The Archaeology of Africa: January 2005

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/316862692
The Archaeology of Africa
Chapter · January 2005
CITATIONS READS
0 4,018
1 author:
Augustin F.-C. Holl

Xiamen University
106 PUBLICATIONS 620 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
The Neolithization Process: À Global Comparative Perspective View project
The Initial History of Africa View project
All content following this page was uploaded by Augustin F.-C. Holl on 12 May 2017.
The user has requested enhancement of the downloaded file.

Eolss Publishers Co. Ltd.,
United Kingdom
Copyright © 2010 EOLSS Publishers/ UNESCO
Information on this title: www.eolss.net/eBooks
ISBN- 978-1-84826-003-0 (e-Book Adobe Reader)

ISBN- 978-1-84826-453-3 (Print (Color Edition))
The choice and the presentation of the facts contained in this publication and the opinions expressed
therein are not necessarily those of UNESCO and do not commit the Organization.
The designations employed and the presentation of material throughout this publication do not imply the
expression of any opinion whatsoever on the part of UNESCO concerning the legal status of any country,
territory, city, or area, or of its authorities, or the delimitation of its frontiers or boundaries,
The information, ideas, and opinions presented in this publication are those of the Authors and do not
represent those of UNESCO and Eolss Publishers.
Whilst the information in this publication is believed to be true and accurate at the time of publication,
neither UNESCO nor Eolss Publishers can accept any legal responsibility or liability to any person or
entity with respect to any loss or damage arising from the information contained in this publication.
All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopying, recording, or any information storage or
retrieval system, without prior permission in writing from Eolss Publishers or UNESCO.
Every effort has been made to trace and credit all the copyright holders, but if any have been
inadvertently overlooked, UNESCO and Eolss Publishers will be pleased to make the necessary
arrangements at the first opportunity.
The above notice should not infringe on a 'fair use' of any copyrighted material as provided for in
section 107 of the US Copyright Law, for the sake of making such material available in our efforts to
advance understanding of environmental, political, human rights, economic, democracy, scientific, and
social justice issues, etc. If you wish to use copyrighted material from this e-book for purposes of your
own that go beyond 'fair use', you must obtain permission from the EOLSS Publishers.
British Library Cataloguing-in-Publication Data

A catalogue record of this publication is available from the British Library.
Library of Congress Cataloging-in-Publication Data

A catalog record of this publication is available from the library of Congress
Singapore
ARCHAEOLOGY – Vol. II - The Archaeology of Africa - Augustin F.C. Holl
THE ARCHAEOLOGY OF AFRICA

Augustin F.C. Holl
Museum of Anthropology, The University of Michigan Ann Arbor MI, USA
Keywords: Africa, Prehistory, Hominids, Modern Humans, Cultural beginnings,

Oldowan, Early Stone Age, Middle Stone Age, Late Stone Age, Pleistocene,
Scavengers, Hunters, Gatherers, Foragers, Pastoralists, Nomadism, Agriculture,
Horticulture, Metallurgy, Emergent Complexity, Archaic state, Chiefdoms, States,
Empires, City-states, evolution, rock art, Urbanization, Social ranking
Contents
Introduction
1. Cultural Beginnings
2. Acheulean Foragers and the Dispersal of Homo sp.
3. Middle Stone Age Hunter-gatherers and the Emergence of Modern Humans
4. Late Pleistocene Specialized Foragers.
5. From Foragers to Food-Producers
6. African Rock Art
7. The Advent and Spread of Metallurgy
8. Pathways to Complexity
Epilogue
Bibliography
Biographical Sketch
Summary
Africa is an extensive continental land mass stretched almost equally on both north and
south hemispheres. Its vegetations and faunas, extremely diverse and rich, are the
product of a long and complex environmental and climatic history. It is the cradle of
humanity, and its archaeology is precisely the archaeology of humanity up to a certain
point. Early Stone Age assemblages, of the Oldowan variant appeared some 2.7 millions
years ago. From then on, proto-humans and later humans spread all over the continent,
colonizing new habitats and inventing new tools. Most of the continent was inhabited at
the end of the Acheulean period with small scattered groups of foragers. The Middle
Stone Age witnessed sustained trends toward regional diversification with more
elaborate tool-kits, more structured settlement-subsistence systems, and the emergence
of Modern humans. A process of standardization and miniaturization took place at the
end of the Pleistocene resulting in the invention of the bow and arrow and other
composite tools. The intensification in the exploitation of wild resources, transfer of
ideas and contact between neighboring groups led to the emergence of food producing
economies, livestock husbandry first, and agriculture later. Mixed farming and village
life developed during the Holocene, with longer lasting or sedentary settlements. Social
inequality, long distance exchange, and differential concentrations of peoples combined
to fuel the development of complex societies from the Middle Holocene period onward.
Africa was never totally isolated from the rest of the world, but the circum-navigation
and European interference shifted the fate of the whole continent.
©Encyclopedia of Life Support Systems (EOLSS) 21

Introduction
The archaeology of Africa is synonymous with the archaeology of humankind.

Remarkably, in Africa archaeological research is the product of all existing academic
traditions. In addition, the continent has the longest archeological sequence of the world,
from the uncertain beginnings of humanity to the present. This entry summarizes the
substantive achievements of archaeology in shedding light on Africa’s past, from its
remotest beginnings to the emergence of early towns and states.
1. Cultural Beginnings
The investigation of human origins involves many fields of inquiry all part of
palaeoanthropology. Archaeology focuses on the analysis of hominid/human-made
artifacts collected from early human sites. The idea of cultural beginnings is relatively
easy to grasp: culture begins with the manufacture of artifacts. This beginning is
however difficult to pin down in the archaeological record. Does the selection of a
specific kind of non-modified rock indicative of “cultural drive”? Does the use of
slightly modified nature-facts qualify for inclusion in the “cultural universe”?
Setting the demarcation line is tricky, as the selection, use, and manufacture of tools, are
part of a continuum at the interface between the hominids/humans and the world they
live in. The systematic use and production of artificially made tools put the emerging
humanity into a peculiar evolutionary path, away from “animal-hood”. The earliest
proto-human-made artifacts were found in a number of localities distributed along the
Rift Valley in Eastern Africa (Figure. 1). The earliest stone artifacts were discovered
along the Gona River, in the Kada Hadar Member of the Hadar Formation (Ethiopia).
The recovered material, stone artifacts as well as a small amount of animal bones, date
from 2.7 to 2.4 m years, The density is low but the clustering is unmistakable. The
artifacts repertoire consists of two basic categories: cores and flakes (Figure. 2). Each
category is divided in two classes: core-tools/chopper and core fragments, and whole
flake and flake fragments. Some cores and flakes show traces of intentional
modification from their use in subsistence activities. In general, the stone pieces are
irregular in shape, and small in size. The faunal material, highly fragmented, presents
spiral and longitudinal fractures but no cut-marks. At least six mammal’s species are
represented, including hippopotamus, elephant, equuid, and several bovids.
Comparable artifacts associated with animal bones and dating from 2.2 to 1.6 myrs were
recorded elsewhere along the Rift valley (Figure. 1). It is the case in member E and F of
the Shungura Formation (Omo valley) in southern Ethiopia, KBS Industry the Koobi
Fora Formation, and Bed I in Olduvay Gorge. This early stage of proto-human
technology is termed Oldowan. The sites are on river banks and lake shores, encased in
fine sandy to silty sediment. Post-depositional disturbance varies but is moderate to low
in the excavated localities. None of the sites is totally isolated, being nodes in a more
extensive proto-human territory. They are partitioned into three broad categories
depending on their “taphonomic integrity”.

Figure 1: The Rift Valley in East Africa with key Early human sites (from J.D. Clark
1982: 188)

Figure 2: The KBS Industry from the Koobi-Fora Formation (from R. Lewin 1993:
122)
(1) Occupation floors are the less disturbed sites. The cultural deposit is generally thin,
0.10 to 0.30 m, and the archaeological remains are more often than not clustered in
high-density spots. Such sites were used frequently by groups of proto-humans, but the
construction of shelter as suggested at DK IA in Lower Bed I, Olduvay Gorge, is still
controversial. (2) Slightly modified sites have thicker deposits, 0.40 to 1.00 m, with
moderate to high density of cultural remains. They result from short distance re-
mobilization of archaeological remains from their original locations. The spatial
relationship between artifacts is disrupted but their preservation is generally good. (3)
Hydraulic Jumbles are not strictly speaking archaeological sites but “finds spots. The
sedimentary matrix is generally a few meters thick and the density of finds very low.
The artifacts as well as animal bones were generally dragged on considerable distance
within a river drainage and trapped here and there in the river bed.
1.1 Scatters of Stones and Bones
The earliest traces of proto-human culture consist of scatters of stones and bones. Some
of the stones are unshaped nature-facts, others are utilized, and others again are shaped
into cores-tools and sharp-edge flakes. Some animal bones were dropped by proto-
humans, others re-deposited by natural agencies, and others again came from in-situ
dead animals. How and why did these two distinct categories of material remains came
to be intricately mixed in proto-human sites. The distribution of proto-human sites

across the landscape and the differential frequencies of stones pieces and animal bones
have been used to draft “Stone Age Visiting Card”. Special purpose sites display a
narrower range of “cultural by-products”. A single animal species site with a small
amount of stones pieces signals a butchering or meat-procurement locale. A dense
concentration of débitage by-products with very few formal tools points to a workshop.
The occurrence of a handful of bones and stones pieces indicates a bivouac. A large
concentration of bones and stones pieces suggests a frequently visited spot used as
“basecamp”, “home-base”, or “central place foraging”. The above Stone Age sites
systematic was derived from contemporary hunter-gatherers ethnographies. Its heuristic
value is however severely limited if the aim of the investigation is the understanding
and explanation of early proto-human sites. How did proto-humans access resources to
sustain their lives?
1.2 The Hunting Hypothesis
Hunting performed by proto-humans was the obvious explanation for the co-occurrence
of stone and bones in the archaeological record. Bones found in association with stone
tools were accepted ipso-facto as the product of proto-human behavior, the remains of
their meals. The influential “hunting hypothesis” dominated palaeoanthropology
debates up to the early 1970s. According to that theory, hunting is the driving force
behind human evolution; it explains the need for tools as well as their constant
improvement, structures the division of labor between “man-the-hunter” and “woman-
the-gatherer”, justifies the territorial imperative, and provides an outlet for proto-human
and later human inherent violent instincts. Accordingly, hunting provided proto-humans
with much needed high grade food and alleviated in-group tensions. The hunters not
only hunted and brought meat back to be shared among women and children, but also
protected the group from dangerous predators and competitors. Social and cultural
evolution was consequently viewed as driven by male competition for resources and
mates. Fundamentally, the hunting hypothesis was based more on a series of
assumptions than supported by facts. Long-term participant-research on hunter-
gatherers life-ways that started after World-War II challenged the hunting hypothesis.
Meat is a desirable food item but its supply is uncertain. People thus rely more often on
gathered plant products. Plant resources offer the certainty and predictability crucial in
the maintenance and reproduction of proto-humans/humans groups.
1.3 The Gathering Hypothesis
The “Gathering Hypothesis” was a reaction against the “Man –the-Hunter” approach to
human evolution. Its main claims were, in fact, more supported by the empirical
evidence marshaled to strengthen its criticism of the then dominant point of view. The
proponents of “Woman, the Gatherer” theory questioned the importance assigned to
meat in proto-humans/early humans diets and the prominent role attributed to men in
the procurement and supply of meat to the rest of the community. Ethnographic research
among inter-tropical hunter-gatherers has shown meat to be a highly desired food item,
of variable availability, and most of the time, unreliable. Tropical foragers diets were
shown to include more than 80% of plants products with minor if fluctuating proportion
of meat. Accordingly, with proto-humans living in the tropics, their diet is expected to
include more plants than meat. In addition, the access to plants resources is more

reliable and less stressful. The gathering of plant resources carried out by women thus
provided more reliable and sustainable food sources to proto- and early human groups.
The ideas of exclusive pair-bonding between mother and child as well as mate-selection
are other crucial dimensions of “Woman-the-Gatherer” hypothesis. In both cases, male
presence is peripheral and dependent upon the will and decision of the female. In social-
evolutionary terms, most of the reliable food resources are provided by women. Women
carry the babies during their pregnancy and develop strong if exclusive relationship with
them once they are born. Women make decisions and select their mates following their
preferences. Considering these interwoven crucial social roles, “woman-the-gatherer”
hypothesis suggests that women were likely the driving force in the emergence of
human social characteristics.
1.4 The Sharing Hypothesis
“Man-the-Hunter” and “Woman-the-Gatherer” hypotheses are antithetic mirror-images.

The polarized debate of the 1960s-1970s had more to do with contemporary social
issues than early human evolution. In the late 1970s, the “Sharing Hypothesis” was
developed, with wide ranging implications in the emergence of human characteristic
behaviors. According to this hypothesis, the members of any distinct hominid group
characteristically cooperate, support each other, and share information and resources.
This cooperation and sharing triggered the development of elaborate communication
that ultimately led to languages. It generated the characteristic division of labor along
gender lines to insured the procurement of much needed resources. Information on the
distribution and timing of resources, as well as potential dangers, collected by each
member of the group is pooled and shared. The home-base or Central Place Foraging
(CPF) was the main locus where the hominid group was supposed to enact the different
facets of the Sharing hypothesis. In simple terms, the sharing hypothesis tried to
reconcile the “Man-the-Hunter” and “Woman-the-Gatherer” hypotheses into one more
encompassing model. Men hunt, women gather, and share the products of their
respective subsistence forays out of the home-base. The sharing hypothesis was widely
adopted as the most “parsimonious” explanation for the emergence of human social
systems. It was however anachronistic, plagued by the reliance upon Homo-Sapiens
analogs to understand and explain proto-human behavioral characteristics. A
devastating critique was launched against the “Home-Base–Sharing, and hunting
hypotheses. It suggested that proto-humans were more or less skilled scavengers,
triggering a new round of debate on the origins of human behavioral patterns.
1.5 The Scavenging hypothesis
As indicated by the archaeological record from Koobi-Fora and Olduvay, protohumans

access to potentially large supply of meat from big-size mammals is compelling. Sharp
flakes with “meat” polish and cut-marks on long bones are clear indications of both the
procurement and consumption of meat. How did such a system operate? According to
the “scavenging hypothesis”, proto-humans relied on scavenging strategies to acquire
meat. They were often secondary feeders exploiting marrowbones and the left-over of
effective and specialized predators. Scavenging is nonetheless a complex procurement
system. It involves spotting animal carcasses across the landscape, avoiding and/or
driving away competitors, collecting pieces of the kills, and moving to safer locations.

However, while scavenging is likely the best explanation for proto-human access to
large animals carcasses, it does not completely rule out the possibility for hunting of
small size game
In summary, the period running from ca. 2.7 to 1.65 million years was characterized
by diversification in proto-human species. At least four Australopithecus species, A,
afarensis, A. Bosei, P. robustus, and A. africanus, and one Homo habilis are
represented in the archaeological record. The “bushy pattern” of inter-species
evolution ended with the emergence Homo ergaster at the 1.7-1.6 million years
boundary. Based on the most recent discussions of the issues pertaining to proto-
human patterns of behavior, plant gathering was likely the most reliable component of
the their food procurement strategies. Meat was accessed from time to time and its
overall availability may have been strongly seasonal. Proto-humans may have
developed ways of conveying and communicating information and emotions to
fellows within and outside their respective groups. It was certainly a kind of language;
spoken language – speech – was a much later development.
2. Acheulean Foragers and the dispersal of Homo sp.
Sites dated to the Acheulean Period, from 1.7/1.6 myrs to 200/150,000 years ago, are
found in almost all parts of the continent (Figure: 3). The regional density of sites is
relatively high in South Africa, East Africa along the Rift Valley, as well as the Nile
valley and the Western Desert. It is relatively low in the rest of the continent,
particularly in the Sahara, Congo basin, and West Africa. In the Sahara, sites are found
in the vicinity of mountains, from the eastern Tibesti to the sand plains of Mali and
Mauritania. In West Africa, they have been recorded in the lower Falémé, the Atlantic
coast of Ghana, and the Jos Plateau in Nigeria.
In general, the earliest Acheulean sites are found along the Rift Valley in Eastern Africa
and the South African plateau. The northwest coast in Morocco and Algeria as well as
the lower Nile and Eastern Sahara appear to have been colonized by groups of Homo
erectus foragers between 1.2 and 1 million years. Most of West and Central Africa were
colonized much later, between 500,000 and 250,000 years ago. The Acheulean lasted
for more than one and half million years, from 1.6 million to 150,000 years. It is
primarily a tools-kit comprised of bifaces and a range of formal tools on retouched
flakes. Three classes of bifaces have been recorded: hand-axe, pick, and cleaver. Their
relative frequency is extremely variable within and between assemblages. The earliest
and “sudden” materialization of the Acheulean complex (Figure: 4) was documented at
EF-HR in the Upper middle Bed II (Olduvay Gorge). The hand-axe appears to have
been a “perfect” multifunctional tool, a characteristic that can explain the technological
stasis and the extraordinary length of the Acheulean “tradition”. There is nonetheless an
increase in the diversity of tools and a significant refinement in the manufacture of
bifaces in Middle and Late Acheulean sites (Figure: 5).

Figure 3: The general distribution of Acheulean sites (from J.D. Clark 1982: 185)

Figure 4: Sample of Early Acheulean artifacts from the Middle Bed II, Olduvai (from
J.D. Clark 1982: 194) 1 – Biface on flake; 2 – cleaver on flake; 3, 4 – bifaces

Figure 5: Sample of late Acheulean artifacts from Sidi Zin, Tunisia (from J.D. Clark
1982: 203)
1 – handaxe with bifacial trimming; 2 – handaxe with one partly trimmed face;
3 – Bifacial handaxe; 4 – small bifacially trimmed tool; 5, 6- double
sidescrapers;
7 – denticulated nosed scraper; 8 – irregular bifacial scraper; 9, 10 –
sidescrapers;
11 – irregular thick scraper; 12 – nosed scraper; 13 – microchopper.

As is the case for proto-human subsistence systems discussed above, patterns of

Acheulean subsistence and settlement dynamics are still poorly understood. The
hunting. versus scavenging debate does also apply to Early and Middle Acheulean
contests, with a significant shift toward hunting during the Late Acheulean. It is very
likely that Acheulean Homo ergaster were “broad spectrum foragers” relying on
scavenging for their optimal supply of meat but able to hunt small size game. A
systematic investigation of Olorgesailie Acheulean sites (ca. 700,000 to 400,000 years
ago) characteristics and location lead to the suggestion of “catastrophe hunting” of
Gelata baboons at DE/89A Site. Accordingly, one or many baboons’ troops were
hunted and butchered at that spot. Because baboons and humans are poorly equipped for
night vision, nighttime was suggested as the optimal hunting window. Baboons sleeping
in trees were “bombarded” with heavy spheroids stone “missiles”. From this
perspective, Olorgesailie Acheuleans were skilled hunters who could mount nighttime
expeditions. This reconstitution of Acheulean hunting abilities was disputed. Many
other taphonomic factors could have contributed to the formation of DE/89A
archaeological assemblage. In this dissenting perspective, the Acheuleans were still
mainly foragers and scavengers, and marginal hunters; a feature considered to have been
dominant in human evolution up to end of the Middle Stone Age.
The use of fire and the manufacture of tools fit for hunting are documented at the end of
the Acheulean period in different parts of the world. Direct evidence is however still
elusive in Africa. The association between Acheulean artifacts and burnt sediments
documented at Chesowanja dated to 1.4 myrs was shown to be accidental. A burnt
hand-axe from Lagreich in the Tilemsi valley was dated to 250,000 BP. The fire-
hardened wooden implement from Kalambo Falls, a possible wooden spear date to
200/150,000 BP. As far as site location strategy is concerned, there is a trend toward
diversification of settlement types. In the south of the continent (Figure. 3), a larger and
increasing number of caves were used. Special purpose sites, for the procurement of raw
material, hunting and/or butchering of large animals carcasses also emerged and became
part of the Acheulean land use system. An Acheulean hunting site has been recorded at
Adrar Bous in the Sahara. Acheulean quarries and workshops are documented at Blaka
Kalia I and Mochi Sounosso in Central Sahara. The intriguing site of El-Beyyed in
Mauritania, consists exclusively of well made hand-axes without any trace of débitage
by-products. And finally, the only case of built feature, a raised rectilinear platform
made of river cobbles interpreted as a “living floor “, is documented at Melka Konture
in Highland Ethiopia,”.
Beyond the readily accessible chert and lava cobbles transported and deposited by
water courses, fine-grained material, like flint, and obsidian, progressively entered
the technological repertoire of the Acheuleans. The territorial range for the
circulation of raw material increased significantly to a few tens of kilometers, in
contrast to the previous stricktly local procurement of the Oldowan.
3. Middle Stone Age Hunter-gatherers and the Emergence of Modern Humans
The Middle Stone Age (MSA) is critical for the emergence of Modern Humans. This
process characterized by a sustained trend toward regional diversification was kicked
off between 200,000 and 100,000 years ago. The main trends as indicated by stone tools

production techniques, settlement types, and sites location strategies are more or less
homologous across regions, but played out differentially and generated distinct
evolutionary pathways. In general MSA assemblages lack core-tools. All tools are made
from flakes or blades obtained through the implementation of sophisticated stone
knapping technique: the Levallois technique. The core element of the Levallois
technique is cognitive in nature. The tool-maker has a flake or blade template in mind
and designed the succession and coordination of the knapping steps to achieve that pre-
determined goal. The selected cobble is prepared through a succession of carefully
removed primary flakes designed to create a striking platform. The flakes obtained are
re-shaped and retouched using a broad assortment of techniques to make tools, such as
points, scrapers, denticulates, burins, awls, etc.. MSA people were effective hunter-
gatherers who developed the earliest intensive exploitation of marines resources as seen
at Klasies River Mouth and Haua Fteah, at the southern and northern tips of the
continent.
3.1 Southern and Eastern Africa
The continent can be divided into five distinct regions, each with a slightly different
MSA sequence. The Southern African MSA sequences, from Angola southward, is
divided into four chronological units, MSA I, II, Howieson’s Poort, and MSA III. The
Howieson’s Poort complex is an early microliths component that emerged between
80,000 and 75,000 years ago (Figure. 6 - 9). In fact, it is inter-fingered with other MSA
III assemblages. By far, South Africa has the largest sample of excavated MSA sites in
the whole continent. Blombos, Bushman Rock Shelter, Duinefontein 2, Eland Bay
Cave, Klasies River Mouth, are some of the key sites that have provided researchers
with challenging material during the last thirty years.

Figure 6: Sample of Middle Stone Age I artifacts (from R.G. Klein 1984: 202)
1, 3, 6, 10, 12 – denticulates; 2 – endscraper with lateral retouch; 4, 7 – scrapers;
8 – retouched flake; 9 – denticulate endscraper; 11 – slab with heavy duty
unifacial retouch; 13 – weathered slab with heavy duty retouch.

Figure 7: Sample of Middle Stone Age 2a artifacts (from R.G. Klein 1984: 204)
1 – flake with proximal retouch; 2 – tip of bifacial point; 3 – sidescarper;
4 – endscraper; 5 – blade with proximal retouch; 9, 7 – denticulates; 8, 9 –
blades

Figure 8: Sample of Middle Stone Age 2b artifacts

(from R.G. Klein 1984: 205)
1, 3 – unifacial points with proximal retouch; 2, 4, 5 – endscrapers; 6 – stemmed
flake; 7, 8 – sidescrapers

Figure 9: Sample of Howieson’s Poort artifacts

(from R.G. Klein 1984: 206)
1-7 – segments; 8-9, 13-14 – truncated pieces; 10-12 – trapezoids; 15 –
trapezoid with notch; 16 – strangulated backed piece; 17 – multi-notched piece;
18-19 – triangles.
The systematic of Eastern Africa MSA is modeled after the South African one with

however local particularities. The Post-Acheulean period is more obscure and seems to
have developed differently on both sides of the East African Rift valley. In the west,
from lake Victoria westward, including the north Central African rainforest, the
discovery of series of heavy-duty core-tools scattered over Sango Bay terraces in
Uganda prompted the “identification” of a new and specific prehistoric culture, the
Sangoan. An issue that ignited a long lasting debate on the significance of this post-
Acheulean industry. More or less standard MSA complexes, with low Levallois
indexes, are found east of the Rift Valley. No equivalent to the Howieson Poort
complex has been documented, with important sites stretched along the Rift valley,
from Olduvay (Ndutu beds) and Laetoli in Tanzania to Gademotta in Ethiopia.
The Sangoan, originally thought as an adaptation to the equatorial forest, and as such
spread as far west as Sierra Leone and Guinea, was re-interpreted as “a special-activity”
Sangoan facies transgressing both time and space boundaries.
The western confines of the Rift valley and the Equatorial rainforest witnessed a
different MSA evolution, starting with the widespread Sangoan tradition. Despite a
sustained improvement, the developmental sequence of North Central African techno-
complexes remains poorly understood. In the east, Katanda sites from the Semliki valley
dated to ca. 90,000 years contain early specimens of bone tools, including harpoons,
along with MSA industry using local quartzite, quartz, and chert. Fish and mammals
bones attest for the practice of fishing and hunting. The gathering of plant resources was
very likely the most reliable component of the food procurement system suggesting that
the MSA ‘Katandans’ were fully-fledged hunting and gathering communities. Sangoan
and Sangoan-related assemblages, dating up to 40-38,000 years, are found elsewhere in
the area, from Kalambo Falls in the southeast to western Congo and southeastern Gabon
in the northwest.
The Lupemban complex replaces the Sangoan and related components in the later
portion of the MSA and appears to range in date from ca. 45,000 to 20/15,000 BP. Its
geographic distribution is however narrower, confined to the west of the equatorial
rainforest. The Lupemban includes flakes, discs, and small retouched stone artifacts.
The characteristics tool that can measure 0.30 m in length is a bi-facially trimmed laurel
leaf-shaped tool, pointed at one or both ends. Pressure flaking was certainly used for the
production of such delicate artifacts as suggested by the narrow and elongated flake
scars. Hafted to long wood shafts, these thin bifacial blades may have been very
efficient hunting spears.
3.2 Western and Northern Africa
There is a genuine dearth of information on MSA sites in West Africa. Exploratory

research was nonetheless carried out at a number of sites, Asokrochona and Tema
West (Ghana), Attingue and Anyama (Cote d’Ivoire), Jos Plateau (Nigeria), Figuil-
Louti (Cameroon), and the Falémé Valley (Sénégal). None of these sites can however
be claimed to be undisturbed.
There are much more commonalities between the Sahara, the Northeast, and the North
of the continent. The terminology itself is strongly reminiscent of that of West European

Middle Palaeolithic. A Mousterian and Mousterian-like techno-complex is widespread,

found with local variations in Nubia, the Nile valley, Sahara, and North Africa. Bir
Sahara-east and Bir Tarfawi data suggest the Mousterian from the northern hemisphere
to date from ca. 200,000 to 70,000 years. The number of sites assigned to the
Mousterian period is relatively small but they are widely distributed. Djebel Irhoud in
Morrocco, Taforalt, El Guettar, Hajj Creiem, and Haua Fteah along the north African
coast, Chemidour, Bilma, Ounianga-Kebir, Esselesikine, and Adrar Bous in Central
Sahara; Khor Abu Anga, Ed Debba, Dakhla Oasis, Kharga Oasis, in the Nile valley and
eastern Sahara.
The Aterian Industry is an off-shot of the Mousterian Tradition. It is confined to North

Africa and the central part of the Sahara, with a single eastern occurrence at Site E-78-
11, in the Nile Valley. It is characteristically a lithic techno-complex with a varying
amount of stemmed tools. As a technological ‘breakthrough”, the stem was devised to
enhance the hafting of stone tools, points, end- and side-scrapers, borers, etc. The
earliest Aterian deposits, found at Uan Afuda cave in the Tadrart Acacus (southwestern
Libya), date from 90,000 to 67,000 years BP. The Aterian is likely a North African and
Saharan phenomenon that lasted up to 25,000-20,000 years ago. Some sites are found
along the Mediterranean coast in the north, as is the case for Haua Fteah (Libya),
Karouba and Bérard (Algeria), the Grotte des Contrebandiers and Dar es-Soltan
(Morocco), Bir el Ater, El Guettar, Wadi Gan, Ain Fritissa and Rhafas in north Africa
hinterland, Wadi Shati, Adrar Bous, and Seggedim in Central Sahara.
o
Seggedim is an unusually large 3.5 ha site located in Central Sahara at 20 12’ N and
o
12 58’ 30” E. It is divided into three distinct areas, one consisting of a 6.90 m in
diameter circle of a fenced feature with a central hearth. There are two subsidiary
hearths in an area filled with burnt stones, charcoal and charred wood. Seggedim’s
wind-break or shelter was built and settled between 40,000 and 35,000 years ago; it is
so far, the most convincing case of intra-site patterning recorded in north-tropical
Africa. The accumulation of débitage waste within instead of outside the delineated
circular space is nonetheless puzzling. The subsistence /settlement systems devised by
the Aterians are still very poorly known. Site like Bir el Ater along the Wadi Djebbana
in Algeria and Seggedim in Central Sahara tend to suggest multi-occupation residential
settlements. Such sites may have been key home bases surrounded by varying number
of special purpose localities. Seen from the angle of their accumulated cultural remains,
the Aterians were probably very effective hunter-gatherers who relied on both
residential and logistic mobility to optimize their resources procurement strategies.
3.3 The Saga of Modern Humans
Despite frequent reports of isolated stemmed stone tools in West Africa, especially in
Mauritania, Mali, Niger, and Chad, there is so far no convincing evidence for the
presence of “Aterians” that far south. Does that mean that modern humans reached
West Africa rather late after spreading all other the continent? The issue of the
emergence of modern humans is hotly debated. The “Candelabra Model”, at one end of
the spectrum, emphasizes regional continuity in anatomy resulting from the
maintenance of gene flow between Africa, Europe, and Asia. Accordingly, there were

neither significant migrations nor population replacement but parallel evolution toward
Modern Humanity. The “African Eve model”, at the other end, derived from
mitochondrial DNA analyses, advocates a new “out of Africa” resulting in the migration
of modern humans and the replacement of other non-modern humans’ population all
over the planet, in Europe and Asia. The recent and spectacular discovery of Homo
floresiensis from Flores Island in the Indonesian archipelago suggests the evolution of
humanity to be much more complex that previously thought. At the same time however,
this discovery makes the parallel emergence of distinct strands of modern humans more
and more unlikely. Fossil remains attributed to early modern humans come essentially
from three sites: Klasies River Mouth (KRM) and Border Cave in South Africa, and
Omo in Ethiopia. The KRM remains were collected from deposits dated from 94,000 to
88,000 years. The Omo specimens are dated to 130,000 years and the Border cave ones
from 62,000 to 82,000 years. The chronology of these remains still needs to be worked
out. For some, it is not certain that modern humans were present in Africa around
100,000 years ago. The human remains from KRM do continue a pattern of change
reflected in the transition from the Kabwe people to the African Transitional Group.
There is nonetheless an increasing body of indirect evidence supporting the early
existence of modern humans in Africa even if not older than 100,000 years; it is the case
for the art pieces from Blombos; the bone tools from Katanda that hint to symbolic
capability and sophisticated craftsmanship.
4. Late Pleistocene Specialized Foragers.
Depending on areas, the MSA drew to its end between 40,000 and 20,000 years ago. It
was followed by different techno-complexes including stone tools, bone tools,
composite tools, as well as fired clay containers. Depending on academic traditions, all
or part of this time-segment is called “Terminal Palaeolithic”, “Epi-palaeolithic”,
“Mesolithic”, and even “Neolithic”. The Late Stone Age (LSA) of the current
archaeological nomenclature is partly confusing. Different remedies have been
suggested to clarify the ambiguity of this concept. In one of the attempts, LSA sites are
partitioned into three successive chronological cohorts, Lower LSA (40/35,000 – 10,000
years ago), Middle LSA (>10,000 – 6,000 years ago), and , Upper LSA (>5,000 – 2,000
years ago). In another, it is referred to as “Advanced Foragers” or, as in the Southern
Africa case, subdivided into a multiplicity of shorter segments like Early Later Stone
Age (ELSA) (before 19,000 bc), Robberg (ca. 7600 – 19,000 bc), Oakhurst (ca. 6000 –
10,000 bc), Classic and early Wilton (ca. 2500 – 7500 bc), Post-Classic Wilton (ca.
2500 bc – ad 200), Ceramic Wilton/Post Classic Wilton (ca. 100 – AD 1870), and,
Smithfield (ca. Ad 1200 – 1870). In general, local “cultural” taxonomy shifts to an
inflationary loop, suggesting an accelerated pace of intra-regional differentiation of
cultural traditions along with a sustained diversification of material products. There is
nonetheless a small number of important variables that can be used to organize the
cultural remains into acceptable if broad categories: (1) the production and use of blade
and bladelets; (2) the production and use of microliths; (3) the production and use of
pottery.
The production of blades and bladelets blanks has deeper roots in the preceding MSA/
Middle Palaeolithic period. The preparation of the selected block of raw material is
geared to create an elongated core, prismatic in shape, from which blades and bladelets

are struck from one (unipolar) or both (bipolar) ends. There is significant local variation
but blade technology resulted in an amazing standardization of lithic tools
manufacturing process. West and North Central Africa do not have convincing
evidence for the presence of blade making and using foragers groups. Shum Laka and
Mbi crater (Cameroon), Bingerville Highway (Cote d’Ivoire), and Badoye (The
Gambia), in West Africa, as well as Matupi cave in north-eastern central Africa, do not
seem to have a significant component of blades/bladelet in their artifacts’ repertoires.
Elsewhere, and in varying degrees, in Southern, Eastern, Northeastern and Northern
Africa, blade/bladelet industries precede microlithic ones. In North Africa, the
Iberomaurusian (Figure. 10), dated from 22/20,600 to 16,000 years ago, is found along
the Mediterranean coast and the northern slopes of the hinterland mountains.
Figure 10: Sample of stone artifacts from the Iberomaurusian site of Tamar Hat (from F.
Klees and R. Kuper 1992: 127)

The Iberomaurusians were mobile foragers exploiting the extensive Mediterranean

coastal plain when the sea was more than 100 m below its present-day level during the
Late Glacial Maximum. They also used some strategically located sites, like Tamar Hat
rock-shelter as a hunting station to intercept the migrating herds of Barbary sheep
(Ammotragus lervia). Tamar Hat fauna consists of more than 95% of barbary sheep
bones, with the samples strongly skewed in favor of juvenile males bones. Such a
pattern suggests the implementation of consistent herds management procedures, in this
case, the intentional culling of un-needed males. Tamar Hat excavator interpreted the
faunal material as convincing evidence for an early experimentation with animal
domestication. In South Africa, the Robberg Industry is stretched on a shorter time
segment, from 19,000 to 7.500 years ago. The eland (Taurotragus derbianus) is the
predominant prey animal, indicating a strong degree of hunting specialization.
In general however, the LGM with its sharpened trend toward increased aridity resulted
in the drastic shrinkage of the equatorial rainforest and the phenomenal extension of the
Saharan desert that expanded some 600 km south of its present-day limit, triggering a
significant drop in Late Pleistocene foragers populations in North tropical Africa. The
situation was different in the southern hemisphere, where late Pleistocene hunter-
gatherers population was on the rise, with parietal art dated to 27,000 B.P. at Appollo 11
Cave in Namibia..
The Sahara appears to have been completely vacated by humans, with some shallow
sites probably located in or around the Central Saharan mountains. A dense string of
Terminal Palaeolithic sites are found along the Nile; “Iberomaurusians” settled on the
Mediterranean littoral and northern slopes of Northwest African mountains; the forest
relic from Western Cameroon was exploited by Shum Laka and Mbi Crater foragers.
Kisese II (Tanzania), Twilight cave and Lukenya Hill (Kenya), provide a glimpse on
the Blade/baldelet complex dated between 18,500 and 15,000 B.P. in eastern Africa.
From 15,000 B.P. onwards, microlithic complexes developed and spread all over the
continent. The bow and arrow were then part of the universal tool-kit of hunters.
Miniature stone tools, generally of geometric shapes – triangle, crescent, trapeze, etc. -
were used as parts fit at one extremity of a straw or wooden shaft to make arrow-heads
with their lateral barbs. The Tshitolian complex in the Equatorial forest and the Eburran
I and II (ca. 12,000 – 10,300 b.p.) in eastern Africa emerged during the Global Warming
at the end of the Pleistocene. In West Africa, sites with microlithic implements area are
found in the savanna, savannah-forest mosaic, and forest, along the Atlantic coast as
well as in the hinterland.
The Capsian complex is widespread in northwest Africa and was characteristically more
developed in the wetlands of the Bas-Sahara. The Capsians were very likely organized
into small highly mobile foragers’ groups. They developed a settlement system
articulated on the exploitation the Atlas Mountains during the summer and the intensive
gathering of aquatic resources of the Shotts during the winter. They had an extensive
and diversified bone industry, decorated empty ostrich eggs used as liquids’ containers,
and hunted small game on a “search and encounter” basis.
There are interesting cultural novelties during the later part of the Pleistocene as far as
territoriality is concerned. There are extensive Iberomaurusian cemeteries in the

Maghrib, smaller ones in the Nile valley(Figure. 11); and isolated burials at Wadi
Kubbaniya (ca. 17,000 b.p,), Ti-n-Hanakaten and Amekni (ca. 9,000 b.p.); and Iwo
Eleru (ca. 11,000 b.p.). The emergence of cemeteries points to a shift in the relationship
between humans and land. The “Kubbaniyans” appear to have a restricted territorial
range, on the west bank of the Nile. The shift toward smaller territories in combination
with environmental change, climatic fluctuations, and population growth, may have
triggered organized violence, in this case, warfare. Jebel Sahaba cemetery, dated from
12,000 to 10,000 B.P., contains the remains of 59 individuals (Figure. 11), infants as
well as adult. Twenty-four of the deceased have a stone arrowhead imbedded in the
bones. Ten of the injured individuals were adult male, 9 female, and 5 children. The
frequency of violent encounters is hard to determine; the possibility of infrequent
raiding with one or two wounded at irregular intervals comes to mind.
Figure 11: Map of the Cemetery of Jabal Sahaba (from F. Wendorf 1968: 956)
As indicated by the archaeological record from Tagalagal, Amekni, and Launay Site,

pottery is used by groups of Late Pleistocene/Early Holocene foragers in and around the
Central Sahara mountains between 10,000 and 9,000 B.P. (Figure. 12). The earliest
manifestations of this significant technological innovation ranging in time from ca.
9,500 to 8,000 B.P. are spread out on a relatively large part of north-tropical Africa,
from the eastern Sahara to the Bandiagara Cliff (Mali), including the Tibesti and
vicinity, Tadrart Acacus, Tassili-n-Ajjer, Ahaggar, and the Air. The vessels found in
Eastern Saharan are simple open bowls used in instances of overt consumption of food
and beverages but not cooking. In Central Sahara, the range of vessels shapes, generally
decorated with comb-impressed wavy and dotted wavy lines (Figure. 12), is broader.
They include large hemispheric pots with round base, hemispheric pots with inverted
rim, and smaller globular bowls (Figure. 13). These vessels were used for the storage of
liquids, wild grain of sorghum, millet, Panicum, and other foodstuff. Pots in the smaller
size range were used for cooking and serving, and the smaller globular bowls for
serving/consumption.
Figure 12: Sample of sherds from Tagalagal with dotted wavy-line and comb-impressed
Decoration (from A.E Close 1987: 222)

Figure 13: Sample of vessel shapes from Tagalagal (from A.E Close 1987: 221)
Most of the Late Pleistocene archaeological sites include a significant amount of coarse
stone artifacts; they are distributed into a range of tools classes with the most frequent
being grindstones, grinders, pestles, mortars, hammer-stones, and infrequently tethering
stones. Wild sorghum is heavily attested in the Wadi Kubbaniya and Nabta Playa
archaeological record. Wild millet and Panicum sp. are predominant in the Tadrart
Acacus. The systematic exploitation of large patches of wild grain may have anchored
Late Pleistocene foragers around some core components of the landscape; this may have
been the case for the eastern Sahara playa lakes, the wadi drainages of the Central
Sahara massifs, as well as the shores of Saharan Early Holocene lakes. Fishing and the
extensive exploitation of aquatic resources were more predictable and reliable than
hunting of large mammals. The intensification in the exploitation of resources confined
to smaller and smaller territories put significant pressures on humans, animals and
plants alike. Late Pleistocene foragers devised different solutions that led later to
different evolutionary trajectories. In many case, the foraging life-style was the best
option carried on for many more millennia. In other circumstances, foragers became
part-time herders and later full time pastoralists. In a third option, groups living in the
humid inter-tropical and equatorial zones adopted horticultural systems that later
resulted in the domestication of yams. And in a fourth direction, mixed farming with a
differential combination of livestock herding and plant cultivation, was invented or
adopted from neighboring areas.

5. From Foragers to Food-Producers
5.1 Early Foragers-Herders
The pioneer stage of the shift toward food-production took place in the northeastern part
of the continent, in an area stretching from the Tadrart-Acacus to the Eastern Sahara and
the Nile valley. A number of archaeological sites dating from ca 9,500 to 8,000 b.p. and
containing small amounts of cattle bones have been excavated, as is the case at Nabta
playa, Bir Kiseiba, and Ti-n-Torha. There is an ongoing debate on the nature and
meaning of these surprising finds. Even the identification of some of these remains
keeps shifting from one species of large bovid to the next. However, when the research
focus is shifted away from the only consideration of bones to the inclusion of other
parameters of human-animal relationships, these scattered remains take on a totally
different meaning.
First and foremost, Cattle mtDNA analyses have shown that African, Eurasian, and
Southeast Asian wild cattle populations belong to distinct stocks that split some 25,000
years ago. Middle Eastern and Indian cattle reached the continent later, making the early
cattle remains the likely result of local processes of domestication. In the Central
Sahara, the occurrence of the disputed large bovid bones takes place within a context of
broadening faunal spectrum. The presence of large bovid (cattle) was followed by a
drastic reduction of the breadth of faunal spectrum, from 40 to 25 species in phase 3 and
4 respectively. In eastern Sahara, the presence of a walk-in-well and the significant size
discrepancy recorded in the faunal spectrum, otherwise composed essentially of small
size gazelles and desert adapted animals, tends to indicated the herding of a small cattle
herd that provided the humans with secondary products, principally milk.
5.2 The Rise of Pastoral-Nomadism
Between 7,000 and 6,000 B.P. livestock herding systems spread in different parts of the
continent, with however important regional variations. Three distinct trajectories have
been documented for the North African and Mediterranean coast.
In one trajectory, sheep-goat herding appeared rather abruptly in the Haua Fteah
sequence of Cyrenaica (Libya). Lambs’ burials recorded in the archaeological
record suggest the emergence of special human-animal relationship. The large cave
of Haua Fteah was settled on a seasonal basis by groups of Mid-Holocene herders
practicing transhumance between the coastal plain and the hilly hinterland. The
whole of the sheep-goat herding system documented at Haua-Fteah resulted from
diffusion from the Near-East, where these species were domesticated one or two
millennia earlier.
The second trajectory encompasses a number of “Cardial Neolithic” sites found in

northern Morocco. Cattle and sheep-goat bones were recorded in the lowest deposits of
sites like El Khril near Tangiers. The finds also include pottery, grinding equipment,
and flint arrowheads but no evidence for the practice of agriculture. The Moroccan case
is the western- and southern-most extension of the Mediterranean Cardial Neolithic
complex which is otherwise found along the southern European coastline. Small groups

of Cardial Neolithic herders spread in northern Morocco around 7,000 B.P. and
practiced short-range nomadism.
Finally, the third trajectory is documented in the Aures mountains in Algeria, at

Cappeleti cave located at 1,540 m above sea level. The cultural sequence spans over
more than 2000 years, from ca. 6,500 to 4,360 B.P., assigned to the “Neolithic of
Capsian Tradition”. Cattle and sheep-goat bones are largely predominant all along the
site’s sequence, with however a significant trend toward decreasing cattle presence.
The site was used as a spring-summer-autumn camping place for small groups of
transhumant herders. They spread all over the lowlands during the rest of the year,
especially in winter times.
Pastoral-nomadic societies, with cattle and sheep-goats herds spread all over the Eastern
and Central Sahara between 7,500 and 6,500 B.P. The mountains and surrounding sandy
lowlands were settled alternately depending on seasonal variation in rain, surface water,
and grass. In the central Sahara, hundred of hearths spread all around the mountains
were recorded, mapped, and dated. These shallow “Neolithic hearths” are the remains of
shifting bivouacs of small pastoralists groups scattered all over the landscape during the
rainy portion of the year. The rest of the time, they used the river drainages as
transhumance corridors, dwelling in rock-shelters, caves, and open-air sites, at Ti-n-
Torha, Ti-n-Hanakaten, Uan Muhhuggiag, Telocoat, etc., along rivers shores as well as
higher up in the mountains. This differential use of the landscape is mapped on the
ground by the distribution of two kinds of archaeological features: cemeteries and rock-
art sites. Cemeteries, with stone monuments are generally found in foothills locations,
river shores, and the periphery of the massifs. Rock art stations on the other hand are
most of the time situated in highland. There is an area of overlap in the regional
distribution of these two kinds of sites. They tend to indicate the assignment of
complementary social, ritual and/or religious potency to distinct parts of the landscape.
In some rare cases however, as is the case at Djorf Torba, there are paintings of horses
found in a megalithic burial. Besides their representation in rock-art imagery, the
involvement of livestock in elaborate rituals is documented from ca 7,500 B.P. by
unambiguous archaeological evidence. Such evidence consists of burials of sacrificed
cattle at Adrar Bous (ca. 7,000 B.P.) and Chin Tafidet (ca 4.500 – 3,400 B.P.) in Niger,
as well as Nabta Site E-75-8 in eastern Sahara. The latter is considered “as a Late
Neolithic ceremonial cente”. The site complex includes cattle burials in stone tumuli.
One, with a “massive “ mushroom-shaped carved stone (Figure. 14), is associated with
“a north-south oriented alignment of nine large sandstone blocks, set at about 100 m
apart, and partially imbedded in playa sediments” with a circle of smaller sandstone
slabs at the north end of the axis.

Figure 14: View of the shaped mushroom’like table-rock from the cow burial of Nabta
Playa (from Wendorf and Schild 1998: 114)
From the north-northeast of the continent, livestock and pastoralists life-ways spread
south and west. The Nile valley and surrounding grass- and steppe-lands were used as a
corridor for the transfer of livestock, and possibly pastoralists to Eastern and Southern
Africa. Evidence for cattle, sheep and goat dating to 5,000 – 4,000 B.P. are found in
Lake Turkana basin and the Kenya rift. The new life-way, termed “Pastoral Neolithic”

involved intensive fishing, livestock husbandry, hunting and gathering by small groups
of highly mobile folks. Livestock herding spread to the eastern and southern part of east
Africa some 1000 years later, around ca. 3,000 BP. Interaction with local foragers was
very likely the rule and not the exception. Livestock expanded further south through two
very distinct processes. Bantu-speaking mixed-farmers brought livestock south to
Zambia, Zimbabwe, Mozambique, and Transvaal and Natal Provinces in South Africa
between ca. 2,400 and 1,700 BP. Khoisan speakers adopted pastoralism and initiated
the spread of livestock in southern Angola, Botswana, Namibia, and further south to the
Cape Province in South Africa.
Livestock herding spread west and south of Central Sahara through the Tadrart Acacus
gateway. The chronological gradient spans some 2000 years, from 7,000 BP at Adrar
Bous , Arlit, and Delebo cave, to 4,000 BP when the Eghazzer basin, Azawagh valley,
Tilemsi valley, Kintampo Culture area, and Dhar Tichitt-Walata, as well as most of the
west African Sahel and Savannah were reached. At the very end of its westward
expansion, in the Ghanaian Savanna, Tilemsi and Dhar Tichitt-Walata, livestock
husbandry was in varying degrees combined with agriculture.
5.3 Mixed-farming and Emergent Village Life
Parallel to the emergence of “transhumant pastoralist traditions in North Africa, Central

Sahara, the Nile valley, and Eastern Africa, mixed-farming, combining the practice of
agriculture and livestock herding developed in the Nile Delta and the Fayum around
7,500 – 7,000 B.P.. In these cases, barley, emmer-wheat, and flax were the principal
cultivars. Livestock was represented by cattle, sheep, goat, as well as pigs, dogs, and
possibly donkeys. Accordingly, the advent of this early village mixed-farming tradition
was strongly influenced by the Near-Eastern complex. However, barley, emmer-wheat,
and other Middle-eastern plants could not spread without a long process of adaptation
to the summer-rain system characteristic of Inter-tropical Africa. Mixed-farming spread
south along the Nile river, with the livestock component more successful along the
north-south gradient, as shown by sites like Kadero, El Geili, Shaheinab, etc, in Sudan
and later in Highland Ethiopia.
Paradoxically, the practice of agriculture is a much later development. The intensive

exploitation of sorghum during the Early Holocene is documented in eastern Saharan
playa sites but domesticated forms emerged much later, around 4,500-4,000 B.P. in
Nilotic Sudan. The Middle Eastern agricultural complex spread along the
Mediterranean coast. The cultivation of African plants has preceded domestication by
a considerable time margin. The difference between wild and domesticated forms of
bulrush millet (Pennisetum sp.) is striking. Grains of the latter are at least ten times
bigger than those from the former. The earliest traces of domesticated millet, dated to
ca. 4000 BP , are documented in the Dhar Tichitt (Mauritania) and Birimu (Ghana). In
the Dhar Tichitt, bulrush millet cultivation appears to have been implemented with the
flood-recessing technique”. Pennisetum sp. is clearly the most important grain crop in
early West African agricultural systems. It is followed by African rice – Oryza
glaberrima -with domesticated forms dated to 300-200 BC in the Inland Niger Delta.
Sorghum is a latecomer, dated to AD 800 at Daima in the Nigerian part of the Chadian
plain. In all circumstances, wild plants were exploited systematically. The horticultural

component of African agricultural systems is best documented in the forest and forest-
savanna mosaic of west and Central Africa. Yams were important crops of the
Kintampo Culture area peoples. Yams phytoliths found in the grove of a stone rasp
from Birimu, suggest that these instruments – terracotta cigars, stone rasps – were used
to peel tubers. Oil plants, like oil palm – Elaeis guineensis -, the shea butter tree –
Butyrospermum paradoxum – and the nuts Canarium schweinfurtii , were an integral
component of the subsistence landscape. Yams – Dioscorea cayenensis, D. rotundata,
and many other wild species – were part of comparable if distinct equatorial African
food-ways that included oil plants, leaves and trees bark, hunted game and later nanny
goats, as well as fish and other aquatic resources. The African horticultural systems
were enriched by the addition of Malayo-Polynesian plants that reached the east
African coast at an unknown time, very likely at the end of the second millennium BC
as indicated by banana phytoliths from southern Cameroon dated to BC 500 . The
adoption of bananas, taros, and other cocoyams triggered a radical modification of the
cultural landscape in the hinterland of Eastern Africa and the Great Lakes region,
boosting population growth. The introduction of American plants, maize, manioc,
cocoa, after the 15th century added another layer in the complex history of African
agriculture.
6. African Rock Art
A number of important and long lasting technological and cultural innovations

overlapped during the last 10,000 years; three of them, rock art, metallurgy, and
urbanization will be dealt with in the remaining part of this entry.
Rock art research is steadily and persistently moving away from the descriptive
approach predominant after the second world. Exploration has been sustained and
numerous new sites have been added to the record. The Sahara and Southern Africa
present the highest density of rock–art sites. In general, such sites are preferentially
found in mountains and plateaus, in rock-shelters and caves. In north tropical Africa,
such concentrations are found in the Atlas, Dhar Tichitt – Walata, Adrar-n-Ifogha, Aîr,
Tassili-n-Ajjer, Tadrart Acacus, Tibesti, Djebel Uweinat, as well as the Nubian
sandstone cliff along Nile river valley. In the southern part of the continent, rock art
stations are found mostly in the Drakensberg and high plateaus of South Africa,
Branberg, and the rocky parts of the Zimbabwe and Zambia. Rock art traditions have
also been documented in Highland Ethiopia, Somalia, Kenya and Tanzania along the
eastern side of the continent as well as Angola, Gabon, Central Africa, and Cameroon in
the west and center-north, and finally, Burkina Faso in West African Sahel.
6.1 The Northern Hemisphere
Rock art traditions are diverse. They include images pecked on rock surfaces –
petroglyphs –, painted in open air, rock-shelters, and caves, and a combination of
both. Engravings, more frequent and widespread, are made by grooving, grounding,
pecking, smoothing, up to virtual sculpting. Paintings on the other hand, seems to be
much more localized, frequently in rock-shelters and caves which have certainly
enhanced their preservation. The chronology debate has mobilized researchers for a
fair amount of time. The Saharan record has been organized into four main stages.

The first and earliest, the “Bubale period”, with images of large bodied wild animals
like elephant, giraffe, hippopotamus, was supposed to represent Late
Pleistocene/Early Holocene foragers. It was followed by the “Bovidian period” with
images of livestock, first cattle and later sheep-goats, indicative of Early to Mid-
Holocene pastoral-nomadic societies. The “Horse period”, the third of the sequence,
features horses in different contexts and situations, with and without carts. And
finally, the “Camel period”, the latest, shows dromedaries in different postures, in war
scenes as well as traveling caravans. This chronological system was adopted with
modifications and adjustments to local situations. An alternative chronology, bringing
all Saharan rock art traditions within the Middle and Late Holocene period was
devised in the mid-1980s. The documented regional traditions were divided into
“schools” based on a range of morphological and pictorial attributes. Recent
approaches attempt to explore the cultural meanings of the images pecked and painted
on the rocks. Elaborate pictorial compositions are found in the Tassili-n-Ajjer in the
central part of the Sahara. A new and robust iconographic approach was developed to
explore the multiple meanings of Central Saharan prehistoric paintings. Detailed
analyses of the paintings from Tikadiouine and Iheren bring forth new understanding
and suggests that they are allegoric representations of pastoralists life cycles. The
Tikadiouine painting encapsulates and recapitulates the key stages in the life of a boy,
from childhood to elder-hood. The Iheren grand composition is an intricate
choreography with more than 500 individual images arranged in a number of
tableaus. The annual transhumance, aggregation, packing, celebrating, defending the
livestock against predators, are all staged in a grand allegory. Saharan rock art
appears to belong to a more or less narrative genre.
Figure 15: The paintings from Tikadiouine, Tassili-n-Ajjer (from A. Holl 1994: 84)

6.2 The Southern Hemisphere
The re-interpretation process is much more radical and advanced in the southern part of
the continent. D. Lewis-Williams and his colleagues and students have initiated a
profound reinterpretation of South African San rock art. This reinterpretation is based
on two distinct but complementary lines of evidence: (1) a neurological one with
emphasis on the universality of entoptic phenomena; and (2) an ethnographic one based
on 19th century compilation of San practices and customs. According to this new
reinterpretation, prehistoric San art features shamans in the multiple sides of their
dealings with misfortune, healing, and the over/under-world. The “interpretative”
approach adopted in this case provides an astounding understanding of the San
worldviews and prominent cultural practices. It is however not clear how far back in the
past one can retrodict this explanation. Southern African rock art traditions date as far
back as 27,000 years ago as indicated by the painted stone slab from Appollo-11 Cave
in Namibia. Did the religious practices remained unchanged during all that stretch of
time? Or does the relevance of the 19th century ethnography stretched too far back in
time? Future research will certainly clarify this issue.
Figure 16: A circular trance dance from Lonyana, KwaZulu Natal in the Drakensberg
(from J.D. Lewis-Williams and D.G. Pierce 2004: 84)

The new trends currently developed in African rock art research emphasize the social
use of icons featured on rock surfaces. The most elaborate compositions of African
prehistoric imagery, in the Sahara as well as Southern Africa, appear to refer to a
world of ideas and symbols; none is a “naturalistic” rendering of foragers’ or
pastoralists’ life scenes. Unfortunately, the strong drive toward the search for deeper
meanings tends to erase imperfections. It is as if all rock art composition, provided it is
prehistoric, is a masterpiece by the simple fact of existing.
7. The Advent and Spread of Metallurgy
The advent and spread of metallurgy indicate the development of advanced pyrotechnic
systems Metallurgy transforms a natural item – the ore – into an exclusively human-
made product, the metal; with the latter used to manufacture a range of artifacts. There
are nonetheless profound differences between kinds of metallurgies; copper and iron for
example have overlapping but different requirements. Native copper can be transformed
into artifacts without any elaborate technology. The piece of ore can be heated in an
ordinary fire, hammered with a simple hammer-stone on a more or less soft anvils, and
made into significant elements of material culture. Meteoric iron can be handled along
the same lines. Both cases are clearly not what can be referred to as metallurgy. The
linear sequence of metal production is relatively straightforward and involved the
irreversible transformation of natural ores into metal. The operation, called the smelting
process, takes place in a furnace, an especially built feature that confines the heat
released by the burnt fuel in its combustion chamber and directs it to the ore to be
smelted. The smelted ore – the bloom - collects at the bottom of the furnace from where
it is picked up after a certain cooling time. The bloom consists of a varying combination
of metal and slag. The latter is removed through hammering and re-heating in what can
be termed a purification process, geared to optimize the quality and quantity of metal.
The metal bloom is then transferred to the blacksmith workshop where artifacts are
made, and distributed later for further use.
Despite the significant improvement in research protocols and the increased output of
material, the archaeology of African metallurgies is still a raging battleground. A
reassessment of the state of research tend to show that iron metallurgy was very likely
developed independently in Africa. Evidence for early copper metallurgy has been
documented in the Eghazzer basin and the Bir Moghrein (Figure. 17). The exploitation
of copper ore, its smelting in low bowl-furnaces, and the production of artifacts ranging
from jewelry to weaponry have been practiced in West Africa as early as the end of the
third millennium BC (ca. 2200 BC). Iron metallurgy is also present in archaeological
contexts dated to the very beginning of the first millennium BC, in the Air –Termit area
as well as the Southeastern end of the Adamaoua plateau in Central Africa. There is a
near synchronism in the emergence of iron technology in at least four distinct areas: the
Air-Termit (Niger), Nsukka zone (Nigeria), Bouar area (Central Africa), and Great
Lakes region (Eastern Africa) (figure. 17), with Egypt and the Lower Nile valley special
cases more directly connected to the Near eastern developments.

Figure 17: Areas of early metallurgical traditions (from A.F.C. Holl 2000: 11)
From ca. 1000/900 BC to 100 BC, the practice of iron metallurgy spread from the
Central Sahara to the river Senegal Valley in West Africa, along the Nile valley in
the Northeast, in Carthaginian North Africa, in and along the north side of the
equatorial rainforest, as well as Eastern Africa. However, no new development and/or
additional facts seem to alter the entrenched conviction that metallurgy was imported
from Carthaginians and/or Egyptians and spread throughout Africa by still unknown
propagators.
Despite persistent but false assumptions, the earliest metal artifacts were not tools
used to booster the production of foodstuff. Such developments occurred later, when the
technology was “routinized” and access to metal “democratized”. Elements of personal
adornment, weapons, and status markers were, for a very long time, the most common

artifacts made of metal, both copper and iron.
8. Pathways to Complexity
The concept of complexity is particularly slippery when used to investigate emergent

stages. In this context, it refers to the increasing diversification and accelerated pace of
social differentiation indicated in the material record by a broad array of criteria. From
the perspective outlined above, the earliest and pristine African cases of complex social
systems have been documented in northeast Africa, along the Nubian and Egyptian
portions of the Nile valley. These early stages took place between the fourth millennium
BC and the fourth century AD. The other parts of the continent, northern, western,
eastern, and southern Africa, followed different trajectories.
8.1 Egypt
In the middle of the fourth millennium BC, during the Late Nagada II/ Maadian period,
complex and stratified societies took root in both the Upper and Lower Egypt (figure.
18).
Figure 18: Early and later states in African past (from J.M. Pritchard 1979: 46)

For the remaining part of the fourth millennium BC, from Nagada III to the
Protodynastic (3300 – 3100 BC), political centers emerged along the Nile at
Hierakonpolis, Naqada, Maadi, Buto, with exchanges of goods, ideas and people taking
place between the Upper and Lower Egypt. These series of processes culminated in the
unification of the Upper and Lower Egypt at the end of the Pre-dynastic period, around
3050 BC.
The formation of the Unified kingdom of Egypt was kicked off during the Middle
Neolithic period by the settlement of sedentary mixed farming communities along the
Nile River. They settled preferentially in areas of fertile and cultivable land rejuvenated
yearly by the Nile flood. They developed patterned interactions with more mobile
herders exploiting the patchy and scattered resources of the steppes away from the
valley. These settlements grew in size and some attracted more people than others. The
elite buried in Nagada II burials added an international dimension to the socio-political
equation, triggering conflicts and spatial inequality. Dominant centers [Nomos] became
decision locales for what was the equivalent of competing chiefdoms. This peer-polity
competition resulted in the emergence of dominant paramount chiefdoms that took
control of the Upper and Lower Egypt. Accordingly, there may have been two states,
each ruled by a king, which were united by King Narmer during Nagada IIIa. (ca 3050
BC).
8.2 Nubia
Kerma kingdom emerged on the south flank of Egypt, during the third millennium BC
(figure. 18). Kerma itself was a large urban center with massive building in mud-bricks,
a royal palace, and an impressive cemetery. This site was the centre of a rich and
powerful kingdom. The kingdom of Kerma was often plundered and under the constant
threat of Egyptian troops. It took advantage of the Hyksos conquest of Egypt to expand
north of the second cataract and colonize Egyptian territories. It was finally destroyed
by the expanding Egyptian New Kingdom state in the middle of the second millennium
BC. During the first half of the first millennium BC. the Kushite early kings and elite
achieved the control of an extensive empire stretched from Palestine in the north to
Central Sudan. This world power was however totally Egyptian even if the rulers were
from Kushite extraction.
The archaeology and history of the Kingdom of Kush within its own boundary from the
1st to the 6th cataract, and between the mid-seventh century BC and the fourth century
AD when is collapsed, is still poorly documented. The general history of Kush is
divided into two major periods, the Napatan from 700 to 250 BC when all the major
pyramids were built, and the Meroitic, ranging from 250 BC to AD 400. The Kushite
state was organized around the king/queen and the central administration. The elite
included office holders, traders, as well as a special corps of priests. It was effectively
the corridor between Black Africa and the Nile delta and the Levant, an artery for trade,
ideas, and people.
All three Nubian kingdoms from the Middle Nile Valley (figure. 18), Nobatia in the
north, Makouria in the middle, and Alwa in the south, converted to Christianity in the
sixth century. Archaeological research was conducted in each of the kingdom’s capitals,

at Faras (Nobatia), Old Dongola (Makouria), and finally, Soba (Alwa). A basilica was
excavated at Faras during the Campaign to save Nubian monuments. The uncovered
religious monument contained numerous inscriptions and murals of Nubian royalty and
bishops. Old Dongola was excavated by a Polish Expedition. Soba, first probed in the
1950s, was later investigated through a larger and longer multi-year project of the
British Institute in Eastern Africa.
In all three cases, churches, elites residences, and burials provided insights into the
dynamics of these distinct Nubian states, from their emergence in the sixth century AD
to the demise of Christianity and the collapse of Nubian polities in the 1500s’. At Soba
for example, a large palatial structure was uncovered. Three churches were exposed,
with “two among the largest in Nubia”. A tombstone with the mention of an Alwan king
named David was found re-used in another building. The expansion of Christianity was
not the product of random travels of zealous missionaries. Missionary enterprise was a
powerful instrument of diplomacy and strategy in the age of Justinian. Christianity and
Empire were inseparable, conversion being equivalent to alliance with the Empire. From
the rule of emperor Justinian and his successors, the Romans needed the backing of the
Nubian to safeguard the south flank of Roman Egypt from Persian invaders. By the late
tenth century AD, Alwa kingdom became wealthier and more powerful than its northern
neighbors, partly because of the exploitation of rich gold deposits. Within the same time
range, the Moslem presence increased significantly, with a Moslem quarter at Soba, in
parallel to an accelerated rate of conversion to Islam.
8.3 Highlands Ethiopia
During the first millennium BC, South Arabic culture expanded across the Red Sea and
took roots in Ethiopian Highlands, in Eritrea and Tigray (figure. 18). This is indicated
by the sudden appearance of writing, monumental stone architecture, and sculpture. Iron
technology was likely part of this new “cultural package” that triggered the 6th -7th
century BC urbanization in Highland Ethiopia. Urban centers emerged at such places as
Yeha, surrounded by at least 30 other known sites among which the Hawelti-Melazzo
complex. Yeha grew out of an earlier mixed farming village to become the main urban
center of the Daamat kingdom in the 5th - 4th century BC. It was a relatively small town,
7.5 ha in size, with however spectacular stone monuments: the temple and the Grat
Beal Gebri. The former is a massive rectangular building, 18.5 m long, 15 m wide, with
preserved plain walls measuring 11 m in height. The latter consists of a series of square-
sectioned massive monolithic pillars probably part of a cultic complex with some
affinities to the “Moon Temple” at Marib.
The Daamat kingdom collapsed during the later part of the first millennium BC but
very few is known of the causes and consequences of its demise. Smaller polities
emerged; stele were used to mark elite burials, and exchanges seem to have been
predominantly with the Nile valley. Aksum was one of the small polities that
developed in the area during the later part of the first millennium BC. It was settled in
the 1st century AD. It achieved regional primacy a few centuries later, in the 3rd-4th
century AD, controlled great amount of wealth, developed a centralized monarchical
system, adopted Christianity as state religion, and launched a far-reaching
expansionist policy. Large areas of southern Arabia were ruled from Aksum at

intervals between the 3rd and 6th century AD. Its political control extended several
times to regions beyond the modern borders of Ethiopia and Eritrea. Meroe, in the
Sudanese Nile valley, may have been conquered by an Aksumite army under King
Ezana; but the nature and consequences of this episode remain poorly understood.
The town of Aksum, at the foot of two hills, was extended in a deep gorge (figure. 18).
The surrounding land was rich, water abundant, building stone ubiquitous. The town
itself was stretched along one mile west-east with its width measuring no more than
500 m. Massive architectural complexes have been excavated. Some were storage
facilities, elite residences, and religious buildings. Elite and royal burials were carved in
the bedrock and marked by lavishly sculpted stele. It became a trade hub linking the
Red Sea to the Nile Valley and the north Roman world to the rest of the continent,
visited by merchants from Egypt, Syria, Arabia, and even India. The economic growth
and expansion of the Roman Empire was one of the key factors in the accelerated
development of Aksum as a thriving economic metropolis. At the peak of its power, the
core of the “Aksumite civilization” extended over a territory measuring 300 by 160 km,
with access to the Red Sea at Adulis.
Aksum kingdom collapsed during the 8th century AD partly because of the success and
fast expansion of Islam and its corollary political and economic isolation in a
predominantly Muslim world. The overexploitation of local resources, intensive erosion
of deforested land, as well as a short arid spell, may have accelerated the depopulation
and demise of the Aksumite metropolis in the 8th – 9th century AD.
8.4 North Africa and the Sahara
The northern Sahara and north Africa were settled by post-Capsian Ancestral
protoberbers; predominantly mobile herders rearing sheep and goats and burying their
deceased in monumental stone tombs. Some of the deceased were buried with elements
of personal adornment, amazonite, carnelian and agate beads and pendants; copper and
iron artifacts including weapons and jewelry. The differential treatment of the deceased
documented in Saharan and North African megalithic cemeteries indicates the
emergence of a mobile pastoral elite among the Proto-Numidians, Numidians, Proto-
Garamantes, and Garamantes. Such an elite is represented in rock art imagery featuring
horses and camels ridding individuals armed with spears. In the late second
millennium/early first millennium BC, the Phoenician colonization and the ensuing
construction of Carthage opened a new era of urban development in North Africa. Later,
during the major part of the first millennium BC, expanding Greek city-states created
new colonies along the Mediterranean coast of Tripolitania and Cyrenaica (Libya). In
North Africa, Carthaginians and Romans were locked in a heated rivalry that ended with
the destruction of Carthage – Delenda Carthago – The pastoralist and nomad
component of the North African population was divided into two distinct parts. One,
living in the urban centers, provided the much needed labor force and became city-folks.
The other, living free in the open land was kept in the desert beyond the limes. A
complex situation of raiding/counter-raiding developed all along the margins of the
Roman North Africa, strengthening the warrior-ethic of all the proto-Libyan and proto-
Berber pastoralist societies. The Atarantes in Ghat, Atlantes in the Ahaggar, and
Garamantes in Central Fezzan, were such groups referred to in Roman sources. From

Roman-City perspectives, all these groups “were nothing more than thieves and
brigands”. The Garamantes emerged in Central Fezzan at the end of the first millennium
BC. Archaeological evidence found in their capital Germa, Aghram-Nadharif, and other
forts along the kingdom’s southern frontier, show a fully developed caravan network in
place at the beginning of the first century AD and may have been in operation earlier, in
the middle of the first millennium BC.
8.5 West Africa
The sociopolitical landscape witnessed important changes during the first millennium
BC with the development of four-tier settlement system in the Dhar Tichitt – Walata
(Mauritania). The documented settlement hierarchy took shape in three main phases
stretched from ca. 2000 to 4/300 BC. During the Late Settlement phase (c. 700 – 400
BC), a large regional center emerged at Dakhlet el Atrous I, at the gravity center of
the study area. It is comprised of 590 compounds distributed into a large main site
and smaller satellite village and extended over 92.75 ha.. The settlement system
consisted of five districts centers, large villages, ranging in size from 3.50 ha to 32
ha, twelve small villages (20 to 50 compounds) and 72 hamlets (less than 20
compounds) scattered all over the region. The Dhar Tichitt-Walata architectural
evidence attests for the emergence of paramount-chiefship in the western confines of
West Africa during the first millennium BC. A severe climatic crisis, dated to the
mid-first millennium BC, accelerated the collapse of this sociopolitical system and
the exodus from the area in the 4th / 5th century BC.
Further east, in the Inland Niger Delta and the surrounding flood plain, proto-urban
centers emerged at such places as Dia (c. 800 BC) and Jenne-Jeno (c. 300 BC). The
population of these early West African towns consisted of mixed farmers, growing
millet and African rice, and rearing sheep/goats and cattle. Hunting and gathering, and
especially fishing played an important role in their subsistence systems. Craft
specialization is indicated by iron metallurgy, pottery-making, as well as weaving.
Interregional exchange is suggested by imported grinding equipment and other heavy-
duty tools. The long occupation sequences of these sites, respectively 2,500 and 1,700
years, signal a shift in West African subsistence-settlement systems. Humans then
tended to congregate in increasing numbers in the same place. The developmental
sequences of Dia, Jenne-jeno, Daboya, Daima, etc., attest for the widespread nature of
this shift. However, the relative position of the above mentioned towns in their
respective sociopolitical landscape is impossible to assess because of the lack of
comparable high resolution archaeological data from surrounding sites.
Such an integrated regional investigation conducted in the Houlouf study area, shows a
more dynamic picture of change in the distributional patterns of sites through time.
Fourteen mounds tested in a study area measuring 400 square kilometers attest for a
4,000 years settlement sequence, from c. 1900 BC to AD 1800. The sequence is divided
into five main phases each split into sub-phases A and B: (1) - Deguesse phase (1900
BC – AD 0); (2) - Krenak phase A-B (AD 0 – 500); (3) – Mishiskwa phase A-B (AD
500 – 1000); (4) – Ble phase A-B (AD 1000 – 1400); and finally, (5) – Houlouf phase
A-B (AD 1400 – 1800). The settlement hierarchy and regional sites distribution and
other evidence indicate the emergence of two competing polities during the first half of

the 2ndmillennium AD with Houlouf and Ble-Mound A as their respective centers.

Houlouf emerged as the regional center during the later part of the sequence, surrounded
by a thick earthen-rampart and a moat; one of the key characteristics of Chadic
chiefdom centers.
During the first half of the first millennium AD, mixed-farming communities dotted the
West African landscape from Lake Chad to the Atlantic coast (figure. 18). Major rivers
drainages like the Senegal, Volta, Niger, Logone, Shari, etc were more or less densely
settled. The Sahel, with its shifting boundaries between the Sahara and the savannah,
was settled by scattered populations of more or less pastoral-nomadic groups as can be
seen in the Eghazzer basin (Niger), the lower Azawagh and Tilemsi valleys (Mali).
Some of the settlements were selected for bulkier and longer lasting occupations. The
urbanization of some portions of the West African landscape has it roots in the second
half of the first millennium BC/first half of the first millennium AD. It was shaped by
the intensification of interregional exchanges in raw materials, manufactured goods,
foodstuff, and others. Copper and iron may have been traded as metal blooms and/or
fully manufactured products, weapons, items of personal adornments, tools, etc.
Foodstuff may have included grain, smoked meat and fish. Products from the savannah
and the forest such as Kola nuts, wild animal skins, pepper, dried leaves and tree barks,
may have been traded for cloth, or other craft items from the Sudano-Sahelian towns.
The role of the Niger River as main thoroughfare for peoples, goods, and ideas all
across West Africa is therefore not surprising. Niani, Jenne-jeno, Dia, and Gao were
connected to the river traffic. The development of the international long-distance trade
amplified the economic and political role of some of these places which became thriving
markets and trade emporia. Gao and Jenne-jeno, for example, were located at the
intersection of two transportation systems; one linked to the trans-saharan trade with
caravans. The other, under the control of “river-people” with dug-outs flotilla. People
and goods moving one way or the other, were transshipped from one transportation
system to the other. Other towns, like Kumby Saley, Awdaghost, Azelik, Marandet,
developed in unlikely and surprising places. Azelik and Marandet were located in areas
with rich copper deposits. Both settlements were important centers for the processing of
copper ore and the production of copper artifacts. The location of Kumby-Saleh, far
from major rivers in an area without salient attracting characteristics has always puzzled
researchers. Awdaghost, the trade entrepot of the kingdom, was built in an area with
easily accessible water resources, from seasonal river and a high water table.
8.6 City-States of the Rainforest
West African rainforest is stretched from Cameroon to Guinea. Archaeological research

has focused on two areas principally, Ghana on the one hand and Nigeria on the other.
In the former, trends toward urbanization are documented at Begho and Bono Manso.
Both sites are comprised of a number of distinct mounds. Begho was a large market
center with four distinct quarters settled in the 11th – 12th century AD. Its wealth and
prosperity peaked in the early 17th century. Its inhabitants traded with the Savanna, the
Sahel, and the Sahara to the north. They were mixed farmers practicing grain and yam
cultivation with livestock husbandry of cattle, sheep and goats. Domestic pigs were also
represented. Hunting as well as the gathering of wild and aquatic resources were
common practices.

Bono Manso occupation started in the 13th century AD and lasted to the mid-18century.
The settlement was comprised of separate and dispersed habitation mounds that later
coalesced into an extensive site. The sociopolitical systems of which these settlements
were part of are still poorly investigated. Both localities were likely small “city-states”
connected to the regional and interregional long distance trade, controlling a
surrounding hinterland of unknown size.
The developments in southern Nigeria were more spectacular, revealing a stunning art
and craft tradition, and an original urban tradition. The site of Igbo –Ukwu, where an
extraordinarily rich burial that may have been that of a “priest-king” was excavated, is
located east of the Niger River, in southeast Nigeria. The deceased was buried in a
rectilinear burial chamber. The grave goods included no less than 685 copper and
bronze objects, 165,000 carnelian and glass beads, three elephant tusks, a crown, and
many other items.
Ife, Benin, Old Oyo, Koso, Ipapo Ile, are located in the Yorubaland, west and southwest
Nigeria. Most of the ancient Yoruba towns are located in the Savanna-forest mosaic.
From a cultural perspective, Ife, located more in the forest than the savanna, and
considered as “the place of the creation of the world”, possess a remarkable range of art
masterpieces. Ife occupation seems to start at the end of the first millennium AD. Floor
pavements made of recycled shards and stones were a frequent attribute of domestic
architecture from the 12th to the 15th century AD. The town, delimited by an intricate
series of walls, includes the Oni Palace at its centre. The precise chronology of the walls
is still poorly known. However, the inner wall, surrounding the Old town, may have
been the first to be built, with the outer one built later following the outward extension
of the city.
Benin City, deep in the rainforest, was settled during the 13th century. Floor pavements
were common in the 14th century. A “sacrificial well” containing the skeletal remains
of at least forty-one young women was found in the Old palace. They were dressed up
with fine clothing, adorned with jewelry, and were dropped in the 12.5 m deep well
during ceremonies that may have been repetitive during the 13th and 14th century.
Offerings including elephant tusks were deposited in the top fill of the pit, highlighting
the importance of the rituals performed in this part of the old Royal Palace. Benin
earthworks are particularly impressive. They cover ca. 6,500 square kilometers, for a
total length of 16,000 kilometers approximately. Their function and use is still debated
among experts. They are the material testimony of patterns of land allocation that
probably involved corporate social units, extended families, lineages, and/or clans.
Their size, shape, and boundaries seem to have been re-adjusted to cope with changing
socio-political situations.
The ancient Yoruba city-states were political centers controlling a surrounding

hinterland. The Oba or king’s palace built at the center of the town also included
ancestral shrines, ceremonial spaces, and ritual features lavishly furnished with
impressive work of art. The city-state officials resided in the inner-sanctum of the town,
at close distance from the rulers’ residence. These members of the city-state
government, the Oba family and relatives, the artists and crafts specialists sponsored by
the palace, were probably granted land and owned farms in the countryside, plots that

may have been delimited by earthworks. Dependents and other family members were
very likely living in these plots, growing yams, palm oil, and other products needed to
feed to towns’ inhabitants. The system of earthworks incidentally strengthened the static
defense systems of each the city-states. The onset of the Atlantic Slave trade
accentuated the competition among Yoruba city-states, and European colonization gave
a deadly blow with the sack of Benin by the British in the 19th century.
8.7 East African Coast and its Hinterland
The archaeology and history of the last three millennia of this part of the Africa are the
product of genetic and cultural recombination of distinct populations of the Indian
Ocean. The Swahili civilization along the coast, the colonization and successful
settlement of the Comoros and Madagascar, resulted from the trans-regional
redistribution of population from the Malayo-Indonesian world, African hinterland, and
the southern portion of the Near East.
The Swahili civilization emerged from an immense interaction sphere 3000 kilometers
long and 20 to 250 km wide, from Mogadiscio (Somalia) to Cape Delgado
(Mozambique). Pastoralists from Cushitic ancestry expanded southward from ca. 4500
BP. Groups of Bantu speakers practicing agriculture, gathering, hunting, and fishing,
trickled in the Great Lakes regions and later along the East African coast from the first
millennium BC onwards. Despite differences in the interpretation of the archaeological
record, the African origins of the Swahili civilization is less and less contested. The
development of Swahili city-states is spelled out in three major phases, with a fourth
one (1500 – 1950) corresponding to foreign domination and colonization. During
Period I (100 BC – AD 300), settlements are small, founded and inhabited by iron-using
farming communities, and characterized by a pottery tradition that can be traced back to
500 BC in the Great Lakes region. This Kwale ware is found at Misasa (Tanzania),
Kwale (Kenya), and Ras Hafun (Somalia).
Period II (300 – 1000) comprised of two phases, the Azanian (300 – 600) and the
Zanjian (600 – 1000), witnessed the kick off of the urbanization process and the onset of
international long distance trade. The maritime trade became particularly active during
the Zanjian phase, with carnelian beads, Partho-Sassanian tin glazed, Chinese, Indian
purple wares, as well as Egyptian glass. The out-going component of the international
exchange was likely made of human cargo, ivory, gold, iron, animal skins, etc.
Settlements sprung at places like Chibuene, Masuguru, Kaole, Kilwa, Manda, Pate, and
Shanga.
The Swahili city-states emerged during Period III (1000 – 1500). Each major city
controlled the coast and part of the hinterland, with docking areas for seafaring
vessels.
The coast was dotted with a string of towns, villages, and fishing stations. Particularly
important urban centers emerged at Kilwa, Manda, and Mogadiscio. Islam, which
started to expand during the Zanjian phase, became a well established religion. The
intensification of long distance trade, the emergence of an elite residing in mansion-like
coral-stone houses, with local copper mint, as well as the development of social ranking,

all contributed to the thriving economies of the East African Coast. Slavery was an
institution geared to satisfy foreign and local demands. Congregation mosques, massive
elite stone mansions, but also wattle and daub houses became distinct but
complementary facets of East African urban landscapes. From 1500 onwards, the
irruption of the Portuguese, the Omani Arab conquest, and finally the European
conquest, sealed the demise of the Swahili civilization.
The Great Lakes region and the Savanna-land south of the equatorial rainforest
witnessed the development of complex chiefdoms and kingdoms during the second
millennium AD. In the Great Lakes region, the development at Ntusi, Munsa, Bigo and
others earthworks sites are remarkable. Large-scale cattle husbandry is documented at
Ntusi. Kibiro witnessed an impressive intensification in the production of salt from the
local brackish springs. And an agricultural colonization took place in western Uganda.
However, tracing the precise evolutionary trajectory of any of the Great Lakes past
polities is still hampered by the lack of sustained long-term archaeological research and
terminological uncertainties. The Savanna, from the Atlantic coast to the Great Lakes,
witnessed the emergence of a succession of states that may have started in the first or
the very beginning of the 2nd millennium AD. It is the context for the development of
Kongo, Loango, Tio, Mbundu, Kuba, Lozi and Luba kingdoms. But archaeological
research is virtually non-existent in this whole area, making the reconstruction of the
past political evolution a “learned guess”.
8.8 Madagascar and the Comoros
The earliest traces of human presence on Madagascar consist Hippopotamus bones with
cut marks made with iron tools, dated to AD 80 – AD 380 at Lamboharana and
Ambolisatra in the southwest. However, evidence for human settlement is particularly
shallow and elusive before the later part of the first millennium AD. Traces of these
early sites found along the eastern, southwestern, and north coast, point to camps
occupied by small visiting crews. Longer term and likely permanent settlements
developed later in the first millennium AD from the 8th century onwards. It is the case at
Nosy Mangabe in the northeast, and the Sandrakatsy phase site of the Mananara valley.
The Comoros islands were settled during this time period, with Dembeni- phase villages
ranging in size from less than 0.5 to 5 ha. In Northern Madagascar, Nosy Mangabe and
Sandrakatsy yielded evidence for iron metallurgy, chlorite schist wares, as well as
silver, gold, glass, and carnelian beads. Dembeni phase groups were organized into
small iron-using farming and fishing communities; growing rice, millet, coconuts,
beans, and possibly citrus fruit.
The beginning of the second millennium AD is a subsistence and technology stasis,

without noticeable trend toward intensification. Major transformations took place
nonetheless, within the context of social ideologies and organizational patterns of
human communities. Larger and more populous centers emerged. In the Comorian
archipelago, Dembeni grew to reach 14 ha and was surrounded by a wall. Villages
measuring 1 to 3 ha as well as numerous smaller hamlets dotted the archipelago
landscape. Comparable change occurred in Madagascar but at a larger scale. Most of the
recorded sites are found in the north and the southeast. The rest of the island appears to
have been very loosely inhabited with isolated village sites found at Fiekena in the

Central Highlands and Rezoky in the southwest. Mahikala in the bay of Ampasindava
on the northwest coast, had a walled area more than 60 ha in size. It was a harbor and
point of entry of the Indian Ocean trade items. The town was likely part of a larger
settlement system that included number of smaller contemporary sites, less than one
hectare in size. The cave site of Andavakoera, and villages and hamlets like Irodo,
Lanivato, Bemanevika, and Sandrakatsy, are located along the northeast coast. In the
remaining part of the island, settlements include small villages or larger sites like
Andranosoa with multiple embankments on some 30 ha with concentrations of mud
houses remains. There was a significant intensification of the Indian Ocean exchange
systems along with the widespread adoption of Islam in coastal areas.
From the 14th to the 16th century, when a Portuguese fleet reached the northwest coast
of Madagascar, the island appears to have gone through comparable transformations if
judged from the settlement pattern perspective. Sites clusters with 5 to 10 smaller
settlements articulated around one larger center of approximately 5 ha emerged. Such
patterns were recorded in the Bay of Antongil, northern Androy, Imerina, and southern
Anosy. Islamic port towns emerged along the northwest and northeast coasts.
Paramount leaders resided at the centers of settlements clusters. From then on, complex
chiefdoms appear to emerged on the island, and fuelled the fierce rivalry and
competition that took place later in the 18th century.
8.9 Southern Africa Hinterland
Food producing communities practicing grain agriculture and livestock husbandry

spread all over the southern part of the continent during the later part of the first
millennium BC. Depending on areas, the newcomers interacted with Khoisan-speaking
foragers, resulting in a complex patterns of cross influences. Livestock husbandry
expanded in most of Namibia. Agriculture and livestock spread in Mozambique,
Zimbabwe, and North/Northeast South Africa, while Khoisan foragers moved away in
parts of the Kalahari. The development to be discussed here focuses on the emergence
of the Zambezian chiefdoms and states.
The Toutswe chiefdom in eastern Botswanna and Mapungubwe in the Shashe-Limpopo

valley are the earliest ranked societies of Southern Africa that emerged at the end of the
first millennium AD. The later developed into a full-fledged state for approximately 200
years, from 1100 to 1280 AD. Excavation at Bambandyanalo, Mapungubwe, and
Schroda indicate a connection to the international trade network of the Indian Ocean. In
exchange, they provided products from the gold mines, as well as animal skins and
hides, and elephants tusks. Cattle was a central element in the Mapungubwe state
political economy. It served as the repository of wealth with success measured by the
herds’ size any potential leader could marshal. Settlements were made of enclosures
organized around cattle byres. The central site that may have been the capital of the
emerging polity shifted from Schroda, to Bambandyanalo (K2), and Mapungubwe.
Bambandyanalo, a mound site with huge rubbish middens, was an important village,
part of a larger settlement system comprised of Pont Drift, Mmamgwa, Schroda and
Mapungubwe. Its inhabitants were predominantly cattle herders and elephants’ hunters
living in sun-dried clay houses with beaten earth floors. It was abandoned during the

second half of the eleventh century, for a new central site at Mapungubwe Hill. The
settlement complex, including a hill site for the rulers and elite members overlooking a
lower site of sun-dried clay-and wattle houses for the commoners, emerged at the end
of the 11th century. Mapungubwe state (ca. 1100 – 1280) participated actively in the
international long-distance trade and controlled a relatively extensive territory along the
Shashe-Limpopo valleys. The settlement hierarchy, with Mapungubwe as the largest
and central site, was divided into five distinct levels, the capital and dominant political
center, hilltop towns of elite members, large villages, hamlets, farmsteads. The state
economy was anchored on cattle husbandry, long distance exchange with the Indian
Ocean coastal towns, and the flow of tribute extracted from dependent polities. The
causes of the decline of Mapungubwe state are still poorly known. Its population started
to decline in the second half of the 12th century. A combination a factors, including
droughts, soils depletion, shifts in international trade routes, and likely over-taxation,
put an end to the existence of the first Zambezi state at the end of the 13th century.
Regional primacy shifted hundred of kilometers north-northeast to the Zimbabwe
plateau. First in the Mateke hills where small villages and homesteads were already
settled at the beginning of the second millennium, between 1000 and 1200.
Great Zimbabwe state emerged at the end of the 13th century on the plateau of the same
name, with settlements distributed between 500 and 1300 m above sea level. Its
territory stretched along ca. 300 kilometers, from the Runde river in the south to the
Save river in the east and northeast. The state economic basis combined intensive cattle
husbandry with grain agriculture. The connection to the international Indian Ocean
trade was however stronger with an intensified export sector involving copper, gold,
ivory, gems, leather, animal skins, etc. The town of Great Zimbabwe, at the center of
the southern half of the plateau, was the capital of the state. Almost all settlements
included prominent stone buildings; they nonetheless differ in size with homestead
clustered around small villages; small villages generally attached to large villages; large
villages part of regional district dominated by small towns; with the state capital at the
top of the settlement hierarchy. The spectacular architecture of the Great Enclosure and
the Hill complex at Great Zimbabwe is a testimony of the prestige and power harnessed
and displayed by the state rulers and elite. At the peak of Great Zimbabwe State wealth
and power, from AD 1300 to AD 1450, the Hill complex included the Western, Cleft
rock, Southern, Recess, and Eastern enclosures. The latter, in which six carved
soapstone birds were found, was devoted to state rituals and ceremonies. The heavy
investment in the construction and the monumentality of the whole southern side of the
Hill Complex that overlooks the lower City of the Valley Complex was designed as an
unmistakable display of the grandeur of the state rulers
The demise of Great Zimbabwe state was gradual but accelerated in the mid-15th
century when emerging competing states of Torwa, Mutapa, and others captured the
bulk of the international long distance trade. The post fifteenth century period saw the
arrival of a new player in the international scene, the Portuguese. They negotiated trade
agreements with many African polities, along the Indian Ocean coast and the Southern
Africa hinterland. However, most of the time competition was fierce involving
successions of attacks, counter-attacks, and retaliation.

Epilogue
The Scramble for Africa started with the Portuguese circum-navigation. It took a
dramatic turn with the colonial conquest, and ended with several colonial regimes that
set the history of Africa along divergent and unpredictable paths. These situations
brought with them new constraints and unsuspected opportunities. The archaeology of
Africa provides a long and unequalled glimpse at the emergence, development and
expansion of human societies. The perspective gained on the development of African
cultural mosaics is unique, if partial and regionally differentiated. Significant portions of
the continent, like north central Africa, are barely explored by archaeologists. Others,
like the Nile valley in Sudan, Nubia, and Egypt, attract dozens of archaeological
projects each year. The achievement of approximately one century of archaeological
research is impressive indeed. Much work remains to be done.
Bibliography
Bisson, M.S., S.Terry-Childs, P. De Barros, and A,F,C, Holl 2000 Ancient African Metallurgy: The
Sociocultural Context. Walnut Creek; AltaMira Press.
Blench, R.M. and K. C. MacDonald (eds.) 2000 The Origins and Development of African Livestock:
Archaeology, Genetics, Linguistics, and Ethnography. London; University College London Press.
Clark, J.D. (ed.) 1982 The Cambridge History of Africa, Volume I: From the Earliest Tomes to c. 500
B.C. Cambridge; Cambridge University Press.
Connah, G. 2001 African Civilizations: An Archaeological Perspective. Cambridge; Cambridge

University Press.
Holl, A.F.C. 2004 Saharan Rock Art: Archaeology of Tassilian Pastoralist Iconography. Walnut Creek;
AltaMira Press.
Insoll. T. 2003 The Archaeology of Islam in Sub-Saharan Africa. Cambridge; Cambridge

University Press.
Isaac, B. (ed.) 1989 The Archaeology of Human Origins: Papers by Glynn Isaac.
Cambridge; Cambridge University Press.
Kusimba, C.M. 1999 The Rise and fall of Swahili States. Walnut Creek; AltaMira Press.
Kusimba, S.B. 2003 African Foragers: environment, Technology, Interactions. Walnut creek; AltaMira
Press.
Lewis-Williams, J.D. and D.G. Pierce 2004 San Spirituality: Roots, Expression, and Social
Consequences. Walnut Creek, AltaMira Press.
Mitchell, P. 2002 The Archaeology of Southern Africa. Cambridge; Cambridge University Press.
Phillipson, D. 1998 Ancient Ethiopia: Aksum: Its Antecedents and Successors. London; British Museum
Press.
rd
Phillipson, D. 2005 African Archaeology, 3 edition. Cambridge; Cambridge University Press.
Pikirayi, I. 2001 The Zimbabwe culture: Origins and Decline of Southern Zimbabwe
States. Walnut Creek; AltaMira Press.

Shaw, T., P. Sinclair, B. Andah, and A. Okpoko (eds.) 1993 The Archaeology of Africa:
Food, Metals and Towns. London; Routledge.
Smith, A.B. 2005 African herders: Emergence of Pastoral traditions. Walnut Creek;
AltaMira Press.
Stahl, A.B. (ed.) 2005 African Archaeology: A Critical Introduction. Oxford; Blackwell.
Van Der Ven, M. (ed.) 1999 The exploitation of Plant resources in Ancient Africa. New York; Kluwer
academic/ Plenum Publishers.
Vogel, J.O. (ed.) 1997 Encyclopedia of Precolonial Africa: Archaeology, History, Languages, Cultures,
and Environments. Walnut Creek; AltaMira Press.
Biographical Sketch
Augustin F.C. Holl is Professor in the Department of Anthropology and the Center for Afroamerican and
African studies, and Curator of West African Archaeology in the Museum of Anthropology. His research
revolves around issues of social evolution, the advent of Food-producing economies, the emergence, and
growth of complex social systems in West Africa and the Levant. He has conducted fieldwork in the
Saharan desert in Mauritania, the Negev desert in Israel, the Chadian plain in Northern Cameroon, and is
currently director the Mohoun Bend Archaeological Project (MOBAP) in northwestern Burkina Faso.
Other facets of Holl's research touch on Mortuary archaeology through the New York African Burial
Ground research project, and the anthropological archaeology of Central Sahara Rock Art. He is editor in
chief of the West African Journal of Archaeology.
View publication stats

The Archaeology of Africa: January 2005

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Archaeology of Africa: January 2005

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

The Archaeology of Africa

Chapter · January 2005

Augustin F.-C. Holl

The Neolithization Process: À Global Comparative Perspective View project

The Initial History of Africa View project

The user has requested enhancement of the downloaded file.

Copyright © 2010 EOLSS Publishers/ UNESCO

Information on this title: www.eolss.net/eBooks

ISBN- 978-1-84826-003-0 (e-Book Adobe Reader)

British Library Cataloguing-in-Publication Data

Library of Congress Cataloging-in-Publication Data

THE ARCHAEOLOGY OF AFRICA

Keywords: Africa, Prehistory, Hominids, Modern Humans, Cultural beginnings,

©Encyclopedia of Life Support Systems (EOLSS) 21

The archaeology of Africa is synonymous with the archaeology of humankind.

©Encyclopedia of Life Support Systems (EOLSS) 22

©Encyclopedia of Life Support Systems (EOLSS) 23

1.1 Scatters of Stones and Bones

©Encyclopedia of Life Support Systems (EOLSS) 24

1.2 The Hunting Hypothesis

1.3 The Gathering Hypothesis

©Encyclopedia of Life Support Systems (EOLSS) 25

1.4 The Sharing Hypothesis

“Man-the-Hunter” and “Woman-the-Gatherer” hypotheses are antithetic mirror-images.

1.5 The Scavenging hypothesis

As indicated by the archaeological record from Koobi-Fora and Olduvay, protohumans

©Encyclopedia of Life Support Systems (EOLSS) 26

2. Acheulean Foragers and the dispersal of Homo sp.

©Encyclopedia of Life Support Systems (EOLSS) 27

©Encyclopedia of Life Support Systems (EOLSS) 28

©Encyclopedia of Life Support Systems (EOLSS) 29

©Encyclopedia of Life Support Systems (EOLSS) 30

As is the case for proto-human subsistence systems discussed above, patterns of

3. Middle Stone Age Hunter-gatherers and the Emergence of Modern Humans

©Encyclopedia of Life Support Systems (EOLSS) 31

3.1 Southern and Eastern Africa

©Encyclopedia of Life Support Systems (EOLSS) 32

©Encyclopedia of Life Support Systems (EOLSS) 33

©Encyclopedia of Life Support Systems (EOLSS) 34

Figure 8: Sample of Middle Stone Age 2b artifacts

©Encyclopedia of Life Support Systems (EOLSS) 35

Figure 9: Sample of Howieson’s Poort artifacts

©Encyclopedia of Life Support Systems (EOLSS) 36

3.2 Western and Northern Africa

There is a genuine dearth of information on MSA sites in West Africa. Exploratory

©Encyclopedia of Life Support Systems (EOLSS) 37

Middle Palaeolithic. A Mousterian and Mousterian-like techno-complex is widespread,

The Aterian Industry is an off-shot of the Mousterian Tradition. It is confined to North

3.3 The Saga of Modern Humans

©Encyclopedia of Life Support Systems (EOLSS) 38

4. Late Pleistocene Specialized Foragers.

©Encyclopedia of Life Support Systems (EOLSS) 39

©Encyclopedia of Life Support Systems (EOLSS) 40

The Iberomaurusians were mobile foragers exploiting the extensive Mediterranean

©Encyclopedia of Life Support Systems (EOLSS) 41

©Encyclopedia of Life Support Systems (EOLSS) 42

©Encyclopedia of Life Support Systems (EOLSS) 43

©Encyclopedia of Life Support Systems (EOLSS) 44

5. From Foragers to Food-Producers

5.1 Early Foragers-Herders

5.2 The Rise of Pastoral-Nomadism

The second trajectory encompasses a number of “Cardial Neolithic” sites found in