. 1
( 12)


This page intentionally left blank
a f r i c a n s , sec o n d ed it i o n

In a vast and all-embracing study of Africa, from the origins of mankind to the AIDS
epidemic, John Iliffe refocuses its history on the peopling of an environmentally
hostile continent. Africans have been pioneers struggling against disease and nature,
and their social, economic, and political institutions have been designed to ensure
their survival. In the context of medical progress and other twentieth-century
innovations, however, the same institutions have bred the most rapid population
growth the world has ever seen. The history of the continent is thus a single story
binding living Africans to their earliest human ancestors.

John Iliffe was Professor of African History at the University of Cambridge and is a
Fellow of St. John™s College. He is the author of several books on Africa, including
A modern history of Tanganyika and The African poor: A history, which was awarded
the Herskovits Prize of the African Studies Association of the United States. Both
books were published by Cambridge University Press.
african studies

The African Studies Series, founded in 1968 in collaboration with the African Studies
Centre of the University of Cambridge, is a prestigious series of monographs and
general studies on Africa covering history, anthropology, economics, sociology,
and political science.

e d i to r i a l b oa rd
Dr. David Anderson, St. Antony™s College, Oxford
Professor Carolyn Brown, Department of History, Rutgers University
Professor Christopher Clapham, Centre of African Studies, Cambridge University
Professor Michael Gomez, Department of History, New York University
Professor David Robinson, Department of History, Michigan State University
Professor Leonardo A. Villalon, Center for African Studies, University of Florida

A list of books in this series will be found at the end of this volume.

Second Edition

john iliffe
Fellow of St. John™s College, Cambridge
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
Information on this title: www.cambridge.org/9780521864381

© John Iliffe 1995, 2007

This publication 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 2007

eBook (EBL)
ISBN-13 978-0-511-34916-4
ISBN-10 0-511-34916-5 eBook (EBL)
ISBN-13 978-0-521-86438-1
ISBN-10 0-521-86438-0

Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
In memory of
Charles Ross Iliffe
Joy Josephine Iliffe

List of maps page xi
Preface to the second edition xiii
1 The frontiersmen of mankind 1
2 The emergence of food-producing communities 6
3 The impact of metals 17
4 Christianity and Islam 37

5 Colonising society in western Africa 63
6 Colonising society in eastern and southern Africa 100
7 The Atlantic slave trade 131

8 Regional diversity in the nineteenth century 164
9 Colonial invasion 193
10 Colonial change, 1918“1950 219
11 Independent Africa, 1950“1980 251
12 Industrialisation and race in South Africa, 1886“1994 273
13 In the time of AIDS 288
Notes 317

Further reading 329
Index 345

List of maps

1 Main physical features page 3
2 The emergence of food-producing communities 8
3 African language families in recent times 11
4 The impact of metals 18
5 Christianity and Islam 39
6 Colonising society in western Africa 65
7 Colonising society in eastern and southern Africa 102
8 The Atlantic slave trade 132
9 Regional diversity in the nineteenth century 165
10 Colonial invasion 194
11 Colonial boundaries 204
12 Colonial change and independent Africa 220
13 Independent African states 254
14 Industrialisation and race in South Africa 275
15 In the time of AIDS 289

Preface to the second edition

David Fieldhouse suggested this book. In writing it, I have strayed far from my
expertise as a documentary historian. John Sutton is partly to blame for that
because he ¬rst interested me in African prehistory through his lectures at Dar es
Salaam. David Phillipson kindly read and commented on my typescript, as did
John Lonsdale, who has taught me so much. John Alexander and Timothy Insoll
helped with books. To the ¬rst edition, published in 1995, I have added a chapter,
current to 2006, and I have extensively revised the chapters on prehistory and
the Atlantic slave trade, together with less substantial revisions to take account
of recent scholarship on other periods. The mistakes that remain are my own.
John Iliffe

a f r i c a n s , sec o n d ed it i o n

The frontiersmen of mankind

the liberation of their continent made the second half of
the twentieth century a triumphant period for the peoples of Africa, but at
the end of the century triumph turned to disillusionment with the fruits of
independence. This juncture is a time for understanding, for re¬‚ection on
the place of contemporary problems in the continent™s long history. That is
the purpose of this book. It is a general history of Africa from the origins of
mankind to the present, but it is written with the contemporary situation in
mind. That explains its organising theme.
Africans have been and are the frontiersmen who have colonised an especially
hostile region of the world on behalf of the entire human race. That has been
their chief contribution to history. It is why they deserve admiration, support,
and careful study. The central themes of African history are the peopling of the
continent, the achievement of human coexistence with nature, the building up
of enduring societies, and their defence against aggression from more favoured
regions. As a Malawian proverb says, ˜It is people who make the world; the bush
has wounds and scars.™ At the heart of the African past, therefore, has been a
unique population history that links the earliest human beings to their living
descendants in a single story. That is the subject of this book.
The story begins with the evolution of the human species in Africa, whence
it spread to colonise the continent and the world, adapting and specialising to
new environments until distinct racial and linguistic groups emerged. Know-
ledge of food-production and metals permitted concentrations of population,
but slowly, for, except in Egypt and other favoured regions, Africa™s ancient
rocks, poor soils, ¬ckle rainfall, abundant insects, and unique prevalence of
disease composed an environment hostile to agricultural communities. Until
the later twentieth century, therefore, Africa was an underpopulated continent.
Its societies were specialised to maximise numbers and colonise land. Agricul-
tural systems were mobile, adapting to the environment rather than trans-
forming it, in order to avert extinction by crop-failure. Ideologies focused on
fertility and the defence of civilisation against nature. Social organisation also
sought to maximise fertility, especially through polygyny, which made genera-
tional con¬‚ict a more important historical dynamic than class con¬‚ict. Sparse
populations with ample land expressed social differentiation through control
2 africans: the history of a continent

over people, possession of precious metals, and ownership of livestock where
the environment permitted it, especially in the east and south. Scattered settle-
ment and huge distances hindered transport, limited the surplus the powerful
could extract, prevented the emergence of literate elites and formal institutions,
left the cultivator much freedom, and obstructed state formation, despite the
many devices leaders invented to bind men to them.
Northern Africa ¬rst escaped these constraints, but the Sahara isolated it
from the bulk of the continent until the later ¬rst millennium ad, when its
expanding economy and Islamic religion crossed the desert, drew gold and
slaves from West Africa™s indigenous commercial system, and created maritime
links with eastern and central Africa. Yet this path of historical development
was aborted by a population catastrophe, the Black Death, which threw North
Africa into nearly ¬ve centuries of decline.
Instead, for most of tropical Africa the ¬rst extensive involvement with the
outside world was through the slave trade, by whose brutal irony an under-
populated continent exported people in return for goods with which elites
sought to enlarge their personal followings. Slaving probably checked pop-
ulation growth for two critical centuries, but it gave Africans greater resis-
tance to European diseases, so that when colonial conquest took place in the
late nineteenth century, its demographic consequences, although grave, were
less catastrophic than in more isolated continents. African societies therefore
resisted European control with unusual vitality and made state formation no
easier for colonial rulers than for their African predecessors. Yet Europeans
introduced vital innovations: mechanical transport, widespread literacy, and
especially medical advances that, in societies dedicated to maximising popu-
lation, initiated demographic growth of a scale and speed unique in human
history. This growth underlay the collapse of colonial rule, the destruction of
apartheid, and the instability of successor regimes. It was the chief reason for
the late twentieth-century crisis.
That population should be the central historical theme is not unique to
Africa. Every rural history must have at its core a population history. Frontiers-
men were key historical actors in medieval Europe and Russia, China and
the Americas. The modern histories of all Third World countries need to
be rewritten around demographic growth. Yet some African circumstances
were unique. Africa™s environment was exceptionally hostile, for the evolu-
tion of human beings in Africa meant that their parasites had also evolved into
unique profusion and variety there. Whereas Russians, Chinese, and Americans
colonised by pressing forward linear frontiers and extending cultures formed in
nuclei of dense population, Africa™s colonisation was mainly an internal process,
with innumerable local frontiers, and its cultures were chie¬‚y formed on the
frontiers “ an experience compounded by Egypt™s failure to export its culture to
Frontiersmen of mankind 3

1. Main physical features.
4 africans: the history of a continent

the rest of the continent in the way that the culture of the Ganges Valley
permeated India. Africa had land-rich cultural traditions even where land was
scarce; India had land-scarce cultural traditions even where land was ample.
Most important of all, the peopling of Africa took place within a unique
relationship to the Eurasian core of the Old World. This is the book™s ¬rst
subtheme. Until climatic change created desert conditions in the Sahara dur-
ing the third millennium bc, Africa held an equal place within the Old World.
Thereafter sub-Saharan Africa occupied a unique position of partial isolation.
It was more isolated than Eurasian fringes like Scandinavia or South-East Asia,
which gradually adopted Eurasian cultures. But it was less isolated than the
Americas, which developed unique cultures unaffected by the iron-using tech-
nology, domestic animals, disease patterns, trading relationships, religions, and
alphabetic literacy that sub-Saharan Africa partially shared with the Eurasian
core. Partial isolation meant that cultural phenomena took distinctively African
forms. Partial integration meant that Africans were receptive to further inte-
gration, which helps to explain both their receptivity to Islam and Christianity
and their disastrous willingness to export slaves, just as the slaves themselves
gained value because they possessed unique resistance to both Eurasian and
tropical diseases.
The slave trade also illustrates a second subtheme. Suffering has been a
central part of African experience, whether it arose from the harsh struggle with
nature or the cruelty of men. Africans created their own ideological defences
against suffering. Concern with health, for example, probably loomed larger in
their ideologies than in those of other continents. But generally Africans faced
suffering squarely, valuing endurance and courage above all other virtues. For
ordinary people, these qualities were matters of honour; the elites devised
more elaborate codes. Historians have neglected the notions of honour that
frequently motivated Africans in the past and are still essential to understanding
political behaviour today. To restore these beliefs to their proper place in African
history is one purpose of this book.
Several general histories of Africa have appeared since serious study began
during the 1950s. The earliest studies emphasised state-building and resistance
to foreign domination. A second, disillusioned generation of historians focused
on market exchange, integration into the world economy, and underdevelop-
ment. The most recent work has concentrated on environmental and social
issues. All these approaches have contributed to knowledge, especially to appre-
ciation of Africa™s diversity. All are utilised here, but within the framework pro-
vided by Africa™s unique population history. The argument is not that demog-
raphy has been the chief motor of historical change in Africa. That may have
become true only during the second half of the twentieth century. Population
change is not an autonomous force; it results from other historical processes,
above all from human volition. But precisely for that reason it is a sensitive
Frontiersmen of mankind 5

indicator of change, the point at which historical dynamics fuse into an out-
come that expresses not merely the actions of elites, as politics may do, nor
merely a surface level of economic activity, as market exchange may do, but the
most fundamental circumstances and concerns of ordinary people. Nor is the
choice of population as the central theme a concession to late twentieth-century
preoccupations or propaganda for birth control. Rather, population change is
the thread that ties African history together at all its different periods and levels.
Yet to choose this theme presses the sources for African history to their limits,
and perhaps beyond. Reliable demographic data scarcely exist before the Second
World War, except in privileged regions. The general history of the twentieth
century can rely chie¬‚y on written sources and the historian™s standard tech-
niques. In Egypt, written materials go back beyond 3000 bc. Arabic references to
West Africa begin in the eighth century ad. But parts of equatorial Africa have
no written records before the twentieth century. In their absence, knowledge of
the past must rely chie¬‚y on archaeology, which advanced dramatically during
the second half of the twentieth century, especially its geophysical methods of
dating by radiocarbon and other sophisticated techniques. Yet archaeology is so
laborious and expensive that it has scarcely touched many areas of the African
past. It can be supplemented by analysis of languages, folklore, oral traditions,
ethnographic materials, art, and the biological evidence surviving in human
bodies. All these have contributed to our understanding of the past, but they
are often surrogates for archaeological research not yet undertaken. One of the
most exciting things about African history is that much of it still waits beneath
the earth.

The emergence of food-producing communities

human evolution
africa is immensely old. its core is an elevated plateau of rocks
formed between 3,600 million and 500 million years ago, rich in minerals but
poor in soils. Unlike other continents, Africa™s rocks have experienced little
folding into mountain chains that might affect climate. Lateral bands of tem-
perature, rainfall, and vegetation therefore stretch out regularly northwards
and southwards from the equator, with rainforest giving way to savanna and
then to desert before entering the belts of winter rainfall and Mediterranean
climate on the continent™s northern and southern fringes. The great exception
is in the east, where faulting and volcanic activity between about 23 million and
5 million years ago created rift valleys and highlands that disrupt the lateral
climatic belts.
This contrast between western and eastern Africa has shaped African history
to the present day. At early periods, the extreme variations of height around
the East African Rift Valley provided a range of environments in which living
creatures could survive the climatic ¬‚uctuations associated with the ice ages
in other continents. Moreover, volcanic activity and the subsequent erosion of
soft new rocks in the Rift Valley region have helped the discovery and dating
of prehistoric remains. Yet this may have given a false impression that humans
evolved only in eastern Africa. In reality, western Africa has provided the earliest
evidence of human evolution, a story still being pieced together from surviving
skeletal material and the genetic composition of living populations. The story
begins some six million to eight million years ago with the separation of the
hominins (ancestral to human beings) from their closest animal relatives, the
ancestors of the chimpanzees. The skull of the ¬rst known hominin, Sahelan-
thropus tchadensis, was discovered in 2001 by an African student examining the
shores of an ancient Lake Chad. Apparently some six million or seven million
years old, this creature is thought to have stood upright and combined other
hominin characteristics with a brain of chimpanzee size.1 During the following
¬ve million years, a wide variety of other hominins, mostly known as Australo-
pithecines, left remains chie¬‚y in eastern and southern Africa. They ate mainly
vegetable food, had massive facial skeletons but small brains, and probably did

Emergence of food-producing communities 7

much climbing but increasingly walked upright, as is demonstrated by their
footprints astonishingly preserved from more than 3.5 million years ago in beds
of volcanic ash at Laetoli in Tanzania.
Australopithecines eventually became extinct, but human beings are proba-
bly descended from lightly-built Australopithecines or an ancestor shared with
them. An important stage in this evolution was the deliberate chipping of stones
to use for cutting. Found at Rift Valley sites in Ethiopia, Kenya, and Tanzania
from 2.6 million years ago, these tools are associated especially with remains
of a hominin known as Homo habilis. Some believe him to be on the main line
of human descent, although others group him with the Australopithecines as
one of several near-human creatures of the period.2
Some 1.8 million years ago, a more clearly human creature entered the archae-
ological record. Homo ergaster (from a Greek word meaning work) was to sur-
vive with remarkably little development for over a million years. Of modern
human height with an easy walking posture and a larger, more complex brain,
these creatures were adapted to life in open woodlands, may have learned to
use ¬re, and made the more sophisticated stone tools known as hand-axes that
were to remain the chief human implements in durable materials until some
250,000 years ago. The earliest examples of Homo ergaster and hand-axes come
from lakeside sites in eastern Africa, but similar stone tools have been found
widely in the continent, although seldom in tropical forest. At an early stage
in his history, Homo ergaster is also found in Eurasia. Each Old World conti-
nent now became an arena for evolution. Europe produced the Neanderthals,
with brains of modern size but distinctive shape. In Africa a similar transition,
beginning perhaps 600,000 years ago in Ethiopia, gradually produced anatom-
ically modern people. The earliest, still with many archaic features, have been
found in the Awash Valley from about 160,000 years ago. Later examples have
appeared at other sites chie¬‚y in eastern and southern Africa. Alongside this
physical evolution went changes in technology and culture as hand-axes gave
way to smaller and more varied stone tools, often designed to exploit local
environments. Some specialists attribute this growing adaptability to the need
to respond to the extreme ¬‚uctuations of temperature and rainfall that began
about 600,000 years ago, owing to variations in the earth™s proximity and angle
towards the sun.
At this point, the study of human evolution has interacted with two lines
of research into the genetic composition of living populations. One line con-
cerns mitochondrial DNA (deoxyribonucleic acid), one of the bodily sub-
stances transmitting inherited characteristics. Because this passes exclusively
(or almost exclusively) from the mother, its lineage can be traced back without
the complication of mixed inheritance from two parents at each generation.
In addition, mitochondrial DNA is thought to experience numerous small
changes at a relatively regular pace. Scientists have therefore compared the
8 africans: the history of a continent

2. The emergence of food-producing communities.
Emergence of food-producing communities 9

mitochondrial DNA of living people in order to estimate the point in the past
at which human beings shared a single female ancestor. Although the details
are controversial, most researchers believe that this was between 250,000 and
150,000 years ago, or in the broad period when the ¬rst anatomically modern
people appear in the fossil record. Initially, these ancestors of modern humans
spread within the African continent, where the oldest surviving lineages of
mitochondrial DNA exist among the San (˜Bushmen™) of southern Africa and
the Biaka Pygmies of the modern Central African Republic. About 100,000 years
ago, some of these anatomically modern people from eastern Africa expanded
brie¬‚y into the Middle East, but apparently they did not establish themselves
permanently there. With this exception, anatomically modern people appear
to have been con¬ned to Africa for some 100,000 years, spreading from the east
to other parts of the continent. A subsequent expansion took them to parts
of Asia by at least 40,000 years ago and from there to Europe. Gradually they
absorbed or replaced earlier hominins throughout the world.3
The mitochondrial and fossil evidence for this ˜Out of Africa™ thesis has
been reinforced by a second line of genetic research. The Y-chromosome that
determines male gender is inherited only from fathers and consequently can
also be traced back to a common ancestor, generally estimated at between
150,000 and 100,000 years ago. The oldest surviving strains of the chromosome
are con¬ned to Africans, especially San, Ethiopians, and other groups of ancient
eastern African origin. After a long period of differentiation, strains derived
from these groups diffused through the continent before being carried beyond
it. All men outside Africa have Y-chromosomes sharing a mutation that is
estimated to have taken place in an African ancestor at some point between
about 90,000 and 30,000 years ago.4
If anatomically modern people emerged in Africa and expanded to repopu-
late the world, a fundamental problem is to identify and explain their moder-
nity, the advantage they enjoyed over earlier hominins. Some specialists suspect
that a crucial breakthrough “ perhaps in the functioning of the brain “ took
place during the period of expansion between 60,000 and 40,000 years ago.
More point to an accumulation of smaller advances over as much as 300,000
years. The best-documented accomplishment was the replacement of heavy,
standardised hand-axes by smaller, specialised tools, eventually mounting tiny,
sharpened stones (microliths) in shafts or handles. Such industries might use
materials brought from scores or hundreds of kilometres away and establish dis-
tinct regional styles, the most remarkable being the Howieson™s Poort Industry
in southern Africa some 80,000“60,000 years ago, whose makers collected ¬ne-
grained stones from long distances to shape the earliest known microlithic
tools. The ¬rst bone tools appeared at much the same period, possibly as
barbed ¬shing harpoons on the Semliki River in the eastern Congo “ although
the dates there are disputed “ and as shaped points at Blombos Cave on the
10 africans: the history of a continent

southern coast of South Africa. Marine environments were among the ¬rst
specialised resources to be exploited, from at least 100,000 years ago in Eritrea
and South Africa. Less tangible innovations included the deliberate collection
of coloured pigments (found at a Zambian site more than 170,000 years ago)
and the use of red ochre and eggshell beads. Many archaeologists regard such
ornamentation as an example of the symbolic behaviour that is a key compo-
nent of human modernity. Another component is artistic decoration, which
may have appeared some 70,000 years ago in scratched engravings on bone
and ochre at Blombos Cave. The most important form of symbolic behaviour
may have been language, but although some believe that human ancestors were
physically capable of speech by about 300,000 years ago, it is not yet known “
although widely suspected “ that language was the crucial advantage enabling
anatomically modern people to repopulate the world.
These advances towards behavioural modernity progressed further within
Africa during a period beginning about 40,000 years ago. Early in that period,
men in the Nile Valley undertook complex underground mining for the stone
preferred for their tools, much the earliest industry of its kind known anywhere
in the world. Microlithic tools were then in use on the fringes of the equatorial
forest. They became common in the East African highlands by 20,000 years
ago, appeared at that date also in southern Africa, spread into western and
northern Africa during the next 10,000 years, and thereafter became ubiquitous.
Arrow-heads, appearing about 20,000 years ago, enabled hunting bands to add
birds and the more dangerous animals to their prey. Forager-hunters, probably
ancestral Pygmies, established themselves permanently in the equatorial forest.
Fishing became an increasingly important activity. Human settlements were
generally still transient, or at best seasonal, but the increasing care given to
burials “ appearing in southern Africa about 10,000 years ago “ suggests a
growing territorial sense. The remains of some 200 people of this microlithic
period excavated from a cave at Taforalt in Morocco show few signs of violence,
but they do show close interbreeding, high mortality among children and
infants, and many routine miseries such as arthritis.
The most striking evidence of symbolic behaviour during the microlithic
period was rock-painting, which dates back at least 28,000 years in southern
Africa. For the future, however, the most important development was the for-
mation of Africa™s four language families. These are so distinct from one another
that no relationship among them has been reconstructed, implying separate
development over many millennia. They coincide to some extent with genetic
differences and perhaps with physical characteristics arising from natural selec-
tion of those best ¬tted to survive and reproduce in particular environments.
Thus the San forager-hunters of southern Africa possessing the oldest strains
of Y-chromosomes and mitochondrial DNA “ together with probably related
Khoikhoi pastoralists “ speak distinctive ˜click™ languages possibly forming a
Emergence of food-producing communities 11

3. African language families in recent times. Source: Adapted from J. H. Greenberg, The
languages of Africa (3rd ed., Bloomington, 1970), p. 177.

loose and therefore ancient family. The only other speakers of these Khoisan
languages are small groups in eastern Africa, where the San may have originated
before spreading southwards as successful forager-hunters. San share the oldest
surviving Y-chromosomes with some Ethiopians, whose languages belong to
a second ancient family, Afroasiatic, which embraces Cushitic, the Semitic
languages of Ethiopia, Arabic, Hebrew, the Berber tongue of North Africa,
the Hausa language of northern Nigeria, and, in the past, ancient Egyptian.
Afroasiatic probably originated in the broad Ethiopian region at least 8,000
years ago and possibly much earlier. Many of its speakers were of the lightly built,
Afro-Mediterranean type depicted in ancient Egyptian art. In this they came
to contrast with the characteristically tall and slender Nilotic peoples whose
languages belonged to a third, Nilo-Saharan family, which may have originated
in the broad Saharan region at least as early as Afroasiatic. Nilo-Saharan may
be distantly related to the fourth family, the Niger-Congo languages, which are
spoken predominantly by Negroid peoples and are thought to have divided into
West Africa™s modern languages over at least the last 8,000 years. As will be seen,
three of these families were associated with centres of intensive food gathering
and production, the exception being Khoisan. Superior access to food may well
12 africans: the history of a continent

have enabled speakers of the three families to expand demographically and
absorb scattered forager-hunters whose distinct languages no longer survive.

savanna herding and agriculture
The addition of herding and agriculture to foraging and hunting economies
permitted larger populations, but the change is dif¬cult to identify in the archae-
ological record, especially in Africa where natural species were so numerous.
What appear to be cattle bones may have belonged to wild rather than domestic
beasts. Remains of root crops like yams rarely survive, while grain may have
been collected from wild grasses rather than cultivated. Pottery is no proof of
agriculture, nor even are grinding-stones, which may have been used to crush
wild grains or pigments such as ochre. The origins of African food-production
are therefore contentious and there is often a wide gap between the linguistic
evidence, which generally suggests early origins for agriculture and herding,
and archaeological research, which usually gives later dates. Nor is it even clear
why people should have begun to produce food at all. The idea that food pro-
duction originated in the Near East and spread through Africa where it was
eagerly adopted by starving hunter-gatherers is untenable. Study of modern
forager-hunters suggests that some can obtain more nutrients with less effort
and more freedom than most herdsmen or agriculturalists. Skeletal evidence
from the Nilotic Sudan suggests that one consequence of food-production
there was malnutrition. Another was probably disease, for several infectious
human diseases were probably contracted from domestic animals, while the
clearing of land for agriculture encouraged malaria and the larger populations
of food-producing societies sustained diseases that could not have survived
among scattered forager-hunters. Given Africa™s abundant wild produce, the
drudgery of food-production can have been tolerable to prehistoric people only
if it offered marked advantage over their previous lifestyle as a result of major
change in their circumstances.
Most experts believe that the crucial changes stimulating food-production
in Africa, as in Latin America, were climatic changes, especially in the north-
ern half of the continent. Africa has no single climatic pattern, but, broadly
speaking, the period from about 30,000 to 14,000 years ago was exceptionally
cool and dry in most of the continent except the south, partly owing to the
angle of the earth™s axis towards the sun. Most of Lake Victoria™s ¬‚oor was dry
as recently as 13,000 years ago, when the Sahara and its environs were probably
uninhabited. This may have concentrated population into favoured areas like
the lower Nile Valley. There is evidence as early as 20,000 to 19,000 years ago of
intensive exploitation of tubers and ¬sh at waterside settlements in southern
Egypt near the First Cataract, soon followed by the collecting of wild grain.
Initially seasonal, these settlements grew larger during the following millennia;
by 12,000 years ago some were permanent and had substantial cemeteries. Yet
Emergence of food-producing communities 13

these developments did not lead to food-production. Instead, the angle of the
earth™s axis shifted, temperature rose in all but southern Africa, and around
12,000 years ago the arid phase in the tropical climate gave way to exception-
ally high rainfall. Devastating ¬‚oods poured through the lower Nile Valley and
drove its inhabitants into the surrounding plains.
From about 12,000 to 7,500 years ago, the northern half of Africa was much
wetter than it is today. The Sahara contained relatively well-watered highlands,
even the notoriously arid Western Desert of Egypt supported sparse grazing,
and Lake Turkana in the East African Rift Valley rose about 85 metres above
its present level. Across the width of Africa from the Niger to the Nile, cultures
with a degree of similarity took shape. Archaeological research shows that their
practitioners formed some permanent settlements; used stone, wood, and bone
tools; and lived by ¬shing, hunting, and collecting vegetable foods, including
wild grains, the exact mixture varying with each local environment. From the
eighth millennium bc, they made Africa™s earliest known pottery in a style,
known as dotted wavy-line, which came to be used from southern Libya and
the Dogon Plateau in modern Mali to Khartoum, Lake Turkana, and possibly as
far south as Lake Victoria. Their most remarkable survival is an 8,000-year-old
dugout canoe, eight metres long, excavated from the shore of Lake Chad, the
second oldest boat known anywhere in the world.5 These people were mainly
of Negroid race and were probably responsible for spreading Nilo-Saharan
languages throughout the region, where they are still widely spoken.
Some analysts of Nilo-Saharan languages believe that the practitioners of
this high-rainfall culture kept livestock and cultivated grain. For livestock, this
may be true; excavators at Nabta Playa and Bir Kiseiba, pond-basins in the arid
western desert of Egypt, believe that they have unearthed remains of domesti-
cated cattle from 9,000 or 10,000 years ago, as early as anywhere in the world,
and the likelihood of independent domestication is supported by evidence
from mitochondrial DNA that African cattle have long been genetically dis-
tinct from those of other continents.6 By about 7,000 years ago, cattle-herding
had certainly spread to highland areas in the central Sahara. It reached North
Africa during the following millennium, somewhat later than the herding of
sheep and goats, which probably came from southwestern Asia because Africa
had no suitable wild species. In North Africa, this pastoral culture was prac-
tised by ancestral Berber peoples. In the Saharan highlands it left magni¬cent
By contrast, there is little if any archaeological evidence to support linguistic
indications of the cultivation or domestication of crops during this high-rainfall
period, suggesting that Africa was distinctive in practising herding before crop
production. In Egypt, domesticated wheat and barley, probably from south-
western Asia, were cultivated in about 5200 bc at the Fayum depression, west of
the lower Nile, and slightly later at Merimde, a substantial village of tiny mud
huts on the southwestern edge of the Nile Delta. Claimed ¬ndings of earlier
14 africans: the history of a continent

domesticated grains in northern Africa have not survived scrutiny. Instead,
by 7,000 years ago, there is evidence at Nabta Playa and in the Saharan high-
lands of increasingly settled populations systematically collecting and grinding
wild grains. That this may have developed into deliberate cultivation has been
suggested especially for settlements in the middle Nile Valley around modern
Khartoum, a summer-rainfall region where wheat and barley could not ¬‚our-
ish and the dominant cereal was to be sorghum. By 8,000 years ago, people on
the River Atbara, northeast of Khartoum, were collecting and grinding wild
grass seeds. At Kadero, twenty kilometres north of Khartoum, a large settle-
ment of the ¬fth millennium bc lived chie¬‚y from cattle and great quantities of
sorghum, to judge from grain-impressions on pottery and ˜tens of thousands of
worn-out grindstones™. Yet the sorghum was wild, for the domesticated variety
has not been found in the Khartoum region until roughly the time of Christ,
having perhaps been domesticated elsewhere in northeastern Africa. One
possibility is that sorghum was cultivated in the Khartoum region for many
centuries without being domesticated. Domesticated cereals differ from wild
varieties chie¬‚y by retaining their grain in the ear until threshed, whereas wild
plants disperse it profusely. Food-collectors probably domesticated wheat and
barley by cutting ears, taking them home, threshing them, and sowing part of
the harvest as seed, thereby gradually selecting those strains that best retained
the grain in the ear. Sorghum, however, had thick stalks easier to harvest by
stripping the grain in the ¬eld, which would not have altered the species into a
domesticated form. Yet whether such cultivation without domestication took
place in the tropical savanna remains uncertain.7
Similar uncertainty surrounds the origins of food-production in Ethiopia.
Domesticated cattle existed there by the second millennium bc and perhaps
as early as the fourth. Evidence from the local Cushitic languages also suggests
early knowledge of millet, wheat, and barley, but there is no archaeological
con¬rmation of this before the ¬rst millennium bc, although Cushitic speakers
may well have cultivated these crops with the plough before Semitic-speaking
immigrants from southern Arabia reached Ethiopia at that time because the
immigrants adopted Cushitic words even for these essentials of their culture.
Moreover, Ethiopians must have domesticated several distinctive local crops:
teff (a tiny grain), noog (an oil plant), and ensete (the banana-like staple of
southern Ethiopia).
Meanwhile food-production had also spread southwards into East Africa.
By the ¬fth millennium bc, the high-rainfall culture of ¬shing, foraging, and
pottery embraced the Lake Turkana region. When rainfall declined thereafter,
Nilo-Saharan speakers may have carried this culture and later the exploitation
of grain southward towards Lake Victoria, although there is as yet no archaeo-
logical con¬rmation of this. Reduced rainfall may also have damaged grazing
lands in the north while reducing disease further south, thereby encouraging
a southward drift of pastoralism that reached the Lake Turkana area around
Emergence of food-producing communities 15

2500 bc and continued southward through the Rift Valley. These pastoralists
may have been Cushitic speakers who spread widely through East Africa, where
isolated groups in north-central Tanzania still speak these languages. Linguistic
evidence suggests that the Cushitic speakers knew of cereals, but there is no
archaeological evidence that they cultivated them. Later, during the ¬rst mil-
lennium bc, other pastoralists penetrated southward from the Sudan region
and occupied the high East African grasslands, probably speaking Nilo-Saharan
languages, although these linguistic identi¬cations are necessarily speculative.
The desiccation that drove food-producers southward into East Africa also
impelled southward expansion in the west. During the third millennium bc,
declining rainfall in the Sahara obliged its pastoralists either to concentrate in
especially favoured areas or to drift southward into the river valleys draining
into Lake Chad and the Niger, free now to exploit regions where the bush had
hitherto been dense enough to support tsetse ¬‚ies carrying trypanosomes fatal
to cattle. By the ¬rst half of the second millennium bc, cattle were herded
close to the top of the Niger bend and on the southern shores of Lake Chad.
Shortly afterwards, the ¬rst strong archaeological evidence of crop domestica-
tion within Africa appears at Dhar Tichitt in modern Mauritania, a large cluster
of stone-built villages where domesticated pearl (or bulrush) millet was culti-
vated for perhaps a thousand years until that region in turn became too dry for
agriculture. Domesticated millet quickly diffused southward. Small quantities
were grown on the southern shores of Lake Chad by 1200 bc and in the north
of modern Burkina Faso shortly thereafter.
Most strikingly, by the middle of the second millennium bc, domesticated
millet, sheep and/or goats, small local cattle, and pottery with Saharan af¬ni-
ties were all components of the economy at Birimi, a settlement close to the
northern edge of the West African forest in modern Ghana. This was an outlier
of the Kintampo culture whose other sites, further south in the forest, show the
exploitation of oil-palm and the use of ground-stone axes, probably for forest
clearance. Savanna food-production had met the distinct culture of the West
African forest.

forest agriculture
The distinctions between food-collection, cultivation, and domestication are
even more dif¬cult to trace in the forest than in the savanna. Animal bones
survive poorly in forest soils. The staple crops that came to be used were not
cereals but yams and bananas, which leave few archaeological traces. Foraging
had a long history in the forest, but the ¬rst indication of more settled life is the
appearance of pottery over 7,000 years ago at Shum Laka in the Cameroun grass-
¬elds, close to the forest edge. This did not necessarily imply agriculture; neither
did the appearance a millennium later of ground-stone axes or the exploitation
of oil-palms from the fourth millennium bc. Linguistic evidence suggests that
16 africans: the history of a continent

yams may also have been exploited, and possibly cultivated, throughout this
period, but this has not yet been demonstrated archaeologically. By contrast,
archaeologists claim to have discovered banana phytoliths (minute mineral par-
ticles found within plants) in southern Cameroun from the last millennium bc,
implying that this Asian plant must have spread through the equatorial region
during earlier centuries despite the lack of evidence of its cultivation further
east. This claim raises such dif¬culties that it awaits further con¬rmation.8
The forest margin of Cameroun and Nigeria was the region from which
Bantu speakers gradually expanded throughout the southern half of Africa. All
Bantu languages form only one sub-branch of the Niger-Congo family. Their
most closely related languages cluster on the border between Cameroun and
Nigeria, so that was almost certainly the Bantu homeland. It is likely that the
Bantu languages were carried by colonists who also took agricultural skills into
regions where they were hitherto unknown, probably often transmitting them
to existing populations. Descendants of these colonists still possess consider-
able genetic as well as linguistic homogeneity. Theirs was one of the greatest
migrations in human history, but it was an immensely complicated and gradual
dispersal across the continent by families and small groups of cultivators, not
a mass movement by organised bodies of pioneers.
The history of this dispersal is contentious and little understood. By about
3000 bc, Bantu speakers with stone tools, pottery, and common words for yam
and oil-palm were probably moving slowly down the western equatorial coast.
They reached the Libreville area of modern Gabon by 1800 bc and continued
at least as far as the Congo estuary. As they did so, some broke away inland
through the forest to reach the middle Ogooue Valley by about 1600 bc and the
upper river by 400 bc. Others penetrated to the River Congo, where some slowly
colonised the tributaries leading into the inner Congo basin from about 400 bc,
while others moved more quickly up the main waterways until, at about 1000
bc, they reached the eastern edge of the equatorial forest in the broad area of the
great East African lakes. There they settled in well-watered valleys permitting
cultivation of their forest crops.
Yet this was only the ¬rst phase of Bantu dispersal. To the east and south of the
equatorial forest lay savanna lands which Bantu speakers could colonise only
if they ¬rst added grain cultivation to their agricultural techniques. Linguistic
evidence suggests that they probably learned to grow cereals (chie¬‚y sorghum)
in the Great Lakes region from Nilo-Saharan speakers who had brought the skill
southward from the Nile Valley. The Bantu probably also learned cattle-keeping
from Nilo-Saharans and perhaps from the Cushitic-speaking pastoralists who
had moved southwards into East Africa through the Rift Valley, although there
is no ¬rm archaeological evidence of either of these peoples in the Great Lakes
region. And it was probably here that the Bantu learned a further skill: to work
iron. To appreciate this innovation, we must return to Africa™s wider history.

The impact of metals

stone-using peoples had pioneered the colonisation of africa.
Their successors carried it forward with the aid of metals: ¬rst copper and
bronze, then iron. Only northern Africa had a bronze age; agriculturalists used
iron to colonise most of eastern and southern Africa.
The earliest evidence of metalworking in Africa comes from southern Egypt
late in the ¬fth millennium bc. At ¬rst pure natural copper was probably used
to make pins, piercing instruments, and other small articles. Smelting of cop-
per ore to remove impurities probably began in the ¬rst half of the fourth
millennium, either invented locally or imported from western Asia. It caused
no discontinuity in Egyptian history, for stone tools were widely used until the
¬rst millennium bc, but the new technique spread until a ¬xed weight of cop-
per became Egypt™s standard unit of value. Moreover, the innovation coincided
closely with the creation of Africa™s ¬rst great agricultural civilisation in the
Nile Valley. It was an African civilisation, for Egypt™s peoples, although hetero-
geneous, contained a core of Afro-Mediterranean race and spoke an Afroasiatic
language. Egyptian civilisation displayed many cultural and political patterns
later to appear elsewhere in the continent, although Egypt also illuminated
wider African history by means of contrast.
The contrast was rooted in the environment. Pioneers had practised agri-
culture in the Fayum depression and on the southwestern edge of the Nile
Delta since about 5200 bc. During the following millennium, desiccation
drove others from the eastern Sahara to settle on ridges bordering the Nile
Valley, where lower ¬‚oods made land available for pastoralism and agriculture.
Dependence on the river made these settlers more amenable to political con-
trol than Africans who retained their ancient freedom of movement. During
the fourth millennium bc, both Lower Egypt (the Delta) and Upper Egypt
(the narrow valley southwards to Aswan) practised a culture characterised by
exploitation of the ¬‚oodwaters, use of copper as well as ¬‚int, weaving of linen
cloth, trade with southwestern Asia, temples dedicated to deities like Horus
and Seth (later prominent in the Egyptian pantheon), a social strati¬cation

18 africans: the history of a continent

4. The impact of metals.
Impact of metals 19

displayed by the plain graves of commoners and the elaborate painted tombs
of the elite, and several small kingdoms with walled capitals of sun-dried brick.
How these kingdoms were uni¬ed remains obscure, but the ¬rst kings to rule
a united country gained power before 3100 bc and were buried at Abydos in
Upper Egypt.
This state, which lasted until the end of the Old Kingdom in c. 2160 bc, was
more centralised and authoritarian than its contemporaries in Mesopotamia.
Its power is often attributed to regulation of the irrigation system, but this was
not so. The Nile Valley had no such system. It depended on the natural ¬‚ood-
ing of the world™s most reliable river to produce a single annual grain crop, for
multicropping probably became signi¬cant only in postdynastic times. Works
were needed to control the ¬‚ood™s power, to remove obstacles to its expansion,
and to retain it on the land, but these were purely local works, directed by
local of¬cials like the provincial ˜canal-digger™ who was among Egypt™s earli-
est administrators. Pharaohs ceremonially inaugurated these works and their
Viziers claimed responsibility for them, but the Old Kingdom™s records do
not reveal a national bureaucracy dealing with irrigation; its natural tendency
was rather to strengthen the forces of provincial autonomy, which remained
powerful throughout Egyptian history and on three occasions “ the so-called
intermediate periods “ triumphed temporarily over political unity.
The connections between irrigated agriculture and pharaonic rule were
rather the system™s productivity “ it has been estimated that peasants could
produce three times their domestic requirements “ its capacity to support a
ruling class, the peasant™s need for order and his vulnerability to exploitation,
the state™s capacity to transport agricultural surplus by water and later to store
it, and especially the temptation that the surplus offered to those greedy for
wealth and power. Pharaohs exercised control by military, administrative, and
ideological means. They were depicted as conquerors, but their agents were
shown as scribes, using their monopoly of the newly invented skill of literacy to
repress autonomy elsewhere in society. ˜Be a scribe™, counselled an ancient text.
˜Your limbs will be sleek, your hands will grow soft.™ These of¬cials collected
tax, sometimes with much brutality; in later centuries the rate seems to have
been one-tenth of the harvest. They propagated the royal culture whose gradual
replacement of provincial traditions was the chief achievement of early dynas-
ties. During the dry season, they managed the rotating gangs of conscripted
peasants who built the gigantic public works of the Old Kingdom, not irriga-
tion channels but the pharaohs™ pyramid-tombs. The largest, built by Pharaoh
Khufu (Cheops) in the mid third millennium bc, was 147 metres high and
contained 2,300,000 stone blocks averaging some 2.5 tonnes. As the pyramids
rose, so peasant tombs disappeared almost entirely from cemeteries, suggest-
ing impoverishment by central power. Pharaohs were semidivine, could alone
communicate directly with the gods, were responsible for the regular operation
20 africans: the history of a continent

of the natural order, and had been preceded on their throne by gods in unbroken
succession since the creation. Although modern research is revealing dynastic
Egypt as a more ¬‚uid society than of¬cial ideologies suggested, with a lively
secular politics and extensive social and intellectual change, nevertheless Egyp-
tian minds were con¬ned by the uniqueness of their environment. The world
outside the Nile Valley was long seen as chaotic, the afterlife was imagined as
the Field of Reeds, and any innovation had to be presented as a restoration of
¬‚awless antiquity.
Although far more densely peopled than any other African region of the
time, Old Kingdom Egypt was still an empty land, with perhaps only one or
two million people, to judge from indications of the cultivated area. The num-
ber may have risen to between 2.0 million and 4.5 million in the late second
millennium bc and to a peak of 4 million to 5 million in the ¬rst centuries
ad.1 These ¬gures imply extremely slow growth rates, well below 0.1 percent a
year, held down perhaps by the contraceptive effects of prolonged breastfeed-
ing (of which there is evidence) and the high levels of mortality suggested by
mortuary evidence and con¬rmed by later Roman census data, which show
that in addition to appalling mortality before age 15, half of those surviving
died in each subsequent decade. Literary evidence refers to fever (presumably
malaria), while mummi¬ed remains show that Egyptians suffered from tuber-
culosis, cancer, bilharzia, arthritis, and probably smallpox, but not (on present
evidence) leprosy or syphilis. Population was most dense where the Nile Valley
was narrowest and most easily managed, but growth took place especially in
the dif¬cult Delta environment, a world largely of marshland and pasture in
Old Kingdom times but the target of systematic reclamation. Colonisation and
permanent cultivation demanded such an investment of labour that private
landownership emerged during the Old Kingdom and a class of great propri-
etors with small tenant-cultivators gradually acquired much of the land. By
c. 1153 bc temples alone owned approximately one-third of Egypt™s cultivable
area. The average peasant then cultivated about 1.25 hectares and showed more
concern to bequeath his rights intact to his offspring than men elsewhere in
Africa would display for another three thousand years.
Thanks in part to royal succession by primogeniture, which protected Egypt
from the succession disputes so destructive to later African states, the Old
Kingdom enjoyed great stability until it came to an end in c. 2160 bc. Under
its later pharaohs, its suffocating authoritarianism weakened as provincial
loyalties penetrated the bureaucracy, diffusing wealth away from the court,
depriving the regime of its capacity to build on the earlier monumental scale,
perhaps undermining its ability to relieve food scarcity in bad years, and gen-
erally robbing it of the Mandate of Heaven. The First Intermediate Period
(c. 2160“1991 bc) came to be seen as a time of civil war, brief reigns, famine,
and an in¬‚ux of desert peoples. This was too negative a picture, for it was
Impact of metals 21

also a time of provincial vitality, greater private wealth, and increased social
concern, but it enabled the restored Middle Kingdom (1991 “1785 bc) to rep-
resent itself in a newly self-conscious way as the embodiment of social order
and collective welfare. This regime temporarily collapsed during the Second
Intermediate Period (1785“1540 bc), only to give birth in turn to the New
Kingdom (1540“1070 bc), the most mature and expansive period of Egyptian
The great pharaohs of the New Kingdom were principally warriors, employ-
ing bronze weapons and the horse-drawn chariots whose arrival during the
Second Intermediate Period had introduced the wheel into Egyptian civilisa-
tion. Egypt™s armies crossed the Euphrates, penetrated southwards into modern
Sudan towards (or perhaps beyond) the Nile™s Fifth Cataract, and made Egypt
the greatest power in the known world. As often happened in later African
history, conquest of an empire changed the central structure of the state. Under
the New Kingdom, for the ¬rst time, Egypt had a militaristic ethos and a large
professional army, mostly composed of foreign mercenaries, whose control
became the key to the throne. There was also a small police force. Pharaohs
reestablished strong central power, aided by the resources in manpower and
material that empire provided. Yet this was also an ancient, wealthy, urbane,
and pluralistic society, for which the Old Kingdom pyramids were already
tourist attractions. Institutions were no longer merely emanations of royal will
but had lives of their own; temple priests, for example, were now hereditary
specialists practising an ascetic code, although their appointment still required
royal approval. Wider experience of the outside world enabled Egyptians to
see at least some foreigners as human beings like themselves. They contem-
plated the possibility that the future might surpass the present. Some even
doubted the utility of elaborate provision for death. Their artists grew more
adventurous, without losing the superb balance and dignity of the past. The
profound contempt for the poor found in earlier elite writings had given way
to the paternalistic social awareness that the ¬fteenth-century Vizier Rekhmire
proclaimed on the wall of his tomb:

I judged both [the insigni¬cant] and the in¬‚uential; I rescued the weak man
from the strong man; I de¬‚ected the fury of the evil man and subdued the
greedy man in his hour . . . I succoured the widow who has no husband;
I established the son and heir on the seat of his father. I gave [bread to the
hungry], water to the thirsty, and meat, oil and clothes to him who had
nothing . . . I was not at all deaf to the indigent. Indeed I never took a bribe
from anyone.2

Social historians seeking to liberate Ancient Egypt™s complexity from the
weight of its of¬cial ideology have found two New Kingdom sources especially
valuable. One consists of papyrus documents and notes written on potsherds
22 africans: the history of a continent

and stone ¬‚akes by a community of sculptors, painters, and plasterers living for
several centuries in a village named Deir el-Medina and working on the tombs
in the Valley of the Kings near Thebes. They were state employees, transmitting
skills and jobs from father to son (often with the help of bribery) and earning a
wage in food suf¬cient to supply their families and provide a surplus to exchange
for other necessities “ for Egypt had no currency and trade was by barter. These
skilled craftsmen defended their interests vigorously. They worked eight hours
a day and only about half the days in a year, enjoying frequent festivals and
often undertaking private commissions on the side. Towards the end of the
New Kingdom, they struck work several times and once organised a sit-in at
the royal tomb when the administration failed to pay their food wages. The
community usually contained between forty and sixty workers and employed
up to sixteen female slaves who did the heavy housework for each family in
turn. Several households also had domestic slaves who were sometimes buried
in the family tomb, for Egyptians sought to acculturate the slaves amassed by
New Kingdom conquests “ Rameses III claimed to have given 81,322 to the
temple of Thebes alone and there was an active market in slaves, although they
were less important in relatively populous Egypt than elsewhere in the Ancient
World. In this mature and settled society, family organisation differed in some
respects from most later African patterns. Elementary households averaging
¬ve or six people were the norm at Deir el-Medina, as elsewhere: husband, wife,
two or three unmarried children, and perhaps the husband™s sister or widowed
mother. Such households maintained close ties with relatives elsewhere, the
family tomb symbolising collective identity, but Egypt had no powerful clans or
lineages collectively controlling property, which was held within the elementary
family. Marriage was mainly monogamous, descent was largely bilateral from
both father and mother, and women had an exceptionally high status, with
full rights to inherit property, preserve the dowry brought into marriage, and
receive one-third of jointly acquired property in case of divorce, which was easy
and common. Conjugal love was a familiar literary and artistic theme. People
of both sexes married early and established independent households, although
so long as children remained under their parents™ roof, they and the family
servants were subject to patriarchal authority. ˜The entire household is like
[my] children, and everything is mine™, the rich peasant Hekanakht of Thebes
reminded his family in letters of 2002 bc. ˜Be energetic in cultivating! Take
care! My seed must be preserved; all my property must be preserved. I will hold
you responsible for it.™3 Although there is little evidence of countercultures in
pharaonic Egypt, the materialism and commercialisation so vigorous in the
New Kingdom threatened to overwhelm its ostensible changelessness.
A second entry into ordinary life in the New Kingdom is through religion
and literacy. The uni¬cation of Egypt had been accompanied by the grad-
ual formation of a common pantheon. Often drawn from the local divinities
Impact of metals 23

of a hunting past, the gods were frequently pictured as human beings with
animal heads symbolising their distinctive natures. Egypt™s extreme concern
with death and regeneration, possibly linked to the regenerating annual ¬‚ood,
also predated uni¬cation; it grew more re¬‚ective with time. The formation
of a countrywide cult was aided by the adoption of literacy at the end of the
predynastic period (c. 3150 bc). The idea of writing may have come from Sumer
(in modern Iraq) where it ¬rst evolved, but the invention of Egyptian scripts
was independent, rapid, and probably encouraged by the state authorities, for
whom they became a major source of power. The state ¬rst used writing to
label possessions. It was con¬ned to administrative notation and royal display
for 500 years before it was separated from oral communication to record com-
plete sentences. Two scripts were invented almost simultaneously. Hieroglyphic
script, the ˜words of the god™ with inherent magical power, was used for formal
documents and inscriptions; it employed a simpli¬ed picture of an object to
represent both the word for that object and other words with the same conso-
nant sequence, a procedure especially suited to an Afroasiatic language. Cursive
script, used in daily life, was a greatly simpli¬ed (almost shorthand) version
of hieroglyphic. The two scripts symbolised the two levels so sharply distin-
guished in Egyptian culture, the one arcane and formal, the other mundane
and ¬‚exible. Yet knowledge of either script required training. Probably no more
than one Ancient Egyptian in a hundred was literate, so that the skill had a
less radical impact on Egyptian thought, religion, and society than alphabetic
literacy had in Greece and in later African cultures. Egyptian thought retained
many preliterate characteristics: it was concrete rather than abstract; each moral
quality was personi¬ed as a deity; no truly historical sense emerged; learning
consisted of a gigantic catalogue of names and attributes; and the law was not
codi¬ed. The state was a mass of individual of¬cials, tasks, and institutions;
unlike the Greek state, it was justi¬ed by antiquity and divine creation, not
by reason. There were no scriptures; the core of Egyptian religion was ritual
veneration of disparate gods never reduced by abstraction to systematic theol-
ogy. Religion remained tolerant and eclectic, adding new gods to its pantheon
especially during the New Kingdom™s imperial expansion. Ritual was seen in
magical terms.
Yet signi¬cant religious change did take place. Among the many gods of the
Egyptian pantheon, the sun god was chie¬‚y responsible for the maintenance
of cosmological order and gradually gained preeminence. Early in the New
Kingdom, the sun god became associated with an invisible and ubiquitous
deity, Amun, around whom the priests at the great temple at Thebes began to
construct a theology. Both drawing on this and reacting against it, the Pharaoh
Akhenaten (1364“1347 bc) instituted a monotheistic state cult of the sun-disc
(Aten), a worship of light to be approached only by sharing the king™s vision.
Other gods were erased, rituals banned, temples closed, and priests dismissed
24 africans: the history of a continent

in a persecution unique in Egyptian history. Such was royal power that this
did not provoke overt resistance. Akhenaten™s successors abandoned his pro-
gramme and eradicated his memory, but the impact lasted. In place of the old
polytheism, Amun came to be seen as the supreme divinity of whom other gods
were manifestations. Both kings and commoners sought Amun™s intervention
in a new mode of personal piety that exempli¬ed the slowly increasing impor-
tance of the individual during the long course of Egyptian history.
These developments supplemented previous patterns of popular religion.
Parents at all periods had named most children after major gods. Symbols and
¬gures of divinities originally con¬ned to tombs of the great had gradually
appeared in those of their inferiors. Votive offerings to temples by ordinary
people multiplied under the New Kingdom, as did the practice of seeking
oracles from gods when carried in procession. Animal worship was immensely
and increasingly popular. Scribes wrote amulets, letters to the dead seeking
aid, and (from late New Kingdom times) letters to the gods themselves. To
compensate for the lack of direct contact with divinity and consolation in mis-
fortune offered by the of¬cial cult, laymen and especially laywomen devised
their own remedies. At Deir el-Medina, for example, workmen erected mon-
uments recording their humility before the gods and their repentance of sins
for which they had been punished by misfortune. Their houses contained
shrines of lesser, popular divinities, often in grotesque shapes. They consulted
˜wise women™ when their children died or they suffered divine ˜manifestations™.
Evidence of these practices multiplied as the dynasties passed.
Like many later African states, the New Kingdom owed its decline to its
empire, which brought overexpansion, militarism, and internal division. Incur-
sions by western nomads from Libya appear to have begun in the thirteenth
century bc. The Asiatic empire was lost under Rameses III (1184“1153 bc) and
Nubia followed a century later. Royal succession became unstable, reigns short-
ened, political authority declined, and of¬ces increasingly became hereditary.
Real grain prices rose rapidly in the later twelfth century, perhaps owing not
only to somewhat diminished rainfall but to weaker agrarian administration,
suggested also by growing evidence of peculation. Power lay increasingly with
commanders of the mutually hostile Libyan and Nubian mercenaries. When
Rameses XI (1099“1069 bc) summoned the Viceroy of Kush and his Nubian
troops from modern Sudan to reassert royal control over Upper Egypt, Herihor
of Thebes “ who was simultaneously vizier, generalissimo, and high priest of
Amun “ used Libyans to repel them. During the ensuing Third Intermediate
Period (1070“664 bc), general militarisation took place, the rural population
frequently took refuge behind walled defences, and Egypt was divided into
regional units “ there were eleven in c. 730 bc, several under Libyan control “
until the Kushitic rulers of Nubia established a military occupation in the late
Impact of metals 25

eighth century bc, only themselves to be expelled during the 660s by forces
from Assyria, the dominant state in western Asia.
Assyrian power rested on cavalry (rather than chariots) and iron, smelted
in western Asia since early in the second millennium. Egypt had neither iron
ore nor wood fuel and its closely regulated craftsmen were slow to adopt the
new metal; the ¬rst evidence of iron-smelting in Egypt comes from Naukratis,
a town in the western Delta founded by Greek colonists in c. 620 bc. Greek
mercenaries enabled the Libyan rulers of Sais in the rich central Delta to
reunite Egypt, ¬rst as Assyrian vassals and then as independent rulers from 664
to 525 bc in the last great age of pharaonic civilisation. The Saites consciously
recreated past glories, decorating their many new temples in Old Kingdom
style. But change continued beneath the archaic surface: the colonisation of
the Delta, the acquisition of land by foreign mercenaries, the use of weighed
silver as a quasi-currency, and reliance on of¬ce and family origin rather than
royal will as sources of local authority. Egypt was now a prize for great powers.
Persian conquerors held it for two centuries after 525 bc, with one long interval
of independence. Alexander the Great took it from them in 332 bc, and one of
his generals created a Greek dynasty, the Ptolemies, who ruled until 30 bc, when
Rome at last added Egypt to its empire. Much of the ancient order survived these
political changes. Greek kings adopted pharaonic styles, patronised the temple
priests who preserved the old elite culture, identi¬ed Egyptian gods with their
own divinities, and were depicted in pharaonic poses on temple walls by an
artistic tradition that survived until the third century ad. They replaced senior
administrators with Greeks and made Greek the language of government, but
they maintained the bureaucratic structure affecting ordinary people. Even the
Romans followed their example, despite their normal preference for municipal
rather than bureaucratic government. Both pressed forward the colonisation
of the Delta, which, by Ptolemaic times, supported perhaps as many people
as Upper Egypt and had supplanted it as the country™s economic core, with
a new capital at Alexandria. The animal-driven irrigation wheel (saqia) to
lift water for dry-season cultivation reached Egypt from the Middle East in
Ptolemaic times, bringing the ¬rst evidence of summer grains and extensive
multicropping. Egyptian grain exports “ ˜the shipments™, as they were known “
were vital to Ptolemaic ¬nances and provided about one-third of Rome™s
wheat supply. Population and agricultural output both probably peaked at
this time of favourable climate. But peasant society was threatened by growing
commercialisation, owing in part to the Ptolemies™ introduction of coinage,
by the dominance of Greek-speaking cities, and by Roman encouragement
of large estates on which tenants paid half their crop in rent, while a growing
class of poor peasants, agricultural labourers, and urban paupers joined the
10 percent of the population who were slaves. In addition to rural revolts in
26 africans: the history of a continent

ad 152 and 172“3, protest found millenarian expression in ancient cultural
[Justice] will return, transferred back to Egypt, and the city by the sea [i.e.
Alexandria] will be but a place for ¬shermen to dry their catch, because
Knephis, the Tutelary Divinity, will have gone to Memphis, so that passers-by
will say, ˜This is the all-nurturing city in which live all the races of mankind.™
Then will Egypt be increased, when . . . the dispenser of boons, coming from
the Sun, is established there by the goddess [Isis] most great.4

nubia and northern ethiopia
˜Egyptian antiquity is to African culture what Graeco-Roman antiquity is to
Western culture™, wrote the Senegalese scholar Cheikh Anta Diop.5 There is little
evidence to support him, for Egypt was remarkably unsuccessful in transmitting
its culture to the rest of the continent, partly because that culture was so partic-
ular to the Nile Valley environment, partly because Egypt™s greatness coincided
with the desiccation of the Sahara, which isolated the Nile Valley from most of
Africa. Saharan rock-paintings show only slight traces of Egyptian in¬‚uence,
chie¬‚y a fascination with chariots. Irrigation techniques, small pyramid tombs,
and an oracular cult of Amun appeared in Saharan oases. Generally, however,
the impact of Egypt™s metalworking skills and notions of kingship was con¬ned
to the Nile Valley itself, ¬rst the ¬‚oodplain immediately to the south, known as
Lower Nubia, and then the narrow valley of Upper Nubia stretching southwards
from the Second Cataract towards modern Khartoum. Perhaps no more than
half a million people lived in this arid region in pharaonic times, with evidence
of high deathrates among young adults. So small a population was vulnerable
to near-extinction in adverse circumstances, especially political circumstances,
for Nubia prospered when Egypt was weak but suffered when Egypt was strong.
Yet Nubian society survived, with a longevity rivalling Egypt™s and a marked
continuity in the physical composition of its people, who inherited the Nilotic
culture of ¬shing, pottery-making, grain-collecting, and early herding of the
high-rainfall period.
Just as the oldest Egyptian tombs of the fourth millennium bc contained
ivory and ebony objects from the south, so Lower Nubian graves of the late
fourth millennium contained pottery, copper tools, and other objects of Egyp-
tian origin. These graves belonged to people known only as the ˜A Group™,
cultivators of wheat and barley who shared in the economic and political growth
that culminated at the end of the fourth millennium in Egypt™s uni¬cation, for
their settlements expanded and some of their leaders were buried in graves
rivalling those of their Egyptian counterparts. But this prosperity was fatally
attractive. A relief of the early First Dynasty shows a Nubian prisoner bound
Impact of metals 27

to the prow of an Egyptian ship, possibly indicating Egypt™s ¬rst known inva-
sion southwards. By Egypt™s Third Dynasty (from c. 2695 bc), Lower Nubia
was only sparsely populated and there was an Egyptian town at Buhen near
the Second Cataract (shortly to be a centre for the smelting of local copper),
while the presence of Nubian slaves and soldiers in Egypt during the pyramid-
building Fourth Dynasty suggests the likely fate of many A-Group people. As
the Old Kingdom weakened, however, Nubians regained living space. Egypt™s
outposts were withdrawn, trade resumed, and Lower Nubia was resettled, per-
haps mainly from the south by people known as the ˜C Group™ who practised a
more pastoral culture. These people suffered further invasion after the creation
of the Middle Kingdom in 1991 bc. ˜I sailed victoriously upstream, slaughter-
ing the Nubians on the river-bank,™ an Egyptian commander proclaimed. ˜It
was burning their houses that I sailed downstream, plucking corn and cutting
down their remaining trees.™6 Egyptians built powerful forts on their southern
border close to the Second Cataract and began to mine the gold of the eastern
desert, which now became central to Nubia™s external relations. For the ¬rst
time, also, Egyptian records mention a kingdom in Upper Nubia, which they
generally describe as ˜vile Kush™.
This, the earliest recorded African state outside Egypt, was centred south of
the Third Cataract, Nubia™s richest agricultural region and the point where a
desert road led away from the river towards the southern lands whose commerce
was one source of the kingdom™s wealth. Its capital, Kerma, took shape about
2500 bc around a religious complex. Its early burials display strong elements of
pastoral and military culture, to which was added the commercial wealth foster-
ing the growth of a state that reached its peak during the Second Intermediate
Period in Egypt (1785“1540 bc), when Egyptian troops again abandoned Lower
Nubia and Kerma™s power replaced them, extending as far north as Aswan and
establishing alliances among Egypt™s warring dynasties. By this time, Kerma
had absorbed much Egyptian culture, using copper extensively for vessels and
weapons, building a massively walled capital, and fashioning its ritual centre
to resemble an Egyptian temple, although the local religion laid a distinctively
African emphasis on sacri¬ce. The huge royal tumulus-tombs of the period,
with their attached chapels, contained ˜piles of ¬ne ceramics, jewels, arms, toilet
objects, chests and beds of wood inlaid with ivory™,7 as well as the remains of
scores and even hundreds of retainers buried alive to accompany their masters.
Kerma was victim of the last and most potent phase of Egyptian expansion,
which sought not only gold but military glory and administrative power. The
reuni¬cation that created the New Kingdom ¬rst permitted the reoccupation
of Lower Nubia and then enabled Tuthmosis III to destroy Kerma in about
1450 bc and to penetrate to the Fifth Cataract or beyond. During the next 400
years, the Egyptian impact was on a new scale. Egyptian temples and noblemen
acquired estates in Lower Nubia, whose C-Group people mostly became tenants
28 africans: the history of a continent

or labourers and were so fully assimilated as to be indistinguishable from
Egyptians in the archaeological record. When Egyptian forces withdrew at
the end of the New Kingdom in 1070 bc, they left a depleted and impoverished
population, perhaps partly because lower Nile levels had meanwhile reduced the
¬‚oodplain™s fertility. New Kingdom Egypt also ruled Kerma, but apparently less
directly and securely, for the great temples built there as outposts of Egyptian
power and culture had to be forti¬ed.
During the ninth century bc, a state reemerged in Upper Nubia, with many
similarities to Kerma (to judge from royal burials) but now, presumably in
response to further desiccation, based further up the Nile at Napata, the point
where the desert road from Kerma again met the river. From this base, in
c. 728 bc, King Piankhy intervened in Egypt, ironically as champion of
pharaonic traditions against Libyan military expansion. Napata™s rule in Egypt
until 656 bc accustomed its kings to an elite culture of Egyptian-style temples,
tombs, arts, crafts, and the use of the Egyptian written language. It also gave
Napata its ¬rst discovered iron object: a spearhead wrapped in gold foil and
found in the tomb of King Taharqa (690“664 bc).
The Saite rulers who expelled Taharqa™s successors from Egypt followed up
their victory by attacking Napata in 593 bc. At some point thereafter, the cap-
ital moved still further south to Meroe, the most southerly junction between
the desert road and the Nile, above the Fifth Cataract. Here the state, already
ancient, was to survive for well over ¬ve hundred years, but in changing form.
Meroe was south of the true desert, on the fringe of the tropical summer rains
where sorghum might grow without irrigation and cattle could graze the plains
in the wet season. Many Meroitic symbols had a pastoral emphasis and cattle
were probably its chief wealth. Its religious system combined the Egyptian pan-
theon, headed by the sun god Amun, with presumably local deities, especially
Apedemak, ˜Lion of the South™. While it was probably men who used the potter™s
wheel to make ceramics that changed with foreign fashions, women used their
hands to make a local pottery that scarcely changed at all. From the second
century bc, twenty-three signs from Egyptian script were converted into an
alphabet in which the still unintelligible Meroitic language was written. Rulers
of Meroe were high priests in the pharaonic manner, called themselves Kings
of Upper and Lower Egypt, and were buried under ever smaller pyramids until
the fourth century ad, but they were chosen from those with royal blood by
the Queen Mother and leading men in a manner wholly African. Meroe sup-
plied gold, slaves, and tropical produce to the Mediterranean and Middle East,
where it was known and occasionally visited as an exotic frontier kingdom.
Its armies rivalled Ptolemies and Romans for control of Lower Nubia, whose
prosperity revived in the early Christian era with the arrival of new crops and
saqia-based irrigation. But the core of Meroe™s economy was the sorghum, cot-
ton, and cattle of Upper Nubia as far south as Khartoum and its surrounding
Impact of metals 29

rainlands. Rather than transmitting Egyptian culture southwards to tropical
Africa, Meroe absorbed it into indigenous culture, as was to happen to foreign
cultures so often in African history. Even the southward transmission of iron-
working is doubtful. The kingdom itself disappeared from Meroe in the fourth
century ad, having perhaps been weakened by a shift of trade from the Nile to
the Red Sea during Rome™s occupation of Egypt. Skeletal evidence shows that
the population survived the political transition largely unchanged, but there
are indications of increased violence, economic decline, and depopulation.
In Lower Nubia, by contrast, new leaders acquired luxuries from the north,
adopted some Meroitic royal regalia, and were buried with their cherished
horses in a manner as spectacular as their predecessors nearly four thousand
years earlier.
There was one further Nubian legacy. While Kerma dominated Upper Nubia,
between about 2500 and 1500 bc, a chiefdom emerged in the Gash Delta to the
southeast, near the modern border between Sudan and Ethiopia, on an impor-
tant trade route to the Red Sea that has left Kerma-style pottery on the western
shore of Arabia.8 The Gash Delta™s trading contacts survived Kerma™s destruc-
tion, but the region was drawn into a new political system centred further
southeast on the northern edge of the Ethiopian plateau in modern Eritrea and
Tigray. Here, in about the eighth century bc, emerged a kingdom known as
Daamat. Its people may have moved on to the plateau to escape the desiccation
of the plains. Its pottery was partly of local Tigrayan origin and partly derived
from the tradition of Egypt and Kerma via the Gash Delta. Its high culture,
however, was largely of South Arabian origin, either by immigration or imita-
tion. A temple of the period to the astronomical gods of South Arabia survives
at Yeha in modern Tigray, probably Daamat™s capital, together with a possi-
ble palace, smaller temples elsewhere, inscriptions in the Sabean language of
South Arabia (although diverging from it as time passed), and sickles and other
objects in bronze, which was probably introduced from South Arabia. Trade
with the Nile continued and Daamat™s queens appear to have adopted Napatan
garments and ornaments, but Meroitic in¬‚uence was generally super¬cial. The
kingdom fragmented between the ¬fth and third centuries bc, bequeathing its
composite culture to historic Ethiopia.9

berbers, phoenicians, and romans
The use of copper in Egypt preceded by more than two thousand years evi-
dence of its use elsewhere in North Africa. Egyptian dealings with people
to their west were with ˜Libyan™ (ancestral Berber) pastoralists of Cyrenaica
and the desert oases, whom they regarded as shaggy barbarians and resented
when they in¬ltrated the Nile Valley as famine refugees, mercenaries, and even-
tually (from c. 945 bc) rulers of Delta states. Further west, in the Maghrib, the
30 africans: the history of a continent

predominant people were also ancestral Berbers. This region, from modern
western Libya (Tripolitania) to the Atlantic, displayed extreme environmental
contrasts: fertile coastal plains merging southwards into arid pasture and even-
tually desert, but broken by cultivable mountain outcrops. Ancient authors dis-
tinguished three main population groups. The most numerous were the Berbers
of the northern plains and especially the more accessible mountain areas, who
were plough-using, irrigating agriculturalists and stock-keepers conventionally
divided into Mauri in the west (modern Morocco) and Numidians in the centre
and east (Algeria and Tunisia). The second group were Berber semipastoralists
in the arid pastures and desert, who adopted horses during the ¬rst millennium
bc; ancient authors knew them mainly as Gaetuli, a generic term for pastoral-
ists. The third category were scattered groups in desert oases and outcrops,
notably the Garamantes of the Fezzan and the ancestors of the modern Tubu
of Tibesti. Roman accounts stressed ethnic difference and con¬‚ict between
agriculturalists and nomads, but modern research has shown much exchange
and symbiosis between them. Both practised a religion centred on the forces
of nature and fertility. Both appear to have had segmentary social and political
systems in which each person belonged to several groups of different size “
family, lineage, clan, tribe, perhaps confederation “ which acted collectively
only when a member con¬‚icted with someone from another group of equiv-
alent size. This segmentary system could limit violence through the threat of
retaliation without needing political rulers, so that ancient authors stressed
Berber egalitarianism. ˜There was a dislike of kings with great authority™, wrote
the Roman historian Livy. At later periods, however, egalitarian ideology often
coexisted with local Big Men, especially during crises, and that was probably
also true in antiquity.
Late in the second millennium bc, Phoenician traders from modern Lebanon
began to colonise the North African coast. Their most powerful settlement was
Carthage (˜New City™), established in the north of modern Tunisia soon after
its traditional foundation date of 814 bc and governed by its wealthy citizens.
The Phoenicians™ chief aim was to capture western Mediterranean trade and
their chief importance in Africa was to integrate the north into Mediterranean
history, just at the moment when the desiccation of the Sahara interrupted
communication with tropical Africa. The Phoenicians™ relations with their
African hinterland, by contrast, developed slowly. Scarcely any Carthaginian
records survive, but tradition says that the colonists con¬ned themselves to
the coast until the sixth century bc, when they extended the city™s territory
nearly two hundred kilometres into the fertile plains of northern and eastern
Tunisia, establishing an enduring pattern of foreign occupation in this region
that left the rest of North Africa to Berbers. Carthaginians also established
trade with the Garamantes, who supplied precious stones and a few black
slaves from the south, although Carthaginians themselves seem not to have
Impact of metals 31

penetrated desert trade. In the northern coastal plains of modern Tunisia,
wealthy Carthaginians established great wheat farms, while on the eastern coast
(the Sahel) they probably introduced the olives for which the region has since
been famed. Ancient sources describe these farms as ˜slave estates™ and record
frequent ˜slave risings™, but some scholars believe that the labourers were rather
the original Berber inhabitants, reduced to labour-tenants and sharecroppers.
Agriculture bene¬ted from the Phoenicians™ skill as metalworkers, especially
in bronze but also in iron, which they introduced to North Africa.
In 241 bc Carthage™s mercenary army lost its ¬rst disastrous war with the
rising power of Rome. War and defeat led the city to demand more tax, tribute,
and labour from surrounding Berbers and to seek greater control over them.
Provincial governors for the ¬rst time ruled the hinterland. ˜Punic Ditches™ were
constructed to defend Carthaginian territory and control pastoral movements.
The most resentful Berbers were followers of the Numidian chief Masinissa in
the coastal plain west of Carthage. They interacted culturally with the colonists:
Berber came to be written in a script derived from Phoenician, while Tanit,
the fertility goddess venerated by Carthaginians as they engaged increasingly
in agriculture, appears to have been of Berber origin. But Masinissa™s followers
also suffered especially from land alienation. In 202 bc he helped Rome to defeat
Carthage again and reduce it to a dependency. In 150 bc his encroachments on
Carthaginian territory provoked them into attacking him, only for his Roman
patrons to raze Carthage to the ground and leave it almost deserted for a
hundred years. Other Phoenician cities survived under Roman rule, but the
chief local powers for the next century were Carthage™s former Berber client
kings. One, Jugurtha, a descendant of Masinissa, fought a long war against the
Romans until betrayed in 105 bc. His Roman conquerors then settled their
troops west of Carthage, but the main period of Roman colonisation began
sixty years later when a chain of settlements was founded along the North
African coast for military veterans. By the early ¬rst century ad, there were
perhaps between ten thousand and twenty thousand Roman immigrants in
the Roman territory stretching from central Morocco to western Libya. Later
emperors added only a few colonies in outlying strategic areas.
Roman power centred in coastal towns surrounded by ˜villa belts™ of estates,
governing and drawing wealth from the Berber hinterland. Rainfall was prob-
ably similar to that of today.10 By the birth of Christ, the coastal plains were
already Rome™s chief source of grain, taken mainly as tax or rent. During the
next three centuries, drier lands became the empire™s main supplier of olive oil.
North Africa was notorious for its great estates, especially the imperial prop-
erties that in ad 422 occupied about one-sixth of Roman territory in modern
Tunisia. They were leased out to contractors who farmed part of the land with
tributary labour from the tenants (coloni) who were left on the remainder and
paid one-third of their crops in rent. Roman villas and Berber villages were
32 africans: the history of a continent

interspersed, the villages gradually predominating as one moved southwards.
Berber cultivators were quick to take advantage of new export markets. In the
predesert of modern Libya, today almost bereft of cultivation, they constructed
¬‚oodwater controls enabling them to grow olives on land whose average rain-
fall was only one-third or one-half of that thought necessary for the crop. The
chief bene¬ciaries were probably the prominent Berber families who increas-
ingly adopted Roman culture and seigneurial lifestyles. At the largely Berber
town of Gigthis in southern Tunisia, for example, Memnius Pacatus was both
chief of the Chinithi tribe and head of a family that, by ad 200, was producing
Roman senators. The Berber goddess Tanit of Carthage became Juno Caelestis,
the Roman Queen of Heaven. Mosaic artists and writers like Apuleius expressed
a vigorous and distinctive North African culture, which was to outlive Roman
government. Even those who resisted Rome™s authority were often in¬‚uenced
by its culture. The Gaetuli formed ephemeral coalitions to resist Roman inter-
ference, but they also relied on grain, harvest employment, and stubble for
grazing in the northern agricultural zones to which they drove their stock each
summer. The Romans sought to control interaction with pastoralists by con-
structing the line of ditches, lateral roads, and strongpoints known as the limes,
which ran parallel with the coast from Morocco to western Libya and served
also to police the more numerous mountaineers.
Beyond the limes, important changes took place. As the Saharan region grew
drier, its former pastoralists clustered into surviving oases. Communication
between them depended on horses and the camels that came into widespread
use during the ¬rst centuries ad. The predominant group in this early desert
economy were the Garamantes of the Fezzan, a people of mixed Negroid
and Berber origin who from the later ¬rst millennium bc constructed several
thousand kilometres of underground irrigation channels in their oasis to sup-
port cultivation of wheat, barley, dates, vines, and olives. Numbering perhaps
50,000 to 100,000, they created a state that clashed with three Roman expedi-
tions before establishing a mutually pro¬table relationship, importing Roman
models and even Roman building materials for its stone-built capital, supplying
in return the slaves, semiprecious stones, and other exotic goods that entered
Mediterranean trade. Their raids extended at least as far southwards as Lake
The Garamantian state peaked during the second and third centuries ad
when the Roman colonies were also most prosperous. Both then entered a
slow decline due chie¬‚y to the instability of the wider Roman empire. By
the late third century, Roman garrisons were withdrawing from the North
African hinterland, although agricultural production was generally maintained
for another two centuries on both the coastal estates and the Berber farms
of the interior. As Roman control waned, Berber chieftains created successor
states on the frontier, exploiting both the military skills of pastoralists and the
Impact of metals 33

taxable capacity of farmers. In 508 one such ruler proclaimed himself ˜King of
the Moorish and Roman Peoples™, although by then the empire was no more.
Vandal forces from Spain had invaded North Africa in 429, taken Carthage a
decade later, and extended their power across the region.

sub-saharan africa
Whether Carthage transmitted metalworking to sub-Saharan Africa is one of
the mysteries of African history. Copper and iron, the two metals at issue, both
occur naturally, but rarely, in pure form. In this state, they can be worked by
beating, especially if heated. Metalworking of this kind began about 8000 bc
in western Asia (modern Turkey and Iran). But copper and iron generally
occur mixed with other minerals as ore and must be puri¬ed of them by
smelting at high temperatures. Copper is easier to smelt; the process began in
western Asia soon after 4000 bc and was discovered independently in several
regions, including pre-Columban America. To smelt iron is more complicated,
for iron is usable only if it has certain physical and chemical properties that
smelting must produce. Pre-Columban America never smelted iron. Western
Asia discovered the process early in the second millennium bc. Some believe
that it was also developed independently in eastern Asia, where copper indus-
tries existed to supply metallurgical skills. But the complexity of iron-smelting
caused most regions to acquire the technique by diffusion. Whether Africans,
partially isolated from the Eurasian core, discovered ironworking indepen-
dently is a most dif¬cult question.
Because Africa™s rocks were so ancient, its natural wealth lay chie¬‚y in miner-
als. Copper was a symbol of opulence used for display, much like gold elsewhere,
but copper was rare (except in Central Africa) and the chief utilitarian metal was
iron, which existed widely as low-grade ore. Iron had an especially great impact
on African history because most of the continent had no prior bronze age. In
much of eastern and southern Africa, moreover, there was no agriculture before
the advent of iron, so that it is little exaggeration to say that only access to iron
allowed Africans to create their distinctive civilisation, a point they recognised
by the special status they often gave to ironworkers, either associating them with
the origins of political leadership or fearing them as possessors of dangerous
mystical power. Yet the origins of African metallurgy are uncertain. The dating
of early metalworking sites generally rests on radiocarbon analysis of charcoal
from furnaces, an unreliable source. All radiocarbon dates need to be corrected
by calibration. Many early iron-smelting sites in the northern half of the conti-
nent provide almost simultaneous dates from a period in the middle of the ¬rst
millennium bc when radiocarbon dates are especially imprecise. At present,
therefore, we simply do not understand the history of African metallurgy. All
that is possible is to outline present ¬ndings in order to set out the problem.11
34 africans: the history of a continent

The earliest known metallurgy in Africa was the use of natural copper in

. 1
( 12)