. 1
( 12)



This history of modern Africa takes as its starting point the year 1800,
because, although by that time the greater part of the interior of Africa
had become known to the outside world, most of the initiatives for po-
litical and economic change still remained in the hands of African rulers
and their peoples.
The book falls into three parts. The ¬rst describes the precolonial his-
tory of Africa, while the middle section deals thematically with partition
and colonial rule. The third part details the emergence of the modern
nation states of Africa and their history. Throughout the 200 years cov-
ered by the book, Africa, and not its invaders, is at the centre of the story.
The authors are as concerned with the continuity of African history as
with the changes that have taken place during this period.
The new edition covers events up to the middle of 2003, and takes
account of the fresh perspectives brought about by the end of the Cold
War and the new global situation following the events of 11 September
2001. It is also concerned with the demographic trends that are at the
heart of so many African problems today, with the ravages of diseases
such as HIV/AIDS and malaria and with the con¬‚icts waged by warlords
¬ghting for control of scarce resources.
Africa Since 1800

Fifth Edition
©¤§ µ®©© °
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press
The Edinburgh Building, Cambridge  µ, UK
Published in the United States of America by Cambridge University Press, New York
Information on this title: www.cambridge.org/9780521836159

© Cambridge University Press 1967, 1972, 1981, 1994

This book is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.

First published in print format 2005

©®-± ·-°-µ±±-±°·- eBook (NetLibrary)
©®-±° °-µ±±-±°·- eBook (NetLibrary)

©®-± ·-°-µ±-±µ- hardback
©®-±° °-µ±-±µ- hardback

©®-± ·-°-µ±-µ·-° paperback
©®-±° °-µ±-µ·- paperback

Cambridge University Press has no responsibility for the persistence or accuracy of
µ¬s for external or third-party internet websites referred to in this book, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.

List of Maps page vii












vi Contents














Suggestions for Further Reading 383
Index 389

1 Northern Africa: geographical features and
vegetation. page 2
2 Northern Africa in 1800. 9
3 Africa south of the equator: geographical features
and vegetation. 19
4 Africa south of the equator in 1800. 22
5 North-East Africa: Egyptian expansion. 38
6 North-East Africa: Ethiopian expansion and
the Mahdiyya. 48
7 North-West Africa, 1800“1881. 54
8 West Africa, 1800“1875. 70
9 Western Central Africa, 1800“1880: trade routes. 79
10 Western Central Africa, 1800“1880: tribal areas
and migrations. 84
11 Eastern Central Africa, 1800“1884. 93
12 Early nineteenth-century migrations in South
and East Africa. 108
13 Southern Africa, 1800“1885: African migrations. 109
14 Southern Africa, 1800“1885: Boer migrations. 113
15 Europe at the time of the partition of Africa. 121
16 Africa on the eve of partition: African states and
European settlements. 124
17 European partition: Western Africa. 132
18 European partition: East Africa. 139

viii Maps

19 Southern Africa: the European partition “ Britain,
France, and Germany. 141
20 Southern Africa: the European partition “ Leopold and
Portugal. 142
21 Africa: the ¬nal stage of partition, 1914. 144
22 Africa: colonial economies and administrations. 152
23 Africa and the First World War. 172
24 The Maghrib: economic development during the
colonial period. 185
25 North-East Africa under colonial rule: economic and
political development. 186
26 The independence of Africa. 227
27 South Africa and the Bantustans. 292
28 Nigeria: four decades of independence. 309
29 Africa and the Cold War. 316
30 The new South Africa. 321
31 Con¬‚icts in the Horn of Africa. 341
32 Crises in Rwanda and Congo (Zaire). 344
33 Warlords in West Africa. 350
34 Sudan: North vs. South. 357
35 Oil in Africa. 379
ONE. Africa North of the Equator

The Sahara and Islam: The Bonds Unifying Northern Africa

The geography of the northern half of Africa is dominated by the Sa-
hara desert. Throughout its vast area, 2,800 km (1,700 miles) from
north to south and nearly 8,000 km (5,000 miles) from east to west,
rainfall is less than 13 cm (5 inches) a year. Except around a few
oases where underground supplies of water reach the surface, agri-
culture is impossible, and the desert™s only inhabitants have been
nomadic herdsmen, breeding camels and moving their animals sea-
sonally from one light grazing ground to another. To the north of
the desert lies the temperate Mediterranean coastland “ its rainfall
concentrated between January and March, with wheat and barley
as its main cereal crops and sheep, the main stock of its highland
pastures. Southward are the tropics, the land of the summer rains,
favouring a different set of food crops from those grown around the
Mediterranean. In the desert and northward live Berbers and Arabs,
fair-skinned peoples speaking languages of the Afroasiatic family.
South of the desert begins the ˜land of the blacks™ “ to the Greeks;
˜Ethiopia™, to the Berbers, ˜Akal n™Iguinawen™ (Guinea); and to the
Arabs, ˜Bilad as-Sudan™.
The desert has always been a formidable obstacle to human com-
munication, but for two thousand years at least “ since the introduc-
tion of the horse and the camel made travel easier “ people have per-
severed in overcoming its dif¬culties. Before the days of the motorcar
and the aeroplane, it took two months or more to cross. Nevertheless,

1. Northern Africa: geographical features and vegetation.
Africa North of the Equator 3

people did cross it, not merely in isolated journeys of exploration,
but, regularly, year after year, in the course of trade, education, and
pilgrimage. The essential intermediaries in this traf¬c were the pas-
toral nomads of the desert itself. They bred the camels, trained them
for carrying, and accompanied and protected the caravans on their
journeys. They also controlled what was, until the twentieth-century
discoveries of oil and natural gas, the one great natural resource of
the Sahara, which was the salt deposited in almost inexhaustible
quantities by the evaporation of ancient lake basins situated in the
very middle of the desert, dating from prehistoric periods of much
greater rainfall. The salt was in high demand to the north, and more
especially to the south of the desert. The nomads brought in slaves to
mine it and supplied the all-important camels to transport it in bulk.
Given the salt caravans, which by the nineteenth century were em-
ploying hundreds of thousands of camels to carry tens of thousands
of tons of salt, the exchange of many other commodities from north
and south of the desert becomes much easier to understand. The gold
from the tributary valleys of the upper Niger, the upper Volta, and the
Akan forest was an early and important element in the trans-Saharan
trade. Slaves, captured all along the southern edges of the Sudanic
belt, accompanied nearly every northward-moving caravan. And, as
time went on, leather goods and cotton textiles manufactured in the
Sudan were carried northwards in considerable quantity. The staples
of the southward traf¬c were the woollen textiles of North Africa; the
cottons and muslins of the Middle East; and the weapons, armour,
and other hardware of southern Europe.
Therefore, long before any sailing ship from Europe reached the
Atlantic coast of West Africa, the Sudanic lands to the south of the
Sahara were in touch with those of the Mediterranean not only
by exchanging produce but also by the sharing of skills and ideas.
Whereas the Latin Christianity of the Roman provinces never crossed
the Sahara, Greek-speaking missionaries, both Orthodox and Mono-
physite, converted the Nubian kingdoms on the upper Nile and the
kingdom of Aksum in northern Ethiopia. In the west, Islam ¬rst
spread through the conquest of Egypt and North Africa in the sev-
enth century, and then moved on across the desert with little delay.
By the ninth century, the nomads of the central and western Sahara
were converting to Islam. By the eleventh century, at least, the new
4 Africa since 1800

faith was beginning to penetrate the Negro kingdoms to the south of
the desert, where it appealed ¬rst and foremost to those who travelled
beyond their own communities and language areas as participants in
an already active system of regional and interregional trade. To them,
Islam offered wider intellectual and spiritual horizons and member-
ship in a universal brotherhood which looked after its members in
very practical ways. Between the eleventh and eighteenth centuries,
at least, the townsfolk of the Sudanic countries learned to be Mus-
lims like the Arabs and the Berbers to the north. Their learned and
pious men studied Arabic, the language of the Holy Koran, and a
few made the pilgrimage to the holy cities of Mecca and Medina,
passing through the great cities of Egypt and North Africa on the
way. The rulers and the rich men on both sides of the desert wor-
shipped the One God, read the same books, and discussed the same
It would, of course, be a great mistake to imagine that all the civil-
isations of the Sudanic belt of Africa were due to contact with the
world of Islam. We now know that a pattern of urban life in walled
towns existed in widely scattered parts of West Africa long before
the spread of Islam, and that the characteristic political formation of
small ˜city states™ grouped in clusters “ each cluster speaking a com-
mon language and observing common customs “ must have been a
development indigenous to the region. The periodic and sporadic in-
corporation of city“states into larger political hierarchies, described
by outsiders as kingdoms or empires, is likewise to be seen as a re-
sponse to various local factors, including differences of economic op-
portunity and military power and the ambitions of individual rulers,
and not as the transfer of political ideas from the north of the desert
to the south. Nevertheless, the growing presence of Islam and the
proximity of the Islamic heartlands as the most obvious point of ref-
erence in the outside world did help to provide a certain element
of unity to the northern half of Africa, extending from the Mediter-
ranean almost to the Atlantic coast of West Africa. Within all this vast
area, despite multitudinous differences of language and culture, in-
terregional trade and travel were practised by a small number of
people and, by the beginning of the nineteenth century, nearly all
of these were Muslims, so that there was a certain pool of common
ideas in circulation from one end of it to the other.
Africa North of the Equator 5

It is debatable just where the southern frontier of northern Africa
lay at different times in history. Until the twelfth and thirteenth cen-
turies, it probably included little more than the open grasslands,
forming a belt 500 or 600 km (300 or 400 miles) wide to the south
of the desert margin, from the Senegal to Lake Chad and eastwards
through Darfur and Kordofan to the Ethiopian highlands. Through-
out this region, beasts of burden, especially donkeys, could circu-
late, and troops of armed horsemen could control and levy tribute
upon the populations of quite large states. To the south again lay the
woodland belt, thickening progressively into dense equatorial rain
forest. Here, because of tsetse ¬‚y in the woodlands and lack of for-
age in the forest, all goods had to be carried by canoes or porters,
and soldiers fought on foot. Markets and states were smaller, and
there were few towns large enough for Islamic religion and learn-
ing to gain a foothold. Nevertheless, by the fourteenth and ¬fteenth
centuries, some interregional trade was beginning to penetrate even
these southern lands. Gold was found in the Akan forest, and kola,
the one luxury stimulant permitted by Islam, was grown exclusively
within the forest belt. When the Portuguese discovered the West
African coast, they found that the trading frontier of the Mande
traders from the Niger bend had already reached the coastline of
modern Ghana. During the three centuries that followed, the Euro-
pean traders operating from the Atlantic beaches pushed the fron-
tiers of interregional trade northwards again, but only by a matter
of 300 to 500 km (200 to 300 miles). By 1800, there was still far more
of West Africa which looked northwards for its contacts with the
outside world than southwards to the seaborne trade with Europe.
And, of course, throughout the whole vast region to the east of Lake
Chad, there remained no other source of outside contacts but the
northern one.

Countries of the Mediterranean Coast
By the end of the eighteenth century, people in the Muslim world as
a whole had lost much of the energy and sense of purpose that had
driven them to produce such a brilliant culture in the early centuries
of Islam. They failed to keep abreast of the new inventions and tech-
niques being discovered in western Europe, particularly in military
6 Africa since 1800

affairs and transport “ such as the improvements made to sailing
ships. This failure to make progress affected all parts of Africa con-
sidered in this chapter in one way or another. It applied especially to
the lands north of the Sahara, from Egypt in the east to Mauritania
in the west. Since the sixteenth century, all these countries, except
Morocco, had formed a part of the Turkish Ottoman empire, with its
capital at Istanbul. By the eighteenth century, Ottoman power had
declined considerably from the peak reached two centuries earlier.
Provincial rulers now acted almost independently of the Ottoman
sultan, and even tribute payments had become fairly nominal. While
the situation varied in detail from one country to another, the ruling
elite in all of them was composed of the descendants of the original
Turkish garrisons, who augmented their numbers in each generation
and their sense of separateness from the local people by recruiting
slave soldiers from the northern con¬nes of the Ottoman empire in
southern Russia and the Caucasus. It was in supplying these recruits
that the Ottoman sultans came nearest to ruling their North African
It was in Egypt that the system of military rule had its deepest
roots. When they conquered the sultanate in 1517, the Ottomans took
over an institution which had been operating since the thirteenth
century; the elite cavalry soldiers, imported as slaves but then edu-
cated and trained to occupy a highly privileged status, were known
as mamluks. At the end of their professional service, they were freed
and generously pensioned. Their children, however, were forbidden
to enter the army. The commanders of the mamluks, the amirs, di-
rected the military and civil services of the state, and each amir on
his appointment imported a fresh supply of mamluks to be his body-
guard or ˜household™. Much of the cultivable land of the country was
divided into ¬efs for the support of the military elite, and for the peas-
ant millions of Egypt who toiled and were taxed Mamluk rule was
a harsh one. But it supported a leisured and educated class which
made Cairo, at least, a centre of luxury and learning as outstanding
as any in the Muslim world.
To the west of Egypt, the countries of Tripoli, Tunis, Algeria, and
Morocco were known collectively to the Arabs as al-Maghrib (the
West). Here, in contrast with Egypt, the authority of governments
rarely extended far beyond the main cities. In the hinterlands lived
Africa North of the Equator 7

¬ercely independent nomadic tribes “ both Berbers and bedouin
Arabs. They could only be rather loosely controlled by playing off
one against the other. In Tripoli, the Ottoman government had been
represented since 1711 by the local Karamanli family, which had con-
centrated its efforts mainly on developing the trans-Saharan trade
from Bornu and the Hausa states. By this route came a steady sup-
ply of Sudanese slaves, who were distributed by the merchants of
Tripoli to Istanbul, Damascus, Cairo, and all over the western part
of the Muslim world. Tripoli was likewise a distributing centre for
the splendid leatherwork of the Hausa cities, already well known in
western Europe as ˜Morocco leather™. Tripoli™s exports southwards
consisted mainly of arms, armour, and Arab horses. They also in-
cluded mercenary soldiers trained in the use of ¬rearms, who joined
the bodyguards of the Sudanese rulers.
Tripoli, however, had no monopoly of the trans-Saharan trade.
Probably, the largest hub of the desert traf¬c was at the oasis of
Ghadames, where the borders of Tunisia, Algeria, and Libya now
meet. Here, caravan routes from the central and western Sudan
converged, and Ghadames merchants were well known in Hausa-
land and Timbuktu alike. From Ghadames, some of the Sudan trade
was carried to Tunis and Algiers. These were busy ports from which
merchants could more easily reach the markets of western Europe
than from Tripoli. The rulers (beys) of Tunis had been drawn since
1705 from the local Hussainid family, whose armed forces protected
a large settled population from the attacks of the nomad Berbers
of the eastern Atlas. These peasant farmers of the Tunisian plain
were some of the greatest wheat producers of the whole Mediter-
ranean basin. In the coastal towns, a sophisticated middle class of
merchants and of¬cials, enjoying a long tradition of Islamic civili-
sation and learning, ran a more orderly system of government than
was possible in any other Maghrib country.
The Ottoman rulers of Algiers were known as deys. Unlike the
rulership of Tunis, this of¬ce had not fallen into the hands of a sin-
gle family, to be passed down from father to son. It was ¬lled on
the death of the reigning dey by election from among a group of
merchants and soldiers who were the most in¬‚uential men in the
city. The merchants, called corsairs by Europeans (from an Ital-
ian word meaning ˜to chase™), traded by sea with the European
8 Africa since 1800

countries, using galleys rowed by Christian slaves. Legitimate trade
was sometimes combined with acts of piracy against European ship-
ping and coastlines, for which they became infamous. During the sev-
enteenth century, Algiers had been one of the most attractive cities of
the Mediterranean. Dr Shaw, an English traveller in the early eigh-
teenth century, commented favourably on its surroundings:

The hills and vallies round about Algiers are all over beautiful with gardens and
country seats, whither the inhabitants of better fashion retire during the heats
of the summer season. They are little white houses, shaded with a variety of
fruit trees and ever-greens, which, besides the shade and retirement, afford a
gay and delightful prospect towards the sea. The gardens are all of them well
stocked with melons, fruit and pot-herbs of all kinds. The natives of Algiers live
extremely happy, for though the government is despotic, it is not so in reality.1

Even at the time of its surrender to the French in 1830, Algiers was
described as ˜perhaps the best regulated city in the world™. The French
conquerors found that the majority of the Algerines were better ed-
ucated than the majority of the local Frenchmen. And this was after
half a century of grave political disorders, due to revolts among the
Arab and Berber tribes who roamed over the high plateaux of the in-
terior behind the coastal plains. These tribes were led by marabouts,
the Muslim holy men, who carried their attacks to the very outskirts
of the cities on the Mediterranean shore.
Morocco had never been a part of the Ottoman dominions, and
its armies were composed mostly of black slaves from the Sudan,
of whom by the mid-eighteenth century there were said to be some
150,000 “ half of them housed in a specially constructed military
town, and the rest dispersed in forts guarding the central lowlands
of the country from the incursions of the Berber nomads of the Atlas.
The extent of territory paying tribute into the sultan™s treasury had
greatly diminished since the late sixteenth and early seventeenth cen-
turies, when for a few years the kingdom had stretched right across
the desert to Timbuktu. Now, as in Algeria, tribal groups from the
High Atlas and the desert fringes were apt to raid the settled areas
and extort their own tribute from the peasants of the plains, and the

T. Shaw, Travels and Observations Relating to the Several Parts of Barbary and the Levant (Oxford,
1738), p. 71.
2. Northern Africa in 1800.
10 Africa since 1800

armies of the sultan were powerless to prevent them. Nevertheless,
Morocco was still the terminus of a considerable trade with the
south. The salt trade conducted by the desert Berbers from near
its southern borders still attracted much of the gold of the upper
Senegal“Niger region, despite the attempts of the French at St Louis
to obtain it. Slaves from the Niger bend still went northwards to
Morocco in considerable numbers. Despite the opening of the sea
routes, a great many European manufactures, especially English cot-
ton goods, were distributed over West Africa by Moroccan merchants
who bought them at Mogador on the Atlantic coast of southern Mo-
rocco. Arabic-speaking Moors lived all over the western fringes of the
Sahara as far south as the banks of the Senegal. Their holy men prac-
tised the characteristic Maghribi forms of devotion, becoming fol-
lowers of such religious orders as the Ramaniyya, which was founded
in 1770 in Kabylia (Algeria), and the Tijaniyya, established in 1781 by
Sidi Ahmad Tijani at Ain Mahdi, near Laghouat, on the desert fringe
of Algeria. Tijani™s teaching was accepted at the great Moroccan uni-
versity at Fez, and his order prospered in Mauritania and among the
Tuareg tribesmen of the central Sahara.

States of the Sudan Region
Most of the larger states existing to the south of the Sahara were
affected in some degree by the stagnation of the Islamic world as
a whole. Generally, there was less security for traders and pilgrims
than there had been and, in consequence, less wealth, less learn-
ing, and less religion. Among the states most affected was Ethiopia,
which, though Christian in religion, lay close to the Arabian heart-
lands of Islam and suffered severely from the decline in the trade of
the Red Sea. More seriously still, Ethiopia had been suffering since
the sixteenth century from the progressive in¬ltration and settlement
of its southern and eastern provinces by its Oromo neighbours. The
Christian monarchy had responded to the situation by moving its
headquarters away from its traditional medieval bases in Shoa (the
region around modern Addis Ababa) to the region north of Lake
Tana, where it ¬nally established a ¬xed capital at Gondar. But this
attempt at regroupment had not been successful. The Oromo pres-
sure continued, and the new frontiers could only be defended by
Africa North of the Equator 11

attracting Oromo contingents into the royal service by granting them
the right to raise tribute from the Christian peasantry. Jesus II (1730“
55) was the last of the eighteenth-century emperors to exercise any
real authority outside the Gondar region. After his death, the feudal
rulers of the other provinces became virtually independent, and rival
emperors were recognised by different factions. By the early nine-
teenth century, one of them had fallen into such abject poverty that,
when he died, there was not enough money in the treasury to pay
for a cof¬n.
Bordering Ethiopia to the west lay the territory of the Funj sul-
tanate, which in the sixteenth century had replaced the southernmost
of the Christian kingdoms of Nubia. The Funj made their capital at
Sennar on the Blue Nile, which became the centre of a rich trade in
Ethiopian gold and Sudanese ivory, which were exported to Jiddah
in exchange for Indian textiles. By the mid-eighteenth century, how-
ever, the dynasty was becoming increasingly dependent on its black
slave troops. When the Scotsman James Bruce, who wrote a marvel-
lous description of his dangerous travels in Ethiopia, passed through
the country on his way down the Blue Nile in 1770, there was still a
standing army of 1,800 horses and 14,000 infantry, but control of the
western province of Kordofan had been lost, and trade had shrunk to
a trickle. Bruce was not impressed by what he saw. ˜War and treason™,
he wrote, ˜appear to be the only occupations of this horrid people,
whom Heaven has separated by almost impassable deserts from the
rest of mankind.™2
More vigorous than the Funj sultanate were now those of Darfur
and Wadai, one on each side of the modern frontier between the
Sudan Republic and Chad. Though nominally Muslim, it is clear
from the early nineteenth-century accounts of Muhammad al-Tunisi,
a distinguished Arab scholar who made voyages of exploration to
both countries, that here, as in many states of the western Sudan,
ancient pre-Muslim ideas of divine kingship still persisted. Al-Tunisi
described an annual ceremony of re-covering the royal drums, when
a boy and a girl were sacri¬ced; he also described the sultans taking
part in almost Pharaonic rituals of seedtime and harvest:

James Bruce, Travels to Discover the Source of the Blue Nile 1768“1773 (London, 1790), vol. 4,
p. 437.
12 Africa since 1800

At the beginning of the planting season the Sultan rides out in great pomp,
escorted by more than a hundred young women, by his slave boys and by a troop
of ¬‚ute-players. When he reaches the open ¬elds, he dismounts from his horse,
takes different kinds of seeds and sows them, while a slave hoes the ground.3

The wealth of these two states were derived from the slaves raided
by the armies of horsemen from among the animist peoples living
to the south of them, who lacked any kind of state organisation.
Both sultanates traded in copper from the rich deposits at Hofrat
en-Nahas near the headwaters of the Bahr al-Ghazal, and both sent
their trading caravans northwards by the Darb al-Arba ˜in (the Forty
Days™ Road), which reached the Nile valley at Asyut.
At the end of the eighteenth century, the ancient empire of Kanem“
Bornu, with its wide territories to the east and west of Lake Chad,
was still the most stable and civilised of the Sudanic states. To be
sure, it had lost some of the imperial outreach achieved in earlier
periods when it had pushed its military garrisons northwards to oc-
cupy the salt mines at Bilma and to touch the southern marches
of Tripoli in the Fezzan. But its rulers were still pious and literate
Muslims and, throughout their kingdom, justice was administered
according to Islamic law. The interregional trade of Kanem“Bornu
was indeed based upon the horri¬c slave raids annually conducted
by the armies of the ruler (mai) upon the stateless peoples of the
Mandara mountains and other peripheral regions to the south; but,
within the state™s boundaries, there was peace and prosperity of a
kind, which impressed all foreign travellers. The capital at that time
was the brick-built town of N™gazargamo, some 95 km (60 miles) west
of Lake Chad “ the ruins of which can still be seen and which cover
an area about 3 km (2 miles) in diameter. Here, the mais lived a dig-
ni¬ed and secluded existence, supported by the tribute collected for
them by their provincial governors from the peasantry of a kingdom
measuring perhaps 1,000 by 500 km (600 by 300 miles), much of it
light grazing land, but including also the densely populated alluvial
farmlands west of Lake Chad.
In terms of interregional trade, the most active part of the central
Sudan was not Kanem“Bornu but the city“states of Hausaland lying

3 ˆ
Muhammad bin ˜Umar al-Tunisi, Voyage au Oudday, tr. Dr Perron (Paris, 1851), p. 159.
Africa North of the Equator 13

along its western ¬‚ank. Though never united, these Hausa cities were
unique in the Sudan in that they possessed manufacturing industries
on a really important scale. Weaving, dyeing, leatherwork, glassmak-
ing, smithing, and metalwork of every description were carried out,
mostly by slave artisans living in the towns and fed by agricultural
slaves living in special villages in the surrounding countryside. At the
end of the eighteenth century, Katsina was still the leading city; soon
it was to be Kano, with Zaria a close third. It was from these cities
rather than from N™gazargamo that the great caravan routes radi-
ated outwards across the Sahara to Tripoli and Ghadames, and from
there to Tunis and Algiers. Bornu, with its cavalry armies, supplied
most of the slaves that were exported northwards. The manufactures
came from Hausaland, and their distribution covered the whole of
northern Africa.
South of the Sahara from the Maghrib, the powerful empires of
the medieval western Sudan had broken up by the end of the eigh-
teenth century into many weak kingdoms. The great empire of the
Songhay Askias, which had stretched in the sixteenth century from
the upper Senegal to the frontiers of modern Nigeria, had come to an
end with the Moroccan invasion of 1591. At the battle of Tondibi, the
Moroccans used ¬rearms for the ¬rst time against Sudanese cavalry
and foot soldiers armed only with bows and spears. The conquerors
settled down, and their descendants formed a new ruling class, the
arma (shooters, gunmen), which soon became independent of the
Moroccan sultan, the soldiers electing their own pashas (rulers) at
Timbuktu and their kaids (governors) in the garrison towns around
the loop of the Niger bend from Jenne to Gao.
After the Moroccan conquest, what remained of the ruling class of
Songhay retreated down the Niger and set up an independent gov-
ernment in the southern province of Dendi. Upstream from Jenne, on
the western side of the Niger bend, the Mande subjects of the Song-
hay empire reverted to their pre-imperial pattern of government as a
cluster of small states, each centred around a little walled town called
a kafu. Here and there, as at Segu and Kaarta, successful warlords,
getting ¬rearms in exchange for war captives, managed to establish
some larger states “ each incorporating several kafus. However, the
most vital region of the western Sudan during the seventeenth and
eighteenth centuries was the far west. Here, in Futa Toro to the south
14 Africa since 1800

of the middle Senegal, there had sprung up during the medieval pe-
riod a people of mixed Berber and Negro descent known as the Fulbe
(French: Peul; Hausa: Fulani). Unlike the states of the Mande, those
of the Fulbe included many groups of nomadic pastoralists, who
herded their cattle in the parts of the country which were too arid for
agriculture even during the wet season and brought them south-
wards into the farmlands to manure the ¬elds after the cereal
harvest had been gathered. Unlike their farming relatives, the pas-
toralists migrated far and wide in search of new grazing grounds
and established similar relationships with the farmers of other lan-
guage groups. As early as the ¬fteenth century, they were spread out
in small communities all over the savanna belt of West Africa to as
far east as Lake Chad. Until the eighteenth century, these scattered
Fulbe were mostly animist in religious belief. However, those Fulbe
who remained in their Futa Toro homeland were converted to Islam
by marabouts from Mauritania in the sixteenth and seventeenth cen-
turies and joined Muslim brotherhoods “ full of zeal for their new
faith. One Fulbe clan, the Torodbe, became the missionary and cler-
ical leaders of the whole of the Fulbe people. Wherever Fulbe were
dispersed, Torodbe preachers would be found teaching the necessity
for conversion and Islamic reform and advocating jihad, the holy
war, as the means of obtaining it. The twin bases of the Fulbe move-
ment were Futa Toro and Masina, a Fulbe-led state to the south-west
of Timbuktu, once tributary to Songhay, which became independent
after the Moorish conquest. From Masina in the eighteenth century,
Torodbe missionaries carried the jihad to Futa Jallon, the mountain
country on the borders of Guinea and Sierra Leone, and eastwards
across the Niger bend to Say. From this background of missionary
zeal and holy war, the great Fulbe jihad of the early nineteenth cen-
tury arose.
It is often said that the seventeenth and eighteenth centuries were
a period of decline in the western Sudan, and certainly it was a time
of great political disorder. But, thanks to the Torodbe and other reli-
gious leadership groups, much of the learning of the medieval Sudan
was kept alive. It is possible that by the end of the eighteenth cen-
tury, both Islam and Arabic education were more widely spread than
they had been during the great days of the medieval empires. De-
spite the political disorders, trade continued to ¬‚ow. There seems to
have been a breakdown in communications between the western and
Africa North of the Equator 15

the central Sudan. This was because, after the defeat of Songhay, no
other power was able to control the ¬erce Tuareg nomads living to
the north and east of the Niger bend. The routes running north-west
from Timbuktu and north from the Senegal remained open and ac-
tive well into the nineteenth century “ more active by far than the
routes from Timbuktu westwards to the Atlantic coast.

States of the Woodland and the Forest

By the eighteenth century, the opening of the Atlantic trade by the
Europeans had made a crucial change only among the states of the
woodland and forest zones to the south of the savanna belt. From
their earliest days, these states had all in their external relations faced
towards the north. The Akan states of modern Ghana and the Ivory
Coast had looked towards Mali, Songhay, and the successor states
of the Mande, while the states of the Yoruba- and Edo-speakers in
southern Nigeria looked towards Hausaland. As Samuel Johnson,
the historian of the Yorubas, pointed out,

It should be remembered that light and civilisation with the Yorubas came from
the north . . . The centre of light and activity, of large populations and industry,
was therefore in the interior, whilst the coast tribes were scanty in number, ig-
norant and degraded, not only from their distance from the centre, but also
(later) through their demoralising intercourse with Europeans and the slave

It was the same in Ghana and the Ivory Coast as it was in Nigeria.
The most important of the woodland and forest states had at ¬rst
been those on the northern side. The smallest and the most back-
ward, populated only by ¬shermen and salt-boilers, had been the
little states on the coast.
The Atlantic slave trade, which was begun by the Portuguese in the
late ¬fteenth century when a trickle of Africans were shipped across
the ocean “ ¬rst to Europe and later to the Spanish and Portuguese
colonies in the New World “ had developed by the end of the seven-
teenth century into a steady ¬‚ood. Most of the European maritime
countries then took part in it, especially Britain and France. The
brisk competition for slaves among the rival Europeans meant that

Samuel Johnson, History of the Yorubas (Lagos, 1921), p. 40.
16 Africa since 1800

the states at or near the coast had easy access to ¬rearms, which they
used at ¬rst in wars of conquest against their neighbours and even-
tually against the formerly more important inland states. Within the
woodland and forest states, there had been considerable change in
the balance of power by the end of the eighteenth century. The rising
states were those based near the coast, especially Asante and Da-
homey, which had grown as a result of the use of ¬rearms acquired
through the Atlantic trade. The most dramatic demonstration of this
shift in power occurred in 1745, when the musketmen of Asante de-
feated the armoured cavalry of Dagomba who, in any earlier period,
would have chased them mercilessly out of any open country they
had dared to enter. This process, begun in the seventeenth and eigh-
teenth centuries, was carried much further in the nineteenth.

The Encircling Power of Europe

The only signi¬cant imperial power in the northern half of Africa
at the end of the eighteenth century was Ottoman Turkey. Its de-
pendencies in Egypt, Tripoli, Tunis, and Algiers were admittedly
almost ˜self-governing™, but they did at least contribute some rev-
enue to the sultans in Istanbul. The European powers trading with
North and West Africa had, in contrast, nothing but a few footholds,
in the shape of forti¬ed trading factories scattered along the West
African coast, from St Louis on the Senegal to Whydah in B´ nin. e
These forts, whether British, Danish, Dutch, French, or Portuguese,
were designed mainly to protect the operations of one group of Eu-
ropean traders from the competition of others. Few of the castles
could have withstood a determined attack by local Africans, and
their governors had to be circumspect in exercising any jurisdiction
outside the walls. They carried on their trade with the help of African
middlemen living in the adjacent towns. Although by the end of the
eighteenth century, they were exporting around 100,000 slaves a year
from West Africa alone, the Europeans seldom captured a slave by
themselves and, save for the French on the Senegal, it was the rarest
thing for any European to venture 20 km (12 miles) inland. Nor did
there seem at any time in the eighteenth century the slightest likeli-
hood of a change in the pattern of these relationships, which were
satisfactory to both the European and the African traders. It is true
Africa North of the Equator 17

that by the end of the eighteenth century, the slave trade was un-
der attack in one or two European countries. A judge in an English
court declared in 1772 that there was no such thing as slavery on
English soil. And, ¬fteen years later, a group of philanthropists in
England bought a few square miles of the Sierra Leone peninsula
for the purpose of settling Negro slaves freed in England and across
the Atlantic in Canada. From this tiny beginning, the result of the
stirring of consciences among a few well-to-do men, truly ˜a cloud
no bigger than a man™s hand™, sprang the ever-growing ¬‚ood of Eu-
ropean interference in tropical Africa during the century to come.
But nobody at the time could have foreseen it.
To contemporaries, the change in the balance of power in the
Mediterranean must have seemed much more impressive than any
growth of European power in tropical Africa. To Britain and, there-
fore, to France, India and the routes to India were already a matter
of the most serious strategic importance. When these two powers
were locked in combat at the end of the century, it seemed a natural
move that Britain should forestall the French by seizing the Cape of
Good Hope from the Dutch. That Napoleon should reply by occupy-
ing Egypt was surprising only in that it ¬nally showed the weakness
of the Ottoman empire in relation to European military might. The
mamluk soldiers surrendered to the French in a single battle fought
near the Pyramids in 1798. The French were removed three years
later only through the powerful assistance of the British. It was all
very well for al-Jabarti, a citizen of Cairo and an eyewitness to these
events and the last of the traditional Muslim chroniclers of Egypt, to

The presence of the French in Cairo was intolerable . . . Muslims died of shame
when they saw their wives and daughters walking the streets unveiled, and ap-
pearing to be the property of the French . . . It was bad enough for them to see
the taverns that had been established in all the bazaars and even in several
mosques . . . The scum of the population was doing well, because it bene¬ted
from the new freedom. But the elite and the middle class experienced all sorts
of vexation.5

What had happened once could happen again. It was only surprising
that it did not happen for another eighty years.

Chronique d™Egypte 1798“1804, ed. and tr. Gaston Wiet (Cairo, 1950), p. 45.
TWO. Africa South of the Equator

The Lands of the Bantu

The geography and climate of Africa south of the equator is much
less simple than that of the northern half of the continent. Very
brie¬‚y, however, high and rather dry steppe country runs south from
the Ethiopian highlands through the middle of East Africa. It then
crosses over towards the western side of the subcontinent, ending
up in the Kalahari desert, with the dry lands of Botswana and the
Orange Free State on one side and those of Namibia on the other. On
the other hand, low-lying and distinctly humid country extends from
southern Cameroun right across the northern half of the Congo basin
to Lakes Tanganyika and Malawi, and from there it continues down
the Zambezi valley to the Indian Ocean coast and round through
southern Mozambique into Natal. In general, the steppe country is
more suited to pastoralism than to agriculture and, therefore, tends
to be only lightly populated. Equally, in the dense equatorial forest,
agriculture is only practicable in clearings and beside riverbanks
where sunlight can penetrate, and so here again population is very
thin and until quite recently was virtually con¬ned to the rivers and
the seacoast, where a little agriculture could be combined with ¬sh-
ing. The best conditions for food production are found in the bor-
derlands between the two zones, mostly therefore in the middle of
the subcontinent. This is where population is densest and where, by
the end of the eighteenth century, the most complex and centralised
political institutions were found.

Africa South of the Equator 19

3. Africa south of the equator: geographical features and vegetation.
20 Africa since 1800

One particularly striking fact about the peoples who live in Africa
south of the equator is that nearly all of them speak closely related
languages of a single subfamily called Bantu (from the common
word muntu, a man; plural bantu, people), of which the nearest rela-
tions outside the area are with the languages of southern West Africa.
The exceptions to this rule “ the people who do not speak Bantu lan-
guages “ are all found in the dry zones of the north-east (parts of
Kenya and Tanzania) and the south-west, where the practice of agri-
culture is dif¬cult. It looks, therefore, as though the Bantu-speakers
were the ¬rst agriculturalists in this part of Africa, and as though
the ˜cradleland™ from which they dispersed was the woodland region
to the north-west of the equatorial forest in what is today central
and southern Cameroun. Starting perhaps around 3,500 years ago,
the early Bantu spread southwards from this area, some groups ex-
panding across the Congo basin by the rivers until they reached the
savannas to the south of the forest, while others kept to the lighter
woodlands bordering the forest to the north, the east, and the south.
From these rather distinct lines of dispersion there developed, ¬rst,
a Western Bantu tradition, in which political leadership tended to be
associated with skill in metal-working “ above all, in the forging of
tools and weapons “ and (later) an Eastern Bantu tradition in which
rulers tended to be recognised by their possession and political ma-
nipulation of large herds of cattle.

The Western Bantu

For the Western Bantu, the region of light woodland extending for
800 to 1,000 km (500 to 700 miles) to the south of the forest of-
fered almost ideal conditions for Iron Age food production. Rainfall
was adequate, but not excessive. There was excellent ¬shing in the
northward-¬‚owing tributaries of the Congo River system and hunt-
ing in the strips of forest which lined the riverbanks, with plenty of
open country for agriculture between the streams. The whole region
was rich in iron ore, and there were large copper deposits both north
of the lower Congo at Mindouli and southwards in Katanga. In the
part of the region to the north and south of the lower Congo were peo-
ples speaking the Kongo and Mbundu languages, who early formed
small states based around dynasties which specialised in working
Africa South of the Equator 21

iron and copper into tools and weapons. At some time around the
fourteenth century, a local migration of well-armed Kongo-speakers
moved southwards across the lower Congo and conquered the north-
ern Mbundu. This resulted in the formation of a large state, covering
most of northern Angola. When it was discovered by the Portuguese
at the end of the ¬fteenth century, the Kongo kingdom was at the
height of its power. A sixteenth-century document preserved in
the Vatican archives describes the authority of the Kongo kings in
the following terms:

At the head of the Kongo kingdom is a king of kings who is the absolute lord
of all his realm, and none may intervene in any of his affairs. He commands as
he pleases. He is not subject to any law. The village chiefs have above all to take
care to collect from their subjects the taxes which are due to the king, and which
they each of them carry to the governor of their province. The governor presents
himself twice in each year at the royal capital in order to pay in the tribute, and
if the king is satis¬ed, he replies with the one word, wote, which means ˜you have
done well™. In this case the governor esteems himself highly favoured and makes
many clappings of his two hands. As a sign of his joy he throws himself on the
ground, covering his body with dust. His servants do the same, and then take
him on their shoulders, and go through all the city crying his praises. But if the
king does not say this word wote, he retreats greatly discom¬ted, and another
time he takes care to bring a larger tribute. The tribute is not ¬xed as to quantity:
each brings as much as he can. But if the governor does not do better, the king
addresses to him a strong reprimand, and takes away his post. Such a man then
becomes as poor as the most miserable of all the blacks.1

By the end of the eighteenth century, the Kongo kingdom was lit-
tle more than a memory. During the ¬rst century of its contact with
the Portuguese, it had overextended its frontiers through wars of
conquest, and thereafter the Portuguese found better slaving part-
ners in the independent Mbundu kingdoms to the south of Kongo,
near which in 1575 they established a permanent colony on the is-
land of Luanda, the nucleus of the future Angola. Other Europeans
traded with the Vili kingdoms north of the Congo estuary, where local
merchants organised successful long-distance trade routes into the
interior, using the rivers wherever possible. Kongo meanwhile disin-
tegrated, ¬rst into its constituent provinces and later into smaller di-
visions still. As the Portuguese colony slowly expanded its footholds

Vatican document cited in J. Cuvelier and L. Jadin, L™Ancien Congo d™apr` s les archives romaines
(Brussels, 1954), pp. 33“4.
22 Africa since 1800

4. Africa south of the equator in 1800.
Africa South of the Equator 23

northwards up the coast, the successor states of the Kongo king-
dom became its commercial hinterland, in which merchant families
wielded most of the in¬‚uence.
Eastwards from the sphere of the Kongo and Mbundu, right across
to Lake Tanganyika, the savanna country to the south of the equa-
torial forest was occupied by another set of Western Bantu peoples,
of whom the most signi¬cant were the Luba and the Lunda. The
Luba lived astride the headwaters of the Congo River, known in
these latitudes as the Lualaba, which along with its many tributary
streams ¬‚ows northwards from the copper-rich Katanga plateau,
which forms the watershed between the Congo and the Zambezi.
A little to the north of the watershed, the Lualaba ¬‚ows through a
sequence of lake basins known as the Upemba depression, where
a happy mixture of ¬shing and farming, combined with advanced
metal-working in iron and copper, gave rise early in the present mil-
lennium to a very dense population, organised in a series of small
kingdoms reminiscent of the Kongo model. The earliest states would
appear to have been formed by the Luba.
By the end of the eighteenth century, however, the two most impor-
tant kingdoms were those of the Lunda, farther to the west and south.
These were the kingdom of Mwata Yamvo, which occupied the whole
south-western corner of modern Democratic Republic of Congo, and
the kingdom of the Mwata Kazembe, astride the Luapula River in
southern Katanga. These great states, however, were but the centre
of a whole cluster of smaller ones, which ¬lled up most of southern
Congo, eastern Angola, and northern Zambia. The rulers of the two
big Lunda kingdoms, and of several of the outlying states, were ˜di-
vine kings™, in the style of those of Darfur and Wadai (see Chapter 1)
and of many other states in sub-Saharan Africa. They ate and drank
in secret, practised ˜royal incest™ with their queen sisters, shared rit-
ual authority with their queen mothers, and used spirit mediums
to communicate with their royal ancestors. As the symbol of their
authority, they used ¬re from the royal forge, from which burning
brands were carried annually to the headquarters of all tributary
The capitals of the Lunda kings, though not as permanent as
the towns of West Africa, were considerable centres of government
and trade. The palace population was large because the kings took
24 Africa since 1800

hundreds of wives drawn from all the main families in the coun-
try. The court of¬cials were numerous, and so were the skilled
craftsmen “ potters, smiths, weavers, basketmakers, brewers, wood-
carvers, huntsmen, and traders “ who congregated around the capital
and lived off the foodstuffs sent in as tribute from the surrounding
countryside. Describing such towns, which he visited in 1906, the
German explorer Leo Frobenius wrote:

When I penetrated into the region of the Kasai and the Sankuru I found villages
still existing whose principal streets were lined on both sides, and for miles on
end, with four rows of palm-trees, and whose charmingly decorated houses were
each of them a work of art. There was not a man who did not carry sumptuous
weapons of iron or copper, with inlaid hilts and damascened blades. Everywhere
there were velvets and silken stuffs. Every cup, every pipe, every spoon was a
piece of artistry, fully worthy of comparison with the creations of Europe.2

The Eastern Bantu

Although Western Bantu peoples settled as far south as Namibia,
in most of the region to the south of the Katanga copperbelt they
were interspersed with Eastern Bantu peoples whose ancestors had
expanded around the north-eastern fringes of the equatorial forest,
bringing their cattle, sheep, and goats and establishing a very dif-
ferent pattern of settlement in dispersed homesteads rather than
concentrated villages. Their earliest settlements in eastern Africa
were in the ˜interlacustrine™ region around Lakes Albert, Victoria,
and Tanganyika. From here, during the early ¬rst millennium A.D.,
they spread eastwards to the Indian Ocean coast and southwards
into central and southern Africa. Where their populations grew dense
enough, the Eastern Bantu, like their Western counterparts, formed
states on a monarchical pattern, and the process of competition and
conquest among the initial small states led gradually to the emer-
gence of some larger ones. In the interlacustrine region, by the end of
the eighteenth century six large states had grown up “ Buganda, Bun-
yoro, Ankole, Karagwe, Rwanda, and Burundi “ which are shown
on Map 4. Though not as big as the states of the Sudanic belt, all of
them probably had populations of half a million or more. As with

L. Frobenius, Histoire de la civilisation africaine (Paris, 1952), p. 15.
Africa South of the Equator 25

the pre-Islamic states of the Sudan, they were ruled by divine kings
who governed through elaborate hierarchies of court of¬cials and
provincial chiefs. As with the kingdoms of the Mwata Yamvo and
the Mwata Kazembe, many smaller states clustered around these
bigger ones, some of them paying tribute to one or other of the large
kingdoms, but most in practice independent of all of them.
The ¬rst European travellers who came upon this region in the
mid-nineteenth century felt that they were entering a new world.
They had walked 1,250 to 1,400 km (800 to 900 miles) from the east
coast along tortuous footpaths never more than a few inches wide,
through sparsely inhabited country where provisions were hard to
come by and even drinking water was often a problem. Along most of
their route, every day™s march brought them into the territory of some
petty potentate with whom they had had to negotiate permission to
pass. And then, suddenly, they found themselves in a world of plenty
and of order. Here a ruler™s writ or authority could run for a hundred
miles from his capital. His messengers sped along wide, well-beaten
roads to the provincial or district headquarters they were trying to
reach. In 1862, Speke and Grant stayed with the ˜ever-smiling™ King
Rumanika of Karagwe while runners were sent to announce their
arrival to Kabaka Mutesa of Buganda. Their passports granted, they
were accompanied for the rest of their journey by royal guides, and
food and lodging were arranged at the end of each day™s march.
Another region where large centralised states were developed by
Eastern Bantu peoples was that between the Zambezi and Limpopo
Rivers, in Zimbabwe and southern Mozambique. This was a region
rich in gold, copper, and ivory; where Indian Ocean traders had been
present from at least the tenth century; and where by about the
same date some concentrated settlements were being built with stone
walling and platforms indicating their function as capital towns. The
earliest examples were in or near the Limpopo valley, but the world-
famous site of Great Zimbabwe was built some 240 km (150 miles)
to the north of it in predominantly pastoral country on the southern
slopes of the Zimbabwe plateau, and was occupied and gradually im-
proved from the mid-eleventh until the mid-¬fteenth century, when
it was suddenly abandoned. A new centre of power then emerged
on the northern edge of the plateau, overlooking the Zambezi valley,
where there reigned a king of kings called the Mwene Mutapa, who
26 Africa since 1800

owned great herds of cattle which grazed over the plateau country,
and took tribute from the elephant hunters and ivory traders of the
Zambezi valley. When the Portuguese colonised the lower part of the
valley in the sixteenth century, they began to intervene in the succes-
sion struggles of the Mwene Mutapa™s kingdom by extending support
to their own favoured candidates. By the late seventeenth century,
the main ruling house had come so much under their in¬‚uence that
a rival dynasty, that of the Rozvi, claimed the paramountcy for their
state of Butwa, which controlled most of the region during the eigh-
teenth century. Although the medieval capital site at Great Zimbabwe
lay within this state, it was no longer a place of any importance. The
Rozvi capitals were built farther to the west, between Gweru and
Bulawayo, where ruins like those at Naletale and Dhlo“Dhlo have
yielded ¬nds dating to this period.

The Trade of Bantu Africa

The most striking fact about these Bantu states was that, unlike their
counterparts in the Sudanic belt of Africa, they were almost com-
pletely cut off from contact with the outside world. The larger Bantu
states were situated in the interior of the continent and, until after
the eighteenth century, they virtually lacked any of the Islamic in¬‚u-
ence which was so important a feature of the Sudan. The infestation
by tsetse ¬‚y was so widespread that beasts of burden were unknown,
although herds could be pastured in limited areas. Everything had to
be carried except on the rivers and, so far as we know at present, the
only water routes running deep into the continent during medieval
times were those which ascended the Zambezi and the Limpopo
from the coast of southern Mozambique. Here, the Swahili Arabs
from Kilwa and points farther north were trading cloth and beads
for gold and ivory at least by the tenth century. In the sixteenth cen-
tury, the Portuguese replaced the Swahili Arabs on the Zambezi and,
henceforward, this line of communication ran through their hands.
The more important part of the Portuguese contribution, how-
ever, was their opening of the Atlantic coast of Bantu Africa to
seaborne trade with Europe and South America. Their ¬rst venture
in this direction was with the Kongo kingdom, where they made
the Kongo kings their allies. They also supplied them with Christian
Africa South of the Equator 27

missionaries and technical and military aid. In the course of a hun-
dred years, however, their interests shifted southwards to Luanda
and Benguela, where they found it easier than in Kongo to obtain
slaves, which were needed in ever-increasing numbers to work the
sugar plantations of their colony of Brazil. The demand of the Por-
tuguese for slaves was supplied on the one hand by the little wars
fought by their Mbundu allies against other Mbundu living a lit-
tle farther away from their forts in the Kwanza valley, and on the
other hand by the purchases made by their African trading agents
(pombeiros, from Pumbe, a market on the Malebo Pool) in the inte-
rior markets. The slave-trader James Barbot observed in 1700 that

These slaves have other slaves under them, sometimes a hundred, or a hundred
and ¬fty, who carry the commodities on their heads up into the country. Some-
times these pombeiros stay out a whole year, and then bring back with them four,
¬ve or six hundred slaves.3

In this way, the Portuguese made indirect contact with the Lunda
kingdom of Mwata Yamvo and its many satellites. By the end of the
eighteenth century, the main Lunda kingdoms had acquired guns,
cloth, and other European luxuries, and their own industries had
developed greatly by learning from European examples. Splendid
axes and cutlasses were made by the Lunda smiths in the period fol-
lowing the European contact. Manioc (or cassava) “ the South Amer-
ican root crop “ was introduced by the Portuguese to feed the slaves
awaiting shipment to the New World, and soon became the staple
food for the whole of the southern Congo region.
The important region which remained right out of touch with the
outside world until late in the eighteenth century was the interlacus-
trine one. To the west of it lay the equatorial forest, to the north the
Nile swamps, and to the east the Kenya highlands, inhabited by war-
like pastoralists such as the Nandi and the Masai. The easiest line of
approach was from the south-east, but it was a long time before even
this was developed. Apparently, the early Swahili Arab communities
of the east coast had no contact with any but the coastal peoples
anywhere to the north of the Zambezi. Certainly, the Portuguese,
when they occupied this part of the coast during the sixteenth and

A. and J. Churchill, Collection of Voyages and Travels (London, 1732), vol. 5, p. 522.
28 Africa since 1800

seventeenth centuries, had no knowledge, even by hearsay, of the in-
terior that lay behind. It was, in fact, only when the Arabs of Oman
(in the Persian Gulf) seized the northern part of the east coast from
the Portuguese at the beginning of the eighteenth century that con-
tact with the interior began to develop; even then, the main agents
of it appear to have been the Nyamwezi people of western Tanzania,
who found their way down to the coast with ivory for sale. Through-
out the eighteenth and early nineteenth centuries, it was still the
Nyamwezi who organised and operated most of the carrying trade in
East Africa. Their caravans covered the whole region from Katanga
to the Indian Ocean. They marched great distances with heavy loads,
eating little and sleeping in the bush. The early European travellers
reported how the Nyamwezi boys used to prepare themselves for this
way of life by carrying small tusks on their shoulders as they went
about their home villages. Traditional history relates that it was in the
late eighteenth century that the ¬rst consignments of plates, cups,
saucers, knives, and cotton goods reached the kingdom of Buganda
at the heart of the interlacustrine region. From this time onwards,
the traditions tell of a steadily growing trade passing to the south of
Lake Victoria and through the Nyamwezi country to the Zanzibar

South Africa: Bantu and Boer

Another part of Bantu Africa which had almost no contact with
the outside world until late in the eighteenth century was the re-
gion south of the Limpopo in what is now the Republic of South
Africa. Two main groups of Eastern Bantu peoples lived here “ the
Sotho“Tswana on the plateau to the west of the Drakensberg moun-
tains, in what was to become the Transvaal and the Orange Free
State, and the Nguni peoples (Zulu, Swazi, Pondo, Thembu, Xhosa)
in the fertile and well-watered coastal lowlands of Natal and the
Transkei. So long as they still had room for expansion by clearing
the bush and occupying the more marginal lands, these peoples cre-
ated no centralised states. The Sotho lived in large, almost urban
concentrations of 10,000 to 15,000 people, their settlements often en-
closed within stone walls. When one settlement grew inconveniently
large, colonists were dispatched to found another. The Nguni lived in
Africa South of the Equator 29

dispersed homesteads, grouped in chiefdoms, each with around
10,000 subjects; and once again, when political units grew too popu-
lous, they divided. Only toward the end of the eighteenth century did
two Nguni groups living at the northern extremity of Nguni coun-
try begin to face the problem of overpopulation by recruiting their
young men into military regiments with a view to conquering and
subjugating their neighbours.
These groups were to emerge during the early nineteenth century
as the Zulu and Swazi nations. Until then, despite the ¬ssiparous po-
litical systems, the South African Bantu appear to have lived fairly
prosperously “ especially after the introduction of maize, another
New World food crop, which spread outwards from the Portuguese
trading stations in southern Mozambique. Such European visitors as
passed through the land in the sixteenth and seventeenth centuries “
mostly survivors of shipwrecks off the notoriously stormy coast “
commented on the large herds of sleek cattle. They noticed, however,
that iron tools were scarce except in the neighbourhood of Delagoa
Bay. The governor of the Cape, Simon van der Stel, wrote a dis-
patch to his superiors in Holland in 1689, telling them of the journey
of the crew of a wrecked vessel, the Stavenisse. The country of the
˜Magossebe™ (AmaXhosa) is described as follows:

Their riches consist in cattle and assegais, also copper and iron. The country is
exceedingly fertile and incredibly populous, and full of cattle, whence it is that
lions and other ravenous animals are not very apt to attack men, as they ¬nd
enough tame cattle to devour. They preserve their corn in cavities under ground,
where it keeps good and free from weavils for years. In their intercourse with
each other they are very civil, polite and talkative, saluting each other, whether
young or old, male or female, whenever they meet; asking whence they come,
and whither they are going, what is their news, and whether they have learned
any new songs or dances. The kings are much respected and beloved by their
subjects; they wear the skins of buck and leopard. One need not be under any
apprehension about meat and drink, as they have in every village a house of
entertainment for travellers, where they are not only lodged, but fed also.4

It was more than a century before the Dutch colony planted at Cape
Town in 1652 made any contact with these south-eastern Bantu peo-
ples. Most of the western Cape province was still the country of the

D. Moodie, The Record, or a Series of Of¬cial Papers Relative to the Condition and Treatment of
the Native Tribes of South Africa 1838“1842 (Amsterdam, 1960), vol. 1, p. 431.
30 Africa since 1800

Khoi (Hottentots) and San (Bushmen) “ the pastoral and hunting
predecessors of the Bantu, now reduced to the south-western corner
of a subcontinent of which they had once been the principal inhab-
itants. The Dutch settlers at the Cape expanded only slowly into the
interior, driving out the Bushmen, and making servants and herds-
men of the Khoi, whose tribal organisation was broken by the double
impact of colonists and smallpox. This labour force was supple-
mented by slaves brought from both the western and eastern coasts
of Africa and the Dutch possessions in the East Indies. The Dutch
settlers and their slaves increased at an almost equal pace “ there
were some 17,000 of each by the end of the eighteenth century. In-
termarriage between all of the racial groups at the Cape “ Europeans,
Negroes, Khoisan, and Malays “ was beginning to produce the mixed
Cape Coloured population. Not until about 1770 did Afrikaner (Cape
Dutch) and Bantu face each other across the Fish River, thieving each
other™s cattle by night and arguing about its return by day. Soon there
were frequent armed con¬‚icts. In 1795, when the British ¬rst seized
the Cape from the Dutch at the time of the Napoleonic wars, the
problems on the eastern frontier were threatening and dangerous,
but might still have been satisfactorily solved, given goodwill on both
sides and a ¬rm determination to maintain a permanent frontier be-
tween the Cape Colony and its Bantu neighbours. But, by the time the
British returned permanently in 1806, the situation on the frontier
had passed beyond the hope of peaceful negotiation and control.

The East Coast: Arabs and Swahili

At the end of the eighteenth century, therefore, Bantu Africa was still
a very secluded region in comparison with most of Africa north of
the equator. The only part of it which had been for any long period
in contact with a literate civilisation and a universal faith was the
coastal belt of East Africa. Here, trading settlements were known
to Greek geographers in the ¬rst century A.D., while early mosques
built of mud and thatch have been found which show the presence
of Islam as early as the eighth or ninth century. Commercial settle-
ments were established by Arab merchants from the Red Sea and
the Persian Gulf; from the tenth century on, some of these were
prosperous enough to have their public buildings made of the coral
Africa South of the Equator 31

blocks abounding on the coast. The Swahili people, in origin per-
haps the indigenous Bantu inhabitants of the Lamu archipelago,
who would have been among the northernmost of the Eastern Bantu,
early learned to make and handle seagoing canoes and small sail-
ing vessels, and so managed the coasting trade, of which the Arabs
supplied the long-distance, oceanic element. It was the indigenous
Swahili, as much as or more than the Arabs, who built and popu-
lated the thirty or forty coastal towns spread out along the coast of
Kenya, Tanzania, and northern Mozambique. Kiswahili is unequiv-
ocally an Eastern Bantu language, though using some hundreds of
Arabic loan-words, many of them recently introduced. By the seven-
teenth and eighteenth centuries, however, if not much earlier, most
inhabitants of the Swahili towns were Muslims, and it was this factor
more than any ethnic or linguistic one which made them feel differ-
ent from, and superior to, the other Bantu peoples who were their
neighbours. Beyond the coastal belt, however, the in¬‚uence of this
Swahili Arab civilisation was very restricted indeed. Swahili Arab
traders had preceded the Portuguese on the Zambezi, where they had
built the riverside ports of Sena and Tete. Here also, Swahili Arabs
resident at the Mwene Mutapa™s court had in 1569 instigated the
murder of the Roman Catholic missionary, Gon¸ alo de Silveira, fol-
lowing which the Portuguese began a policy of extermination against
them. Elsewhere, however, until late in the eighteenth century, there
is scarcely a reference to Swahili Arab activities more than a few
miles from the Indian Ocean coast.

The Portuguese in Africa South of the Equator

After the Swahili Arabs came the Portuguese, whose direct in¬‚u-
ence was con¬ned to the Kongo kingdom, the Kwanza and Zambezi
valleys, and to a few offshore islands, including Luanda, Mozam-
bique, Kilwa, and Mombasa. In Kongo, some thousands of people,
including the royal family, became Christians. The Portuguese king
corresponded with the king of Kongo as an equal, addressing him
as ˜Most high and powerful prince and king, my brother™. Many
of the Kongo people remained Christian for eight or nine genera-
tions, until the last links with Europe were cut by the quarrelling
and ¬ghting which broke out inside the country from the end of the
32 Africa since 1800

seventeenth century onwards. On the Kwanza, and again on the Zam-
bezi, some tens of thousands of Africans came to regard themselves
as the subjects or allies of the Portuguese rather than of any indige-
nous African state. Of these, perhaps the majority became in some
sense Christians, but only a tiny handful acquired any literary educa-
tion or became assimilated to the Portuguese way of life and culture.
In any case, the example of Portuguese manners in such isolated set-
tlements in Angola and Mozambique was not very inspiring. In one
way or another, by conquest or by taking people under their protec-
tion, the Portuguese destroyed most of the African states with which
they came into direct contact. Much more important were the indi-
rect effects of their presence upon those living a little farther away
from them. Undoubtedly, the opening of the Atlantic trade encour-
aged the expansion of African states in the hinterland of Angola and
Mozambique, as it had done also in West Africa. While Portuguese
interference diminished the Mwene Mutapas, it created the condi-
tions for the Rozvi dynasty of Butwa to take their place and to es-
tablish the ¬rst effective customs control upon the gold production
of the Zimbabwe plateau. Again, the Portuguese destroyed the king-
dom of Kongo, but their presence in Angola certainly assisted the
rise of the Mwata Yamvo dynasty farther inland. Whoever had guns
had power. Whoever had cloth had the prestige goods with which to
reward loyal subjects. These things were obtained in exchange for
slaves and ivory, which could best be supplied and transported to the
Portuguese frontier markets by a state with well-armed hunters and
protected caravan routes. Such states could be, and usually were,
established and run on completely African lines. The Mwata Yamvo,
the Mwata Kazembe, and the Rozvi Mambo were all ˜divine kings™
and their political institutions were of a fully African kind. But it is
unlikely they would have developed as far as they did without the
Portuguese presence in Luanda and on the Zambezi.


The large island of Madagascar, lying some 500 km (300 miles) off
the coast of Mozambique, furnishes yet another example of how,
in the early stages of trade with Europe, it was better to be liv-
ing just beyond the range of direct contact than within it. Until the
Africa South of the Equator 33

early Christian era, Madagascar had remained uninhabited, and it
was ¬rst colonised by immigrants from maritime South-East Asia
in much the same way as the islands of the Paci¬c. The newcomers
travelled in outrigger canoes, bringing with them seedlings of their
native food plants, notably bananas, rice, and taro. Their Malagasy
language, the nearest relatives of which are found on the island of
Borneo, established itself so ¬rmly that all subsequent immigrants
came in time to adopt it. Although Madagascar had neither ivory
nor precious metals, it supplied foodstuffs for the coastal trade of
the Swahili Arabs, who had established settlements on the island by
the eleventh century. Perhaps as a result of this contact, the western
coast of the island came to be settled by Africans from the Mozam-
bique coastlands, who brought their own food crops and cattle. The
Portuguese and the Dutch called there for supplies on their voyages
to the Far East, but it was not until the eighteenth century, with the
introduction of coffee-planting on the neighbouring islands of Mau-
ritius and R´ union, that Madagascar began to play a signi¬cant part
as a supplier of slaves. The response of the Malagasy to this trade
was similar to that of the African peoples on the mainland. The best-
armed among scores of small states grew at the expense of their
weaker neighbours and sold their war captives to the Europeans, a
motley collection of pirates drawn from all the maritime nations of
the West. With the aid of European ¬rearms, some powerful states
emerged, in particular that of the Hova people, whose homeland was
a small, intensively irrigated area around the capital of Tananarive,
high on the central plateau. Between 1787 and 1810, a great warrior“
king, Nampoina, embarked on a career of systematic conquest and
political expansion. By the time of his death, the Hova-ruled state of
Imerina controlled most of the centre of the island. From his inland
fastness, and using the advantage of interior lines, Nampoina laid
the foundations of a powerful, centrally administered kingdom.
At the end of the eighteenth century, therefore, the southern half of
Africa, no less than its northern half, was still standing very much on
its own feet. If European in¬‚uence was somewhat more noticeable
in the south than in the north, the in¬‚uence of the Islamic world
was much less so. With one exception, the foreign-ruled enclaves
on African soil were of negligible size and small signi¬cance; on
the whole, African political systems had proved capable of adapting
34 Africa since 1800

and regrouping in response to the new economic opportunities of-
fered by the outsiders. Thus far, at least, most of the modern weapons
which had entered Africa through trade with the outside world had
found their way into the hands of Africa™s traditional rulers and had
been used to strengthen existing institutions. The exception to the
general rule was, of course, the Dutch colony at the Cape, where
the absence of any African occupation denser than that of hunters
and pastoralists had left a local power vacuum that was unique in
the continent. It was not that any European government yet wanted
anything more than a refreshment station at the Cape for its oceanic
shipping, which could be defended and denied to competitors. It was
rather that the few hundreds, and later the few thousands, of com-
pany servants-turned-settlers were able to establish a bridgehead for
local population growth of the same kind that Europeans had estab-
lished in the coastlands of North America. The enclave of Mediter-
ranean climate around the Cape peninsula, which made it unsuitable
for African cereal farmers, was to make it also the Achilles™ heel of
Africa, but no one in 1800 had foreseen it.
THREE. The Opening up of Africa:
(1) From the North-East

H istorians have often written of the nineteenth century in Africa
mainly as a period when Europeans were increasing their in-
¬‚uence and power in preparation for later conquest. It should rather
be seen as a period when, with the aid of a greatly increased sup-
ply of ¬rearms from the new coal-¬red furnaces and of cheap textile
manufactures from the steam-driven factories of the industrialis-
ing countries, world trade was pressing remorselessly into Africa in
search of whatever might be exportable with the limited means of
transport available. For most of the nineteenth century, the agents
of this pressure were the rulers and traders of the coastal regions,
including, ¬rst and foremost, those of the predominantly Muslim
countries bordering on the Mediterranean and the Red Sea.

Muhammad Ali (1805“1849): The Revival of Red Sea Trade

In all this part of the world, the French conquest and occupa-
tion of Egypt between 1798 and 1800 had marked a turning point.
Napoleon™s decisive victory at the battle of the Pyramids had revealed
the weakness and technical military inferiority of the whole of the
Ottoman empire, and, like many other such victories in history, it
provoked a compelling desire in the conquered to learn the skills of
the victors. Soon after the end of the Napoleonic wars, French mili-
tary instructors were to be found throughout the Ottoman lands, and
in addition to their military manuals, the intelligent and ambitious
among the Turkish of¬cer class were reading widely in European

36 Africa since 1800

history and political philosophy. Above all, in Egypt the Napoleonic
episode threw up an outstanding leader, able to understand the sig-
ni¬cance of events and to apply the lessons he had learned. The his-
tory, not only of Egypt, but of the whole of North-East Africa and the
Red Sea area, was dominated during the ¬rst half of the nineteenth
century by the ¬gure of Muhammad Ali Pasha. This remarkable man
combined the talents of an oriental despot with a shrewd under-
standing of the very different world of Europe. Muhammad Ali was
a man of great charm and utter ruthlessness, an able administrator,
and a cunning diplomat. Although he himself was not an outstand-
ing military leader, several of his many sons were extremely ef¬cient
commanders. He was concerned, above all, to secure his position as
Ottoman viceroy in Egypt and to make the of¬ce hereditary within
his own family. In addition, however, he restored the beginnings of
order and prosperity to the Red Sea area and provided Egypt with
the framework of a modern state.
Born in 1769 in Macedonia, one of the Ottoman territories in
the Balkans, Muhammad Ali ¬rst entered Egypt as an of¬cer in the
Ottoman forces sent there to deal with the French invasion. With
the support of a body of Albanian soldiers loyal to him personally
rather than to the far-off Ottoman sultan, he made himself by 1805
the most powerful military boss in Cairo. The following year he was
appointed vali (Turkish: viceroy, governor) by the sultan. His power
was at ¬rst tenuous, there being many other military groups opposed
to his own. In 1811, he dealt with this opposition with characteristic
ruthlessness by inviting the principal Mamluk amirs to a banquet in
Cairo and then having some 300 of them massacred in a narrow alley
leading out of the citadel on their way home. His power in Egypt now
secure, for the next eight years Muhammad Ali devoted his main ef-
forts to the paci¬cation of the Red Sea area. First and foremost, this
involved the suppression of the Wahhabis, the followers of a puritan-
ical Muslim sect which had arisen among the bedouin tribesmen of
the Arabian desert. The Wahhabi leaders denied the authority of the
Ottoman government and disrupted the annual pilgrimage caravans
travelling to the holy cities of Mecca and Medina. Muhammad Ali™s
armies cleared the Hijaz of the Wahhabis and, in 1818, ¬nally over-
came the dissident tribesmen in the heart of Arabia. The holy places

. 1
( 12)