}V1ESI_J RETURNING MATERIALS:
P1ace in book drop to
LJBRARJES remove this checkout from
.—:-—. your record. FINES will
7 » be charged if book is
returned after the date
stamped below.

 

 

 

 

 

NORTHWEST BENIN SITES AND
SOCIALLY DETERMINED ARTIFACT DISTRIBUTION

By

Idemudian Omokhodion

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Anthropology
1987

5" 7' 3773’

Copyright by
Idemudian Omokhodion
1987
All Rights Reserved

ABSTRACT
NORTHWEST BENIN SITES AND
SOCIALLTY DETERMINED ARTIEACT DISTRIBUTION
By
Idemudian Omokhodion

The mode of evolution of the ancient kingdom of Benin
southwestern Nigeria, is of research interest in the way oral
historical data has been related, for example, to
archaeological evidence. Carbon 14 dates and related
evidence from the city center indicate that the kingdom was
established at about 700 years ago. Specific determinants
such as the territoriality associated with the unique pattern
of earthworks have been offered as explanation for this
origin.

New evidence provided by this research is in the form of
artifact (especially pottery) distribution on the northwest
of the Benin City center. The genera1.variability of pottery
design attributes has been found to be explained by the
mobility of women in post-marriage residence and the
characteristic non-sororal polygyny. Values placed on
residence and marriage to several wives are vividly.
illustrated in Benin architecture, and elaboration of these
values in mythology and office regalia could account for the
distinctive kingship and caste systems. This deduction is

not inconsistent with the evidence of Benin-Ife relationship
since Benin shares several cultural features with the Yoruba

states.

Dedicated to
Dr. Aku Omokhodion

ACKNOWLEDGEMENTS

I wish to say this, that it would be hard for me to
capture in words the gratitude I feel towards my Department
(Anthropology) at Michigan State University, under the
leadership of Dr. B. Gallin, for the ready assistance I got
throughout my graduate program. I wish I had something to
give back in return.

Secondly, I could not have gone half way if I did not have
my academic advisers, Dr. Cleland, Dr. Dwyer, Dr. Lovis and
Dr. Robbins, who understood my problems in a personal way.
They provided me with very helpful hints and comments before
and during the preparation of my dissertation. In
particular, I am personally indebted to Dr. Robbins who used
his time and money, outside of the official requirements, to
help me through the program. Dr. Robbins also readily read
through my drafts and I am inspired by his recognition and
commitment to scholarship.

I must express my gratitude to the Department secretaries
who have done a lot to make my stay in the Department a lot
easier, and in particular, Barbara Knuth. I always felt
reassured by her smiles.

vi

Without a research grant from the L.S.B. Leakey Foundation
it would have been very difficult to carry out my fieldwork.
I am grateful for this and also to the University of Benin
for the grant I obtained from them at the initial stages of
planning.

Thanks also go to Dr. Fred Anozie who used his money to
travel all Ithe way from Nsukka to act as my local field
supervisor in the Benin project. At various points of my
fieldwork in Benin, several assistants were very helpful and
I thank them all.

And lastly I must thank these my editors for their
assistance. They are Dr. Kofi Egbo, Ms Carol Bacak, and Mrs
Virginia Bacak.

vii

TABLE OF CONTENTS

List of Tables .................................... x
List of Figures .................................. xii
CHAPTER
I INTRODUCTION
1.1 Background to Historical Reconstruction
in Nigeria ..................................... 1
1.2 An Anthropological Perspective ................. 5
1.3 Purpose of the Study .......................... 10
II BENIN RESEARCH
2.1 Ecology ....................................... 13
2.2 Benin Society ................................. 32
2.3 Benin Art ..................................... 40
2.4 Benin in West African Architecture ............ 44
2.5 Historical Sites .............................. 50
2.6 Benin Archaeology ............................. 55
2.7 A Summary of West African Ceramics,
Historical Perspective and the Place
of Benin Materials ............................ 60
2.8 Benin in West African Trade and
and State Formation ........................... 66
2.9 Summary ....................................... 73
III METHODOLOGY
3.1 Assumptions ................................... 75
3.2 Fieldwork on the Northwest 1 .................. 79
3.3 Eko Abetu Site ................................ 94
3.4 Uwan Wall Site ................................ 98
3.5 Evboneka Farm Site ............................ 99
3.6 Summary ...................................... 100
IV FIELDWORK ON THE NORTHWEST 2

4.1 Uwan Living Site ............................. 104

viii

VI

VII

VIII

CLASSIFICATION AND ANALYSIS 1

5.1 Northwest Benin Pottery ...................... 115
5.2 Quantitative Analysis ........................ 117
5.3 Qualitative Analysis ......................... 156
5.4 Dating ....................................... 162
5.5 Summary ...................................... 163

CLASSIFICATION AND ANALYSIS 2

6.1 Qualitative Analysis ......................... 164
6.2 Quantitative Analysis ........................ 195
INTERPRETATION

7.1 Bias ......................................... 199
7.2 Interpretive Models .......................... 203
7.3 Historical Implications ...................... 215
7.4 Discussion ................................... 217
SUMMARY AND CONCLUSIONS .......................... 221
Footnotes ........................................ 226
List of Reference ................................ 230
Appendix A -- Plates ............................. 243

ix

E
e

coooqmmpweop

10

11
12
_ 13
14
15

16
17
18
19

LIST OF TABLES

PAGE
Trench B1, Tool Inventory, Eko Abetu Site ........ 101
Trench X2, Tool Inventory, Eko Abetu ............. 102
Trench Y2, Tool Inventory, Eko Abetu ............. 103
Trench 410, Tool Inventory, Uwan Site ............ 113
Trench 420, Tool Inventory, Uwan Site ............ 114
Rim Nominal Scores, Trench B1, Eko Abetu ......... 121
Rim Nominal Scores, Trench X2 Eko Abetu ......... 122
Rim Nominal Scores, Trench Y2, Eko Abetu ......... 123
Jaccard Coefficient for Rim Variables,
Eko Abetu ........................................ 124
T-Test Sample Means for Lip Thickness,
Eko Abetu ........................................ 125
T-Test Scores for Lip Thickness, Eko Abetu ....... 126
Body Decorated Scores, Trench B1, Eko Abetu ...... 127
Body Decorated Score, Trench X2, Eko Abetu ....... 128
Body Decorated Scores, Trench Y2, Eko Abetu ...... 129
Chi-Square Determinations for Body
Decorated Scores, Eko Abetu .................. 130-135
Diagnostic Vessel Count, Eko Abetu ............... 136
Diagnostic Vessel Count, Uwan Site ............... 169
Rim Nominal Scores, Trench 410, Uwan Site ........ 170
Rim Nominal Scores, Trench 420, Uwan Site ........ 171

X

20

21
22
23

24
25
26
27
28

Jaccard Coefficient for Rim Variables,

Uwan Site ........................................ 172
Body Decorated Scores, Trench 410, Uwan Site ..... 173
Body Decorated Scores, Trench 420, Uwan Site ..... 174
Chi-Square Deterinations for Body

Decorated Scores, Uwan Site .................. 175-178

Weight of Sherd Analysis,
Weight of Sherd Analysis,
Weight of Sherd Analysis,
Weight of Sherd Analysis,
Weight of Sherd Analysis,

xi

Trench B1, Eko Abetu...179

Trench X2 .............. 180
Trench Y2 .............. 181
Trench 410, Uwan ....... 182
Trench 420, Uwan ...... .183

LIST OF FIGURES

FIGURE PAGE
1 Map of West Africa ................................. 9
2 Geological Map of Benin ........................... 17
3 Map of Nigeria - Geography and Archaeological

10
11
' 12
13
14
15
16
17
18
19

Sites ............................................. 18
Rainfall Map of Nigeria ........................... 19
Edo-Speaking People ............................... 20

1st and 2nd Benin Walls and Historical/

Archaeological sites .............................. 22
Benin Area ........................................ 24
Escorting an Important Person (Benin Art ) ........ 25
Motifs in Benin Art ............................... 26
West African House Types .......................... 27
Plan of Ogiamwen’s Palace, Benin City ............. 29
Benin House Types ................................. 30
Ogiamwen’s Palace (photo) ......................... 31
Uwan Wall and Northwest Sites ..................... 80
Eko Abetu Site .................................... 81
Grid Plan for Eko Abetu Site ...................... 82
Bl Trench profile, west face, Eko Abetu ........... 83
B1 Trench profile, north face, Eko Abetu .......... 84
X2 Trench profile, west face, Eko Abetu ........... 85

xii

20
21
22
23
24
25
26
27
26
29
30
31

32

33
34
. 35
36
37
38
39
4o
41
42

X2 Trench profile, south face, Eko Abetu .......... 86

Gun flints, level 4, X2, Eko Abetu ................ 87
Y2 Trench profile, east face, Eko Abetu ........... 88
Y2 Trench profile, south face, Eko Abetu .......... 89
Uwan Wall site .................................... 90
Uwan Wall Trench, east profile .................... 91
Uwan Living Site ................................. 108
Trench 410 profile, west face, Uwan site ......... 109
Trench 410 profile, south face, Uwan site ........ 110
Trench 420 profile, east face, Uwan site ......... 111
Trench 420 profile, south face, Uwan site ........ 112
Dendrogram cluster for Jaccard Coefficient

on Rim nominal variables, Eko Abetu .............. 137

Schematic Representation of Decorative

Techniques ....................................... 139
Correlated Stratigraphy, Eko Abetu ............... 140
Level 111 rim cross sections, Eko Abetu .......... 141
Level 112 rim cross sections, Eko Abetu .......... 142
Level 113 rim cross sections, Eko Abetu .......... 143
Level 121 rim cross sections, Eko Abetu .......... 144
Level 122 rim cross sections, Eko Abetu .......... 145
Level 123 rim cross sections, Eko Abetu .......... 146
Level 133 rim cross sections, Eko Abetu .......... 147
Level 132 rim cross sections, Eko Abetu .......... 148
Pottery types, Eko Abetu ......................... 150

xiii

43
44
45
46
47
48
49
50
51
52

Level 411 rim cross sections, Uwan site ......... 184
Level 412 rim cross sections, Uwan site ......... 185
Level 413 rim cross sections, Uwan site .......... 186
Level 414 rim cross sections, Uwan site .......... 187
Level 421 rim cross sections, Uwan site .......... 188
Level 422 rim cross sections, Uwan site .......... 189
Level 423 rim cross sections, Uwan site .......... 191
Level 424 rim cross sections, Uwan site .......... 192
New Pottery Types, Uwan site ..................... 193

Dendrogram cluster for Jaccard Coefficient,

Uwan site ........................................ 194

xiv

43
44
45
46
47
48
49
50
51
52

Level 411 rim cross sections, Uwan site ......... 184
Level 412 rim cross sections, Uwan site ......... 185
Level 413 rim cross sections, Uwan site .......... 186
Level 414 rim cross sections, Uwan site .......... 187
Level 421 rim cross sections, Uwan site .......... 188
Level 422 rim cross sections, Uwan site .......... 189
Level 423 rim cross sections, Uwan site .......... 191
Level 424 rim cross sections, Uwan site .......... 192
New Pottery Types, Uwan site ..................... 193

Dendrogram cluster for Jaccard Coefficient,

Uwan site ........................................ 194

xiv

CHAPTER I
W

Some historians have often challenged archaeologists in
Nigeria to demonstrate historical relevance in archaeological
analysis. In some instance, out-right pessimism has been
expressed about the significance of archaeological
classification and typology for historical reconstruction in
West Africa. The aim of this study is to demonstrate that on
the contrary, there are new approaches to archaeological
analysis which can be used as means to "decode" aspects of
past social organisation in ways that would potentially be of
great interest to historians and others. In this
dissertation, this historical point is initially addressed
through a brief survey of Afigbo's paper (1977:81-98), titled
"Facts and myths in Nigerian Historiography". The point is
made through a consideration of three historiographical
traditions in Nigeria. The three traditions are a)

Precolonial, b) Colonial and c) Nationalist historiographies.

1.1 WMWWMW-

1.11 W W.
Afigbo (1977) has suggested that precolonial

historiographical tradition in Nigeria can be discussed by

reference to the Forest peoples on one hand, and the

Grassland peoples on the other. Forest historiography is of

two types. The first consists of "myths and legends" in the
1

2

folk tradition of the people, in which origin is attributed
to the actions of the supernatural, animals and inanimate
objects. This can be observed for example, amongst the Igbo,
Yoruba and the Bini, whose legends recall the role of nature
and supreme deities in different acts of creation of their
lineages and clans. Each group, explains its own creation to
include the creation of the entire world - "an amazing
combination of the narrowest egocentrism and localism with an
all embracing universalism" (Afigbo 1977:84)

Afigbo sees this historiographical technique as partly
"irrational", but that, it is often complemented by a second
type of traditional historiography, which includes “accounts
of actual heroes and heroines". (Afigbo 1977:85) This second
type of accounts, comes in the form of ritual incantations,
praise and festive songs, poems, proverbs, chants, folktales
and so forth. While, the myths and legends deal with the
remote past of the supernatural, the second type deals with
the recent past of real human acts and deeds.

Amongst the Grassland peoples, on the other hand, Islamic
influence from about the 7th century, introduced with it,
Arabic literacy, including the ”Mecca-centric" tradition of
historiography. In this approach, various groups claim they
“migrated" from , for example, Yemen, Iraq, Egypt or Mecca,
perhaps as a result of the prestige to be gained by the
association. Even Forest peoples such as the Efik, Igbo,

Bini and Yoruba, have at one time or the other traced their

3

ancestry to these locations. Like the myths and legends, the
factor of the supernatural is very dominant in Islamic
historiography. For example, the rise of Islam corresponds
to the rise or origin of communities and peoples. In fact,
this Islamic rise of people and history, went into decay
after the first four caliphs (Roberts 1980:321-326), and
individuals within the community of Dar-al-Islam should work
back their histories and societies for the restoration of the
golden age. Arabic literacy helped to reinforce this
ideology. Afigho sees irrationality in Islamic
historiography, in that real human factors and genius in
different local settings are ignored or undermined.

1.12 WW-
Colonial historiography in Nigeria, marked a watershed.

For the first time, an integrated account was needed for
the purpose of administration. Colonial historiography
however introduced its myth, -- the "myth of race".

Different Nigerian peoples were seen to be at different
levels of history or civilization based on their racial type.
This was a "man-centred" historiography, and its racial
principle was Just as “irrational" as the supernatural myths
of precolonial times. ’It was believed that different races
passed through cultural stages, from primitive to complex,
advanced or civilized society. Evidence for this was
observable in the political ability or "cultural virility" of

the light-skin, Hamitic peoples of the northern savanna,

4

compared to the “naked savages" especially prevalent in the
forest region (Perham 1937z3). This view was reinforced by
the presence of centralized political administration amongst
the Islamic peoples of the Sudan. Peoples of the Forest who
similarly had centralized political organizations, such as
the Yoruba, Benin, Idah, Nri and the "organizers of the long
JuJu (Ihini flkpghi) oracle of Arochukwu“ were regarded

as being ruled by descendants of "small bands of the
Hamites", which penetrated into the forest. Non-centralised
peoples, such as the Igbo, Ibibio, and the Benue Plateau
peoples, came on the lower end of the racial ladder.
Colonial administration applied this principle in the
Indirect Rule system, where the centralized peoples were
governed through their traditonal rulers, while the non-
centralized peoples were administered through locally

stationed colonial officers.

1.13 W W-

‘ Post-colonial or “nationalist historiography" presents
different facets, and partly reflects the previous colonial
definitions of the racial principle. Post-colonial
historiography again, introduced its own myths and
subjectivity which manifested as ethnic nationalism, in
which newly Western educated elites strived to write down the
histories of their ethnic groups in such a way as to create
the impression of a long standing histories of occupation of

their localities. Ethnic historiographers also wanted to

show the hegemonic dominance of their groups over others.
They also saw the "beauty" or relevance of their traditions
or institutions in modern society. In terms of methodology,
they relied more on "oral tradition", which includes the
myths and legends, folktales and so forth of the precolonial
period. (Alagoa 1972). In more recent times some have
applied the evidence of Archaeology. (e.g. Onwuejeogwu
1981).

The point, however, has remained as to the objectivity of
historiographical method. The archaeological and
anthropological perspectives can be shown to offer more
valid approaches for an objective presentation of history
not because methodological and theoretical problems do not
exist, but because it is possible to work with competitive

and potentially valid assumptions or hypotheses.

1.2 MWW
’ In this section the point is that, archaeological and

ethnographic methodologies have been important
anthropological approaches to historiography in West Africa.
To illustrate this point, some key cultural motifs during the
Early and Later Iron Ages in West Africa, are briefly
surveyed.

An important point made by Shaw (1976:159), in a
tribute essay in honour of Basil Davidson, stated that

African Archaeology had progressed from a consideration of

6
"What Happened" to the establishment of "How it Happened".
The issue here was that an outline, or a framework or
chronology had been established for cultural study on the
continent.

In the case of West Africa however, although far little is

known of its Early and Middle Stone Ages, when compared with
the rest of the continent, there is good evidence that the
subregion was populated at least not later than 10,000 years
ago, (McIntosh and McIntosh 1981), by Late Stone Age peoples.
From this point on, progressive development transformed the
peoples from hunter/gatherers to societies and cultures who
practised agriculture and animal husbandry from about 7000
years B.P. Early iron using peoples such as the Nok culture
(from about the 8th century B.C.) had settled villages.
Through the development of long-distance trading networks,
towns centers for the control of trade items and resources
developed earlier within the transitional savanna zones.
. In Nigeria, the Nok culture complex is a good example of a
tradition that was continuous from the Later Stone Age to the
Early Iron Age. Nok has been known for its terra cotta art
styles and it is dated from about the 8th century B.C. to
A.D. 200 (Shaw 1981). Nok culture has been recognized over a
wide area of the Middle Belt of Nigeria and it’s cultural
influence in art styles and pottery are known from the Later
Iron Age societies of Ife and Benin.

Another Early Iron Age motif culture complex, comes from

7

the Mauretania-Senegal area of West Africa. This is the
construction of funerary megaliths and tumuli. The tumulus
is a burial mound, and one for which there is an Arab written
record was constructed in A.D. 1067 for the king of the
empire of Ghana (McIntosh and McIntosh 1981:612). Megaliths
on the other hand are single stone structures erected at
burial sites and evidence of such structures indicate a
fairly complex caste society. Some archaeological sites
associated with these structures include Tegdaoust and Koumbi
Saleh, reputed to have been the capital city of ancient
Ghana. The site of Tegdaoust in Mauretania has a rich
cultural deposit (and may date) from about A.D. 810 to 1800
associated with brick and stone architecture, local and
imported ceramics and stone and metal artifacts suggesting
trading links, agriculture, smithing, hunting and tailoring
(Es’Andah l976:4, Hill 1978:604-605).

In later West African societies, Arab and European
influences, are particularly important in the development of
trading relationship. Some of these relationships can be
specifically described as follows: In the context of Arab
and Islamic influence from about A.D. 700, West African
agricultural productive techniques became mixed with trading
or mercantile activities. The long distance trading networks
across the Sahara involved the exchange of salt, cloth, and
beads for West African gold, ivory, and slaves. Well-known

trading centers of this period, include Timbucktu, Gao, and

8

Jenna Jeno in modern Mali; Sokoto, Katsina and Kano in
northern Nigeria and perhaps as far south as the forest
states of Benin in Nigeria; and Begho in modern Ghana (see
Figure 1; Davidson 1974:62; Calvocoressi and David 1979:2).
West African currencies included gold, brass, salt and cowrie
shells (Cowrie shells were used up to colonial period, and
were traded from the Maldive Islands of Sri-Lanka and the
coast of East Africa. ) (Johnson 1970:17-49, 1970:331-353).
Accompanying this mercantile influence was the use of horses,
spears and swords, but the plow, the wheel, and the shaduf
(an Egyptian device used along the Nile for drawing water)
although penetrated into the Sahara, did not reach West
Africa. Agricultural production techniques, therefore,
remained simple with the use of the hoe and the slash and
burn method. The land use system was also simple; vassalages
and the landed fiefs were either absent or very unlike those
of feudal Europe. During European contact, the mercantile
economy was intensified first by the trans-Atlantic slave
trade and later by trade in palm (oil and so forth) produce.
European contact brought with it the use of the gun which
partly reinforced the powers of coastal or forest kingdoms.

Jack Goody’s (1971) ethnographic studies have classified
the recent West African societies into types namely:
Acephalous societies, mass dynasties of the savanna, and
forest kingdoms. The acephalous systems consisted of

segmentary groups whose important cultural equipment included

 

 

 

' Kano-

o‘\ .v
\
‘1131 1‘
\’ o\’—\

\ \ ‘ ~ ,
QI/QI/QI/QI/QIa/«Ia (‘0880110 .

1

a, I O
\ o \
I I \/ \/ \/ \I.’\’: l— / / anlfl / .

 

 

 

 

\‘\ \~.~. \‘\‘ \‘x \ \ \
I’ I/ l/ l’ I/ I/ I/ I/ o. I/ I
s \ \ \ \ \ \ \ \ \ \ \

I \I \l I s v s’

\ (ll: ‘

FIGURE 1: Map of West Africa. (McIntosh, 1974, 1976)

10

the bow and arrow for hunting as well as the iron for
agriculture. The societies lacked centralized government and
are sometimes also described as egalitarian. They were
present in both the savanna and forest, often living side by
side with centralized polities. Such groups include the
LoDagaa, ermbe and Bole of Modern Ghana, and the Segou and
Kaarta of Mali. The "mass dynasties" on the other hand, were
organized around control over trade routes in the Sudan zone.
The states depended on a mobile cavalry forces which were
sometimes used to raid the ill-equiped infantries of the
acephalous peoples. Leadership within this group appear to
have rotated within an elite class of Islamic lords. In some
states, such as was in the Mossi (Bourkina Fasso), there was
control over the production or smithing of spears and swords
or other items, symbolic of state power. The centralized
kingdoms of the forest, were distinct in that they initially
based their powers on mythological sanctions, but they became
despotic during the slave trade, from access to guns which
they used to raid the hinterland for slaves. 'Such kingdoms
included Dahomey, Benin and Opobo (Dike 1956).

L3213£22§22£ihi§m

It would appear from the brief overview above, that
specific cultural determinants are responsible for the growth
of society. Such determinants include change in the
technoeconomy, militarism, trade and religion. Can some or

all of these determinants account for the evolution of the

11

ancient kingdom of Benin, (south-western Nigeria)? What
criteria are to be applied to address this problem
objectively? In Chapter II, this question is brought more to
focus through a critical survey of Benin research and related
studies. In Chapter III, this study focuses on the

archaeological methodology.

CHAPTER II
mwmwm 13

In Chapter II, information about Benin is sifted from
various study foci that could throw light on the uniqueness
of the people, their environment, society and culture. The
purpose of this is to provide an understanding of the setting
of the archaeological research, review the literature, and to
provide interpretive information which will be drawn on in
later chapters. Information has been drawn from ecology,
society, art, architecture and archaeology.

Benin City is located within the high rain forest zone of
West Africa. It has two clearly marked rainy and dry seasons
which partly regulate the hoe agricultural economy, with root
crops as the main-stay. The people speak a language that
belongs to the (Kwa subgroup) Niger-Congo branch of the
Niger-Kordofanian family (Greenberg 1970). The society is
organised in two ways, - a village social organization, on
one hand, where age is the criterion of authority, and a town
or state social organization, where authority is hereditary
and based on the "rule of primogeniture“ (inheritance of
first son). The sex.principle cuts across the two, for only
males can participate in the two modes of social
organization.

Historically, the art and architecture more vividly help

to illustrate this Benin patriarchy. Previously, Benin

12

l3
archaeology has been addressed mainly to the city center, or
to the symbols of the state system. It has also provided a
more reliable chronological framework for the tradition. As
will be shown, it is possible to provide further evidence
for this mode of social organization outside the city center
and to demonstrate that this was part of a universal

phenomenon.

2-1 3221252..

The unique ecological conditions in the Benin area must
have played a significant role in the development of the
ancient Benin society as it continues to be important in the
area. For example, agriculture is still heavily dependent on
seasonal rainfall farming and, traditional builders relied on
the red Benin soil for house construction, clay image
building and pottery making. (This has however been changed
significantly since the introduction of concrete during the
colonial period). The red Benin sand or Benin Formation, as
it is geologically called, underlies Benin City itself, and a
considerable area around it (Figure 2). The Formation is
sedimentary, and is of Miocene/Pleistocene in age. It is
composed of yellow and white sands with iron stained gritty
beds (Freeth 1975:225-228). The Benin Formation probably
derived from a northern sandstone deposit, which is of
Paleocene to Lower Eocene in age. The oldest sediment in
the area, is the crystalline basement complex, whose age

ranges from Precambrian to the Lower Paleozoic (Ogunkule g;

li

g1 1980:1-3). North of Benin City, and in many parts of
Nigeria, this basement complex outcrops as many folding
inselbergs. The Benin Formation is overlain on the southern
coastline by recent beach and alluvium deposits which also
include decaying organic swamp materials. Since some
materials such as chart and flint, which are alien to the
Benin Formation may occasionally occur around Benin City, it
may be possible that such stones could come from the
sandstone deposit or the basement complex to the north, and
reports from the south eastern part of Nigeria indicate, that
there are ancient brass or bronze alloy deposits on this
sandstone, suggesting it was exploited traditionally for
metallic art at some point (Chikwendu and Umeji 1979:155-
161).

Geophysically, Benin City land is less than 1.000 feet
above sea level, a low-lying plain which is an extension of
the coastal plain. The Benin low land rises northwards to
the Ishan Plateau, which is the source of most of the rivers
and streams of the area. Except for the hills of the
Cameroon-Adamawa ranges, on the south eastern edge of
Nigeria, there are no major physical barriers to population
movement in the country and the larger rivers must have in
the past provided communication routes especially under
forest conditions (Figure 3). Nigerian vegetation follows
annual seasonal patterns of rainfall. Benin City is located

within the tropical rainforest, with annual rainfall between

15

1,524 and 2,032 mm (Figure 4). The double maxima rainfall
regime, in July and September here, or generally in the
south-west of Nigeria, makes it possible to raise two crops
of maize and vegetables annually (Udo 1980:11). Between
November and March, there may be a marked period of dry
season. There is a steady gradient of drier conditions
towards the north, and wetter and more humid situation in the
coastal swamps. The mangrove and the fresh water swamps of
the coastal region, are inhabited largely by such tree
species as Bhizgphgzg :gggmgfig, with aerial breathing roots,
and the raphia palm, which the people exploit for locally
'brewed gin. In the rainforest, the natural tree cover
consists of giant canopy trees such as iroko (thgzgphgzg
W) and obeche (W M), which have
been exploit for the timber industry. ‘Excessive ancient
farming in this area and tree felling now results (in some
parts), in a plagioclimax of palm bush (Jones 1956).
Evidence for ancient tree felling in West Africa, has
recently been related to the beginnings of agriculture in
this region which has been dated to 2,800 B.P. (Sowunmi
1985:129). About 100 kilometers north of Benin City, the
rainforest gradually gives rise to the savanna, but the
practise of bush burning in the region has been influencing
the enchroachment of the “derived savanna" ,further south, by

the elimination of tender and susceptible plant species,

16

resulting in the spread of typical savanna fire-resistant
trees.

The land use pattern here is still very traditional.
Except in government reserves and plantations, such as the
nearby Nigerian Institute for Oil Palm Research (NIFOR), 20
kilometers north-west of Benin City, individuals and family
groups own and raise gardens and farmlands to plant seasonal
crops and cash crops, or to protect semi-cultivated species
such as the oil palm, kola nut and timber. The peasant
farmers do not use any particular water or soil controlling
techniques but rely on the traditional method of bush fallow
and/or shifting cultivation for such arable crops as yam,
cassava, and cocoyam. Other crops may include rice, maize
and beans. Except for the indigenous guinea yam, most of the
other crops are introduced from outside especially since
European contact. Introduced cash crops here include cocoa,
rubber and coffee, which are also largely in the hands of
peasant farmers. Along the rivers and coast, fishing,
hunting; and collecting are known. Coastal fishers often
exchange their sea and river foods for farm produce from the

hinterlands.

l7

 

 

      
 

 

’. '8 3' g
' ‘ C ‘ ‘.’ I
s . I p I j o s \ a a i \
.§"'. ’8: .‘I 91". \. \".\'O'\I""’ ‘5’ z.) "\ f,‘ "‘ I
‘9’gl" ".§’ a\"\o"~|’\l '\’\‘\"
’ ‘o's".‘. 1‘, slg~ 'vte'l‘sxos".\|’ (I: ’3‘".\ I \l
5).].S" “1" ’\O\l\.’\" l""
[:‘l ~'~" I ".\’\l"
\‘l ~"~ ’I‘ ‘.O \ ’| 5’ ‘5 ’1‘ I "’
I I ‘I- I ’ I 1" l
"P ’ ‘ ‘ ~"t‘ I ’n‘ ’ UR ' " ’l‘
'\ ’Q‘ "‘ .8! s’.‘ |’§/‘\ ‘8‘ "\I \l‘l
5". <‘:‘<'l“.'l‘ I ' g‘!’ ’ I '
D Q ’ I . I ‘ I I ‘ \ o
' ~"l‘ 1‘: ~"|“"~ ’ '3 4‘: s': :|> 'l‘ ' s': ' “’3‘"
,slsl ”.39 a ,gl, I .s,‘l, 1\ol\,‘
"| '1sl"! \, g'rsllsvsa [‘o|"‘l’ .\
s I" \‘ ‘I, ~ 1" 1'.‘I," 4"}, ‘0‘ 'n‘ , ‘l' ‘,: .p‘\
I '1‘. ‘ ' .o‘sa‘sv‘I‘o.sv'~l," .\
C :‘p \‘II‘ ’3‘ 3, ‘IO'\ 1"" ‘n\"\o‘\', ‘ > < \ ,
u- " I’g"\" ""'\’\I':.""'\ll ‘ ’
~ t Is"‘ I ‘.o Q’JI I '.\'\o‘l '\
I ’I II ‘ I ’5 ‘I I “ ‘ ‘ ' \‘o I ’g"! I
(‘"I" Q"“ '1: “"‘ ‘ :': \ :z‘3’I‘ ’9“; ‘I: ‘8‘ . ‘
| ' '5 | ~ ’ \’ ' ‘0 \
"I ‘ "3 “’t 1 I ~‘vu‘ ’u‘ ’ i“‘l~":~ ' s': '3‘” s" (u‘
‘U‘ "s’u b "'\o\’§" 9.“,
‘I‘. .‘If t; I x I, I I I l‘ '|\ o \l
|’§’\ " ‘\ ""I‘ o‘\.'~' o
g ' I‘\I\I'\l'\o'\’ . .
a \

 

I

 

    

 

4t
4 I . , . ','.':. 1'.-
N :-.'-:-3-3:-3-':-.'°3:-.‘°1:-.'°1:
| . .,-. . -.-.~. .:-.-:-:-:;1:-:;I:-:-:;-:°I:.:-1:-:-3:-'~I:-:-I;.
".'.::.:.::.:.:: #45,...- A RI {2}}:-':I'-I{'-I{'-}."-}322-3:
-. -' '-.:.:-.;.:-.;-:-.;a:-.=:-.-.:-:-.:~.-:-I--:-I=::I°-:~I=:€°-:4'9?-

' .. 0.. 20" 2". .o.'.. ‘0'. .. 0...; .. o. :..e.:.. e

 

 

ml
RECENT ‘—"CRET EOU
- - .- Aauvium o Beach 88905513 ‘mfiflas

. f '— BENIN Foamncu *

.Soncs

 

 

l

 

 

 

 

 

FIGURE 2: Geological Map of Benin (Freeth, 1975:227)

l8

 

 

 

 

 

. -~ .. - .1-
”- a
.~ ~-
I- .. .090 . .:"... .‘ .~ 0. m m2)": { <0 .
s 0 Ore 0-0- ”. 1,. .
.... M ~. M. .~ ‘\ \ N‘GER‘I \
0. ~ ~'- O.- 0* ~ _ e
‘ 0... ~00 .- 0. en'— , ~', . .
“I H-l 0‘- w ~ /
'- I‘ O... o. o -~.I ~.. mm.
- °- . - o -- no. a”
' ‘ 0“. o-ofi-o 0e..- ' "' m‘ . '2
. "0.. “‘~. .~ - .‘u.“—'.. v.‘ W . ,«0.
° «a. a... ...- ~— ~— . . (3.1-m9‘. mfg"
- w~.* . - - 0* ~.
. "R;N:F'=... ...:".... ... ._.:";..f";.."fi... [:355-ntllbllllillbilti .
‘ .o 0.“ .. w 0-. a- - a..- an. ..- —‘
o 0.- - 0”. ‘~ ~.- -.g. ~.
”0 Q. '0- 0.; ca. - -- . . U D a
.00 0-— .... I.- ~~ a. . ‘. gggggggggggggg g I ‘!‘
Q0. 0 I -0 a ‘0. ~e a-

Q—

‘I!’ It.“

FIGURE 3: Map of Nigeria -- Geogra h and
Ar
Sites (Shaw, 1969) p y chaeological

19

,,,,///,, -

"/, /'/, -'

///

,/ /

 

FIGURE 4: Rainfall Map of Nigeria (Udo, 1980)

20

- NUPE

 

FIGURE 5: Eda-Speaking Peoples (Ryder, 1969)

21

FIGURE 6: let and 2nd Benin walls and Historical/

Archaeological Sites (Connah, 1975a)

Site Label:

a.
b.
c.

d.

h.

Oba's Palace (modern)
Ogiamwens Palace

Ugbekun Palace

Ogiso Ere Palace

Usama Palace

Edaiken Palace

Museum Archaeological site

Clerks Quarter's Archaeological Site

i-J. Inner waI1 Archaeological Sites

22

 

 

J miles

 

 

23

FIGURE 7: Benin Area (Bradbury, 1957)

Village Label:

a.
b.
c.
d.
e.

£0

9.

Use

Utekon

Uwan

Evboneka

Ekiadolor

Iwu .3

Urhonigbe

24

 

O
10
._ . 3° ‘30milos
.6
o 04
e“ .0
o d ob
" a i c:
. enn (y
. I
O“. “.60
a
. gramme
‘9

 

o 9
0
9,. .°\
0‘.
b \
.’
w"

 

l r

25

 

FIGURE 8: Escorting an Important Person (Benin Art)

 

 

 

26

o...“
0..
no.
J.“
.00
. O 00.
«3.1.55: 93.32.
a. 00...,00‘ .000 o
- a... 00 0° 0
.. o”°° 9.... . n
. ...° “”9. 0.0.7 0..
m' . ' “-2.3; ‘
:2? : -..
0.. '

.0.....:
49“ ”0.0%
.0. ‘

 

.-.-.o-a-.-
, I

FIGURE 9: Motifs in Benin Art

a . Leopard
b . Crossroad

‘3. Leaf

(1‘9 Geometric Shapes

h ‘ Maltese Cross

 

 

Ir

FIGURE 10: West African House Types

a. Round House
13. Rectangular Terraced Roof
c. Hausa Dome

d . Rectangular Prismatic Roof

 

 

28

FIGURE 11: Plan of Ogiamwen's Palace. Benin City.

nga h. unbuilt courtyard
a,b,c,d. built courtyard - ugha - room
e,f,g. enclosed rooms - ugha

Erie x,y,z. Erie quarters or rooms
i. Ikhinmin tree

k. kitchen

NOTE: Measurements were plotted as if the rooms form perfect
squares. Some of the wall lines had to be reconciled
to make them fit. The plan is however a fair
representation. \

 

 

29

 

 

 

 

 

 

 

 

1:!)
.J

 

 

 

 

 

 

 

.....Le

 

Reception Courtyard

 

L

__ L a
‘1

91 2.3m.

 

0 l.,______, .
_ LLJ

 

 

 

 

 

 

 

 

 

 

30

‘FTGURE 12: Benin House Types

ii.

.13

Commoner

Chief

 

Efle'

 

 

 

nga

 

 

 

 

 

Rfiwmyw...

 

 

 

{If

31

 

IFIGURE l3: Ogiamwen's Palace (photo)

BQOTE: Notice the Horizontal Grooves or Flutes on the wall.
Only Chiefly Elites can treat their wall surface in
this fashion.

32
2-2 32mm
The people who live in Benin City today and their

neighbours to the north and south (who speak closely related
dialects or languages) were first described by Thomas (1910a)
as Eda-speaking (Figure 5). Edo language belongs to the (Kwa
subgroup) Niger-Congo branch of the Niger-Kordofanian family
(Greenberg 1970). Sometimes, linguists classify the language
into the following (Bradbury 1957:14):

I Edo proper and the Ishan and northern Edo dialects.

II The Urhobo and Isoko dialects to the south.

Traditionally, the people of Edo proper who live mostly
in Benin City and around it, refer to themselves and the
language they speak as "Edo". They also call the Benin City
center as "Edo". In more recent times, the Edo proper has
been called Bini, perhaps to distinguish it from other “Edo”
dialects.

The origin of the people is obscure in myths and
legends. Popular Bini myths speak of founding father to have
come from the sky, being the youngest son of ngnghgg (the
Bini high God or supreme diety). This youngest son also
‘became the first ruler of the Bini people or Edo world, and
Ihim and rulers that followed him became known as Ogiso (ruler
of or from the sky). The traditional Bini historian, Jacob
JEgharevba, puts this Ogiso dynasty as before A.D. 1170. At
.at about this time, a practise of maladministration led to

cirisis and revolt and the people unable to find a suitable

33
ruler, sent to the ruler of Ila-Ife, to the west, in the
Yoruba country, to help provide a ruler (Egharevba 1968:1-8).
Oranmiyan, was the ruler that was sent from Ife, but it is
his son by a Bini woman, called Eweka I, that is generally
regarded as being the first 9b; of Benin (as the rulers
came be called.) It is thought that this second period in
Bini history, lasted until 1897, when the then oba
Ovonranmwen was deposed by a British punitive force and
exiled to Calabar (on the south-east corner of Nigeria.).
The third period or the most recent, falls into the colonial
period and after. Evidence from archaeology more or less

corresponds to this chronology.
Individuals are grouped politically in Edo society at
various levels from family to village or town political

units. The nuclear family consisting of a man, his wife or

wives and their children (who may own a house or be located

in rooms in the house of a man’s father or senior brother),
constitute the smallest socio-political unit. There are also
"joint family" and extended family groups, usually consisting
of a man and his married brother and sons with their wives
and children (Bradbury 1957:27-28). Members of the family
groups co-operate with each other in economic pursuits as
:farming, house construction etc. The head of the localised
patrilineage settles disputes and punishes misbehaviour, and

as he is often a member of the senior age grade in the

34 .
village council, he represents the family and he is
responsible for their actions. The support for the authority
of the family head, lies in his position relative to the
deceased patrilineal ancestors. The "rule of primogeniture"
(inheritance of first son) in Benin (Bradbury 1964, 1973),
makes this matter a complicated affair, for it is the eldest
son who sets up an altar at which to worship and communicate
with the deceased father; and his warefare, and that of his
siblings, wives, brothers and agnatic descendants is thought
to be dependent upon the goodwill of the deceased father and
his lineal ancestors who punish wrong doing with sickness,
death and other catastrophes. The eldest or senior son thus
has access to powerful supernatural sanctions for his
authority, but this authority is not complete so long as
there are surviving dead fathers junior brothers, who do not
have their own altars but have to be represented in that of
their senior nephew. The intense conflict that may arise
from this vig-g-vis, property and residence, tends to
weaken lineage affiliation amongst the Edo people, and
descent is rarely traced beyond the extended family group,
unless there is the prestige to be gained by belonging to an
important title holder. 0n the other hand, the people have a
.more "unitary" village social organisation, in which although
the “clan" may become dispersed, members of the village see
themselves as one and co-operate as such in economic and

cultural pursuits (Bradbury 1957:30-31).

35

The pan-village three-tier age grade system cuts across
kin groups and makes age rather than lineage affiliation, the
criterion of authority. The age grade system consists of
lzgghae, consisting of boys in their teens who perform
communal task of clearing and cleaning; the lghglg,
entered by adult men who perform the heavier skilled communal
tasks, as an executive arm of the political organisation; and
the Edign, who are exempted from communal labour
because of age and who form the village council for political
and judicial purposes. There are two kinds of village
headman namely: hereditary and non-hereditary. The non-
hereditary head of the Egigp grade is called Qgignggrg,
found in all villages and age is the principle factor of
appointment. Not all villages on the other hand, have an
hereditary headman or nggig, who may rule over a number of
villages which thus constitute a chiefdom. The "Qgigngggg
concerns himself with age-grade organisation and the
internal affairs of the village while the Qnggig is the
community’s representative vig-g-vis other villages and the
central authority of the kingdom" (Bradbury 1957:33).

This central authority in Benin City center is generally
regarded to be co-terminous with the arrival of the Obaship
from Ife, although this is a matter of debate (Obayemi
1976, Ryder 1975 ). Benin City itself with it’s
‘traditional urban character, (a city with a network of

<iefensive or territorial moats, whose cummulative length has

 

“F.

36

been put at about 105 kilometers (Connah 1963, 1967); ancient
art works, and a divine kingship which has survived to
present (Nevadomsky 1984a. 1984b; Nevadomsky and Inneh 1983),
symbolises a unique indigenous state . The influence of the
kingdom in the past was felt by people of different
linguistic groups, such as the Yoruba to the west, Igbo
people to the east, and the Igala to the north. The king was
addressed as the Oba. Since it was traditional that the Oba
may not leave the city until death, he depended upon
representative titled chiefs (enigie), some of whom are found
in the city center or in outside villages. Some of these
gnigig, claim descent from chiefs who ruled before the
introduction of Obaship, or before the village or chiefdom in
question was incorporated into the Benin kingdom. The
majority however, trace their descent to junior sons of the
past Oba who were made enigie on the ascension of their
eldest brother - a practise which still persists (Bradbury
1957:33).

The plan of the ancient city center was such as to have the
Oba’s palace (Egngg) at a focal point, and clustered around
which were the service wards or lgnnngn, in which were
skilled guilds of workers - drummers, blacksmiths, builder
etc. (Figure 6). The traditional chiefly castes are
Idivided into the seven king-makers (Uggmg pLhirgg); the
:Palace chiefs (Egbggzbg pngbg), ‘who reside in the palace;
Eund the town chiefs (Eghggxhg nLgpg). These three castes are

 

37

further divided into complicated system of hereditary and
non-hereditary titles.

The palace inhabitants are also further divided into
associations or 933, whose membership may cut across age.
There are three palace associations (gtn;ggn§e), namely,
Igghg, who are in charge of the Oba’s wardrobe and state
regalia; Ingguge, who provide the Oba’s personal and domestic
servants, and Ihiwe, who are the caretakers of the erie. or
harem. Titled castes and associations of men are symbols of
the state or kingdom. The age grade is a village social
organisation. Throughout the Edo speaking area, the age
Grade is present, and except amongst the northern people, an
hereditary titled chiefship may become super-imposed.

One cultural feature of all of these, is that the Edo
society is markedly male-dominant and this is expressed
variously, such as for example in the great deal of respect
and obesiance which is required of younger members, women and
untitled people. No one, for example, may address an older

Person, and for that matter, a titled chief or the Oba, by

hifi name (an example of name taboo). The Oba takes a new
”We on ascension. In fact, it was a taboo to say that the
013‘ dies, sleeps, eats, washes. All these were only
°xPressed through metaphors. Secondly, the rule of

primlogeniture, apart from being exclusive to men, also

r°~111forces the birth principle and the patriarchal nature of

 

38

the family head. Ryder (1969:4) reports that in the earlier
Ogiso political system, before A.D. 1170, some of the rulers
were women. What this could mean in terms of the shift in
the political system is hard to say. The rule of
primogeniture with particular regard to property, appears to
create incessant friction and weakens collateral
relationship. A family head commonly refers to dependants as
my "servants" or "little ones"

The religious system of the people is closely linked to the
divine kingship, since the Oba himself, is related to the
BuPreme diety. There are lesser dieties, such as hero-
dieties, spirits of the departed, personal spirits, etc. The
Olokun, is a popular cult among the people, and in Bini
NYthology, Olokun is the senior son of Osanobua, but now the
God of the sea, whose source is identified with the Ethiope
or Benin river. In the Urhonigbe village (Figure 7), where
this river is believed to rise, there is a large temple which
houses life size clay figures of Olokun (Bradbury 1957:52-
53). Olokun is represented as an Oba with his retinue of
Wives, and he may be consulted in cases of barrenness,
sicfltness and the desire for wealth.

Bradbury (1973:182-209), has explained that cult beliefs
Land, the festivals addressed to them), amongst the Edo
paOple, partly dramatise the separation of the sexes, and the
rationale behind it. Cult beliefs also help to articulate

within. a ritual setting, the male age-grade system by

39
demanding pledges or vows, as well as goods and services to
the secular order; and the bond that arises from this, help
to set one village membership apart from another. Some
motifs of this can be illustrated in the Ovia mythology of
the people. The Ovia myth is connected to a very beautiful
Bini girl, (named Ovia), who later married the king of Oyo
(in the Yoruba speaking area to the west). She became a
favourite of the king, for apart from her beauty, ”Ovia had
always worn a single white cloth trimmed with the scarlet
material called ngdg" (Bradbury 1973:187). This was to the
envy of the other wives who tricked the king into displeasure
with Ovia. The plot went like this: The senior wife of the
king of Oyo, summoned all the wives to go to the bush and
collect snails, but unknown to Ovia, she had cunningly told
the other wives to take bags along with them in which to put
the snails. When they got to the bush, they collected the
Snails, but Ovia not having any bags to put hers, put them on
her beautiful white dress which became stained. When the
Wives got to the palace, the other wives showed off the mess
t0 the king and explained to him that Ovia had a disease
‘Which made her to menstruate continuously. The king believed
the' story and Ovia in her sorrow, melted away in her tears,
and became the Ovia river which is on the west side of Benin
CM”vytoday (Figure 7). In the ritual drama, which is called
miafestival by the people, women are barred because
°f their treachery towards Ovia. During the period of ggig,

—-

 

40

the men go into seclusion in the Ovia groves located in the
bushes, (where they assume the spirits of patrilineal
ancestors), for three months, but from time to time come out,
masked, to perform the ugig dances which are later talked
about and boasted about, by the people. Sometimes, the women
would sing and pray that their hushands may dance well or
even excel others, including the patrilineal spirits, for not
dancing well, could bring shame and despair. At the last day
of the festival, the epinbhig (which the masked dancers are
called) are “killed" by the women who touch or embrace them
(for the first time since the period of seclusion started),
to bring them back again to the secular domain. Cults of
this nature still exist in Benin City today, although
membership may have expanded to include trading or
"professional" groups.

In Benin art and architecture, some visible and more
permanent symbols can also help to further illustrate some

historical and cultural elements in Benin society.

2-3 33mm.

The origin of Benin art is a matter for debate, but both
oral history and archaeological evidence appear to agree that
by at least the 13th century, Benin City was inhabited by
people associated with solid "bronze" working or smithing
(Connah 1975 :247-253; Egharevba 1968:11). This early
73mithing technique, which produced such materials as the

aeolid manillas, that have been subjected to spectroscopic

41

analysis (Shaw 1975:232), eventually evolved by about the
15th century into the hollow lost wax cast technique. Benin
memorial heads were produced by this method. Even the early
"bronze" heads were comparatively thinner than the later ones
as Portuguese influence after the 15th century may have led
to more copper supply which is an important component of
bronze or brass alloy (Eyo 1977:140).

Another uniqueness of Benin, is the range of corpus of
the art objects.(1). Benin art forms include memorial heads,
altars, plaques in metal; ivory gongs, coral beads, ebony
staffs etc. These forms and the features they contain
provide varied cultural information , but the general purpose
was to depict the court life and political history. The art
has been described as "court art’ because legend has it that
Benin artists were commissioned by the Oba to specially
produce the objects for commemorative use in the palace; as
badges of title; in ritual settings and as objects that are
of diplomatic use, when it was necessary to reinforce
alliances between Benin and external states or chiefdoms.

The political art of Benin includes plaques which often
show the attendance of chiefs and servants at the Oba’s
palace, sometimes the hands of the chief are held by his
attendants. In Benin City today, the Oba or chief, going or
coming from a ritual or social function, similarly have their
hands held in the fashion of tradition. (see Figure 8).

biotifs in Benin art include nature objects such as water leaf

42

( e- e) or the leaf called 93;; lb; 9g; kg (Figure 9c),
which means “the sun never misses a day" (Ben-Amos 1980:28).
Others include symmetric pieces such as the leopard (Figure
9a), or geometric features, including triangles, circles,
lunates, polygons, rectangles, and_trapezoids (Figures 9d-h).
Other art forms are of historical and mythological
importance. For example, horses did not appear in Benin
plaques until the 17th and 18th centuries, suggesting that
horses must have got to Benin as a result of Portuguese
introduction after the 15th century. The Maltese cross
(Figure 9h), was in use in Benin art before the time of the
first Portuguese expedition to Benin, in the latter half of
the 15th century. It’s use in Benin art is of interest
especially in ellucidating the relationship between Ife and
Benin (Ryder 1965:25-26). A form known as aghadgghgdg or
”crossroad" (Figure 9b), recurs in Benin and Ife mythology.
Shrines devoted to the trickster god (Eshu or Egg) were
usually placed at crossroads and door entrances (Fraser
1974:l; Thomas 1910a:32). In present day Benin, sacrifices
are still performed at crossroads, and some indigenous
christian churches commonly perform “spiritual work" at road
juctions.

One of the closest relatives of Benin art is Ife,
(Figure 3), and tradition puts it that it was Oba Oguola

(about 1280) who introduced brass-casting to Benin by

43
requesting the Q31 of Ife to send someone to teach his people
the art of brass casting (Egharevba 1968:11). Direct
derivation of Benin art has been questioned by some authors
(e.g. Ryder 1965, Eyo 1977), on several grounds. For
example, although the art of Ife is somewhat older (Classical
Ife art is between 12th and 14th centuries), than Benin, the
fact that Ife art is not particularly associated with the
kingship argues against this political relationship. Ife art
is distinctively naturalistic as against the more
"conventional" or political art of Benin. Ninety percent of
Benin art is in metal while terracotta manifest more in Ife.
The extinction of Ife art at about the 15th century, raises
doubts about the continuity of the ancient art with present
Ife. On the other hand, Benin art is continuous to present
day. This is not to say that there are no relationships. In
fact, it is now thought that given the broad relationship in
techniques, forms and motifs, the ancient art of southern
Nigeria, is related to a common, more ancient and more
northern center. A northern center which has been described
as the "Beny confederacy", of which the Nupe nation is a
survivor, has been suggested as the watershed for the other
later art traditions of Igbo Ukwu, Ife, Owo, Benin and Idah.
(see Figure 3 and 5; also see Ryder 1965). Igbo Ukwu, which
is dated to the 9th century is older in this complex (Eyo
1977, Shaw 1969:193-196). The significance of this, is still

lhard to judge, but archaeological information in general

44
indicates that Igbo Ukwu cultural materials are significantly
different from those of Benin and Ife, and that a
relationship via Igbo Ukwu, for example, could not fit the

data.

2.4 BenininflestAfrimArohitaoturslw)
Although architectural study in West Africa, is

problematic because of rapid mud wall decay, (McIntosh 1974),
it’s comparative study is probably more sensitive to symbolic
and historical information. Comparatively, West African
peoples share a common tradition of reliance on locally
available raw materials in their housing programs, although
form may reflect both ecological and symbolic conditions.
(Fletcher 1977:47-160; Prussin 1971, 1976; McIntosh 1974;
Haberland 1981; Vlach 1976). In West Africa, mud
architectural techniques were largely of three types, namely,
wattle and daub, terra pice, and tomb.

Wattle and daub is a rapid technique, involving the
erection of a framework of poles and twigs, which are laced
with ropes in the manner of basketry. Wet mud is then later
pounded onto this framework (McIntosh 1974:161-162). This
method was in use in West Africa, especially under conditions
of low plasticity of clay, such as the riverine deltas of
south eastern Nigeria. Benin sand for example, is very
plastic and readily sticks when wet.

The terra pice is a more prevalent technique. The

45

process involve the digging up of plastic clay usually within
the vicinity of the building site. This is then puddled with
water and used to pile up successive courses of mud wall of
approximately 60cm in height, and 26cm in thickness (McIntosh
1974:159). Each pile is allowed about two weeks to dry
before another course is added. In most cases, four such
courses are needed to the top, but in some hierarchical
societies such as Benin, a chief could have five or six,
while the Oba could have as many as he wanted (Goodwin
1957:67).

The tomb technique, commonly used in the Western Sudan,
reflects Arab and Islamic influence from about the 8th
century A.D. Tomb consists of dried mud bricks made with
wooden molds, and then later laid as walls with wet-mud
mortar (Prussin 1971:132). The Arab influence with regards
to architecture is significant in the Western Sudan and
Central Sudan. In the Western Sudan, the Madinka or Mending
speaking people helped to propagate Arab and Islamic concepts
cf dried mud masonry as is reflected in the mosques or
553123. In the Central Sudan, the Hausa played a similar
role. Hausa architecture features a distinctive dome, unlike
the rectangular or square houses with decked roofs of the
Mending. Centers in the Sudan where this kind of
architecture can be recognized include Kano in northern
Nigeria, Begho in modern Ghana, Timbuktu, Gao, and Djenne in
Mali (Figure 1).

46

West African architectural forms provide a clue to the
evolutionary pattern. Although it has been suggested that
one may not talk of architectue in West Africa, until
linguistic concepts were able to distinguish a "home" as an
idea from a "house“ as a physical manifestation (Prussin
1971:130-136), one may add that West African architecture was
both a home and a house in an adaptive sense. The
specialised post-Islamic mosque or aegizg, however, has it’s
unique symbolic quality. From archaeological evidence
therefore, it would appear that early architectural forms in
West Africa, were round. There is evidence for the possible
construction of round huts on stone bases, from the Nok
culture, dated from 925 1 70 B.C. to A.D. 200 1 50 (Angela
Fagg 1972:75). These early round houses (McIntosh 1976) were
probably adapted like the present Fulani round huts of
temporary structures of twigs and grass. Round mud houses
(Figure 10a), are widespread especially in the grassland
zones of West Africa today. It has been suggested that
Islamic influence must have helped to transform this round
concept into the "directional" or rectangular form (Figure
10b, 10d). Roundness is thought to be related to
autochthonous beliefs of emergence from a hole (Prussin
1976:10-11; Posnansky 1984:149), but that Islamic
directionality or orientation to Mecca, helped to create a
linear format in West architecture. There is a similar

Isituation to this in southern Africa, although, Islamic

47

influence may not have been directly responsible for it.
Centrality and directionality symbols are known in the so-
called "Bantu cattle culture" (Kuper 1980, Huffman 1982:140-
142), were the village is organised around a central cattle
byre. This central cattle byre also expresses male authority
as it may be regarded as the political and judicial meeting
point of the men. On the other hand, there is a left and
right arrangement of houses according to seniority including
a front-secular/back-sacred orientation of the building. The
symbolic order in the use of space may prove to be a unique
approach in West African architecture.

However, the rectangular form of West African
architecture, is of two types, namely, the terrace type
(Figure 10b), found mostly among the Madinka peoples of
Western Sudan, and the one with a prismatic roof (Figure
10d), common amongst the Bini, Igbo and other forest peoples
of West Africa. This rectangular form with prismatic roof,
of the forest region, could have been directly influenced by
the savanna type, but had to be adapted to condition of
intense rainfall, where a lighter roof which ensures
immediated run-off is needed. The Hausa vault or dome
(Figure 10c), is round but it represents a variant of Islamic
architecture which was popularized in the Central Sudan by
the Hausa people.

The point in all these is that the symbolic order in

West African architecture goes beyond a simple directional

48

orientation. Islamic laws as enshrined in the Malekite
rules, with respect to spatial provisions, for example,
placed emphasis on visual privacy and formalised male/female
distancing (Prussin 1976:10). This often meant the provision
of separate house units for each adult occupant. Each
household member may thus be formally excluded from house
units they do not normally occupy within the same compound.
The separation of male/female dwelling in West African
housing can be vividly illustrated by the spatial character
of Benin architecture which is probably not a direct result
of Islamic influence.

It was Goodwin (1957), who first provided insights into
Benin architecture. Benin houses were constructed as clay
courses, as was mentioned earlier, and this was then roofed
with thatch of broad leaves. On structural and ecological
grounds, the architecture could be seen as an adaptation to
the intense hot and humid condition. The heavy roof of
thatch was supported by series of walls which form around the
rooms and courtyards. Traditional Benin houses lacked wall
windows, so that the opening of the roof into the courtyards
served to admit light into the living floors. Water was
drained from the courtyards through sunken impluvium.
Adjacent to the living spaces were series of "mud beds“ or
raised clay platforms with some serving as tables or altars.
The horizontal grooves or flutes was a symbolic surface wall

”treatment of the Oba and chiefs (see Figure 13)

49

One such structure, that is still surviving presently,
is the Ogiamwen’s palace (Figure 6b), built before the 1897
punitive Expedition to Benin City. The palace is now a
national monument under the protection of the Nigerian
Federal Department of Antiquities. The spatial structure of
this palace can be divided into two major locales, namely,
Qggg and Exig (Figure 11). The Qgflg is the male quarters
while the E313 is the harem or female quarters. The chief
and his unmarried sons normally lived in the gggg, and each
:11; room may be occupied by one woman and her unmarried
daughters. Male dominance is symbolically expressed here by
the disparity between the areas of gggg and gzig, but the key
point is that the gggg also expresses the exclusivity of the
male. The female may be forbidden from coming near
particular locations of the gggg and at some times, she may
be totally barred from the qgga. This is largely because
each Bini room is also a "chapel" (Goodwin 1957:69), and
there is usually a space regarded as "holy" to which contact
by the female may lead to "pollution". This sacred character
of the ggug is unique, and clearly has political
implications, and is also related to mythological beliefs
already discussed.

In terms of direction, the ang and grip in this palace,
manifest on the left and right, depending on the direction
one is facing. In other Bini house plans, particularly those

of the commoner, the gggg and 91;; commonly express on a

50
front/back axis (Figure 12a), but in all, there is a certain
element of ggwg-sacred/exie-secular, since erie is
theoretically not forbidden to both men and women. Secondly,
it may be that the commoner form is more ancient and that the
elite type (Figure 12b) became super-imposed during the
evolution of the chiefship or kingship in Benin.

The Oba or king’s palace (Figure 6a) was larger than
those of the chiefs. The Oba’s palace was said to consist of
201 rooms, a number which may be mythical. In front of the
Oba’s palace, is located a market, Eki-Qpa a symbolic feature
of the political nature of the ancient economy. Such a plan
also suggest relationship between Benin and the Yoruba
palaces or Afin, which also have markets in front of them.
(Ojo 1966:35).

Benin archaeology has in the past been addressed to such
palace sites, some of which are historically known. Some of
these historical sites are therefore initially discussed
below.

2.5 Histgziggl Sites; (3) (Goodwin 1957, Egharevba 1968 ,
see Figure 6).
2.51 21.19.21.232 center: 31m:

Pro-13th century Benin is considered the first period of
Benin history. It witnessed the initial organisation of a
political system under the Ogiso, at the close of which a
weakness set in, requiring solutions which were dramatic.

For this period, there are two most important foci in Benin

51
history. These are the Ugbekun and the Ihumwum-Idunmwum
palace sites.
Maintains

Ugbekun is regarded as the first palace site of the
rulers of the Benin empire. It is today about five
kilometers from the Ring Road, along Sapkoba Road, on which
spot has been erected a sign by the National Antiquites
Commission declaring the site a National Monument. This site
of the first palace was built about A.D 904 by Igodomigodo,

one of the early Ogisos of Benin.

9221mm

Legend says that the palace was transferred from Ugbekun
to Ihumwun-Idunmwun, 36 years after it had been founded at
Ugbekun, indicating a brief period at the Ugbekun site. This
transfer was done by Ogiso Ere.. Ihumwun-Idumwun is today the
area in the 2nd East Circular Road between the Old Western
Boy’s High School site and the Electrical Power station down
to the slopes of Ikpoba river. Since this is a wide expanse
of land, historical sources appear to be in disagreement as
to the actual focus of this palace, and archaeological
investigations within this area have failed to yield accurate
information (Goodwin 1957). However, the palace is thought
to have remained within this locality until about A.D. 1170.

2.52 1.3132933322333333;
At about A.D. 1170, the Benin Ogiso organisation

52
witnessed a series of upheavals as a result of misrule or
mal-administration, culminating in the imposition or adoption
of the obaship political institution by the Bini people
through their relationship with Ife, the spiritual center of
the Yoruba nation. The person generally regarded as the
first Oba of Benin is Eweka I, who ascended the throne about
A.D. 1200. This is also regarded as the beginning of the 2nd
dynasty of the Benin empire, which lasted until 1897, when
Oba Ovonranmwen, the 35th Oba on that line, was deposed and
exiled to Calabar, by a British punitive force, attempting to
impose control over the southern area of what is now Nigeria.
During this period, the Oba’s palace appear to have been
moved essentially to two sites - these are the Usama site and
the area more or less around the present Oba’s palace. Two
other permanent palaces have in addition come into focus,

these are the Edaiken and the Ogiamwen palaces.

2.513933332313931
Usama is the palace site of the first Oba of Benin,

Eweka I, who constructed the palace at the time of his
coronation about 1200. The other Obas, Uwakhuahen (about
A.D. 1240) and Ehenmihen (about 1242) are known to have lived
here, but Oba Ewedo (about 1255) moved the palace from this
point to another site at Ogboka, on the location of the
present Felix Idubor’s cultural center, along the Sakpoba
Road. The Usama site is at the beginning of Siluko Road on
the‘ Ekenwan Road junction, opposite the Mobil Petrol

 

53
station. It is surrounded on all four sides by roads, with
Usama street on it’s western end. It remains unbuilt on,
and it has now been demarcated apparently to discourage land

speculators.

2.54 9123;321:921

The site of the Oba’s palace was occupying a much larger
land area, than what is the case at the present. Although
the palace, built by Oba Ewedo mentioned earlier, had a focal
point at the present Idubor Art center, the palace consisted
of ten Qghgghg or streets, and each street was named after a
particular Oba. This is an indication that the palace was
originally quite extensive, including the present Ring Road,
part of Mission and Forestry Roads, Akpakpava and Sapele
roads, as well as it’s present site. Evidence suggest that
the contraction of the palace over time could have been due
to the recurrent fire outbreaks in the city and palace,
causing extensive damages and requiring rebuilding at
different points in the general area of the site. Around
1816, for example, Oba Ogbebo, who had usurped the throne,
burned down the palace and hung himself, following a crises
which ensued in which he was vanquished. The palace was
rebuilt by Oba Osemwende, but was destroyed again by fire
immediately after the British punitive expedition of 1897.
Part of the palace was rebuilt by Oba Eweka II. Fire again

razed the palace in 1922, after which the palace was roofed

54
with corrugated iron sheets, instead of Ekhggrg and eke;
gkggkgg thatch leaves which when dry, burn fiercely. The
present palace plan is actually a creation of Oba Eweka II

(1914-1933).

2.55 Qsiaarenia sales: (see Figure 13)

The Ogiamwen’s palace is located on the northern side of
Sakpoba road, between the let and 2nd East Circular roads.
This palace, which has been taken over by the National
Antiquities Commission, as a national monument, is unique for
a number of reasons. The first is that it is the only
building of comparable status which has survived the great
fire and destruction sequel to the punitive expedition of
1897. Secondly, it typifies the distinctive Benin
architecture. Some members of the late Chief Ogiamwen’s
families are still living in the quarters so that the palace
offers a real opportunity for the understanding of some
palace practices, which otherwise would be unavailable to the
uninitiated. Legend has it that the Ogiamwen political
institution is a survivor of the early Ogiso system, before

the introduction of Obaship from Ife.

2.56 Edaiken’g pglgge=
The Ediaken title was created by Oba Ewuare the Great

(about 1440), to commemorate the death of Iken, the ruler of
Uselu, who was killed in active service against the Owo

people. This title was given to the first son of the Oba,

55
and the practise has survived to this day. It is largely a
bush site and there are a few large trees and some lkhinmip
trees which are used to mark points or areas of traditional
importance. The land, which is about 300 meters by 300
meters in area, is probably also used mainly during the
installation of the Edaiken. There is a building within the

site.

2-6 WW

Pioneer archaeological studies in Benin were first
undertaken by Goodwin (1957), who, following oral traditions,
identified some of the palace sites mentioned above. This
was followed by series of surface surveys and excavations at
the Ugbekun site, Ogiso Ere site, Usama site, Museum and
Clerk’s Quarter’s sites (the Museum and the Clerk,s Quarter’s
sites fall within the most recent location of the Oba’s
palace (see Figure 6). Goodwin’s excavated artifacts
included a slab of cast reptilian bronze head, plaque
fragments, rolled brass plates, bronze or brass rivets beads
and a number of pottery. The overall stratigraphic
significance of these materials at the time was lacking.
Goodwin however, complained at the time that the Benin area
was a difficult place for excavation partly as a result of
the wetness and plasticity of the soil. The artificially
high water table (as a result of the "prodigal local use of
water") (Goodwin 1957:72), in addition to the known Benin

architecture of steps altars and sumps also helped to cloud

56
stratigraphic analysis. Archaeologist Connah, on the other
hand, performed extensive excavations between 1961 and 1964,
within the palace locations identified by Goodwin earlier.
Connah’s excavated sites can be described as follows (see

Figure 6):

2-61=Eanin were;

The site is located at the city center within the pre-
1897 palace area. The site
had a number of pits and within one of the pits, came two
isolated human skulls, thought to be the remains of human
sacrifice, buried face down. Other objects include pottery,
a few beads (?glass) and tobacco pipes of African and
European manufacture. There is also a living floor with the
potsherd pavement technique, a technique known in Ife and Old
Oyo sites, indicative of a relationship (Connah 1963:468).
Connah reports that there is “no local tradition nor
historical reference to the previous existence of such

pavements in Benin" (Connah 1963:468).

2-62=9.l_e£lsa W are;

This site is also within the pro-1897 palace. The site
turned out to offer the deepest stratigraphy with about five
occupation phases. The stratigraphy runs through a well
shaft that was 58 feet deep. At the early period, the site
is associated with a mass burial of 35 individuals, located

at the bottom of the shaft. The bones had been in a solid

57

mass, having been adversely affected by weathering or erosive
action. The fact that parts of the bones such as the skull,
mandible and limbs were missing, also suggested human
sacrifice. This site is rich in pottery, and iron; and
bronze and glass objects are mainly located in the "Late
Phase“ of occupation. A date in layer 14 from the Early
Phase is 1490 1 90, while other dates from the mass burial
are 1310 1 90 and 1180 1 105.

2.63 91322331133393;

At the Reservation Road and Ogba Road intersections, on
the Innermost Wall, thought to have been built by Oba Ewuare,
in about 1440 to 1470 (Egharevba 1968), Connah cut trenches
across the wall at the Reservation Road and below the ditch
bottom at the Ogba Road site. The Ogba Road bottom of the
wall gave a date of 1340 1 105. The wall had been
constructed over earlier occupation deposits in which are

located pot fragments and charcoal.

2.64: flgggg gig;

In search of pro-16th century information, Connah selected
a site thought to have been the palace location of the first
four Obas or kings of Benin. This is the Usama site
mentioned earlier. The site consisted of pits producing
pottery, a miniature iron or 5Q; (ceremonial sword), and a
solitary bead, and some charcoal, animal bones and burnt oil-

palm kernels. Pit 5 on the site has a date of 1500 1 105

58
(Connah 1975 :182)

On the basis of radiocarbon dates, datable European
imports, bronze art tradition and pottey types, it is clearer
that Benin area was populated by at least the 13th century
A.D., by which time, there is evidence for solid brass
working and smithing (Connah 1975 :251). The 14th to the
16th centuries may be called the golden age, represented by
city walls, hollow "bronze" casts, and potsherd pavements.
By the 17th and 18th centuries, European influence is
indicated by the presence of imported smoking pipes, and
perhaps leaded bronze. In 1897, the colonial period sets in,
as the Oba is barnished to Calabar by a British punitive
force.

The Benin City walls, unlike the free-standing walls of
other northern towns like Bida, Kano, and Zaria, are
described as "dump rampart" in that materials forming the
walls were dug from the ditch or moat and then heaped on it’s
inner side (Connah 1967), forming a bank. This walling type
is considered as an adaptation to the wet forest environment
and are similar to the walls of Ife and Old Oyo (Soper rand
Darling 1980:61-81). Two of the major walls of Benin city,
form concentric rings around the city center, enclosing the
Oba’s palace on it’s southern side (Figure 6). Legend has it
that the Innermost or 1st concentric wall was built by Oba
Ewuare between 1450-1500, while the 2nd concentric wall and
3rd walls were built by Oba Oguola about 1280 to 1295

59

(Egharevba 1968). The first wall is considered to have been
a defensive structure, while the second and third walls are
considered to be territorial or to have been used for
demarcation between chiefdoms or as agricultural boundaries.
In a chain and compass survey of these walls, Connah
(1967:606, 1975 :105) estimates that the inner wall is about
17 meters high and 9 kilometers in circumference, while the
2nd wall is about 8 meters high and about three times the
Innermost wall in circumference. The cumulative length of
all Benin City walls is about 105 kilometers.

Walls have also been recognized outside Benin City
center, and the studies of Darling now indicate that there is
a wide geographical distribution of the walls or 123 as the
Bini people call them (Darling 1976:148, 1984), and the walls
may have been associated initially with competitive and more
or less equivalent chiefdoms. One of such walls is the Uwan
Wall which is located outside the city center about 20
kilometers to the northwest of Benin (Rees 1975:237-242).
The wall was found to be associated with pottery, charcoal
and oil palm seed shells, suggestive of ancient settlement
(Jones 1956). The wall itself appears similar in it’s
general characteristics to the 2nd concentric wall in the
city center, and has a height of between 1 and 6 meters, with
circumference of about 26 kilometers (Figure 14). This
research is centered on this Uwan wall area, in an effort to

provide fresh clues as to the relationship between ancient

60
Benin City and it’s outskirts.
The Benin system should however, be seen as the outcome of
a long process of cultural development, particularly from the
West African Late Stone Age ceramic tradition. Archaeological

evidence for this is reviewed below.

2.7 AWQMWWW

Wandthenlaeeeffiaainmm
Early pottery traditions in West Africa developed within

the context of a microlithic stone tool technology. Earlier
than 10,000 years B.P., not much is known about the Early and
Middle Stone Age traditions as compared with the rest of the
continent. Later than this date however, there is evidence
that Late Stone Age people, using microlithic tools were well
established in West Africa. They did not live in the extreme
forest conditions, but rather in the forest-savanna ecotone
environments. At the the Iwo Eleru site, near Akure in
Nigeria, and only about 50 kilometers into the rain forest,
there is the first evidence of pre-pottery microlithic
tradition dated to between 5000 and 11,000 years B.P. (Shaw
1969:191, 1984). Evidence from several other sites (e.g. Old
Oyo, Rap Rock shelter, both in Nigeria, Rim in Bourkina Faso,
Bosumpra in Ghana, Yengema Cave in south central Sierra
Leone, Kamabai and Yagala both also from Siera Leone), shows
that between the time when the tradition first appeared at
Iwo Eleru and the beginning of the last millennium B.C.,

61
microlithic traditions of hunter/gatherer peoples had become
more wide spread in West Africa (Fagg 1972, Rosenfield 1972,
Coon 1968, Andah 1978, Atherton 1972, Eyo 1972)

Then, a new cultural innovation appeared. This was the
use of pottery which was in some cases associated with food
domestication. It’s point of origin or introduction into
west Africa is a matter of debate, but earlier than the time
pottery first appeared in West Africa proper, there is
evidence that pottery was being used in the sahara (Camps
1977). Two such known saharan traditions are called the
Neolithic of Capsian Tradition and the Saharo-Sudanic
Neolithic. These traditions coincided with a dry climatic
phase which may have stimulated new settlements and
subsistence adjustments such as response to the need for
permanent water sources.

The Neolithic of Capsian Tradition originated from the
highlands of the Maghreb on the northwest sahara. It is
characterised by a Late Stone Age assemblage (geometrics,
mixed with bifacially flaked arrow-heads, polished stone axes
etc) in association with pottery, some of which are of ovoid
in shape with conical or “appendicular” bases, decorated with
a simple comb motif and incised rims (Camps, 1977:7). The
tradition also has ostrich eggshells. sometimes with
decorative engravings. The dates of the Neolithic of Capsian
tradition are between 6th and 4th millennium B.C.

62

South of the Maghreb, evidence from sites of Dhar
Tichitt in south central Mauritania suggests that influences
from the north-west sahara were moving south. Tichitt has
eight cultural phases (Munson 1971). The lithic assemblage
includes, end scrapers, small stemmed projectiles, highly
polished axes, and milling stones. Tichitt pottery is
decorated with dentate stamping (comb), fabric marking and
diagonal cord impression. Early Tichitt food sources
included fish, crocodile, and aquatic molluscs which were
later replaced by large animals. The Chebka phase or phase
6, marks the beginning of cultivation as indicated by the
abundance of pennisetum impressions on pottery. The date for
this period is about 1000 B.C. .

The Saharo-Sudanic Neolithic Tradition on the other
hand, was widespread in the central and southern sahara, and
fairly contemporary with the Neolithic of Capsian Tradition.
Unlike the Neolithic of Capsian Tradition, ostrich eggshells
in this tradition are rare and never decorated. Also, true
geometrics are rare or absent (Camps 1977:8).

Characteristically, the pottery is decorated with wavy-
line and dotted wavy-line. There are also grindstones or
milling stones. In some sites, bone harpoons indicate
fishing; in some other sites, there is evidence for zebra,
hippopotamus, wild pigs, mudfish and crocodiles as well as
domesticated cattle (Camps 1977:8). It would appear then

that the central saharan cultures directly changed from a

63

hunting and fishing subsistence to that of pastoralism at
about 8000 years B.P. (McIntosh and McIntosh 1981:606).

In West Africa proper, it appears that elements from
these two saharan neolithic traditions become represented.
At the northermost bend of the Niger, in Mali, the sites of
Karkarichinkat produced grindstone axes, projectile points,
and harpoons. Pottery includes simple jars decorated with
wavy-line, rocker impression (with comb) and fibre rouletting
(Smith 1974, 1984). Fibre rouletting is new and could have
been introduced from some southern sources such as Iwo Eleru,
where pottery appears earlier in about 5000 years B.P.
Karkarichinkat also has clay figurines of cattle, including
domesticated Egg and goat bones. Karkarichinkat is dated to
between third and second millennium B.C.

At the Iwo Eleru site, there is a ceramic phase in the
post 5000 years B.P. period, associated with trapezoids
(stone tools) made of chalcedony similar to the "sickle
flint“ (Shaw 1984:153), and is perhaps evidence of grass or
cereal harvesting. Iwo Eleru ceramics are continuous with
present day Yoruba ware.

In modern Ghana, there are ceramic sites at Kintampo,
Ntereso and Begho. The Kintampo Neolithic is preceded by an
initial Punpun phase in which pottery is present although
scarce, and is decorated with twisted rouletting. Kintampo
is associated with artifacts such as chisels, lip plugs, bone

harpoons and arrow heads. Pottery is decorated with comb-

64

stamping and rocker impression (Rahtz and Flight 1974:15).
The associated "terra cotta cigars" are thought to have been
used as potting tools (Posnansky 1984:147), but it may be
that they were also used in cult rituals. There is evidence
for dwarf cattle and dwarf goats in the form of clay
figurines. Ghanaian neolithic sites cluster in the middle
and later half of the second millennium B.C.. Some of the
Kintampo materials are located close to major Iron Age sites
which are continuous with historic societies. Such sites
include Begho, a medieval town founded about A.D. 1000 and
whose legend speaks of founding ancestors having emerged
"from a hole in the ground within a grassy plain" (Posnansky
1984:149).

Neolithic sites are known from western West Africa and
such sites are collectively described as ”Guinea Neolithic"
(Atherton 1972:39). Many of the sites have elements akin to
other West African evidence.

The West African ceramic Late Stone Age tradition can be
seen as a major break-through in this region. Apart from the
unique invention of pottery which has become an important
research focus, the period also marked the introduction of
plant cultivation and animal domestication and perhaps a

settled way of life.

65
2.71 1193 Agg

During this period, there is concrete evidence that West
African peoples were now inhabiting settled villages often
associated with symbolic art. Pottery continue to be used
and earlier influences can be observed . Perhaps the Benin
evidence illustrates that the people were in some
circumstance quite territorial, suggesting caste or social
differentiations.

As has been mentioned in Chapter I, the Nok culture
complex marks the earliest evidence for iron using peoples in
this region. Apart from it’s famous terra cotta figurines,
Nok materials include stone grinding tools, stone beads,
stone lip plugs, hooks, bracelets, pieces of knives,
arrowheads, spearheads and pottery decorated with carved
wooden dot roulette and incision. The people would have
lived in non-centralized villages practicing some form of
agriculture and colleCting. Contemporary villages would have
been linked together by the distinctive art tradition which
has been recognized over an expanse of 900 kilometers, east
to west. The art itself is made in the technique of pottery
and earlier similar practices are known from Daima (near Lake
Chad) (Connah 1981), Karkarichinkat, and Ghana, but these
earlier art styles were in animal form. In Nok, we have the
first evidence of clay figures in human form south of the
sahara. Influences of Nok art have been recognized in the

art traditions of Igho Ukwu, Ife and Benin.

66

2.8 figngn 1n Wes; Africap trade and state formation,
It has been suggested in chapter I that Islamic

influence in the post-700 A.D. lead to increased long
distance trading contacts and the growth of social wealth and
distinctions in West Africa. This does not mean that trans-
saharan contacts and trade were not known before Arab
influence. In addition, evidence in general suggest that
though production, transportation and trading of economic
items and resources did in some cases leave visible
expression of social differentiation and the growth of
controlling centers, this must be viewed in the context of
the contemporary belief systems (such as the Islamic ideology
of the savanna theocracies and the role of art symbols in the
forest kingdoms).

One item of early long-distance trade that must have
been essential, but rare, to human communities in West Africa
and "therefore a breaker of the barriers of self-
sufficiency", (Alexander 1983:1) is salt. Salt (Sodium

Chloride) was needed at four levels of consumption, namely,
a) as a dietetic necessity - human beings need at least 2
grams of salt in a day to avoid sodium deficiency, and
although this could be obtained from meat, blood, milk and
wine, largely vegetarian (or cereal) communities would have
needed a supplement. b) as luxuries - certain religious or
social functions may require particular kinds of salt either

as condiment or for offerings. c) industrial necessity -

67

trade in meat, fish or hides and skins may also require
quantities of salt. Salt was sometimes considered as a
preservative. d) as currency - a secondary use of salt was
more indirect, such as an easily stored wealth, often used to
pay tribute or as a general currency (Alexander 1983:1-2).
Archaeologically, it is often difficult to identify evidence
for salt directly because of the problem of preservation, but
it is thought that the absence of the camel before A.D. 400
makes it unlikely that the desert sources of salt could have
been available to the West African savanna and the forest
zones. Salt in these zones at this period could have been
produced by boiling salt-water on the coast or by solar
evaporation of salty earth in the lake Chad, for example.
The introduction of the camel however made it possible
probably for the first time to trade salt (from the rich rock
salt deposits of northern Mali (Taghaza, Taoudeni, Erebeb)
and southern Algeria (Amadror), to West African savanna and
forest communities. (Alexander 1983:5). This salt trade with
the sahara appear to continue up to the 19th century at the
time of which mechanised production of salt in Europe brought
cheap salt to the West African coast.

Other long-distance trade items in West Africa include
metallic goods (iron, copper, gold), sea shells, and glass.
Copper alloys used in metallic art in Igbo Ukwu, Ife and
Benin probably originated from such early sources as the

copper mines of Ajoujt in Mauretania (mined about 1000 B.C.)

68

(Posnansky 1973:150). Other possible sources for forest
copper include such sahelian towns as Derfur, Takkeda and
Sous (Eyo 1977:88). West African gold, ivory and skins were
in fact items of trade in the Greek and Roman towns of north
Africa. Evidence that the sahara was being crossed around
this time enroute trading centers on both sides of the desert
include the distribution pattern of engraved and painted
pictures of horses and chariots.

It would appear that on the sahelian foot-steps or
forest-desert transition zones, effects of trading contacts
between north Africa and West Africa were being expressed in
the growth of centers of resource control and management
along the trade routes. Such evidence include the rich
funerary megaliths and tumuli in the Senegambia area.
Megaliths are single stone structures and tumuli are burial
mounds. Wealth and social class are suggested by the rich
grave goods (which include gold and carnelian beads, iron
weapons, ornaments of gold and copper etc ). It has also
been suggested that some of the tumuli sites such as Kumbi
Saleh and Tegdaoust in Mauretania may have been associated
with the capital of ancient Ghana (Posnansky 1973:152,
McIntosh and McIntosh 1981:611). Eight century dates
from some of the sites suggest that the structures pre-date
Arab influences which are not felt in the archaeological
context until about the 10th century. Report of an Arab

record of the social distinction at about this time reads as

69

follows:

But in this case the idea that burial treatment
reflects status is supported by an Arab
chronicle written in A.D 1067 which describes the
construction of a large tumulus for the African
king of the empire of Ghana. According to this
account, the dead king’s body, his ornaments
weapons, food, drink, and slaves (still living),
were placed in a wooden chamber with a domed roof.

The chamber was then closed and earth heaped upon
it until it resembled a hill (McIntosh and
McIntosh 1981:612)
Sea shells from the Indian Ocean have been traded into
West Africa in the last 1000 years. The most popular of
these shells - Express mgngig - obtained from the Maldive
Islands of Sri-Lanka were probably first introduced into West
Africa from about the 10th to the 13th centuries by the
Arabs, who exchanged these for West African ivory, slaves and
gold (Johnson 1970:18-37). This trade appear to continue up
to the arrival of the Portuguese in the 16th century who
joined in by importing large quantities of cowries
to the coastal trade system. Their ready acceptance in
Benin and the Forcados River indicate that the cowries that
were already in use there were similar to the ones introduced
by the Portuguese. Another form of these cowries is the
92211;; gnnulgg, which are larger and more cumbersome to
transport, and hence of inferior quality as a currency.
9221113 gang1g§ were traded from the East African coast and
Zanzibar. However, on the basis of cowrie type, route of

introduction and method of counting, Johnson (1970a, 1970b)

has indicated that cowrie was an important item of trade,

70

ritual and social wealth and used alongside metallic currency
(e.g. gold) in West Africa.

Now to return to the ancient kingdom of Ghana, although
its kingship was a "pagan" institution, there is evidence
that the succeeding empires or states (such as Mali and
Soughay) were a result of Islamic conquest. Islamic conquest
in West Africa continued up to the last century and some of
the method by which this was done can be illustrated as
follows: Traditionally, the ethnic nations of Senegambia
were governed by monarchs who depended for their power on
access to guns and horses which they obtained in exchanged
for slaves. The monarchs were supported by an army of slaves
such as the ngddg which often raided peasant muslim
agriculturalists and nomadic herders. As the slave trade
declined after 1800, trading in "legitimate“ commerse or
agricultural products (such as peanut oil, hides and wax) now
made the peasant able to purchase guns and other European
products. This, thus titled the balance of power in favour
or the muslim peasantry. From amongst the muslim clerics,
powerful leaders emerged to overthrow the traditional
monarchs and introduce Islamic principles of morals and
government (Gamble 1957, Klein 1972:423-433). The same basic
principle of access to resources can also explain the
establishment of the Masina state (Mali) (Johnson 1976:481-
495), and Sokoto (Last 1976). The point however is that a

71

commitment to Islamic ideals remained a critical factor -
need to enthrone sharia law, Islamic symbolism etc, with the

economic and technological facilities acting as means to an
end.

In the Yoruba states as well as Benin, the role of early
trade is poorly understood and there is little or no direct
evidence of major or intensive extraction of energy from the
hoe agricultural economy. Formation of the kingdom is
attributed to the arrival and/or spread of politico-ritual
elites (god-kings) over the area (Obayemi 1976, Egharevba
1968:6-7). Kingship and the title systems were often
expressed in the art symbols. The Yoruba bead-embroidered
crowns with veil and bird decorations is said to have been
initiated by Oduduwa, the father of the Yoruba nation
(Thompson 1972:227). It may even be that this veil motif in
Ife art could have been an adaptation of Islamic symbolism in
the forest. Also, naturalism in Ife art has been suggested
as an indicate of divine kingship worship.

The religious content is so clear that we might
reasonably characterize the art of Ife as hieratic
rather than aristocratic, especially since the
majority of Ife leaders had priestly function
(Willet 1972:224)
Also, notice that political offices may originate by the
proliferation and elaboration of the regalia systems, such as

is present in the 919 title art forms amongst the Ibo people
(Cole 1972:88). In Benin, the role of art as a politico-

72

ritual medium has been discussed earlier. Others (e.g. Ryder
1969, Darling 1984) have however raised the question of
military conquest as a possible determinant for the evolution
of Benin. How this may have been related to resource
appropriation (especially prior to European influence) is
hard to say. Benin kingship predate European influence.

The coastal states ( e.g Dahomey and Opobo), including
Oyo, to the north of Dahomey appear to emerge as dominant
centers in the attempt to control mainly European trade items
(such as horse, salt, cowries, glass, textiles, guns etc)
(Argyle 1966, Dike 1956). Ikime (1969) has documented the
growth of a merchant class on the Niger delta particularly in
the 19th century following the decline in slave trade and the
need to exchange forest resources such as palm oil, ivory etc
for such European goods as glass beads, cowries, manillas,
tobacco, caps, alcohol, and textile. The picture is that
merchant ”chiefs" such as Nana Olomu arose as a result of
their personal abilities to organise trading deals and
attract privileges by creating alliances not only with
European traders on the coast, but also with local chiefs and
agents in the inland. But these merchant ”chiefs", such as
Nana, also appear to belong to some dominant lineages or
related by blood to the 913 or king.

One universal feature of indigenous trade and market in
West Africa is the sex principle which appear to define the

range of trade items and method. Amongst the Akan peoples of

73

southern modern Ghana, for example, women control the trade
in local foodstuffs and crafts (Daaku 1971:177). Only men
could participate in the long distance trade networks as is
revealed by the trade diaspora mechanism of the Hausa cow
distributors. (Cohen 1971:266-281). The mode of organisation
of the Hausa trade diaspora in Ibaden (Nigeria) require that
individuals (butchers) who wished to get access to the cow
meat, become converted to Islam, which form part of the

communication network of this trading system.

2.9 Summers.

Benin research and related studies have been surveyed
through the unique social organization, religion, art,
architecture, archaeology and trade. Archaeologically, a
series of ancient or historical sites have been excavated.
Results from these excavations based mainly on radiocarbon
dates, European imports, bronze or brass art tradition and
pottery types indicated that Benin palaces were established
by at least the 13th century. Some of the non-local goods
suggest the importance of trade in this early development.
Also, the series of defensive earthworks have suggested that
territoriality or military defense could have been a

determinant in the evolution of Benin as a dominant center.

These are some of the points or problems critically
examined in this research. This research has been addressed

mainly to the Northwest, about 20 kilometers from the Benin

74

City center with the aim of obtaining new archaeological or
cultural information. The choice of the northwest was partly
based on the previous report of the Uwan Wall in this
location which is similar to some of the walls of the city
center. (Maliphant g1 e1 1976, Figure 14). The potential for
finding settlements around or related to this wall was
therefore likely. There was also the added chance of
obtaining single or unit component sites which could be
compared spatially as against the more historical or
chronological approach adopted at the city center. In
chapter III, the methodological models are spelt out as five

stated assumptions.

CHAPTER III

METHODOLOGY
In Chapter III, a method for the research and excavation
program on the northwest of Benin is developed by suggesting
a set of operating assumptions between artifacts and the
social system. Subsequently, a descriptive report on the
actual field work undertaken on the northwest between
December of 1985 and April of 1986 is presented. An
excavation report on the second phase of fieldwork is
presented in Chapter IV. The second phase excavation was
carried out in November of 1986 to provide additional
information on certain aspects of the first excavation

results.

3.1 33332211233
Fieldwork undertaken on the northwest of Benin followed

the studies of Maliphant g; 51 (1976) and others (e.g Rees
1975) with the primary goal of obtaining cultural materials
associated with the Uwan Wall. (see Figure 14). Such
materials hopefully could be compared with the city center in
a manner that certain questions related to the origin and
nature of Benin tradition could be addressed.

Against the anticipated data, questions about the Benin
cultural system were assessed on the basis of five

assumptions as follows.

75

76

Agsgmptign 1: Associations between artifact types or
design attributes on artifacts can reflect
temporal or historical significance.

This assumption follows several studies (e.g. Huffman
1970, 1982), which have related artifact types to historical
or chronologically distinct episodes. Emphasis is both on
typology and obtaining carbon 14 dates which aids
chronological correlation. Assumption 1 makes it possible to
provide new information on dating and to offer suggestions
about the cultural sequence of the northwest sites as well as
to clarify the chronological relationship between the
northwest and the city center.

Assumpi1gg 2: Distinct ethnic or population groups are
expressed in archaeological contexts as lack
of association between artifact types.

Closely related to the problem of chronological
significance, is the question of the relationship between
artifacts and culturally distinct groups of people. The
expression of group diversity in archaeological artifacts has
for example, been an important focus of study (e.g. Huffman
1970, 1982, Robertson and Collett 1983, Hall 1983, 1984a,
1984b), in the prehistory of East and southern Africa,
especially in connection with the problem of Bantu population
migration. The question this poses with respect to the
northwest of Benin is whether a continuity or discontinuity
in artifact tradition can be related to distinct groups or

population, and hence to some patterns of early settlement.

 

77

Assgfiption 3: There is a relationship between artifact
variability or variability of design
attributes on artifacts and the settlement
and residence type.

Assumption 3 is an attempt to seek for artifact variability
on the northwest and to relate this to modes of social
organisation. It goes beyond the question‘ of cultural
distinctiveness, to the question of whether artifact
variability could be, for example, due to particular modes
of post marital residence. Deetz (1965) similarly observed
that variability in pottery design attributes can be
explained by the changes in post-marriage residence amongst
the Arikara American Indians.

Agggmptigp 1: The relative occurrence of artifact type
(or categories), between sites, can indicate
a kind of exchange configuration which
includes class and privileged access, between
sites or centers.

The occurrence of non-local, exotic or valued objects,
could indicate a hierarchical relationship between sites or
centers. This principle has been applied with some success
to Woodland, Mississippian and other American sites (Struever
and Houart 1972, Caldwell 1964, Brown 1971, Goldstein 1980,
Upham e; g; 1981). Recent studies (e.g. Lightfoot gt g;
1982) have also suggested that leadership at potentially
strategic centers would have been stimulated by the need to
maximise privileged positions. In the case of Benin

northwest, apart from the distribution pattern of the

earthworks; the frequency of non-local goods, or occurrence

78

of particularly valued vessel forms, could also indicate a

hierarchical relationship between the northwest sites, and

between the northwest and the city center.

Asggmptign 5: Artifact densities can suggest differential
population sizes and social complexity.

The problem of the intensity of site use is a recurrent
issue in Archaeology. Although, an absolute index that can
suggest population size may be difficult to obtain directly,
the recent study by Shapiro (1984) attempts to compare site
population on the basis of artifact frequency or density. An
ability to suggest population differential between sites on
the northwest, at least, could be relatable to Assumption 4.
That is, a high population would probably reflect a more
complex settlement.

The problem of function, or activity mode, has not been
stated as an assumption here, though it can be inferred
indirectly from for example, the type density. A specific
relationship of type to function would depend on a good set
of data, and this is unlikely to be properly addressed here

because of the limited scope of the study.

79

3.2. Fieldwgrk 29 Egg Northwest 1

The northwest Benin sites, are located on top of the
Benin Sands as the parent material. The topography is
generally flat (between 200 and 500 feet above sea level),
although there are some irregular and gentle slopes present
both on the high grounds and lower grounds. The Okhuo river
to the north of the NIFOR station which flows southwest, is
the major drainage valley in the area, and there are minor
streams towards the southeast (Figure 14). Average annual
rainfall is 1,900mm. Soils in the highlands are generally
heavier than those of the slope bottoms due to vertical
eluviation or percolation of clay components and the
downslope wash of the sandy particles. This results in a low
send to clay on the high ground, while the reverse is the
case on the low ground. Correspondingly, plant nutrients and
soil acidity are higher on the uplands than below (Ogunkunle
e; 31 1980:8-40).

The lack of physical barriers and the flatness of the
surface with a good supply of river and flood water could
have encouraged settlement over the area in the past. In
more recent times, the area has been extensively farmed with
both permanent (rubber, oil palm, citrus, plantain) and

arable crops (yam, cocoyam, maize, cassava).

80

 

     
 

A JUwan Wall

9’“

“f

 

Ek"‘°l~ A.
. \
A Eko Abetu Slic-
B Uwan Wall Slla
L__J1mlla C Evboneka Farm Silo

D Uwan Ilvlng alto

FIGURE 14: Uwan wall and Northwest Sites (Maliphant et al.,
1976)

81

 

 

 

 

 

 

 

 

 

 

 

 

 

anomalies 9"“

 

 

Eko Abetu Site

FIGURE 15:

82

 

 

 

__ I L 1 laseJJne

 

 

 

 

 

 

 

 

 

 

FIGURE 16: Grid Plan for Eko Abetu Site

83

an.)

3:5 3 m

L
r-
{a
a
3
m

 

.54 .

 

 

. 6‘ '0 Q . . . I .0 .. C. O ' ‘

 

.710.

 

to
t1
)

 

 

. Very Dark grayish brown
'””‘4 wmmul+ roouh
ntflrreddhflibnmnISamhknay

 

FIGURE 17: 81 Trench Profile, west Face, Eko Abetu

84

 

 

 

 

Very Dark grayish brown
4 humus + roots. ‘

' "1; Dark reddish brown Sand/Clay
A] Bed Sticky wet clay.

  

 

FIGURE 18: El Trench Profile, North Face, Eko Abetu

85

3 1 5 2-5 9 3;s 9 415 s m

L.u__i_nm_i

I“

 

 

 

 

 

 

 

Dark grayish brown
. loose humus + Root.

f:ggpgggzcollapae materials

_. .' ,looae co la e material
'. ,. Reddisl'r-gri 4 elayu:

_ Gritty Sand.

 

 

‘*.Layer bearing gun flints

FIGURE 19: x2 Trench Profile, West Face, Eko Abetu

86

 

 

 

 

 

 

 

C
‘1»21/~—~/ ' I“ “f;
54“" . - . .‘- . o . ‘ ‘ \
'. . . ' ° 3 s 0 s
s o '. - . '
.7. . r . ’ b , f a '
.. '9 . . o '
8* -. . .=. ' ' . ,9 . " .
gafL-N- ~ ' . __.
103-d". ‘: °' 0' 'o '7-WI - n.
q . o ' ' ° 3 . u o o .
'."r—_~N__/—_f
:11

Dark grayish brown
loose humus + Root.

Park Re Collapse materials

 

 

FIVE: Pcoap‘c

'31-: 9'3 loose collage e material
13,-3.1. Reddislr- trig. + clayu:
5, no Gritty Sand.

‘ 990d"

* Layer bearing gun flints

FIGURE 20: X2 Trench Profile, South Face, Eko Abetu

87

 

FIGURE 21..¢,3 — ~ .cu. — flints , Level 4, x2,
Eko Abetu

cod fl «med .o.< no once a .enus cosaooou geocheno :\NM\<M on» msuhaoo nosed a
sauna axe .ooaa anon .oanuoua access «a u- museum

 

H 2... .»a8 3:
use.
Leanna» Ema nhmwooumgwum
awaken—.3. eanuaaow «woodman. 1.
crown raglan made has» .
svoow + uses; oaoou

95m. mas.

 

 

 

Ta
10.“

 

 

88

 

w. mu““““a.w r.
a

“a u. a: u mu

 

 

E
a a
I!
6
'4
”’4
0
0
I,

89

 

 

 

 

 

 

 

@ wall Stun ‘ .

% loose humus + roots

__ Very dark grayish brown
amnaét collapse material

. * i" no brown earth)
I; + Charcoal ’snail sheIls

-. .: [Red Clay.

 

 

 

FIGURE 23: Y2 Trench Profile, South Face, Eko Abetu

90

 

 

 

 

" 1 t} 9 I? u
acct” 4”. a $
«$.15\ \ W Evmfifi‘mv‘wfi'
' “no?
‘TTSL é?‘ gig?
\34 K
\3

; «184-’7- qtfl -_ Wm ”on-9-

rr“ I g kg tffi
I: g “$6 «3.1% .9 «8“ ‘00’3‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.Siee
. £5enk$>

 

' new
Rood .

10

 

 

 

 

JET

JO

 

Uwan wail Site

24:

F I CURE

Test Trench

T'l':

UVT: Uwan wall Trench

Datun height (100 meters) arbitrary

91

 

 

 

 

 

UVAN‘ WALL- TRENCH . :- . A ‘-
smsvs . —P
(b
1L5 Q
4.
'F'

       
 

'2‘ A Specimha
“WT/41:10 um.
4'60 mflrs Pam

 

 

 

2'5 5“” if""’I-‘.1e.. - ..
uni-3 ‘ -
“ ‘Humusi-Roei-s: :Do-m Radish 30010“.
3.5 Loose sand: Seclch‘sh e.”°‘°“‘
“ . 9!be {- (cam-Bug ad“; 516- __
4 ‘ Comp“? “N3 Dam flu "
1r- 4

 

 

FIGURE 25: Uwan wail Trench, East Profile

 

92

Ancient settlement sites are closely associated with
existing villages. These sites appear to have been
abandoned, by the local people, in the early part of this
century due to resettlements along newly constructed
highways. When reconnaissance surveys were first carried out
in the area in 1982, several recently abandoned sites were
identified. The sites are presently located in secondary
forest plagioclimax of herbs and palm bush. Some of the
locations were occupied with recent farm or garden plots,
and, unable to sieve through the forest mat, we took
advantage of farm clearings to make collections. In some
area, where some of the cultural plants are protected,
several tree crops could be observed. Also, the
characteristic Bini ritual plant called lkhinmig (4), can be
seen. A shrub plant, which is evergreen, the people regard
ikhinmin as “the oldest tree in the world" (Nevadomsky and
Inneh 1983:52). It was planted to establish a new settlement
and almost each compound in recent settlements has one or
more planted ikhinmin. Ancestral and earth sacrifices and
other rituals, such as purification rites and oath taking
particularly by women, were often addressed to the foot of
this tree. When old settlements were abandoned, the plants
were often protected on the spot. Apart from plants, the
northwest sites also commonly had wall embarkments, ditches,
cisterns, as well as pottery, cowries and other cultural

objects. Not all of these can be observed in all the sites.

93

Perhaps, the very ancient sites can only be recognized by
relatively few sherds which are characteristically worn. On
the basis of these observations, the northwest sites were

initially segregated as follows:

1. Sites with worn sherds: These appear to be mostly
located north and northwest of the Uwan wall and the NIFOR
station. We collapsing house walls or ditch system were

observed. (Figure 14).

2. Sites which appear to be contemporary with the Uwan wall.
These sites are located mostly within the Uwan wall. They
show a high frequency of pottey on the surface, with charcoal
and often a humus mound or bill. A reassessment of this in
the 1985/86 field-work, indicates that these type of sites
are similar to the type 3 sites, and that the chronological
relationship of these sites with the Uwan wall needs to be
objectively based on absolute dates, before firm conlusion

can be reached.

3. These sites (commonly on the south of Uwan Wall) also
have a high frequency of pottery, charcoal shells, including
humus mounds, series of ditches and walling features.
Sometimes, the walls are semi-circular or linear, and they
are commonly discontinuous.

The excavation programs (5) were based on a number of

considerations, including the available funds, facilities,

94
(e.g. equipment, laboratory.) personnel, and available time,
on the basis, of which, one living site (Eko Abetu) and the
transect through Uwan wall (Uwan Wall site), were excavated.
Test trenches were dug on a third site - Evboneka Farm site.
This was the first phase.of the field work on the northwest.
The second phase fieldwork was under-taken in November of

1986 as has been mentioned.

3.3 Eko Abetu 51:: (Figure 15).
This site is situated on a hill crest, about one kilometer

north-east of the present Ekiadolor village. Informants
claimed that this was the old settlement of Eko Abetu, a
village now about two kilometers towards the east. The
settlement appeared to have been abandoned in the 1920s, when
the people moved to the old Benin/Akure highway, constructed
by the colonial administration. This site appear to be on
one of the highest points (400 - 500 feet above sea level) in
the area of flat to gentle slope topography. Cultural
features in the form of coconut palms, mounds, walls and
ditches are clearly visible within the site, although the
site may not be readily visible as a result of bush cover,
from along the Eko Abetu/Ekiadolor road. The site which is
located within a citrus farm, was first located on the 17
August 1982, during which, some pottery and other material
collections were made on the surface. It is difficult to be
certain of the extent of the site, but on the basis of the

series of walls or mounds, it seems to fall within about

95

15000 square meters. Some of the mounds or walling features
consist of collapsed materials from house or domestic walls.
In some other areas, the mounds appear big enough to resemble
the territorial or defensive walling banks.

Within the site, these features created a contrast
between a west side of walls and ditches and an east side of
slope and level ground. The east side appear relatively
undisturbed. A 10/10 meter grid system was therefore layed
out to sample areas of high ground (or mound) and location on
the low ground. Numbers were assigned to the site grid as in
figure 16 (Joukowsky 1980:130, Fladmark 1978:84). The
excavation grids were established in such a way as to cut
through one of these walls (see Figure 14, trenches X2 and
Y2), 'as well as digging a trench on the relatively
undisturbed slope (trench Bl). X2 and Y2 trenches each

consisted of a 5/3 meter, while Bl was 4/4.

3.31: IIQEQL 21‘

This is the trench on the east side of the datum, and on
the gentle slope that appear relatively undisturbed (see
Figure 15). The 4/4 meter trench was positioned to avoid the
orange trees and as much as possible their roots. Digging
was done in 20 cm. arbitrary spits through 4 levels.(6). A
fifth level was dug to half, on the north end of the trench.
No features were noticed on this trench, and three distinct

layers can be recognized on the profiles (Figure 17,18). The

96

uppermost layer consists of very dark grayish brown humus
materials and roots. The materials are generally loose.
This is the most recent occupation layer. Layer 2, is of
dark reddish material, and fairly more compact sand and clay,
but, fewer roots. This layer is also cultural. The third
layer is very wet and sticky and perhaps has .more clay
components. This layer is largely sterile and corresponds to
the Benin Sands. Only level 1 appears to yield bottle glass,

a European product (see Table l).

3-32= 1mg}! XL

X2 and Y2 trenches were established close to each other
to cut across one of the low mounds which (by the presence of
a wall stump on Y2), appeared to represent collapse housing
features (Figure 15). The trench was dug as in Bl, through
about four levels. This trench presented features which
would have been harder to read, if Y2 had not been dug. The
upper humus layer is represented (see Figure 19 and 20), and
this is followed by very hard compact dark collapse
materials- This is in turn followed below by very coarse or
gritty sands. It was significant that the dark reddish brown
(hearth) layer, clearly noticeable in Y2, (see Figure 22,
23). was absent in X2 in spite of the closeness of the
trenches. The deposit here, therefore, suggest disturbance,
since also pedological studies of northwest deposits
(Ogunkunle g; 31 1980:31), indicate a significant increase of

97

clay component to a reduction of coarse sands, below 52 cm.
depth. Cultural artifacts from X2, (see Table 2) include
European or bottle glass, from levels 1 and 3, and then 2
distictive gun flints (see Plate 1, Figure 21) from level 4.
At first, these stone artifacts were thought to be belong to
prehistoric period, or the West African Late Stone Age, but
experts in historic artifacts (Jim Robertson, Dean Anderson.
and Professor Cleland ) of the Department of Anthropology,
Michigan State University, confirm that these were gun flints

and therefore European products.

3.33: Trench 12...

The Y2 trench, appeared to contain more interesting
features. A wall stump had been present on the mound surface
and this was traced into level 2, where it appeared to have
been established on hard concrete-like foundation. (see
Figure 22). But the surrounding materials are part of the
wall collapse, on top of which is the humus mantle. Below
this compact collapse house features is a dark reddish brown,
hearth layer which includes several snail shells and
charcoal. A carbon 14 date of A.D. 1432 1 100. has been
obtained from this component, which appears to be securely
sealed below the last building occupation. (specimen sample
number: EA/Y2/4). Below the hearth layer, is a red clay
layer which appears to corresponds to the Benin Sands.

European items are absent from this trench and apart from the

98

pottery, there are 3 fragments of locally made smoking pipes

(Figure 35f), and a round polished stone in level 4)

3-4= Emmm

The Uwan Wall site was located following the works of
Maliphant gt a; (1976), whose mapping indicated similarity
between this wall and the 2nd and 3rd walls of Benin City in
terms of magnitude, or height. The height of the Uwan wall
is between 1 and 6 meters. The height of the wall was
probably never quite even and also erosion and weathering
activities could have progressively altered it’s height.
Informants indicate that this wall acted as demarcation
between the chiefdoms of Uwan within the wall and Iyowa to
the south of it. At the wall, excavation strategy was to cut
a transect across the wall essentially to obtain dating
materials to provide chronological information between the
relationship of the wall and living sites within and outside
of the wall. The wall was encountered at the Evhoneka
village, on it’s southwest (Figure 14), from where it was
traced towards the east. A point was established within a
rubber plantation in an area which was fairly open and where
the wall was not too high (see Plate 2, 3). A grid system
was layed out from a datum and a 10/2 meter excavation grid
was layed on the Uwan wall as the Uwan Wall Trench (UWT.
Figure 24). Two test trenches (TT) were also established on

both north and south of the UWT. The site slopes markedly,

99

and the slope becomes very steep towards the south, in the
direction of the new Benin/Akure highway.

Excavation on the wall was done in steps (Figure 25).
On the sloping face of the wall, the deposit is quite
disturbed and drifting. The deposit appears sterile, with
only occasional (small fragments of ) charcoal turning up.
(Connah 1975 :85). A few charcoal specimens have been
obtained for dating (from layers which appear more stable
and not too close to the surface.) The test trenches also
turned out to he sterile, apart from a few pottery fragments

and some charcoal.

3.5: mmm

Far less time was spent on this site than would have
been necessary due to time constraints and the limited nature
of funding. The site was located for the first time on 4
April, 1986, through informants who said the old Evboneka
village site was located towards the west. The site turned
out to be situated within recent maize and cassava farms, and
also close to the Uwan wall on it’s inner side (Figure 14).
Only 4 pottery sherds were located on the surface in spite of
much searching. For two days, two trenches, one 5/3 and the
other 2/2 meters, were tested. The 5/3 trench turned up 22
shards and a few stones through a depth of 40 cm. The sherd

‘consisted of 2 rims and 20 very small body fragments.

100

3.6 Seams
In summary, one habitation site outside the Uwan Wall

and a transect across the Uwan wall have been excavated. A
third site within the Uwan wall was tested. Cultural
features and artifacts have been recovered. It would appear
from a general point of view and based on the excavation
alone, that at least 2 occupation deposits or phases are
represented on the Eko Abetu site. There could be more, but
such a decision would have to be based on further analysis of

the artifacts.

101

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

w m 8 e. .a a a

_ nutmeg. mum one pm. a» .a a. n

a... wag-fl in”... “85. 8.... 2.“. Rm awn mm a on u

...=.2uonuw 8: as. as A m. a a ,
.o.«x .eoam . -.;m .con _ s...: .newusm;uuuu w.wn.uwuuuu . “Hun" .eu.uw.muon .ea.¢m.“eon .cu.aa no. ~.p.s

 

 

ouqm spend oxm .huouco>cn Hoes .Hm cocoua "H Hands

102

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. a - .
~.su:oo. m . . . .
Bathuum . m “.85. . 8.. u 2.... R .r «a as o ..
lam-45;» . m u
e. . "nun“mmw “fiancee. “ _ “scum“. H can. on; me. up an a. n
. as w .
. «gunman» . w u m can on. .e. on an .. «
Imam-algal“ _ H
033.»; . .
at..._- m _ m cam. o.n up. an up a. .
. . . A.”
. m ” 26M...- azuas 38 as. 88 as»
8:. - .83 :25 88 3.: T“. 23.... J. 23... 3. as“: 633. .223 5.. ~25
sauna oxu .auouco>cu Hooa tux cocoua an sands

103

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

28;
. e.g.oaoa . “Ha-cave emu cu m . .. .. a
. Ina-aha s, - d
”axon—”am .3st as w . m 0.. M n
«soon-Au cocoon m .5 33 u 80 on p «0 mm m :—
qdflflflflullJmHmfl M
:2: m .23. m 32.an . A85 a 8% 8a.. a... on. «3 .3 a
"88.8 o8. p e: am 5 .- a
.8... at; :23 e3 3.: a.“ 3a 25.5.. 38. .2...» ran .23» 88 o
to causes as sees-s ou.su eases .o.a .euog» use a poa

 

sound oxu .huouco>cu Hooa .N» zucoua ”m man‘s

CHAPTER IV
FIELDWORK 93 Egg NORTHWEST g
The second phase excavation on the Benin northwest was
carried out in November of 1986. In spite of efforts, only
the Eko Abetu site yielded any good analytic materials from
the first phase of excavation on the northwest. There were
therefore no comparative artifacts from a different site.

The excavation program of November 1986 on the northwest

was designed essentially:

1. to look for and excavate a living site that could offer a

relatively abundant pottery sample.

2. to compare the new materials with those of Eko Abetu and

to see if a spatial relationship exist between the sites.

3. to collect more dating samples.

4-1 Eran Liziaa Site (Figure 26).

This site was first located on 30 August 1982, and it
was excavated between November 22 and 29, 1986. The site is
located off the Benin/Akure road, opposite the present
Uwan/Ogbogiobo villages (Figure 14-D). In November of 1986,
the old' farm plots where we made collections in 1982, had
been over-grown with weeds (7). We therefore entered a cocoa
plantation within the general area of the site to look for a

more open space which was convenient for laying the

104

105

excavation grids. Uwan village is mentioned by Egharevba
(l968:3) as one of the oldest Benin villages to have been
founded in the lat period (before 1170) of the Benin history,
but Chief Ogiezughanmen, the owner of the cocoa plantation
explained that Uwan and Ogbogiobo villages which are now
almost merged together, were founded at about the same time
during the reign of Oba Ewuare (about 1440). The founders
were members of the Benin royal family. Ogbogiobo village
has remained more or less on the same spot, but the Uwan
village had moved closer to it’s sister Ogbogiobo when the
motor high-way was constucted in the early part of the
century. Chief Ogiezughanmen explained that the presently
abandoned Uwan settlement was very extensive, but not as big
as the present village. The size of the ancient settlement
is difficult to determine on the ground because of the dense
vegetation, but judging by the mounds and sherd scatter, on
the plantation and the farm plots, one could estimate that
the village must have been about 250,000 square meters in
area. The site map (Figure 26) illustrates that like other
northwest sites, mounds and ditches are present and the areas
are often associated with farm plots based on shifting
cultivation.

The Uwan Living site is located on high ground about 500
feet above sea level. (The lowest ground areas here are
about 200 feet above sea level. Such areas are close to

river channels). The site is about 2 kilometers northwest of

106

the present Uwan/Ogbogiobo villages. The excavation grids
were established as shown on the site map (Figure 26). A 2
by 7 meter trench (410) was established inside the cocoa
plantation, and a 4/4 meter (420) trench was established just
outside the plantation, open enough to avoid the cassava
crops recently under cultivation. The owner of the site
showed us some iron slag from ground surface close to trench.
420. As usual, excavations were done in 20 cm spits. The
materials were not seived but efforts were made by probing
with a trowel to reduce the chance of losing some cultural

materials. (6).

4.11 Imm=

This trench was established on a low mound top. On its
south end, the deposit is quite disturbed with evidence of
two or more pits. These pits were traced to level 3. On the
whole, the materials are loose with coarse sand and humic
materials (dark grey on top, turning to red below). In
levels 2 and 4, on the northern end, the deposit is more
consolidated or compact. Isolated bricks occur now and then
and levels 1 to 3 revealed an adequate sample of pottery.

(see Figure 27, 28, Table 4 . also see Plate 4).

4.12 Trench $29:
This trench contrasted with 410 in that the deposit is
hard. It is on a lower ground and could have been affected

by erosion and other weathering factors leading to lateritic

107

concretion. This reddish laterite was present throughout the
deposit appearing to mask the layers. In level 4, and on the
south east end, there is a lens of charcoal/hearth material.
Even though the level itself is largely sterile, this lens
contains numerous pottery sherds. The trench in general also

yielded a good pottery sample (Figure 29, 30, Table 5 ).

108

 

 

ml'l

 

spatusnuu .eoose aspen

 

 

ovum ocu>de as): new museum

 

messaging!» PAS—
eguo couusbsesm §

‘

E O. a,

 

 

m

.ssrswa: roc
onsx<\cucom on

3

 

 

 

scausucsao soooo

 

 

o":

 

 

 

 

109

o
yum cm}: .oumm who: .oHeuOum adv nucoue new museum

.53.. ea.— »03 mesa-So a
ens. as.— too

026.: £320." ens: V\\\

58.5 .3195 dosage—:3 + uses: 0-09. + .333 %

 

 

110

 

 

1m;;f”;rf . . .fi;;‘;‘-;

Litter + loose humus + ssnd+roots. Grsyish brown
Hard reJclay. ?house
° :3' Dark red sand

" Compact wet red clay.

 

 

 

 

FIGURE 28: Trench 410 Profile, South Face, Uwan Site

lll

.4m

 

 

 

 

 

 

 

   

 

 

 

P
3"};

O“8||
.0 .

h

3

l

a...
man
8
.s ..x
km...
”a.”
D s

.3
t I.
cm...
W m

0..»
we“
Oy
/r+o
Ale 7..
r v.1 a
a 31
h .06
.37..

a e
8 h r
m c:
u.m.e
h ahw
n my
158?
h.ee
rmav

 

Trench 420 Profile, East Face, Uwan Site

FIGURE 29:

112

 

 

 

Thin humus + hard/compact. Dark Red.

Very compact clay. Reddish brown.

9.

Hard and

Dark Red.

charcoal + wet clay.

 

 

 

 

incarth.

y
a
I.
O
.4...
e
r
t
e
W
y
r
e
v

 

Uwan Site

Trench 420 Profile, South Face,

FIGURE 30:

113

 

 

me o co: once .1:

a _ m. es s:

mw~_e o.ssom _ _ . _ as. .m. m_.

as.» :2. s N z 8“ u:
ma..e o.ssom _

a N as. men __.

«:02;— 33... 0:03 0:00 am. u been 5 a: .96..—

ouum cos: .auouco>cn Hoes

.oev goeoue "v memes

114

 

 

 

 

m . men n a3 is.
so a. .N.
a. ma. "a.
_
_ m__ as» «a.
a._m . .
cor. N a _ on. so” .N.
assoc... .3... «new econ 3.2» n... 335 $8 5.... .26..
eu.m ems: .suon:o>ca .ooe .cm. rezone “m sauce

CHAPTER V
CLASSIFICATION AND ANALYSIS 1
Temporal Significance 91 North-West Pottggz
In this chapters V and VI, various statistical analytic
methods are applied to the pottery arising from the Eko Abetu
and Uwan Living sites (8). Although, the analytic approaches
have been segregated into two chapters, one exploring
temporal significance and the other spatial significance,
this is done for practical purposes. The idea is to discover
variability or the lack of it. In Chapter VII, artifact
variability is related to specific behavior modes, based on

the assumptions proposed in Chapter III.

5.1 WWW.

Traditionally, Benin pottery was hand made, as was the
general process in West Africa. (Thomas 1910b, Nicholson
1929, Tremearne 1910:103, David and Henning 1972). There
were two most important pottery villages near Benin namely,
Use, (now within Benin urban), and Utekon to the north of
Benin City, and east of the northwest sites (Bradbury 1957,
Willet and Connah 1969. see Figure 7). The potters obtained
very finely grained clay from Ekiadolor or Iwu villages
towards the west or near the Ovia river. The grayish clay
imposed very finely grained fabric on Benin pots. The clay

was processed by mixing with water. A new pot was built from

115

116

this. The potters started with a mould, usually an old pot
or calabash and by turning this around with one hand and
moulding with the other, the desired form was obtained. This
was allowed to set or dry for a while before the shoulder,
rim and ledge or carination might be added. The potter later
smoothened or burnished the pot surface with a snail or
coconut shell to reinforce the rigidity and imperviousness of
the pot. Various decorative motifs ( e.g. roulette, comb,
and groove) may then be applied, before the pots were
subjected to intense firing usually in a burn-fire of logs.
Pottery materials collected from the Benin northwest reflect
some of these patterns.

From the northwest of Benin, 2063 pottery sherds from
the trenches (B1, X2 and Y2) on the Eko Abetu site were
sorted and analysed. The sherds were coded according to the

trench and levels as follows:

Trench B1 (110) level 1 - 111
2 - 112
3 - 113

Trench X2 (120) level 1 - 121
2 - 122
3 - 123
4 - 124

117

Trench Y2 (130) level 1 - 131
2 - 132
3 - 133
4 - 134
Two major levels of analysis were applied, namely,

quantitative and qualitative analysis.

5.2 anntitgtive Analysis.
The quantitative method relies on the ability to define

an attribute list which reflects morphological and stylistic
variables, within an assemblage (Krieger 1944, Rouse 1960,
Gifford 1960, Deetz 1965, Brashler 1978), and to be able to
transform frequencies of occurrence with a mathematical
procedure. There is much debate about the quantitative
approach, regarding applicability and context, but the
studies of Read (1982:72-73) provide confidence that an
attribute definition is a qualitative operation but once
defined, it may be manipulated quantitatively. That is, that
criteria are really dependent on the problem at hand, and
should be formally defined.

The northwest pottery sherds were treated as analytic
units from which both metric and nominal variables were
extracted. Analysis has been addressed separately to the rim

and decorated body sherds.

118

5.21: Rim Sherdg
By an inspection of the rim sherds and on the basis of

previous knowledge, the following nominal variables were

defined for rim analysis.

3.1.1; mum 14.1.4.1:
Flat lip

Round lip

Grooved lip
Smoothened interior
Smoothened exterior
Everted rim

Hatched decoration
Carinated interior
Carinated exterior
Grooved decoration
Incised

Stab and drag

slashed decoration
Comb stamping

Carved wooden roulette
Twisted cord roulette
Punctation

Wavy-line

Perforation

119

More than one variable can occur on a particular sherd
and each variable is scored presence/absence. Secondly, no
distinction is made at this level between pot and bowl, since
this may be subjective, but commonly, everted rims come from
pots although the inward slanting rim of a shouldered bowl
(see Figure 42F) can be confused for a pot, using this
criterion. Thirdly, it is the technique of application of a
variable that is considered more important here than the tool
of application. For example, punctate decoration could have
been done with different kinds of tools and it would be
nearly impossible to determine the particular kind of tool
involved. However, carved wooden roulette (because of the
sometimes angular or regular nature of the lines) and the
twisted cord roulette (because of it’s bead pattern), cannot
be missed in an assemblage. Stab and drag appears as a
particularly variable form. On the rim ledge, stab and drag
appears to be applied with a pointed blunt object. On the
body of some sherds, it is sometimes applied with a comb. No
case was encountered where two kinds of stab and drag
appeared on the same sherd however, and such would have
introduced a problem. But there were cases where the
variable called "slashed" occurred with stab and drag
particularly along the raised ridges of the rim. The slashed
decoration appears to be a successive horizontal movement of

the hand, as if trimming the raised ridges.

120

Some of these decorative motifs have been illustrated in
the photographs (see Plates 5 - 9). The Plate labels are as

follows:

Plate: 111 body sherds

121 rim sherds

5

6

7 - 132 rim sherds
8 132 body sherds
9

133 rim sherds

Some key or essential attributes have been marked on the
photographs as follows:
a : carved wood roulette
b : punctation
c : slashed
d : grooved
e : wavy-line
f : twisted cord
One point of note is the variation that exist even
within the units, The carved wood for example, may consist
of vertical lines associated with polygonal patterns, zigzag
or diagonal lines. The stab and drag pattern does not come
out clearly in the photograph, but is commonly located on the
lower ridges that run on top of the in-curving rims of some
bowls. On the higher ridge as in in Plate 6c, is the
"slashed" pattern. (also see Figure 32 for schematic

illustration of the decorative patterns)

121

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Rim Nominal Scores, Trench Bl, Eko Abetu

 

 

 

 

 

TABLE 6

 

*
O
.3 5‘ ~- _-.
M
unmanned «-
sum n?!
“V
surname w
anus '
azmmma
mascara: «a
susmmimmmo
aumumsmmo i
unseen: 2
:mmcmvrns 2 *
GSSDNI ' ~ .—
s 2'
'030 asaooao. .'
:nmaxaosmmrmmu ‘8 a
“firearms "
.s N
Inumnans- '
mammal °‘ :
cmamam
‘rman: "’ F; a
mamas
an. o
-drxmumm9- “ ' .
no 2 i
amnmmm .
cairn:

 

 

 

Ls vol

 

 

 

 

 

* One sherd may have more than one motif

uaaoa eco can» once obs: has chose

0:0 I.

 

 

 

 

 

«soon.

 

 

 

 

' .

p

 

122

 

 

Iago
can

 

[cm

mm: mm

sun LAVA

scum

mos mm

 

 

tun-"V ‘w’

 

calms m

1*..- __

“WE‘VE-

 

 

 

Gasman
891m
norm
moss
Em

' 'un -SV'IS *
muons

’0! moons
mm:

'33! m1!

 

 

 

 

 

 

mmoouo

an mmoa
‘ at am

 

 

 

 

 

 

 

 

suon< oxm .Nx access .mououm umcuaoz 5.x up uqmda

Rim Nominal Scores, Trench Y2, Eko Abetu

TABLE 8

123

Short
Sample

25

10

 

smrgnm

 

EOIIVIDNfld
SCTIYIS

 

SJLSZCOH
0303 GELSIHI

 

anzmmi
‘.amo

 

auanusamfi

 

°OSH GZHSVII

‘0‘

 

DYE! GS? HYTS

 

8851331

 

’OSH GZAODHD

 

sugar:
anvmm

 

enema
mamamm

 

'33:! «mud

 

 

NIH QSIXEAS

12

 

83183113
QEEJOOHS

]?

 

HDIHEJKI
CSEJOOHS

5L

 

€11 TYAOSED

1]

 

""‘ d1? GKDOH

5

 

 

 

 

 

Lmnl

 

 

 

 

* One sherd may have more than one motif.

124

TABLE 9: Jaccard Coefficient for Rim Variables, Eko Abetu

 

B11. B12 B13 X21 X22 .X23 .XZh 122 Y2}

I

I

I f
. i

1

 

B11

B12 .95

 

 

s13 .79 .88 n
121 .88 .76 .ah I
x22 .73 .89 .86 .87
X23 .77 .97 .83 .88 .91.
x21. .85 .91 ‘87 .81. .86 .91.‘
r22 .98 .93 .80 .69 .85 .83 .90
Y23 .78 .92 .73 .85 .9h .97 .98 .87

 

 

 

 

 

 

 

 

 

 

125

TABLE 10: T-Test Sample Means for Lip Thickness, Eko Abetu

.-—~—-._ - _ ._

 

 

Lavel - n E Z! ; i (if!) ; i(X~i)2

0 ° I

. , . ,j ,3 .. 3
p11 J 42 E 300 g 7.14 5 149.03
812 f 39 ; 292 7.49 89.74
813 _ 14 107 7.64 43.12
x21 j 13 f 99 7.62 1 45.02
x22 ; 10 , 78 5 7.80 ‘ 32.51
x23 ' 12 102 8.50 49.00
722 ; 28 192_ § 6.86 55.44
723 j 13 99 7.62 1 26.74

 

 

126

TABLE 11: T-Test Sccores for Lip Thickness, Eko Abetu

 

 

f 2 . T' . ' T _ {Decision
Level/Level i 8 .df .Observed Critical (flog é;01)
311/312 3.02 79 0.90 I 1.67 Accepted
911/513 5.56 54 0.88 I 1.67 Accepted.
311/121 5.66 :5 0.79 1.67 iAccepted
p11/x22 3.65 :50 0.99 1.67 Accepted
p11/x23 5.80 52 2.15 1.67 _ Rejected
n11/Y22 5.50 :68 1.40 1.67' iAccepted
1311 /123 3.51 53 0.84 1.67 Accepted
312/315 2.61 f 51 0.29 1.67 Accepted
p12/x21 2.70 f 50 0.25 1.671 Accepted
n12/x22 2.60 ’ 47 0.5 1.68 i Accepted
p12/x23 2.85 g 49 1.80 1.68 5 Rejected
212/722 2.25 1 65 1.70 1.67 3 Rejected
812/725 2.52 ' 50 0.27 1.67 ; Accepted
p15/x21 5.52 .25 0.05 91.70. f Accepted
313/322 5.44 22 0.20 1.71 i Accepted
p13/723 3.84 24 1.12 1.71 3 Accepted
. 313/722 2.46 40 1.5 1.66 2 Accepted
213/723 2.79 , 25 0.05 1.70 i Accepted
‘x21/x22 5.69 § 21 0.22 ; 1.72 , Accepted
x21/x23 4,09 25 1.09 g 1.71 3 Accepted
x21/222 2.58 59 1.41 g 1.68 ‘ Accepted
x21/Y23 2.99 24 0.00 f 1.71 Accepted
x22/725 4.09 20 0.81 i 1.72 Accepted
2122/ .2 2.44 36 1.65 g 1.68 Accepted
X22/Y23 2.01 21 0.25 g 1.72 Accepted
725/722 2... 58 1.54 ; 1.68 Accepted
725/725 5.29 25 1.22 1.71 Accepted
722/725 2:10 59 1.59 1.68 'Accepted

 

 

 

127

TABLE 12: Body Decorated Scores, Trench Bl, Eko Abetu

SEER!)
SAMPLE"

.79

139

h1

 

 

$033250.

313m

OI
1

19-

.2»

 

weenxpcem

.-111|A

 

caved—5x
choc dove“;

.32

85

 

03333“
cocoa: cornea.

22

 

donwocH
33103:

 

ecceccH

 

dugouu

22.

 

 

m

 

 

 

 

 

* One sherd may have more than one motif.

128

Body Decorated Scores, Trench X2, Eko Abetu

TABLE 13

- .—..—— .-.. 2‘_ .,,

*

w
..W

S

HLhD

72 ,

3h

58

23

 

codpdpocaa
maamum

1b

15

 

 

 

mcecxpdcn-

 

 

wcaanaam
paou

 

Oflflogom
choc uceceua

.38

13

17

11

 

ovuoflsom
cocoos ampnmu

11

17

 

nowfiocH

 

oo>oouu

 

 

_ m.

 

 

9.

 

 

h

 

One sherd may have more than one motif.

1"

129

Body Decorated Scores, Trench Y2, Eko Abetu

1UU3L52114

882m: '
SAMPLE *

.117

222

11

 

couvavonsm
ncfinam

12

62

12

 

mc£\ccem

 

Aaoo mic—ac:

I. 0.00.!' '1‘.

 

 

oavoasom
choc concurs

 

11“.

cfiucHzom
Q¢UOO3 60:60

 

coaqocm ocuq1h>c3

 

conuocn

 

cabooao

 

 

 

 

 

VI. Ill... 10

 

 

 

* One sherd may have more than one motlf

130

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

«a ma am mm m H ...c.
. Tn a... o.. o. w .... m...“ . 3.. n. . me. o. . ...» 2.23.. ...
«a p ”r c:. o... m. 4.. m.- wn m.. ~.o~ - n.. u.~ a oo. ..o n ..n
...... c... .n 8 .31 ... .. ....c... .
mo 9... a... a. we. ...... em 2. c5. 8 o. a. 8. a... m C» 8.28.. ._
... .... .... m. .8. «.2 Na 3. «A... 2 2.. o... ... ... ..o m ..a
N. ... S. 8 4 a. .28
.2. .9 8. «.mm 3 mm. 9.8:... mm. flow 3 ...... a... n S. o6 c 2» 3.98.. ._
.c a.. a... o. R. ..R ~n ... cé. a... m. .... N ...... ..n o ....
m... H «w .4 a .. 7.8
a... 3 3. m... a. S. a... on ...... m... a me. .... a me. e... u .2 e323. ..
... 8. ...e. a. .o. o... R ..n. ...! a... 2. m... ... ... a.» . ...
8. ._~. 3 a :1 .... ...c...
S... m... we. ca m 36 ..o. .. Penn ...m . 4?. a. ... SA 9... o ..2. 9:27.. ._
.a 3. To. m. 86 a... an m... m... 2 an. .... w 3.. .... o ..-
e... an 2 mm a ... .38.
am... am no. a... m. 3. c6. ... 3. ...o. ... 2. e; « .... a.“ o .98....183 .—
.. .c. c... a. 1 c... 09.. en 8. m... an 3. n... ... ..m. ..o c ..n
... n... n we a... ..r . 5 .36..
36. an ...... To ._ ....” 8. a 5. m5 n. a... ..a m S. .... a m... m... a «a. 3.3.... m
3 8. cc. .... n; n; o .o. ...: ..n ..m. e.g. ...... 9.. ...e ... 3.. ...o. c ..-
__w. 3 . em «a 0 fl! .3...
3.... 2. u...» ca. ... an. S. . 2.. ..2 an an; «.m. .. ... .... . . a... e.g a .3 3:33 m
... .... a... .... m. m. 1 S. a... .... Q... n... - 2. .... ~ 2. n... o ..n
a. Q a mm a... ..F . .. .39.. .
“e.g ... ...... ...... a... 2..” .. ... .5... ..n. .. ... n. . R. T. ... ..m. «A m 9- 2.2.... m
... ..n mam m. n.. T. o 3.. 99. .3 0A «.m. 2 n.. u... N 9... n... u :-
cR an 1. .... mm :1 a... .28
an. a... o2: a. a... m... a 3.. 32. 3 SJ 3.. p 0.... m... a up... «A. 2 2a
3.: ... 3.. ma. .... a... a... .. 3.... e.g R 9...... v.9. «n mm... _.... n 2... To. u ..n 2.2.; a
c2525 .33 e. ... c. ... .. ca ... .. c. .. ... c. .. .. e. L e. c.
L206... Fauna: “7.6: matte... ......c... «7.8.: ..
.3333...» $9551.; .3026: «Son qua-.39 0:026: coo) .u ...-«05. v: .95
space oxm .mououm

noumuoooo xnom now mco.umc.sumumo mumsam1.zo "ma mummy

 

 

 

 

 

 

 

 

 

 

1.311

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

OWN mm _ w -
Nc.N. .. N.. N.. o N. N. N.N . o ..JN n MN .=.oe
. . . . on o N m. m. . N.n N. . NN. .. .
.. N .. 2 . .... .
36w mm .m... o.a «N «.N ~.N 2..— o. w w HM ~52.
o o o .0 mm ON PPom— Won ON Pom-w o d
“Mm cN . .N mu N. N N a me. m.oN w“ Nm.e n.o. N Na. N.. N :N. m.u. WN mm“ N..»...N m
3.3 ..N ...N "was ...N ...N 2.. EN. 3... m a ..N ......
. . . . . ..m a... N.. 3. ...N N ...N ..N.
... .. .. .. ... ... ... .... W. ...; .... .... .. N. 5:.
Na... a; N..N ma.N N. o.. o.. . . . 4.. m. ...o»
. . . . . m . «N . N.wN oN mo.m ..a a o.. o.. N N . .
”M" 3 o N.. n... . ... ..N “n N... ...... WM N..... ...: N m. N.N N w. 3N N M” .52....m
.... mN ..m. SA m w. u. o... . _.. .m . ..N 3°...
. . . . . N .. .. 8. N.. . ....N m. N NN. N. . N
ao.NN mm Nn.N "..o. m. . . . .N . cm. .=.oN
. . . o m n oN N. N..c. e.g N. nN. ... N mo. a.
. .... .. H. .. N .. .. ... N .. ...... ......
Choc? mm ~00 .won- 9— am. Now 0 o F Am FF Panda?
. . . . . N NN N N N. n. .o.N a.N o.. co. N ... N. N
Wm... N.. w. .. Mm . ... N .. m N..... N..: m. N.. .... N .... ... N Mm. N....N MN Mm ... .3. m
....» NN N. n... ... R. N.. . n... n.N..m.w.n. RA ..w W” 8. .... .q .m m cm 2...... N
- O o O o o o o OOF 0
mm” .o .N .N my N N n u .N . N 00 mm oN. m.o.N co. N.. m NN. «.m. MN mm“ a . «cu. m
wee . ..— c.N~ Ta 4... R. n; m . . ._ m 7:...
. . . N m a cN N mm. o.. an.. a. N u. . . N . ..
.9 a.” N R .... N..“. on .. ...“. o... N.. ... ..., N.. .... N ...m. .....NN WN M” g . ...... m
Ow “ I O“ IIIIII Oh oh UN , oh . 6h
ootoopu «x ~38. «A9795. 0..— ou ”MA: 9.3 rnomlcnb ~7u$b at ..n “fly on cumwaouuauv. o..— o.—
..a:
a... .
.3332.)— _ ugh—)3..." ewnmaaqou owwmmsw" ...-«2; m «.2595 on 2.
bl

 

.fl.u:00. ma wand?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

132

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_. ., N. .52. m
.m. a. . _ . n . . . NNx .oa.N.ouN m
.9 . ... . N. Nq .... N.N. .. 8. ..N v N . N. N u.” ..N..m._m . ..NN M
mh.u MM mm. .0.”— H“— M.” O.“ — Own. 4..“ on 8. E M. mac 0 N—N _0 p.509 . m
J. r . m u
c N N . . . . . N.. . . :N» oNNNNuu.
.N _ .. % .. .. 8. N.. N N. .... . .N. ... . w. .... "N .... .
N..... : a. . m w . N.. N 8. N.N N ... .N.. . N.. . N.N .... .50...
pu— — 8. ......N _N M9 N M” .—N o o —N GP. v.” ”m .MNF COJOUflbK m
...... Na . .. N.N ....N NN a... N.N._,.N N . N.. ... ....N ....
an mm m... o.- ~— N... M. m .6 m.m— p N.» .6 NI a..." N. m N. 7.9—.
..— n... mé. .3 q. a a mm. N... .\ .... _ .N Tu We “wmu 039...... m
m...» we . . . mm; wAON m: mm. ....mn es Nu. m r s N. .N .m A;
...... ..N S... N.NN N. N.. m.. .. N.N N.N. N ..N a... . N.N N. N N N N .... .32.
TN; N.Op «on: an M m m H“ Op .. . o J :1 ncm .m pm? Goaounafl m
. a. . w. No. N N... oN ...N N.N N .y: W. N N No. Mug. N N..
.3... ...." NNHM FM" .Nm ...... N. .N N.. N ...N. N .N.N .... ... No. _ . .N. .. ....NN N
—. p L N Ow o o '0.- “ q—NK EOdN—UOU‘ .
- mo NN . .. N.. N. . N.. .m 1 N ac. ..
Nu... m NN ..n Nno. m nx. N.. N m“ m.w. N no. .N.. . «a. ...N N N; N w m. ”page
. .. N.N o . _N N m 4N _ c. . . a ma. N.N a NNN ....oonog m
8. .N N.N ...: N. .... ...: c 8. _N N N . N.N .. N... .
N..... NN TN. NM ”M N.. N. N N.. ...N . m... N .... . 2. .... .... 8 .... . ......h
.4 . . oN .e . . . . . . NNx no.9oauz m
...... N New ....m" “N ...u. ....N . .N.. N..: .N _ N.. N.. .. 8. ...NMI ... N. .5...
.. ..N.- .. .. . . N. . . .
4.. N.. . . N N.N _o..N mm. N. . N.N _.._ N... _N.. . _ .o. .N... m . Mm" “...... N. N
NZNN mm... m... min .... 5 m.. . .m N.. .o. N u m. N n... _. m 0.. 6.. N no ..in can
. . ... Us. w... ..n ....N 0»..— ML N on .... ..N .... 0.: o... a... ch u o “ :nc
.. am... .. ...... . .3”... ...... ...... .... _. .
cote-no «x .72... w .. Sac—:0..— .ynou taco—3...: ‘ uu-qocn m 35.3. “ ...
_ cozzoga 995333..» c3323 .30.... u . . ?

 

 

 

 

 

Ac.u:oov ma wands

1.313

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NNN NN N NN. .N a. NN. ....N .
NN.NN .NN on. N.NN No N.. N.. o N.. ..... N.. No. N.N; N. N. ;.N N N.N o.N. N NNN .....NNN N
NN N;.N N.N N N.N. N. N Na. N.N. N. Na. N.N N N.a N. N N.NN N.. N NNN
no N. N NNI a. 4“ .2 ....N
N.N a. No. N.N N. N. N. o No. o.N. oN N.. e.g N No. N.N N N... e.g N .NN .....ooN N
NN No.. N.N .4 No. N. N MN. N.N. N. N.. N.N N 0.. N.. N No.. N.N N NNN
oN .N N :N N a N. ....e
No.N NN .a. NN.N N cc. N. o.. a. .. oN. N.N . o.. N.. N cc. N.N N 4N. .....oNN N
MN NN. NN.N a NN. ... N. No. .. N. a». N.N N No. N.N N .c. Nu». N NNN
NN N. N on. NN _ a N. ....N .
NN.N. NN on. N... N. N.. N.. o N.. N.N. N. 4N. N.N. N. ... ”N.N N Na. N.o. o NNN N.NNuuuN N
NN NN.. o.N a N.N N. N NN. .... N. N... ..N _N MN. N.. N . N... N.N N NNx
.... oN . o; N. N _ N .50... f
N.... .. 44.4 N.N N N.. N.. o NN.. N.N N N.. .. . N.N N. . _ oo. ... _. .NN ....NNN N
NN NN. N.N. :. No. N. . .N. N.;N on No. N.N. .. NN. N.. . co. N.N N .Nx
on. NN . NN .N . .N. ...oe
N...» N.. 8. N.N. N. ... ... NN. o..N ..N NN.. N..: N ... ... N.. ..N N NNN T233 N
NN oo. N.N.... N. N. . NN. N.aN on NN.. N.N. .. N. N. . .N. N.N a .Nx
:cN .NN . NN. NN. N _ ;. ..NoN
NN.N. .NN NN. N.NN NN N. N. o 8. .2. N.. N.. N.N; .. .... N.NN NN.. N.N. N NNN T383. N
NN NN. N.N. a. N.N N. . go. N.NN an N.. N.N. .. N.. N. . NN.N _N.N N .Nx
.N. .NN . NN. oN .N“ N. ....N
NN.N ca NN. N.o..N. N. N. NN. o.NN oN co. N.N N NN. ... N co. ..N N .NN N.NN-on. N
NN N.. N.N._.. NN. N. . No. N.:N an no. o.N. .. NN. N.. . cc. N.N N .Nx .
NN .N. . N: N. a. ....N
N.N NN oo. N.N N N. N. o N.. N.N. .. NN.. N.N . N.N N. N NN.. N.N N .NN N..N...N N
nN co. .... a. N.. N. . No. N.NN on no. N.N .. No. N.N . NN. ..o. o .NN
NN. NN . NN JNN N. ....N
NN.N NN an. N.N..N. a. a. o NN.N N.NN N. NN.. N.N. N. NN. N.. N ... ..N a NNN
NN NN. ..N. ..N N. N. . N.. .....R on ma; ...N. .. N..? N..“. N..... a... .o .Nx 8.38.. N
..N N.N PNM‘ ..N oh 0... oh N ..N oh N.N uh on on ..N ea c... ..N .0..—
Notte. .33 N.....o: _ $7.70.. N.....oN: NN.N...»N. i NN.N-oN. _N.o...o... N..... ....
N . _ . N . ..
N . .....aoz o....ao¢ o...o: a pa No
:935055 N b95\paam Boo coon—)9 0003 .u n _ o _

 

 

 

 

 

 

AG.UCOUV mH mdm49

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

134

 

 

 

 

 

 

 

 

 

 

mm _ , N.. _ u N :26.
Nw.m .. . oN.N _ _ Na. oN. a. a. . Na. ..N . :N» . ug.n.uo<
mN - no. . _ :4. co. _n.. o.N N on. a.; » sNN
o” _ . ...—n _ N . Z 75o...
mm.m. me u 09. _ so. . mm.. ..m. =.. o eo._ N.N m nN» No.99nos
mN . .o. . cc. a Nc.n a.m m. N o.n c.n o 4Nx
3N _ a . N. "5°...
No.cn .nN . mo. No. m on. =.N m.n N m ..N e.c » NNN c..o.a.x
“N :m. _ n.. n N.N m.; tn. N . o.,N ... w :NN
mm a " Nfi Edda
ma.“ o; No. _ oo. n.oN «a. m.» o.N N m N». N.N .m 1.N» o..N.uu<
mN ... _. .o.. m.o. mo.. a.m .m.. N N4.. N.N .o .ng
2. .. fl 2......
N.m .. oN.N ” NN. N.N m... N.N a. . ... 4.. . :NN c..N.ooN
mm m.. . no. e.g. . NN. ..up m.N N no. m.N a .Na
4N. m N . ...oe .
oo.m no . .n. w on. m.nN No. n.e. o.. a Na. a.» m N» oo.N.uu- .
N.. n R. _ 8. .3... N... e...” N. N . a. ..o a n2 _
omN . d a. afll.c.oe m
......N :u. . 8. 3. ..mo. an. 19.. ..n N . ...”. .... w N» 2:82;:
mm No. mm.n muoN Nn.. N.N. N. o. N ” mo.m N.N o . Nx
No. . 4 _ a» .a.oe _
mm.N a; co. 4N. m.w. em. w...mm N.. N _ ... N.“ m .NN N....ouN .
mm 9: . R. aéN ofl néfi: %N N. :. .N a Na _
. =0 [I‘M .0, a aw ...09
.3... “N.N 2. __ w E. N.N 22 n... .« ... N 2. .... e .5 . 3:8:
an N? , a? 9N. NW. oauN. ad N «T a; a Cu .
e. . N a n 9 23a-
an.o. .. ma.“ 4. _4. a an. m.n ... ;.. w. N. . nN. n.N N 4N» . ...Nouo<
mm of. wt _%.N N? £2. m9 msmm NN N “Q a; a fix .
so, . N mmN N .m4_..oa
N4.n. mw . aN. n.. _n.. o No. o «N _ no. mn¢ moN N. o :o.. ..N m nN» ...o.n.z
an . .n:. .;.N .N. N Np. .. M o».— “N.N m N.. N n:.n N.4 N NNx
. .N .N ..N c an on u. .N «N “on .N on .N pm as
o.»u..no ..uoe ..A.u.cav WAcuncu. _ ".on.cuv V which. _ . nucuv NANN-939 b _ _ L
N ‘- 0....aam - z . _ -
u _ codaaaofifi no.5}3m choc-33.» coo... .u ...-«..N; 9.2.2.5 _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ac.u:oo. ma mama?

 

 

1135

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. op _e. _ “a u —.N _ : _ N ~=.o."

N..N. .. 44.; u.N w _ «N.. n.a N _ NN.N o.n _. ..o ... . me. _a. . 3N» ”cocoon.“ a
mo _N . M.m_ N. NN. o.mN =N ” 4N. m.~..cN m. o. u. on. ..w m an» .
NJN mw . N... «N4 .1 u N .caoN M

N~.m, .. o.n N.N u m o.N m.m N m N“. ..N _. ..o .. . mm.. . . 4N» "..¢o.q.¢= a
.nN N.. N.4u.Nm m ow. w.... m... me. =.;4_oa _ NN. N.N 1 mo. .¢ m 9.x.
mo“ am ” ta." mg m m m ..Namr.9N

aw.» me o.. N.w. N, h ma. _;..n NN . NN.N a.aN N a. g. . .m.N ";.N m «N». o..g..c« a
:N NN. was m 2. mg: ...... s. N..; .... N.. m; N :. ...» “LN.” ,
mm mo. m . NN “a. . w e am«.a«.

mn.N .. oN.N n.n .o No.N_..a N ma. 0.. ... m. m. w. ca. _... . .:N» .....oua a
n: ma. e.g..N._ . NN. .m.N. oN o.. ,..m .N ;.N awm[N co. .m.a .m ..N» . .
my. «JNN. . .c; _ .NN N . .a.w_e.op.

an.m mo NN. a.n..N.. . oo. o.NN N mo. e.g. oN m—.. w... a no. MN.m ”m _NN» “...“..oc 4
... a. 12%.; 8. To.“ ..N 3. N.N. N ... .... .N E. N... W "..N»
NNN .daN_ mm. «um am « _N.M.a.op

8.: RN 8. N..:Nu 8. N..: m: 8. m4... 2; .m. nAWN no; ”..N—w NN» 9:23.. ._
ma no. puu,nw_ . :4. N.NN oN .o. a.a m «N.N e. ”N o¢.m e.. mm .N» .

.N .N .ou ch .N on .N am on cu «a on .N on 0N own es en
0.70.3 m wackuom: Naohlou: TthL: “sauce“; «AfiMusC. a
oatmuéo an ~33. . r . ....
. . . uuaunaom «uomanm a a
n m 5:325.— _ L95\£3m v.30 e553. «.9... .u _ c032; a: on:
o

Ao.u:oo. ma mqmdk

136

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mN. m N N N .- NN u 8»
om. 9 N 4 N z N N ” NNN
NN.N o. N N 4 N m NNx
me. N u N m M NNx
ohm. m N d . NNx
1431. N N m N SN
8; \ NN N N .. a .. NN NS
8. N N m N N a N m IN
as..." asao 2:85 2:25 2.63 2.85 2.63 2.5.: A- .23 $3: _
1.253 332. No.5. o 05 No.5. :5. n 25. o 2:9 N neg. < 25:

 

suond ONE sue—sou Hommo> unnmocomNo NwH mqmfir

137

 

 

 

 

 

 

 

 

 

0) q- . . a '- '-
X >. m x x >. m .
1°" ' y r
L.’ _.1 i1
09.. ' 0.] I
l .87 I
up I! .82 .
' "81—1.
084-
-78
afl-

 

FIGURE 31: Dendrogram Cluster for Jaccard Coefficient on Rim
Nominal Variables, Eko Abetu

138

FIGURE 32: Schematic Representation of Decorative Techniques
Key:
1. Grooved
2. Slashed
3. stab/Drag
4. Punctate
5. wavy-line
6. Carved roulette
a. Diagonal lines
D. Zigzag
c. Diagonal lines/zigzap
d. Basket work/vertical lines
e. Vertical lines/V-shaped
7. Frond impressed

8. Twisted cord roulette

140

 

 

 

 

 

 

 

~31 . x2 Y2
Level 1 Level 1
~Level 1 Level 2 Level 2
Building Building
1 ” Gap.
Level 2 '~ Disturbance.
Level 3
Level 3 Level 3
_ Crude pottery Heartkmaterial
14: ' e 0
Level 4 ' _ Level 4 32 I 100 A D
- Gun flints. '1
Chart.

 

 

FIGURE 33: Correlated Stratigraphy, Eko Abetu

142

FIGURE 35: Level 112 Rim Cross Sections, Eko Abetu

143

sauna oxm .ueouuuom «mono.edm NHN No>oa .NN mmseN
. .N

Eon N o

 

 

 

 

 

144

“\
-~\~

 

FIGURE 37: Level 121 Rim Cross Sections, Eko Abetu

145

 

3 5 :77 D‘
If // K t ‘ W
/ / ° ., ° .

0 1 2cm

FIGURE 38: Level 122 Rim Cross Sections, Eko Abetu

146

 

d
’\
/ ?\
/
I ( [\V‘R /
< \9 f / t
h b\

/
x 1" {O V
\ / '
/.
’ i
1

’ /

FIGURE 39: Level 123 Rim Cross Sections, Eko Abetu

147

2cm

 

 

 

 

Eko Abetu

Level 133 Rim Cross Sections,

FIGURE 40:

149

FIGURE 42: Pottery Types, Eko Abetu

Type Label:

A.

B.

C.

D.

E.

F.

G.

H.

Water Pot

Narrow Neck Pot
Ledge Bowl
Inturned Rim Bowl
Hemispherical Bowl
Shouldered Bowl
Globular Bowl

Flat Rimmed Heavy Bowl

 

 

 

 

 

3:” :3;
:0 ’.| ' I
o':;/_’ D

150

 

 

151

Scores for rim attributes are presented in Table (6-8)
and for comparison between excavation units, this nominal
data was transformed applying Jaccard’s coefficient.(9)
Jaccard, which is given by the following convention, from a 2
by 2 contingency table

SJ = a
a + b + c

expresses a similarity between paired components based on a
comparable pair of attribute scores. "a" represents co-joint
presences, and it is expressed as a proportion of the sum of
all presences, ignoring co-joint absences. Jaccard’s
coefficient lies between 0 and 1, (Doran and Hodson 1975:140-
142), where the lower end of the scale expresses maximum
dissimilarity, while 1 indicates maximum similarity. Jaccard
matrix is presented in Table 9 and this is clustered by the
average linkage method(10) (see Figure 31). The cluster
shows a clear distinction between the lower levels and the
upper levels of the trenches (which appear to cluster on
opposite poles), confirming the basic nature of the
excavation deposits. Beyond this, very few other inferences
could be drawn at this point.

The T-test, a powerful test for interval scale
variables approaches normal distribution for large samples
and corrects for the bias in the deviation of small
samples.(1l) The Null hypothesis for the paired comparison

of sample means is set to determine (a) Ho: that differences

152

between means, due to chance and (b) Hi: that differences

between sample means are not due to chance, or that there is

significant patterning in the materials under observation.
The t-test formula for the estimation of mean

differences is given by:

GE.- fz) - (m- )21)

 

t = .1.
[5" + s:
n: “z
where i represents sample mean

,D population mean
S2 "pooled estimate“ of variance between means.

n number of values in a sample.

S = Lx.- 2}) + (33' 3&2

n. + 112+

 

where:
n\+ n1+ 2 = degrees of freedom (df)
In the t-test formula above, the Ho Null assumption
has it that population means are equal or that,fl = ,D, such

that the formula becomes

 

(§:“§i) ’ 0
Sf + _§;
.3: n;

The alternative hypothesis (Hi) is that‘p f ,H

T-test determinations which are presented in tables 10 and
11, show significant patterning between the pairs of 111/123,
112/123, 112/132, when alpha level is set at 0.01 (two-
tailed).

153

Although, no distinction was made between pot and bowl
rims in the t-test determinations, there seems to be a
general tendency for earlier or older materials to be thicker
and more crudely made. Age therefore, could partly account
for these differences. This should not also preclude
acitivity variations of a spatial character, that may have
required the discrimination between one kind of pot and
another. In some of the levels, as would be presented later,
there is sometimes a marked differential frequency between

pot and bowl vessels.

5.22WMM
For the analysis of the decorated body sherds, a
separate attribute list was segmented from the one already
used for rims, as follows:
WMMWMM-
Grooved
Incised
Wavy-line
Carved wooden roulette
Twisted cord roulette
Stab/drag
Punctation.
This attribute list was decided upon for convenience and
the fact that though a great deal of variation may occur

within such groups as stab and drag, and punctation, these

154

variations could not be objectively isolated. The issue of
the stab and drag has been mentioned in connection with rim
attributes, and as would be seen in table for attribute
frequencies of decorated body sherds (Table 12 - 14), stab
and drag decoration is minimal on the body of pots.
Punctation on the other hand is widely applied to the body of
pots, and bowls, particularly towards the basal portions.
The impression sometimes appear circular, lunates, oval,
dotted etc. (Plate 5 - 9). What is common in all however, is
that even though they may appear singly, or in groups, they
appear to have been done with a single tool as distinct from
punctates done with a comb for example. There was no
particular case however, where there was doubt as to whether
the punctate was done with a single tool or a comb. Comb
appears to be absent. If such a case had existed,. it would
have probably been more convenient to group them together
since the idea of punctation is the same. But when a comb is
used for stab and drag, the idea is different. The point
does not contradict the difference between grooved/incised
and carved wooden roulette/twisted cord roulette for example.

Carved wooden roulette and twisted cord roulette
can be conceptually separated. As was indicated by
informants in 1982, at the Use pottery village near Benin,
women potters cannot make the carved wood themselves. It
requires special skill akin to stone tool making, to be able

to carve the patterns on the wooden object (12). Only men

155

have this skill and this is an example of a point at which
male contribution to the pottery process can be recognized on
the pot. Women could make the twisted cord by themselves,
since women were often better at plaiting fibres together.
Yet, within these decorative techniques, there is a great
deal of variation, depending on the kind of fibre and the
particular inclination of the wood carver for example (see
Plates 5-9).

The point of convenience as made above, is related to
the method that was to be adopted for the transformation of
the nominal scores arising from these attributes. The X2
statistic was decided upon. It is simple and convenient, but
become "cumbersome as the number of variables increases"
(Doran and Hodson 1975:55). Like the t-test, the chi-square
provide a method by which relationships or the lack of it
'between variables can be judged with some degree of
confidence, following the works of Spaulding (1953). And
like the t-test, the chi-square has been applied to the
components from the excavation levels which
are treated as units. As a non-parametric determination, the
chi-square statistic does not assume like the t-test
statistic, a normal population distribution to which the

sample mean may fit.(13)

156

The chi-square statistic is given by the following

convention:
2
x2 =£(o - E)
E

or, the sum of observed frequency (0), minus the Expected
frequency (E), squared, divided by the Expected (E).
Comparing each level with every other level, at the Eko Abetu
site, a total of 55 chi-squares were computed as is shown in
table 15, with chi-square critical set at 0.05. Degres of
Freedom (df), is given by (row - l)(column - l). The result
turns out to be very interesting and can be summarized as

follows:

1. Trench B1 levels show marked disagreement between each
other, partly suggesting a well stratified sequence. One can
therefore say with some degree of confidence that the Bl
levels are chronologically ordered and that this may relate
to distinct historical formations. This agrees with the

excavation information.

2. Across the X2 levels, there is no significant difference
between the components. The near-by trench Y2 however, does
show some significant disagreement between 131/132, 132/134
133/134. These two trenches as was revealed above are
located within an area where building processes may have led
to the disturbance of the deposits. This is probably what is

reflected in the chi-square determinations. The

157

identification of the floor levels in trench Y2 and the fact
that these appear to securely seal a rubbish or hearth
provenance within level 3 below, appear to indicate that
disturbance across the levels was only partial in Y2. X2,
therefore appears to have been more disturbed. The markedly
large sample of level 132 could also partly explain the
differences involving 132. Between trenches X2 and Y2,
differences between the levels are at best sporadic related
in general to disturbance, a more stratified sequence in Y2

and the large level 132 sample.

3. Levels 112 and 113 do not appear to go with each other
and with every other level in the site except with a few
exceptions, namely: 112/121, 112/122, 112/124 and 113/124.
Given the disturbance in X2, one could yet say with
confidence that the two lower levels of trench B1 are
markedly different from the upper level of B1 and the upper
and lower levels of X2 and Y2.

4. Level 111 does show agreement (with some exceptions) with
X2 and Y2 levels. This could mean that the last occupation
(building phase of Y2 and X2 or the upper levels ) is
contemporary, or at least more closely related to 111 and
that relationship between 111 and the lower levels of X2 and
Y2 could have been indirect through disturbance or other

sample characteristics.

158

5. These correlations, therefore suggest at least three
occupation phases for Eko Abetu. An earlier phase
represented by the lower levels (3 and 4) of X2 and Y2. The
fact that levels 112 and 113 (together), do not agree with
this early occupation, and the later and more recent levels,
suggests a historical displacement of activities from X2 and
Y2 to B1 location, during the deposition of 113 and 112.
This ordered sequence has been graphically represented as in
figure 33.

Some questions would however be raised as to the way in
which this result would be affected had the bowl and pot rims
been treated separately. The materials have therefore been

sorted in a qualitative way addressed to a formal analysis.

53mm

It was necessary to produce a general outline .of the
pottery forms and types from the northwest which can be used
for comparative purposes. Cross sections of the diagnostic
rim sherds have therefore been illustrated in figures 34 to
41. The pottery forms have been group into 8 types as

follows : (see Figure 42).

Ing A: Represented generally by everted rim sherds (Figure
34f; 35e, i, m, 0; 36a, c; 38a, b, e; 39b, c, J; 41a, c, k,
m, o, p). Variation in this form relates to size (height,
volume etc) and the nature of the everted rim. Sometimes,

the evertion may be very slight or the rim itself may be

159

slight or very angular and a tendency to an expanded or more
voluminous body (Figure 39a) (Connah 1975 :122-124). This
type, reconstructed in figure 42A is commonly decorated with
twisted cord roulette followed by carved wooden roulette,
below which may be various forms of punctation. The lip may
have grooving. From personal knowledge, this type was used

as a water pot for storage until recently.

Ing 3: Only one sherd of this type occurred in the sample,
(Figure 41i), and cannot be mistaken because of it’s long and
narrow neck. The form probably looked like figure 42B and
may have served as a handy flask for taking water to the

farm. No decoration is present on the lone rim.

Ing 9: This is a ledge bowl. It is for general purpose,
with a ledge, carination or ridge around it’s greatest
circumference. The height of the ledge may be much varied,
associated with grooved patterns, stab/drag and slashed
decorations (Figures 34g, r, u, w; 41f, h, e). see Figure

420, The rims are uncurving or constricted.

Ing D: The type resembles type C and it is also general
purpose bowl. It may contrast with type C because of a low
ridge which may be very reduced or absent. The pot generally
looks smoother or more refined (Figures 34a, g, h, l, m, o;

35c, h, j, k; 40b; 41b) Figure 42D,

160

Ing E: This is a popular grooved bowl, and may be very
large or deep for storage or general purposes. This may also
be called a hemispherical bowl. (Figure 341, n, t, w, x;
35q, v, d; 40a; 411, q, r). see Figure 42E.

1229 E: This is a shouldered bowl for cooking soup. Connah
(1975 :117-118) reports that this type also used to be placed
in the bottom of house walls as foundation deposits. The
bowls are also commonly used in shrines. The bowl features
a sharp or angular shoulder, from where the rim leans inwards
(Figure 42F). Unless care is taken, the rim may be confused
for a necked pot (Figure 34b, c; 41d).

Ing Q: A globular little bowl, with a generally smooth
curvature and slightly constricted mouth. This was a secure
dish for picking seeds and grains by the women and also used
for washing the hands or even for serving food (Figure 34p;
35u; 36f, g; 41f) (also see Figure 426). No decoration was

observed.

122: E: This is a coarse flat rimmed heavy bowl, which is
probably very ancient. Connah (1975 :117), reports three
possible uses for it: For washing hands; as a lamp container
on the shrine of the god of medicine; and as container for
wax by brass casters. (Figure 35n; 39a, 1) (also see Figure

423). No decoration was observed.

161

The vessel count from these types have been scored in
table 16. The following points emerge from the frequencies

of the types in the levels:

1. The thick walled type B vessels appear in level 3 of X2
and level 2 of B1, which may indicate more archaic occupation
of these levels. In the Benin City center, this vessel has
also been recognized as an archaic form (Connah 1975 :117,

196. Form F6).

2. Type A water pot is almost absent in 111. Pct/bowl ratio
for this level is .09. On the other hand, the frequency of
pot/bowl in 123 is comparatively high (2.33), and less so in
112, where pot/bowl ratio is 1.20. The remaining levels show
a generally lower pot/bowl ratio. The ratio of .09 in level
111 is markedly low. The sample could be a problem in
general, but since the use of these pots are
sometimes differentiated, activity differences may be
suggested. However, in association with stratigraphic
information, statistical results, and design attributes on
pottery, presented above, there is a clear distinction
between the upper and lower levels of trenches X2 and Y2, and

the way they relate to levels in trench Bl.

162

5-4 Dalian

The following date was recieved from BETA ANALYTIC INC.
Coral Gambles, Florida, USA (14).

Lab Number Sample Number Carbon-14 Age A.D.

Years B.P. i 100
Beta-17078 EA/Y2/4 470 i 100 B.P. 1480 1 100
1/20/86

The calibrated figure for this date with 95 per cent
confidence is A.D. 1432 (range 1280 - 1650) (Stuiver and
Becker 1986:863-910).

Specimen EA/Y2/4 consists of charcoal from the Hearth
material on top of level 4 of trench Y2, Eko Abetu site. The
above date was supplied with the following information:

the charcoal was pretreated by first picking out

any rootlets, and then given our series of acid-

alkali-acid soakings to eliminate carbonates and

humic acid. The following benzene synthesis and
counting of the radio-carbon activity with a liquid
scintillation spectrometer proceeded normally.

Although one should not put too much weight on a single
date, it however reinforces the correlation results of the
excavation levels which suggest 3 phases of occupation of
the Eko Abetu site. A date of 500 years ago could mean that
a 100 year time period could represent a phase of occupation,
and this is probably consistent with the cycle of birth and
decay of mud wall villages in the forest region of West
Africa. Also, the date is quite comparable to dates from

occupation sites in the Benin City center (see chapter II).

163

5.5 Summary.
1. The Eko Abetu pottery has been analysed from both the

quantitative and qualitative designs.

2. . The correlation of T-test comparisons of rim thicknesses
between excavation levels; cluster of Jaccard matrix,
chi-square determinations, carbon 14 date, stratigraphy, and
pottery typology has allowed this study to suggest three
occupation phases for the Eko Abetu site - an early
occupation of crude pottery; a distinct upper levels of most
recent occupation, associated with house construction; and in
between which, is a middle phase occupation, with materials
which do not appear to go either with the lower or upper

phase.

CHAPTER VI

CLASSIEICATION AND ANALYSIS g

Spatial significance 9.: We we t P_____Lotter -
In this chapter, analysis is addressed largely to the

pottery recovered from the second phase of excavation from
the Benin northwest.(8) At the Uwan site, a total of 2842
sherds were recovered from two trenches. As was done in
Chapter V, for the Eko Abetu pottery, the Uwan materials have
been analysed both quantitatively and qualitatively. As will
be shown, the major result of the analysis reveals that there
is a high degree of variability in the northwest Benin
pottery. This variability has both temporal and spatial

significance.

6-1 WM
Qualitative analysis was initially addressed to the

isolation of pot forms, and types from the Uwan excavation.
What was immediately apparent even from the excavation, was
that the mateials were quite different from the Eko Abetu
types. In particular, pottery from the trench 420 was
distinctively crude or coarse, lacking the surface finish or
smoothening of the Eko Abetu forms. At first, it was
suspected that the crude appearance was due to the lateritic
weathering of the deposit, but the form and decoration
analysis showed clearly that other factors could have been
responsible.

164

165

Cross secions of the diagnostic forms have been illustrated
in figures 43 to 50, These forms have been grouped into
types (see Table 17). There are distinctively three new
types but some types are shared with the Eko Abetu site (see
Table 14 and Figure 42). The new types include the coarse
pot (I), flared rim bowl (J), and the short neck bowl (K)
(also see Figure 51). These forms or types and some
decorative attributes they contain have been further

illustrated in the photograph plates as follows:

Plate 10 Level 411 rims

11 412 rims
12 413 rims
13 414 rims
14 421 rims
15 422 rims
16 423 rims
17 424 rims
18 411 body
19 421 body
20 424 body.

Some decorations have been labelled on the plates as follows:

Decoration: a. carved roulette
b. stab/drag

c. frond impressed.

166

The new types can be described as follows:
122; 1: This is a commonly grooved lip, and thick walled
coarse pot. It is most prevalent in trench 420. This pot is
similar to type A of Eko Abetu, in general form, but is so
different in its roughened surface, thick wall and often

grooved lip (see Figure 511).

Ing Q: This is the flared-rim bowl, characterised by a
pronounced rim hanging over a small body (Connah 1975 :118,
and 128). (The type is illustrated in Plate 11, Figure 44,
46, 49). Connah reports that this bowl was used for washing
the hands or eating or also sometimes used to cover
"medicine", when this is set in a hole on the ground within
the house compound. This pot form, which has an external
ridge below the rim was not encountered in the Eko Abetu

site.

Ing K: This is the short neck bowl. Type K neck and rim
are characteristically slight or reduced, almost in opposite
form to the Type J rim. It is illustrated in figure 44K,
Plate 12 and figure 51K.

Materials scored into types D and E (see Table 17) were
quite varied particularly with respect to rim shape (e.g. the
ridges, carination or grooving). E, which is the
hemishperical bowl, was however distinguished from D by the

larger depth of E. At the Uwan site, both appear to have

167

rims that are incurving. Many of the type E also have
straight rims characteristic of the deep hemispherical bowls
in the Eko Abetu site (see Chapter V).

On the whole however, one could see that the pots from
the Uwan site show marked differentiation from the Eko Abetu
site (about 8 kilometers away). The type frequencies scored
in table 17 were further analysed into pot/bowl ratio which
reveal as in the Eko Abetu case, (see table 14), that there
is a general tendency for bowl forms to be better represented
than the pot. Bowls have far more domesitc use than pots in
cooking, serving, washing etc. Where pots show a higher
frequency such as in level 414 and 422, for example, it must
be explained by an alteration in behavior or some kind of
localised activity. Pots were commonly used for storage
(water, seeds etc) or for "heavy duty" cooking (such as
cooking yams or oil palm seeds). Such activities may at times
be communal such as during house building or early farming
activities or also during festivals. The mode of deposition
of refuse can also partly account for higher pot/bowl
differentiation.

With respect to decoration, however, there is a general
similarity between the Uwan and Eko Abetu sites. The
decorations of grooving, incision (often with comb),
stab/drag, twisted cord roulette, punctation, are present.
This does not preclude proportional differences. Some of

these decorations have been ilustrated in Plates 10 to 20,

168

(also see Figure 32 for schematic representation of
decorative techniques). One key difference however, concerns
the presence of the decoration that has been described as
frond impressed. (15). This decoration sometimes looks like
large twisted cord patterns, but apart from it being rather
large, when compared with twisted cord, one could notice
striations in the impressions. These striations are the
negative impressions of palm frond veins (see Plate 18-20 c),
This is probably what some other workers have described as
coarse string roulette. (Darling 1984:223). This decoration
was absent in the Eko Abetu site and it’s presence at the
Uwan site is a major difference because of its frequency.
Some other decorative motifs have been illustrated in the
Plates. The stab/drag for example, which did not come out
very well in the Eko Abetu photographs can be observed in
Plates 10-17 "b“ (The "slashed" form of Eko Abetu appears
absent here). Also see the carved wooden roulette

illustrated (Plate 18a)

169

TABLE 17: Diagnostic Vessel Count, Uwan Site

 

 

Level A E H l G Vessel Count Pot/Bowl Ratio
411 8 1 3 19 0.05

412 8 2 4 28 0.12

H: 9 4 2 23 DJ]

414 1 3 4 3.0

421 24 7 1 36 0.24

422 24 25 59 1.04

423 27 6 2 40 0.17

424 14 a 1 _gg 0.50

tom 233

 

Rim Nominal Scores, Trench 410, Uwan Site

TABLE 18

 

170

TVLOI
uotnemiogiad
saurq Akan

passaidml
DUOJJ

uornenauna
sn153s

anaatnou
P100 93:51n1

auiatnou
paAJnQ

Sutdmcns qmog
'390 PBQSPIS
3810 pus 01235
pasroul

-oaa paaooza

Jorzanxg
paneurzeq

JotzaauI
paneuIJeo

'aaa paqaaea
mtg pauses;
saaoozg

punou

9 39I3

 

167

17 7

28

21

15

5

31

11

101

10

26

l3 l7 1

8

217

12

39 18 7

36

23

9

16 34

3

27

 

Uwan Site

Trench 420,

Rim Nominal Scores,

TABLE 19

 

171

TVLOI
uotnezo;iaa
sautq Akan
passaidmI puozg

notuenauna
sntfias

annatnox
P103 Paasrhl

anuatnoa penzeo
Surdmeas qmoo

'330 pausets

891a pus 01235
pastauI

-aaa paAoozs
iotzanxa peanutzeq
Jorzanul panautzeq
-aaa paqaqen

mtg 93313A3.

drq panoozs

dti punou

I9 39Id

 

233

15

l6

l6 2 36 41 17

ll 53 23

l

205

ll

9 30 53 l

36 30 30

2

218

15

21

9

39 52

ll 32 15 15 7

3

160

16

3

29 36

19

ll

9 33

4

 

TABLE 20: Jaccard Coefficient for Rim Variables, Uwan Site

1372

 

 

411 412 413 414 421 422 423 424 111 112 113
411

412 .94

413 .96 .97

414 .85 .88 .91

427 .98 .98 .98 .93

422 .92 .93 .94 .97 .96

423 .88 .94 .94 .90 .95 -.97

424 .95 .95 .94 .97 .98 .96 .95

:11 .81 .97 .92 .87 .95 .93 .94 .85

112 .84 .88 .86 .96 .88 .95 .84 .92 .92

113 .72 .63 .79 .92 .77 .83 ,.83 .89 .72 .81

 

173

 

mm m a“ m a m a“:
Mfifi Nm on mm m m ma:
omH ea c: ma o“ m o N”:
mow m cm" mm” mm um N mg m "a:
ouauasom
ozaa opuodzom choc ouaoasom .oun

Amway a>m3 ccoum noduauocsm cognate uo>uuo mmua\pmum

womaocm no>oouo .qo>og

ouflm ems: .oav zucoua .mouoom unannouua acom “Hm mgm<a

174

me o: ma : g N N am:

 

baa a: ms nu mm:
:on and odd mm n N m N am:
mom a no um an m n a n «N:
. . oauoasom
on“; ouncaaom vuoo ouuoauom .ouo

dance abs: uzoum codaauocam unawaxs vobumo mmua\naum vouaoza vo>oonu ao>oq

oudm emza .omv succue .mououm cmumuoumo xcom "mm mgm<e

175

 

 

 

 

 

 

NNN 3 N.. NN N. N N
ANN N... N.NN NN NN. NN ..N NN. N.NN NN NN. N.N N N... N.N N N.N N.N NN.N
..N. N... NN NN .N. ..N ..N .N. N..N N. NN. N... o. .. N.N N ...N ...N N N N.NNN:
. NN.N. N 8.- - NN. NN NN N N. N
8 .N. NN. N.N NN ..N NN. N.NN N. N.N N.N. N N.N N.N . NN. ..N. N... N.N N NN. ... N N.NN
NN.N 3. N.. N ..N. NN. ..N. N.. NN. NN. ... N.NN N.. .N. N.NN NN N... N.. N 8. N.N. N. 8. ...N N N NNNN..N
8N N NN. NN. N.. v... NN N N. N
NN. N. N. NN.. N.NN NN ..N. .N NN No. ..NN NN N.NN.N. N. N. N.N N.N NN.. NN.. NN.NN
no: N.. NN.. N N.N NN ..N. ... .N. NN. 8. N..... ..N ...N N.NN NN N.. NN.. N 2. N... N. NN. N... N N N32...
NN.N N NNN NNN N.. NN N N. N
.....N NN. NN. N.N N.. ..N. N.. N... N: N... N.NN NN N.N ...N. N N... N.. N ...N N.N N NN. N.N N NN-NN
no: NN. ... N ..N ...N. ..N. N... ..N. NN. ...N 2.3 N.. N.N N.N. NN .N. N.N N ...N N.N N. N. N.N N N NNNN..N
N N N.N NN. NN. ..N N N. ..
wa 8. .... . ...N .N N. .N. N.NN NN N... ...N NN ...N N.N. N NN.. N.. N NN.N N.N . NN. N.N N 8.:
no: 8. N.. N N..N ...N. ..N. NN. NN. NN. .N. N... N.. 3.. N.NN NN .N. N.N N NN.. .N.... N. N.. N.N N N NNNN..N
N.. N NN. NN. NN NN N N. N
RN N.. N.. NN.. N... N N.N N... N. N.. ..N N N.N N.N N.. N.. NN. N.. . N.N NN. N NN.N.
no: N... N.. N ... N..... ..N. N.. N.NN. NN. 8. N.NN N.. .N. N.NN NN N... N.. N 8. N.N. N. NN. N.. N N N.NN..N
N NN. NN. N.. NN N N. N
N.N NN. NN. No. N.NN NN ...N N.NN ..N No. N.NN NN NN. N... N NN. NN. NN. N.N N N.. N.. NN.N
.N.“ N.. N.. N .N. N..... ..N. NN. NN. NN. N... N... N.. N.. N.NN NN N.. N.. N NN. N.N. N. NN. N.N N N N.NN..N
NNN N NN. NN. 8. NN N N. N. N
..N. N. N. 8. N.NN NN ..N. N..N ..N N.. N.NN N. S. N.N o. N. N. NN. ..N N ...N N.N N NN.N
8.. N.. N.. N 8. ..N. ..N. ... ....N. NN. N. N.NN N.. 8. N.NN NN N.. N.. N .. ..N. N. ... N.. N N N.NN..N
I O O O I O O O I O C O O
0.— oh .— h u u h u L h u u u u on“: on oh ..N $250.30
N..
OHUOQ
... 22?... 3:33 38.. 83:25.. .38 ...-HE 338 2.2:: N..-Nun. N388

 

ouNm can: Nnouoom
coumuoooo xcom now ncoNuchsuouon cumsumnugo "mm mamas

176

 

 

 

 

 

 

 

 

8.: ”N..“ ..N. ..N. ..N N N N N N
N. N.N NN. NN. .N. N.. N.. NN.N NN NN N.. N.N N NN. N.. N NN. N.N N NN. N.. N
N.“ N.N .N NN ... NN ..N N.N N. NN N.N N.N N N. N. N.. N.. N ... N. 5::
.N.: NNN NN. . N NN. NN. NN N. N N N
. N N ..N ..N NN N.. NN NN ... N N N... N.N N NN. N.. N NN. N.. . N.. N.N N N
N.. R R. N.N NN NN N.N NN ..N NN. N.N NN N. N.N N N.. .N.. N.. N.. N N.. N.. N 5::
NN. NN NN NN N N N
N . NN NN. N.N N NN. N. N. No. N.N N N.. N.. N.. NN. . N.. NN. N N
N N N.. .N. NN NN .N. .N ..N N... NN NN NN. N.N N N... N.N N .N N.N N :53
NN. NN NN NN .. N N N
N. NN N.N NN 8 N. NN N. N.N N.N N ..N N.N . NN. N.. N N. N.N N
N. ..N. N NN NN N. .N NN N.. N.N. N. N.. ...N N. N... ..N N N.. ..N N N NNNNN.N
NNN NN NN NN N. N N N
.N.. N.. N. N NN NN. NN NN ... NN NN N.N N.N N.. N.. N.N N.N
N N.. N. ..N NN 8. NN NN NN. NN N ...N N. N.. N.. N N ...N N N NNNNN:
NNN NN. NN. NN N. N . N N N
NNN N.N NN. NNN 8. ... N.. N.. N.NN NN N.N N.N N NN. N.. N N. N.N N ..N N.N N 3.:
..N. ...N NN NN NN. NN NN N.N N. N. N... N.N N. N. N. N.. N.. N N.. N.N N N N923
NNN . . N NN. NN. NN N. N N .. N
NNN NM. N.. N ..N NN NN NN. N.NN NN .N. N.NN NN .N. N... N N. N.. N N. N.N . NN. N.. N NN.N.
..N. N. R ...N NN NN N.. N.NN ..N N... N.N. N. N... N.N. N. ... ... NN.. N.. N N.. ...N N N :53
NN. NN NN NN N. N N N N
NN N.. ...N N N. N.N. N. 8. N.N N ..N ..N N... NN. . NN. N.. N NN.N
..N. ..N. N.NN ..N N.. ..N NN .N. N.N. N. NN. N.N N. 8. ..N N ... ...N N N.N.N...
on 0. NN ..N N. on NN o.— ou 2 N. ..N N. 3 am». up on as vet-008
.N . ....
. w «NN.-N6: l-
vole-No 3N.— bia use... 8339:... :3 ...NNNNF vet-3 .8313 3.33 vol-0.5

 

.N.»:ou. NN mamas

177

 

 

 

 

 

 

 

 

N.N . NN. N... .N N N .. N
N.. NN. NN. NN. ..N NN ..N NN NN ... NN NN N.N N.N ... ... NN. NN. N.. N.. NN.N.
NNN NN. N. . NN. NN NN NN. NN NN NN. NN NN N.. ...N N N. N.. N NN. N. . ... ..N N N 2.53
NNN . .NN 8. .N .. N N N
NNN N. N. NN. NN. NN. N.. 8. N.. NN.N NN NN N.. N.N N NN. ...N N 8. N.. N ... N.N N 8..
NNN N.. N. . NN. 8. NN N.. NN NN ..N NN NN ..N N.N N N.. N.. N 8. ... . N.. N.N N N NNN:Nv
.N. NN NN N. . N N

.N.N. NN NN.N N.NN ..N ..N N.NN N. N.. N.N N N.. NN. . 8. N.N N .N. N.N N
NN N.N N.N. N N.N N... N. NN.N N.N N .N. NN. 8. NN. . N.. N.. N N NNNNNG
NN. NN NN NN . N

2.: N.. NN. NN NN N.. NN NN No. N.NN NN NN. NN. NN.N NN.N
NN NN.N .. N NN. N.N. N. N. NN.N N NN.N NN. . N.N NN. N N NNNNNG
NNN NN. NN. NN N N N N

NN.NN NNN NN. NN. NN. N.. N.. N.. NN. N.NN NN NN. N.N N N... N.. N N.. N.N N NN.. N.N N
NN NN.N ...N. N N.. N.N. N. NN.N N.N N NN. NN. N.. N.. NN.. .NN. . NN.N. NN. N N NNNNNG
NNN . NN NN 3 N N N N

NN.N. NNN NN. N. . NN. NN NN NN. N.NN NN 8. N.NN N .N. N.N N .. N.N N NN. N.. . NN. N.N N
NN .. .. N.N N.N. N N.N N.N. N. N... N.N N ... ... N. N. N.N N. . NN.N ... N N 3.5%
.N. NN NN NN N N N
NN N.N NN ..N N.. N N. N.N N. N NN. N.N . N.. N.. N NN. NN. N

2.3 N.. N.N NN NN ... NN ..N ...N N. NN NN. N.N N NN. N.N N N.. N.. N NNNNNG
NNN NN NN ..N N N

3.3 N.. NN. NN NN NN. NN NN N.. NN N N.N N.N N.. N..

. N.. NN. NN NN N.. NN ..N N... NN NN N.N N.N N N.. N. N N NNNNNQ
2. ..N N. NN NN ..N N. a. I... N.. ..N N. N..
o.— ou on on 3 on .N....
EN”... p 3:16: 2.03
o 33 Fun 392 c.3325. vuou van-NZ. vet-nu .3533“ .033 $2530

 

Ac.ucouv mm mqmde

 

 

178

 

 

 

 

 

 

 

N.. N. NN Nn N N N N

NN ..N ..N NN. NN N. N... N N.. NN. . N.N NN. N N N.NN

NNN an 3 3. Ne 3 ...N N." no. no. on: N~.N n N333

NNN NN. NN. .N . N N N N

NN NN NN NN. N.NN N. N.N N No NN. . NN. N.N NN. N N ...N

3n NNN 3N nu. SN 0: an N.N 3. «.n n N. 3. . «A n N N N323

NN.N «on NNN an n a N

N.. NN NN NN. NN NN N. NN NN. NN. NN. NN. NN. 3.8

«On no. .2 no. NNN oNN «.3 N" on. N. n an. an. N.N n n N «:23

N.NN NNN ..N on N N N N

3 an. an co 3. «N 2 N.. N no. u. N 2. o.~ NN. ... n 8.:

«on NN. no no No. on N... 3 on an. N..- . on. no. «.u N N N .323

N. NN o. N. o. NN ..N NN N. ..N

.3239. Nun stucco

h INN-N .90: 3.9: 60330.5.- 33 NN.N-Nth v2.53 ‘0. N

 

.o.N=ou. NN mamas

179

TABLE 24: Weight of Sherd Analysis, Trench Bl, Eko Abetu

 

 

 

 

 

 

 

 

Level ; Total Sherd gueéggat Weight/Sherd height/cc. x 1o-~.I+
Q : R

1 . 3:5 2 5516 2;»; 17.2

2 f 558 _5§10 10.1 17.7
_ 3 42L ! 122i 42.9.. 4.7

g, ' 6 3 so 8.3 .10

 

 

 

180

TABLE 25: Weight of Sherd Analysis, Trench x2, Eko Abetu

 

3

level; Total Sherd; weight weight/Sherd . weigh cc

 

 

 

 

1 g (3 ~ 2 10’

T I . .
1 1 176 : 1900 . 10.7 5.3
2 f 101. ! auo ' 8.3 - 2,3
3 : 11.3 I 1700 ; 11.8 ; 5.5‘

h i 53 5 670 12.5 s 2,2

 

181

TABLE 26: weight of Sherd Analysis, Trench Y2, Eko Abetu

g

level gTotal Shards. Wei ‘8 ’ Weight/Shani ; k'eithPO-A
- (53h . '

 

 

 

 

l x 10-4
1 i 110 ; 1000 .0 *—
i . ~__ 9 ‘ 3.3
2 i 1421; .3 75.30 17.6 25.0
.3 i 11.0 2180 ' 15.5 . 7.2 ._
u 5 20 3 250 i 12.5 .80

 

 

182

TABLE 27: Weight of Sherd Analysis, Trench 410, Uwan

 

 

Level Total Sherd lelght leigth/Sherd weight/cc x 10"
1 708 7998 11.3 28.5
2 371 4738 12.7 16.9
3 300 3410 11.3 12.1
4 59 1050 17.7 3.7

183

TABLE 28: weight of Sherd Analysis, Trench 420, Uwan

 

 

Level Total Sherd 11111911: neigtn/snara weight/cc x 10"
420
Level

1 490 7150 14.5 . 22.3

2 51s 9930 19.2 31.0

3 298 5540 21.9 20.4

4 167 4450 26.6 13.9

 

@WWW>(
((12811

111W
1181f”'

187

 

It? T)

o 1‘ 2 cm

FIGURE 46: Level 414 Rim Cross Sections, Uwan Site

11111111
«WVVWV1
11283 971
11118111

E 47- Level 421 Rim Cross Sections, Uwan Site

193

 

 

 

 

 

 

   

FIGURE 51: New Pottery Types, Uwan Site
Type Label:
I. Grooved Lip Thick-walled Coarse Pot
J. Flared Rim Bowl

K. Short Neck Bowl

194

 

 

 

 

 

 

113

 

 

V’ (V ‘9
:393NNN333
rm 1: v' '0 u' q' '¢ 1¢ ‘9 .. IF
l 9§1 98 [-93 j
‘ ' .95 .
94 .m3
9 so so -39
4+ .79
I
i
.7“

 

FIGURE 52: Dendrogram Cluster for Jaccard Coefficient,
Uwan Site

195

6.2 Quantitative Analzsigt
On the basis of the qualitative attribute dimensions

discussed above, and following from the attribute list
already generated for the Eko Abetu materials, (see Chapter
V), the following rim attributes were used for quantitative
purpose for the Uwan site pottery:

Flat lip

Round lip

Grooved lip

Everted rim

Hatched decoration

Carinated exterior

Grooved decoration

Incised

Stab/drag

Slashed

Comb stamping

Carved roulette

Twisted cord roulette '

Punctation

Frond impressed

Wavy-lines

Perforation.

Each attribute present on each sherd was scored 1, level

by level, and tables 18 and 19 show the frequency of

attributes for trenches 410 and 420 respectively . This data

196

was transformed with the Jaccard coefficient (see Chapter V).
As a form of control, materials from the trenches 410 and 420
were compared with one another and with materials from trench
B1 ~of the Eko Abetu site. Bl, from Eko Abetu was chosen
because it has the best stratified sequence. Jaccard
coefficients are scored in table 20 and this result was again
clustered by the average cluster method. (see Figure 52),
The result is quite remarkable in that materials from each
trench essentially cluster separately. In a sense, this
reflects what was ordinarily visible (i.e. differentiation).
Apart from the demonstration of this differentiation, it
is important to explain why it exists. Is this
differentiation due to temporal or spatial factors? Which
is the older part of the chain? These are difficult
questions to conclusively answer. An absolute date would
help matters here but comparing the vessel forms on tables 14
and 17, from the Eko Abetu and Uwan sites respectively, one
could say that pottery forms from trench 410 are more similar
to those of Eko Abetu. For example, type C bowl with
pronounced external ridges or carination is absent from the
420 trench but fairly represented in trench 410 of Uwan and
the Eko Abetu site. Even some materials from trench 410 have
surface smoothening like the Eko Abetu pottery, but unlike
the coarse forms from trench 420. This is suggestive that
trench 410 deposit may be younger than 420. (note that the

Eko Abetu time sequence is fairly well established).

197

However, apart from this time dilemma, the differentiation
shown by the northwest materials has other cultural or
behavioral implications. It could reflect behavior addressed
to localised points. This point is reinforced below by the
statistical test results on body decoration.

The second qantitative analysis was performed on the
data for body decorative scores from the trenches (see Tables
21 and 22). The body decorative attributes are as follows:

Grooved decoration
Incised

stab/drag

Carved roulette
Twisted cord roulette
Punctation

Frond impressed
Wavy-lines.

As with the Eko Abetu materials, the data was
transformed with the chi-square statistic, and with alpha set
at 0.05 (see Table 23). Out of the chi-square statistics
computed for 28 pairs of comparison of the excavated levels,
only 7 results show significant similartity. The rest show
significant dissimilarity. This result further illustrates
the general high level of variability in the northwest of

Benin pottery.

198

Looking at the pottery density on the whole, further-
more, one sees that the sites may have also supported varying
degrees of activities in the past. In tables 24 to 28, the
pottery sherds are compared with their weights on one hand,
and weight of sherd is compared with unit volume of space on
the other. Fragmentation could partly account for the weight
per sherd, which says very little culturally, but in the case
of weight of sherd per unit volume, the results reflect a
general tendency for high weight index as one goes up the
levels. This higher index could suggest higher intensity of
occupation, the growth of local population, or possibly other
unknown factors. When put together, these results indicate a
changing pattern of papulation, characteristic of growth, as
well as a general mobility of perhaps a segment of the
population associated with pottery making. One is however
not able to say whether the growth pattern in population may

have been dramatic as in migration for example.

Will...

W191!
In Chapter VII, the biases in methodology and analysis

are initially identified so as to isolate the strengths and
weaknesses of the results. In spite of some inevitable
problems, the point is made that corroborative evidence from
different ~ angles allow the investigator to suggest
interpretive models for the results. This is done by
relating specific results and other evidence to the five

assumptions proposed in Chapter III.

7-1 Bias
The following set of biases can be identified in the

methodology and analysis of the northwest artifacts.

7.11 shamanism
As has been pointed out, the northwest was decided upon

for this research on the basis of previous work in the area
and the existence of a base map (see Figure 14). Secondly,
the choice of particular site location such as Eko Abetu and
Uwan Living site was based on the reconnaissance survey of
1982. The surveys were carried out by methods that might
be considered as subjective. Regional survey data on
archaeological sites do not exist in the area. Sites were
identified on the basis of information provided by local
informants, walking through farm paths and clearings and

199

200

the presence of cultural features and artifacts. This was
the most practical approach, but it is informative to
indicate that methods can be applied which seek to sample and
identify sites objectively, both on a regional basis and for
within site decisions. More objective probabilistic methods
have been applied to wooded conditions (e.g. Lovis 1978) that
are of potential interest to tropical forest environments..
Also, surface distribution of artifacts can be objectively
related to subsurface materials through the systematic

intensive surface collection method (Redman and Watson 1970)

7.12 WWMW

This is a recurrent problem in archaeological
methodology. That is, how much an archaeological sample
represents the archaeological population. It is assumed in
this study, that the assemblage samples represent their
respective populations. This is not necessarily correct.
Depending on the complexity of the cultural environment under
consideration, specific sampling strategies (probabilistic in
nature) can reduce or eliminate skewness in the bias related
to representativeness of archaeological sample. (Also see

Lovis 1976).

Titiwmxatigndsmif
This problem has already been partly answered by saying

that by the experience of the investigator in working on

comparable sites in Nigeria, it was considered appropriate to

201

probe carefully with a trowel. This method would have been
sensitive enough to the mainly expected pottery artifacts.
It can also be added that practical considerations of time,

money and personnel were in general, constraining factors.

7.14 wmmmdnmm.

Arbitrary excavation levels were dug as much as possible
following the shape of the ground surface. Mixture of
materials between the natural layers was however an
expectation, since arbitrary levels may not necessarily
follow the natural profile. The role of analysis is partly
to identify this kind of bias or correct for it. For
example, in the Eko Abetu site, although technically, 11
arbitrary levels were dug in the 3 trenches, the 11 levels
were correlated into 3, based on several analytic deductions
e.g. the stratigraphy, chi-square results, t-test, Jaccard
coefficient, carbon 14 date and vessel typology. Even
disturbance by previous artificial processes can be detected
in an analysis. Notice for example that the disturbance of
pits in the trench 410 of Uwan site, (which were identified
in the first 3 excavations levels) is clearly expressed in
the chi-square results. Chi-square results for levels
411/412, 411/413 and 412/413, all show significant
similarity. If the investigator did not have the independent
stratigraphic information, other explanations (e.g. function

or persistence) for the significant patterning of the levels

202

might have been suggested. The point however, is that the

arbitrary levels were not treated in isolation.

7.15 Over-representation of attribute units

There can be low or over representation of artifact or
attribute units within a sample or an assemblage. An example
of over-representation is a condition where a' large pot
breaks into several small sherds. It’s design attributes
would then tend to be over-represented (Shapiro 1984:700-
701), especially when sherds are the units of analysis as has
been done in this study. This bias could be corrected by
working with percentages. Then, the mathematical
transformations the investigator has carried out on nominal
data may not be applicable, but since this problem could be
assumed to exist across the levels, it could be ignored, as

the investigator has done in this research.

7.16 Vessel count

In establishing the vessel count in an assemblage, this
invesigator usually wanted to establish the minimum vessels
based on the diagnostic rim sherds. Sherds that were too
small were discarded, but there could be a tendency to pay
more attention to small sherds in a small sample and ignore
similar sherds in a large sample. Attempts were made to

avoid this bias in the study.

203

7.17 Begsgnal bigg

Personal bias can occur due to differences in judgement.
Also, bias of judgement can occur when certain distinct
attributes also appear to exist in a continuum. An example
of this is related to the frond roulette decoration, which
may merely look like large twisted cord roulettes. 'In this
study, the investigator was fairly confident that frond
impressed apart from a coarse or large nature, tend also to
have the negative impression of palm frond veins on them. In

general, efforts were made to avoid personal bias.

7.2 Interpretigg uodelg

In spite of the difficulties and inevitable bias in this
research, results and existing evidence on Benin can be
related (to some extent) to the assumptions proposed in
Chapter III.

7.21 Assumption 1; Associations between artifact types or
design attributes on artifacts can reflect temporal or
historical significance.

In relation to this assumptions, stratigraphic evidence
alone shows that in both the Eko Abetu and Uwan living sites,
two distinct cultural deposits are commonly represented over
a depth of about 1 meter. Some of the trenches show a third
mantle of humus layer more than others. Except in trenches
where disturbance create artificial mixture, such as trench
X2 of Eko Abetu and 410 of Uwan, the two cultural deposits

clearly relate to distinct historical formations. A' 15th

204

century carbon 14 date from the lower deposit of Eko Abetu,
indicates that the lower deposit of Eko Abetu probably dates
to the last 500 years.(16) It does not follow that one
cultural deposit follows the other without a break or that
the deposits or components are necessarily contemoporary, or
that other components may not exist on the northwest. In
fact, the absence of any absolute date from the Uwan Living
site, has made it difficult to stratigraphically relate the
Eko Abetu and Uwan sites together. Dating evidence such as
the frond impressed/twisted cord roulette, suggested by
Darling (1984), is, in the opinion of the invesstigator not
secure. Stratigraphic evidence in the Benin area, needs to
be provided to prove that the twisted cord (fine) is earlier
than the frond impressed.

However, chi-square determinations on body decoration
has allowed the investigator to group the Eko Abetu deposits
into 3. By the association of the two lower excavation
levels (which. can be assumed to correspond to the lower
cultural deposit) of X2 and Y2 and the non-association of
these with both their upper levels and the levels of B1,
these lower levels of X2 and Y2 can be said to belong to the
early occupation of the site which occurred about 500 years
ago. By the lack of association of the lower levels of B1
with its upper levels, and the levels of trenches X2 and Y2,

these two levels of trench B1 represent a gap in the

205

utilisation of the X2 and Y2 location. This was therefore a
middle phase occupation of the Eko Abetu site. The recent
occupation of the site relate to the building phase of X2 and
Y2, and the recent pottery in level 1 of trench B1. A 3-
phase occupation of the Eko Abetu site would be consistent
with an average of about 100 years life span for each phase.
This time span may correspond to the life and decay of mud
wall villages in West Africa. The t-test and cluster of
Jaccard matrix generally support this correlation (see figure
33). This sequence relates well to the Benin City
chronology.

The investigations of Connah (1975, Ryder 1975:243),
indicates that Benin palaces were established at about the
13th century. The Benin system of territorial and defensive
earthworks are thought to have been constructed by about the
15th century by Oba Ewuare. From about the 17th century, the
kingship was transformed from a ”warrior" ruler into a
"ritual figure rigorously secluded within the palace" as is
suggested by the intensive use of the palace during the
middle phase occupation of the Clerk’s Quarter’s site. In
1897, the colonial period sets in as the Oba was barnished to
Calabar by a British punitive force.

It would therefore appear that the Eko Abetu was first
occupied around the period of Ewuare. It is in fact usual
for people on the northwest and north of Benin City to

attribute the origin of their settlements to the reign of Oba

206

Ewuare. (Egharevba 1968). Informants told the investigator
that the Uwan and Ogbogiobo (close to Uwan) villages were
established by members of the Benin royal family during the
period of Ewuare. However, it has been possible in this
research to suggest a chronology for the Eko Abetu site on
the basis of artifact associations. The Eko Abetu sequence
looks like a reasonable chronology for the northwest, going
by the Benin City sequence. Oral historical data in general
agree that settlements around Benin post-date Benin City
itself.

7.22 Agggmptigp 2; Distinct ethnic or population groups are
expressed in the archaeological contexts as lack of
association between artifact types.

In Iron Age studies in East and southern Africa,
stylistic cultural attributes have been used successfully to
identify ethnic or linguistic groups and their historical
movements. For example, East African Early Iron Age peoples
are often distinguished by comb stamped decoration on
ceramics as distinct from roulette decoration of the Nilotic
peoples of the north (Huffman 1982:132). The point here is
whether the cultural phases identified on the northwest can
similarly be said to relate to distinct ethnic traditions.
Several lines of evidence argue on the contrary, or that
there is continuity of the northwest deposits to present
local Bini people.

The point has already been made that some of the

northwest settlements were established by people from the

207

city center. Also, although the Eko Abetu and the Uwan
living sites can be distinguished by some pottery types (see
Figures 42 and 51) and decorative motifs, such as the twisted
cord and frond impressed, for example, these attributes are
not unique to the northwest. All. the pottery types
identified on the northwest have been similarly identified at
the city center, and roulette and comb stamped decorations
are widespread in the Edo speaking area. This is indicative
of the continuity and homogeneity of the tradition.
Fortunately, historical records which date to the late 15th
century, also suggest continuity in the tradition in spite
of some conflicting reports (Connah 1967:597-598, Ryder
1965). At least the pottery tradition does not suggest a

break or ethnic differentiation.

7.23 Assumption Q; There is a relationship between artifact
variability or variability of design attributes on
artifacts and the settlement and residence type.

Closely connected to the idea of continuity in the
tradition is the need to explain what is responsible for the
observed variability in the northwest pottery. In the
analysis, pottery variability, based mainly on chi-square
determinations and cluster of Jaccard matrix, highlighted two
key explanatory variables, namely, temporal and spatial. But
the point is that the attribute distinctions both temporally

and spatially relate to behavioral differentiations within a

continuous tradition.

208

Although the variability does not specifically identify
pottery types, it can suggest site function indirectly, for
example. But what kind of function? Bini people identify
pot form and it’s function as follows: (Connah 1975:115-133,
Thomas 1910b, Omokhodion 1986b)

Akhg: This is commonly a necked pot but may also be a
globular pot or large forms of type E of Eko Abetu site (see
Figure 42). The general purpose is for storage of water or
grains. They are also used for heavy duty cooking e.g.
cooking yams for a community workforce or for cooking oil
palm seeds. The small form of the pot (gybi;gkhg - meaning
small pot) is regularly used to cook yams for the immediate
family, depending on the size of the family unit.

Egan; This term generally refers to bowls, used for cooking
soup and serving food (see Figures 42C,D,F and 51J, K).
Similarly, a small bowl is described as vai-gwggg.

Qkpgtg; The word literally means ”for washing the hand".
Forms (Figures) 426 and 51K can be used for this purpose.

They are usually small bowls.

The point in the above is that pot use amongst the Bini
people is quite ltmttgg (from narrow options) and peggigtggtt
The regular use of a pot is for cooking yams and soup. This
is a persistent cultural behavior. Traditionally it was
common for families to cook and eat yams three times in a

day.(17) Repetitive cultural function cannot confer

209

variability. Because of the degree of regularity of design
elements, tegombigatigg (which causes variability in design
attributes) is a fggtign of the mghilitz of pot ptgfiggezg and
ggztiggg. Specialised purpose use of pots, for example, as in
ritual contexts, (ritual function is often culturally mute,
and hence difficult to identify) is unlikely to create a
distinction in pottery analysis, since the same regular pots
are often used for ritual offerings. Models of seasonal
activities would be difficult to apply to highly sedentary
yam farmers. Temperature, which is a severe seasonal
determinant in temperate climate is not a significant
variable under tropical conditions. Even if someone wished
to relate the limited function to the pottery type, the chi-
square calculations were based on body decoration, and
decoration. may not be expected to express function.
Differences in pot form are more expressed on the rim than
on body decoration. Also there is no evidence for labor-
intensive production of pots in Benin. (Feinman gt 3; 1981,
Upham gt a; 1981). Production of Benin pottery was part-
time, unspecialized and a village activity.

A compelling explanation for pottery variability on the
northwest therefore, must account for the Igggmhingtign of
peggigtggt design attributes. This would be the case where
women who pxgdngg and gggzz the pottery have limited design

vocabulary and are mobile in W maidens... This
point is consistent with a pgtzilgggllz based social system,

210

and Benin tradition is a typical example. Mobility in post-
marriage residence does not imply descent. Patrilineality
may only be expressed as an inference of patrilocality, given
the context. The conclusion draws confidence from the
studies of Deetz (1965) who has similarly related stylistic
variability in American Indian Arikara ceramics to the
changes in marriage and residence pattern.(18).

Cultural persistence, pottery variability and
patrilocality in Benin, fit very well the continuity in
tradition already identified. Variability is reinforced by
the characteristic non-sororal polygyny. Different pot
carriers come to reside in the same location or different
women produce different pots in the same location. However,
the problem of representativeness of sample, makes this a
tentative conclusion.

7.24 Agggmptign A; The relative occurrence of artifact
type (or categories), between sites, ,can indicate a
kind of exchange configuration which includes class
and privileged access between sites or centers.

Relationships between the northwest and the city center
on the basis of occurrence of exotic or valued objects are
fairly clear. There are few long-distance trade items in the
northwest deposits. In level 4 of trench X2, Eko Abetu site,
there is a small cowrie (probably of the moneta specie), and
two gun flints. European bottle glass is present in the
most recent level of trench B1 of Eko Abetu, and levels 1 and

3 of trench 410 of the Uwan living site. The spaces sampled

211

on the northwest and the city center (Connah 1975) are not
directly comparable. Larger trenches have been excavated on
the city center, but the occurrence of beads (?glass) in the
city center and the absence of it on the northwest is quite
significant (see tables 7, 10 of connah 1975). Glass beads (
a European import) were valued as ornaments, or often as
symbols of rank to present day. In association with other
lines of evidence, however, the city center was clearly a
controlling center and the northwest like other outskirts
settlements must have acted as service communities, providing
farm produce to the city center or providing men for the
military. Bradbury has illustrated the controlling mechanism
adopted by Benin in trading relationship with Europeans.

No “foreign" traders from the interior were

permitted to operate in the Benin kingdom itself,
and stringent controls were exercised over the
waterside commerce with European and Itsekiri
merchants. Heavy dues were demanded from visiting
ships, the Oba’s monopolies in certain exports
strictly enforced, and general trading was allowed
when he and his chiefs had completed their business
(Bradbury 1973:49).

Although trade control during European contact must have
added considerable power to the Obaship, it is unlikely that
similar trading relationships prior to European influence
could have been significant to account for the evolution of
the kingship. Pre-European trade items that have been
recognized in Benin include the cowrie, metals, and perhaps

pottery. The extent of these early trades is poorly

understood.

212

Other lines of evidence which mark out the city center
from the northwest include the earthworks which show more
complexity at the city, the elite architecture and art
systems.

7.25 Agggmptign fit Artifact densities can suggest
differential population sizes and social complexity.

Population characteristics are important indices for
explaining the degree of complexity of the cultural system.
Cultural complexity could mean the growth of rank and control
centers; development of new techno-economy; efficient energy
extraction or increased per capita (imput) energy out-put
(Harris 1977). It is possible, for example, to explain the
relationship between population features and the labor cost
for the construction of the Benin earthwork.

The population census for 1952 indicates that Bini
speaking peoples (Benin Division) numbered 292,000. Out of
this figure, 52,000 people lived in the city center, the rest
lived in several villages of large size (6,000), medium size
(500) and small hamlets of about 20 people. The great
majority of villages had the medium size figures of about 500
(Bradbury 1973:51). These figures are important from two
angles.

First, it is possible to suggest a significance for the

Uwan wall on the basis of pottery densities and therefore the

213

associated population within and outside the wall. Looking
at per c.c. pottery weights for the Eko Abetu and Uwan living
site levels (Table 24-28), one could say that pottery density
at Uwan is about twice that of Eko Abetu. Pottery density
could reflect population or intensity of site use or pottery
quality (e.g. coarse or fine). If one assumes a 2:1
population ratio, this would be comparable to the present
conditions of the villages. Uwan is larger than Eko Abetu
and they both fall within the medium size villages of
Bradbury. A population ratio of 2:1 is not significant to
suggest any difference in social complexity, and medium size
villages, even in their present form could have been unable
to organise for labor for the construction of the Uwan Wall.
This point raises doubt about the relationship of the Uwan
Wall with the northwest settlements identified so far.

The second point also relates to the population of the
city center and the construction of the defensive earthworks.
Bradbury estimates that the population of the city center in
1897, must have been half it’s 52,000 figure in 1952. The
implication of this is that at the time the innermost or lst
concentric wall was constructed about 500 years ago, the
population of the city could have been less than 10,000 or
about the size of the existing large villages from the 1952
census. For the construction of the innermost wall, (which
is about 9 kilometers in circumference and 17 meters high in
recent condition) Daniels estimates (Connah 1967:606-608)

214

that it would have taken about 5000 men 97 days, working for
10 hours a day, to merely dig the ditch. The estimate does
not include labor taken to heap the earth up into a high
bank. The estimate probably assumes that agricultural hoes
were used for the digging purpose.. Now, this work force
estimate is difficult to apply to the population data and no
evidence of organised labor-intensive slave service is known.
Connah indicates (1967:608) that he could not say whether the
labor force needed, could have been of "slaves or other
origin". In addition, "tradition has nothing to say
directly about these earthworks" (Ryder 1975:243). The
construction of the earthworks of such magnitude, since these
would have been major cultural achievements, ought to be well
represented in mythology and the political art. The
traditional names for the earthworks (125 in Bini and lzglg
in Ishan) refer in general to a valley of two hills,
sandwiching a ditch or depression. There is therefore a
problem about the continuity of the earthworks (if they are
artificial) with present population.. The point cast doubts
on the military significance of the earthworks (Darling 1983,
1984).

215

7.3 Historical Implication

In explaining the evolution of the ancient kingdom of
Benin, there is no need in my opinion, to rely on insecure
interpretations of military conquest or control of trade
resources, when evidence from mythology (oral tradition) and
social organisation are very strong pointers. This research
has illustrated evidence for birth and sex principles in
marriage, inheritance, architecture, and religion. Some of
the evidence are surviving features of the people.

The appointment of knowledgeable leaders to act as
governors over people who may be unrelated to the original
point of origin of the political institution is a common
feature in political history. Colonial experience in Africa
is a living evidence for this. It is therefore easy to rely
on the evidence for Benin-Ife relationship and recognize
features of Yoruba influences in Benin. Kingship mythology
and regalia in Benin and the Yoruba states are similar. The
term Qt; is a Yoruba word. In Benin City today, there is a
rivalry between this introduced obaship and descendants of
the original rulers of the people, of which the Ogiamwen
institution is a survivor. The Qggnmila divination
equipment, which is widely used amongst the Edo-speaking
peoples today is clearly a Yoruba introduction. In Ifa
mythology, Orunmila, otherwise known as gig, was one of the

401 divinities sent to the earth by Olodumgre, the Yoruba

216

High God. Orunmila was the one charged with the power of
divination because of his great wisdom (Abimbola 1977:35).
The ancient architecture and town plan of Benin is almost a
copy of Yoruba designs. (Ojo 1966)

But the more important or novel evidence from this
research is the application of archaeological methods and
techniques to identify the patrilocal mode of social
organisation. This was based on the measure of variability
which is an expression of pot production and carriage.
Although the result is tentative (subject to confirmation),
it is quite consistent with the characteristic sex symbolism,
in which value is also placed on the taking of several wives
(non-sororal polygyny). This, in fact may be one of the most
important symbols in the evolution of the Benin kingship.
When for example, Oba Ovonranmwen was deposed and exiled in
1897, he was found to have had about 80 wives ‘(Egharevba
1968:58). Historically therefore, Benin tradition is a
patriarch institution, like many such traditions in Nigeria
today.

Apart from the nature of Benin-Ife relationship,
evolution of a patriarch institution is a quite possible by a
mere imposition, locally, of a divinity (e.g. a chiefship
over previously segmentary tribal groups) who can convince
others that the oracles have decreed his leadership. Very
little research has been done in the Edo-speaking area on

the great influence oraclular belief has on decisions and

217

action, and therefore its role in cultural change. The
investigator has been told several times in divination
clinics to begin to wear beads on his hands because he is a
chief. There are examples in Nigeria today of very powerful
religious patriarchs who are more like kings, and who explain
their great power and influence to relationship with the
divine. Influence, prestige, control of resources (including
women), and even military suzerainty are usually the external

expressions of this kind of underlying ideology.

7.3 Disgu§§ion

It is necessary to comment further on the question of
Benin-Ife relationship, since this has been used as a
supporting arguement for the military determinism in the
evolution of the kingdom. In this regard, the recent
observation made by Darling (1984), regarding correlates
between linguistic evidence and the distribution pattern of
pottery and the earthworks, is a significant point. From
linguistic evidence, it is thought that the origin of Benin
is related to the split in the proto-Niger-Congo language,
thought to have been spoken around the confluence of the
Niger-Benue, some 3-6,000 years ago. (Darling 1984:63). From
this split, ancestors of todays "southern Edo" (Urhobo/Isoko)
(see Figure 5) must have migrated into their present
locations in the swampy deltas about 2-3,000 years ago.

Waves of migrations into the Edo speaking area followed this

218

and the territoriality associated with the movements resulted
in linguistic differentiations presently noticed and the
building of 125 or earthworks. The Bini and Ishan earthworks
are therefore related to such migrants who penetrated into
this area between 1 and 3 thousand years before present.
Benin City emerged as a dominant center thereafter. Surface
distribution of design elements on pottery such as the coarse
string roulette, and the nodular carved wooden roulette
appear to support the migration view. Secondly,
archaeological evidence for iron using, immediately north of
Benin and at Abakaliki to the east which generally date to
the 7th and 9th centuries A.D. argues for a similar early
period for Benin. (see Anozie 1976, for similarly early dates
from the deltas). If this were so, then the association of
Benin with Ife whose dates are largely post A.D. 800 would
then be tenuous.

This doubt about the historical relationship between Benin
and Ife has also been expressed by Ryder (1965). Ryder had
noticed that early Portuguese records on Benin did not
mention Ife. In fact, Ife did not become current in Benin
historiography until the last century. And that indications
from historical records, and art tradition, suggest that
Benin must have been related to the "Beni confederacy" (Ryder
1965:32), located around the confluence of the Niger and
Benue rivers. That is, that the Portuguese were probably

confusing “Beni” (thought to have been an ancient state on

219

the Niger/Benue confluence), with the present Benin City.
The Maltese cross for example, (which was mentioned by the
Portuguese visitor to Benin in the latter half of the 15th
century). and the ritual associated with it, is not known in
Ife art, which further argues against the Benin-Ife
relationship.

1 Only one clear parallel to the Benin use of the
Maltese cross has so far come to light in adjacent
areas of Nigeria. This occurs in the standing bronze
figure at Tada, a village a little to the south of
the Niger and no more than ten miles from Jengi, said
to have been the capital of the Gwagbaji- Beni (Ryder
1965:33).

It is the opinion of Shaw (1973:237) however, that on
technical and stylistic grounds, this particular standing
bronze figure at Tada, and in fact the so-called Tsoede
bronzes along the Niger, belong to Ife art. Also, the
investigator has in this research, raised the problem of the
continuity of the Benin earthworks with present populations,
which casts doubts on their cultural significance. This
does not nevertheless remove the doubts that have been raised
concerning the relationships between Benin and Ife, but the
point remains that no real good historical or archaeological
evidence exist that seriously negates the Benin-Ife
relationship. On the contrary, archaeological evidence
support close ancient ties. Forms and decorative styles on
pottery between present day Benin and Ife are quite similar.

They share a common potsherd pavement flooring technique; the

art and architecture are very close. In addition, the dates

220

of Ife are earlier than those of Benin, but even if earlier
dates for Benin are established, this may yet not remove the
close typological similarity between the cultural systems.
Looking at typology for example, Igbo Ukwu is nearer to
Benin, than Benin is to Ife. The dates of Igbo Ukwu are-in
the 9th century and Igbo Ukwu cultural materials are
significantly different from those of Benin and Ife. For
example, although Igbo Ukwu pottery is "so highly decorated,
roulettes have not been used at all" (Shaw 1977:85, 1970).
This is not the case in Benin and Ife where rouletting
constitute perhaps over 60 per cent of design elements.
Also, metallic art of Igbo Ukwu consist of pure copper and
heavily leaded bronze (high tin), while the Benin alloys are
of brass (high zinc content), and Ife has almost pure copper
and leaded brass, showing that the Benin materials are closer
to Ife. However, a northern influence for these cultural
traditions is not in doubt and this has generally been
accepted, and in fact it is the opinion of Shaw (1973) that
only a long distance network of trade to the north and the
sahara could have ensured the provisions and introduction of
the alloy for metallic art and the lost wax or Qigg perdue
method of casting. However, debates about the typological
and chronological distinctions between Benin and Ife have
previously left unexplained the unique similarities between

them.

CHAPTER VIII
WWW

This research has been concerned about the problem of
the origin of the ancient kingdom of Benin, through the study
of artifact distribution on the northwest of the Benin City
center. The study was put into perspective through a review
of Benin research and related studies.

The unique rainforest ecological conditions of Benin,
was considered. The ecology partly influence the hoe
agricultural economy, architecture and settlement. The
people speak a language that belongs to the Kwa subgroup of
the Niger-Congo. The patrilineage society is organised on
two levels, namely, a village age-grade social organisation
in which age is the criterion of authority, and the town or
state social organisation in which authority is based on the
rule of primogeniture (inheritance of first son). This
unique social organisation is also based on sex because only
men can participate in the two modes of social system. The
society placed value on non-sororal polygyny (taking of wives
from different locations), including the isolation of male
and female in Benin architecture and in cult organisation.
These values are based on mythological or religious belief
systems.

Ancient trading or diplomatic relationships involving
Benin is reflected in the metallic art, ornaments (e.g. glass
beads) and other items such as cowries. The importance of

221

[E1

222

trade in the evolution of Benin was considered. Also, recent
studies (e.g Connah 1975, Darling 1984), on the distribution
pattern of the Benin earthworks and its military implication
have been critically assessed.

This study adopted an approach in which new
archaeological sites have been identified and excavated on
the northwest of Benin City. The sites yielded mostly
pottery materials which have been analysed . Inspite of
methodological problems, this research has been able to

suggest the following substantive conclusions:

fit; A 3-phase occupation of the Eko Abetu site, which dates
from about 500 years ago has been suggested, based on
deductions and correlations of stratigraphy, chi-square
results on body decoration, 'cluster of Jaccard coefficient,
t-test, pottery typology and the single calibrated date of
1432 1 100, from the lower levels of trench Y2. The result
is consistent with oral historical data which indicate that
some of the northwest settlements were established during the

reign of Oba Ewuare about 500 years ago.

8‘2 The research has also suggested a continuity of the
northwest cultural deposits with present local population by
the persistence of pottery forms. This conclusion isolates

the possibility of a cultural break.

223

fitfi The variability discovered in the chi-square
determinations on body decoration has been related to the
recombination of persistent design attributes. RecombinatiOn
of persistent design attributes on body decoration is only
explained by the known patrilocal mobility of women in post-
marriage residence. Other possible causes of variability
such as function, seasonal activity and labor-intensive
production have been isolated. It is stated that variability
is reinforced by the characteristic non-sororal polygyny
(different carriers of pot come to reside in the same
location, or different women produce different pots in the
same location). It is stated that this sex symbolism in
which value is placed on the taking of several wives may be
an underlying principle in the evolution of the Benin
kingship. Political offices can originate by the elaboration
of symbols which articulate their values. It is recognized
that the deduction based on pottery variability is tentative

because of the problem of representativeness of sample.

fitfi Population characteristics have been related to the
pottery density on the northwest. There is no evidence that
these northwest settlements were more complex than what the
1952 census population census suggests. It is therefore
suggested that medium size villages with populations of about
500 would have been unable to organise for labor for the

construction of the Uwan wall. This point raised doubts

224

about the continuity of the Uwan wall with the northwest
sites identified so far. The same point has been made
concerning the city center. Apart from the fact that the
methods by which the Benin City earthworks were constructed
have not been identified, it is significant that the
construction of earthworks of such magnitude (which would
have been major cultural achievements) are not represented in
mythology or the political art. The point therefore cast
doubts on the continuity of the earthworks with present
populations and hence their military implications. On the
other hand, ancient diplomatic relationship between Benin and
Ife is considered to be a likely explanation for the origin
of the obaship. Several cultural features which are shared
between Benin and the Yoruba states make this a compelling
explanation.

Methodologically, this research has been able to suggest
three occupation deposits on the northwest of Benin City.
The occupations could be found isolated in some sites, and
there may be more components. Conclusion in this regard is
also tentative subject to confirmation from other studies.

In the pottery analysis, the Jaccard coefficient and
chi-square determinations were found to be very convenient
and also helpful in arriving at quick decisions as to the
kind of relationship or association between assemblages. The
value of objective methods such as these can not be

overstated. It would be of interest if statistical

225

analytical techniques such as has been presented are
frequently applied in West African Archaeololgy. It could
stimulate students. The t-test determinations were not so
successful and appear as suspect. It is not clear from this
research whether a metric approach is appropriate, but one
problem was that the lip thickness was difficult to read
because of the curvature of some rims. Also it was a
frustrating task to read the color of sherds on a Munsell.

Further excavations are called for, to provide more
evidence from the northwest. It would also be necessary to
work on the south and north of Benin City, outside the Bini
speaking area to provide clues as to relationship between
Benin City with other Edo-speaking neighbors.

W

1. Material illustrated here, under "Benin Art" has been
published under the title "Features of Benin Art".
Omokhodion, 1986.

2. Information here is based on previous original study and
construction by the author. The report has been accepted for
publication by AMAN. Journal of society, culture and

environment. Vol. 5 no. 1.

3. In 1982, a new town planning program for Benin requested
me to prepare a report on the historical sites for the city.
The report was titled "Sites and Buildings of Historical
Importance". The materials are unpublished government
documents, but information here is from the report.

4. The technical name for lkhtngtg is flewbogldia lggzis.

5. It was important for me to obtain a formal approval for
the excavation from the Federal Department of Antiquities. I
obtained this at the middle of December of 1985, from Mrs.
A.K Fatunsin of the National Museum, Ibadan. Part of the
condition under which this approval was to be obtained was
that an archaeologist from one of the universities should
agree to be my local field supervisor. This led me to travel
to Nsukka on the 2nd of December to ask Dr. Fred Anozie of
the Department of Archaeology, University of Nigeria, if he
could act in that capacity for me. He kindly agreed. On 1,
February, 1986, Dr Anozie was able to take time off his
pressing duties at Nsukka to visit me on the sites.

6. 20 cm spits (arbitrary levels) were considered
appropriate for the excavation on the northwest, on the basis
of previous experience on some Nigerian sites. Secondly,

seiving of the deposits was not carried out since careful
probing with a trowel would be sensitive enough to the mainly
expected pottery artifacts.

7. Old farm plots in this area are easily taken over by a
nuisance weed which is locally called Awolowo (named after a
political leader). The technical name of this weed is

WW

8. Apart from pottery, no other significant cultural
materials have been recovered from the northwest. Bone
materials, for example, have been scanty and very
fragmentary.

9. Jaccard coefficient was decided upon for it’s convenience
in the transformaion of non-weighted nominal data.

226

227

10. Average linkage has the advantage of a rapid organizing
principle, where new clusters are simply joined to existing
cluster, by computing the average similarity of all relevant
units (Hodson 1970:307). This cluster method also avoids the
problem of chaining. The method was designed to cluster the
excavation levels and not the attribute set as such.

11. T-test determination vis to determine association
(patterning) or the lack of it, between excavation units on
the basis of lip thickness.

12. Some carved wooden roulettes have been illustrated in
Plates 5-9. The interesting point that would be obvious is
that a great deal of error can arise if one wished to isolate
the differential patterns typologically. For example,
Darling (1984:267) typologically associated his "simple grid
pattern" (CR1) with early Edo settlements. One point is that
carved roulette patterns are rarely single. They are often
multiple. In Plate 7-8, carved roulette is illustrated as
either zigzag with vertical lines or vertical lines with
basket work pattern. (Note Darling (p.225) illustrates
"basket work pattern" as carved roulette 2 (CR2). Now, a
nodular pattern could appear with a grid pattern etc, but in
a sherd that is broken, an erroneous significance of a motif
could be suggested (more so when attribute sample is low and
unstratified). The typological and chronological relevance
of the patterns are therefore basically suspect. I would
prefer to go with David and Vidal (1977:44), and see carved
roulettes as essentially illustrating the use of iron.

13. Probability chi-square.computation is fairly easy and

straightforward to understand. Like the t-test, it was to

search for patterning between excavation levels on the basis

nominal pottery attributes. Note, that the patterning sought

in these organizing principles is of a polythetic nature.

That is, groupings are based on a sufficiency of conditions.
(see Clarke 1968:37-38)

14. This dating sample has been kindly submitted on my
behalf by Dr. Larry Robbins, Department of Anthropology,
Michigan State University.

15. This motif has been described as coarse string roulette
by others (e.g. Darling 1984:223, 266-267). It is
illustrated in Plate 18, 19, 20. Specifically, the following
is the way Darling has described it’s chronological
importance, "Although CSR1 (coarse string roulette) is
present on most of the sites which on other evidence appear
to be early, both the early vessel forms (SQE 24/MSC 27) and
the culture sequence chronology of the computer analysis

228

suggest that it occurred later than the bulk of the finer
string roulettes. The change from the finer string roulettes
(possibly made from savanna grasses to the coarse string
roulettes (made from oil-palm leaf fibres) may have merely
been the adoption of the new material from a different
environment" (Darling 1984:266-267). A date from trench 420
of the Uwan sites would have been helpful in contributing
further to this point about whether the frond roulette is
earlier or later than the twisted cord roulette on the
northwest. The point that can however be made here is that
apart from the fact that the forest environment presents an
array of rope fibres which can be plaited into fine or
coarse roulettes, the frond roulette is better represented at
the Uwan site, while the twisted cord roulette (fine string
roulette) is more prevalent at Eko Abetu. Secondly, in my
isolation of twisted cord from frond roulette, I was able to
establish that the frond roulettes are characterised by not
only a large (coarse) impression, but also the
presence of negative impression of palm frond leaves, (These
impressions are visible on Plate 20). If I was not able to do
this, I might have grouped them under twisted cord. The
following is the way Darling (1984:223) addresses the similar
problem: "As no examples of twisted rather than plaited
roulette were identified (Soper:1977; Pers comm 1979; 9-12),
this group was categorized into degrees of coarseness for
analysis". That is his plaited string roulettes were grouped
into a continuum of sizes even without any objective
measurement. The argument here is not that this could not
be done, but that what is done, is influenced by the problem
and techniques at hand.

16. The gun flints (which date to the post 17th century), in
level 4 of trench X2 has been considered intrusive from the
upper levels due to disturbance of this trench. Disturbance
in X2 has been illustrated by the stratigraphy and the
significant association of its excavation levels based on
chi-square results (see table 15).

17. Yam is often processed as pounded yam (ignfign) and this
is normally eaten with vegetable soup, such that the pot and
bowl forms would tend to associate functionally.

18. Other workers (e.g. Hodder 1979) have drawn attention to
the role of competitive exclusion (related to resource
control) as a determinant of artifact variability especially
across inter-tribal boundaries. Within-tribe variation can
occur as result of symbolic constraints such as censureship
or public opinion insisting on conformity (Hodder 1977:269).
These interpretations in my opinion assume that the artifact
traits in question are produced and used locally. In the
case of Benin pottery, the importance of transportation or

229

carriage (Nicklin 1971:47) has been established, since Benin
pots were produced in two pottery villages - Use and Utekon -
(see Chapter V), from where they were most likely transported
to other villages. There is no evidence for pottery
production locally in the Eko Abetu and Uwan villages. Apart
from this point, it should be recognized that the variability
established in this research does not portray mutual
exclusivity but a differential in the combination of traits.
Patrilineal institutions however do maintain territorial
village boundaries expressed in settlement and farmlands and

perhaps in specific ancestral worship. But there is no
direct relationship between these dimensions and pottery
production and dispersal. Also exchange modes are really

secondary factors.

230

LIST OF REFERENCES

I."
H
U)
*3
lg
H
E!
2
U)

Abimbola. Wande.

1977

Afigbo, A.E.
1977

Alagoa, E.J.
1972

Alexander,
1983
Andah, B.E.
1978
Anozie, F.N.
1976
Argle, W.J.

1966

.Atherton, J.
1972

Ben-Amos, P.

The Ifa Divination system. Nigggign
Magazine. 122-123:35-68

Facts and Myth in Nigerian Historiography.
Nigerian Magazine. 122-123:81.

Afliamgfibsflimnalts

Ibadan University Press.

"The salt industries of West Africa" Paper
presented at the Congress of Pan African
Association of Prehistory and Related studies.
Held at the Jos Musuem, Nigeria, 11-17, Dec.

Excavations at Rim, Upper Volta. Eggt 5131933
Jgutngl 91 Agchaeologz

Onyoma and Re: a Preliminary Report on
Archaeological Excavation in the Niger Delta.

{Egg-grim Malgfmm

The £23 91 Dahgmey. A history and Ethnography
of the old kingdom. Oxford. Clarendon Press.

A.
Excavation at Kamabai and Yagala Rock Shelters,

Sierra Leone. West Attican Journal gt
Archaeology 2, 39.

1980 In: Azt 9f 32812. Thames and Hudson, London.

Bradbury, R.
1957

E.
mmmmmmwm
Qt south-wgstgtn Niggtigt Ethnographic survey

of Africa. London.

231

232

1964 The Historical Uses of Comparative
Ethnographic Data, with special ref. to Benin

and the Yoruba, in, The Historigg ta Iggpiggl
Afgica, (ed) J. Vansina gt gt. London.

1973 Bgnig Stufiigg. Oxford Univ. Press.
Brown, J.A.
1971 The Dimensions of status in the burials at

Spiro, Memoirs 9: the Society gt Amegicag
Atchgeologz, 25

Burke, K., Durotoye and A.J. Whiteman
1971 A Dry phase south of the Sahara, 20,000 years

ago. West African Journal gt Archagglogy, 1:1.

Caldwell, J.R.
1964 Interaction Sphere in Prehistory. In, ngeggllign
Stgdies (ed) J.R. Caldwell and R.L. Hall.
Illinois State Museum Scientific papers. vol XII

133-150
Camps, G.
1977 Ten years of Archaeological Research in the
Sahara- fleet Afrieea lessee; 2f Areheeeleszi
7:2

Chikwendu V.E. and A.C. Umeji.
1979 Local sources of raw materials for the Nigerian
bronze/brass industry: with exphasis on Igbo

Ukwu. ngt Afgiggg Jougggl gt Agghaeglogz
9:141.

.Clarke, D.L.

1968 Anelziieal Areheeelesz- Methuen and Co
Ltd. London.

Cohen, Abner.
1971 "Cultural strategies in the organisation of
trading diasporas" In. Meillassoux, Claude (ed)
The dexeleemeat ef indiseaeue trade and gaskets
in Wegt Aitiga. London, Oxford Univ. Press.

Cole, H.M.
1972 Ibo Art and Authority. In Attiggn Art gag
legdgrship, editor. Douglas Fraser and Herbert
M. Cole. Madison. The Univ. of Wisconsin Press.

233

Coon, 0.8.
1968 Xengene Ceve Repeyt. Philadephia, Univ. of
Pennsylvania.
Connah, G.
1963 Archaeological research in Benin City, 1961-
1964.;t Hist, Soe. Nigeria, 2, 4:465-77.
1967 New light on the Benin city walls. it Hist.

Soe. theyte. 3, 4:593-609.
1975 The Ayeneeelegy 91 Benin Clarendon Press. Oxford.

1981 Three Iheusend sears in Afriea. Man and His
Environments in the Lake Chad Region of Nigeria.

Cambridge Univ. Press.

Daaku, Kwame, Y.
1971 Trade and trading patterns of the Akan in the
17th and 18th centuries. In. Meillassoux, Claude
(ed). The dexeleemeni ef indisenens trade and

maykets in fleet Afyieet London. Oxford Univ.
Press.

Daniels, S.G.H.

1975 The Middle and Late Stone Age, In, Lectnres en
Nigerian Prehistory eng AEQBEQQLQEZI Ibadan Univ
Press.
Darling, P.J.
1976 Notes on the Earthworks of Benin Empire. fleet
Afrieen Journel e; Archaeotegy 8, 143.
1983 Surveys of the Ancient Linear Earthworks in Benin

and Ishan, southern Nigeria. Paper presented at
the Congress of Pan African Association of
Prehistory and Related Studies. Held at the Jos
Museum, Nigeria, 11-17 December 1983.

1984 Arehaeelesz and Eisterx in

The ancient linear earthworks of Benin and Ishan
Cambridge Monographs in African Archaeology.
BAR Int. Series. 215. vol I and II.

David, N. and H. Henning,
1972 The Ethnography of pottery: a Fulani case seen in
Archaeological perspective, Module, 21 1029.

 

234

David N. and P. Vidal
1977 The Nana-mode village site and the Prehistory
of the Ubangian-speaking peoples, West Atyieen

dedrael 2f Areheeelesz 7=17.

Davidson B.

1974 Aiyiee in Bietezy. Collier Books. New York.
Deetz, J.

1965 The Dznemiee ef stzlistie ehanse in Arikara

eeyenieet Illinois Studies in Anthropology. 4.
Doran J.E. and F.R. Hodson
1975 Mathenetiee end Qennnteye in Ayenaeology

Harvard Univ. Press. Cambridge,

Egharevba, J.

1968 A fiber: 3153922 91 Benin Ibadan UniV. Press.
Es’Andah, B.W.

1976 An archaeological view of the urbanization in the

early West African states,Jt Hiett 5291 Nigeyiet
Eyo, E.

1972 Rop Rock Shelter 1964. fleet Afyieen Jenynei efi
Aycnaeoiogy, 2:13.

1977 Tue Theusend Years 91 Nigerian Art. Federal
Department of Antiquities, Lagos, Nigeria.

Fagg, A.

1972 Excavation of an occupation site in the Nok
Valley, Nigeria. fleet Afgieen Jeuynel ef
Areheeelesze 2 75.

'Fagg, B.
1972 Rop Rock Excavations, 1944. fleet Afyieen

leurael ef Areheeelezz 2=1

Feinman, G.M. Steadman Upham and K.G. Lightfoot.
1981 The production step measure, an ordinal index
of labor imput in ceramic manufacture.

Amerieen Aaiisuiiz- 46:871.

Fladmark, K.R.

1978 A guide Le Basie Areheeelesieel Field Breeeddree-
Dept of Archaeology. Simon Fraser Univ. 4.

235

Fletcher, R.
1977 Settlement Studies (Micro and semi-micro). in,

gnetiei Arehaeeiogy ed. D.L. Clarke, London.

Academic press.

Fraser, D.

1974 Aiyieen Ant 3; Ehiieeephy Interbook, New York.

Freeth, D.
1975 Geological report on the rock samples., in,
Connah. 6- The Areheeelezz 2f Beaiai=225-228
Gamble, D.P.
1957 The fleiet e: fienegemhiet London, Int. African
Institute.
Garlake, P.S.
1974 Excavations at Obalara’s land, Ife, Nigeria.
Test Afrieen dedrdel 2f Areheeeleszl 4 111
1977 Excavations at Isoya, near Ile-Ife. Nigeria, in
1972 Feet Afrieen ll 21 Arehaeelezzi7=57-96
Gifford, J.C.
1960 The type variety method of ceramic classification

as an indicator of cultural phenomena. Aneyieen
Antignity 25:341-347

Goldstein, L.G,
1980 . A case study
of two cemetries, in the lower Illinois valley,

Evanston Archaeological program.

Goodwin A.J.H

1957 Archaeology and Benin Architecture. it e: fliet=
Soea fligeyiet 1, 2:65-85
Goody J.
1971 Teehneleszi Treditien and the state in Afriedl

Hutchison Univ. Library for Africa.

Greenberg, J.H.

1970 The Languages 91 Afriee- Bloomington
Indiana Univ -

Haberland, E.
1981 West African mud Architecture. Research of the
Frobenius-Institut. Atriean Ayts XV, 1:44-45

236

Hall, M.

1983 Tribes, traditions and numbers: the American
model in southern African Iron Age ceramic
studies. 522th Afrieen Arehaeelt Built
38:51-7.

1984a Pots and Politics: Ceramic interpretations in
southern Africa. Woyld Archaeology, 15
3:262.

1984b The Burden of tribalism: The social context of

southern African Iron Age studies. Aneyieen
Ahtidhitz 49. 3=455.

Harris, Marvin.

1977 Qennihele and Kings; The erisine ef eulturee.
Vintage Books, NewYork.

Hill, H. M.
1978 Dating Senegambian Megaliths: a correction.
Cnryent Anthropology 604-805
Hodder, Ian
1977 The distribution of material culture items in the
Baringo District, western Kenya. Men 12, 2:
239.
1979 Economic and social stress and material culture
patterning. Aneyieen Antiguity 44, 3:446-454
Hodson F.R.
1970 Cluster analysis and archaeology: some new
developments and applications Werld Ayeheeeiegy
1. 3:299-320
.Huffman, I. N.
1981 Archaeology and Ethnohistory of the African Iron
Age. Annnei Review 91 Anthyopeiogy, 11:133-50
1970 The Early Iron Age and the spread of the Bantu.

fienth Aizieen Ateheeeli Enlli 25=3-21.

Ikime, Obaro.
1969 Niger Delta Bi_elrx: Tteekiti: Urhehe Beletiehe
_nd the E.reeeeh Breeenee 1884- 1936

Johnson, M
1970 The cowrie currencies of West Africa. Part I
g African Hist. XI, 1: 17-49

 

237

1970 The cowrie currencies of West Africa, Part II
it Afrieen Bietl XI. 3:331-353-
1976 Economic foundations of an Islamic theocracy:
The case of Masina. Jeurna; e: Aftican Histogy
XVII, 4:481
Jones E.W.
1956 Ecological studies on the rain forest of

southern Nigeria.gt et Eeeiegyt44:83-117
Joukowsky, M.

1980 A eenelete mantel ef field Areheeelesz.
Prentice-Hall Inc. Englewood Cliffs, New Jersey.

Klein, Martin, A.

1972 Social and economic factors in the muslim
revolution in Senegambia. ienynei e: Africen
Hietepy XIII, 3:419-441
Krieger A.D.
1944 The typological concept,. Amepieen Antignity
9, 3:271-288
Kuper, A.
1980 Symbolic dimension of the southern Bantu

homestead. Atriea 50:8-23.

Last, Murray

1967 The Sehete Ceiiphete. Longmans, London.
Lightfoot, Kent G. and Gary M. Feinman.
1982 Social diferentiation and leadership development

in early pithouse villages in the Mogollon
region of the American southwest.

Amerieeh Antishitz 47, 1:64

Linares De Sapir, O.
1971 Shell middens of lower Casamance and problems of
Diola protohistory. West Atyieen A e:

Areheeelesz 1:23-

Lovis, W.
1976 Quarter sections and forests: An example of
Probability sampling in the northern Woodlands.

Amerieen Ahtidhitz 41. 3:364.

Maliphant C.K, A.R. Rees and P.M. Roesse
1976 Defense system of the Benin empire - Uwan.

fleet Afrieeh 21 2f Areheeelesz 6:121

McIntosh, R.

1974

1976

238

J.
Archaeology and mud wall decay in a West African

village. World Aychaeology 6:154-171

Square huts in round concepts: prediction of
settlement features in West Africa,.

Archeeology 29:92-101

1977

McIntosh, S.
1981

Munson, P.J.
1971

The excavation of mud structures: an experiment

from West Africa. flepie Aeeheeeiegy 9,2:185-199.

K. and R.J. McIntosh.

West African Prehistory. Anepieen Seientiet
69:602

TheTiehitTreditien; Aleteenehieterie
gccgpgtien 91 the eeutheeetern ethere-
Ph.D. Thesis. Univ. of Illinois. Urbana-

Champaign.

Nevadosky, J.

1984a

1984b

Kingship succession rituals in Benin. 2.
Aftideh Arte XVII. 2:41

Kingship succession rituals in Benin. The
coronation of the Oba. Atyieen Ayte XVII
3:48

Nevadosmky J. and D.E. Inneh.

1983

Kingship succession rituals in Benin . Becoming
a crown prince. Atyieen Ante XVII 1:47.

Nicholson, W.E.

1929

Nicklin, K.
1971

Obayemi, A.
1976

The Potters of Sokoto, N. Nigeria. Men, 29
No. 34:45-50.

Stability and innovation in pottery manufacture
Woyld Ayeheeeiegy 3:13-48.

The Yoruba and Edo speaking peoples and their
neighbours before 1600, in. Ajayi A.E.A. and

Micheal Crowder (ed) flietepy pi fleet Afipiee I

New York. Columbia Univ. Press.

 

239

Ogunkunle A:O., U. Omoti, D.O. Ataga, A, Eseiche.

1980

Field tours to parts of the Afemai hills, the
Benin lowlands and the alluvial plains of Bendel
State for soil examination. 8th Joint Annual
Conference. The National Soil correlation
committee and the Soil Science Society of
Nigeria. Unpublished.

Oguagha, P.a. and A.I. Okpoko

1984

Ojo, A.
1966

Omokhodion,
1986a

1986b

Histogy ene Ethnoarcheeology in Eeetern fligeyie;

A Study et Igbo-Igala relations with specia;
reference te the Anambye valley. Cambridge

Monograph in African Archaeology, 7. BAR Int.
series. 195

Teruhe Eeleeeei A etddz ef Afine ef Xerhhe
ldhfi- Univ. of London Press.

D.

Features of Benin Art. Atpieen fitndy
hehesreehe 6=67-70-

Benin Pottery. Paper presented on April 20
1986, at the Department seminar, Dept of
Sociology and Anthropology, University of Benin.

Onwuejeogwu, M.A.

1981

Perham, M.
1937

Ah late Qixilizetien; Nri Kingdom and
Hegemony, Ethiope Publishing Co. Benin City.

Netixe Administratien in Nigeria. Oxford

Univ. Press.

'Posnansky. M

1984

1973

Prussin, L.
1970

1976

Early agricultural societies of Ghana, in,
From Hunters te Farmers. ed. J.D. Clark and
S.A. Brandt, 147-151.

Aspects of Early West African Trade. flepig
Aychaeology, 5, 2:149-162.

Architecture and the Manding. Attican Arte
3:12-19

Fulani-Hausa Architecture. Atpieen Ante
X, 1:8-19.

 

240

Rahtz, P.A. and Colin Flight.
1974 A Quern Factory near Kintampo, Ghana. Weet
African it gt Archaeology 4:1-31

Rees, A.R.
1975 A wall and ditch system north of Benin City,
in. Connah G. The Areheeelesz 2f Behin-
Read, D.W.
1982 Toward a theory of Archaeological classification
in, Eseays gn Archaeolggical Typology ed.
Read, D.W.

Redman C.L. and P.J. Watson
1970 Systematic intensive surface collection.

Amerien Antiduitz 35:279-91.

Roberts, J.M.

1980 A Belieeh Hieterx hi the Eerld- Pensuin
Books.

Robertshaw, P. and David Collett.
1983 A new framework for the study of early pastoral
communities in East Africa. Journal gt African
Histoyy 24:289-301.

Rosenfield, A.
1972 The microlithic industries of Rop rock shelter.
West African it gt Archaeology 2:17

Rouse, I .
1960 The classification of artifacts in
Archaeology. Ametican Antiquity 25:313-23.
.Ryder, A.F.C
1965 A reconsideration of the Ife-Benin relationship.
it African Histoyy. 6:25-37
1969 Benin egg the Europeene 1485-1897. London.
1975 Historical implications. in, The Archaeology gt

Benin ed. Connah G. 242-246

Shapiro, Gary
1984 Ceramic vessels, site permanence and group size:
A Mississippian example. American Antiquity
49, 42696-712.

Shaw T.
1969 Archaeology in Nigeria. Antiggity. XLIII.

241

1970 Igbo-Ukwu: An eggount gt epgheegiggigei
dieggveriee in Easteyn Nigeyia, Faber and Faber.

1975 Analysis of ”bronzes" from Benin. in. The
Archaeology gt Benin G. Connah.

1976 Changes in African Archaeology in the last forty
years. in. Afrieeh Sthdiee einee 19$§= A

tribute to Basil Davidson. ed. C..Fyfe,.

1981 The Nok Sculpturs of Nigeria, Scientific
Ametican 244, 2:154

1984 Archaeological evidence and effects of food
producing in Nigeria. in, From Hunters tg
Fathers. ed. J.D. Clark and Brandt. Univ. of
Californania. Press.

1973 A note on trade and the Tsoede bronzes. West
Afticen it gt Archaeology. 3:233-238
1977 Unearthing Igbo Ukwu: Archaeological

Discgveries in Eastetn Nigeria. Oxford Univ.
Press. Ibadan.

Smith A.B.
1974 Preliminary report of excavations at
Karkarichinkat. Mali. West Africen it gt

Archaeology 4:33-55.

1984 Origins of the Neolithic in the sahara. in,
Etgn Hnnters tg Eermers ed. Clark and Brandt.
Univ. of California.

.Soper, R. and P Darling.
1980 The walls of Oyo Ile, Oyo State, Nigeria. West

Atrican it gt Archaeology 10:61

Sowunmi, M.A.
1985 The Beginnings of Agriculture in West Africa:
Botanical Evidence. Nature 26, 1:127-129

Spaulding, A.C.
1953 Statistical techniques for the discovery of
artifact types. American Antiquity 18:305-313

Struever, S. and G.L Houart
1972 An analysis of the Hopewell Interaction sphere.
in, Sgcial Exchange Interaction ed. E.N.Wilmsen
Museum of Anthropology. Univ. of Michigan.

242

Stuiver, M. and B, Becker.
1986 High-precision decadal calibration of the
radiocarbon time scale, AD 1950 - 2500 BC
Radiocarhon 28, 2B:863-910

Thomas, N.W.

19108 Ahthreeelesieel reeert en the Ede;_2eekih£
peoples gt Nigetia. New York: Negro Univ. Press

1910b Pottery making of the Edo-speaking peoples,
southern Nigeria. Man No. 53:97-98.

Thompson, R.F.

1972 The sign of the divine king: Yoruba bead-
embroidered crowns with veil and bird
decorations. In. D.Fraser and A.M. Cole ed.
African Att egg Leadership. pp. 227-260.

Tremearne, J.N.

1910 Pottery in northern Nigeria. Men 10, 57:102-103
Udo, R.K.
1980 Environments and peoples of Nigeria. A

geographical introduction to the History of
Nigeria. in. drehndherk ef Nigerian Hieterz

ed. Obaro Ikime. Heinemann Books.

Upham, S.; K.G. Lightfoot and G.M. Feinman
1981 Explaining socially determined ceramic
distributions in the prehistoric Plateau

southwest. Ametigen Antiqnity 46, 4: 822- 833

Vlach, J.M.
1976 Affecting Architecture of the Yoruba. Atyigen
Arts X. 1:48-53

Willet, Frank.
1972 Ife, The art of an ancient Nigerian aristocracy.

In. Attican Att egg Leedership (ed). D. Fraser ,

and Cole. Madison. The Univ. of Wisconsin Press.

Willet, F. and Connah, G.
1969 Pottery making in the village of Use, near Benin
City. Nigeria.Baessler-Archiv. (N.F.) 17:133

 

243

APPENDIX A -- PLATES

244

 

Plate 1: Gun Flints. Level 124 Eko Abetu.

245

   

Plate 2: Uwan Wall. Figure in photo walks down slope along
the ditch. Main embarkment of the wall on his right hand
side

 

246

 

in photo stands on top of main

Uwan Wall. Figure

Plate 3

embarkment.

247

 

Plate 4: Excavation in progress on trench 410. Uwan site.

248

 

 

Plate 5: Level 111 body sherds. Eko Abetu. Note the two
carved roulette motifs in this Plate may belong to a single
pattern on one vessel, but when found isolated someone could
type them differently.

Decoration label: *

a' carved wood

b' punctation

c: slashed

d' grooved

e wavy-line
twisted cord.

f
* Note not all the motifs are illustrated at the same time
in each Plate.

 

249

 

 

Eko Abetu.

Level 121 rim sherds.

Plate 6

carved wood
punctation
slashed
grooved

e: wavy-line
twisted cord.

f

Decoration label.
b
c
d.

250

   

 

Plate 7: Level 132 rim sherds. Eko Abetu.

Decoration label
: carved wood
punctation
slashed
grooved

wavy line
twisted cord

HOG-00‘“

 

251

 

Plate 8: Level 132 body sherds. Eko Abetu.

carved wood
punctation
slashed
grooved
wavy-line
twisted cord

 

252

 

 

Plateq: Level 133 rim sherds. Eko Abetu.

Decoration label.
a: carved wood

b: punctation

c: slashed

d: grooved

e: wavy-line

f: twisted cord.

253

 

Plate 10: Level 411 rim sherds. Uwan site.

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved-rim pot
J: flared rim bowl
K: short-neck bowl

 

 

254

 

Plate 11: Level 412 rim sherds. Uwan site.

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved-rim bowl
J: flared rim bowl
K: short neck bowl.

 

255

 

Plate 12: Level 413 rim sherds. Uwan site

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved rim pot
J: flared rim bowl
K: short neck bowl

 

 

 

 

 

Plate 13: Level 414 rim sherds. Uwan site

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved rim pot
J: flared rim pot
K: short neck pot

257

 

Plate 14: Level 421 rim sherds. Uwan site.

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved rim pot
J: flared rim bowl
K: short neck pot

258

 

Plate 15: Level 422 rim sherds. Uwan site

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved rim pot
J: flared rim bowl
K: short neck bowl.

259

 

 

Plate 16: Level 423 rim sherds. Uwan site.

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved rim pot
J: flared rim bowl
K: short neck bowl

260

 

 

 

Plate 17: Level 424 rim sherds. Uwan site

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved rim pot
J: flared rim bowl
K: short neck bowl

Jk

261

 

Plate 18: Level 411 body sherds. Uwan site

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved rim pot
J: flared rim bowl
K: short neck bowl

 

 

Plate 19: Level 421 body sherds. Uwan site.

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved rim pot
J: flared rim bowl
K: short neck bowl.

263

 

 

Plate 20: Level 424 body sherds. Uwan site.

Decoration label
a: carved wood

b: stab/drag

c: frond roulette

Form label

I: grooved rim pot
J: flared rim bowl

K: short neck bowl.

      

MIC

77:17:77le1173777777177“