, 404 iron, 3, 136, 229, 260, 277, 312, 195, 196
Hinda dynasties, 396, 401 327,334,338,382,383,410- Joos, 143, 144
historiography, 81, 127 11,412 Jula, 157, 164, 165, 193, 195,
Hlengwe, 324, 325 Iron Age, 100, 302, 335, 338, 417 200,207,214
Holland see Dutch irrigation, 27, 38, 252
Hormuz, 373, 376 Iru, 397, 399, 404 Kaabu kingdom, 134, 136, 137-8,
Horn of Africa, 35 Islam, 345, Isandra, 429-30 146, 152, 191, 196,200,201
347, 349-51, 355-6 languages, Isera, 387, 388, 389, 390 Kaarta, 151, 158-9, 163, 200
349 migrations, 345 religious Islam, 13, 14, 15, 16, 60, 64, 67, Kabunduli, 307, 309
wars, 345, 347, 350-1, 357-8 70, 72, 124, 126,218,248-9, Kâdirîyya order, 94, 163, 165
states, 347, 349 see also 251,252,255,436,437,445- Kaffa (Gomar), 359, 365, 442
Ethiopia; Somali 6 East African coast, 370, 371, Kafula, 302
horses, 239, 241,243, 252 375, 379 Ethiopia, 345, 349- Kagera plateau states, 400-1 , 403,
hunter-gatherers, 26, 27, 29, 30, 51, 358-9 Hausaland, 236, 404
260, 325, 326, 336-7, 388, 243, 245, 246, 250 Morocco, al-Kä'im bi-' Amir Allah. 104
409 109, 111, 114, 117, 119, 153 Kakalala Musawila Gondwe, 313
hunting, 20, 27, 207, 263, 292, Senegambia, 134, 140-3, 146- Kakwa, 382, 383
314, 334, 336, 338, 410 w 51, 152 Sudan, 89, 91,93-4, Kalenjin, 407, 409, 410, 416, 440
also hunter-gatherers 98,99, 156-7, 165,200 Kaluluma, 307, 309
Hutu, 397, 399, 403, 404 Isinâ'îl, Mawläy, 113-19, 160, 164 Kalundwe kingdom, 292
Ismä'il Bey, 85, 86 Kamba, 407, 414
Ibibio, 221,229-30 Ismä'il Gabarti, 356 Kameme, 308, 309
Iboina, 48, 427, 428, 429, 434, Iteso, 381,387, 388, 389, 390, Kamusi, 446
438 406 Känem, 18,232,248,251,256,
Ibrahim Bey, 85, 86 Itsekiri, 37, 226-7 258 see also Borno, Känem-
Ibrahim Kazdoghlu, 80 ivory trade, 3, 6, 136, 200, 208, Borno
Ibrahima Sory Mawdo, 148-9 269-70, 277, 279, 280, 300, Känem-Borno, 13, 19, 22, 232,
Idris Alawoma (1564-96), 249, 304,308,311,312,313,314, 244, 247 see also Borno
250-4 333-4, 378, 379, 402, 409 KanEmbu, 14,252,258
Idris Katakarmabe (c.1497-1519), Iyasu 1, emperor (1682-1706), Kano, 208, 234, 235-6, 237, 238,
249, 250 362
239,240,241,242,244,245,
Ife, 224, 226 246, 247, 249
Ifrandj Ahmad, 79 Jaga, 25, 32, 268, 277-8, 282 Kantor, 193, 196
Ifrikiya, 123 Jahanka, 30, 37 Kanuri, 232, 242, 443
Igala, 33, 226, 229 Jaki, 386 Kanyenda, 307, 309
Igbo, 29-30, 38, 221, 224, 227, Jallonke, 147, 148, 149 Kanyok, 290, 295, 296
229, 231 Jamaica, 60, 62 Karagwe, 397, 399, 400, 401,
Ihangiro, 396, 401 Janissaries, 72, 74, 77, 78, 79, 80, 403, 405
Ijebu, 226 123, 124, 126, 130, 132 Karamojong, 34, 35, 440
Ijo, 221,226, 227 Janjira Island, 67 Karamokho Alfa, 148
Ile-Ife, 223, 224 Jaxanke, 147, 148 Karanga, 307, 315, 321, 323, 331
Imbangala, 32, 33, 278, 289, 290, Jenne, 16, 150, 154, 156, 158, Kanmojong, 387, 389
298 160. 161, 164, 165,207 Karou, 201
Index 485
Kasai, 285, 290 274 religion, 283-4, 287-8 362-3
Kasanga, 137, 191, 196, 198 slave trade, 273, 277, 278, Loango, 10, 264, 265, 266, 269,
Kasanje, 278, 280, 281,289 280, 281-2, 286 social 273, 274, 276, 278, 279-80,
Kâsïmïya, 77, 78, 79 structure, 274-7, 282-3 see also 281-2, 285, 286-7
Katanga, 302,313 Angola Loggar, 143, 144
Katsina, 208, 234, 235, 236, 237, Komagui, 137, 146, 149, 192 Loko, 190
239,240,241,242,244,246, Konko Bubu Musa, 145-6 Lolo, 300, 301,302, 305, 310
249 Kono, 190, 191 Lomami, 29
Kawar, 242, 244 Kora, 342, 344 Lomwe, 300, 302
Kayor, 22, 134, 138, 139, 141-2, Kordofan, 92, 93, 94, 96, 97, 99 Lotuhu, 382, 383
143, 144, 150, 151, 152 Koten, 387, 388 Lower Guinea Coast Atlantic
Kayra dynasty, 98, 99 Kru, 190, 191, 193,203 trade, 208-9 centralization,
Kayrawän, 123, 126 Kuba, 30, 32-3, 36, 285 442 Christianity, 445 cultural
Kazembe kingdom, 37, 297, 298 Küchük Mehmed, 78 changes, 217-20 economic
Kebbi, 234, 236, 237, 239, 240, Kuku, 382, 383 changes, 206-10 impact of
242, 244, 246, 250 Kulango, 213, 214 slave trade, 210, 215, 217
Kel-Awllimiden, 157, 159, 160 Kumam, 381,388, 390 Islam, 445-6 languages, 206
Kel-Tadmekket, 157, 159, 160 Kumasi, 209, 210, 211, 213, 214, migrations, 204-6, 211,217
Kenya, 33, 35, 36, 352, 386, 388, 218,220 political changes, 210-17
392, 407-17 Kunta, 157, 159, 160, 164 social structure, 206, 211, 212,
Kete, 290 Kurfay, 237 217-18 trade, 207-10
Kctu.21,223 Kwa languages, 187, 190, 221, Lozi, 290
Khäyr Bey, 72 268 Lualaba river, 296
Khoi, 29, 335, 337, 338, 339, Kwango river, 273, 274, 278, Luanda, 21 , 269, 270, 276, 277,
341,342,343-4 280, 281, 289, 290, 297, 298, 278,280,281,283,288,289
Khoi Khoi (Khoisan), 340, 341, 299
Luangwa valley, 312, 313
344 Kwarafa, 232, 234, 235, 246 Luapula river, 292, 297, 299
Kigeri Mukobanya, King, 402 Luapula state, 297-8
Kikondja, 291,296 lagoons, 217 Luba, 22, 297, 303 origins, 290-2
Kikuyu, 407, 410, 411-12, 414 Lagunaires, 206, 211,217 political organization, 291,
Kilifi, 374 lamana system, 14, 16 292, 295-6
Kilwa, 7, 20-1, 22, 66, 312, 313, Lamu, 21, 30, 370, 373, 375, 377, Luel, 100
330,368,371-2,373,378, 378, 379 Lulua, 290, 291
379, 380 lançados, 201-2 Lunda, 32, 37, 289, 290-2, 296-9,
Kimbu, 407 land, 14-16,26-8 311,312
Kingabwa, 268 Landuma, 189, 196 Lundu, 305
Kinshasa, 268, 269 Langi, 34 Luo, 31,34, 35, 36, 381-2, 383,
Kisi, 149, 189,203, 412 Lango, 381,383, 406 407,409,410,411,412
Kiswahili language, 370 Larache, 108, 116 language, 386, 389 migrations,
Kitara, 389, 390, 392, 393, 394, Lasta, 350, 362 385-8, 440 political
396
Lat Sukaabe Fall (1695-1720), organizations, 391-2, 393,
Kivu, lake, 402, 403 143
415-16 in Southern Sudan,
Koc state, 385-6 leather, 18, 136,252 101
Kololo, 314 Lebna Dengel, emperor
Koly Tengella, 136, 138, 147 (1508-40), 350, 351 Maalik Sy, 147
Konde Burama, King of Sankaran, Leo Africanus, 122, 208, 242 Maasai, 29, 35, 36, 375, 386, 407,
148-9 Liberia, 187, 188, 190, 191, 193, 410,411,412,413,414,417,
Kong empire, 164, 200 201 440
Kongo kingdom, 18, 22, 264, Libya see Tripoli, Tripolitania Macina, 159, 161, 165
268, 298, 442 agriculture and Limbia, 189, 196,230 Madagascar, 25, 37, 38, 66, 418-
food, 280 art and culture, lineage system, 17, 33, 187, 204, 35, 437-8 centralization, 442
284-5 Christianity in, 276, 206,218,397, 404 eastern and southern islands,
277,281,282,283,284,285, Linta, 426 433-5 Europeans in, 418,
287-8, 445 disintegration of, Liptako, 159 423-6, 435 interior states,
281 languages, 273 origin of, literature, 81-2, 109, 150, 360, 429-33 newcomers and
486 Index
coalitions, 420-3 northern and Manandriana, 429, 430 Mediterranean region, 1, 17, 19,
western states, 426-9 slave Manda, 375, 377 39, 164
trade, 418, 423, 424, 425, Mandara, 249, 253-4, 258 Meknes, 114, 117
426, 434, 435 Mande (Manden), 17, 18, 30, 37, Menabe, 424, 427-9
Madi, 100,382,383,386 136, 137, 138, 139, 147, 148, Menarandra, 426, 427
Mafamba, 268 156, 187, 188-9, 190,191, Mende, 34, 192
Maghrib, 1,4, 16, 18,19, 111, 192-3, 195, 196, 198,200, Merina, 418, 420, 430-1, 437
156, 161,250,253; 201,207,209,213,214,445, mestizos, 201 , 202
eighteenth century, 130-3; 446 metals, 3, 10,17-18,45,49,70,
and Europe, 128-30, 131, Mane, 25, 32, 37, 190-1,202 327
133; land ownership, 14-15; Manganja, 301, 306, 307, 311, Mexico, 47, 49
pnvateenng, 124, 127-8, 131, 321 al-M'hammad. Mawläy. 112, 113

132-3; Regencies, 123-4, Mangbetu, 33, 262 Middle East, 1, 39, 50, 65-6, 70,
126-7ÍF.; seventeenth century, manioc, 12 83 see also Arabs; Islam
124-30; sixteenth century, Mankhamba, 305 migrations, 25-6, 28-33, 69, 188,
124-30; social structure, 125- Mansfield, Lord, 57, 61,62 204-6,217,227,258,260,
6; Spaniards in, 122-3, 127; Manthimba, 305, 306 261,262,304,305,321,327,

trade, 122, 126,128-30, 131, Manyika kingdom, 315, 318, 322, 345, 352-4,381,385; band,
132, 163-4 324,328,331,333 28, 32, 33, 34, 37; by drift,
Magos, 388 marabouts. 17, 119, 134, 140-3, 28-31, 37; élite, 28,32-3;
Mahafaly (Mahafales), 418, 422, 146-51,165 mass, 28, 31-2, 33,34, 38,
423, 426, 433 Maravi state, 37, 300, 303-7, 309- 440; see also population
al-Mahddiyya, 104, 116 14,372 movements
Mai, 13, 15 Marinids, 104 Miji-Kenda, 35, 374, 375, 378,

maize, 12, 207, 231,241,246, Maroserana, 37, 418, 422, 423, 407,413,414
265, 280, 324, 330, 338, 367, 426, 427, 429, 430 Milansi dynasty, 415
409, 444 Marrakesh, 16, 104, 106, 107, mining, 18, 49, 207, 317, 327,
Makhazen, 24m,124, 125 111, 113, 114, 117, 118, 160 328-9,331,333
Makua, 300, 301, 302, 305, 310, Masa-si, 158, 159 Mirambo, 415
311 Mascarene Islands, 321, 418, 426, missionaries, 283, 284, 319, 359,
Malabar, 67 433, 435, 438-9; see also 445
Malagasy language, 420 Bourbon Isle (Réunion); Mocha, 65

malaguetta pepper, 190, 195, 200, Mauritius Mogadishu, 349, 355, 373, 424
203 Massawa, 75, 347, 348, 349, 351, Mogador,119, 164
Malam Djibrll dan 'Umaru, 246 362 Moheli, 436, 437
Malam 'Uthmän, 245, 246 Matamba, 275, 280, 283, 286 Mombasa, 3, 7, 21,351,368,
Malawi, 37, 300, 302, 303, 307 Mauritania, 14, 30, 31, 142, 144, 371,372,373,374,375,377,
see also Northern Zambezia 160, 163,446 378, 380, 424
Malawi, lake, 300, 302, 303, 304, Mauritius, 66, 379, 424, 426, monarchies, 143-6, 237-9, 241,
311,313 438-9 244, 264, 286-7, 308-9, 364
Malebo Pool, 266, 268, 269, 271, Mawanda (c.1 674-1704), 394 see also anstocracies
272, 274, 277, 280, 286 Mayombe, 273, 274, 281, 286, Mongo, 29, 30, 36
Malemba, 285 287 Monzón (cl 789-1 808), 159
Mali, 13, 37, 136, 188, 190, 193, Mayotte, 436, 438 Moors, 56, 144, 145, 146, 150,
195, 196,200,232,247,250 Mazagän, 106, 109. Ill, 119 424
Malik Ambar, 67-9 Mbanza Kongo (S3o Salvador), Morocco, 1, 5, 6, 11, 18-19, 22,
Mâliki order, 93-4 268,274,277,281,284 24, 145, 253 'Alawite dynasty,
Malindi, 7, 21,368, 370, 371. Mbomu, 262, 263 112-19, 160 art and
372-3, 374, 378, 424 Mbonacult, 306, 307, 310 architecture, 109 battle ofthe
Malinke see Manden Mbongo, 263 three kings, 107-9 invasion of
Mamba, 412 Mbum, 260, 261 Songhay, 153-7, 158 Islam,
Mamlüks, 15, 70, 72, 74, 75, 77, Mbwela, 297, 298 109, 111, 114,117, 119
79,80,81,83,85,86,97, means of production, 14 militias, 1 17-18 Portuguese
371,373 Mecca, 30, 72, 74, 76, 83, 119. in, 104-9, 111, 116 reign of
al-Ma murä see al-Mahddiyya 128 Mawläy Ismä'il, 113-19 Sa'ädi
Index 487
expansion in, 109-12 Naatago (1756-66), 144 358 Ottoman influence in,
Shbanäte revolt, 111-12, 113 Nafana.213,214 75-6, 351,356, 358 seealso
trade, 163, 164 Nalu, 192, 196 Ethiopia; Horn of Africa,
Moronou, 213 Nama, 342, 344 Somali; Sudan
Mossi, 37, 154, 164,208 Namibia, 36 Northern Manden, 192
Moura, José Lopez de, 202 Nandi, 416 Northern Zambezia peoples of,
Mozambique, 7, 21, 37, 67, 289, Nâsir al-Dïn, 140, 141, 142, 147, 300-2, 305-11 prazos, 311
300,302,303,310,312,318, 150, 446 religions, 303, 304, 306, 308-
319,320,331,371,377,378, natron, 252 9, 310 slave trade, 312 social
379, 424 Nchare, 261 structure, 302-3, 305-6 trade,
Mpongwe, 263, 266 Ndebele, 322 303-4,308,310-13,314
Mporo state, 381 Ndenye, 213 Nsenga, 300, 302, 306-7, 311,
Mpororo, 392 Ndongo, 274, 276, 277, 278, 281 313,321
Msukwa, 309 Nembe, 226, 227 Nuba, 101
Mughogho, 314 New Calabar, 227 Nubia, 76, 91-2, 96
Muhammad al-'Alem, Mawläy, Ngbaka, 262 Nupe, 18,224,226,232,239,
118 Ngbandi, 25, 36, 262, 263 242, 243
Muhammad Al-Hadjdj, askiya Ngil society, 267 Nyakyusa, 300, 309
(r. 1493-1 528), 154,232,234, Ngizim, 251,252 Nyamwezi, 407, 412, 414, 415
250 Ngobi, 387 Nyanga, 325
Muhammad al-Mubârak (c 1653- Ngonde, 300, 301 , 309 Nyanza, lake see Victoria, lake
87), 234 Ngulube, 314 Nyiha, 309, 415
Muhammad al-Shaykh. 104, 106 Nguni, 31, 314, 336, 338-9, 344 Nyilak, Princess (c. 1517-62), 383
Muhammad 'All Pasha, 102 Niger bend, 1, 11, 136, 146, 153, Nyoro state, 389, 401,402
Muhammad Bey, 83, 85 156-61, 163, 193 Nzabi, 27, 29
Muhammad Kisoki (1509-65), Niger-Chad region, 13, 14, 17, Nzima, 204, 206, 212, 213, 214
235, 249 18, 19, 20, 22
Muhammad Kurra, 98, 99, 100 Niger delta, 221, 226-7, 229, 230 Ogowe delta, 29, 37
Muhammad II Benkan Kiriai Nigeria, 16, 18,52,221,224-6 Oman, 7, 22, 65-6, 351, 368,
(1531-7), 234 Nile valley, 20, 22, 37, 76, 91, 96, 376-9
Muh arrimad SJjärif (c. 1 703-3 1 ) , 97, 101,262,382 Ometo, 355, 365
236, 247 Nilotes, 89, 100 central-eastern, Omolo, 386, 387, 388
Muhammad Tayräb, 98, 99 381,382-3,387,389,390, Omoro, 413
Muhammad III, 119 391-2 southern, 35 Omukama Isansa, 393
Muhammad Zaki (c.1582-1618), Njaag Aram Bakar (1733-57), 144 Onilahy, 426, 427
235, 245 Nkomi, 266 Opoku Ware, 215
mulattos, 201-3, 218, 283 Nkore, 389, 393, 396, 397, 399, oral traditions, 32-3, 188, 213,
Murad Bey, 85, 86 400
227, 230, 259, 263, 304, 364,
Muradites, 126-7 nomads, 26-7, 36, 90, 122, 125, 422
Muscat, 376 148, 159, 165, 252, 355 sec Oran I, 106, 123, 132
music, 245 also pastoralists Ormans, 146
Muslims, 1,14, 20, 56, 67, 89-90, North Africa, 1,252,255 Oromo (Galla), 25, 31, 32, 35,
93, 111, 140-3 eighteenth architecture, 16 feudal system, 36, 345, 348, 352-5, 356, 358,
century revolutions, 146-51, 15 Ottomans in, 72 359, 374, 375, 378, 380, 382,
446 see also Arabs; Islam population movements, 30, 410,417,440
Mutapa empire, 13, 22, 303, 315- 35, 36 predatory economy, Orungu, 266
19,321,325,326,328-9,330, 18-19 slave trade, ll,39iee Osei Tutu, 215
333, 442 also North-east Africa Osu, 210
al-Mutawakkil, 107 North America, 9, 41, 43-7, 59, Otondo, 377
Mwafulirwa family, 313 60-1 see also Americas; Ottoman empire, 1, 7, 18, 19, 99,
Mwahenga, 313 Atlantic economic order; 106,109, 119,153.250,253,
Mwaulambya, 309 Atlantic slave trade; United 351 , 356, 358 conquest of
Mwene Mutapa, 13, 423 States
Egypt, 70-88 decline of, 86
Mwenekisindile, 308 North-east Africa, 345, 440, 442 and East African coast, 373-4,
Oromo people, 352-5, 356, 376 feudal system, 15, 16, 24
488 Index
fiscal problems, 78-9 m 30-1,33, 34, 37 drift, 28-31, 198,202

Maghnb, 120-7ÍT, 133 and 37 expansion, 28-30, 31,38 Rozvi empire, 315, 319, 321-4,
Mamluks, 70, 72, 74, 75, 77, individuals, 28 and land use, 326, 328-9, 333
79, 80, 83, 86 m North-east 26-8 and language, 34-5 Rufisque, 6,7, 139, 143
Africa, 75-6 and Nubia, 91-2 reasons for, 27-8 sources for, Ruganzu Ndon, 402
Ovambo, 336, 337 33-5 and trade, 30-1,37,38 Rugomora Mahe. 401, 405
over-population, 32, 36 Upper Guinea, 188-93 see also Rumfa (c.1463-99),235
Owiny, 386-7, 388 migrations Rund, 290, 292, 296-7
Owo, 226 Poro, 203 Russia, 65, 119
Owu, 226 Porto Novo, 10 Rwanda, 381, 391, 395, 396, 397,
Oyo, 11,223,224,226 Portudal, 6, 7, 139, 143 399, 400, 402, 403, 404, 405,
Özdemir Pasha, 76, 92 Portuguese, 1, 3, 4, 7, 18, 19, 20- 406
1,37,50,56,67,70,75,92, Ryangombe, 396, 397, 400

Padhola, 386 154,208,221,226,286,288,
Paez, Pedro, 359-60 298,335,351,359,441,445 Sa'ädi dynasty, 5, 104, 106, 107,
painting, 363 and Cameroon, 263, 264-5 108, 109-13, 117,153, 160
Pajulu, 382, 383 and East African coast, 368, Sahara, 13, 14, 17, 19,22,26,30,
Pakoyo, 392, 393 371-4, 376-8, 380 and Kongo 43,52, 134, 154, 165
Pakwac-Pawir, 382, 383, 385, kingdom, 276-80, 283 and Sahel, 14, 16,38, 130, 136, 161,
386,388 Madagascar, 418, 423-4 and 163,200
Paluo, 385, 390, 392, 393 Morocco, 104-9, 111,116, Saint-Louis, 139, 140, 141, 142,
paper, 18 153 and Northern Zambezia, 143-4,145,151,163
Pare, 407, 412, 414 305, 307, 310-11, 312 and St Thomas island, 4, 5, 7
pastoralists, 100, 147-8, 252, 327, Senegambia, 134, 136-7 Sakalava, 418, 422-3, 424, 427-8,
336-8, 342, 343, 344, 433 and settlers, 201-2 and slave trade, 429, 434, 437, 438, 442
agriculturalists, 15, 407, 409, 4,5-6,9, 10, 57 and Southern Sakatovo, 426
413-14 in Great Lakes region, Zambezia, 315, 316, 318-21, Salima, 96
388-9, 397, 399, 403-4 and 322, 323-4, 329 structural salt, 3, 136, 153, 164,207,252,
population mobility, 25, 26-7, weakness of, 3-4, 6 and 264,268,270,291,313,327,
29,30,31,36 Upper Guinea, 195, 196, 198 334,411,412
Pate, 21, 374, 375, 376, 377-8, pottery, 18, 218, 227, 242, 268, Salum, 138, 143, 144, 152
379, 380 338, 412 Samba Gelaajo Jeegi (1725-31),
Pawir, 387, 388, 392 prazo system, 311,319-21,324, 145-6
peasants, 15, 16, 19-20, 266, 326- 341 San, 7, 36, 335-6, 337, 339, 342,
7,329,330-1,364,366,443, privateering, 19, 124, 127-8, 131, 344
446 132-3 Sanga, 290, 291
Pemba, 374, 375, 376, 377, 378 proto-Bantu languages, 260 Sankola province, 196
Pende, 299 Pygmies, 26, 29, 260, 262, 269 Sanwi, 204, 206, 212
pepper, 6, 50, 330 see also Sao Tomé island, 264, 276, 277,
malaguetta pepper Quoja, 192, 201 278, 283
Phiri, 304, 305, 306, 314 Sapes kingdom, 190, 196
pilgrimage, 30 al-Rashid, Mawläy, 112-13, 116, sarki, 237-8
pirates, 425, 437; see also corsairs 160 Sarkin Mazum, 238
plantation system, 5, 9, 45, 46, Ratsimilahoe, 434 Sartsa Dengel, emperor (1563-
55,60, 136,321,438-9 Ravahiny, Queen (c. 1767-1 808), 97), 356, 358, 359-60
poetry, 81,82, 109 428 savannah, 221, 232, 259, 260
Pokomo, 375 Red Sea, 1,43,52,64,65,75, scholars, 93, 99, 238, 241, 243,
Pokot, 407 92,347,351,356 244,245-6,251,255,446
population density, 26, 27, 28, 29, religious brotherhoods, 14, 80, sculpture, 285
36, 38, 259, 445 94, 104, 163, 165 Sefuwa dynasty, 248, 249, 250,
population growth, 38, 43, 44, Réunion see Bourbon Isle 252, 255, 256, 258
45,217,444 rice, 27, 46, 231,241 Sefwi, 204, 206,211,212.212.
population movements, 25-38, Rid wan Kähya, 80 214
299, 440-1 Cameroon Rift Valley, 349 Segeju, 373, 374, 378
highlands, 260 diaspora, 28, Royal African Company, 10, 140, Segu, 157, 158-9, 161, 163,200
Index 489
Sèkko, 346, 365, 442 217,230-1,243,244,252, also Mutapa empire; Rozvi

Selim I, Sultan, 70-2, 74, 81 264-5,269,277,278,281-2, empire
Semakokiro, king, 394 298,312, 321,438, 443-4 and Soyo, 276, 279, 280, 281,283
Sena, 298, 311,315, 322, 330, African traders, 10-11 and Spain, 1,4,5,6,7,9, 18,24, 44,
331,333,372 Madagascar, 418, 423, 424, 56-7,59, 131, 198,250,279,
Senegal, 6, 10, 17, 195 425, 426, 434, 435 to Asia 376, 441 and Maghrib, 122-3,
Senegal river, 134, 136, 139, 140 and Middle East, 39, 43, 52, 127 and Morocco, 104, 106,
Senegambia, 4, 6, 7, 21, 196, 441 64-9, 379, 380, 440 see also 107-9, 111,116
and Atlantic trade, 134-8, Atlantic slave trade; domestic Spanish Morocco, 4-5, 9, 41, 44,
140, 141, 143-6, 148, 149, slavery 45, 47-50, 55, 59, 60
152 ceddo rule in 143-6 slaves, 4, 117, 118, 150,241,243, state(s), 17,22,24,33,37,210-
Marabout War, 140-3 Muslim 244,282,283,319,320,342, 17, 247
revolutions, 146-51, 446 366, 438 Sublime Porte see Ottoman
partition ofthe coast, 139-40 social change, theories of, 41-2 empire
political structure, 137-8 Sofala, 3, 7, 22, 30, 318, 321, Sudan, 4, 16, 18,30, 111, 134,
sources for, 188 324,329,330,331,368,371, 136, 155-60,249,253,262,
Sennar, 91, 92,93,96,97, 102 372, 380 347, 441, 442 'Abdalläbi state,
Sereer kingdom, 14, 143, 144 Sokoto caliphate, 15 90-1,92, 97 and Atlantic
Sese Islands, 401 Somali, 25, 35, 36, 345, 347, 355- trade, 200 Fundj sultanate, 89,
Shaba, 290, 291,292, 297 6,410,440 90,91-4,97, 446 Für
Shambaa kingdom, 33 Somalia, 350 sultanate, 89, 90, 94, 96, 97-
Shambala, 410, 414 Songhay empire, 13, 15, 17, 19, 100,446 Islam, 89, 91,93-4,
Shandî, 94, 96 196, 207, 208, 250, 253 and 98,99, 156-7, 165,200,248-
share-cropping, 14 Dendi, 156, 158 and 9,250, 446 languages, 103
Shaykïyya, 96, 97 Hausaland, 232, 234, 236 north-south divide, 89-90
Sherbro, 190 Moroccan invasion of, 153-4 population movements, 89, 90
Sherbro Island, 195, 200 Songye, 290, 295, 296 slave trade, 90, 94, 96-7, 153
Shila, 292, 295, 297-8 Soninke, 146 trade, 94-6, 98, 164 seealso
Shilluk, 89, 92, 96, 101-3, 383 Soso, 37, 147, 189, 190, 192, 289 Central Sudan; Känem-
Shirazi dynasty, 370, 374, 436 South America see Spanish Borno; Southern Sudan,
Shire nver, 305, 306, 307, 310, America Western Sudan
311 Sotho, 336, 337, 338, 339, 344 Sufism, 94
Shoa, 348, 350, 354 Sousa Coutinho, F. I de sugar, 4, 5, 7, 20, 45, 49, 59, 109,
Shona, 18,307,315,318,319, (1664-72), 288, 289 153,324,349,438
323, 324-5, 326, 327, 328, Southern Africa, 22, 27, 29, 36, Suku kingdom, 298
329-34, 339 37, 440 agriculture, 336, 338, Sukuma, 407, 412,414
Shuwa Arabs, 251, 252, 258 341 changes in, 335-6 Dutch Sukwa, 300
Siddis, 67 in, 336, 341-3 hunter- Suläyman Baal, 147, 150
Sidi M'hammad, 117, 118 gatherers, 336-7 pastoralists, Sulaymän Solongdungu, 98
Sidi Muhammad, Sultan (1757- 336, 337-8 population, 337, Sulaymän the Magnificent (1520-
1790), 160 338 social structures, 338-9 66), 91
Sierra Leone, 3, 10,34,37, 61, Southern Manden, 187, 190 Sunnites, 80, 436
187, 188, 189, 190-2, 195, Southern Rivers, 134, 136, 137, Sus, 104, 112, 113, 118, 156, 163
198,200,201,202-3,441 140, 146, 152, 189, 196, 198, Susenyos, emperor, 345, 360, 362
Sihanaka, 418, 431 201,202 Susu see Soso
Sun, 138, 139, 143, 144, 152 Southern Sudan, 100-3 Swahili, 18,30,37,314,330,
Silanga dynasty, 402 Southern Zambezia, 303, 441, 368,370,371,373,374,377,
silver, 3,7, 18, 21, 44,328 442 agriculture, 324-6, 334 378, 380
Siyu, 374, 376, 377 animal-rearing, 326-7 luwveld Swahih-Arab traders, 319, 320,
slave hunters, 320 economy, 325-7 mining, 327, 326,329-30,331,423,424
slave revolts, 60-1, 64, 264, 321 328-9 political control, 317- Swellendam, 343
slave societies, 13-14, 319, 320 18 Portuguese in, 315, 316, Syria, 70, 83, 85
slave trade, 1,3, 4-6,7, 19,20, 318-22, 323-4, 329-34 prazo
28,90,94,96-7, 143, 144, system, 319-21 slave trade, Tafilälet, 112-13, 163
146, 153, 164,208-9,215, 321 trade mdfeiras, 329-34 see Taghäza, 153, 154, 163
490 Index
Takrür, 13, 17, 90, 249, 250 trappers, 27 for, 187-8, 189
Tanala, 418, 431,433 Trarza emirate, 142, 144, 145, urbanization, 28
Tanganyika, lake, 302, 395, 396, 150 Uteve kingdom, 315, 318, 322,
401, 404 Tripoli, 1, 18, 19,120, 122, 123, 328,329,331,333
Tangier, 109, 116 125, 126, 127, 129, 131, 132- Uvinza, 412
Tanoe, 187 3, 163,250 Uyombe, 314

Tanzania, 33, 36, 37, 339, 407-17 Tripolitania, 22, 24, 125, 126,
Tatoga, 402 128, 132-3 Vai, 190, 191
taxes, 15, 16,20,24,74-5,77, Tsimihety, 418, 428, 429 Victoria, lake, 385, 386, 393, 395,
122. 131,240,241 Tsotsebaki states, 237 396,397,401-2,410,412
Taza, 112 Tswana, 336, 337, 338, 344 Vih, 37, 264, 279-80, 289, 298
Tebandeke (cl 644-74), 394 Tuareg, 25, 157, 159,161,238, Vinza, 412
technology, 22, 42, 265 246, 253, 256, 258 Virginia, 60, 64, 208
Tejek, 143, 144 Tubu, 251,252, 258 Volamena, 427, 428, 434
Tekidi, 382, 383 Tukuloor, 135, 165 Volta basin, 208. 214, 217, 252
Tema, 204, 210 Tumbuka, 300, 301, 302, 303, Vumbu Kuu, 375, 378
Temne, 189, 192, 195, 196, 198 307,309,311,313,314 Vungara, 262-3
tenant farming, 14, 16 Tundjür, 97, 98
Tenda, 137, 192 Tunis, 1, 18, 19, 107,120, 122, Waalo, 138. 141, 142, 143-4, 150,
Tete, 298, 307, 311,316, 372 123, 125, 126, 128, 130,131, 151, 152
Teve, 318, 331 132, 133, 163 Wadai, 94, 96, 97, 98, 256, 258
Tewodros (Theodore II), 345, Tunisia, 18,99,126, 130-1 Wädi al-Makhäzin, battle of, 108-
363 Turkana, 35, 387, 389, 407, 410, 9, 153
textiles, 3, 6, 7, 10, 15, 18,21, 417, 440 Walowa of Yimbo, 41 1

128, 164, 327 see abo cloth Turks see Ottoman empire Wanga, 414
Tldjaniyya brotherhood, 163 Tutsi, 397, 399, 403, 404 Wangara, 232, 242, 243
Tigré, 345, 348, 350, 351, 365 Tüwät, 153,163, 164 wamors, 32, 52, 53, 141, 142,
Tikar, 260-1 Twa, 415 143-6, 151,156, 165,297,
Timbuktu, 19,150, 154, 156, Tyapi, 189, 191,196 305
157-8, 159, 160, 161-3, 164, Tyokosi, 32, 34 wars, 329, 390
165 Washana, 410, 414
Tio, 22, 269, 270, 271-2, 274, Ubangi, 30, 35, 36, 261-3, 265, Wassa, 206, 211,212, 214
277,278,281,287 268 Wasuya, 410-11,414
Tlemcen, 1, 106, 122, 123, 126 Ubangian languages, 29, 103 Wattäsids, 104, 106
tobacco, 9, 45, 46, 265, 270, 280, Uele, 261-3 We',' 193
367 Uganda, 36, 100, 103, 385, 386, weapons see firearms
Togo(c.l690-1710), 102 396 weaving, 18,220,242,252
Tonga, 36, 300, 307, 311,315 Ugweno, 414 Wenya state, 313-14
Toposa, 387, 389 underdevelopment, 41, 42, 47-50, West Africa, 1, 3, 10, 232, 242-3
Toro, 390 53, 210 architecture, 16 Atlantic trade,
Torodo revolution, 17, 147, 150-1 United States, 9, 24, 45, 46, 49, 196-200 Christianity, 445
Torwa state, 315, 322, 323, 327, 50,55,60,61-2, 119, 132-3 Dutch in, 7, 195, 198 growth
328,331,333 see also North America of states, 442-3 industnes, 17-
Toussaint L'Ouverture, 62-4, 69 Upare, 415 18 Islam, 150 mulatto groups,
trade routes, 15, 37, 38, 76, 95, Upemba Depression, 290, 291 , 201-3 population movements,
157, 161-4,209-10,242 296 30, 37 slave trade, 4, 196,
trading companies, 6, 7, 9, 10, Upper Egypt, 76, 77, 85, 96 208-9,221,223-4,440
139-40, 198, 200, 202, 203, Upper Guinea Coast Atlantic West Atlantic peoples, 187, 190,
418, 438-9 trade, 196-200 establishment 192, 195-6
trading posts, 21-2, 24, 139-40, of sea routes, 193-5 growth of West Indies see Caribbean
143, 195, 207, 210, 223 states, 442 Islam, 446 Mulatto Western Europe, 1 and Atlantic
transhumance, 25, 26, 29, 100, groups, 201-3 population economic order, 41, 43-7fT.,
326-7, 338, 404 movements, 188-93 443-4 and slave trade, 9-11 see
trans-Saharan trade, 19, 140, 141, Portuguese in, 195, 196, 198 also Europe
163, 164-5. 224. 235, 249 societies, 187, 195-6 sources Western Sudan, 1, 13, 14, 15, 16,
Index 491
17, 18, 22, 207, 232 famines, Yao, 37, 300, 301,302,311-12, Northern Zambezia; Southern
161 Islam, 156-7, 165 and 314,372,378,379 Zambezia
outside world, 160 post- Yarse, 37 Zambia, 290, 297, 300, 302, 303,
Songhay period, 156-60 Yemen, 66, 76, 93, 351 306, 31 2 see abo Northern
predatory economy in, 19 Yerim Kode, 141, 142 Zambezia
trade, 161-5 Yoruba, 13, 14, 16, 17, 18,221, Zande, 36-7, 103, 262, 263
White Nile, 35, 100, 101,102, 223,224-6,231,242,243 Zanzibar, 20-1, 66, 368, 371, 376,
103 377, 378, 379
Whydah, 223 Zafikasinambo, 420, 421 Zarabehava, 422
Wolayta, 347, 364, 365-6, 442 Zafimanara, 435 Zarma, 237
Wolof, 6, 14, 18, 143, 151 Zafindraminia, 420-2, 425 Zawila, 244
world economy, 55, 122, 198, Zîhir 'Umar, 83,85 Zayyänids, 122, 123
335, 440, 441-2 role of slave Zaire, 266, 268, 290, 292, 293, Zera-Yakob, emperor (1434-68),
trade in, 39-4 Iff. 303 347, 348
Wuh, 136 Zaire nver, 29, 30, 273, 274, 286, Zimba, 32, 305, 373-4, 378, 380
287 Zimbabwe, 28, 35, 36, 37, 304,
Xaaso, 146, 151 Zamfara, 234, 236, 237, 238, 239, 327, 372, 422-3
Xhosa, 29, 338, 339, 343, 344 242, 245, 246 Zinza, 407, 412
Zambezi nver, 18, 290, 297, 298, Zulu, 344
YaasinBubu, 141-2 300, 302, 305, 306, 307, 310, Zumbo, 316, 322,331,333
Yaka kingdom, 286, 289, 297, 298 311,315, 320 seealso
General History of Africa
The History sheds light both on the historical By taking stock of our knowledge of Africa,
unity of Africa and also on its relations with putting forward a variety of viewpoints on
the other continents, particularly the African cultures and offering a new reading
Americas and the Caribbean. For a long time, of history, the History has the signal
all kinds of myths and prejudices have advantage of showing the light and shade
concealed the true history of Africa from the and openly portraying the differences of
world at large. opinion that may exist between scholars.
Reviews of complete edition

'The UNESCO first fruits should be acclaimed as 'These volumes are sumptuously produced but
a real contribution to scholarship.' remarkably cheap. . .They must represent some
Roland Oliver, Times Literary Supplement of the best value for money any book buyer will
find. . .'
'One ofthe most ambitious academic projects to Basil Davidson, Third World Quarterly
be undertaken in this century.'
West Africa
V Africa from the Sixteenth to the Eighteenth Century
Volume V covers the history of Africa distinct social classes and, often, a
from the beginning of the sixteenth strongly feudal character. Traditional
century to the close of the eighteenth religions continued to coexist with both
century. Two major themes emerge: first, Christianity (suffering setbacks) and Islam
the continuing internal evolution of the (in the ascendancy).
states and cultures of Africa during this
Along the coast, particularly of West
period; second, the increasing
Africa, Europeans established a trading
involvement of Africa in external trade -
network which, with the development of
with major but then unforeseen
New-World plantation agriculture,
consequences for the whole world. became the focus of the international
In North Africa, the Ottomans conquered slave trade. The immediate consequences
Egypt and established Regencies in Tripoli, of this trade for Africa are explored, and it
Tunis and Algiers. South of the Sahara, is argued that the long-term global
some of the larger, older states collapsed consequences include the foundation of
(Songhay, Western Sudan, Christian the present world-economy with all its
Ethiopia), and new power bases emerged inbuilt inequalities.
(Asante, Dahomey, Sakalava). Highly
210x148 512pp
centralized political and administrative Bibliography Index
structures developed and societies with Illustrations, Maps, Diagrams throughout
JAMES CURREY- CALIFORNIA- UNESCO

Berkeley 94720
ISBN 92-3-102497-3
Printed in the United Kingdom 789231"024979

African Histoty V

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

African Histoty V

Uploaded by

Copyright:

Available Formats

UNESCO International Scientific Committee for the Drafting of a General History of Africa

V Africa from the Sixteenth

to the Eighteenth Century

JAMES CURREY CALIFORNIA UNESCO

Africa from the

Sixteenth to the Eighteenth Century

UNESCO General History of Africa

Volume I Methodology and African Prehistory

Volume II Ancient Civilizations of Africa

Volume HI Africa from the Seventh to the Eleventh Century

Volume IV Africa from the Twelfth to the Sixteenth Century

Volume V Africa from the Sixteenth to the Eighteenth Century

Volume VI Africa in the Nineteenth Century until the 1880s

Volume VII Africa under Foreign Domination 1880-1935

Unabridged Paperback Edition

Volume VIII Africa since 1935

Assistant Editor C. Wondji

is published by the following publishers

In Ghana, Sierra Leone,

P.O. Box CT3499

Cantonments - Accra, Ghana

P.O. Box 45314

Claremont 7735, South Africa

Dar es Salaam, Tanzania

P.O. Box 488

P.O. Box 32379

In the United States of America and Canada by

General History of Africa V

Africa from the

Sixteenth to the Eighteenth Century

JAMES CURREY CALIFORNIA UNESCO

James Currey Ltd

University of California Press

© UNESCO 1992 and 1999

ISBN (UNESCO): 92-3-102497-3

British Library Cataloguing in Publication Data

Library of Congress Cataloging-in-Publication Data

Typeset in Monotype Bembo 11/11 pt by Long House, Cumbria, UK,

Preface by Amadou-Mahtar M'Bow, former Director-General of vii

2 African political, economic and social structures during this period 13

3 Population movements and the emergence of new socio-politicalforms 25

5 The African diaspora in the Old and New Worlds 56

6 The Ottoman conquest of Egypt 70

7 The Sudan, 1500-1800 89

10 Senegambia from the sixteenth to the eighteenth century: 134

11 The end ofthe Songhay empire 153

12 From the Niger to the Volta 166

13 The states and cultures ofthe Upper Guinea coast 187

14 The states and cultures ofthe Lower Guinea coast 204

15 Fon and Yoruba: the Niger delta and Cameroon 221

16 The Hausa states 232

11 Känem-Borno: its relations with the Mediterranean Sea, Bagirmi 248

18 From the Cameroon grasslands to the Upper Nile 259

19 Tlie Kongo kingdom and its neighbours 273

20 The political system of the Luba and Lunda: 290

21 The northern ZambeziaLake Malawi region 300

22 Southern Zambezia 315

23 Dependence and interdependence: southern Africa, 1500-1800 335

24 The Horn of Africa 345

25 East Africa: the coast 368

26 The Great Lakes region, 1500-1800 381

28 Madagascar and the islands ofthe Indian Ocean 418

29 The history of African societies, 1500-1800: conclusion 440

Former Director-General of UNESCO (1974-87)