TH E West African society dealt wit h in this book is half a millio n strong. It is far from being either
‘ simple’ or ‘ primitive’ . Its social and economic complexity is com• parable only wit h the civilization of Imperial Rome, Byzantium, or Medieval Europe. But very few historical records exist; the people of Nupe have not yet risen to the stage of education which can pro• duce writte n records of important events or institutions. Th e present book is a piece of pioneer work, the result of two years in the field, spent in the course of two ex• peditions, one in 1934, the other
A
BLACK BYZANTIU M
THE KINGDO M OF NUPE I N NIGERIA
B Y
S. F. NADEL , PH.D .
Government Anthropologist
Anglo-Egyptian Sudan
WIT H A FOREWORD B Y
TH E RIGH T HON . LOR D LUGAR D
G.C.M.G. , D.S.O. , O.B.E .
Formerly Governor General of Nigeria
Published for the INTERNATIONA L INSTITUT E O F AFRICA N LANGUAGE S & CULTURE S
by the OXFORD UNIVERSITY PRESS
LONDON NEW YORK TORONTO
TO
M Y W I F E
W H O S E B O O K T H I S IS .
A S M U C H A S M I N E
F I R S T P U B L I S H E D 1 9 4 2
R E P R I N T E D 1 9 4 6
I N T E D I N G R E A T B R I T A I N A T T H E U N I V E R S I T Y P R E S S , O X F O R D B V J O H N J O H N S O N , P R I N T E R T O T H E U N I V E R S I T Y
FOREWORD
By LORD LUGARD
HI S exhaustive treatise on one of the most important communities in Nigeria owes its inception to the generosity of the Rockefeller Foundation of New York, which placed a large sum at the disposal of the International Institute for the Study of African Languages and Cultures. Fellowships were instituted to enable specially trained students to undertake scientific research into the social organization of a selected unit of African society, the factors of social cohesion which operate to prevent its disintegration by contact with European civiliza• tion, the openings for co-operation, and the tendencies towar4s new
groupings. Others were specially concerned with linguistic studies.
Dr . Nadel, of the University of Vienna, was one of the original Fellows, and he selected the Nupe of the middle Niger for his field of research. He brought to his task not only a thorough mastery of the technique of scientific investigation, but an original mind, and a sym• pathy with and understanding of the African, which won the confidence and friendship without which scientific qualifications can effect little.
His exceptional talents as a linguist enabled him not only to con• verse, at first in the lingua franca (Hausa), but in less than six months to acquire, and speak with fluency the very difficult ‘tone’ language of the Nupe. Many an English reader of this book may even envy his command of idiomatic English. The Nupe seem to have regarded him as an unusual phenomenon and conferred upon him a high title of tribal rank.
As an ‘Anthropologist* he is concerned at the outset to explain that the term—whether ‘functional’, ‘practical’, or ‘applied’—is ill adapted to the research he had undertaken. But surely the study of ‘a native society in all its complexity’, and the factors which had affected its social evolution are better described as ‘Sociology’ than as ‘Anthro• pology’—which is defined as ‘the study of man as an animal’ ? A clear understanding on the other hand of the connotation of the term
‘ Culture’ as used by a highly qualified sociologist is essential to the appreciation of his work.
‘Culture (he says) varies with environment and . . . expresses the changing dominant interests involved in social, political, and economic conditions…. It is not a sum total of discrete traits (but).. . a conscious unity created by and embodied in a common group life (by) . . . the people who form the group, . . . specific to it and only to it irrespective of variations within, or similarities with groups outside, the tribal boundaries.’ (pp. 14-15.)
Th e distinctive individuality of the ‘tribe’ depends in turn on the pre• sumption that it is a ‘group possessing more or less the same culture’ and language.
The ‘social and economic complexity’ of the Nupe State as it exists
I V FOREWORD
to-day is epitomized in the comparison he has chosen as the title for his book. Affiliations with neighbouring tribes, migrations to found new settlements, and above all the influence of foreign rulers have introduced modifications, but without radical change in the traditional culture. Th e complete absence of any written records, and the reticence of the people until complete confidence had been established, accentuate the difficulty of elucidating the origin of the tribe before and up to the death of Tsoede, the mythical founder of the kingdom. From the end of the sixteenth century, however, the history can claim comparative accuracy. Comparison of the Nupe language with those of neighbouring tribes is complicated by the existence of five distinct dialects.
Dr . Nadel’s first object was to discriminate between the essential characteristics of Nupe culture, and the variations from the typical pattern in ‘a heterogeneous society divided by gulfs of culture, ethnic extraction, community, and class’. For this purpose he con• sidered that what he calls ‘the anthropological quadrivium—kinship, political organization, economics, and religion’—was unsuitable, and decided to base his research on ‘the two inseparable aspects of culture, political and economic organization’, with special reference to religion and law.
This involved in the first place an examination of ‘the factors of social cohesion’ upon which the claim of a community to rank as a unit of self-government is based, and in the second place the effect of changes brought about, inter alia, by contact with Europeans and especially by the British Administration—the two main objectives in his ‘terms of reference’.
Singe it is the declared policy of the British Government to help the different units of native society to govern themselves in accordance with civilized canons of justice, and of impartiality between rival claims arising from ‘the interpenetration of essentially different social sys• tems’, such as Islam and paganism, it goes without the saying how valuable such an objective study would be to the Administration. We find that, in fact, it has been utilized in conjunction with the researches of District Officers.
Among the binding forces which constitute the state, kinship, says Dr . Nadel, has no part. Its influence is limited to component groups only. Nor can economic interdependence exert a binding influence. It is more likely to be disruptive. He finds that the factors which have contributed to the solidarity of the Nupe State are primarily the attach• ment of the people to the principle of Kingship, of which the Etsu (king) and the holders of the important Offices of State are the symbol.
Th e Etsu personifies the heroic achievements of the mythical Founders of the Kingdom whose glories are enshrined in legend and song, and in which the people take pride as citizens of the state. A second and potent centripetal force was the acceptance of the brother• hood of Islam, the creed of the ruling race, and the avenue to title and dignity.
Th e annual festival of the Sallah (Id el Kebir), attended by immense crowds, became a day of national ceremonial and national unity. To
FOREWORD v these influences may be added the common participation in a uniform system of law and justice. It is surprising that the cohesive effect of a language radically different fro m the lingua franca is not mentioned in this connexion, since it is surely one of the most important of the characteristics whic h confer a distinct individuality upon the Nupe State and a common bond among its people ?
Th e tw o view-points of political and of economic organization inevitably involved research into ‘urban and rural culture, Mohammed• anism and paganism, the organization of the state, and the organization of the village community’. It is only possible here to name a few of the varied aspects o f native life, of whic h the reader wil l find a very full analysis in these pages. The y include the modifications of traditional law and custom brought about by the advent of Islam and later by the British Administration and the abolition of slavery.
Th e tenure of land ; the place of women in the body politic ; the relation of the rulin g classes to the peasantry; the psychological effect of the multiplicit y of titles of rank in every class and every vocation (not wholl y peculiar to Nupe), are all exhaustively described. Th e com• paratively advanced methods adopted in agriculture include manuring and rotation of crops, and Dr . Nadel has some shrewd comments on the system of ‘mixed farming’ introduced by the Government.
We learn how completely this African State has accustomed itself to a ‘money-economy’—the extent of its trade, the contrasts between urban and rural economy, the system of collective assistance (egbe) and ‘labour-units’, and of financial expedients such as dashi, wit h careful estimates of the cost of livin g and of the family budgets of different classes. Th e account of the different Guilds—of blacksmiths, glass-makers, weavers and the rest, including the ‘luxury-crafts’ of bead- workers, and of artistic ornaments, is specially interesting.
Th e book is an exhaustive and meticulous picture of the life of an African community whic h has maintained its individualit y through many cycles of change. It wil l be a very valuable addition to the parallel researches by other Fellows of the Institute. We may congratulate the Sudan on having secured the services of so competent a sociologist.
ABINGER COMMON
September 1941
INTRODUCTION
HIS book has been conceived, written, and published under the shadow that darkens the face of the world to-day. We have long believed that the furtherance of knowledge carries its own justification. But in recent years the legitimacy of pure, self-contained, detached science has become more and more dubious.1 For no field of science is this more true than for the science of society, which has been pressed into the service of ruthless political aims, persecution, and suppression. Historical falsifications, racial doctrines, distorted dogmas of social necessities, have produced weapons no less deadly than those manu• factured in factories and laboratories. A science that could be thus abused must no longer hope to recover its spurious detachment. I t will find redemption only in closeness to the problems of our existence as
society and civilization.
A student of society professing this belief in the social mission of science could not wish for more congenial material than the African society described in these pages. Indeed, I have not been satisfied with the bare description of a remote and strange social existence. I have tried, on a modest scale, to contribute to our understanding of the forces that shape the fate of societies—of our own society to-day as of that West African polity. Wars, suppression, persecution, and the reactions against these forces—you will find them all in this book. And you will find them so presented that their relation to the problems of our own social existence is never obscured. I believe that some lessons might be read from these facts: at least, I have tried to write them out as legibly as I could.
And another thing. Large parts of this book are dedicated to what has become known as ‘practical’ or ‘applied* anthropology. It has-been said that modern anthropology is destined to be of great assistance to colonial governments in providing the knowledge of the social structure of native groups upon which a sound and harmonious Native Administra• tion, as envisaged in Indirect Rule, should be built. Let me say that I for one firmly believe in the possibility of such co-operation between anthropologist and administrator. Indeed I hope that this book will bring evidence of the fruitfulness of the co-operation. The form which it should take in the concrete case is perhaps not as yet well defined or universally accepted, and a few words must be said about the view taken in this book.
Obviously, the description of a native society as it might appear if it had never been touched by modern changes, even a mere compilation of facts to which the administrator could turn as to a dictionary or gazetteer, would be misleading because artificially simplified. It must be the duty of the anthropologist to demonstrate the native society in all its complexity, fashioned as it is by a multitude of social causes and effects, in which the efforts of administration represent causes and effects
1 See J. D. Bernal, The Social Function of Science, 1939.
INTRODUCTIO N VII among others. It may be debatable how far the anthropologist can go in this demonstration withou t introducing values of his own, withou t venturing criticisms or committing himself to concrete suggestions, and thus going outside his domain. But there can be no doubt about the legitimacy of his task to demonstrate alternative developments that have arisen, or are likely to arise, from administrative measures or their omission, from a new extraneous change or the unbroken continuance of an indigenous institution. If such a view is called implied or indirect
criticism, I shall bear the blame lightly . For this indirec t criticism* is nothing but a legitimate scientific device (used throughout the book) to assess the effectiveness of social forces—internal or extraneous, historical or modern. Moreover, all criticism stands or falls wit h the conception of principles and necessities of whic h it is an expression. Anthropologists and administrators may differ about these principles and necessities— we anthropologists are the first to realize this. But we feel that this divergence of view—i f divergence there be—should be placed in the ful l light of knowledge; if our ‘criticisms* are discounted, as indeed they may wel l be, the final decision should yet be inspired by them, and take the form of conscious choice.
Moder n anthropologists wil l not lightl y surrender their claim to the name ‘explorer’. True , their work no longer takes them to peoples who m no traveller has yet encountered, and to countries whic h no white man’s eye has yet seen. But they probe more deeply into the society of primitiv e man than any of the early explorers. The y have set them• selves the task of liftin g the knowledge of primitiv e cultures beyond casual acquaintance, on to the plane of true scientific insight. They are explorers, not of the surface of the globe, but of an unknown social realm. As an explorer, the anthropologist is faced wit h the difficulties of method and technique that beset all pioneer work. Anthropology in this new intensive sense is a young science, whic h is still evolving its methods of approach and has to find its instruments of research as it goes along. It has not yet been able even to map out wit h sufficient exactitude all the possible forms of society whic h its field-workers might encounter. Every new field-work may mean the discovery of unknown social structures unapproachable wit h the techniques hitherto used. Thi s is indeed the plea that this book makes: a plea for understanding— and consideration—of its nature as a pioneer work, and for the difficul•
ties and shortcomings inherent in such work.
Th e present book can claim to be regarded as a pioneer work in more than one respect. Th e technique of modern anthropology (‘functional’ anthropology as it is called by some) has been evolved on small islands in the Pacific, and in the study of small groups and cultures infinitely simpler than our historical civilizations. Th e West African society dealt wit h in this book is half a millio n strong. It is far from ‘simple’ or
‘primitive’ . Its social and economic complexity is comparable only wit h the civilizations of Imperial Rome, of Byzantium, of medieval Europe. Yet if the technique of anthropological field-work in the accepted sense must thus prove inadequate, the method of approach normally applied
vm INTRODUCTIO N
in the study of these historical cultures is equally barred. Very few historical records exist, and they are hardly informative from the sociolo• gical point of view. The people of Nupe have not yet risen to the stage of education producing written records and accounts of important events or institutions. Wit h the exception of a few educated Mohammedan scholars whose co-operation I was able to enlist in the recording of folk• lore and historical traditions, I have been unable to use a single literate informant, for example, to keep diaries of important events, or to handle questionnaires. Lastly, the most essential evidence in the study of com• plex societies, namely statistics, scarcely exists. Th e statistics in this book were, with few exceptions, compiled single-handed and on a scale of the inadequacy of which no one can be more aware than the author. I have described elsewhere the technique of inquiry which I applied, and need say nothing on that point here.1 But it is necessary to describe the general plan of research I pursued, for it illustrates the methods by which I attempted to circumvent some of the difficulties just mentioned. I spent two years in the field, in the course of two expeditions in 1934 and 1935-6. From the outset I was confronted with the problem how to cover, in intensive study, the vast area inhabited by the Nupe people. Th e political configuration forbade the selection of one section to the disregard of others. The widespread political system had clearly to be studied at different points, in its centre as well as on its periphery. Moreover, a short preliminary survey already revealed the existence of considerable cultural differences within the tribe. Th e further my studies proceeded the more manifest became the significance of these differences as variations of a common pattern, the study of which was essential to the understanding of the forces that moulded the culture of the people. Th e investigation thus became perforce a comparative study within the framework of one tribe and one culture. At a certain stage of my investigation I even had to go outside this framework and devote some study to neighbouring groups—the Kakanda, Gbari, and Bassa— which, closely akin to the Nupe and partly their fellow citizens in the Nupe kingdom, might throw light on certain aspects of Nupe culture.
Th e main problem was to discover representative sections and centres for the study of this widely spread social system and its varying culture. I will not speak of the difficulties which I encountered in this task, helped as I was only by records and information which soon proved superficial and inexact. In the first six or seven months I was reduced almost to the method of trial and error: a very short stay in a new place would have been useless even merely to discover whether or not this place might prove a useful centre; and once I decided on a long stay the time was gone and could not be recovered even if the place turned out to be less profitable than I had expected. Let me admit that this did indeed happen in two villages (Mokwa and Sakpe) which I had made my centres upon the advice of various informants and which, as I discovered later, were not as significant or essential as certain other places would have been. However, certain general facts about the culture and the
1 See “The Interview Technique in Social Anthropology’, in The Study of
Society, chapter xiii, 1939.
INTRODUCTIO N I X
people could be obtained there as wel l as anywhere, and on my second expedition I could to some extent correct the mistakes of the first. On the other hand, I was unable to study one or two villages (e.g. Yeti , in the oil-palm area, and Kutiwengi , where an interesting variety of a certain religious cult is reported to exist), because I had discovered their significance too late and my time di d not allow the inclusion of further centres and communities.
Th e main centres on whic h I based my investigation were: Bida town , wit h its surrounding belt of hamlets and small villages; the large villages of Kutigi , in the centre of Nupe country, and Mokwa , in the west; and finally Jebba Island on the Niger, whic h commands an important stretch of the river area. Each of these centres I visited several times. Th e repeated periods of study enabled me to check my material and to remedy oversights; often they led to the discovery of new points of view whic h only emerged from the comparison of the cultural variations typical of different areas and local communities. In addition I stayed for shorter periods in certain other communities in order to study a specific question or to collect certain comparative evidence. Thus , for example, I paid several short visits to the village Lemu , twenty miles nort h of Bida, because a new commercial crop had been adopted that year by the people of Lemu ; and once I spent tw o weeks in Patigi (a quite uninteresting visit, by the way), because there the last descendants of the old dynasty of Nupe kings hold court.
Bida, the large capital of the country, required the longest time. Here also I began my studies of the Nupe language wit h the help of the English- and Hausa-speaking teachers of Bida Middl e School. Durin g these first months I carried out my anthropological work through the medium of Hausa, whic h I had studied before going to the field, and whic h is widely understood among the people of Bida, but from the fifth or sixth mont h of my stay I used Nup e exclusively. Durin g the first expedition I spent altogether six months in Bida and the Bida area, in the course of three visits; three months in Kutigi , in the course of two visits; and one mont h each in Mokwa , Agaie, and Jebba Island. Durin g the second expedition I spent five months in Bida, in the course of tw o visits, two months in Kutigi , in the course of tw o visits, three months, in one visit, in Jebba and on the river, and two months on travels to the neighbouring area of Gbari country and to Ilorin , Ibadan, and Lokoja, where I made a special study of the large Nup e colonies in these towns. Th e difficulty of discovering in this large and complex culture repre• sentative areas and group sections, and of distinguishing them from others whose cultural life must be regarded as a variation of the main pattern, arises in a different for m wit h regard to documentation. Th e problem here is how to prove wit h scientific exactitude the typical or non-typical nature of cultural facts and events, and how to present them convincingly as representative or as in some degree deviating from the norm . Th e distinction between typical and non-typical facts is primaril y a question of frequency, and the ideal method of documentation woul d be one in whic h all descriptions and analyses were backed by statistical evidence. Fo r obvious reasons such a method of documentation is rarely
INTRODUCTIO N
applicable. So far as it was possible I have made efforts to supplement all descriptions wit h quantitative evidence to show the frequency and regularity of social facts and events. It must be understood that where I limite d myself to a few descriptions or single instances these are pre• sented (unless stated otherwise) as typical instances, and that there exists sufficient evidence of their representative nature. But I must admit that in certain cases no such evidence could be adduced: if I record the absence of a certain cultural fact or its occurrence in a certain degree of frequency, this may only mean that the fact di d not occur at all, or di d not occur in any higher degree of frequency, in the areas and groups wit h whic h I was mainly dealing, and during the time of my investiga• tion . Unhappily, the orbit of the anthropologist is limite d by time and space. Th e larger and more complex the group whic h he studies, the more strongly he feels his dependence upon that treacherous purveyor of facts, chance. It was chance that brought to my knowledge the fact that in one small and unimportant Nup e village succession to chieftainship is matrilineal—an isolated occurrence in that otherwise conspicu• ously patrilineal society. An d it was chance that led me to the dis• covery, in a small riverain village, of a certain ritua l whic h amounted to a symbolic ‘king-killing*—again a usage for whic h there seems to be no parallel in other parts of Nupe. I could discover no evidence that woul d allow me to explain these anomalies, and I have pu t them down as such, wel l aware of the possibility (slight though it may be) of such evidence existing, undiscovered, in the villages and group sections of Nupe to whic h my studies di d not extend. Th e general knowledge of the cultural constellation gained from the available material allows us to a certain extent to formulate theories, to predict, as it were, whether or not a certain cultural fact is likely to occur, or to occur in this or that frequency. But we can predict only a likelihood; we can delimit the possible sources of error: we can eliminate them as littl e as we can eliminate chance itself.
Th e presentation of the material involves also difficulties of a more technical nature. It s very quantity forbade treatment in the for m of a comprehensive monograph. No r di d the material lend itself to a neat division into a number of separate treatises each dealing wit h a different aspect of the anthropological quadrivium—kinship, political organiza• tion , economics, religion. Th e solution could only be found in a com• promise. My original intention was to buil d the book round the relation between the two paramount group configurations, political association and family—’Kingdo m and Kinship*. But this woul d stil l have meant a very unwieldy book. Th e final result, then, is a ‘rump ‘ monograph, dealing wit h the culture of Nupe viewed from the tw o angles of political and economic organization—two aspects of culture whic h must be regarded as inseparable. Even so, I had to leave out, or to reduce to a short quasi-preliminary outline, the description of two domains of culture the bearing of whic h upon the problems of the book I wil l not minimize : religion and law.
I must anticipate two further criticisms. One may be provoked by the
INTRODUCTIO N X I
rather unusual dichotomous arrangement of certain chapters. It was necessitated by the nature of the material, by the contrasting configura• tions whic h dominate the cultural life of Nupe—urban and rural culture, Mohammedanism and paganism, the organization of the state and the organization of the village community. One dichotomous formulation I have made almost the leitmotif of this analysis, namely, the twofold classification of social systems, formulated most clearly by Maclver , into Communities and Associations. Le t me quote his definition: Com• munit y is a ‘circle of people who live together, who belong together, so that they share, not this or that particular interest, but a whole set of interests wide enough and comprehensive enough to include their lives’. An d Association: ‘a group specifically organized for the purpose of an
interest or group of interests which its members have in common.’1
These two categories are used in this book consistently in Maclver’s
sense, that is, as descriptive categories defining, not rigidly separated
social entities, but rather two poles of a more fluid social reality of many
shades and grades.
Again, it might be objected that the descriptive accounts in this book are frequently interspersed wit h theoretical discussions. I shall offer no apology. Description is meaningless and misleading unless supple• mented by a clarification of the concepts and categories wit h whic h it operates, and by an assessment in terms of theoretical generalizations of the concrete data which it produces. Thi s is doubly true of the complex and, from the viewpoint of traditional anthropology, unusual culture described in this book. Thus I was forced to devote comparatively long theoretical discussions to the examination of such concepts as ‘tribe ‘ (in Chapter II ) and ‘state’ (i n Chapter VI) , to the analysis of social class in Nupe (Chapter VIII) , and to an inquiry into the relation between economics and sociology (Chapter XX) .
In the research programme of the International Institute of African Languages and Cultures, under whose auspices the present investigation was carried out, the study of culture change under the impact of modern Western civilization is assigned a prominent place. Culture contact in Nupe is not restricted to this most recent historical phase; nor does it affect only certain well-defined domains of social life—i t pervades the whole history and the whole living culture of the people. No r can it be said (wit h some anthropologists) that the study of modern culture contact entails specific methods of approach, applicable only to this one aspect of social development. It di d not therefore seem feasible to con• centrate its examination in special chapters. Every description and analysis contained in this book bears to lesser or greater extent also on the problems of culture contact and culture change, ancient or modern. Only in a few instances di d the material lend itself to a separate treat• ment of modern culture contact. They wil l be found in Chapter X, under the heading ‘Problems of Adaptation’, in Chapter XVII , whic h is entitled ‘Industrial Labour’, and in Chapter XIX , in the section on
‘Moder n Business Methods’.
1 R. M. Maclver , Society, Its Structure and Changes, 1931, pp. 9, 12, 13.
Xl l INTRODUCTIO N
A minor technical point, in conclusion. In the orthography of ver• nacular words I have used the phonetic script generally accepted to-day by students of African languages, which, in the case of Nupe, contains only two not very common symbols: ~ for nasalized vowels, and z for the fricative sound (like s in ‘lesion’). I have not distinguished open and closed e and o vowels since this distinction is not of great importance in Nupe. And, for the sake of simplicity, I have not marked tones, although Nupe is a pure tone language; or father, I have reduced the symbols for the five tones of Nupe (low, medium, high, rising, and fall• ing tone) to a single accent (‘) indicating the relative higher tone—a mark which the reader unfamiliar with tone languages can interpret a9 accent indicating stress. In words where no accents are marked, the different syllables are of equal tone or stress. I have marked the low tone only in the case of the negation, which consists of the suffix a pro• nounced on the low tone (a). In the spelling of geographical names I have kept largely to the spelling used in official maps (e.g. Kacha instead of Kaca). Only where the official spelling is at variance with the correct pronunciation have I corrected the spelling (e.g. Gbako river for Bako river, Egga for Eggan, &c) .
There remains the last task of Introductions—acknowledgements. Th e first acknowledgements are due to my teachers and friends, to whom I owe most of my scientific equipment, the late Professors Malinowski and Seligman; to the International Institute of African Languages and Cultures, as whose research fellow I went to the field and whose subsidy rendered the publication of this book possible; and to the Government of Nigeria and to the many officers with whom my work has brought me into contact for the generous help in many forms which they extended to me throughout my stay in Nigeria. I am greatly indebted to Professor Radcliffe-Brown for many fruitful suggestions and criticisms. I owe special gratitude to Sir Gordon Lethem, Governor of the Lfceward Islands, late Lieutenant-Governor of Northern Nigeria, who first drew my attention to the Nupe tribe, and to the Rev. Dr . G. P. Bargery, who initiated me into the secrets of Hausa and the world of Nigeria. Mr . T. E. Alvarez of the Church Missionary Society in Bida and Rev. Ira W. Sherk of the United Sudan Mission in Mokwa have been most kind friends and helpers to the new-comer in a country of which they had lifelong knowledge. Last but not least I owe deep gratitude to all my native friends without whose freely given help my work would not have succeeded: Etsu Ndayako, the Emir of Bida; Mallam Aliyu, the Head• master of Bida Middle School, who was my teacher in Nupe and my first guide through Bida life; Mallam Ndayako, whom I was proud to call my friend; old Ndakotsu, who is dead now, and who was so close a friend of mine that the Bida people knew me as the nasara Ndakotsu— Ndakotsu’s European; and many others whose words and opinions are quoted in these pages.
CONTENT S
FOREWOR D b y Lord Lugard . . . . INTRODUCTIO N
I . ENVIRONMEN T
Th e country—vegetation and natural resources—climate—health and disease—population movement
I I . CULTURE , TRIBE , SOCIET Y .
Problems of definition—the tribe—the sub-tribe
I I I . INTRODUCTIO N T O NUP E KINSHI P .
IV . TH E COMMUNITY : VILLAG E AN D TOW N .
V . POLITICA L ORGANIZATIO N O F TH E VILLAG E .
Th e council of elders—chieftainship—village government—balance of power—binding forces
VI . POLITICA L HISTOR Y O F NUP E KINGDO M Th e state—foundation of the kingdom—the Fulani conquest
VII . POLITICA L HISTOR Y O F NUP E KINGDO M (continued)
Kingship—a day of court routine—the royal nobility—officers of state—slavery—war
VIII . POLITICA L HISTOR Y O F NUP E KINGDO M (con- tinned) . . . . . . . . .
Th e feudal state—patronage and clientship—social class in Nupe—
binding forces
IX . TH E POSITIO N O F TH E WOME N
Ranks and offices—marriage and status—polygamy and sex-morality
X . GOVERNMEN T I N MODER N NUP E
Th e administration of the Emirate—town administration—taxation—
law—problems of adaptation
X L ECONOMICS : RESOURCES Possession of land—live stock
XII . AGRICULTURE : PRODUCTION . • .
Country and crops—farming technique—rotation of crops—disposal of crops—tree cultivation and fruit industry—innovations in pro• duction—technique and knowledge—by-products of agriculture
XIII . AGRICULTURE : ORGANIZATIO N O F LABOU R .
Individual and family work—maintenance of labour—collective labour—wage-labour—-division of labour between the sexes (I )
iii-v vi-xii
I
12
27
34
44
69
87
147
157
180
205
241
xiv CONTENTS XIV . INDUSTRIES
Organization of craft-guilds—blacksmith work—brass- and silver•
smiths—glass-workers—weavers—beadworkers
XV . INDUSTRIE S (continued)
Individual crafts—meaning and future of guild organization—
division of labour between the sexes (II )
X V I . FRE E PROFESSION S
Barber-doctors—drummers and dancers
XVII . INDUSTRIA L LABOUR
X V I I I . WEALT H
Money capital—pawning
X I X . ECONOMI C BALANCE : EXCHANG E .
Currency, measures, and prices—market and trade—modern busi• ness methods—trade as a profession—division of labour between the sexes (III )
XX . ECONOMI C BALANCE : MEAN S AN D WANT S
Th e problem—standards of living: necessities of life—standards of living: cultural requirements—standards of living: status and prestige
—wants and cultural assimilation—wealth and status
X X I . ECONOMIC S AN D SOCIA L STRUCTUR E .
Economic mobility—risk and insurance—economic relations and the group
X X I I . EDUCATIO N FO R CITIZENSHI P
Mallam schools—age-grade associations—the meaning of Nupe life
APPENDICE S I- V . . . . . . .
Bibliography—Tribal marks of Nupe sub-tribes—Historical chart of Nupe rulers—Genealogy of the Fulani Emirs of Nupe—Notes on the Nupe calendar
INDE X
257
286
298
304
309
314
335
366
378
404
411
LIS T O F ILLUSTRATION S
Th e Etsu Nup e
Surrounding s o f Bid a
Th e Kadun a rive r near Gbar a
Compound s i n Kutig i
Jebba Islan d
A villag e Katamba
A Katamba in Bid a
Th e palace o f th e Roya l Hous e o f Masab a Th e palace o f th e Roya l Hous e o f Usma n Tw o villag e chiefs . . . . Th e sacred bronz e figure s o n Jebba Islan d Th e Chain of Tsoede
Etsu wit h Bodyguar d in Sallah procession \ Displa y of Horsemanshi p durin g Sallah i Cour t Hous e i n Bid a I
Th e Alkali of Mokw a I ‘
A famil y head repairin g hi s granar y
Granar y an d cookin g place i n a villag e compoun d
Efako wor k on a far m near Bid a \ Farmin g Egbe I Blacksmit h \
Silversmit h I
Mat-weaver s o n thei r wa y hom e fro m th e far m
A machin e tailo r
A n ol d potte r woma n \
Th e women’ s loo m 1
Guild-hea d o f Bid a Barbers \
Grou p o f Drummer s I On e of th e Etsu’s Kakati players
A professiona l singe r wit h he r ban d
A foo d marke t
A woma n trade r sellin g her husband’ s silverwar e
A Bid a Malla m an d one o f his pupil s
A n Age-grad e Associatio n dance
Frontispiece facing p. 4
36
38
42
46
72
144
170
226
248
264
290
296
298
300
330
382
Triba l Ma p o f Nup e
MAP S
Bida City at end
CHAPTER I
ENVIRONMENT
THE COUNTRY
HE people who call themselves Nup e live in the heart of Nigeria, in the low basin that is formed by the valleys of the two rivers Niger
and Kaduna, between 90
30′ and 8° 30′ N.L . A line drawn fro m Leaba,
on the Niger, eastward to Kataeregi marks the northern boundary of
Nupe land, another line drawn eastward from Shari (or Tsaragi) to
Abug i and on to the Niger south of Baro the southern boundary. Th e
Niger, flowing almost straight north-south between Leaba and Jebba,
divides Nupe country from Yoruba in the west; the slowly rising
country, east of Lapai and Gidi , sloping upward towards the hills of
Gbari country, forms the eastern boundary of Nupe .
It is a low-lying , softly undulating country, not more than 200 feet above sea-level in most parts. A few low ridges of broken, rocky hills stretch out, rib-like , from the river valleys, and curiously symmetrical flat-topped hillocks rise abruptly here and there from the plain . Th e highest points are no more than 800 feet above sea-level. At Jebba the Niger forces its way through a narrow passage formed by rocks, the ragged end of a short, broken chain of hills. Farther south, at Kpatagba and Baro, and above Yankyedya in the north , hills and rocks close in upon the river. Rabd, the ancient river capital of Nupe kingdom, stands on hig h cliffs of red sandstone. Between cliffs and rocks the river bed is bordered by low banks on whic h stretches of thick rain-forest alternate wit h shrub and low bush. Durin g the rains both forest-covered banks and open bush are flooded far inland and turned into vast swamps; the level of the Kaduna at Wuya rises by 30 feet; and in the Niger valley villages whic h lie on high ground in the dry season have the water of the river washing the threshold of their houses. Part of Jebba Island is flooded every year; and Nupeko, near the confluence of Niger and Kaduna, about two miles from the river in the dry season, can be reached only by canoe in the rains.
Th e bi g rivers Niger and Kaduna form the two axes of the country, an efficient, natural system of communication. It is no accident that the early Nup e kingdom grew in the area near the confluence of the two rivers, whic h were its first high-roads as wel l as its natural boundaries. A net of smaller rivers and creeks, ric h in fish, covers the country, pro• viding an efficient system of natural irrigation. In most parts of Nupe land the soil is ric h and fertile, especially in the thinl y inhabited districts west of the Kaduna, the only part of Nupe which still possesses large areas of virgi n forest. East of the Kaduna, especially round Bida, a dense population and intensive cultivation have reduced the fertility of the land. It is the typical soil of Norther n Nigeria, red earth mixed wit h laterite and ric h in silicates—a fact whic h becomes important for the native glass industry in Bida. In a few districts, round Sakpe and near
B
2 ENVIRONMEN T
Lemu , pure white sand is found, a muc h more fertile soil as far as I can judge, whic h makes the country in the eyes of the native ‘more beautiful*
—by whic h he really means ‘more useful’.. Lastly, the low broken hills of Nup e are ferruginous, and on their slopes iro n ore is mined by the natives.
VEGETATION AND NATURAL RESOURCES
Th e geographical position of Nupe country makes it a typical area of transition fro m the southern forest belt to the arid savannahs of the north . Narro w strips or patches of thick rain-forest-like vegetation accompany river valleys and watercourses; bu t then, inland, we find a country of thi n forest, shrub, and ‘orchard-bush, alternating wit h beautiful open park-land, such as roun d Bida and Agaie, where inten• sive cultivation has cleared bush and shrub. Th e plant-life, again, reflects the geographical position of Nupe midway between nort h and south. Nupe is one of the northernmost countries in whic h the oil-palm is found, and yam and the southern variety of maize are cultivated; it is, at the same time , the southernmost habitat of the date-palm, the shea- nu t tree, and mille t crops. Th e trees and plants whic h are important for Nupe production and diet are these, in the order of importance. In the drier inland areas we find the locust-bean tree (Parkia ftlikoidea), in Nupe lonci; the shea-nut tree (Bassia Parkii), koci; the indigo tree or vine (Lonchocarpus cyanescens), eci and indigo grass (Indigofera arrecta), tsentse; baobab (Andansonia digitata), huci or kuka; silk-cotton tree (Malvaceae), lembubu; date-palm, dobina; and finally the mango, man- gorof a European importation. On irrigated land and in the thick groves near watercourses grow the oil-palm and other, less important, varieties of palm-trees, kola-nut tree (Sterculia acuminata), ebi, and banana,yaba. Onl y the kola-nut tree, the mango, and partly the banana are cultivated, all other trees growing wild . Nup e cultivation comprises the follow• ing kinds of crops and edible plants and roots: three millet varieties, bulrush-millet (Pennisetum), mayi, ‘late millet ‘ (Spicatum), kpayi, and guinea-corn (Sorghum vulgare) eyi; yam, eci ; cassava, rogo; maize, kaba; ground-nuts, guzya; rice, jenkafa; sugar-cane, kpasanako; and onions, lubasai in the river marshes; further sweet potatoes, duku; beans and ground-beans, ezo and edzu ; red pepper, yaka, and various kinds of greens whic h are eaten as vegetable. Of plants utilized commercially and in native industry cotton, lulu; henna, lali ; hemp, rama or egba, rope’; and gourds, evo, are the most important. In this enumeration of crops and plants of Nupe we have left the field of physical environment in the narrow sense. Thi s ric h and varied system of cultivation reflects already a standard of civilization, technical achievement, and inventiveness of a people whic h has learned to master its environment. Some of the plants and crops mentioned have been adopted comparatively recently and introduced into Nupe from the nort h and south. As the Nup e names for hemp, henna, or cassava are borrowed from Hausa the plants themselves represent ‘borrowed cul • tural goods. Th e cultivation of onions was introduced into Nupe from
ENVIRONMEN T 3
Hausa, and partly from Bornu , in historical times.1 No r is the cultiva• tio n of yam an original cultural possession of the Nupe , whic h is proved by tw o facts: first, that the Nup e nort h and east of the Niger are un • familiar wit h certain important ways of preparing yam common in all
‘real* yam countries, e.g. the dryin g of yam ; and second, that yam, though to-day an important staple food, has stil l no place in their sacrifices, ritua l meals, and first-crop ceremonies: only corn and mille t may be used for these ritua l offerings.2 Cassava, finally, has only been introduced in the last thirt y years; thirt y years ago, I was tol d by a missionary, it was almost unknown in Nupe. Thu s communication and culture contact have helped to moul d the physical environment of to• day. Thi s manifold agricultural system, however, could not have been evolved if the natural conditions of the country had not been favourable. Th e meeting of south and nort h in the vegetation of Nupe , the coexist• ence of marsh-land and savannah country, the natural irrigation and the strips and patches of thick forest, al l this was as essential to the develop• ment of the productive system of the country as was the inventiveness and stage of civilization of its people. We cannot attempt to decide whic h factor, environment or* civilization, has had the larger share in this development. Th e rigi d distinction between environment and culture has long been abandoned as misleading and futile. For our investigation it is irrelevant. We can wel l regard the environment as it appears at present, the natural resources as they are utilized by the people to-day, as a unitary complex of factors the effect of which , upon the life and well-being of the people, we have to examine.
CLIMAT E
Nupe , lower lyin g than any of the neighbouring districts, wit h its two big rivers, their dry , blazing sand-banks in the dry season, and their vast swamps in the rains, is one of the hottest and at the same tim e most humi d districts of Nigeria. Th e river valleys especially, and the hollow in whic h Bida is situated, are damp, airless, and unhealthy. Th e following figures give the average temperature, humidity , and rainfall in Bida for 1933:3
Jan.
Feb. Mar.
Apr.
May
TABLE I
June
Aug.
Sept. Oct. Nov.
Dec.
Annual total
temp.
(per cent.)
1 It is still practised to-day by the Bornu settlers of Kutigi.
2 There is one exception: Mokwa, in the north-west corner of Nupe, where
yam is sacrificed at certain annual rites. But the exception proves the rule, for
Mokwa is the gate for cultural influences from Yoruba.
3 This table was copied from the readings taken regularly in Bida Middle
School.
4 ENVIRONMEN T
Th e highest temperature measured in Bida was 106 in the shade, only a few degrees lower than in the hottest (but driest) district of Nigeria, Bornu . Th e humidit y of Bida, on the other hand, almost equals the humidit y in the rain-forest belt. Th e combination of features from south and nort h whic h seems characteristic of Nupe country has its most unfavourable aspect in climatic conditions. Th e influence of the unfavourable climate upon the health of the natives is evident in the large proportion of people suffering from lun g trouble and rheumatic diseases. Climate and soil combine to even more deadly effects: full of malaria-carrying mosquitoes, the river forests infested wit h tsetse-fly, and all watercourses polluted wit h bilharzia, Nup e is indeed an ill – favoured country.
Th e tsetse-infected areas have played their part in Nupe history. Th e forest belt in the south, of whic h the strips of river forest in Nupe are the northernmost outposts, marks the southern boundary of cattle and horse in Nigeria. Th e parts of Nup e whic h are free from tsetse are also safe for horses and cattle (more precisely, the northern breed of cattle, whic h is not immune from trypanosomiasis). Nupe country, wit h the neighbouring Ilorin—th e southernmost country in whic h horses can be kept and to whic h the nomadic Fulani herdsmen can take their cattle— became also the southern outpost of the Fulani Empire, the southern• most stronghold of that race of cattle-people and horsemen whose spectacular conquest of Nigeria was stemmed only by the natural bul • wark of the forest belt.
HEALT H AN D DISEASE
In the Census of 1931 Nupe appears among the districts wit h the highest proportion of blindness, leprosy, and insanity.1 In the case of leprosy the Census estimate of over 5, and possibly over 10, per thousand is hardly exaggerated—even casual observation shows this, although the fear and secrecy that surround this disease among the Nupe make a more precise estimate impossible. Information whic h I obtained on the spot, from the Medical Officer in charge of Bida Native Administration Hospital, rather went to show that blindness is not more frequent in Nupe than elsewhere in Nigeria. As regards the proportion of insanity, I have seen too few cases to venture any definite opinion. A remark on the nature of the mental disorders, however, may not be withou t interest. Of the five cases whic h I have seen three showed clear symptoms of paranoid schizophrenia—strongly delusional thinking , to the extent of persecution mania, combined wit h violent aggressiveness whic h appeared and subsided wit h marked suddenness, and involving no
general degeneration of normal intellectual functions. Tw o cases were simple forms of imbecility.2
1 Census of Nigeria, 1931 (London , 1933)- Th e Census figures refer partly to the tribe , i.e. to random samples of Nup e (ii , pp . 110-11), partl y to the political district of Bida Emirate (ii , p. 391), and partl y to Nige r Province as a whole (‘Nup e Province* in the ol d terminology), the population of whic h consists to two-third s o f Nup e (i , pp . 41-2, ii , p . 38).
2 See p . 402.
SURROUNDINGS OF BIDA
TH E KADUN A RIVER NEAR GBARA
ENVIRONMEN T 5
Sleeping sickness is the great scourge of the country. A Sleeping Sickness Commission whic h visited Nupe in 1936 found 70 per cent, of the people infected—the highest figure in the whole of Nigeria. Bil – harzia and other wor m diseases are no less widespread. Th e prominence of venereal diseases is characteristic of Nupe as it is of the whole of Nigeria and indeed of the whole of semi-civilized Africa. Dysentery, malaria, and yaws complete the unhappy picture. It is wort h noting, however, that typical deficiency diseases, including boils and ulcers, are comparatively rare. A statistical table, compiled fro m the attendance figures of the N.A . Hospital and Dispensaries in Bida Division , may illustrate the distribution of the more important diseases occurring in Nupe.
TABL E I I
Th e figures for Bida Hospital refer to in-patients treated during April , July, October 1935 and January 1936. Th e figures for the Dispensary of Lemu give the attendance in January 1936. For another dispensary (Kutigi) only the order of frequency of the various diseases could be ascertained; it is indicated in the third line by roman numerals.
Bida Hos• 35 pital
Dispensary 2 1
Lem u
Dispensary I I
Kutig i
14 15 13 9
18 3 54 2
I V II I V I ?
.I II
3 5
2 68
V II
1
!
2
4
10 cases
80 per cent, of all cases
?
?
128
106
?
Th e description of health conditions has a more special bearing on the anthropology of the Nupe people. First, it is of a certain significance in connexion wit h certain cultural elaborations of the theme of disease among the Nupe; 1 second, it involves relevant questions of mortality and population movement—a subject to whic h we shall tur n presently. Finally, wit h respect to mental disorders: psychology and sociology have long realized that certain tensions and maladjustments in the social system tend to manifest themselves in the sphere of psycho-pathology. We shall see that this correlation is true also of Nupe society.
POPULATION MOVEMENT
It is unnecessary to stress the importance of population studies for the understanding of a society and the nature and working of its institutions. But it may be necessary to justify the inclusion of a discussion on popu• lation movement in a chapter on environment. A first justification lies in the fact that, in a study of social organization, environment and popula-
1 Most instances of this kind, however, I must leave for future publications. To one typical instance, the part which leprosy plays in the mythology and ritual of the gunnu cult, I have referred in the description of this cult in Gunnu> A Fertility Cult of the Nupe, in Northern Nigeria’, in J.R.A.I. Ixvii, 1937, pp.
91-130 .
6 ENVIRONMEN T
tio n movement are to some extent methodologically of the same order. Both represent in certain respects ‘data’, given facts—factors external to the field of social phenomena whic h we investigate. Moreover, popu• lation movement and environmental conditions are inseparable from one another. Population movement reflects the effects of a complex of heterogeneous factors of whic h physical environment is one of the most important. It bears on possibilities and means of production, on-nutri• tion , health, and livelihood in general.
But we shall also encounter the effects of other factors upon popula• tio n movement, namely, specific ‘cultural* factors—social and economic conditions, ‘standards of living’ , moral concepts—and certain historical events whic h have influenced the distribution of population in Nupe country.
Birth-rate and Fertility
Th e Nigerian Census places Nupe birth – and fertility-rate at the bottom of the scale.1 Th e proportion of live children per mother among the Nupe is i-8 (the Yoruba are the only other tribe wit h an equally low rate), as against an average for all tribes of 2-6, and a fertility-rate of 24 in the tribe wit h the next lowest figure. Fifty-eight per cent, of Nupe women in the age group 30-39 are sterile, as against the total average for Nigeria of 227 , and the next highest sterility rate of 27-1 among the Yoruba. These figures contain certain inconsistencies, and leave certain relevant questions unanswered. Thu s the rate of live children per mother is equal among Nupe and Yoruba, while the proportion of sterile women among the Nupe is exactly twice the corresponding figure among the Yoruba. Thi s would impl y that infant mortality among the Yoruba is twice as large as among the Nupe—whic h seems hardly probable. Moreover, the proportion of live children per mother does not indicate to what extent the low figure is due to still-birth , infant mortality, con• stitutional sterility or, possibly, birt h restriction. Finally the statistics ignore the sociologically most important distinction between rural and urban areas.
Th e last point can to some extent be answered indirectly. Th e follow• ing figures give the birth-rate of three Nupe towns and one district (Kontagora) whic h has a very large proportion of Nup e among its in • habitants. Thoug h only approximate these figures are wel l wort h quoting in this connexion as they refer to localities representing different degrees of urbanization: the purely urban area of Bida tow n side by side wit h the much less urbanized village-capitals of Agaie and Lapai Emirates.
TABLE II I
.
. .
1 Census of Nigeria, i, p. 55.
ENVIRONMEN T 7
In the course of our investigation my wife and I attempted to ascertain further statistical data whic h could thro w light on these points. The y were obtained by the method of house-to-house investigation in three selected localities, in the villages Kutig i and Jebba Island, and in one tow n ward of Bida. Th e scale of this statistical research, carried out single-handed, was of necessity very limited . Nevertheless, certain significant facts could be ascertained (Tables IV and V) .
Villages
Tow n
Mothers
(age group
30-40)
101
144
Total children born
206
230
TABLE IV
Children alive
108
126
Children dead
98
104
Births per mother
2 i’5
Live children per tnother
1• 06
0•8
TABLE V
Children died at age of
Mothers Birth 1-2 3-5 6-8 9-15 Total
Villages 101 29 25 27 8 9 98
Tow n 41 21 15 27 7 4 74
Table V shows, first, the distribution of infant mortality over different age groups. Infant mortality in the villages seems, on the whole, con• siderably lower than in the town , remarkably so at, or shortly after, birt h and in the first five years of the infant’s life. Thi s seems in the main due to the spread of venereal infection in the town , and also to the effects of abortion and artificial birt h restriction which , as we shall see presently, is practised in Bida. Th e highest mortality, both in urban and rural districts, occurs at the age of 3-5. Thi s peak of mortality seems due primaril y to malaria, whic h is especially fatal at this age.1
Table IV (examining a larger group from the urban area) attempts to establish a comparative birth-rate for urban and rura l areas. For the low fertility-rate of the Nupe women in general two physiological explana• tions suggest themselves: first, reduced fertilit y may be the result of certain endemic or widespread diseases, such as sleeping sickness and gonorrhoea; or second, it may be the result of prenatal deficiency of vitamin E (a vitami n contained, for example, in palm-oil). Th e effective• ness of these tw o physiological factors cannot, be denied.2 However, the striking difference in the birth-rate of tow n and villages demands a different interpretation. Th e distribution of venereal diseases, definitely higher in the tow n than in the rural area, woul d fit into the picture. But of sleeping sickness rather the opposite is true ; and the vitami n deficiency
1 For this information, as for the other medical information in this chapter, I am indebted to Dr. M. Ellis, of the Nigerian Medical Service, who at the time was in charge of Bida N.A. Hospital.
2 A study of diet among the Nupe has shown that the women’s share of the
daily meals (i.e. the meals cooked with palm-oil) is only about one-fourth of the men’s share.
8 ENVIRONMEN T
should be less marked in the town , where ric h food is the rule. Physio• logical factors are thus not sufficient to explain the low fertility-rate in Nupe. Thi s is confirmed by the data of Table IV . For in the context of this table, whic h refers only to women who have had children and ex• cludes sterile women, the difference in birth-rate precludes to some extent the explanation on the ground of constitutional sterility and suggests instead the influence of another factor: the existence of birt h restriction in the town , in other words, the effects of social motives.
Nup e religion and morality, as we shall see later, emphasize very strongly the idea of progeny. Nevertheless, in the higher social strata, especially in Bida town , the women often prefer childlessness, or at least do not want more than one child . Childlessness opens to them certain economic opportunities of whic h women wit h children, or many children, cannot avail themselves as easily. Th e Bida women know how to brin g about abortion, and even know the use of contraceptives—or rather of certain practices and drugs whic h are believed to work as contraceptives. Man y of them are clearly mere superstitions and the expression of magical beliefs; whether certain drugs whic h one can buy (i n the greatest secrecy) fro m medicine-men and barber-doctors are more effective, I am unable to say.1 However this may be, sterility of the Bida women is, to speak wit h Durkheim , a typical ‘social fact*. In our own terminology, it represents a typical case of that ‘indirect ‘ influence of environment whic h determines population movement, working through considerations of economic benefit and standard of living . An d this influence becomes evident in our case not only as an abstract social trend , derived from statistical correlations, but stil l more in the for m of conscious and tangible behaviour.
Sex Ratio
Statistical figures show a surprising fact: a marked preponderance of females among the adult population, bu t hardly any preponderance of females—indeed, in Bida town , the reverse relation—in the sex ratio of non-adults.
TABLE VI 2
Adults Non-adults
Females per head of male population
Bida Emirate Agaie-Lapai Emirate . Bida town .
Bida Emirate excl.
of Bida town .
Male
57,759
20,148
7,714
50,045
Female
66,546
23,915
11,602
54,944
Male
22,949
6,139
3,240
19,709
Female
24,204
5,938
2,675
21,529
Adult
115
1•I
1•5
1•1
Non-adult
I- I
0-9 o-8
1 1
1 The Nupe ‘doctors* know two kinds of contraceptives: temporary contra• ceptives of a magical nature (e.g. a magical belt), and permanent contraceptives— powders which are taken internally. These drugs are considered ‘evil’, immoral me2dicines, but are made and sold all the same.
Census of Nigeria, ii , pp. 176 et seq.
ENVIRONMEN T 9
Thi s discrepancy is partly explained by the fact that girls marry at an earlier age than boys, and also by the fact that polygamy, especially the large-scale polygamy in Bida town , implies marriage of young girls to much older men. Marrie d girls being counted in this table as ‘adults’, these factors must upset the ‘normal ‘ sex ratio in the higher age groups of non-adults. But this explanation only underlines the general social problem emerging from these figures: how are we to interpret the diver• gence between the sex ratio among adults (implie d in polygamous marriage), and the sex ratio among non-adults? In other words, how can the sex ratio in actual polygamous marriage be harmonized wit h the sex ratio on whic h such marriage potentially depends? There is, first, the heavier mortality among men, whose numbers, only a generation ago, were constantly reduced in the almost permanent wars in Nupe . But the comparison between the figures referring to Bida tow n and to the rural districts in the Emirate respectively reveals that there is also another influence at work. In Bida tow n the difference between the two sex ratios is 0-7, in the rural districts the figures are identical. Tha t means that marriage in the villages—monogamous marriage, or poly• gamous marriage on a small scale—can rely on local women, whereas Bida must import its wives from outside. Formerly slave girls con• stituted a large contingent of this urban importation of wives-to-be; to • day, in smaller proportion, girls from the country drif t into Bida. We shall see later that genealogies and marriage statistics fully endorse this explanation.
Population and Emigration
Translated into terms of general population movement, the strikingly low birth – and fertility-rate among the Nup e can mean only one thing : a decreasing, or at least a stationary, population. Th e general population statistics in the Census can, on the whole, be taken to confirm this con• clusion. Bu t the nature of the available statistics does not allow more than this indication of a general trend. Th e total population of the Nupe tribe was given as 349,008 in the Census of 1921;1 ten years later the figure is 326,017. These figures, implyin g a fall of over 20,000 in ten years, cannot be accepted as anything bu t a very crude approach to the true situation.
Th e difficulties whic h stand in the way of exact population statistics in a West African territory need not be stressed specially. Yet the earlier Census figures were likely to have underrated rather than overrated the population; wit h the improved technique of registration (e.g. for taxa• tion) in the last ten to fifteen years one woul d have expected an increase rather than a decrease in the numbers that appear in the Census, even if the population had remained stationary or had decreased slightly. Certain other sources of error, however, seem not to have been elimin • ated, above all , inconsistencies in triba l classification. Th e native and British officials who compile the Census lists are frequently insufficiently acquainted wit h triba l and sub-tribal names and distinctions.2 Thus,
1 See C. K. Meek, The Northern Tribes of Nigeria, Oxford, 1925, vol. ii .
2 It may therefore well be that the 1921 Census which was compiled under the
10 ENVIRONMEN T
for example, the Gana-Gana (Dibo ) sub-tribe of the Nupe is classed sometimes under ‘Nupe ‘ and sometimes as a separate group. In the Census of 1921 their number is given as 19,700, in the Census of 1931 as
2,897’/ m t 0 whic h triba l category the rest of them disappeared I have been unable to discover. A last example: only in 1936 di d the Adminis • tration of Ibadan Divisio n discover that the people who were listed in their statistics under the heading ‘Tapa ‘ (Takpa) were identical wit h the
‘Nupe* of the Norther n Provinces, Takpa being the Yoruba name for
Nupe.
Th e difficulties and sources of error whic h render triba l statistics so uncertain do not affect in the same degree population statistics com• piled for certain single Provinces, political Divisions, and especially townships. Here the improvement in registration has had its ful l effect; and I may add that in the preceding discussion I have used only these more reliable statistics. Here are certain figures illustrating population movement in Niger and Ilori n Provinces—the tw o Provinces in whic h the bulk of the Nup e people are living .
TABLE VI I
General Census ig2i General
Census
1931 Local
Census
1934
Bida Emirate . 171,958 171,458 168,247
Bida town 29,848 25,231 23,286
Nupe in Ilorin Province . 64,652 65,040 53,600
Nupe in Niger Province . 233,840 226,132
Again we find a marked decrease in population and a decrease whic h cannot be dismissed as due to miscalculation. Evidently, a decrease in population even considerably smaller than that shown in these figures cannot be due merely to unfavourable birth – and death-rate. Emigration seems the only alternative, and I have indeed found evidence of this. A continuous emigration on a small scale and the existence of a floating population of traders and craftsmen driftin g to the prosperous centres in the south could be established beyond doubt.2 Th e study of life-histories and genealogies among the Nupe shows further that emigration of considerable importance occurred, roughly, in the last six to ten years. Th e drift of the emigration was again mostly south, to Ilorin , Lokoja, Ibadan, Oshogbo, and Lagos, that is, to the wealthier and, above all , less heavily taxed districts of Southern Nigeria.3 In • direction of an anthropologist, Dr. C. K. Meek, is in many respects more correct than the more comprehensive and elaborate recent Census.
1 Census of Nigeria, ii , p. 70 ; on p. 30, however, there are, again, 18,220
Gana-Gana.
2 A census of Nupe people living at present in Lokoja, for example, showed the following distribution of professions among 60 family heads: 2 farmers, 4 mechanics, 5 barbers, 15 tailors and embroiderers, 4 black- and brass-smiths,
9 Koran teachers, 3 mat-makers, 2 carpenters, 5 canoemen, 5 N.A . messengers, interpreters, and other small officials.
3 These are the comparative tax figures for 1934-5 per adult for Bida Emirate
and Ilorin and Ibadan respectively: Bida town, ys.-ios. 6d.; Bida Emirate,
ENVIRONMEN T I I
vestigations among the Nup e community of Ilori n and Ibadan have elicited the information that in 1934-5 a bi g wave of Nup e emigration had reached these towns.1 These tw o years, my informants explained, had been bad years, wit h a severe food scarcity in the north , high prices, and shortage of money. Th e reason is indeed convincing: 1933-4 marked the nadir of the economic depression in Norther n Nigeria, whic h almost paralysed trade and native industry. To the mobile professions of trader and craftsman emigration appeared in many cases as the natural solution.
Density of Population
Th e following table illustrates the density of population in different parts of Nupe country. Th e first three figures are taken fro m the Census of 1931, the other figures, referring to districts of Bida Division , are compiled from the local Census of 1934.
TABLE VII I
Density per Density per
Locality sq. mile Locality sq. mile
Bida Division 35*22 Mokwa district 12
Agaie-Lapai Division 217 0 Kutigi „ . . . 13
Lafiagi-Patigi Division 30-00 Districts west of Kaduna
Badeggi district 2 2 river (‘Trans-Kaduna’) 17 6
Gbangba ,, 5 0 Districts east of Kaduna
Jima-Doko „ 53 river (‘Cis-Kaduna’) . 55
Kacha ,, . 60
Th e striking difference in density of population east and west of the Kaduna river needs explanation. It lies in the historical fact of the Fulani conquest of Nupe and the immigration of the Fulani rulers and their huge army of warriors, slaves, courtiers, and other dependants into the area east of the Kaduna, where they settled, occupied the land, and buil t their capital and numerous villages. Th e history of the conquest and the political reorganization of the country whic h it entailed wil l be described later in more detail. Bu t here already we realize the far- reaching influence whic h this difference in population density must have upon the social organization of the country, and upon the distribution of natural resources on whic h its economic life depends.
districts, 9s. 6d.-155.; Ilorin town, 9s.; Ilorin Emirate, districts, ios.-ns.; Ibadan town, 7s.
1 During 1935 200 people from Bida Emirate migrated south. (Personal
information received from the District Officer, Bida Division.)
CHAPTE R I I
CULTURE, TRIBE, SOCIETY
HE description of environment is concerned wit h a solid, easily defined, almost self-evident reality. A description of the general demographic and cultural constellation deals wit h the fluid entity of groups and systems of unity , the mere circumscription of whic h involves already theoretical definitions and classifications. These definitions and classifications must be based on facts whic h only the ful l analysis of the society can establish. Thi s chapter wil l thus have to anticipate certain results of the general description whic h forms the main body of this book. It wil l not attempt to anticipate the descriptions themselves. It s aim is only to sketch in a background, to stake out a field of demographic and cultural facts by whic h is bounded that social reality wit h whic h this
book is dealing.
PROBLEMS OF DEFINITION
Le t me draw a triba l map of the part of Nigeria where Nupe country lies: in the nort h the neighbours of the Nup e are Kamberi , Kamuku , and Hausa; in north-east and east lies Gbar i country; in the south, on the river, the country of the Kakanda; and west of the river, Yagba (a section of the Yoruba) and Ilori n Yoruba ; and in the north-west, finally, the people of Borgu and Bussa. Colonies of Nupe , small sections, split off from the main body of the tribe , are scattered all over Nigeria, especially in villages and towns on the Niger. Th e largest of them , a group of about 15,000 people who call themselves Bassa, or Bassa Nge, lives near the confluence of Nige r and Benue. The y appear to have emigrated from the mother country about a hundred years ago, before the Fulan i conquest; they settled first on the right bank of the Niger , on the hills of what is to-day Lokoja, and moved later, when the new tow n of Lokoja was founded by the Niger Company, to the left bank where they are livin g to-day.
Th e Nup e are known as Nufawa (sing. Nufe) among the Hausa, as Abawa (sing. Bawd) among the Gbari of Kuta , and as Anupeyi and Anufawhei among the Gbari of Paiko and Birni n Gwari ; the Yoruba know them as Takpa and the Kakanda as Anupecwayi. Th e Nupe call themselves Nupeci or, in the plural , Nupecizi, their language Nupe, and their country kin Nupe, Lan d of Nupe . No special wor d for * tribe ‘ exists in Nupe ; the Nupe speak of za Nupe, Nup e man, Nup e peoples.
Thi s book is concerned wit h the culture of the people that calls itself Nupe. But what does ‘people* mean here ? What is the social reality to whic h the ter m Nupe people or Nup e tribe refers ? Th e three meanings of Nupe, people, language, and country, do not coincide. Disregarding the split-off sections we may say that the Nupe tribe inhabits, roughly, the same territory. Bu t it is not, and never was, ‘one country* in the sense of a group united politically, or by facts of close contact and
‘common life’, thus marked off from other, similar areas or groups. O f
CULTURE , TRIBE , SOCIETY 13 such facts of common life whic h woul d lin k the whole tribe we find only
one—a weak, potential link , determining individual action only, namely,
the rule of hospitality whic h every Nup e must extend to his fellow tribesmen (but whic h he woul d not refuse to strangers).1 Otherwise the common life that exists in Nupe takes its course in a very muc h more narrow compass, in scattered villages, towns, or groups of villages— small circles on the spacious triba l map.
Politically the Nupe people are scattered over several ‘countries’, seven Emirates and chieftainships to-day, and four or five in pre- European times. Even the ancient Nup e kingdom has not, as far as our evidence goes, included the whole group ; sections of Nupe have always lived outside its boundaries, and other tribes or triba l sections beside the Nupe have formed the nationals of these countries of Nupe. To-day the largest of them is Bida Emirate wit h a population of 171,458, the core of the ancient kingdom of Nupe whose present rulers stil l take the titl e Etsu Nupe, Kin g of Nupe. Some of the other, smaller Nup e states are ruled by independent kings (i n modern terminology, Emirs) : Patigi and Lafiagi, comprising an area of 3,023 square miles and a population of
90,702, and Agaie and Lapai, wit h an area of 2,587 square miles, and populations of 24,174 and 25,000, respectively. Other Nupe territories are but small chieftainships, political districts, under the rulership of Hausa, Yoruba, or Gbira kings: Zuguma (wit h a population of 5,000) under the Hausa Emi r of Kontagora, Shonga and Shari (appr. 17,000) under the Fulani-Yoruba Emi r of Ilorin , and Egga and its hinterland (5,000) under the Atta of Gbira. In Bida, Patigi, Lafiagi, and Agaie the Nupe represent the vast majority of the population; in Kontagora, Lapai, Ilorin , and Gbira they are a small minority , livin g side by side wit h Hausa, Yoruba, Yagba, Gbari , Kakanda, and Gbira .
If the Nupe tribe is thus not a local group in the strict sense of the word , it is not a linguistic uni t either. Nupe is spoken to-day by the bilingual Gbari of Paiko and Lapai and the Kakanda on the Niger. On the other hand, some of the split-off sections and colonies of Nupe, though stil l laying claim to triba l membership, have abandoned and for• gotten their mother tongue. True , these facts reflect only comparatively recent historical changes. But language as such, in its very structure and organization, proves the same divergence between the two social realities, tribe and linguistic group. Nupe language belongs to the group of Sudanese languages, and withi n this category forms, together wit h Ibo , Edo, Ijo , Andoni , and Yoruba, the ‘Kwa-group’ of West African languages.2 Nupe is related more closely, in structure and vocabulary, to Gbar i and Kakanda (I am speaking here only of languages whic h I have studied myself). But to speak of Nupe language pure and simple is incorrect. There exist to-day five markedly different dialects: a dialect which we wil l ter m Nupe proper, and the dialects of Ebe, Zitako, Bassa,
1 There exists a fine distinction, though: fellow-tribesmen who claim hospitality are, where possible, given a sleeping hut to spend the night in, and are often invited to share a meal with their hosts; strangers sleep in the entrance hut of the compound.
2 See D. Westermann, Africa, viii, p. 145, 1935.
14 CULTURE , TRIBE, SOCIETY
and Kupa. Nupe proper differs from Ebe, or Bassa from 2itako, no less than Nup e differs from the languages of the other, related, tribes, Gbari and Kakanda. Dialects, language, language group, or linguistic class— these are all fluid categories, shading over into one another, and none of them capable of being identified wit h tribe or people.
Political group and language group thus impl y the existence of the tribe, they do not explain it . Th e objective reality of political and linguis• tic organization leaves undefined the nature of that elusive human sub• stratum, tribe or people. No r can we count on any external symbols of triba l unit y to help us wit h the definition. No triba l costume, or triba l emblems, exist, no tattoos or face-markings characteristic of all Nupe.1
Yet to the Nup e man himself the Nup e tribe or the Nupe people is a clearly defined, unmistakable entity. H e wil l rarely hesitate whether to describe the population of a district or village as Nupe or to put it down as alien, non-Nupe. Thi s is, of course, an expression of traditional, historically accumulated, and transmitted knowledge. People and tribe themselves represent but the product of complex historical processes, contacts, and ramifications. These processes are no longer traceable to • day ; but even if we could trace them they woul d tel l us nothing about the nature of this social fact whic h we witness to-day, the classification of groups whic h do not speak the same language and do not live together in
‘one country’ unified by political organization, under the heading of ‘one tribe’ . Th e only definition of tribe that seems possible is the tautological statement that a tribe or people is a, group uni t the members of whic h claim to for m a group unit . Bu t perhaps there exists some other, objective, social reality whic h can help us to define this evasive concept ?
We have not examined yet that paramount social reality, culture. But culture no less than language or living-together seems a fluid criterion of triba l unity . Th e only definition of tribe that we can give in terms of culture is to call the tribe a group possessing more or less the same culture—as it possesses more or less the same language or lives more or less in contiguity and ‘togetherness’. In Nupe the uni t of culture is in certain respects smaller than the uni t of the tribe. Th e different
‘countries’ of Nupe represent different, and differently organized, cul • tura l ‘provinces’. Culture varies wit h environment and the productive organization whic h it entails; it expresses the changing dominant interests involved in social, political, and economic conditions. We can thus speak of a riverain Nupe culture, a hunters’ or peasant culture, and distinguish tow n from rural culture, or the culture of the rulin g group from the culture of the subjects and commoners. Th e cultural uni t is also wider than the uni t of the tribe. Nupe shares its political and social system wit h many tribes in West Africa, its traditional religion wit h neighbouring groups in the north , east, and south, its modern religion, Islam, wit h the whole Sudan. There is, in fact, good sense in speaking of a West African culture, or a culture of the inland groups of West Africa (as against the culture of the groups in the rain-forest and near the coast). Finally, culture appears also, as it were, crystallized in the for m of
1 See below, p. 22.
CULTURE, TRIBE, SOCIETY 15 a tribal culture, and the area of cultural unity reveals itself in some
respects as coextensive with the tribe.
This is not merely an enumeration of isolated ‘ traits’ selected at
random. Each cultural factor which we reviewed affects, and is inter•
dependent with, a wide range of other cultural factors. Culture, as
anthropologists are realizing, is not a sum total of discrete traits; it
possesses a definite structure and identity; its traits or elements, inter•
dependent and mutually determined, build up a ‘pattern’ that tends
towards coherence and consistency. This consistency—a functional con•
sistency—reveals its existence to the student of the society in those
cultural characteristics which he learns to regard as ‘typical’ or ‘non-
typical’ of a culture. Each of the three cultural units which we enumer•
ated shows, in some measure, the features of this consistency and
identity. We will not here probe deeper into the nature of this con•
sistency which seems to crystallize in different degrees of intensity and
comprehensiveness. But we must clarify our concept of culture. So far,
we have used the word in a rather vague sense. It is not the same culture,
the same category of culture, which we discovered in these three rela•
tions between cultural and tribal unit. Cultural unity is, speaking
generally, unity and correspondence in forms of social behaviour. When
speaking of culture units wider than the unit of the tribe, we referred to
a correspondence of behaviour among disconnected groups which we as
students of cultures and peoples discover, classify, and name; it is a
correspondence upheld by the unconscious processes involved in tradi•
tion and the perpetuation of custom; it impresses upon the groups a
certain external uniformity. When speaking of the smaller cultural units
which reflect common environmental conditions and common interests,
common spheres, that is, of organized social life, we refer to something
that goes beyond an externally visible uniformity. We mean a conformity
of behaviour, a conscious unity created by, and embodied in, a common
group life and the mutual control and adjustment of behaviour among the people who form the group.1 The identification of cultural group and tribe implies, in the observer’s interpretation, something of both.
But were this all it would enable us only to define the tribe as a natural species is defined—unknown to the subject of the definition. It would amount to colouring a certain space on a cultural map and calling it
‘Nupe tribal culture’. It would not explain the crucial point in our argument, that the people are themselves conscious of the reality of the tribe. But culture has also another, new, meaning: it represents the conception of the people themselves. To the Nupe the tribe possesses a certain culture, specific to it and only to it, irrespective of variations within or similarities with groups without the tribal boundaries. Their awareness of this unity is expressed in the form of almost axiomatic statements, ‘the Nupe do this or that’, ‘this and that is a Nupe custom’,
1 Th e historian may attempt to link the two, and derive ‘uniformity’ from an original ‘conformity’, that is, explain the present cultural correspondence among disconnected groups by cultural diffusion from a centre in which culture was identical with organized common life. But considerations of this kind remain outside this discussion.
i6 CULTURE , TRIBE, SOCIETY
contrasting Nupe usages wit h customs and forms of behaviour character• istic of other tribes. Yo u discuss, for example, the worship of human figures and idols of the kin d you see in every village and every house in neighbouring Yoruba ; or the face-like masks whic h the Yoruba and many southern tribes wear in their ritua l dances; or perhaps the cere• mony of spirit possession whic h the Hausa wh o live in Bida tow n per• form on Bida market. You r Nupe informant wil l tell you positively: no
—we have none of this ; only Yoruba do it , or Hausa do it , as the case may be. Or you ask a Nupe man what all Nupe have in common. He will , as a rule, enumerate the same few traits. Al l Nupe , he wil l say, have the enagbarufuzi (age-grade associations), all Nup e have the same system of tici (ranks), and all Nup e have the gunnu cult. Of the Nupe community in Lagos, a group whic h has forgotten its traditions and abandoned its language, I was told that they were stil l Nupe, because they stil l practised the Nupe kuti (ritual) of the ndako gboyd (a ritual closely linked wit h the gunnu). Yet on closer examination this shib• boleth of Nupedom proves only half true. Worship of idols and human figures exists also among the Nupe , in the riverain area; mask dances wit h face-masks are practised in a large section of the Nupe , in Mokw a and the neighbouring villages. Th e age-grade associations are common both to Nupe and Hausa. Th e rank system is common all over Norther n Nigeria, bu t the details in its organization (very significant details) vary considerably among different groups of the Nupe. Th e gunnu cul t exists also among Gbari, Gara, Basa Kom , and, what is more important, appears in Nupe itself in strikingly different forms in different parts of the country.1 Finally, as regards the Nupe colonies outside Nupe country : I have visited the Nup e communities in Lagos and Ibadan. They are descendants of Nupe slaves and serfs, completely Yoruba-ized to-day in language, custom, and everyday habits. It is true they have the ndako gboyd (which they call by the Yoruba name igunnu). Bu t the ndako gboyd is completely absent among the eastern section of Nupe , the Dibo , and also, I believe, among most groups of the Ebe, the northern section of the Nup e tribe.
Objectively, then, the rigid distinction of Nupe and non-Nupe on the basis of cultural characteristics often proves incorrect, and must do so, since culture is both wider and narrower than the tribe. True , there exist certain traits, or ‘patterns’ of traits, whic h are typical of all the Nupe, and possibly only of the Nupe—facts of material culture, social organiza• tion , religious beliefs. Lik e the Nupe who mean to uphol d the identity of their tribe, I shall speak of things ‘the Nupe do ‘ or ‘do not do’. Tru e it is that the people admit the existence of certain cultural variants whic h qualify the cultural unit y of the tribe . Bu t the essential fact to us here is not the objective constellation of cultural traits. Th e identification of Nupe tribe wit h a core of Nupe culture represents a concept, a spiritual reality, rather than a critical interpretation of cultural facts. It reflects not the quasi-scientific calculation of more essential against less essential traits, or of a majority of common characteristics as against a minorit y of
1 See J.R.A.I. lxvii, 1937, pp. 91-130.
CULTURE , TRIBE , SOCIETY 17 dissimilarities, bu t a phenomenon of a different order, a theory and
ideology whic h ignores divergences ex deftnitione. It represents, in one
word , the theory of the Nup e tribe . Ou r tautological definition of trib e
then reads thus : a tribe or people is a group the members of whic h claim unit y on the grounds of their conception of a specific common culture.l To define the unit y as conceptual does not make it less real or less significant. Th e awareness of unit y is the basic element in al l social unity . Consciousness of commonness is the primary element in social commonness. ‘Communities, for all their external marks, are not objective things, they are spiritual realities.’2 Th e tribe is not a com• munit y in the strict sense of the word : it lacks, as we have seen, the element of ‘common life ‘ and co-activity. But communities exist in different degrees of intensity; they have a different compass of social inciusiveness; there is not one community, but a complex hierarchy of communities.3 Th e consciousness of a unifor m culture defines the widest, loosest uni t in the hierarchy of communities—potential rather than actual ‘common life.’ One turns easily int o the other. ‘One of the great processes of society .. . is that thereby the common is buil t out of the like, common interests growing out of what were at first merely like interests.’4 We add that the consciousness of uniformit y is
the first step towards conformity of culture.
At the same time this uniformit y of conduct whic h is conscious of itself is more than merely a potential stage of social existence. It repre• sents, in itself, a specific social reality. Le t me define here a vital concept of social existence whic h wil l occur throughout this book: social integration. Th e essence of integration is co-activity—potential or actual; social integration is predictable co-adaptation of behaviour in a group. It reveals itself in the certainty wit h which , in the group, certain actions of individuals or group sections wil l call fort h appropriate response actions of other individuals or group sections. No w uniformit y of behaviour is, sociologically, meaningless. Conscious uniformity , however, is predictable uniformity ; the knowledge that a certain familiar pattern of activity wil l be repeated endlessly throughout the group represents a first form , and a degree at the same time, of social integration.
These tw o types of integration, uniformit y and conformity, do not exhaust the forms of unit y of behaviour existing in society. We shall meet in this analysis also wit h a thir d interpretation of cultural common• ness or likeness, namely, the co-adaptation of behaviour involved in all organized associations. We shall speak of it as co-ordination of behaviour, for it represents essentially a planned,controlled organization of behaviour in the group, regulated and enforced by specific social mechanisms. In
1 This definition claims to hold true for all large tribal groups (‘peoples’) of a similarly complex structure. It does not apply to smaller and more homo• geneous groups. Th e factor of size and numbers is of importance. Evidently, for a small group on a Pacific Island tribe equals local group; or among American Indian tribes the tribal unit may coincide with the political group (see A. L. Kroeber, Anthropology, 1923, p. 232).
2 R. M. Maclver, Society, p. 66. 3 Ibid. , p. 10; Community, p. 23.
4 R. M. Maclver, Society, p. 8.
C
18 CULTURE, TRIBE, SOCIETY
this book we shall be dealing mainly with the political association, which seems characterized by two sets of factors: first, by its in- clusiveness, being the relatively widest unit of co-ordination, and, underlining it, by the conception of a collective identity emphasized, above all, in inter-group dealings. And second, by the nature of its regulative mechanism (here the ‘Government’), which rests, in the last resort, on the employment of force and coercion, and which controls both co-ordination for the purpose of internal stability (‘law and order’) and co-ordination for the purpose of external, inter-group, action (‘offence and defence’). Cultural groups and communities have fluid boundaries; the association (in our case the political unit) is or is not rigid, but its boundaries—the boundaries of the group and sphere of conduct within which the co-ordination is to be valid—are by definition rigid. Th e consciousness of this validity defines the membership of the political association—in modern terminology, nationality. Th e political unit, unlike culture and community, is unequivocal: one either is or is not
a citizen of a country; one either comes or does not come under its laws and restrictions. However, one remains at the same time a representa• tive of a culture which may not simply coincide with the cultural uni• formity upheld by the political system. One remains also a member of a community or of communities which may exist apart from the ‘com• mon life* embodied in political co-ordination. Planned co-ordination of behaviour cannot ignore these other forms of integration—cultural like• ness and commonness. Whether it utilizes them, creates them anew, or simply accepts their existence, they are the cultural raw material out of which it must fashion its unity.
Th e three planes of integration, and the group units which they define, thus cut across each other, overlap, and possibly grow out of each other, It will not always be easy to separate them clearly in the concrete de• scription. Th e preceding analysis has to some extent defined their theoretical boundaries. In what follows we can attempt to isolate, in our concrete material, these three social realities: the integration of (con• scious) cultural uniformity or the tribe; the integration of cultural conformity or the community; and the cultural co-ordination typical of organized society, above all, the political system and its paradigm, the state.
THE TRIBE
The conception of a common tribal culture is kept alive, as part of Nupe tradition, by the transmission of traditional knowledge. It forms part of the general ideological background of all institutionalized be• haviour. For many forms of customary behaviour membership of the tribe constitutes a convincing mison d’etre. The tribe, lacking all unity of comprehensive co-operation, also lacks the unity of a common centre of cultural teaching. Th e consciousness of a common tribal culture is thus maintained by the scattered, disconnected mechanism to which the transmission of traditional knowledge is entrusted in Nupe. It is implied in the nature of this transmission that the same raison d’itre—’we do this
CULTURE, TRIBE, SOCIETY 19 or that because we are Nupe—is quoted in support of widely different and divergent cultural activities. In the riverain districts they tell you: ‘The Nupe know only two funeral rites> the ceremony of the Eight Days and the ceremony of the Forty Days; but the Mohammedans have in • troduced a third ritual, the Hundred-and-twenty Days.’ Go to the Nupe round Bida or in Trans-Kaduna, and you learn ‘that all Nupe perform the Hundred-and-twenty Days’. In Bida they say that ‘no Nupe eats eggs, for they make the women sterile—but in other districts
I have seen people eat eggs and think nothing of it. A contradiction ?—
only so far as concepts which are never checked against reality can imply
contradiction.
I have said already that for certain cultural practices the people admit that no general rule can be formulated and that different sections of Nupe follow different practices. But this admission of differentiation is, with regard to actual facts, no less inaccurate than the comprehensive statement about customs that are ‘typically Nupe. Th e cultural theory which allows for variants in the tribal culture is no less a theory and no less an ideology. This qualified ideology of cultural unity becomes specially important in connexion with another fact, namely, that the Nupe theory of tribal or cultural unity does not imply a belief in common stock or origin of all members or sections of the tribe. On the contrary, the Nupe assume different origin, and subsequent processes of fusion and sometimes differentiation, to account for the existing tribal sections. The Nupe count eleven or twelve such tribal sections or, as I propose to call them, sub-tribes. Here are their names: Ebe, Gbedye or Gbedegi, Kusopa, Benu, Beni, Dibo or Zitako (also known as Gana-Gana among the Hausa), Kyedye or Kede, Ebagi, Bataci, Kupa, Cekpa, and perhaps also Gwagba—a section formed but recently, something between sub- tribe and political faction. One group in Trans-Kaduna, which is believed to represent the oldest section of Nupe, has no second sub- tribal name; the people refer to themselves only as Nupe, or Nupe zam, ‘pure Nupe’. In this area no real conception of the sub-tribe exists; the different villages regard themselves as independent cultural units, of independent origin, united only by the one, widest link of triba member• ship. Th e different sub-tribes are supposed to be not only of different origin but also of diiferent age and to have evolved as sub-tribes at different periods. According to Nupe tradition there were five ‘original sub- tribes in existence at the time of the culture-hero of Nupe, Tsoede, of whom we shall hear more presently, namely the Ebe, Beni, Ebagi, Bataci, and Dibo. Th e origin of the Kyedye and Gbedegi is linked, in this tradition, with the coming of Tsoede. Th e riverain Kyedye are sup• posed to have come with him to Nupe country and to have been settled by him all along the river Niger and established as the Lords of the Water and as local rulers over the aboriginal inhabitants of the area, the marsh-dwelling Bataci. Th e Gbedegi are believed to have been of Yoruba origin, and to have been Nupe-ized by Tsoede and his men. Cekpa and Kupa, the sub-tribes on the southern bank of the Niger, are said to have lived first in the area north of the river, in Nupe proper, but
2 0 CULTURE, TRIBE, SOCIETY
to have crossed later to the other side where they settled and became, in the course of time, independent sub-tribal sections.1 In the case of the Kup a the tradition is more concrete; it describes the m as Ebe people who have crossed the Kontagora river (called in Nupe Eku , to cross the river is, in Nupe, Id ‘ku pa, and this is supposed to be the origin of their triba l name) in the time of the bi g Nup e wars of conquest, about 1800. Th e Kusopa, the Forest Tenders, appear in Nupe tradition out of noth • ing . Somehow they grew into a separate sub-tribe the identity of whic h is based on their habitat in the thick inland forest west of the Kaduna and also oh their special occupation, whic h is the cultivation of kola-nut trees. Gwagba is a deprecatory nickname given originally to a small section of ‘ Nupe zam’ from Gbara, followers and henchmen of the Nupe kings who , in the wars for the Nupe throne, fled to Zuguma. Gwagba means, literally, Those who threw up their Hands, the Defeated, but it is applied to-day to the whole group of Nupe livin g in and around Zuguma town . Th e Benu, finally, represent another sub-tribe of recent origin , formed in historical times. They are people of Bornu origin (Benu being the Nupe corruption of Bornu), from Kukawa, the ancient capital of Bornu, who came to Nupe country towards the end of the eighteenth century, at the time of Etsu Ma’azu and settled, wit h his permission, in Trans-Kaduna. They came as wealthy traders, bringing wit h them certain new agricultural techniques—onion growing—horses and cattle (the latter soon disappeared in the tsetse-infested country). They soon became the dominant group, economically as well as politi • cally, in the region where they settled, bu t they Nupe-ized themselves, intermarried wit h the Nupe, forgot their language, and abandoned most of their original cultural traits. Thus they are to-day a section of the Nupe, speaking the Nupe language, and conforming in their whole social structure to the standards of Nupe culture. Only their great annual festival, the gani, and the title of their chief, Zonuwa, are reminis• cent of the ancient culture of Bornu.2
Th e name Nupe has also been extended to other originally alien groups livin g in Nupe country, whic h have become Nupe-ized and are hardly distinguishable to-day from the people among who m they live. In many villages and towns of Nupe we find groups known by the name of Konu, Prisoners of War,3 or Eyagi-Nupe, Yoruba-Nupe, people of Yoruba origin , who are the descendants of slaves captured in the wars of the Fulani kings of Nupe against the Yoruba ; they were later freed and, by order of the king , settled in various districts of the kingdom. In Bida tow n I met the makers of horse-trappings—a craft imported from Hausa—who assured me that they were Hausa no longer. ‘Yes, our fathers were Hausa,’ they admitted; ‘but we are Nupe’—claiming Nupe-
1 They speak, in fact, the same dialect—proof that the tradition is not without historical foundation.
2 See Sir Richmond Palmer, The Bornu Sahara and Sudan, 1936, p. 148. I
have given a more detailed description of this process of cultural fusion in Man,
1938, 85.
3 Konu means ‘war’ in Fulfulde; the adoption of this term reflects cultural conditions in the early times of the Fulani regime, when the rulers still used their own language.
CULTURE, TRIBE, SOCIETY 2 1
ization after barely one generation. Th e people in the nort h of Nupe country, near the boundary of Gbari , are known as Gbari bautd, Gbar i slaves. They represent the original Gbari population of an area conquered by the Nupe and incorporated into the kingdom about 1800. In the north-east, too, Nupe culture is absorbing groups of Gbari , in a smooth and silent development whic h is stil l observable to-day. In Kataeregi, for example, in the north-east corner of Beni country, one or two ol d men stil l speak Gbari and remember their triba l origin ; but the younger people speak only Nupe, regarding themselves, and being accepted, as Nupe and Beni ful l and proper. On the river, in the south, we have another group whic h has for some time been under Nupe political influence and has partly adopted Nupe language and culture : the Kakanda. The y are not, however, regarded as part of the Nupe tribe
—or not yet. Whether or not the consolidation of their cultural links wil l advance further and wil l eventually lead to an adoption into the tribe we cannot say for certain; perhaps the Kakanda will , wit h the comple• tio n of cultural assimilation, some day be regarded as another, thirteenth, Nup e sub-tribe.1 In one or two cases, finally, people of one sub-tribe have emigrated into the district of another sub-tribe, livin g there as settlers in a ‘foreign’ land. Thus we find a group of Beni villages in Dib o country, emigrants fro m Beni durin g the Fulani wars, and also tw o small groups of Ebe in Trans-Kaduna, in the area of the section whic h calls itself ‘pure Nupe. Thu s one can ‘become a Nupe—a ze Nupe, they have become Nupe, is what the Nupe say of their cultural proselytes. It is, of course, im • possible to define precisely the point when cultural assimilation warrants, or migh t warrant, inclusion in the tribe. But it may seem as if this elastic application of the triba l name, whic h takes into account de facto cultural similarities, contradicted the ‘theoretical’ and ‘conceptual’ uni t in whic h we saw the definition of the tribe. Th e unobtrusive assimilation of a small group on the fringes of some Nupe section and its eventual absorption in this section (as in the case of the Kataeregi Beni) can be easily understood. But the so-to-speak official inclusion of a new section, under a separate heading, seems to presuppose something like a cultural tribuna l whic h grants triba l membership to new, ‘naturalized’ groups. Thi s contradiction is due not to an obscurity in our definition bu t to an obscurity in the facts themselves—and an obscurity whic h cannot be eliminated. Th e substratum wit h whic h we are dealing has, as it were, changed under our hands. Th e concept of tribal culture has changed into the concept of a culture vested in a political uni t and its deliberate, conscious expansion. Th e proselytes of the Nupe tribe or Nupe culture were in reality proselytes of the Nupe kingdom.2 (I t is,
1 A few informants, in fact, already classified them as a sub-tribe of Nupe— but these were informants from sub-tribes adjacent to the Kakanda, that is, from groups which, so to speak, could directly watch the cultural changes. Groups farther away kept to the established, traditional definition of the Kakanda as a separate tribe.
2 This is the conception of the people themselves. Al l informants spoke to
me of the various sub-tribes as having ‘become one through Tsoede’ (the ancestor-
22 CULTURE , TRIBE , SOCIET Y
in fact, doubtful if a tribe not backed by any political organization could make proselytes on any larger scale.) Later we shall see more clearly how it happens that tribe and kingdom become, for the people, one and the same thing. But the mark of the ‘theoretical* tribal distinction is not obscured completely. It is visible in the qualified adoption of these new
‘naturalized* sections. They ‘have become Nupe; but they are still called ‘Gbari slaves or Yoruba-Nupe. And perhaps now, since the expanding Nupe kingdom no longer exists, there will be no thirteenth sub-tribe of Kakanda-Nupe, after all.
TH E SUB-TRIB E
Th e fact of common origin which is excluded from the concept of the tribe becomes an essential element in the definition of the sub-tribe. Th e sub-tribe is thus not a ‘tribe* on a smaller scale; it differs essentially from the wider unit of the tribe. If the Nupe tribe possesses no external symbols, costumes, or body markings, such emblems exist to some extent for the sub-tribe. Th e Nupe practise body and face markings, etsa, ‘patterns, which are cut into the skin of newborn babies, male and female, and which differ in the different sub-tribes.1 Th e sub-tribe, too, has its ideology in which the conception of unity and differentiation from other groups is expressed. Th e ideological criteria are two: one we have just discussed, the unity of origin and of historical continuity. The other is, again, the criterion of cultural unity. But the physical fact of the smallness of the group as compared with the tribe makes ‘culture here a more concrete, so to speak more concentrated reality. No Nupe term which corresponds to our ‘sub-tribe* or ‘section* exists. Th e Nupe refer to the various sub-tribes as to ‘lands, they speak of kin Beni, land of the Beni, or kin Kusopa, land of the Kusopa, and even of kin Kyedye, land of the Kyedye, a ‘land’ which consists of a narrow strip of territory running across river banks and islets. Th e sub-tribe is thus, like the tribe, a territorial unit. But again the fact of physical size changes the meaning of the concept. Th e smaller locality is, in fact, more of one ‘land; it implies a more unified environment. This is partly expressed already in the names of the sub-tribes, some of which refer to the nature of the environment in which the group lives and to the nature of the occupation by which it earns its livelihood. Th e Bataci or Marsh Dwellers are a group of marsh-land farmers and fishermen whose hunt• ing-grounds are the pools and backwaters of the Niger. Th e Kusopa or Forest Tenders work the famous kola-nut plantations of Labozi and
king of Nupe), or explained the development as due to the fact that ‘one Etsu
established his rule (or law) over the whole country*.
1 I must add, however, that even here these external symbols of group unity are not quite unequivocal. Fashion is changing the markings to-day—and there is no reason to assume that this has not always been so. I have seen fathers and sons whose faces bore different markings. Th e markings of the Bida people, like other fashions associated with the capital of the kingdom, have been adopted by the people of the districts. And the modern Bida and Beni markings are in turn said to have been adopted originally from the Yoruba, whose less conspicuous facial marks the Nupe found ‘nicer’ than their own crude slashes. (See Ap• pendix II. )
CULTURE , TRIBE, SOCIETY 23
Kud u and occupy the ric h oil-pal m district of Yeti . Th e name Kyedye, finally, is often used synonymously foi? all canoe-men and fishermen on the main river. Unit y of environment, a unifor m productive system, a unifor m material culture, represent a first aspect of the cultural unit y of the sub-tribe.
The n there is language, certain linguistic differences whic h thro w into relief the cultural boundaries between sub-tribes. Th e Zitako, Ebe, and Kup a stil l speak dialedts whic h differ considerably from each other and from the Nup e spoken in the rest of the country. To-day most people in these sub-tribes are bilingual and understand and speak bot h thejr own dialect and ‘Nup e proper*. Th e Gbedegi, too, are said to have spoken a special dialect, only a generation or two ago. No trace is left of their language; I found only one old man in Jebba Island who could speak a littl e Gbedegi. In the area where Nupe proper is spoken only insigni• ficant dialectal differences exist. Thi s linguistic unification withi n Nupe is the counterpart of the expansion of Nupe language (Nupe proper) and culture beyond the triba l boundaries. Both reflect the same develop• ment. Th e linguistic unification proceeded from the political centre of Nupe kingdom, the area of the Beni, Kyedye, and Bataci sub-tribes. Thei r dialects, identified wit h the language of the Nupe state, gradually came to embrace all other triba l sections. Th e language map of Nupe reveals the phases of this process, the stages in the growt h of a supreme, unified culture. We can see the resistance to assimilation increase as we move away from the centre of expansion: the dialects spoken on the borders of the kingdom are losing their identity more slowly and more reluctantly than the dialects of the groups in the centre. Th e last phase of this development is not yet completed. We are yet to see the ful l Nupe-ization of Zitako, Ebe, and Kupa .
Bu t in the centre of this expansion the differentiation between cultural centre and periphery gains a new meaning. It is conceived of as a differ• entiation not of dialects representative of triba l sections, bu t of dialects, or forms of patois, representative of a cultural and social milie u and up • bringing. To-day we have a Bida-Nupe, a ‘higher’, more refined Nupe (a Nupe which , for example, has replaced many Nupe words by words adopted from Hausa and Arabic) ; and we have, besides, the cruder,
‘low’ , peasant Nupe—sometimes called ‘pagan’ Nupe—spoken in the districts. However, this new attitude towards cultural differences, this evaluation on the scale of a ‘lower’ and ‘higher’, a more or less refined culture, belongs to another chapter.
Th e cultural unit y of the sub-tribe manifests itself also in the religious field. Certain cults and rituals are peculiar to single sub-tribes, and inter-tribal contact and assimilation have hardly obscured these cultural boundaries. Certain cults, bound up wit h conditions of environment, seem linked naturally to special areas: a sacrifice to the mythical ancestor of kola-nut planting among the Kusopa; cults whic h have to do wit h the water, wit h floods, wit h the dangers of the river, and success and safety in fishing, among the riverain groups. Th e elo ritual , and the mamma cult, masked ceremonies, are known only among the Gbedegi; the Zikinta among the Ebe, the gani among the Benu, the ‘white ‘ gunnu
24 CULTURE, TRIBE, SOCIETY
among the tribes on the southern bank of the Niger, and (a negative evidence mentioned already) the ndako gboya, which is regarded as
‘typical* Nupe by Hausa and Yoruba, is unknown among Dibo and Ebe. There are also other rituals and cults which cannot be assigned (at least not to-day) to a certain sub-tribe, and finally rituals and cults which represent the cultural possession of single village communities. In brief, the possession of a common cult means only an external uniformity of culture. None of these rituals implies co-operation and actual contact in the group at large. They are still but the expression of parallel, ‘like’, behaviour and not of concrete co-activity and ‘togetherness’.
This is equally true of the other cultural features on which the cultural identity of the sub-tribe rests. Th e consciousness of a cultural uni• formity is supported by the actual existing uniformity. But there is still no organized centre of co-activity behind this uniformity or behind the transmission of the consciousness of this uniformity. We can only say this: the sub-tribe, much more than the tribe, constitutes a potentiality of such co-activity and intensive ‘togetherness’. Th e links of a common language, of similar economic pursuits, similar interests, parallel re• ligious ideas, may grow easily into links of co-operation and ‘together• ness*. Only in two cases, however, has this potentiality become reality— among the two sub-tribes Kyedye and Beni. In no other section are there contacts and forms of co-activity which make the sub-tribe ‘one country’, and its people ‘one group’. How, then, does that cultural uniformity crystallize into the unified, conscious commonness of ‘one country’ ?
I have described elsewhere the growth and organization of one of the two sub-tribes which have evolved such unified structure.1 Let me repeat here the main facts. Th e growth of the Kyedye organization was the growth of a colonizing and conquering group. Expanding from one centre—Muregi, near the confluence of Niger and Kaduna—the Kyedye established themselves in the course of time all along the river banks. They built their settlements on the land of the aboriginal sections, Bataci and riverain Gbedegi, and claimed rulership over them. Twi n villages arose, half Kyedye and half Bataci or Gbedegi, from which the Kyedye administered the district, assuming judicial authority and levying taxes. This colonization is still going on, and their rulership, rigidly centralized, is still alive and undisputed.2 Close contact, intermarriage, and regular new emigration link the settlements, old and new, capital and colonies. Delegates from Muregi, relatives of the Kyedye chief, reside in the settlements, and these administrative ‘posts’ are correlated with the ranks which the delegates hold in the Kyedye hierarchy; promotion in rank involves promotion to a more important as well as more lucrative post. This closely knit ruling group precludes all permanent links be• tween the cultural groups which this political development has thrown
4
1 See “The Kede : A Riverain State in Norther n Nigeria’ , in African Political
Systems, Oxfor d Universit y Press, London , 1940, pp . 165-95.
2 Only twent y or thirt y years ago have the Kyedye established themselves on the river Kaduna, or on Jebba Island, a place whic h had only then become of economic importance.
CULTURE, TRIBE, SOCIETY 25 together. There is no intermarriage, and almost no cultural give-and-
take; to-day the Kyedye are Mohammedans (or semi-Mohammedans),
but the Bataci and Gbedegi have remained pagans. Yet in one signi•
ficant field cultural unity has evolved, over and above the differences of
culture, tribal origin, and social status, and over and above also the co•
operation embodied in the political structure. Wit h the political ruler-
ship the Kyedye have assumed spiritual authority as well; myths and
sacred insignia confirm the position of the Kuta, the Kyedye chief, as
the ruler over the whole riverain area and its people; and two most im•
portant sacrifices to the river Niger, which secure large catches and
safety from floods, the ketsd and ndaduma, are linked with his person
and office. Every new appointed Kuta will send one of his near relations
to Jebba and Buka, the two places where these sacrifices are performed,
with two white bulls to be sacrificed to the river spirits by the priests of
these cults. By these sacrifices the chief secures ‘greatness* for himself,
and success and welfare for his subjects.
Th e Beni organization presents an entirely different picture. Unlike the Kyedye system it exists no longer; it was broken and destroyed by the Fulani conquest. It took the form of a confederacy formed by the twelve ‘towns’ of the Beni: Bida, Tafie, Esa, Doko, Towagi, Egbe, Gaba, Nupeko, Eda, Panjuru, Ewu, Yesa. Th e political bonds were twofold: the twelve towns recognized the supremacy of one town, and paid a small tribute to its chief, or ‘town-king’; further, in case of war, the twelve towns were bound to fight together and to support each other, again under the leadership of one town.1 Mythology, and the existence of certain common sacred insignia of the twelve towji-kings of Beni, supply the spiritual background for the political unity. Moreover, the gunnu cult, this most important religious cult of Nupe, appears among the Beni in a special form, namely, as a cult which utilizes comprehensive co-operation between the priests of the whole country, thus underlining, in the religious sphere, the unification of the sub-tribal section.2
One might ask why it was that just these two sub-tribes evolved a comprehensive political and cultural unit. Th e origin of the Beni con• federacy is hidden in the dim past of a Nupe of which we have no records. As regards the Kyedye we can, to some extent, answer the question. In the environment of the river valley, and in the pursuits of a riverain people, lies the germ of territorial and political expansion. Long fishing expeditions, an extensive river trade, on which the liveli• hood of this riverain group depends, demand extension of political in• fluence as security for the very wide area of their economic activities. Trading-posts become political outposts, and fishing colonies become strongholds of territorial expansion. Th e spirit of adventure and enter• prise, finally, which alone could have built up this efficient organization, is part of the’ life of this riverain people.
Of the enterprising spirit of Kyedye chiefs, born colonizers and
1 This chief town of the Beni seems to have been Tafie” till the Fulani con• quest ; after that the headship passed to Bida, till it disappeared finally with the absorption of the Beni confederacy in Nupe Emirate (see p. 82).
2 See J.R.A.I. lxvii, 1937, p. 98.
26 CULTURE, TRIBE, SOCIETY
pioneers, the history of the Royal Niger Company can give many in• stances. But let us read these examples correctly. They show, first, that political integration tends to widen its basis so as to embrace all spheres of cultural activity; political unification calls into being other forms of contact and co-activity, and thus cements its own unity with the help of a more comprehensive and more intimate ‘togetherness V But these examples show also that the creation of such co-activity and together• ness does not simply follow from the potentiality of cultural uniformity. In the case of the Beni confederacy cultural uniformity appears intensi• fied into co-activity and togetherness; among the Kyedye the political fusion of different sub-tribes, different cultural areas, created the wider area of co-activity and common life. Obviously, the nature of this co- activity and common life is different in each case. The commonness that grows from uniform culture and the commonness that is superimposed upon a cultural diversity represent two different social realities. What we have said above of the links of cultural uniformity which ‘grow easily into links of co-activity and togetherness* was not, therefore, quite correct. We have before us not a simple quasi-automatic process of
‘growth’, but two specific, alternative forms of social development.
CHAPTE R I I I
INTRODUCTION TO NUPE KINSHIP
EFOR E attempting the analysis of the Nupe community we must examine certain essential aspects of the kinship system of the country. This description cannot claim to be more than a brief outline. It is concerned only with the main principles of kinship structure among
the Nupe so far as they bear on the subject-matter of this book.
Th e basic structure of the Nupe kinship system is the extended family: all individuals who can trace their descent to a certain initial, nucleus family—a husband and his wives—regard themselves as related and as forming one kindred. Th e Nupe have only one word for kinship, dengi, which defines relationship in the widest as well as in a more restricted sense. Th e vagueness of the terminology reflects the inherent vagueness of this concept of relationship. Th e extended family is a fluid unit in two respects. First, as the women of the family marry outside, into other houses and families, their connexion with the family from which they come, although never admitted to have ceased entirely, becomes weaker and less effective generation after generation. Secondly, the constant growth and expansion of the kinship unit which this principle of reckoning descent entails leaves the boundaries and ‘ fringes’ of the family unit only vaguely defined. People cannot, in fact, trace their descent back farther than two or three generations, and in every family there are a certain number of ‘relatives’ who are known to be dengi, related, without anyone being able to define the degree or nature of the relationship. Moreover, individuals and sections may split off from the main family group, move to other parts, and thus disappear from the view of the rest of the kinship group.
It is implied in these facts that the effective kinship unit is represented, not in the vaguely defined ‘theoretical’ extended family, but in that section of it which remains living close together—whose members keep, as it were, in each other’s view—in the particular case, the kinship section inhabiting one house. This kinship section is, in fact, referred to as a ‘house’—emi in Nupe. Th e Nupe reckon relationship in ‘houses’, saying, for example, that people belong or do not belong to the same
‘house’, that is, are or are not related. Technically, the ‘house’ is a com• pound of a number of houses (to be described later in more detail), enclosed by a common wall and sharing a common gateway or entrance hut, called katamba. The term katamba is frequently used synony• mously with ‘house’: in enumerating the kinship groups of a village the Nupe will tell you that the village numbers so many katamba.
To some extent the house-group, too, is subject to the natural fluctua• tions of increasing or decreasing population. An overflowing population may necessitate the building of new houses and compounds, mostly placed close to the original home, but in some cases at some distance from it, in another part of the village, perhaps in another locality
28 INTRODUCTIO N TO NUPE KINSHI P
altogether. Thus one ‘house* (in the abstract sense) may be found dis• tributed over a number of adjacent compounds, occupying a part or the whole of the separate cluster of compounds which constitutes the sub• division of the Nupe village, and is called efu—a term best translated with ‘village ward \ Or we may find separate small hamlets, called tunga, which are regarded as dependencies of the original house and village.
HOUSE OF A VILLAG E CHIE F
O Sleeping huts CKatamba A., partition of family head. B. .partition of unmarried sons. C partition of unmarried daughters. D. partition of married sons and younger brothers E… sleeping hut of family head. ©Granaries.
Both terms, efu and tunga, will play a considerable part in our discus•
sions of the Nupe community. Th e fluid transition between ‘house’ and
‘ward* is reflected in the often synonymous use of the two terms: thus
the quarter of the blacksmiths (tswata) in a certain village will be known
both as erm and efu tswata, or the part of the village in which the chief
(Zitsu) resides both as emi and efu Zitsu, Decrease in the size of the kin•
ship group (owing to death or emigration), on the other hand, may cause
the break-up of the house-group and lead the people to abandon their
home altogether. Individuals who are left without a large family may
join friends or neighbours as adopted members of their house.
In Bida the situation is altered by a number of circumstances peculiar to the urban society. Space for local expansion being limited, families are much more widely scattered than in the village, even beyond the
INTRODUCTIO N T O NUPE KINSHI P 2 9
single efu, and the family section inhabiting one house is reduced to the smallest size. Moreover, independent economic position and indepen• dent, possibly even divergent, political interests and connexions cause adult family members to leave the paternal home and to set up house• holds of their own . Even the wider local unit , the efu, is no longer based on kinship, bu t is largely replaced by a local uni t the composition of which is determined by the factor of political allegiance and partisanship. Th e size of the house-group varies widely : the largest group whic h I encountered numbered about sixty individuals, including men and women, adults and children. Th e smallest groups are found in Bida,
and may comprise not more than six or eight people.
Th e effective kinship unit—the ‘house* in the abstract sense of the word—expresses itself conspicuously in the recognition of a common family head, who is called, significantly, emitso, Owner of the House. We shall learn later of the political, religious, and economic prerogatives vested in the person of the family head. Here I wil l only add that where a section of the house-group splits off from the main body and estab• lishes itself in a new, separate, house it cannot claim to be at once recognized as a new unit, entitled to a separate etnitso, but continues to acknowledge the old family head. In the course of time the head of the new group may gain the position of an independent etnitso. Th e follow• ing diagram illustrates a case of this kind . Th e members of the family group placed to the right of the dotted line separated from the main family body in 1934 and moved to a newly buil t house in another part of the village (Kutigi) . Th e reason given was that the old house had become ‘too crowded*. Th e head of the new group expected to be recognized as etnitso after the death of the old family head (marked FH in the diagram).
Th e ‘house* is subdivided into a number of family sections, each of which inhabits a division of the compound. Th e single section repre• sents an even more effective uni t than the house-group and its members enter a more intensive co-operation: they keep house together, eat to• gether, and also form, a common productive or labour-unit. No special
30 INTRODUCTIO N T O NUPE KINSHI P
name exists for this group qua group, only a name referring to its organization as a labour-unit, or efakd. To-day these smaller sections are often reduced to the small individual family consisting of a father, his wives, and unmarried children; as often, however (and regularly, under traditional conditions), they embrace larger units and represent
‘sections’ in the strict sense of the word, cut from the house-group, and repeating its composition on a smaller scale. Such a section embraces, for example, a number of brothers, their wives, children, and grand• children. Th e attached genealogy may illustrate the relation between the ‘house* and the smaller kinship unit.
This genealogy shows another household in Kutigi whose family head is a certain Kolo, a man of about 50 or 55. Brackets indicate deceased family members. A dotted connecting line means that the nature of the relationship cannot be ascertained concretely but is only referred to vaguely as implying relationship of * younger brothers’. Family members working together in the same efako group are marked with the same letters (e.g. all individuals marked a form one efako group, all individuals marked
6 another, &c) . The family members marked SH have left the original family home and have built themselves a separate house. Women and small children are ignored in the chart.
W]
Each of the sections of the house may one day contribute the family head for the group at large. Th e position of the family head falls to the senior male member of the house-group, who is succeeded by his younger brother in the classificatory sense, i.e. the man next in seniority of the same generation, and, when the generation is exhausted, by the oldest man of the next generation, i.e. the eldest son of the eldest brother. This rule, however, is not applied rigidly and allows of some latitude so that personal qualifications can also be taken into account. Especially is this so where succession to family headship entails the assumption of a certain important position in the community (e.g. priesthood, or position of an elder). This brings us to the question of the transmission of rights and duties vested in family membership and to the method of reckoning descent.
INTRODUCTIO N TO NUPE KINSHI P 31
Th e family has been regarded by many anthropologists as a typical bilateral unit—a group in which the transmission of rights and duties follows both the paternal and maternal line. We know now that such family structures are extremely rare, and that in most societies succes• sion in paternal and maternal line supplement each other and are assigned different spheres of validity, certain rights and duties being transmitted in the paternal and others in the maternal line. For the Nupe family we
HOUSE OF A BIDA NOBLEMA N OF LOW RANK
can say that rights and duties which are transmitted through the father greatly outweigh in scope and importance those transmitted through the mother. Kinship membership itself and the access to the political and religious rights vested in it are determined by paternal descent. As women in Nupe neither own nor work land, this most important property is inherited through the father. Th e rights to certain forms of financial support (e.g. for the payment of bride- price or support in old age) are equally derived from descent in the father’s line. As RadclirTe-Brown has shown, the division into paternal and maternal succession depends also on the nature of the benefits and duties that are transmitted from individual to individual, some being transmitted more satisfactorily through men (i.e. the father) and others through women (i.e. the mother).1 Thus the personal property of the Nupe women is inherited by their daughters—as the personal property
1 R. A. Radcliffe-Brown, ‘Patrilineal and Matrilineal Succession*, Ioiva Law
Review, xx, 1935.
32 INTRODUCTIO N TO NUPE KINSHI P
of the men is inherited by their sons. As in the Nup e economic system the women frequently enjoy a financially more advantageous position than the men we find also that, in the event of economic stress, a mother may take over some of the financial duties normally discharged by the father.
Generally speaking, matrilineal descent in Nupe, including here bot h descent from the mother as an individua l and fro m her family, has essentially the significance of an ultimate safeguard in case the patri • lineal arrangement proves insufficient. In Bida society, wit h its great inequality of wealth and status, ric h or powerful maternal relations may be as valuable as paternal relations; political connexions, especially, through the mother’s family wil l often be utilized profitably. We realize that these ‘matrilineal’ rights do not follow rigi d rules of descent and succession; rather they represent a ‘second line of defence’ on whic h individuals can fall back i n case the ‘first line of defence’—claims derived through the father—has failed. If we speak of the Nupe ex• tended family as ‘patrilineal’ it wil l be understood that we refer only to the preponderance and the more rigi d formulation of rights and duties transmitted in the father’s line. Th e kinship organization as such, in whic h the alternative methods of reckoning descent stand in that relation of ‘first ‘ and ‘second line of defence’, can be most aptly described by a term whic h was suggested by R. W. Firt h as ‘ambilateral’.
Th e rule of succession to the rights and duties derived in the paternal line, which , as we have seen, impl y a succession running through one whole generation before descending to the next, is expressed very clearly in the Nup e kinship terminology, about whic h I wil l only say a few words. It is of the classificatory type and corresponds to the generation pattern: it subsumes all members of the kinship group who are of the same generation under a common term . Thu s members of the same generation are all ‘brothers’ to each other, ‘sons’ to the generation above them, ‘fathers’ to the generation below, ‘grandsons’ to the second generation above, and so forth . Withi n the generation seniority is specially defined: elder brothers are distinguished from younger brothers, and also the father’s elder brothers from the father’s younger brothers. Th e former, who represent the potential family heads senior to one’s own (real) father, are called ‘father’, the latter, who may be placed as junio r relations under one’s father, ‘littl e father’.
Le t me summarize the main kinship concepts wit h whic h we shall be concerned in the course of this book. We shall largely disregard the fluid and vaguely defined large kinship uni t formed by all individuals who are regarded as being dengi. Whe n speaking of kinship group, or large family group, we shall always mean the ‘house-group’. To the smallest unit , finally, we shall refer either as the family section, or as the household or labour unit , according to the aspect under whic h we are considering it . I must add, however, that the Nupe also recognize a much wider uni t derived—in a very vague sense—from kinship factors. Thu s a whole village is sometimes spoken of as representing one dengi. and ad having ‘one ancestor’. Thi s interpretation does not refer to con• crete common descent from an original family, bu t rather reinterprets
INTRODUCTIO N TO NUPE KINSHI P 33 the existing political structure of the community in a kinship sense. Th e
ancestors of the chief who are regarded as founders of the village are
also thought of vaguely as ancestors of the whole community. These
essentially mythical kinship links exercise no influence upon concrete
kinship claims and obligations.
Political factors are also responsible for the development of the large
units whic h characterize kinship among the rulin g family. We shall see
that the rulin g house of Nupe crystallized in three dynasties, whic h trace
their descent from different sons of the founder of the rulin g house, and
divide between them the rights and duties vested in the rulin g house.
Th e ancestors of the three dynasties lived only two or three generations
back; but already certain religious observances (prayers on their grave)
mark them off from all subsequent royal ancestors. In this rigidly defined
system of reckoning descent in the father’s line back to these almost
sanctified ‘first ancestors*, and in the relationship wit h one another
which is in the nature of mutual obligations for the sake of the larger
unit , the royal house itself, the three dynasties correspond from kinship
point of view to incipient clans—the only analogy to clan structure
which we find in Nupe.
Le t me say in conclusion that kinship structure in Nupe is not that dominant, all-enveloping structure which we know from many primitiv e societies. Although it bears upon religious life, education, and the organization of production, it is greatly overshadowed as a determinant of social behaviour by the factors of political organization and, in economic life, by economic interests proper. Indeed only because of its lesser significance in social life could we omit the detailed description of Nupe kinship from the sociological analysis wit h whic h this book is concerned.
CHAPTE R I V
THE COMMUNITY: VILLAGE AND TOWN
HE description of ‘areas of common life’ must start with the description of the external arrangement, that is, forms of settlement.
Th e Nupe live in large villages or towns. They have only one word for both, ezi, thus defining the essentially identical nature of the two forms of settlement. There exists another Nupe word, tunga or kangi, which is applied to very small settlements. But this name refers not so much to the size as to the different sociological nature of the settlement; for the tunga, unlike the ezi, does not represent an independent local unit but a
‘daughter-settlement*, a ‘colony’ of a village or town.
Th e size of the ezi varies considerably. Nupe Emirate counts three villages with a population of over or near a thousand: Doko with 800 inhabitants; Mokwa with 2,000; and Kutigi with 3,000. But there exist also smaller ‘towns’, villages the population of which had been reduced by emigration and which may be, in size, hardly bigger than a large tunga. Th e local arrangement of the Nupe village, large or small, is nearly always the same: clusters of compounds scattered comfortably over an area, each ‘cluster’ consisting of a number of walled compounds,
‘houses’ in native terminology, and forming what the Nupe call an efu, or ward. These village wards are separated by stretches of open land, spacious cultivated fields, studded with clumps of trees. More rarely the efu stand close together and are separated only by narrow thoroughfares between the walls of neighbouring compounds. This latter arrangement is typical of many riverside villages where, in narrow space, house presses upon house and compound upon compound. Each efu has its own name, derived either from a topographical peculiarity, e.g. Takogi, meaning ‘the one below’, or Kpacinefu, ‘the river bank efu\ or from a sociological feature, namely, the rank or profession of the people living in the efu, e.g. efu Zitsu, ‘efu of the chief, or efu Tswata, ‘efu of the blacksmiths’, and so forth; or, finally, the name of the efu may refer to the origin of its inhabitants, betraying them as immigrants from another village or tribe, e.g. Yafu (abbr. for Eyagi-fu), ‘Yoruba efu’, or efu Dazi,
‘efu of (the people of) Dazi’.
Th e Nupe ezi with its scattered efu clusters is encircled by a large town wall, or more correctly, used to be encircled by a town wall. It exists no longer, but its ruins can still be seen in many places, and all the people insist that formerly all ‘towns’ were protected by such walls. These crumbled ruins of the old town wall provide us with a piece of important historical evidence, which tells us, I think, why the Nupe have chosen just this type of settlement, in large populous ‘towns’, and not, for example, settlement in loosely linked, scattered homesteads; it allows us to grasp at least one of the motives behind this social develop• ment: the dangers of war.1 Large villages represent in a country of
1 Th e other motive is economic and is entailed in the organization of the Nupe system of production and exchange. (See p. 366.)
TH E COMMUNITY : VILLAG E AN D TOW N 35 constant warfare the only safe method of settling. Th e small outlying
settlements could quickly seek the protection of the walled town ; a large
concentrated population meant comparative security from raids, com•
parative safety in unsafe times. We have more historical data to support
SKETCH MAP OF DOKO
this interpretation. One or two Nupe villages have sprung from the amalgamation of small groups, and in one significant case this fusion of component groups into a single large ‘town ‘ can definitely be traced to the time of the Fulani wars. What is to-day the village of Doko origi • nated then out of the fusion of a small hill-side village, called Edoko, wit h a village situated originally about two miles away, in the plain, by the name of Dazi. To escape from the Fulani raiders the people of Dazi moved to Edoko, into the protection of the hill-side. Th e new twin – town was called Doko, wit h Edoko and Dazi as two of the five efu which form the village to-day; and the chief, who comes from the stock of the Dazi people,still takes the title Etsu Dazi, king of Dazi.1 We possess also
1 How far a development such as this, with a large village growing out of the fusion of two or more component groups, is typical of Nupe settlement in general
3 6 TH E COMMUNITY : VILLAG E AN D TOW N
negative evidence to show that wit h the pacification of the country the reverse process has set in : Jima, once a capital of Nup e kingdom, a huge tow n to judge from the ruins of its tow n walls, is reduced to-day to a tin y village; most of its people have moved outside, and the eight former town-wards, stil l known by their old typical e/w-names, are to-day scattered outposts, village ‘colonies’—tunga.
Th e origi n of the tunga settlements of Jima represents an unusual development. Normall y the tunga reflects the expansion rather than the breaking up of the ‘mother-town*. But the ter m tunga is not unequi• vocal. It refers to three different types of settlement. Th e first, the release of small groups for the foundation of ‘daughter-settlements’, is a direct consequence of the concentration of a large population in a single town . A growing population in the village, the need of new land, may lead to small groups movin g into the uninhabited bush, to occupy virgi n land and to found new settlements. A small section of one family at first, a few young men wit h their wives and children, the tunga may grow to a small hamlet embracing two or three large family groups. Or immi • grants fro m another part of the country may choose to settle on the land whic h belongs to a certain village, and under its political protection: this settlement, too, wil l then be called tunga. Finally, the bi g landlords of Bida have settled their slaves and dependants on the land all roun d the capital, in small hamlets, called again tunga. But these settlements are in a different category, they represent not colonies of villages or towns bu t private property, farm settlements of individual landlords. A tunga is thus easily identified, for it is always referred to as a tunga a ‘colony’, o/a certain village or, in the case of private settlements, of a certain land• lord . Tunga settlements seldom grow beyond three family groups. Should the population expand further, the tunga woul d again release some of its members who woul d leave the ol d settlement and found a new colony near-by. But it seems to have happened very rarely that a tunga grew into a ‘town ‘ proper—Gbangba, a prosperous village nort h of Bida and to-day the seat of a Distric t Head, is the only example I can quote. Gbangba, however, was not a village colony bu t a tunga founded by people from Bida. On the other hand, I have seen several cases of tunga settlements whic h have grown out of each other, dependencies of dependencies. Th e tunga is itself a much more mobile settlement than the village proper. A tunga is often abandoned, and the people move to a new place, if there have been two or three bad harvests, or repeated cases of illness.
We understand the mobilit y of the tunga when we realize that these small farm settlements, founded by younger branches of families, can have none of the spiritual bonds—religious and ancestral—that woul d tie the ‘real’ village more permanently to the locality.1 Th e same factor
I will not examine here. I may say that I have found two other typical instances; one I have described in Man, 1938, 85. There may be many more, of which the traces have become much more obliterated.
1 An exception is provided by certain cults which are linked with individual
families and not with the village as such; a sacred shrine of this kind may thus be carried to a tunga. Th e tunga Shebe Woro, for example, harbours the sacred chain and the shrine of Tsoede—two paraphernalia of the ndako gboya cult.
COMPOUNDS IN KUTIGI
JEBBA ISLAND
TH E COMMUNITY : VILLAG E AN D TOW N 3 7 also explains why tunga settlements rarely (i f ever) develop into real
‘towns’. Th e factor of size and numbers, too, plays its part. In order to
provide that self-contained social and political life that is typical of the
Nupe ‘town’ , and in order to afford its inhabitants the political security
whic h is part of the raison d’Stre of that for m of settlement, a much larger
concentration of people is necessary than the village tunga is likely ever
to achieve in a sparsely populated country wit h so low a birth-rate. Bu t
the attitude of the people is as important: they cannot conceive of a tunga
becoming an ezi—a rival settlement to their mother-town. Th e village
tunga has no social life of its own : it possesses no market, no independent
political organization, often no age-grades of its own , and celebrates its
feasts and ceremonies wit h the ‘mother-town’. ‘Towns’ , on the other
hand, are independent, self-contained units, which have littl e in common
wit h each other, and often know nothing of their next-door neighbours.
To become a ‘town ‘ would thus mean for the tunga to deny the very
essence of its social identity.
Every village then counts, ‘owns’ as the Nupe say, a certain number of colonies, whic h surround it in a radius of six to seven miles, right up to the point where the area of the next village begins. Kutigi , for example, possesses six tunga settlements founded between i860 and 1910. Five were founded by people from the two largest efu of Kutigi , one by people from Etsu, a neighbouring ‘town’ , who preferred to settle on the more open and fertile land of Kutigi . Doko , a muc h smaller village bu t situated in a more densely populated area, owns nine large tunga settle• ments and a number of smaller, secondary, dependencies. Tw o of the large tunga settlements were founded by people of the Cekpa sub-tribe who fled from their homes corin g the Yoruba wars and settled in more peaceful Beni country. Th e other seven are village colonies proper. Interspersed between these colonies of Doko are four ‘private’ tunga settlements, founded by slaves and other dependants of Bida landlords.
Th e following examples may illustrate the organization of two com•
paratively recently founded tunga settlements:
ZaborUy a tunga near Sakpe, consisting of two compounds, was founded by the man Zaboru, together wit h his elder brother, the present Village Head of Sakpe, about thirt y years ago. Th e two men went out together to find a place wit h new farm-land, and buil t first temporary, later permanent huts. Th e elder brother returned to Sakpe after the death of their father to assume his position as the new head of the family. In his place a third , youngest, brother, who had just married, went out to the tunga, about ten years ago, where he is still living .
Shebe Woro, belonging to the ‘town ‘ Shebe and consisting of two com• pounds, was founded in 1933, when a certain Kol o left Shebe ‘because the place was too full’ . Together wit h some relations and two friends he occupied a new place, about three miles away, deeper in the bush, and founded there the settlement now called Shebe Woro, which means, New Shebe.
Le t us now go inside a Nupe village. I have described already its typical lay-out—the scattered efu and the walled compounds or ‘houses’
38 TH E COMMUNITY : VILLAG E AN D TOW N
which make up the efu. Th e Nupe house consists of a number of huts, mostly round, built of clay and thatched with grass, which are sur• rounded by a high mud wall. Th e wall opens upon the path or open space in front of the house in the katamba y a hut more spacious than the rest, with big doors which make the katamba a gateway or entrance-hall of the compound. Inside the house are smaller partitions and a number of smaller katamba, all invisible from outside. You only see the tops of the grass roofs protruding from behind the compound wall. Sometimes the house of a chief is built in a different, more imposing style with a large square or semicircular katamba which has a high gabled roof and windows let into the walls. Often huts or entrance-halls are decorated with ornaments worked in relief into the clay. Some houses are more beautifully decorated than others—a sign both of difference in prosperity and in ambition or aesthetic leanings. There are the signs of occupa• tional specialization: one or two blacksmiths’ furnaces; a house decorated with crude plastics in relief of animals, or with a pair of horns surmount• ing the roof—a hunter’s place; a platform of dyeing pits in one corner of the village, and perhaps the half-open sheds of the weavers. To-day one small mosque at least is found in every village of Nupe. Pagan sacred shrines or sacrifice places are mostly hidden inside the houses; only in the riverain villages did I find a sacred shrine placed in the open, in the middle of the village. Th e sacred groves where the annual sacrifices are performed are always some distance away from the village, hidden in the thin bush and brushwood which surround the village. Most villages are situated near a watercourse, be it only a little stream or brook. From here the women fetch the water, and here you may see them at all hours of the day washing dishes and calabashes or laundering their clothes. There are also wells in most of the villages, well-dug deep wells—the Nupe are excellent well-builders—with a strong timber framework and a solid, clean platform round the mouth of the shaft; standing on the platform, the women draw water in buckets or earthenware pots tied to long ropes of native hemp. Th e big, round roasting-ovens of clay which look like huge red eggs scattered on the outside of the village, and in the bigger places the square of open booths which forms the market, com•
plete the picture of the Nupe village. Once or twice a week this square is overflowing with people, men and women of all ages: they buy and sell, they listen or dance to the music of drummers and flute-players, they have their heads shaved, they talk, chat, cry out their wares. They are a mixed crowd, and the nearer the village is to the main roads of the country, the more mixed it is: people from the village, peasants from the tunga near-by, traders from Bida, travellers and occasional visitors from places and countries far away.
Th e market gathering is not the village community. Yet the boundary of what we have called the area of common life is, of necessity, fluid. It is redrafted and reset with every new bond of contact and every new channel of communication. But we may say this: the village and its dependencies form the core of this area of common life. Between them takes place the intensive life of the community, of regular and habitual co-operation in all spheres of life. Within this area marriages are arranged,
A VILLAG E KATAMBA
A KATAMBA IN BIDA
TH E COMMUNITY : VILLAGE AND TOW N 39 pacts of friendship made, visits, feasts, and meetings organized regularly.
Contacts and forms of intercourse may stretch beyond this area—they
regularly do in trade connexions or in the contacts involved in the
political system of the country. Moreover, modern communications
have greatly extended their scope. Yet these contacts and single lines of
intercourse hardly affect the self-containedness of the community. A
sign of the tendency to seek regular contacts with places and groups
without is travelling. And here we see that the villagers evince little interest in travelling and in visits to other towns and countries.1 In the small Beni village Pichi, about five miles from Bida, only three men out of fifty have been across the Kaduna river or outside the area of their sub-tribe. Most peasants, at least the older men, with whom I have dis• cussed travelling, said in so many words that travelling had no interest
for them.
Again, the self-limitation of community life expresses itself in the parochial outlook of the villagers. There is a striking lack of interest in the life and customs of the peoples ‘outside’; or the interest is of that lofty and self-centred kind that will credit anything unusual reported of
- foreigners’. When discussing with villagers the customs, for example, or the religious rites, of neighbouring communities you will be told: ‘We don’t know; we do this and that, but they may have different customs’, or you will be given unreliable, second-hand information, often biased and sensationally made up. Certain of these opinions have become quite firmly rooted—commonly accepted popular theories about other com• munities. The people of Jebba Island professed to know nothing about the religious practices of Gbajibo, a village some twenty miles up the river—’they have quite different customs there’ is what they told me. A visit to Gbajibo revealed that the religious practices of the two places were almost identical. In Bida and Doko I was told that the people of Etsu were blacksmiths by origin and possessed special blacksmith magic. The people of Etsu themselves had a quite different tradition of origin and their magical rites had nothing to do with blacksmith work. In several Nupe villages the people told me that while they themselves per• formed the gunnu ritual in the open, other villages possessed special sacred gunnu huts. I examined every case, but found no trace of this reported difference. This tendency to assume non-existent and exaggerate insignificant cultural differences between one’s own village and other communities represents, on the ideological plane, the assertion of the cultural identity of the community. As in the case of the tribe, we find the social group sustained by an ideology which ignores and overrides empirical facts.
We must, however, qualify these statements about the self-contained nature of the village community. There exists one kind of interest and knowledge—and this time keen interest and on the whole well-informed knowledge—which embraces groups outside of the village community, namely, interest and knowledge bearing on the kingdom, its countries
1 I am speaking here of the inland villages and not, of course, of the riverain Nupe, whose very existence depends on travelling. No r do I refer here to migra• tions and similar permanent changes of residence.
4 0 TH E COMMUNITY : VILLAG E AN D TOW N
and peoples, its organization and its personalities, and above all its centre, the tow n of Bida. It seems characteristic that of the people of Sakpe, a backwood village in the centre of Bida Emirate, a large number have been to Bida, thirt y miles away, but no one to LaboSi, whic h is only ten miles away bu t lies in the opposite direction to Bida. There is only an apparent contradiction between this parochial outlook wit h respect to the community next door and this interest for the kingdom and its capital. It reflects the centripetal structure of the system of cultural ‘co• ordination ‘ whic h we identified wit h the Nupe state. But let us examine first, under the aspect of community and common life, its capital, Bida. Bida as it exists to-day is a comparatively young town . It was founded
by the first Fulan i Emi r of Nupe about 1860. Th e ol d Beni village of
Bida whic h stood on its site, a small village of four efu, forms to-day the
‘ol d city ‘ of Bida, its innermost part, known as ba nin, the Inside. At the
time of the Britis h occupation Bida is reported to have had a population
of 60,000. Later it dropped rapidly : the enormous households of the
feudal lords dissolved, slaves and private armies disappeared, and
absentee landlords moved out into the country to live on their land. Th e
ol d trade-route to the south whic h ran through Bida was diverted by the British , who suggested the shorter route fro m Zaria to Lokoja.1
Trade began to avoid the defeated, impoverished capital. Some ten years ago Bida counted over 30,000 inhabitants; the last figure, for 1934, is 23,286. Bu t the cosmopolitan character of the population stil l betrays its character of a metropolis: all triba l sections of Nupe are represented, and besides almost every tribe of Nigeria: Hausa, Yoruba, Yagba, Bunu , Kukuruku , Dakakari, Kamberi , Koro , Asaba, Gara, Tiv , and many more. Th e Nupe of Bida total over 20,000; of the strangers, Yoruba wit h nearly 600 and Hausa wit h 400-500 constitute the largest contingent in this medley of tribes and races.
An d now the tow n itself. Fro m a broken-off plateau that forms its northern edge, from the foot of two flat-topped hills , the tow n sweeps down into the plain. Th e tow n wall embracing the wide expanse of Bida runs over the edge of the plateau, climbs down into the plain , crosses tw o rivers and opens its gates upon four roads. Thi s wal l was estimated to measure twelve miles in circumference. To-day it has lost its former importance. It has broken down in many places and has been left un • repaired. A go-ahead Administrative Officer some tw o years ago had ordered all tow n gates to be pulled down—to make room for the in • creased motor traffic (which never came). Also, the tow n no longer fills the space withi n the wall , and many ruins and deserted places inside the wal l betray the changes that have overtaken Bida and its people.
Four bi g roads lead through Bida, roughly in the shape of a cross, from east to west and north to south. Th e road from the Kaduna river enters Bida in the west, by the ‘ Wuya gate’, so named after the village on the banks of the Kaduna. Just outside the west gate is ‘Ne w Bida’—a nucleus of Ne w Bida—consisting to-day only of a few official buildings, among them the Emi r of Bida’s newly buil t residence. We leave New Bida behind, wit h its brick buildings, its ti n roofs, and glass windows, and
1 Dr . W. B. Baikie, R.N., 7 R Geogr. Soc, 1867, p. 362.
TH E COMMUNITY : VILLAG E AN D TOW N 4 1 enter the native town . In the tow n as it presents itself to the eyes of the
visitor there are bu t few traces of modern influence. There are the
modern roads, metalled, wit h stone parapets and bridges of concrete.
But the houses whic h line the streets are stil l the same century-old Nupe
houses: the big , naked wall enclosing the compound and, hidden behind
it , the array of small thatched huts. As in the village, the compound wall
opens upon the street in the katamba, the gateway or entrance-hall of the
Nupe house. It s style and lay-out and the state it is in indicate better
than anything else the status and position of the people who live there.
There are the round gateways of the wealthy, some painted black, some
in gay colours, wit h ornaments worked in relief into the clay, and the
roomy gateways of the royal princes, wit h windo w recesses, and pillars
and buttresses to support the high-gabled roof. Some katamba have
raised platforms in front of the door, carefully levelled, and stained wit h
the juice of the locust-bean whic h gives them the appearance of smooth
asphalt. Others have coloured earthenware plates inlaid in the clay of
the wall ; others show, worked in relief, figures and emblems which took
the builder’s fancy or whic h are to indicate the owner’s profession, such
as the likeness of the wooden writin g slate—the ‘trade-sign* of the
learned Koran-teacher. An d then there are the naked, sober katamba of
the simple craftsman or farmer, and the dilapidated, neglected, and roof•
less entrance huts of the poor. Katamba follows katamba, house follows
house, as we walk along the road. Side-streets branch off, windin g
between the compounds, a maze of small streets and paths, whic h are
often jus t wide enough for a man on horseback to pass through. Here
and there the walls recede and give room to an open square wit h a few
trees, a littl e grass or a patch of cultivated ground, where goats and
sheep graze, children play, and a few men, sitting in the shade, enjoy the
hours of leisure.
We pass a blacksmiths’ quarter, easily known by the open forges wit h their many doors and small triangular windows all round the walls. A littl e off the road live weavers and bead-makers. Th e road opens into a large square. To the right is a market, one of the three bi g day-markets of Bida, noisy and crowded wit h people who are buying and selling: foodstuffs, firewood, cloth, small pots. To the left, occupying one whole front of the square, is the residence of the Masaba dynasty. Tw o small bronze cannons formerly flanked the entrance-hall as symbol of the royal might. In the centre of the square stands a small throne hall ; it used to be a pretty building , wit h painted and decorated pillars all round. But to-day, only a year since the death of the last Etsu from this house, it is already in a state of disrepair and neglect whic h symbolizes, more convincingly even than the disappearance of the bronze cannons, the passing of royal power. Houses of nobilit y and high office-holders occupy the other sides of the square. Walkin g on, past the long wal l of the royal residence, we come to the street of the brass- and silversmiths. At all hours of the day their forges are busy, but not too busy to greet you wit h a littl e flourish of hammering on their anvils. Thei r women• folk are sitting outside the huts wit h baskets ful l of brass- and silverware in front of them , offering the products of the street to those who pass by.
42 TH E COMMUNITY : VILLAG E AND TOW N
It is a little over a mile from the west gate to the centre of Bida where the four main roads meet. Here stands another royal residence, the house of Etsu Usman Zaki—to-day almost in ruins. But the ruins, the enormous lay-out, the faded paintings and decorations, still speak of the ancient greatness. It is only ten years since the last ruler from this house, Etsu Bello, died. But ten years in this country, ten years with no royal household, no army of servants and slaves to take care of upkeep and repair, have sufficed to turn the king’s palace into a desolate, half- deserted ruin in a corner of which the king’s last descendants find ample room. Behind the house of Usman Zaki is the big town mosque where the Friday services are held, a building of red clay, with a tin roof perched on top, bare and singularly unattractive. In front, and all along the north wall of the royal residence, filling the wide open square, stretches the big ‘night-market’—the centre of Bida’s pulsating life. It starts in the late afternoon, when peasants return from the farms, crafts,- men stop work, and the women leave the houses, dressed for the occasion. Soon the large square is packed with people: men, women, children; townspeople and villagers, buyers and sellers, and those who come to the evening market to watch the crowds, meet friends, or to enjoy the excite• ment that goes with such public gatherings. The market is easy to sur• vey. Every important commodity has its proper place. Horses and live stock, saddlery and fine coloured grass mats are sold behind the mosque; leather work, caps and straw hats, timber for roofing and coarse mats, in front of Etsu Usman’s house. Th e butchers have a hut of their own, and so have the barbers, while traders and brokers who deal in cloth, gowns, ornaments, and swords occupy the long rows of booths which form the one rigid structure in this moving, seething multitude. Women are sitting in long rows all along the street, so close to each other that you can hardly move, selling food, native soap, kola-nuts, ornaments, silver bangles and glassware, beads and European cloth—anything that can be bought and sold in West Africa. Sometimes there are drummers, or a group of girls will sing and dance somewhere in the crowd; the
- young men about town’ walk up and down the street, shouting greetings to friends, and joking with girls. When it becomes dark the big trade ends; only the women continue their trade in the light of little clay lamps which they place in front of them—a truly enchanting illumination.
But then, in the dark of the night, the market changes its face: the respectable meeting-place of buyers and sellers becomes a market of Venus Pandemos. Th e women who remain so late are not dressed for work or trade: they wear their most beautiful dresses, light-coloured and embroidered cloth; they have bangles on their wrists and rings on their fingers; their eyes are painted with antimony, and their fingers and hands stained with henna. They never carry loads themselves—they always have a small girl with them to carry their baskets and bundles. They all sell kola-nuts—a man’s article. They are always surrounded by men with whom they gossip and joke, and then, a whispered appoint• ment, one of the men leaves, followed after a while by his companion for the night. One light after the other is extinguished. Th e market grows
TH E PALAC E O F TH E ROYA L HOUS E O F MASABA
THE PALAC E O F TH E ROYA L HOUS E O F USMA N
TH E COMMUNITY : VILLAG E AN D TOW N 43 dark and quiet. After 9 o’clock the market-place and the streets are
almost deserted.
But let us return to our daylight inspection of the town: we were standing at the crossing of the four main streets of Bida. Tur n to the north, and you come to lalemi, the strangers’ quarter of Bida, where Hausa and Yoruba are living in houses and huts built in their own tribal style and fashion. They are traders by profession, and shops and stalls line the long road to the north gate—shops in which you can buy every• thing at every hour of the day, from native food to a European wrist- watch, from the powders and amulets of the Hausa medicine-man to brass buttons or bicycle spare parts offered for sale in a Yoruba man’s shop. Th e road to the east bridges a stream and a stretch of marsh-land that runs across the town, with its fields of rice and sugar-cane; it runs past another quarter of brass-smiths, skirts the quarter of the masagd, the Bida glass-makers, and leaves the town by the Badeggi gate, so called after a village thirteen miles from Bida on the small rickety railway line (the oldest railway line in Nigeria) which links Baro on the Niger with Minna on the main line of the Nigerian Railway. Th e south road crosses another stream, with yellowish, dirty water in which children are bath• ing, grooms watering horses, and women washing their clothes and cleaning pots and pans. It runs on through a thickly populated part of the town, past a beautiful old mosque perched precariously on the edge of a deep trench; another day-market, and another royal residence. Another weavers’ quarter, and another group of blacksmiths. Old Bida and New Bida mix for a while: still inside the town walls are Prison and Hospital; on the other side the Native Administration Workshops and the Middle School. And beyond this, the European Reservation, the Government Offices, and then open fields and farms.
This, then, is Bida. Like all large towns it is both more and less than one community. More, in that contacts, communications, and forms of intercourse stretch far beyond the town boundaries; and less, in that within the town walls of Bida there exists a vast number of different, almost independent, strata and ‘areas of common life’. Some, perhaps the weakest and least tenacious, are denned by tribal origin. Th e social differentiation of rich and poor, of social status and political rank, create others—strata in which life is lived in different fashion, in which different values are valid and different patterns of behaviour accepted as normal. But we have also observed the division of the town into three large districts, almost exact copies of each other, each with its day- market, its quarter of craftsmen, and its royal residence. The three districts are separated, geographically, by the two streams which run through Bida, and they are spoken of as three ekpa, ‘banks’, and given the name of the royal house that stands on their ground. Here we deal with units of a different order: they represent political factions, organiza• tions of partisanship and patronage, which cut across the other social, cultural, and tribal strata. They reflect specific political and economic interests, they are vested, ultimately, in that paramount political organ• ization, the state. But that we shall understand more clearly as this description proceeds.
CHAPTE R V
THE POLITICAL ORGANIZATION OF THE VILLAGE
H E political organization o f the Nup e village differs i n certain respects in different parts of Nupe-land , bu t the general outline of
th e system remains, wit h one o r tw o exceptions, the same. Ever y Nup e village, as it constitutes a separate community , also constitutes a separate political system. Th e area over whic h its sovereignty extends is forme d by the village an d its tunga, the ‘town* and its ‘dependencies’. At its head stands th e ‘town-king’ , Zitsu, who , to use the language of th e people,
‘owns* the village—u wu zi—and the land on whic h it stands. By this righ t of ‘possession* he controls th e lives of his communit y and the mai n resource fro m whic h it derives its livelihood , land. Th e close identifica• tio n of chief an d locality is often expressed in the titl e whic h the chief assumes an d whic h is identical wit h the name of the village : the chief of Jebba (the correct name of whic h is Gebba ) is know n as Gebba, of Bele as Bele, and the old Ben i ‘town-king ‘ of Bida styled himself Bida. I n th e chief, then , th e collective existence o f the village communit y i s represented as in a symbol.
Thi s political configuration is to some extent though t of as having evolved gradually i n the course o f time . Bu t the Nup e village (unlik e the kingdom ) possesses no elaborate mytholog y whic h tells of the origi n an d the growt h of the1 present polity. We have spoken of the vague interpretatio n in a kinship sense of the past of th e village, according to whic h th e people of th e village have all descended fro m some commo n ancestor. As regards the specific political features, we find equally vague traditions of half-forgotten ancestors wh o wer e the founders of a village an d th e first settlers in th e locality (w e can disregard here the recent memories o f fusion an d separation o f villages). Th e following tradition , referrin g to the village Doko , is a typica l instance:
Tw o brothers came fro m beyond the river, fro m a place called Gaze (a village on the river Kaduna) . The y came to a place no w called Kop a [Kop a is about five miles fro m Doko , on the other side of the hil l on the slope of whic h Doko. is situated]. Ther e was nothing in this place then save bush. The y settled there, first as hunters, later they began to work the land. Afte r tw o years the younger brother left Kop a and came dow n the hil l to the place where Dok o is to-day. He settled there and called himself chief. He became the first Zitsu of Doko , Zitsu Lefiya.
Thes e traditions o f origi n become mor e articulate onl y wher e the traditio n o f the village can b e linke d wit h th e semi-mythica l history o f Nup e kingdo m an d its ancestor-hero Tsoede . I n a numbe r o f Nup e villages th e people trace thei r line of chiefs back to one wh o was con• firme d i n his office b y Tsoede , and invested b y hi m wit h certain insignia an d magical emblems. O f bot h Tsoed e and these insignia w e shall hear mor e at a later stage. Her e it may be said that the ancestral chief thus
TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E 45 marked off by tradition is not, as a rule, identified wit h the first chief and
founder of the village. It is mostly the second in the line of chiefs who m
tradition associates wit h the ancestor-king of Nupe. Thu s Lefiya, the
first chief of Doko , is said to have been succeeded by a certain Dib o Saba
upon who m Tsoede conferred insignia and chiefly magic. Or in Pichi
tradition has it that a certain Dikw a founded the village and assumed
chieftainship first, and that it was his successor, Goga, wh o was invested
by Tsoede wit h the emblems of rulership. Thi s mythological feature is
of great significance, for it proclaims, projected in the manner of myths
into the di m past, two facts in whic h we shall discover the very essence of
village society: the fact that the village is conceived of as possessing a
social identity of its own , and at the same time the fact that it depends
for its continuous existence on the sanction of king and kingdom of Nupe.
We notice also another interesting fact, namely, that while normally
Nupe families cannot trace their ancestry back more than three or four
generations, these chieftain families proudly record a long list of an•
cestors, embracing ten or more generations. Th e explanation lies, again,
in the political significance of this lin k wit h the past which , like the
mythology that surrounds it , endorses the existing political constellation.
In the case of ordinary kinship groups the genealogy has no significance
beyond that of establishing, rather vaguely, present relationships; in the
case of (hereditary) chieftainship it is to establish a convincing—and
unique—pedigree for political claims.
THE COUNCIL OF ELDERS
Th e chief rules over his community assisted by the village elders, the nusazi or Ol d Ones, or, as they are also called, the ticizi, the Title d Ones. They are the heads of families or of groups of families whic h live to • gether in one compound or one efu. Thei r appointment to the position of an elder is expressed in the titles whic h the chief confers upon these members of his council. Th e office of the elders is only very loosely specialized. Certain titles reflect the special occupations whic h nusazi and their families follow—titles of hunters, drummers, blacksmiths. When the community is in need of the services of these professions— for the purpose of organizing a large-scale hunt or of preparing a cere• mony at whic h the village drummers are to perform—it wil l secure them through the elders who represent the profession on the village council. Sometimes a religious office is combined wit h a political rank. But more frequently the important priestly offices, for example the priesthood of the gunnu, the main cult of Nupe, although bestowed on family heads, are not connected wit h a political position. Th e holder of this religious office has no seat or voice in the village council except when questions of ritual—his ritual—are discussed. Apart from these more specialized offices the rank of the nusa is exactly what the name implies: the rank of a man whose age and experience and the headship over a large family entitle hi m to assume an active part in guiding the fate of his community.
Titles and composition of the council of elders vary, bu t the same rule is valid in’every Nupe village: so many ‘houses’ or family groups, so
46 TH E POLITICA L ORGANIZATION OF TH E VILLAG E
many elders. No ‘house’ may be excluded from the share-out of village offices, nor may one ‘house’ ‘own’ more than one title—at least not in theory. Th e senior members of the house sometimes accompany their family head to the meetings of the council in the chief’s house. Or if an elder is himself too old, or an invalid, a younger relative may do his work for him ; but there would not as a rule be two holders of a nusa rank in the same house.1 These rules, however, are not, in fact cannot be, rigid. A powerful chief may appoint a friend or favourite to a rank to which he would not normally be entitled, or refuse a rank to the legitimate claimant for reasons of private enmity or jealousy; an immigrant group may demand that a rank be conferred on its head as a confirmation of established citizenship; or an expanding family may claim additional titles for its members. New ranks may thus be created, new titles intro• duced, which do not correspond to the traditional system of village ranks.2 On the other hand, titles may disappear with the family in which they were vested or with the special profession to which they referred. Th e title Ndace, Father of Hunting, originally a title common every• where in Nupe, is missing to-day in many villages; the titles Etsu Dza, Head of the Drummers, and Map Tswata, Master of the Blacksmiths, have disappeared here and there where strangers (Hausa) have taken over the work of drumming and smithing and have disorganized this originally family-owned profession.
The equation: so many houses, so many titles, may also be upset (at least temporarily) by the opposite trend: in an expanding family group sections may split off from the main body, and build themselves a new house, without being awarded a separate rank. Th e tunga very rarely
‘owns’ a nusa rank. Founded by young people, it usually keeps its status of a ‘junior branch* even in subsequent generations. However, a son or younger brother may be called to succeed the family head in the village and leave the tunga to assume his inheritance and position in the ‘town’.3
Thus the institution of ranks of elders has two aspects: one is the official recognition, by the community at large, of the headship of certain indi• viduals over the component groups of the community—kinship groups and ‘houses’—a headship which implies both a certain moral control and control of the economic resources of the group; and the other the division of the political executive of the village between a number of selected individuals. Th e closer examination of the first aspect belongs to the chapters on kinship, land-tenure, and economics. We are concerned here more especially with the second aspect. But we shall see that the two are closely intertwined. The acceptance of a family head into the village executive enhances the position of authority which he normally occupies in his kinship group; but it signifies at the same time that he may be called upon to use this moral authority in the service of the village at large. Appointment to a village rank also strengthens the measure of
1 Thus an old, blind nusa in Jebba no longer took part in the meetings of the village elders. But no successor was appointed until his death.
2 In Mokwa, for example, the new title Maji Koshagba (lit. Master of Grave
Digging) was conferred on the head of a Hausa family which had settled there during the Fulani wars. 3 See p. 37.
TH E POLITICA L ORGANIZATION OF TH E VILLAG E 47 economic control which he exercises, by granting him access to special
economic resources (above ail, fruit-bearing trees) unattainable to the
other members of his group. Thus supported, the titled family head is
qualified to assume his double role, as official agent of the community at
large and as representative of his group vis-a-vis the village community.
Th e relation between the two roles and their relative preponderance varies with the rank which an elder may attain. For the nusa ranks are not of equal order. They are defined by strict order of precedence, and there is never any doubt which title ranks higher or lower. Th e measure of political responsibility closely corresponds to the place in the scale of village ranks. In some villages we even find a formal distinction between the ‘great* and ‘small* ranks, the former three or (more often) four in number, forming as it were the ‘inner council. Where the village is arranged in a number of separate large clusters of compounds (as in Doko, Kutigi, or Sakpe), the distinction between ‘great’ and ‘small ranks also implies a distinction between headship over the whole cluster of compounds, the efu, and headship over single compounds or ‘houses’ only.1 Th e professional ranks of blacksmiths and drummers are always at the bottom of the list of village ranks. The political importance of the rank of the Head of the Hunters and of the ranks which are combined with priestly offices differs in different communities.
It would be wrong to think of the village council of elders as of an elaborate, rigidly formalized, instrument of government. Th e full council is rarely summoned except on important occasions, such as the election of a chief, or the visit of the Etsu and his delegates or the District Officer. The normal routine work of village government is carried on by the chief and his ‘inner council’ of two or three elders of highest rank. And here again we have no formal appointed assembly, but rather casual meetings in the houses of the chief or of one of the elders, at which current questions are discussed and the necessary decisions taken. If more elders have to be consulted or instructed, a messenger will be sent to their houses, or they will call individually at the house of the chief or the head nusa. If in the case of the nusazi of high rank their position of authority transcends the component groups to the headship of which they have risen in virtue of their kinship position, this latter aspect pre• ponderates in the case of the nusaii of lesser rank who are excluded from the ‘inner council*. They are executive organs of the village in the indirect sense of intermediaries through whom decisions of the chief and the leading elders reach the various kinship and local groups, and through whom, in turn, these groups can voice claims and interests so far as they call for action on the part of the community.
The graded scale of village ranks does not command an unequivocal corresponding system of appointment and succession. Succession to the lowest, professional, ranks is invariably hereditary. So is succession to the priestly ranks or the ranks linked with a priestly office: the secret
1 In eastern Nupe, especially among Dibo and Beni, the number four plays an important part in the arrangement of efu as well as in the order of ranks. We find four efu and four ranks, or four head ranks and four secondary ranks, or four ranks plus the chief’s rank.
48 TH E POLITICA L ORGANIZATION OF TH E VILLAG E
of the ritual office remains vested in one family group. But succession to
the other ranks involves a certain ambiguous situation. Only a few com•
munities have worked out the principle of precedence consistently,
evolving a system of promotion from lower to higher rank—Jebba
Island and Mokwa are examples. In these communities every vacancy is
made an occasion for a reshuffle of the ranks. A nusa whose seniority
and standing qualify hi m for the rank that has fallen vacant is appointed,
his own rank going to another nusa deserving promotion, and so forth,
til l finally one of the lowest ranks becomes vacant, which is then duly
bestowed on a fitting, as yet untitled, candidate. But in other communi•
ties all promotion from one rank to another is excluded, the titles in
almost caste-like rigidity remain linked wit h ‘houses’, and a man can
claim only this one title whether or not it is one for whic h his age, experience, and personality would qualify him. 1 The people are them• selves well aware of this ambiguity. It is overcome, in practice, in various ways. First, the appointment to the rank which has fallen vacant may be postponed til l the candidate has reached the age and gained the status appropriate to the rank; a ‘young man* of Kutig i (he was about 45) had been waiting for ten years for the title which his
father’s brother had held before and which had been ‘reserved’ for him ; every time I saw hi m and asked how his affair stood, he replied wit h the same words: n’wa ke degi, ‘just a little longer’. Or one may introduce new titles, intermediary ranks, by which device the rigid order of traditional ranks is circumvented. Finally, hereditary succession in Nupe is never understood in the narrow sense of a quasi-automatic appointment. Often it is a selection from a number of candidates of suitable seniority—younger brothers or eldest sons (i n the classificatory sense) of the deceased. Though seniority counts above all, old age and debility wil l disable a prospective successor to family headship. Apart from seniority, other attributes are also essential: good reputation, skill and experience as farmer, a friendly temper, and good health; a leper or a man suffering from elephantiasis or blindness is ineligible; so is a com• plete invalid. But so long as his impediment (e.g. a short leg, a crooked arm) does not interfere wit h his work on the land and his ability to move about, it does not matter. The main attribute is dnkali—intelligence. In Kutig i a serious quarrel had broken out between the chief and some of
his elders who refused to give their consent to the appointment to a high rank of a man whom they considered ‘to have no brains’.
The following genealogy illustrates the succession to the posi•
tion of a certain titled family head, the Ndagoro of Mokwa. We
1 An interesting transition from hereditary ranks to ranks of the promotion type has evolved in Faze, a village in Ebe country. Only recently the former system of hereditary ranks was replaced in this village by a promotion system, the depopulation of the village in the Fulani wars having made this step necessary; for some of the families in whic h the most important ranks had been vested had died out, and only promotion from lower to higher ranks could pre• vent this untoward vacancy. Thi s local development cannot, for various reasons, be taken as affording the explanation of the origin of the promotion system of nftsa ranks in Nupe. But it clearly shows that the people are aware of the inherent conflict between the rank system based on precedence and the strict rule of hereditary succession.
TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E 49 observe the general scheme of Nupe inheritance, one generation being
ru n through before the next generation is admitted to the succession;
bu t we notice also that the progress of succession left out two men of the older generation.
Th e same qualifications would also make a stranger who has settled for some time in a Nupe village eligible for a higher or lower nusa rank. Bu t ‘good reputation’ woul d be understood according to local standards, whic h may not be the standards of the applicant. A lute-player in Doko , who had been livin g there for some twenty years, a fairly prosperous and well-liked man, stil l went without title. Lute-playing, though appreci• ated in Bida, was not in the eyes of the villagers a respectable profession. Ha d he been a farmer, or even a drummer, he woul d have been awarded a title long before.
If a ‘new man* desires to be granted a title he broaches the subject first wit h one of the titled nusa wit h who m he is intimate; the latter wil l bring i t before chief and council of elders, who wil l then discuss the question at their leisure. If , on the other hand, a rank falls vacant, especially one of the higher ranks, the official discussions wil l start at once, and all the men of the village wil l be talking about i t and surmising likely successors. In the lowest ranks succession is a simple matter. In hereditary ranks, particularly in small families, the decision may also take but littl e time. When the Ejuko of Jebba Island—the hereditary Keeper of the Chain of Tsoede—died, his successor was known a week after his death, and the titl e was conferred upon hi m withi n a month. But sometimes the appointment of a successor is a serious and long-drawn-out business, and often enough, as we have seen, feuds may break out over the appoint• ment of the most suitable candidate. Discussion follows discussion, often for weeks and months, especially if it happens to be the rainy season when everybody is busy and can spare only an evening or two in the week. Th e eiders meet first, in small groups, in their own houses; later when the discussion has reached a more advanced stage they meet in the house of the nusa of highest rank and finally in the house of the
chief. As always in meetings of this kin d the assembly gathers in the entrance hut of the house; the younger men sit near the outer door, the elder men on the inner side of the hut. They are rarely given anything to eat or drink except perhaps a few kola-nuts in the house of the chief.
E
50 TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E
Th e meetings (which I have had many opportunities of attending) are quiet and dignified, the discussion flows slowly, no loud wor d is spoken; the young men are mostly silent, only the elder ones take part. If con• troversies or disagreements occur, they are never apparent in the dis• cussion itself. Rather they reveal themselves in the disconcerting practice that what had seemed to be happily settled in one discussion wil l be questioned again i n the next; and men whose agreement seemed to be successfully secured to-day wil l call another meeting of their ow n to-morrow or the next day and start the discussion all over again.
Th e appointment to a new rank is a matter of littl e ceremony. It is identical wit h the modest ceremonial that accompanies succession to the position of a family head (to be described in a later chapter) wit h which , in fact, it frequently coincides. Appointments to high nusa ranks usually fall in the long period of mourning for the late holder of the rank, and so exclude any festivity or general celebration. Appointment to a new rank, whic h woul d mostly fall to a comparatively young man, is sometimes made the occasion of a littl e feasting: beer or palm-wine for the guests and a dance in the afternoon and evening to whic h neighbours, friends, and above all the comrades from the age-grade association are invited. Bu t more often there is nothing ; the selected candidate woul d be sum• moned to the chief’s house before the assembled council of elders, and woul d be informed that nw’a gi Tsowa, ‘he had been appointed (lit . had eaten) Tsowa\ or some other such rank, and that is all . Onl y by the self- importance wit h whic h the new nusa carries himself in the next few days or weeks wil l you be able to say that this had indeed been an im • portant occasion. If by mistake you address hi m by his ol d title , he—or others who know—wil l pu t you right at once.
Perhaps I should mention that I have myself gone through this ex• perience when the people of Jebba Island decided to bestow a rank on me, who had become i n their eyes a ‘ful l Nupe’ , and who at any rate was insisting on being treated as one when it came to religious meetings or confidential discussions of village matters. I had to make a few small presents (from whic h even a ‘real’ nusa-candidate might not be exempted) to chief and elders; they were ‘sitting ‘ for three days, discussing the most suitable title, til l eventually the senior nusa informed me that I had been given the titl e Shaba (a very high rank but, needless to say, introduced ad hoc, and copied from the rank-list of other villages). A day or two after my appointment I walked through the village and met some people who used the customary greeting for Europeans, zaki\ but my com• panions called the people sharply to order, informing them that I was Shaba of Jebba now, whic h information at once elicited the greeting corresponding to the Shaba rank, gbako nyenkpa (body of iron). In a week or two there was nobody withi n ten miles of the place who di d not know of my new rank.
CHIEFTAINSHIP
Chieftainship in Nup e is, like the system of nusa ranks, characterized by the coexistence of two systems of succession: promotion and heredi• tary succession. In Jebba and Mokw a the chief’s rank is merely the
TH E POLITICA L ORGANIZATION OF TH E VILLAG E 51 highest in the order of village ranks, and one of the senior elders of the
village is promoted to the position of the chief. In all other areas chief•
tainship is hereditary and bound up with one * house. Chieftainship of the promotion type goes invariably hand in hand with a rank system of promo• tion type. But hereditary chieftainship does not invariably imply heredi• tary nusa ranks. In Dibo country, for example, chieftainship is hereditary, but the nusa ranks in some villages follow promotion. Th e elected chief is by no means regarded as less closely identified with the village which he ‘owns than the hereditary chief. The rule that he assumes the name
of ‘his* village as title holds good both for elected and hereditary chiefs:
the Gebba of Jebba (Gebba) is, as we have just seen, an elected chief.
Neither chieftainship by election nor hereditary chieftainship implies an automatic, rigidly fixed, succession. Under the former, any elder who has reached an important rank is eligible for the final promotion. The latitude which this arrangement allows is illustrated in the succession to the chieftainship in Mokwa, where, in the case of the last five chiefs, the following rank-holders were elected as chiefs: Sdmaza, the fourth on the village rank-list, Sojeko, the second on the village rank-list, Maji Bagidi, the eighth, and Sarkin Zongu; this last rank, borrowed from Hausa, has been abandoned during the regime of the present chief. It was attached to the position of a caretaker in the camp prepared for the trading caravans stopping at Mokwa—a responsible and lucrative, though from traditional point of view inferior, post. (See chart I, p. 54.) Hereditary succession to chieftainship follows the same elastic rule
which governs succession to hereditary nusa ranks.1 A few communities have adopted the practice of appointing beforehand the successor to chieftainship, an ‘heir presumptive, mostly known by the title Shaba. But it can be shown that this practice is of comparatively recent origin, in fact, a copy of the system of succession characteristic of Nupe king• ship. Th e larger the family group or ‘house of the hereditary chief, the more scope there is for the selection of a fitting successor. Expansion of the chieftain family may lead to a splitting-up of the ‘house* as a result of which the chief’s office, vested in two or more related houses, is taken in rotation by these families. In at least one case hereditary succession itself was abandoned, and replaced by free election: in Doko the line of hereditary succession was interrupted two generations ago, and a ‘new man, a man of outstanding ability, was appointed as successor by the old chief and accordingly ‘elected later by the elders of the village. The genealogy of the Doko chiefs (p. 52) up to the appointment of this ‘new man’ may illustrate the latitude allowed even in strictly hereditary chief• tainship. We see that sons, grandsons, younger brothers, and brothers* sons are all equally admitted to the succession of village chieftainship.
Th e appointment or election of the new chief is a matter for the council of elders. It is presided over by a certain nusa of high rank who is in many places called Ndeji (a corrupted form of Ndezi, viz. nda ezi),
1 In one village, Etsii, I found matrilineal succession to hereditary chieftain• ship, through the mother’s brother’s son, although in all other respects the rules of inheritance were the same as everywhere else in Nupe. I am unable to explain this isolated Instance, unless perhaps historically.
52 TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E
Genealogy of Doko Chiefs
Father of the Town , and who stands somewhat apart from the other village ranks. He plays the part of the chief ‘elector’ or ‘king-maker’; he is never himself a candidate for chieftainship, bu t is entrusted wit h the rule of the village durin g the interregnum. We shall hear more of hi m presently. Th e Nupe dislike a long interregnum: as a rule only a few weeks, filled out wit h discussions and the consultation of diviners, elapse before the appointment of a new chief.1 Thi s appointment wil l eventu• ally need the confirmation of the Etsu Nupe and, to-day, of the Adminis • trative Officer. Thi s last rule, however, applies only to those chiefs who are recognized as official Village Heads under the modern Native Administration. In the reorganization of the Emirate under Britis h rule only the larger villages have been created seats of official Village Heads. Thei r number does not comprise all the places whic h are ‘villages’ or
‘towns’ in traditional native conception.2 A significant terminological dualism has been the result: the people speak of Zitsu, ‘town-king’ , when referring to their traditional, officially unrecognized, chief, and call the officially appointed Village Head Etsu Nyenkpa, ‘Mone y Chie f (i.e. tax chief).
I can offer no explanation for the coexistence of these two different systems of chieftainship, of hereditary and promotion type. There is no evidence of either having developed from the other. No r can local conditions, economic factors, or the factors of size and population be made responsible. But I could discover the sociologically interesting correlation between the system of succession and the religious elements whic h surround Nupe chieftainship. In the villages where chieftainship is hereditary we find that vague mythology of origi n whic h tells of the first occupation of the land, and of the foundation of the village by the
1 I have been told that formerly unanimity between the various diviners of the village was considered necessary. In Mokwa my informants stated that the normal duration of the ‘interregnum* is ten days.
- In connexion with a scheme to raise the salary of Village Heads the number of Village Headships in Bida Emirate has recently been further reduced.
TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E 53 ancestor of the present chief. A certain special magic and sacred insignia
are linked wit h the house in whic h chieftainship is vested. Thu s the
chiefs of various Beni villages and of S’akpe possess the brass bangles of
Tsoede; chiefs in Dib o country an iro n staff; chiefs in the riverain area
the sacred chain of Tsoede. Th e magic ritual whic h is associated wit h
chieftainship varies as widely : in Doko it is a harvest rite , the ‘eating of
the bulrush-millet’ ; in other yillages it appears as magic against barren•
ness, as a rain sacrifice, an ordeal, or as a general rite to secure health and
prosperity of the community. Bu t invariably the chief’s ritual , whatever
its more specific aim, wil l be performed as funeral rite and ceremony of
installation of chiefs, and also whenever certain great calamities befall
the community—a drought, famine, or epidemic. Th e chief is not him •
self the priest of this magic, and often has not the insignia in his own
possession. Insignia and magic are entrusted to another member of the
chief’s family. Th e office of the zigi, the village priest, is kept in the
junio r branch of the chieftain house, and Zitsu and Zigi, Town-kin g and
Town-priest, appear in the genealogies as elder and younger brother
(i n the classificatory sense).1
In the communities, on the other hand, where the chief is elected from the senior rank-holders, his house ‘owns’ no special magic or in • signia, and there is no mythology of a founder of the village, but tradi • tio n speaks of the village as having ‘always been there’. Th e religious duties fall, instead, to the nusa who acts as chief ‘elector’ and who is entrusted wit h the rule of the village during an interregnum. Thi s man’s office wil l be hereditary and sacred, he wil l own insignia or a special powerful magic, whic h is, again, invoked at the death and the appointment of chiefs or in the case of great calamities. A supernatural agency, a mystic safeguard, is thus always assumed behind the chief’s authority. But Nupe village organization has evolved two different solutions, in accordance wit h the two different systems of chieftain succession: one makes the chief or his family the agents of the super• natural power whic h supports the claim to chieftainship; the other separates the political rule from the mystic safeguard and, instead of a hereditary king , brings into being a hereditary ‘king-maker’. (See chart I I , p . 55).
Thi s division of religious and secular office seems to eliminate effectively conflicts between chief and priest. Th e priests of the village cults, experts in religious matters only, have no influence in political affairs; they are, as we saw, not even granted the general prerogative of their age: a seat on the council of elders. Th e priest of the chief’s cult, if of the chief’s family, is classed as a ‘younger brother’ and thus equally barred from the position of an influential elder. Onl y the priest-king• maker (where he exists) could gain the position of a rival power and interfere wit h secular affairs. However, by his office the main ‘elector* of the chief, he should be a supporter and ally of the chief, whose office reflects to some extent his own authority, rather than his adversary or rival. He may prefer a puppet at the head of the village—I saw an instance
1 See the genealogy given in chart I I , p. 55.
54 TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E
I
Rank System from Mokzva (Promotion Type)
Locality
Rank (‘efu’ or Insignia,
order Title Office ’emi’) Succession magic, &c. Remarks
i Lile Chief Efugbo or
efu Lile
Promotion Present chief was promoted to chief• tainship from rank Sarkin Zongu
2 Shaba Lile
Gunnuba (so called because gunnu priest lives
here)
Title introduced un• der Fulani influ• ence; replaced ori• ginal title Sojek6.
3 Ndeshi ‘King•
Ilancita Hereditary Sacred chain, Title Ndeshi corre•
(Sheshi) maker’ priesthood of Tsoede
sponds to Ndeii,
‘Father of theTown’.
3 Guye Successor to 3
Etisheshi and Ko – rofugi
Hereditary in family of Ndeshi’
3b Tstvashe Successor to 3a
Ntaci As za
4 Sdmaza Gbave and Promotion Was promoted to
Yafu
5 NJagoro Goro If
present rank from
Tsonktva.
Newly introduced
6 Maii Kosh- eghe
7 Tsonktva
(Tsoiva)
Kpege ,,
emi Nda- kogi
rank.
8 Maji Bagidi Ilancita
9 Maji Deride ,, ,,
1 0 Maji nya
Kola
Goro ,,
I I Maji Guye Kpege
1 2 Maji Tsivaci Head of Black• smiths
Tswata Hereditary
13 Etsu Dza Head of Drum • mers
13a Shaba Dza Successor to 13
emi Dzezi
emi Dzezi
“
Hereditary in house of Etsu Dza*
Wangtoa efu
Wangwa
This rank no longer
exists. The position of the head of the efu has fallen, quasi- unofficially, to the Liman, the Moham• medan priest, who is not yet, however, officially admitted into the village council.
1 These secondary ranks defining succession to the hereditary rank of ‘king-maker’ I have found only in Mokwa. The Sheshi, Guye, and Tswashe regard each other as elder and younger brothers respectively. I was told that originally the office of Sheshi was inherited matrilineally, and that this is the explanation why it has become divided between three kinship sections living in different efu. Though this explanation is, logically, incorrect, and although these three offices have actually been inherited in paternal line since the people can remember, the fact remains that the three groups which succeed each other in rotation are vaguely thought of as of matrilineal origin. The aloofness from the other village ranks which characterizes the office of the ‘king• maker’ is thus further enhanced by the different conception of succession.
- Secondary professional ranks of this kind are found also in other craftsmen groups, e.g. blacksmiths. Sometimes there exists also a third rank of this order, with the title Kpotu. This organization of craftsmen ranks will be discussed in Chapter XIV .
THE POLITICAL ORGANIZATIO
I I
Rank System from Doko (Hereditary i Type
Rank Locality Insigma,
order
Title
Office ( ‘efu’) Succession magic, &c. Remarks
—Etsu Daii Chief Daii Hereditary Sacred bangles, chief’s
Zigi (priest) is of chief’s family.1
2 Shaba I n tunga
A relation of chief,
promoted to pre•
ritua l
Newly introduced
sent rank from title, copied from
3 Lile Formerly
‘Heir
Presump•
tive*
Lile at death of late Shaba, 1935
Da2i Present holder chief’s younger brother
Bida ranks.
4 Kpotu Present holder chief’8 elder broth• er (died 1936)
5 Maku Tentatively intro •
duced (see p. 56)
6 Ndefie Edokd Hereditary in one
Newly introduced title, copied from Bida.
As 2 and 4
7 Dzoko Takogi
8 Boke Kpacinefu
9 Shabaeii Ndadofu
10 Bodo Dai i
11 Zitsu Bebi Edok6
1 2 Maji
‘house’ ••
13 Maji Egbe
I Head’of
M » •
(or Ndace)
Hunters ,, ,,
1 Th e typical relationship between Zitsu and Zigi, chief and priest of the chief’s ritual , as de• scribed above exists no longer, as the present chief comes from a new family. Th e relationship between chief and priest as it existed in the original chief’s dynasty is shown in the attached genealogy.
of this : a young man, son of the late chief, but inexperienced, timi d and unintelligent, had been put forward as candidate by the ‘king • maker*, himself a shrewd and keen man, when the Administration looked for a successor to the vacant chieftainship of Sakpe. (Th e young chief, whose rule not unnaturally proved a failure, was deposed after two years.) But such cases are rare. As a rule the division of secular and priestly office works out harmoniously. Only once di d I find an open conflict; bu t it revealed, not a failure of the established system, bu t rather a departure from it , brought about by extraneous influences.
It was the case of the chiefs of Doko , to whic h I have already referred. Here the kinship bond between chief and priest, Zitsu and Zigi, had been dissolved when the new family succeeded to the chieftainship. Th e priest of the chief’s ritua l became, from the priestly partner in the two • fold office of chieftainship, an independent, and thus dangerous, agent. Th e descendant of the original chief’s family, a young man, Ndak6 Gana by name, *used the spiritual power vested in his house as a weapon
56 TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E
against the rulin g chief who m he regarded as an usurper. He refused the chief the use of the sacred insignia—the bangles of Tsoede—without whic h the annual chief’s ritua l cannot be performed. Finally, in 1934, he asked me to intervene for hi m so that he should be given official authority to keep in his possession the invaluable insignia, whic h the chief was demanding back not so much for religious reasons (for some of his brothers and sons were already posing as Mohammedans),- as in order to deprive the riva l family of this convincing evidence of their claim to the rule of Doko. Th e case remained undecided, and the rivalry between the two men unrelieved, til l i n 1936 the chief orTerdd the ‘pre• tender’ official recognition of a sort, namely, the title Maku. Th e highest rank next to the chief’s, Shaba, had jus t fallen vacant, and a reshuffle of titles had taken place. Ndako Gana, however, rejected the title Maku— a new title , introduced ad hoc (copied from the dynastic ranks in Bida)— as a meaningless, makeshift sort of rank and demanded instead the rank of Shaba itself. Th e deadlock was finally overcome by an appeal of both parties to the Etsu; he managed to persuade the young man to accept the offered rank, at least as an earnest, and as more fitting to his age.
VILLAG E GOVERNMEN T
Th e tasks of government whic h fall to the village chief are these:
- He appoints the village elders from the ranks of the family heads and is thus bound, as by a constitution in the modern sense, to accept their co-operation and advice in all matters of village government.
- He controls the land resources of the community and in this capacity apportions farm-land, trees, fishing rights, and levies certain tributes and land dues from the families holding village land. Wit h this aspect of village government we shall deal later on in detail, in the chapter on land-tenure.
- He is not, as we have seen, the religious head of the community.
But he organizes, directly or indirectly, and partly finances, certain
religious rites linked wit h chieftainship. - He arbitrates in litigation and wields (though now only on a limite d scale) a certain informal judicia l power.
- He organizes all large-scale co-operation and concerted action of the village community.
We have already spoken of the composition of the village council and the co-operation between chief and elders. Le t me add some concrete data to illustrate the workin g of the village * constitution’.
A Dok o farmer had quarrelled wit h a neighbour over some trees; people from a village near Mokw a came into Mokw a to borrow masks for the elo ritua l whic h they were going to perform, having lost their own masks; a family head in Sakpe accused his sister’s son who lived in his house of having stolen rice from his farm ; a gir l in Labozi denounced tw o strangers who had been given hospitality by a neighbour and who m she saw stealing their host’s clothes; a young man in Jebba accused his wife of adultery; another man in Jebba complained that a certain hunter
TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E 57 who was reputed to possess strong and deadly magic had threatened his
life ; a family head in Kutig i that his family was too highly taxed; a man
in Pichi that a neighbour had beaten his goat (which the other man had
found eating his grain): these are some of the cases whic h a village chief
and his elders may daily be called upon to decide.
Th e procedure is nearly always the same. Th e various news, requests, or complaints are communicated at once to the chief, directly, if the supplicant or bearer of the news has the chief’s ear, that is, is himself a titled elder, and indirectly, through the family head or the efutso (head) of the efu in whic h he lives, if the bearer of the news is an ordinary peasant. Upo n receiving the news the chief wil l summon the head nusa (the Sheshi in Mokwa , the Lite in Doko , the Ndeji in other villages), and he in tur n wil l call together the other members of the council whose co-operation is required: two or three of the senior elders—the members of the * inner council*—and always the nusa to whose kinship group or efu the litigant or supplicant belongs. No t only does this nusa act, as we have seen, as the official informant of the chief on all matters concerning his group and locality, but his intimacy wit h the people whose case is to be heard, and his moral influence, make hi m an important witness and, when a decision has been reached, a guarantor of its execution. Before this tribunal , then, supplicants and litigants in the village appear. Th e discussion that ensues is informal, and devoid of the stiff procedure that characterizes, for example, the Native Court. Supplicant or plaintiff state their case in a few sentences—the Nupe are not fond of long speeches and elaborate expositions. Obscurities in the statement are unravelled in patient cross-examination, in whic h everyone present can take part. Then , at one point, the chief or a nusa may suggest a solution, whic h may be agreed to by the others wit h the stereotyped expression u ge, ‘i t is well’ , or revised in renewed discussion. Very rarely do tempers flare up and the discussions become heated. Th e final decision of the chief and his council is pressed home in the for m of advice, persuasion, warning. A discussion of this kin d rarely lasts longer than tw o or three hours, after whic h time it wil l be interrupted to be resumed on a follow• ing day. Mostl y one or two sessions suffice to arrive at the final decision.
But then, in more difficult cases, in whic h many witnesses have to be called and thorny problems of arbitration settled, the decision is taken out of the hands of the village executive. For the chief is not entitled to dispense formal justice and to enforce arbitration. He must send all legal cases to the professional Mohammedan judge (Alkali) of the dis• trict ; moreover, litigants who are dissatisfied wit h the chief’s decision may themselves brin g their case before the professional judge or the Distric t Head (a point to whic h we shall return later). Thi s limitatio n of village jurisdictio n makes it impossible to pronounce in more detail on the efficiency of the village council and the way in whic h it handles, or could handle, more severe conflicts. Th e gamut of advice, persuasion, and warning is the only one whic h it can use: the more strident notes of authoritative orders and severe reprimands are outside its scale.
Tha t they, too, have a rightfu l existence in the administration of village affairs is shown in the organization, by chief and elders, of all concerted
58 TH E POLITICAL ORGANIZATION OF TH E VILLAG E activity that exists in the village: common labour and communal cere• monies and sacrifices, fights and wars in pre-European times, and tax- collection for the modern Native Administration. Tax-collection in the village has become a little ritual of its own, interesting enough to be described here. Some time before the N.A . lorry is expected from Bida on its periodical tax-collecting visits you will find the chief sitting in his house, or in a booth in the market, surrounded by the people who*have come to pay their tax. The two or three head nusazi will sit with him. Th e ‘tax Mallam’, who is attached to the Village Head as scribe and
‘accountant’, makes entries in his little book and is busy writing out tax receipts. Little heaps of coins pile up before them. On this day the normal routine of village life is interrupted: the men do not go to their farms and often forgo their midday meal at home, and even a ceremony that might fall in that time will be postponed until the ‘tax ritual’ has been concluded.
A large proportion of co-operative village work is done to-day fyy the order of the N.A. : the repair of a rest-house, the clearing of a motor- road, &c. But the concerted action of the village community is still as efficient in spontaneous, unsolicited work. I have seen the men of a village repairing a bush bridge leading to their farms a day or two after the news that it had been swept away by floods was reported to the chief. And although there exists to-day an efficient body of N.A . police, in an emergency it falls frequently enough to the people of a village to organize the pursuit of a criminal and to escort him to the District Head—a task which the young men at least undertake with efficiency and relish.
BALANCE OF POWER
To repeat: the village elders are the mouthpiece and the agents of the chief, through whom his orders reach the community; and they are also, in turn, the informants on whom the chief depends, and the mouthpiece of the people whom they represent both in their official capacity and as heads of kinship and local groups. Again we may call the chief a symbol of the collective identity of the village community, in the sense that, through his orders, he expresses its collective will and interests. Yet the degree of perfection which this political system, with its well-planned harmonious co-operation between chief and elders, attains in practice varies considerably. Autocratically inclined chiefs find means to evade the check on their power which an efficient council of elders implies. We have spoken of certain chiefs who appointed their friends and favourites to the position of titled elders in disregard of more legitimate claims, thus creating for themselves a personal following in the village council. I was told of chiefs who in the past have overridden the authority of the elders, levying tribute and refusing legitimate grants of land, until the elders appealed to the overlord of the country, the Etsu Nupe, against their overbearing ruler. If they failed to obtain the support of the king, my informants added, they would carry on as if there were no chief—go on strike, as it were. I have myself witnessed all variations possible in this fight for balance of power: an autocratic chief who
TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E 59 seemed to have managed perfectly to turn his elders into mere puppets;
a chief (the chief of Doko) whose overbearing behaviour caused a tunga
group to refuse him the customary ‘tithe’ of fish caught in a pond on
their territory; and also a chief who had tried, in the appointment of a
rank-holder, to go against the majority of his council, but was brought to
heel by the organized ‘strike* of the elders: this happened in Kutigi,
where, during the great Mohammedan festivals, the elders declined to
attend the traditional ceremonial gathering in the chief’s house until
he gave in rather than prolong this.assault on his prestige. That this
religious ‘strike’ concerns a Mohammedan ceremony is incidental. Re•
fusal to co-operate in established ceremonial activities may well check a
chief’8 high-handedness (though I have not seen another case of this
kind): the congregation is as indispensable as the ‘owner’ of a special
magic and the chief organizer and ‘financier’ of communal rites.
Into the problem of political balance enters yet another factor—the economic aspect of chieftainship. Th e ability of the chief to free him• self from all checks on his authority may depend ultimately on the economic resources of which he can avail himself—of the economic resources, that is, allowed him by the village constitution. (We may leave out of account for a moment the existence of the supreme control of village chieftainship by the overlord of the country.) Having to appoint the heads of family groups to the position of village elders, the chief cannot eliminate completely these agents of ‘public opinion’. But decisive economic superiority may spell political power uncontrolled— the ability to win over, bribe, or force opponents into submission.
Th e political power of the chief is sustained by a number of important economic privileges. First, the chief receives (or rather used to receive) a land-tax in kind, known by the Arabic-Hausa name dzankd, ‘tithe’, from every family head. He also receives a certain gift in kind, or some• times money, when allotting land to new tenants. Th e nature of this payment as a gift and not as a regular tax is expressed in the rule that on its receipt the chief has to make the usual small return gift with which the Nupe always dismiss a bearer of presents. A special right to fruit- bearing trees assigns to the chief a source of income not available to other members of the community. Th e communal farm-work, egbe, which the Nupe farmers carry out for one another is always performed first, and on the largest scale, on the chief’s farms, so that he can count regularly on a larger and more docile team of workers than any other citizen of the village. Finally, the chief has special rights to the fish caught in the ponds or backwaters in his territory, and to certain game killed by the hunters of his village. As regards the former, he again claims dzankd, tithe, of every catch; hunters have to surrender to him the skins of leopards or lions, or meat of leopard and lion, or one leg of a buffalo—here the rules vary locally.
Some of these prerogatives have disappeared to-day. Th e general money tax of the Emirate has replaced the former land fees and tithes. Besides, the chiefs of larger places, official Village Heads under the N.A. , receive a fixed salary. Only in a few small villages the annual gift in kind has survived, though reduced in amount and only in the nature of a
60 TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E
voluntary present. It is, I was told , only the ‘ol d men who are stil l doing it*. Th e value of the right to communal farm-work has also lost its former importance. A number of salaried Village Heads have abandoned almost completely their former agricultural work , being too busy wit h their official duties. One village chief, who is Distric t Head at the same time (i n Kacha), has found it impossible to spare even his sons and brothers for farm-work, needing them as messengers and the like. In • stead, he has adopted the most recent innovation in Nup e economics, and engages wage-labour to work his land. Th e chiefs of small villages, however, who draw no wages, and have been given no official position under the N.A. , have lost the most important of the ol d economic pre• rogatives—unless their people voluntarily continue the ol d practice— without benefiting from the new. Onl y the right to special fruit-bearing trees, game, and fishing has survived everywhere without change. It is not surprising, therefore, that the ol d term Zitsu is more and more losing its splendour, overshadowed by the matter-of-fact ter m Etsu Nyenkpa, Money or Ta x Chief. An d often the small chiefs, who have been passed over in the political reorganization of the Emirate,1 tell you that to-day there is littl e difference between chief and commoners.
Ho w far di d the former economic prerogatives in reality establish a definite, unassailable economic superiority of the chief? Th e approach to this question must be through the comparison between traditional income and expenditure of chief and elders respectively. Here we see that some of the economic prerogatives of the chief are shared (though in lesser degree) by the other family heads and title d elders. Th e nusa&i, like the chief, have a right to the possession of fruit-bearing trees; they too can avail themselves of communal farm-work—though only after the chief. An d if they have a large family, many sons and younger brothers helping in the farm-work, their normal income may not be very much lower than the chief’s. Of the economic prerogatives of the chief only the land-tithe (or its modern counterpart, the chief’s salary) and the game and fishing rights are exclusive to his official position. Th e game rights contribute mainly to the symbols of chieftain status—the posses• sion of a leopard or lio n skin, allowed only to the chief. Th e additional capital resources vested in chieftainship—additional in the sense that they are unavailable to ordinary family heads—were thus derived in the main fro m the tithe.2 I have calculated that the annual land-tithe of a village the size of Doko would amount to about seventy bundles of corn
—twice the annual yield upon whic h an average peasant household could reckon.3 Now , on this evidence we are justified in saying that the
1 See p. 52.
2 Certain other irregular forms of income it is impossible to estimate to-day :
occasional land fees fro m new tenants, fines or gifts from parties between who m the chief was arbitrating. Moreover, certain chiefs received a small share of the tax money they levied for the Etsu Nup e (see p. 116); but this last income, derived from the kingdom , no longer reflects the village system as such.
3 There is a certain margin of error in this calculation. Th e constant extension
and subdivision of Nup e family groups and households make it impossible to define wit h any precision the size of the households originally responsible for the payment of the tithe. I have taken an average of the larger groups, such as are
TH E POLITICA L ORGANIZATION OF TH E VILLAG E 61 chief’s special income di d not, normally, constitute an insurmountable eco•
nomic barrier between himself and his subjects. Thi s is further shown
by the fact that the characteristic possessions of a chief—possession of
things generally valued in the society—were not unattainable by others.
His surplus income the chief uses as follows: some of it he must devote
to special expenditure attached to his office, the purchase of a goat or
fowl for the annual sacrifices, or some other large contribution (larger
than that of the other family heads) towards the cost of the village
ceremonies. Occasional liberal presents to the elders, food, kola-nuts,
beer, gowns are also indicated; a niggardly chief is much disliked. Th e
larger part of his income the chief spends on himself, his family, and
generally on things the possession of which are demanded by his status
and prestige. Some chiefs, wit h a view to profit as well as prestige, have
bought slaves or taken pawns. Every bigger chief possesses at least one
horse. His dress is always of the more expensive kind—embroidered
trousers, a turban, and an indigo-dyed gown, possibly an embroidered
Bida robe. Many chiefs possess the beautifully carved staffs which are
made by the craftsmen in the river valley and sold for sums up to two
pounds. One or another chief owns an old flintlock gun, and some
chiefs pride themselves on building bigger and more ornate nouses than
the majority of the people possess. Finally, as the most conspicuous
symbol of economic advantage in this country of extremely high bride-
price, two at least of the bigger chiefs I met had what among the Nupe
peasants is considered a very large number of wives (5 or 6 as against the
usual 1 or 2), and so had some of their sons and younger brothers.
Set against this the economic position of a well-to-do family head and rank-holder. He too used to own slaves or pawns. The higher rank- holders often possess large, imposing houses. They carry carved staffs, own a horse, turban and state gown, and a gun. Some of them may have three or four wives. Wit h regard to the accoutrements of his office—the religious elements linked wit h chieftainship, its insignia and symbols of rank—the chief stands on a plane by himself; as regards things ‘that money can buy’, or liqui d capital, he is placed only one step higher than his citizens of rank. Th e surplus capital which he derives from his office is not, as we can see, large enough to be turned into an effective instrument of political ambition.
Involved in this is the question how far the temporary economic benefits linked wit h the chief’s office can be used so as to be turned into a lasting economic advantage, enjoyed by the descendants and the whole family of the chief. For the communities in which succession to chieftainship follows promotion or rotation the question answers itself: the surplus income vested in the chief’s office is not sufficient, once acquired, to endow his whole ‘house’ permanently wit h economic superiority.1 In
presided over by the nusa ranks, and the smaller family groups (approximately
100 in number) wit h whic h we should have to reckon to-day.
1 Here is a significant illustration: in Mokwa and Jebba, two villages of
‘promotion type’, the imposing houses of the late chiefs have fallen into dis• repair; the descendants are no longer able to command the communal labour without which the upkeep of the large house is impossible.
62 TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E
hereditary chieftainship, the accumulation of surplus renders such de• velopment more likely. But even here the existence of the special chief’s cult, which imposes additional expenses on the chief (higher than in the case of cults to which he need only contribute his share), should tend to adjust more equitably the economic balance between chief and elders.1
However, we have other evidence to show that the chief’s position, economically privileged though it may be, is not meant to lead to the formation of an essentially different, sharply separated, permanently privileged ruling class; evidence, in other words, to show that the political organization of the village is meant to preserve the social homo• geneity of its community. This evidence lies in the rules of etiquette defining the behaviour of the people towards their chief. People every• where show reverence to their chief: they knee! down by the door when entering his presence. Often there are raised platforms in the chief’s entrance hut, near the door leading to the inner part of the house, on which he would sit, while the others are squatting on the ground in front of him. Or the chief would sit on a leopard skin or at least a cow-hide, while the others are assigned seats on grass-mats. Th e chief, finally, is addressed by special formulae of greeting reserved for his title. Yet again, this attitude of reVerence represents only the highest grade of respect as it is shown similarly to all nusazi, graded in accordance with age and rank. In every house young men or visitors of lower status and seniority would squat near the door while the head of the house, or honoured guests, occupy places on the inner side—the more honoured side. Special greetings of the same symbolic nature as the salute re• served for the chief are typical of the various ranks. Here as in the case of economic privileges etiquette marks off the chief not as standing on
‘a plane by himself* but merely as the most prominent in a graded order of prominence. If on state occasions, wearing his state robe and turban, the chief is conspicuously marked off from the rest of the people, when going out to his farms there is nothing to distinguish him from an ordinary farmer; like him he carries his hoe across the shoulder, a straw hat on his head, and wears only the cheap blue cloth slung loosely round the body. There is thus ample ‘neutral ground’ in the matter of etiquette and distinction of status between chief, elders, and the rest of the people.
But above all, the attitude of reverence and the special chief’s salute are reserved for the chief’s person; they are not enjoyed equally by his relations merely in virtue of their being his relations. The exalted social position of the chief does not extend to his family—there exist, even in hereditary chieftainship, no symbols of a hereditary class differentiation. Later, in the discussion of social classes in the kingdom, we shall under• stand fully the significance of this ‘democratic’ element in Nupe village chieftainship.
1 In 1936 the chief of Doko bought five sheep for sacrifices connected with the
‘chief’s cult’, contributed beer to other ceremonies, and presented two gowns to men with whom he was anxious to keep on good terms—not counting occasional small gifts and contributions, e.g. to family feasts, &c. Altogether an expendi• ture of approximately £2-3, or 30 to 40 bundles of corn, that is, roughly half of his income from tithe.
TH E POLITICA L ORGANIZATION OF TH E VILLAG E 63
There is one exception—which proves the rule. In Kutigi all the members of the chief’s family are addressed as waci, ‘Great Ones’, and though not shown special reverence are regarded somewhat as a class apart. Th e explanation is this: the rulers of Kutigi come from an alien group, those Bornu immigrants to whom I have referred already. Owing to their economic and educational advantages (they were Mohammedans in a pagan and rather primitive country) they had established themselves as the dominant group, and the leaders of the immigrants became the political rulers of the original inhabitants. In this position they were confirmed by the Etsu Nupe. In the Fulani dynasty of Nupe the Benu had an example close at hand how an alien group could arrogate to itself the position of a hereditary ruling caste. They followed it in every detail: from the Fulani kings they adopted the salutation waci; and like them they managed that the recognition of social superiority was extended not only to the actually ruling chief but to his whole kinship group—a whole hereditary ruling class.
BINDING FORCES
Let us try, in conclusion, to define the specific nature of the political unity achieved in the Nupe village. It must be emphasized, first of all, that it is a specific unity, and that it is achieved by mechanisms peculiar to itself. In other words, political organization is not—as is the case in many primitive societies—simply absorbed by religion, magic, or kin• ship, and the mechanisms of control which these domains of culture entail. Nor is it, on the other hand, possessed of any of the conspicuous instruments of political control which characterize other societies: a police or military executive, a disproportionate chieftain power, or an elaborate apparatus of coercion. Moreover, the connexion between political control and religious and kinship agencies is very close. Political organization delegates to the kinship system the important task of pro• viding the executive organs for the community in the persons of the fiusazi, who, as we have seen, are village elders and family heads in one. And religious beliefs and practices supply the spiritual sanction of chieftainship and authority in general. But as important as these specific contributions of religion and kinship are certain less specific influences, by means of which both religion and kinship help to sustain the system of political control. Thus encouragement of intermarriage within the village community adds the strength of interlocked kinship relations to political ties; and the communal religious rites exercise, and underline in their own impressive fashion, the unit of the village group.1 They do it consciously and openly: their dedication to the support of the village community expresses itself in the prayer formula which occurs in every religious ritual whatever its avowed purpose or the occasion of its per• formance: ezi u de lafiya, ‘may the village prosper’. Political control, kinship, and religion thus define coextensive social units. Only because
1 This interdependence of political unit and religious congregation is analysed more explicitly in the case of the gunnu cult. (J.R.A.I. , lxvii, 1937, pp. 91 –
130 )
64 TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E
of this coextensiveness of the spheres of religious and political control could the village elders of Kutig i to whose ‘strike’ we referred above effect that political adjustment by religious pressure. An d again only be• cause of this coextensiveness of the spheres of political and kinship control could, in another instance, a family head be excluded from the ranks of the title d elders on the grounds that one of his sons had twice been caught thieving. Th e family head was identified wit h the moral standards obtaining in the kinship group the headship of whic h woul d have entitled him , in the normal case, to a position in the political executive.
‘Ho w can a man who cannot keep order in his ow n house/ said my in • formant, ‘keep order in the village?’ Bu t these three agencies of control are only three in a multiplicit y of coextensive forms of social organiza• tio n : economic pursuits, the contacts of everyday life, recreation, educa• tion . Th e political activities take place withi n this one area of common life. Expressed in one formula : the political system and the closely kni t social life of a single community organize an identical human substratum.
Here we may, in parenthesis, touch upon the theoretically interesting question whether this system of ‘multipl e control’ whic h the political organization is capable of utilizin g can be taken to explain the absence, in the Nupe village, of the more conspicuous, specifically political instruments of control. However, this question cannot be solved on the basis of Nupe material alone. For here the issue is confused by the fact that the state organization, superimposed upon the village system, obviates the necessity of the more essential of these instruments, especially wit h respect to legal and military executive. Indeed, the fact that the state removes graver legal cases from village jurisdictio n re• leases the social system of the village from certain social pressure whic h must otherwise have n *cessitated a more strongly marked political authority.
Thi s brings us to the question of the resistance of this delicately balanced system of multiple control to outside influences and, generally, the influence of change. It s efficiency is clearly a function of the uni • formity of social interests obtaining in the community. It wil l fail as soon as the closeness of community life is broken by a strong divergence of interest. Extraneous changes creating new aims whic h are apt to dis• sociate the political executive from the rest of the people may disturb its equilibrium , whic h could then be readjusted only by a corresponding strengthening, at the cost of the other agencies in the multiple machinery of control, of political power as such. Th e most radical adjustment of this kin d is that brought about by the modern political changes which , in accordance wit h the system of Indirect Rule, have turned the village chief into an official agent of the Administration. In this capacity he may have to give orders, enforce laws, and exact actions unpopular wit h his people. He is bound to stand for aims and interests whic h may threaten’ to estrange hi m from the community. An inconsiderate Ad • ministrative Officer may still more widen the rift . Twic e I have come across chiefs who , in their official capacity and by the order of a Distric t Officer, had to forbid all dances and festivities in a district whic h was
TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E 65 behindhand wit h the tax, at a time when he should have organized, as
chief, the village rituals that precede the harvest.
As long as a chief in his capacity as a paid official of the N. A. organizes common labour for road-work—the value of whic h the Nigerian peasant is soon coming to realize—the capture of a criminal, or certain obvious hygienic practices, he wil l be backed by public opinion. But when en• forcing an unpopular tax, pressing children for the Government schools (which was frequently enough the case a few years ago), or—a case I witnessed myself—sentencing to a fine in accordance wit h some am• bitious by-law a man who had been defecating on the public road, his authority no longer serves the accepted common interests and can no longer reckon wit h common approval.1 Here the backing by the Ad • ministration must provide the extra power whic h is to weigh the scales in favour of the chief and to reset the political balance between chief and community. Here and there the Administration even felt compelled to replace a traditional chief, who proved incapable of carrying out the tasks whic h his position under the N.A . assigned to him , by a more will • ing or more enlightened man, possibly one who had had the advantage of education and could thus better see the ‘white man’s’ viewpoint.2 But this solution was not an unqualified success: Government backing was not capable of making up for the loss of the moral unit y whic h supports the traditional chief in the execution of his traditional duties.
Th e case of the present chief of Jebba Island is an illuminating in • stance—although here the Administration is only indirectly responsible for the departure from the traditional rules of chieftainship. After the death of the last Gebba, who was said to have been a rather inefficient Village Head, a complete outsider was appointed as chief, a stranger who had been livin g for some time in the late Gebba’s house and had been acting for hi m as semi-official messenger in his dealings wit h the Distric t Head at Mokwa . It was on the advice of the Distric t Head that he was made the new Village Head. Th e Distric t Head, himself a stranger who di d not know the local situation, preferred to have a man he knew for his subordinate, and a man who had been to the Government School and who was, incidentally, a Mohammedan like himself. Elders and ‘king • maker’ were not consulted. Th e village elder who would have been appointed chief if the latter had had their way held at the time the titl e Wangwa. He set up something like an independent rule of his own. Except for official business—tax-collection in particular—the people ignored the other, shadow chief, completely. An unhealthy situation arose and was left to exist for two years, during whic h time there was no real authority in the village, and legal offences and unruly conduct in general were increasing rapidly. In 1935 I had the opportunity of draw• ing the attention of the Administration to this fact.* Th e man Wangwa was recognized officially as a ‘second-in-command* and prospective
1 Thi s case happened in a community half Yoruba and half Nupe. Th e pro • gressive chief was a Yoruba, the victi m (who came to me to complain) a Nupe . I could, in fact, hardly imagine a Nup e chief to be sufficiently appreciative of hygienic principles to take such a step.
2 Thi s happened twice, in Jebba Island and Sakpe.
v
66 TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E
successor to the chieftainship (i t being thought inadvisable to depose the once appointed chief). At the same time the elders promoted hi m to the highest village rank next to the chief’s, Tsadu. Havin g been granted this responsible position in the village government, the ex-rival of the chief seems prepared to use his influence and moral power in support of the official Village Head who m he may succeed some day.
Thi s book cannot venture to answer the question how far the authority of the traditional chief can be strained withou t alienating hi m from the people and the elders, or withou t forcing the latter to abandon their role as free spokesmen of the community and become mere pawns of the chief.1 But it must be said that in many respects the adoption of new governmental duties by the traditional chiefs is proceeding smoothly and efficiently. Thi s readjustment seems due less to political measures as such than to the comprehensive effectiveness of culture change, whic h gradually diminishes certain divergences in the views to whic h the political executive is committed and those of the people themselves. Already many Nupe villagers appreciate medical help and education, and are sending their children to school withou t pressure from the chief.
For completeness’ sake we must add that the balance of ‘multipl e control* may be endangered also by the opposite development, namely a separation between political executive and community in whic h the latter represents the new extraneous influences. For individuals or sections of th * community may dissociate themselves from the unit y of common life, alienated by modern education, new economic enterprise, or conversion to Christianity, and may thus no longer be amenable to the traditional authorities. However, this danger has not yet arisen in the Nupe village, chiefly because the ‘progressive* elements flock to the tow n rather than stay in the village, and also because Christian influence is inconsiderable.
Yet it would be a mistake to assume that the factors that weaken the homogeneous village community are all of recent date and the results of modern colonial development. Disturbances of the ‘balance’ very similar to those just described, and like them entailing a strengthening from outside, as it were, of the chief’s authority, were also at work in the pre-British Nupe kingdom. The y are embodied in the very existence of the kingdom as a political and social superstructure whic h is the carrier of new values and new, manifold, cultural influences.
First the purely political factors. We have heard of the insignia and emblems of chiefly power which, according to tradition, were bestowed on village chiefs by the ancestor-king of Nupe ; the most ancient for m of
1 It is possible that in communities in which the ‘multiple* system of control is less delicately adjusted, and the political authority of the chief more independent of the other principles governing community life, the political power of chief and elders can be used more freely for new tasks. This adds, I think, a new meaning to the following statement by Sir Donald Cameron: ‘The Native Authorities Ordinance also makes provision for the limitation in particular instances of the powers which may be exercised by a Native Authority, for some of these are well organized and experienced while others are of . . . weaker mould.’ (‘Native Administration in Nigeria and Tanganyika’, Journal of the Royal African Society, xxxvi, Supplement, p. 13.)
TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E 67 village chieftainship is thus thought of as an office held ‘b y the grace of
the king*. I have also mentioned that the appointment of the village
chief is subject, to-day as of old , to confirmation by the Etsu Nupe.
Th e chief thus becomes the official representative of the king , respon•
sible to hi m for law and order and the collection of tax, but also able to
rely on his sovereign’s spiritual or political support. Th e people of the
village can appeal to the Etsu if they are dissatisfied wit h their chief, and
in the case of riva l claims to chieftainship the kin g woul d be called upon
to decide between the claimants—a role whic h he di d not always under•
take impartially ; bribes offered by a chief threatened wit h the loss of his position or a new claimant woul d easily influence his decision.1 Th e control of the kin g over his chiefs is expressed already in the ceremonial of the appointment. Th e chief-elect, accompanied by the head elders, woul d be summoned to the capital to receive his appointment and to be vested wit h a gown, turban, and sandals as insignia of his office, in a
ceremony whic h closely resembles the investiture of officers of state or members of the royal nobility. Th e Etsu Nupe may also be called in to arbitrate in internal disputes, e.g. over the succession to local ranks
—we remember the example from Doko . Lastly, legal cases come under the jurisdictio n of the Etsu and his appointed agents, thus removing from the orbit of the community the right to see offences punished and retaliation achieved.
Another source of ‘disturbance’ lies in the impact of Islam. It led in many cases to a spiritual separation of chief and community, for many chiefs embraced Islam (at least nominally) long before the rest of the people. Thu s these chiefs can no longer organize, and rarely even show interest in , the traditional religious ceremonies whic h weld the com• munit y into a single congregation. No r can they base their authority, hereditary or not, on that spiritual commonness of whic h it was once an expression. They rely instead more or less openly on the closer associa• tio n wit h the Mohammedan rulers of the country wit h who m they now share the creed—at least one feature of their exalted position. In its ever- increasing scope of proselytizing, Islam drives the wedge of another class distinction into the structure of village society. Th e distinction between Mohammedans and pagans in a village often has this effect, that the two sections differ not only in religious observances, bu t also in general mode of life, dress, important food habits (Mohammedans do not drink sorghum-beer, the traditional ceremonial drin k of the Nupe peasants), their—often imaginary—educational qualification and, again, in the degree of the affinity they could claim to the highest social stratum in the country.
In other spheres of life—economic, educational—similar influences are at work, striking at the very roots of social homogeneity. But we may cut short our description; we shall meet wit h these influences again later. Ou r present discussion has established the main facts. For at least a century and a half the Nupe village di d not exist in that self-centred, secure isolation i n whic h the balance o f its social system could fulfil
1 Bribes to the Etsu are said to have played an important part in gaining the assent of the king when the new chiefs of Doko usurped the chieftainship.
68 TH E POLITICA L ORGANIZATIO N OF TH E VILLAG E
itself ideally. Throug h the person of its head, first of all , the village organization is bound to duties towards, and linked wit h the fate of, the wider political organization—the kingdom. Throug h religious, economic, and educational factors whic h follow in the wake of political absorption the one-ness of community and political organization whic h proved the essence of the village society is finally broken. Th e com• munities themselves, their areas of integrated social life, are broken .open and forced partly into a system of integration, under the dictate of a concept of social unit y whic h is that of the centralized political system of the state. I say ‘partly’ , for the latter has to some extent left intact the social identity of the village. Thi s is not meant only historically, bu t refers to the relation and actual inter-penetration of tw o essentially different social systems. No r does it refer merely to a distinction be• tween political autonomy of the village, or certain villages, as against political domination by the kingdom. One may perhaps speak of a partial autonomy of the villages under or withi n the kingdom, bu t the essential point is the different solution of the problem of political integra• tio n in one and in the other case. We have met wit h the analogous difference when we discussed the tw o different systems of political unification evolved by the two sub-tribes Kyedye and Beni. Le t me formulate it again: it is the difference between an organization growing out of and utilizin g the potentialities of unifor m culture and common life, and an organization relying on the extraneous machinery of political control ; between a political integration whic h remains true to the concept of Community , and an organization whic h forces unit y upon diversity of culture and common life—the State.
CHAPTER V I
POLITICAL HISTORY OF NUPE KINGDOM
THE STATE
T wil l have become clear that we regard the state as a specific form o f political organization. No t every political system represents thus a
state. Thi s book can do no more than suggest a descriptive workin g definition of that difficult and much discussed concept. If we accept the definition of a political system suggested above, on the basis of its in – clusiveness and the nature of its ultimate sanctions, we may regard the state as a social system in which these two characteristics crystallize in specific forms. Thu s we arrive at the following threefold definition: (i ) Th e state is a political unit based on territorial sovereignty. Member • ship of the state, i.e. citizenship or nationality, is determined by residence or birth-righ t in a territory. Th e state is thus inter-tribal or inter-racial, and distinguishes nationality from triba l or racial extraction.1 (2) A centralized machinery of government assumes the maintenance of law and order, to the exclusion of all independent action.2 (3) Th e state involves the existence of a specialized privileged rulin g group or class separated in training, status, and organization from the main body of the population. Thi s group monopolizes, corporately, the machinery of political control. It is regarded as a ‘representative’ group, and its actions and decisions are accepted by the people as actions and decisions on behalf of the country at large. These three factors must appear combined to justify our terming a political system a state.3 Th e three factors of state structure represent interdependent social necessities whic h follow from one another and can be reduced to the basic principle of territorial sovereignty maintained over groups of different cultural and ethnic extraction. Such sovereignty must involve a certain measure of centralization, powerful enough to counteract separatist tendencies (Lowie) , and also the creation—or continued existence—of a group of rulers and administrators who, by special training and tradition, are raised above the complexity of cultural and ethnic groups. Historically, an inter-tribal sovereignty of this nature can only have arisen from the ascendancy of one ethnic group over others. Certain students of society have interpreted this fact as implyin g the essential identity of state
1 Cf. Sir Henry Maine’s statement that the ‘most nearly universal fact con• cerning the origin of known states* is that they ‘were formed by the coalescence of groups’ (Lectures on the Early History of Institutions, p. 386).
2 Political order in the state is ‘distinguished most of all by this trait, that it is
a peaceful order which allows of no individual violence and only permits itself to enforce justice’ (N . M. Korkunov, General Theory of Law, New York, 1922, P- 341).
3 Obviously, there exist transitory forms in which only one or two of these
political principles are manifest. Th e Beni confederacy represents such an in• cipient state organization. Yet if Lowie says: ‘Some sort of state is a universal feature of human culture’ (The Origin of the State, 1927, p. 44), we feel inclined to alter this to ‘some sort of political organization’.
70 POLITICA L HISTORY OF NUPE KINGDO M
origin and conquest, state organization and alien domination.1 I wil l not go too deeply into this controversial question. But we cannot pass withou t comment the fact that the structure of the Nup e state as it exists to-day, wit h its rulin g class of Fulan i conquerors, and also its growth so far as we can wit h any claim to accuracy trace it back in history, indeed support this theory of the origi n of the state.
Ou r discussion of the principles of state organization has thus led to an historical interpretation. Indeed the description of Nupe kingdom woul d be incomplete withou t the description of its growth and development. It s inter-tribal and inter-cultural nature cannot be fully understood withou t the inclusion of the historical viewpoint. Our three• fold conceptual system of cultural unit , community, and state must itself be envisaged under the aspect of a dynamic process- rather than of a static differentiation. For the deliberate political co-ordination of forms of conduct must tend to influence other spheres of cultural activity as well . In Nupe as everywhere, social patterns, common ideal types, spring up by themselves as * natural products of long association’.2 Political unification goes hand in hand wit h dissemination of culture and wit h an extension and reinterpretation of racial and community boundaries. Th e ‘ inter-cultural’ ‘and ‘inter-tribal ‘ state gives way to a new ‘state culture’ and a new paramount group-unit, the nation.
It wil l be for a later chapter to examine how and how far this unifica• tio n has been achieved in the Nupe state. But we must point out at once that this absorption of heterogeneous elements could only achieve a relative unity . Th e antagonism between the original heterogeneity and the subsequent unification is never allowed to disappear completely. It survives in certain forms of cultural and racial discrimination. Th e culture of Nupe , more correctly the culture typical of that section whic h formed the core in the historical growt h of the kingdom , became the representative culture of the whole country, set up as a standard and
‘ideal type’ for the heterogeneous population. Th e sections of the Nupe tribe whic h were the carriers of this culture became the representative group of ‘real’ Nupe, sharply distinguished from the groups whic h have become Nupe-ized and absorbed in the state in the process of political expansion. A triba l and territorial section thus gained the position of a cultural dHte, considerations of political status came to determine the evaluation of cultural features, and facts of cultural and triba l extraction were re-interpreted in the sense of political class distinctions. Here we understand the anomalies in the application of the name Nupe whic h we discovered in our discussion of Trib e and Culture : wh y the people in the capital regard the Nupe spoken in the districts and on the borders of the country as an inferior brand ; or wh y the Nupe stil l speak of ‘Gbar i slaves’ although culturally this triba l section has become indistinguish• able from the Nupe proper. Th e ‘real’ Nupe call themselves kintsozi, Owners of the Land , as against all who have come later into the country
—settlers, aliens, citizens of Greater Nupe. For all the groups outside of
1 F. Oppenheimer, The State, 1923.
2 A. E. Ross, Social Control, p. 222.
POLITICA L HISTORY OF NUPE KINGDO M 71 the ‘core* of the kingdom, whether they are Nupe, aliens, or Nupe-ized
aliens, the Nupe have a common name: zazi kati, ‘ Outside Peoples’, and
the cultural evaluation implied in this ter m is felt acutely by these
groups. It was extremely interesting to see how strongly the people of
Lemu , Nupe-ized since three generations only, resented being called
zazi kati, although they frankly admitted their Gbar i origin . ‘Thi s
name’, they said, ‘does not apply to us; we have been Nupe and Beni for
a long time. Th e Gbari in the nort h and east, in Wushishi and beyond
Kataeregi, they are Outside Peoples.’
Th e Fulani conquest added a new chapter to this ambiguous history of cultural and political unification. But here ‘representative* and
‘absorbed’ groups have exchanged places. Th e Fulani conquerors of Nupe , numerically an insignificant minority , were absorbed completely by the culture of the people who m they subjugated. Some of the Fulani came from nomadic herdsmen stock; the majority belonged most probably to the ‘tow n Fulani ‘ (as they call themselves), who had long been settled in the northern Hausa states and whose culture had become identical wit h that of Hausa. Th e former surrendered their identity completely; the latter had little , if anything, of their own to give. Th e Hausa traits visible in Nupe culture to-day di d not reach the country through the Fulani ; and Mohammedanism, the religion of the Fulani conquerors, was firmly established already at the tim e of the early Fulani infiltration. In Nupe the Fulani lost their language1 and the last characteristics of their race and original culture. Racially, after genera• tions of intermarriage, they are indistinguishable fro m their subjects; culturally, they are as much Nupe as the Nupe themselves. Man y are even eager to step into the cultural heritage of the country over whic h they rule. The y read and quote the history of ancient Nupe wit h the same enthusiasm and pride wit h whic h they learn or talk about their own historical heroes, the first Fulani emissaries.2 Yet they remain a separate social group, conscious of its alien origin , and stil l distinguished by a special triba l name: they call themselves, and are called by their subjects, goizi (a name whic h distinguishes these settled, ‘tow n Fulani ‘ from the nomadic cattle people, who are known as bororozi), never Nupe. They are a ‘representative’ group in a different sense—the small ilite of conquerors and rulers. Th e historical memory of their alien origin—for it is only this to-day—buttresses their detached social position. Racial antagonism whic h might have disappeared in a homogeneous community here, in the framework of the state, blending wit h the antagonism of class, remains alive, and even strengthens wit h its sentiments the inner tension of the society. It flares up wit h every open friction between rulers and ruled. We have again a political and social cleavage in the guise of ‘racial’ discrimination.3
1 Etsu Mohamadu (1897-1915) was given the nickname ‘Etsu with ,two tongues’, being the last Fulani ruler who—already a striking exception—could still speak Fulfulde.
2 For some time I carried on a correspondence (in Hausa) with two of the most prominent members of the Nupe-Fulani community who had asked me to let them have all information I could obtain on the ancient history of Nupe.
3 Th e following is a significant anecdote: the Etsu Nupe once confided to me
72 POLITICA L HISTORY OF NUPE KINGDO M
Visualize this intricate racial and cultural pattern : tw o itites of different order—kintsozi and Fulani—and surrounding them, almost in concentric circles, ethnic groups of varying cultural dissimilitude, each cultural grade also a grade on the scale of political evaluation. Thi s is indeed the exact picture of the historical growth of Nupe kingdom. Throug h conquests and expansion, wit h fluid, ever-changing boundaries, Nupe kingdom maintained its conspicuous centre, the core of its political and cultural growth. Lik e a tree, it grew from the centre, and you can stil l trace the rings whic h mark its growth.
FOUNDATION OF THE KINGDOM
Before embarking on the historical account we must clarify the con• cept of ‘history* as we propose to use it . For there is a certain ambiguity in the ter m ‘history’ as it is used by anthropologists. History can mean two things: first, what we may call ‘objective’ history, that is, the series of events whic h we, the investigators, describe and establish in accordance wit h certain ‘objective’, universal criteria of connexion and sequence. Second, the ‘ideological’ history, whic h exists in the society whic h we study, and whic h describes and arranges events in accordance wit h accepted traditions about past happenings and developments. Th e two may coincide (as they do in modern history); or the second may at least supply some source of information for the first. But also it may have significance only as a sociological datum per se, i.e. as a system of accepted beliefs and traditions influencing social behaviour in the group. As such, it is often indistinguishable from mythology and from what Malinowski has called significantly the ‘mythical charter’ of customs and existing cultural usages.1 Different societies may stress in a different measure either the mythical, supernatural or the rational, pseudo- scientific nature of what they accept as an account of their past. Th e Nupe are what we may call an historically-minded people. Th e highest, constantly invoked authority for things existing is, to the Nupe, the account of things past—but an account whic h is clothed in terms of a sober, pseudo-scientific ‘history’. Such is also the nature of the Nupe tradition of the origi n of their kingdom. Only for the more recent periods do we possess ‘objective’ sources; the earliest period of Nupe kingdom is described only in their own ‘ideological history’.
Th e earliest history of Nupe centres round the figure of Tsoede or Edegi,5 the culture hero and mythical founder of Nupe kingdom. Th e genealogies of Nupe kings whic h are preserved in many places in Nupe country, and whic h have also found their way into the earliest writte n records of Nupe history which were compiled by Mohammedan scholars
his grievance against a certain rather unruly relative of his whom he believed to have intrigued against him. He added: ‘All Nupe are bad people, you know.’ Th e said relative, himself, of course, as much Fulani as the Etsu, retaliated by informing me that ‘all Fulani were liars’.
1 B. Malinowski, in Encycl. Soc. Sc., London, 1931, vol. iv, p. 640.
2 Tsoede is, linguistically, a contraction of Etsu, or ‘tsu Ede; the Nupe more often use the name Tsoede, the Hausa the name Edegi.
TH E SACRED BRONZE FIGURES ON JEBBA
ISLAN D
POLITICA L HISTORY OF NUPE KINGDO M 73 and court historians, place his birt h in the middle of the fifteenth century.1
At this time , the tradition runs, there was no kingdom of Nupe , only
small chieftainships which , among the Beni, were united in a con•
federacy under the chief of Nku , a village near the confluence of Nige r
and Kaduna. At that time the Nupe people were tributar y to the Atta
(king) of the Gara, at Eda (Idah), far down the Niger. Th e tribute was
paid in slaves, and every family head had annually to contribute one
male member of his house. These slaves, as tradition has it , were always
sister’s sons. It so happened that the son of the Atta Gara came hunting
to Nk u in Nupe country. Here he met the daughter of the chief of Nku ,
a young widow , fell in love wit h her and lived wit h her for some time.
When the death of his father recalled hi m to his country, to succeed to
the throne of the Gara, this woman was pregnant. He left her a
charm and a rin g to give to their child when it was born . Thi s chil d
was Tsoede. The n the old chief of Nk u died, his son became chief,
and when Tsoede was 30 years of age the new chief sent him , as his
sister’s son, as slave to Eda. Th e Atta Gara recognized his son in the
new slave by the charm and rin g whic h he was wearing, and kept hi m
near his person, treating hi m almost like his legitimate sons. Tsoede
stayed for thirt y years at his father’s court. Once the kin g fell victi m to
a mysterious illness whic h nobody could cure. Th e court diviner
prophesied that only a fruit from a very high oil-palm outside the town ,
plucked by one man, woul d cure the king. Al l his legitimate sons tried ,
in vain, to obtain the precious fruit . Finally Tsoede made the attempt,
and succeeded. But in this attempt he cut his li p so badly that he looked
almost like a man born wit h a split lip . Fro m this time—and this still
holds true to-day—all hare-lipped boys bor n in Nupe are named Edegi.
Tsoede’s achievement, whic h made hi m stil l more beloved by his father
and honoured by the court, evoked the jealousy of his half-brothers.
Thus , when the Atta felt his death coming he advised his son to flee, and
to return to his own country the rule of whic h he bestowed on hi m as a
parting gift. He assisted hi m in his flight, he gave hi m riches of all kinds
and bestowed on hi m various insignia of kingship: a bronze canoe ‘as
only kings have’, manned wit h twelve Nup e slaves; the bronze kakati,
the long trumpets whic h are stil l the insignia of kings in the whole of
Norther n Nigeria ; state drums hung wit h brass bells; and the heavy iron
chains and fetters which , endowed wit h strong magic, have become the
emblems of the king’s judicia l power, and are know n to-day as egba
Tsoede, Chain of Tsoede.2
No w comes the story of Tsoede’s adventurous flight from Eda, travel-
1 See Appendix I I I . There exist several versions of the myth. I am giving here the version which I learned from the Etsu of Patigi, the last living descendant of the Tsoede dynasty.
2 Th e villages where these chains are found (all lying in the Niger valley) were
entrusted with the special mission of acting as the ‘ King’s Hangmen*. I have described this organization in Man, 1935, 143. Tw o places on the Niger, Tada and Jebba Island, possess sacred bronze figures which, like the chains, are linktd with the tradition and worship of Tsoede. According to some versions of the Tsoede myth they too were a gift from the Atta to the departing Tsoede; the Etsi of Patigi, my informant for the version given here, denied this.
74 POLITICA L HISTORY OF NUPE KINGDO M
lin g up-river, hotly pursued by his half-brothers and their men. On the way he is helped by tw o men, who m he later rewards by making them chief and second-in-command of the Kyedye tribe . Whe n he reaches the Kaduna river he turns into a creek called Ega, and lies here in hidin g til l his pursuers, tired of their fruitless search, return to Eda. Tsoede and his men leave the canoe and sink it in the river ; the people of Ega stil l perform an annual sacrifice on the spot, where, as the traditio n has it , Tsoede’s canoe was sunk, and at these ceremonies they are able, yo u are told , to see the brigh t bronze of the canoe glitter in the water. Tsoede the n went to Nupeko, a village near-by, kille d the chief, and made him • self chief of the place. He conquered Nku , the tow n of his maternal uncle, made himself the ruler of all Beni (or Nupe—all informants are vague on this point), and assumed the titl e Etsu, king . He made the twelve men who had accompanied hi m from Eda the chiefs of the twelve towns of Beni and bestowed on them the sacred insignia of chieftainship, brass bangles and magic chains. Th e present chiefs of Beni (as far as their * towns’ stil l survive) claim descent from these twelve men, and stil l treasure bangles or chains as insignia of chieftainship. Tsoede carried out bi g and victorious wars against many tribes and kingdoms, con• quering in the south the countries of Yagba, Bunu (two sections of the Yoruba), Kakanda, as far as Akoko, and in the nort h the countries of Ebe, Kamberi , and Kamuku . He resided first in Nupeko, whic h name means ‘Great Nupe’ , for eight years, and when Nupeko grew too small, he buil t the new capital of Gbara, on the Kaduna, whic h was to remain the e£i ‘tsu, the King’ s Town , til l the Fulani conquest. A t that time his residence is said to have counted 5,555 horses, so many in fact, that there was no room for them in Gbara, and one of Tsoede’s sons, Abdu , who was in charge of the horses, crossed the Kaduna and founded on the opposite bank a place whic h is stil l known by the name of Dokomba, ‘Horse Place’.
Apart from the royal insignia and emblems of magic Tsoede is said to have brought to Nupe from Eda certain crafts and techniques hitherto un • known in the country. He brought wit h hi m blacksmiths and brass- smiths who taught the crude blacksmiths of Nupe their more advanced technique; the canoe-men who came wit h hi m imported into Nupe the craft of buildin g large canoes of whic h the Nupe are said to have been ignorant at that time . Tsoede is also believed by some to have intro • duced human sacrifices—there is a gruesome story of the first sacrifice, the victi m of whic h was Tsoede’s mother’s brother—and to have established first the custom of bride-price.
We can to some extent ascertain what there is of historical trut h in this tradition. Certain indirect evidence tends to confirm the main facts of the Tsoede tradition . Th e alien, southern provenance of some of the cultural goods, whic h in the tradition are linked wit h Idahan d the coming of Tsoede, seems fairly certain. Some at least of the sacred bronze figures in Tada and Jebba, cast in cire perdue, a technique unknown in Nupe and the north , bear the stamp of the culture whic h we are wont to associate wit h Benin—and Benin, as we know, has for a long period been the political overlord of Idah ; the same is true of the beautifully cast brass
POLITICA L HISTORY OF NUPE KINGDO M 75 bells—among them the square-shaped bells typical of Benin work—on
the state drums found all along the Niger. Th e heavy iro n chains and
fetters, on the other hand, are most probably of Portuguese origin ; they
may have been brought in by slave-traders from the coast and have
found their way up the river into the hinterland—the home of these precious trading goods.1 Even the approximate date of the immigration of Tsoede and his band can be fixed wit h some accuracy. For on this point all the various genealogies of Nupe kings, obtained fro m quite different sources, tally. Givin g the years of reign of every Nupe king , back to Tsoede, they all place the life of Tsoede in the same period, the
early fifteenth century. It is interesting to note that the royal genealogies in two great neighbouring kingdoms, Yoruba and Ewe (i n Togoland), also trace back the history of their royal ancestors to the same period, about 1400.2 We may not be mistaken therefore in assuming that the beginning of the fifteenth century marks the end of a bi g wave of immi • gration whic h eventually led to the fir m establishment of a number of new West African societies—kingdoms, in our three instances. For Nupe we can go farther than this. Th e same period saw the first contact of the Guinea Coast wit h Europe. In 1498 the Portuguese established themselves in Benin. But at that time they found a Benin already on the verge of collapse, the once powerful kingdom weakened and in obvious disintegration. Idah , once tributary to Benin, and master over Nupe as of many other dependencies, must have weakened wit h it . Th e myt h of Tsoede’s delivery of Nupe from the yoke of Idah may well contain a kernel of historical trut h (whatever form the events may have taken in reality); the foundation of an independent Nupe kingdom may have been one of the historical results of the disintegration of Benin and its vassal states. Th e ancient contact of Ida h and Nupe as wel l as the former sovereignty of Idah , are undoubtedly historical. Th e tradition of Tsoede is as wel l known in Ida h as in Nupe. An d when some years ago the Atta of Ida h came to Bida—he was keen to visit that country of whic h the tradition of his tribe had so much to tell—a significant political imbroglio threatened the meeting of the Nigerian potentates. It arose from a question of precedence; the Ida h man insisted on being given a higher seat than the Etsu Nupe, basing his claim on the unforgotten historical fact of Nupe’s ancient vassalship to Idah.
Thi s bare framework of facts—if facts they are—has absorbed many other features, legends, traditions, embellishments; whic h of them contain any historical trut h it is impossible to say. But we are not con• cerned here to determine what is and what is not historical trut h in the rich cycle of traditions round the figure of Tsoede. Th e myt h as it stands, and as it is treasured by the people of Nupe, is to us a sufficiently significant social reality. It represents the typical * mythical charter’ of
1 Exactly the same chains and irons have been found on the Gold Coast; small models of these fetters, of brass, iron, or gold, are sold as charms all down the river and the creeks of Southern Nigeria. A common origin suggests itself forcibly.
2 See pishop Johnson, History of the Yoruba, and D. Westermann (personal
information).
76 POLITICA L HISTORY OF NUPE KINGDO M
the kingdom , and its common knowledge constitutes the first and fore• most of these common beliefs and forms of cultural ‘commonness’ whic h supply the background of political unity . I have found the con• sciousness of this tradition developed most strongly in the river area and among the Beni, that is, in the area that marks what I have called the
‘core* of the expanding kingdom. It is totally absent to-day, and most probably has always been absent, among the Dibo , who have never formed part of the ol d kingdom, and among some of the Nupe-ized groups (such as the ‘Gbar i slaves’) which , absorbed into Nupe in more recent times, have remained culturally ‘Outside Peoples’, Th e distribu • tio n of the ‘mythical charter’ of Nupe kingdom and the boundaries of the kingdom itself coincide.
To go on wit h the tradition of Tsoede. For sixty years he reigned in Gbara; he died, 120 years old , on one of his militar y expeditions to the north , near Gbagede. Some of his belongings, two stirrups and his sword, are stil l kept as almost sacred treasure by the people of Gbagede (who are Kamberi , not Nupe). At the time of Tsoede’s death the king • dom of Nupe was expanding towards the north , and his four sons who succeeded hi m as kings of Nupe, one after the other, are said to have founded Mokwa , in the north-west corner of Nupe, as their temporary capital. Another of his descendants, Abd u Waliyi , founded the capital Jima, on the Kaduna opposite Gbara.1 Th e fifteenth kin g after Tsoede, Jibiri , who reigned according to Nupe genealogies about 1770, was the first Nupe king to become Mohammedan. He was deposed by his son and died in exile, in Kutigi , where his grave can stil l be seen. Under Etsu Ma’azu, the nineteenth king of Nupe, Nupe kingdom reached its greatest power. Durin g his reign, about 1810, Malla m Dendo, also called Manko , ‘Great Mallam’ , a Fulani from Kebb i who was destined to change the whole fate of Nupe kingdom, is said to have first appeared in Nupe country as an itinerant preacher, diviner, and seller of charms. Wit h Etsu Ma’azu we also reach, for the first time, in this account, the plane of ‘objective history’; his death is recorded by one of the first European visitors to Nupe as having occurred in 1818.2
THE FULANI CONQUEST
We enter now a period of Nupe history whic h is illuminated by the ligh t of ‘real’ history. It is described wit h many details in the writte n records of Nupe court historians, the earliest of who m was a certain Abduramani, a Mohammedan scholar who seems to have lived in Raba in the early days of Fulani rule. It was during this period, too, that European travellers first penetrated into Nupe.3
1 In addition to the historical capitals of the kingdom , Gbara, Mokwa , Jima, and finally Rabd (see below), the Nup e records also mention certain other places at whic h the Nup e kings are said to have resided temporarily at various times, viz . Etsu, Jengi, and Yeti .
2 Journal of Rev. J. F. Schoen and Samuel Crowther, Expedition up the Niger,
1842, p . 191.
3 I owe to the kindness of the Administratio n the opportunit y of having been able to consult unpublished Government records, compiled by the first officers who were serving in this area; I must acknowledge specially the very complete
‘Notes on Nup e History’ , by Majo r Burdo n (1902).
POLITICA L HISTORY OF NUPE KINGDO M 77
Fro m that time Nupe history is linked indissolubly wit h the fate of the Fulani conquerors of Nigeria who had made themselves rulers of the main Hausa states in the beginning of the nineteenth century. Th e history of the Fulani Jihad has often been described,1 so I need say only a few words. Th e Fulani established their suzerainty over Norther n Nigeria in a loosely organized, decentralized dual empire, the western part under the supremacy of the Sarkin Musulm i of Sokoto, and the eastern under the Emi r of Gwandu. Nupe kingdom fell to Gwandu’s share. Th e Fulani conquest was achieved in two phases and by two methods: a gradual infiltration of the foreign country by Mohammedan Fulani, and then, ultimately, a militar y conquest. Th e same policy was adopted in Nupe. Malla m Dendo was one of those Mohammedan preachers and emissaries to ‘heathen countries’ who was to collect an ever-increasing group of followers as the nucleus of future conquest. Hi s followers seem to have included men of all stations: cattle Fulan i to who m in their semi-nomadic life Nupe had become a second home; Fulani and Hausa mercenary soldiers who took service in the Nupe armies; merchants from northern towns; and Mohammedan priests and missionaries; I estimated their number at the time of the conquest at not more than 1,000 or 1,500. When Malla m Dendo became strong enough to venture the decisive blow, he applied to the Emi r of Gwandu, his overlord, for military help and for a tuta, flag, which signified his official recognition as a leader and accepted feudal chief of the Fulani Empire.
Th e opportunity for the Fulani to make themselves masters of the country soon came. Almost immediately after Etsu Ma’azu’s death internal trouble broke out. Th e claim of Maazu’s son Jimada, the legi• timate heir to the throne, was disputed by his father’s brother’s son, Majiya I I . Th e details of this feud are not clear. According to some accounts Jimada was Majiya’s father’s sister’s son, who had been appointed regent for the boy-king Majiya ; later, when Majiya grew up and demanded his kingdom, Jimada refused. Other accounts make the two claimants sons of elder and younger brother respectively, namely, of Etsu Ma’azu(the nineteenth king of Nupe)and EtsuMama (the twenty- first kin g of Nupe). Majiya claimed the throne by the right of general Nupe succession, being the elder brother’s eldest son; Jimada, however, was regarded as the legitimate heir according to the ancient Nupe rules of royal inheritance whic h decreed that the son ‘bor n in the purple’ should become king.2 Whatever the origi n of this dynastic feud, it led to a temporary division of Nupe kingdom into a western and eastern half: Jimada reigned in Gbara, the ancient capital of Nupe, and Majiya buil t himself a capital in Raba, on the Niger. Majiya gained the friendship of Malla m Dendo and the F’ulani group in Nupe, and wit h them on his side he became by far the more powerful, and soon snatched the other half of
1 See A. C. Burns, History of Nigeria, Oxford, 1929; A. W. Bovill, Caravans of the Old Sahara, Oxford, 1933; C. K. Meek, The Northern Tribes of Nigeria, Oxford, 1925, i, pp. 98-102.
- In Africa this rule of inheritance is unique, and cannot, according to Nupe
tradition, have applied to Tsoede himself and his four sons who succeeded him. Al l informants insisted that it became, at some unspecified period, the rule of succession’in the Nupe dynasty.
78 POLITICA L HISTORY OF NUPE KINGDO M
Nupe kingdom fro m his rival . He defeated Jimada in the war, killed hi m at Ragada, near Jengi, and thus made himself the undisputed kin g of Nupe , about 1820. Idirisu , Jimada’s son, fled wit h the rest of his followers to Labozi, and later across the river, to Egga, where he stayed, an exile and fugitive, powerless against Majiya’s army. Nupe was not to remain at peace for long. Majiya’s gratitude to the Fulani who had helped hi m to defeat his riva l soon changed to jealousy and fear. He turned against them , only to be himself defeated, and to lose his throne to his sometime friends and allies.
Thi s is the story that is tol d in Nupe of the rise of the Fulani and of the final struggle for the Nupe throne. Majiya, as king of Nupe , had learned of the fame of Manko as a preacher and Mohammedan scholar and diviner. He invite d hi m to his court and installed hi m there in the influential position of his private counsellor and diviner. He trusted his prophecies and his advice and was convinced that his victories in the field were due to Malla m Dendo’s asiri (lit . ‘secret’, that is supernatural powers); he rewarded hi m magnificently, gave hi m money, slaves, horses. He took Malla m Dendo’s youngest son Masaba into his house as his foster-son, as the Nupe do wit h sons of friends. Mor e and more of Malla m Dendo’s ki n and fellow countrymen came to Raba to take service under hi m or under the king, til l at last Manko’s power and influence were more than equal to that of the kin g himself. Majiya now thought it high time to ri d himself of the dangerous rival . He drove Malla m Dendo and all his followers from Raba, and chased them fro m his territory , across the Niger, into Ilori n country.
Here another Fulani emissary, Malla m Alimi , had already been estab• lished for some time as kin g over the Yoruba of Ilorin . He now gave protection to the Fulan i refugees from Nupe . But Etsu Majiya, desirous of uprooting for ever the power of his dangerous vassals, decided to attack Ilori n and to conquer this southern stronghold of the Fulani invaders. He attacked Ilori n wit h 4,000 horsemen and 10,000 foot- soldiers. He marched for two days fro m Raba to Ilorin , and halted his army at the gates of the town . Th e victory seemed his. Thi s war of Majiya against the Fulani of Ilorin , as it is tol d in Nupe, is a magnificent epic. There were five Fulani emissaries in Ilori n at that time : Malla m Baba, who had fled wit h Dendo fro m Raba* and later became the first Emi r of Agaie; Malla m Dendo himself; Malla m Musa, another fugitive from Rabd; Malla m Maliki , who had resided in another Nupe town , Lafiagi, and had been driven from there by Majiya ; and Malla m Alimi , the Emi r of Ilorin . Dendo was the youngest of the five, bu t his keen intelligence made hi m at once their leader. He first sent wor d to Idirisu , Jimada’s exiled son, asking for his support in this fight against Majiya, the murderer of his father. The n Dendo prepared his most powerful asiri. Tw o brave men were sent out at night to steal into the enemy’s camp and to brin g back some earth. Dendo dug a large pi t close to the tow n wall . He put the earth into this pit , and covered its mout h wit h grass. On a Friday morning he entered the pit , taking wit h hi m fourteen dates. He stayed there til l the hour of evening prayer, but what he di d in the pit , what secret magic he performed, no one knows.
POLITICA L HISTORY OF NUPE KINGDO M 79
Whe n he reappeared, he called again the two men who had brought the earth the previous night, and ordered them to scatter it by night all round the tow n walls. Next mornin g Majiya attacked. A thick mist had fallen over the country. Majiya’s troops were unable to move in the mist. The n a sand-storm sprang up , blindin g men and horses. Th e horses bolted, and Majiya’s cavalry, placed (i n accordance wit h Nupe militar y principles) behind the infantry, trampled down their own men. Whe n the small army of Ilori n sallied fort h from the gates of the town , it met an enemy already half defeated. Majiya was beaten, he fled across the Niger, back to the protected Raba, wit h only 1,000 horses and 1,070 men left of his huge army. Thi s was not the end. Idiris u and his followers joined forces wit h Malla m Dendo. Together they pursued Majiya’s troops, marching their men up the Niger as far as Gbajibo, fifty miles nort h of Raba, where they were able to cross the river un • molested. Here Malla m Dendo is said to have used a clever ruse: for a week he collected horse-dung, and then threw it into the water; driftin g down the river it reached Raba; and the Nupe, who had hardly recovered from their defeat at Ilorin , thought that an enormous army of horsemen was approaching Raba and had already crossed the river. • Majiya fled into the interior, to Zuguma, leaving Rabd unprotected, an easy prey to the Fulani army.
Malla m Dendo repaid Idirisu’s help badly. He let hi m remain a puppet king of Nupe exiled to Egga,1 while he installed himself in the capital Raba* as the real ruler. Dendo and his people collected taxes and tribute ; they had the command of the larger part of the army ; they began to dispense justice, and generally to direct the affairs of the king • dom. A revolt of Idiris u ended in his being killed by Malla m Dendo’s troops, whereupon Dendo, ‘true Fulani’ (say Nupe informants), turned back to Majiya, previously his enemy, offering hi m friendship and recognition as kin g of Nupe—another puppet king , in far-away littl e Zuguma. Tw o years after this unification of Nupe, in 1833, tn e Y ea r when Oldfield and Lander visited Rabd, Malla m Dendo died.2 He left four sons (two having died before their father) and two daughters, the sons being Abd u Gboya, Usman Zaki , Mamud u Gwogi , and Masaba (Mama Saba), the daughters Gogo Sabaci and Gogo Wodiko .
Durin g Malla m Dendo’s lifetime Malla m Baba, one of his companions on the flight to Ilorin , set out to conquer a kingdom for himself. Follow• ing Dendo’s advice, he took twelve men, crossed three rivers, travelling by night and hidin g by day from the l^upe troops who migh t intercept his party. Th e small troop raided* the country, and growing as it marched, finally reached Agaie, whic h Malla m Baba made the capital of another (much smaller) Fulani Emirate in Nupe country. Revolts of the Nup e of that part, and attacks of Majiya, whose sovereignty had extended over Agaie, were in the end defeated, and soon Malla m Baba could enlarge his small kingdom by new victories: over part of the Bataci tribe on the river ; over Kacha, whic h subsequently fell to Bida ;
1 In 1833 Oldfield, whose ship, the Quorra, was anchoring off Egga, met there the exiled Idirisu (Laird and Oldfield, op. cit. ii , pp. 38-9).
2 Laird and Oldfield, op. cit. ii , p. 76.
8o POLITICA L HISTORY OF NUPE KINGDO M
and over the Gbari of Payi, whic h became later the independent Emirate
(again under a Fulani) of Lapai.
Malla m Dendo is said to have advised his sons not to seek, after his
death, official secular power, bu t to remain what he had been: the un •
crowned king , nominally only an emissary of Islam and the spiritual
head of the country. But the division into real and nominal, spiritual and
secular rulership di d not work. Tw o years after Majiya’s death (1835 ?)
Tsado, his son, like hi m puppet kin g of Nupe in Zuguma, revolted again
against the hated Fulani. He was defeated and killed by one of the
Fulani generals near Egbako. Hi s son, Dzurug i (The Red One), re•
turned to Zuguma. Bu t now Usman Zaki , the second son of Dendo,
himself adopted the title Etsu Nupe, demanding the ancient regalia from
the exiled shadow kin g in Zuguma and making Rabd again the official
capital of the kingdom. Fro m this time the legitimate dynasty of Nupe
abandoned all legal claim to the rulership over the countrv, and the new
Fulani dynasty assumed full powers. O f the other descendants o f
Malla m Dendo, Abd u Gboya, Usman Zaki’s elder brother, whose
leanings were more scholarly than warlike, became Alkali (judge) of
the kingdom, and Mamud u Gwogi , Usman’s younger brother, Shaba,
or Hei r Presumptive.
Fro m this time the history of Nupe kingdom is the history of constant intrigues, fights, wars, and rebellions. Fights between the sons of Malla m Dendo, wars against other tribes and against factions in Nupe, and rebellions of the descendants of the old Nupe dynasty, who di d not give up hope of recovering the throne from the Fulani, or rebellions of powerful generals of mercenary troops who claimed the rulership for themselves, and often rebellions of both , one using the other as a pawn in this game for the throne of Nupe. Masaba, the ambitious and most popular youngest son of Malla m Dendo, born of a Nupe mother and educated at the court of the Nupe Etsii Majiya, claimed the throne on the grounds that he was a real Nupe and not, like Usman Zaki, a Fulani by descent. He wo n the support of Tsado in Zuguma and Idiris u in Egga, the tw o shadow kings of Nupe, and finally even the support of the Emi r of Gwandu, the spiritual head of the Eastern Fulani Empire. Masaba led two revolts against Usman Zaki , the first from Raba, shortly after the latter had assumed power, and the second, from exile in Lade, south of the Niger, in 1841.1 It was successful and forced Usman Zaki to leave Nupe kingdom together wit h Umaru , the son of his eldest brother Majigi , and to return to the home of his ancestors, Gwandu. Masaba became king of Nupe.
Le t me enumerate the main facts in the involved history of Fulani- Nupe, wit h its revolts, its constant bloodshed, and its fratricidal wars. Durin g Masaba’s reign Nupe kingdom grew extensively. Masaba, re• siding first in Lade and later in Rabd, conquered the Kamuk u in the north , the Gbari in the east, and the Kakanda in the south; he entrenched himself firmly on the Yoruba side of the Niger, snatching the country from other Fulani chiefs who had gained a foothold there. A revolt of one of his mercenary generals, the Bornu general Uma r Bahaushe, drove
1 Schoen-Crowther, op. cit., p. 161.
POLITICA L HISTORY OF NUPE KINGDO M 81
Masaba from Rabd, and the king had to seek refuge in Ilorin. Umar Bahaushe, for some time undisputed master of Nupe, attempted to make himself Etsu Nupe. All Fulani rallied against him, Usman Zaki and
SKETCH MAP SHOWING TH E DIVISIO N OF ROYAL LAN D
- Royal Estates (esozi)
Umaru Majigi returned from Gwandu, their self-chosen place of exile, and joined forces with Masaba against the usurper. For three years the Bornu general drove the Fulani armies before him and beat them in battle after battle. At last the luck turned. Th e Fulani army under Umaru, son of Majigi, fleeing from the usurper’s army across the Kaduna river, found protection in the walled town of Bida. Wit h the arrival of Masaba’s troops from the south, and with the assistance of
G
82 POLITICA L HISTORY OF NUPE KINGDO M
the town-king of Bida and the twelve Beni towns whose confederacy is brought into play for the last time in Nupe history, Uma r Bahaushe’s army was defeated after a three months’ siege of Bida. Th e rebel general was killed , and when the Dayspring reached Raba, in 1857, Dr . Baikie saw the head of the unfortunate Uma r Bahaushe impaled at the landing- place.1
The n came a time of comparative peace. Bida became the centre of the new kingdom. Usman Zaki, now reinstated, was the first Etsu Nupe to reside there. He died in 1859 or 1860, and was succeeded by Masaba, whose second reign lasted til l 1873. Under his rule Bida was transformed fro m a huge war-camp into a capital worth y of the most powerful king • dom of Central Nigeria. Man y great buildings in Bida date from this time : the three royal palaces, the great mosque, and the lay-out of the night market. Masaba, suppressing revolts of the Gbedegi in Mokwa , of the riverain Kyedye, fighting wars against the Kukuruku , Kakanda and Bassa Nge in the south, and Gbar i and Agaie Fulani in the east, consolidated the country internally and expanded its boundaries farther and farther afield. Of the two descendants of the original Nupe dynasty, one, Yisa, of Jimada stock, had fought on the side of the Fulani against the usurper Uma r Bahaushe, and had then settled, half guest, half hostage, in Bida. Th e other, the descendant of Majiya, after having revolted, in vain, several times, fled again to the refuge of his father, Zuguma, in Kontagora. In Etsu Masaba’s time, Nupe kingdom is said to have stretched from Muly e and Awawu , in Ebe country, in the north , to Akpara in Bun u in the south, and from half Yoruba Tsaragi (Shari) in the west, to Payi, in Gbari , in the east.2 After Masaba’s death Umaru , the son of Majigi , became Etsu. He reigned from 1873 til l 1884. He is said to have been a fanatic Mohammedan, and he, too, helped to extend the boundaries of Nup e kingdom. He fought against Bunu and Gbira— and in this fight used for the first time gunpowder bought from the first Niger Company trading-post whic h had newly been established in Egga, on the Niger. He quelled a revolt of the Kusopa under a chief from Yet i who joine d forces wit h the exiled Nupe king of Zuguma, and another rising of the Kyedye. In this fight he had the help of the Kakanda, now tributar y to the Nupe, and of the Niger Company troops whic h then for the first time played an active part in Nupe politics.
Etsu Maliki , who succeeded Umar u Majigi , reigned from 1884 t o
- He is said to have been the wealthiest Etsu of Nupe, but (which
means, perhaps, the same) avaricious and cruel. Under his reign taxes
reached an almost insufferable height. His reign was on the whole
peaceful except for the usual border raids and punitive expeditions
against recalcitrant vassal tribes. Abubakari, 1895-7, followed. It was
under his reign that the Niger Company troops were fighting against the
1 A. C. G. Hastings, The Voyage of the ‘Dayspring*. London, 1926, p. 90.
2 About that time Dr . Barth, travelling in Hausa, was told of the great war• like kingdom of Nupe, whose boundaries he records; they are almost exactly the boundaries given above: Faze (in Ebe) in the north, Bunu country in the south, Tsaragi in the west, and Liffe (?) in Oban, in the east. (Henry Barth, Travels and Discoveries in North and Central Africa, London, 1857, vol. iv, Appendix V, p. 545.)
POLITICA L HISTORY OF NUPE KINGDO M 83
Nupe, finally taking Bida, in 1897, deposing Etsu Abubakari and appointing Mohamadu in his place.
Th e story of the Niger Company campaign has been often written . Between 1870 and 1890 the Royal Niger Company had established trading-posts all along the river, its northernmost posts being situated on Nupe territory. Shortly after the accession of Abubakari trouble broke out in the south of Nupe kingdom, in what is to-day Kabba Province, where the Yagba and Gbira had revolted against Nup e suppression. Th e Niger Company, whic h had many important trading-posts in this part of the country, promised them protection fro m Bida. Etsu Abu • bakari, no friend of the white man, decided to kil l two birds wit h one stone, to ‘pacify* Kabba country and to oust the hated Europeans. He moved his army south, under the command of Maku Mohamadu. But the latter, who nursed a grievance against the Etsu for having refused his claim to the title of Hei r Presumptive, made a pact wit h the British ; in exchange for the promise of Sir Willia m Wallace, who commanded the Niger Company troops, to make hi m Etsu of Nupe, he agreed to keep his army out of the fight. Th e Kyedye, always ready to revolt against the central government, manned their canoes, placed them at the disposal of the Niger Company and, under the command of a gunboat, under• took to guard the river so that no reinforcements should reach the beleaguered capital. Th e Niger Company troops marched on a city depleted of soldiers and arms. After twelve days’ fighting the stubborn resistance of Bida was broken. Abubakari was deposed, and Mohamadu, Sir Willia m Wallace’s ally, appointed Etsu. Th e chief of the Kyedye was rewarded wit h the grant of the semi-independent status whic h his country had lost under the Fulani , and was placed directly under the authority of the Niger Company (a prerogative whic h was later withdrawn). Moreover, acceding to the request of the Kyedye chief, Patigi, the territory on the southern bank of the Niger opposite the Kyedye capital Muregi , was detached from Bida and given, as a separate Emirate, to Idirisu , the exiled Nupe king of the Jimada branch, who had been livin g in Bida as the ‘guest’ (hostage, in reality) of the Fulani kings.
Th e Niger Company and later its successor, the Government of Norther n Nigeria, continued for some time this policy of breaking up the too dangerous bi g Nupe kingdom. Kol o Yisa, a grandson of the exiled Majiya, had thought of using the defeat of Bida for his own ends. He hurried to Zuguma, where his ancestors had lived as puppet kings of Bida, made himself king over the territory , and was later confirmed in this position by the Government. Yagba, Bunu , Bassa, and Kakanda had become independent districts and provinces immediately after the conquest of Bida. No w also the Nupe possessions on the southern and western bank of the Niger were taken away from Bida and placed in a different province, Ilorin , where they were divided up between the Emirs of Ilorin , Shonga, and Lafiagi (1900-5). Of the Nupe Empire that had reached from the Benue in the south to Gbagede in the north , only Bida Divisio n remained, bordered on all sides by countries populated by Nupe , butTule d by different sovereigns.
84 POLITICA L HISTORY OF NUPE KINGDO M
Etsu Abubakari, who had been banished to Lokoja after the conquest of Bida, made one attempt to return to his country and to recover throne and independence. He succeeded in collecting a following large enough to threaten Bida. He defeated Etsu Mohamadu and drove him from Bida (1906). But the British, by this time firmly established, helped Mohamadu, brought him back to Bida, and sent Abubakari back to Lokoja, where shortly afterwards he died in exile. Mohamadu died in
1915, succeeded by Etsu Bello, 1015-26, Etsu Mallam Saidu, 1926-34, and finally the present Etsu, Mallam Mohamadu Ndayako.
Viewing the historical growth of the Nupe Empire as a whole we discover the familiar two trends which, closely intertwined, force political systems into an endless twofold development: from expansion to military and political consolidation, in the course of which the necessity of further expansion seems invariably entailed. Th e roots of this de• velopment are no less evident in Nupe history than in the political developments of our own history. First, newly conquered territory demanded in due course conquest beyond its boundaries in order to ensure the pacification of the conquered areas and to prevent inter• ference from tribes or states without. And second, the expanding terri• tory absorbing more and more alien groups, all to be controlled and governed from a small nucleus-country, by a population less than one- fifth the size of the population that was brought under political control, necessitated continuous ‘re-conquests’ and wars of consolidation. In the field of social organization this development becomes manifest in the rise of the feudal system; in the military sphere, in the ever-increasing employment of mercenary troops.
Not every step in the expansion of Nupe kingdom was carried out by war. There is evidence of peaceful expansion as well. Certain com• munities and tribal groups are said to have ‘submitted’ (gba) voluntarily to the Nupe state, accepting the position of tributary groups and claim• ing at the same time the political protection*which the Nupe overlords extended to their subject groups. But accepting this ‘protection’ meant largely avoiding the more disastrous subjection. Though not the direct result of war and conquest, it was yet a submission to the threat of war and conquest which the momentum of a steadily expanding Nupe must have seemed to entail to her weaker neighbours.
To discover with equal certainty the first stimulus of the whole de• velopment, the force which, as it were, set the stone rolling, is impossible. Th e origin of Nupe rule and expansion is buried in the mythical past. Whether it was the immigration of a large tribal group (condensed in the myth to the small band of men round Tsoede) or a small ilite of the kind Pareto likes to discover in every political achievement, we have no means of deciding. We do not even know what the original social structure and economic system of this immigrant group was, or whether it contained any stimulus to territorial expansion or conquest. Only in the Fulani conquest does this first ‘stimulus’ lie open to historical examination. We have described to some extent the gradual infiltration of Nupe by the Fulani, who came as herdsmen, mercenary soldiers, and Mohammedan preachers. Here a distinct tribal section, a semi-nomadic group on its
POLITICA L HISTORY OF NUPE KINGDO M 85 wanderings from west to east and south-east, was the carrier of conquest.
It was separated from the groups whic h it subjugated by ethnic descent
and social structure, consisting apparently of a loose aggregate of kinship
groups and self-constituted warlike bands, and only partly by its system
of livelihood; for the predecessors of the Fulani, the Nupe rulers, had
already been soldiers and had themselves evolved an urbanized civiliza•
tio n in whic h Mohammedan culture played its part. But the analysis of
Fulan i political history goes beyond the scope of this book.
We can study some of the stimul i which , though perhaps it di d not initiate, yet certainly influenced and directed the political growt h of Nupe . Limitin g ourselves to the periods that come under the heading of
‘real’ history, we can point out the essential difference in the policy of expansion before and after the Fulani conquest. Th e expansion of the pre-Fulani kingdom appears to have been most marked in the north , and of the Fulani era, in the south. Tsoede’s grave, symbolic to all Nupe kings of what we migh t call his ‘political testament’, lies in Gbagede,
100 miles nort h of their northernmost capital, Raba. Th e Fulani Emirs , on the other hand, drove their armies farther and farther south; only in recent times, less than two generations ago, di d they begin to ‘con• solidate’ the Gbari provinces, thirt y or forty miles nort h of Bida. Th e conquest of the northern territories, inhabited by primitiv e pagans who possessed no handicrafts or remarkable wealth, had evidently tw o main aims (apart from the necessity of pacifying the border countries): first, to tap the rich , easily accessible reservoir of slaves whic h the Nupe found among these primitiv e and poorly protected groups; and second, to pro • tect and attract the trade from the wealthy centres of civilization in the Hausa states. Of this latter aim we have concrete proof. Th e main trade- routes from Hausa passed through, and partly ended in , Nupe, crossing on their way the territories of unruly hostile tribes. Raba was the goal of the huge caravans that carried regularly kanwa (potash) fro m Lake Chad over Kano to the countries of the south.1 Whe n the first European explorers, in the times of the last indigenous Nupe kings, reached Nupe country and Raba, they found it a tow n ric h in cattle (the humped, northern, breed), horses, camels, and ostriches—which must all have come from the north—and the markets ful l of precious goods from Hausa, Bornu , and Nort h Africa.2
Th e turnin g to the south represents a new political interest, bound up wit h a later phase in the history of West Africa. Fulani-Nupe, as the southern outpost of the Fulani Empire, carried its wars of expansion into the pagan south and no longer into the northern countries whic h had come under the sovereignty of various Fulani potentates to who m Nup e was tied by common allegiance. Th e south offered, over and above new densely populated areas for slave raiding, the command of the main route for the slave-trade, down the river to the coast whence ships took the slaves to foreign ports. No t unti l the F’ulani subjugated the Kakanda
1 See Barth, op. cit. ii , p. 115.
2 See J. and R. Lander: Raba had large herds of ‘horned cattle remarkable for
size and beauty’ and a ‘prodigious number of excellent horses’ (op. cit. ii , pp.
8 6 POLITICA L HISTOR Y O F NUP E KINGDO M
trib e could Nup e canoes, wit h slaves an d various goods, trave l unmolested between Nup e an d Lokoj a an d Onitsa , having had , before then , t o stop a t Egga an d t o tranship thei r loads t o Kakand a boats.1 Th e highly developed trades and crafts of the south, capable of supplementin g the norther n trade , prove d another attractio n to a kingdo m that ha d grow n ric h an d fastidious. W e have hear d o f th e Yorub a slaves importe d b y the Fulan i Etsu Masab a because o f thei r skill i n weavin g an d dyeing. Finally , the south became of paramoun t importance as th e mai n door to Europea n imports . Thi s importanc e reached its clima x whe n Etsu Umar u bought gunpowde r for his long-barrelled flintlock guns o f Ara b patter n fro m the store o f the Nige r Compan y o n his territory—nort h an d south matche d again in most profitable manner .
Ou r sketch o f th e growt h o f this Wes t Africa n kingdo m has shown political expansion dominate d b y economic interests. I f other motives— a collective desire for militar y adventures an d glory—wer e decisive in shaping it , the y are no w invisible i n the historical perspective. Th e im • petus of religious zeal, whic h drove th e Fulan i armies across Norther n Nigeria , was spent when , a s rulers o f Nupe , the Fulan i turne d thei r faces southward . I wil l no t pursue farthe r this question of political philosophy, o f the ultimat e stimulus behin d political expansion. Bu t w e shall see that economic considerations were involved in th e buildin g and maintenance o f the kingdo m i n man y other, mor e subtle, ways. Th e incentive of gain wil l appear not only as an abstract, impersonal, histori • cal stimulus , bu t as a concrete element, directly affecting individua l action an d directly shaping the motives an d interests by whic h the in • dividua l agents of the political system are animated .
1 See African Political Systems, Oxford University Press, 1940, p. 170, n. i.
CHAPTER VI I
POLITICA L HISTOR Y O F NUPE KINGDO M
(continued)
KINGSHIP
N the traditio n o f Tsoede, that ‘mythica l charter’ o f Nup e kingdom , the essence of the Nup e state is already clearly expressed, sanctioned, almost sanctified, by that paramoun t authorit y tha t attaches to pre • historic events: it outlines a system of political domination , growin g by conquest and expansion, and governed by a group detached in origi n an d status fro m the rest of the population , an d claimin g its supreme position b y the righ t o f might . Th e small rive r kingdo m whic h figures , in the Tsoed e tradition , as the nucleus of the great Nup e of later times, an d the vast empir e unde r the Fulan i Emirs , are essentially of the same nature . Th e twofol d process of expansion over alien groups an d cul • tures, combine d wit h cultura l assimilation and absorption, is reflected in both , in the ideological history of the Tsoede myt h as wel l as in the ‘real ‘ history o f the Fulan i kingdom . I n one there i s Tsoede , wh o conquers Nup e wit h the help o f alien magic, wh o brings into Nup e the insignia o f alien rulership and culture , but who , himself hal f Nup e b y descent, creates a new , independent, and unite d Nupe . In the other there is this remarkable piece of empire-policy whe n Masab a claims succession to the Nup e throne o n the grounds o f his being half Nup e b y birt h an d ful l Nup e b y education. I f the mythology o f Tsoede establishes the essence and growt h of kingship, later history shows its vicissitudes involved in such system of dominatio n whic h forces unit y upo n heterogeneous groups an d cultures, revolts and feuds, kings deposed an d exiled,
usurpers and shadow kings.
W e kno w little about the system o f kingship i n pre-Fulan i times. I have mentioned one piece of informatio n already whic h I obtained fro m various sources, namely , the ancient rul e of succession which made the eldest son’bor n in the purple’ , that is, the son firstborn durin g the father’s reign , the heir to the throne of Nupe . A few other facts concerning the person of th e kin g I was able to ascertain fro m the Etsu of Patigi . Certai n ritua l rules obtained at the king’s court—an d some of these rules are still observed in Patigi. Thu s the Etsu ma y only wear white ; he can trave l an d go out whenever he likes, bu t ma y not eat any food save that prepared by his wives ; nor ma y any stranger watch hi m eat. Finally , no one ma y be kille d in the presence of the kin g or in the tow n in whic h he is residing.1
1 Thi s helps us to understand the organization of the ‘King’ s Hangmen’, the crucial fact of whic h was that the royal executioners kille d their victims in places far away fro m the king’s capital. Anothe r effect of this observance, whic h is stil l visible, is the rule whic h forbids the performance of the ndako gboyd, a cul t believed to lead to the death of witches, in Patigi, the present ‘king’s town’ , or anywhere withi n a distance fro m whic h ‘the shrieking of the spiri t could be heard in Patigi*. (See Man, 1935, 173.)
88 POLITICA L HISTORY OF NUPE KINGDO M
Althoug h the present-day political system of Nupe appears to repre• sent in many respects only a copy and continuation of the pre-Fulani system, the characteristic features of ancient Nupe kingship—the former rule of inheritance, the ritua l taboos, and the sacred emblems and form of magic linked wit h kingship—have disappeared under the Moham • medan Fulani Emirs. But already a new background of religion and myt h is formin g behind the modern kingship. Th e memory of the ancestor of the Fulani rulers of Nupe, Malla m Dendo, has become sanctified. Th e stories of his life and exploits, emphasizing his super• natural faculties, have the flavour of myths. An d his tomb at Raba, over whic h a mosque has been built , became a place of worship and pilgrim • age. At the time of the election of the present Emir , when a number of important political posts were due to change hands, the Rani, a Fulani prince anxious to obtain one of these posts, went to Raba, fasted for seven days, sacrificed in the grave-mosque, and prayed to his ancestor for help.
Th e disappearance of ancient Nupe kingship reflects the far-reaching historical changes whic h Nupe kingdom had undergone in the Fulani conquest. In pre-Fulani Nupe the lin k wit h magic and myth , ritual rules and taboos, and possibly even the law of succession, separated the king from the rest of the people, even fro m the members of his own house. Fulani rule, only recently established by undisguised conquest, and de• pendent upon the co-operation of a band of war-chiefs and faction- leaders, turned semi-sacred kingship into rulership of the strongest. Th e king becomes a primus inter pares, the highest rank-holder in a royal nobilit y ruled by precedence and promotion. Th e exclusiveness of primogeniture gave way to a system of succession that allowed for balance of power and could satisfy rival claims. Thi s final balance was not achieved at once. Of the seven sons of Malla m Dendo three only were destined to found royal houses, two directly, Usman Zaki and Masaba, and one in the second generation, Majigi , through his son Umaru . Having once concentrated militar y and political power in their own hands, these three families eliminated the other sons and grandsons from the succession, paying them off wit h political offices the importance of whic h decreased from generation to generation. Thi s resulted in three dynasties whic h succeed to the throne in strict rotation. In the genea• logy given i n Appendix I I this development i s clearly visible. Thi s historical explanation of the present system of kingship is borne out by the views of the people themselves. One of my most enlightened infor• mants, a very learned man and himself a member of the royal nobility , defended the present for m of succession against the old primogeniture on these grounds: the old system could work in ancient Nupe , when there was a powerful kuti (magic) behind the Etsu’s authority. For the Fulani dynasty, whic h came to power as a large group, composed of jealous relatives and partisans, a divided succession was the only solution. Th e country too, he added, could only benefit from such balance of power.
We may mention here another change from pre-Fulani to Fulani king • ship whic h concerns the political status of the Nupe state. Whil e the Nupe kings were absolute masters in their own country, the Fulani
POLITICA L HISTORY OF NUPE KINGDO M 89
Emirs were, in a loose sense, vassals of a larger state, the Fulani Empire of Gwandu. Beyond an annual tribut e by Bida to Gwandu (reduced to • day to a nominal gift) and the merely symbolic confirmation by the Emi r of Gwandu of every newly appointed king of Nupe , expressed in the traditional bestowal of a flag, the vassalship involved littl e inter• ference wit h the internal government of Nupe. Th e Emi r of Gwandu acted mainly as the guarantor of the various political systems under his suzerainty. In this capacity he was sometimes appealed to in boundary disputes and suchlike, and he woul d support his arbitration if need be by militar y action, as happened in the dispute over rights to raid between Bida and Agaie (see p. 115). In tur n Bida, like all other Fulan i vassal states, could be called upon to send militar y help to her suzerain. Th e overlordship of Gwandu was also appealed to in the early dynastic disputes between the descendants of Malla m Dendo. It was due to the advice of Gwandu that Usman Zaki ceded the throne to Masaba; Gwandu helped the Fulani against their rebellious general Uma r Bahaushe; and Gwandu supported later the reinstatement of Usman Zaki. Th e influence of Gwandu upon dynastic affairs dis• appeared, characteristically, wit h the gradual establishment of a fixed system of succession which could itself safeguard balance of power and the conciliation of rival claims.
Th e rival character of the threefold dynastic structure is abundantly manifest in present-day Nupe. Every possession or privilege linked wit h Nupe kingship reflects the tripartite arrangement. Th e three highest ranks in the hierarchy of the kingdom are vested in the three royal ‘clans*: the rulin g king , head of one royal house, takes the titl e Etsu; the head of the royal house next in succession, the titl e Shaba,
‘Hei r Presumptive’; and the head of the thir d house, the titl e Kpotu. Th e land of the Beni round Bida where the Fulani settled after their decisive victory, appropriating all vacant land, is divided into three parts: the family of Usman Zaki owns the country east of Bida, the family of Umar u Majig i the country in the south, and the family of Masaba the country in the west.1 Bida tow n is divided, as we have seen, into three ekpa, each wit h its royal palace, and each bearing the name of one of the three royal families. Every king adding more land and power to his king • dom is adding at the same time more land and power to his own family, and distributes land and wealth among his own followers and favourites. Every Etsu Nupe uses the period of his reign to increase to the utmost the wealth and power of his house. Th e only balance of power lies in the rigi d system of succession. For favouritism and extortions from the rulin g Etsu may wi n for the house of his successor the favour of the masses; oppression from one may mean help from the other; the increase of power i n one royal house wil l eventually be balanced by the growth o f the house that succeeds it to the throne. It is a precarious balance, and its history, whic h is the history of Bida to-day, is writte n in rivalries, feuds, intrigues, and revolts, whic h threaten constantly the peace of the country and tear the people of Bida into rival factions.
Th e struggle for dynastic power is as apparent in the economic field.
1 See sketch map, p. 81.
90 POLITICA L HISTORY OF NUPE KINGDO M
Th e control over the wealth and the potential resources of the country, though not exclusively the king’s, is yet sufficiently his to render his economic position unassailable. Th e king of Nupe derived his income firstly from the taxes and tributes whic h his officers collected, and of whic h he received the largest share, and secondly, from the wars and raids whic h the king’s armies carried into foreign lands. Th e king’s private estates constituted a thir d important source of income. These estates were worked by slaves who lived in small hamlets on the royal lands called eso£i. We shall discuss the organization—past and present— of these royal estates later, in the chapter on land tenuie. Here we need only say that not all the Emirs of Nupe were equally keen farmers; some, Etsu Mohamadu for example, let things drift ; they took little interest in their estates, neglected their esoXi, and cared only about the regular payments of their share in the yield of the farms.
Other sources of income are less clearly defined—profits made from commerce (above all , from slave-trade), in which an Etsu Nupe, like every Nupe nobleman, might sometimes engage; or proceeds from bribes, extortion, and the confiscation of estates belonging to rebels and other ‘enemies of the country’. One interesting variety of this latter kin d is the exploitation of a religious institution, a ‘king’s cult’ of com• paratively recent origin, the anti-witchcraft society whic h is known in Nupe as the ndako gboyd. Th e Fulani Emirs had ingeniously grasped the control over this powerful and lucrative secret society and shared the proceeds wit h its members; the share of Etsu Bello in one year is said to have amounted to over £100.1
Another source of income, which wil l gain the economist’s special appreciation, lay in the royal monopoly of the kola-nut trade. Yo u wil l find to-day two kinds of kola-nuts on the markets of Nupe (or for that matter Nigeria): the small cheap kind , which is imported from the Gol d Coast and Southern Nigeria, and the large, most expensive kola-nut which is grown in Nupe country, on the farms of the kola-nut planters of Labozi. Til l the days of British rule this latter kind , famous all over Nigeria, could be procured only by the members of the nobility ; no commoner or peasant had any opportunity of obtaining Labozi nuts. Th e trade in these nuts was a strict monopoly of the Etsu Nupe, the kola-nut growers being forbidden to sell their crop to anyone except his own agents.2 Stories are current of traders smuggling the precious fruits out of Labozi by night, lest they should be seen and caught by the men of the king. In the kola-nut monopoly an instrument of sharp class- prerogative was cleverly combined wit h an efficient economic measure.3
But to what extent the Etsu Nupe disposed of the kola-nuts as gifts to his courtiers and favourites, and to what extent he made money by sell• ing them to the nobility of the country, I am unable to say.
1 See p. 142 and Africa, 4, 1935.
2 The Etsu Nupe in turn supplied the kola-nut planters with slaves to help them in their work. In pre-Fulani days the kings of Nupe are reported also to have presented one slave every year to Laboii to be sacrificed in the annual ritual with which the Labozi planters secured the thriving of their crop.
- See p. 129.
POLITICA L HISTORY OF NUPE KINGDO M 91
Ou t of this large and varied income the kin g gave directly very littl e to the common weal: he ‘financed’ certain public works, the repair or the buildin g of the city mosque, his own palace, and the tow n walls and gates. Th e work itself was organized by the members of the lower nobilit y and carried out by their slaves and serfs; the kin g contributed only food and drin k for the labourers, and small presents for the men in charge of the work. Fro m his income, but assisted by gifts from other members of the royal house, the Etsu also paid the annual tribute to the Emi r of Gwandu, by whic h the kings of Bida recognized the sovereignty of that ancient centre of the Fulani Empire.1 Finally, the king , equipped his army wit h horses and arms, and formed of his personal slaves the police force whic h had to keep order in markets and streets. But indirectly the kin g contributed very considerably to the economic life of the country. Th e court was the centre round whic h the economic and cultural life of the capital revolved. Th e splendour of the court life of Nupe , famous in Nigeria, meant work and income for thousands, incessant production in the workshops of the craftsmen, and rich business for the merchants. Beautifully buil t houses to live in ; gorgeous gowns, horses, saddles, arms, to be given as presents to courtiers and members of the household; food and drink , music and dances, for the great feasts celebrated at court, these were (and still are) the daily needs of the royal household. Presems play an important part in the etiquette of Nupe kingdom : from a small bowl of kola-nuts to a silk gown or a horse range these expressions of the king’s favour. An d at the great Mohammedan festival of the year, the SaUah as it is called in Nigeria, every one of the hundreds of people who by the right of birt h or status are entitled to the king’s attention are honoured wit h a smaller or larger gift. To be left out means disgrace; and the sending of a present to a man wit h who m the Etsu was known to be on bad terms expresses conspicuously reconciliation and new favour.
Of the former sources of royal income only two have survived under modern conditions: profits from occasional trade, and the proceeds of the royal estates (materially reduced by the transformation of slave labour into labour on the metayer system, or even wage-labour; see p. 200). To this we have to add to-day the Emir’s considerable salary. On the other hand, his expenses, too, have been greatly reduced: public works, the upkeep of a police force, even the buildin g of the royal residence, now for m items on the list of Native Administration expendi• ture. But ail the current expenses, all those demanded by status and prestige, by the etiquette of the kingdom, and by the necessity of win • ning supporters for himself and his house, are stil l as much part of the unwritte n obligations of an Etsu Nupe as they ever were.
A DAY OF COURT ROUTINE
Every mornin g the Etsu mounts his horse and, accompanied by members of his household and some of his officers of state, rides round
1 In the first period of British rule this tribute consisted of 100 gowns worth
£300, 200 gowns worth £125, and one silver sword worth £10.
92 POLITICA L HISTORY OF NUPE KINGDO M
the walls of his house. Thi s morning ride is stil l part of the daily routine of the Etsu Nupe, even up at his modern brick residence whic h stands by itself on the hil l outside the town. He may extend his ride to visit his eso&i, a short distance away, and inspect his farms, or possibly to inquire after his guests—for visitors of standing who have come from far away are assigned quarters in the esozi. Once when I accompanied the Etsu on this ceremonial morning ride he went to see a Nupe Malla m who had returned fro m the pilgrimage to Mecca and stopped in Bida as the king’s guest on his way home. Another time two noblemen from Sokoto, relatives of the Sultan, had come on a visit to Bida and were invited to stay in the esozi. Later in the morning, back in his house, the Etsu receives the first official visitors of the day: the Etsu Dogari and the Sarkin Yandoka, the chiefs of the two police forces of the Emirate, who come to brin g hi m the ‘news of the town’—arrests made on the market the night before, a case of small-pox in one of the tow n quarters, and so forth . Th e Alkali and his Chief Assistant drop in to pay their respects, possibly to report on an important law-case, before they go down to the Court House in the centre of the town . Th e guild-head of the traders may present himself in order to infor m the Etsu of the arrival in Bida of traders or a trading caravan wit h new attractive goods whic h the Etsu might care to inspect. Messengers come and go—from the Native Ad • ministration Treasury, or the Divisional Office, the Bida Tow n Council, or from a Distric t Head. In the meantime the highest officers of state, the members of the Emir’s Council, have arrived in his house for the daily nkd, the Council of State. At the nko all matters of importance are discussed; or there may be legal cases on the day’s agenda. For the nko has also judicia l functions and tries cases referred to it from a lower court. Th e session of the Etsu’s Council concludes the official duties of the day. In the afternoon the Etsu might see friends or visitors in his house, or drive through the tow n in his splendid motor-car—one of the modern symbols of royal prestige.
Th e climax of Bida Court life is the celebration of the Friday. On Thursday night and again on Friday afternoon the Etsu rides in great state to the mosque in the town , and on Friday at his return he holds a reception in his house whic h is attended by every man of rank and standing. Envisage this Friday procession: king and courtiers on horse• back, in their sumptuous gowns, the king under the great silken state umbrella, the horses wit h trappings of silver and beautiful cloth ; a body• guard carrying swords, and police wit h their staffs, are running ahead, shouting the Etsu’s name or blessing formulae or littl e proverb-like sentences whic h illustrate the king’s greatness; drummers are beating their drums, three mounted trumpeters blow the huge bronze kakati in an incessant deafening chorus; wit h it blend the shril l notes of the algeita, the Hausa oboe, while another musician is beating an iro n double bell wit h a wooden stick, shouting and singing at the same time . Al l streets are lined wit h people, adults and children, who watch, fascinated, this procession, bearing witness to the impressivenessof this display of royal power. An d then, back again in the royal palace: the wide place in front of the house and the courtyard are crowded wit h people; the talakazi—
POLITICA L HISTORY OF NUPE KINGDO M 93 the commoners, men without name and rank—sit and stand outside,
chatting and watching what there is to watch; the members of the
privileged classes, titled nobility, and Mohammedan priests, are sitting
in the inner courtyard and in katamba (entrance-halls), each katamba
reserved for a special group. Only royal princes and the highest officers
of state are admitted to the presence of the Etsu; they sit in the inner
part of the house together with personal friends and favourites of the king.1 Al l these people come to pay their respect to their sovereign, to see and to be seen, to exercise prerogatives, or to proclaim their con• nexion with the privileged and mighty. This external arrangement of the Friday ceremonial is indeed a symbol of the whole social structure of Nupe kingdom with its rigid system of etiquette and precedence, its
differentiation of status, rank, and prerogatives, and its display of wealth and power. This structure we must now analyse in detail.
THE ROYAL NOBILITY
Th e descendants of the house of Mallam Dendo, under their three heads Etsu, Shaba, and Kpotu, form the royal nobility of Nupe, the ena gitsu£i, the Group of Those who will become King. They are gitsu%i in the sense that from their ranks the new king is elected. The titles of nobility which they hold by the privilege of birth may lead every one of them, by slower or more rapid promotion, eventually to the highest rank, kingship. In most cases the royal prince who succeeds to the throne has reached first the rank of Shaba; the change from old to new king is thus made almost automatic, a dangerous interregnum is avoided, and the fight between the rival factions is confined to the less dangerous sphere of competition for the ranks next to the king. Only twice was a man who had not reached the title of Shaba (or, if the Shaba had died, the title of Kpotu) made Etsu.2 In 1936 there existed, or more precisely, were actually held, 41 royal ranks, of which 10 fell to the house of Usman Zaki, 17 to the house of Masaba (the house which had supplied the last
1 Th e arrangement at these receptions, the way in whic h the people wh o are admitted to the presence of the Etsu group themselves, is invariably the same. It is shown in this diagram:
Tw o bodyguards wit h ostrich fans
Etsu (on state bed)
GitsuXi Ndeji Sarakizi
Mayaki
Liman-Alkali
Mallams
2 These exceptions were Etsu Mohamad u who, only Maku, was appointed Etsu by the Niger Company, and the present Etsu who had been Ndgenu, bu t had held the important office of a Distric t Head of Jima. It is the present tendency of the Administratio n to abolish the office of Shaba and to make the appointment to kingship a free election fro m the royal ranks. After the death of the late Shaba in 1935 no new Shaba has been appointed. It is not yet clear what effect, if any, this innovation wil l have upon the internal dynastic organization. Th e threefold system is so deeply rooted that I believe the effect wil l be only to abolish a name, and that the rank next i n order i n the dynastic section wil l assume the main function of the abolished rank, that is, headship of the dynastic clan.
94 POLITICA L HISTORY OF NUPE KINGDO M
Etsu), and 14 to the house of Umaru , the present rulin g house. Th e death of a high rank-holder and the subsequent appointment of a successor is frequently made an occasion for a far-reaching reshuffle of ranks, called in Nupe wuruwuru. It goes hand in hand wit h a large-scale exchange of houses, younger brothers or sons moving into the vacated house of their older brothers or fathers (vacated by death or promotion), for the state residence of the highest gitsu ranks goes wit h the title . Rarely would a new house be buil t in Bida to-day for a waci, a Great One, as the members of the royal nobilit y are also called; ‘only talakazi, Poor Ones’, said an informant, ‘must buil d themselves new homes.’ Appointment to a first rank and promotion to a higher rank is granted by the king in council wit h royal princes and the other officers of state. An y able- bodied man in the Fulanifamilies who has reached senior age and who has proved his ability and loyalty can claim to be admitted to the ranks of the title d gitsuzi.x But there is no rule. Favouritism, intrigues, bribes play their part in appointment and promotion. Th e ranks whic h a family can accumulate reflects its influence at the court and its favour wit h the rulin g king . It is no accident that the royal house of the last Etsu (who had only just died) counts more ranks than either of the other two dynasties. Unlik e the system of ranks which we found in the village these royal ranks do not leave an ol d man or an invalid who can no longer fulfil his duties i n the possession of his office. I n such a case a younger man is appointed to the rank, and the ol d holder is know n as the ‘Ol d Kpotu\ or whatever the rank may be. A similar difference between the rank systems of village and state lies in that perfect health is an indis• pensable condition of appointments to any state rank. Deformity , even of the slightest kind , is ruled out. Th e younger brother of one of the Emirs of Bida, who had a withered arm, was unable to obtain any title at all. In the state system of ranks, whic h was based on service in the war and on offices vita l to the safety of the state, respect for ol d age or con• sideration for even minor disablements has no place.
Th e investiture—it is the same for royal princes and for officers of state—is an interesting ceremony, an impressive proclamation of the main facts on whic h the feudal kingdom is built : social promotion, the oath of fealty and loyalty and responsibility towards kin g and country. A propitious day for the investiture is first fixed by Mallams or diviners. The n follows the invitation to the ceremony—gifts of kola-nuts sent by the candidate to all men of rank and standing. Drummers in the house of the Etsu announce the beginning of the ceremony; all rank-holders are assembled in the Etsu’s house. He addresses the candidate cere• monially, proclaiming three times his appointment. Th e Alkali (or Liman), as representative of the clergy of the kingdom, clothes the candi• date in an embroidered gown and ric h cloak—gifts of the Etsu—and places a turban on his head. The n he leads the prayer, and the Etsu addresses a final exhortation to the new officer of state: ‘Practice for-
1 Th e Nup e of Patigi have preserved a linguistic distinction whic h has dis• appeared in Bida, namely, between royal princes who have received titles and those who have not. Th e former are called b’rama (their traditional salute is lemma, i.e. behold the rama), and the latter gitsuzi.
POLITICA L HISTORY OF NUPE KINGDO M 95 bearance; do not set the hearts of your men against each other; beware
of conceit, and do not fail the people.’ Al l present repeat, like a formula
of confirmation: Hakika, hakika, ‘verily , verily’ . Drummers and
musicians intone the songs and dru m rhythms whic h are associated wit h
the family and rank of the newly knighted; he mounts his horse and
rides, accompanied by all rank-holders, through the town , dismounting
at each house of a prince or saraki, making his greetings to the head of
the house. If the newly appointed (or promoted) nobleman is a member
of the royal dynasty he stops at each of the three royal houses and says a
prayer at the grave of his ancestors. A great festivity in his house, lasting
for several days, to whic h everybody in the tow n is invited , noblemen
and commoners alike, concludes the ceremony.
Descendants of the men who conquered the country and appropriated all vacant land, the gitsuzi may become, through ordinary inheritance, owners of land and considerable wealth. But the appointment to a royal rank meant more than this chance of inheritance: it meant status, in • fluence, and access to increasing power and wealth. Th e princes of rank were admitted to the nko, the Council of State, but had littl e influence, it seems, upon the actual government—this was entrusted to the council of the more competent, appointed, officers of state. Th e power of the gitsuzi rested in the main on two things: the part they played in the wars, and their position as egba, royal fief -holders and delegates placed in charge of the districts of the kingdom. We shall speak presently in more detail of these services whic h the royal nobilit y rendered to king and state. I wil l give here a list of all the ranks held at present among the gitsuzi of Bida. To illustrate the system of promotion I have indicated certain changes in the distributio n of ranks whic h had been anticipated by the Etsu and his council after the death of the Shaba in 1935; as already said, the Administration was not prepared to endorse this reshuffle, whic h woul d impl y the reoccupation of the post of Shaba (see p. 93, n. 2). Th e list also shows some of the fiefs whic h were connected wit h the ranks in pre-European times, and finally the offices under the present- day Native Administration whic h are linked wit h certain gitsu ranks.
DYNASTI C RANK S I N BID A 1936
- Ranks held by Members of the emi Hsu Maru (House of Umaru Majigi) Etsu, appointed 1935, former title Ndgenu.
Shaba, died December 1935; til l 1934 District Head of Sakpe. Fief i n
pre-European times: Mokwa.
‘Old ‘ Maku, til l 1934 Distric t Head of Kacha, then retired; previous title
Lukpa. Fief in pre-European times: Kacha.
Makii, younger brother of ‘Old ‘ Maku; no official position.
Ginya, present Distric t Head of Badeggi.
Cekpa
Gara
Lafyene (Lakpcne) Uban Doma
Tswayi
9 6 POLITICA L HISTOR Y O F NUP E KINGDO M
Tdka, hel d by eldest son of Shaba. Sontwdraki
Waci Taba, held by younger brother of Shaba.
Nda Iya
Ndgenu, hel d by younger brother of Etsu, appointed 1935; no previous
title ; Distric t Hea d o f Jima.
‘Old* NatsUy retired fro m office 1934.
Natsuy son of a former Kpotu, younger brother of Ginya; on the Bida
Tow n Council.
Shabarabd, a new ad hoc introduced rank, bestowed in Ma y 1936 on the
eldest son of the late Shaba, as no higher rank was vacant.
- Ranks held by Members of the emi ‘tsu Usman (House of Usman Zaki)
‘Old ‘ Kpotu, died in 1936; for some tim e Distric t Hea d of Sakpe. Fie f in pre-Europea n times: Edupka (countries beyond the Niger) .
KpotU, son and nephew of former Kpotu; Distric t Hea d of Kutigi . He wil l shortly be asked to resign, that is, to become an ‘old’ KpotU, on the grounds that his health is not very good. Th e Etsu does not contemplate promotin g hi m to the vacant post of Shaba in accord• ance wit h the general rules of succession.*
Ndgya, til l 1935 Distric t Hea d of Mokwa , now membe r of the King’s Council . He is chosen to succeed to the rank of Shaba. Fie f in pre – European times: Edukpa.
Rani, since 1935 Distric t Hea d of Mokwa .
Kusddu, younger brother of the late Etsu Bello, living in his Bida house.
Fie f i n pre-Europea n times : Kamber i country.
Wad Tszva
Shaba Gini Fogu Lafarma
Kafa
Cata
- Th e ground of ill health is a pretext rather than the true reason—and this is very significant. Th e KpotU does not belong to the house of Usman Zaki, although he is generally regarded as belonging to this dynasty. He is a descen• dant of Abdu Gboya, one of the sons of Mallam Dendo whose offspring had failed to secure succession to the throne of Nupe. He is, in fact, the only member of these ‘second-rank’ royal families who had obtained, still in the fourth genera• tion, one of the three highest dynastic positions. His premature resignation is a striking evidence of this dynastic policy to restrict succession to the throne to the three ‘houses’.
- Ranks held by Members of the emi ‘tsu Saba (House of Etsu Masabd)
(Lat e Etsu Saidu, died 1935.)
Wacimbe
Ndkorji, Distric t Hea d of Gbangba. Chosen to succeed to office of KpotU.
Fie f i n pre-European times: Yeti .
Lukpa
Tsaduya
Tsaci
Tswasha Giraba†
Tsuroma†
POLITICA L HISTORY OF NUPE KINGDO M 97
Benii, Fief in pre-European times: Ki n Agbadya (Lokoja).
Tswanku Tsoyida Tswankdku RofiS Cecekd Nkoci Tswashako Yitsa
Nddma
† Tw o established gitsu titles, Fokpo and Gbaruku (a Dibo word, meaning Old One), are not represented in this list, i.e. are not held at present. On the other hand, new titles are constantly introduced to cope with an increased demand; e.g. Tsuroma is a Bornu title (Ctroma). The rank Tswasha Giraba, lit. ‘Tswasha bestowed in Raba’, seems to have been a newly introduced, additional, rank at one time, in Raba, as the meaning of the title indicates. Th e Etsu bestowed a new title of this kind shortly before I left Bida on the younger son of the late Shaba, making him Shabami (lie. ‘Shaba of the House’).
OFFICERS OF STATE
Placed under the royal nobility in rank but its equal in political in • fluence is the title d office nobility of the kingdom. It s members are known as ticizi, the Title d Ones, or rowni, the ‘Turbans’, so called because the gift of a turban by the king is the symbol of the bestowal of rank and office. There are three groups of officers of state, each of which is thought of as a sharply defined social order or estate’—ena in Nupe, which means ‘group*, ‘society’, ‘order’: first, the sarakizi, a two • fold group of civi l and military nobility, consisting of the ena Ndeji, the Order of Tow n Elders, and the ena ‘ku, the Order of War. Second, the ena manti, the Order of Mallams, comprising the judiciary and the clergy of the kingdom, the Alkali , judge, and Liman, religious head of the Nupe Mohammedans, wit h their scribes and assistants. Th e thir d class of state officials, lowest in rank but still possessed of great influence, is the ena wuzi, the Order of (Court) Slaves; we shall deal wit h this group later, in the chapter on slavery.
We shall see that many of the titles of the saraki nobility are identical wit h the ranks bestowed on the village elders. Of some ranks we know that they have been copied comparatively recently by the villagers. Whether this adoption of state ranks by the village communities explains all the existing correspondences, or whether they are due, historically, to the evolution of the kingdom from a village community, we have no means of deciding. To-day, it is a correspondence in name only. Th e identical symbols of status in the kingdom and in the village refer to very different social contexts. Th e differences in the official duties and in the rules of appointment wil l become clear presently. Here I wil l only mention the difference in the social conceptions upon which the two rank systems rest. Both must be interpreted as systems of selection by means of whic h the society obtains its leaders. But selection in the village means offering 13 ranks to , say, 200 adult men or—expressed more correctly—13 families; in the kingdom, it means admitting to
H
98 POLITICA L HISTORY OF NUPE KINGDO M
leadership 30 men out of a population of hundreds of thousands. In the village, the position of importance and responsibility already attained by the individual who qualifies for the rank (through seniority, experience, and headship of a kinship group) is at least equal to the new benefits entailed in the appointment; in the state ranks these benefits far out• weigh any position one could have made for oneself previous to the appointment—often the latter is negligible. At the same time many of the state ranks, subject to changing political fortunes, are held more in • securely. In the village, then, rank is essentially a form of assured public recognition: in the state it represents a hotly contested and often capricious elevation to the privileged ilite.
I have called the saraki nobility an office nobility to contrast it wit h the hereditary nobility of the gitsuzi. Th e saraki ranks are, at least in theory, always open to new blood. Only one or two of the higher ranks are to-day hereditary in practice, having become the prerogative of certain families which , once established in their influential position, secured these offices for their descendants and relatives. Civi l and military nobility are mutually exclusive, and are ruled by rigi d pre• cedence and promotion. Promotions and first appointments are granted by the king in council wit h his princes and other officers of state. These ranks, unlike thegttsu ranks, cannot be accumulated in the same family; they are bestowed exclusively on one member of the family group, and no other close relative can claim a title or office in addition to the one already bestowed on his * house*. Only the less rigidly defined benefits of status and economic privileges that go wit h the elevation to the nobility extend, to a lesser or greater extent, also to the other members of the family group. Outsiders, even commoners, who have won the favour of the king may be admitted to the nobility of the sarakizi. Success in war and outstanding military achievement constituted formerly the main claim to saraki rank. Wealth and financial influence, which, “under the old system were bound up wit h big land ownership and the slave- trade—prerogatives of princes and noblemen—have come into their own in modern times and are demanding recognition. Th e granting of titles to rich traders in return for huge presents has, in fact, become a new feature in the Nupe nobility.1 Th e result is that the saraki group em• braces people of very different origin : descendants of the old office nobility of pre-Fulani Nupe as well as families of Nupe commoners who had become partisans of the Fulani conquerors, and also strangers, Hausa, Yoruba, Fulani mercenary soldiers, or rich traders who have earned the gratitude of this or that Etsu Nupe. Th e titles themselves equally reflect the situation, new titles being frequently invented or borrowed from Hausa or Bornu.2
1 Etsu Bello and, Etsu Saidu are said to have ‘sold’ certain titles for prices varying from £5 to £2 0 to ambitious, wealthy traders; they are even reported to have conferred important state offices, offices to which the candidate was entitled by birth and standing, such as the office of the Alkali, on the payment of a ‘fee’.
2 In present-day Nupe Hausa names have replaced to a large extent the
original Nupe titles. Th e traditional Nupe name, for example, for sarakizi
(which is a corruption of the Hausa sarakuna) is agabi, but this name is only pre-
POLITICA L HISTORY OF NUPE KINGDO M 99
Civi l and military officers received no salary. Th e payment for their services took the form of a share in the booty of slaves, horses, or other property, made in the war by their household troops; of landed property and fiefs bestowed by the king upon his loyal servants; and, finally, of status and exalted social position linked wit h title and office. Every pro• motion to a higher rank, as it involved a larger measure of political in • fluence, involved also claims to larger and more lucrative fiefs.
The Ndeji,’ Father of the Town’ , is the head of the civil nobility. He is in charge of the affairs of Bida, and all reports or complaints are brought to hi m first. He is closest to the king, always about his person, the most powerful of all the officers of state because he ‘has the king’s ear’. He and the other civil rank-holders are the shawracizi nya minin, the Councillors of the Inner House, so called because they hold their daily nko (council) wit h the king himself.
The ena ‘ku, a creation of the kingdom, has no parallel in the village organization. Service in the war, open to every man in Nupe, nobleman or commoner, is compulsory for this class of professional soldiers. Th e
‘war-ranks’ imply tasks for experts, and special responsibility for the execution of these tasks to king and commander of the army. Thus the different titles have specialized meanings: the Maiyaki is the General- in-Chief, the Uban Dawaki and Dokoyiringi are in command of the main body of horsemen, the Sokyara is in charge of the mounted vanguard, the Ejuko of the archers, the Sonkali of the foot-troops, &c. The ena Maiyaki, too, holds its daily nko, but not in the presence of the king. They, too, receive reports about the affairs of the town , and also of the country at large; but their main work is the discussion and preparation of military expeditions.
Only for the four or five highest ranks of the civil and military nobility is the order of precedence really rigi d and generally valid. Th e signifi• cance of the remaining ranks changes frequently; bestowed as they are by the king on individuals to whom he owes gratitude or recognition for their services, they may lose their importance under another king, who has other, new men to reward and to distinguish. On pp. 100-1 is a list of the saraki ranks which exist in Bida to-day. Where possible I have indicated previous ranks and promotions, and also the first appointment under one or other Etsu.
Th e ena manzi derives its name from the fact that the judicial and religious offices which it comprises are bestowed on Maliams, men of learning and scholarship. Profound knowledge of the Mohammedan literature, which is religious and juridical in one, is the only way to these ranks whic h are, like all other ranks in the kingdom, bestowed by the king. But the specialized training, the very necessity of finding a man of so special an education, limits the number of possible candidates. Wit h knowledge and scholarship the office is handed on from father to son, political influence and power of certain families do the rest, and the result is that the offices of the Alkali, Judge, and Liman, Head of the Mohammedan clergy, become hereditary almost more rigidly than
served in Patigi. Or again, the highest military rank to-day is called by the Hausa name Maiyaki, instead of Somasu, which is the old Nupe title.
100 POLITICA L HISTORY OF NUPE KINGDO M
the saraki ranks and offices. Thu s the present Chief Alkali of 6id a as wel l as his assistant is himself the son of an Alkali, Th e same few families supply to-day the judges for the district courts in Nupe and indeed for many courts outside Nupe country, where the reputation of Bida scholarship has found recognition. And , as a final result of this tendency to make these offices the prerogative of a small group, we find since the time of Etsu Mohamadu the tw o offices of Liman and Alkali combined in one person.1
Th e office of the Alkali in Bida, the highest Mohammedan judge of the kingdom, dates back to the first period of Fulani rule. Its impor• tance and distinction is expressed in the fact that llVIallam Dendo’s eldest son chose for himself the office not of a temporal ruler but of judge of Nupe. Man y of the men who held these offices were of great learning and outstanding ability. Some have left behind records of the history of
i. Ranks of the Ena Ndeji
Previous
Title Tribal origin ranks Biographical remarks
Ndeji Nupe from Magajin Gari Grandfather was Ndeji, appointed Dibo by Etsu Masaba; the father had no title. To-day highest official of
the N.A. ; head of the Emir’s council.
Nydfyenye Fulani Magajin Gari Father and grandfather both Ndeji; first appointment under Usman Zaki.
Magajin Gari Nupe
Sonfadako >» Gadza
Father was Nydfyenye and later
Ndeji.
Father had no title; his first
appointment was under Etsu Maliki , his second under Etsu Mamadu.
Gadza No holder at present, but f ormerly a very important title.
Sakiwa Fulani Father was Ndeji, titled under
Umaru Majigi.
MamasU Father held a lower title; he him•
self was titled under Etsu Maliki .
Etsu Basa Benu-Nupe
Etsu Gbari Nupe
A hereditary title held in one family.
Father was Uban Dawaki.
Tsadza Fulani Galadiman Father and elder brother have held
Gari same title.
NdasQ Hausa Newly introduced title.
Galadiman Fulani Father was Gadza; he himself was
Gari appointed under Etsu Bello. He became later a ‘ward-head* of the modern town-administration.
Kotonkomu Nupe Father held same title.
1 There exist two different kinds of Liman offices: the Liman Ratibi, the minister of the weekday service held in the various mosques of the town, and the Liman Jemaa, the /ninister of the Friday service, which is held in the big city mosque, in the presence of the king. In every part of the town there is a Liman of the first kind ; but there is only one Liman Jemaa. And it is to this highest priestly office that we referred above.
POLITICAL HISTORY OF NUPE KINGDOM
A. Ranks of the Ena Maiyaki
Previous
Title Tribal origin ranks Biographical remarks
Maiyaki Hausa Dokoyiringi, Father received title of Maiyaki
Uban Dawaki in the war. To-day official of the
N.A. ; on the Emir’s council.
Uban Dawaki Nupe Sokyara Father Maiyaki and Uban Dawaki, To-day on the Bida town council.
Sokyara >» Wambayi Father was Sokyara, Uban Dawaki,
and finally Maiyaki. He himself
was formerly on the Bida town council.
Dokoyiringi >» Wambayi
Ejukd >» Tswakoko
Father was Uban Dawaki and Mai•
yaki.
Father already Ejukd.
Sondzamitsu ,, Father had no title; son got his under the late Etsu because of his wealth.
Sonlawu ,, Sarkin Ba• Father had no title; son received
rayi his in the war.
Sarkin Barayi Hausa Father was Maiyaki. Tswakoko Nupe Father was Nagun.
Nagun >» Father was made Nagun in the
war.
This title is hereditary in one family, followers of the Hausa general and relative (by marriage)
Wambayi »»
of Manko, Handi Boshi.
Father was a serf (bard) of Etsu
portant title.
Luci Nupe Formerly a powerful man; lives to-day on a farm in Trans- Kaduna.
Rwafi Hausa An hereditary title, held by rela• tives of the Hausa general of Manko, Handi Boshi.
Sonfadakwa Nupe
Barize ,, Father was Sojetsu.
Sojetsu ,, Received title from Etsu Bello;
rich landowner.
Magayaki Received title in the war, under
Etsu Maliki .
Songubi >i
Sonkali >»
Manlawo >» .,
MamasU ,, Father was Sonlawu, a title which
he received in the war.
Nupe. Th e judicial organization of the kingdom left them free hand to interpret and apply, even to alter and create laws. Th e names of past Alkalai are known to everybody, and are associated with certain new
10 2 POLITICA L HISTORY OF NUPE KINGDO M
laws or special judicia l practices. Of this aspect of their activity we shall hear more at a later stage. But one point (already mentioned) must be stressed here. Th e office of Liman and Alkali could become hereditary; many judges and Mohammedan scholars are of Fulani stock; bu t this reflects only the political influence and partly also the tradition of learn• ing in these families. As representatives of Mohammedan scholarship Liman and Alkali are only the most exalted exponents of a professional class whic h comprises men of every status and birth—the Mallams,
‘learned men’, who are held in high esteem by everyone. Even withou t rank or title , when of littl e and insignificant learning, they are placed above the ordinary commoner. They may marry the daughters and sisters of the high-born. Th e Etsu himself, descendant of a Mallam , puts a Dallam ‘ before his name. Lik e his ancestors he honours and receives at his court all Mallams of reputation. They are a class between the classes, a mobile, privileged intelligentsia.
Th e description of the hierarchy of Nupe kingdom is incomplete withou t the mention of a last group of minor rank-holders, composed of the heads of the Bida craft-guilds. Th e guilds are closed occupational groups, the members of which live together in one locality and practise certain hereditary crafts such as glass-making, brasswork, weaving, &C 1
Th e guild-heads receive their titles, together wit h the usual gift of a turban, from the Etsu. Thoug h low down in the scale of ranks, they have yet an official standing, and are raised above the ordinary commoner and peasant. Thei r headship implies authority over all the members of their craft in town and village. Formerly the guild-heads received a small annual tribute of products of their craft from their fellow craftsmen. Lik e the members of the nobility , the titled guild-heads are summoned to the Etsu’s Council at the Sallahy the great Mohammedan festival; and, like the privileged nobility, they are exempted from taxation. Thei r only tribute takes the form of an annual gift—an emisa, or ‘salute’— again of products of their craft, the cost of which is borne jointl y by the whole guild , and whic h is presented to the Etsu at the Sallah.
Certain craft-guilds are linked wit h the dynastic division of the capital and are divided into three sections each of whic h resides in a different ekpa. Th e groups of craftsmen also bear more special alle• giance to the royal house in whose ekpa they live and wit h the ancestor of which they associate their coming to Bida and the foundation of their guild . But over and above this party allegiance guilds and guild-sections are closely linked wit h the ruling king-of Nupe. For the king and the court represent the most important customers of the Bida craftsmen. Al l orders which the court intends to place are given through the guild – head, who guarantees their execution and acts as an official * controller* of his group of craftsmen.
A position similar to that held by the guild-heads is also assigned to the head of the Bida brokers, the Etsu Dilali, who again plays the part of an agent of the king in all matters concerning commerce and trade in the capital.
1 The organization of the Bida craft-guilds will be described in more detail in
Chapter xiv.
POLITICA L HISTORY OF NUPE KINGDO M 103
We shall see later that the explanation of this incorporation of artisans and brokers in the rigi d bureaucracy of the kingdom lies in the control whic h the state exercises over crafts and merchandise; only by means of a control of this kin d could the head of this warlike state secure an un • impeded supply of the commodities that were vita l to the court and the state: arms, clothes, and saddles for the soldiers, food for the huge capital, and the luxuries that were needed for the display of royal power and wealth. We find convincing evidence for this interpretation in the fact that the different crafts are not of equal standing. Some rank higher than others, and this grading is not correlated wit h economic success : the very prosperous butchers are last on this list, blacksmiths and brass- smiths, on the other hand, head the list, ‘for they are makers of arms’. Onl y the glass-makers, a strongly independent and important craft, stand aloof from this order of craft-guilds. Thi s is a list of the craft-guilds in Bida, wit h the titles of the guild-heads, arranged in order of precedence.
The figure in brackets shows the number of ekpd in which they live. The titles of the guild-heads are given in italics.
Brassworkers (2): Muku.
Blacksmiths (3): Dokodza.
Carpenters and Builders (1): Egbaba. Barber-Doctors (3): Sokyara. Beadworkers (2): Map Lantana.
Leatherworkers (scattered in the town) : formerly Map Kiitipa.
Weavers (3): Leshe, Map, Ndakd,
Butchers (3): Sonfawa.
Glass-makers (1): Masagd.
SLAVERY
Th e lowest in the complex order of social classes in Nupe are the slaves (wuz’i, as against egi, freeman, lit . child). But in this one class there are shades and distinctions of social status. Some slaves have risen to positions of wealth and influence. Others have remained menial workers in the households and on the farms of their masters. Pre- Britis h Nupe numbered its slaves by the thousands. Twent y to thirt y slaves was a moderate average for the household of a man of con• sequence. Th e prosperous craftsmen of Bida town , blacksmiths and weavers, always had slaves to work for them ; even in villages one found households wit h one or two slaves.
Th e people of Nupe obtained their slaves by two methods: first, there were the slaves captured in the war, wh o became the property of the man to whose share in the booty they had fallen; and second, there were slaves bought in the markets of Bida.1 Th e majority of the latter, it seems, were slaves sold by their captors—noblemen who had taken more slaves than they needed for their household. Slaves were also sold abroad, mostly to the south. Man y Nupe men of rank were engaged in
1 Lander reports that 200 slaves were on sale in the market of Raba, the then capital of Nupe (op. cit. iii , p. 313).
104 POLITICA L HISTORY OF NUPE KINGDO M
professional slave-trading; they acted as middlemen between the slave markets of the north, Kano, Katsina, and others, and the buyers in the south, on the river and coast.1
At the time of the Fulani conquest the price paid locally for an able- bodied male slave was 40,000 cowries, i.e. £1 , and for a female slave
50,000 cowries, i.e. £1 . 5s.2 Th e price of slaves rose very high in later times. Shortly before the British occupation it had reached the sum of
£7 for male, and £7.10s. for female slaves. As my inquiries from various informants showed, this price had become almost standardized in the last decade before the British occupation and was valid throughout the country. It is interesting to note that this rise in price took place at a time when the largest market for slaves from Nupe was disappearing, shipments of slaves to countries overseas, especially to the Americas, having come to an end. We cannot say how this rise in price started, nor what might have been its cause. I can only suggest that it was due to increased difficulties of supply. Half a century of wars and raids had denuded the country of people who could with rio great difficulty be made slaves. Slave-raiders were forced to obtain their human chattels in more distant areas and by more serious warfare—in the language of the economist, under conditions of higher ‘cost’. Whatever the cause, this rise in price meant that the commoners and peasants of Nupe could derive from the wars and raids by which their rulers gained their for• tunes not even the benefit of cheaper slave labour. Commoners and peasants would suffer when their country was defeated and their villages invaded by the enemy, or when, in civil wars, rival troops fought on their land. But the fruits of victory—rich booty in slaves—they could share only at a price: the same price which foreigners had to pay when buying slaves from Nupe in the open market.
Slaves were, in the strict sense of the word, the property of their masters. If anyone killed a slave belonging to somebody else this was not regarded as a case of murder or manslaughter, but only as an offence demanding restitution of the ‘destroyed* property, i.e. the full purchase price had to be paid over to the owner. But the slaves who were kept in the houses of the Nupe nobility were on the whole well treated. Th e master of a slave was quite within his rights if he chose to maltreat him, sell him again, or even kill him. But there is no doubt that this happened rarely—if ever. Many slaves, in fact, preferred to stay on in the houses of their ex-masters when slavery was abolished, in the position of an unpaid servant or retainer, rather than to start life on their own, especially if they were of advanced age. Custom even guaranteed the
1 Bart h states that most of the slaves fro m Kan o were carried by small caravans to Born u and Nupe ; ‘the best of slaves now go to Niffee (Nupe) to be there shipped for America. .. . Slaves are sent fro m Zinde r to Niffee. Indeed it no w appears that al l this part of Africa is pu t under contributio n to supply the South America n market wit h slaves’ (Dr . Barth, op. cit . ii , pp . 132 and 135). Lander mentions that the slaves brought to Raba were purchased by southern peoples, traded dow n the Niger , and fro m there ‘delivered fro m hand to hand til l they at length reached the sea’ (op. cit . ii , p . 298).
2 See Schoen-Crowther, Journals, p. 194, and J. and R. Lander, op. cit . ii , p.
298.
POLITICA L HISTORY OF NUPE KINGDO M 105 slaves a not inconsiderable measure of economic and social indepen•
dence. An adult male slave would work for his master in the same way as
sons and junior relations work for the head of the household. If he was
put to farm-work, he would be given a small plot of his own to cultivate
in his spare time. Or if he helped his master in a craft or in trade he
would be able to count on a regular commission. Other slaves, who made
good soldiers, were put into their master’s private army, in which they
would serve as foot-soldiers, armed and equipped at their master’s
expense. In the wars they would have their share of the booty. Slaves
could thus acquire property, and the more fortunate owned horses,
sheep, cattle, money—even slaves of their own. But when they died their property fell not to their wives and children but to their master.1
Female slaves became personal servants of their mistresses or possibly concubines of their master. Unlike the male slaves they had no income of their own; what they might earn—by handicrafts or trade—would be their master’s property.
Frequently the master would have to find wives for his male slaves; he would buy either a slave girl from the market, or possibly a girl belong• ing to another master—a slave girl to whom his own slave may have lost his heart. In the latter case a very reduced price would be paid, but the children would belong to the owner of the girl. But a slave might make money enough during his service to pay the bride-price for a free woman
—mostly a widow, for whom a low bride-price is paid. In such marriages the free wife would in no way share the conditions of slavery, she would have her own house and work (a farm or trade) and might in due course be able to ransom her husband. Or a slave who had done well might be able to buy his own and his children’s liberty. In this case he had to pay the full purchase price, except in the case of very old slaves who were no longer good for work—they would be released for half the amount. But if the offer to ransom a slave came from the slave’s (free) relations, the owner would ask an exorbitant price, up to £20 or £30. Very rarely would a master let a slave free without the payment of the ransom, for example, as reward for faithful service. In all these cases the slave would bring two men as witnesses of the act of manumission.
A slave girl often became the wife of a free man—her own master or another man who bought her from her master, the purchase price re• placing in this case the usual bride-price. This was not, however, real marriage; the slave girls became only concubines, not wives, and no religious or legal ceremony accompanied the marriage. Once they had borne their ‘husband’ a child, both mother and child became automati• cally free (though the status of a concubine remained unchanged). Th e families of the Fulani nobility count an innumerable offspring from this type of concubinage.
Children born of a free father and a slave mother were thus free
1 To-day, more than thirty years after the abolition of slavery, this rule still holds in a modified form, with respect to ex-slaves who die without issue. In a court case heard before the Chief Alkali in Bida in 1936 the considerable property (cattle) of an ex-slave who had died without heirs was thus awarded to the son of his ex-master.
106 POLITICA L HISTORY OF NUPE KINGDO M
themselves; so were children born of a slave father who married a free woman, and also of a slave father who bought a gir l slave for himself and married her. If , however, slave married slave, the marriage being arranged—i.e. the purchase price for the gir l paid—by the slave’s master, the children remained slaves. But these slaves in the second generation were slaves of a different kind . The y were no longer called wuzi, but mangi, ‘bor n as child’ (i.e. as freeman). Thoug h not really free they were exempted fro m many of the disabilities of slavery. Th e children of a mangi who married another mangi were themselves mangi slaves again. But a mangi could not be sold by his master, nor could he be maltreated or killed—i n the latter case his parents could obtain legal redress and woul d be paid the usual purchase price as ‘blood-money’ At the death of a mangi only half of his property could be claimed by the master, the other half was inherited by his own family. Th e mangi also had family farms of their own , and were generally treated like ful l members of the household in everything except that they were not entitled to any share in the inheritance of their master’s property. A mangi was brought up together wit h the children of the house, was given the same education, and, like the sons of the house, served in the mounted household troops. Man y mangi would free themselves, or at least marry free women. But the names of these house-born slaves in • variably betrayed their origin ; for they were named by their masters, and given not the usual Nupe names but names made up of some well-known blessing formula, of the first words of a proverb, and such like, for example: Adetsua, from a de ‘tsu a set Soko, ‘there is no Lor d but God’ ; Barsu, from bar su ga Allah (Hausa), ‘leave them
t o God’ .
It woul d be wrong, however, to assume that slavery in Nupe carried wit h it a stigma. True , the people of Bida stil l speak wit h contempt of the ‘pagans’ and ‘slaves’ (the tw o terms are almost synonymous) into whose countries their fathers had carried their profitable raids. But , characteristically for Nupe, the stigma attaches to the humble status rather than to the misfortune of having been made a slave. It dis• appeared wit h the signs of success and prosperity, open to slaves as to anyone. Th e wretched slave who could never rise above a menial labourer had to bear the ignominy of his position. But in the case of the trusted slave in a noble household, and above all the successful, freed slave, the taint of slavery meant littl e more than, in their own words,
‘luck of war’. Slaves who served in the houses of the high nobilit y di d indeed share some of the splendour that surrounded the position of their masters. Abl e and faithful slaves were rewarded wit h the position of head slaves, overseers, bailiffs; their masters bestowed upon them ric h presents and gifts of land—more land than they cared to work wit h their own hands, having become men of wealth and importance. Thei r masters bestowed upon them, above all , special titles, household ranks— the ubiquitous symbol of social rise in Nupe, whic h in this case obliter• ated even the last external symbol of slave status, the typical slave- names.
Th e royal houses of Nupe, like every nobleman’s house, had their host
POLITICA L HISTORY OF NUPE KINGDO M 107 of slaves: slaves for menial work—i n the house and on the land—as well
as the title d slaves of respected position. At the court this latter group of
slaves formed the ena wuzi, the Order of Court Slaves, a distinguished
slave ilite into whic h only few selected slaves could ever hope to be
admitted. Fro m the slaves captured in war those of high birt h and
status, chiefs or distinguished warriors, were selected for the court.
These, too, were given ranks and titles, and could rise high in the service
of the king . Theirs was the specialized service of officials and trusted
royal agents. There were those who looked after the king’s vast house•
hold ; those who were in charge of the Etsu Nupe’s no less imposing harem—both offices formerly entrusted to eunuchs.1 Other slaves formed the king’s bodyguard and police force; they were provincial delegates, tax collectors, trusted messengers, 4liaison officers’ between the king’s palace and the houses of his noblemen, and between the king and the army in the field. These slaves, like the other rank-holders, owned land and held fiefs. Th e Nupe kings in Rabd are said to have had only four title d court slaves; the slave ranks that existed at the court of Bida some thirt y years ago (and the ranks stil l exist, although their holders are no longer slaves) were more than three times that number. Under Etsu Maliki , one of the last independent kings of Nupe, the most responsible state offices were taken from royal princes and sarakizi and
entrusted to slaves. Th e position of the slaves as officers of state thus seems to have overshadowed more and more all other state ranks. Slave ranks were ranks bestowed ad personam, and ranks bestowed by the kin g personally, independent of the council of princes and ministers; and this gives us the explanation for their rise in importance. Th e offices linked wit h civi l and militar y rank also depended to some extent on the per• sonal favour of the king. But once appointed, these men stayed on, servants of the kingdom rather than of the king. The y had behind them the claims of powerful families, they were exponents of political factions, and could play on that ‘balance of power’ between the dynasties. Th e slaves were personal servants i n the ful l sense of the word. The y were brought into office, not by family claims or promotion or a rigi d rule of succession, but by personal appointment, selection, and trial . The y were bound to the king’s service generation after generation, by the strongest bond : their position of unfree men. It frequently happened that a new Etsu took over the slaves who had served his predecessor and left them their ranks and possessions. Th e rise of the slave officials at the Fulani court is the counterpart of the rise of a powerful, dangerously independent feudal nobility.
Th e following is a list of slave ranks held to-day, not by slaves, but by former slaves or descendants of slaves at the court of the Etsu Nupe. I have again trie d to indicate in what manner they received their title , and also in certain cases the office whic h some of them are holding to-day in the Bida Tow n Administration. Tha t so many of these former slaves
1 In the pre-Fulani kingdom the court is said to have employed only a few slaves—all eunuchs. Al l slave ranks which have a nda in their title, meaning
‘father’ (ranks that are still in use—see attached chart), were originally ranks reserved for* eunuchs.
108 POLITICA L HISTORY OF NUPE KINGDO M
had been placed in responsible positions proves what I have said about the selection and rise of slave officials at the court of Bida.
Slave Ranks
Present
Title Tribal origin office Biographical remarks
Gobi Seidi Kupa ‘Officer of A rich slave who under Etsu Bello
Liaison’ be• tween Etsu and District Officer
bought his title for £16.
Ndatvjaki Yagba On the Bida Father was NdatwaH under Tow n Etsu Maliki ; he himself received Council the same title under Etsu Bello.
Dzwodfu »> ,, ,, An hereditary slave title, first granted under Etsu Usman.
Ndamdraki Bunu ,, „ Father was Ndamdraki under
Etsu Masaba.
Sdntali Benu-Nupe Grandfather was a slave of Usman
Zaki ; they remained slaves of this
house.
Masdntali Yagba Grandfather was a slave of Etsu
Umaru Majigi ; they remained slaves of this house.
Swajiyd’tsu ,, . , Father held same slave rank.
Ndatddo »» Father held the same title and also title of Ndamdraki.
Manfdda Another slave title which has be•
come hereditary in one house.
SonmdH Kusopa- Nupe
Id .
Ndddari Nupe Id .
Shangbo Nupe from Father was a saraki (Somasti) of
Raba Etsu Majiya. He was made prisoner in the war and became a slave of Etsu Masaba.
(Sarkin Dogari Hausa Head of To-day not regarded as slave title.
police force Formerly bestowed on a personal
slave of the ruling king.)
WAR
It will be understood that this description is based entirely on informa• tion and the historical recollections of informants. However, the wars that were fought in Nupe and by the Nupe are not so far distant that there are no longer eyewitnesses left, or people who have themselves borne arms in these wars. Thus information could be checked and cross-checked. For certain details I could, moreover, count on most authoritative informants, who had played a leading part in the last wars of the kingdom; the description of the military organization, for example, is based on information obtained from the late commander of the ilite troops of Nupe, the king’s bodyguard.
War was essentially a concern of the Nupe state. No warfare or organized fights occurred between the villages or tribal sections of the
POLITICA L HISTORY OF NUPE KINGDO M 109 country.1 But villages and tribal sections were frequently involved in the wars of the kingdom, either themselves causing wars by rebelling
against the central government, or being forced to take sides in the wars
which the kingdom waged against other groups. Thus the Beni and
Gbedegi sided with Bida against the rebellious Kusopa, and again in the
Kyedye war, while other tribal sections, such as the riverain Bataci,
Kupa, and Ebagi, came to the aid of the Kyedye. In the later times of the
kingdom, to which my records chiefly refer, the wars were indeed largely
caused by internal disturbances and were directed against rebellious
provinces or feudal lords—being reconquests rather than conquests.
Historical records mention ten major wars in the thirty years that preceded
the British conquest, between 1864, and 1895. Most of these wars are
still vividly remembered, and old men would even recall characteristic
incidents in the various campaigns. My informants, not unnaturally,
greatly extolled the soldierly virtues of the Nupe people. But there is
also Barth’s testimony that the Nupe army, especially the cavalry, was well known in the Hausa states.2 Twice, in fact, (in 1880) the Nupe were called to Hausa to help their allies, the Emirs of Kontagora and Gwandu, to put down rebellions in their countries.
The Nupe army was comparatively loosely organized. It possessed only a small nucleus of regular troops, represented in the king’s body• guard; the rest of the army consisted of levies of slaves, volunteers, and mercenary soldiers, raised by the houses of the feudal nobility. But these household troops, too, were largely composed of trained and ex• perienced soldiers.
The king’s bodyguard comprised both infantry and cavalry. Th e foot- soldiers, called dogari nya dakdri (i.e. ‘guards on foot’), were equipped with guns or bows and arrows, and were under the command of the Etsu Dogari. The troops who carried guns were mostly slaves, the sons of slaves from the king’s household, the others were free men, mercenary soldiers—among them.many foreigners, from Hausa, Bornu, and Yoruba
—who had taken service under the Etsu Nupe. Under Etsu Maliki this royal bodyguard numbered 80 men. Th e royal horse troops, a slightly larger group, were called lefidi, after the cotton-wool armour which they wore. Their weapons were swords, spears, and big leather shields. They were commanded by the Etsu Lefidi (who was also called Zarmi). One of my informants for the present description was the Etsu Lefidi of the royal house of Usman Zaki. He was of Hausa origin, a freed slave and the son of an Etsu Lefidi who had served under Usman Zaki himself. The horse troops were composed again of slaves and mercenary soldiers. The slaves who served in the royal cavalry were the sons of titled slaves, every court slave of rank having to send one of his sons into the Etsu’s bodyguard, and the volunteers were all foreigners. In the reign of Etsu Maliki the lefidi numbered 100 men.
The troops which were contributed by the various houses of the
1 Th e only exception was the warlike Kyedye sub-tribe.
2 Barth, op. cit. iv, p. 545. Lander reports that Malla m Dendo could ‘command
1,000 horsemeri, well equipped, and an uncountable number of foot-soldiers’ (op. cit. ii , p. 296).
110 POLITICA L HISTORY OF NUPE KINGDO M
nobility were of three categories: ordinary foot-soldiers, called dakdri, foot-troops carrying guns, or bindigacizi, and horsemen, or dokocizi. Th e dakari were composed entirely of volunteers, i.e. men who had entered the service of a feudal lor d as soldier-tara or ‘clients’.1 The y were all young men, and their ranks comprised peasants and townsmen, Nupe as well as foreigners. Every noble household had its own Etsu Dakari, Chief of the Foot-Troops, a bar a of higher rank, who was in charge of these volunteers and responsible for their training. Th e dakdri wore a sort of leather apron, called walke, and caps, and were equipped wit h shields and bows and arrows or spears. Th e troops which carried guns were, like the king’s bodyguard, slaves. Th e cavalry represented the elite of the Nupe army. It was composed of senior bar a—men whose sons migh t be fighting in the foot-troops—and younger relations, sons and brothers, of the feudal lor d himself. Th e horsemen wore short shirts of strong coarse material, called ewogi, i.e. small gown, into whic h leather amulets were sewn, caps or turbans, and, over the ezvogi, a sulke, a chain mail. The y carried shields, swords, and spears. In the Kyedye war certain feudal lords could muster the following troops: the Kusodu io o foot-soldiers, 200 guns, and 200 horses; the Kpotu 100 foot-soldiers,
100 guns, and 20 horses; the Ndeji 100 foot-soldiers, 200 guns, and 10 horses; theBenu 50 foot-soldiers, 50 guns, and 20 horses; the Waziri 50 foot-soldiers and 5 horses. Man y of the feudal lords—princes and civi l nobility—woul d themselves joi n the army at the head of their household troops; but they woul d have to place themselves under the command of the ‘professional soldiers’—the militar y nobility—who would joi n the troops in a body.
Warfare in Nupe possessed its etiquette, its conventions, and rules, whic h show how highly developed and specialized a technique it had become. Tw o kinds of war are distinguished: the great wars planned and organized by the central government, and smaller raids and military expeditions carried out by individual feudal lords. In the latter case, the feudal lor d would request the permission of the Etsu for undertaking a campaign against this or that village or tribe. Withou t the sanction of the king his military expedition woul d amount to cin amana, a breach of the peace (lit. : devouring friendship), and might itself become the cause of a punitive expedition sent out by the central government. In the ‘real’ war, before hostilities are opened, a messenger of the king, called kuru, is sent to the enemy’s camp to deliver the ‘ultimatum’ , that is, to inform the enemy of certain demands of the kingdom the rejection of whic h woul d be followed by war. A titled royal slave of high rank acts as messenger. Hi s person is inviolable—in theory; bu t during the Kusopa revolt three royal messengers were killed , one after the other, by the rebellious tribesmen. Th e next step is the appointment of a commander-in-chief of the armies, in the person of a royal prince who wil l then be known as Nna ‘Ku, Mothe r of War.2 Sometimes soothsayers and Mallams are
1 For the definition of bora and ‘client’ see p. 123.
2 Owing to erroneous information I have mentioned this title as Nda ‘Ku, Father of War, in a previous publication (Africa, 1935). I must add that I was unable to obtain any explanation of this rather surprising matrimonial touch.
POLITICA L HISTORY OF NUPE KINGDO M 111 asked to divine a victorious general. Invariably his rank woul d corre• spond to the importance and scope of the war that is planned. At the same time his office is only nominal. Although he woul d take the field at
the head of the army, and although he would bear the responsibility for success or failure, the actual organization of the campaign is left to the professional soldiers: in an important war the Maiyaki himself woul d take command, in a minor war his next in rank, the Dokoyiringi.
No w the council of war meets in the house of the Maiyaki to discuss the question of raising the levies. Th e various feudal lords are informed that they are expected to produce so many guns and horses. Preparations take one or two months—the troops must be raised, recruits trained and equipped, and food supplies provided for the campaign. Finally diviners and Mallams are summoned again by the commander of the army to decide on a propitious day. On the appointed day, drummers and trumpeters sound the assembly, and the troops leave Bida on their march into enemy country.
Th e Nupe never fought by night. Thei r method was to move up as close as possible to the enemy, hiding in bush and scrub, camp for the night, and attack in the early hours of the morning. In open battle the attacking troops woul d be arrayed in a long straight line facing the enemy. Bu t often the war woul d tur n into the long siege of a walled village or town , close to which the army would put up its war camp, the troops surrounding the town and cutting the population off from their farms, supplies, and reinforcements.
Thi s was the typical formation of the Nupe army in open battle: first came the foot-soldiers, the various household troops each under their Etsu Dakari, commanded by an officer of the king, the Sonkali. Behind them, at a distance of a quarter of a mile, a first troop of horsemen halted, under the Sokyara, and wit h it the lesser militar y and civi l nobility. Behind this group again another, larger, detachment of cavalry was posted, under the highest generals of the army, the Uban Dawaki and Dokoyiringi. Wit h these troops were the higher civi l and militar y nobility and certain princes of lesser rank. Th e infantry woul d move first, sent into the battle by the shouts of the Sokyara: ‘Kay Dakari’, ‘Forward ! foot-soldiers’. When they had engaged the enemy, the cavalry of the Sokyara would charge after them, the Dokoyiringi giving the signal wit h shouts of ‘Kay Sokyara\ Th e Dokoyiringi’s own troops would charge last, to the signals of the long bronze trumpets whic h only the troops of the highest militar y ranks may carry. There would also be drums and flutes, and the soldiers woul d shout the kirari of their leaders and sing their war songs.
In the rear of the troops the reserve was placed, which woul d come into the fight only if the other troops were unable to master the situation. It was commanded by the Maiyaki’, wit h whom were also the princes or highest rank, including the commander-in-chief. In the last bi g war, against the Yagba (the same war whic h ended in the conquest of Bida by the British), the infantry is said to have numbered 6,000, the cavalry of the Sokyara 400, those of the Dokoyiringi 1,000, and the reserve 2,000 men.
112 POLITICAL HISTORY OF NUPE KINGDO M
An unsuccessful army would never retreat, or return to Bida with the goal unachieved. One would call for reinforcements from the capital, which would be sent out together with a new Nna ‘Ku to replace the unsuccessful commander-in-chief. If these troops too were unable to bring the war to an end, the Etsu would take command personally, at the head of a third army composed largely of his own bodyguard and of the household troops of noblemen who had not joined the troops previously. During the king’s absence from the capital a regency council, headed by the Ndeji, would be appointed to look after things at home. In the Yagba war the Etsu did not join the army but stayed behind, with his body• guard of 200 men, which, together with the household troops of four noblemen who had remained in Bida, were the only trained soldiers to oppose the British attack. The king and the noblemen of high rank took the field accompanied by their whole household staff of slaves and servants, and often by their wives as well. They stayed in the Great Camp—consisting of huts of grass and matting—which was put up behind the fighting line or established in a hamlet near-by. A number of Nupe villages betray by their names that at one time the Great War Camp, edoko, had stood on their site.
Th e Nupe war was carried on till the capital of the enemy, that is, the village or walled town in which the chief resided, was captured. If the chief and the other leaders fled, they would be pursued until they were captured or until further pursuit became impossible. Often a fugitive chief sought refuge with the king of a neighbouring country whose territory the Nupe troops would have to respect. Al l captured places were looted and burnt, and the whole enemy country devastated. But slaughter was avoided as far as possible—the defeated enemies were more valuable alive, to be sold later as slaves. Only when rebellions had to be put down by force of war the enemies, above all chiefs and leaders, would be killed mercilessly, either in the battle, or later in Bida, whither they were taken to be executed.
No less well defined than the etiquette of warfare were the rules and conventions governing the economic aspect of the Nupe wars—the division of loot and booty. All booty taken before the decisive battle— the capture of the enemy’s capital—belonged to the king and was dis• patched to him immediately. Upon the capture of the town wasoso was declared, free looting, in which everyone who fought in the battle could join. Th e shares in the spoils of war varied according to rank and position. The slave soldiers had to deliver everything to their masters but were rewarded with a small share in the booty they had taken. The bara soldiers could keep all booty in kind—(called dukiya, wealth)—but were expected to present some of it to their overlords as ‘gift of thanks’; of the slaves whom they captured they had to surrender half the number to their overlords. Th e independent fighters, i.e. the sons of noble houses, kept whatever they captured, both ‘wealth’ and slaves, except for a voluntary gift to the head of the house. If I have at various occasions in this book called war in Nupe a profession and a source of income, these facts substantiate that view.
Let me conclude this description with the report of an eyewitness of
POLITICA L HISTORY OF NUPE KINGDO M 113 standing—the youngest son of Etsu Masaba—who had fought in one of
the last great wars of old Nupe, against the Kyedye. I have referred before
to the origin of this. war. It was the answer of the central government to
the capture by the rebellious Kyedye of the town Kacha, which, having
submitted to Fulani rule, appealed to Bida for protection against their
unruly fellow tribesmen. It was, incidentally, also the last war in which
the Niger Company assisted the ‘Bida government. In the next war
Niger Company troops were to turn against the Fulani, and (helped
by the same Kyedye) to defeat the armies of the kingdom.
Since the time of Manko, goy and Kyedye’ were always at war. It happen• ed like this. The Kyedye refused to recognize the Fulani kings. They con• tended that they were independent ‘kings of the water*. So Manko sent his son Ibrahima from Raba to the Kuta, to come to an arrangement and to stop hostilities. But the Kyedye killed the messenger, cut his body to pieces, and threw it to the fish. Thereupon Manko went to fight them, but could not do much. So the wars’ between the two went on and on.
Later, at the time of Etsu Umaru, the Kyedye attacked Kacha under their Kuta Usman, because Kacha had submitted to the Fulani and accepted their egba (delegate). The Kacha people fled to Bida and asked for help. Etsu Umaru sent out a big army led by Shaba Mamudu. Most of the gitsuZi went with him, and I was among them, also many sarakizi both from the ena Maiyaki and ena Ndeji. We were three days on the way, stopping twice for the night (Kacha is about 40 miles from Bida). We made camps of grass huts; some of us lived in a tunga near-by. For the great men we built huts with grass fences round. The army had guns and men and horses—you could not count them. In Kacha the water was covered with Kyedye canoes. We attacked at once, shooting from the banks down upon the canoes in the water. The boats of the Niger Company were helping the Fulani then. They closed the Kacha river in the back of the Kyedye. The Kyedye fled, and we crossed the river in canoes which they had left behind. The Dokoyiringi led the pursuit, accompanied by Shaba Mamudu. The rest of the troops followed later. We pursued the Kyedye on the other bank, both on land and water, for three days, as far as Lafiagi and Shonga. This was a war not only against the Kyedye, but also against Kupa and Ebagi, who were helping the Kyedye. AJ1 villages and hamlets were burnt down, numberless men and women captured. We made more than 2,000 slaves. Later they were all sold (although the Niger Company had asked for the release of the prisoners of war). At Shonga the war ended. Many Kyedye had been run down and captured or killed, but some hid in the creeks and the river forest. The chief and the seven elders of the Kyedye fled with their wives and children to Ilorin. It did not help them. Emir Aliyu (the Fulani Emir of Ilorin) sent them back as prisoners to Bida. Here they were held as hostages till Etsu Umaru’s death. But shortly after his death Etsu Maliki condemned them to death, and they were strangled secretly at night in the houses where they lived.
In war, which is political action par excellence, the essence of state structure must reveal itself most clearly. We observe what we have recognized as an element of state organization: the monopoly position of the ruling group with regard to the employment of the political machin• ery, in this pase military action. We see, too, that this monopoly of control is vested in the ruling group corporate^: its individual members,
I
114 POLITICA L HISTORY OF NUPE KINGDO M
though they control severally the executive side of the political machinery, can only use it subject to the agreement of the administration as a whole (of whic h they for m part), else the militar y machinery is directed against them as breakers of the ‘king’s peace*. Finally, wars are planned and the militar y machinery is set in motio n by the rulin g group as a representa• tive group, that is, on behalf of the country at large. In wars against out• side enemies the interests of the country and the people are perforce identified wit h the interests of the rulin g group—expansion, conquest, or the destruction of a riva l power. But in wars waged against internal enemies—rebellious sections or dangerously independent members of the government—the militar y action is taken as muc h in the interest of the rulin g group (safeguarding, above all , the revenue) as in the interest of the state and the people in that it re-establishes order and public security. Thu s in the Kacha war, referred to above, the rulin g group, as the representative group of the state, protected one subject group against abuses by another. Th e government can take sides and place the political machinery at the support of one section against another only because it is vested in a group (as I pu t it in a previous section) raised above the heterogeneous component sections of the state. Th e monopoly control over the machinery or force thus reveals its twofold significance: as a privilege or prerogative, in the sense that it can be used (i n war proper) for conquests and expansion, thus serving the specific interests of the rulin g group; and as a constitutional duty , in the sense of being used internally, for the purpose of maintaining the ‘king’s peace\ These two functions are not, however, securely balanced, and we shall discover the effects of this unstable equilibriu m in the internal administration of the country, to the description of whic h we must now
turn .
CHAPTER VII I
POLITICAL HISTORY OF NUPE KINGDO M
(continued)
TH E FEUDAL STATE
E have seen how in the history of the kingdom a certain territorial nucleus evolved, an area of ‘Nupe proper’. Thi s concept of a territorial Nucleus’ also forms the basis of the division of the country for the purpose of government. For the purposes of political, judicial , and tax administration the kingdom was divided into four zones. First , the area of the conquered tributary countries outside the boundaries of the
‘kingdom proper’, in Nupe terminology, the countries of the zazi kati, the ‘Outside Peoples’. Second, withi n these boundaries, an innermost core, the countries of Beni and Kyedye; having formed the nucleus of the rising kingdom they are now assigned the privileged position of royal domains. Th e thir d zone comprised the wider ‘ring* of remain• in g towns and districts of the kingdom, most of whic h lie in Trans- Kaduna. Bida town , the seat of the central government, formed an administrative district of its own.
Under the name of Outside Peoples were comprised the conquered territories of Gbari , Gbira, Kamberi , Kakanda, Yagba, Bunu , and certain sections of Nupe (Dib o and Ebe mostly). These districts were not administered fully by the central government. Thei r political dependence entailed two things: an annual tribute to Bida, and the recognition by other Emirates and chieftainships of the political status of these territories as ‘under the protection’ of Bida. It was a one-sided
‘protection’: Bida itself was free to raid her own outside peoples, but no other country could do this or levy taxes in these territories wit h im • punity . Th e guarantor of these ‘treaties’ was, in the last instance, the Emi r of Gwandu, the highest feudal overlord of all Fulani Emirs in this part of the world. 1 Th e central government in Bida di d not assume jurisdiction over these countries, nor di d it interfere in their own political administration. TJbe tribute was paid in slaves and was collected annually by royal messengers, tuci, or as they were also called ajele (from the Arabic ajala, to dispatch), who visited these countries at the head of small armies, ready to quell any disturbance or to brin g recalcitrant tax• payers to heel. Of the tribute levied a littl e more than half went to the king , the rest to the royal messenger. Th e office of tax-collector in the outside areas of the kingdom was entrusted mostly to the royal nobility , and the districts over whic h they were put in charge were linked wit h
1 Such an ‘international’ agreement existed, for example, between the Etsu Nupe, the Sarkin Sudan (the Emir of Kontagora), and the Emir of Zaria with respect to Gbari country. These treaties were not always kept by the parties concerned. Thus the Etsu of Agaie used to raid the territories of his brother Emir, the Etsu Nupe, and had eventually to be stopped by a big war which was led by the Emir of Gwandu himself.
116 POLITICA L HISTORY OF NUPE KINGDO M
certain ranks. Thus Edukpa, the Countries beyond the Niger, were always the province of the Nagya, the country of the Kamberi of the Kusoduy of the Gbari of the Etsu Gbari, of the Yagba of the Maku, and so on. Political influence, however, could alter the rule, and a rich dis• trict would sometimes go to the more powerful royal prince irrespective of his rank. Thus Etsu Maliki, to win over the powerful Kpotu Abuba- kari (who became Etsu after Maliki’s death) transferred the rich fief of Edukpa from the Nagya to him. Power and the possession of a strong private army was essential for the post of an ajele. An ajele who failed in his task was dismissed from rank and office and replaced by a ‘stronger* man.
It is no accident that the two royal domains of Nupe kingdom were the two political organizations which embraced the whole tribe, the two small ‘states within the State’, Beni and Kyedye. The political unifica• tion which they had evolved was recognized and utilized by the political administration of the kingdom. These areas were placed directly under the king. The chiefs of Beni and Kyedye acted somewhat as royal agents. Their appointment and succession was subject to the confirma• tion of the king, and a gift of sandals, a gown, and a turban which he presented to the new iitsu symbolized, as we have seen, the fact that the local chief held his office ‘by the grace of the king’. Th e town-kings of Beni and the Kuta of the Kyedye administered law in all cases which came under the category of gyara, ‘reparation’, and were required to refer graver cases, cases involving sheri’a, ‘judgement’, to the central authority—a point to which we shall return later. Th e town-kings also collected the tax in their areas on behalf of the king. This tax, edu, levied in these areas as in the whole kingdom, was in money, thus differ• ing from the dzanka, the local tithe in kind. Th e tax was collected first by the head chief of the district, i.e. the Kuta among the Kyedye, and the town-king of Tafie and (in Fulani times) Bida among the Beni, who placed the money into the hands of the royal tax-collectors appointed for these two royal domains. Of the tax levied in the domains the Etsu received two-thirds, the royal tax-collectors and the head chiefs of the two districts sharing between them the remaining third. In the time of Etsu Masaba, however, the Etsu took all and the town-king of Bida, in order to obtain his share, levied an additional tax of his own. The total tax paid by a Beni village in Fulani times is said to have amounted to
£2 under Etsu Masaba and £3 under Umaru Majigi ; later, under the greedy Maliki, it rose higher and higher till u de iyaka a, ‘there was no limit’.1
The rest of Nupe kingdom was divided up into smaller and larger
1 Th e Borord Fulani who came regularly to Nupe with their cattle formed another such royal domain with respect to taxation. Th e organization of this nomadic group evidently lent itself to no other method of administration. Thei r tax was collected by their own group head, the Diko, and then handed over to the royal tax-collector, who was known as egba Bororo (Delegate to the Boror<5). He was himself a Fulani, who lived—and is still living—in Bida. When the Etsu Nupe recently wanted to buy cattle for his new experimental farms he arranged the deal through the man who used to be egba Bororo, making use of the old connexion for the new purpose.
POLITICA L HISTORY OF NUPE KINGDO M 117
‘countries’ comprising each a town with its dependent villages and tunga which were administered as fiefs through feudal lords or egba. The egba—a member of the royal house or of the office nobility, or a court slave—received the fief from the king together with his rank with which the fief remained permanently linked. Mokwa was known as the town of the Shaba, Egg3 as the town of the Ndeji, Kutigi and Enagi be• longed to the Ndatwaki, and so on. Promotion to a higher rank brought with it appointment to another, more lucrative, fief. Th e feudal lord himself lived in the capital and rarely visited his possession. Th e local affairs, the collection of tax, were in charge of his representative, one of his slaves or serfs who was residing on his lands as egbagi, Small Dele• gate. Of the tax collected in these fiefs, again a money tax, a quarter went to the king and three-quarters were retained by the egba, who returned a small share to his egbagi, salary and commission in one. Sometimes the towns or districts also paid an additional tribute in kind both to Etsu and egba. I have been able to collect some data about the tax levied in these areas. The uniform information which I obtained from different sources leads me to believe that the facts are substantially correct. Th e town Kutigi, according to my informants, paid under Etsu Umaru a money tax of 20,000 cowries (about 105.) to the Etsu through the egba; to the egba himself 10,000 cowries (5s.) and to the egbagi 5,000 cowries (2$. (yd.). In addition to this, the people of Kutigi sent the Etsu 20 mats of the kind which has made Kutigi the centre of a famous mat industry, worth over £1, 4 bundles of corn worth about 7s. ‘for the king’s horses’, and 10 mats and 1 bundle of corn to the egba. In modern money, the total tax of the town which at that time had eight efu and a population of perhaps 500, amounted to between £2 and £3.1
Th e feudal lord in Bida was responsible to the king for law and order in his district. The appointment of the village chief was, again, subject to royal confirmation. The overlordship of the king also implied the right of appeal of the villagers to the king against oppression or injustice from their feudal lord.2 Lastly, the egbagiy together with the village chief, administered justice in the smaller cases and sent all bigger cases to Bida, where the egba himself would plead the case of his tenants before the court of Etsu and Alkali.
A variety of the fiefs just discussed were bestowed on individuals, relatives, courtiers, or favourites of the king, as personal fiefs, indepen• dent of rank and office. Even women could hold fiefs of this kind. These fiefs were sometimes hereditary and sometimes reclaimed by the king after the death of the holder. They were very small, single towns or villages, but one person could hold two or more such fiefs—which, how• ever, were never adjacent lest the power in the hands of one person
1 Oldfield reports that in 1832-4 (i.e. under Etsu Usman) Egga, one of the largest towns of Nupe at that time, paid an annual tax of 120,000 cowries, that is, about £2 (Laird and Oldfield, op. cit., ii , p. 39). This fits in very well with the data I obtained.
2 Th e people of Mokwa once availed themselves of this right to lodge a com•
plaint against their feudal lord. When I asked my informants what the result had been, they shrugged their shoulders and said: ‘The Etsu promised to help us; but does one Fulani go against another Fulani ?’
11 8 POLITICA L HISTORY O F NUPE KINGDO M
became too great. Th e towns or villages whic h were granted as such personal fiefs could lie anywhere withi n the boundary of the kingdom , on the royal domains as well as in the areas held as office fiefs by egba. Th e personal fiefs involved no legal or administrative authority, only the authority of levying an additional tax called edugi (small tax) over and above the royal tax, eduko (big tax), after the latter had been collected. There was no limitatio n to this additional tax, whic h must oftea have exploited the peasants ruthlessly. Yet I have also seen close personal attachment between villages and their lor d still existing to-day, bearing witness to the existence of more friendly ties between feudal lor d and tenants.
It is clearly impossible to state in more detail how this administrative system worked, and especially how far the administrative and judicia l authority of the feudal delegate worked in harmony wit h the existing traditional authority vested in village chief and eiders. Later on, I shall give more data about the workin g of this legal machinery. But it is easy to understand that the powerful feudal lor d and his delegate could uphold political and legal authority by the very fact of their power and influence. They could, for example, secure redress or punishment where the village chief’s limite d power might be too weak to enforce judgement or arbitra• tion . But this intervention migh t easily distort justice rather than further it . As aliens, never linked permanently wit h the people over who m they rule, the fief-holders were bad guardians of triba l law and traditional justice. This , in fact, reflects the fundamental conflict in • volved in the system of feudal administration. Th e feudal lords were officially responsible for the well-being of their subjects; they were agents of a central authority whic h was vitally interested in the main• tenance of peace and order; they themselves, like their royal master, depended on the people of the fief for tax, supply of food, and supply of men for their armies and wars. But the connexion of fief wit h rank, in • volving as it di d a change of the fief wit h every promotion, di d not allow the mutual attachment between the feudal lord and his subjects to grow really strong and permanent—too strong and permanent for the safety of the king and the unit y of the state. To balance the danger that might arise fro m a too firmly entrenched feudal lord , the kingdom had to risk the other danger of maladministration and discontent among the people on the fiefs.
Thi s policy of preserving the balance of power between dynastic and feudal interests whic h had to be re-established and readjusted wit h every change of regime is reflected in the many changes whic h the administra• tio n of the kingdom has undergone in the hundred years of Fulani rule. Etsu Masaba, intent on buildin g up an enormous dynastic power, divided the whole of Beni and Kusopa country, tow n for town , among his sons and daughters and other followers as personal fiefs. There was no place in the kingdom, the Nupe say, whic h had not its private ‘owner’. Th e Nupe nicknamed this generous king nda na ma ga na, Father of Dis • tributing . Etsu Umaru , who succeeded him , adopted the opposite policy in order to wi n the support of the native tribes against the feudal lords who had supported his predecessor. He abolished all private fiefs in
POLITICA L HISTORY OF NUPE KINGDO M 119
Beni and re-established the integrity of the Beni confederacy in the for m of a royal domain. Etsu Maliki , levying tax upon tax on a country depleted by civi l war, and having to defend a weakened kingdom bot h against powerful feudal lords and a restive population, invented a new system of government: he now applied the ajele system whic h was formerly restricted to the Outside Peoples to the kingdom itself. He divi • ded the country into six large districts placed in charge of his delegates— trusted slaves, who , accompanying the fief-holders on their taj-collecting expeditions from fief to fief, collected the royal tax at the head of an armed force strong enough to cow into submission rebellious subjects as wel l as any too independent feudal lords.1 When Etsu Abubakari came to the throne, he found all the wells of supply dried up. Th e taxing system of his predecessors had left nothing in the country that could be made into money—nothing except human beings. He started raiding his ow n provinces to find slaves for the market, breaking down the most funda• mental distinction of political status, that between the citizens of the kingdom proper and the conquered zazi kati. But, my informant added philosophically when he finished his sad story: ‘A gown is dirt y
—yo u wash it ; behold, the water is dirty—wit h what are you going to wash the water?’
In this ever-changing pattern of feudal administration the capital stands somewhat apart, being administered directly by the king and his officers of state. Th e threefold territorial and triba l division of the king • dom into an innermost core, a kingdom ‘proper’, and an outside area reappears on a smaller scale in the division of Bida and its population : here we have ba nin, the Inner City , where the Beni of pre-Fulani Bida live ; the Greater Bida, and finally the Strangers’ Quarter, Ialemi. But the order of status and prerogatives is not the same. Greater Bida in • cluded all the people who had come to Bida when it became the capital of the kingdom—the members of the rulin g class, Fulani and Nupe nobility , wit h all their followers, hangers-on, wit h traders, Mallams, warriors, drummers. A privileged class here again, this group was free fro m all taxation, and this privilege also extended to all settlements whic h they founded, the tunga in the country, whic h were worked by their dependants, slaves, and serfs. Th e Nupe craft-guilds whic h had come wit h the Fulani rulers from the former capital of the kingdom, Raba, were equally exempted from tax save for an annual .’salute’ to the Etsu. To the people of ba nin the people of Greater Bida, Fulani and Nupe alike, are ‘strangers’. But in the kingdom they themselves, kintsoii, Owners of the Land , were a group deprived of all privileges. They were taxed exactly as the other peasants in the country; and so also, by gate- tolls and market dues, which were collected by the king’s police force, were all Nupe from the country who came to sell their goods in Bida market.
Finally, there are the strangers—’real’ strangers—of who m this cos• mopolitan capital was (and is) always full : people from Hausa and Bornu , Arabs from the Sudan and Tripoli , Yoruba from the south.
1 These districts were: Cekpa (the country of Patigi), Gbakokpa (Badeggi and
Kacha), Egga, Beni, Kyedye, and Kusopa.
POLITICA L HISTORY OF NUPE KINGDO M
They come to Bida as traders or itinerant artisans, some for a short period only, others to stay. Th e trading caravans from the north stop outside the town gate in the zongu, camp, reserved for them. There are three such zongu in Bida, each situated on the territory of one of the royal houses; the caravans must stop in the camp that belongs to the ruling house—another expression of that comprehensive dynastic division into three. Th e traders from the south, individual traders and the strangers who settle down in Bida for longer periods, perhaps per• manently, five in Idlemi under their own headmen, the Sarkin Hausawa (or Madugu, as he was originally called) and the Sarkin Yorubawa (or Asari), the Chief of the Hausa and the Chief of the Yoruba. These two men are the recognized ward-heads of the aliens’ quarter, and latterly have held an official position on the Bida Tow n Council. Strangers from the rich countries of north and south were always welcome in Nupe. They were treated with special consideration by the rulers of Bida, they paid no regular taxes, only comparatively small market dues, and they were looked after by their own headmen, who were appointed to watch the interests of their fellow countrymen, as, for instance, in all matters concerning financial transactions. This treatment of strangers contrasts curiously with the ruthless treatment of the native peasants. It reflects the vital part that trade played in this country, in the households of this society which depended for its existence on foreign markets where im• ported arms, horses, and all the paraphernalia of court life might be bought, and where the valuable commodities of the kingdom, embroid• ered gowns, beads, brass- and silverware, and slaves might be sold.
Cutting across this threefold division of Bida there exists another division, still more important for the life of the town: the division into small local units, factions rather than organized town wards, which form themselves more or less spontaneously round the houses of the power• ful nobility. This is, in fact, how the Bida people refer to the different parts of the town; they call them by the title of the most prominent member of the nobility who has his house in that locality, efu Ndeji, or efu Dztvodfu, efu Sokyara, and so forth. This link of locality is explained partly by the fact that powerful men tended to settle their henchmen, slaves, and serfs in their neighbourhood, and partly by the fact that the people of the locality were ready to acknowledge as their patron the member of the titled nobility, living in their midst, who at the time had the greatest influence, the man whose rank and power promised protec• tion and support to those who rallied round him. But it is not the strong, permanent link of the kind we found in the Nupe village. Nor is there real continuity in this voluntar allegiance of a group of commoners to a noble patron. As the ranks change hands so the allegiance will change; every change on the political horizon may dissolve the local faction and reset the face of the town.
Just as these rank-holders are themselves dependent upon the favour of the king and the royal house, and as they group themselves in accor• dance with that tripartite dynastic division, so these small local factions are subordinated to the threefold arrangement. We remember the three ekpa, the three ‘banks’, on which Bida is built: the large number of efu,
POLITICA L HISTORY OF NUPE KINGDO M
linked with a title or name, encased in these ekpa which bear the name of a royal house, the group of commoners rallying round a noble patron, and the nobility rallying round a dynastic head—this is one of the most conspicuous symbols of the political relationships which rule the capital. Th e political administration of Nupe kingdom (like every political administration) can be regarded as an organization guaranteeing order and security to the subjects of the state—individuals and groups—in return for specific forms of submission: submission of individuals to the established rules of conduct; surrender, by groups, of all independent political action; and surrender, by individuals and groups, of portions of their property in the form of taxes and tributes. Th e law-abiding, tax- paying citizens are thus promised protection against enemies at home and abroad, and general security for carrying on their daily work, hold• ing markets, using the roads, and enjoying the fruits of their labour. Wit h regard to individuals, a legal machinery safeguards their civic rights, that is, their claims to this protection. Wit h regard to subject groups, we may speak of constitutional rights in which this safeguard of security is embodied. Th e element of reciprocity implied in these con• stitutional rights, namely, the promise of security in return for a sur• render of autonomy, is clearly understood by the people. This is shown in the current interpretation of such political measures as the suppres• sion of the Kyedye attack on the town of Kacha described above. The intervention of the central government on behalf of Kacha is positively linked with the previous surrender of autonomy (a gba ya Bida, ‘they
submitted to Bida’).
A striking feature of the internal organization of Nupe are the con• siderable regional differences in the completeness and directness of ad• ministration and the frequent changes which it has undergone in the fifty years of Fulani rule. These variations and changes have a threefold explanation: (i ) They interpret, in terms of political status, the cultural and ethnical provenance to which we referred in the preceding chapter. (2) They are a compromise between the superstructure of the state and certain claims to autonomy, admitted mainly because of the difficulty of establishing a more complete control—for example, over the strongly organized river-state of the Kyedye, or over the newly absorbed ‘outside’ groups on the periphery of the kingdom. (3) They are a compromise also between centralization and delegation of power. The delegation of power represents an essential structural characteristic of the Nupe political system. Th e ever-expanding kingdom, increasingly dependent on the military help of the nobility, had to secure their services more and more by granting them landed property in the form of fiefs, and thus investing them with a portion of territorial sovereignty. But then, only through this indirect—feudal—administration could the expanding state hope to reach and to control its entire population, above all those scattered village units which never evolved a larger political system than could be fitted easily (like the Beni confederacy or the Kyedye political organization) into the large-scale framework of the Nupe state.1
1 Compare the similar situation in Imperial Rome: ‘Th e state had to appeal to private persons of wealth and influence because it was not able to transmit its
122 POLITICA L HISTORY OF NUPE KINGDO M
But to ‘reach’ and to ‘control* means often wholesale exploitation rather than protection and security. Here the fundamental weakness of a sovereignty vested corporately in a ruling group, the members of which are entrusted severally wit h executive power, is clearly revealed. In other words, in the employment of the political machinery the aspect of duty towards the state is not invariably harmonized wit h the aspect of class privilege. Thus the peace of the country was often disturbed by the rulers themselves: one feudal lor d breaking into the territory of another, levying illegal taxes; a royal messenger misappropriating tributes which he was to collect for his royal master; or a royal prince raiding on his own the countries which had paid their tribute and thus bought their claim to the ‘king’s peace’. Kin g and notables alike would rally against the man who thus betrayed his loyalty to the kingdom. Punitive expeditions, loss of office, and heavy punishment were the reward of those who broke the ‘king’s peace’ and committed what the Nupe call cin amana (see p. no) .
Revolts were quelled, raids answered wit h raids, war fought wit h war, and the ‘king’s peace’ eventually re-established. But under conditions like this, protection and public security could be, at the best, only very unequal and unstable. Th e stress and tension of the system revealed itself in the various attempts on the part of the peoples of Nupe to secure better safeguards for their civic rights. Direct rebellion was the crudest, and, in fact, least efficient, weapon. Siding wit h one feudal lor d against another, wit h pretenders against the rulin g king, and princes against their royal overlord, meant obtaining, if not permanent security, the promise of protection and favours. Villages and districts could not choose their overlords or feudal delegates: but there are one or tw o indications that they could change their political status. Thu s the villages of Sakpe and Kutig i claim that they obtained under Etsu Masaba, in return for heavier taxation, the legal franchise normally granted only to royal domains. I have mentioned the frequent efforts of the Kyedye to secure a semi- autonomous position. Th e Kusopa of Labozi, taking advantage of
their exceptional economic position as the kola-nut planters of the kingdom, were also able to secure from Etsu Masaba the status of a royal domain equal to that of Beni and Kyedye.
PATRONAGE AND CLIENTSHIP
Th e privileges granted by one Etsu could be abolished by another. Security granted by charter could be infringed by powerful feudal lords. Where the group as such, village, or district, failed to obtain permanent security, individuals could seek protection of a different order: the declaration of voluntary allegiance to a feudal lor d or an influential man in the capital bought his and his family’s friendship and personal pro• tection. Thi s allegiance, whic h often went hand in hand wit h the voluntary adoption of Mohammedanism, woul d be initiated as a rule by
commands to the inert masses of the population in any other way’ (P. Vino- gradoff, in Cambridge Medieval History, i, p. 653).
POLITICA L HISTORY OF NUPE KINGDO M 123 presents whic h the applicant brought his patron-to-be. A short tria l
period established the readiness of the client’ t o fulfil his obligations,
and the ceremonial gift of a sword and turban by the patron to his new
henchman sealed the pact. Such a pact involved on the part of the client
gifts, services, and a regular tithe paid from the produce of his farm ; on
the part of the protector, help against claims, jus t or unjust, of other
people, against the abuse of power by other feudal lords, sometimes
grants of land, and invariably assistance in all matters that had to do wit h
the central authority. In legal affairs, e.g. court cases brought before the
central authority in Bida, the protector, residing in the capital, wit h his
knowledge of state affairs and his private influence, woul d give invalu•
able help to the powerless ignorant peasant. Even to-day, peasants from
the country, when summoned before the court in Bida, invariably go
first to see their protector in town , ask for his help or advice, and then
appear in court accompanied by one of his men—an outward sign of their
close connexion wit h the powers that be. Thi s system of patronage is
involved in the very structure of this rigi d class society; it grows to an
institution whic h impresses its stamp on the whole social life of Nupe
and indeed all the Emirates of Norther n Nigeria. We may call it by the
native name, bara-ship, from the Hausa and Nupe wor d bara, servant or
serf, or by the name by whic h the same institution was known in
Imperial Rome and in the feudal system of medieval Europe: the
patrocinium, or clientage.1
Clientage, this sacrifice of freedom for the sake of political protection, is common to country and capital. But in the capital it has a variety of meanings. First, there are the men who , anxious to obtain political pro• tection, attach themselves loosely to an important personage as his loyal followers withou t surrendering their independence. These men, who do not work for their patron nor brin g hi m gifts or tributes, are called egi kata, ‘sons of the house’—the Nupe wor d for adopted children. The n there are men who become bara—real bara—because they are poor, and sometimes also because they shirk hard work on the land. Impecunious craftsmen thus hope to obtain a prosperous customer; Mallams a wealthy patron who can use their services; and ex-peasants become servants in the house of their master, are given food and clothes, money for tax, and have their children educated by their patron. Here
‘protection’ comes to mean freedom from all more serious responsi• bilities. An d lastly, over and above securing material assistance, attachment to a man of rank and influence is a way to secure social pro• motion , the chance of becoming ‘somebody’, of being lifted from the ignominy of a commoner’s life into the sphere of rank and importance.2
Th e man in tow n who has no land to offer, offers himself and his services. Th e bara attaches himself to the household of his patron; he is servant, messenger, major-domo, soldier in his master’s private army. He may rise to the position of a trusted friend and intermarry into his
1 See P. Vinogradoff, op. cit . i, p. 563, and v, pp . 460-2.
2 One of my Bida informants, himself a man wit h numerous bara, divided the bara into tl\ree categories: those who ‘are afraid’ and seek protection ; those who shirk hard wor k and responsibility; and those who are out for ‘greatness’.
124 POLITICA L HISTORY OF NUPE KINGDO M
patron’s family. If he has been accepted as bar a by an influential feudal lord, he may hope to be rewarded with the grant of land or a sub-fief. A title, finally, bestowed by patron on client proclaims publicly the bora’s position as a henchman of a man of rank. These are the titles which are usually bestowed on bara: Sonfada, Mizindadi, Tsowa, Tswanya, Tsadti.1
In all these cases the benefit is mutual. Th e bara receives material support and possibly a rank; the patron gains servants, soldiers for the contingent which he is to muster for the king’s army, and followers and henchmen to add to his prestige. Th e patron will choose his clients and will not accept everyone. A young man of about 25, who had become a labourer, told me that he intended eventually to Attach himself as bara to a certain nobleman, but he was as yet too young and no one wanted him. To have gathered around oneself a large number of bara and egi hata is the final, most undeniable, mark of success. Whenever you visit the house of a nobleman, at any time of the day, you will find him surrounded by some of his bara or egi kata: possibly they have just returned from an errand dr have come to discuss important questions with him, or simply to pay their respects and share, as they regularly do, his meal.
Often the loyalty of clients to their patrons extends to the whole family of the bara and is transmitted from generation to generation. But loyalty and ambition may come into conflict. Sons may choose a new patron for themselves, hoping for increased benefits from the new association; or the bara himself may leave his own protector when the influence of the latter decreases and look for another more influential patron.
Th e ittte of clients are the bara of the royal princes. Not only are they servants of masters who have ample land and sub-fiefs to bestow, but they may also rise, with their masters, to the highest positions and become, some day, bara of a ruling king. These royal bara are almost equal in rank to the nobility of the officers of state. Th e Etsu Nupe has only two bara, the Mizindadi (lit. equal to major-domo) and Sonfada (lit. companion on travels—the present holder of this rank sits beside the chauffeur in the Etsu’s car), not counting the freed slaves who play to• day more or less the role of bara. But the lowest saraki titles which the Etsu may bestow on his favourites are nearer clientship than nobility proper. Social climbers, rich and ambitious men, could buy themselves a clientship in the house of a royal prince or the king: Etsu Saidu, the late Emir, sold the title of Mi£indodi three times over. Wealth buying a serf’s title—this is indeed a paradoxical result of a social system which, intent on promotion and social rise, turns even the humiliation of client- ship into a vehicle for a social career.
Th e complex pattern of clientship and the various motives which make free men seek patronage are illustrated in the following list of bara who were in the service of the late Shaba.
1 Political ranks are frequently distinguishecKfrom bara ranks using the same title by adding the suffix -hd, meaning great, orth e word tsu, meaning ‘of the king’, e.g. Sonfada and Sortfadako, Soje and Sojetsu.
POLITICA L HISTOR Y O F NUP E KINGDO M ” 5
List of Bara in the Household of the late Shaba Usman
(Bara titles are printe d in italics)
i . Ibrahim a Madawaki—in Bida. Formerl y warrior b y profession, now matmaker. Mad e submission to Shaba; his father an d the rest of the famil y are not bara of the Shaba.
- Etsu Ndako–i n Bida. Wealth y trader. He is the type of bara wh o is out for social promotion— a ‘bara for greatness’. He began as egi kata of the Shaba, wh o had no ran k at that time . Whe n Usma n became Shaba, Etsu Ndak o made ful l submission, in the hope of profiting by this allegiance to the ‘heir apparent* wh o migh t become Etsu one day and then bestow a saraki rank upo n his client.
- SonmdZi—in Bida. Ma t and straw-hat maker. Hi s father was a bara of Etsu Umar u (of the same dynasty). Th e son followed in his footsteps, an d served tw o Shaba in succession. He marrie d a widowe d daughter of his present master.
- Yis a Gbodu—i n Bida. Mad e submission to Shaba. Poor, a small trader by profession. Plays the role of a handy-ma n in the Shaba household.
- Jiya Fugi—i n Bida. Wa s formerly a bara of Etsu Mohamadu , the
Shaba’s father. - Yisa Sharu—i n Bida. He is a Malla m by profession, and became bara
of the Shaba before the latter obtained his rank.
7 . Mamud u Yakagi—i n Bida. A s No . 1 . - Umar u Zainu—i n Bida. Wa s formerl y a bara of Etsu Malik i (of a
different dynasty). He is a tailor by profession. Hi s patro n is at the same tim e his most profitable customer, having all dresses needed b y the famil y made b y him . - Babban Ndeji—in Bida. Hi s father held Ndeji rank. Whe n he died, his wido w marrie d the Shaba and took her son wit h her, wh o then became a bara of his stepfather.
- Ets u Saba Manzo—in Bida. As No . 1. Acts as personal messenger of the Shaba. Th e title Manzo means messenger (i n Hausa).
11 . Suleiman Gadza—in Bida. Hi s father was a title d slave of the Etsu.
He gave his son as egi kata to Shaba. Whe n the father died Sulei • man’s elder brother inherited the title, and the younger brother, having no prospect of obtaining a saraki rank, remained wit h Shaba as his bara an d received fro m hi m the bara rank Gadza. - Idiris u Madawaki—in Bida. A very wealthy trader wh o became bara
in order to obtain a title. (Cf . No . 2.) - Malla m Yusufu—i n Bida. Hi s father was Alkali of Bida, he himself formerly Alkali of Mokwa . Late r he lost his position and became
bara of the Shaba, for who m he now acts as a private ‘chaplain*. - Malla m Mamudu—i n Bida. Also a private chaplain of the Shaba
household. - Jiya Man—i n Bida. Anothe r Mallam . Hi s father was already bara
of the Shaba. - Ndak 6 Tetengi—whic h means ‘little Ndako ‘ t o distinguish hi m fro m his father, wh o was called Ndak 6 and was also a bara of the Shaba.
A t that tim e they lived i n Bida. Late r they buil t themselves a
far m settlement (tunga) on land given the m by their patron. The y
live there now, paying an annual dzankd to the Shaba.
126 POLITICA L HISTORY OF NUPE KINGDO M
- Alhassan—younger brother of Ndako Tetengi. Lives wit h hi m on the farm.
- Suleiman Maidunki—i n Bida. His father was a bara of Etsu Umaru.
He is a tailor by profession-—Maidunki means tailor (i n Hausa). Cf. No . 8. - Kolo Langbogu—on a farm near Wuya. He obtained his land through
his patron.
Th e social significance of clientship reveals itself, finally, in the formation of political factions. Every man of rank and standing collects round hi m a smaller or larger group of followers, henchmen, and clients, their families, their sons, brothers, and friends, who come to look upon this man as their protector and patron. Th e people of rank themselves are organized similarly, on a higher plane: we have seen how they attach themselves to a royal prince from who m they expect favours, support, and, some of them, promotion to an important office or ric h fief. Le t me tr y to reduce to figures this manifold system of integration that is the result of patronage and clientship. Th e old Shaba of Bida (as our list shows) had stil l in the days of his waning influence nineteen bara. The n there were his sons and younger brothers wit h their bara—some forty men altogether (not counting women and children). Of the other nobility , members of the royal house of Umar u Majig i to whic h he him • self belonged, and members of the saraki nobility bearing allegiance to this house, some eight or ten were attached to the ol d man in per• sonal allegiance, each of them wit h his own bara, bringing the total to over a hundred. Th e number of peasants, commoners, small traders, and Mallams who , livin g in his neighbourhood or on one of his tunga, regarded themselves as his clients in a looser sense, I can only guess. In Bida itself their group included all the streets round his house, twenty to thirt y houses, and the people from this group woul d come to the Shaba or any important member of his faction, whenever they were in need of advice or protection. Th e following of the Shaba must thus have amounted to over 400 men, a thousand if we count their whole families, in a tow n of 30,000.
What was said earlier in this chapter about the division of Bida into local factions or ‘wards’ is now fully explained. Yo u can map out the whole of Bida in such areas of personal influence and patronage. Rivalry and feuds between the leaders of the local factions may thro w the whole town , indeed the whole country, into a civi l war. Thi s is what happened not long ago. In 1931 the tria l of a witc h in Bida had deeply stirred the passions of the people. A certain woman in the efu of the Sokyara, a powerful nobleman, though not at the time in favour at the court, was believed to be a witc h and was accused by the people of the efu of having caused several deaths and cases of mysterious illness. Th e people feeling themselves menaced by her presence appealed for help to their patron, the Sokyara, wh o decided to expel the woman fro m his efu. She appealed to the Etsu and the Alkali; summoned before the court, she swore on the Kora n that she was innocent, whereupon the Alkali let her go and dismissed the case. Thi s to the enraged people appeared as a farce. Th e anti-witchcraft society to whic h the people of Nupe for-
POLITICA L HISTOR Y O F NUP E KINGDO M 127
merly used to entrust the discovery and punishment of witches had been forbidden some ten years previously. Th e modern method of obtaining protection against a witch , through the medium of the law courts, had evidently failed. Moreover, the people were aware of the enmity that existed between their patron and a powerful faction at the king’s court, especially between the Sokyara and a certain favourite of the rulin g Etsu, the JJban Dawaki, wh o had recently obtained his rank by what had seemed to many gross favouritism. Th e action of the Alkali, who was known to belong to the same faction, was supposed to be prompted by the wish to please the Etsu’s favourite and to discredit his rival. En • raged, and despairing of any other way out, the people took the law into their own hands and stoned the alleged witc h and her daughters on their way home from the courthouse. The n the fight between the two factions flared up openly. Althoug h certain members of the king’s household had, as it was revealed later during the official investigation, taken part in the stoning of the witches, the Etsu accused the Sokyara and had hi m and his whole family arrested. Th e Sokyara and his brother managed to escape, and hi d in their house. When this became known in the tow n the people of the Uban Dawaki faction thought this a good opportunity to deal the riva l faction their death-blow. They set fire to the house of the Sokyara, who sought safety in flight while his brother died in the flames. Later the Uban Dawaki party trie d to throw the blame on the Sokyara and accused hi m of having himself started the fire. Bida was in upheaval, the Etsu powerless, or pretending to be powerless, and only after the sternest measures could the government establish peace.
But we need not go to this extreme case to demonstrate the reality of this system of political factions.1 Extend the ‘circles withi n circles’ of patronage and clientship wider and wider til l they are bounded by the widest circumference—the three dynastic ekpa—and you understand their ful l import . Patronage and clientship are only small meshes i n a texture whic h envelops the whole social life of the kingdom.
SOCIAL CLASS I N NUP E
Th e foregoing description has revealed a system of social gradation of remarkable thoroughness and indeed conspicuousness. We can speak of Nupe as a typical class society. But hitherto we have touched upon only one aspect of social stratification, namely, social differentiation so far as it results from differentiation in political rights and duties. We have studied social barriers—the barriers of rank and grade—where they are pivoted in the political structure. We have yet to learn the ful l extent of their influence upon the life of society at large and their relation to other trends of social stratification.
Social class is a complex phenomenon, the resultant of many factors,
1 There is another hardly less drastic illustration: in April 1936 the same Uban Dawaki fell ill. At once rumour had it that he had been poisoned by the Maiyaki, an old enemy of his who was said to be jealous of the large following which the Uban Dawaki, though his junior in rank, could boast. Tha t the medical officer diagnosed lung trouble did little to abate the rumour.
128 POLITICA L HISTORY OF NUPE KINGDO M
chiefly political, economic, occupational, and educational.1 But so is the widest constellation defined on the basis of similarity or diversity of conduct—culture. Th e essence of social class lies in the specific fashion in which these various factors appear combined. To understand it fully we must view social class under two aspects. Th e first is that of firmly established and comparatively permanent differences in actual conduct— differences in standards of living, in habitual behaviour, occupation, and moral values.2 Th e second aspect is that of the social agency which attaches to these differences of conduct the index inferior’ or’superior’, and makes them reflect different grades on a common scale of values. Were we dealing only with differences of the first order we should be describing merely different cultural groups in which life is lived in different fashion, in which different customs are observed and different values valid. What turns mere cultural differences into differences of conduct typical of social class is the loss or absence of their self-con• tained and largely unconscious nature. Their permanence is no longer, as it were, self-chosen. The ‘cultural’ differences are now conceived of as results of unequal advantage and opportunity, and they are never allowed to lose their significance as different grades in the ‘common scale’. It lies in the nature of things that this common scale, which depends essentially on the demonstrability of unequal opportunities, should rest above all on political and economic factors. Th e essence of class (as of any group) lies in the consciousness of group membership; the consciousness of class is the consciousness of this ‘common scale of values’.
Let us examine first the social mechanism which maintains and pro• pagates the consciousness of the common scale. It is essentially an educational mechanism (in the widest sense of the word). While per• petuating the various cultural traits characteristic of social class it per• petuates at the same time the common perspective from which these traits are set off as traits valued more or less highly by the society. One of the most efficient instruments of this mechanism is the convention of class symbolism and class etiquette, which not only regulates the inter• course between the classes, but also serves to demonstrate the reality of class barriers.
Nupe society is plentiful in such symbols of class distinction. The collective name for nobility, roumi, ‘turban’, marks already one of these outward signs of class: it refers to the turban of glazed blue cloth which only members of the nobility may wear. Another prerogative of rank is the right to carry a sword, or rather, have it carried by one’s sword- bearer, the zagi (lit. ‘little person’, i.e. ‘page’), and to appear on horse• back on all official occasions. Wit h regard to these emblems of rank the Nupe have kept abreast of the times: a novel symbol of status is an alarum clock, put in a little wooden case, which body-servants of men of rank carry after their masters.
1 M. Ginsberg, Sociology, 1934, p. 162.
2 Cf. M. Ginsberg, op. cit., p. 159: ‘T o the outward observer class is primarily a matter of behaviour, speech, dress, education, and especially habits of social intercourse.’
POLITICA L HISTORY OF NUPE KINGDO M 129
Or , in the sphere of etiquette: men of equal rank greet each other by cowering low (women go down on one knee) for half a minute or longer while exchanging the formulae of greeting and, if they are well acquainted, stretching out their hands several times in succession and lightly touch• ing each other’s fingers. A man of lower rank when meeting his social superior wil l bow very low or kneel down, and only if the difference of rank is slight or obviated by personal intimacy (as between a bora and his patron), wil l he venture to offer his hand. If he is on horseback, he wil l dismount; i f he is wearing sandals, he wil l take them off, though i t may be on the road, and put them beside him ; if he enters the house of his social superior, he wil l keep close to the door, squatting there til l he is summoned to come near. Th e man of higher rank wil l stand or sit still , or make a short perfunctory gesture of bowing, and hardly move his arm to meet the other man’s hand. When sitting together in a group, the man of low rank always sits slightly behind the man of higher rank. In the presence of the Etsu, Alkali,the highest ranks, or royal princes, you may not rest your arms on the floor—this would be disrespectful: the proper posture is to squat cross-legged, the arms folded in your lap, and to bend forward a littl e so that it should not seem as if you were looking straight at your superior. Th e Ndeji and Maiyaki were the only men who m I have seen behave really freely in the presence of the king ; all others, even the Alkali and the royal princes, acted in a conspicuously submissive and slightly constrained manner.
Tha t one addresses one’s superior always through a thir d person is a common feature of African etiquette. So is, I believe, the custom of pulling the sleeve or a cloth over one’s hand when offering a gift to a person of superior rank. Of gifts as a symbol of Nupe class etiquette there is more to say. In Nupe (as in the whole of Northern Nigeria) kola-nuts are the commodity most universally used for gifts and pre• sents. To announce your visit or to invite somebody’s visit, to wi n the favour of another man or to express your appreciation of another man’s services, you send a present of kola-nuts. Thei r quality and amount reflect precisely the social status of donor and receiver.
I have already mentioned the existence in Nupe of two kinds of kola- nuts, the cheap southern variety, and the expensive kola-nut from Labozi ; the latter, as we have seen, could be obtained only by, or through, the Etsu himself—presents of Labozi kola-nuts were thus symbols of royal favour, and unobtainable by anyone who was not in one way or another connected wit h the court. Whe n I attended the Sallah at the EtsiCs court in 1936, he had two bowls (teapots, as a matter of fact) ful l of kola-nuts in front of him , one filled wit h Labozi nuts and the other wit h the small cheaper kind . To watch hi m distribute the kola-nuts— the kin d and amount as wel l as the gestures which accompanied the gifts—was to watch the rank system of Nupe come alive. He presented ten Labozi nuts through a servant to all the higher rank-holders, and five in the same manner to the less exalted officers of state. He then gave one especially big one wit h his own hands to the men of highest position— the Ndeji, Maiyaki, Alkali (and myself)—and threw one into the lap of favourites-and friends whose rank called less for formality. Finally,
K
130 POLITICA L HISTORY OF NUPE KINGDO M
small kola-nuts, in fives or tens, were handed out to messengers, dancers, guild-heads, and other people of lower rank.
Even language and music are used as symbols of class. There exist in
Nupe special drum-rhythms, take, greeting formulae, kirari (Hausa, lit .
‘calls’), and songs whic h are linked wit h certain noble families and im •
portant ranks. As soon as a family comes to power or a man receives
rank and title , the drummers of Bida wil l compose such take and kirari
in their honour. Yo u can, in fact, have them made to order, at various
prices; bu t an influential person wil l not need to order them, for the
drummers, counting on his patronage, wil l supply hi m at once wit h this
musical salute. Some of the take and kirari whic h one can hear to-day
date back to pre-Fulani times; others were composed recently. When •
ever a member of these noble families attends a festivity, the drummers
woul d announce his arrival by playing ‘his’ take and singing ‘his’ kirari,
and the ‘master of ceremonies’, who is always present at such
gatherings, woul d recite a long list of illustrious ancestors to who m the
nobleman can lay claim, the ranks whic h he had held, and the deeds
whic h he achieved. Lastly, the Nup e address each other by titles only,
and never by name, if the person addressed bears a title . Special
formulae of greeting are linked wit h every rank, and thus mark position
on the social scale by the most obvious system of signs—everyday
speech.
Class symbols and class etiquette shade over into the more substantial differences of class. Behaviour whic h is pressed into the service of class manifestation becomes indistinguishable from behaviour whic h results from differences in social and economic advantage. Th e righ t to carry a sword or appear everywhere on horseback presupposes the economic ability to acquire these ’emblems of rank’. Submissive behaviour of a client towards his patron, or a member of the lower class towards a man of rank, is determined as much by the convention of class etiquette as by the special interest of gaining favour in the eyes of a person of consequence. Here we are dealing wit h the other aspect of class differen• tiation, those ‘cultural ‘ differences whic h are, or become, typical of differ• ent classes. The y revolve round the two pivots: political and economic structure.1 The y involve the existence of political prerogatives and the varying measures of individual liberty and mastery over one’s own destiny, enjoyed by the different classes. Here we can outline the exist• in g classes of Nupe society. We have met their protagonists already: the rulin g class of the kingdom, ric h absentee landlords, men whose occupa• tio n was war and slave-raiding, and, low down on the social scale, com• moners and peasants, heavily taxed, and gaining a scanty livelihood rendered insecure by the very facts that add power and prestige to the rulin g class. We have also met the craftsmen in the town , who share some of the privileges of the nobility , and the Mohammedan intelli • gentsia who can attach themselves to the highest stratum irrespective of
1 Th e two structures do not define identical strata, nor is the economic differentiation invariably reducible to political prerogatives. However, we will leave the examination of the exact relation between the two ‘pivots’ of social stratification for a later chapter.
POLITICA L HISTORY OF NUPE KINGDO M 131 their class origin. There is no sphere of life, not even an ordinary social
gathering, in which the representatives of the various classes can meet, as
it were, on ‘neutral ground*.
Th e social stratification stretches beyond the formalized rank system embodied in the political structure. It embraces the whole country and absorbs and reinterprets in its own sense the cultural configuration of Nupe at large. On this wider scale the commoners of the town, the people of no name and title , rank higher than the people and tribes out• side the capital, wit h their petty village titles and offices. Th e citizens of Bida look down upon the gberizi, the ‘heathen’ in the villages, ridiculin g their habits and character, their fashions and the way they talk, their lack of education and ‘urbanity’.
These differences of customs and habits are real enough. But they are underlined on the part of the Bida people by that credulity and fondness of exaggeration whic h we encountered already in the discussion of the village community and which, here even more strongly, expresses the biased self-centred interest of a group strongly conscious of its cultural identity. I have recorded many sensational accounts of the religious customs of the gberizi in which the Bida people firmly believe, unaware, or unwillin g to admit, that they are quite untrue. Bida informants tol d me that among the peasants and pagans women practise the craft of medicine and not men as in Bida—another statement whic h I could easily disprove. Or I was warned not to believe anything that villagers say—’they are all liars’. Even if it were possible to test quantitatively this collective libel, it would , I am afraid, hardly reveal any real differ• ence between town and country. What I have called, in the discussion of the village community, an ideology sustaining the assertion of group identity subserves here the assertion of (i n a loose sense) class distinctions.
To return to the real, concrete, cultural differences linked wit h class. Th e framework of status which we have outlined is filled out wit h a multitude of minor traits, differences of conduct, extending even to walks of life which may appear only remotely connected wit h the central political and economic structure. We shall meet wit h these differences in every stage of our investigation; let me anticipate here a few typical instances as they woul d strike the observer of Nupe everyday life. He would note class differences even in one of the most elementary wants— nutrition : rich and poor have different food habits. Th e poor eat littl e meat, chicken very rarely (unless for religious reasons); but meat cooked in palm-oil is eaten every day by the wealthy, and chicken-and-rice, the most delicious food the Nupe know, comes on the table at least once a week. In Bida the sorghum or millet grain is always milled to fine flour; in some peasant districts it is boiled unmilled—a preparation of food whic h the more refined despise, although it is undoubtedly of higher nutritive value. Th e people of Bida drink palm-wine, the peasants in the districts feast on sorghum-beer.
Again, in the matter of recreation: among the peasants the men dance to the music of flutes and drums, or the women sing to the clapping of hands or the beating of upturned calabashes. Among the sophisticated
132 POLITICA L HISTORY OF NUPE KINGDO M
people of Bida professional women singers, accompanied by their own drummers, perform for an audience which pays for its pleasures. Th e uneducated pagans like to sit together on afternoons and evenings in front of their houses, chat or weave mats, or just enjoy the peace of these hours of leisure. Th e educated Mohammedans meet in the house of a Mallam and read with him the holy books of Islam, using even their leisure hours for furthering their education.
As regards dress: a man of standing would always, even in the privacy of his home, wear trousers and the flowing white robe, and would never appear before guests and strangers without turban. Peasants and com• moners normally wear a loin-cloth and a blue cloth thrown over the shoulder, and the blue Phrygian cap which is said to be the traditional head-gear of Nupe. Only on state occasions would they don robe and turban.
People of rank have adopted, with Mohammedanism, many features of Hausa culture. In their families the Hausa custom of removing the uvula of small children is practised—a custom absent among the lower classes and unknown among the peasants. They are fond of using Hausa words and phrases; for expressions concerning the delicate topic of sex they invariably use Hausa or Arabic circumlocutions. They despise the peasants and uneducated people of the lower classes who use the crude literal expression, and who cannot embellish their speech with the more elegant phrases borrowed from Hausa.
Th e various features enumerated above are clearly not of the same order. Characteristic though they are of the standards valid in the differ• ent strata of Nupe society they are not equally reducible to the central facts of class structure—political and economic conditions. Th e evalua• tion of certain habits as inferior or superior is as much a matter of con• vention as a result of social or economic inequality. That the poor man cannot afford meat or chicken-and-rice is due to his poverty; but the preference for milled or unmilled grain or for palm-wine or .sorghum- beer respectively, is a matter of taste and habit, i.e. a ‘cultural* trait pure and proper. Yet rendered permanent by education, linked with habitual group-behaviour which in other, more vital, spheres of life positively reflects economic or social prerogatives, these smaller traits, too, gain the significance of class-determined behaviour. Here we grasp another aspect of class differentiation. I have said that where the attribute of inferiority or superiority is disregarded, differences of habit and living conditions in general are reduced to cultural differences in the narrow sense. This gains special significance in a society like Nupe, which has grown out of a conglomerate of various tribal groups, each with its own more or less specific culture. Some of the food habits of the ‘poor’ peasants are indeed the original food habits of cultural groups which have been incorporated into the kingdom. I have mentioned that the Phrygian cap, worn by peasants and commoners, is regarded as the traditional head-dress of the country. Their ‘uncouth’ manners, their ‘primitive’ customs, are part of the pagan peasant culture of the country. Similarly turban and flowing robes, or the appreciation of professional musicians and Mohammedan learning, reflect the preferences of a specific cultural
POLITICA L HISTORY OF NUPE KINGDO M 133 group, the ‘tow n culture’ typical of Norther n Nigeria. These habits and
customs exist, as it were, in their own right, quite apart from questions of
economic or political opportunity. But as in the growt h of the kingdom
certain customs and habits have become representative of the elite, they
have set the standard of cultural desirability and superiority in general.
The y have imparted to habits and customs whic h differ from the
standard the stigma of lower, less ‘respectable’ ways of life. In this
sense, then, class differentiation as we find it in present-day Nupe can be
said to represent a direct result of the growth of the state.
Th e intensity of class cohesion and the vividness of class conscious• ness are determined by ‘the amount and ease of social mobility’ , that is, inter-class movement.1 We have denied to the economic and social stratification that obtains in the Nupe village the character of a social class, on the grounds that the higher and highest social orders were poten• tially attainable to all—universal mobilit y contradicts the conception of social class. Exclusion of all social mobility , on the other hand, woul d tur n social class into hereditary caste. Th e Nupe state, a true class society, steers a middle course.
Th e social classes in Nupe represent to a large extent self-recruited, closed groups. A man born of commoners as a rule remains a com• moner, a craftsman a craftsman, and the ranks of nobilit y are filled wit h men of noble birth . But there exists promotion from grade to grade withi n the class, and a certain amount of inter-class mobility . Inter-class mobilit y in the fullest sense, that is, formal adoption into the higher orders, is attained, as we have seen, through service in war and, to less extent, through wealth and clientship. Its scope is narrow and closely controlled: there exist only thirt y to forty ranks to whic h an ‘outsider’ may aspire, and only seven out of the thirty-seven civi l and military ranks enumerated in our lists were granted to men whose families had held no rank previously. Moreover, adoption into the title d classes means only gaining a first foothol d on the long social ladder. Al l social rise possible in Nupe is rigidl y canalized by the hierarchy of ranks and grades. Here, viewed from the angle of social mobility , the Nupe rank system reveals its ful l significance: it controls and restricts as much as it aids social advancement. In the interest of governing the extensive king • dom, carrying on its wars, and administering its subject groups, political power had to be distributed and delegated. Individuals whose services were of value had to be attracted by the promise of admission to privileged positions. But by making social rise subject to the formal in • vestiture and, above all, by admitting individuals only gradually, through a step-by-step promotion, to increased power, the rulin g class could regulate the infiltration of new blood as wel l as the accumulation of power by its members. We have seen that ranks and promotion are subject to changing political fortune; even the meaning of ranks and offices may change under different sovereigns. Recognition can be with • held or withdraw n from men who migh t prove dangerous or from families already too powerful. Th e hierarchy of ranks thus helps to pre-
1 M. Ginsberg, op. cit., p. 163.
134 POLITICA L HISTORY OF NUPE KINGDO M
serve the precarious balance of power in the feudal state. It becomes, in the hands of the established ruling group, an efficient weapon to safe• guard their class privileges.
Social mobility of a more fluid, indirect kind, which admits individuals to a close association wit h the privileged classes and indirectly to a certain share in their privileges, reaches a much larger proportion of the population. It is upheld by the institution of clientship and by the existence of the Malla m intelligentsia ‘between the classes’. Th e possi• bilities of social mobility , above all, the extensive possibilities of this indirect mobility, account for the lack of cohesion among the lower classes of Nupe, or at least of Bida. Th e conspicuous symbolism of rank and the rigi d definition of the social estates cannot but stimulate class- consciousness: commoners wil l always describe themselves as tdlaka, and men of rank wil l leave you in no doubt about their consequential position. But this class-consciousness is blurred by thoughts and pros• pects of a social career that might raise anyone to a higher estate. Con• cern about betterment rather than sentiments of group solidarity define the nature of class-consciousness in Nupe. Thi s can be seen most clearly in the attitude of the younger generation, who are just planning their future career. Only in two social groups do we find a more strongly pronounced and self-contained consciousness of common fate and common, class-bound interests; they are characteristically groups in whic h mobility, direct or indirect, plays hardly any part—the closed group of guild-orgamzed artisans and (i n the framework of their scattered communities) the peasantry.
Ho w far social mobilit y took the form of local mobility, that is, of a movement to the tow n in the attempt to bridge over the cultural and social gulf between townspeople and peasantry, we cannot say. If it existed, it could only have been of very limited scope. For there were strong natural limits to such a development, above all, the scarcity of land near the capital.
No w to be complete and convincing, social rise must go hand in hand wit h assimilation to the habits and modes of life typical of the higher classes. This , in Nupe, proves no stumbling-block. Th e familiar in • stances of ‘parvenus’ who, though adopted into the higher social order, do not fit into the picture of its social life are unknown. Th e explanation is that cultural assimilation of this kin d is much more widespread than actual mobility . Th e mechanisms of social mobilit y whic h we have enumerated all accord their benefits to individuals, not to groups; on the contrary, individuals may succeed where groups (local communities, triba l sections) would fail to achieve that improvement of social con• ditions and that increased security bound up wit h classes. Bu t the assimilation of whic h we have just spoken must be regarded as a qualified for m of social mobilit y whic h includes groups and whole sections of the population. Whole groups are free at least to imitate the external characteristics of the higher social orders. Where economic conditions allow it commoners and peasants adopt habits and modes of life of their betters—fashions in dress and housing, their way of talk, their food habits, and the forms of entertainment whic h they patronize. The y also
POLITICA L HISTORY OF NUPE KINGDO M 135 adopt, even more eagerly, the religious practice of the higher and highest
classes, Mohammedanism. Thi s is not ‘real’ social mobility ; it serves
only to blu r the border-line between the strata. It involves attachment
to a higher stratum only in the sense of external closeness and
assimilation.
Th e assimilation achieved is irregular and of very unequal signifi• cance. Yet it is prompted by a definite motive running through the whole society—self-enhancement. Th e readiness of groups to identify them• selves culturally wit h the representative stratum reveals an incentive to overcome that heterogeneity of cultures and groups in whic h we have grasped the essence of state structure. Potentially, this incentive may carry forms of cultural reorientation capable of wielding more intensive, more real, commonness. To examine how and how far this potentiality is utilized by the political system is to examine the crucial problem of this discussion: the dynamic relation between political integration and the heterogeneous raw material out of whic h it must fashion its unity.
BINDING FORCES
Wha t then is the specific apparatus of control by means of whic h this heterogeneous society, severely divided by gulfs of culture, ethnic ex• traction, community, and class, maintains itself as a unit? An d to what extent does it draw into its orbit and use for its own support these other non-political configurations and spheres of social life ? Whe n examining the political organization of the Nupe village fro m this viewpoint we discovered that there the basis of political control is extended to its widest possible limit : political organization utilizes the integrative forces of community life at large; it organizes into a uni t what is integrated into the same uni t by every existing institution. It migh t very wel l seem from our foregoing analysis of the Nupe state that here we have the exact opposite. A state machinery which concentrates political power in the hands of a small ilite and ignores—save in order to transform into class barriers—tribal and cultural identity, seems committed to the narrowest system of political control : by coercion and physical force.
However, this section wil l show that such conclusions woul d be in • correct. One might, on general principles, reason that no state could afford to rely solely on coercive measures; and that it woul d rather utilize coercion only as a ‘short cut’ to bring about sentiments which , ‘once established and sanctified, . . . may well flourish withou t compulsion, glorified as loyalty to a sovereign king or a national flag*.1 A state that woul d disdain to call to assistance these bonds of inner unit y seems to jeopardize continuously its safety, and to risk being shattered and broken up by unrest and revolts, possibly beyond repair. Such tension and unrest seems indeed characteristic of the political history of Nupe. But the historian woul d have to point out also that wit h all these grave disturbances Nupe kingdom yet maintained itself as a unit . Even when the Fulani took over the rule, testing most severely an already strained
1 R. Lowie, The Origin of the State, 1927, p. 117.
136 POLITICA L HISTORY OF NUPE KINGDO M
political structure wit h the additional weight of alien rulership, the state di d not break down but on the contrary appeared to have taken out a new lease of life.
No w we shall be able to prove the existence of strong trends of in • tegration which form withi n the political unit, incidental to it or fostered by it , and which, bridging over cultural diversity and the rif t of social and economic inequality, endorse in their fashion the uni t wrought by political control. However, in our search for these ‘bindin g forces’of more subtle and conciliatory type we shall discover that they do not all point the same way. The y are indeed ‘binding* forces—but what they tend to bin d and draw together is not invariably the same group substratum.
We must examine first a significant negative instance—kinship. Kinshi p has no part in this integrative scheme, either ideologically or factually. No attempt is made to invest the conception of the state wit h that fiction of genealogical or kinship unit y which is linked wit h the concepts of the sub-tribe and the village community. Wit h the increased freedom of movement which is one of the results of political unification, kinship relations become more widely scattered: marriage between distant localities is increasing and is regarded more and more as normal. But these new kinship relations draw only a patchy, irregular pattern that can have little influence upon the integration of the people at large. Moreover, kinship relations are often subordinated to political interests; patronage and the adherence to political factions may even break up family units. Th e widest gulf, between rulers and subjects, and between the elite culture of Bida and the peasant sections, is bridged only by the sociologically less significant form of intermarriage between men of the upper and women of the lower classes. We shall see later that ful l intermarriage is rigidly restricted and rendered specially diffi • cult, unless it follows already existing political ties of patronage. We shall, in fact, have to speak of a definite class endogamy among the rulin g classes. Marriage and kinship ties thus cement the solidarity of the com• ponent groups of the state, not of the state itself.
We continue our search in the economic field. Political unification carries in its wake expansion of economic co-operation and interdepen• dence. Economic areas were re-designed in accordance wit h the new, wider political uni t and its centre of gravity, the populous capital, which depends for food and supplies on the agricultural districts. But the con• tributio n of this widened economic co-operation to national solidarity is negligible if not wholly negative. In this reorientation of economic links the boundaries of the state have frequently been crossed and disregarded: trade became ‘international’; craftsmen, traders, and even peasants are always ready to seek profitable new markets outside Nupe, and even to emigrate for the sake of better business prospects. Economic inter• dependence and co-operation even introduce new conflicts and forms of antagonism which set group sections against each other. We shall speak of the village artisans who openly admit (and partly resent) the competi• tio n of the superior Bida crafts. In times of crisis conflicts of this kin d tend to become more strongly marked. Durin g the depression of
1934-5 I could notice distinct resentment on the part of the peasants
POLITICA L HISTORY OF NUPE KINGDO M 137 against the people of Bida, on who m they were dependent for the mar•
keting of their produce, and who m they now held responsible for the
disastrous fall in prices.
Legal unification goes a step farther. Dependence upon a common forum of morality evokes a certain group cohesion, a common orienta• tio n of min d at least i n that all the component groups wil l share common hopes—and fears—and a common interest in whatever affects the judicia l organization. If we can test the strength of economic co-opera• tio n by the facts of volume of trade and traffic and the relation between productive and consuming areas, we can test this moral reorientation by examining the measure of interest whic h the various groups evince in legal change. I shall show later how keenly the people everywhere in Nupe follow changes in the law. But I shall also show that this interest, all the more intense because such vital issues are involved, does not invariably foster the dissemination of true, or reassuring, knowledge: often it spells suspicion and fear of a biased jurisdictio n and legislation. Whole sections feel themselves tricked and threatened by this dangerous law and reject co-operation wit h its agents and the groups which it seems to favour.
Dynastic and political changes engage the attention of the people no less strongly. Whe n the late Emi r died and a new Emi r was appointed, and when the appointment of the highest royal ranks was to take place in Bida, the events elicited comments from many groups from whic h one woul d not have expected such interest in ‘higher polities’. One day I was standing in a Bida street, talking to a brass-smith, when the Kpotui one of the disappointed claimants to the throne, rode past. Th e brass-smith looked after him , shook his head, and said to me: ‘I t must have been a great shock to hi m when someone else was appointed Etsu.’ In Kutigi , the district of the Kpotu, the older men knew all about the rumour of the Kpotu’s illness, whic h had been given out in Bida as an explanation or excuse for his not having been promoted to a higher rank. An d in Mokwa , the district of the Rani, the elders tol d me wit h amusement of the abortive attempt of their Distric t Head to obtain promotion by supernatural intervention: how he had gone to Raba and prayed at the tom b of Malla m Dendo for the success of his political aspirations.
‘Unit y of interest in this sense obtains also wit h respect to all things that affect the people, and the news of such events woul d travel fast from place to place. In Marc h 1936, at a Bida wedding feast, I happened to hear girls sing a song the words of which sounded unfamiliar to me. I found it to be a new topical ‘stanza’ which referred to the activities of the Sleeping Sickness Commission which even then was working in eastern Nupe, i.e. Agaie or Dib o country. Th e song gave voice to the suspicions and fears of the people and their anxiety to escape, by what• ever means, the unpopular treatment. Th e song runs as follows:
The needle is coming,
The Mallam of the needle, The father of threading.
O let me go,
An d not only for my money [i.e. also for pity’s sake].
138 POLITICA L HISTORY OF NUPE KINGDO M Th e needle killed the men of Dibo ,
An d also the men of Baro [a town in that district].
O father, save me and let me go. If you must use your needle,
Do not let it bite me.
Th e ‘needle’ is the injection needle, and the ‘Malla m of the needle’ the native medical assistant who was performing the injection and who, according to the testimony of the song, could be bribed to spare the patient the (very painful) injection. Th e testimony, incidentally, was true ; some time later I learned that one of the native medical assistants had actually taken bribes fro m the people, and had been apprehended not long after I had heard the song.
To summarize: political co-ordination calls into being a co-ordination of interest—actual interest units—which in certain respects overcomes the heterogeneity of tribe, culture, and even class, and includes the whole people. Yet as economic co-operation goes its own ways, often disregarding political frontiers, so the other forms of ‘reorientation’ may tend to further the formation of new, even antagonistic, groups rather than to cement the political unit . Th e ‘interest units’ are, paradoxically, units that spell disunity. Th e paradox is easily solved: the common interests round whic h these units are focused are such that they show a different aspect and portend different things to rulers and ruled, or to adherents of different political factions. Th e units whic h they create may efface triba l and cultural divisions, and temporarily even the barriers of class: bu t i n critical moments they wil l only pu t forth more forcibly the existing disparities and forms of antagonism.
We turn , then, to another group of integrative factors whic h seem to succeed where the interest units fail. The y wiel d integration whic h endorses unconditionally the political unit . Th e secret of their success lies in that they attempt unit y and integration in spheres in whic h the political interests suffer such integration to exist, and in which the diver• gent political interests are overcome by some stronger incentive to unity.
Th e study of the unit y whic h the Nup e state has achieved, or is achiev• ing , reveals the effectiveness of two main groups of such incentives. To formulate them tentatively: one is self-enhancement—especially a self- enhancement that (i n the language of psycho-analysis) ‘sublimates’ the humiliating experience of political dependence and inferiority ; the other, the appeal of the mystic and supernatural. Th e first incentive reveals itself in the progressive identification of the people wit h the cultural values typical of the social elite; the second, in their readiness to adopt new myths and embrace religious beliefs and practices, whic h are vessels of political * propaganda’ at the same time. Th e two incentives may joi n forces, self-enhancement being offered in the realm of religion
—as in the spreading of Islam, the religion of the rulers and conquerors, among the subject groups of the kingdom. We can also discern the agency of a thir d incentive, more unspecific and subsidiary in nature, which , as it were, canalizes the influence of the other two , namely, the
POLITICA L HISTORY OF NUPE KINGDO M 139 appeal to an historical (or quasi-historical) tur n of mind . I have already
called the Nupe an historically-minded people: here is more evidence
for this bent. We shall see presently how under its influence ritual
emblems become relics, and self-enhancement takes the form of exulta•
tio n in past glories.1
Self-enhancement, then. These are the historical facts: the Nupe state has bound its many groups and peoples to its own fate, made them share its adventures, its conquests, the fruits of its victories and defeats. But the element of compulsion behind this commonness of fate dis• appears in its mental reflection. Th e commonness of fate is lifted up into the sphere of things eminently glorious and desirable. Th e people are eager to identify themselves wit h the kingdom and its fortunes, and share in a collective pride that ignores the social and cultural distance between rulers and ruled. Even the wars which deprived the Nupe of their own rights, chiefs of their independence, peasants of their wealth and land, become a cherished heroic memory. I was discussing the role of strangers in Bida wit h old Ndakotsu, one of my great friends, and I asked hi m whether there had been any Kamberi, Kamuku , or Gbari among the strangers who visited Bida regularly. Ndakotsu, last descendant of the Beni town-kings of Bida, who never became quite used to the idea that the goizi (Fulani) had robbed hi m of ‘his’ town , said ironically: ‘Kamuk u and Kamberi? Yes, but only as our slaves’—the
‘our’ evidently meaning a Nupe which credited to itself the successes of its Fulani rulers. When I first came to Mokwa my informants first of all showed me proudly the historical places near Mokwa where some of the battles in the wars of the Fulani against the usurper Umar Bahaushe had been fought, and insisted on describing them all in detail, as if this were the most important thing I had to learn. Thi s di d not prevent them later on from complaining bitterly about the tyranny of the Fulani and their indifference to the fate of the people.
Thi s new, proud consciousness of a unit y whic h embraces the whole kingdom is kept alive by every song telling of old times and every story relating the adventures and feats of the kings of Nupe. Wit h fashions and cultural practices, songs and stories spread from the capital through• out the country. Th e songs composed in honour of the king or one of his-noblemen by the drummers of Bida find their way into all the bigger villages of the country. In Mokwa I heard a song which had been com• posed in Bida in 1918 in honour of Etsu Bello’s first motor-car, the first motor-car in the country. I put this song in here as an example not of the force of poetic expression which it possesses (i n spite of the un • congenial topic), but as an example of its efficient ‘propaganda for the kingdom*.
1 There exist also certain other subsidiary stimuli, such as aesthetic appeal, which we may disregard here. One may put the question whether these various stimuli are to be regarded as universal or as peculiar to the culture which we are studying. More comparative material is needed to answer this question finally. However, one can fairly safely say already that the appeal of the supernatural is a universal trait of the human mind, while the appeal to self-enhancement, and still more so the appeal to historical appreciation, differs in effectiveness in different cultures; it is eminently typical of Nupe culture.
14 0 POLITICA L HISTORY OF NUPE KINGDO M
Song of Etsu Bello
Th e name of Alla h is the beginning of all learning.
Le t us speak of the day whe n Etsu Bello bought a motor-car.
On the day whe n Etsu Bello bought a motor-car,
Th e whole of Bida wen t out to buil d a road.
Younger4 brother and elder brother, they wen t out to buil d the road. Al l the people on the farms wen t out to buil d the road.
Th e young bride, she went to buil d the road. Th e bridegroom, he went to buil d the road. Bu t wh y di d they all go out to buil d the road ?
It was because Etsu Bello had bought a motor-car.
Because of that all people wen t out to buil d a road.
The y said: Le t us build a road on whic h the car can travel.
The n the ma n wh o wants to go to Baddeggi, in this car he wil l go.
An d Zungeru , the ma n who wants to go to Zungeru , in this car he wil l go. An d Wuya , wh o wil l travel t o Wuya , i n this car h e wil l go.
Th e ma n wh o wil l go to Jima, in this car he wil l go. Th e ma n wh o wil l go to Kacha, in this car he wil l go.
Th e ma n wh o wil l travel beyond the river, in this car he wil l go. Thu s he wil l go, they said, and therefore let us make the road. Etsu Bello has the money to bu y the car,
Bu t the car wil l benefit all the people of Bida, fro m the farms and fro m the
city,
Th e great ma n and the servant.
The y thank Etsu Bello because he bought the car.
Etsu Bello, wh o has horses, and wh o now has a car,
Etsu Bello can say: Thoug h the horse may break his legs,
We shall go all the same.
On the day whe n Etsu Bello bought the motor-car, Th e bottles of the glass-makers turne d into red beads. An d the kernels of the gombara grass became necklaces.1
In that year whe n Etsu Bello bought the motor-car, Th e me n were beating the ground,
An d the woman , she was sitting at home.
Ho w di d it happen that the woma n was sitting at home ?
Of old it was the woma n wh o was beating the ground,
An d the ma n he just stood and watched.
Bu t because of this motor-car of Etsu Bello,
Al l the me n were beating the ground, and that is that.2
An d all the people, the whole of Bida,
The y were saying their thanks to mace da ciki.3
An d all Bida was giving thanks to karan giya,4
Wh o has sold the car to Etsu Bello.
1 i.e. all the people who went out to see the motor-car were dressed in their most beautiful dress, and the glass-makers of Bida were busy manufacturing ornaments for the occasion.
2 By ‘beating the ground* is meant the levelling of the ground in the road
building. Th e same term is also used for the levelling of the floor in the Nupe huts, a work which is reserved for the women. Therefore this comment on the change of times.
3 Mace da ciki (Hausa, pregnant woman), the rather irreverent nickname the
Bida people had for the District Officer at that time.
4 Karan giya (Hausa, a grass with prickles—Pennisetum), another nickname for the District Officer at that time.
POLITICA L HISTORY OF NUPE KINGDO M 141
An d all the people were giving thanks to Etsu Bello, Because he gave his money to buy a motor-car,
Tha t motor-car that wil l become a thing of benefit
To all the human beings.
In myt h and religion the ‘ bindin g forces’ of the kingdom gather them• selves to fullest strength. Th e Etsu Nupe is not, as we have seen, the religious as wel l as the political leader of his people; unlike the village chief, he cannot even claim kinship links wit h the priests of his country. Bu t Nupe kingship is linked in so many ways wit h religious usages and forms of ritua l symbolism that they inevitably become instruments of political solidarity. Rites and beliefs are focused upon the person of the king. Appeals for the help of supernatural agencies become appeals to agencies more or less clearly symbolic of the political system. Th e welfare and continuity of the community, unthinkable withou t these supernatural safeguards, become unthinkable withou t the state.
Out of the weak and irregular uniformit y of beliefs and rituals and out of disconnected local cults and traditions the kingdom thus creates a new religious community. One of the centres of this religious unification is that ‘mythical charter’ whic h made the dynasty of Nupe kings descen• dants of the mythical culture-hero, Tsoede. Relics of Tsoede, regalia, instruments of ordeal, magical emblems, whic h exist in every tow n of the Emirate, are the symbols of this ancient sacred pact. Th e name and tradition of Tsoede is also associated wit h cults whic h secure childbirth , fertility, and well-being, wit h certain rituals which prevent floods of the Niger, drive away disease, or brin g punishment upon a criminal.1 To take one example only. In Mokw a exists a certain ritua l called Tsoede or sara (Hausa, lit . malicious accusation, and also snake bite ; refers here to the use of the ritua l as ordeal), whic h is practised as ordeal, or whenever trouble befalls the community. It is performed by the Sheshi, that ‘king • maker’ of who m we have heard before; he places the short ‘chain of Tsoede’, the insignia of his office, on the ground in the sacred grove, makes a sacrifice of beer and fowl, and speaks thus:
God, we have come.
God, everybody has come. Thi s ancient thing of Tsoede Whic h is lying on the ground, It says we should bring beer.
Because of the old saying which we have heard, Because of that we brin g beer and fowl.
Tsoede, make the whole town prosper [or whatever is the special aim of the invocation].
A powerful secret society, organized to combat witchcraft, the order of the ndako gboyd, became a kuti ‘tsu, a King’ s Cult , only half a century ago. There is no doubt that the ndakogboyd cult itself existed long before that. Bu t under Etsu Masaba the head of this order, the grandfather of
1 This latter cult I have described in Man 1935, 143. The association of
Tsoede with the main fertility cult of Nupe, the gunrtu, is discussed in J.R.A.I.,
1937, P. 118.
142 POLITICA L HISTORY OF NUPE KINGDO M
the present head of the ndakogboyd society, was invested wit h the pom • pous titl e of Map Dodo, Master of the Monster, and wit h this royal recognition the organization of the society was made subservient to Nupe kingship. One after the other the heads of the society received their titles from the Etsu Nupe. Villages and districts whic h sought the help of the ndako gboya against witches had to apply first to the king. Thi s new royal prerogative also entailed a considerable economic benefit: for the king shared in the income of the order whic h was derived fro m the gifts collected in the villages whic h had called in the ndako gboyd.1 But besides tapping a new source of income the co-operation between Etsu and anti-witchcraft society meant adding to the centralized temporal jurisdiction the power of spiritual, mystic jurisdiction over the supernatural enemies of society. Thi s powerful prerogative of the king was, i n fact, maintaine4 by the Mohammedan Fulani Emirs til l very recently.2
Every performance of these various rituals and cults in the villages and towns of Nupe—and many are stil l performed every year, long after the disappearance of the Tsoede dynasty—revived and perpetuated the consciousness of this unity whic h has grown into far more than an external bond of political unification. Wit h the Islamization of the country many of these factors lost their power. Th e Fulani rulers could use the secret organization of the ndakogboyd; they could seize by force, as they actually did , the royal insignia from the last Nupe king. The y could not transfer the authority of the former ‘mythical charter’ to their own alien rulership. Bu t they could introduce another, younger, and perhaps more powerful uniting force, Islam—Islam as creed and Islam as education. Mosques, a Liman and Mallams in every village, these are the instruments of propaganda.3 An d in every religious service, in every babbling little Malla m school this propaganda is enacted. Teaching Islam means teaching the holy cause which sent the conquerors down to this country. Praying to Alla h means praying to the Go d of the powerful and sharing wit h them, if not their power, their beliefs. But Moham • medanism even becomes a real means of social approach between rulers and ruled. In a country still to a large extent pagan the adoption of Mohammedanism creates something like an intermediary social stratum, placed nearer to the rulers than the rest of the population. Formerly it involved—at least in theory—certain concrete benefits: an increased measure of security, above all, immunit y from slave-raids (as the Koran forbids the enslaving of Mohammedans). Thi s rule, however, was never strictly obeyed—unless the religious qualification was supported by the
1 Th e ndako gboyd society soon became a means of ruthless exploitation. Th e king and the members of the society were extorting money from the peasants all over the country on the grounds of ridding their villages of witches. This became so intolerable that eventually the Administration of Bida Emirate had to forbid all performances of the ndako gboyd. (For a fuller description see Africa, 1935,4.)
2 It is interesting to note that when the independent Emirate of Patigi was
created by the Government this new Nupe ruler appointed a second Maji Dodo
for his country.
3 Mokwa, a village of 2,000 inhabitants, possesses eight mosques, one of them built in 1934 during my stay in the village.
POLITICA L HISTORY OF NUPE KINGDO M 143 more solid one of political protection. Attachment to the rulin g
group thus fused wit h conversion to its religion. In the early days of
Fulani rule hundreds of pagans are said to have flocked into the capital
to seek out an important person as their patron and to declare to hi m
—and not to the religious officers of Islam—their readiness to em•
brace Mohammedanism. The y received their turban and sword,
insignia of clientship and new creed in one. Of such men the Nupe
say not ‘they become Mohammedans’ but ‘a ze goi%V—’they become
Fulani*.
Whatever the true essence of Mohammedan faith may be in Nupe, and however sparsely Mohammedanism has so far spread, Fulani ruler- ship di d create, and is creating, another ‘mythical charter’ of the king • dom. It tries again, wit h its new means, to moul d the religious elements into the intensive uni t almost of a State Church.1 Th e effect of Islam as a ‘bindin g force’ of the precarious political system that the Fulani had erected cannot be doubted. A striking example is that under Etsu Umar u the whole Trans-Kaduna revolted, refused to pay the tribute, and finally made war on Bida. Th e rebellious peasants crossed the Kaduna river, marched against Bida and almost reached it , being defeated at a place called Bidakoko, a few miles outside.2 Th e whole Trans-Kaduna, that is, wit h the exception of one town—the strongly Mohammedanized Dabba, the same town where Malla m Dendo started his proselytizing work in Nupe, and whic h now refused to joi n in the war against the rulers wit h who m they shared their creed.
Th e identification of State and Church is demonstrated in the annual Mohammedan festival, the Sallah, which is as much a pageant of the might and splendour of the kingdom as a religious feast. There are two Sallah feasts, the Small Sallah and the Great Sallah, separated by a tw o months’ interval.3 Thei r celebration may be called the climax of the whole year’s cycle: the two feasts give the months in which they fall their names—etswa salagi and etswa salako, ‘mont h of the Small Sallah1 and
‘mont h of the Great Sallah1; and to these the people look forward and look back, in the long-drawn preparations and anticipation, and in recol• lections and memories. Th e Sallah feast is like a Friday celebration on an immensely enlarged scale. On the first day the kin g on horseback under the state umbrella and surrounded by his noblemen and state officials, leads the vast procession to the praying-ground outside the town , where the Litnan holds the service. After the service the proces• sion returns to the royal residence. Here a great display of horseman• ship takes place, at which the bodyguard of the king, his police, and officers of state parade before hi m and his guests of honour—the fore• most among who m are the Distric t Officer and all the Europeans livin g
1 See C. K. Meek, The Northern Tribes of Nigeria, 1925: ‘Islam . . . has con• verted isolated pagan groups into nations; . . . it is largely due to Islam that numbers of tribes are now, before our eyes, forming themselves into nations’ (ii, pp. 4, 10).
2 It is said that this name Bidakoko was later given to the place of the battle,
because koko a tu Bida o, ‘almost they (the rebellious army) had reached Bida*.
3 Th e Nigerian Sallah is the Beiram or Id of other Mohammedan countries, which feasts are also celebrated twice, as ‘Little Beiram’ and ‘Great Beirani.
144 POLITICA L HISTORY OF NUPE KINGDO M
in Bida Emirate. On the second day of the Sallah the Etsu holds the great nkd of the year, a reception at which everyone who holds rank and title and every person of consequence comes to salute his sovereign, to offer gifts of loyalty and to receive counter-gifts expressing the king’s favour. For two days tens of thousands of people crowd the streets of Bida and assemble in front of the royal residence—Bida people as well as peasants from the villages who flock to the capital to see the Sallah. There they watch their rulers ride past in all their splendour, equipped with their paraphernalia of rank; they listen to music, drum signals, and songs of praise; they join in the general rejoicing, made more joyful by the largesse and gifts of food and kola-nuts which a gracious sovereign causes to be distributed among his loyal subjects. Here, then, it fails to the elements of worship and secular display, inextricably fused, and supported by the stimulus of mass experience, to mobilize periodically sentiments of loyalty and the consciousness of unity.
From these manifold sources springs that common mental and moral orientation which we call national solidarity and loyalty to throne and country. Here are born the sentiments of unity which, in the words of Lowie, ‘flourish without compulsion’. They crystallize largely round the figure of the king, thus giving substance to the self-description of the people of Nupe kingdom as ‘king’s men’. Yet the king is not simply the state—the factors checking his authority, his varying powers, show clearly that he is only an organ of the state. Nor is he a symbol of the power and unity of the state in the same concrete and personal sense in which the village chief can be regarded as a symbol of his community. Kings have been deposed and exiled; they have to fight and intrigue for their position; they are subject to the openly working balance of power. All this is well known—too well known for the identification of the person of the king with the unity of the society. Loyalty to a personal king of Nupe is factional; solidarity and loyalty on a national scale rallies round the throne and round the abstract principle of kingship rather than round the actual king.1
But the /binding forces’ are not wholly pivoted on kingship. They yield also a commonness referring to the country as such, the people as such, to a largely abstract and timeless collective identity which no longer needs the sustaining symbol of kingship. These ‘binding forces’ implant in the minds of the people of the state the realization that, over and above tribal sections, cultural and class divisions, they form a single human material, living an identical life. They teach a new meaning of the word Nupe, which is the meaning of nation. Collective pride, otherwise vested in the tribe or the local community, and its counterpart, the aloof or critical attitude towards other groups and their culture, fuses with and is supplanted by the sentiments of national pride—patriotism. Co-ordination of behaviour loses its aspect of a dictated order sanctioned by force, and becomes indis-
1 It is possible that this was less so in the pre-Fulani kingdom, with its un• divided rulership, invested with ritual functions and based on a unique, auto• matic, system of succession. Here kingship may well have been accepted in a much more concrete and personal sense as the symbol and pivot of political unity.
ETSU WIT H BODYGUARD IN SALLAH PROCESSION
DISPLAY OF HORSEMANSHIP DURING SALLAH
POLITICA L HISTORY OF NUPE KINGDO M 145 tinguishable from the self-willed conformity born of a closely kni t
community life. Th e state itself becomes that twin-creation as whic h
Maclve r conceives it , and, wit h one side of its being, a community.
Th e process is not yet complete. Existing social divisions and diver• gent ‘interest units* are not wholl y submerged in the new consciousness of unity . Sociologically, the incompleteness of the development is visible in the reluctance wit h whic h certain sections, especially among the peasantry, respond to the cultural reorientation. Psychologically, it manifests itself in the ambiguous attitude of many individuals in the lower classes, vacillating between bitter resentment against the rulin g class and readiness to identify themselves wit h its glory, of whic h we have seen examples. It is an attitude familiar in many societies similarly split by sharp distinctions of political and social inequality.1 One might doubt whether this process of intensive unification wil l ever be com• pleted now, wit h new influences coming in , new cultural and class divisions being formed whic h disturb the steady growth and blu r the old border-lines. I do not myself thin k that this wil l be so. Bu t in a wider sense the growth of the Nupe nation, even if it were to continue undis• turbed, woul d never be complete: for the nation is a dynamic, not a static, unit , a uni t whic h must constantly readjust itself and constantly absorb new ‘raw material’—new generations, newly-formed classes, strata, groups.
Th e achieved integration of the state is then in some measure, like that of the village community, the integration of a ‘multiple system of control*. Yet the state and the village community differ not only in the intensity and completeness but also in the quality of the integration whic h they achieve. We noticed that the ‘bindin g forces’ of the state appear in varying degrees of concreteness, from actual co-operation to moral orientation, and that those of most abstract appeal were the really effective ones. In this, I believe, lies the essential difference between the integra• tive mechanism operating in the village community and this new mechanism of creating nation-wide unity. Evidently, the very difference in size and numbers must play a decisive part. In the village all in • tegrative agencies can address themselves to an identical group sub• stratum, and can envelop it in the network of actual collaboration. Th e state, wit h its huge human substratum, its subdivisions and sections, precludes the ful l effectiveness of such direct and concrete integration. Direct contact and direct co-adaptation of behaviour are subordinated to indirect methods of bringing about conformity, through ideological per• suasion. Actual concerted action gives way to efforts at securing parallel action, and ‘common’ interests in the strict sense are replaced by
‘identical’ interests. Instead of communal rites at whic h groups that in other contexts act as political units are welded together by the influence of religious experience, we have the spreading of common beliefs and practices designed to cal l fort h identity of outlook and to foster, though not an identical group, a group conscious of identity.
Ultimatel y and rather surprisingly, then, the binding forces of the
1 M. T. If erskovits has called it, by a suggestive term, ‘socialized ambivalence’
(Life in a Haitian Valley, p. 295).
L
14 6 POLITICA L HISTORY O F NUPE KINGDO M
state can be reduced to the fact of cultural assimilation. Yet it is not that random assimilation whic h is incidental to the expanding state, and whic h only superficially obscures social division, achieving but an in • secure, external, likeness. Thi s assimilation is designed to carry the foci of cohesion and solidarity—ideologies, and the institutions by whic h they are sustained. I said * designed’. Bu t how far are we justified in regarding this assimilation and the incentives behind it as deliberately conceived and employed by the political system and its agents ? Certain practices such as the encouragement given to Mallams proselytizing among the pagans, the reward of increased security held out to converts, or the engagement of musicians and poets to sing the praises of the king and the kingdom, represent actual ‘propaganda’, stil l carried out under our eyes. Others, such as the supplanting of a local cult by a’ king’s ritual’ , must have amounted to the same in former times (i n the case of the royal interest in the ndako gboyd society it can be proved to have been so only two generations ago). Other forms of persuasion, again, may have worked through the unconscious mechanism of habits and views formed imperceptibly. Often their agents are men who have been in the Fulani wars or who have in some other way been associated wit h the rulin g group and then returned to their villages, where their stories and memories aid the other, institutionalized, * propaganda’ of the kingdom. In a wider sense, however, they all are deliberately conceived and em• ployed : they are agencies of an assimilation conscious of itself and of the message whic h it carries.
CHAPTE R I X
THE POSITION OF THE WOMEN
RANKS AND OFFICES
T speaks for the consistency of the political system of Nupe that the position of the women, too, should be drawn into its orbit and even into the framework of official grades and ranks. Amon g the women of royal blood in pre-Fulani Nupe three women’s ranks existed: Nimwoye, Sagi, and Wogbo. Precedence and succession in these three ranks corresponded to the succession in the three highest gitsti titles. At the court of ancient Nupe the titles were bestowed by the kin g on his mother or father’s sister, elder sister, or daughter. Th e holder of the titl e had to be a nytzagt gbako, a ‘strong woman’, that is, a woman of about 40-50 years. She had to be unmarried, or else her marriage woul d be formally dissolved on her appointment to the rank, and she woul d forthwit h be regarded as unmarried. She lived, not in the house of her husband or father but in the Etsu’s own house. Th e Fulani adopted the first two ranks, abandoning, however, the condition of celibacy or nominal celibacy, and made the two daughters of Manko the first Sagi and Nimwoye. Th e two ranks have remained in the royal house of Nupe since then. After the split of the dynasty into three branches the women’s ranks followed the new dynastic division, the titl e Nimwoye being assigned to the women of the royal house of Masaba, and the title Sagi being taken, in rotation, by the women of the royal houses of Umar u and Usman. Thi s rule stil l holds good : the present Sagi, for example, is a sister of the late Etsu Bello of the house of Usman, and the Sagi
before her was an elder sister of the Shaba, of the house of Umaru .
Th e women rank-holders occupied an extremely influential position at the court. The y took part in the king’s council, they could joi n in the war wit h their own troops of slaves and serfs, they held fiefs and owned land. Theirs was the position of ‘kings over the women of Nupe’ . Althoug h these political prerogatives have disappeared, the position of authority vested in these women’s ranks is unimpaired. Al l quarrels between women, especially women of the noble families, are brought before them ; the women of Bida and the villages send them presents and receive counter-presents from them, much as is the custom between the noblemen of Nupe and their king . Sagi and Nimwoye are often the official representatives of the court at the family feasts of the Bida nobility. At the name-giving ceremony which was celebrated in the house of the Shaba in 1936 the Sagi came to brin g presents and con• gratulations from the Etsu—an honour whic h was duly appreciated by all who were present.
Amon g the Bida women of low birth , the wives and daughters of commoners, we find another women’s rank, Sony a.1 She is, as the Nupe
1 Th e Nupe word Sonya corresponds to the Hausa word (and the equivalent women’s rank) saraunyia, lit. queen. So or So is a Nupe root meaning head or chief.
148 TH E POSITION OF TH E WOMEN
pu t it , the ‘Sagi of the poor’; what the royal princesses do for the women of the nobility the Sony a does for the women of her own class: she advises them and arbitrates in their quarrels, taking, however, the more serious cases to the Sagi. Th e Sonya might almost be called a female officer of state. Al l large-scale women’s work that is done in Bida by order of the king is organized and supervised by her—for example, the beating of the floors in the houses of the king and the royal nobility: Th e Sonya is, above all, entrusted wit h the supervision of the market. She is elected by the Bida women who are regular traders on the market, and her rank is confirmed by the Etsu. Th e respect and deference wit h whic h all Bida women greet their Sonya illustrates the importance of her position. Formerly she was a royal fief-holder on a small scale, holding land of her own ; she was also allowed to levy a small due on all business transacted on the market. Of this levy the Sonya returned one-third to the Sagi or Nimwoye as a symbol of her submission to the highest authority of these royal princesses over the women of Bida. To-day, the Sonya is stil l supervising the market and the work of the women. She has kept her status and her official position as well : for she has been made an official of the Native Administration and granted a small salary.
In every Nupe village, finally, we find a title d ‘head of the women’, called Sagi, like the royal princess in Bida. She is usually a woman about or over 50. She is elected by the married women of the village and her appointment is confirmed by the chief. In larger villages there may be a Shaba Sagi as well, an appointed successor to the headship of the women. Th e title Sagi itself seems to be of comparatively recent origin and was probably, like many other village titles, copied from the Bida rank system. Th e original title which it replaced was Lelu. Thi s title, though no longer used, was explained to me in most villages to be the exact equivalent of the modern title Sagi. Yet this is not quite true: wit h the change in the title a subtle change in the significance of the rank has taken place. As I have described elsewhere, the Nupe chiefs used to bestow the rank Lelu on the woman who was believed to be the most powerful witc h in the village. Her secret knowledge could thus be made to benefit the interests of the community. Turne d into an organ of village administration, this woman could be used, in the opinion of the people, to check and control the subversive activities of the other witches.1 Wit h regard to witchcraft, then, the Nupe have adopted the proverbial policy of making a (reformed) criminal chief of police, or rather an imaginary criminal controller of an imaginary crime. Thei r belief in the efficiency of this measure has not changed. Yet the curtailed power of the present-day village chief, who is no longer free to act upon the advice of his female ‘controller of witchcraft’; the general fear in the country, born of the experiences in the witchcraft tria l in 1932, even so much as to mention the wor d witc h or witchcraft; and in certain places also the growing importance of other more prosaic business for the head of the women, have all contributed to suppress the magical aspect
1 See Africa, 1935, 4.
TH E POSITION OF TH E WOME N 149 of the office of the village Sagi and to emphasize its harmless business
side instead.
Thu s we find that in many Nupe villages an efficient business-like head of the women has succeeded the Lelu or Sagi of the old ty£>e. In other places, again, the people woul d not admit the magic significance of the woman’s rank. Yet the strength of the belief, concealed though it may be, was shown in a witchcraft case whic h I happened to witness in Gbajibo, a small village on the Niger above Jebba. Here the Sagi, the very business-like head of the local women, was yet behind the public outcry against an alleged witch . Th e women much more than the men were leading in this persecution; they had been the first to ostracize the victim ; and when I questioned them about their motives, they all referred me to the Sagi. He r own motive appeared to be jealousy of the other woman, the alleged witch , in who m she saw a dangerous rival , possibly even for the position of a Sagi,
Th e office of the Sagi, modern or traditional, is invariably a much – coveted one. Her moral hold over the women, as our example has shown, is complete. It is equally strong in the less sinister spheres of communal life : like the Sonya in Bida, the village Sagi advises the women, organizes women’s work, arbitrates between them, and receives small presents from every business transaction which they carry through.
MARRIAGE AND STATUS
Th e ranks and offices already discussed all belong to married women. But we have seen that they represent forms of status which the women acquire in their own right, on the grounds of descent, or personal ability, or both, independent of their status as wives of men who occupy this or that social position. We shall see later that the same principle applies also to the status won by wealth. In particular the status by descent of the Nupe women is affected but littl e by marriage, i.e. by the social status of their husbands; in so far as it is so affected, this change is in the direction of social advancement. Women of low birt h who marry above their status into the aristocracy can claim the status of their husbands; and women who marry beneath their station do not lose their status but raise their husbands to their own social level. Th e men wil l be allowed to share most of the social and political privileges of the class into whic h they marry. Tw o cases were reported to me in which the men could even acquire saraki ranks after marrying a woman of the saraki nobility. But even if the men are not admitted fully into the title d class, their children woul d always acquire the status of their mother. Daughters may thus inherit the status of their fathers, and bot h sons and daughters that of their mothers; husbands may be adopted into the social sphere of their wives and parents-in-law. Thi s social regulation may seem surprising in an otherwise strongly patrilineal society. But this one-sided interpretation of the change of status that goes wit h marriage is not an expression of matrilineal kin • ship tendencies, nor of ‘feminist’ leanings in Nupe society (though we shall find ample evidence of these, too); rather it expresses political
15 0 TH E POSITION O F TH E WOME N
interests, deep-seated class sentiments, which cannot admit of a lowering of status in any one high-born. Yet if , owing to the rigi d class System, marriage between the classes is thus turned into a potential instrument of social advancement, its utilization in this sense is also rigidl y controlled.
For marriage in the aristocracy, especially in the royal nobility , is to a very large extent endogamous—’family marriage’ as the Nupe call it . As the Nupe-Fulani regard themselves all as related, and as most families of the nobilit y are more or less closely related by marriage, practically all
- family marriages’ of this kin d are at the same time marriages withi n the rulin g class. Th e Fulani in Nupe, moreover, following Mohammedan practice, encourage parallel cousin marriage (which is forbidden by the traditional marriage rules of the country) as wel l as cross-cousin marriage, and thus widen enormously the scope of ‘family marriages’, especially between descendants in the male line. Family marriage in Nupe, finally, is characterized by a much lower bride-price than that paid in ordinary exogamous marriage. Thi s leads to the rather para• doxical situation, on whic h my informants invariably commented, that the wealthiest and proudest classes of the country favour ‘cheap* marriage, while the humble commoners and peasants are committed mostly to a very high bride-price. Th e explanation is that the low bride- price in this case supplies an additional incentive for endogamous marriage—’class endogamous’ marriage, as it were—and thus for the preservation of the social identity of the rulin g classes (as wel l as, in • cidentally, a strengthening of their group solidarity).1
Indeed the counterpart of the incentive of a low bride-price in ‘class endogamous* marriages is the ‘deterrent’ of an exceptionally high bride- price demanded in the case of marriages between the classes, i.e. between men of the tdlaka class and girls or women of the nobility . Such cases are accordingly rare. Yet we may note that if a chance of such marriage exists, if the gir l loved the man sufficiently to overlook the difference in social status, the high bride-price alone proves no obstacle. For friends and relations woul d joi n forces to find the money and to enable the ambitious suitor to make his luck by this advantageous marriage. Th e benefits of his new position woul d be considered sufficient promise of an eventual ample gratification. One of my Bida servants was an example: the son of a commoner, he had succeeded in winnin g the love of the daughter of a very high saraki and eventually, after considerable difficul• ties, in marrying her wit h money borrowed from friends and relations. Th e endogamous marriage rules only ensure that the women of the rulin g classes should marry withi n the class; they do not attempt to exclude the less dangerous for m of dilutio n implied in the marriages of the men to women from other classes and strata. Indeed, the poly• gamous nature of marriage in Nupe, whic h forces the men to seek wives
1 One of my informants who belonged to the saraki class even went so far as to explain the whole principle of having a different bride-price for ‘family’ and ‘out• side1 marriages as a political measure taken by the saraki and Fulani nobility with the purpose of preventing commoners from marrying into their class. Though incorrect as regards the facts, this interpretation is significant as an expression of the dominant class sentiments.
TH E POSITION OF TH E WOME N 151 from outside their social group, precludes an application of ‘class endo•
gamy* in this sense. Marriage of this kin d between the strata has reached
enormous dimensions. An d this shows, incidentally, that the rule of
‘class endogamy’ serves, even in the case of the Fulani rulers, the interest
only of a social identity and solidarity and not, as one migh t perhaps
assume, of racial purity .
POLYGAMY AND SEX MORALITY
In Nupe as in most polygamous societies in whic h marriage is by bride-price (and by hig h bride-price at that) the number of wives a man possesses becomes an infallible index of wealth and status. In the peasant districts, where we find comparatively littl e inequality of wealth, the range of polygamy also varies in narrow limits . Genealogies whic h I collected in the two Nupe villages Doko and Kutig i show the following range of polygamy (the figures refer to middle-aged or ol d men, both of the livin g and previous generation):
Number of wives
1 2 3 4 5 Over 5 Total
Number of cases 68 2 0 1 1 1 1 179
Th e average is thus 2.1 wives per man.1 To quote a similar average for Bida woul d be meaningless: the great inequality of wealth and status is reflected in the widely varying range of polygamy. In the houses of the talakazi, the ‘poor ones’, monogamy is the rule ; in the ‘middl e class’ the conditions are much the same as in the peasant districts; but as regards the men of substance and rank, though it woul d be correct to say that they have rarely less than four wives, the upper limits of their polygamy are very fluid , and cannot even be ascertained wit h any accuracy. Popular notions attribute to certain royal princes and, above all, to the kings of Nupe, numbers of wives varying between 100 and 200. Even the late Etsu Saidu is reported to have had sixty wives. Discretion for• bade any direct inquiries. Besides, the vagueness in the legal definition of ‘wife ‘ and ‘marriage’ in these cases woul d have detracted from the value of the information. According to Mohammedan law a man may have four wives only. But this rule is easily circumvented by limitin g the number of legal wives who m one marries wit h religious rites to four (or four at a time), and relegating the rest to the position of mere con• cubines. In the houses of the higher as wel l as of lesser nobilit y I found
1 Th e range of polygamy varies to some extent locally. In Mokwa, for example, I obtained the following figures:
Number of wives
r 2 3 4 ! 5 Over 5 Total
Number of cases 44 28 3 3 60
Th e average is, again, 2 wives per man.
15 2 TH E POSITION O F TH E WOME N
the practice of divorcing ageing wives and replacing the m wit h more attractive girls. Th e ex-wife, especially if she had borne children, woul d be allowed to stay on in the house and woul d continue to be supported by the husband, or ex-husband—a wife on pension, as it were. Th e late Etsu of Agaie was partial to a different method: he used to marry off his pensioned’ wives, at very low bride-price, to talaka£i and peasants, who were either glad to find a ‘cheap’ wife or else afraid to refuse so exalted a gift. Amon g the upper classes of Bida ol d men, long past marrying age, woul d stil l add young girls to the number of their ‘wives’. Aphro • disiacs (which one can bu y from the barber-doctors in Bida) and stimulants (above all, kola-nuts) keep alive their sexual appetite. Ol d S., who died in 1935, 70 years old, had married a new young wife only the previous year, and a son was born to hi m (‘to him ‘ at least in the legal sense) posthumously.
Thi s classbound polygamy deeply affects the sex morality of the women and, indirectly, of the society at large. Respect for chastity and the sanctity of matrimonial ties dominates the traditional moral con• cepts of Nupe. But this ‘double morality’ of a social system whic h legitimizes unrestricted sexual desire of the men in one privileged stratum, but woul d force similar desires in other strata into the realm of illegitimate and indictable action, is in danger of defeating itself. In Bida it has clearly provided a strong stimulus to extra-marital sex rela• tions in all classes. Moreover, the starved sexuality of the women in the large polygamous households strives to find an outlet.1 Th e insecurity of their legal position tends to remove the barriers of conventional matri • monial morality. Thu s many of the younger wives or concubines seek satisfaction in illegitimate relationships and even semi-professional prostitution.
Adultery , including adultery in this extreme form of semi-profes• sional prostitution, is extremely widespread among women of every status. Prostitution in the strict sense (practised as a profession and in brothels) does not occur in Bida. But we have heard that the kola-nut vendors on Bida night-market, all married women, are prostitutes at the same time.2 We shall learn later of the close association between women’s trade and prostitution (and the large majority of Bida women are traders of one kin d or another). ‘Every woman’, an informant tol d me,
‘who m you see beautifully dressed in the streets of Bida after dark, is a “bad ” woman.’ Informants, themselves married men, were all agreed that there were hardly any ‘good’ women left in Bida. But when I asked them what the husbands were doing about this they said: ‘Nothing ; what could we do ?’ Indeed, they can do little but shut their eyes. Th e obvious alternative, divorce, is impossible in most cases. Poor men cannot afford to divorce their wives, for this woul d mean forfeiting part of the bride-price and thus their only chance of finding another wife ; moreover, the economic dependence of many husbands on their wives (of which we shall hear more in a later chapter) forbids such a course.
1 In the royal ‘harems’ homosexual practices—called madigo or lutu (the
Arabic word for homosexuality)—are, in fact, said to occur.
2 Thei r ‘tariff’ is commonly known: it varies between 6d. and 1s.
TH E POSITION OF TH E WOME N 153
Me n who have to uphold their prestige and social position woul d never risk the publicity and ridicule of divorce proceedings in whic h they woul d have to appear in the role of the wronged husband.1 Th e amorous affairs of Bida noblemen supply, in fact, an ever-welcome topic for the tow n gossips. Whe n some years ago two popular noblemen quarrelled, everyone knew that it was about a gir l and had the details at their finger• tips. Onl y the men of very great wealth and highest status are in a posi• tio n to restrain or threaten their unfaithful wives. Fo r one thing , they can afford to forfeit more than one bride-price, and can thus divorce their wives (according to Nupe Mohammedan law) withou t invoking elaborate public proceedings. Besides, wives and paramours woul d hesitate to provoke the enmity of these powerful personages. It is not unnatural, then, that (as my informants said) the wives of these men supply the only example of unimpaired marital fidelity in Bida—another no less striking instance of * double morality’.
If the older married men are either resigned or complacent, the younger men see in illegitimate adventures only an exciting and tempt• ing aspect of Bida life. We shall meet later wit h their ideal, the dashing young man ‘about town ‘ who can ‘have’ all the women he wants and can boast of his conquests. Th e women who avails herself of the loosened marriage ties may be talked about and called (behind her back) a ‘bad woman’—nyizagi banza, a ‘worthless woman’. Bu t in their ordinary behaviour towards these women, when meeting them in the street or i n the house of a mutual friend, the men wil l never fail i n their respect towards them or betray by a sign their moral prejudices. Thi s ambiguous attitude is symptomatic of a society which , though stil l con• scious of its traditional dogmatic appreciation of chastity, is yet adapting itself to a spreading laxity of morals.
There is indeed evidence to show that sexual licence has been greatly increasing in the last generation or two. Bu t there is also evidence of the fact that sexual licence on a smaller scale and withi n strictly defined limit s enjoyed a legitimate existence in traditional Nupe society. Th e anomaly of public morals whic h woul d condemn sexual licence yet at the same time sanction it in a certain well-defined social domain is familiar to the student of society. In Nupe as in all other societies this qualified admission of sexual licence is due to various social causes. A first cause lies in the nature of sex itself, namely, in the reaction to the restraint on all pre-marital and extra-marital sexual relations whic h characterizes the traditional sex morality of the people. Another cause is the large floating population in Bida, composed of traders, warriors (formerly), labourers (to-day), that is, men who come to Bida withou t their wives and depend on prostitution for the satisfaction of their sexual appetite. A thir d cause is embodied in the institution of female trade. We shall see later that women’s trade involves a loosening of matrimonial and family ties and tends to become identified wit h licentious living . In female trade Nupe society thus has grasped a ready occupational framework for that domain of sanctioned
1 Clear evidence of this is the scarcity of divorce cases, which form two-thirds of all court, cases in Bida, from the houses of the nobility.
154 TH E POSITION OF TH E WOMEN
sexual licence towards whic h the other social forces appear to be driving . Th e result is the close association between professional female trade and semi-professional licensed prostitution whic h we can observe to-day.
Yet admission into this ‘ domain’ was in tur n made dependent upon a certain qualification, namely childlessness. Here we can add a final, ideological motive to the social causes behind the moral development. It is embodied in the Nupe conception of barrenness as a state outside conventional morality, and its contribution is twofold : it helps to define more narrowly the group in whic h sexual licence shall be sanctioned, by making sterility of the women a criterion of admission* and it provides at the same time the spiritual framework for such social sanction, by link • in g it wit h certain religious or quasi-religious beliefs. To some extent the ideological association between barrenness and sexual licence can be reduced to certain obvious rational motives, whic h are, in fact, present in the minds of the people: childlessness is of great advantage in the life of the women traders whose profession involves constant travelling and long absence from home; moreover, the sexual indiscretions of sterile women are socially of less consequence than those of other women, being comparatively safe from discovery and precluding the more dis• turbin g legal complications (i.e. birt h of illegitimate children). But the social interpretation of barrenness also expresses certain more deeply rooted beliefs. To the Nupe marriage has no real meaning withou t progeny. Barrenness is regarded as a great curse and misfortune, and many cults and forms of magic are designed to secure fertility of the women or to cure barrenness. Traditiona l religious beliefs place barren women in a category apart and exclude them from participation in the most important annual cult of the country, the gunnu.1 As the barren woman fails by the common standard of marriage and womanhood, she is also exempted from the standards of common morality. Adulter y and unchastity count less in her than in other women. An d although the stigma of immorality woul d stil l attach to the unchaste barren woman (she woul d stil l be called a ‘worthless woman’), it woul d be, as it were, overshadowed by and fused wit h the other paramount stigma of barrenness itself.
Th e growt h of sexual licence far beyond its traditionally sanctioned limit s is again due to various causes, economic, political, and religious. Th e primary cause appears to be economic and wil l be discussed i n that context. Its effectiveness, however, is greatly strengthened by the political and social factors whic h foster what I have termed the ‘double morality’ in Nupe society. We can also see that the Nupe conception of barrenness must itself indirectly aid these tendencies and defeat the limitations whic h it was meant to uphold . For in a country wit h spread• ing sterility and even birt h restriction, a religion which withholds its sanctions of matrimonial morality in the case of childless women can prove bu t a weak bulwark against the tide of sexual licence. Moham • medanism in Nupe, committed as it is to that ‘double morality*, under• mines rather than supplies forces of moral restraint.
1 For a description of this aspect of barrenness see J.R.A.I. 1937, p. iii .
TH E POSITION OF TH E WOME N 155
A comparison between sex morality in the tow n and in the villages wil l brin g the final proof of our interpretation of the social causes o f moral laxity and publicl y condoned sexual licence. For in the villages, where the political and economic factors to whic h we have attributed this development are muc h less conspicuous than in Bida, sexual licence in this sense is absent also. Th e same informants who declared that there were no ‘good women* left in Bida maintained that the opposite was true of the country. Withi n limits , this is correct. Semi-professional prostitution is absent and adultery is far less frequent than in Bida. Above all , it is not glossed over for fear of publicity and ridicule : it leads to quarrels and frequently to divorce. No r are illegitimate amorous adventures applauded and bandied about. Th e more closely kni t village life renders sexual promiscuity both more difficult and more risky, and the danger of provoking serious conflicts forces it into the sphere of clandestine and disreputable activities. Th e ideal of the Bida Lothario is absent in the Nupe village.
It is largely true to say that among the peasant women also the Bida ideal of the glamorous ‘fast* woman finds no following. Or, more correctly, it finds no following among the peasant’s wives and daughters who have chosen village life. For especially in the larger trading-places there is a certain (increasing) minorit y of women who leave the village to joi n the army of’loose’ women traders. However, they do not introduce their ways into village life itself. To quote again from my informants :
‘Peasant women are on the whole better wives than our women in Bida. Yo u see it when you come to a village: you wil l not find women wit h who m you’can sleep. But who would want to sleep wit h them? The y are badly dressed, they do not wash, use no scent, and smell of sweat. There are a few exceptions, though, for example the women in the house of the chief of K.’—referring to a village in whic h the chief’s family was i n many ways emulating the standards of Bida life. There is good sense in these cynical (though not cynically meant) remarks. Har d work, lack of leisure, the inability to afford nice dresses or the help of’cosmetics, make the peasant woman comparatively unattractive; she ages more quickly than her more fortunate sister in town . It is not surprising that she should come short of the sophisticated urban standards of taste. If the peasant woman thus succumbs less to temptation, it is also true that less temptation comes her way. Having chosen peasant life she can but choose peasant morality also. Th e different sexual morality in town and village is thus essentially an expression of differences in economic condi• tions and of inequality of wealth and status.
It is tempting to dwell on this relation between economic and political factors and the attitude of the society towards the problems of sex and morality. There has been the tendency in recent anthropological writings to regard the elaboration of the sex theme in different cultures, the temperamental and personality traits which , in a society, appear associ• ated wit h sex, as the expression of some ultimate quality in culture itself
—the genius or ethos of a culture. Tha t women in one society are
‘gentle’ and ‘aggressive’ in another, that one culture regards them as
masculine workers and another as feminine mothers or partners in love,
156 TH E POSITIO N O F TH E WOME N
is all due to the mysterious workings of that genius of culture.1 In a criticism of these views I have pointed out that the sex morality whic h a society makes its own and the whole ethos whic h its culture expresses are largely functions of the economic conditions prevailing in the society.2
Of this the preceding discussion has, I think , provided conclusive evidence.
1 M. Mead , Sex and Temperament, pp . 298, 310, and passim.
2 ‘Th e Typological Approach to Culture’ , in Character and Personality, 1937, p . 280 et seq.
CHAPTE R X
GOVERNMEN T I N MODER N NUPE
F T E N i n the foregoing description I have used the present tense as if the Nup e kingdo m of pre-Britis h days still existed. True , this
was to a large extent an historical chapter; yet it described a develop• men t the results of whic h are still vali d and still moul d society to-day as they moulde d it thirty , fifty, or even a hundre d years ago. I was not merely reconstructing dead events and past phases. Th e Nup e kingdo m and its society still exist, not only very vividl y in the memor y of the people bu t to a ver y large extent in concrete reality. Thi s is not denying the force of social changes, bu t stressing the tenacity of a social structure tha t survives although its contents have changed materially. Th e recent economic an d political changes have affected fundamentall y the
‘contents’ of the social life of the country, the mai n interests and in • centives roun d whic h it revolved. Ther e are no more wars, raids, and conquests; slaves and feudal fiefs exist no longer. Bu t ranks and titles and state offices linke d wit h ranks and titles, still exist. Slaves have turne d into bara or Nativ e Administratio n officials; fiefs into administra • tive districts in the charge of royal princes. Tha t the framewor k of the political organization could survive almost unchanged is in large measure due to the Administratio n of Nigeri a unde r Indirec t Rul e and its ten • dency to utilize as far as possible the traditiona l social system. Bu t a stronger proof of the tenacity of the Nup e social system is the fact that wher e the Administratio n had departed fro m i t and trie d t o make wa y for a new development the old tren d broke through , undeflected.
Tw o years ago the Administratio n buil t a ne w official residence for the Etsu Bida, wit h a brick wall , an imposing tower, a garage, and a flower garden—a n emble m of moder n Nigeria . Incidentally , for all its modernit y the new house is but a replica of the hereditary palace of the Nup e kings, wit h special katamba for the ena Ndeji and ena Maiyaki, for gitsu$i an d bara, th e ena manzi, the me n of the Etsu Dogari, and the commoners wh o come to wai t on the Etsu. Now , this ne w house was buil t on ‘neutral’ ground , outside the tow n gates, as a symbol of the Government’ s intentio n to break wit h the division into three royal houses, whic h the Governmen t considered harmfu l to the unit y of the country.1 Th e Administratio n believed itself to have overcome success• full y the opposition of the late Etsu. An old and sick man , he was reported to have said—and his words wer e all over Bida—tha t he had agreed to the new site only because he kne w he woul d not have to live in the new house. Whic h was tru e enough—he died before the buildin g was finished. Bu t the ne w Etsu, wh o willingl y took up his residence in the splendid new house, one day surprised the Distric t Officer by demandin g that his ‘proper house’ in Bida, in his ‘own ‘ part of the town , should be pu t in repair at the expense of the Nativ e
1 See p. 93.
158 GOVERNMENT I N MODER N NUPE
Administration. Ne w house or old, in his min d the position of the Emi r remained inseparable from that traditional personal lin k wit h the local tow n faction.
THE ADMINISTRATION OF THE EMIRATE
Le t me describe the modern Government of Nupe Emirate, in official nomenclature, the Native Administration (N.A . for short) of Bida Division . * Ou t of the multitude of traditional ranks the modern Emirate has taken a certain number and embodied them in the N.A . as official posts, carrying a salary and to some extent also implyin g specific, new tasks. Th e salary is graded in accordance wit h the importance of the office, and the Etsu himself heads this new bureaucracy wit h a salary of
£2,000.2 Th e Etsu Nupe is stil l elected from the ranks of the royal princes by gitsufo and sarakizi, or more precisely by those gitsuz’i and sarakizi who hol d appointments under the N.A . Th e Etsu, assisted by a council of four, looks after the affairs of the Emirate; he watches over law and order in the country, carries through measures of administration, tries certain legal cases, and controls appointments to offices and titles. Advised and guided by the Distric t Officer in charge of the Division , the Etsu is the responsible head of his government. Th e agreement of both Etsu and Distric t Officer is necessary for every administrative measure.3
Th e appointment of a new Etsu is subject to confirmation by the Governor of Nigeria, and in extreme cases the Governor may depose the Emi r on the advice of the Distric t Officer and his superior officer, the Resident of the Province. Th e Gobi Seidi, formerly the head- slave of the Etsu, is to-day the official messenger between the Ernies residence and the Divisional Office.
Th e Etsu’s council consists to-day of the following members: the Ndeji, Maiyaki, Ndgya, and Etsu Tafie. Th e Ndeji, the old’Father of the Town’ , is the ‘president* of the council. Th e Maiyaki, from a ‘Ministe r of War’, has become a ‘Ministe r of Public Works’ and a representative of the Etsu on tours of inspection. Th e Nagya, a royal prince, was formerly Distric t Head of Mokwa , and was then co-opted int o the council for the ability which he had shown in the administration of his district. Th e Etsu Tafie, finally, was taken into the council in 1931, after the Bida disturbances, as an official representative of the talakawa (this is the Hausa for m of talakaz’i, commoners), who formerly had no voice in the management of state affairs. I may say a few words about this appointment. Justified as was its aim, the appointment itself seems rather problematic. It was justified officially on the grounds that the
1 Th e Administration of the other Provinces and Divisions where sections of the Nupe tribe are living is organized on identical lines; we may disregard them here.
2 These were the salaries of the Etsu and the higher officials of the N.A . in
1934-6: Etsu £2,000; Ndeji, Maiyaki £180; Ndgya, Etsu Tafie £120; District
Heads £150; Chief Alkali of Bida £180.
3 No fixed rule can be laid down as regards the relative share of responsibility and initiative in this co-operation of District Officer and Emir. It varies con• siderably with the personalities of these two men, and also with the general trend of policy that might obtain in the Province at a given time.
GOVERNMENT IN MODERN NUPE 159
Etsu Tafie had been in pre-Fulani times the chief of the * aboriginal population* of Nupe, the Beni tribe1—a justification which , as this chapter has shown, takes into account, out of the whole complex social structure of the kingdom, but one comparatively irrelevant fact: an external historical connexion wit h one isolated village community. Typica l of what I have called the ‘parochial outlook* of the village, the Etsu Tafie, bearer of one of those ‘petty village titles’, having reached this new position through no acknowledged channel of ‘social career’, seems out of place in a council on whic h he could hardly hold his own against the traditional leaders of the country’s fate, the highest ranks of the Nupe hierarchy. Th e insignificant role which I saw hi m play on the council seems to support my view.
Bida Emirate is divided for the purpose of administration int o a number of districts; at the head of each is a District Head, appointed by the Etsu as a rule from the ranks of the higher royal nobility ; he himself, as Ndgenu, was formerly Distric t Head of the Jima-Doko district (see table, p. 95). Some time ago the number of districts was reduced from ten to seven by amalgamating certain districts, making the area in charge of one Distric t Head rather larger, and perhaps also more hetero• geneous, than migh t seem advisable. Here is the list of districts and rank-holders appointed as Distric t Heads.
List of District Heads
Former Districts Present Districts District Heads
Gbangba Gbangba Nakorji
Dakmon-Magw e t > » » Badeggi Badeggi Ginya Kacha Kacha Sheshi
Jima-Doko Jima-Doko Ndgenu
Sakpe Kutig i Kpotu
Egbako ,, >>
Laboz i
Mokw a Mokw a .Rani
Kyedye Kyedye Kuta
Tw o Distric t Heads are not members of the royal nobility : the Kuta of the Kyedye, whose privileged position as a provincial chief directly under the Etsu had been recognized in the early days of British rule and had been renewed recently by the Administration, and the Sheshi of Kacha, who was appointed in 1934. He had been the hereditary village chief of Kacha, and his appointment to an office which had been held previously by the Maku, a royal prince, was designed, like the appoint• ment of the Etsu Tafie, to open responsible state offices to the Nupe
‘commoner’. Unfortunately this appointment was once more based on insufficient knowledge of the country. Th e territory over whic h the ancestors of the Sheshi used to exercise their rule as ‘town-kings’ of Kacha has been placed in the political rearrangement of the country mostly into a different Division , the Agaie Emirate. Also the town and
1 See p . 25.
160 GOVERNMENT IN MODERN NUPE
territory whic h he is called upon to rule to-day, the richest trade centre of the country, is inhabited to a very large extent by strangers, traders, craftsmen, and canoe-men fro m other parts of the country, many of them from Bida, that is, by people who reluctantly bow to the rule of a culturally ‘inferior* peasant chief. It is not surprising, therefore, that his administration of the district proved unsatisfactory, and that the Etsu felt justified in urging the Distric t Officer to reinstitute in this respon• sible office a member of his own class.
Th e Distric t Heads live in their districts; they reside in the largest tow n and hol d there their littl e court, an exact copy of a nobleman’s household in the capital. The y have to make frequent tours of inspec• tio n of their area, and at regular intervals they visit the capital to report to the Etsu, especially at the time of the great Mohammedan festivals when all officers of state assemble at the king’s court. Distric t Heads are sometimes transferred from one district to another, but the tendency to-day is to let them stay in one district as long as possible so as to become thoroughly familiar wit h the people and the local problems of administration. The y are responsible to the Etsu for law and order in their districts, their main task being the supervision of tax collection; bu t they are also charged wit h such non-political tasks as finding pupils for the elementary schools of the Emirate. Thei r subordinates in the administrative machinery are the various Village Heads of the district, of who m we have heard already.1 Th e Distric t Heads are further assisted in their work by a staff of official and semi-official helpers— scribes, messengers, policemen, and the host of bara and henchmen who attach themselves to every man of rank.
TOWN ADMINISTRATION
Th e administration of Bida tow n has undergone many changes in the thirt y or forty years of British Administration . Th e original system was based to some extent on the division of the tow n into a number of
‘wards’, each associated wit h an influential member of the Bida nobility . Th e tow n was divided into twenty efu (or, as they were known officially, ungwa—the Hausa wor d for tow n ward) ; five of these wards were grouped together as one large ungwa, and these four ungwa were placed under four men of high rank: Ndeji, Maiyaki, Galadiman Gari, and Gabi Seidi. Th e four ungwa bore no relation to the dynastic division of Bida into three ekpa; their arrangement simply followed the four main roads which cross the tow n from nort h to south and east to west. However, the four rank-holders in charge of the large ungwa as wel l as the other subordinate ward-heads in the small ungwa (or efu) were actually livin g in the part of the tow n over whic h they had been placed; the lin k of locality whic h is so relevant in Nupe life was thus recognized officially. Later, however, this system of ward-heads lent itself to certain abuses, bribery and extortion of tax-money; it was abolished by the Administra • tio n and replaced by a new system whic h now divides the tow n into four
1 See p. 52.
162 GOVERNMENT I N MODERN NUPE
ungwa (the same four ungwa), defined geographically as Bida North , South, East, and West. Four new officials were appointed who are no longer inhabitants of ‘their’ ungwa but are supposed to be jointl y re• sponsible for the whole town , irrespective of their personal connexions wit h one or the other locality. Th e Hausa and Yoruba communities of Bida> who live in Ialemi, were placed under their own ward-heads. These six officials for m the present Tow n Council. A royal prince, the Natsu, was appointed as its chairman, and at the same time as the representative of the Etsu on the Tow n Council. Th e attached sketch- map illustrates the changes in the tow n administration.
A certain inner weakness of this system becomes apparent at once. It lies in the disregard for the factor of locality and for those direct contacts and bonds of ‘protection* and personal attachment which alone establish in Nupe a relationship of mutual confidence between the social classes, between ruler and ruled. It is not surprising therefore that the new division into four geographical areas ‘di d not work’, and that in 1936, only three years after its introduction, a fundamental reorganization was contemplated.1
TAXATIO N
Th e taxation of the Emirate is fixed by the District Officer, who tours the country and, assisted by District Head and Village Head, assesses the income of the various villages and districts. Th e tax uni t is the village area.2 Th e distribution of the tax among the people of the village and its dependencies is left to the Village Head and his elders. Taxation varies partly from district to district and is also different in town and country. Complaints against too heavy taxation, or reports on default• ing, are forwarded through the Distric t Head to the capital, that is, to the Emir’s Council and the District Officer. Ta x defaulters can be imprisoned.
Of the total tax revenue of the Emirate (to which are added fines, licence fees, &c.) 30 per cent, goes to the Central Government of Nigeria and 70 per cent, to the N.A . Treasury.3 Fro m the N.A . share in the revenue are paid the salaries of all (native) N.A . officials, the expenses of all Public Works for which the Emirate is responsible, the maintenance
1 In a memorandum for the reorganization of the Bida Tow n Council, whic h I drew up on the request of the Administration , I have suggested (1) a tow n division based on the division into three ekpa, whic h is so eminently character• istic of the social life of the kingdom, and (2) a system of election of ward-heads by the people themselves. Th e latter would , in my view, utilize and render more valuable the channels of personal connexion between patron and client, trusted ‘protector’ and the ignorant, intimidated commoner who stil l feels that he could only benefit from the association wit h a man of rank and learning.—A paper on the same subject was published in abridged form in the Summary of Lectures, Oxfor d University Summer School on Colonial Administration , 1937, pp . 97-8. See also p. 123.
2 Lor d Lugard , Colonial Reports Misc. , No . 40.
3 Thi s scale corresponds to the position of Bida Emirate as a First Class Emirate. In 1935-6 the scale was slightly altered in favour of the N.A . share, because the buildin g of the Etsu’s new official residence—a typical item of N.A . expenditure—would not have been possible withou t this additional revenue.
GOVERNMENT IN MODERN NUPE 163 of N.A . Hospital and dispensaries, schools, experimental farms, &c .
Part of it is used to buil d up and maintain a substantial reserve fund to
be used for unforeseen expenses, in the case of a famine, and suchlike.
Th e apportionment of the tax revenue, above all the allotment of funds
for new enterprises, is subject to the join t approval of Etsu and District
Officer. It is, in fact, in matters of this nature that the harmonious
collaboration of the two men is often put to its most severe test.
Th e far-reaching social change that the introduction of modern taxa• tio n has meant for Nupe kingdom cannot be easily exaggerated. Th e universal taxation under British rule has completely overthrown the former prerogative of tax exemption enjoyed by the nobilit y and the population of the town . It is not surprising therefore that these people should stil l strongly resent this abolition of their privileges. But those who resent it most strongly are not the men ‘on the top’, but rather the people of inferior status, such as guild-heads and Bida craftsmen, who had not been recompensed as many of the others had been in the share- out of new, well-paid official posts. Th e old guild-head of the Bida woodworkers once quite violently insulted and attacked me, and wit h me all White Men , for having robbed hi m of his former status and privileges. ‘Formerly’, he shouted at me, ‘the other craftsmen would come numbly to my house prostrating themselves before me and would present me wit h gifts to show their loyalty. To-day I have to pay the same tax as they—can you wonder that they have become arrogant and simply ignore me?’ One would expect, then, that peasants and small craftsmen and traders would appreciate the equalizing effect of the
, present system. Thi s is not wholly so—partly for reasons to be ex• plained presently, but partly also because the change had done away wit h the former system of voluntary gifts on which so much of their social security used to depend. By these gifts they established a personal lin k wit h the men of rank and influence, on whose assistance they could count in many situations when such assistance was of utmost importance. These situations have not and perhaps never wil l disappear completely. It is not surprising that the people cannot yet fully appreciate the ‘im • personality’ of the modern system, nor realize fully that in return for their tax-money, paid to an ‘impersonal’ Administration, they receive things no less valuable than the former personal lin k wit h powerful individuals: public security, law courts, roads, hospitals, schools.
Taxation is the most sensitive spot in any governmental system. It is perhaps not to be wondered at that the administration of taxes proved the most dangerous stumbling-block on the path of the young N.A . Native official after official, often in the highest position, has lost his post, and even gone to prison, for embezzling tax-money or levying illicit dues. However, there are two more serious weaknesses inherent in the taxation system—more serious because affecting more deeply the inner structure of the society.
Th e first concerns the (i n my view) disproportionately high taxes in the Emirate. Bida Emirate as it exists to-day is not a rich country, the masses of the people are not well off; the discussion on the economic system wil l bring ample evidence of this. Nupe country is i n this respect
164 GOVERNMENT IN MODER N NUPE
much less fortunate than, for example, the neighbouring Emirate of Ilorin , a much richer country, yet paying lower taxes. One at least of the reasons for the high taxation in Bida Emirate appears to be the high cost of the N.A. 1 To compare it again wit h Ilorin : there, wit h a muc h higher revenue, the expenditure on the Emir’s court and the salaries of the higher N.A . officials are almost the same as in Bida.2 To understand how such a situation could have come about we must envisage the historical frame of min d wit h whic h a Government committed to the policy of Indirect Rule is often tempted to approach the problem of administra• tion . Historically, Nupe Emirate was one of the most important Emirates in the north of Nigeria. Culturally it stil l is so. Politically, however, it has shrunk to a small country, shorn of many of its former dependencies (e.g. the whole south-west bank of the Niger wit h the ric h rice-growing district of Patigi and the large tow n of Egga). Lon g wars and longer political unrest have impeded its economic recovery. Yet it could hardly be made anything else but a First Class Emirate, and its ruler had to be given the corresponding status, weighing heavily though it does on the country’s revenue.
Th e second weakness lies in the method of assessment. Th e tax in the more advanced areas of Nigeria is—i n theory—an income tax. But how, one is teiripted to ask, can any native Distric t Head and group of Village Heads acquit themselves satisfactorily of such a formidable task of assessment ? In Bida town , where Distric t Officers themselves have computed incomes and taxes, the taxes are sadly out of proportion wit h the income (at least to-day, after the economic depression of 1932-4). In certain cases where property and capital are easily assessable some sort of rough approach to an income tax on a sliding scale is possible, e.g. among the Kyedye canoe-men where the amount of tax can be assessed in accordance wit h the number of canoes owned by an individual. But among the large majority of the people whose taxable income consists exclusively of farm products such a method must fail. Th e result is a qualified poll-tax assessing the tax per head of the population bu t admitting two or three different grades: the higher tax is paid by able- bodied married men roughly between the ages of 22 and 50, the lower by men younger and older than this. Children, women (unless they are independent traders), and old men or invalids who can no longer work are free from tax. Now , whereas the traditional productive uni t among the Nupe peasants is embodied in the family, taxation takes into account only the individuals. At the time of my stay in the country the last tax- receipts made out for the whole family, to be paid by the family head— as one can imagine very large receipts—were disappearing rapidly, giving
1 A British trader who knew the economic situation of the country intimately framed his impression in the following words: ‘The overhead expenses of the N.A . are too high.’ Bluntly expressed though it is, the statement is far from unsound.
2 In Ilorin, with a revenue of £28,885, the salary of the Emir is £2,250, and
the salaries of 17 District Heads amount to a total of £2,539 (an average of £149 per head). In Bida, the revenue of which is £18,161, the Emir’s salary is £2,000, and the salaries of 7 District Heads amount to £1,110 (an average of £159 per head). Nigerian N.A. Estimates, 1936-7.
GOVERNMENT IN MODER N NUPE 165 way to the (much smaller) tax-receipts made out for individuals. It is, in
fact, the present official policy of the Administration to encourage the
issue and acceptance of individua l tax-receipts throughout the country.
Taxation thus ignores and even denies in its own powerful province the
existence of the family unit—wit h what results we shall see later.
LAW
Le t me tr y first to give a picture of the judicia l system of Nupe king • dom as it existed in pre-European times. Thi s is, naturally, an historical reconstruction. But the history to whic h it refers is not the history of a mythical past. It is a history well documented, and obtained from infor• mation whic h seems reliable and significant and which , collected from different sources, shows remarkable agreement on all main points. My chief informants were village elders and chiefs and the Mohammedan judges of Bida Emirate who, in charge of modern jurisdictio n to-day, have witnessed and partly practised themselves the old system.
Th e theoretical difficulties wit h whic h the student of law in primitiv e cultures is faced are absent in our investigation. Th e problem, espe• cially, whether to define the province of law on the narrow basis of systematic social control exercised by a politically organized society (Radcliffe-Brown) or, more comprehensively, as embracing all social sanc• tions that make for uniformity of conduct (Malinowski), is decided for us by the nature of the society which we are studying in the sense of the former essentially modern interpretation. La w in Nupe kingdom is quite conspicuously a concern of the political organization and forms part of the elaborate coercive machinery of the state. Thi s does not mean that no sanctions or forms of redress exist whic h are applied out• side the political framework and left to the more diffuse workings of public opinion and individual or group ‘self-help’. But the scope of this non-political (or extra-political) law is extremely restricted. It comprises only two types of offences: religious offences—desecration of sacred objects or places, both Mohammedan and pagan—and kinship offences—litigation over inheritance, offences against traditional mar• riage rules (i.e. marriages in forbidden degrees), and also incest. I cannot describe here in detail the treatment whic h Nupe society meted out to individuals guilt y of these offences. Suffice it to say that public ostra• cism whic h might force the culpri t to self-exile or to confidence in his vindication by the deity was its main feature. Th e society thus exempted from its otherwise comprehensive scheme of formal political jurisdictio n the two offences whic h were either of a nature transcending human justice (as in the case of religious offences, and also partly incest) or concerned only internal frictions in individual kinship groups. Al l offences that affected the relations between kinship groups or individuals from different kinship groups, and all crimes against human life, were regarded as a concern of the community at large.
But the comprehensiveness of political jurisdictio n admitted of two degrees. Traditiona l Nupe law operated wit h a legal distinction which , corresponding in certain respects to the modern distinction between
166 GOVERNMENT I N MODERN NUPE
civil and criminal law, defined two classes of delicts: simpler delicts, which were settled by ‘repairing* the damage that had been done, and here the Nupe did not speak of sentence or judgement but of gyara, reparation, or arbitration; and graver delicts, ‘criminal’ cases in the modern sense, which called for formal judgement and punishment— sherVa. The following table summarizes the various crimes and offences according to this twofold classification. Let me emphasize that* these data are widely known, and that every village elder or older man in Bida is able to enumerate and explain the types of offences and the nature of the punishment. This wide-spread legal knowledge is clearly the psycho• logical counterpart of the systematic co-ordination and centralization of Nupe jurisdiction.
Offences involving gyara Offences involving sheri’a
Debts, small. Debts, large.
Theft on a small scale and during Theft on a large scale, theft of cattle the day; theft of fowls, sheep, and or horses, and all theft committed
goats. during the night.
Adultery. Adultery leading to fights and blood•
shed.
Seduction of girl by unmarried man Seduction by married man who refuses who is willing to marry her. to marry the girl.
Murder and manslaughter. Highway robbery.
Arson.
High treason, i.e. rebellion of feudal lords against the king.
Lhe-majeste (gi toko nya tsu, lit. abuse
of the king).
Th e first type of offences was a matter for the local authority. In the royal domains it was dealt with by the chief and the elders, in the fief by the delegate of the feudal lord who resided in the village, in consultation with chief and elders. The sentences which this local tribunal could inflict were fines, flogging, and sheld, a practice closely akin to pillory: the culprit was marched through the village accompanied by drummers who proclaimed his offence.
Th e graver crimes, crimes which demanded sheri’a, were referred to the central authority—the court of the king and Alkali in Bida. They were, in fact, also called leifi nya tsu, Crimes of the king’. This highest court of Nupe could inflict sentences of every order, above all, the death penalty. The offences were brought to the notice of the central authority through the feudal lords and delegates. Th e villagers would report to their overlord in the town, or the feudal lord would have the news from his local delegate, and would then inform the supreme court of Nupe. Th e criminal was apprehended by the king’s slave police and taken to the capital where the court was sitting. This system which made the feudal lords and delegates the ‘ recorders’ of crimes and offences laid itself open to many abuses. Th e feudal lords could be influenced in favour of a better paying claimant, or prevented by bribes from report• ing a particular case. Th e usual procedure as regards, for example, theft or debts was described to me thus: Up to the value of ios. nothing
GOVERNMENT IN MODERN NUPE 167 would happen: ‘you give the egba half this sum, he wil l quash the case,
and you wil l still have a profit of 55. Of course, if a larger sum is involved
this is not so easy, for such an amount is difficult to hush up, and too
many people woul d know about it.*
Th e judicial procedure i n the capital followed, til l recently, tw o different principles : the sheri’a gboro, ‘straight judgement*, for pagans, and the sheri’a nya kurani, ‘judgement according to the Koran’ , for Mohammedans. Th e former represented the ‘customary law’ of the country, whic h woul d do justice to the local pagan customs; it also im • plied the oath on the sacred objects of pagan Nupe, especially the chain of Tsoede. Th e latter followed the writte n Mohammedan law and implie d the oath on the Koran . Punishment of ordinary criminals was performed in the capital, on the open market—be it flogging, shela, or capital punishment. In the case of political offences and in the case of men of rank, royal princes, or members of the high nobility , the execu• tio n took place under the cover of night and in the secrecy of their ow n houses. Whether this ‘preferential’ treatment expressed a last privilege granted to the rulin g classes or whether such secrecy was thought to uphol d the morale of the feudal kingdom I cannot say.1
Al l the various crimes and offences listed in our table became action• able only when there was a plaintiff. Th e state itself di d not undertake to prosecute unless it was appealed to first by a person bringin g a charge against some other person—with one exception: the crime lese-majeste. Th e widespread knowledge wit h regard to this crime and its punishment in districts and groups whic h certainly never came into personal contact wit h the kin g precludes the assumption that in this case, too, the legal executive waited for a concrete charge being brought by the plaintiff in question, i.e. the king . Rather we must assume that the crime became actionable in an absolute sense, whenever and wherever it was com• mitted , provided that witnesses of the act existed. Whether the informa• tio n about the crime reached the central authority throug h private information or denunciation, or through the official channels of delegates and sub-delegates, I cannot say; bu t the important fact is that here the state took the initiative in prosecuting the offender. We have here the instance of an indictable offence in the modern sense of the word—a legal definition unparalleled in any known primitiv e system of law.
In this comprehensive and strongly centralized jurisdictio n the political system of Nupe achieved its strongest hol d over the peoples of the kingdom. Here the co-ordination of morality, the uniformit y of cultures of whic h we have spoken, is secured by the most effective weapon, the coercive machinery of state jurisdiction . We possess certain evidence whic h shows that this comprehensive co-ordination was achieved gradually, and that the political system extended its judicia l
1 I have referred already to the ancient, pre-Fulani, judicial organization of the leduy the ‘King’s Hangmen’, who lived in villages on the Niger and there, far from the king’s town, executed the criminals who were condemned to death (see p. 73). It is interesting to mention here certain stories current in Bida which record that the ledu organization was still used in the time of Etsu Usman against rebels of royal blood, who were imprisoned in Giragi, the ledu town closest to Bida.
168 GOVERNMENT I N MODER N NUPE
control further and further in the course of time into domains of social life hitherto unaffected by state law. Thi s extension proceeded on two planes: centralized control superseding local jurisdiction, i.e. moving the line of division between ‘offences’ and’ crimes’, and formal political law usurping domains that had been left originally to the informal work • ings of public opinion or self-help. An instance of the former is the treatment of witchcraft: we have heard of the secret anti- witchcraft society whic h the Nupe villagers could summon to detect and punish witches, and we have heard also that this secret society was brought later under the control of the Fulani rulers of Nupe. An instance of the second for m of extension appears to be the treatment of adultery. In • formants tol d me that formerly, at least in pagan communities, adultery was dealt with , not by legal measures but either by a religious sacrifice meant to safeguard matrimonial sanctity, or by self-help on the part of the husband, who was withi n his rights to beat up the paramour who m he caught in flagrante, strip hi m of his clothes, and drive hi m out into the street naked.
No w this strengthening of state jurisdictio n is understandable from the viewpoint of the dynamic state endeavouring to extend its control over new groups and new aspects of social life. But here we note the rather surprising feature that there are indications of the subject groups themselves having welcomed and invited the extension of state jurisdic • tion . I was tol d in Mokw a that during the Fulani regime fathers of headstrong sons sometimes appealed to their feudal lord , and that the latter would lend his authority to the more severe forms of domestic punishment (i.e. flogging). Thi s official cognizance of what is essenti• ally a ‘family affair’ clearly represents an incipient stage of state jurisdic • tion . It is only one step from this appeal for moral support of the official organ of the state to a full acceptance of political jurisdiction i n the sphere of kinship offences or litigations, for example litigation over inheritance or divorce, whic h has become reality to-day. Ho w are we to explain this readiness on the part of the people to surrender the sanctions of custom and self-help to state jurisdiction ?
Th e explanation lies in the evolution of the divided legal authority whic h had made this development inevitable. Tw o competing authori• ties of different range of power existing side by side and concerned wit h the same or closely similar domains of social life must be in danger of frustrating each other, as individuals dissatisfied wit h the decisions of one may be tempted to obtain help fro m the other. Th e elimination or (as in Nupe) surrender of the weaker authority is the result. Family heads could maintain their paternal authority only by appealing for support to that authority whic h had superseded their own in other social domains. We may formulate it almost as a social law that an expanding system of jurisdiction , once it begins to supersede another weaker system of jurisdiction , is bound to expand further; culture change in this sphere appears to be carried on by its own momentum. But inevitable as is this development, so are the conflicts inherent in
it . Th e surrender of new domains of social life to the judicia l authority of the state means potentially surrender to an alien law. In our example
GOVERNMENT I N MODERN NUPE 169
the family heads invited the intervention of the political authority in the interest of preserving the existing code of conduct. Bu t once this extraneous authority is admitted, it wil l have to be obeyed also should its decision ru n counter to the traditional code. Moreover, the rival authority may be appealed to equally by individuals or groups dissenting from the traditional conceptions of right and wrong. Ultimately , the issue wil l depend on whether the representatives of state jurisdiction are prepared to uphol d customary practices and codes of conduct, or are committed to promote a unifor m state law. Th e trend towards close political co-ordination and centralization must greatly limi t the scope of such concessions as might be granted to the divergent legal conceptions in the various subject communities. Th e extension of state jurisdictio n over new domains of social life may thus come to mean the substitution of existing traditional codes and doctrines by a new unifor m state law. I have stated this as a hypothetical case, being unable to produce con• crete evidence to show that precisely this development occurred in the case of the various delicts and offences over which we found state jurisdic • tio n already firmly established. I can only point to certain evidence of the reaction of group sections against what must have appeared to them as an excessive unification of the law, namely, certain attempts to change state jurisdictio n back to independent local jurisdiction : thus the chief of the Kyedye is reported to have obtained from Etsu Abubakari the right to exercise ful l and independent jurisdiction in his tribal group ; and the people of Sakpe and Kutig i claim to have been granted a similar legal fran• chise by Etsu Masaba in exchange for heavier, taxation. However, the progressive extension of state jurisdiction and the trends underlying this development are still visible and effective in present-day jurisdiction , to whic h we shall tur n presently. The y appear, characteristically, in that domain of social life whic h was the last to be absorbed by state jurisdic• tion—kinship—and concern the relation between the spreading-modern
state law of Nupe, Mohammedan law, and pagan customary practice.
Before we describe modern jurisdiction in detail we may examine for a moment the causes of the spreading of Mohammedan law in Nupe. It has no doubt been consciously fostered by the Mohammedan rulers of the country. But to some extent Fulani rule, stil l in British times, appears to have avoided excessive legal unification by allowing for customary pagan law—the ‘straight judgement’ in the case of pagans. Here, however, we find that the same effects that accompanied the
‘ divided authority* occur again in the case of the ‘divided law’. Th e coexistence of two legal systems, one backed by the prestige and supreme power of the rulers of the country, the other reserved for the pagan subject groups, is again reduced to uniformity , the ‘stronger law’ ousting the ‘weaker law’. We have concrete data to illustrate this process. Th e Mohammedan judges used to ask the parties appearing before them whether they were Mohammedans or pagans. Th e large majority proclaimed themselves Mohammedans, and desired to be tried according to Mohammedan law. Administrative Officers in areas where pagan and Mohammedan courts exist side by side also said that the people woul d generally prefer the Mohammedan court. The y explain
170 GOVERNMENT IN MODER N NUPE
this attitude as due to the reluctance of the people to be classed wit h the more backward sections of the population. But there is also a more fundamental reason. Offences and litigation involve by their very nature dissension fro m established rules of conduct. It is thus often in the interest of offenders and litigants to dissociate themselves from tradi • tional justice and to appeal to the rival law existing in the community. We shall see that in certain extreme cases the ‘divided law’ almost puts a premiu m on dissension from the traditional concepts of right and wrong.
In modern Nupe, then, Mohammedan law is paramount. Th e legal administration of the country is entrusted to Alkalai, judges trained in Mohammedan law, who reside in Bida and in the districts. Th e Distric t Courts, of whic h there are seven in Bida Division , are courts of a lower grade—in official terminology, ‘grade C courts’—that is, they are en• titled to deal wit h civil cases involving fines up to £50 and criminal cases involvin g imprisonment up to six months, in case of theft up to twelve months. Bida tow n possesses a ‘grade A court’, the chief Alkali’s court, whic h combines the two different functions of a local court for the tow n and an appeal court for the districts. It has unlimite d jurisdictio n over all cases whic h under the Native Courts Ordinance are subject to the jurisdictio n of Native Courts, i.e. excepting such cases as sedition, charges against Europeans, treason, tria l by ordeal, and corruption of officials. Th e jurisdictio n of the Bida Court also excludes homicide cases—an exceptional limitation whic h was imposed upon Bida after the disturbances of 1931. Th e Etsu wit h his councillors forms another
‘grade A court’ whic h deals wit h cases concerned wit h land, and consti• tutes at the same time a higher court of appeal for cases from the districts. Th e Distric t Officer represents a final court of appeal for all cases. Cases excluded from native jurisdictio n come before the Hig h Court, whic h sits fro m time to time at the Headquarters of Niger Province, presided over by a professional British judge.
It is interesting to dwell for a moment on the description of a Native Court in action. Picture the court building : a spacious round house, wit h a gallery runnin g round its outer wall . Th e Alkali has his place on a raised platform near the wall , books and papers spread out before him . On the floor, leaning against the platform, his assistant is sitting. In the Bida Court there are also scribes who take down the minutes of every case. Ushers, messengers, and sometimes policemen are standing at the doors. Th e Bida Court sits daily from 10 a.m. til l 2 p.m., the Distric t Courts whenever there are cases. The- court procedure is very rigid . Plaintiffs and defendants are squatting in the centre of the court room, facing the judge, their hands folded in their laps; to rest one’s hands on the ground means disrespect towards the Alkali. Th e place of witnesses is to the side and back of the plaintiffs and defendants. Women must sit in the outer gallery (at least in cases concerning divorce, adultery, and suchlike); they speak through a small windo w i n the wall. Yo u wil l recognize at once the peasant from the country: he sits there shy and frightened, often quite bewildered by the questions of the judge ; often too, he has to be reminded to raise his voice, which is almost inaudible.
COURT HOUSE IN BIDA
THE ALKALI OF MOKWA
GOVERNMENT IN MODER N NUPE 171
But in the Bida Court temper may ru n high, especially when people of standing and wealth are involved who not infrequently wil l argue and remonstrate wit h the judge. I have seen judges lose their patience wit h such unmanageable parties, shout at them and threaten the m wit h punishment for this ‘contempt of court*.
Before the pretentious Bida Court wit h its office hours and fixed rules of procedure, its crowd of subordinate officials, assistants, scribes, ushers, policemen, a multitude of cases are brought daily whic h fi t wel l into its up-to-date routine: offences against the by-laws of a modern tow n (ridin g a bicycle withou t light, or riotous behaviour in the street or the market, &c.) and all the various disputes bound to occur where yo u have an intensive trade, complicated money transactions, and suchlike. Bu t before the same court come the cases (and in this book we are only concerned wit h these) that are the expression of tensions and conflicts of the society in its old unchanged form—cases of inheritance, divorce, witchcraft. Th e extension of native jurisdictio n over the problems and conflicts of a modern urban community, whic h is merely a question of applying jurisdiction and executive under the guidance of the Britis h Administration, is of less interest to us than the progressive penetration of formal political jurisdiction into the domains of traditional native life, whic h is a problem of legislation that the native society must itself solve. In this, the modern Native Court is the legitimate heir of the pre-British system of law—heir both to its strength and its weakness.
We may say that the modern legal system is both more comprehen• sive (embracing, above all, the domain of kinship) and, viewed fro m the standpoint of the groups which submit to its jurisdiction , more alien than the ol d system. To explain this we must for a moment go back to a question discussed previously. I have said that every extension of a unifor m jurisdictio n and legislation over new social domains involves potentially the substitution of a new law for original codes and con• ceptions, and thus a certain disparity between the recognized law and the culture of whic h the old conceptions were an expression. Thi s disparity is clearly a phenomenon of transition: every society wil l tend in the course of time to eliminate such a hiatus—either by reacting against the expanding judicia l system and forcing it to adjust its too extravagant claims (we shall presently encounter instances of this)—or by complet• ing that process of cultural assimilation of whic h the co-ordination of legal concepts represents only one aspect. We may assume that our table of jurisdictio n in pre-British times, whic h shows a certain system of law firml y accepted throughout the society, refers to aspects of culture in whic h this process of mutual assimilation has been completed. Whether or not my hypothesis is historically correct, we must admit that where the process is stil l in motion, where the politically-backed law is only jus t extending its hold over new domains, the disparity of whic h I have spoken must be visible. Thi s is indeed the case to-day, when Moham • medan law is being applied to kinship offences.
It woul d go beyond the scope of this book to study the relation between Mohammedan law as it is practised in Nupe and the traditional concepts of kinship conduct and kinship rights and duties. I may be
172 GOVERNMENT IN MODER N NUPE
allowed to reduce this relation to a simple formula: if the former is bqr n of the social conditions of an urbanized stratified society and reflects the specific concepts of property and morality whic h have evolved there, the latter is typical of the peasant community, its economic pursuits, its values and interests. Th e Mohammedan law as it is applied in Nupe has undergone a number of significant changes in the comparatively short period of Fulani and later Britis h rule. Th e knowledge of the more im • portant changes (e.g. in the treatment of seduction and adultery) has spread throughout the country and the changes remain linked in the memory of the people wit h the names of the judges who were their originators. Some changes are regarded as good, others as bad, and accordingly judges have earned or forfeited the reputation of being ‘wise men’. Th e lawgivers themselves have been aware of the necessity of bringing their extraneous legal machinery into at least partial harmony wit h existing conditions and conceptions.
Here a new aspect of the * disparity’ between law and culture reveals itself. Th e very fluidity of this evolution of a new legislation must remove the law from the orbit of the average citizen. He is no longer able to keep track of it as he could when the legal system could be reduced to the short list of * civi l offences’ and ‘crimes of the king’ . I have encountered ample evidence of this kind—erroneous beliefs of individuals as to particular laws or forms of legal procedure. Some ten or fifteen years ago unscrupulous judges made use of the ignorance of the peasants and prosecuted, for * neglect of their parental duties’ the parents of girls and young men suspected of illegitimate sexual inter• course. Th e repercussions of this abuse are stil l felt, and many peasants dare not bring a case of seduction before the court for fear of being fined themselves. A conspicuous evidence of this gul f between the law and the people is the continued practice of peasants and commoners to seek out a Malla m or a person of standing and have hi m accompany them to the court when they are about to commit themselves to the vagaries of an unknown law.
Th e disparity between law and culture is naturally most marked in the slow-moving peasant communities, where social and economic condi• tions diverge most fundamentally from the social life of whic h Moham • medan law is an expression. Here the unfamiliarity of the judges, who come from the highest families of the kingdom and who are Bida born and bred, wit h the local situation and specific customary practices (e.g. marriage rules) may add yet another source of conflict.1 Even the religious emblems upon whic h the Nup e were wont to take the oath are absent in the modern Distric t Court which , instead, is using the Koran. 2
Pagans wil l always be ready to swear by the Kora n i f asked to do so by
1 Some of the local Alkalai may be very reluctant to regard themselves as per• manently attached to their station. The y may be hoping for a speedy promotion , eventually to the court in Bida.
2 In a court in the Nup e district of Ilori n Province a very enlightened Alkali
used both Kora n and the ‘chain of Tsoede’ to do justice to Mohammedans as wel l as pagans. In this as in similar respects muc h depends on the personality (and special training) of the individua l judge .
GOVERNMENT IN MODERN NUPE 173 the Alkali, said one of my informants, and wil l thin k nothing of perjur•
ing themselves.
Le t me quote a significant instance. An Ebe man had refused to let
his daughter marry a Hausa who m she loved, having a different suitor
fro m Ebe country in min d for her. Th e Hausa brought the case before
the Alkali of Kutig i (some 50 miles away); the Alkali instructed the
father to give his blessing to the marriage on the grounds that the Kora n
decreed that no father should stand between two young people who
loved each other. Th e father swore an oath to that effect on the Kora n
and returned home. But after a mont h the Hausa came back, reporting
that the girl’s father had renewed his opposition to the marriage. Three
times they were summoned to court, and three times the father agreed in
the presence of the Alkali, only to revoke his agreement afterwards. At
last I happened to be in the court and asked the man if perhaps the Ebe
disliked letting their daughters marry outside the sub-tribe or tribe.
Evidently relieved, the man said that this was indeed the case. Curiously
enough, the girl , who had til l then stubbornly maintained her desire to
marry the Hausa, declared that she woul d obey her father as soon as the
customary marriage rule was brought up in the court.
Thi s example also illustrates the effects of what I have called above the
4 divided law’ : the gir l and her suitor, well knowing that if they submitted to the customary law as it was upheld by her father and most probably the village elders woul d not be able to marry, appealed to the Moham • medan court, whose decision, as they anticipated, was favourable to them. Here we see what I have mentioned above—that in districts in whic h the customary conceptions of law stil l obtain, modern legislation puts a premiu m on breaches of the traditional usage. Thi s is shown most clearly in litigation over inheritance. According to Mohammedan law the property and land left by a family head should be divided equally among the sons; traditional rules of inheritance hold that land and property should be inherited by the younger brother, and that sons and brothers’ sons should stay on the land workin g under the heir as under their sociological father. Thi s rule expresses the interest of a farming community in maintaining landed property so far as possible intact, while the Mohammedan rule does justice primaril y to owners of movable property, money, trade goods, or cattle. Now , for reasons which we shall examine in a later chapter modern economic developments have led increasingly to the adoption of the Mohammedan solution. But were individuals stil l to attempt to enforce the traditional rule, dissatisfied family members could easily obtain a reversal of this decision by appealing to the Mohammedan court.
Again we observe the effectiveness of the ‘social law’ which we have formulated above, that a stronger law ousts a weaker law. Th e existence of the divided authority whic h favours ‘dissenters’ from traditional law, must force chief and elders—the guardians of traditional usage—to surrender in ever larger measure the authority whic h they are still hold • in g to the stronger authority of the state law, even to invite its inter• vention—as di d the family heads who appealed to their feudal lord for support bf their paternal authority. It is important to emphasize this, as
174 GOVERNMENT IN MODERN NUPE
many Administrative Officers thin k it desirable that the village chiefs and elders should continue to practise their informal arbitration side by side wit h the existing official courts. Ou r data preclude the likelihood of such a parallel development; they show that the tendency is rather towards the virtua l surrender of one to the other. We have heard that in the present-day judicia l arrangement the Village Head is expected to send all legal cases whic h cannot be settled by arbitration to the Distric t Alkali’s court. No w I have found the chiefs quite satisfied wit h this state of affairs, which freed them of the responsibility in more serious cases. Some seemed, in fact, quite frightened at the idea of taking over this responsibility themselves, and were only too ready to leave a larger and larger proportion of their judicia l tasks to the professional judge.
‘To-day’ , a chief said to me, ‘ people want writte n judgements in such cases as divorce, inheritance, debts; ama mi de takarda a, but I have no papers to give.1 Tapers’ are only the modern symbols of the more powerful and more permanent authority wielded by state jurisdiction ; in his own words the chief expressed the essence of that principle of
‘divide d authority’.
Shall we acquiesce wit h the chief in these results of an inevitable process of change ? I thin k not. Politically, the chief is divested already of much of his former power. To divest hi m also of all judicia l authority is to deprive hi m of the last vestige of the most vital prerogative of his office, which is the enforcement of a certain line of conduct upon his community, and to sanction all deviations from it as acts detrimental to the welfare of the community. It would amount to turnin g the chief completely into a petty executive official, entitled only to carry out orders, but wit h no right to voice public opinion. Besides, the disregard for local customs by the professional Mohammedan judges is felt as a real hardship by many. Even if more change is on the way, bound to promote the final assimilation of Mohammedan law and customary practice, its pace should be set by the people themselves, and not by an alien, un • interested judge. In the nature of things the solution must lie in a com• promise. In my view the best compromise woul d be to vest the Village Head wit h the right (and, more important, the duty) to act as an assessor
i n the court, wit h full consultative authority, whenever cases from his community are brought before the Alkali.1
Thi s does not contradict what I have said in a previous section about the ‘parochial outlook’ of the village chief. Th e parochial outlook may unfit hi m for responsible posts in the administration of the Emirate at large, such as Distric t Headship, whic h call for a wide view and an appreciation of policy on, as it were, a national scale. But it is just this parochial outlook whic h qualifies the village chief eminently for dealing wit h legal cases, bound up as they are wit h local conditions. However, there is a wider issue involved in this question—the most intricate problem which modern administration has to solve, the problem of centralization versus local autonomy.
1 This solution is envisaged in the Native Courts Ordinance, which leaves it to the discretion of the Native Court to appoint assessors from the ranks of village chiefs and elders.
GOVERNMENT I N MODERN NUPE 175
PROBLEMS OF ADAPTATION
Th e description of the political system of modern Bida has illustrated the relation between the social structure of Nupe society and the new Native Administration, the points of harmony as well as of divergence and conflict. I have just formulated one of the graver conflicts; we recognize it as one of the central problems of this analysis, the problem of cultural and social diversity versus political unification.
Moder n conditions have re-designed the cultural map of Nupe country. Roads, railways, and river steamers demarcate new areas of communication and intercourse. Th e scope of economic co-operation has widened immensely, far beyond the political boundaries of Nupe. Importan t trade centres, nuclei of a new, cosmopolitan Nigeria, wit h their large colonies of strangers, have arisen on railway lines, motor- roads, near river ports or river bridges: Badeggi, on the railway, is such an example, or Baro on railway and river, and Jebba where the railway bridge crosses the Niger. New cultural centres are forming’i n the villages which are the seats of Distric t Heads, Native Courts, schools, dispensaries, and trading-posts of the big merchant companies of Nigeria. But this development above all emphasizes anew the cultural and social gul f between Bida and the country. Bida is more than ever the seat of
‘civilization’ and learning, art, and fashion, the residence of the wealthiest traders and of the new rulin g class—the salaried officials of the Native Administration. Bida harbours the highest school of the Province, a hospital, the workshops of the Public Works Department which give employment to many, the large ‘canteens’ of the European trading companies. It represents again, on a new plane, the ‘core’ of the kingdom, the area of social life close to the ‘powers that be’. Moder n Nupe has re-defined the old conflict between cultural diversity and political unifica• tion, it has not eliminated it . In the political sphere the Government is attempting to do justice to this vital problem by admitting commoners and peasants to responsible offices in the central administration. I have ventured to point out certain weaknesses inherent in this experiment. But the question of the success or lack of success which is likely to meet this administrative experiment is less significant than the fundamental question whic h of the two principles of social and cultural development these administrative reforms are envisaging, a strengthening of unifica• tio n and centralization, or support and preservation of the social identity of village communities and cultures. Th e former tendency has been the trend of the historical development of the Nupe state (as of any state); it has also become almost an accomplished fact in the other spheres of present-day development, in economic progress and in the centraliza• tio n whic h is part of educational and health policy. Th e choice is thus, I think , in reality taken out of the hands of the Government. But what• ever policy is adopted, it can, for the present, mean littl e more than a preliminary arrangement. Th e final solution cannot be found in the realm of political organization; it wil l arise—and this our survey of the political history of Nupe kingdom has taught us—from cultural com• monness itself. One of the main instruments for achieving this cultural
176 GOVERNMENT IN MODERN NUPE
commonness in modern Nupe is planned education. A twofold plan of education: an ever-increasing spread of general education, disseminating the elements of modern civilization whic h so far have remained a prerogative of the few, among tow n people and villagers, people of rank and commoners alike; and also a reformulation of the higher education in the capital, so as to make it less specialized and brin g it more into harmony wit h the culture of the country at large, to make i t Jess an expression of exclusive, class-conscious ‘tow n culture’, and more an instrument of understanding for Nupe, wri t large. Only when this development is complete wil l the rigid , exclusive class boundaries be dissolved; only then wil l the tdlaka, the Poor One, be able fully to stand on his own feet, to have his share in the administration of his country, and to dispense finally wit h those links of clientship and patronage whic h so far have been his only weapons to overcome the vulnerability of his posi• tion . On this new plane the question of centralization versus independent local administration may then be reopened, and wit h more prospect of a permanent successful settlement. A peasant once said to me: ‘I send all my children to school, then they wil l be able to read and they wil l know what i t says on tax-receipts and fine-tickets. N o more cheat• ing or extorting money then!* Put it in this crude and a little aggressive manner, or call it education as a ‘means of cultural unification’, it means the same thing .
In the tow n and its rigidly stratified society our problem takes a slightly different form. Th e question how to transform exclusively political domination into a system wit h more equal, more ‘democratic’ distribution of political influence and responsibility reappears as the problem of social mobility . Th e social mobilit y of the Nupe class system, restricted as it was under traditional conditions, has been re• stricted further and at the same time diverted into new channels by the modern development. Of the vast bureaucracy of the kingdom only comparatively few ranks and offices could find a place in the new society and could thus maintain together wit h the symbols of rank their social significance. Tru e there are a number of newly-created offices and posts under the newly-created Native Administration : the post of Mayaji> Treasurer, posts as teachers in Middl e School and elementary schools, posts of clerks, scribes, N.A . ‘Mallams’, messengers attached to the various offices, and so forth. But the more important of these modern posts demand modern education, whic h at present has been enjoyed only by a small minority . It is a convincing proof of the fact that this minorit y is stil l to a large extent drawn from the classes of the nobilit y that almost all important N.A . offices, down to scribes and messengers, are in the hands of what have been for a century the privileged classes of Nupe.
War, the paramount factor of promotion and inter-class mobility, has disappeared; free competition, individual success, are almost excluded as channels of a social career. Th e system of offices is more exclusive to-day than it had been in feudal Nupe. Th e comprehensive scheme of promotion which used to embrace the whole scale of Nupe society is reduced to insignificance by the exigencies of specialized training and by
GOVERNMEN T I N MODER N NUP E 177
the changed conception of office and responsibility. Th e Administration itself is undecided in its attitude to the question of promotion and career. Certain distinguished state offices were granted in accordance wit h the traditional conception of a social career, that is, to men who had attained a hig h position on the native rank scale. Such are the appointments of Maiyakiy Ndeji, and the Tow n Councillors of Bida. Bu t other offices have been bestowed ad personam, for special ability, irrespective of the traditional rank or promotion. Thus , a former Head Teacher of Bida Middl e School was raised to the position of something like a ‘ Chief Secretary’ of the Emirate, and given the (Mallam- ) titl e Naibi. Or the Ndgya, a member of the gitsu nobility , has been made a member of the Emir’ s council, although according to traditional practice membership of the Etsu’s council was reserved for the saraki£i, the militar y or civi l ranks. Th e only promotion wit h universal application (and sure of universal appreciation) is promotion in salary. Fixed wages have be• come almost the only officially recognized symbols of social career. Ho w then wil l class distinctions and inter-class movement re-establish them• selves i n modern Nupe? Ho w wil l the native society adjust its con• ception of those paramount values in their cultural life, promotion, and status to the new situation ?
Th e attitude of the people towards these questions is ambiguous and undecided—the typical attitude towards changing values. It is interest• in g to study the attitude of the young men of to-day, including those who have enjoyed the highest education the Emirate can supply, in Bida Middl e School, towards the occupations whic h are open to them. It seems of littl e account to them whether they obtain the position of a clerk, an office messenger, or some other obscure post, or whether they are called to fill the post of a teacher or a responsible political official, as long as their appointment carries a salary. Again and again men of intel • ligence and hig h standing have said to me: ‘ Al l our native ranks and titles mean nothing to-day; what has meaning are only offices under the N.A. , posts whic h carry real influence and wealth.’ Yet it woul d be incorrect to infer from such statements as this that the traditional system of ranks and the whole concept of promotion and mobilit y are losing their vitalit y and significance. Th e class distinctions of Nup e are reduced to-day to distinctions of status devoid of their former economic and political significance; what seems to have remained intact is only the outer shell, a rigi d empty structure. Yet I have often been surprised to see how littl e the meaning of this structure has been affected.
Whe n I was in Bida a ric h trader was jus t making all possible efforts to entrench himself in the position of a favourite of the king, and he was pu t down accordingly by ‘those who knew’ as the next candidate for promotion to a saraki rank. I have seen three vacancies occur among the higher ranks of the nobility ; I have witnessed the ensuing party feuds, the passionate play of interests, the bartering for support and influence
—al l these signs of a charged social atmosphere whic h can be character• istic only of a system stil l intensely alive. An d one of those enlightened, sophisticated informants who deprecated the traditional ranks and titles was the hero of the following story: I have jus t mentioned that the
N
178 GOVERNMENT IN MODERN NUPE
former Head Teacher of Bida Middl e School, a very able (though un • scrupulous) man, had been attached for some tim e to the Administration in a very distinguished official position, wit h the titl e of Naibi. When he was dismissed fro m his post for malversation the new Head Teacher definitely expected to be appointed Naibi and even approached his superior officer officially to help hi m secure this post, to which , accord• in g to precedent, he felt entitled. Thi s example is deeply characteristic. It shows that that fundamental trend in Nup e society is stil l alive: it reveals the anxious search for traces of a routine of promotion to whic h the people had adjusted their life, and the eagerness to rediscover in the new society in whic h they are livin g the paramount social value of their culture.
If the restriction of social mobilit y and the reduction in the number of official positions under modern rule have caused this moral uncertainty in the rulin g classes, indirectly they threaten to unsettle the morale of the people at large. No t only because they render promotion from the lowest orders of the society so difficult, bu t because they have throw n out of gear the whole machinery of inter-class relationship. Man y of the channels through whic h public opinion—the desires and grievances of the people—used to find its way to the ears of the rulers of the country have been rendered useless; clientship and similar forms of political attachment can no longer promise protection when the large majority of the men who m native society had endowed wit h status and rank can no longer claim any official standing. Or rather, the channels stil l exist and stil l continue to function in their way, bu t they have been forced under• ground, and they function in the subterranean sphere of intrigues and scheming. Th e Government, supporting those who m it has appointed, cannot stoop to recognize an authority resting only on the fluid, ill – defined loyalties involved in patronage and clientship. Th e former wide distributio n of political influence, combined wit h the system of compre• hensive promotion, had secured a fairly efficient ‘balance of power’, and this ‘ balance of power’ the powerless masses could tur n to their advan• tage through the institutio n of clientship. Th e rigi d present-day system ranges its authority exclusively behind the few men to who m it entrusts its offices; it must brand all other efforts at political influence as dis• turbin g and even dangerous.
I have a concrete case in mind : a man of highest status in the native society, highly educated (i n the Mohammedan sense), who was regarded by a large number of people, in the tow n as wel l as in the districts, as their patron and adviser in many questions. However, holding no official position, he was regarded by the Etsu and his entourage as an interloper, an unwelcome meddler in state affairs; and an over-cautious Administrator described hi m officially as a ‘subversive element’. In a sense, he was that ; he had been forced into this position, since the in • fluence whic h he attempted to exercise in favour of his clients and fol • lowers had to wor k outside, and not infrequently against, the established system. Ye t had the rigidit y of the present system not excluded hi m from an official position I have no doubt that he woul d have been able to use his influence over the people to the benefit of the country.
GOVERNMENT IN MODERN NUPE 179
I need not enlarge upon the dangers of such subterranean activity— subterranean through no fault of the. people themselves. Th e terrible flare-up in the witchcraft tria l of 1931 is a lesson not easily to be for• gotten. A remedy? I have suggested one above, couched modestly in a footnote: brin g these channels into the open daylight ; let the people themselves elect, periodically, at least in Bida town , their official repre• sentatives—Town Councillors in official nomenclature, People’s Repre• sentatives in a wider sense.
Tha t the people are equal to this task I need not elaborate further; nor woul d the group of likely candidates be too large and thus render an election difficult. Th e dangers involved in the solution whic h I suggest
—the dangers of faction politics and intrigues—would be easily con• trolled once the whole network of relationships lies on the surface. How • ever, one migh t object to my suggestion on the ground that it tends to keep alive the serf-like lin k of patronage whic h should rather be dis• couraged. To this I woul d reply, first : the personal lin k of patronage, littl e as it may appeal to the modern moral conscience, is for the present the only lin k of political representation that promises success. For the masses of the Nupe people as they are to-day, unenlightened, only too conscious of their weak position, economically and socially, it woul d be futile to attempt to find representatives in their own midst—they woul d be puppets in the hands of the experienced and infinitely better equipped rulin g group. An d second: a solution such as I have suggested woul d be most easily adapted to improved social conditions, that is, to a com• munit y wit h a universally higher level of enlightenment. Muc h more easily, in fact, than a system wit h carefully selected representatives whose position of confidence among the people is never put to a test. For the solution here suggested looks towards the future; it embodies one of the most vita l elements on whic h the social achievements of the native peoples wil l finally depend, namely, political education—that training i n political responsibility to the achievement of whic h the colonial policy of this country is irrevocably pledged.
CHAPTE R X I
ECONOMICS: RESOURCES
ALFRE D MARSHALL , discussing possible definitions of Economics, A says: ‘Th e less we trouble ourselves wit h scholastic inquiries as to whether a certain consideration comes withi n the scope of economics, the better.’l If any student of society is entitled to follow this practical advice it is the anthropologist. Clearly, his working definition of economics must be as wide and fluid as possible. In his field every fact referring to what Marshall woul d call the ‘business side of life’ must claim his attention; even apparently irrelevant details in the technique of production, exchange, or consumption may be significant because new and not hitherto examined under the aspect of economic problems. Such a wide definition of economics must do justice to three realms of facts: (i ) the primary resources, in the sense of the physical substratum on whic h all productive activity ultimately depends; (2) the technical processes and forms of group organization which make up production in the narrow sense, i.e. whic h actuate and bring into existence the re• sources ; and (3) the specific needs that exist in a society and on the satis• faction of which the efficiency of a productive system must prove itself.
This , then, wil l be our definition of the realm of economics: the utilization of resources for the satisfaction of culturally (or socially) standardized needs. Th e description of the needs themselves, elements of culture as they are, is part of the general description of the society. Th e relation between resources and the satisfaction of the needs wil l be examined under the heading of Economic Balance. In the description of the technical processes, however, we shall have to limi t our field. In accordance wit h the general theme of this book we shall be concerned more wit h the group organization involved in production than wit h technology proper; we shall view productive labour under the aspect of human occupation rather than under the aspect of technical develop• ment. Th e description of the primary resources and their distribution wil l form the subject of this first section. It wil l examine above all that primary and ultimate resource of every productive system, savage or civilized, land.2
1 A. Marshall, Principles of Economics, 1930, p. 27.
2 For various reasons I have thought it better to omi t from this book a discus• sion on huntin g and fishing. Huntin g is indeed of littl e importance in the present- day productive system of the country; it has been reduced everywhere to a secondary additional activity in the nature almost of a leisure-time occupation. In some parts of the country (e.g. in Bida and surroundings) huntin g has dis• appeared altogether. Fishing stands in a different category. It is of paramount importance in the riverain area. Bu t a description of the greatly varying fishing techniques in use in Nup e country, and of the very specific forms of labour organization involved in fishing, woul d lead too far. For the economic system of the country at large fishing becomes important in so far as it produces one of the most highly valued trading-goods of the country both in local and export trade. Thi s aspect of fishing wil l best be discussed at a later stage, i n the chapter on trade.
ECONOMICS: RESOURCES 1S1
POSSESSION OF LAND
In the discussion of the rights to land obtaining in Nupe tw o facts must be borne in mind . First, the Fulani conquest, whic h resulted in the appropriation of part of the vacant triba l land ; and second, the very low birth-rate and sparse population of the country. These two factors limi t each other’s effectiveness; moreover, they affect different parts of the country in different measure. We have seen that the appropriation of triba l land took place only in the area east of the Kaduna (Cis- Kaduna).1 In the area west of the Kaduna (Trans-Kaduna),2 land- tenure on the traditional triba l or village basis stil l obtains, and here the population factor shows itself effective in that it eliminates almost all pressure on the land and all more severe competition for landed posses• sion. Owing to the political factors the opposite is true of Cis-Kaduna, and here a complex system of land ownership and land transfer, whic h in the extreme case amounts to outright purchase for money, reflects the attempts of the people to counteract the pressure on the land. Tha t land disputes are extremely rare in Trans-Kaduna, but very frequent in the
‘conquered’ territory, is evidence of the same development. One ex• ception may be mentioned at once: the especially valuable marsh-land groves and plots near river-courses, which the Nupe call bata (the Hausa kurmi). Disputes about this valuable kin d of land occur now and again even in the land-rich Trans-Kaduna, and they are fought wit h special bitterness in Bida and the crowded ‘conquered’ territory . Th e following figures may be quoted in support of this statement: in the years 1934 and 1935 24 cases of land litigation came before the EtsiCs Court in Bida ; 21 referred to Cis-Kaduna and 3 to Trans-Kaduna; of the latter none was a land case in the narrow sense of the word—one was a dispute about trees, and two were disputes over marsh-land plots.
Th e description of Nupe land-tenure may start most profitably wit h an enumeration of the various methods by whic h a man, stranger or native of Nupe, may acquire land or has been able to acquire land in the past. It must be said first of all that under traditional conditions only men can hold or claim possession of land. Th e dynastic policy of the Nupe kings has first led to women being invested wit h fiefs and landed property. Mohammedan law, besides, is inclined to confirm land hold• ing by women in their own right. But the holding of land by women is stil l extremely rare, and limite d to Bida and the surrounding districts
I have come across one case only—a court case in whic h sons and brothers of a recently deceased woman contested the inheritance of her landed property. Needless to say, none of the parties concerned were peasants; the land had been acquired first through the channels of cliencship, and the present claimants—Bida Mallams—wanted it only in order to sublet it for profit.
To return to traditional land-tenure. For a Nupe man who comes
1 Th e territor y between Kaduna and Gbako river, i.e. Beni country, and the districts east of the Gbako river, i.e. Badeggi and Kacha districts.
2 Trans-Kaduna is the literal translation of the Nup e name for the area:
Lavunkpa (Lavun being the Nupe name for the Kaduna river).
182 ECONOMICS: RESOURCES
from farming stock, that is, whose family is holding land already, the usual way to acquire land is to exercise his claim to a parcel of the land that is held jointly by the family group, and which is administered and apportioned by the family head. This would be the procedure, for example, in the case of a young member of the family who had founded a family himself and now requests land for a farm of his own. More or less the same situation would arise in the case of a stranger, a native of Nupe though not of the particular village, or even a native of Nigeria (a Hausa or Yoruba) though an alien in Nupe, who, having been adopted into a household, may request land on which to settle. There is, how• ever, the possibility that the family land available at the time is insuffi• cient ; or there might be a man desiring land whose family had not held land previously or had lost the claim to the land, e.g. through overlong absence. Here two main methods offer themselves: first, the man desir• ing land can exercise his right as a member, native or adopted, of the village community to request land (possibly additional land) from the chief, which may be either land that has been under cultivation pre• viously but has fallen vacant, or virgin land within the village territory. Second, he can acquire new land through an arrangement with an in• dividual landlord (using the term here in the widest sense); in other words, he can lease land temporarily for a rent in kind—a method known in Nupe as aro, ‘borrowing’. In both these methods the land thus acquired is inalienable.
A third method closely related to borrowing is a lease for life or for an indefinite period, in most cases against a single payment in money to which is added a fixed rent in kind. This arrangement, which definitely recognizes the money value of land, may be tantamount to real purchase. Fourth, there exists hereditary tenantship, which gives the tenant absolute right over his land, including the right to dispose of it again or sublet it. This tenantship I shall call (for reasons which will become clear presently) secondary tenantship. It implies not a fixed rent but a share in the produce—a form of metayer system.1 Finally, there exists SL primary tenantship, in which a landlord who himself owns the land by the right of conquest or appropriation may cede complete and absolute rights over land to another person, Nupe or stranger, in recognition of (political) services rendered; here again a share in the produce is implied. Th e last two methods form part of the feudal organization of the Nupe state.2
We have thus five methods of land acquisition, which fall in two main categories: (i ) Acquisition in virtue of membership of a group (kinship
1 See A. Marshall, op. cit., p. 643.
2 Different from these traditional methods of acquiring and transferring land is the method of land-tenure under the Native Land Ordinance Act, which, ad• justed to modern requirements, involves the lease of native land to non-natives, i.e. non-Nigerians (a merchant who plans a trading-station, a missionary who intends to build a school or church, &c) . In such cases application for a lease of a parcel of the land that has been retained as Crown land must be made to the Administration. This method, and all questions entailed in it, will be disregarded in the following. It is, in fact, of very little significance in Nupe country where no commercial utilization of land and no settlement of non-natives exist.
ECONOMICS: RESOURCES 183 group or village community), holding ‘corporate’ rights to land. (2) Ac •
quisition of land in virtue of a contract between individual landowners—
a short-term contract such as the ‘borrowing ‘ of land, or a long-term, or
indefinite, contract such as is embodied in tenantship and in the granting
of land for services rendered. Th e methods summarized under the first
category occur both in Trans- and Cis-Kaduna; of the methods sum•
marized under the second category only ‘borrowing ‘ occurs in Trans-
Kaduna; all acquisition and transfer of land for indefinite periods is
restricted to the ‘conquered’ area east of the Kaduna.
Village Land
Th e village land is administered, that is,held intrus t and apportioned
—i n Nupe terminology ‘owned’—by the chief. Linguistically the specific
character of his administrative right to the disposal of village land is not
defined very clearly. Th e Nupe refer to the chief’s control over land in
exactly the same terms whic h they use for the direct utilization and
ownership of land by a family head or any individual farmer. Th e Nupe
would use the same term also when speaking of the territory belonging
to a village, i.e. the land that is ‘owned’ by the people of the village collectively.1 Understood in the widest sense of the word , this village land embraces a territory surrounding the village in a rough circle. It comprises three kinds of land, each called by a different name: (1) occupied land fallow or actually worked, in the hands of families or individuals—lati, ‘farm’ ; (2) vacant land whic h has once been occupied
bu t has fallen back to the common fund of land held in trust by the chief for future allotment—mostly referred to as gonta; and (3) a fringe of virgi n land claimed by the village community as a land reserve—called cika, ‘bush’, or, again, gonta. Th e second and thir d kin d of land we wil l call ‘village land’ in the narrow sense, as against land of the first kind ,
‘family land’.
Th e border between the fringe of virgi n land which is regarded as tjhe land reserve of the village and the stretch of no-man’s-land which separates it from the territory of the next village is fluid and never rigidl y defined. There are no marks or boundary signs. But in the sparsely populated area of Trans-Kaduna this never leads to disputes between villages. Th e fluidity of this boundary is, in fact, an essential part of the elastic system of Nupe settlement, wit h its star-like expansion and founding of new ‘colonies’ which , step by step, extend the village terri • tory. Th e area of village land rarely stretches beyond a radius of three miles: in other words, the Nupe farms are situated withi n an hour’s walk from home. In some cases, where the farms are farther away, the farmers buil d small grass huts on the farms, in whic h they can find shelter in a rain or storm or spend the night if it is too late to walk home. Should the people of the village, in their search for new land, go beyond this easy reach, they woul d found new, permanent settlements, village
1 Th e Nupe would say, for example: Zitsu wu kin kpata, the chief owns all the land; or Dokocizi wu kin babo, the people of Doko own the land round here; but also Kolo-ga wu lati-o, it is Kolo who owns this farm.
184 ECONOMICS: RESOURCES
Colonies’ or tunga. Th e tunga thus possesses land in its own right , but at the same time pushes its territorial possessions further into no-man’s – land.
Access to Village Land
There is no general rule for the distribution, withi n the area of village land, of the farms belonging to the various family groups and individuals. Roughly one may say that the small individual plots (called bucd—a term to be explained presently) are situated near the village and the larger family plots (efako plots) farther afield. Whe n old men own plots of their own they, too, are given plots near the village so that they should not have to walk too far. Th e different plots whic h belong to the group of people livin g together in the same house (as a rule, a single family group), whether the plots are worked jointl y or independently by the various household members, do not necessarily lie together, but are frequently scattered. No r have the people who live together in the same efu their land on adjoining sites, or in the neighbourhood of the part of the village where they live. We may add that, again, there exist no special signs or boundary marks to distinguish the property of different farmers. A narrow path, a border of shrubs, or a tree, serve as landmark.! Bu t the different cultivation that may be practised on the different fields, or the different stage of cultivation whic h the different fields may repre• sent (fallow land, land that has been worked for one year, or tw o years,
&c) . make these boundaries more obvious to the eye.
Thi s irregular distribution of farm-plots appears to be due to a variety of causes. To some extent it is entailed in the complex agricultural tech• nique of the country, wit h its many different types of cultivation and its regular system of rotation of crops. Every farmer owns a certain number of fields—some under cultivation while others are lying fallow—which he tries to distribute over different sites so that he can utilize different types of soil, e.g. the marshy sites near watercourses for the cultivation of rice or sweet potatoes, dr y sandy soil for ground-nuts, heavier soil for yam and cotton, and so forth. Thi s arrangement also has the advantage of distributing more evenly the risks of failure. Yo u often find farmers complaining that one site on whic h they have a farm is less good than another, or has perhaps suffered special injury that year through birds or through thieving baboons, or has been washed away by a flooded river. Bu t the irregular distribution of farm-plots is also a result of the very method of Nupe settlement: it could not be otherwise wit h the farm-land lyin g in a circle round the large villages in whic h a numerous population is concentrated. There woul d not be room enough for all family groups in the village to have their lands arranged in compact blocks in equally favoured positions, both wit h regard to the distance from the village and to the quality of the soil. Moreover, some plots more heavily worked than others ‘tire ‘ sooner and must be replaced by virgi n land lyin g farther afield; and growing families must provide more land for their
1 A characteristic exception are a few marsh-land plots near Kacha, which are properly fenced in—’because of the many disputes’ that arise over this kind of land.
ECONOMICS: RESOURCES 185 adult members. Al l these factors have in the course of time contributed
to make the map of Nupe village land the irregular patchwork it is
to-day.
At the beginning of this section I have mentioned the possibility that a family head may have to request additional land from the chief for a junio r family member for who m he cannot find sufficient land on his own property. To-day, this is rarely put into practice, usually only in the extreme case when a whole section of the family, finding the avail• able land near the village insufficient or too poor in quality, sets out to colonize a piece of virgi n land and eventually to settle there as founders of a new tunga. In the normal case—especially in Trans-Kaduna— when land has to be found for an individual family member, the family head is as a rule expected to possess sufficient land to effect this readjust• ment. Th e family head, owning three to four different fields, each of about 4 acres, can, in fact, well afford it in a society wit h so low a birth • rate. Th e worst that may happen is that for a time the family head may have too much of his land lying fallow to be able to find additional land for the family member who desires a new farm. It is then that the method of ‘borrowing ‘ land is practised.
Le t us consider now more closely the position of the chief wit h respect to the question of access to village land. Control over the land is part of his political authority. It is, as I called it , an administrative control only, and this is expressed in a balanced, twofold, limitation of the claims to land of chief and community, of the power of the chief to apportion or withhol d land, and of the power of the family head to claim and keep land. As regards the former limitation, the chief cannot reasonably withhol d land from any claimant of whom the other elders of the com• munit y approve. No r is the chief’s own farm-land out of proportion to that of any other family head wit h an equally large working-group at his command. As regards the latter imitation , the control which the chief retains over all reapportionment of village land is acknowledged when• ever the main family land changes hands, that is, in the case of succes• sion to the position and property of a family head.1 Th e chief must be informed of all such changes, and in some parts of Nupe he summons at this occasion the family members wit h their new head and reminds them, of their obligations towards each other: the family members, that they wil l now work on the land of the new family head, and the family head that he has now assumed the responsibility to feed and clothe his people and, some day, to supply them wit h land of their own. Sometimes a gift to the chief of one pail of whatever crop is just ripe seals the pact. Th e annual village tax in kind , dzanka, which, as we know, was formerly paid to the chief by every family head, is itself a land-tax, a recognition of the chief’s control over the village land. It consisted in a gift of one’to three bundles of millet or sorghum, the amount varying wit h the condition of the crop. A complete failure of the crop meant exemption from the dzankd. Th e dzanka was thus not regarded as a fixed tax, but rather as
1 Th e succession to the position of family head being, in most cases, identical with the succession to a nusa rank, the procedure described here holds good for both (see p. 50).
186 ECONOMICS: RESOURCES
albarka (lit . blessing), a gift meant to express the ‘blessing’ that lay on the land whic h the family received from the chief.
Another expression of the chief’s control over the reapportionment of land is that all land of families or branches of families whic h have died out, and similarly the land of families who have emigrated, falls back to the common fund of land reserve administered by the chief. Th e period withi n which the land of emigrants stil l remains theirs is fixed everywhere in Nupe as three years. After that time the chief is entitled to reappor• tio n the land. Should an heir from the emigrated group return after a longer period, he could (like any other stranger) claim land in the village, bu t not as a rule the same land that his family had held previously. Th e dying-out of a family, or a branch of a family, seems a fact not easy to establish and always open to doubts in a social system in whic h small sections repeatedly split off the main body of the family. There is always the chance of an unexpected or belated claimant turnin g up, representing a distant branch whic h chief and elders may no longer have reckoned as likely to identify itself wit h the original family. But this is mainly a theoretical speculation; there occur few such cases in Nupe
—none in Trans-Kaduna, where land is plentiful and vacated plots unlikely to be much competed for.
In Cis-Kaduna I witnessed an interesting case of this kind . A family head in Doko had died, leaving a small boy, who was then adopted by a childless brother of the deceased, livin g in a tunga on a farm of his own . Th e boy grew up on this farm and lived there contentedly unti l recently, after more than twenty years, the chief reapportioned the apparently unclaimed land that had belonged to the boy’s father to two elders of high rank, the Lile and Shaba (both of the chief’s family). No t unti l then di d the young man return to Doko and make a formal claim for his ancestral land. But political motives rather than economic interest were behind this move; the parties in this case were, again, the chief of Dok o and young Ndak6 Gana, the ‘dispossessed’ descendant of the line of original chiefs. Th e chief had reapportioned the land in order to blot out the last visible connexion between the ‘pretender’s’ family and the chieftainship of Doko ; and the young man disputed the reapportionment on these very grounds. In the end he received back some of the original land and other new land in addition.
Here we discover a new significance in the apportionment of village land—its connexion wit h the bestowal of rank or office. Thi s is not sur• prising in the communities (such as Doko) where the village ranks, like the land, remain in the possession of one ‘house’. Later, we shall see more clearly, in the discussion of inheritance, that the Nupe kinship system tends indeed to couple attainment of family headship (which goes wit h rank) wit h inheritance of ancestral land. Considering, however, that the ordinary village land does not differ widely in quality and pro• ductive value, we can understand that this political aspect of land owner• ship is rather obscure—save in a case such as the one just quoted where political aspirations have attached to a specific property more than its proper share of symbolic significance. Th e political aspect becomes more prominent in connexion wit h a natural resource that shows this
ECONOMICS: RESOURCES 187 difference in quality and value to a muc h greater extent, namely, fruit –
bearing trees.
In most districts the chief has a special titl e to certain economically
valuable frui t trees wherever they grow, on vacant land or on people’s
farms. In Dok o and Shebe this special chief’s tree is the locust-bean, in
Kutig i and Wun a the oil-palm . In Mokw a there exists another regula•
tion , namely, that all shea-nut and locust-bean trees inside the town
belong to the chief, while the trees outside are the full property o f the
man on whose ground they stand. Also there is in many places the
general rule that no frui t tree on village land may be cut down withou t
the special permission of the chief. Sometimes family heads exercise a
restricted right to the’chief’s trees’ growing on their land, i.e. they return
two-fifths of the crop to the chief and keep the rest. Bu t the chief may
also grant the full possession o f these trees as a separate right ; thus he
bestows the property of a certain number of valuable trees on ah
individual—young or old—as a recognition of higher civic status, to •
gether wit h the promotion to rank and title . Th e righ t to the trees,
coupled wit h the rank, remains the property of the ‘house’, conditional
upon the renewal of the appointment (to the same or a similar rank) after the death of each holder.1 Th e shea-nut trees—to-day of considerable commercial importance—occupy everywhere in Nupe a special position. Th e trees on vacant land (virgi n village land or no-man’s-land) are common property. Th e women who collect the frui t go out in small parties and share the crop among them. Th e shea-nut trees on occu• pied farm-land belong to the owner of the farm.2 In either case the exploitation of the crop is the exclusive prerogative of the women.
A few words may be given to a special variety of village land, the plots that are situated in the village itself, roun d the compounds and between the village efu. Thi s land is owned by the people who live in the com• pounds adjoining the plots. It s inheritance and apportionment is ruled by the same regulations as the inheritance and apportionment of the houses themselves and wil l best be discussed i n a later section. Th e land inside village and tow n differs only in one respect from the village land without : it is never left uncultivated. Manure d as it is by the refuse fro m the houses it can be used continuously withou t the necessity of being left fallow for a period. If the owner of such plots leaves the village or the tow n and does not return for some time, leaving no one behind (which has happened frequently in Bida durin g the last twenty years), the neighbours wil l use the farm-plots; they are always ready,
x A certain complication arises when the modern official Village Head is not identical with the traditional chief, for the former may assume control over the land only, not over trees—he may indeed not possess any ‘chief’s trees’ himself, at least not at the time of his appointment. There is no evidence as yet to show how this contradiction will be solved in the course of time—except an indirect evidence to be mentioned presently (see p. 189, n. 2), which tends to show that this separation of land- and tree-rights is allowed to last for several generations.
2 Th e facts mentioned here will correct the statement made by some students
of West African agriculture, that ‘shea-nut trees are generally regarded as en• tirely common property’ (O . T. Faulkner and J. R. Mackie, West African Agriculture, 1933, p. 27).
188 ECONOMICS: RESOURCES
however, to give them up as soon as the former owner or his heir may return, even if he had been absent for a longer period than three years— the period of absence stipulated in the village. I have heard of no dis• pute that has ever arisen over these house-plots, either in the village or i n Bida town .
Strangers as Settlers
Fro m the land whic h the chief holds in trust he apportions land on request to strangers who come to settle in a Nupe village and equally, as we have seen, to natives of the village who, through long absence, have forfeited the claim to their family land. Th e procedure is the same for both. It is interesting to examine these regulations, especially because this for m of granting land implies a previous adoption of the applicant into a local household. Generally speaking, the Nupe like strangers— respectable strangers—to settle in their villages. Thi s fact reflects both the lack of pressure on the land and the attitude of a commercially highly developed country whic h is ready to welcome new citizens, very likely people who brin g wit h them new techniques and new valuable know• ledge.1
Th e stranger who decides to settle i n a Nupe village wil l first stop for some time in the house of a family head, perhaps a friend, or friend of a relation, who is ready to accept hi m as a temporary member of the house• hold. Be i t a short or prolonged stay, strangers wil l never be expected to pay for their ‘lodgings’. A small present of food or kola-nuts is the only compensation. Th e guest wil l keep his own minage. Bu t after a time he comes to be regarded as an adopted member of the house, and then it is time to make his request for land and a site for a house of his own. Thi s request wil l always be done through the emitso, the Owner o f the House, who wil l brin g the case of his guest to the notice of chief and elders. Th e emitso wil l have to vouch for the character and reputation of the appli• cant—a bad reputation woul d invariably lead to a rejection of the application. Th e procedure (which varies slightly in different places) is this : a preliminary gift, called emisa, ‘salute’, consisting ofte n to twenty kola-nuts, a fowl , a calabash of beer, or—more rarely—money, accom• panies the formal request to the chief. Another gift of kola-nuts is made later to the chief’s representative (one of the elders), who wil l go wit h the applicant to stake out the farm-plot and the site for the house, and who may become later an important witness if any dispute should arise. No w the new landowner sets to work. For two to three years he is exempted from tax. Then , when the crops have been good, he pays his first dzankdy not yet to the chief himself, but to the family head in whose
1 We have mentioned evidence of this already, e.g. the Born u settlers wh o have introduced onion-growing into Kutigi , or the Yoruba-Kom i who introduced weaving and indigo dyeing in Yoruba fashion. At the hearing of a court.case—it was a rather involved dispute over a house in Kacha that had been ‘sold ‘ by the late owner in order to obtain money for tax arrears (the only case of this kin d that has ever come to my notice), as a result of whic h his wido w had now no place to go to—the Etsu’s Cour t officially recommended that ‘strangers should always be adopted into one’s household and given a hu t to live in’ .
ECONOMICS: RESOURCES 189 house he had been staying before, and the latter returns half of the gift to the chief. Th e next dzanka wil l be the ordinary land-tax of every
member of the village community, presented to the chief directly. To •
day the first part of the procedure has remained unchanged, except that
the application (and sometimes the presents as well) may also be addressed to the Distric t Head.1 But the annual dzanka is now replaced by the Government tax.
Th e grant of land to strangers never includes the trees growing on the farm-land. The y remain the property of the village chief or the other native inhabitants, and this makes the analogy between strangers and adopted household members full : for like the strangers junio r family members cannot own trees, although they hold land of their own.2
Access to Family Land
Thi s question wil l be examined in detail later, and no more than a short summary is needed here. Three methods regulate the access of the individual member of the family group to the family land held jointl y by the group and administered by the family head. (1) Temp • orary and restricted access to this land is embodied in the right of every male member, boys as well as adults, to a small plot of his own, which he can work independently. Such plots are called buca, and the individual work on the plots is called buca work. (2) Adul t male members of the family group, who have already founded a family of their own, may claim a large farm of the kin d that is worked jointl y by a number of family members. Thi s farm and farm-work is called efako. (3) Th e fullest access to family land is embodied in inheritance and in the succession, to the position of family head. Thi s position implies control over all family land that the family head holds in his hands. Th e first two forms of access are ‘family affairs’ arranged with • out any formality; the last transaction implies, as we have seen, a formal reference to the chief. Th e chief’s control over the occupied village land is thus made conspicuous not wit h respect to land that is apportioned on a small scale to individuals for direct use, but wit h respect to the land that is to be held in trust for future distribution-by the new family head. Th e chief, in other words, as he is bound himself to trusteeship, guarantees the trusteeship undertaken by others. Th e analogy in the position of chief and family head wit h respect to land is clearly recognized by the Nupe. Both are heads of corporate groups, and both administrators of the join t rights vested in the groups. Both, moreover, hold the same special, economically significant, right to trees.
1 I am not sure if this is really prescribed by the regulations. I know that in Agaie Emirate all peasants thought it was. In any case it woul d save a lot of trouble, should disputes arise later, if one had the Distric t Head for a witness of the original transaction.
2 Thi s rule also holds good wit h respect to large groups of alien settlers. In
Kutigi , for example, it led to the anomalous position that the Benu settlers, although they had become the political rulers of the communit y and supplied the chief of ‘Greater Kutigi’ , owned no trees at all in the beginning, and stil l own very muc h less than the family heads of the kintsogi, the Owners of the Land , that is, the original inhabitants.
190 ECONOMICS: RESOURCES
No junio r family member, even when already holding efako land in his ow n right, may claim any share in the produce of the fruit-bearing trees. Onl y succession to the family head’s position implies the usufructory righ t to the trees that stand on his or any other parcel of the family land.
Borrowing of Land
Lan d can be borrowed only for limite d periods, as a rule for periods fro m two to five years, that is, in order to help a temporary shortage of land. Th e family heads or other ‘landlords’ arrange the transaction between themselves, not necessarily informing the chief, bu t always in the presence of witnesses who should be able later to testify to the nature of the transaction. Every landowner who has enough land is ready to
‘lend ‘ it in this way. In Trans-Kaduna the transaction is regarded as a friendly act, a ‘helping out’, quite natural between neighbours or fellow villagers. Th e borrower pays annually a fixed rent in kind , as a rule one bundle of mille t or sorghum. In Cis-Kaduna the transaction gains a definite commercial significance: the borrower pays a small sum of money down, varying wit h the quality of the land (from Sd. for about one acre of ordinary land to 2s. for a much smaller plot on valuable marsh-land—the highest amount paid I found to be 6s.), and again the annual rent of one bundle or the equivalent in money.1 In one case I found that regular farm-work for the owner of the land had taken the place of the annual rent in kind .
A landlord is free to cancel the lease before it has lapsed if the tenant is not working the land—as happens frequently; the court woul d con• firm his right to do so, only ordering hi m to return the initia l payment. But it may also happen that the landlord dies or emigrates while the lease is stil l running, or that for some other reason the lease is not recalled. In such cases the lessee often continues to work the land, although he has naturally ceased to pay the annual rent ; he even comes to regard the land as his own whether he is working i t or not, til l one day it may be demanded back by the original owner or his heirs. In Trans- Kaduna, where no long-term transfer of land exists, any evidence as to the nature of the original transaction wil l at once settle the case. I n the conquered areas of Cis-Kaduna, where land is let and sublet, and where short leases and leases for an indefinite period coexist, it is often difficult to establish clearly whether temporary or permanent lease had been arranged originally, specially when the witnesses of the original contract have died. What type of lease was originally intended is shown by the amount of the initia l payment, whether it was high or low, that is, a ‘real’ payment, or only a kyuta, a gratuity. Th e latter, combined wit h a re• peated rent in kind , establishes the nature of the temporary lease, the ji aro, ‘borrowing’ . Bu t the continued working of the ‘borrowed’ land, and the failure of the original owner or his heirs to call in the lease, clearly alter the situation considerably. Th e attitude of the court in such
1 Characteristically this annual rent is called in Bida—quite logically—dzankd
‘tithe’, but in the villages kyuta, ‘present’.
ECONOMICS: RESOURCES 191 disputes is this : if the original owner or his heirs had allowed the rent to
lapse withou t calling in the lease, and if the other party can prove that he
has worked the land for a long period, this is generally considered
sufficient evidence that the original owner has abandoned his claim to
the land, whether the new owner is at the time actually working the land
or keeping it as reserve. Lon g occupation is thus recognized as tanta•
mount to ownership. In summing-up a land case in 1935 the Etsu
Nupe said: ‘W e do not agree that a man should be evicted from the
place whic h he has been workin g for a long time.’ An d in another case,
heard in 1934, the claim of the new ‘ borrower’ to the land was confirmed
by the court ‘because of the many years the farm had been in so-and-so’s
hands’.
Litigatio n about ‘borrowed’ land comes up constantly before the court in Bida. Thi s is a typical case: A. suddenly demanded back tw o farms whic h his grandfather and father had ‘lent’ to another man, because A. was going to be married and needed the land for himself. He based his claim on the fact that only aro had been intended, although the rent had been allowed to lapse. Evidence of the original transaction was difficult to obtain, so the court suggested the following compromise: the transfer of the farm of A.’s grandfather was confirmed, but the farm whic h A.’s father had ‘lent’ to the other man the latter was ordered to return—a decision whic h is again based on the length of occupation. Th e lack of precision in the concept of land-borrowing in Bida and Cis- Kaduna is illustrated stil l better by the following case, not’itself a court case, but, as it were, a typical potential cause for such land-disputes. A certain ric h Malla m had ‘borrowed’ land fro m the Cece, a bar a of L. , a nobleman of high rank. He had paid 8d. down, and every year one bundle of sorghum as rent, for five years. Th e Cece died in the mean• time, and the present owner maintained in discussion wit h me that the land Was his and could be inherited by his sons after his death. Whe n I discussed the case later wit h the Alkali he confirmed my suspicion that this was not so; the land could in theory be recalled any day after the five years by the heirs of the Cece. But the Alkali admitted that this may not be the case for ten or even twenty years; by then the rent woul d have been allowed to lapse and the details of the original transaction woul d be difficult to establish.
In the course of the Fulani conquest certain farms whic h were at the time withou t a definite ‘owner’, the original owner having emigrated before the lease had lapsed, and the lessees having ceased work on the borrowed farms, were simply appropriated by the Fulani landlords withi n whose territor y this ‘ownerless’ land was lying .
Trees are included in this short-term lease or ‘borrowing’ . Bu t the strength of this regulation, in harmony though it is wit h the general principles of Nupe land-rights, has been weakened under present-day conditions. Economic stress has made fruit-bearing trees a very valuable resource, and farmers have in a few cases ignored the traditional practice, given away trees wit h the land, and even ‘lent ‘ trees independently of the land they stand upon. Thi s sort of temporary transfer of trees is a simple means of obtaining money quickly (for example, for the payment
192 ECONOMICS: RESOURCES
of tax) : the owner of oil-palms or locust-bean trees woul d sell a year’s yield of his trees to another person, who in return woul d pay the tax for him . Th e issue is stil l more confused where the transfer of the land wit h whic h the trees go appears to be in the nature of a lease for an indefinite period. In this case the original owner or his heirs could claim that the trees, at least, were essentially inalienable, and could be transferred only temporarily, that is, coul d only have been ‘lent’ and borrowed’ . Thi s is itself a compromise, and the attitude of the court in these disputes constitutes a further compromise. In 1934 a farmer had ceded a farm-plot wit h locust-bean trees on it to another party. After the death of the original owner his heirs reclaimed the trees (though not the land) on the plea that the transfer of the trees could in the nature of things have been meant only as a temporary transfer. Th e decision of the court was that the new owner should share the crop, fifty-fifty, wit h the former owners.
Long-term Lease and Purchase
We come now to forms of land-ownership whic h are typical of Bida and the conquered areas east of the Kaduna and whic h are almost com• pletely absent in the Trans-Kaduna districts.1 Amon g the people of Bida tow n who derive their livelihood from the land, a few descendants of the original Beni population stil l ow n land under triba l conditions, that is, under the conditions described under the heading of village land. Thei r farms are scattered over the land around Bida—the village land that used to belong to the old , pre-Fulani Bida. But most of this land had been expropriated early in the conquest, under one or other pretext, and the former owners and their descendants had left Bida and moved into the country to other villages or tunga settlements, where they found vacant land. Thu s of the descendants of the Beni ‘town-kings’ of Bida only Ndakotsu, the eldest of a number of brothers, had remained in Bida and continued to work the small plots that he had been able to keep; the other brothers had left the tow n and settled in the country.
Lon g absence, or the dying out of a family branch, too, gain a totally different aspect here in Bida, as compared wit h conditions in the village. For, the vacated plot is as a rule at once appropriated by one of the bi g landlords whose lands may surround it , or possibly by one of the three royal houses on whose territory the plot may lie. Th e latter procedure was, as we shall see, the practice in pre-British times and in the early days of British rule. A somewhat similar case happened durin g my stay in Bida: a Bida man had returned to the tow n after a long absence and, claiming his ol d land, found that it had gone to another man. According to Nupe customary law he had forfeited his righ t to the land, having been away for such a long period. However, the customary law is not adapted to cope wit h conditions as they obtain in the tow n and in ‘ con• quered territory’ : for here no spare land exists that could be offered to
1 Th e exceptions are a few tunga settlements which some Bida noblemen have founded outside the ‘conquered territory’, for example on the Wuya-Kutigi road. There are no tunga of this kind west of Kutigi.
ECONOMICS: RESOURCES 193 the claimant in compensation for the land he had lost. Besides, the
man’s old land had changed hands several times in the meantime, and
could not simply be taken away from the present owners who had spent
work and money on it. The man was unwilling to move to a village,
where he could find land, and unwilling, too, to take land in Bida for
which he would have had to pay a small rent. He felt that he had been
wronged, and being of a rather violent temper and a slightly unhinged
mind he attacked the judge and threatened the Etsu, and his search for
a new place where to live ended under lock and key.
The village ways of obtaining and securing land, rendered impossible
by the social conditions obtaining in the capital and in what we have called
the ‘conquered territories’, are replaced by a large variety of other, much
more complicated—and costly—methods of land transfer. All land is in
fixed hands; land cannot simply be claimed on the grounds of citizen•
ship from the Etsu Nupe. Nor is land around Bida, especially the valu•
able marsh-land, often given away in short-term lease. But land can be
acquired by the method of lease for life or of tenantship for an indefinite
period. In Bida a large number of dependants of the royal household,
fief-holders, former warriors, slaves, or Mallams who were attached to a
noble house, possess land granted to them by the king or a powerful
feudal lord. These landlords, usually not farmers themselves, are
frequently prepared to let their land (or sometimes to sublet it if they
themselves have obtained it by a similar arrangement) to men who are in
need of land for their farms. If the landlord is of high rank the trans•
action is never carried out by him personally but always by an
agent, who expects a small gratuity from the tenant for his services.
The rent itself consists in a comparatively high initial payment in money
—£1 to £2—and, again, the fixed annual rent in kind. M. , a wealthy Bida
man, for example, acquired two years ago a farm for his newly married
son from a former slave of Etsu Mohamadu. He paid £1 in money,
and undertook to pay an annual rent of one bundle of millet. The
money was pocketed by the former slave himself, but the rent in kind
went to the family of Etsu Mohamadu—a sign that the former owner
was himself only a tenant (though a tenant of a different category). This
transaction is definitely spoken of as sale and purchase of land. The
lease itself is, in fact, ‘sold’ and ‘bought’; for the original tenant has ceded
all his rights, for a single payment, to the new tenant who takes over the
lease from the (real) landlord. The concept of sale and purchase is
emphasized further by the practice of the court rarely to bring up the
matter of the annual dzankd in kind (although it is almost invariably
implied in the transaction). The initial, single payment, and the nature
of the right to the land which this payment entails, are the only points
with which the court is concerned. The importance of the annual rent
in kind, however, apart from the ‘purchase’ price, reveals itself in the
practice of recalling the contract if the tenant leaves the land unworked
(that is, is unable to pay the annual tithe). Also, if the tenant has for
some other reason discontinued the tithe, the lease may be recalled
(though here, as we shall see presently, modern court practice would
not sustain the claims of the landlord).
o
194 ECONOMICS: RESOURCES
Th e lease is only for a lifetime—at least in theory—and has to be renewed by another payment in money if the heirs of the tenant intend to keep the land. If tenants and landlord are on good terms, or are social equals, or if the landlord is under some obligation to his tenant, the renewal is a mere formality and may be granted for a small gratuity. Thu s tenants often feel justified in regarding their righ t to the land as implyin g absolute ownership, ownership har dbada, ‘for ever’. Bu t the large number of disputes of this kin d whic h come before the court prove that the arrangement is open to misinterpretation as well as misuse. Landlords who for any reason are dissatisfied wit h their tenants, or have found a higher bidder, may unexpectedly call in the lease, either at the death of a tenant or durin g his tenancy, or demand on the threat of eviction another, higher, payment from the tenant or his heirs. Here are two examples: A. , a ric h landlord, had ‘sold’ the same farm twice, first for 105., and afterwards for 16s. E. transferred a farm-plot that had fallen vacant after the death of the tenant twice in two years, pocketing
£2 altogether.
Lik e the practice itself the attitude of the court towards these cases of
‘purchase’ or indefinite lease is ambiguous and uncertain. Only in one
respect the issue is certain: all evictions from planted land, between the
times of sowing and reaping, is declared illegal on whatever grounds the
eviction may be based. Every man is entitled to the fruits of his im •
mediate work. As regards the wider issue the tendency of the court
seems to be this : if the transfer of the lease has taken place fairly long
ago, the fact of the long occupation (and actual working) of the disputed
land is taken to entail permanent ownership; further payments of
dzankd rent are discouraged, and thus the purchase of the land is
recognized as an accomplished fact. (As already said, in cases of this
kin d an original short-term lease—’borrowing’—is often confused wit h
transfer for an indefinite period or for lifetime.) In other cases, where
the transfer took place more recently, the court suggested the alternative solution of a return either of the money or the land.l In one or two cases of this kin d the Etsu even made it clear that he wished to discourage sale of land for money altogether. In summing up a land case that came before the court in 1934, he said that ‘i t was righ t to divide up land for the benefit of people who had insufficient land ; but this should not be
done for money’—a pronouncement which , though morally admirable, does not quite do justice to actual facts. For how, one might ask, can land be ‘divided up ‘ for the needy in an area where such strong pressure on land, and such multitude of vested rights in land, obtains, if no allow• ance is made for some sort of compensation ? In the Nupe economic system, firmly based as it is on money, a monetary compensation, sale and purchase, seems the only solution, unless the Administration is pre• pared to repudiate the whole system of land ownership as it exists in the
‘conquered territory’ as well. 2
1 This decision, however, made no allowance for the diminished value of the land, in the sense that the man to whom it was returned might have to leave it lying fallow for a considerable period before being able to cultivate it again.
2 In one dispute of similar nature, which came before the court in 1935, the
ECONOMICS: RESOURCES 195
I t wil l have been seen already that the land disputes whic h we have jus t described reflect more than merely disputable, equivocal land- contracts. Th e landlords who ‘sell’ the land are withou t exception men of the title d classes, courtiers, favourites, or former slaves of the king or of powerful noblemen, and the ‘ buyers’ mostly commoners, peasants who have had no other chance of obtaining additional land. To-day they are coming to realize that their righ t to the land whic h they acquire is protected by the Native Court. Bu t til l very recently, i n order to obtain their rights, they woul d have had to bribe their way through or to find a protector equal in rank and influence to the landlord who threatened to evict them—a typical source of the institution of clientship, whic h we discussed in a previous chapter.1 A final example. A certain farmer had
‘bought’ land from D. , a bora of the late Etsu Saidu. Th e Etsu, for reasons of his own, refused his confirmation of the deal—D., being a tenant of the Etsu, could not transfer the lease withou t the latter’s per• mission—and the tenant had only one way open: he attached himself to another, more influential, bar a of the Etsu and through hi m made various gifts to the Etsu, over £3 in money, a valuable civet cat, and promised besides an annual rent of two bundles of corn.
Hereditary Land: Secondary Tenantship
Clientship, combined wit h a metayer arrangement, becomes a factor in land-tenure also in another respect. A commoner who desired more land, or a new piece of land, could give himself into the service or
‘ patronage’ of a landowning overlord and thus secure land as well as political protection. Th e conditions were: an initia l gift in kin d as in all pacts of clientship, occasional presents, visits and other expressions of homage, and finally a tithe (dzanka again) from his mille t or sorghum crop. Thi s submission to patronage was often the only means by whic h the peasants in Cis-Kaduna could hope to retain their property when threatened wit h expropriation, and also the only means by whic h they could secure new land, that is, access to what under triba l conditions woul d have been a reserve of village land. Th e necessity of obtaining land and enjoying its secure possession thus constitutes another of the main roots of clientship. Th e lease of land obtained by these methods is hereditary, and remains in the family, a comparatively safe possession— unless political factors interfere. Th e tenants are rarely allowed to for• get that the relationship between landlord and client implies more than merely a correct commercial relationship. Lack of respect, an apparent failure to render due homage, or behaviour suggestive of changed
Etsu himself suggested that the disputed piece of land (a river grove that was claimed by two neighbouring villages) should be bought for money by one of the parties. Th e purchase price fixed was 155.
1 Til l very recently no peasant tenant would have dared to invoke the Etsu’s or
Alkali’s help against his landlord ; he would have been looked upon as somebody who had broken away from a firmly established tradition of loyalty, one-sided though it was. Th e tenant in the last example quoted above kept silent for fifteen years; not till 1934 did he venture to bring his case before the court and thus to obtain a final confirmation of his ownership.
196 ECONOMICS: RESOURCES
allegiance, suffice to make the landlord refuse a renewal of the tenancy or even to cancel a running tenancy. Several cases of this kind have come to my knowledge; one even found its way to court: L. , a nobleman of high rank, evicted his tenant (and this in the middle of the planting season!) because he had lacked in loyalty towards him and ‘ betrayed his confidence’.
Political changes have brought about an interesting development. To-day the initiative in this kind of tenantship has to some extent passed from the tenants to the landlords themselves. Th e landlords were, as we have seen, mostly warriors and courtiers who, living in the capital, never worked their land themselves, but settled their slaves on it, and had it worked by their labour. When slavery was abolished and wars came to an end, when the ‘absentee’ landlords no longer needed bora as soldiers for their private armies and could no longer support a large household of henchmen and hangers-on in the town, they offered their land to their bara, and so new settlers went out into the country, occupying the land and founding new farm settlements. Sometimes the noble landowners even settled some of their relations in this fashion on the land, nephews and younger brothers, for whom no place or job could be found in town.
Al l these farm settlements, whether they were for slaves or bara, for peasants turned client or clients turned peasant, are called tunga—the same name that is applied to the Nupe village ‘colonies’. But the ‘private’ tunga has no name of the kind villages possess, but bears the name of the noble patron to whom it ‘belongs’. These tunga settlements are scattered all over Cis-Kaduna, and surround Bida in a dense belt. There are in• numerable hamlets called Sokyara, Maiyaki, Ndeji, or similar titles of nobility—the titles of the landlords under whose patronage the settle• ments were founded. To our description of the village tunga given in a previous chapter we will add here a few examples of the ‘private’ tunga.
Etsu-Kolo (about four miles south of Bida), consisting of two compounds, two families which are distantly related. Called after the founder of the tunga, the father of the present family head. This Etsu- Kolo had been, like his father before him, a bar a of the Cekpa in Bida. Later, however, Etsu-Kolo transferred his allegiance to the then ruling Emir, Etsu Mohamadu. The change of allegiance proved good policy: Etsu Mohamadu procured for him the position of a Village Head for the whole district with its many tunga settlements. Since then the family had remained faithful to Etsu Mohamadu’s house, even after the Village Headship had been abolished again; they are still paying their regular dzanka to his successors, at present to the Shaba in Bida.
End Bwori (‘House of Bwori’, one mile north of Bida), called after a certain Bwori, the head of the guild of Bida butchers, and a bar a of Etsu Umaru. This Bwori settled his own slaves in this tunga. The present owners of it regard themselves as bara of the descendants of Bwori, but have abandoned paying dzanka to them; but they still go to ‘salute’ them whenever they come up to town.
Lumanen (four miles west of Bida), a group of very prosperous tunga settlements’ founded at different times by relations and noblemen attached to the royal house of Etsu Masaba. One settlement has been built by the
ECONOMICS: RESOURCES 197
father of the present Maiyaki for his bara, and is accordingly called Maiyaki, A second settlement is called Emi ‘tsu Saba, and was founded in the time of the father of the present family head, who was a slave in Etsu Masaba’s house. A third settlement was built recently, by the present family head, who was a bara of the present Etsu Saidu (also of the Masaba dynasty). All three settlements still pay their regular dzanka to their patrons. This year’s dzanka amounted to one bundle of corn and ten yams.1
Maku (one mile south of Bida), consisting of three compounds, called after the noble patron of a certain Ndatazi, a peasant who had had a farm somewhere else first and then moved to the present site, obtaining the land with the help of the Maku. He lives in one of the compounds; of the other two one belongs to his brother, and one to a distant relative by marriage, who left his old farm and came here, where the land is better, also be• coming a bara of the Maku. Each of the three houses pays an annual dzanka to the present Maku. It amounted this year to three bundles of corn and ten yams per compound.
Saci (three miles east of Bida), called after the Saci of Bida (a nobility rank). It consists of five compounds; in one a son of the Saci is living— the other members of his family are living in Bida—in the other com• pounds bara of the Saci. They have discontinued paying their dzanka, but still pay regular visits to their patron in town.
One word about the economic arrangement obtaining in the tunga. Th e slaves had worked the land entirely for their masters; that is to say, they had no efako (family) land of their own, but were fed and clothed by their masters, and were also allotted smaller, buca plots, which they were free to work for themselves. The arrangement between the landlord and the bara whom he settles on his land is the same metayer system which also regulates the tenantship of the free peasant who accepts clientship in order to obtain land. To-day, the difference between former slaves and bara has disappeared; for to most of the freed slaves their change of status meant only a promotion from slave to (free) client. Both, in a good many cases, still acknowledge this form of dependence and, as we saw, still return the annual tithe from their crops to the patron, or at least still pay their respects dutifully. Th e dzanka is no longer a com• pulsory tribute, and would not be upheld by the court. It is a voluntary gift which confirms the relationship of patronage and protection—a relationship which is as real to-day as it was in former times: political protection for oneself and one’s children is bought cheaply with a bundle or two of corn a year.
Hereditary Land: Primary Tenantship
Th e acquisition of land through clientship follows the general pattern of the Nupe feudal hierarchy. A man who obtains land as a bara obtains it from a man of higher rank, either a royal prince or nobleman who holds the land in his own right, or from a rank-holder or titled slave or a king’s bara who has himself obtained the land from another landlord higher up on the hierarchical scale. These ‘primary tenants’, however, have acquired their land by a different method, namely, through certain
1 This census was taken in 1934, when Etsu Saidu was still alive.
198 ECONOMICS: RESOURCES
services which they have rendered to their overlords. Ultimately all land is derived from one source: the royal house, whi,ch again bestows land for services rendered. Th e title of the royal house to the possession of land is by the right of conquest. We must examine this more closely. In the conquest of Nupe country the members of the royal family appropriated three categories of tribal land: all no-man’s-land between villages, i.e. virgin land in the cika, the uninhabited bush; also all village land which, once occupied, had been lying fallow for long, either because of the death of the owner or his emigration, or because a short-term lease (borrowing) had lapsed; finally, under certain conditions, the Etsu also seized the village land on the outskirts of the village territory, what we have called the ‘land reserve’ of the village. Th e first two kinds of land were divided up geographically between the three royal houses, as we have already seen, one receiving all land to the east of Bida, one all land to the west and north-west, and the third all land to the south of Bida. (See sketch-map, p. 81.) The last category of expropriated land (the village
‘reserve’) was sometimes appropriated by a ruling Etsu arbitrarily and irrespective of its geographical position in one of the three royal terri• tories.1 Of the land thus owned by the royal houses by the right of conquest the Etsu or royal princes of highest rank bestowed smaller or larger parcels on their followers, members of the nobility or slaves and bar a ‘for services rendered’; the land grant gave them complete and permanent rights to the land, in return for which they had to pay the usual annual dzanka in kind. Th e royal tenants—’primary tenants’ as we have called them—then settled their own men in various ways on this land, either founding tunga settlements for them or subletting the land (or part of it) in any of the ways discussed above. Every landed property which we find in the ‘conquered territory’, excepting only the scattered islands of original village land, has thus been derived directly or indirectly, that is, through the hierarchy of feudal nobility, from one of the three royal houses of Nupe.
Apart from these possessions which were given away to rank-holders and henchmen the royal families have kept certain royal estates proper. This land is not transmitted in the ordinary way of inheritance to younger brothers or sons, nor is it alienable like the other lands appro• priated by the royal houses, but can be claimed only by the three titled heads of the royal dynasties. Right to this special royal land is thus part of the promotion to rank and status within the dynasty and the political system at large. It implies absolute and permanent ownership; it implies more than merely the ‘administrative’ control over family land that is vested in the family head in the Nupe village. Yet it reflects the most essential element in Nupe peasant ownership of land—what I have called ‘access’ to the land in yirtue of headship of the kinship group. This surprising analogy between feudal monarchy and peasant ownership is rooted in an analogy in the structure of the two groups as such, village
1 Thus , for example, all the land in the district of Jima, south of Bida, be• longs to the house of Umar u Majigi . Bu t the village land of Kucigi , a village in this district, had been appropriated by Etsu Malik i (o f the house of Usman Zaki) when he was Etsu Nup e and bestowed on one of his slaves, the Cece.
ECONOMICS: RESOURCES 199 kinship group and royal ‘house’, different though they are in many other
respects. Th e possession of land is only one instance of the joint, ‘cor•
porate’, rights which characterize the two analogous groups. But of this
we shall say more later.
Royal Estates
Th e three royal estates are situated immediately outside the town walls of Bida. On each of the estates we find a small hamlet, called eso£i, where (former) royal slaves are living, who were appointed to work the land and look after the royal property.1 Each of the three eso£i is placed under the headship of a head slave, styled Sokyara—a titled overseer of the royal estates.
The first cultivation in the year—the clearing and first hoeing—was always done by large-scale communal labour in which over a hundred people used to join—slaves, bar a, peasants from the neighbourhood— and which lasted for one or two days. The Etsu himself arranged for food and drink (sorghum beer) and for the customary band of drummers and flute-players to play during the work and at the dance afterwards. After this first cultivation the care of the land was left entirely to the slaves. Originally the whole yield of the royal estates went to the Etsu and to the other heads of the royal houses. The esozi used to supply over a hundred bundles of corn annually. The royal landlord gave his slaves in return food and clothing, and cared for their various needs. Th e married slaves (as everywhere in Nupe) were given fairly large individual farm- plots of their own which were not only sufficient to provide food for their families but as a rule also allowed some surplus for trade.
Under modern conditions the position of the royal slaves in the esoii has been assimilated to that of clients who only pay an annual tithe from the produce of their farms. But the different Emirs dealt differently with the question of adapting the management to modern conditions. Etsu Mohamadu, who was not very interested in farming, abandoned the communal work of the first cultivation; he also lost all personal touch with his esoziy and insisted only on the annual dzanka. Etsu Bello, on the other hand, maintained the communal work of the first cultivation; he demanded one-third of the total produce as ‘tithe’, leaving the rest to the people of the esozi. Under the late Etsu Saidu and under the present Etsu every family head in the esoii paid one to two bundles of sorghum or late millet to the royal landlord, the Sokyara acting as intermediary— the total tithe amounting to about ten to fifteen bundles a year. When the Government built the new residence for the present Emir a final change took place. Th e land outside the Wuya gate of Bida, which for• merly belonged to the house oiEtsu Masaba, was bought by the Govern• ment as a site for the new building, and turned into Crown land.2 The
1 Durin g the dr y season the three royal estates, being situated directly outside the tow n gates of Bida, are the natural camping-ground for the donkey and cattle caravans that pass throug h Bida. No w it is the customary practice that these camps—rather primitiv e grass-shelters erected by the people of the Etsu—are always pitched on the farm-land belonging to the rulin g Etsu. Hi s farms have thus the benefit of the well-manured soil.
2 In accordance wit h Nup e practice the fruit-bearing trees were not included
20 0 ECONOMICS: RESOURCES
present Emir, who comes from the royal house of Umaru Majigi, thus owns, unlike any of his predecessors, two royal estates: his own family estate near the Jimagate, south of Bida, and the new estate on the Crown land near the Wuya gate, which, like the residence itself, belongs to the modern accoutrements of his office. Th e esozi people on his family estate were prepared to work for him on the usual dzanka arrangement. But the esozi of Etsu Masaba—the esozi on the land that had now become his official estate—refused to work for their new landlord on the old terms, and demanded money wages. Finally, when the Etsu in 1936 started an experimental farm for mixed farming on his family estate, the esozi workers there also denounced the old agreement and demanded wages; for this was work of a new kind altogether, work carried out entirely for their landlord, no longer work after their own fashion, from which they could hand over the traditional share in the produce.
Non-agricultural Use of Land
A few words must be said in conclusion about rights to land that is used for other than agricultural purposes, such as clay-pits from which clay is obtained for pottery or for the sun-dried bricks used in house• building, the iron-ore deposits which are used by the native iron-ore smelters, or the space needed for the dye-pits of the indigo dyers. Al l these stretches of land, whether they are situated close to the house (like the dye-pits) or some distance away (like the clay-pits and the ferruginous hill-side) are accessible to all.
Th e thick strips of rain-forest near the rivers are specially valuable
‘no-man’s-land’. Here woodcarvers and carpenters find the timber they
need for their work. Th e branches of certain palm-trees are used for
rafters in roof building; a certain species of reed is used for the making of
the curtain-like mats which the Nupe call eshegi (the Hausa asabari);
other species of grass, and the leaves of the dum-palm, are used for mat-
making and the making of straw hats. Th e fact that these materials are
found only in certain places, far away from the town where the craftsmen
who are using them are living, is turned to economic use, and the pro•
curing of the raw material thus gains independent commercial signifi•
cance: the big palm-branch rafters are sold on Bida market by the men
who cut them; and men and women frequently use the time of com•
parative leisure in the dry season to travel down to the river in order to
collect dum-palm leaves and marsh-grass, which they sell later, for a
little money, to mat- and straw hat makers.
Th e collecting and selling of firewood (which is done by women) and the cutting and selling of grass for roofing (done by men) belongs to a slightly different category. For in this case the non-agricultural use of land does not apply to ‘no-man’s-land’; the farmers and their families
in this transaction and remained the property of the original landlords, i.e. the head of the house of Etsu Masaba. Thus in January 1936, when the new official residence of the Emir had been standing already for almost a year, people from the house of Etsu Masaba were collecting the fruits of the trees on the new Crown land.
ECONOMICS: RESOURCES 20 1
themselves make use of these secondary resources which they find on their land.
We may mention, finally, the various uses to which uncultivated land in the town—streets and open spaces in Bida or the larger villages—is put: women may put up little ‘kitchens’ by the roadside, where they cook food and sell it to passers-by; traders or craftsmen may build themselves a small hut to be used as workshop or booth; or neighbours may get to• gether and build one of the little mosques on the open space between their houses, which you find scattered all over Bida town. No permission is necessary for these modest buildings. For larger enterprises, a big mosque, or a market, the Etsu in Bida or, in the village, the chief, would himself allot the necessary space. This practice includes modern re• quirements as well: thus Etsu Mohamadu, who had shown great sym• pathy for the work of the C.M.S . Mission, gave them a space for building a school and a compound inside Bida, in his own ekpa of the town, free of charge.
LIVE STOCK
Cattle-keeping plays a very secondary part in Nupe economics. This is shown most clearly in the very low per capita figures quoted in the statistical table at the end of this section. To-day cattle are owned only by the butchers who buy for the kill from the Bornu and Fulani cattle- traders who, with their huge herds, pass regularly through Nupe and Bida, on their way from the north and the south. I say ‘to-day’, because formerly many of the members of the Nupe aristocracy and the more wealthy among the Bida people used to own large herds of cattle.1
Slaves taken from the Bororo-Fulani, the nomadic Fulani herdsmen, looked after the cattle. Th e cattle were always kept outside the town, frequently a long distance away; for they had to be moved from pasture to pasture according to the season, up north, away from the tsetse-infested area near the rivers, in the rains, and south again in the dry season. This meant, obviously, much trouble and cost, and this is mostly quoted as the reason which made the Nupe eventually abandon cattle-keeping altogether. U de wahalla, ‘it is a great trouble’, was the explanation given to me by the Alkali of Bida when I expressed my astonishment that it was so difficult nowadays to find people ready to buy a herd of cattle. I have known two cattle-owners—the last in Nupe; one was an ex-slave who died without heir in 1936 and left a herd of thirty-two head of cattle which at the time was tended for him by Fulani herdsmen who had their camp near Bida. It was at his death that the question arose where to find a buyer for the herd. Th e other cattle-owner was the late Etsu of Agaie, a Fulani by birth, who owned a large herd of very beautiful cattle, of over 200 head, which was tended by Nupe shepherds, the sons of ex- slaves of the Etsu.
I may, perhaps, add here an historical note: Th e cattle of the Etsu of
Agaie were not, like the cattle to be seen at present in Nupe, of the
1 Laird and Oldfield refer to the huge herds of cattle which they saw in the capital of Nupe kingdom, Raba.
20 2 ECONOMICS: RESOURCES
northern, humped, breed. They were humpless and of small build, evi• dently related to the dwarf cattle which exist to-day in certain areas in eastern and south-eastern Nigeria. These dwarf cattle are said to have been very widespread in eastern and northern Nupe country before the Fu – lani conquest; the Gbari, for example, maintain that they had owned, originally, large herds of these cattle which, to-day, have almost disap• peared in this region. Th e traditions of Nupe speak of a magic connected with a bull or a cow with which the ancestor-king Tsoede endowed the first Beni chiefs. Th e ketsd ritual on the Niger river demanded the sacri• fice of a bull, to be presented annually by the chief of the Kyedye. Evi • dently, these rites and sacrifices could only have referred to these dwarf, pre-Fulani cattle. Anyhow, there is no evidence that cattle have ever played any important part in the economy of the peasant areas, and pos• sibly in the whole pre-Fulani kingdom.
The disappearance of cattle, however, from the Fulani-Nupe, and from the households of the rich, demands explanation. No doubt the
‘great trouble’ has played some part in this change. Wit h the disappear• ance of slavery the problem of finding suitable and sufficiently cheap labour presented an additional difficulty. But the decision to abandon cattle-keeping was also due to a decided change in the attitude of the people towards this commodity. Th e herds of cattle were, in a word, capital. Th e Nupe do not live on the cattle, basing their diet on milk and butter, as the Fulani and other cattle people do. Nor are the cattle in Nupe linked to a whole system of religious beliefs, taboos, and cere• monial usages. Th e Etsu of Agaie (on whose herd I could examine these questions in detail) occasionally sold one or two head of cattle when he was in need of money—this was the only concrete use to which he or any of the Nupe cattle-owners would put their cattle. I cannot judge how far status and prestige may have formerly entered into this assessment of the value of cattle. I gathered that it could not have been of much sig• nificance; the possession of horses counted for much more. But the existing incentives to possess cattle must have been further weakened, and even reversed, by two changes in the present-day economic situa• tion. First, by the enormous rise in the price of cattle, especially in the countries south of Nupe, which came with the increased (European and Europeanized) demand for meat and skins. Second, by the appearance of new means of economic, enterprise, and the prospects of more profitable, large-scale trade, which necessitated a more liquid form of capital. It is a fact that a large number of rich cattle-owners soon sold their herds, attracted by the high price, and also realizing most probably the greater advantage that, under modern conditions, lay in transforming their stock into ready money.
This tendency to keep the capital ‘liquid’ is shown also in the readi• ness of the Nupe to experiment with different kinds of live stock. Thus we find here and there—and the same thing is reported from Nupe a hundred years ago—ostriches, which the Nupe buy from traders from the north, feed and tend and finally sell again with a profit. An enter• prising chief had even taken to breeding horses; it was done on a very small scale, and was in Nupe an isolated case, but still characteristic of
ECONOMICS: RESOURCES 203
the general trend. The same trend is, finally, evident in the Nupe method
of keeping sheep and goats.
Sheep and Goats
Very rarely do the Nupe farmers (or town people for that matter) breed sheep and goats. It is, once more, the principle of ‘liquid ‘ re• sources that counts. A great many farmers buy a sheep or goat when it is young, or when sheep and goats are cheap, and sell it again later on wit h profit, when they are in need of money. Certain regular fluctuations in the price of goats form the basis for this calculation. During the season of the great Mohammedan feasts, when rams are killed in almost every house, prices go up, to fall again during the rainy season, before the harvest, when there is little money coming in and accordingly little demand for animals.1 This method of husbandry accounts for the very irregular distribution of sheep and goats at a given time, varying greatly wit h the individual need for money and the ability or inclination to in • vest money in animals. The general figures per capita of sheep and goats are surprisingly low (see the statistical table). The figure for Bida town is o-i2 and for the districts 07 1 per male adult. These figures reflect the very small money resources the Nupe have for engaging in this sort of seasonal trade. Throughout the year, and especially during the time of the Mohammedan festivals, a large number of sheep and goats have, in fact, to be imported from Hausa. But the figures also show that sheep- and goat-keeping is more frequent in the districts than in Bida. This bears out what I have said above about the influence of varied and expanding economic enterprise upon stock-keeping. Bida, though richer than the districts, is poorer in live stock; the centre of trade and of manifold economic interests keeps its capital more liquid than the country wit h its limited scope of economic prospects.
It is interesting to note in this connexion that in goat- or sheep-breed•
ing the young animals always belong to the owner of the mother-goat or
-sheep; the owner of the female animal may ask the owner of the male
animal to let him have the animal for breeding, but this would be done
without payment or return of any kind, merely as a personal favour.
Female animals are accordingly more expensive than male animals,
and the inducement to part wit h the mother-goat thus often overweighs
the interest in the breeding itself.
Other Domestic Animals
The Nupe keep no pigs and eat no pork—an avoidance which is not, as far as I see, due only to Mohammedan influence. The few pigs one sees in the country belong to Yoruba.
1 Prices for sheep and goats vary from 5s. to 15s., according to the season (see p. 316). The great seasonal changes are also shown in the figures for sheep and goats slaughtered in Bida in the different months of the year: January, 130; February, 150; March, 82; April, 151; May, 129; June, 68; July, 58; August,
114; September, 116; October, 119; November, 120; December, 122. The highest figures refer to January/February and April, the months of the Mohammedan festivals of Ramadan and Muharram; the lowest figures to June and July, in the middle of the rains, before the first harvest of the season.
204 ECONOMICS: RESOURCES
There is little to be said about fowls. In every house chickens are kept; men and women, grown-ups and children, have their own chicken. Often one sees a young boy going to the market to sell ‘his’ chicken, given him by father or mother. Th e fowls are bought and sold regularly in the food-markets of every larger village. Fowls represent the most important sacrificial food of pagan (and to some extent Mohammedan) Nupe; almost every religious ceremony, communal as well as house• hold-rites, involves the killing of a fowl. Chickens are, besides, the most desirable food the Nupe know. Eggs are not marketed and in certain parts of Nupe not eaten. Th e Beni do not eat eggs because ‘they make men sterile and women barren’. Th e Nupe also keep pigeons, for which they put up a bowl of water on a forked pole inside the house. Pigeons are not as a rule traded in the market, nor are they eaten regularly, but sometimes the eating of pigeons is ordered by diviners for ritual reasons.
Donkeys and Horses
About three-quarters of the donkeys which are found in Nupe belong to Bida town. They are transport animals owned mostly by traders, pack animals used in their caravans which carry palm-oil and salt from the river steamers up country, and bring natron down from Hausa and Bornu. In 1935 the only districts possessing no donkeys were those which have other means of transport: the riverain area, and the district of Mokwa through which the railway runs. Again, the largest number of horses is found in Bida, Horses, as I have said already, are a symbol of rank and status. Th e Etsu Nupe may, as a sign of his highest esteem, present a man with a horse. But every man who can afford it will buy one for himself or for his sons. A horse represents, if not the most profitable, the most noble expenditure or investment.
Yet, profitable in the narrow sense, or in the wider sense as implying enhancement of prestige—all these possibilities of investment of capital in live stock are determined by the amount of liquid capital available; in other words, by the volume of money in circulation. Th e economic stringency of the country which, starting from the decline in the market prices for native produce in 1934-5, as well as the subsequent recovery in 1936, is reflected clearly in the figures of the census of live stock for Bida Division for the last four years.
Sheep Goats Horses Donkeys
1933
1934
1935
1936
5,788
6,600
5,0101
5,702
28,856
29,445
25,876
34,75i
1,196
1,160
917
1,007
III
157
137
149
CHAPTE R XI I AGRICULTURE: PRODUCTION COUNTR Y AN D CROPS
Y far the largest proportion of the Nupe population work on the land. In the purely agricultural districts in Trans-Kaduna 9,000 out
of a male population of 11,000 are farmers. Even in Bida town, the centre of crafts, industry, and trade, more than a quarter of the popula• tion earn their livelihood on the land.1 This shows clearly the pre• dominant position of agriculture in the productive system of the country. We have already spoken of the rich and complex character of Nupe agri• culture, due both to the favourable configuration of the environment and to the advanced civilization of the people, who were always ready to adopt and introduce advantageous techniques of production from other groups. The wide scope and the complexity of Nupe agricul-‘ ture enables the people to rely in their nutritive arrangements on a large variety of crops which, as we shall see, is well distributed over the year.
But these facts must be correlated with the facts of local distribution and specialization of agricultural techniques. The most important local specialization refers to marsh-land farming. The marsh-land comprises only perhaps one-tenth or less of the total land under cultivation. Th e intensive cultivation, however, and the concentration on certain specially valuable crops (above all rice) make it a most important factor in the productive system of the country. Other forms of specialization which, on a smaller scale, influence the productive and nutritive balance of Nupe, are due to differences in the quality of the soil and, indirectly, to population factors and conditions of landownership. Thus cotton and ground-nuts thrive specially in Trans-Kaduna, the most fertile region of Nupe. The comparatively poor selection of crops round Bida is due both to the more exhausted soil and the comparative scarcity of land in this most densely populated area. Less yam, for instance, is grown round Bida than elsewhere, as yam demands a new plot every year (see p. 209). Nor is bulrush-millet grown near Bida, although in this case the reasons put forward by the people are different: it is no good, I was told, to try planting this early crop near a large town; the multitude of birds would devour the crop as soon as it appeared. The difference in the quality of the land in Beni country and in Trans-Kaduna is also reflected in the varying length of the period during which a plot can be kept under con• tinuous cultivation. In Mokwa a plot is generally worked for four to five years; round Kutigi for six to seven years; in Doko, in Beni country, only for three to four years. In this last district, with its heavy popula• tion and its restricted land-resources, a rudimentary system of animal manure, with horse- and goat-dung, is practised by the people. Asked to explain why it was that only this one section of Nupe was practising animal manuring, the people said: ‘Look for yourself; you will see that
1 Of a total of 7,714 Bida men 1,664 are farmers. Th e figures mentioned here are taken from a local census taken in Bida Division in 1930.
206 AGRICULTURE : PRODUCTION
we have little land, much less than people in other parts of the country. We even have to make our farms up on the hill-side since there is no vacant land left in the valley.’ This is true enough. When I visited Doko on my second expedition I saw that a few enterprising men had started farming on the steep stony hills at the back of the village—a thing nobody dreamed of doing when I stayed in Doko on my first expedition two years before.
Th e appreciation of manure has led in Beni country to an interesting co-operation between the population of a village, or a certain landlord, and the groups of nomadic Fulani herdsmen who come down regularly in the dry season to find pastures for their herds. Th e Nupe landowners invite the head of such a nomadic group, and sometimes induce him by presents of food or assistance in the building of the camp, to make his camp on their fallow land. This procedure is repeated every year with a new plot of fallow land. Th e year after the land has been thus manured it is taken under cultivation. I have myself seen the enormous differ• ence in the growth of the crops between a plot on which the Fulani had made their camp and other, ordinary farm-plots.1
Cultivation is also to some extent carried on inside the villages and even inside Bida town. In the villages, where large open spaces separate the different efu, the cultivation inside the village is fairly extensive, and one grows there the main crops, millet, guinea-corn, ground-nuts, &c. In densely populated villages where there is little open space left, and in Bida town, the people still grow a few things. Something will be planted, in fact, wherever there is room, all along the town walls, round the com• pound wall, on any open space in front of the house and even inside the house. Here are grown chiefly greens and vegetables, beans, sorrel, red pepper, also sweet potatoes, cassava or maize, and hemp. As a rule the small plots in or near the house are planted with crops which are easy to tend, and this work is left to the old men who can no longer walk long distances and do heavy work. Even inside Bida there exists a certain specialization; on the small stretch of marsh-land that runs through the town, and in the thick groves on the banks of the two streams, typical marsh-land crops are grown: sugar-cane, rice, and bananas.
In this large variety of crops cultivated in Nupe not all are, of course, of equal importance from a nutritive or general economic point of view. Again, it is impossible to give a simple formula. The staple crops, the crops on which the food supply of the people primarily depends, are millet, sorghum, yam, and, to a lesser extent, rice. From the point of view of trade the most important crops are rice, ground-nuts, cotton, but also sorghum and yam and a few lesser crops. Cassava, maize, and sweet potatoes (grown inland) are of secondary importance only. This is shown clearly in the attitude of the Nupe farmer towards these crops. Th e pride of the farmer is always his millet, sorghum, and yam farm. Cassava and maize, the Nupe would say, grow anywhere, and anybody, farmer or no farmer, can grow them. A man who would rely on this
1 Among the Bida landlords it is an accepted arrangement to place one’s fallow land at the disposal of the Boror6 previous to leasing it to a new tenant. Th e landlord can then obtain a much higher price for his land.
AGRICULTURE: PRODUCTION 20 7
type of cultivation only would lay himself open at once to the criticism of his fellow farmers. As regards cassava, it is known to have little nutritive value—’only yam or millet or sorghum make a man strong’. Both crops may, however, provide a useful stopgap for the time between the harvest of the main crops.
We have said earlier in this book that the complexity of the agricul• tural system of the Nupe is due partly to the close contact they have had with other cultures, in particular with neighbouring groups. We realize here that this was more than an occasional contact, and more also than some obscure process of diffusion. It was to a large extent part of that assimilation and cultural ‘learning’ that goes hand in hand with in • tensive political unification of long standing and all that it entails— regular intercourse, a wide net of transport and traffic, and an ever- extending economic co-operation, As the kingdom brought together peoples from the marsh-land and from the inland forests, farmers from the ‘yam countries’ and farmers from the ‘millet districts’, it made them learn from each other and at the same time utilize each other’s special resources. Assimilation and specialization are both features of an ex• panding economic area. We have, in fact, a threefold correlation; for these two features involve another, third, factor: a large-scale system of organized distribution, that is, trade and marketing.
FARMING TECHNIQUE
Like most African tribes the Nupe practise hoe-agriculture. They use two kinds of hoes, a large heavy type called zuku, and a small type called dugba. The zuku has a heavy handle, about 16 in. long, bent somewhat like a hook, into the angle 0/ which the broad shovel-like blade (8 in. long) is fitted. Th e dugba has a straight handle, 20 in. long, with a less broad, slightly bent blade inserted at the end. The heavy hoe is used for throwing up mounds and ridges, the lighter hoe mainly for weeding and other less heavy work. Th e Nupe also use a long bush- knife, gada (now mostly of European manufacture), for cutting grain stalks or digging out yam.
When using the heavy hoe the farmer stands across the ridge he is making, with legs wide apart, the body bent forward, and, gripping the handle of the hoe with both hands, he lifts it to the level of his head and shoulders and brings it down again between his legs, with one swing digging up the earth and throwing it behind him on to the mound or ridge he is preparing. When the mound or ridge has reached the proper height, he moves a few steps forward to repeat this action, until the ridge extends to the edge of the field. Having finished one ridge, the farmer turns round to throw up the next, parallel, ridge.
Nupe cultivation works both with large round mounds and low long ridges; the mounds are 2-3 ft. high and about 3 ft. in diameter, and are arranged in long rows; the mounds are spaced (measured from top to top) about 4-5 ft. in the same row and 6 ft. from row to row. The ridges are 1 ft. high and are drawn, more or less straight, right across the field in parallel lines about 3 ft. apart. The two types of cultivation are used
208 AGRICULTURE : PRODUCTION
for different kinds of crops. Ground-nuts, sweet potatoes, and certain vegetables are always planted in ridges, yam always in mounds. As regards the other crops, the rule is that in a first cultivation, that is, on a plot planted for the first time after having lain fallow, mounds are throw n up ; in the subsequent year the mounds are turned into ridges, as a result of the practice of ‘shifting’ the ridges and interplanting in the second half of the planting season. Th e throwing up of mounds is by far the heavier and longer work. ‘Lazy ‘ farmers—in the eyes of the people—thus sometimes prefer the easier jo b of making ridges even in a first year’s cultivation.1 Big trees are always left standing on the farms and cultivation is carried out round them. Th e Nupe farmers appreciate the shade of big trees during the heat of the day. Small trees on the farm-land whic h do not bear fruit are lopped or cut about 4 ft . from the ground. Thi s becomes important in the period of lyin g fallow; for the tree-stumps and undergrowth whic h the farmer leaves standing on his fields help to pu t the fallow land under a growth of ‘bush’ rather
than grass, thus assisting and speeding up the regeneration of the soil.2
Th e cultivation on the marsh-land farms is adapted to the conditions
of a natural annual irrigation by the flooded rivers and watercourses.
Th e irrigated area is divided into square plots, approximately 6 ft . by
9 ft. , each consisting of from four to six ridges running parallel and two
ridges at right angles on the upper and lower end. Th e ridges are flat
and narrow; they do not connect at the corners, but leave small channels
open; by means of these ridges the water from the flooded river is dis•
tributed evenly on the cultivated plots, and at the same time kept stand•
ing there.
In the lay-out of the Nupe farms the aesthetic aspect is strongly pro•
nounced. Th e farmers pay much attention to drawing their ridges nicely
parallel and as straight as possible, and to distributing their mounds
symmetrically over the field. Th e nice rounding-off of the border of a
field is specially appreciated. Frequently one finds ridges and cross-
ridges arranged in a symmetrical and even conspicuously ornamental
pattern. Similarly, the skill and keenness of the farmer is judged by the
clean look of his farm, wit h all weeds, leaves, and brushwood cleared
away. An overgrown, untidy looking farm invariably elicits the com•
ment: za-gd de kokari a, ‘that man makes no effort.
ROTATIO N OF CROPS
Th e Nupe system of agriculture is based on shifting cultivation com• bined wit h a well-worked-out rotation of crops. Plots are kept under cultivation for from four to seven years (see p . 205), i.e. til l the land
‘grows tired’ , and then abandoned temporarily. Th e individual farmer does not, however, always wait unti l failing crops have warned hi m of
1 Th e mounds represent the more thorough cultivation, utilizing a deeper and, I believe, also heavier layer of the subsoil. This is shown in the fact that yam, which demands a heavy and more humid soil, is always grown on mounds, whik ground-nuts, for which light sandy soil is best suitable, are invariably grown on the low ridges. (Cf. Faulkner and Mackie, op. cit., pp. 147-8.)
2 See op. cit., p. 44.
AGRICULTURE : PRODUCTION 209 the ‘tiredness’ of his farm-plot, bu t as a rule simply follows the general
routine obtaining in his area, varying the standard period of cultivation
occasionally by one year, in accordance wit h the special requirements of
his farm-land. For the abandoned field a new plot is taken under cultiva•
tion, while the former is left to lie fallow, unti l eventually it wil l again be
fit for cultivation. Th e period of lyin g fallow, dependent as it is on local
conditions of land-tenure and individua l land-holdings, varies muc h
more widely than the period of cultivation—between four and fifteen
years.
Durin g the period of cultivation a strict rotation of crops is observed. There exists a comparatively large number of different types of rotation, of whic h I shall mention only the most important. Th e rotation depends on the crop whic h is planted first on a plot taken under cultivation. Th e following chart illustrates the most common types of rotation :
Table of Four- Year Rotation
A B C D
I Ya m (Bulr.-mill. ) late mille t and sorg• hu m
Ground-nuts (and cas•
sava)
Cotto n
2 Maize and sorg• (i) bulr.-mill. , (ii ) maize, Bulr.-mill.,
hu m o r bulr. – late millet , late millet ,
mill. ; late mil • let and sorg• hu m
3 Bulr.-mill. , late mille t and sorg• hu m
and sorghum sorghum
,, >> >> >> >> ,, ,,
4 ,, >» ,, ,, Cassava, sweet potatoes, Cassava,
or fallow, or cassava, sweet potatoes, and
beans
and beans sweet po • tatoes, and beans
Th e main rules in this rotation are these: yam, cotton, and ground• nuts are ‘one-year crops’, i.e. they are invariably the first crops in the cycle of rotation and are not as a rule planted a second time on the same plot. On the contrary, it is well known that mille t and sorghum do particularly wel l planted as a second year’s crop after cotton, ground• nuts, or—i n a less common rotation—red pepper. Where the rotation starts wit h mille t and sorghum bulrush-millet is sometimes left out in the first year and added later, as it does not do very wel l as a first crop on a new field, especially if the rains happen to be late. Also it is sown mostly only where there are a number of neighbouring fields under cultivation ; on isolated fields the danger of birds picking the crop is too great. On the other hand, land on whic h the two mille t kinds and sorghum are planted in succession may ‘tire ‘ sooner than land planted wit h varying crops and thus have to be abandoned after only three years. Th e alternative choice, finally, after a first crop of ground-nuts (C (i) and (ii ) in jthe table) depends mainly on the energy and keenness of the farmer; the crops listed under (i) , i.e. the combination millet-sorghum,
p
21 0 AGRICULTURE : PRODUCTION
though more valuable, involve the heavier work, namely, a complete re- hoeing of the ridges.
Jebba Island, which is half marsh-land, is the only area where I found yam planted twice on the same plot. This type of cultivation necessitates a different system of cropping: between two yam crops sorghum alone is planted, without the additional two millet crops, which is said to exhaust the soil far less than does the mixed cropping. Even so, the field sup• ports only three successive crops, and is left fallow after the three years. Before the second yam crop, i.e. after two years’ cultivation, the opera• tion of ‘shifting the ridges’ takes place, which we shall describe later in detail, and which, in the normal rotation, has to be carried out every year. As an alternative the people of Jebba Island practise a five or six years’ rotation with four or five sorghum crops succeeding a first year’s yam crop, the ‘shifting of the ridges’ being again inserted after two or three successive crops.1
The table shows further that certain crops are planted together on the same plot. This combined cultivation is much more extensive and much more varied than is shown in the table. Th e practice of ‘interplanting’ certain crops between others is one of the most characteristic features of the intensive and economical farm technique of the Nupe. Th e people speak, in fact, in connexion with this interplanting oiewo banza, ‘useless ridge’, meaning here the furrow between ridges or mounds; it is the
‘ useless ridge’ that is utilized in the interplanting. This combined cultivation constitutes at the same time one of the most complicated aspects of Nupe farm technique because of the wide range of variations which it permits. These are the main combinations:
Simultaneous Interplanting. In the purely agricultural districts the two millet kinds and sorghum are planted together on the same mound, sorghum on the top and bulrush-millet and late millet on the flanks of the mound. The three crops are not planted quite simultaneously, bulrush- millet being sown a little before the other two crops (see chart, p. 213). The farmers near Bida practise various other combinations, e.g. late millet-sorghum, maize-sorghum, or late millet-sorghum with ground-nuts interplanted in the furrows.
Yam and okra are frequently planted together, yam on the top of the mound and okra at the foot.
Ground-nuts and cassava may also be planted together, ground-nuts on the top of the ridge and cassava in the furrow between the ridges. The same arrangement obtains with sweet potatoes and cassava.
Successive Interplanting. After the reaping of bulrush-millet or maize beans are sown in their place, i.e. on the same ridges or mounds, between the still growing late millet and sorghum plants.
Melons are frequently planted irregularly between various crops.
Cassava may be interplanted with any crop, e.g. yam or millet and sorghum, and at any time.
On the farm-plots close to the compounds, which are manured with the refuse from the house, no rotation is practised. Nor does such com• plicated rotation exist on the irrigated marsh-land farms. Here the same
1 In Yoruba country, where the soil is much richer, three or four successive yam crops are said to be the rule.
AGRICULTURE: PRODUCTION 21 1
plot can be cultivated for ten years or even longer. Th e cultivation follows every year the same pattern: two alternating root-crops, cassava and sweet potatoes, and, after they have been harvested, on the same plot, rice. On certain marsh-land farms in the Kaduna valley two succes• sive crops of maize are grown. Other marsh-land farms again (on the rivers Gbakoand Kaduna, and on a smaller scale near Kutigi) are devoted to the cultivation of onions and partly sugar-cane. Farmers in inland areas who possess a field close to a watercourse plant rice in addition to the usual inland crops.
To return to inland cultivation. Of the four types of rotation listed in our table (A B C D ) the farmer selects a certain combination. Most farmers, as we shall see presently, cultivate two or three crops at the same time, the most frequent combination being ABC , ABD , and AA C (D) , i.e. a rotation which includes the maximum proportion of staple food crops, millet, sorghum, and yam, together with one or two com• mercial crops. By a judicious overlapping arrangement, starting the different series of crops in successive years, the farmer secures an even distribution of these crops. This is an example from Mokwa, recorded in 1936.
1935
Plot 1 Plot 2
Yam
Plot 3
Sorghum, millet
Plot 4 Plot 5 Plot 6
1936 Yam, ground- Maize, sorg•
beans hum , beans
Sorghum, millet
Ground-nuts
1937 Maize, sorg•
hu m
Sorghum,
millet
Sorghum, millet
Yam Cassava Ground-nuts
1938
Sorghum,
Sorghum,
(?) Maize, Cassava Cassava
1939
millet
Sorghum,
millet sorghum
Sorghum, ? Cassava
1940
millet millet Sorghum, millet
Cassava
Th e main problem in determining a suitable system of rotation are the
‘one-year crops’. It can be seen that, allowing for a minimum period of
four years for lying fallow, a farmer must have at least four different
plots under cultivation in order to obtain one of the one-year crops in
every year of the rotation, or eight plots, in order to secure a continuous,
uninterrupted, cultivation of these crops. Where such an extensive sub•
division proves impossible, the number of one-year crops will vary from
year to year, in accordance with the progressive stages of the rotation. In
1936 I found that of thirty-seven farms on which one or more of the
three ‘one-year crops’ were grown only two farms combined all three,
six combined the cultivation of yam and ground-nuts, seven of yam and
cotton, and of the remaining twenty-two the cultivation was restricted
to one of these three crops.
Th e selection of crops on which the farmer eventually decides and the cycle of rotation which he will accordingly adopt will thus depend primarily (ignoring here the question of the quality of the land) on the size of his farm-land and the possibility, or profit, of subdividing it into plots for different cultivation. Evidently, land cannot be subdivided indefinitely to make room for a large variety of crops. I have found that
ai2 AGRICULTURE : PRODUCTION
the farms under cultivation in Nupe varied in size from 2 to 10 acres, according to the number of men who are workin g together; the average size of farm corresponding to one able-bodied male adult is from 2 to
2 ½ acres. A farm of this size (i.e. a one-man farm) can stil l be, subdivided into two or three plots, for the cultivation of different crops; larger farms can be subdivided into three or four different plots for cultivation. In addition to the field cultivated at a given time the farmer as a rule possesses from one to three fallow fields of the same size to whic h he turns when it becomes necessary to take a new field, or new plots on this field, into cultivation.1 Th e following examples, whic h were collected in Lemu , a prosperous village nort h of Bida, may illustrate the relation between size of farm and selection of crops. These examples represent, of course, only stages in the four-year rotation, whic h explains the reason for the absence of certain one-year crops in the selection of crops whic h were grown that year.
- Five men working together; total land under cultivation approx. 9½ acres. Grain (millet and sorghum), 6 acres (half of this plot was under yam cultivation the previous year); red pepper (formerly ground-nuts),
1½ acres; cotton, 2½ acres. - Three men working together; total land under cultivation 6 acres. Grain, 5½ acres; rice, ½ acre.
- Tw o men working together; total land under cultivation over 4 acres. Grain, 2 acres; yam, 2 acres; ground-nuts, ¼ acre.
- One man working ; total land under cultivation under 2 acres. Grain,
1½ acres; rice’, ( acre; red pepper, ¼ acre (½ acre under yam cultivation the previous year). - One man working ; total land under cultivation over 2 acres. Grain,
1 acre; yam, 1 acre; rice, 200 sq. yds. - One man working (a young man, recently married); total land under cultivation 1 acre. Grain, J acre; red pepper, ¼ acre.
These examples brin g us to a further condition determining the selec• tio n of crops and type of rotation, namely, the size of the workin g group at the disposal of the farmer, and the efficiency of its organization; on its co-operation depends the carrying out of the complicated and at times overlapping schedule of planting, weeding, &c , withou t any of the crops being neglected. Finally, the selection of crops is also determined by the prospect of return, i.e. by the satisfaction of specific wants whic h the farmer may expect from the different forms of cultivation. We wil l for the present ignore this last point and concentrate on the second condi• tion . In order to understand full y we must study more closely the agri• cultural calendar of the country. In the attached chart I have tabulated the schedule o f Nupe agricultural activities. I t wil l be best i f I supple• ment it wit h a description of the more important agricultural activities as I have observed them in actual cases. First, however, a few comments are necessary.
1 It can be calculated that for the continuous cultivation of two one-year crops, e.g. yam on 1 acre and ground-nuts on ¼ acre, in addition to an ordinary crop, e.g. grain on 1 acre, a total land possession, i.e. worked as well as fallow land, of 9-10 acres is necessary. Th e figures quoted above show that actual practice falls considerably short of this standard.
AGRICULTURE : PRODUCTIO N 21 3
The Agricultural Calendar of Nupe
As mentione d elsewhere (see Appendix , p. 411 ) the Nup e peasant calendar counts 12 luna r months starting fro m the mont h whic h sees the beginnin g of the rain y season. It wil l be realized that the various dates given i n this synoptic chart cannot b e fixe d exactly. T o simplify the chart it is assumed tha t the rains start punctuall y on the 1st of April , and tha t th e plantin g cycle thus begins wit h the first mont h in native counting. Needless to say tha t in practice the whol e time-tabl e wil l frequentl y be foun d to have move d forwar d or backwar d by as muc h as a fortnigh t or three weeks as the rains set in earlier or later tha n ou r key date. Besides, there occur certain other variations in the tim e sequence whic h are due to local conditions or th e varyin g efforts of individua l farmers ; these varia • tions have onl y partl y been include d in the chart.
Calendar
Planting operations I
Eur op.
Apri l
Ma y
June
July
August
September
Native
1st month
2nd month
3rd month
4t h month
5th month
6th month
7th month
Inland farms (incl. irrigated plots)
Belated preparation of yam plots and planting of yams by farmers who have left i t til l the new season. Sowing of bul • rush-millet and maize. Sowing of late millet and sorghum. Sowing of ground-nuts. Interplant- ing of okra, melon, and cassava.
First weeding (takes 2-3 weeks).
Short period of rest (5-7 days).
Yam is sprouting leaves, bulrush-millet and maize
2 ft. high. Second weeding.
Sowing of cotton. Plant• in g of sweet potatoes, rice, and red pepper. Also of cassava grown on separate plots. Latest date for planting ground-nuts. Harvesting of bulrush- millet and maize. Inter- planting of ground-beans and late beans.
First yam harvest. Picking of okra.
Thir d weeding, and ‘shift• ing the ridges’. (This operation may be post• poned by 2 months.) Beginning of first ground• nu t harvest.
Fourth weeding and trans• planting of late millet and sorghum.
Sweet potatoes and melons harvested.
Some farmers carry out the
‘shifting of ridges’ now instead of two months earlier.
Harvest of early maturing
rice.
Marsh-land farms
Planting of cassava or sweet pota• toes.
Harvesting
of cassava and sweet potatoes; sowing of rice. Sow• ing of first maize crop.
Transplant• ing of rice. Harvest of first maize crop.
Trees
Frui t of lo • cust-bean tree picked, also mango.
Locust-bean harvest finished.
Shea-nuts picked and dried.
Shea-nut harvest finished.
Preparation of shea- butter.
Second man•
go crop.
21 4 AGRICULTURE : PRODUCTION
Calendar Planting operations
State of
Inland farms (inch
Eur op. Native
Marsh-land
Trees
October
November
December
January
February
Marc h
7t h month
8t h month
9th month
10th mont h
11th month
12th month
irrigated plots)
Second yam harvest. Har• vest of ground-beans, sweet potatoes, and cas• sava grown on separate plots.
Preparation of new yam- plots and planting of yam (may be postponed for 2 or 3 months).
Second ground-nut harvest begins.
Harvest of late millet . Harvest of sorghum. Picking of cotton and red pepper.
Harvest of late beans.
En d of ground-nut harvest. Burning of farms.
Some farmers prepare new yam plots now instead of
3 months earlier. Burning of farms goes on.
farms Trees
Sowing of second maize crop.
Harvest of rice (late maturing crop). Harvest of second
maize crop.
First mango crop be• gins, also harvest of locust-bean.
food stores
New yam, from second harvest, for storage.
New sweet
potatoes
and cassava. Yam from first harvest finished.
Sorghum giving out.
Sorghum and late millet for storage.
Selection of Species and Varieties
Most of the crops enumerated in the chart occur in more than one variety. Ya m is the only exception; although grown in a large number of different varieties in Southern Nigeria, it is known only in one variety among the Nupe, which is incidentally another proof that Nupe is not a
‘real yam country’. Sorghum (eyi) was originally grown in Nupe in a number of varieties. I could collect names for three main sorghum varieties: ktiyi, dindorogi, and ekpa. Of these three varieties kuyi has disappeared almost completely, and dindorogi is fast disappearing. Th e latter variety was formerly specially valued for the black and red dye that was obtained from the stalk. Th e importation of European dyes has reduced considerably its importance, so that to-day ekpa is grown almost exclusively. Ekpa is also agreed to be the finest and best tasting of the three varieties; its survival thus appears to reflect a certain grow• ing fastidiousness with regard to food, which may well have influenced market prices and demands, and thus indirectly the policy of the sorghum growers.
Tw o varieties of rice are grown: the ‘sweet’ rice, a six-months’ crop, in the river marshes, and a variety which matures in ninety days, on the less well irrigated inland farms. Til l six years ago only the native variety of maize was cultivated. To-day the American variety (brought from Europe), which is sweeter and has a softer husk, is more and more widely adopted by the natives. This new variety is easily recognized since it is grown as a separate crop, in separate ridges, while the indigenous variety is interplanted through millet and sorghum or yam. Th e two millet varieties, finally, bulrush and late millet, have been mentioned already.
AGRICULTURE : PRODUCTION 215
Th e proportion of maize and the various other mille t and sorghum varieties in the productive scheme of one household is illustrated in the following typical example: on a farm whic h produced a total mille t and sorghum crop of twenty bundles the different varieties were distributed as follows:
Bulrush-millet . . 2 bundles
Maize . . . . 2-3 „
Sorghum and late millet are clearly shown to be the main staple crops as compared wit h bulrush-millet and maize, the importance of which , as we shall see later, lies in their subsidiary, ‘stopgap’, character.
Some enterprising farmers, and farmers in certain districts of Nupe, have added to this table of crops certain special, commercially valuable, plants or have given over a large portion of their farm-land to the cultiva• tio n of plants whic h normally woul d play only a secondary, unimpor • tant, part. Thu s we find henna (lali) cultivated in many parts; indigo (eci) is collected and prepared especially in the districts of Sakpe and Kutigi . Recently farmers in Lemu , nort h of Bida, have started the cultivation of red pepper (yakd) as a main crop, in the place of ground• nuts. Th e first two kinds of produce have a considerable marketing value in the country, the last kin d is meant for European export.
Clearing
Th e dry season is not completely a season of leisure. Man y Nupe farmers wil l visit their fields more or less regularly even durin g that time, perhaps once a week, to ‘look over the farm’ as they say, to see that animals have not done damage to ridges and mounds, or to collect brushwood for fuel, or leaves for whic h there is always use in the Nupe household.
In the dr y season comes also the first agricultural activity of the year, the nwa tsaka, the clearing and burning of the farms. On farms (fallow or virgi n land) whic h are to be taken under cultivation for the first time, grass and shrubs are cut and afterwards the farms are burned. Ol d farms also are cleared summarily by burning the dead stalks, leaves, and undergrowth that have been left there from last year. Later the first rains wil l wash the ashes into the soil, thus completing the process of
‘green manure’, whic h the Nupe farmers consciously utilize (see p. 221). Th e burning of the farms takes place at any time between January and May ; new plots, the clearing of whic h involves the heavier work, being as a rule burned earlier in the season, cultivated plots (on whic h the stalks and leaves are given ample time to dr y and rot) towards the beginning of the rains. Th e heaviest and lengthiest work is the clearing o f virgi n land. I f carried out early i t wil l curtail seriously the short leisure season o f the Nupe farmer, and i f left til l late wil l add to the work of an already very busy time of the year. Clearing virgi n land is, in fact, considered so heavy and inconvenient work that the farmers, especially those who cannot count on the assistance of a large working group, make all efforts to avoid taking virgi n land under cultivation.
2X6 AGRICULTURE : PRODUCTION
Thi s reluctance must be regarded as one of the main factors responsible for the growt h of the institution of ‘borrowing land’, whic h we have described in the preceding chapter.
Sowing and Planting
Th e Nupe, possessing an elaborate calendar system whic h works wit h six seasons and twelve lunar months, have no difficulty in determining the coming of the rainy season. Besides, they regard the appearance (i n April ) of a small red insect, whic h is significantly called gbama, ‘early rains’, as a sign of the approaching rains. Th e first light showers, whic h as a rule are accompanied by thunderstorms, are ignored. After the first heavy rain, whic h may last for several hours or a whole night, has thoroughly softened the soil, village and tow n spring into activity. In the mornin g after the rain you wil l see the people wandering out to their fields. The y walk in groups, neighbours or families together, fathers wit h their boys. Dogs trot behind their masters, and often a man or boy wil l lead a goat which wil l then feed on leaves and shrubs on the field. Th e men wear working dress: a loin-cloth and, if it is cold, a heavier cloth slung round the shoulders, and a straw hat as protection fro m the sun. The y carry their hoe over the shoulder, and gourds wit h water, food and seeds, over the arm. In many parts farmers carry their bows and arrows wit h them when they go out to their fields as ‘protection fro m wil d beasts’—a habit whic h is dying out but slowly, although there is hardly any need for it to-day. I have never seen these weapons used.
Th e nature of the first farming activity depends on whether or not the farmer has done what ‘good’ farmers are expected to do, namely, planted his yams already in the rains, immediately after the harvest. If he has not done so he wil l now at once start preparing his yam field; if he has, he is free to tur n his attention to the planting of bulrush-millet and maize. Of five farmers whose daily activities I was recording in Mokwa , three had planted yam in the rainy season and two (who both had to wor k single-handed) waited unti l the beginning of the new farming season. Wit h the ‘good’ farmer, we wil l start wit h the planting o f millet.
Again there are certain variations. Where mille t and sorghum are planted as successive crops no hoeing or re-hoeing as a rule precedes the planting ; as we shall see later the plots are normally prepared durin g the previous season. Where mille t and sorghum are planted after other first-year crops or on a new plot, ridging, occupying the first few days of the new season, precedes the sowing. Th e technique of planting, moreover, varies slightly according to whether one plants on mounds or ridges, but is the same for both millet varieties and for sorghum. I wil l now describe the scene one wil l meet on the Nupe fields when planting is in progress. Th e farmer has divested himself of his superfluous clothing, his food and water gourds, and has placed his things out of the way, under a shady tree. Perhaps he has his small boys wit h him , who m he lets play by themselves in a corner of the field whil e he goes on wit h his work . Tie d to his left arm the planter carries the seeds in a small
AGRICULTURE : PRODUCTION 217 gourd or his straw hat, keeping a ready supply in his left hand. In his
right he carries the light hoe used for planting. Wit h it he scratches the
top or flank of the mound, while with his left, almost simultaneously, he
throws the seeds into the newly made hole. Th e hole is then flattened
out with the back of the hoe. Thus the farmer proceeds from mound to
mound and row to row, working with amazing speed and ease: one man
is able to sow about one acre in from five to six hours, with hardly any
rest. This is the method used for sowing on mounds; sowing on ridges
is even easier and quicker. No hoe is used, and the farmer can make use
of both hands for the sowing, the seeds being pressed into the soft soil
with the heel. The distance kept between two plants is a little over 2 feet.
The sowing of one field with millet or sorghum is invariably done in
a day or two, and is nearly always carried out single-handed. Where
normally a larger group would be working together—brothers or a
father and sons—the rest of the team would be occupied on another
field, carrying out ridging, i.e. work for which co-operation is more
urgently required. But when for one reason or another a farmer falls
behindhand with his planting, ridging and sowing may have to be done
at the same time. Once I saw a man and his two sons of 10 or 11 years
carrying out this combined and hurried work: the boys were hoeing,
preparing the ridges; the father followed in their tracks, sowing. After
every ridge they had completed the boys waited for the man to catch up
with the sowing. The ridges were crooked, irregular.
In these first days of the planting season the Nupe keep short days :
they start comparatively late, about 8 a.m., and rarely return later than
2 or 3 o’clock. As the season wears on their working day extends. They
start at daybreak and often do not return till 5 or 6 in the afternoon.
They have their midday meal out on the farm, some bringing their own
food, others having it brought out to them by their wives or daughters.
At that time of the year, when fresh leaves and herbs are plentiful on the
farms, the men rarely return home without a fair supply of various
kinds: broad leaves which the Nupe women like for wrapping food in,
or a special kind of aromatic leaf which the women use for their bath, or
occasionally medicinal herbs used for various purposes. Th e men tie the
leaves round their straw hats, and thus pleasantly and ornately attired
wander home. A bath, a change of clothes, and afterwards the evening
meal concludes their day.
To return to sowing. A day or two after bulrush-millet has been sown, maize is planted in very much the same fashion, again mostly in one day. Then one has to wait for another heavy rainfall before the next crops on the list, late millet and sorghum, can be sown. As a rule all four crops are sown within a week or ten days. These voluntary or involun• tary waiting periods have their use: for often it becomes necessary to get one’s tools repaired, perhaps a new blade fixed to the hoe, or to find new tools for the young boys who will, this year, join their fathers for the first time in the farm-work. By the time sorghum is planted the rains fall fairly regularly, and there is no need for further waiting.
Th e farmer turns directly to the planting of ground-nuts, which is reckoned to take from five to ten days. Ground-nuts must be planted on
218 AGRICULTURE: PRODUCTION
a new plot, and new ridges must therefore be prepared. Th e planting itself is also a slow process, the seeds are placed into the soil with the fingers and the earth then stamped down with the heel.
Th e planting of yams demands no less skill and patience. Yam is like•
wise planted on a fresh plot and involves the preparation of new mounds
—the work of several days. The planting itself is also spread over a
number of days, one man needing two or three days to plant one acre.
Hal f a tuber is used as seed (called ebu). Th e planter carries the seed
yams in a cloth tied round his waist; he inserts the tuber about i foot
deep into the loose earth of the mound, and then pats the top of the
mound back into shape either with his hands or with the light planting
hoe. Having covered a section of the field, he goes over it again and
places a grass cap over the top of each mound, weighting it down with
earth; the grass cap protects the young plant from the sun and at the
same time provides a convenient sign by which the planter can dis•
tinguish planted from unplanted mounds. About two months after
planting stakes 4 feet high are put up on the side of the mounds, which
will serve as support for the vine; sometimes a conveniently placed tree
may be used for the same purpose.
Now, assuming that planting has started early in April, our time-table
has taken us to the end of the month or the beginning of May. It is by
then high time for the first weeding, which is to clear the ridges for the
growing plant and to remove roots, leaves, and undergrowth which the
burning of the farms has but imperfectly destroyed. Not until after this
weeding, which takes the best part of three weeks, will the farmer be
able to enjoy a short period of rest, between five days and a week. It
reaches as a rule into June, and into the short break in the rains which is
known as the ‘ little dry season’. Th e period of rest inaugurates a more
leisurely period of work. In June many farmers work on their large
farms only two days in succession, and rest on the third day, or devote it
to the lighter work on the garden-plots round their houses. Th e ‘second
weeding’, which takes from five to ten days, and is to clear the young
millet plants of the high grass, is also a task allowing of comparative
leisure, except when it coincides with the planting of cotton, a delicate
and lengthy operation, which falls in the same time. Farmers who
possess marsh-land plots are sowing rice in June; on inland farms
cassava is planted—but cassava can be planted at any time, and needs
little skill or attention: cuttings of the stem are simply put into the
ground; one person can plant a half-acre field easily in a day. Th e
beginning of July heralds a period of increased activity. It starts with
the reaping of bulrush-millet, to be followed immediately by the sowing
of beans and another, third, weeding. Let me illustrate the change from
comparative leisure to renewed pressure of work by quoting a few entries
from my diary for 1936.
Doko. 15 May.—Bulrush-millet 2-3 feet high. Yam vines fairly long. Yam stakes are being put up. Cassava and sweet potatoes are being planted. Doko, 1 June.—The people have begun to sow cotton, and some farmers also rice. Grass on millet-sorghum farms very high. Weeding (‘second
weeding’) in progress in many places.
AGRICULTURE : PRODUCTION 21 9
Kutigi, 5 June.—Cotton is planted, weeding of millet-sorghum farms going on at the same time.
Kutigi, 15 June.—Weeding still in progress.
Kutigi, 5 July.—Harvesting of bulrush^millet and maize begins.
Kutigi, 12 July.—Bulrush-millet harvest finished; crop all dried and stored. ‘Third weeding’ in progress. Beans are sown.
Kutigi, 18 July.—First yams dug out.
Weeding and Transplanting
Let us for a moment interrupt our description and examine in more detail the technique of weeding and certain related farming operations. Th e Nupe practise four main weedings or hoeings, each of which is known by a special name. Th e first weeding, called gongi (‘small grass’ ?), is carried out three or four weeks after the first planting and serves, as mentioned already, to clear the ridges from roots and undergrowth left on the farms from last year. At the same time the ridges which have not been newly made this year are re-done and put in proper shape. Th e second weeding, nuvu, follows about three weeks later, ‘when the bul• rush-millet is arm-high5, and clears the millet plant of the newly grown grass. The third weeding or hoeing is called sa wo gi, ‘making small ridges’; it combines a thorough cultivation with an operation best termed the ‘shifting of the ridges’. Carried out shortly after the harvest of bulrush-millet, it consists in hoeing away the original mounds or ridges and throwing them up again on the side, in what used to be the furrows. Late millet and sorghum, which after the bulrush-millet harvest had been left growing from the flanks of mounds or ridges, are thus transferred to the new position, in the furrows between the new ridges. Th e sa wo gi is a delicate operation which may easily damage the growing plant. When it is carried out the original mounds are already much reduced in size by the rains, and when they have been shifted they have the appearance much more of long ridges than of separate mounds. Th e shifting of the ridges besides weeding the maturing plant provides fresh surface soil for the new ridges, which are then left, all prepared, for the next planting season.
Th e fourth weeding, called muya (‘pulling out’), consists mainly in a thinning out and at the same time transplanting of late millet and sorghum. It is carried out four or five months after sowing, when sorghum is about 4 feet high. From the eight to ten shoots which grow from the seed about half are pulled out, their leaves are cut and the shoots transplanted either to a separate plot, if sufficient farm space is available, or to additional ridges thrown up along the edges of the original plot. Another method is to interplant the shoots through the original ridges or mounds, which in this case have been spaced specially wide apart. The transplanted shoots soon surpass the mother-plant in growth. Th e rice planters in the river marshes, I may add, who cultivate the late maturing variety, also practise thinning out and transplanting, three months after sowing.
An operation somewhat akin to the’shifting of theridges’, which,infact, is often referred to by the people as a ‘fourth hoeing’, is the sa ci wot the
22 0 AGRICULTURE : PRODUCTION
‘making yam mounds’. Man y farmers wil l tr y to prepare their new yam plots soon after the yam harvest, or at any rate before the beginning of the dr y season.1 The y clear new plots, thro w up the mounds, and plant the yams at once, thus having everything prepared for next year. In Nupe opinion one obtains bigger yams if one follows this practice, plants yam early, and leaves the seed yam in the ground through the whole dry season.2
Th e early preparation of ridges or mounds offers, on the other hand, tw o very concrete advantages: first, it means comparatively easy work, carried out as it is in the soft soil ; second, it enables the farmer to start his planting without further delay as soon as the rains break. In the case of yam, particularly, this early planting removes one difficult crop from the already crowded time-table of sowing and planting.
Reaping and Harvesting
Our description of the agricultural activities of the year has taken us to the second phase in the farming cycle: harvesting. A fortnight after the harvest of bulrush-millet, following closely upon the thir d weeding, the first yam crop is dug up, and shortly afterwards the first ground-nut crop. There are two yam harvests. In July the farmers di g out yam for the first time, cutting one tuber fro m the plant and leaving the rest in the ground to sprout farther. Thi s young yam is considered a delicacy; unlike the late yam it is eaten ‘mashed’. Onl y a littl e of it is sold ; most of it is consumed by the household. Th e fresh yam, however, lasts only a short time ; the second yam harvest, in October-November, is the main harvest, at whic h the remaining tubers (three or four in each mound) are dug out.3 Ground-nuts, too, are harvested twice. Th e first ground-nut crop is dug out between mid-July and the end of August, and used for food only ; the second ground-nut crop, dug out between the end of December and the end of January—a much drier crop than the first—is used exclusively for sale to the European trading companies.
There is hardly any pause between the end of the first yam harvest and the fourt h and last weeding. After this, however—it is by then the end of August or beginning of September—there is a certain lul l in the farm • in g activities, at least i f the time-table has been kept fairly strictly. Til l the next bi g item on the list, the second yam harvest, the farmer can enjoy a short respite of three or four weeks, durin g whic h he can take his wor k more leisurely. Even so, he remains fully occupied: he is stil l going regularly to the farm to clear away leaves and brushwood; those who have planted rice or sweet potatoes as additional crops harvest the m now ; the women pick the melons whic h have been interplanted wit h yam or corn. Certain farmers who were pressed for time earlier in the year have postponed the important thir d weeding unti l now. Th e lul l continues to some extent after the second yam harvest. No w is the time
1 In Bida this operation is often postponed till January or February.
2 I cannot say how far this opinion is substantiated by fact. Agricultural
officers whom I consulted on this question were unable either to confirm or to refute this native theory.
3 In Leaba, in the extreme north of Nupe, i.e. close to the line which marks
the northern boundary of yam cultivation, yam is harvested only once.
AGRICULTURE : PRODUCTION 22 1
when ‘keen’ farmers prepare the yam plots for the next year, and re-hoe the yam mounds into ridges to be ready for next year’s planting.
Th e harvest of late millet, whic h falls in November, is the signal that the last ‘rush-period’ of the year has begun. Le t me first describe the harvesting of millet—the method is the same for both mille t varieties and for sorghum. Th e millet and sorghum varieties cultivated in Nupe grow to a height of 6-8 feet. To reap the corn, the stalks are broken and cut close to the ground, and then laid in bundles in the furrows, where they are left for fro m five to seven days to dry in the sun (an exception is bulrush-millet, whic h may have to be dried artificially). Afterwards the heads of the plants are cut off wit h the bush-knife and carried home in large bundles, 50-60 lb . in weight, while the stalks are left on the fields. These bundles, called epay are the Nupe unit of reckoning for calculating the yield, and also the standard ‘load’ (kara) used in the marketing of unthreshed grain.
Th e harvesting of late millet almost coincides wit h the beginning of the second ground-nut harvest. A fortnight or so later sorghum is brought in . The n follows in rapid succession, frequently overlapping, the picking of red pepper, cotton, and beans, while the ground-nut harvest is stil l going on. Lastly, grass on fallow farms is cut ; when the last batches of ground-nuts are dug out, at the end of January, the agri• cultural cycle is completed.
Le t us return for a moment to the harvesting of millet and sorghum. I have mentioned above that the dead stalks of millet and sorghum are left on the farms. Some of the stalks are retrieved after a fortnight, and used for mat-making (see p. 239). Th e rest, however, is left lyin g in the fields, til l they are destroyed later i n the ‘burnin g o f the farms’. Th e same applies to the stalks of maize, the cut-off vines of yam, and the leaves or stems of other plants. Thus the dead plants are allowed to rot on the fields, and fertilizing substance is washed into the soil by the rains whic h may fall in the weeks or months between the harvesting and the following dry season. Thi s is the native system of ‘green manuring’ to whic h I have referred before. Th e Nupe are well aware of this effect, and say that these stalks and leaves ze taki, ‘become dung’.
Th e time-factor in reaping and harvesting is less sharply defined than in sowing and planting. Harvesting activities are spread more irregularly over a longer space of time. Certain crops, e.g. yam, cotton, sweet pota• toes, and rice, are harvested in stages as the individual plants or batches of plants mature. Durin g the (first) yam harvest, for example, I saw the farmer open two mounds only to close them again, having discovered that the frui t inside was not yet ripe. Again, the picking of certain crops, e.g. ground-nuts, cotton, red pepper, and beans, is a very long-drawn-out operation. But it may be speeded up—and a quick harvest of the’ cash crops’ (cotton and ground-nuts) may be an urgent necessity in times of financial stringency—if outside help can be enlisted. Th e picking of beans, whic h involves no such motives, is often left til l the very last, and carried out leisurely by the family after all the other crops have been brought in .
The’harvesting that falls in the middle of the agricultural season, on the other hand, must be carefully fitted into this schedule of farm-work.
22 2 AGRICULTURE : PRODUCTION
For example, cassava and sweet potatoes grown on the marsh-land farms must be dug out quickly as rice is planted on the same plots almost immediately. Th e harvesting of bulrush-millet and the first yam harvest, if delayed, may interfere with the planting of cotton, and the third weeding. A quick harvest, for a different reason, is imperative also for late millet and sorghum; here it is the danger of monkeys and birds damaging the standing grain which forbids delay. Th e Nupe have various ways and means to combat this danger: they put up scarecrows and erect high platforms on the fields from the top of which boys try to scare off monkeys and birds, by shouting, throwing stones, brandishing long sticks, or swinging long ropes tied between the platform and a near-by tree. In districts where the danger is very great the farmers may have to spend the night on the farm shortly before the harvest to guard their fields from these unwelcome guests.
It is interesting to supplement this large-scale survey of the agricul• tural cycle with a view from closer range which could do justice to the varying individual arrangements. To illustrate this aspect of Nupe farming I have compiled the farming diaries of three Kutigi farmers for two typical weeks—one between the end of Ma y and the beginning of June, when the work is temporarily easing off, and the second towards the end of June, when work is at its highest. These farming diaries show the scope of the individual variations of the general time-table. We shall see that they are fairly considerable—in details; but in general lines the activities of the three farmers conform closely to a common pattern. During the more leisurely season all farmers are ready to interrupt their work, and rest for two or three days in succession; but during the busy season only a very weighty reason will make them stay at home and leave their fields even for a day. Indeed at that time of the year you will find the Nupe villages deserted from early morning till about 3 or 4 in the afternoon, save for women, children, and old men. Another inter• esting fact emerging from the diaries is the disregard, under pressure of work, for the traditional day of rest, Friday. To work six days and rest on the seventh is a common practice in Nupe (in fact a practice said to have been common already in the times before the spreading of Moham• medanism). But a farmer who knows himself behindhand with his work will disregard the rule. Finally, we see that in labour groups consisting of several people the family head may sometimes enjoy a short rest while the young men carry on the work; or that if one of the workers is called away, for example, by some social obligation, the work of the group goes on unimpaired. An arrangement of this kind is evidently impossible in the case of the farmer K., who works on his own.
N, M. K.
Labour group: father and Labour group: father and Labour group: one man three sons one son
315193 6
Goes out to farm about Goes out to farm to- Is away from Kutigi ;
8 a.m. Eats midday gether with N., whose went to a neighbour- meal on the farm, and farm is close to his own. ing village for the turns about 4 p.m. Also returns with him. funeral of his sister.
AGRICULTURE : PRODUCTIO N 223
N. M. K.
Labour group: father and three sons
1.6.1936
Labour group:father and Labour group: one man
As previous day.
2.6.1936
Stays at home, together with sons.
3.6.1936
Stays at home. Sons go out to farm in the morn• ing, but are back at 2 p.m. in time to attend the market, which is held to-day.
4.6.1936
Goes out to farm at 8 a.m., returns for mid- , day meal; sons stay longer.
5.6.1936 (Friday)
Stays at home, together with sons. Attends mosque in the after• noon.
6.6.1936
Stays at home, says he is tired. Works on gar• den-plots, but will go to farm to-morrow. Sons go out to farm.
22.6.1936
Goes out to farm at 8 a.m., returns at 4 p.m. One son goes to Sakpe to attend wedding of a friend.
23.6.1936
Goes out to farm at 8 a.m., returns for mid • day meal. He hurt his hand and cannot work properly. His sons stay on; one son still in Sakpe.
24.6.1936
Goes out to farm at 8 a.m., but returns early as his hand hurts. Son who went to Sakpe re• turns in the evening.
As previous day.
Stays at home, and visits
N. in the evening.
Stays at home, works on garden-plots round the house. Son goes out to farm.
Goes out to farm at 8 a.m., returns after mid• day meal.
Stays at home. Visits N.
in the evening.
Goes out to farm at 8 a.m., returns for mid • day meal.
Goes out to farm at 7 a.m., returns after 4 p.m.
Goes out to farm at 8 a.m., returns for mid • day meal.
Goes out to farm at 7 a.m., returns after 4 p.m.
Returns in the evening.
Goes out to farm about
10 a.m., and returns at
4 p.m.
Stays at home; says he is tired from the jour• ney.
Goes out to farm at 8 a.m., returns after midday meal.
Stays at home. His brother, who lives in a hamlet, comes on a visit.
Goes out to farm at 8 a.m., returns for mid• day meal.
Goes out to farm at 8 a.m., returns after 4 p.m.
Goes out to farm at 8 a.m., returns after 4 p.m.
Stays at home as he is tired and has stomach trouble.
224
N.
AGRICULTURE : PRODUCTION
M . K.
Labour group: father and Labour group: father and Labour group: one man three sons one son
25.6.1936
Stays at home, his hand not yet well . Sons go out to farm .
26.6.1936 (Friday) Stays at home.
27.6.1936
Goes out to farm at 8 a.m., returns at 5 p.m .
28.6.1936
Stays at home because tax clerk of chief had summoned all elders. Sons go out to farm.
Goes ou t to farm at 7 a.m., returns after 4 p.m .
Stays at home.
Goes ou t to farm at 8 a.m., returns after 4 p.m .
Stays at home for same reason as N. Son goes ou t to farm.
Goes out to farm at 7 a.m., returns after 5 p.m .
Goes out to farm at 8 a.m., returns after 4 p.m .
Goes out to farm at 7 a.m., returns for mid • day meal.
Goes out to farm at 7 a.m., bu t returns early to meet the tax clerk.
CONCLUSION S
Throughou t this chapter I have stressed the technical aspect of farm • ing rather than its sociological context; I have laid particular emphasis on the time-factor. Nupe agriculture, complex and ambitious as is its scope, depends for its success on a very hig h degree of technical efficiency, both wit h regard to the organization of labour and individual skil l and efficiency. Th e most conspicuous gauge of this efficiency lies in the successful use of the intricate and finely cut schedule of planting, weeding, and harvesting. Thi s may not appear too easy: in Kutig i I learned—from the best farmers of the country—that the careful sowing of cotton may interfere wit h the second weeding, whic h is particularly important for late millet ; this competition of two main crops for the attention of the farmer has often forced the planter to choose between them. Another complication in connexion wit h the growing of cotton is that the careful tending of cotton may hinder the yam harvest and the immediate preparation of the new yam-plots—an operation which appears preferable to many. But even the most rigi d schedule of work must allow for a certain latitude. I have already given a few examples showing this. However, the essential point is that these deviations fro m the
‘standard’ times are recognized as such; there is a ‘standard’ schedule of farm-work and of its existence all farmers are equally aware, those who follow it as wel l as those who avail themselves of the latitude allowed in it . A t every phase of farming operations a farmer wil l be able to tell whether or not he is ‘o n time’ . Th e consciousness of the ideal time • table is clearly expressed in the common distinction between what ‘good’ and what ‘bad’ farmers do or are expected to do. Th e latitude whic h the ideal time-table allows carries its own corrective wit h it , be it merely
AGRICULTURE: PRODUCTION 225
moral or of a more concrete order. Ground-nuts, for example, may be planted at any time between the end of Apri l and the end of May ; but the late planner, as all agree, wil l be ‘late for the tax money’. I n Bida I watched farmers making new ridges for millet-sorghum in a field that had been already under cultivation the previous year rather late in the season; when I asked why they di d not use the ‘shifted’ ridges, they explained that they had not performed the ‘shifting of the ridges’ the previous year, adding rather apologetically that they, the Bida people, were not as keen and efficient farmers as the country people, who as they well knew would rarely omit that operation.
Thu s theje is at least a moral premium put on the keeping of the delicate time-table and the overcoming of its various difficulties. Pride in farming achievements, publicly recognized and applauded, becomes a stimulus for attaining that high degree of efficiency on which Nupe agricultural production depends in so high a measure. Its final com• pensation lies, of course, in the concrete success of its ambitious scheme. Wel l carried out, it promises to satisfy the nutritive as well as the com• mercial demands of the community.
Here we are brought back to the thir d of the three conditions which appear to determine the Nupe farmer’s agricultural planning and his choice between different crops and systems of rotation, namely, the con• sideration of the return—adequate return—which he may expect from his cultivation. If in his efforts to combine the cultivation of a certain number of crops the cultivator must take into consideration the efficiency of the labour at his disposal, he must, conversely, also consider how far the exercise of that highly efficient labour organization, or its direction to a particular task, is warranted by the prospects of an adequate return. Th e role played by considerations of this nature in Nupe agricultural economy is shown in the following examples. In 1934, the year of the
‘slump’, the majority of farmers in the ground-nut-growing districts declined to dig out the second ground-nut crop of this year as no sales were possible and it seen^d hardly wort h while to undertake the long and (since outside help would have to be called in) expensive task with • out the prospect of monetary compensation. Again, in 1936, a far better year, I learned that a number of Bida farmers who for various reasons had to carry out their farming single-handed were selling their harvest of ground-nuts together wit h the seed ground-nuts whic h they ought to have kept for the next planting season. Th e explanation was that one man alone could not grow any quantity wort h selling. What they di d therefore was to grow ground-nuts every second year only, and to buy seed ground-nuts in the intervening year, thus concentrating their modest ‘labour-organization’ on food crops and commercial crops alternately.1
1 In figures this calculation works out as follows in the case of one of my informants: to have sold ground-nuts exclusive of the seeds would have brought him is. 6d; instead of which he sold the lot for 4s., and bought back ground• nuts for seeds the next year, from profits made on mat-weaving, for 2s. 6d. This might not look a very convincing calculation to us, but one must realize that 4$. was almost half the tax money in his case (9s. 6d.), while a sum of 1s. 6d. would only have been a negligible contribution towards that most important expense.
Q
226 AGRICULTURE: PRODUCTION
It is natural, though from a certain point of view unfortunate, that to• day examples of this kind will almost invariably refer to modern com• mercial crops. The consideration of ‘adequate returns’ is thus given a narrower, and very specific, meaning. It cannot be denied that the extensive cultivation of these modern crops has accentuated the element of economic calculation in Nupe agricultural planning. Not only has the introduction of these crops considerably complicated the planting schedule, but it has also immensely emphasized the monetary aspect of the question of adequate returns. The ‘all-or-nothing’ viewpoint, or in the language of the economist, the ‘marginal’ considerations, which our examples have revealed, are no doubt due to its influence., I have said that these modern influences have ‘accentuated’ the element of economic calculation—they have not introduced it for the first time. To be satisfied with what the Nupe call, by a stereotyped formula, farming ‘for the mouth only’ is characteristic only of those sections which practise agriculture on the most modest scale, as a secondary occupation in addition to a craft, a trade, &c. We shall learn presently that even in its original form, before modern economic influences were brought to bear on it, Nupe agricultural production included a large measure of produc• tion for marketing and exchange.
Storage
DISPOSAL OF CROPS
We find both storage on the farm and in the house. Yam, sweet potatoes, cassava, and in some places the portion of millet and sorghum meant for sale are stored on the farm, either in the small huts (evu) which serve the farmers as temporary shelters, or in special store-huts (kpyakpya) built of sticks and branches put up, tent-like, against a tree and covered with leaves or grass. Inside, the fruits are heaped on a raised platform. The yam fruits are stored whole, and the fruits needed for food or marketing are collected from time to time, about twice a week, mostly by the men. The storing arrangement in this stick-and- branches structure is adapted to the need of the root crops for a drier and more airy storage than the clay granaries in the houses could provide. As regards the grain crops intended for sale, they are stored in these huts only for a short time, being disposed of very soon after the harvest. In the districts where little yam is grown, especially in the country round Bida, all crops, including yam, are stored in the house. Of cassava only small quantities are stored; as fresh cassava does not keep, it is first dried in the sun. But as the harvesting of cassava is not bound to any special season it is frequently dug out and used for food at odd times through• out the year; often in fact, dug out and eaten raw, it serves as a con• venient ‘quick lunch’ to the farmers while they are working in the fields.
In the granaries in the house are stored millet, sorghum, maize (so far as this crop, grown in small quantities, involves any special storing), beans, ground-nuts, and all greens, each crop being kept in a separate granary. These granaries, called edo, are distributed between the living huts of the compound. Of egg-like shape, about 7 feet high and 4 feet in
A FAMIL Y HEAD REPAIRING HIS GRANARY
GRANARY AND COOKING PLACE IN A VILLAGE COMPOUND
AGRICULTURE : PRODUCTIO N 227 diameter, the y are buil t o f clay, wit h strong, thic k walls. Ove r th e dome •
like to p a grass roof is placed as protection fro m sun an d rain . Th e
granary is raised above th e ground , resting on stones or solid blocks of clay, to mak e it safe fro m termites. On one side, a littl e above the middle , is a rectangular opening whic h can be closed completely wit h a clay lid . At th e beginning of the rain y season yo u wil l see th e people everywhere busy repairing thei r granaries, or buildin g ne w ones, whic h are to take the ne w crop. Later , whe n they have been filled, the li d is closed, and the granaries the n left unti l thei r contents are require d for food or sale.
Mille t an d sorghu m are stored unthreshed, i n whol e heads. The y keep remarkabl y wel l i n these granaries. Th e metho d o f storing un • threshed grain ensures a certain measure of ventilation an d also prevents the grai n fro m being packed too tightly in the granary, and thus becom• in g dam p an d mouldy. 1 A s already mentioned , late mille t and sorghum are left to dr y on the fields for a few days before being taken home to be stored. Bulrush-millet , whic h is reaped durin g the rains, ma y have to be drie d in special roast ovens before it is fit for storing. Thes e ovens, yakuy whic h are used for various purposes besides th e dryin g of millet—soap-makin g and the dryin g of shea-nuts—are buil t of clay, of roughly cylindrical shape, about 5 feet hig h an d 3 feet wide , an d are divide d into half horizontally b y a sieve-like grill . Th e botto m part o f the oven takes the fuel (firewood), the uppe r half, whic h is open at the top , is the receptacle for the stuff to be drie d or roasted. Frequentl y a small grass roof is pu t up over the oven to provid e shade for the wome n workin g there. Ever y Nup e village possesses a large numbe r of these ovens. The y are placed outside the compounds, often on the outskirts of the village. The y are not the property of individua l households, but are buil t by the me n of the village or village section in collective work , an d used by th e women—th e dryin g and roasting being thei r work .
Th e seed plants are as a rule at once divide d up fro m the crop. Th e seed yams are stored on the farm—unless ya m is planted immediatel y after the harvest. Th e heads of millet , sorghum, and maize whic h are later to supply the seeds are stored, not in the granaries, bu t in the hu t of th e famil y head, wher e they are kep t on a shelf of bamboo or woo d underneat h the roof.
Th e crops undergo n o further treatmen t durin g the tim e o f storage. A s mentione d already, the Nup e d o not dr y yam. 2 Sorghu m and mille t are not threshed unti l the grain is being prepare d for cooking or sale, although heads of unthreshed grain , packed in large bundles, are also offered for sale on the markets. Th e threshing is done in the courtyard
1 Thi s is actually the almost universal method of grain storage in Norther n Nigeria. An indirect testimony to its efficacy I chanced to obtain from Hausa settlers in the Anglo-Egyptia n Sudan who m I had occasion to visit in 1939. Chiefly for lack of proper buildin g material for the construction of the Nigerian type of clay granary, these settlers have adopted the Sudan method of storage, i.e. storage of threshed grain. The y all agreed that this method was inferior to the Nigerian, and preserved the grain muc h less satisfactorily,
2 It is interesting to note that the Nup e colonies in Yorub a country and the
Nupe communities whic h are livin g side by side wit h Yorub a communities in
Ilori n Province have adopted the dryin g of yam.
228 AGRICULTURE : PRODUCTION
of the compound or under a shady tree in front of the house. Th e grain is either spread out on the floor and beaten with sticks or pounded in a big wooden mortar. Th e former method can be used only in houses which possess a hard, clean floor in front of the huts; where there is only sandy ground (as is the case in many smaller villages) the second method has to be adopted. Th e first method, besides, is only possible when large quantities of grain are threshed. The small quantities norm• ally needed for daily use are done more quickly by pounding. In either case at least three people (women and girls) will be working together. After threshing the grain is winnowed. This is done with the help of two platters of basket-work, by repeatedly pouring the grain slowly from one into the other; the woman who does the winnowing walks up and down in front of the house, letting the breeze remove the chaff. Finally the grain is milled, either on a grinding-stone or, again, in the mortar. Th e latter is the heavier but quicker work, the former is slower but gives the finer flour. The empty heads of grain left over after threshing are used as fodder for horses; the chaff, collected after winnowing, as fodder for sheep and goats.
Food Crops and Market Crops
Th e foregoing description of methods of storage and preparation does not embrace all the crops grown in Nupe. Before we can make our list complete we must consider another aspect of the question, namely, whether the produce is used for household consumption or intended for sale. No comprehensive simple distinction is possible: we can only mark extremes—a few crops which are used exclusively for consumption or for marketing, respectively, while the rest, the large majority of crops, may be used for either purpose. The produce used for sale comprises first of all the non-food crops: cotton (for European export), indigo, and henna. Food crops grown almost exclusively for the market are: late ground-nuts and red pepper (for European export), and rice and onions (for internal markets or export to other countries in Nigeria). Th e crops which are rarely sold and used primarily for household con• sumption are few: cassava, beans, a few vegetables, and sweet potatoes and maize grown in small quantities on inland farms.
In the case of crops grown for export the question of storage does not arise, at least not for the producer himself, for this produce is sold almost immediately after harvesting (e.g. late ground-nuts, rice). Instead, we have two new questions to consider. First, that of transport and pack• ing. But this also is mostly a matter for the trader and middleman, not for the producer. Th e native traders provide the big grass bags in which the rice is packed and transported, the European trading-companies the sacks for cotton; the other produce is transported in the ordinary receptacles of the country, large baskets, gourds, or tin basins of European manufacture.
Second, there is the question of the special preparation, an elementary manufacturing process, which is frequently applied to the agricultural produce intended for local trade before it is put on the market. In the markets of Nupe agricultural produce is sold either as it is (e.g. yam,
AGRICULTURE : PRODUCTION 22 9
rice, and partly ground-nuts) or after it has undergone a rudimentary
‘refining’ process (such as the threshing and milling of grain, the shelling
of ground-nuts, &c) . Side by side with this trade there exists an
extensive local trade in the ‘ finished’ product, which means various
kinds of prepared food. I cannot attempt a detailed description of the
intricate and varied art of Nupe cooking. I will only enumerate the more
important products of this native ‘food-industry’, which, like most food
preparation, is in the hands of the women. It is of considerable import•
ance for the economy of the Nupe household. There are three main kinds
of such prepared food which are sold everywhere in the markets and
streets of the towns and villages of Nupe: (i ) a gruel of millet or sorghum,
made with water, and called ekiva in the more jelly-like and zambu in the
more solid form. It is eaten for breakfast or lunch, and nicely done up in
banana leaves, bought by labourers, canoe-men, travellers, and some•
times farmers going out to their fields—men who are unable to have their
meals with their families or to have food brought out to them from home.
(2) Yisa (Hausa, fura), bread-like balls made of late millet or sorghum
and rolled in flour. Me n and women attending markets are fond of
buying this food and eating it with milk which they buy, there and then,
from the Fulani women. (3) Masa, kuli-kuli, bdnkuru, and kara—
savouries of different kind, made of beans, sorghum- or millet-flour or
ground-nuts, fried in shea-butter and prepared with various spices.
They are eaten at any time of the day, practically by everybody who can
spare an anini (a tenth of a penny).
To turn to the food crops proper. Our chart has established the even distribution of food crops over the year’s cycle, a distribution which should, normally, leave every household well provided with food from harvest to harvest. Old yam should last till the new yam is dug out, sorghum and late millet till the new sorghum harvest—or at least till bulrush-millet—
is reaped. I have found that June and July are the critical months: yam may begin to give out by the end of Ma y or June, about six weeks or two months before the new yam harvest, and late millet and sorghum may become scarce by the middle of June. But bulrush-millet, which is harvested about that time, and maize, both from inland and marsh-land crops, make the transition easy.1 So do also fresh ground-nuts, new potatoes, and partly the new rice, all harvested before the late main crops; I have mentioned already the ‘stopgap’ character of cassava.2 A dry year„ a late or failing harvest of early millet or first yam, may, of course, upset the balance temporarily in a district—till imports from the river area where yam matures earlier, or other districts, which have fared better, set it right again.3 But the nutritive balance of the country must always remain dependent on the contribution of trade and markets:
1 See p. 213.
2 Cassava-flour as a substitute for grain-flour appears on the Nupe markets only for about two months, during July and August. Corn-gruel, a daily food in many Nupe households, can be made with rice-flour instead of sorghum- or
millet-flour, if the latter has become scarce. This substitute is more expensive than ordinary corn-gruel, and is also said to be less pleasant to the taste.
3 In 1936 the first yams appeared on the river markets on 6 July, while inland
the first yams were dug out a fortnight later.
23 0 AGRICULTURE: PRODUCTION
yam, maize, and early millet are not grown everywhere; special crops are cultivated on marsh-land and inland farms; and besides, there is the large population of Nupe which does not live on the land and can obtain its food only from the markets of the country. Th e Nupe agricultural system, as we have realized already, is adjusted essentially to a highly developed system of exchange and marketing. Th e surprisingly small number of crops grown exclusively for household consumption is perhaps the most striking proof.
Th e degree to which the individual farmer, and the farmer in the different districts of the country, adjusts his planting to the principle of exchange economy varies considerably. In the marshes of Niger and Kaduna almost the whole production of rice is intended for sale and export. But at the other end of the scale we find farmers who, able to cultivate a few crops only, have just enough to provide food for their families. We have already examined some of the conditions determining this changing balance between trade crops and food crops. Ultimately it only reflects that greater fundamental balance that underlies Nupe peasant economy: between the need of food for the group, and the need of commodities which are obtainable only by means of exchange, that is, money—which can itself be obtained only from the sale of farming produce.
TREE CULTIVATION AND FRUIT INDUSTRY
As stated in the first chapter, most of the fruit-bearing trees in Nupe are wild growing. Only two or three kinds are cultivated: the kola-nut tree, ebi, and to much lesser extent the banana, yaba; to these two indi• genous plants we must add to-day the European imported mango tree, tnangoro. In the case of trees the main productive activities refer there• fore not so much to the planting and tending of the plants as to the collection of the fruits and the preliminary manufacturing process of the same order as that described in the preceding section. Only in a few cases can the fruits of the trees be utilized directly—for food, cooking, or marketing. Most fruits have to undergo a special ‘refining’ process before they become marketable or ready for use.
Banana
Only in Bida, Patigi, and Labofcidid I see real banana plantations. They are small and of little importance, for bananas have little commercial value in Nupe; they are sold for an extremely low price (a bunch of about twenty for \d. or less)—if they are sold at all. Bananas are eaten raw, and play in Nupe a part comparable to our occasional eating of fruit. In the banana plantations the plants are arranged in a circle of about
6 feet in diameter. Th e important thing is to plant them very deep in the ground (from 3 to 4 feet), and the whole circle has thus to be ex• cavated to that depth. It is this heavy and slow work—unprofitable work from a commercial point of view—that, according to my informants, dis• courages the Nupe farmer from growing bananas. In Bida, where the soil is poor, the banana plot has to be manured with animal dung; in the bush plantations and in the forest groves where the banana grows wild the
AGRICULTURE : PRODUCTION 231 leaves are allowed to fall off and rot on the ground—an efficient system
of natural fertilization, though not, apparently (unlike the ‘green
manuring’ practised in agriculture), applied consciously.
Kola-nut
We have had occasion already to speak of the Kusopa, the kola-nut planters in the forest groves of Labozi. Th e Kusopa of Labozi and the neighbouring districts devote the larger part of their agricultural work to the cultivation of the kola-nut tree, growing little else besides—only a little millet and sorghum. Th e method of kola-nut cultivation is inter• esting. Th e trees are planted at the beginning of the rains. They flower for the first time after five years, about July, and after six years bear fruit, towards the end of October. Th e fruit is picked in February. From every harvest a gourdful of nuts is kept to be used as seeds, for nearly every year new trees are planted to replace old and barren ones. When planting a new tree the seed is buried hand-deep in the ground. To protect the seed the place where it was planted is covered with twigs; moreover, four banana plants are planted around it, in a square, about 6 feet by 6 feet. As the banana grows faster than the kola-nut the four banana plants provide shade for the tender young tree, and at the same time secure the necessary distance between newly planted trees. When the kola-nut tree reaches a certain height and no longer needs the shade of the banana plants, the latter are dug out and removed. Th e tending of the young kola-nut tree involves constant weeding—in which no hoe is used; only the ‘ bush-knife’, so as not to hurt the delicate plant— and much watering. During the dry season the seedlings must be watered every four days. The need for humid soil explains the fact that the kola thrives only in the most humid part of Nupe, in the thick forest groves of Kusopa country.
This environmental factor is fully understood by the people. Th e failure of kola-nut cultivation in other districts of the Emirate (on which we shall have more to say presently) is partly attributed to the less favourable soil. But the traditional monopoly of kola-nut planting in Labozi is not, in their eyes, fully explained on such empirical grounds. A legend that has grown up round the cultivation of the kola- nut tree which tells about the discovery, by a Labozi woman, of the secret of kola-nut planting, supplies another explanation more con• vincing to their way of thinking. Let me quote this legend in full.
A woman was threshing grain for beer when a Hausa from Borgu, on his way back to his country, happened to pass. He crossed the threshing- ground and put down the load he was Carrying. The woman went into the house to fetch water and to cook porridge for the guest. She brought the food and gave it to the Hausa. He ate the porridge and liked it very much. He opened his load, took out two kola-nuts, and gave them to the woman to thank her for the food. At that time there were no kola-nuts anywhere in Nupe country. The woman took the kola-nuts and put them on the threshing-ground. Later, when she took the grain into the house, she for• got the kola-nuts and left them lying there on the threshing-ground. One day, when she came out to the threshing-ground again, she saw these kola- nuts. She said: ‘Look, the nuts which the Hausa gave me have come
23 2 AGRICULTURE : PRODUCTION
up. ‘ She went and told the men of the house. They all came and looked. The y cleared the ground around the young plants; they had two kola-nut trees then. After three years the trees bore fruit. Th e men picked the fruits and took them home. They broke the nuts open and ate them. The y picked out the kernel, took it into the bush and planted it . When the kola-nut trees bore fruit again the men picked the fruits and took them to the king. Thi s was the beginning of the kola-nut in Nupe. When they had taken the fruits to the king, he said: ‘The y are very nice indeed.’ But the following year the trees di d not bear fruit. So the men went to see the soothsayers. The y said that the trees woul d not bear fruit again unless they sacrificed a human being. So the men went and tol d the king, and said: ‘I f we do not sacrifice a human being our trees wil l never bear fruit again.’ Therefore the king gave them permission to take a human being and sacrifice it . They di d so, and the trees bore fruit again. Thi s was the beginning of human sacrifices in Nupe, wit h the permission of the king.
As the legend explains the origi n of kola plantation in the country of the Kusopa, certain magic rites whic h are practised by that group—and by that group only—emphasize the unique position of the kola-nut planters. Onl y wit h the knowledge of that special magic can the success of the difficult and precarious cultivation be assured. Th e Labozi people practise tw o kinds of this professional magic. At the time when the kola trees are i n bloom every planter wil l collect the dead leaves on his planta• tion , place them in a mortar, and pound and then bur n them on the plantation. There is no spell or special ritua l linked wit h this act, bu t by it , as by a magic act, the planter believes he wil l ensure a ric h harvest. Th e second magic connected wit h kola-nut planting is performed for the whole group by the Map Tsuba, the Master of Threshing (i n the legend the first kola-nuts were growing from the threshing-ground), the village priest of Labozi. Annually, about two months before the picking of the kola-nuts, he performs a sacrifice in front of the hu t in which , as the legend has it , the woman lies buried to who m the Kusopa owe the know• ledge of kola-nut planting. To-day a white ram and cock are sacrificed, the blood together wit h some beer is poured on the threshold of the hut, and Soko, God , is invoked that he may let the kola trees thrive. Formerly this sacrifice is said to have demanded a human victim . (See p. 90, n. 2.)
Thi s connexion between planting technique and mythical and magic ideas, common all the worl d over, needs no special explanation. Bu t in the case of the kola-nut planters it gains an additional significance. Th e special nature of the rite, its infrequent occurrence, correspond to the special monopolized technique. In the eyes of the people who, not kola- nu t planters themselves, accept the existence of these special rites, it confirms and endorses the exceptional position of the Kusopa as the masters of a unique craft. Abou t fifteen years ago the Administration planned to introduce the cultivation of kola-nuts into the other districts of Nupe. A few expert planters from Labozi were sent round to teach the farmers in places where the soil looked promising the almost secret art of kola planting. These efforts met wit h but meagre success. Loca l conditions were to a large extent responsible: the soil in the other dis• tricts seemed capable of producing only the inferior kin d of kola, and
AGRICULTURE : PRODUCTION 233 frequently the trees bore no frui t at all. Bu t to some extent the failure of
the experiment was due to psychological factors. Th e lack of interest in
their trees whic h these novices in kola-nut planting stil l display reveals
the lack of incentive for the adoption of this new, difficult, and lengthy
enterprise, officially encouraged though it was. Th e adoption of the new
cultivation ran counter to a century-old specialization and monopoly, to
an organization of production which , firml y rooted in legends and re•
ligious practice, has come to be accepted by the people as part and parcel
of that inert, unalterable, entity : their culture—a point to whic h we shall
return.
Locust-bean Tree
Thi s is one of the most important trees in Nupe. It grows both inside the tow n and in the open bush. Every part of its frui t is utilized. Th e pods (elo) are sold, in large basketfuls in the markets of Bida. Both the whole fruits and the empty skins are marketed. Th e infusion of the skins, a black sticky liqui d (longoro in Nupe, mdkuba in Hausa), is used as dye for pottery, leather and woodwork, also as ink, and for staining and making ant-proof timber-rafters, walls, and floors; it gives them that neat polished appearance as if they had been treated wit h tar. A yellow powder found inside the pods (elo nuwa) is used to sweeten milk , water, and other drinks. Th e fermented seeds (elo kosu), pounded and kneaded into small flat cakes (kula), are sold on the markets; strongly tasting, and even more strongly smelling, they for m an indispensable ingredient of Nupe cooking, being used to flavour practically every dish. Th e fruits of the locust-bean tree are plucked by the women, wit h the help of long poles whic h have a forked piece of wood, almost like a pair of shears, tied to the top.
Shea-nut Tree
No less important than the locust-bean tree, the shea-nut tree grows wil d everywhere in the thi n bush country. Th e picking of the fruits is again women’s work. Small groups of women go out together, help each other, and eventually divide the fruits whic h they have gathered. Th e women rarely stray very far from roads and paths, and never into the uninhabited bush. Th e fruit , eko, is either sold to the trading-companies or manufactured by the women themselves into shea- butter, an important fat in Nupe cooking, and an important commodity in the markets of the country. Th e shea-nuts intended for export are first dried in the sun or in the roast-ovens described above. Shea-butter (emikote) is prepared in the following fashion :
After having been dried thoroughly, the nuts are shelled by hand. As a rule this lengthy work is left for one day, and next morning the inner, finer skin is peeled off and the nuts are pounded in a mortar to a coarse• grained flour. Plenty of water is added, and the dough is kneaded wit h hands and feet for a whole day. In the evening the preparation is covered and left standing overnight. On the next day the dough is transferred into a big clay pot. A little water is added and the pot pu t on the fire, where it is left boiling for the whole day. Durin g the boiling the pure fat comes to
234 AGRICULTURE : PRODUCTIO N
the surface; it is skimme d and placed in another pot, where it hardens to shea-butter. Th e residue of the pure fat is used as a fuel for the Nup e clay lamps.
Oil-palm
Thi s grows in th e thic k groves near watercourses and , covering a larger area, i n th e dense forest countr y roun d Yeti , i n th e nort h o f Nupe , whic h i s inhabite d b y th e Kusopa , th e Keepers o f th e Forest. Th e Kusop a of Yet i are accorded a position somewhat similar to tha t of th e kola-nu t planters of Labozi . The y too are regarded as ‘ experts’ fro m most ancient times, an d thei r chief holds th e titl e Yikunu Dzuru, ‘ Oil – palm’ , as a symbo l of his tribe’s close association wit h th e tree.1 Bu t the y are ‘experts’ on a muc h smaller scale an d clai m no monopol y wit h regar d t o th e products o f thei r trees. Moreove r th e palm-oi l produce d i n Kusop a countr y plays a n unimportan t par t i n the economy o f Nup e a t large. I t contributes only ver y littl e towards the needs o f th e country , an d palm-oi l has to be importe d regularly—an d has most probabl y been so importe d for centuries—fro m th e south, i.e. th e great oil-producin g countries i n th e forest belt. Th e export o f palm-kernels fro m th e Emirat e i s negligible. Th e fruits o f the palm-tre e are plucke d b y th e men , wh o clim b up th e trees wit h the hel p of a hoop mad e of supple boughs, an d thro w the fruits o n the ground . The y are later collected b y th e women . Th e manufactur e o f oi l fro m the palm-kernels i s women’s work .
Th e kernels of about twenty fruits are extracted and the n boiled in water for one day. The y are left to cool overnight. Nex t mornin g they are placed in a mortar, and the wome n knead the soft mass by treading on it . Later , cold water is poured over it , and a first oil comes to the surface. It is skimmed and boiled again, in another pot. Th e oil whic h now appears on the surface is the pure palm-oil , end dzuru (lit . red oil) . Th e residue is drie d in the sun, later kneaded again in a mortar, and finally boiled. Fro m this boiling a second, less pure, oil is extracted, called edi. Twent y palm-fruits give tw o small pots of pur e palm-oi l (about 4 gallons).
Palm-oi l plays an importan t part in the Nup e household; the pure palm – oil is the chief ingredient in all soups and sauces whic h are eaten wit h the universal Nup e dish porridge-and-fish an d porridge-and-meat. Th e less pure oil, whic h is gained as a second extract, is used for cooking yam , potatoes, and beans, bu t also serves—a rather indiscriminate ingredient— as a hair-lotion, ointment, laxative, and lamp-oil . On the Nup e markets only the pure ‘red’ oil, not the whole frui t nor the less pure second extract, is offered for sale.
As palm-oi l is also used in the preparation of native soap we ma y add the description of this process; it is again women’s work . Wate r is filtered through hot wood ashes, collected in a pot and boiled. In another pot a bowlful of palm-oi l has been prepared. To this an equal amount of boiHng ash-water is added. Th e mixtur e is again boiled, and is constantly stirred til l it hardens. Thi s mass is left to cool, and is afterwards kneaded into large balls. Th e quality of the soap depends on the purit y of the oil used.
1 Th e Kusopa of Yeti are also said to ‘cultivate’ the oil-palm. What this term implies in reality, and how far it refers to real tending of the trees, weeding, thinning out, &c , I cannot say, not having been to Yeti myself.
AGRICULTURE : PRODUCTION 235
But as even the purest native-made soap can hardly compete with im• ported soap it is not surprising that soap-making is gradually disappear• ing, and has disappeared completely in Bida and the more progressive villages, such as Kutigi or Mokwa.
Indigo
Indigo, eci, is gained from two different plants, the indigo grass (tsentse) and the indigo tree (eci or baba). Th e grass grows wild in the thinned-out bush that surrounds most Nupe villages, but is also culti• vated in some parts (e.g. Mokwa) and grown as a first crop before millet- sorghtim.1 Th e dye is gained from the decomposed grass which, after having been cut, is left lying in the open for from two to three months. Th e powder which is obtained by this method is mixed with ashes and placed in a sieve-like basket. Water is filtered through, and this gives the dye, which is allowed to drip directly into the deep clay-cemented dye-pits (called marina, as in Hausa). This technique of dyeing follows the Hausa fashion; and in Nupe as in Hausa the dyers, who also manu• facture their own dye, are men.
Th e indigo tree is found scattered everywhere in the open bush, most densely in Trans-Kaduna. In the district between Kutigi and Sakpe we find even something like indigo-tree plantations which are tended by another group of’tree-experts’. Here every village possesses large num• bers of indigo trees, which are to some extent cultivated, i.e. cut and thinned out when necessary; they are not, however, planted or trans• planted, but left to fertilize themselves. Among these indigo planters the making of indigo is men’s work. But in the districts where indigo is gained from the scattered, wild-growing tree, the picking and preparing of indigo is done by women. Th e method of preparing the dye is in both cases the same; the leaves are pounded in a mortar, then left to ferment in water, mixed with ash, and finally rolled into balls. Th e finished product is used exclusively by the women dyers, that is, in a practice of dyeing in pots, and which appears to represent the original dyeing technique of Southern and Central Nigeria; according to Nupe tradition it was introduced into the country by the Konu weavers and dyers (see p. 188, n. 2).
Nupe country offers only a restricted market for indigo. Th e indigo manufactured by the wives of the Konu weavers, the total quantity of which is small, is either used by the women themselves or is sold in the markets—in the districts more than in Bida—where modern products have ousted almost completely the native technique of dyeing. Th e much larger output of indigo dye which is produced by the cultivators of the indigo tree is disposed of with the help of middlemen who trade it to Yoruba, the most favourable market for indigo, or is taken by the producers themselves to the nearest market of that country (Ilorin). This latter method secures the better price but may interfere with the normal agricultural activity. A family of farmers near Kutigi who
1 This method of indigo cultivation is practised on a large scale by the
Gbari.
236 AGRICULTURE : PRODUCTION
owned twenty-five indigo trees had to take their produce to Ilorin market just at the time of the second yam harvest. However, they pre• ferred to sell indigo first and to postpone the digging out of yam for a month or two, although this meant losing the always profitable market for the first yams of the season. Th e element of economic calculation, to which we have already assigned its place in the planning of Nupe agri• culture, reveals its efficacy in this new context as well.
Not much need be said about the other fruit-bearing trees. Th e fruit of the mango tree is eaten when and wherever it can be picked or gathered up from the ground; it is not traded—like the banana it is ‘just fruit’. Th e tender young leaves of the baobab (muci) are used as food; they are pounded and sold in the markets as a vegetable which is added to soup and meat. In order to obtain the tender leaves the owner of the tree lops the big branches so as to make them sprout new leaves. Th e cotton• wool-like kapok which is contained in the capsules of the silk-cotton tree (lembubu) is traded in one or two markets of the country; it is bought by the Bida cushion-makers and also by the canoe-men on the Njger, who use its fine fibres to stop leaks in their canoes. Th e commercial use of this tree and its fruits has not yet penetrated into the more remote districts: while in Kutigi, on the main road, natives were collecting kapok to sell it to the Bida cushion-makers, I found in Sakpe, only ten miles away but right in the bush, the cotton-wool left unused, lying on the ground to be carried away by the wind. When I asked a number of men why they did not collect it and take it to the market—I could call upon a boy from Bida whom I had with me to bear out my contention that this stuff possessed commercial value—they replied, not wholly convinced, that it would be too hard work. Another example of rudi• mentary ‘ economic calculation’, with leisure weighed against money- profit.
INNOVATIONS IN PRODUCTION
From this discussion an interesting question arises: why have the Nupe, who have mastered the cultivation of kola-nut and banana, not taken to the cultivation of other trees as well, but left most of their fruit-bearing trees to the ‘care of God’. If you ask a Nupe farmer he will give you one of two answers: either the vague theological excuse that God has planted the trees, and that God will look after them; or the more ration• alistic but no less acquiescent explanation: ‘I f I planted a tree I should never see its fruit, only my grandchildren would profit by it. Our grand• fathers have not planted trees for us, why then should we plant for our grandchildren?’ Th e fact, well known to everyone, that the Kusopa plant and cultivate trees is yet easily dismissed. This is one of the cultural features typical of certain tribal sections which are simply taken for granted by the other groups. Th e myth that explains how it came about that just this and no other sub-tribe happened to become kola-nut planters, the special magic which the members of that sub- tribe are believed to own provide the spiritual background for this attitude. This acceptance of an existing productive specialization as if it were a final and unalterable fact was effective enough, as we have seen,
AGRICULTURE : PRODUCTIO N 237
to check the attempts to extend kola plantation in Nupe. Whether or not it wil l check other attempts at agricultural innovations one cannot say.! Often in discussions of this kin d the reluctance of a native group to adopt a certain innovation is summarily dismissed as being due to the inveterate, almost proverbial, ‘conservatism’ of the native races. Clearly, such an explanation does not meet the case. As regards the Nupe we should have to contrast their alleged ‘ conservatism’ wit h respect to trees wit h the readiness wit h whic h they adopt new crops and agricultural techniques. Evidently, to adopt a new crop whic h shows results in a year’s time, and whic h is easy to tend, often easier than the traditional crops (as in the case of maize or cassava), and to take up the difficult work of planting trees whic h may not bear frui t for six or seven years, is an entirely different thing . One can be easily fitted int o the framework of established agricultural activities while the other involves a complete readjustment of the productive system. Th e Nupe are, withi n limits , extremely progressive people, and keen on tryin g out new things. These limit s are determined by two factors: first, by the time that an ordinarily crowded workin g year could allow for additional tasks; other work, or precious leisure time migh t have to be given up for the sake of the in • novation—a change that may or may not appear wort h while to the Nupe farmer. Th e second factor is the measure of change whic h a readjust• ment of the native production woul d entail in this closely kni t economic system, wit h its exact balance of specialization and exchange. Every radical change in the production, such as the adoption of tree-cultivation in a district, may upset this balance; it may, in other words, make the innovation less profitable. Th e fact that the Nupe farmer is go-ahead wit h respect to certain farm techniques and ‘conservative’ wit h respect to others is not due to any triba l or racial ‘conservatism’, but reveals itself
as a clear instance of the working of free economic choice.
TECHNIQU E AN D KNOWLEDG E
So advanced a technique and so conspicuous a standard of husbandry must involve a body of theoretical knowledge whic h goes far beyond the counsels of peasant lore, and could avouch statements of very nearly scientific authority. Thi s is no doubt the case in Nupe farming. Th e understanding of the basic principle of manuring, the fixed rules wit h regard to rotation and lyin g fallow, the practice of thinning-out the growing plants, whic h is correctly explained as being intended to give the plant more space, they all confirm this.2 Moreover, while in many African
1 Rev. Ir a W. Sherk, the head of the U.S . Mission at Mokwa, has been successful in making the people of a small neighbouring village adopt the cultiva• tion of grapefruit. This innovation, carried out on a very small scale, and in• volving hardly any difficult readjustment of the native productive organization, bears out what is said later in the text about the limits that are set to all such innovations.
2 I must record that one of my Bida informants, an experienced farmer, denied the ralue of ashes and the burning of fields as manure. He held that animal manure (with the regular use of which he as a Bida farmer was well acquainted) was the only effective manure. This critical attitude only emphasizes the element of empirical knowledge in Nupe farming technique.
238 AGRICULTURE : PRODUCTIO N
communities rulings on agricultural technique and the organization of agricultural labour appear in the guise of magic devices and rituals, there is none of this in Nupe agriculture, which remains matter-of-fact and consciously based on empirical knowledge.1
In a productive organization of this kind special emphasis must be laid on the position of authority of the old and experienced farmer. It can be studied in action whenever groups comprising both older and younger men co-operate in farm-work. In family work as well as in non- family co-operation the older men always lead the work or supervise the work of the younger men; they point out mistakes to them and teach them the important rules of farm-work. I have frequently seen men scold boys for damaging a plant when weeding, or for planting seeds too close. Boys are made to work on the farms from the age of 6 or 7, first with toy• like light tools, later with the real heavy hoe, in order to learn through practice and under constant supervision. In Nupe farming very little is left to chance. I have, for example, never heard of anyone starting farm-work too early or too late in the season and thus wasting the seeds— a mistake not uncommon among more primitive agriculturalists. Th e quasi-scientific aspect of Nupe agriculture is made most evident in the
‘ theoretical’ rulings which the people themselves are always ready to lay down. Thus informants formulated the following rule: there exist two kinds of rice, one taking 90 days, the other 180 days to mature; or this: the latest date for the planting of ground-nuts is the beginning of the third month (counted from the beginning of the planting season). Not all such rules are fully convincing to the European agriculturist—we remember the rule about planting yams before the dry season in order to make them grow bigger. But, on the other hand, there is plenty of evidence to show that the existing body of native farm-knowledge admits of experiment, and is constantly assimilating newly gained ex• periences. There is, for example, the new ‘rule’ about cotton and red pepper being excellent first crops before millet and sorghum. Experi• ment has besides been forced upon the native cultivator by the introduction of modern crops or varieties. Thus the people have tried with varying success to grow cotton on different kinds of soil; cotton-growing has also caused new difficulties with regard to the distribution of agricultural work, which can be solved ultimately only by experiment.2
Long experimenting, however, which does not lead to the incorpora•
tion of the new experiences into that solid body of farming-knowledge
1 Th e one exception is the magic linked with the cultivation of kola-nut trees, which itself represents an exceptional form of cultivation. Apart from this magic there exist only the annual fertility cults which set the spiritual frame• work for farming activities in general, and certain varying, unsystematic forms of magic—charms applied to hoes, or a specific magic performed over a field— which are to ensure success in individual cases.
2 In 1936 the Agricultural Officer of Bida Division travelled round the district
to instruct personally the cotton-growers in the more important centres about the proper date for planting cotton—information which, coming many years after cotton-growing had been introduced in this area, only confirmed a point of technique which the people had already discovered for themselves.
AGRICULTURE : PRODUCTION 239 easily turns into discouragement. In Kutig i and Sakpe I have spoken to
many farmers who, discouraged by the increasing difficulties that appear
to accompany cotton-growing, seemed determined to give it up alto•
gether. We may, I think, draw a valuable lesson from this. If new agri•
cultural schemes are to be introduced they must be brought in already
wel l tried-out and fully adapted to the existing conditions. Th e adoption
of a new crop or technique is to the native in itself a large-scale experi•
ment ; to make the people adopt a technique whic h has not been fully
tried-out means to double this element of experimenting. I t wil l pre•
vent the new knowledge from being absorbed quickly in the ‘body of
Nup e farming-knowledge’ on whic h the people feel they must be able to
rely, and may eventually give them over to disappointment and dis•
couragement.
BY-PRODUCTS OF AGRICULTURE
Agriculture and forest-cultivation produce as by-products certain raw materials whic h are used in Nupe home industries and handicrafts. Thu s the grain-stalks whic h are left on the fields after the harvest are used for the making of fences, doors, and bedsteads: cut in fine strips, grain and maize stalks are used for mat-making; a vegetable dye, black and red, is gained from the stalk of a certain species of sorghum (dindo- rogi). Small trees whic h do not bear frui t are occasionally cut down by the owner of the field on which they stand and used as timber for house- repairs and suchlike. In all these cases it is the cultivator and his family themselves who use the plant or tree; the by-products of agriculture have no marketing value. Thi s is different in the case of firewood and grass for thatching which, as I have mentioned already, is collected or cut by the Nupe women and marketed by the men (see p. 200).
It is frequently quoted as a characteristic feature of African agricul• ture that every part of a plant or tree that can be used in any way is actually utilized by the people. Thi s is only partly true in Nupe. Only a small proportion, for example, of the grain-stalks is used in the various ways just described. A great many farmers are either unskilled in the craft whic h makes use of these by-products, or devote their spare time to other, to them more attractive or profitable, occupations. Thi s com• parative failure on the part of the Nupe farmer to utilize to the fullest possible extent the products of his land appears to be connected wit h the complexity and specialization of the productive system of the country. As we shall see later, only certain groups or individuals have specialized in mat-making, the making of bedsteads, or dyeing. Th e influence of specialization involving a clear-cut division between the agriculturist and the craftsman who utilizes the by-products of agriculture is most conspicuous in this case; bi g trees, when not fruit-bearing, may also be cut down and used as timber. But no farmer woul d ever attempt to cut a big tree himself for fear of bungling the jo b and being injured or even kille d by the falling tree. He calls in the professional wood-cutter, who possesses both the necessary skill and experience and the special magic whic h alone can prevent the spirit of the tree from revenging himself on
24Q AGRICULTURE . PRODUCTIO N
th e perso n wh o i s destroyin g hi s abode . Thes e bi g trees hav e a money – value , an d are sol d t o th e wood-cutter s an d carpenters , wh o use th e timber in their own industry. On the whole the Nupe are reluctant to cut down big trees, especially in the town , for they appreciate the shade whic h they give. Bu t it happens occasionally that the owner of a barren tree, in need of money, has it cut down and sells the timber to wood• cutters or carpenters.
In the interest of forest-preservation the felling of certain trees has been made subject to the payment of a licence fee varying in amount fro m 5s. 6d. to £i or more. Thi s new rule involves an economic re• adjustment whic h the people—professional timber-workers as wel l as ordinary farmers—have not yet been able to achieve. Th e carpenters of Bida almost went on strike when the new licence was introduced, and refused to manufacture any more wooden stools and other products of their craft for fear of being unable-to sell their goods for the higher price whic h the licence woul d necessitate. Another illustration is the case of an ol d man, a carpenter, who, unaware of the new rule, had bought a barren locust-bean tree in Bida for felling, for 7s. 6d. Whe n he was requested to pay the licence fee of 5s. 6d. he was unable to do so. Hi s case came before the Bida Court, but in consideration of his poverty and age the court reduced the licence fee for hi m to is. 4d. It is too early yet to say what effect this licence wil l have on the Nupe wood-working industry : whether i t wil l lead to a technical readjustment and a change in the kin d of timber used, or to a gradual deterioration and eventual dis• appearance of the indigenous wood-carving technique, whic h is already feeling the competition of the cheaper, and infinitely uglier, products in
Western style: packing-case chairs and tables, deck-chairs, and the like.1
1 It is to be feared that it will be this latter case with which Nupe wood-carving will meet. In other crafts these changes in material and partly style are already an accomplished fact: thus the Bida glass-workers have almost completely abandoned making their own glass, and make their bangles and beads of smelted down bottle glass of European manufacture; similarly the Nupe blacksmiths buy cheap sheet-iron at the trading stores, and no longer smelt their own iron ore.
CHAPTE R X I I I
AGRICULTURE : ORGANIZATIO N O F LABOUR
ABOU R organization i n Nupe , i n every field o f economic activity, i s characterized by a twofol d principle : the distinction between in • dividua l wor k an d famil y wor k or, i n native terminology, between buca and efako. As the division of labour implie d in thjs distinction also involves a corresponding division of income an d property , our discus• sion of the organization of agricultural labour must at th e same tim e be a discussion of the forms of propert y that obtain in relation to agricul • tura l production . Th e terms buc a and efako have, i n fact, this wide r meaning . Ever y kin d of wor k done individually , its products and the income—individua l income—derive d fro m it , are buca; an d every wor k done jointly , i n collective famil y work , producin g outpu t and income over whic h the famil y group or a section of the famil y group exercises join t rights, is efako. Thu s we find tw o different kinds of far m property : efako farms whic h are owne d and worke d jointl y by the famil y group , and buca farms worke d and owne d by individuals. Similarl y there are efako granaries to hol d the produce fro m the famil y farms, an d buca granaries to hol d the produce whic h the individua l member s of the famil y group ow n and utilize. Th e famil y group or, mor e precisely, the section of the famil y group whic h forms the efako-umt, owns fro m ten to twelve granaries whic h are as a rul e placed close to the centre of the house. Ever y buca owner has, besides, his ow n granary, whic h is usually foun d close to the wal l of the compound . Obviously , in a system of labour based no t on wages bu t on join t wor k an d a share in the pro • ceeds, the discussion of the labour organization cannot be treated separately fro m questions o f ownership an d land-tenure . O n the other hand , a co-operative group consisting of a section of the kinship group cannot be described withou t constant reference to kinship organization. Frequen t cross-references, and overlapping wit h other sections of the
book, are thus unavoidable.
INDIVIDUA L AN D FAMIL Y WOR K
Le t me begin wit h the buca, whic h can be briefly described. Ever y mal e membe r of the famil y owns a buca, wit h the exception of the famil y head wh o being the administrator of the efako has no buca himself. Fo r his personal requirements he is entitled to dra w on the efako income. Youn g boys of 8 or 10 already possess buca farms of thei r own . Bucd farms var y considerably in size, and grow wit h thei r occupants. Th e largest buca far m whic h I foun d was about one acre, i.e. about hal f the size of the plo t of lan d whic h on the famil y possessions is proportionate to every adul t worker . Th e average buca yiel d I foun d to be between one-thir d an d one-half of the individua l adult’s’ share i n the yiel d o f the famil y farm . Th e buc a owner can wor k on his far m only ‘out of business hours’, whe n he is no t needed for efako
R
242 AGRICULTURE : ORGANIZATIO N OF LABOUR
work. Tw o different arrangements exist. Th e one found more frequently is this: every man works on the efako farm from early morning till 3 or 4 in the afternoon; after that he is free to work his own individual plot. Th e second, less frequent, arrangement requires married men to work five days in the week, and bachelors six days in the week on the efako, and allows them one or two days—including the weekly day of rest, Friday—on the buca, Th e selection of crops which are grown on the buca is of necessity very much simpler than that grown on the efako \ moreover, no collective work, not even occasional help, is possible for buca work. Thus farmers grow mostly only the staple crops, sorghum, millet, or yam, on their buca farms but sometimes, though rarely, also cotton or ground-nuts. Th e income from the buca, farm is used entirely for personal expenses, for clothes and additional food which are not provided by the family, such as savouries or kola-nuts, beer or palm- wine ; for contributions to parties of friends, certain personal gifts, and so forth.
Buca farm-work is not practised everywhere, and not even in every farming community. Bucd work and income of some kind, however, always exists. In certain areas where other forms of production (e.g. certain handicrafts) are practised besides agriculture—mostly forms of production involving less heavy work and higher money-return—these additional productive techniques supply the buca work to which the men turn ‘ after hours’. Thus we find in Kutigi, the centre of Nupe mat- industry, mat-working adopted as buca; in Mokwa, indigo-dyeing; while on Jebba Island, which is inhabited by the mixed farming and fishing community of the riverain Gbedegi, fishing with the drag-net (carried out in the evenings, i.e. after the day’s farm-work) replacing the farming buca.
Th e joint work on the efako farm is carried out by a section of the household group, i.e. by a section of the family group which lives to• gether in the same ‘ house7 or compound. Only once have I found an efako group comprising individuals who were living in different com• pounds. It was a transitional arrangement which had occurred in the course of the expansion of one household group, a section of which had split off and was preparing to take new land, farther afield, under cultivation. Th e ’emigrants’ had already built a house in the new locality but were still, for the time being, working with the rest of the original household group on the old land. Th e whole household group, with its twenty-five to thirty male members, has most probably never formed a single efako group, although informants, here as in so many cases only too ready to idealize the past, will often give exaggerated accounts of the size of the groups that formerly used to work together. Their estimates vary widely, between thirty and a hundred. However, even reliable informants put the size of the efako group in pre-British days at a figure considerably higher than that observed to-day, namely, at from ten to fifteen. Th e average efako group to-day numbers only two or three adult members. There is no doubt, then, and this is borne out by the evidence of genealogies, that the size of the efako group has been steadily decreasing for some time. These are the figures relating to
AGRICULTURE : ORGANIZATIO N OF LABOUR 243 the size of the efako working group in the case of fifty Nupe households
which I have analysed (non-adult members of the working group are
ignored in the table):
Number of men working together .
1 2
3 4
5 6 Over 6
Number o f cases . . . . 13 1 1 6 2 1 1
Th e smallest efako group, of one man, has obviously lost all the essential characteristics which distinguish it from the buca group. In these extreme cases the buca itself disappears, and we have single individual holdings, worked by one man, helped only perhaps by his boys who will stay with him until they are grown up, when they too will demand farms of their own.
Th e normal efako group consists of a number of close relations— brothers and sons who are either unmarried or only recently married, and have one or two small children—working under the headship of the senior man, a father, father’s brother, or eldest brother. Older men who are childless may also remain in the efako group. In larger efako groups and especially in groups the head of which is an old man, the family head may not himself join in the actual farm-work. He supervises it, he makes frequent tours of inspection to the farm, and he lays down the arrangement and order of the work for the day. As a family head he stands more aloof, and is free to use his leisure in his own way. Th e team-work which the efako implies is of two kinds: actual working to• gether, or co-ordinated working on different plots performing different tasks. Actual working together on the same plot is practised in the tilling of the soil and the preparation of ridges, in transplanting and harvesting. In this team-work a strong stimulus is embodied: the more experienced workers set the pace, and the others follow as well as they can, always conscious of the element of competition and display of efficiency em• bodied in the joint activity. Th e leader of the team is not, as we have seen, the head of the efako group himself but the member next in seniority, in the traditional arrangement usually the younger brother or eldest son of the family head. Sowing, ordinary weeding, and the regular tending of the plants may be done by one or two workers of the group only, while the rest may be assigned tasks on other plots or sometimes a different kind of work altogether, such as the task of helping in the house, or taking farm-produce to a distant market.
Whatever the actual division of work, the consciousness of the in• dividual share in the common labour is strongly pronounced, and with it the competitive element in the co-operation. Every co-worker in the team is aware of the number of ridges, for example, which stand to his credit in the common tilling; similarly everyone, at least among the adults, is able to describe precisely his share in the yield produced on the
1 Th e proportion of efako groups comprising more than four adults is in reality much smaller than is indicated in the table. A more superficial census of over
100 households, designed specially to ascertain the occurrence of large efakd groups, produced the same three instances of households with efakd-units of five or six adults.
244 AGRICULTURE : ORGANIZATIO N OF LABOUR
common land. Th e system of individual buca work whic h runs parallel to the family work no doubt fosters such reckoning; no doubt also that this for m of reckoning, essentially competitive as it is, must enhance the energy devoted by the individual worker to the common task. But such an attitude must also foster the tendency to split up and eventually dis• solve the working-unit. In a small hamlet near Bida I discovered an interesting form of transition between ful l co-operation, whic h yet recognizes the individual share in the common work, and complete dis• solution of the working-uni t into independent agents. In the case to whic h I am referring the efako group was divided into two working – units, each being engaged permanently on different fields although the yield produced by the two sections was stil l collected as the common efako yield. Th e efako group was formed by a father—a man of over 50
—and his two married sons—of about 25 and 30. Th e father was work• in g one group of plots and the two sons, together, another. Th e two groups of plots incidentally produced equally large crops, whic h was due to the fact that the ol d man could enlist outside labour on his farms and thus equalize the difference in working capacity between himself and his tw o sons.
MAINTENANCE OF LABOUR
Th e produce of the efako is handed over in toto to the family head. It is stored in the efako granaries, and only the family head has the righ t to dispose of it . We shall discuss later in detail the various uses to whic h he may put it . But let me describe here one aspect of the employment of the efako income, namely its use for the benefit of the efako members themselves who helped to produce it , speaking more generally, for the maintenance of the labour organization. Th e question of maintenance of labour has, first, a strictly material aspect, namely, the supply of the means of production—the agricultural tools. Th e main agricultural instruments used in Nupe farming, as we have seen, are two kinds of hoe, bush-knife, and axe. The y all have to be bought from the black• smiths or, in the case of the bush-knife, from the European store, and must be repaired periodically—expenses whic h are by no means negligible. A heavy hoe costs between is. and is. 6J., a light hoe be• tween 6d. and 1s., axe and bush-knife between is. and 1s. 6d.; repairs may amount to between 3d. and 4d. each time, that is, perhaps 1s. a year. Now , as long as the individual worker remains in the efako group, the head of the group supplies new tools and pays for repairs. A man who leaves the efako group to work on his own may take away the tools which he has been using—they have become ‘his’ tools—but from now on he must himself pay the cost of all repairs or the purchase of new tools.
Th e second, human, aspect in the problem of maintenance of labour is the maintenance of the workers themselves. The y are housed, clothed, and fed by the family head. Th e head of the efako group is responsible for all expenses that occur in connexion wit h the repair of the house and the huts in whic h his co-workers live ; he provides their clothing and food—the section of the family group whic h is regularly
AGRICULTURE : ORGANIZATIO N OF LABOUR 245 eating together and constitutes one menage coincides in fact wit h the
efako-unit. Th e efako head further pays the tax for his co-workers and is
also expected to make himself responsible for establishing them socially
when they have reached ful l adulthood, i.e. he wil l enable them to get
married, by paying their bride-price. By allotting to the co-workers
from earliest yout h an individual plot for bucawovkox allowing them the
time for other buca enterprises, he also puts them in the position of earn•
in g an independent income. The y use it , as I have said, for their per•
sonal wants over and above those catered for by the family head—what
we may call the small luxuries of life. Whe n they desire to leave the
efako group and set up for themselves the family head finally provides
the members of his group wit h the necessary ‘capital’ in the for m of the
tools whic h they may take wit h them, food for the first year whil e their
new land does not yet bear crops, and, above all, in form of the parcel of
family land whic h he allots to them as their own property. In order to
do all this the family head must be able not only to draw on regular
supplies of agricultural produce from the family land but also to avail
himself of certain commodities of exchange and that universal medium
of exchange, money. He must plan his money expenditure and accumu•
late reserves whic h can alone enable hi m to pay the large bride-price
(which is always in money) when the time comes. Th e resources of the
family working-uni t thus gain the significance of capital proper, invested
in the maintenance of labour; withou t it , withou t the prospect of ful •
fillin g all these economic obligations towards the members of the
working-unit, the efako group cannot be maintained satisfactorily.
Here we come back to the decline or disintegration of the family working-uni t under modern conditions whic h we have outlined in the beginning of this section. Now , every co-operative group must to some extent reckon wit h a periodical and temporary disintegration, that is, wit h the necessity of periodically readjusting its body of workers, re• placing those who have died or are too ol d or who for other reasons desire to leave the group and to admit new, young members in their place. Thi s is done through the twofold mechanism of inheritance and succession, and of release wit h regard to members who leave the group to work on their own. We shall examine inheritance and succession later; here i t wil l suffice to say that i n the Nupe system o f inheritance the family farm goes to the relation next in seniority, younger brother or eldest son (i n the classificatory sense). As regards the release of members fro m the group, the rule used to be that junio r members remained in the efako unti l they had more than two wives and more than two or three children. In the last five to ten years, however, many large efako groups were dissolved after the death of the family head, and the family land, instead of being worked jointl y by the remaining members of the family under the new head, was divided up among the various individuals who now work independently each on his own land. Moreover, nowadays young men who have just married frequently decline to stay in the efako- uni t and demand to be given a separate farm. I have noted many cases of this kind , in all parts of Nupe, in whic h the dissolution of the efako group occurred between six and ten years ago.
246 AGRICULTURE : ORGANIZATIO N OF LABOUR
How are we to explain this comparatively sudden change ? Viewing it from the general economic point of view all the advantages seem to be on the side of a large labour-unit: our discussion of the agricultural technique of the country has taught us that a large labour group repre• sents, in Nupe, the only means of cultivating a large variety of both use• ful and profitable crops. Moreover, a large efako-unit assures a certain measure of leisure to the family head and the old men in general, by enabling them to leave a large proportion of the actual work to the younger generation. It assures also a better distribution of responsibility, counting more adults among its members than does the small working- unit. It utilizes more fully the important incentive of competition and is, finally, of great educational value; for only in large labour-units are the young boys likely to work regularly in the company of others of their own age.
But if from point of view of group efficiency and security the large working-unit appears superior to the small (say) one- or two-men
‘group’, it is not necessarily so when looked at from the viewpoint of the individual. Take the question of leisure and distribution of responsi• bility. In one of the instances in which I found a large efako-unit having been dissolved recently (six years ago), after the death of the family head, I was given this as the reason of the change: the new family head, the eldest of five brothers, had availed himself liberally of the privilege of leisure which his position implied, and frequently went hunting while the others were working on the farm. Now, while they had accepted the same arrangement under the old family head, their father, they declined to accord this privilege to their eldest brother, and preferred to dis• solve the whole unit and work each for himself. To-day, working on their own, the individual family members have a comparatively larger measure of freedom and leisure, which they are free to use as they like. They can use their own discretion in planning the farm-work: of the five brothers two have taken to spend their spare time in making mats, a leisure-time occupation which, they assured me, is financially very advantageous to them; one has left the village; and the family head him• self can still spend part of his time in hunting. Against the considerations of group efficiency thus stand individual considerations of gain, which provide another, and in some respects very different, standard of appre• ciation by which to measure the value of the large labour-unit. Th e word
‘gain’ should not be understood in too narrow a sense. Freedom of movement, freedom to choose a different profession (we remember, for example, the openings which the political life of the capital offered to young men of adventurous spirit), even freedom to enjoy one’s leisure independent of the dictates of family requirements—these spiritual values appear under certain circumstances as tempting to the younger generation as does the prospect of material gain.
Th e two standards of appreciation will coincide where the efako group is in a position to guarantee the satisfaction of all needs and demands which the group-members might feel. Nupe family co-operation in farm-work, once apparently more satisfactorily adjusted, is now failing in this respect. This failure can be attributed to two different develop-
AGRICULTURE : ORGANIZATIO N OF LABOUR 247 ments: to an increase in the needs and demands on the part of the in •
dividual members wit h whic h the system of family co-operation failed
to cope; or to a decline in the efficiency of the group to maintain its
former standards of maintenance. Th e Nupe efako system has, in fact,
failed on both accounts. As regards the first, a large variety of new needs
and wants, temptations and values have spread, primaril y from the
town , and have begun to transform the original peasant culture; as
regards the second, the family head is to-day frequently unable to meet
out of the family income two financial demands whic h are regarded as
essential to the efako concept, the payment of the bride-price and the
payment of tax for the individual members of the efako group.
Certain of the influences whic h have thus led to the decline of the large working-uni t are no doubt due to factors of modern culture change. Yet it woul d be wrong to regard the change in the structure of the working – uni t merely as a result of the outside influence of culture contact. Th e structural change forms part of a much wider phenomenon; it is the result of an intrinsic economic development, namely, the readjustment between productive organization and needs existing in the society. If the change were due exclusively to culture contact we should expect to find the larger working-unit preserved only in backward, ‘unsophis• ticated’, districts. Thi s is not the case. Th e table on p. 243 showed that three households out of fifty had working-units composed of more than four adults. Of these three households only one belongs to 2, backward area—the household of the chief of Doko. Th e two other examples come both from one of the most progressive places in Nupe, Lemu , and from a group of farmers who, as mentioned before, have bu t recently undertaken the cultivation of a new commercial crop, red pepper, to replace another commercial crop, ground-nuts, whic h had not proved an economic success. A large workin g group is essential to the cultiva• tio n of this new crop. In these two households the young men have remained in the efako, content wit h their position, while the family head, in turn , is able to provide bride-price and tax-money. Here, then, culture contact so far from being responsible for the dissolution of the large working-uni t has on the contrary supplied a type of production whic h enables the large working-uni t to maintain itself; paradoxically, culture contact has placed the efako-umt in a position to satisfy those very demands and needs whic h culture contact has newly imposed upon the society. However, these instances from Lem u are exceptional: neither in Mokwa , the centre of ground-nut cultivation, nor in Kutigi , where cotton is grown on a comparatively large scale, has the large efako group survived. Th e exceptional cases whic h I have discovered are too few to warrant a more far-reaching, systematic explanation. Moreover, whereas in this case attractions and temptations appealing to the younger generation are partly responsible for the social change, in • dividual differences must play a considerable part. We can only tenta• tively point to a certain negative evidence: the absence of some of these ‘temptations’ in Lem u whic h have made the younger generation in Mokw a and Kutig i claim an early economic independence. Lem u is neither a centre of native trade, nor does a profitable alternative
248 AGRICULTURE : ORGANIZATIO N OF LABOUR
occupation to farm-work (e.g. a handicraft, carried out in one’s leisure- time) exist there, nor, finally, has its connexion with the political life of the town ever been as strong as in those other places.
COLLECTIVE LABOUR
Th e co-operation of the efako group is more or less regularly supple• mented by a system of mutual assistance and large-scale collective labour among the villagers, an institution which the Nupe have in common with all other Nigerian tribes. Its help is enlisted in all cases in which the efakd group by itself would be insufficient to secure efficient cultivation
—be it because of its smallness, or an unforeseen obstacle such as illness among its members, or irregular rains,1 or simply because its ambitious programme demands larger co-operation than is at its disposal. Th e Nupe have two systems of collective labour devised to operate in differ• ent situations: egbe (the Hausa gaya) and dzolo.
Th e egbe is the more general type of collective labour. It is not limited to farm-work, but is organized for evei-y kind of work which demands large-scale co-operation and specially for work done, not for one household alone, but for the community at large. Thus the building or roofing of a big house, for example the chief’s house, repairs of a bush- bridge or a stretch of road which is maintained by the village, are carried out in egbe work; repairs of the town wall, formerly, and repairs of the building of a Native Administration rest-house, to-day, are equally typical occasions for the egbe. However, in Bida and in many larger places a large proportion of work of this kind has now been taken over by professional labour. But in Bida as well as in the hinterland farm egbe is still practised regularly.
Th e help of egbe for farming is enlisted mostly for tilling and the first and second weeding, more rarely for the third weeding and the clearing of new yam-plots. A few days before the egbe is to take place the young men of the family go round to their friends and neighbours and invite them to the egbe on such and such a day. Egbe work is on the whole work for the younger people. A ready frame for their co-operation exists in the age-grade associations of which all the young men of a district are members. As they arrange throughout the year meetings and feasts in which they all join, they now organize the collective farm-work for one of their fellow member’s family. Th e family head sends out a special invitation to drummers and musicians to play at the egbe. The accom• paniment of music plays a very important part. Finally, the host pre• pares food and beer for the feast with which the egbe invariably ends.
Early in the morning drummers and flute-players walk out to the farm on which the egbe is to take place. They play and sing while they walk, beating on their drums the characteristic rhythm of egbe music, and everyone who meets them knows at once their destination. Some of the
1 I once witnessed a thir d weeding carried out by collective labour in Septem• ber, after an unexpected dr y spell. Th e farmer explained to me that if the rains had been norma l and ordinary farm-wor k not delayed, he woul d have carried ou t the thir d weeding i n ordinary famil y work .
EFAK6 WORK ON A FARM NEAR BIDA
FARMING EGBE
AGRICULTURE : ORGANIZATIO N OF LABOUR 249 working party follow the musicians, the rest meet later out on the farm.
When all have assembled the young men start work. They work in a
group of ten or twelve, starting on the same side of the field, leaving
plenty of space between each other. Everyone is working his own ridge
and, having finished one, he turns to the next with hardly any pause.
Th e drummers and flute-players walk alongside the field abreast of the
workers. Older men stand at the side, watching, criticizing, commenting.
Often you can hear them call one of the workers by name, exhorting him
to ever greater efforts, or shouting impatiently: ‘Quicker, quicker!’
When one field is finished some of the young men, the captains of
the age-grade association among them, will execute a short dance,
balancing their heavy hoes in the air and heavily stamping on the tilled
field, while the drummers are arrayed behind them, beating the rhythm of the age-grade take.1 Then workers and audience—and every egbe attracts a large audience—move on to another field, while some of the older men may stay behind to sow the newly tilled plot. On another occasion I saw the young men working in two groups of about equal strength, each under its own leader—one of the captains of the age-
grade association. Th e groups started on opposite ends of the field, working towards each other. Now and again a worker would get up, survey with one glance the work which his group had been doing, and cry out in a long high-pitched yell; others were shouting with deep hoarse voices at every stroke of their hoe. Th e speed with which they worked, the whole atmosphere of the egbe—the constant drumming, the shouts from the audience—had something deeply exciting, almost feverish. I have seen a team of fifteen young men, the oldest between 19 and 20, the youngest 14, clear and till a farm-plot of one acre in two and a half hours.2
Th e egbe work still more than efako work is animated by an intense spirit of competition. Th e worker who has quickly finished his ridges will stand up and look round, proudly showing off his splendid work; he will chaff his fellow workers about their clumsy progress, their crooked ridges; he executes a little dance, singing, and shaking his hoe to show how leisurely he can work and still beat the others. When all the work is finished the whole party, musicians, workers, and onlookers, repair to the house of the owner of the farm. Here the feast which concludes the egbe is celebrated with plenty of food, corn-beer, and dancing, and much boasting and prancing of the young men, who are the heroes of the occasion. Nor is their boasting and prancing without significance. In the village communities where the traditional age-grade system is still practised, one of the important side-issues of the egbe is that it represents a regular test in which the young men establish their qualification as farmers. From egbe to egbe, every year, their skill and their achievements are watched, compared, and judged by the older people. In the last egbe of the season, or some other outstanding egbe (e.g. on the farm of the chief),3 the results of this competition are made known and the young
1 See p. 395.
2 In another case I saw six young men clearing a densely overgrown bush- plot for yam-cultivation of about a third of an acre in one hour.
3 An egbe of this kind near Bida at which new titles were bestowed was
250 AGRICULTURE: ORGANIZATION OF LABOUR
farmer who did best is given by his age-grade association the title Sode, the second highest title that can be bestowed on young men. A man who is once made Sode keeps his rank until some other claimant snatches the title from him, or if he had no competitors, until he joins the ‘old men’ and retires from the arena of the egbe. In the urbanized age-grade associations of Bida this title and the whole usage have disappeared. In other large places which follow Bida fashion, the title Sode> though still in existence, has become a title of minor importance, a long way down the list of age-grade ranks.
The primary advantage of collective farm-work lies evidently in the time-saving which it makes possible. The second weeding, for example, which normally takes from five to ten days, can be done by egbe in one day. But this valuable help must be paid for. Th e cost of an egbe of from fifteen to twenty workers—the cost of food and drink for the guests and the pay of drummers and musicians—is between 6$. and 8s. Egbe work, then, implies a calculation of investment and return, cost and profit. To-day, farmers try to lower the cost of egbe. In an egbe near Bida (the egbe just described, in which the workers were grouped in two teams) the owner of the farm contributed music and food for thirty people, but no beer or palm-wine. Th e fare was simple—no meat or fish—but plentiful: porridge cooked in palm-oil for the men, half a bowl per head (i lb.), and corn-gruel for the women. This cheap egbe cost only 4$. yet proved still sufficiently attractive to make a large group of workers attend.
Yet there is one element of ‘profit’ in this calculation which cannot be put down in shillings and pence, and which is of utmost sociological significance. Egbe work is performed only on family farms, i.e. on the farms of family heads and old men in general. Younger people, whether they still work under the family head or have made themselves in• dependent, are not entitled to the help of the egbe. Thus the egbe pro• vides a twofold balance, in favour of the old men who cannot work as efficiently as the young, and in favour of the sociological head of the family group who, though perhaps no longer able to command the services of a large body of co-workers, can still command the assistance of collective farm-work.1 How real this balance is was illustrated in the example given before (p. 244) of an old man, the family head, cultivating fields equal in size and yield to the fields worked by his two adult sons. Wit h the splitting-up of families and the reduction of the e/ako group, with the disappearance, that is, of the clear-cut distinction between family and individual farms, the significance of the egbe, too, must weaken. Information gained from a large number of independent wit• nesses testifies to the gradual loss of importance of egbe work in Nupe productive life. Once practised regularly on every farm it has now been
performed on the new experimental farm of the Agricultural Department which was tilled in collective work.
1 Recognition of this privileged position of the sociological head of the group
in a wider sense is expressed in the traditional rule, still valid in most peasant communities in Nupe, that the first egbe of the season is performed on the chief’s land.
AGRICULTURE: ORGANIZATION OF LABOUR 251 reduced to a system of merely occasional assistance. According to my
informants the change is due primarily to the high cost of the egbe, which
to-day few farmers can afford. This is no doubt true if we understand
4 cost’ in the relative and not in the absolute sense. In the very closely balanced budget with which most Nupe farmers must reckon to-day the comparatively large single outlay for the egbe may prove an insuperable difficulty. But this is not the whole picture. A cost is high only measured on the profit it can realize. And here we have seen that one of the most essential kinds of profit—moral profit—which the egbe can realize, namely, its assurance of the economic balance between the generations, has itself lost much of its significance in the life of the community.
Th e second type of collective work, the dzolo, is less important than the egbe. It is practised only for harvesting of those crops which demand a large working team (ground-nuts, cotton, and beans) and consists in the arrangement among a number of farmers to work in turn on each other’s farms. The host provides the meal after the work. But there is no drumming or dancing, nor is any competition involved in the work of the dzolo. Neither is the mechanism of the age-grade associations utilized for this collective work, which is carried out on a much smaller scale than the egbe, and only between friends, i.e. families whose heads are friendly with one another. In the dzolo the tie of friendship evidently supplies that incentive to collective work which, in the egbe, is entrusted to the appeal of music and dance and the stimulus of competition.
WAGE-LABOUR
Assistance in farm-work by means of paid labour plays on the whole (we must add as yet) an insignificant part in Nupe production at large. It is called kantsu in Nupe but is often referred to by the Hausa name lada, i.e. wages. It is employed by the owners of farms who are not them• selves farmers or who for other reasons cannot themselves look after their land. Paid farm-labour is a new thing in Nupe. It has no place in the traditional system of land-tenure and efako work. Characteristically, it is almost completely absent in the rural districts; it is employed mostly in the large commercial centres in which a considerable proportion of the population belongs to the well-to-do ‘professional classes’, such as Kacha, Jebba, and, above all, Bida.1 Let me give you a h’st of the cases I recorded in which wage farm-labour was employed. Th e employers were a number of traders in Kacha who had bought some land a few years ago but did not work it themselves; a number of traders in Bida; two Mallams in Bida; a Mallam in Jebba whose health did not allow him to work on the land (he was employing wage-labour for the first time); Native Administration officials in Bida—a few policemen, and teachers at the Bida Middle School; and lastly, the village chief of Kacha (who is at the same time District Head) whose time is taken up by official
1 One of my Bida informants, himself a well-to-do member of the ‘intelli• gentsia’, estimated that six out of ten farmers in Bida employ wage-labour. Though I should regard this estimate as much too high it is significant for the outlook of the urban bourgeoisie.
25a AGRICULTURE : ORGANIZATIO N OF LABOUR
work, and whose relations also assist him in this work as messengers and the like.
Th e labourers are all young men, mostly the sons of farmers, but occasionally also sons of impoverished bourgeois families (one of the labourers on my list was the son of a Mallam, another the son of a Bida saraki). Some labourers come from far away to the place where they expect to find employment: thus I found a young man from Lapai working in Jebba, and one from Bida working in Ilorin. Their working day is from early morning till about 4 o’clock in the afternoon, that is, the normal working day of the efako worker. They get 2d. or 3d. a day and food; they work mostly for from four to ten days, some for a longer period, till they have collected a substantial sum of money. In no case has wage- labour in Nupe become a permanent, exclusive occupation. These young men seek only temporary employment and return after or in between their engagements to the family farm, where they carry on with the ordinary work of the season. Even a temporary absence of this kind may, of course, render the management of the family farm difficult and even precarious. For instance, the labourer from Lapai whom I have just mentioned, was working in Jebba as late as June, and could not return home before July when all the early planting and weeding would be completed. However, the need of money seems to outweigh all other objections. The main motive in undertaking wage-labour is invariably to obtain money for tax—either for oneself or as contribution to the parental household; next in importance is the motive to obtain money for bride-price.
It must be said in conclusion that though wage-labour represents a comparatively recent growth in Nupe society, the situation which created the need of wage-labour is not. Nor is the system of employed, i.e. non-family, labour the product of a recent development. Part of the social stratification of Nupe was, as we have seen, the creation of an upper class which owned land but did not work it. Th e labour which this class had to employ was of three kinds: slave labour, the labour of pawns, and, to a lesser extent, of clients. Th e first and the last kind of employed labour, both involving a social and political significance far beyond the aspect of labour, we have discussed already in a previous section. Th e second kind—labour of pawns—we will discuss later when analysing the concept of wealth.
DIVISION OF LABOUR BETWEEN THE SEXES (i )
Th e role which the women play in the agricultural production of the country lends itself to a simple formula. Wit h very few exceptions the women do not perform any primary productive activity: they do no farm-work; they do not help in cultivation except occasionally when assisting the men in the harvesting of cotton or beans or the digging out of cassava—the easiest crop of all. Their sphere of work is, first, the
‘refining’ of agricultural produce to make it marketable, and secondly, the marketing itself. If the general formula is simple, the details of the division between ‘primary’ and ‘secondary’ productive activities, one
AGRICULTURE : ORGANIZATIO N OF LABOUR 253 assigned to the men and the other to the women, are involved and
intricate. We wil l discuss them here only i n so far as they refer to the
household arrangements and leave the discussion of marketing proper
for later.
Th e most elementary ‘refining’ process in this respect is the cooking for the household. In our general description we have reached the point when the various crops from the family farm have been harvested and stored. Fro m these stores the family head wil l hand out regularly a certain amount to the wives for cooking. Th e arrangement varies wit h different crops: grain and beans, for example, are dispensed by the family head as a rule once a week or once in ten days; yams are brought in from the farm (where they are stored) about twice or three times a week, in the case of a family of three or four adults four to six yams each time, i.e. a supply sufficient for from two to three days; rice, the most expensive of all foodstuffs, is handed out only for the day when it is needed for cooking. Th e women plan the meals, but the men being in charge of the stores always know beforehand what they are going to have; occasionally they even order the ‘menu’ for the day. Th e amount of food which the men allow for cooking is adjusted to the exact require• ments : for example, if the men are having their midday meal on the farm, where they cook their own yam or cassava, they wil l hand out a proportionately smaller supply of yams, two or three, for the house, whic h must last for three days. In the town and among non-farmers the family head gives his wife or wives money for buying the necessary food about once in a week or ten days. Th e family head, however, does not provide for all meals, nor for every foodstuff needed in the household. He contributes food only for the main meals, midday and evening meal, and then only the bulk of the meal: grain, yams, rice, fish, and meat. Of other ingredients such as pepper, okra, or certain greens, he provides only those whic h he has grown himself. Other even most essential in • gredients the wife must invariably procure herself—even though they may have been obtained from trees growing on the land of the husband— namely, kula (made of locust-bean), kuka (made of the leaves of the baobab tree), palm-oil, shea-butter, and also salt. These things the wife has to buy in the market wit h her own money. Unless food has been left over from the night before, the mornin g meal consists mostly of corn-water or gruel, sometimes eaten wit h ground-nut or bean pastries, and is made of foodstuffs invariably supplied by the women. The y wil l share the morning meal wit h their husbands, but wil l only sell it to the other members of the household, even including the children, for whose meal their father has to pay. Pastries and savouries are always made of foodstuffs whic h the women have to buy wit h their own money. The y are then sold, and wit h considerable profit, in the markets or streets,
except the titbits which a mother wil l always pu t aside for her children. If the husband feels he woul d like one of these delicacies he has to buy it , either from his wife or some other woman pastry-cook.
Th e Nupe women have also other, more important, sources of income whic h more than balance the financial obligations whic h they have to undertake in the household arrangement. Th e services of the women
254 AGRICULTURE : ORGANIZATIO N OF LABOUR
are indispensable, and are accordingly rewarded in the disposal of crops intended for marketing. There are various methods in the disposal of crops. One is the following: the family head hands the produce he in• tends to sell over to his wife, either the whole produce at once, immedi• ately after the harvest, or small quantities at different times, according to whether he is in need of ready money or can wait for a better price later in the year. His wife will then sell the produce in the market and return the money she received for it to the husband. He gives her a little of this money, half as present, half as commission. His gift varies with the price which she has managed to obtain, and will express his appreciation of her more or less clever marketing. Th e woman may, of course, have to try a good many different markets if she wants to obtain the best price; and the men may not always be in a position to check the price which their wives quote to them after the sale. My information is that the women do not hesitate to take advantage of this lack of control; but the men, it seems, are as well aware of this and regard it as a legitimate way of making money. Another method, possible only if the women have sufficient money in hand, is this: the family head sells the whole produce directly to his wife after the harvest. She pays for it, stores the produce in her own granaries, and sells it later in the year when prices have gone up, possibly in small quantities at a time to obtain the best price. This method is as a rule applied in the disposal of palm-kernels and the fruits of the locust-bean tree. Another method in the disposal of crops does not enlist the help of the women of the house—although it may have to enlist the help of other women in the person of the traders. It is limited to a few kinds of produce only: rice, kola-nuts, and also palm- kernels in case one’s own wife is not an expert in oil-making; these goods are sold by the family head directly to professional traders.
So far the role of the Nupe woman—more precisely the wife of the Nupe farmer—is that of a ‘broker’ and ‘middleman’. In the disposal of the produce of certain fruit-bearing trees, however, the women are assigned direct and absolute property rights. Th e fruit of the locust- bean so far as it has to be plucked from the tree is disposed of in the way described above; but the fallen fruits of that tree belong entirely to the women of the house. They can do with them what they like, and the money which they obtain from the sale of the raw as well as the ‘refined’ produce belongs to them. Also, the whole produce of the shea-nut tree belongs to the women. Th e exploitation of the shea-nut tree, the fruit of which is of comparatively high commercial value, is regarded as a definite monopoly of the Nupe women, and as representing a direct compensation for their contribution to the expenses of the household.
Th e picking of the fruit of the shea-nut tree by small groups of women also furnishes an example of a co-operative organization among the women beyond the single household or family, somewhat similar to the system of collective labour organized by the men. There exists another form of collective female labour which is regarded by the people them• selves as the exact counterpart to the men’s egbe. As the men, in egbe work, build or repair houses and roofs, the women’s collective work is
AGRICULTURE : ORGANIZATIO N OF LABOUR 255 employed in the beating of the floors of the huts or the platforms in front
and between the huts. Thi s wor k is done in the evenings, when the
women have finished their own housework. It has all the elements of the
egbe: the generous meal awaiting the workers, the singing and dancing—
the beating itself is done rhythmically, to the accompaniment of songs
(no musical instruments are employed in this or in any of the women’s
dances), and in its movements the motions of wor k and dance are
attractively combined. Th e group of women fills the hut to overflowing.
The y are arranged in concentric circles, pressing closely against each
other. Bent low over the floor they beat the ground three times wit h the
heavy wooden flail whic h they hold in their right hand, then pul l them•
selves up, lift both hands i n the air, and take a measured dance step
towards the centre and back again, and then bend down once more to
resume the beating of the floor. Or they fill this rhythmical pause i n the
work wit h the verse of some interminable song, and while singing they
move a few steps along in the circle; when the moving rin g comes to a
halt they bend down to their work, to interrupt it once more in the
rhythmical motion of this ‘ring-dance’. No r is the incentive of competi•
tio n absent in this egbe of the women. House-groups compete wit h
house-groups; girls of different age are assigned different tasks: small
girls of 6-7 are workin g slowly and silently on the outskirts of the group ;
girls of 12 or so are singing wit h the women and work towards the inside,
tryin g to emulate the speed and ease of the grown-ups; and the
adult women, leading in dance and work, spur their teams to almost
breathless speed. As they work and dance, closely packed into the
small dark hut, stripped to the waist, their bodies glistening wit h
sweat and spattered wit h mud , beating, singing, stamping, and
shouting, they seem to be actors in some exciting savage ritual rather
than in ordinary work.
To return to the more prosaic aspects of the division of labour between the sexes, and to the question of the ‘ compensation’ for her contribu• tions to the household whic h the exploitation of shea-nut trees allows the Nupe woman. Th e clearly defined attitude of the people wit h regard to this question of ‘compensation’ and the economic rights of the women in general is illustrated in the following example. We remember that the roast-ovens in whic h the shea-nuts are dried before being marketed or manufactured into shea-butter are buil t by the men. Now , in Leaba I found that the women had to dry shea-nuts in the sun, a much slower and more laborious procedure, as all their ovens were dilapidated and sadly in need of repair. I inquire d wh y the men di d not buil d new ones. Th e men explained they had no tim e and money. I suggested, half jokingly , that the men might be given a share in the profits made on the shea-nuts and in return repair the roast-ovens. Whereupon the women, realising that their own special interests were involved, joined in the discussion, whic h is interesting enough to be given here wor d for word .
Women: ‘Wh y should we give the men part of our money ? Do we not cook p6rridge for them?’
I: ‘Bu t he gives you the grain for making the porridge.’
256 AGRICULTURE : ORGANIZATIO N O F LABOU R
Women: ‘Yes, bu t only the grain. We have to contribute kula, kuka,
salt, pepper and palm-oil. ‘
I : ‘ Surely yo u have locust-bean trees and re d pepper i n Leaba?’
Women: ‘No , there are no locust-bean trees here, nor pepper.’
One woman: ‘Thi s is not true, we have pepper.’
Another woman (rather defiantly): ‘Well , there still remains salt, kula,
and palm-oi l for whic h we, the women , have to pay.’1
1 Another example of the clearly denned economic rights of the women is this: in the Nupe commumties in the forest groves on the Niger, where there are no shea-nut trees, the banana is the property of the women and supplies them with fruits which they can dispose of in their own right.
CHAPTE R XI V
INDUSTRIE S
ORGANIZATION OF CRAFT-GUILDS
N the Western worl d Nupe native industries woul d fall under the heading of ‘home industries’. The y are carried out in the home, or in workshops connected wit h the home, and their labour organization is to a large extent composed of members of one household or a number of related households and families. Al l the industries are carried out by men. But we find a certain differentiation wit h regard to the organiza• tio n of labour, between industries carried out in individua l work and industries based on co-operative group production or, in other words, between individual crafts and guild-organized crafts. For the co-opera• tive organization is of a specific type. It represents a closed group, the membership of whic h is to a large extent hereditary in a number of families; outsiders can joi n it only through a formal adoption into the profession, after a period of apprenticeship. Th e group, moreover, is defined not only wit h regard to the productive pursuit but also socially and politically, as a specific social group, almost an artisan class, whic h enjoys official recognition and certain political privileges. Its internal organization, finally, is marked by a certain hierarchy of grades and ranks. These crafts thus present a picture closely resembling the organi•
zation of craft-guilds in Imperial Rome and the Middl e Ages.
Of the native industries in Nupe the following are organized in craft- guilds: blacksmith work and the minin g of iro n ore; the craft of brass- and silversmiths; glass-making; weaving; bead-work; building ; wood-work and carpentry; and, lastly, the profession of butchers. Tailorin g and embroidery, leatherwork, indigo dyeing, straw-hat- making, mat-making, and basketwork are practised as individua l crafts. Th e majority of Nupe crafts, especially guild-organized crafts, are con• centrated in the town. But even where these crafts occur both in town and village (as in blacksmith work and weaving), they appear in their most typical form only in Bida, and are indeed symptomatic of the civilization whic h evolved in the capital of the kingdom. I cannot attempt here to describe i n full all these native industries, neither from the technological point of view nor wit h regard to the details of their organization. I shall deal only wit h the most typical and, sociologically, most significant crafts.
I t wil l be best, however, before entering into the description of the various crafts, to anticipate certain general features common to all of them. I have called their productive organization ‘co-operative group production’. Bu t co-operation in these crafts exists in two different forms and on two different levels: we have, first, co-operation of the group at large, i.e. the guil d or guil d section; and, second, co-operation in the narrow sense, i.e. the actual workin g together, of the individual labour-unit. We shall leave the description of the former to the sections dealing wit h the various specific industries and concentrate here on the
S
258 INDUSTRIES
latter, narrower, group co-operation. We shall see that it varies widely in scope and closeness, owing partly to the different nature of the pro• ductive technique and the varying demands upon co-operation which it entails, and partly also to extraneous factors, not involved in productive technique. Th e labour-unit in craft-work is called efako, i.e. by the same name as the labour-unit in farm-work, to which, in fact, it corresponds in all essentials. Again we find a section of the family group working together under a senior relative who, as the common head, directs group co-operation, supplies the means of production (tools and raw material), and disposes of the proceeds from the group work. He also undertakes the other efako obligations of paying tax for his co-workers, providing them with food, clothing, and bride-price. At the death of the efako head the group member next in seniority takes his place. If old age disables the efako head for active work he can retire from the labour- unit without giving up his position as the head of the group, i.e. without abandoning his control over the group income. Old men who have not reached the position of an efako head when they are forced to retire will be supported by the labour-unit. Th e old men of the group are, moreover, often allotted a little farm-work close to the house, which allows them to make their contribution, even in old age, to the group income. We can see already that the ability of the group to ‘pension off’ their old men and late leaders depends essentially on the size of the labour-unit; a very small labour-unit is clearly incapable of making the necessary provisions, and we shall find that in the case of these labour- units the support of the old men is undertaken, rather irregularly in fact, by their family as a family obligation, and no longer forms part of the regular services of the labour-unit towards its members or ex-members.
Wit h regard to the junior efako members we find, again, the practice of releasing from the efako-xiriiX. adult workers who desire to make them• selves independent and to set up a separate labour-unit. Th e question always entailed in the release from the labour-unit, namely, of the acquisition of new means of production, is complicated in craft-work by the fact that here the means of production are to some extent immovable (e.g. the blacksmith’s furnace and anvil, the glass-maker’s furnace, the weaver’s loom, &c) . It appears to be due to this difficulty that the release from the labour-unit in craft-work does not invariably involve withdrawal also from the workshop in which these means of production are fixed, but that, on the contrary, there is the tendency to stay on and share workshops, furnaces, and so forth, with the ex-fellow-workers. Only if the workshop is overcrowded will the worker or workers who are leaving the efako-unit look round for another place, for example, avacant furnace in another workshop, or the entrance hut of a compound, or (more rarely) build themselves a new workshop, with the assistance of family and fellow craftsmen. Th e withdrawing workers will be allowed to take with them the (movable) tools which they had been using, and will also be supplied as a rule, by the old efako head, with raw material for the start. This setting-up of independent workers is in craft-work as in farm-work the last act of the functioning efako.
Th e co-operative organization of efako work also enables the group to
INDUSTRIES 259 train its adolescents and to bring them gradually into the system of co•
operation. Young boys are employed first in easy auxiliary work, such as
working the bellows in blacksmith and glass-work, or the spooling of
cotton thread in weaving; sitting all day long in the workshop they learn
a good deal from watching; when they are old enough they are given
more serious work, which they carry out under the guidance and tuition
of the efako head till, eventually, they become fully fledged members of
the labour-unit. In most cases the efako work is supplemented by special
individual, buca, work. Th e relation between the two varies widely,
with the specific nature of the craft and its productive technique.
BLACKSMITH WORK
Th e blacksmiths are called tswata in Nupe, or sometimes tswata
gbagba, ‘ heavy smiths’, to distinguish them from the tswata muku, the
brass- and silversmiths. Nearly every village in Nupe possesses at least
one or two blacksmith workshops. In Bida we find three groups of
blacksmiths, whole blacksmith ‘ wards’ in fact, each with its cluster of
workshops, and each situated in a different ekpa of the town. According
to tradition the three blacksmith groups are of different origin. One,
in the old centre of Bida, ba nin, claims descent from the original group
of blacksmiths which was established in Bida when it was still a Beni
village. At that time the head of this group, the Maji Tswata, the Master
of the Blacksmiths, is said to have held authority over all the blacksmiths
of Beni country. The blacksmiths living in the Etsu Masaba quarter, the
smallest of the three groups, came to Bida comparatively late, from
Trans-Kaduna. Th e largest blacksmith group in Bida, in the Etsu
Umaru quarter, traces its descent to the blacksmiths of Raba, the former
capital of the kingdom. They came to Bida together with the king and
court when Bida became the new capital of Fulani Nupe. As we shall
see later, this last group has established itself as the supreme guild
authority over the blacksmiths of Bida and all Nupe. Th e craftsmen of
this group call themselves dokodzazi, after their head, who bears the title Dokodza.1
Technology
Th e blacksmith wards of Bida are of very different size; the smallest contains two workshops, the largest seven. In some cases the entrance- hall of the compound in which the families of the craftsmen are living serves also as workshop or forge. But usually the forge is in a separate hut which stands by itself a small distance from the compound; with its three or four wide doors, and the many small window-holes let into the wall to serve as passage for the smoke from the forge, the workshops can be easily identified. Inside are two furnaces, each screened in by an angular mud wall, about 3 feet high. On the inner side of the screen, facing the open fire, sits the blacksmith himself; on the other side of the
1 It is interesting to note that the modern Emirate of Patigi, where the descendants of the old Nupe dynasty are reigning to-day, also possesses its group of ‘royal blacksmiths’, under another Dokodza. The y came to Patigi together with the family of the present rulers, from Gbara, the ancient residence of this branch of the Nupe dynasty.
260 INDUSTRIES
screen another man is sitting, who works the bellows. Th e bellows are of the split-bag type common in Norther n Nigeria ; they are made of two bags of goatskin whic h end in a clay tube and are worked wit h sticks fixed on the top of each bag. Th e clay tube is buried in the ground and, pass• ing underneath the screen, ends in the shallow furnace on the other side. In this furnace charcoal is burned, whic h the blacksmiths make them• selves. The y also forge their own working tools (apart from a few European tools, hammers and files, whic h have strayed into some of the blacksmiths’ huts): a club-shaped hammer, a long pair of tongs, and a few chisels and pointed rods. Th e anvil, too, is of their own production ; it is a solid piece of iron , shaped like a toadstool, the foot of whic h is buried in the ground. Of other working implements there is a bowl of water in whic h the red-glowing iro n is tempered, and a heap of fine ashes in whic h the finished product is left to cool.
Th e technical process is simple. Th e bul k of crude iro n is held over the fire unti l it becomes red hot and pliable. It is then placed on the anvil and gradually hammered into shape. Th e change from furnace to anvil is repeated several times, unti l the object has been given its final shape. Sometimes it is finished off wit h the help of files and chisels. Th e main products of the Nupe blacksmiths to-day are blades for hoes, bush- knives and crude axes, big nails and hooks for the big plank doors of the Nupe houses, and small brackets used in canoe-building. Some• times the blacksmiths also make the wooden handles for hoes and axes; sometimes the customers prefer to buy only the blade and to make their own handles. One man, I found, can produce two hoe-blades a day, working from 9 a.m., wit h a pause for the midday meal, til l 4 p.m. , when the Nupe working-day finishes.
Iron-wor k thus involves four subsidiary forms of production: the manufacture of the bellows, of charcoal, of the tools, and finally of the raw material used in blacksmith work, i.e. the minin g of iro n ore. Wit h the exception of the last all are undertaken by the blacksmiths themselves. Th e blacksmith buys the goatskins on the market and sews them together into the skin-bags used as bellows; he also produces the mouthpiece of clay for the bellows. Ordinary tools, such as hammers and tongs, are made by every worker individually ; should it become necessary, however, to replace the anvil, the whole group of blacksmiths in the quarter (forty in all) woul d have to work together.
In charcoal-burning two or three men who share the same workshop work together, and produce a store whic h is to last the whole workshop for some time. Charcoal is burned in a clearing in the thick bush at some distance from farms and habitations. Woo d of two different kinds may be used. Pieces of timber of about arm’s length are piled 3 feet high . A thick layer of grass is placed on the top, and the whole heap is covered wit h sand, leaving only a small hole open. A dr y piece of wood is li t and pushed into the opening to kindle the fire inside, after whic h the hole, too, is covered up , and the heap is left to bur n for five or six days.
Th e production of the raw material of blacksmith work deserves special description. Formerly all iro n used in blacksmith work was
INDUSTRIES 261 mined and smelted in the country. Th e production of native iron ore has been greatly reduced in recent years. To-day a large proportion of the crude iro n is bought from the European stores—the Nupe call it kansana, ‘dr y iron’ , to distinguish it from the native mined nyenkpa ziko
‘black iron’ . Th e competition of cheap European iro n has made the up • keep of the large native foundries unprofitable; many have been abandoned and allowed to fall in disuse, and even where they are still in use only two or three out of perhaps six furnaces are stil l workable. But in the busy season, shortly before the rains, when everybody needs hoes and axes, the supply of European iron frequently runs out, and the resulting shortage has to be relieved by calling on the supplies of native- mined iro n ore. Nupe iron-smelting can, in fact, no longer meet heavier or more regular demands, although many Nupe farmers stil l prefer a hoe made of native-mined iron—i t is stronger, and lasts better, they say; European iro n is fit only for repairs, for patching hoe-blades, &c. Th e tools of the blacksmiths and, above all, the anvil, are invariably made of native-mined iron .
Th e iron-ore miners, called edudacizi (meaning ‘workers in the deep’), form a professional group distinct from the blacksmiths, although some of them are blacksmiths as well as iron-ore miners. Th e minin g is done on the slopes of the low flat-topped ferriferous hills of the country. I have heard of five different minin g places: near Mokwa , near Shari, near Patigi, near Kutigi , and near Pichi (about 8 miles from Bida). I have seen only two ; one in work, though on a reduced scale (near Pichi), and one abandoned (near Shari). Th e iron-ore foundry near Pichi is the largest in existence. Lik e the blacksmith groups in th 2 districts round Bida the iron-ore miners of Pichi submit to the authority of the guild – head of the blacksmiths in Bida. His authority goes so far that, five years ago, he could order two additional furnaces to be buil t in order to increase the Bida supply of iron ore; these new furnaces, however, are no longer in use.
Th e ore is scratched off the surface of the hil l or dug out of shallow pits. Close to the hill-side are the furnaces in whic h the iro n ore is smelted. Three furnaces are placed side by side under a large grass roof which protects workers and ‘machinery’ from sun and rain. It wil l be best to describe the very elaborate arrangement by means of two dia• grams, the first of whic h represents a bird’s-eye view of the furnaces, and the second a vertical section through a furnace. We have first, a large pit , 6 feet square and 6 feet deep. On one side a pot for water is let in the ground ; on the other the furnace proper is erected. It consists, above ground, of a kin d of chimney (a) made of thick clay, 5 feet high, buil t over the mouth of an airshaft (b) whic h serves both as ventilator and as opening for the fuel and the iro n ore. On the other side of the chimney are the two bellows (c), clay tubes formed somewhat like horns, the wide mout h of whic h is covered wit h goatskin. Th e narrow bottom ends of the bellows tubes are joined, under ground, wit h a long straight clay tube (d), a littl e over 5 feet long, whic h is fixed in position by means of wooden wedges. It opens into a cave-like space whic h forms an ex• tension of the square pit . On the bottom of this space the iro n ore is
FIG . I
INDUSTRIES 263 placed, wit h charcoal piled over it . Th e ‘cave’ (e) is connected, by means
of a narrow airshaft, wit h the mout h of the chimney (b). Cave and square
pi t are separated by a semi-circular clay plate (f) , 1½ feet high, whic h is
cemented int o the opening so that it leaves only a small triangular hole
open at the bottom. Throug h this hole the slag pours into the pit . One
worker sits on a crude wooden bench whic h is supported by a timber
structure (g), behind the bellows; he is relieved every hour. Other
workers, tw o or three in number, act as stokers, feed the furnace wit h fuel
and iro n ore, and remove the slag by means of a long wooden stick wit h
iro n points, so that the small hole in the clay plate should be kept open.
Th e iro n ore is broken into small pieces by means of stones or hammers
before it is placed in the furnace. Th e smelted ore remains behind the
clay plate unti l the whole process is finished; then the clay plate is broken
and the iro n removed. Th e clay plate has to be made new and fixed in
position every time the furnace is put to use. Wit h the furnace all pre•
pared, a lum p of iro n of the size of a large yam frui t can be produced in
five to six hours’ work.
We have already seen that the organization of blacksmith work and iron-ore minin g (as of all guild-organized craft-work) involves two differ• ent types of co-operation, the actual working together of the labour or efakd-unit, and the wider co-operation in the framework of the occupa• tional unit, the guil d itself. We wil l examine first the composition of the labour-unit.
Th e men who m we find working in the same workshop are all related and belong to the same family group or section of the family group. Th e common workshop does not, however, invariably establish. the labour-unit; the workshop group is the wider unit , for the men who occupy the same workshop are not necessarily also co-workers in the narrow sense. Th e labour-unit itself corresponds to the small individua l family or section of the family—a father and his sons, or a number of brothers, who divide all work among themselves and in certain respects share the proceeds. Al l more important work is regarded as efakd work, i.e. the family head buys the raw material and receives the proceeds from the work carried out by his group. Less important work, mainly small repairs of hoes or axes, is regarded as buca, and the money received for it belongs to the individual worker. Onl y adults, however, practise buca work ; small boys who are yet too young to carry out serious blacksmith work are assigned the work of blowing the bellows. In workshops where there are no young boys, the adult workers may take the bellows in turn ; occasionally a woman, the wife of one of the blacksmiths, may help out.
Th e relation between these various units, workshop group, family group, and efako group, is involved and variable. I t wil l be best to illustrate it wit h actual examples. Th e first two are taken from the largest blacksmith group in Bida, the Dokodza people. In one of their workshops we find two adults and two boys workin g together—the boys blowing the bellows. The y all belong to the same family and live in the same house. Th e two adults are brothers, and the two boys the sons of the elder brother. Th e four workers for m a single labour-unit, under
264 INDUSTRIES
the elder of the two brothers, who is efako and family head in one. His younger brother practises buca work in addition to the family work. In another smithy (the smithy of the guild-head himself) we find five men sharing the workshop. The y belong to the same family and live in the same house; their common family head is the Dokodza, the guild-head. Th e workshop group is divided into two labour-units. One, consisting of three men, is headed by the Dokodza’s younger brother, who works together wit h his two sons; the other, consisting of two men, the sons of the Dokodza, is headed—nominally—by the guild-head himself; being an ol d man he leaves the actual work to his sons. Each of the workers also has his own buca work.
In the other blacksmith groups, especially in the villages, we find a more loosely kni t co-operative organization wit h much smaller labour- units. In a workshop of the Beni blacksmiths in Bida I found four men workin g together, two adult brothers and the two young sons of one of the brothers. Th e kinship configuration is thus the same as in our first example; but here the workshop group is divided into two labour-units, one formed by the father and his sons, and the other, a ‘one-man-unit’, by the man’s brother. Amon g the Kutig i blacksmiths one workshop numbered four men. The y were tw o brothers and tw o more distant relations (a father’s brother’s son and a father’s brother’s grandson). Th e workshop was divided into, four labour-units, each man working on his own. I t wil l be seen that i n these last two examples the division between efako and buca work has disappeared.
In addition to the craft the blacksmiths as a rule also practise a littl e agriculture on the farm-plots close to their houses, where they grow a few crops ‘for the mouth’ . These farm-plots are tended mostly by the ol d men of the family, who can no longer carry on the heavy work in the smithy. Th e crops belong to the family head, and are regarded as efako property.
Blacksmith work, closely associated as it is wit h the requirements of agriculture for whic h it produces the main tools, is subject to the same market fluctuations. It has its busy season in the early rains, and is slack durin g the dry season. Fro m December to February, the quietest season of the year, one smithy in Bida was producing four hoes a month ; other workshops were doing only minor repairs. For two months, durin g the planting season (Ma y and June), the same smithy turned out four hoes a day. At that time of the year the blacksmiths lay in a large store of hoe- and axe-blades, whic h is quickly disposed of in the village markets, while during the dry season they must often send their wives to the surrounding villages and hamlets to find customers for their products. Le t us examine the question, of profits and cost of production. A blacksmith buys, for example, a lum p of iro n ore for 2s.; this quantity is sufficient for the making of six heavy or ten light hoes, whic h he sells at the price of 6d. and is. a-piece respectively, exclusive of the handle (which woul d be another 1½d.). On one hoe he makes a profit of between zd. and 3’2?. In the busy season, when the blacksmiths have their hands ful l wit h work, the prices go up. A hoe which woul d be sold for 6d. in the dr y season fetches 8d. in the rains, and one that woul d be sold for u.
BLACKSMITH SILVERSMITH
INDUSTRIE S 265 i n the dr y season wil l sell for is. \d. i n the rains. I n these two busiest
months of the year, then, the profits of the blacksmiths may be as much
as yd. on every hoe, or over is. a day (one man can produce two hoes a
day). Th e blacksmiths of the Dokodza group in Bida seem very satisfied
wit h their economic situation. But the blacksmiths in the country, and
also the blacksmiths of the other, smaller, guil d sections in Bida, are
bitterl y complaining of the decline of native blacksmith trade. Some of
these country blacksmiths have turned farmers, or have at least taken up
farming on a larger scale than formerly, in order to balance the losses of
their industry. Thi s retrograde movement is due to a variety of causes.
One whic h has affected all blacksmith groups in the country equally, is
the competition of European trade. Thu s the Nupe blacksmiths have
lost almost their entire trade in bush-knives to the European trading-
stores; nor does the once very extensive trade in Bida hoes whic h the
Hausa caravans used to take nort h exist any more. Th e iro n smelters
have suffered most severely under this competition. As we have seen,
their work has been almost completely disorganized under the pressure
of European competition. Other causes affect the blacksmiths in the
country more than in the town, and smaller groups more than larger.
Th e scarcity of money among the peasants has forced certain farmers in
Kutigi , for example, who were unable to pay in money for the hoes
whic h they needed, to offer a half-day’s work on the farms of the black• smiths in lieu of payment.1 Closer intercourse and the increased facili• ties of transport between tow n and country have also deflected some of the former custom of the village blacksmiths to Bida, where, owing to the presence of the European trading-stores, there is less danger of a shortage of raw iro n occurring in the busiest time of the year. A final,
more subtle, cause of the decline in blacksmith trade, whic h attacks smaller groups more severely than larger, lies in the failure of the former to meet successfully the demands of what we have called Main • tenance of labour’. Thi s influence wil l become clear when we examine the second aspect of co-operation in native industries, namely, the co• operative organization of the professional group at large.
Guild Organization
Th e term ‘guild ‘ is applied correctly only to the Bida blacksmiths, whose organization shows the characteristic features of the guil d system
—the closed profession, centralized control, the hierarchy of ranks and grades. However, the individual blacksmith groups in Nupe, in Bida as well as in the villages, are all modelled on very similar lines—although the guil d sections in Bida are organized on a larger scale, and also more closely and efficiently. Moreover, the control of the Bida guil d system extends in certain respects over the whole country and the various inde• pendent local blacksmith groups. Thi s widest professional uni t is the result of political rather than economic factors; it is expressed mainly in
1 Prices in the country are higher than in Bida. In Kutigi a heavy hoe is sold for 1s. to 1s. 6d. Th e difference in price seems chiefly due to the fact that the Kutigi blacksmiths must obtain their iron through middlemen who bring it from Bida, as there is no European store in the village.
266 INDUSTRIES
the allegiance to a common head, the master of the Dokodza section in Bida. As this group of Bida blacksmiths is associated, historically, with Nupe kingship its head also holds his rank and position by appointment of the king of Nupe. Th e Dokodza claims that the blacksmiths in the whole of Nupe ‘must obey him’ in all matters concerning the profession. As an external expression of this allegiance he receives annually, at the occasion of the Great Sallah, a gift of crude iron ore from the various blacksmith groups of the country. Although this gift has now been abandoned the village blacksmiths still acknowledge (at least in theory) the supreme position of the guild-head in the capital, and never fail to pay their respects to him when they are in Bida. In return they count on his help and influence if they are in need of official intervention in the capital or at the court. The closeness of the co-operation under the guild- head, and the efficiency of his control over his fellow craftsmen, increases with the nearness to Bida, till, in the town, it takes the form of guild organization in the strict sense of the word.
Th e most effective unit, however, is not the guild at large, but the single guild section. The people of each group are of one dengi, one kin• ship group, related both by descent and frequent intermarriage. Every male born into the group is expected to become a blacksmith—’how can he become anything else’, said my informants; ‘we know no other job’. But, on the other hand, they would maintain that no man from outside could become a blacksmith. When I asked the Dokodza people why this should be so, they answered with a laugh: ‘How can anyone who was not born a blacksmith become one?’ Nevertheless, in another part of the town I discovered a man who, born of a family of traders, had adopted the profession of a blacksmith. He seemed to have been a rather more unruly youth than the rest of the family; for he had firmly declined to become a trader, like them; he had first studied the Koran and thought of becoming a Mallam, but later changed his mind, and decided to become a blacksmith: he ‘liked it better’ than either trading or studying. Certain maternal relations of his were Beni blacksmiths in a village near Bida. There he went, stayed with them for some years till he had learned the trade, and then returned to Bida, where he is now keeping a little workshop of his own in the entrance hut of the house of his paternal relations. Among the Beni blacksmiths in Bida I was told that formerly apprentices were taken regularly. Th e pupils came mostly from villages near Bida, and returned to their home as fully fledged craftsmen. Th e training took from three to five years, during which time the apprentice lived in the house of his master. No money was taken for’the tuition, but after the completion of the apprenticeship the parents of the pupil sent a present of food and kola-nuts. Adoption into the profession went hand in hand with adoption into the guild, and the new craftsman would regard himself as a member of the guild section in which he was apprenticed. However, there have been no adoptions into the craft for many years now.
Each guild section has its own titled head and, besides, one or two other ranks which are bestowed on the family heads of the other large family groups in the section. Th e position of the head of the guild
INDUSTRIES 267 section is taken in rotation by these families, the ‘second-in-command’
succeeding to the position of the head at his death. Among the Dokodza
people we find a threefold elaborate hierarchy which settles the succes•
sion even to the rank of the ‘second-in-command’: Dokodza, Dokpa, and
Shaba Dokpa (lit. Successor of Dokpa). Th e smaller blacksmith section
of the Bida Beni possesses two ranks: Maji Tswata and Shaba Tswata.
Th e blacksmith groups in the villages, and also the groups of iron-ore miners, have similar rank systems.1 The heads of the blacksmith groups, and to a lesser extent the other guild ranks, are respected, and their orders are obeyed by all co-workers. Th e main significance of their position lies in the fact that it is their task to organize all co-operative work within the group.
Co-operation and mutual assistance within the section takes many different forms. If the people of one workshop, for example, are unable to fulfil a certain order, they pass it on to another workshop; or if the occupants happen to be absent when a customer calls, their neighbours will always be ready to take the order on their behalf. When the work• shops ran short of iron ore, the guild-head used to send messengers to the iron-ore miners to bring back a sufficient supply for the whole group; this form of co-operation, however, no longer exists, and to-day each workshop procures its own raw material. Co-operation of a more far- reaching order is concerned with the various tasks which we have placed under the heading of ‘maintenance of labour’. We have already seen that these tasks involve co-operation and mutual assistance on a scale which transcends in varying degree the efako-unit. Thus the division of labour in the preparation of charcoal implies co-operation of the whole workshop group. When it becomes necessary to forge and put up a new anvil an even larger group must co-operate in active work. We may add that the upkeep of the workshop itself, or the building of furnaces—especially the large furnaces used by the iron-ore miners— demands a co-operation much larger than that which the efako group can provide. In none of these cases is the co-operation recompensed with a share in the profits (as is the case in efako co-operation); nor is this co-operation organized on the basis of collective work of the egbe type; nor can it be explained, finally, by the fact that the co-workers may belong to the same family and household. Rather is it a co-operation based on membership of the same professional union—the guild or guild section—and mobilized by the appointed head of the group.
Th e head of the guild section is also responsible for maintenance of labour in a wider sense. His professional advice may be sought by all workers in the group if they have some difficult or unusual work in hand. He will also occasionally lend his people money if they need it for pro• fessional purposes (e.g. the purchase of crude iron ore), in other words, he provides them with the necessary working capital. Certain similar services towards the group are vested in the position of the Dokodza alone, that is, of the head of the guild at large. He arbitrates in all internal disputes among the blacksmiths of Bida, and is thus able to
1 See chart of Village Ranks, p. 57.
268 INDUSTRIES
prevent friction in his body of workers. He is also the agency through whic h important large orders reach the group. On a smaller scale this task also falls to the heads of the other blacksmith groups. But in the case of the guild-head it gains special significance; for these ‘royal black• smiths’ worked for the largest customer of the country, the Nupe court. The y could be certain of a constant large demand, e.g. for tools for the farm-hands workin g for the king , or various iro n implements needed in the royal household, for horse bits and stirrups for the king’s horses, or flint-locks for the guns of the bodyguard. These orders are given by the kin g or court officials to the Dokodza, who then distributes the wor k among the craftsmen of his group according to their capacity and skill , or, if the order proves too large, among the workers in other guil d sections as well . Th e Dokodza also receives the payment for this work , and keeps the largest share; the rest of the money is divided equally among thewholeguildsection,whetherthey have beenworkingon this particular jo b or not.1 These royal orders have been greatly reduced recently, not only because many of the implements needed are obtained nowadays from the European stores, but also because many of the requirements of the royal household are now catered for by the Public Works Department in Bida. Bu t when the present Etsu buil t his new house, in 1935, he ordered nails and brackets for all doors from the Dokodza people.
Th e part played by the guild-head (and to a lesser extent by the sectional heads) in providing Jabour and capital for his group explains the authority whic h he exercises over his fellow workers, and also his ability to act in certain circumstances as their ‘financier’. On the other hand, the close co-operative organization of the guil d section reveals itself as a device to meet these regular demands upon the production of the group. Here we can attempt to explain wh y the smaller guil d sections in Bida (and similarly the village groups of blacksmiths) have suffered more severely under modern economic changes than di d the Dokodza group. Th e size of the groups is itself significant. Only the privileged group of the ‘royal blacksmiths’ evolved a large-scale professional organization; the other groups, whic h di d not have the same access to constant large-scale demands upon production, remained small, and more loosely organized. Now , we can assume that these smaller groups also obtained formerly a sufficient demand for their products from various sources. It may also be that their sources have failed sooner, and more completely, than the source wit h whic h the Dokodza people could reckon—the continued orders from the royal household are evidence of this. But these unequal changes are less significant in their direct effect upon production (which may be negligible)2 than in their indirect effect,
1 How, on the other hand, the authority of the guild-head can be adapted to new situations is illustrated in the following instance. When I went to inspect the furnaces of the iron-ore miners near Bida, I had to obtain a guide through the Dokodza. On our return to Bida the guide at once showed the money which I had paid him to the guild-head; though he was allowed to keep it he could not have done so without first informing the guild-head, who could have deducted a part of the money.
2 Th e order for nails and brackets for the royal house which we mentioned
above brought in only 2s. and a gift of food.
INDUSTRIE S 269 through the medium of labour organization. Th e royal orders, though
at present small and irregular, stil l mobilize the group organization of
the Dokodza people in the traditional way. Th e general weakening of
the guil d organization brought about by recent economic and political
changes, whic h is clearly visible in the decline of the authority of guild –
head and sectional heads, has thus affected the largest and strongest
group less than the smaller, and already weaker, guil d sections. In their
case the group organization has, in fact, been reduced beyond the margin
of safety. We have seen that Nupe blacksmith wor k demands com•
paratively large group co-operation for a successful maintenance of
labour (i n all its aspects). In a weakened organization, a markedly re•
duced co-operative group, maintenance of labour becomes increasingly
difficult. Thi s is what happened in the case of the smaller blacksmith
groups. An d this difficulty, while not sufficient in itself to disrupt the
labour organization, woul d yet weaken its resistance to the effects of
economic change—a decline in trade, the competition of European
goods—effects whic h (at least for the time being) have passed by the
larger and more strongly organized section.
We can quote several instances of this : the division of labour whic h enables a workshop, say, of four men, to continue work while two of the men prepare charcoal is impossible in a small workin g unit , and a valu• able workin g day, possibly in the busiest time of the year, must be sacrificed. Or a workshop may have to interrupt its work when iro n ore runs out and new raw material has to be fetched from the iron-ore miners—a task which , in large-scale co-operation, can be performed by one man for a number of workshops withou t deranging the work of the group. I have even witnessed cases in whic h work had been interrupted, or reduced to small repairs, because the workers had no money to buy raw materials; in a large co-operative group money would have been found more easily, especially wit h the help of the guild-head. In Kutig i I saw one of the two local blacksmith workshops left dilapidated for a whole year because the workers could not muster sufficient labour. Th e most conspicuous example of all is that of the iron-ore miners who had let their means of production, their furnaces, fall into disuse.
BRASS- AND SILVERSMITHS
Th e Nupe brass-smiths—tswata muku—reside almost exclusively in Bida. Th e only place outside Bida where brass-smiths have settled is Kacha, the ric h trading-place, where these craftsmen from Bida hoped to find a profitable market. There are two groups of brass-smiths in Bida, one in the quarter of Etsu Masaba and one in the Etsu Usman quarter. Th e latter has lost its former importance. In technical skill both groups are equal, as I could test for myself; as regards work and design of the traditional type the group in the Etsu Usman quarter seems the more efficient. Yet its decline, unmerited though it was, is not difficult to understand. It was partly due to political factors. Th e group in the Etsu Usman quarter flourished durin g the regime of Etsu Bello, who was of the dynasty associated wit h this part of the town , and also under Etsu
270 INDUSTRIE S
Mohamadu before him, whose royal residence was not far from the brass-smiths’ quarter. Here, as in many other instances, the invaluable royal patronage was extended primarily to the craftsmen living in the part of the town traditionally associated with the dynasty. Wit h the change of the dynasty the royal patronage changed also, and under the late Etsu Saidu, of the house of Masaba, the brass-smiths in the Masaba quarter captured the largest and most lucrative trade. A contributory cause lies in the lay-out of modern Bida. Th e motor-road to the centre of Bida and the big market leads through the street of the brass-smiths in the Masaba quarter, while the other brass-smiths’ group is situated in an obscure corner of Bida, away from roads and markets. Now, the brass- and silver-ware manufactured in Bida consists to a large extent of ornaments for men and women and other ‘luxury’ objects. A large pro• portion is intended for export, and certain goods are even made specially for tourists and European ‘curio’-hunters. In either case the nearness of the big trading centre and the main thoroughfare of the town, with its opportunity for display and for attracting customers, must prove an advantage which is denied to the other, almost inaccessible, group of craftsmen.
The character of this luxury industry also explains its concentration in Bida. Its products find their way to villages and peasant districts as well. But a constant and large-scale demand exists only in the capital. Moreover, the concentration of this industry is, in sophisticated Nupe, a trade-mark of refinement and value, and a guarantee that its products express the latest fashions.
The technical equipment of the workshop of the brass-smiths differs in certain important details from that of the blacksmiths. The work• shops of the brass-smiths, with their smaller and less smoky furnace, are more often housed in the compound itself, i.e. in the entrance hut. The tools are less heavy and adjusted to the finer technique. European hammers, chisels, and files are used to a much larger extent. The main products of the Nupe brass industry are sword-hilts and daggers,1 horse-trappings, bowls, jugs, trays, ladles, and receptacles of various kinds, made mostly of beaten brass or copper; bangles and rings, which are cast in open clay moulds of bars of brass or silver or (nowadays) aluminium and then bent and twisted into shape; also hairpins and silver chains. Artistic decoration, especially ornamentation of the surface of these various objects, plays an important part. From old Nupe brass- and silver-work preserved in Bida and in European museums we learn the different styles in decoration which have succeeded each other in the course of the last hundred years. The style in fashion to-day is orna• mentation in pointille technique.2
The Nupe brass-smiths also practise soldering with tin, and one or
1 Th e blades are old, mostly of German or Austrian manufacture, and have been traded for centuries to Nigeria and Nupe, across the desert, from North Africa.
‘ Th e oldest style appears to be a fluted design which I have seen on certain brass bowls; a later style which flourished at the end of last century, consisted in embossed ornaments which made special use of the flower motif. Th e pointille style appears to have come in about 1910.
INDUSTRIES 27 1
tw o have specialized in repairs of watches, sewing-machines, and such• like. The y also practise a crude technique of cire-perdue casting. Thi s technique is applied in the making of a kin d of signet-ring, very popular bot h among men and women, and, by one expert, in the making of small brass figures of animals, whic h have become quite famous in Nigeria. These figures are traded by many of the itinerant Hausa traders and are bought by Europeans who take them for products of an interesting indigenous art. I cannot prove wit h certainty that these littl e figures— zasagiz’i, as the Nupe call them , ‘likenesses of man’—have never had a deeper significance. Bu t I am inclined to thin k that they have not. The y have no place in Nupe culture as it presents itself to us, either as re• ligious symbols or as symbols of wealth (i n whic h for m similar figures are used, for example, in Dahomey). I know, moreover, that in the case of the one expert wh o produces these figures to-day in Bida it was his ow n enterprising spirit that made hi m adopt and develop this (for a time at least very profitable) ‘tourist industry’. To examine the cultural channels through whic h the suggestion of this production, the artistic motif, has reached Bida, is not part of my present task. I have, however, been able to trace two objects produced by this expert brass-worker to their original source. It is a case of what ethnographers call ‘culture borrowing’ , and is sufficiently interesting to be quoted here. Thi s brass- worker had for some time been producing ‘dancing figures’, littl e human figures which , by a clever utilization of stable equilibrium , were made to revolve and ‘dance’ on a small brass stand withou t falling off. He got the idea for these figures from a travelling India n trader, who had shown hi m pictures of the model and explained the principle to him . He had paid the India n 2s. for the ‘lesson’ and then proceeded to manu• facture similar figures himself. Th e same brass-worker also produced the figure of a cobra of typically India n design, whic h he had copied from a figure he had seen in the house of a European.
Th e raw material used by the Bida brass-smiths is all imported. Th e Nupe , it appears, never mined copper, nor knew how to produce the alloy. In pre-European days, I was informed, brass and copper were brought up the Niger from Onitsa and Lagos. To-day copper is obtained from English pennies, and brass fro m rifle cartridges whic h Hausa traders buy up in the garrison towns of Nigeria and then sell to the brass-smiths of Bida and other towns. Silver is gained by smelting down silver coins whic h the brass-smiths procure from a Kaduna bank. A few examples may illustrate the relation between cost of material and the profit made on the finished article. Brass ladles, an article for whic h there is great demand in Hausa country, are sold for 3d. apiece; for one ladle the brass from twelve cartridge-cases is used, costing 1½d.; the brass-smiths can produce two ladles a day, thus making a daily profit of 3d. Silver bangles are sold for 12s. a pair; the cost of the material is
10s., and the workin g time from two to three days, allowing a daily profit of 9d. to 1s. Aluminiu m bangles, whic h are made to cater for the cheaper tastes, are sold in large quantities to traders and middlemen ; half a dozen, whic h can be produced in one day, are sold for 2s.; the material (bought in sheets at the European store) costs is. 6J., and the profit per
272 INDUSTRIES
day is thus 6d. Th e littl e brass animals, four of whic h can be done in from one to two days, are sold for 2s. to 3s. apiece; the material comes to only 8d. Th e highest profits are thus made on work that demands special skill , taste, or invention, such as the silver bangles or the brass figures.
Amon g the brass-workers as among the blacksmiths three or four men, often close relations, share the same workshop. But it happens not infrequently that people who are not relations work together in the same workshop, or that close relations wor k in different workshops. These are typical examples:
- In this workshop three old men are working together, two of whom are brothers, and the thir d a distant relation who lives in a different household.
- Here we find two distant (maternal) relations, who live in different houses, sharing the same workshop together wit h a young boy who is the pupil of one of the men.
- Thi s workshop is next door to 2. In it are working the two brothers of one of the men in workshop 2 together wit h another man, a distant relation, and a young pupil .
- Thi s is the largest brass-smith workshop in Bida. It belongs to the head of the guild section, and is shared by the guild-head himself, who, however, does not work there regularly, his two married sons, a married younger brother, and a stranger.
Th e labour-unit is much smaller than the workshop group, and also smaller than the labour-unit among the blacksmiths. Grown-u p or married sons or brothers invariably work on their own, so that the work• ing-uni t consists as a rule only of a man and a boy, or frequently of one man only. Thu s workshop 1 numbers three one-man labour-units, workshop 2 two labour-units consisting of one man and of one man and his pupil , respectively. In workshop 3 we have three labour-units, two consisting of one man and one of a man and a boy, and in workshop 4 five one-man labour-units. In these small labour-units the difference between efako and buca work must naturally disappear. Th e explana• tio n for the smallness of the groups lies in the very nature of brass- smit h work. It requires much less actual co-operation than the heavy work of the blacksmiths—in most cases, for example, the brass-smith blows the bellows himself. Brass-smith work, wit h its delicate technique and the large variety of objects whic h it produces, is also more highly specialized. Certain objects, such as bangles and ladles, are produced by all brass-smiths; but workshop 1 specializes in addition in sword-hilts and daggers; in workshop 2 one of the men produces the brass figures of whic h I have spoken; the men in workshop 3 are experts in the repair of sewing-machines, and so forth . A craft, finally, whic h follows fashion so closely and is always on the look-out for new attractive things must foster individual skil l and inventiveness rather than collective enterprise.
Th e guil d organization at large also differs in certain essentials from that of the blacksmiths. Th e training of apprentices plays a muc h greater part in this prosperous and up-to-date industry. Th e training is longer and may take up to ten years. After the completion of the period
INDUSTRIE S 273
of training the new member of the profession joins one of the workshops in the quarter, while continuing to live in the house of his family. Large- scale co-operation is also of muc h less significance, since the furnaces are simple and easy to build , since no heavy anvils are used, and the forges mostly situated in the entrance hu t of the house. Th e varied enterprises, and the spirit of ‘ individualism’, render the advice or practical help of the guild-head less important. His organizing influence, however, is stil l an important factor. When , for example, the Bida brass-smiths first began to produce goods for Europeans, the guild-head took it upon him • self to propagate this new profitable enterprise, and at the same time to assist the craftsmen in perfecting their technique. A Distric t Officer who was a keen supporter of native arts and crafts had explained to hi m the kin d of things that Europeans woul d buy—trays, ash-trays, finger-bowls, and suchlike. Th e guild-head himself led the way in the new production, and devised a system of guidance for his fellow workers. Yo u can stil l see scratched into the mu d wal l of the guild-master’s workshop circles, ovals, and squares of different size, whic h are the outlines of the objects the craftsmen were going to produce; they served as standard measures, anc\ are indeed called by the Nupe etna, measure.
There are two ranks in the brass-smith guild , each presiding over one section: the guild-master, called Muku, and the ‘Successor of the Muku’, the Shaba Muku. The y are the heads of the two large family groups whic h are said to have been the earliest representatives of the craft in Nupe and to whic h most of the brass-smith families trace (at least in theory) their descent. Th e two ranks are taken in tur n by the two family groups. Th e present guild-master, a very old man, is livin g in the brass- smith group in the Etsu Usman quarter; the Shaba Muku is at the head of the more prosperous section in the Masaba quarter. Wit h the grow• ing prosperity of this guil d section the authority and influence of the ol d guild-master have declined, and to-day the Shaba Muku is the more in • fluential person. Even the most successful and most ‘individualistic’ young brass-smiths who have created their own special style treat the head of their guil d section wit h the respect due to his position and ex• perience.
Here, as in blacksmith work, the most important service whic h the guild-master renders his fellow workers is that of acting as an inter• mediary between the craftsmen and the court. Here, too, the kin g of Nupe places all his orders through the Muku, who then distributes them among the different workshops. Th e large orders of former days, for arms, sumptuous horse-trappings, ornaments of many kinds, have dis• appeared. Th e costly ornamented bowls, used formerly in every Nup e household of consequence as drinking-vessels or receptacles for food and kola-nuts, have been replaced by enamelled teapots of European or Japanese manufacture. But on a limite d scale the court is stil l the most im • portant customer of the Bida brass-smiths. To recall themselves to the kin g and to solicit his custom the brass-smiths make hi m an annual gift of products of their craft whic h is presented to hi m at the occasion of the Great Sallah. Th e present whic h the brass-smiths presented to the new Etsu at the first Sallah celebrated under his reign was, not unnaturally,
T
274 INDUSTRIE S
especially costly. It was in form of a brass tray for kola-nut6 which had been made in the workshop of the Muku himself and to the cost of which (65.) every brass-worker had contributed his penny or halfpenny. Th e Etsu accepted the gift with thanks and sent a return gift of a large bowl of kola-nuts (worth 10$.), which was distributed among all the brass-smiths of the town. If here the organizing influence of the guild- master becomes again visible, the ‘individualistic’ element is present also. For, in addition to the collective gift, individual workers are free to send their own personal gifts and to attract the attention of the court to their own particular style and skill. Thus one of the Bida brass-smiths presented the Etsu at the Sallah with a covered brass bowl, worth 3$. He was rewarded shortly afterwards, when the Etsu placed an order for new horse-bits, in European style, with him.
GLASS-WORKERS
The glass-worker guild represents the most strongly organized and most conspicuously self-contained of the Nupe craft-guilds. In the whole of Nupe there exists only one group of glass-workers, the masagd, who live in Bida, in a separate quarter of the town which is named after them masagd fu, efu (quarter) of the glass-workers. They regard them• selves almost as a distinct tribal group. They have a tradition that they came originally from Egypt, and brought their art with them from the old home. Their tradition has also preserved the names of the various towns and countries through which they passed on their long journey which took them, eventually, to Gbara, the ancient capital of Nupe kingdom.1 They had stopped at none of these many towns and villages as they could not find there the kind of earth necessary for the making of glass. They settled then in Gbara, where, as their tradition has it, they arrived at the time of Tsoede. Later, when the kings of Nupe moved their residence to other places, the glass-workers followed the court, to Jima, Lade, Rabd, and, finally, Bida.
Th e tnasagd obtain their raw material both by making their own glass and smelting down bottles of European manufacture. Th e latter kind of raw material, easy to obtain, and more attractive because of the variety of colours which it offers, has to a large extent replaced the original native-made glass.2 Bida women, wives of the masagd as well as women from outside, regularly scour the town for empty bottles from European stores, households, offices, and dispensaries (where they can obtain the highly appreciated amber-coloured medicine bottles). Bottles of various kinds from other parts of Nigeria also find their way to the tnasagd quarter, bought up locally and transported to Bida, partly by railway, by the enterprising women traders.
Th e masagd nevertheless still produce their own glass as well. A
1 My informants mentioned the followin g towns and countries: the ‘lan d of
the Arabs’, the ‘land of Bornu \ Kano , and Zozo (Zaria).
2 Th e actual cost of production is lower in the case of native-made glass. Bu t this fact is balanced by the time-saving made possible by the use of bottle glass and also by the greater demand for coloured glass-ware.
INDUSTRIE S 275 certain type of glass, black in colour and of rather crude texture, which
is bought widely in Nigeria, is made exclusively by the traditional
method. I have had opportunity of studying this traditional technique
of glass-making—a technique which is kept a secret by the glass-
workers and which has never yet been recorded. Several writers on
Nigeria, in fact, maintain that it has long since fallen into disuse and is
even no longer remembered by the people. This native glass is called
bikini, as against the glass gained from smelting down bottles, the name
for which is kallaba, ‘bottle’. Th e process of glass-making is long and
complicated; it takes from three to five days according to the quantity of
glass made. Th e glass can be made in the ordinary glass-maker’s work•
shop, but the furnace must be re-dug especially for the purpose, since
the furnaces used normally for glass-smelting are too wide and too
shallow. The same men who ordinarily work in the workshop also carry
out the making of glass. The glass-maker’s workshop is built somewhat
like the Nupe smithy, but has only one round furnace in the centre of
the hut. For the purpose of glass-making the furnace is dug 5 feet deep
and 2 feet wide. Throughout the process its mouth is covered with pot•
sherds and broken calabashes which leave only a narrow passage for the
smoke. As good Mohammedans, the masagd invariably start their glass-
making on Friday afternoon, after their return from the mosque. Th e
various ingredients have been collected and arranged inside the work• shop : earth (the same earth that has been dug out of the furnace),1 soda (kanzva) of the kind that is brought by Hausa caravans from Lake Chad, water in pots, and a vast quantity of firewood. Th e soda, which has been pounded to a fine powder, is mixed with sand and a little water till it gains a mud-like consistency. The mixture is placed on the bottom of
the furnace, with firewood and a layer of grass for the lighting of the fire on top. Th e fire is lit, and the furnace is kept ablaze, day and night, till the process is finished. The men work in shifts so that there are always four workers tending the furnace; adding fuel, and occasionally probing the bottom of the furnace with long iron rods. Th e heat is so intense that the workers have to put up mat screens as protection. Six or seven hours after the lighting of the furnace large quantities of water are added, together with a handful of slag from a blacksmith’s forge. After another twelve hours the substance in the furnace becomes liquid, ke nuwa na,
‘like water’, and after eight hours again the process is completed. Th e glass, a red-glowing semi-liquid mass, is fished out with long iron sticks and left to harden and cool on the sand outside. The cooling takes five to six hours, after which time the glass is ready for use in the other fur• naces, where the finished products—bangles and beads—are manu• factured. Th e colour of the raw glass is a deep glossy black. Th e masagd
1 Th e sand is called jikana, which is the Nupe name for ordinary sand. Th e glass-workers informed me that the sand everywhere in the locality was suitable for glass-making. When I returned from Nigeria I brought with me samples both of the Nupe glass and the ingredients used in glass-making. Thanks to Professor Seligman the Courtauld Institute in London undertook to test the glass and the glass manufacture, by attempting to produce glass from these ingredients. Th e attempt was fully successful. For a fuller report on this experiment, and also on the technique of Nupe glass-making, see Man, 1940, 107.
276 INDUSTRIES
have never possessed the art of colouring their glass. The y have to do it indirectly, by mixin g it wit h coloured bottle-glass. They also ornament their dark glass wit h dots, lines, and circles in various bright colours: the red colour is obtained by smelting down thi n iro n sticks, other colours by smelting down coloured glass or coloured glass beads (Venetian beads mostly).
Glass is, as a rule, made only in one workshop at a time, while the other workshops carry on the ordinary routine work. Me n from other work• shops who need new glass contribute soda and sometimes also firewood (or else pay in money for their share in firewood), and receive a pro• portionate quantity of raw glass. The y pay the men who were tending the furnace in kind , returning to them a small quantity of the newly made glass.
Th e further treatment of the raw bikini glass corresponds in every detail to the treatment of bottle-glass, except that the latter has to be crushed into small pieces before being used. It is smelted down, again, over the ordinary furnace, unti l it is semi-liquid and can be formed into beads and bangles of different shape and size. Thi s finished glass is called karuwa. Th e furnace in which the raw” glass was made is later rebuilt and adapted for this new work. It is made a littl e wider, banked up wit h a mu d wall i foot high , and provided wit h bellows.
Nupe glass-making implies a strict organization of work and a well – planned distribution of the various tasks among the whole community of glass-makers. Th e assignment of tasks varies from day to day, according to when a particular workshop has completed the work in hand. Th e masaga quarter numbers thirteen workshops, and each workshop per• forms its assigned tasks. In one hut you may perhaps find the making of bikini glass going on, while in another molten glass is formed into black bangles; in a thir d workshop coloured bangles are manufactured, or small black beads; another workshop, again, may specialize in blue and red beads that day, and so forth. Th e workers, four to six in number, sit roun d the furnace. One of them, as a rule a small boy, is blowing the bellows. Th e raw glass is throw n into the furnace, where it is left to liquefy. Th e glass-worker sticks small lumps of the semi-liquid glass upon the point of a long thi n iron rod. Grippin g the rod in his right hand, he holds it over the mouth of the furnace and, while the glass liquefies further in the heat, he rotates the rod rapidly, throwin g the glass lum p wit h quick, calculated movements into any shape he wants. A second rod , held in the left hand, helps in the process, and is used to produce the widt h of the bangle or to tap a bead into shape. In the single workshop the work is, again, carefully divided up : while one worker prepares the liqui d glass, another shapes it into bangles or beads, and another adds the coloured ornaments.
Th e scheme of collaboration may change from day to day; every worker, frequently even including the man or boy at the bellows, is, in fact, a fully fledged all-round craftsman. Th e trained worker performs his tasks wit h ease and rapidity and, thanks to the smooth co-operation between the workers, the whole workshop equally maintains a very high speed of production. It takes one worker between three and four
INDUSTRIES 277 minutes to produce a medium-sized bangle; I estimated the total daily
output of a single workshop at from fifty to sixty bangles.
Th e finished articles, the coloured and ornamented bangles and beads, are left to cool in a heap of ashes inside the hut to be collected later in large baskets and to be put aside for the traders who call regularly at the masagd quarter. Th e glass-workers have a small but busy market of their own in the centre of their quarter. To this market the traders in bottle-glass bring their goods, women from neighbouring hamlets and farms the daily supply of firewood, and other women traders grain and various food-stuffs for the households of the glass-workers. Th e traders in the finished article, however, buy the bangles and beads directly from the producers, i.e. the individual workshop. These traders are both men and women and, among the women, both wives of the glass-workers and women from outside. Th e masagd themselves never trade their glass• ware. Bida glass is bought and sold all over Nigeria—it is to be seen in every larger market of the country, in Lagos in the south, and in Kano in the north . There is a steady demand for it , and the masagd work regularly six days a week (resting only on Friday), all through the year, always certain of disposing easily of their output. They have neither time nor inclination for additional agricultural work, not even in the re• stricted form of farming ‘for the mouth ‘ whic h is common among all other Nupe craftsmen.
Th e people who work together in the same workshop are, as in the case of blacksmiths and brass-smiths, mostly close relations. But among the glass-workers the group which shares a common workshop also forms a single labour- or efako-unit, workin g under a common head who organizes the group work, disposes of its output, and, in efako manner, controls the income produced by the group. Even distant relations or strangers who have joined a workshop fall under the efako rules. Th e labour organization of the masagd, then, does not allow for an individual buca income in addition to the group income. Instead we find a new arrangement in the form of small regular ‘wages’ (at the rate of zd. a. day) whic h the efako head pays his adult co-workers oiit of the income pro• duced by the group. He keeps the rest and, wit h it , defrays the costs of production, i.e. the costs of fuel and raw glass or the occasional purchase of new tools from the blacksmiths, as well as the various family expenses entailed in his position as efako head. Thi s financial arrangement is clearly necessitated by the special conditions of labour: it woul d indeed be extremely difficult to calculate the share of the individual worker in this strongly co-operative form of production. Th e intensive co-opera• tio n itself is explained by the nature of this ‘mass production1 and by the demands of efficiency and economy: the use of a common furnace and a common store of fuel and raw material, the specialized tasks of the in • dividual workshops and, in the single workshop, the elaborate division of labour, are all calculated to ensure a high degree of efficiency and economy. We note that the means of production are partly produced by join t labour (e.g. the buildin g of the furnace), and partly purchased by the leader of the labour-unit out of his share in the group income (e.g. the iro n tools, the fuel); and we note further that this method of production,
278 INDUSTRIES
which, in the interest of efficiency, has evolved centralized control of labour and means of production on a comparatively large scale, comes closer than any other Nupe craft to the principle of the modern factory system.
We have already seen that in the group at large, i.e. in the glass^ workers’ guild, a similar intensive co-operation obtains. Its organization falls to the guild-head, who assigns the different tasks to the various workshops and is responsible for the general planning of production. This planning must take into account the varying conditions of the market, and we find, in fact, that the guild-head, without interfering with the commercial transactions of the individual workshop, regularly con• sults traders and middlemen who visit the masagd quarter, and maps out accordingly the scheme of work for his community. On my many visits to the masagd I found that the guild-head could always tell me what kind of work was carried out in the various workshops, or when a particular workshop would start on a new task. Th e guild-head, who styles him• self Masagd, has his own workshop, where his sons and brothers are working. He does not often work there himself, but certain difficult tasks, such as the making of very big, heavy bangles, which fetch the highest price, he alone can perform. The guild-head assists his co• workers with advice and practical help, also with small loans of money for the purchase of raw material or fuel; he arbitrates in internal quarrels, and he is, like all Nupe guild-heads, the accredited representative of his guild at the Nupe court.
Th e ever-expanding business of the Bida glass industry involves an almost permanent need of new labour. Formerly the glass-workers bought slaves and trained them in the craft in order to enlarge their machinery of production. To-day they are regularly accepting appren• tices. In 1936 five young men were being trained as glass-workers. An apprentice applies to the guild-head, who will then place him in charge of a certain workshop. Th e training is calculated to take three years, during which time the apprentices continue to live in their own homes and come t6 the masagd quarter only to work. But when their apprentice• ship is completed they not only move into the masagd quarter, but are expected to marry masagd women (as did also the slaves in former days). This adoption into the kinship group is considered essential and is as much a qualification for the practice of the craft as is the technical train• ing itself. Thus the monopoly of the glass-workers is kept intact, and no new branches can be established outside their organization. Kinship interests and kinship co-operation are enlisted in the service of profes• sional co-operation and solidarity and in maintaining the continuity of a highly specialized, monopolized and almost secret, technique of production. Economic interests, in turn, are seen to be sufficiently powerful to effect a radical adjustment in the kinship structure, and to create, in a patrilineal society, a matrilineal enclave.
WEAVERS
Th e weavers, called edeluciZi, present another instance of the tripar• tite organization of Bida crafts. Each of the three ekpS. of the town harbours one of the three guild sections of the weavers. However,
INDUSTRIES 279 according to Nupe tradition , weaving was not originally a guild-organized
craft, bu t a free profession carried out individually, like mat-weaving,
for example, to-day. It became a ‘closed profession’ and a guild –
organized craft when the Nupe kings imported craftsmen from the
south, who brought with ‘ them their own special technique and who,
aliens in Nupe country, kept to themselves, neither intermarrying nor
coming much in contact wit h the rest of the population. These im •
ported groups of weavers were konu, freed slaves of war. The y for m two
of the three groups of weavers which we find in Bida to-day. Th e first is
composed of people from Salu, a tow n in Yagba country, who were
settled in Bida by Etsu Masaba. Th e members of this group, whic h is
the largest of the three weaver guilds, live scattered over the whole
town ; their guild-head, however, used to reside in the Etsu Masaba
quarter. Th e second konu group is of Yoruba extraction, and is con•
centrated in the Etsu Usman quarter. Thi s group has representatives
also outside Bida, in a number of Nupe villages: Kutigi , Enagi, Sakpe,
and Dabba. Th e thir d group of weavers, in the Etsu Umar u quarter, is
composed of Nupe proper. It is a very small group, consisting only of
one compound. Thi s Nupe group of weavers forms a section of a large
kinship group whic h comprises farmers, who live outside Bida, as well .
The y trace their profession to a man who came to Bida from Rabd, as a
bora of Etsu Umaru . Originally a warrior, this man later settled in the
tow n and learned weaving from a konu craftsman. At that time , [ was
told , weaving was a lucrative and highly respected profession. Weaving
has been the profession of his family for three generations now, and the
family heads are received at the court and treated as official representa•
tives of their group like all heads of guilds or guil d sections. Small
groups of weavers of Nupe extraction are also found in various villages
in the Emirate. Thei r labour organization closely resembles that
of this last group of weavers, but they remain outside the Bida
guil d system.
Th e technique of weaving is the same in all groups. Thei r loom is of the narrow upright type, worked wit h pedals, whic h is common all over Norther n Nigeria. These looms produce narrow strips of cloth 5 inches wide. Th e looms are placed in a half-open shed, six to seven in one row, in the inner courtyard of the compound, each loom being worked by one man. Th e weavers have to buy a large proportion of their tools from other craftsmen: the shuttle and the wooden posts on whic h the loom rests, and whic h are frequently beautifully carved and painted, are made by the carpenters; the iron rods used as pedals and the iro n bar upon whic h the finished strips of cloth are wound , are made by the black• smiths. Only the weaver’s reed is produced by the weavers themselves. A complete outfit for one man costs from 6s, to 12s., according to whether cheap plain or expensive carved wooden posts are used. Th e Bida weavers use mostly thread of European manufacture, both plain and coloured, and only very rarely native-ginned and -dyed cotton. Th e village weavers use mostly the latter kind .
Th e men who occupy a common workshed are members of the same family and household group. In one house, for example, I found
280 INDUSTRIES
thirteen looms, of whic h only ten were actually used, as the family numbered at that time only ten adult men. Th e individualistic nature of this work , however, in whic h no actual workin g together of several individuals is involved, is reflected in the restricted efako arrangement, whic h comprises only a man and his young, as yet unmarried, sons. As soon as the sons are married, they wor k on their own , although they may continue to use the looms and tools whic h the family head had allotted to them when they were stil l working in the efako group. Th e efako head also supplies them wit h cotton thread for the start. Whe n they are fairly launched on their new individual work they have to buy their ow n raw material from the proceeds of their work. Only the looms are stil l regarded as family property, and are kept in repair or replaced if neces• sary by the head of the household. In this case the individual worker pays the family head a regular ‘rent ‘ (called ladat lit . wages) of 2d. in every 35. whic h he earns wit h his work—an arrangement whic h clearly takes into account the high cost of new looms. Thi s arrangement, how• ever, has been dropped; the young workers make themselves responsible for the cost of repairs or the purchase of their means of production, and the family head only receives occasional presents of food and clothing. I n a number of cases I found the efako-ximt still further reduced, and even unmarried sons workin g on their own as soon as they reached the age when tax had to be paid for them. Th e difficulty whic h many family heads have to-day in providing tax money for a large family explains this disintegration of the labour-unit. Th e steadily declining demand for native woven cloth has been the main cause. In many houses looms stand unused, because weaving is no longer profitable, and young men, released from the efako-umt, use their newly wo n leisure to tur n at least temporarily to other, more lucrative, professions. In one weaver’s family one of the three sons had given up weaving and become a messenger in the Distric t Office; in another family, two of the six sons became servants in houses of Europeans, and only one of the two returned to his ol d profession after a number of years, having collected enough money for his bride-price and for a new start in his former profession.
Th e individual weaver is free to plan his work independently, and there are a number of different methods whic h he can adopt. Th e long narrow strips of cloth whic h the Nupe looms produce are always woven to a standard length of ten kpere, i.e. arm’s lengths, the weaving of whic h takes five days. Th e most expensive, striped, cloth is then cut into five strips of two kpere each and sewn together, edge to edge, to for m the complete ede, cloth, of the kin d wor n by women round the waist or as kerchiefs over the head. It is reckoned that one man can produce three to four lengths every month . Th e Nupe weavers work every day except Friday, but much time is taken up in getting the cloth sold. A weaver can entrust his goods to one of the brokers who have their shops on the Bida night market; bu t nowadays this means a long wait—even if one is lucky enough to get the cloth sold at all. Man y younger men prefer therefore to do their own trading, and wander through the villages, often for three or five days, selling their cloth in village markets where there is less competition both in native-made and European goods. One piece of
INDUSTRIES 281
cloth of this expensive kin d is sold for 4s. 6d.; as the cost of the material is 3s. 4d., the profit made in five days’ work is is. 2d., or jus t under 3d. a day. If cheaper, plain material is used the profit is slightly smaller; the cost of the material for one piece of cloth is 1s. 6d., and the cloth is sold for 2S. 6d., six pieces being again produced each month , whic h gives a profit of is. per cloth, or 2d. a day. But instead of cutting up and sewing together the cloth in form of the ede, the weaver can also sell the full – length cloth as such to the cap-makers, or, better still , can make it into caps himself. A weaver can produce one cap (made of one kpere of cloth) each day, that is, twenty caps a month, allowing for tw o intervals of five days each for the weaving of the cloth. Th e white cloth caps, called/w/a, are very popular in Bida and are sold at prices from gd. to 1s.; the cost of the material being 6s. 8d. the profit on twenty caps is thus between 7s. and 12s., or from 3d. to 5d. a day. It is interesting to note that this last kin d of work , although more profitable and also in much greater demand, is nevertheless regarded as a slightly degrading occupation. Fula making, my informants pointed out, is tailor’s work, and not the proper work for trained weavers; they undertake it , as they frankly admit, only in order to obtain money for tax or bride-price. In this respect, then, the dissolu• tio n of the efako-umt offers another benefit to the young members of the group who are out for quick money: for, relieved of their duties towards the head of the group, they are also relieved of their duties towards the ethical standards of the profession, to whic h the older men stil l cling.
In the villages only cheaper qualities of cloth, mostly native-dyed cloth , are produced; striped cloth whic h is sold for is. 3d., the material costing 9d. and the dye 3d., and plain indigo-dyed cloth (a kin d of cloth that woul d find no buyers in Bida), sold for 1s. 9d., and produced at a cost of 1s. 2d. But even when the village weavers produce the same kin d of cloth as the Bida craftsmen, they sell it cheaper, being content wit h a smaller profit. Thu s striped cloth of the kin d sold in Bida for
2s. 6d. is produced by the village artisan for 2s., allowing a profit of gd. instead of 1s. 2d. On the other hand, the work and material of the village weavers are known to be inferior to that of the Bida craftsmen. Th e village weavers frankly admit it . The y use the cruder indigenous dye instead of European dyes, they gin their own cotton and work wit h this inferior cotton-thread whic h breaks frequently during the weaving. The y take ten days for the weaving of a piece of cloth whic h the Bida craftsmen finish in five. But just because they limi t themselves to this cheaper and cruder material and cater exclusively for the less wealthy customers, they can stand the competition of European cloth better than the Bida weavers, wh o have specialized in more ambitious and expensive work. Here are the comparative prices: a striped cloth of European (or Japanese) manufacture is sold in Bida for 55., as against 4s. 6d. for Bida- woven cloth and 1s. 2d. for cloth produced in the village.
Formerly, when the efaka system in weaving was stil l capable of sup• portin g a whole family, the younger members of the efakd group prac• tised a littl e farm-work as buca. Th e family head di d not joi n in this additional work and often, if he was an old man, he di d not even take part in actual weaving. Hi s income was supplemented by training
282 INDUSTRIES
apprentices. This source of income, formerly quite considerable, dis•
appeared when weaving became less profitable. Young boys of about
10 years, mostly sons of farmers, were the pupils; they stayed with the
weaver for from four to five years, and the work which they produced
during the time of their apprenticeship belonged to their master; when
their training was completed the parents of the pupil would bring one
bundle of corn as payment.
Th e loose organization of the labour-unit has its counterpart in the organization of the profession at large. Among the weavers no co-opera• tion on a larger scale exists. We have even seen that individuals are quite ready to abandon the profession for other, more attractive, work—an attitude unprecedented among blacksmiths or glass-workers. Yet the symbols of guild organization, the guild ranks, are still preserved. Each group of weavers has its titled guild-head: the head of the Salu (Yagba) weavers is called Maji, and the head of the-Yoruba weavers Leshe. He is also regarded as the head of all konu weavers in Nupe, and used to receive from them small annual gifts or tributes. But this common allegiance and co-operation reflects less the professional union than the tribal and cultural unity of these scattered Yoruba groups in Nupe country. Th e head of the Nupe group of weavers in Bida has no official title but is called simply Ndako, Grandfather. Th e position of the group-head carries to-day none of the influence and authority which characterizes the guild-heads of other crafts; he neither receives gifts from the fellow craftsmen nor is he expected to advise or help them in their work. As in all other guild-organized crafts the official position of the guild-heads of the weavers was based originally on their role as intermediaries between the king and the craftsmen of Bida. But this role has lost very early all practical importance—earlier than it did in the other Nupe crafts. To quote the guild-head of the Yagba weavers: ‘Etsu Masaba still used to come to our workshops and order cloth personally from the Maji. But Etsu Umaru, who came after him, already bought what he needed on the market, and so did all the later kings of Nupe.’
BEADWORKERS
As a last example of guild-organized industries I will describe the organization of the beadworkers in Bida. Their craft shows certain specific features which make it specially interesting and distinguish it from the other crafts of Nupe. Beadworking represents, first of all, a typical refining industry, which involves a system of re-exportation com• parable only to the most complex forms of economic enterprise found in Western civilization. It involves further the use of money capital and credit operations on a scale unparalleled in Nupe industry. It is, finally, a pure luxury industry, producing more expensive goods than either glass- or brass-work. These various factors have contributed to making the beadworking industry of Bida extremely vulnerable; it has suffered more than the other crafts from the economic stress of the recent years and, a significant result of this, presents the picture of a complete dis• integration of co-operative production.
INDUSTRIE S 283
There are three groups of bead-makers, lantana, as they are called, in Bida, one in each part of the town. Their work consists in ‘refining’ the crude or plain beads which are brought into the country from the north by Hausa and Arab traders, that is to say, they polish and sometimes bore these beads and grind them to various new shapes. The grinding and polishing is done on large flat sandstones especially brought to Bida from the river Niger. Th e grindstones and a few small iron tools for boring represent all the utensils used by the beadworkers; they are com• paratively cheap—the grindstones are sold for is. in Bida—and the full equipment of a worker costs not more than two or three shillings. Th e beads are finished in different styles according to the nature of the bead and also according to the prevailing fashion and the demands of the market. Th e following styles of beads are in fashion to-day: long slender round beads, used for the egbogi, the chains of beads which the Nupe women wear round the waist; small ball-like round beads, with highly polished surface; and small flat triangular beads, worn as necklaces. Th e Nupe dislike edged beads and beads with dull surface; beads of this kind are polished and ground down to the desired shape and appearance. Th e grinding and polishing of the beads is a long and delicate work. Th e average production per worker is three beads a day. Th e finished beads are strung on cotton thread and arranged in long chains in which beads of different kinds and styles are often mixed, and are then ready for the market.
Th e beadworkers have no special workshops, but carry out their craft in the entrance hut of the compound. Four or Rve men are working in one hut; each man works independently, sitting on the ground behind his grindstone. As in all other crafts, the men who share the workshop are usually close relations who live in the house of which the entrance hut serves as workshop. Occasionally there is a distant relation or a stranger amongst them who lives in another house and comes there only to work. More rarely an individual beadworker who has no relations practising the same profession would work alone in his own house. Traders and buyers, who bring beads to be polished or bored, or who collect the finished article, are constant visitors in the workshops of the lantana.
In the economic arrangement which obtains in the Bida beadworking industry the majority of workers are reduced to the position of mere labourers, working to order and paid at a piece-rate. One of the bead- makers in the workshop may act as contractor: he buys the crude beads from the Hausa or Arab trader and has them worked by his fellow workers. He divides the work among them and pays them \d. per bead. Th e finished chain of beads is entrusted to the wife of the contractor; she will sell it in the market or to one of the traders in beads, who will then ‘re-export’ these products of Bida fashion industry to the north. A chain of crude beads bought for 7s. will fetch 10s. as the finished article;
2d. to 3d. go to the woman ‘middleman’ as commission, and about 15. 8d. (for a chain, say, of forty beads) goes to pay the wages of the labourers, so the’contractor can make a net profit of is. 2d. on work that took three to four days to complete; the individual labourers make in the same
284 INDUSTRIES
time $d. Evidently, only a man who possesses some capital can assume the role of contractor, and the ordinary beadworker can hardly hope, wit h his daily wages of 1½d., ever to attain this position. In a few cases I have found two contractors in the same workshop; but many other workshops numbered no ‘capitalist’ among their workers. In a situation of this kin d two other methods are open to the beadworkers: either they buy the crude beads on credit from the traders and pay them back after they have sold the finished article; or they undertake the defining’ work on a piece-contract for a trader in beads—the Hausa or Arab trader, or sometimes a ‘capitalist’ woman of the beadworker’s family. In neither case does the work of the beadworkers include efako work, even on the smallest scale. Al l work is done individually, and even unmarried young men are workin g on their own, being assigned by the contractor a task suitable to their skill and experience. Th e beadworkers, finally, do not practise farming in addition to the handicraft.
Formerly it was the family head and, on a larger scale, the guild – master who assumed regularly the role of contractor and financier. As in all other crafts the guild-head made himself responsible for ‘mainten• ance of labour’ in the group, whic h means that he undertook to provide the necessary workin g capital and credit and to distribute the work effectively among the craftsmen. Th e guild-head, called Maji, Master, also acted as the agent of the king in all matters concerning the trade in beads. Whe n the foreign traders in precious stones arrived in Bida, they woul d not take their goods at once to the public market, but woul d show them first to the guild-head of the lantana, who woul d select what he needed, and what he thought suitable for the kin g and court. He woul d take the collection of newly arrived beads to the kin g and advise hi m on the purchase; then, receiving the order for the ‘refining’ of the beads, he woul d distribute the wor k among the workshops of the lantana paying the workers the usual piece-rate and keeping a large commission for himself. Th e constant demands of the court—the requirements of the fashionable ladies in the harems of Nupe royalty, bride-price and dowry (of whic h chains of beads form an essential part) whic h the king pays for innumerable relations and dependants—kept the beadworkers of Bida busy. Orders amounting to £5 and more at a time , whic h secured the Map a commission of £I, were by no means rare. Th e profits made in this fashion enabled the guild-head to finance the ordinary work of his group and to offer them credit for the purchase of their ‘raw material’. In these times of prosperity the individual worker who desired to act himself as contractor could withou t difficulty obtain large credits (up to
£1 or £2) from the Hausa bead traders. Th e guil d organization and the strong financial position of the guild-head were evidently regarded as sufficient security.
Al l this is a thin g of the past. Th e court no longer buys beads from the beadworker guild . European beads, cheaper and muc h more attractive in shape and colouring,1 have crushed the native industry. Wit h the decline of the financial position of the guild-head, outside credit
1 Some of the European beads which are sold in the markets of Nigeria are, in fact, specially manufactured as copies of the native style.
INDUSTRIE S 285 has also disappeared. Th e result is a complete disintegration of the
guil d system. Th e guild-head, no longer able to act as financier and
contractor, has lost his position of authority ; as his contribution to the productive organization of the group was concerned exclusively wit h these financial and credit operations, the very raison d’etre of a guild – head has disappeared. Thi s is evident in the manner in whic h the Majl is treated by the other craftsmen; no one pays the slightest attention to him . Since he is too old now to earn money, he is supported by his son
—bu t this is entirely a family matter, and in no way concerns the pro• fessional group at large. He receives no gifts or shares in the profits, not so muc h as an occasional present of kola-nuts, whic h even in the greatly weakened guil d organization of the weavers stil l marks the privileged position of the guild-head. If I have called this change in the productive organization of the beadworkers a decline and disintegration, it is also felt as such by the people themselves. To them, it appears as the sad decline of a once highly respected and well-organized craft: as one of them said, ‘the beadworkers to-day live from hand to mouth ; their Maji has no longer any authority; theirs is an unlucky trade’ (gwa zi, lit . an unlucky hand).
CHAPTE R X V INDUSTRIE S (continued) INDIVIDUA L CRAFTS
HE guild-organize d crafts represent, as we have seen, a specific social grou p or class—the class of skilled artisans—which onl y occasionally enlarges its membershi p by adoptin g individuals fro m other professional o r social groups. Th e individua l crafts, o n the other hand , present a vaguely defined an d ever flui d group whic h recruits itself fro m every stratu m of Nup e society, th e peasant class as wel l as th e ‘intelli • gentsia’, slaves an d clients as wel l as member s of th e nobility . We shall see presently tha t me n of th e title d classes wh o have lost thei r forme r sources of income frequentl y adopt an individua l craft as a ne w means of livelihood . Th e technical conditions whic h contribut e t o mak e th e guild-organize d crafts int o ‘ closed groups’, organized on th e lines of mor e or less intensive co-operation, are absent in the industries whic h are carried ou t as ‘free professions’. Tailors , straw-ha t makers, or mat – weavers need no workshops; the y can wor k anywhere , in any par t of th e house, in sleeping-huts as wel l as in a corner of the katamba. Some craftsmen even wande r fro m place t o place an d practise thei r wor k whereve r there is a prospect of trade . Man y Bid a tailors, for example , have mad e a practice of visiting neighbourin g villages on th e marke t days o r a t the tim e o f the annua l ceremonies an d festivals, whe n the y can b e sure o f findin g plent y o f work . No r does thei r wor k involve co-operation, or necessitate the use of expensive tools. IVfat-weavers wor k wit h thei r hands, straw-ha t makers use a block of clay or woo d (called ekugi) over whic h the y shape th e crow n of th e head , a too l whic h the y can easily mak e themselves. Tailors , embroiderers, or leather-
workers need, i n additio n t o th e material , onl y needle an d thread .
Althoug h ther e exist, specially in Bida , experts in these crafts wh o practise the m as thei r onl y profession, th e techniqu e itself is simple enough t o b e learned b y anyone i n a ver y short time . W e re • membe r th e farmers wh o carry ou t a handicraft as buca wor k in thei r spare time . People in th e tow n similarl y practise these handicrafts as leisure-time occupations, in additio n to othe r work , or as a lucrative pastime. Man y learned me n an d students o f th e Kora n d o a littl e tailor • in g beside. Embroidery , a mor e delicate work , involvin g taste an d a certain skill , has become th e pastime occupation of man y member s o f th e ‘intelligentsia’. O f hat-making , o n th e other hand , th e Nup e say tha t it is th e ideal occupation for ‘lazy people’, wh o can do nothin g else, meanin g particularl y those impoverished ‘people of quality ‘ wh o are neithe r skilled no r resourceful enoug h to find other means of livelihood .
Th e varied demands o f the Nup e markets have also called int o bein g various other, minor , handicrafts, some of whic h are so easy tha t youn g boys of 10 or 11 are encouraged to take the m up as a first buca occupa-
INDUSTRIE S 287 tion , for their pocket-money. Fo r example, a father woul d give his boy
2d. to bu y cotton-thread whic h the boy woul d then plait into trouser
strings and sell for 3’/.; or boys wor k thi n strips of leather into the
lacquered amulet strings whic h the Nupe wear roun d the neck, making
again a profit of id.
A few examples may illustrate the economic scope of these handi• crafts. Straw hats are made of narrow strips of plaited grass, sewn to • gether in spirals. Th e making of a good straw hat takes fro m three to four days if the hat-maker buys the plaited strips of straw ready-made, and five days if he plaits them himself. A plain hat is sold for 4d. to 6d., the material costing between 1d. and 2d. A more expensive type of hat, in whic h part of the crown is covered wit h leather-work, is the product of tw o craftsmen, straw-hat maker and leather-worker. A hat of this kin d sells for is. Th e Bida tailors produce a variety of goods: caps bot h plain (machine-made) and embroidered; gowns, trousers, saddle-cloths. A plain white cap is sold for 4d., the cost of the material being 2d.; a tailor can produce four such caps a day. Th e embroidered caps are sold for ur., the material costing 6d.; one cap is made in three days. Plain whit e gowns, also machine-made, are sold for ys. 6d., the cost of the material being 55.; it takes one or two days to make. A simple saddle-cloth takes fro m three to five days to make; the cost of the material is 6d., and it is sold for is. or 1s. 6d. A more elaborate cloth, embroidered and dyed in various colours, takes twenty days to make; the cost of the material is 3$., and it is sold for 10s. An embroidered gown is made in fifteen to twenty days. Embroidered Bida gowns are famous all over Nigeria ; they are sold for £2 to £5 IN Bida, and for much more in the north . Th e cost of the gown itself (whic h is made by the ordinary tailor) and of the material used in the embroidery comes to between 15s. and 30s.; the profit of the craftsman is thus between £2 and £3 . Bu t then, he may not be able to sell a gown of this kind , or receive an order for an embroidered gown , oftener than once a year.
We have already mentioned that certain villages, e.g. Kutigi , are famous for their mat industry. Every man and boy in Kutig i works mats in his spare time—and spare time here means jus t as muc h the leisure season of the year as every hour that can be spared from farm-work. Whe n sitting in front of their houses in the evenings, even when walking home from the farms, the people are working : they carry a supply of dried grass dyed in different colours, in a kin d of quiver under the arm and, while they walk or talk, their fingers are busy plaiting the grass into long, narrow strips. Th e strips are 3 or 4 inches wide, and up to 15 yards long. Th e Nupe make oval mats, in whic h the long strips are sewn together in a multicoloured spiral pattern; or square mate, in whic h the strips, cut to the same length, are sewn together edge to edge, again in a multicoloured design. Th e Nup e also make square mats of split sorghum and maize stalks, often dyed in red, yellow, and black, whic h are harder of texture and have a glossy surface. Th e oval mats take fro m five to seven days to make; the cost of the material—grass brought from the river swamps (see p. 200)—and the dye is 5d.; the mat is sold in Kutig i for 9d. or 10d., and in Bida, after it has passed through
288 INDUSTRIES
the hands of a middleman, for 11d. or 1s. Th e square grass mats take ten days to make; the material costs 7d. or 8d., and the mat is sold for from is. 3d. to 2s. Th e mats made of stalks take a fortnight to make; the material costs nothing, as the stalks are found on the farm of the crafts• man, and the dye is produced fro m the sorghum stalks; a plain mat of this kin d is sold for 8d., a dyed mat for 15.
If we reduce these various figures illustrating cost and profit to the average profit per day made by the individual craftsman (assuming that he is always working to ful l capacity) we obtain the following table:1
Commodity Profit per day
Straw hats . id. Grass-mats, oval . id. Grass-mats, square lid. Mats made of stalk 1½id. Caps, embroidered 2d. Saddle-cloth, plain 2d. Saddle-cloth, embroidered Ad. Caps, machine-made U. Gowns, machine-made . is. lod. Gowns, embroidered 3s. od.
I t wil l be seen that the Nupe price system is highly discriminative bot h wit h regard to the taste and skill needed in the production of a particular article and to the factor of risk involved in the production of more expensive and ‘luxury ‘ goods, the demand for whic h is less wide • spread. Hat-making, regarded commonly as the easiest craft of all , and at the same time a craft workin g for large constant demand, is remunerated most cheaply; the embroidering of gowns, whic h involves considerable taste and skill , and whic h can reckon only wit h a comparatively small and irregular demand, is remunerated most highly. Thi s calculation of price and profit holds good similarly for the other products on our list. However, there are two conspicuous exceptions: machine-tailored gowns and machine-tailored caps command a price quite out of pro • portio n wit h the other products of Bida handicrafts. We shall retur n later to this anomaly.
Even in these individual crafts we find whole families or household groups following the same profession. Bu t just as often we find house• holds in whic h three or four different professions are represented, and houses in whic h tailors, hat-makers, embroiderers, and, say, cushion- makers practise their craft at the same time, in the same katamba. In these crafts whic h involve no special training and in whic h the novice derives no special benefit from partnership wit h the more experienced older men, individua l preferences have free play and outweigh possible tendencies to follow the example of one’s family or one’s elders. Bida
1 At the time when these figures were recorded prices were lower than usual (see p. 292). Th e relation between the various prices, however, had remained unchanged. Th e profits quoted for mat-making in Kutigi supplement the state• ment made previously that mat-making represents a lucrative leisure-time occupation for the Nupe farmer. Th e demand formats is, in fact, increasing, and latterly Kutigi mats have been exported to England through the United Africa Company.
INDUSTRIES 289 men are always ready to tr y their hand at something new, and changes of
profession are frequent. Th e record is held, perhaps, by a Bida man
who started as a mat-maker (the occupation of his older brother),
changed over to cushion-making, then to tailoring, and eventually to
leather-work, whic h he found more lucrative and whic h he is practising
at present. Efako co-operation on the narrowest basis, between father
and unmarried sons, occurs occasionally. Thu s in a family of em•
broiderers a father and two sons worked together; workin g on the same
gown, the father woul d execute the more elaborate designs, while his
two sons carried out the simpler embroidery; the sons were maintained
by the father, and all earnings went to him . Wit h simpler kin d of work,
this co-operation is reduced further, and we find, for example, this
arrangement between a father and his unmarried son, both straw-hat
makers: the father took half of his son’s earnings, and in return main•
tained him , and would in due course pay part of his bride-price (he
would be unable to bear the full expense). But i n the majority of cases
this rudimentary co-operation, too, disappears, and even unmarried
sons work entirely on their own.
One might wonder whether these ‘individualistic’ crafts have not, at some previous stage, shown a larger measure of co-operative organiza• tion . Remembering the weakening of guild-organized crafts under the pressure of modern conditions, we might, in fact, ask whether the in • dividual crafts as we find them to-day do not represent merely a final stage of that incipient disintegration. Thi s is indeed the case in one of the industries practised to-day as individual crafts—leather-work. Leather-work involves specialized technique and skill in much higher measure than do the other individual handicrafts. But it also used to involve comparatively large-scale co-operation of a group of workers, in the sense of an efficient division of labour, whic h was meant to ensure the execution of large and frequently urgent orders. A guild-head, called Magaji, organized and controlled group production; he also acted as the official intermediary between the craftsmen and the king , who was, again, their most important customer: for the Bida leather-workers supplied a large proportion of the militar y equipment of the Nupe army, such as saddlery and leather shields, sword-sheaths, and quivers for arrows.
Th e individual craft whic h is most a product of modern conditions is machine-tailoring. It introduces certain new features into the pro• ductive system of Nupe whic h are of special interest to the student of culture contact. Machine-tailoring does not demand long training or special skill—young men learn to use a sewing-machine in a week. Th e machine-tailor can reckon wit h a very constant demand for his services. Th e number of machine-tailors in Nupe has, in fact, greatly increased even durin g the periods which I spent in the country. In 1936, for example, I found six sewing-machines in Kutig i market, against three in 1934. Machine-tailors, like ordinary tailors, wil l regularly visit village markets (walking perhaps ten miles or more, carrying the sewing- machines on their heads), and appear in villages a fortnight or month before an annual festival, when their services wil l be greatly in demand.
u
29 0 INDUSTRIE S
They make as much as 1s. or 2s. on a market day; and shortly before the Great Sallah, for a week or ten days, they can reckon upon daily earnings up to 4s. Most machine-tailors do not buy their own material, but prefer to make up material brought by the customer, whom they charge a
‘wage’ (lada) of 2d. to 4d. a cloth. It is not surprising that the men who have taken up this most modern industry are mostly young men. They come partly from tailor families, but partly also—like many other ‘free’ craftsmen—from other professions and other social classes. Th e con• tribution of the ‘gentry’ is, again, considerable. Tw o of the five machine- tailors from Bida who are now working on the village market of Jebba Island are descendants of titled families.
You can buy a sewing-machine in Bida (second-hand or new) for £2 to £3 . Some young men have themselves bought their own sewing- machines with money made in wage-labour. More often the sewing- machine is bought by the father of the young man (or occasionally his mother), who by this purchase sets up his son in an independent trade— as blacksmiths or weavers equip their sons with tools and looms to start them on their independent work. Unlike these craftsmen, however, the father of the machine-tailor may provide the machine before his son is married, and expect him then to find himself the money for his bride- price. Th e son who is yet unmarried may hand over a half or a third of his earnings to his father—but this is not a rigid rule. Married sons work invariably on their own; an occasional gift of five kola-nuts (never more) to the father is an act of filial piety rather than a recognition of
‘business relations’.
Th e introduction of sewing-machines gives a new meaning to the question of maintenance of labour. Sewing-machines represent by far the most expensive means of production used in native crafts. They must occasionally be repaired, spare parts have to be bought, and eventually the whole machine will have to be replaced. Th e cost of maintaining these means of production seems more than balanced by the regular profits derived from the use of the sewing-machines. Th e question of obsolescence and replacement has not yet become acute. But already the problem of maintenance of labour in this new sense is felt. Blacksmiths and glass-workers, as we have seen, invest only a little money in their means of production—their main capital is labour. Weavers have to purchase looms and thus to invest a certain capital in tools; but specialized skill and training under experienced masters represent as much a ‘working capital’ as does loom or money. In machine-tailoring the money capital (the purchase money for the sewing- machine) represents the only investment of the craftsman; training and special skill are negligible factors. Th e money returns are, in fact, dis• proportionately high, measured by the traditional standards of skill or length of training. Here means of production and profit-bearing capital become one and the same thing. Th e machine must itself produce the profits out of which it will have to be replaced some day; and, through the medium of the machine, capital must reproduce itself, almost auto• matically. We discover, in a typical native home industry, the first traces of the modern industrial system.
MAT-WEAVERS ON THEIR WAY HOME FROM TH E FARM
A MACHINE TAILOR
INDUSTRIES 29 1
MEANING AND FUTURE OF GUILD ORGANIZATION
In examining the different industries of Nupe we have met with co• operative organizations of widely varying degrees of closeness and in• tensity. Th e individual crafts appear in this respect as one extreme, the zero point, as it were, of this scale of intensity of co-operation. What, then, are the conditions determining the degree of closeness in productive organization ? Or, more specifically: Wh y are certain native industries guild-organized, and others left to independent individual enterprise ? We can discover four conditions of the closeness of productive co• operation: (1) Th e nature of the productive technique and the measure of actual working together which it entails. In this respect the work of glass-makers and iron-ore miners must be placed at the top of the list, and the work of bead-makers or individual craftsmen at the bottom. (2) Special skill and experience involved in the productive technique. Th e glass makers, with their jealously guarded trade secret, again, top the list; the individual crafts represent the other extreme, of techniques which can be easily acquired by anyone. (3) Th e regularity of the demand for the goods produced by an industry, and thus, indirectly, its economic success. The work of glass-makers or brass-smiths presents one extreme, the work of bead-makers or weavers the other. (4) Th e comparative unimportance of originality and inventiveness on the part of the individual craftsman. In this respect brass-smith work shows the most ‘individualistic’, and blacksmith-work, the most closely integrated,
form of production.1
The balance which, in the actual case, is achieved between these four factors determines the final structure of the productive organization. Th e issue is, however, complicated by the fact that different factors bring about similar results. Thus both a high degree of ‘individualism’ (as among brass-smiths) and a low measure of economic success (as among beadworkers) tend towards a dissolution of the co-opera• tive organization. On the other hand, we find a certain closeness of organization visible in the tendency of craftsmen to work together in the same locality and perhaps in the same workshop, when most factors are pressing towards dissolution. Thus beadworkers of different families and households still tend to share a common workshop and to live to• gether in the same quarter although their technique of production does not necessitate actual working together, nor the economic success o their craft favour co-operation. This qualified co-operation appears to be due to the advantage of keeping in close contact with one another for the purpose of profiting from each other’s experience, and of keeping abreast of fashions and new styles.
Efficiency in group production appears, then, as the main motive for co-operation on a guild scale. But while the first and third condition refer to efficiency in the narrower, technical, sense of the word, i.e. ease, speed, and economy of production, the second condition reveals a more
1 Wit h the weavers we find an additional motive for closeness of co-operation in the tribal and cultural unity of these alien, ‘Nupe-ized , groups. We can dis• regard this type of co-operation, which is not limited nor peculiar to labour organization.
2Q2 INDUSTRIES
specifically social and cultural aspect. Efficiency in this sense means continuity of the craft and of craft knowledge; it means the ability to raise, through the medium of an hereditary, imperceptibly changing labour group, successive generations of craftsmen. If in these respects production is planned for an unchanging future, the agency of the thir d factor must expose it to the disruptive effects of time and change. For it may appear as if modern economic changes directed their most destructive attacks against this thir d condition of closeness and co-operation, whic h is dependence upon regular demand. Thi s influence is especially visible in the gradual disappear• ance of orders from the main customer of the country, kin g and court. Th e Great Sallah, once the occasion for numerous orders from the court, whic h kept the workshops of Bida busy for weeks, is now an occasion for making purchases at the stores that trade in European goods. At the Sallah of 1935, the first Sallah of the newly installed Etsu, the traditional royal gifts of kola-nuts were sent out, not in ornamental brass bowls of native make, but in newly bought enamelled teapots; the king’s bodyguard, newly equipped, wore, not Bida gowns and turbans or native-made straw hats, but red blankets and Trilb y hats; and many notables who woul d have appeared formerly on horses wit h costly trappings and gorgeous saddle-cloths, rode in on new, shining bicycles. Conspicuous though the competition of European goods appears, it is not the only reason for the decline of Nupe craft-guilds. Th e very organization of guild-organized crafts has contributed to lowering their resistance to outside pressure. Th e inertia inherent in the rigi d group- organization wit h its traditional standards of production prevents the craft-guild from readjusting itself to new demands wit h the same ease as the individual craftsman, who is free to change his mode of production and to gain success by new methods. Th e decision of the younger men among the weavers to abandon the traditional production and to tur n to the more profitable production of cap-making, the success of the most
‘ individualistic’ of all crafts, brass-work, are instances of this.
But readjustment involves in reality two different and independent processes: adjustment to modern conditions may mean both adjustment in price and cost of production, and adjustment wit h regard to the kin d of goods produced. A close examination of this question is of special interest as attempts to preserve and help native arts and crafts are to-day officially sponsored by the educational departments in many African colonies. In these attempts the second type of adjustment is frequently taken into account exclusively, the first being either ignored or artificially eliminated by turnin g native crafts into tourist and ‘curio ‘ industries whic h can command prices quite out of proportion to the native economic system. Tha t the Nupe crafts themselves have adjusted prices to demands is clearly shown in the general fall in prices which has taken place durin g the last five or ten years.1 Thu s mats of the best quality were sold, nine to ten years ago, for 3s.; in 1934 for 9d. and 1s.; in 1936,
1 In the two years I spent in Nupe country—two years which saw a partial economic recovery—I witnessed a corresponding slight rise in prices of native craft products.
INDUSTRIES 293 after a slight recovery, for 15. 3d. Straw hats ten years ago fetched
2s. 6d. and 5s.; in 1934 8d. and is. ; in 1936 5½d. and 1s. 3d.; embroidered
gowns sold ten years ago for £5 to £6 ; in 1934 they could be bought for
30s. or £2. But in the attempt to lower prices Nup e guil d industries
cannot go beyond cutting down profits; they cannot reduce the costs of
production—or else they must weaken maintenance of labour and de•
stroy the labour organization itself. One method of reducing what we
might call in a broad sense costs of production was to discontinue the
payments to family heads and guild-masters, as in the case of black•
smiths and weavers; another, to reorganize production on the basis of
individual paid labour, as in the case of the bead-workers. Another, to
save on the upkeep of the means of production : this was the case in
native iron-mining . Native iron-minin g is, in fact, the most convincing
instance of the limits set to the reduction in costs of production. Native
iron-ore minin g was, as we have seen, not destroyed by the competition
of a more satisfactory raw material; nor woul d the price of the native
product alone have been responsible for its disappearance from the
market. What was responsible was the relation between a low price and
a ‘hig h cost of production’ whic h could not be reduced. In this primary
industry in the strict sense of the wor d costs of production equal crude
labour costs, i.e. time and human energy. Whe n low prices made it
impossible for the iron-ore miners to make enough money even to
ensure sufficient food supplies, cost of production had to be reduced, i.e.
time and energy taken away from craft-work and devoted to agriculture.
We could study the effects of these elementary attempts at reducing
‘labour costs’, whic h were forced upon the craftsmen; only one way was
open to them, and that was to work their furnaces wastefully, irrespective
of gradual deterioration, even to the degree of allowing them to fall into
complete disrepair.
Yet it woul d not be true to say that the economic system of Nupe has never evolved more satisfactory methods to meet these contingencies whic h every economy must reckon with . I have said that formerly the Nupe craftsmen employed slave-labour on a considerable scale. Now , slave-labour provided an important mechanism to safeguard the flexi• bilit y of the economic system wit h regard to labour costs and profits. Slave-labour represents an investment of capital almost in the modern sense; for by investing capital in slaves one invested surplus capital in a profit-bearing concern, and the surplus made in times of prosperity could thus be made to serve as means of reducing labour-costs in times of stringency. Th e craftsman who bought slaves could not only use them as cheaper labour in his workshop, but could also put them to agricultural labour, to produce food for himself and his family—food whic h otherwise woul d have to be bought out of the profits made in the craft. Th e Nupe industries have lost the primitiv e for m of industrial capital, slave-labour, withou t having yet evolved the modern substitute, machinery. The y must thus fall short of the most vita l necessity in industrial productive organization, its adaptability to changing demands. These conclusions must not be ignored when planning the preservation or revival of African crafts and craft-guilds.
294 INDUSTRIES
We can now place the two systems of industry side by side and com• pare their respective assets and defects. Against highly efficient co• operation and mutual assistance embodied in the guil d organization we have to set the free play of competition allowed in individual crafts. Against the principle of continuity and planned training we have to set higher adaptability and larger scope for individual initiative and in • ventiveness. Broadly speaking, it cannot be doubted that uncontrolled,
‘individualistic’, industries stand a higher chance of developing produc• tion , possibly on new lines, and of achieving technical progress. More • over, efficient training and continuity of craft knowledge can, to some extent, also be assured in the narrower form of co-operation vested in the small labour-unit; this is shown in the village groups of blacksmiths and weavers. The question whic h we have asked at the beginning of this section thus stil l stands: Wh y has Nupe society chosen, of these tw o possible methods of organizing industrial enterprise, the method that assured group efficiency and continuity on the largest scale, and group efficiency rather than progress and individual initiative ?
Th e main reason lies in the political system of the country. Th e guilds are, as we have seen, buil t closely into the political framework. Le t me recall the main facts: the guild-head is the agent of the king in all matters concerning the group ; like the members of the nobility , he is the holder of an official rank, and is admitted to the nko, the King’ s Council. Th e guilds themselves enjoy a privileged position in the Nupe state in that they were exempted from tax and were assigned a social position superior to that of other commoners, peasants, or traders. Yet they were, in other respects, very much controlled; they were bound to their pro • fession and could not freely change it . The y were even bound to the locality in whic h they practised their trade and could not move from it withou t permission of the guild-head. If a village chief desired to have a certain Bida craft represented in his village, he had to apply to the guild-head in the capital; this happened some years ago in Kutigi , when the chief requested the guild-head of the carpenters in Bida to send hi m a craftsman to settle and practise in his village. Th e craftsmen, then, are the king’s craftsmen. No t only are they dependent on the king for their prosperity, but they are also controlled and supervised by hi m and his official agent, the guild-master. Thi s organization of the craft-guilds amounts to a ful l control of the political system over all the more im • portant industries of the country. We may even go further and say that this control was dictated by the needs of the state: based on constant warfare, committed to uphold the splendour of a huge court, the political
system has to guarantee a dependable, uninterrupted supply of all that is needed—arms, tools, clothes, saddles, as wel l as the many symbols of wealth and status. Upo n this dedication of the craft-guilds to the service of the state the seal is set by the discriminative social appreciation which the Nup e people extend to the different crafts.1 Th e noblest craft is that of the brass- and sword-smiths, for they, as an informant put it , are ‘the makers of arms’.
1 See p. 103.
INDUSTRIES 295
DIVISION OF LABOUR BETWEEN THE SEXES (II)
In none of the industrial activities described on the preceding pages women take any part, except, indirectly, in weaving. But there exist also certain special women’s crafts—again, ‘individual’ crafts—the technique of which is known only to the women and the proceeds from which are women’s property. We may note that the men are much more ignorant about women’s crafts than they are about the special activities of the women in connexion with agriculture and housework, i.e. cooking, palm- oil manufacture, and so forth. I could always obtain very reliable in • formation on cooking or shea-butter and palm-oil manufacture from men; but information obtained from men on, say, pottery, proved super• ficial and misleading. Often, in fact, the men refer you to the women rather than venture any information themselves. Woman’s craftwork is thus much more specifically and exclusively a woman’s concern. It is also more self-contained than cooking and similar activities, being hardly connected with the productive activities of the men. This is most clearly shown by the fact that in craftwork the women must produce their own means of production, unlike, for example, in shea-nut manu• facture, in which the women use roast-ovens built by the men, and that, for this purpose, the women have evolved special forms of com• paratively large-scale co-operation. At the same time women’s crafts are much less widespread than women’s other activities. Th e three crafts peculiar to women, indigo dyeing, pottery, and weaving (a special type of weaving), occur only in certain localities and social groups; there are large numbers of women in Nupe who do not practise any handicraft.
We have seen already that indigo dyeing by women is practised only in the few konii groups in Nupe country, where the men are the weavers and the women the dyers (see p. 235). Th e women’s craftwork goes beyond this partnership; the women dye not only the cotton-thread which is used later by the men in weaving, but also any kind of ready- made cloth, old and new, that is brought to them by outside customers. For this work they charge 2d. to 3d. per cloth. Dyeing is done individually, every woman having her own dyeing-pot and making her own dye. But the women of one household often help each other in this work. Among the women dyers in Kutigi and Sakpe we even find a co-operation which goes beyond the single household. Th e pots in which the dyeing is done are arranged on a platform outside the compound or part of the village in which the dyers live. When building this platform and putting up the dye-pots all the women of the locality work together—it has been said already that the building and beating of platforms is women’s work; the dyeing ‘platform’ itself thus represents a close analogy to the large work- shed in which the weavers work. This co-operation among the konii women does not represent a specific and, as it were, additional productive organization, for it takes place within the existing framework of a closely integrated kindred, formed by these craftsmen and women of alien tribal origin. But here the interesting sociological question arises how this limitation of a woman’s craft to a local and kinship group is achieved and
296 INDUSTRIE S
maintained in a patrilineal and patrilocal society. Th e explanation is supplied by the endogamous organization of the konu groups, whic h marry mostly into other konu families of the same or a neighbouring village. In one or tw o cases in whic h intermarriage of a konu woman wit h a Nupe man has taken place, the wife took her craft wit h her and is practising it in her new home, but now as a completely individual craft, withou t the help and co-operation of her fellow craftswomen.
Pottery is widespread in Nupe, both in Bida and in the villages. Certain places are famous for their pottery, such as Jebba Island, Baro, Badeggi, and also Bida. Th e best clay for pottery is found near the river, and this may explain the superiority of the pottery produced in these localities, whic h are all situated on or near the river. Bu t there exists also a regular trade in potter’s clay; women from Badeggi take the excellent clay whic h is found in that district to Bida and sell it there to the local potters. Other inland places, even rich , busy villages such as Kutig i and Mokwa , have no pottery at all. But Mokw a is only ten miles from Jebba, the great pottery centre, and Kutig i market, on the main trade-route to the west and south, is amply supplied wit h pots from everywhere. Facilities for trade and communication thus make for specialization, and tend to eliminate inferior local production—a point to whic h we shall return in another context.
In Bida, too, we find a certain specialization in pottery. Clay lamps and the small pitchers whic h the Mohammedans use for their ablutions are made everywhere in Bida. Other kinds of pots are made in one efu only (the efu of the masagd, whic h is situated close to the river). These are large, wide pots used for cooking; tall pots, shaped somewhat like a column, whic h are in reality three round pots placed on top of each other and formed into one large receptacle—Nupe men and women store their clothes in these pots, whic h are called etso; and small roun d platters which , broken into small pieces, are used for making tessellated floors. We note that the women are here producing the ‘raw material’ for another typical woman’s work, the making of floors. Th e making of these pots and platters is generally recognized to be much more difficult and to involve more special training than the manufacture of clay lamps and pitchers. Th e secret of the technique—for it is, in Nupe eyes, a secret—is only possessed by the women in three compounds in the glass-workers’ quarter.1 Mothers teach their daughters this technique, and we are, again, faced wit h the question how to account for the limita • tion of a technique whic h is transmitted in the female line to a local group formed according to the rule of patrilocal residence. Th e answer is more or less the same as in the case of the women dyers. Th e women potters from the masagdfu mostly marry into families of the same or a neighbour• in g efu. Th e girls who occasionally marry into the masagd quarter fro m outside are instructed in the art of pottery by their mothers-in-law. Th e daughters of the women potters, however, who marry outside (I recorded
1 This technique involves the use of a crude potter’s wheel—the only known case in Africa. Of the different groups of women potters in Nupe only those in Bida and Badeggi use this device. For the description of the technique I must refer to a future publication.
A N OL D POTTE R WOMA N
TH E WOMEN’ S LOO M
INDUSTRIES 297 three cases) abandon the craft altogether. Pottery also involves a certain
co-operation between the women whic h goes beyond the single house•
hold . Th e pots are made in the house, individually , but the burnin g of the pots takes place in an open space outside, and in this the women from different compounds work together, buildin g the fire of grass and fire• wood and using it jointl y for the burnin g of their pots. Thi s co-opera• tio n neither expresses kinship and triba l unity , as in the case of the konti women dyers, nor is it based on an organized occupational unio n as in the case of the handicrafts of the men. Rather is it an occasional, free co-operation for the sake of expediency whic h is suggested by the particular technique of this handicraft.
Th e question of the limitatio n of a woman’s craft to a particular section of the population reappears, wit h a different meaning, inweaving. Th e women’s weaving is done on the upright broad loom (21 in . wide), which , though produced to-day by the Bida carpenters, appears to have had its origi n in the south, in Yoruba country, where I have also seen it used as a woman’s loom. Thi s technique of weaving is said to be com• paratively new, and not traditionally Nupe, and may indeed have come to Nup e from Yoruba in the times of the Yoruba wars, throug h the medium of Yoruba wives and slaves of the Nup e nobility . For women’s weaving is restricted to the womenfolk of the upper classes of Nupe society; only wives and daughters of the nobilit y or the ‘intelligentsia’ possess a loom and practise the craft. There is, in fact, no house of the upper classes i n whic h you wil l not find several of these looms being used by girls and women. Conversely, looms and craft are completely absent in the poorer man’s home. Th e technique is handed on from mother to daughter. In very few cases only was I informed that a wife had been taught weaving by her mother-in-law. Th e profit involved in this work corresponds to the average profit of the men’s handicrafts. One bi g embroidered cloth, 2 yards long, is sold for 10s. to 15s., the material costing 5s.; the weaving takes thirt y days (a cheaper, not em• broidered, kin d of cloth can be manufactured in half the time), and the profit per day is thus between 2d. and 3d. Thi s gives us the explanation for the limitatio n of this craft to a specific social stratum. Only women who have servants (bara and formerly slaves) to help in the housework and to ru n errands for them can find the leisure for this long and delicate work . But the restriction of this woman’s craft to one social stratum is only made possible through the existence of a unilateral endogamy— class endogamy—on the part of the women, wh o rarely marry beneath their station. If I am right , and these looms have been imported into Nup e from Yoruba, we have an interesting instance of class-restricted culture contact, that is, of cultural ‘borrowing ‘ whic h is at once fitted into , and made to subserve, the existing system of rigi d social stratifica• tion .
CHAPTE R XV I
FREE PROFESSIONS
T is not surprising that a rich and complex culture like Nupe should ossess its clearly defined artes liberates—its professional scholars, scientists, and artists, or, in a terminology more akin to the native con• ception, its Mallams, barber-doctors, and drummers and dancers. I will leave the discussion of the profession of Mohammedan scholars for the chapter on education, and deal here with the last two of these ‘free pro•
fessions’.
BARBER-DOCTORS
Tre e professions’ in the sense in which we are using the term here does not necessarily mean ‘individualistic’ professions or professions un• attached to any organization or co-operative unit. If what I have said in the preceding chapter about the meaning of guild-organization is correct, we may expect to find it developed also in the field of these native artes liberates. In fact, we do find it, adjusted to the new field of activity, in the barber-doctors of Nupe. We can no longer speak here of home industry; nor is the element of co-operation and mutual assistance the same as in the industries. But the other factors that tend to call into being a closely co-operative organization are clearly visible: we find again the necessity of long, specialized training, even to the degree of adoption into a trade secret; we also find the existence of a constant
‘market’ for the products and services of the craft; and lastly, we find a close relation between the group of craftsmen and the political system of the country.
As in medieval Europe and the Orient, the barber of Nupe,gozan, is also doctor and surgeon. His activities are numerous: he cuts the hair of men, women, and children; he shaves, the head and cuts the tribal marks of newborn babies at the name-giving ceremony. For a tenth of a penny he cuts finger- and toe-nails: a few barbers have specialized in tattooing. Th e barbers also dispense medicines (mostly worthless1 infusions and charred powders made of herbs) for every known disease; they perform
certain minor surgical operations—circumcision or blood-letting—and administer the turare, a potent inhalation of the smoke from certain medicinal herbs. Certain barbers know the secret of contraceptive medicines, others know cures against barrenness; others, again, are ex• perts in the treatment of lunacy. This far-reaching competency is the most characteristic feature of the craft of Nupe barber-doctors. Leaving aside the ordinary routine work of the barber and certain simpler opera• tions such as the cutting of tribal marks, we may say that every barber
1 This judgement is based on an analysis in the Wellcome Research Labora• tory in London, for which I am greatly indebted to this institution, of certain drugs made by the Nupe barber-doctors which I had brought home from Nigeria. Th e discussion of the more specifically medical questions I must leave for a future publication.
GUILD-HEAD OF BIDA BARBERS
GROUP OF DRUMMERS
FREE PROFESSIONS 299 has his own speciality, his own expert knowledge, his cures, and his
special field. No wonder, then, that the Bida barbers are not of equal
standing: there are famous ‘ celebrities’, and there are also the smaller fry.
Th e public is aware of these distinctions. It is well aware also of the
special field of certain foreign barber-doctors who frequently come to
Bida to practise. Thus, for example, it is generally known that the
operation of removing the uvula of children—an operation said to help
growth and secure better nourishment—is performed only by Hausa
barbers. Th e more surprising, then, is the fact that this array of
specialists and special cures is not equally accessible to every citizen of
Bida or Nupe. For the barbers are organized in three independent
guilds, one in each ekpa of the town. Neither doctors nor patients ever
ignore these boundaries. Each guild, or guild section, is under its own
titled head, who as a rule has also a ‘second-in-command’ (who is ‘heir
designate’ to the guild-headship at the same time) to assist him in guild
work. In the Etsu Umaru quarter guild-head and ‘second-in-command’
are called Ndeji and Ndabaki; in the Etsu Usman quarter Tsadzo and
Ndamaraki\ in the Etsu Masaba quarter we find only one guild rank,
who is the present head of the guild at large, the Sokyara. This highest
guild office is taken in rotation by the three guild sections.
Th e Bida barbers, busy and prosperous, do not practise farming in addition to their main occupation; nor do sons of barbers ever follow any other profession. On the contrary, men of other professions may become barbers. I have recorded three recent cases: two were adults, one a farmer and the other a (Fulani) herdsman; the third was a young man, the son of a straw-hat maker; the two adults settled later as barbers in villages; the young man, who came from Bida, remained there to practise his craft. To learn the barbers’ craft proper does not take long: the apprentice works only a year under a master (without changing his habitation). Th e master, during this time, pockets the larger part of the apprentice’s earnings. Th e apprentice himself buys the instruments of his new trade—a collection of old-fashioned razors, costing about 4s. to
$s. But to learn the medical side of the craft requires (I was told) a five- years’ apprenticeship. One of the two men just mentioned had done this, and is to-day a practising ‘doctor’. Let me add that neither the ‘trade secrets’ of the barber-doctors nor the training in the craft involve any magical or religious conception and form of procedure. It is purely a craft technique.1
Th e special cures and forms of treatment are secrets of individual barber families. They are handed on from generation to generation and are the common property of all adults in a family (though not of the
1 Nupe society also possesses a certain group of ‘doctors’, or, more precisely,
‘medicine-men’—cigbeciii (cigbe means ‘medicine’)—whose craft is based on magical ideas and is closely linked with divining and the religious system of the country. I cannot describe here their organization more fully. I wil l only say that they practise mostly in the pagan districts, and that they are united in a loose4order’, acknowledging a common head, the Etsu Cigbeciii in Bida. To-day the modern conception of the barber-doctor is gaining ground also in the pagan villages. I found a significant effect of this transition in two Nupe villages—
‘medicine-men’ and diviners who were practising barbers at the same time.
300 FREE PROFESSIONS
larger group, the guil d section or guild) . Th e wor k of the barbers is organized on the familiar lines of efako and buca. Althoug h their work does not entail actually workin g together, a certain kin d of wor k is regarded as a concern of the family group as such and as falling under the rules of efako. Th e efako principle itself is interpreted in an unusual sense. In the work of the barbers orders or summons do not usually reach the individual barber directly, but come to hi m via the head of the household, who allots the various tasks to the different family members. Th e proceeds fro m all work carried out on these lines are divided among the whole efako group, in accordance wit h a certain schedule whic h takes into account the seniority of the different family members. Th e tasks whic h come under efako rule are the most highly paid jobs : the shaving of the hair and the cutting of triba l marks at the name-giving ceremony, and the circumcision of boys. Th e first task is performed at the occasion of an important household feast, and the barber-doctors rarely fail to benefit from the generosity that goes wit h festal mood. These simpler operations are executed by the younger men of the family, bu t the family head and the older men wil l also be present, as guests quite as muc h as in their professional capacity—which latter may reveal itself in no more than an occasional hin t throw n out to the young colleague. Th e barbers wil l all sit together, and in front of them there wil l be a bowl into whic h every visitor is expected to drop a coin, the largest contribution coming from the parents of the newborn child . Le t us follow the fate of a shil • lin g dropped into this bow l of the barbers at a name-giving ceremony in a Bida house: 2d. is retained by the grandmother, who held the infant durin g the operation; 1d. goes to each of the five junio r members of the efako group (including the young man who performed the operation);
1 \d. each to tw o older men of the family, the family head and his younger brother; and 2d., finally, to the head of the guil d section.
Al l other work of the barbers is regarded as buca: ordinary haircutting and shaving as wel l as medical treatment.1 Th e latter, more lucrative, work is a prerogative of the older men, who alone know herbs and drugs and can perform the turare. Th e younger men wil l be taught these craft secrets when they reach a certain age. An d thus another principle of family work is realized in the organization of the Nup e barber-doctors, namely, the balance in favour of senior family members whic h we have observed in the organization of farm-labour and native industries.
We have seen that in the distribution of efako income a certain con• tributio n goes to the head of the guil d section. This , then, is a first instance of economic co-operation of the large professional unit . Thi s co• operation is visible also in another, external, feature of this professional organization. In each of the three day-markets of Bida there is a special barbers’ ‘hall’ , a spacious open hut , where all the barbers (specially the family heads) meet in the afternoon, and sit and wait for clients and summons. Thi s hu t is buil t and kept up fro m a common fund. Th e existence of this common fund is itself evidence of a close co-operation in the financial field. Int o this fund , whic h is called dashi, every barber
1 The barbers are usually paid in money; but when the Bida barbers tour the country villages they often take firewood or food in lieu of payment.
ON E OF TH E ETSU’S KAKATI PLAYER S
A PROFESSIONA L SINGE R WIT H HE R BAN D
FREE PROFESSIONS 301 pays a small amount of money at regular intervals; the fund is adminis•
tered by the guild-head; he deducts from it money for possible common
expenses, and pays out the rest i n full to every member of the guild
section in turn , after every pay-day. Thi s financial arrangement is fairly
common in Nupe, specially among traders and labourers, and we shall
discuss it in more detail later. Th e barbers, however, are the only pro•
fessional group whic h has embodied it in its professional organization
and utilizes it as a system of mutual financial assistance vested in the
craft-guild.
Th e guil d organization is restricted to Bida town . In the villages, even in the larger places, one or two barbers are sufficient to meet the normal demand. Besides, the medicine-man of the pagan districts woul d con• stitute a serious competition to the Bida doctors. But so far as barbers and barber-doctors are wanted in the country, the organization in the capital supplies them, either permanently or temporarily. Certain Bida barbers, for example, have settled in villages; others tour the country regularly, stopping in this or that place for longer or shorter periods. Even in these tours of the country districts round Bida, the barbers keep to the territory of ‘their ‘ royal house, i.e. the royal house on whose land in Bida their home stands. Although it clearly cannot be said of the barber-doctors that they ‘serve the state’ in the same way as blacksmiths or leather-workers, this arrangement is a striking evidence of the fact that their association wit h the political structure of the country is as strongly marked, and that they, again, are in certain respects the ‘king’s crafts• men’.
DRUMMERS AND DANCERS
Th e drummers in Bida, true to the more ‘advanced’ urban conception of art, are individualists; in the village, where the art of drummin g be• longs not only to the recreational side of social life, but also forms part of the organization of ritual and ceremony as well as production (we re• member the part drummin g plays in the farming egbe\, it is less an ‘art’ in the modern sense of the wor d than a ‘craft’, workin g for well-defined institutionalized demands, and charged wit h responsibilities of a graver order. Drummin g in the village thus moves in the framework of a close organization on guil d lines. In every larger Nupe village we find a special group of drummers and musicians who , though belonging tp different families and houses, yet recognize the authority of a common head, the Etsu Dza, the Kin g of the Drummers. Certain other, sub• ordinate titles, are distributed, in guil d fashion, among the other members of the craft. Th e art of drummin g remains in the family and is handed on from father to son. Yo u may see small boys practising drummin g on littl e toy-like instruments under the eye of the father. They wil l go wit h hi m wherever he is called to perform, watching hi m and, when they are sufficiently advanced, accompanying hi m on their own drums. Sometimes boys from other families desire to become drummers. Thei r father wil l then get in touch wit h the head of the drummers, and give hi m his son as apprentice. Th e young apprentice wil l visit the drummer’s house every day for his lesson. I n the beginning
302 FRE E PROFESSION S
he will be allowed to use the drums in the house; later he will buy him• self a drum of his own. No payment is taken for the tuition; but when the novice later performs at ceremonies or feasts, he will send his former teacher a small gift of money or food. The drums are made by the wood• workers on the river, who are alone in the position to obtain the suitable wood. One large drum costs 10s. Th e drummers themselves cover the drum and provide the goatskin for the diaphragm.
But all this is a thing of the past. To-day apprentices are hardly ever taken. Th e drummers, like so many other Nupe craftsmen, complain bitterly about the bad times. Formerly, they told me, they used to be paid 6s. to 8s. a night; to-day they receive 3s. for the night or, for one
‘song’ (played, for example, at a family feast), one kola-nut. This decline appears to be due to two factors: first, to the impact of Mohammedanism and the gradual disappearance of rituals and ceremonies in which drum• ming formerly played a prominent part and the songs of which were stored in the memory of the village drummers; and second, to the spreading of fashions from the capital, and the competition of the more
‘refined’, famous Bida drummers and musicians. Th e organization of the village drummers has survived only in the names of the guild ranks. Th e drummers, who once devoted only their spare time to farming, are to-day mainly farmers and carry on their old profession only as a side• line.
In Bida drumming and music were never tied to the narrow scope of limited traditional occasions; nor was the development of the artistic style rigidly hemmed in ritual. Both the profession and the style were allowed to develop freely, and the result is an impressive variety in artistic production and in its social counterpart, the organization of the craft. Th e Bida practice also introduces a completely novel element into the tradi• tional craft, namely, music and dancing executed by professional women dancers and singers. A number of performers, drummers and women singers (who rarely come of one family or household) associate with one another, and form themselves into a ‘band’ or ‘company’. In this form they become known, are invited to perform at feasts and similar occasions, and also tour the country, sometimes even going abroad to other large towns in Nigeria where Nupe music always finds an appreciative public.1
In constant co-operation the ‘company’ builds up its special ‘repertory’ and evolves its individual style. There are several such ‘companies’ in Bida, each with its special repertory, style, and technique. They are known by the name of their leader, or by the name of the style in which they have specialized. Thus we find the group of the drummer Sogba, or the group of the woman singer Fati, or a group called zawdro (lit. ‘new man’), so called because it performs in a style only developed since the days of the ‘new men’, i.e. the British. One or the other ‘company’ is known to be patronized by a certain noble family and will be summoned at every occasion when the services of drummers and dancers are re• quired, a name-giving ceremony, a wedding, or an age-grade festivity. Certain drummers even enter a kind of bara relation with a nobleman
1 I met a company of Bida drummers and women singers, whom I had known in Bida, in Lagos, where they were ‘on tour’.
FREE PROFESSIONS 30 3
whose patronage they seek. Often the patron may order special com• positions from ‘his’ drummers, for example, a personal drum signal, called take, for a newly acquired nobility rank, or a song of praise on himself, or a friend or relation.
Between these different groups there is no co-operation. Each group is, economically, a closed unit. Al l earnings are divided equally among the members. This division of the income has no bearing on factors of family or household organization. Individual group-members may con• tribute to the expenses of their own families; but the ‘company’ is purely a professional association, to which the categories efako and buca do not apply. Professional and family concerns are, in fact, kept strictly sepa• rate, occasionally the two overlap, and we find a woman singer and a drummer of the same ‘company’ who are man and wife; but just as often a drummer has a wife, and the woman singer a husband, who are not members of the ‘company’, nor even of the profession.
Drumming in Bida is a full-time occupation (sometimes a drummer practises a little farming besides, but this is very infrequent). Each company regularly adopts new members as old ones retire. The novices serve their apprenticeship as minor performers of the ‘company’, until they become fully fledged members of the craft. Sons of drummers mostly follow the paternal profession; but many singers and dancers do not come from families whose women have practised this profession previously. Music and dancing in Bida is a vocation rather than a pro• fession. The free competition of talent and individuality rather than family tradition determines professional success, and thus the profes• sional association in which the craft organizes itself. Indeed the profes• sion of Bida musicians and drummers represents the only instance in the economic organization of the country of a free association purely on a business basis, and without the support of the framework of kinship and hereditary tradition.
With regard to the profession of drummers and musicians town and country have exchanged roles: we find the closed and hereditary profes• sional group (though of much smaller scale than guild organization proper) in the village, and the detached individual groups of craftsmen in Bida. But the principle of hereditary craft organization has remained essentially the same. We need only read village chief for king, and village community for state, and we understand the new situation. For the drummers, as we have seen, are as essential to the corporate life of the village as are the craftsmen proper to the political life of the state. In town-life, on the other side, drummers and musicians have no traditional duties assigned to them which concern the common interests of the group at large. Indeed the more varied social life of Bida, and the appreciation of new fashions and special styles typical of the cosmo• politan town, foster that spirit of originality and inventiveness which here as in all craft-work counteracts efforts at a closer co-operative organization.
CHAPTE R XVI I
INDUSTRIAL LABOUR
HE influences of modern culture contact whic h we have seen at work in almost every field of Nupe production gain their fullest expression in this most recent native occupation. Th e employer is the Bida Native Administration ; the employees are recruited from certain specific sections of the Nupe people, and the nature of this selection is
deeply significant of the structure of the society.
We must distinguish two different categories of paid industrial labour: skilled permanent, and unskilled temporary, labour.1 Skilled labour is employed in the workshops of the Native Administration Public Wprks Department, and comprises blacksmiths, mechanics, carpenters, and bricklayers, i.e. occupations whic h (wit h the exception of the brick• layers) are closely associated wit h the traditional native crafts of the country. Temporary unskilled labour is employed partly in addition to permanent labour in the N.A . Workshops, but mostly in occasional work demanding a large body of unskilled labourers such as construction work of various kinds. Th e ter m ‘temporary’ must be understood in a very loose sense; men who have been working as ‘temporary’ labourers, off and on, for ten or fifteen years figure prominently on my lists. Some move from place to place, wherever there is a chance of finding employ• ment. As there are large numbers of alien, non-Nupe, labourers among those engaged in temporary work in Nupe country, so we also find Nupe workmen in other districts and provinces of Nigeria. A few of these labourers have never even had another, more or less permanent, profes• sion, but have adopted their present occupation in earliest youth. None of them, however, are skilled and specially trained labourers. By far the largest proportion of temporary labourers go home and return to what• ever their original profession was when their temporary employment comes to an end. In the sense, then, that their work as labourers con• stitutes an alternative to some other original occupation, they can all be classed under the heading of ‘temporary labourers’.
Th e data on whic h are based the statistical tables whic h follow were compiled in and near Bida in 1936. It was a fortunate coincidence that at the time a large buildin g programme was being carried out—the con• struction of new office buildings in Bida, and the rebuilding of a big bridge on the Badeggi road—in whic h a large body of labourers was employed. My census of temporary labour was taken among the work• men who were employed in this work. I have placed bricklayers in the same category wit h temporary labourers, although, technically, they represent a group of skilled, permanent, workmen ; but the technique of
1 To be quite precise we should distinguish four different classes of labour: permanent and non-permanent, and skilled and unskilled. As, however, all skilled labour is permanent, and all non-permanent labour unskilled, and as unskilled permanent labour represents only a small minority (8 individuals) it is simpler to deal with the two sub-divisions as one.
INDUSTRIA L LABOUR 305 their work is in no way connected wit h a traditional trade or craft, whic h
means, that they all had to abandon another, original, occupation (either
of themselves or their families) and that their present wor k constitutes
an alternative profession in the sense just explained.
TABLE 1
Temporary Labour
Length of employment as labourers
Under ‘N o alternative occupation’ I have listed in brackets the labourers who are studying the Kora n in their spare time , bu t withou t intention of practising the profession of Kora n teachers. Th e heading
‘seasonal.labour’ refers to farmers who, for various reasons, seek work as labourers only in the dry season and return to their farm-work in the rains. It is natural that their number should be much larger among people from the country than among Bida men. Al l peasants, inciden• tally, who have become labourers durin g the last two years belong to this class of seasonal labour. Our figures further show that their number is greatly on the increase. Labourers pure and simple, as migh t be expected, come almost exclusively fro m Bida. Th e only exception is a man from Badeggi, the village on the urbanized railway, twelve miles fro m Bida ; he is also the only country man who has been a labourer for over ten years. Under ‘alternative occupations’ I had to list a great many different occupations; Table I I I wil l show the various original, o r
‘alternative’, occupations o f temporary labourers. I n Table I I w e shall examine the professional groups from whic h skilled permanent labour is recruited.
TABLE I I
Permanent Labour
Occupation of the labourer’s family
Occupation Guild crafts
Individual
of the labourer Same Different crafts Farmers Mallams Traders Nobility Total
Carpenter s 6 1 1 1 9
Blacksmith s 5 2 2 10
Tota l 11 1 1 3 3 19
Th e large majority of skilled workmen come, as we see, from families whic h have practised the same crafts on indigenous lines both in Bida and in the villages. Th e few ‘outsiders’ appearing in this group come mostly from families wit h no specific occupation, such as Mallams, and nobility .
Th e largest contingent of country people among the labourers is, naturally, composed of farmers and farmers’ sons. Comparing Table I I I
x
3o6 INDUSTRIA L LABOUR
TABLE II I
Temporary Labour
Guild
Home Farmers craft
Indi•
vidual
craft Mallams Traders Nobility
No altern. occupa•
tion Total
Occupation of father
Alternative occupation
Bida 4 1
Country 10
Total 14 1
Bida 1
Country
13 9
6
13 15
2 0 8
2
1 11
2
1 13
39
18
57
5 38
1 19
of labourer
Total 2 0
1 2 0 10
6 57
with Table I we can see that of the 16 farmers who have sought work as labourers only 1 has abandoned farming permanently for labourer’s work, while 15 remained seasonal labourers. Turning to the Bida con• tingent we see that among the different professions which contribute to temporary labour individual crafts come first and Mallams second—the most loosely organized occupational groups thus contribute most heavily to occasional labour. The occupations of the families of which these labourers come appear in a slightly different order of importance : individual crafts are, again, first, with ‘nobility’ second and Mallams third. ‘Nobility’ means here mostly the lower political ranks, above all
‘household’ and bar a ranks. Their share in the contribution reflects the great political and economic changes that have deprived many members of this ‘leisure class’ of a livelihood mainly derived from warfare or political attachment.
But what are the incentives which led the men from the other occu• pational groups to give up their previous occupation or to undertake the work of labourers in addition to their ordinary profession ? This question is answered by the data of the last table.
TABLE IV
Married Unmarried Tax as reason
Permanent labour , 9
Temporary labour ( ; 9
Age 20-5 . . . . 4
12
13 4
9 5
1 0
8
4
Th e majority among permanent as well as temporary labourers is un• married—although many of the men, as we can see, have reached, and partly passed, the age at which it is normal for the Nupe man to marry. The work as labourers in their case was undertaken with the intent to obtain money for the bride-price, a plan which did not always material• ize as our figures show, especially in Bida. For there are many other uses fcr money, and some, as we shall see later, may interfere with the laudable intention of working for one’s bride-price. In a number of cases the reason for becoming a labourer was to obtain money for tax,
INDUSTRIA L LABOUR 307 either for oneself or for one’s family. One young labourer from the
country gave his father 5$. out of his weekly wages of 7′. to put by for his
bride-price; another kept 2s. to 3s. a month for himself, and gave the
rest to his father. Many Bida labourers, on the other hand, used the
whole money for themselves; some explained that they had become
labourers because there was no money in their old profession, others
because they wanted to be independent and because their fathers or
elder brothers could not give them as much money as they liked to have.
To summarize: as one would expect, paid labour, especially temporary
labour, does not represent a ‘real’ profession in the sense of an occupa•
tion involving special training and selection of workers. It is essentially
a stopgap occupation chosen to supplement or replace a failing normal
source of income. Owing to the specific composition of Nupe society
the new profession is being adopted by certain social groups and strata
more than by others. Yet in some respects paid industrial labour must
be regarded almost as a modern continuation of forms of occupations
which, in the traditional society, bore the same supplementary and stop•
gap character. As such it appeals strongly to the individual craftsman,
to Mallams, and baray who depend for their livelihood on similar forms of unskilled and partly temporary work. It differs from these other occupations, however, in that it is carried out on the road and in the workshop yard instead of in the home, and that it represents heavy manual labour, and not light work which it is easy to combine with
a dignified social status. Th e labourers, unlike the individual craftsmen, can no longer pretend that thfcy belong to the leisured class. If they receive better and more regular pay, they also lose caste. This explains why of the many hundreds of men who, once members of the privileged classes, are to-day men ‘without occupation’,1 gaining a precarious livelihood from occasional work, only those coming from the lowest grade of the ‘upper classes’ have chosen the profession of labourers.
In the case of the peasants we can no longer speak of any inherent congruity between their work and the new temporary occupation. To them, especially to the peasants who do not practise a handicraft beside agriculture, paid labour in addition to farm-work,aformof labour,more• over, which keeps them for months from their homes, must appear as something quite new and unprecedented. If to-day paid labour is gradually attracting the peasant class as well, twenty or thirty years ago the (very prosperous) Nupe peasants had to be coaxed by all available means to undertake temporary paid labour, for example, in railway con• struction. The more ‘sophisticated’ Hausa and Yoruba were already then flocking in large numbers to work of this kind. Missionaries who knew Nupe country at that time told me that it was often impossible in the peasant districts to obtain even porters to carry loads; the promise of money did not attract them sufficiently to make them leave their homes even temporarily, to upset their whole routine of work or forgo their well-earned leisure. In certain backward and prosperous districts this
1 An unofficial occupational census of Bida lists 1,130 men under the class ‘no occupation'”. They belong without exception to the titled leisured classes of the Emirate.
308 INDUSTRIA L LABOUR
attitude has still survived: when I trekked through the country round Sakpe in 1936 I found myself in the same predicament, and although it was in the dry season and farming was at a standstill, I had eventually to agree to women carrying my loads as men were not available for this task. That the appeal of paid labour as a seasonal occupation for peasants is rapidly gaining ground cannot be doubted. Th e stopgap occupation can, in fact, be said to satisfy a definite want in present-day peasant economy.1 Th e strength of its appeal still varies greatly locally. Th e large majority of the peasants whom I found employed as labourers in the construction of Badeggi bridge were local men or men from the neighbourhood of Bida. Of 19 labourers 14 came from Badeggi itself,
3 from hamlets near Bida, 1 from Kacha, 1 from a village farther afield (Jima), and none from Trans-Kaduna. Th e local selection does not merely reflect varying degrees of progressiveness or ‘sophistication’; it also reflects the varying urgency of the wants for the relief of which paid labour is undertaken. The nature of these wants we shall discuss in detail later.
1 To-day many Nupe peasants are ready to try any kind of paid labour, but wil l abandon it quickly if it falls short of their expectations of a regular and sub• stantial income. Thus the Nupe of Leaba tried their luck in gold-washing, which is carried out on a considerable scale in that district. The y had heard from Hausa labourers that one could make as much as 9d. or 1s. a day in this work. They had not realized, however, that the payment was not per day, but depended on the amount of gold which the individual worker can produce. As the Nupe were rather clumsy and slow, unlike the experienced HaUsa labourers, their pay was small and irregular, and, gravely disappointed, they soon gave up gold-washing.
CHAPTE R X V I I I
WEALT H
H E examination o f the proble m o f labou r ma y b e fittingly con•
cluded wit h a discussion of the role played by wealth , free capital,
i n Nup e economy. I t i s i n keeping wit h Nup e productio n i n general, wit h its emphasis on individua l propert y an d mone y profits, tha t th e concept o f wealt h should b e clearly pronounced. Th e principl e o f accumulation of wealt h as free capital pervades Nup e economics and , in particular, serves to define the economic (and , indirectly , social) strati • fication of th e society.
MONE Y CAPITA L
Th e appreciation o f wealt h figures prominentl y i n th e educational conceptions of the people. Youn g boys of 8 or 10, whe n the y are give n a separate buca plot to farm , are at the same tim e taugh t the value of mone y income an d mone y capital. Mother s wil l help the m t o marke t thei r small produce for money , o r wil l bu y thei r produce directly fro m them . An d fathers an d mothers, as we have seen, equi p thei r childre n wit h a first diminutiv e workin g capital, a penny or two , to start a profitable trad e (see pp . 286-7 , 33 0- Thes e profits enable boys and girls to defray the expenses of the social obligations incumben t upo n this tender age, namely , contributions to the costs of the age-grade associations of whic h the y are members . Mone y expenses, partl y involvin g the accumulation o f reserves, accompany Nup e me n an d wome n all throug h thei r lives: mone y to meet the costs of kinship rituals—birth , weddings , and funerals
—mone y for the maintenance o f labour, the purchase o f tools, &c , o r for personal wants and the luxuries of life. Mone y also frequentl y forms par t o f the inheritance left b y me n o r women .
Far m produce o r gifts i n kin d i n general cannot full y replace mone y expenditure . Thi s is shown clearly in the nature of the universal gift in Nupe—kola-nuts . Gift s of kola-nuts are mor e or less regularly ex• changed between friends an d relations; a gift of kola-nuts serves as the traditiona l announcement, and invitatio n at the same time , extended to kinsfolk at the occasion of forthcomin g kinship ceremonies. Gift s of kola-nuts also represent the traditiona l expression of political allegiance. And , wit h the insignificant exception o f the kola-nu t growers, the donor o f th e gift has t o bu y the kola-nuts o n th e marke t for money. Th e largest single expenditure of Nup e economy, finally, bride-price , is pai d entirely i n mone y (see th e discussion o f bride-pric e i n Chap . XX) . I t involves, moreover, accumulation o f mone y reserve, or, i f the bride – price is pai d in instalments, as it frequentl y is, careful plannin g of money expenditur e over a considerable period .
Althoug h none of the larger forms of traditiona l money expenditur e wer e ever replaceable by gifts or contributions in kind , it is also tru e that th e practice of mone y expenditur e has been increasing wit h the spread• in g o f cultura l influences fro m th e town . I t i s natura l that Bid a society,
31 0 WEALT H
with its large population of non-farmers, should have developed the practice of money expenditure more extensively than the villages, and adopted money expenditure in certain instances in which, in the village, gifts and contributions in kind were the custom. Th e universal practice nowadays of giving, and accepting presents of kola-nuts, for example, appears to have spread from the town, and to have replaced gifts of food or sorghum beer. Of similar changes in other fields we possess even more concrete evidence: thus the name-giving ceremony or funeral feast involves to-day contributions in money by guests and relations; in Mohammedan districts it is also customary to distribute alms (again in money) among the local Mallams. In the traditional form these kinship ceremonies implied gifts in kind, i.e. contributions of food and sorghum beer for the feast. Formerly, also, every household used to produce its own beer; to-day the manufacture of beer is limited to certain house• holds so that the other families are obliged to buy it for money. More• over, sorghum beer has been replaced in the town by palm-wine, which, as palm-kernels are largely imported from districts outside Nupe, must again be bought (see p. 234).
It is necessary to emphasize that this change represents a change of customs and fashion, and a change in the form of expenditure, rather than a change, or an increase, of expenditure itself. We shall, in fact, see later that the modern Mohammedan burial, for example, may come cheaper than the pagan ceremony, although in the latter contributions and gifts were largely in kind. In other words, the symbols of wealth are chang• ing and becoming more unified, being adjusted to the symbols of wealth which have been adopted and standardized by the social and cultural ilite of the country.
How, then, are money reserves and wealth in general accumulated? One method, which is widespread in Nupe, is to hoar<J«raoney by burying it. Sometimes valuables, above all, precious beads!’, take the place of money, and are buried or concealed otherwise in a safe place. In a number of court cases over inheritance the question of discovering the hidden wealth (money or beads) of the testator played an important part. Another illustration of the practice of hoarding money came to my knowledge in connexion with inquiries into the continued use of cowrie- shell currency.1 It is generally known in Nupe country that when cowrie-shells were finally replaced by English money and the former currency declared worthless many people buried their cowrie-shells in the hope that, some day, they might become legal tender again. Once I asked certain informants in Kutigi if they could show me any cowrie- shells. They said that they had none, and that cowrie-shells were no longer to be found in the country. I said that I was prepared to pay money for them, whereupon large baskets full of cowries suddenly turned up, and, the rumour having spread that the white man was buy• ing cowries, I had to do my best not to be inundated with the suddenly rediscovered currency.
Another method of accumulating capital is the investment in interest-
1 See ‘ A Ritual Currency in Nigeria—A Result of Culture Contact’, in ‘Notes and News’, Africa, 1937, 3.
WEALT H 31 1 bearing enterprises (interest’ understood here in the widest sense). Th e
nature of the enterprise varies with local conditions, and differs above
all in town and country. Peasants mostly invest their money in live•
stock (see Chap. XI) . We have seen that Nupe inventiveness is always
on the look-out for new opportunities promising a higher ‘interest’,
even in the narrow field of stock-breeding as they know it : we have
mentioned the buying and selling of ostriches, and the isolated experi•
ment in horse-breeding of one village chief—incidentally the chief of
a typical backwood village, Sakpe. In the town, and in the large trading-
centres of the country, money is invested chiefly in trade, even by men
and women who do not personally engage in trade. They would trust
a certain sum to a trader of their acquaintance for a particular purpose,
e.g. to help him finance a trading expedition to Kano or Bornu. Th e
trader will use their capital together with his own, and will return it to
them increased by the profits which it has brought when his business
has been successfully concluded. Th e trader, of course, also derives
a certain profit from these investments, as they enable him to enlarge
the scope of his enterprise and to secure more favourable prices. Man y
most highly placed notables in Bida, not excluding the king himself, use
this profitable method of investment.
Certain traditional and formerly highly important methods of invest• ment, above all war and slavery, have disappeared under the new political order. I have described in an earlier chapter the contributions of the houses of the Bida nobility to the wars and military exploits of their country and have also discussed slavery and slave trade. It remains to discuss the closely related traditional form of investment in ‘ human capital’ which has become a thing of the past only quite recently, namely, money investment in human pledges (sofa, in Nupe).
PAWNING
Besides the pawning of human pledges, ordinary pawning—pawning of valuable goods—also occurs. My information is that beads, costly gowns, a sword, or saddle, all things of silver or gold, are frequently pawned as security for a loan. No interest is paid on the loan itself, and the transaction is regarded purely as an act of friendship. I have, how• ever, not discovered any actual instance of this practice. My impression is that it is not, and has never been, very widespread. Pawning of people, on the other hand, has grown to enormous dimensions. When pawning of human pledges was finally suppressed in Nigeria, in 1933-4, 3,000 pawns were released in Bida Emirate alone.
As slavery reflects the insecurity of the political rights of the con• quered subjects of Nupe kingdom, pawning reflects the economic in• security under which a large proportion of the population of the Emirate lives. Th e lack of flexibility, more particularly, in their economy tends to foster such radical means of escape from financial difficulties. Under the pressure of unexpected demands or certain imperative heavy ex• penses which cannot be matched by normal efforts—payment of bride- price; a heavy fine; money with which to bail a family member or bribe
CHAPTE R X I X
ECONOMI C BALANCE : EXCHANG E
N th e discussion of economic balance we approach a grou p of questions whic h moder n economic theor y regards as the principl e subject-
matte r o f economics, viz . the relatio n betwee n means whic h exist i n some degree of scarcity an d thei r alternative uses for th e satisfaction of needs an d wants. W e shall examine ‘balance’ unde r tw o aspects: first, th e relation betwee n means an d thei r uses in a wide r sense, as they exist an d are balanced i n th e economic system o f the society a t large. I n this sense economic balance can be equated wit h exchange an d exchange- mechanisms. Secondly, economic balance can be approached fro m th e angle of th e economic unit , an d understood in th e sense of th e relatio n existing betwee n means an d wants i n th e narro w compass o f individua l or grou p economy. Economi c balance here can be equated wit h th e relatio n betwee n capacity to produce an d the abilit y to satisfy existing wants , i.e. betwee n income an d expenditure .
Althoug h th e tw o aspects refer to the same phenomenon , an d are thus clearly interdependent , it is necessary to poin t out tha t th e relatio n betwee n th e tw o does not invariabl y lie o n th e surface. Certai n want s whic h greatly influence individua l economy do not enter openly int o general exchange relations. I refer here to wants whic h flow fro m specifically social sources—the dictates of religious obligations, of eti • quette, status and prestige, political interests, an d so forth . Ove r an d above th e overt exchange-mechanisms whic h we study unde r the first aspect of economic balance we shall have to introduce , in the discussion of the second aspect, a certain ‘invisible’ exchange—exchange against such imponderables as mora l values, or forms of social appreciation. We wil l examine these tw o aspects of economic balance in th e order in whic h w e have enumerated them .
CURRENCY, MEASURES, AN D PRICES
Th e description o f Nup e productio n has clearly established the natur e of Nup e economy as a highl y complex exchange-economy based on money. A discussion of exchange-mechanisms ma y thus profitably begi n wit h th e examinatio n o f the units o f exchange, the currency, an d certain fixed measures used in trade .
Before th e adoption of Britis h mone y cowrie-shells constituted th e currency of th e country. It s use still lingers—not in th e currency tha t is actually used (although this, too, has been mentioned), 1 bu t in an abstract sense, i n th e native methods o f reckoning money . Certai n standardized amounts, e.g. bride-price , certain traditiona l gifts an d payments, are expressed mor e often i n cowrie-shells tha n i n moder n money. I n th e countr y figure s i n cowrie-shells are quote d commonly , an d i n connexion wit h every kin d o f payment , includin g food-prices an d suchlike. Cowrie –
1 See p. 310.
ECONOMIC BALANCE : EXCHANGE 315 shells are a clumsy currency: yo u have to carry them about in bagfuls,
and to count in thousands and tens of thousands. Th e numerical system
of Nupe is fully equal to this task. It s elaboration and lucidit y is not the
least contribution to the efficiency of the exchange system of the country.
It s scope is practically unlimited . A figure like 3,600,000 is expressed by
this comparatively simple symbol : gba kpotwani (lit . tw o thousand times
eighteen hundred, or, in more detail, two thousand times tw o hundred
times nine). Thi s facility in forming large numbers is due to the exis•
tence of a number of simple root words and multiples whic h are easy to
handle. Omittin g here the simpler figures 1-9, we find root words for
100,200, and 2,000, multiples of 100 and 1,000 being formed by a simple
multiplicatio n of the root figures. For example, 400 is 200 . 2, i.e. kpa
guba,contracted to kpoba; 600 is 200 . 3, i.e. kpa guta, contracted to kpota;
4,000 is 2,000.2, i.e. gbaguba, contracted to gboba; 6,000 is 2,000.3, i.e.
gba guta, contracted to gbaotd; 10,000 is 2,000 .5 , i.e. gba gutsu, con•
tracted to gbotsu’y and so on. A new uni t is introduced as an alternative
symbol for 20,000 cowries in the for m of the gura, whic h is the name for
the grass bags in whic h cowries were formerly packed. Thu s 200,000
cowries (the traditional bride-price) make 10gura; 400,000 20 gura, and
so on.
Th e exchange rate between cowries and Britis h money appears to have undergone some change. Th e original exchange rate, whic h I as• certained from the account of bride-price and tribute paid formerly, was 8d. for 2,000 cowries. Later the exchange value of cowries rose to is. for 2,000 cowries. Thi s exchange rate is applied to-day when figures are quoted in cowries. Althoug h cowrie-shells represented a cur• rency in the strict sense of the word , they were used also for certain other purposes beside exchange, and had, above all , a certain value as ornaments. Gourds, bags, and musical instruments were formerly ornamented wit h cowries; but this practice has fallen into disuse.
To this elaborate system of units of exchange corresponds an almost as well-defined system of measures applied to commodities. We have spoken already of the kpere, the ‘arm-length’, a standard length used in measuring cloth ; 10 kpere give, as we have seen, one full-length cloth. Th e measures used in selling agricultural produce are even more inter• esting, as quantities of these commodities do not lend themselves easily to simple forms of standardization. Th e uni t for unthreshed grain is the bundle, epa (from pa, to tie up) or load, kara. It weighs very nearly
60 lb. , bu t its size and weight vary a littl e according to the quality of the grain : of poorer grain there woul d be a larger quantity in the bundle, of better-class grain a proportionately smaller quantity, the main point being the equality in the value. Threshed grain is sold in large or small gourds, for whic h to-day ti n bowls and basins are usually substituted. A large basinful of grain, evo (lit . gourd), contains twelve small bowls, kpanu (the latter taking a littl e over one pound). One bundle of un • threshed grain is reckoned to correspond to one evo of threshed grain. Wit h maize five to six cobs are reckoned to give one small bow l of maize- flour. Rice is sold in large gourds (evo) or, more frequently, in small bowls, ten of whic h as a rule give one gourd. Beans are calculated in
316 ECONOMIC BALANCE: EXCHANGE
baskets (kasa); fruits and ground-nuts in kasa or kara (loads). Th e calculation is fairly precise, one ‘load’ of shea-nuts, for example, being reckoned to bring 2s., or one load of locust-beans 3d.
These examples suffice to illustrate the clear conception which the Nupe have of measures and marketable quantities. Combined wit h the universal use of money they result in a well-defined system of market prices. The calculation of market prices in Nupe corresponds closely to variations in supply and demand, above all, to seasonal fluctuations. We have already given instances. I wil l add here a few more which refer to the staple produce of the country.1 In Apri l and Ma y 1936 one bundle of unthreshed sorghum was sold for 6d., in August and Septem• ber for 9d. to 10d., and in certain places even for 1s. Durin g the same time the price of bulrush-millet rose from 6d. to 1s. 6d. In Apri l ten yams were sold for 6d.; at the beginning of June the price rose to 10d., and in July, shortly before the yam harvest, to 1s. 8d., to fall again in August to 6d. In the case of bulrush-millet and yam, the cultivation of which is restricted to certain districts, the fluctuations of price are markedly greater than in the case of the universal food-crop, sorghum. For rice, on the other hand, an expensive foodstuff for which there is only limited demand, and new supplies of which appear on the market three times a year, the price remained unchanged at 2d. for a small bowl.
These prices refer to village markets. Prices in Bida are always higher. When grain in the country sells for 6d. a bundle, the Bida price is Sd.; when the highest price in the country is 10d., it wil l reach in Bida
1s. and 2s. When yams are sold ten for 6d. in the country, the Bida price is ten for 9d., and so forth. Industrial products manufactured in Bida, on the other hand, are dearer in the districts, as the middlemen add a certain amount to the original price (between 1d. and 2d. in the shilling). Goods produced in the country which pass through the hands of the Bida trader on their way to Bida market or to markets abroad become markedly more expensive: thus Kutig i mats produced locally for 11d. are sold in Bida for is. or 1s. 2d., and abroad, for example in Lokoja, for
- 6d.
These regular variations and fluctuations in price must, to some extent at least, be based on a systematic common knowledge of the conditions which determine supply and demand in the country and which can thus sustain definite expectations of price developments at a given moment. Such a body of common knowledge indeed exists in Nupe. It is em• bodied in the universal experience that certain commodities are always plentiful in the dry season and get more and more scarce towards the close of the rains. It is also reflected in the generally accepted fact that prices are higher in the capital, which has to import a large proportion of its food-supplies, and which is also wealthier and possessed of a higher standard of livin g than the rest of the country. These considerations imply a further important fact viz. that distance between area of produc• tio n and market, and the additional labour and loss of time involved in
1 Certain fluctuations are determined by increased consumption during the festival season. Thus one sheep fetched 55. in January and 10s. to 15s. in March, i.e. shortly before the Great Sallah, to fall again to 5s. duringthe rains (see p. 203, n. 1).
ECONOMIC BALANCE: EXCHANGE 317 transport, enter int o the calculation of price and profit.1 Finally, the economic significance of ‘waiting ‘ is fully recognized, and goods sold are
at a proportionately cheaper price when they can be disposed of in
larger quantities. I found, for example, that a trader on the market
woul d sell two bowls of grain for i\d., but one bow l for o.8d.; a large
evo (containing 12 small bowls) of the same grain could be had for $d.
One large evo of rice (containing 10 small bowls) is sold for 1s. 6d., while
a small bow l woul d be priced at 2d. Th e factor of ‘waiting ‘ plays an
important part also in the economy of the individual household. When
money is urgently needed, or when for other reasons the whole crop is
sold in one, the seller must be prepared to accept a less favourable price.
A farmer in Lem u sold his whole crop of bulrush-millet and late millet
for 4d. a bundle, in order to have sufficient money in hand for certain
urgent household expenses (the marriage ceremony of one of his sons).
Wit h his sorghum crop he could afford to go more slowly, and obtained
for it prices varying from 6d. to 9d. per bundle. Another farmer, near
Bida, had to sell twelve bundles of late millet and sorghum for 1s. per
bundle, as he was pressed for money for tax, well knowing that if he
could afford to sell his grain in smaller quantities he could easily obtain
- 6d. a bundle.
Thi s common knowledge of price-conditions seems also to explain the lack of haggling in Nupe marketing—to the student of Africa n peoples a rather surprising feature. Cooked food, all victuals in • cluding grain, rice, palm-oil, &c , as also the common products of tailors, blacksmiths, mat-makers, and so forth , have ‘fixed prices’, and buyers wil l never tr y to bargain. They might, when buying yams or fish, a mat or a hoe, poke about i n the goods displayed for sale til l they find the biggest or nicest-looking specimen offered at the price they are will • ing to pay. Real haggling and bargaining is common only wit h ex• pensive, more rarely sold, commodities, the price of whic h is always more or less fluid, and commodities the quality and perfection of whic h show many gradations: embroidered cloth, a woven gown, and all products of artistic value.
Th e common knowledge of price-conditions is limite d by existing communications and forms of economic contact. Thu s we find a uni • for m regulation of prices only withi n the area in whic h regular contacts of this kin d exist; outside it the calculation of prices and profits, and productive planning in general, is determined by the varying local conditions of supply and demand. Th e most striking instance of this lack of co-ordination is perhaps the backward village of Sakpe to which I have referred in a previous chapter, where the people were unaware of the fact that the kapok from their silk-cotton trees had a money value in Bida. Th e following example, though referring to an exchange con• nexion going beyond Nupe country, is no less significant: in the same village, Sakpe, I met two men from Igbira country who had come to
1 It is not unnecessary to mention especially this aspect of Nup e economic knowledge. For it is frequently found in more primitiv e societies that the economic value of time is not recognized and that native producers sell their goods at a distant market for the same price as they woul d locally.
318 ECONOMIC BALANCE: EXCHANGE
Nupe to sell dogs, and were greatly disappointed to find that dogs had no market value in that country. It «ieed not be stressed that modern development of communications has greatly enlarged the area in which common price-conditions obtain. Railways, steamers, and road trans• port have become regular factors in native marketing and in the planning of native production.1 Th e stores of the European trading-companies, which are the centres for the marketing in commercial crops, have helped to expand further this framework of common economic condi• tions. But these last factors have at the same time introduced a new element of insecurity into the body of native economic knowledge. Th e new, -wider, framework of reference includes conditions of supply and demand which go far beyond those embodied in the native productive system, and calculations of price which are no longer accessible to the experience of the native producer—the conditions of supply and de• mand, and the calculations of price, determined by the state of European or American markets. Th e erratic and, to the native, incomprehensible fluctuations in the price of ground-nuts or cotton are typical instances; we have seen how, in the extreme case, they may completely upset native agricultural planning. Th e price policy of the trading-companies may also place a certain native productive area in a more unfavourable position than others: thus the price for ground-nuts paid by the United Africa Company in Nupe (that is Niger Province) was 5s. 6d. per ton as against 11s., for an inferior quality, in Zaria or Kano. Th e conditions in cotton marketing were, till recently, similar.2 Many enterprising native traders avail themselves to-day of the modern facilities of transport to visit these various towns and centres of trade: how can they be anything but bewildered by that unwarranted difference in price; or rather, they have learned to accept it as one more of the many incomprehensible arrangements brought in by ‘the white man’, and find a ready explana• tion for it in the fact that ‘money is more plentiful in Zaria and Kano’. We must consider here the relation between commodity prices and labour. Th e prices for the products of native industries given in an earlier chapter reflect, within limits, the varying amount of labour necessary for their production. Th e lowest daily profit with which the various Bida industries reckon is, as we have seen, between 1½d. and 2d. It is interesting to note that this amount corresponds to the lowest cost of living per day of an adult in Bida (see below, p. 342). Th e highest profits, on the other hand, may be determined not only by the amount of work, but also by the skill and artistic qualities which the work involves. We have seen that the artistic work of embroidery is paid more highly than ordinary tailor’s work. If we compare, however, the average profits made in blacksmith work with the profits with which brass-smiths
1 A Nupe trader known to me bought £5 worth of cheap kola-nuts in Lagos early in the season, stored them for 6 months, and then transported them by rail to Bida, where he sold his stock for £10 .
2 I was informed that this discrepancy in price was due mainly to the mono•
poly position of two large trading-companies in Niger Province, while in the north several companies have to compete with one another. Th e recently adopted control of the cotton markets by the Administration tends to maintain cotton prices at approximately the same level in the whole of Nigeria.
ECONOMIC BALANCE: EXCHANGE 319 reckon, we make the somewhat surprising discovery that here the more
refined work is not only not more highly paid but partly even worse paid
than the cruder industry. Both as regards profits per day and profits per
article produced the brass-smiths work on the whole with lower profits than the blacksmiths.1 Why, then, does the calculation of price and profit allow for an appreciation of the specific artistic qualities of the work of tailors and embroiderers but not in the analogous instance of
brass-smith work? But is it an analogous instance? In the case of tailors and embroiderers both the cruder and the more refined technique belong to the same class of production: they are both what we may loosely call luxury industries. In blacksmith and brass-smith work we have fundamentally different industries, one supplying necessities, and the other a typical fashion and luxury industry. Th e explanation thus seems to be that indispensability may equal aesthetic value; and that where these two diverse types of industry compete side by side the factor of necessity may balance or outweigh the appreciation of artistic qualities.
MARKET AND TRADE
We have had frequent opportunity to speak of the complexity of Nupe economics and of the far-reaching specialization of Nupe production, which must express itself most forcibly in the wide scope of trade and marketing. In a wider sense Nupe country itself is one big market. For Nupe is a country which exports and imports, and represents, besides, an important station on the route of transit-trade across Nigeria. Th e ancient trade-routes have to some extent been adjusted to-day to the new forms of communication, motor-roads, steamers, and railways. But to a large extent they have remained the same, for now as of old a large proportion of this transit-trade is carried to, and by, the water-way of the river Niger.
Th e following are the traditional trade-routes through Nupe. They run mostly roughly from north to south. One, coming from Sokoto and Birnin Kebbi, goes through Kontagora and reaches Nupe north of Mokwa ; from here it goes either west, crossing the Niger at Jebba (formerly at Buka, five miles up the river) into Yoruba, or east through Labozi, Sakpe, to the Kaduna, which it crosses at Gbara, then south to Muregi and, after crossing the Niger, on to Yoruba and Kabba. A second route from Kontagora reaches Nupe Emirate north of Kuti – wengi, passes through Dabba and Kutigi, and crosses the river Niger at Ogudu, on the way to Ilorin. An alternative to this route branches off at Dabba, crosses the Kaduna at Wuya, and ends at Bida. A third trade- route passes through Birnin Gwari and Wushishi, and reaches Bida via Gbangba. There are three routes from Bida to the south and south-east: one through Jima and Dokomba, down the Kaduna by canoe to Muregi,
1 Th e profits of blacksmiths per object are: 3d., 3½d., 6d.; per day, 6d., 7d., is. Th e profits of brass-smiths per object are: 1½d., 2d., per day, 3d., 6d. Higher profits are only realized in the casting of bangles in precious metal, in the casting of brass figures for tourist trade, and also in the production of ladles in the modern cheap material, aluminium.
320 ECONOMIC BALANCE : EXCHANGE
where the route crosses the Niger and goes on to Ilori n and Yoruba ; a second, to Badeggi and Kacha, where it crosses the Niger to Egga and Yagba country ; the thir d route goes fro m Bida to Kacha, or Giddi , farther down the river, where the trade-goods are loaded on canoes on their way south to Lokoja, Idah , and Onitsa. In addition to these land routes we have also an ‘all-water’ route, the nort h terminus of whic h used to be at Rabd or Ogudu.1
Le t me summarize the mai n facts concerning the ‘ international’ com• merce in Nupe. Nupe country exports a certain proportion of its agri• cultural produce, especially rice and Labozi kola-nuts, smoked fish, and a large variety of its industrial products. It imports various foodstuffs, especially palm-oil and European salt, cheap southern kola-nuts, potash, and live stock. It re-exports potash and beans from the nort h to the south, and palm-oil , European salt, and kola-nuts from the south to the north . Thi s commerce follows certain interesting traditional patterns. A simple export trade from Nupe to another country is the exception: of this kin d is the trade in smoked fish from the riverain districts of Nupe by rail to Ilori n and Ibadan, or the trade in Nup e glass, beads, brass, and silverware, whic h goes by canoe to Lokoja. In the majority of cases the impor t or export trade in a certain commodity is combined wit h a definite return trade in another commodity. Here we find the following regular
‘patterns’: donkey caravans from Hausa brin g beans to Bida and return wit h kola-nuts (either the local Labozi nuts or the cheap southern variety). Other donkey caravans go nort h from Nupe ; they are loaded in Kacha wit h palm-oil and salt, whic h arrive on the Niger Company steamers from the south, and return from the nort h wit h kanwa (potash) fro m Bornu . In Kacha the riverain Kakanda buy Nupe-grown rice, whic h comes down by canoe from Patigi and the Kaduna, and take it south to Onitsa, whence they return wit h palm-oil . Hausa traders from Kano brin g horses and sheep and goats to Bida, to return to their country wit h Nupe straw hats. Another exchange pattern between Hausa and Nupe is the trade of potash against Bida embroidered gowns. In this trade the imported goods are invariably first sold for money, wit h whic h the goods for the return voyage are then purchased. Al l the variations that are theoretically possible in this threefold export and transit-trade are found to occur in practice. In other words, the Nupe merchant plays three different roles: he is either the local broker who arranges the sale of goods between importer and exporter; or he acts as wholesale trader and buys himself the imported goods to re-export them (himself supplying the transport) to their eventual destination; or he acts as retail trader and sells the imported goods either directly to the customer (on the Bida markets) or to other retail traders, who wil l dispose of them on the various village markets of the country.
We wil l begin our study of the marketing system of Nupe wit h the
1 To-day the ‘all-water’ route goes to Jebba, the new railway and steamer station, and even farther north, the water-way north of Jebba having been made navigable for canoes. I have described this river-trade more fully in ‘The Kede: A Riverain State in Northern Nigeria’ (in African Political Systems, Oxford University Press, 1940, p. 165).
ECONOMIC BALANCE : EXCHANGE 321 study of the village market. No Nupe tow n or village is withou t its
market and its regular market-days. Larger places hold their markets
once a week; smaller villages every fifth day (irrespective of the seven
days’ week). Thi s arrangement, illogical though it may appear, is wel l
adjusted to the needs of the country, where local transport is stil l to a
large extent by porterage. It enables more people, and people from more
distant places, to attend the markets in the bigger villages. I estimated
the attendance of the big market at Kutigi , a village of 3,000 inhabitants,
at fro m 400 to 500. Th e distance fro m whic h people come varies con•
siderably. Kacha market, the largest market after Bida, attracts men
and women traders fro m the capital, forty miles away. In Kutig i I met
women traders from Lemu , whic h is thirt y miles from Kutigi , who were
selling pots made in Gbar i country, another fifty miles north . Profes•
sional, full-tim e traders, men and women, regularly attend all the larger
markets in the country, stopping in one village for five or six successive
market-days, and then moving on to another place. Th e peasant women,
on the other hand, who practise trade only as a ‘side line ‘ and whose
marketing is concerned chiefly wit h the disposal of the produce from
their houses and their husbands’ farms, rarely attend markets whic h are
more than six or ten miles from their home. These small circles of inter-
village exchange form no less part of the comprehensive marketing
scheme of the country than do the extensive journeys of the professional
traders. For the system of local exchange areas is partly subordinated
to the main axes of trade and traffic. We find an organization almost of
relay stations, half-way houses of exchange, situated all along the main
trade-routes of the country, above all on the roads leading to the capital.
On the western road to Bida from Wuya (which is 12 miles from Bida) there are three such ‘ relay stations’, used for the different kinds of produce of whic h Bida is constantly in need. Duzk o Wuy a (lit . Market Wuya) on the right bank of the Kaduna is used by the women of Batati, a village fifteen miles inland, for trading their cotton to Bida traders who m they meet at this place. Zantica, six miles from Bida, and Lum – anla, four miles from Bida, serve as exchange places for agricultural produce and firewood, whic h goods are brought to these places by the local women, and then bought and carried to Bida by traders from the town . A similar arrangement, wit h regard to yam and other farm – produce, obtains on the northern road to Bida from Lemu ; the ‘relay station’ is a tin y village half-way between Lem u and Bida, the only im • portance of whic h lies in its use as an exchange market.
Similar relay systems whic h exist in the river area, following and branching out fro m the water-ways, serve as centres for the extensive trade in smoked fish. Here the local exchange circles are primaril y determined by the varying conditions of transport. Thu s the wives of fishermen on the lower Kaduna take their fish (already smoked) in their small canoes to the nearest riverside market, e.g. Jima or Muregi . Here a wholesale trader fro m Jebba buys up fish in large quantities, takes them , again by canoe (this time a large river-going canoe) to Jebba market, and sells them to another wholesale trader who wil l transport the fish (by rai l or lorry) to the final destination—the inland markets of Bida,
Y
32 2 ECONOMIC BALANCE : EXCHANGE
Ilorin , Oshogbo, or Ibadan. Another method is that the wholesale trader who buys up the smoked fish at the local riverside markets also trades the m directly to Bida, Ilorin , &c. ; certain enterprising traders wit h plenty of capital may also charter canoes and, travelling from place to place, buy up the local produce before it comes to the riverside markets.1
I have emphasized on various occasions the close correlation between markets and specialization of production. Here is the place to state the case once more. Generally speaking, trade and transport whic h brin g the products of one area withi n reach of another involve competition, and through competition the elimination of inferior forms of production. Trade and transport must thus tend to foster (as they are in tur n fostered by) local specialization. Certain developments in the agricultural and industrial production of Nupe reveal the effectiveness of these influences also in that culture. Th e people of Jebba, for example, do not grow sugar-cane or onions on their land, although on a similar type of land in Raba and Ogudu these crops are cultivated. The y explain this situation as being due to the fact that they are farmers on a less extensive scale, spending muc h time on fishing and (formerly) hunting ; they prefer to impor t onions and sugar-cane fro m these other places. Fo r certain handicrafts our evidence is even more striking , for here competition owing to an expansion of trade has only comparatively recently replaced an indigenous craft by imports from another area. Th e women of Leaba formerly practised pottery; bu t when the regular canoe traffic, whic h used to stop farther down the river, was brought up to Leaba, and pots made in Jebba and Gbajibo reached Leaba market, the Leaba women gave up their craft; their own pots, they explained, were inferior in quality to the imported product, owing to the inferior potter’s clay found locally.
In small villages no trade is done, and markets are deserted, on days other than market-days. In large villages the markets do a littl e trade— in food only—also on ordinary week-days. On the day after a market there is often quite considerable trade in goods whic h di d not find a buyer the previous day. On market-day the people begin to arrive between 9 and 10 o’clock. Th e market is in ful l swing by noon, and fro m 1 to 3 is the busiest time. After 4 the attendance begins to scatter, and soon after that the traders remove their goods, and the market is left empty. Every kin d of trade and craft is represented even in the smaller village markets. In the roofed booths whic h every market possesses the more valuable goods are displayed—products of native industries as wel l as European goods. Tailors and leather-workers wor k in open huts. In the open square framed by these booths and huts agricultural produce, food-stuffs of every kind , live stock, pots, and tools are offered for sale.
1 Lately the wives of the fishermen have attempted to break down the ‘relay system’ and to dispense with the services of middlemen, having begun to market their fish directly in Ilorin and Ibadan. A place like Ogudu, on the road from the river to Ilorin, which formerly formed an important exchange place, has now lost this transit trade in fish. Th e chief of Ogudu, to whom I owe this informa• tion, remarked regretfully that the women have ‘begun to understand’ (lit. ‘have seen light’).
ECONOMIC BALANCE: EXCHANGE 323
In one corner of the market barbers put up their open-air shop. In another corner butchers sell their goods, a drummer endlessly repeating the same rhythmical motif—the signal that the butchers have slaughtered that day—to attract the attention of the crowd.
SKETCH-MAP O F KUTIG I MARKE T
Markets in the wider sense of the word , places for exchange and trade, grow spontaneously wit h the needs of the country. Bu t markets of the kin d just described, wit h permanent buildings, booths, and stalls, have to be buil t and laid out specially. Moreover, provision must be made to guarantee the protection of the large attendance, especially of the traders from other districts and countries. Th e element of planning is clearly visible. Th e villages whic h decide to have a market of this kin d wil l go to wor k methodically; formerly, they appealed first, through the royal delegate in their district, to the king , who woul d designate some of his private slave police to take charge of the supervision of the market. In the Mohammedan and semi-Mohammedan villages a well-known Malla m woul d be summoned to consecrate the market and thus ensure its prosperity. Thi s practice is said to be stil l in use. Certain markets become, in fact, known by the name of the Malla m who dedicated them. A sketch-map wil l best illustrate the well-planned arrangement o f
a typical village market—the market of Kutigi , to whic h I have referred
324 ECONOMIC BALANCE: EXCHANGE
above. A census of the same market will give quantitative data with regard to the attendance of the market and the goods which are offered for sale on an average market-day. Th e figures quoted, which are of necessity elastic, were compiled from a large number of regular observa• tions during the months of February and March. Th e kind and quantity of goods traded on the Nupe markets reflect the seasonal fluctuations in supply and demand. It must be understood, therefore, that at other times of the year, especially in the rainy season, the com• position of the market will be somewhat different. In a large centre like Kutigi, however, which can count throughout the year on imports from distant places, these changes remain within more narrow limits than in a small market which relies largely on local supplies. For the purpose of comparison I have also added the corresponding figures referring to ordinary week-days.
Th e chart shows that, in marketing, certain goods are regularly com• bined. Th e same women who sell grain also sell rice, red pepper, kuka, and partly palm-oil and shea-butter. Th e sale of yam is regularly com• bined with the sale of cotton, and sometimes of gourds; or cotton is sold by the traders in native cloth. Such goods as smoked fish, native tobacco, or cooked food, on the other hand, are always traded separately. Th e combination of goods offered for sale does not invariably betray the character of the seller, whether she is a farmer’s wife, marketing his produce, or a professional trader or ‘middleman’. The two categories, in fact, overlap to some extent. As a rule the women who sell grain, rice, kuka, red pepper, yam, and cooked food are only ‘occasional’ traders, i.e. are marketing produce from their husbands’ farms. In the case of the women who combine the sale of grain, red pepper, rice, and palm-oil, grain and red pepper may well come from the farms of their husbands, while rice and palm-oil may have been bought by the woman from some• one else. Or the woman who sells cloth and cotton may be selling her husband’s cotton together with cloth which she purchased in Bida. Th e women, on the other hand, who sell tobacco, smoked fish, and various Bida goods are invariably professional traders or ‘middlemen’.
Th e distinction between seller and buyer is no less difficult to elicit from the chart. The two roles are often combined in the same person. Many women who come to the market to sell agricultural produce or cooked food, or men who bring live stock or hoes, at the same time pur• chase various goods from other traders. But the market, in Africa as well as Europe, is more than merely a business institution. Th e people who attend it, whether they come to buy or to sell, also come for the sake of company and entertainment. In the market they meet friends and acquaintances with whom they can gossip and exchange news. There also the sexes meet, girls and their suitors. Rarely will there be a market- day on which drummers and musicians do not turn up also, and young men or girls join in a dance. This ‘social’ aspect of the village market must appeal more to the local people, than to the professional traders from Bida or other more distant places. It presupposes common interests, a familiarity and intimacy of long standing. Th e group of peasant men and women who are brought together regularly on market-
ECONOMIC BALANCE: EXCHANGE
Census of Kutigi Market
325
Attendance and quantities of goods
Goods On market-day On ordinary days
Fulani butter and 4 Fulani women from camp in the 2-3 Fulani
milk neighbourhood, each with 1-2 gals, of milk and 1 lb. butter
women
Grain, rice, red 30-40 women, each with 10-12 kpanu 2 women pepper, kuka of grain and rice, 1 lb. of red pepper
and a small bowl of kuka
Palm-oil, shea-but• 50-60 women, partly the same as those 2 women ter, and salt who sell grain and rice, each with 2-4
gals, of palm-oil (in old petrol tins)
Cotton and yam 10-12 women, each with 4-5 heaps of yam, each containing 6-8 yams. Quantity of cotton very variable
Smoked fish 30-40 women, each with one basket containing 15-20 fish
Various greens and 15 old women herbs (partly used
as medicine) !
2-3 women
3 women
Native tobacco 5 women 3 women
Live stock Me n with 4-5 sheep or goats
Pottery of different 10-15 wome.n, each with 10-15 smaller
size or 4-5 larger pots
Gourds 7-8 women, partly the same as those who sell cotton
Hoes 1-2 men, each with 4 hoes
Mats Very variable, 2-10 mat sellers, mostly
Cooked food (pa• men
4-5 women
stries, corn-gruel,
2s0e-l3l 0grwaoinmoernp, tahlems-aomile, eaascthhowsietwh h2opatsls.o
corn-water, &c.) corn-water, 20-25 packets (in banana leaves) of corn-gruel, and 10-15 pastries
Meat 3 butchers, average 2 sheep or some• The y slaughter times 1 bull or cow every second
day
Straw hats 5 men
Cloth, various 5 men in the booths, and 15 women fancy goods (Nupe and Yoruba) in the square
Tailors 6 men, in the booths
Barbers 4 men
Leather-workers 4 men, in the booths Charms, medicines, 4 men, two of them Hausa and potash
Roast meat 1 man, sitting on the road-side opposite the market
days indeed constitute a substratum of community life at large. Th e social life at the village markets utilizes, and creates or revives, at the same time the contacts and social relationships which characterize the ‘area of common life’. It is no accident that this circle of five to ten miles radius from which the local attendance of village markets is re• cruited coincides with the area which is covered by local knowledge and interest in each other’s affairs,1 and within which intermarriage between different villages takes place.
1 See p. 38.
326 ECONOMIC BALANCE: EXCHANGE
I have described already at some length the markets of Bida so that there is little to add here. Bida possesses six markets, three main day markets close to the 3 royal houses, two small day markets—one near the west gate, one in the strangers’ quarter—and one night market. In the distinction between day and night market the village distinction between the restricted trade on ordinary week-days and the extensive trade on market-days is in some respects repeated.1 Multiply by ten the largest village market, and its attendance and the quantity and variety of goods displayed, and you have the Bida night market. Its average attendance has been estimated at 6,000, and as regards the goods which are bought and sold, I can only repeat that there is practically nothing that cannot be obtained there, from kola-nuts to beams for roofing, and from native- tanned goatskins to watches, scissors, or bicycle spare parts. Besides, there are scribes who will write letters for you, Yoruba money-changers (who charge a commission of one penny in the shilling), cooks who roast small pieces of mutton over the fire and sell the meat on wooden skewers to passers-by, drummers, barbers, tailors, loungers, not to forget the policeman who regulates the traffic and keeps watch (not always a sinecure) on disturbances of the peace.
In certain respects, however, marketing and trade in Bida differ essentially from their counterpart in the village. A first difference con• cerns the local distribution of marketing. Bida possesses, in addition to its four markets, also more or less regular shops which are open all day, such as the wayside shops in the street of the brass-smiths, where the womenfolk of brass- and silversmiths display the goods manufactured by the men, or the permanent shops in Idlemi, the crowded strangers’ quarter. Many traders, moreover, who will attend the night market keep during day-time little wayside stands on cross-roads or busy thoroughfares. Th e wide local distribution of shops corresponds to the typical demands of a large, decentralized town, in which the groups of potential customers are concentrated in widely scattered localities. At the same time it expresses a keener spirit of competition, and that doctrine of ‘seeking out the customer’ which is familiar to all more advanced systems of marketing. Tw o new stands which have only appeared during my stay in Bida will illustrate what I mean. One way• side shop established itself opposite the workshops of the Public Works Department, and its customers were the labourers, who would come out to buy food, sweets, and cigarettes; the other was installed close to the Bida Middle School, and catered for the occasional wants of schoolboys and teachers. In both cases two or three young women had joined forces and were keeping shop together. These wayside shops and stands do not pay rent; street corners and cross-roads are free to everyone; nor have any quarrels over claims to a particular spot arisen—as yet. But this is essentially a fluid process; and it is not unlikely that quarrels or litigation of this kind might arise in the future if competition becomes
1 Th e day markets reflect the different needs of the large town as compared with the village markets. Firewood, for example, which is never seen on village markets as the women collect their own firewood on family farms, appears in large quantities on the day markets of Bida.
ECONOMIC BALANCE: EXCHANGE 327 more severe and the advantageous positions secured by the enterprising
first-comers are threatened by later recruits of the ‘seeking-out-the-
customer’ movement.
Wayside shops also line the road leading from the north gate to the
two large European stores—Canteens’ in the popular idiom. Here the
inventiveness of Bida traders has seized upon a new device to attract and
‘seek out’ the customer. Many peasants from backward villages who
come to Bida to buy certain essential European articles, such as salt or
sugar, either want to buy them in very small quantities—smaller
quantities than they could obtain at the store—or are too shy and
frightened to enter the stores themselves. These are the customers of the
little stands outside the ‘canteens’: for here the peasants can buy the
same articles in the smallest quantities and without having to undergo the ordeal of a purchase in the unfamiliar, forbidding, store.1
MODERN BUSINESS METHODS
Th e last example brings us to another essential difference between the marketing system in Bida and in the village. Th e large volume of native trade in European goods and the close business connexion between native traders in Bida and the stores of the trading-companies have introduced certain completely new factors into Nupe marketing. They are visible in local trade on the smallest scale as well as in the large-scale enterprises of wealthy Bida merchants. Trade in certain European goods allows a more rapid turnover and more certain profits than any branch of native economy. Th e native trader or middleman can, as we have just seen, turn to his profit the fact that he can make certain essential and highly valued goods accessible to the consumer by retailing them in such small quantities that their price (though relatively high) is low enough to be within the means of peasants and small craftsmen. It is, for example, a regular practice of Bida women traders to buy tins of
50 cigarettes at the ‘canteen’, and to sell them singly (or even cut in half); to buy boxes of matches, and sell them in little bundles of five; packets of sugar, and sell the sugar a piece or two at the time; or to buy 4-gallon tins of paraffin, and sell it in pint bottles. It is hardly surprising that the profits, though absolutely small, are relatively high, up to 100 per cent.2
1 This fear is really a compound of shyness and suspicion—one is suspicious just because one is shy and unfamiliar with the ways of these modern institu• tions. A most characteristic instance, concerning a Bida man who, one would think, would hardly need to be shy or afraid, is the following: a Bida Malla m with whom I was very friendly came to see me one day and asked me if I would do him a favour and order spare tyres from the ‘canteen’ for a lorry belonging to a friend of his. His friend, he explained, disliked ordering them himself, for he knew he would be cheated and given the wrong kind of tyres. What had happened was that his friend had some time previously ordered tyres without, however, knowing that it was necessary to quote the exact size. He had been given tyres of the size most frequently found on the lorries running in the country, but unfortunately they did not fit his lorry.
2 One tin of 50 cigarettes costs 15.; one cigarette is sold for 0.3d., allowing a profit of 3d. in 1s. One box of matches, containing approximately 50 matches, costs 1d.; five matches sold for o.2d. give a profit of 0.8d. in the penny. One
328 ECONOMIC BALANCE: EXCHANGE
It is impossible to discuss in detail the business enterprises of the big merchants of Bida. Their business activities hardly differ in essentials from the business activities of Western civilization. They are the whole• sale traders who, working with large capital, buy goods from the Euro• pean stores and supply the small retail trader with commodities which the latter sells in small quantities in the town market or peddles in villages and local markets. Some big merchants are contractors who run lorries for the transport of passengers and goods, others importers of goods from countries outside Nupe. Their business is to a large extent based on credit which they receive from the European trading-com• panies, and there are several Bida merchants who figure on the books of the trading-companies with debit accounts of £200 or more. I will only outline the effect which the introduction of commercial credit in the modern sense has had upon Nupe economy. The concept of credit itself is not alien to Nupe economy; but this new type of credit, which can be so readily obtained, opened to the Nupe trader a source, not only of working capital, but also of liquid capital as such. Native business men have learned to use this new economic device to the full, and in their own way—which is not always the orthodox way of credit operations. This is an actual and, I may add, typical instance: a Bida trader bought European cloth for £3 in the store, on a three months’ credit. He then sold the cloth at a loss, for £2 , to a retail trader who was not unnaturally very ready to pay in cash for these goods the normal price of which was well known to him. Wit h the £2 our trader went round to the villages buying in advance the cotton crop of the year (this happened two months before the harvest was due). He paid an extremely low price—in certain cases only the annual tax-money of the farmer—and two months later sold the crop to the trading-company for an amount which was sufficient to pay his debt and to make a profit of £1 in addition.1
Th e increasing volume of European trade in Nigeria as well as in other parts of Africa is due to a large variety of causes. Th e cheapness and, generally, better quality of the European goods is often mentioned, and also the change of taste which is brought about by education and by the example of European standards of living. Our short survey of the in • fluence of European trade upon the native economy in Nupe has revealed another aspect of the problem, the importance of which has hitherto been hardty realized. Trade in European goods, qua trade, is a highly lucrative enterprise which opens up a chance for business and profits unrivalled in purely native economy. It is to a considerable extent the native trader himself who is pushing trade in European goods. He is its keenest propagandist, for he is the one who benefits first and most.
-gallon tin of paraffin costs 7s. 6d.; one bottle is sold for 6d., giving a profit of
8s. Th e higher the investment, the higher evidently the profit.
1 Th e device as such, to sell goods at a loss for the sake of ready money, and to
make the loss good on a subsequent sale, is not new. Hausa traders, for example,
sometimes sell their (native) goods at a loss in Bida in order to obtain money,
and with this money to buy certain Bida goods which they can sell in Hausa country at a profit large enough to cover their previous loss.
ECONOMIC BALANCE: EXCHANGE 329
TRADE AS A PROFESSION
Th e majority of professional traders in Nupe are Bida men and women. For these professional traders the restriction to the small areas whic h are characteristic of inter-village trade does not apply. Th e whole country, and indeed the whole of Nigeria, is their field of activity. Trade, business, is independent of the boundaries of a community, a tribe, or language. Inside Nupe one of the results of this extensive trade is the regularization of hospitality towards traders in all larger villages of the country. Onl y a few traders may be able to find relations in the distant places to whic h their business takes them. Th e others are free to stop in every house. Th e rules wit h regard to the compensation for this for m of hospitality whic h does not fall in the category of kinship duties vary locally. In Mokw a the guest maintains himself and is expected to give his host a parting gift of 10-15 kola-nuts (representing about
1s. 6d. in money). In Kacha, a much larger business centre, the popula• tio n of whic h consists to a very large extent of traders, a stranger who has no relatives in the place may go to the chief, who wil l allot quarters to him . Th e guest receives lodging as well as food, but is not expected to make any gifts in return ; he is expected only to return the hospitality in his own village should his host be taken there on his business trips.
Outside Nupe the Nupe traders are frequently more than merely birds of passage. The y stay at a place for several months or even years at a time, and may possibly settle there permanently. We find such colonies of Nupe traders in every larger tow n of Nigeria, in the northern Emirates as well as in the Europeanized centres of trade in the south. In the north they speak Hausa; in the south they can mostly get on wit h Nupe (only a few Nupe men and women have learned Yoruba). No t the least convincing evidence of the century-old route of Nupe trade to the south, down the river Niger, is that Nupe language is understood all along this route. In places on the lower Niger where the lingua franca of the north, Hausa, is no longer understood I found that the people would speak and understand Nupe. Although in the life of the traders triba l and cultural affinities are subordinated to business interests, these links re-emerge in a new form . In these foreign parts where the Nupe live as aliens, they are drawn together by the lin k of their common language and common origin . They form small cultural enclaves, Nupe colonies, and tend to live together in the same part of the town . These colonies do not for m social units or communities in the narrow sense of the word ; rather, they are fluid, semi-temporary, groups brought together by the chance of common enterprise. The y can hardly be homogeneous; they contain men and women from different sub-tribes and parts of Nupe country. Moreover, they spring up suddenly, and may disappear again as sud• denly, owing to changes in the economic conditions. But while they exist they show a strong tendency to develop something like an organized community life. They all feel themselves as Nupe ; they recognize the ranks and’forms of status valid at home and elect a head or chief in Nupe fashion. Man y men and women in these colonies are joined in the course
330 ECONOMIC BALANCE : EXCHANGE
of time by their family and relations; and many will , or hope that they will , some day return to their home country.1
In this description I have dealt wit h professional and non-professional traders as if they represented sharply distinct classes. There is, in fact, good sense in speaking in Nupe of cewoce or koragba, ‘trader’, as against other professions. Th e line of division, however, is fluid. It is more fluid among women than among men. Amon g the men we can at least define the extremes: the big trader is nothing but a trader, and the culti • vator or guil d craftsman is never a trader. Bu t trade on a small scale goes hand in hand wit h the individual crafts and ‘free professions’, and an occasional business deal is welcome to everybody. But here a further distinction is necessary. Th e Nupe distinguish between traders proper and brokers, dilali. Th e brokers have as a rule their booths on the market, and undertake to sell goods for their clients for a small commission. It is interesting to note that while the Nup e traders are ‘free’ professions in every respect, the brokers are organized in a guild-like organization. The y elect a guild-head, the Etsu or Sarkin Dilali, who is confirmed by the kin g and keeps in close contact wit h the court. As soon as a new trade caravan arrives at the zongu (the traders’ camp outside the gates of Bida) he wil l go and see the traders and inspect their goods. It is his duty also to infor m the kin g of caravans and goods that have arrived. He must ascertain the king’s wishes wit h respect to these goods before he allows them to be placed on the open market. Here, then, we find again the element of political control whic h is so closely associated wit h guil d organization. Th e Etsu Dilali, placed in control over all Bida brokers, levies a small fee from them, in return for whic h he guarantees them equal access to the valuable and profitable goods whic h arrive in Bida fro m abroad. He , and through hi m the whole organization of brokers, thus guarantee to clients and fellow members an efficient marketing organization, the avoidance of unsound competition, and business in • tegrity. Th e organization of brokers differs in certain respects essenti• ally from the guil d organization proper. It does not utilize the principles of kinship (real or fictitious); nor does it attempt in any other way to ensure the solidarity and continuity of the professional group. It repre• sents a ‘free’ association of ‘free’ business men, coming very close to the principle of the modern professional union.
DIVISION OF LABOUR BETWEEN THE SEXES (Ill )
Th e market census given above has shown that marketing and trade, too, are subject to the rules regulating division of labour between men and women. We can formulate certain general rules: all food-stuffs and agricultural produce, and of native industrial products manufactured by men those that are typical women’s articles, are traded by women. To the first category must also be counted the native stimulants, tobacco for
1 These remarks refer to a typical Nupe traders’ colony in Ibadan which I have studied at some length. An instance of a trading-colony which has sprung up quite suddenly is Agege, a village ten miles north of Lagos, which, in 1936, had been in existence for five years only.
A FOOD MARKET
A WOMAN TRADER SELLIN G HER HUSBAND’S SILVERWARE
ECONOMIC BALANCE: EXCHANGE 331 snuff and smoking, its modern counterpart, cigarettes, and kola-nuts,
although they are used mostly by men. To the second category belong
the various products of Nupe fashion industry, rings, bracelets, chains,
ladles, glass-ware, beads, and also the low carved stools, made by the
carpenters, on which the Nupe women are wont to sit. Th e products of
women’s crafts are traded by women, irrespective of their use: thus the
women sell cloth woven by women (used chiefly for women’s dresses),
indigo dye (which may be used by and for the men), native soap and
pots, including the pitchers which the men use for ablutions. Al l goods
produced by men and used by men are marketed by male traders: hoes,
axes, arms, gowns, straw hats, books, leather-work, &c. Goods used by
both sexes are traded by both men and women: for example, mats, and
certain kinds of cloth of native or European make. Live stock is always
marketed by men, only chickens are sometimes sold on the market by
women.
Throughout this description the term ‘woman’ means married women as well as girls, mothers and wives as well as daughters. Small girls of 6 or 8 will already ‘keep shop’ in play, sitting behind a basket in which they have arranged a few pebbles, little rags of cloth or cigarette tins filled with sand, and silently display their ‘goods’, as they have seen their mothers and elder sisters do. Mothers always take their young girls to the market, letting them sit beside them and, if they are big enough, carry some of the loads on the way there and back. Girls of 12 or 13 will assist their mothers or deputize for them when they are called away; or they will take some of the cooked food which the women sell, and walk through the market and adjoining streets, crying out the wares, while the mother remains sitting behind her basket. This is, incidentally, an interesting division of labour between mother and daughter, and old and young trader in general. For it is against the dignity of grown-up women to cry out their wares—only young girls may do that. A girl of
14 or 16 is already fully initiated into the secrets of commerce, and will soon start to trade on her own, with a capital of a few pennies or a small store of grain or ground-nuts given her by her mother. By degrees the capital and the scope of her business will grow. I remember a court case in Bida in which two girls, between 14 and 15 years of age, had been summoned because they had fought on the market. Th e fight was about a piece of cloth which one girl had taken from the other, agreeing to buy it for a certain amount; but later she declared it to be too dear, and paid only half the agreed price. Th e price of the cloth was 6d., and the people In the court were very amused about this commercial transaction in (as the Alkali put it) ‘little girl’s rags’. However, ‘little girl’s rags’ become real cloth, and even more valuable goods, in the course of time, and when the girl marries she is a fully fledged trader, and fully versed in the art of marketing which will be one of her duties as a married woman. Her earnings remain her own property and cannot during her lifetime be touched by her husband. Only when she dies will the husband be entitled to inherit her property.
This brings us to the question of the economic inequality of husband and wife. It is implied in this division of labour between the sexes which
332 ECONOMIC BALANCE : EXCHANGE
makes the women the main traders of the country that the women can as a rule command more liqui d capital than the men. In this sense, refer• rin g to the liqui d resources, it can be said that the Nup e women are on the whole wealthier than the men. Thi s does not mean that their financial position is entirely unaffected by the limite d resources of the husband. Th e wives of small farmers and craftsmen or, generally, im • pecunious husbands, have littl e scope for their commercial activities, and may be unable to derive fro m the household resources the necessary workin g capital for a profitable independent enterprise. Where there is littl e to sell there can be littl e profit ; and this small profit may be swallowed up by the normal household expenditure. Even the utensils and goods whic h constitute the wife’s dowry may disappear gradually in a household where there is no money wit h whic h to replace the stock of pots, clothes, and ornaments—the woman’s initia l ‘capital’ (see p. 351). Th e household of a diviner in Bida, who m I knew well , was a typical case. Th e man was of poor health, unable to do any serious work, and had to rely entirely on his income from divining , whic h was small and irregular. Th e wife maintained the household, poor and simple as it was. She was about 40 years old ; she had nothing left of her dowry save the few rags whic h she wore, and a few damaged pots. Every day she bought grain for 6d. on the market, made corn-gruel and corn-water, and sold this food in the mornings in the street, wit h a total profit of 2d. Wit h this money she kept herself and her husband and a small child .
Bu t usually the wives even of men who m one woul d hardly call pros• perous attain an impressive degree of solvency. Solvency in this con• nexion is a relative term , as the pecuniary demands whic h the wives have to meet are muc h smaller than those whic h devolve on the men as the heads of the household. Ye t in many cases the women come to the succour of their men even in expenses of the latter kind . It is by no means rare, especially in Bida, to find families in whic h a mother under• takes to pay the bride-price for the sons. Muc h more widespread is the practice of wives lending their husbands money, for example, for the payment of tax. Of this practice we shall hear more later. Le t me emphasize here the dependence of the husband on his wife wit h regard to these monetary obligations—a dependence whic h must increase as the independent economic enterprise of the women, commerce, is in • creasing in scope. Thi s fact is essential for the understanding of certain moral issues involved in the expansion of women’s trade in Nupe .
I have stated earlier that the line of distinction between professional and occasional traders is muc h more fluid among the women than among the men. Wives of farmers, craftsmen, Mallams, noblemen, traders, officials—they all are, or may be, traders. Th e enormous variety in the trade whic h is carried out by the Nupe women, and in the degree of their ‘professionalization’, cannot be reduced to a simple formula. Th e woman who sells every morning two or three dozen packets of home-made kamu (corn-gruel) for a tenth of a penny each, and the woman who buys £5 wort h of rice, kola-nuts, or European cloth, do not really differ in anything except in the amount of capital they can lay their hands on. Th e difference in the distance whic h they cover when
ECONOMIC BALANCE : EXCHANGE 333 disposing of their goods—the peasant woman who walks eight to ten
miles to the nearest local market, and the professional woman trader
who travels to Lokoja or Ilori n by canoe or train—may seem bu t an
external feature. Yet this factor of distance possesses a deeper sig•
nificance. Th e essential difference between the professional and the
occasional woman trader is embodied in the extent to whic h their trading
and marketing becomes an independent, self-contained, and self-centred
enterprise. Both the occasional and the professional trader are out for
profit. Bu t only the latter has adjusted her life fully to this aim. ‘Proper’
trade necessitates a freedom of movement, a full-tim e devotion to the
task, whic h only women who have made themselves independent can
afford. It means extensive journeys, and long absence fro m home.
Obviously, the wife of a farmer or craftsman who must look after the
house is less able to do this than the wife of a trader or a man of the
leisured classes.
Bu t the profession demands even more than comparative leisure. Th e long absence fro m home, the life on the road, in foreign villages and countries to whic h none of the normal contacts of the community stretch, means ultimately a dissolution of the bonds of marriage, household, and family. Th e disappearance of all moral control of the husband over his wife, or the local community over its women, expresses itself in the sexual licence claimed by these women. Female trade and immorality are in Nupe conception two things closely related. Th e women traders who go on these long expeditions and stay away from home for weeks, months, or even years, are commonly regarded as ‘loose’ women, as women who cannot help leading a ‘loose’ and ‘immoral ‘ life. Th e women traders prostitute themselves for money in the villages and towns where they stop on their trading expeditions. Th e money which they make is regarded by the people as legitimate a form of earnings as that gained fro m trade itself. When they return to their homes husband and family wil l receive them withou t question, and no blame woul d attach to them for the life whic h they had been leading while away. Public opinion, though outspoken in its general condemnation of moral laxity, yet countenances the sexual licence of women traders and accepts it almost as an institution . If people—men, as the spokesmen of public opinion—express their indignation at this danger to matrimonial morality, they attack, not the practice itself, bu t its disproportionate growt h in recent times.
We have, in a previous chapter, discussed the social significance of this qualified admission of sexual licence in a certain limite d social domain (see Chap. IX) . Le t me recall the salient facts: female trade, closely associated as it is wit h licentious living , provided the scope for that socially condoned sexual licence whic h other social forces seem to have made necessary. Thu s female trade became in some respects a closed profession, the entrance int o whic h was made dependent on the qualification of sterility. We may add here that, not unnaturally, women who were past child-bearing age could also legitimately joi n the profes• sion of women traders. By means of these tw o qualifications an occupa• tio n whic h harboured some danger to common morality was limite d to
334 ECONOMIC BALANCE : EXCHANGE
what were (to the society) the least dangerous groups. Thi s limitatio n is no longer true. Bu t the information whic h I collected on this point, fro m every part of the country, and every social stratum, leaves no doubt that the limitatio n di d obtain, and that it is stil l regarded as the more adequate practice. Formerly, I was told , no peasant woul d have dreamed of letting his young wife go to distant markets the journey to whic h woul d involve spending the night on the road, away from home. No r di d women wit h children ever leave the home for these long expeditions to foreign parts. To-day, an ever-increasing number of Nupe women, young women and women wit h children, travel about the country, in • dependent, and unrestricted by village morals. In the large trading- places in the south such as Baro, Lokoja, Ibadan, I have met many— they lived there, some wit h their children, as well-known traders and as equally well-known prostitutes.
We have learned that this development is due to a variety of social factors. W e wil l discuss here the factors whic h are embodied i n the economic changes, and which , in fact, appear to represent the main moving force. It is difficult to say for certain whether the floating popu • lation of wifeless men has considerably increased durin g the last genera• tion . It seems to have done so. But speaking more generally, the rapi d development of trade since the pacification of the country after the Fulan i wars and specially after the Britis h conquest cannot be doubted. It appears to have overtaken Nupe society unawares, as it were, withou t givin g it the chance to readjust itself gradually to the new conditions. Trade, in the war-disturbed times of the past an occupation for men who could carry arms, fell now easily to the women. It called for new recruits, and seized upon the existing system of division of labour be• tween the sexes—men as producers, and women as ‘middlemen’ and traders. Yet this organization, whic h was designed to fit into the narrow scope of village life, was adapted too readily to the purposes of the new widespread trade that developed in the Emirates. No other system of division of labour and specialization, it appears, could be evolved as quickly to take charge of the new tasks. It could not but disorganize social life. Family organization and the traditional concepts of morality fell victims to economic progress. Th e profit motive, here as in many other fields of Nupe economy, outstripped and overruled the interests of group cohesion and group stability.
CHAPTE R X X
ECONOMI C BALANCE : MEAN S AN D WANT S
TH E PROBLE M
N th e discussion of economic balance in this new sense we meet wit h certain methodological difficulties. Th e wants o n whic h resources are expende d for m par t of culture—the y reflect customs, conventions, tastes, fashions. Th e ideal solution woul d obviously be to includ e here a description of th e whole culture , political life, social organization, law, religion, education, recreation, and so forth , fro m th e economic aspect. Ye t even i f this wer e possible the question woul d remai n whethe r w e coul d not produce an arrangement mor e adequate to th e present proble m tha n this enumeratio n based on th e qualitative contents of th e different domains o f culture . A s w e are dealing here wit h questions o f equilibrium ,
‘the mor e adequate’ arrangement woul d clearly have to be in th e nature of a weighte d or graded order, i.e. a scale of economic values.
Ho w are w e t o buil d u p this graded order o f values? Economists agree tha t economic incentives can be arranged in an order of intensity.1
Ye t economics can express th e intensity of incentives only in terms of the concrete efforts expended on thei r satisfaction. Broadly speaking, the existence or intensity of wants is derived fro m th e fact tha t the y are satisfied, an d tha t certain wants are satisfied rather tha n others.2 Thi s circular argumen t precludes the possibility of studying wants apart fro m thei r concrete satisfaction. As anthropologists, havin g defined economic relations as relations between means an d th e satisfaction of culturally accepted needs, we must presuppose tha t th e scale of values has to some extent at least an existence per sey tha t is, is of some permanence, an d is buttressed b y cultura l factors independentl y o f actual demands. With • out this theoretical separation of wants an d thei r concrete satisfaction we could never attemp t to fram e such questions as wh y certain forms of satisfaction are highl y in deman d whil e others appear to be neglected althoug h thei r ‘value’ is recognized by th e society. We could never poin t ou t tha t certain needs ma y be highl y value d in themselves, an d ma y yet be disregarded because of an insufficiency of means. Thes e are questions of th e rationality of wants, of the motives behin d wants, and o f th e efficiency o f socio-economic systems i n general (the mai n mani • festation o f whic h i s th e distributio n o f povert y an d wealth) . Thes e questions, thoug h admittedl y beyond th e scope of economics in th e strict sense, deeply concern th e student of society.3
W e cannot demolish th e circular argumen t o f economics, whic h is, i n
1 L. Robbins, An Essay on the Nature and Significance of Economic Science,
- P- 86.
2 D. M. Goodfellow, Principles of Economic Sociology, 1939, p. 28.
3 Th e concept of poverty and wealth depends on the possibility of comparing the magnitude of incomes and forms of satisfaction secured by the income of different individuals. An inter-individua l comparison of this kin d is not ad• mitte d by theoretical economics. See L. Robbins, op. cit. , pp . 137-40.
336 ECONOMIC BALANCE: MEANS AN D WANTS
fact, intrinsic to all generalizations about human behaviour. But for the purpose of sociological analysis we must attempt to extend this circular argument over as wide a range of interconnected phenomena as possible. Wants are discovered in the way people behave; but in this discovery we are, unlike the economist, not bound by statistical evidence alone, but can with the help of our field methods study the actual behaviour of individuals in all its aspects. Behaviour comes to include also verbal behaviour and expression of beliefs, convictions, and ideals. Verbal behaviour may thus also supply us with evidence of the intensity of incentives. We are driven to admit the existence, side by side with ‘concrete’ evaluations as revealed in economic decisions, of diffuse
‘theoretical’ or ‘moral’ evaluations which may be at variance with the former.
We are, then, operating with a two-dimensional system of wants and values. It has, I admit, no validity in the economic universe of discourse.l Seen from the economic point of view the ‘theoreticar evaluation merely represents the weaker incentive which, in the eventual economic decision, has been subordinated to the stronger and thus relegated to the sphere of ineffectual theoretical concepts. But the acceptance of the two-dimen• sional scheme is indispensable to the adequate description of human conduct. Th e reality which it defines is primarily of a psychological nature: it embraces the motivations and repressions behind the economic decision, and the significant attitudes and responses which accompany it—hope and regret, disdain or resentment, expectations of betterment or the consciousness of failure. Yet every one of these psychological reactions may become a concrete determinant of conduct and thus a potential drive for readjustment either in the narrow sphere of social mobility, i.e. in attempts of individuals to improve their position, or in the wider sphere of collective attempts to bring about (by force, if necessary) changes in the socio-economic system. We need say no more on this point: it is illustrated abundantly in the history of society.
We will make two further admissions: first, that the divergence between theoretical and actual evaluation is frequently the symptom of a merely temporary lack of adjustment, and thus of social change; and second, that the divergence is partly due to a certain ambiguity in the concept we are using of ‘wants’ or ‘needs’. Th e first does no more than underline the significance of the phenomenon to the student of society. As regards the second, the following examination of the nature of the divergence will show that the ambiguity in the concept of ‘wants’ is due, not to inexact formulation, but to the nature of society. The divergence between actual and theoretical evaluation occurs in three different forms.
(i ) The simplest and most familiar form of this divergence is that owing to time-lag. Th e actual evaluation frequently takes account of changed wants, while the theoretical evaluation still maintains pre- change standards.
1 I suspect, though, that this dual definition of ‘concrete’ and ‘theoretical’ evaluation forms a parallel to a similar distinction made by economic theory, namely, between the ‘real value’ of commodities and their ‘exchange value’.
ECONOMIC BALANCE: MEANS AN D WANTS 337 (2) Th e divergence may be due to a differentiation withi n the struc•
ture of the group whose economic decisions we are studying. To put it
schematically: individuals act always as members of groups; but every
individual is at once member of different groups of progressive inclusive-
ness—family, kinship group, local group, tribe , state, &c. His conduct
may thus be influenced by possibly conflicting standards of evaluation
obtaining in the different groups. Hi s economic decision may follow one
standard of evaluation, while he relegates an alternative standard whic h
he is either unable or unwillin g to carry into action to the plane of
‘theoretical’ values. We may then find that the ‘concrete’ values adopted
by individuals as members, for example, of the narrower group, appear
as ‘theoretical’ values in the larger, comprehensive, group, and vice
versa. Disregard of the alternative standards of evaluation owing to
inability is typical of most forms of class stratification and, potentially,
of class consciousness (i n the economic sense). For here we see groups
act according to evaluations derived from the sufficiency of means in
their section, but yet remain conscious of wants and satisfactions obtain•
ing in other, more fortunate, strata of the group at large. Disregard of
alternative standards of evaluation owing to unwillingness, on the other
hand, wil l mostly be found to indicate differences in social insight, whic h
varies wit h position in the social group, wit h age, and experience. Thu s
the young labourers who preferred to purchase clothes and other
temporary attractions rather than save money for marriage acted accord•
ing to the ideals of the younger generation and disregarded the value
which , as members of the kinship group, they should have recognized—
values, moreover, whic h are upheld by the executive organs of the
group, the elders and the family heads. We shall base our examination
of the Nupe scale of values mainly on the widest range of social insight
and (unless expressly stated) shall assess economic wants fro m the view•
point of the family head.
(3) A final, most elusive, divergence is that between felt wants and abstract needs.1 We must realize that the social significance of organized human conduct, shaped as it constantly is by convention and custom, is not exhausted in the attainment of a certain single satisfaction (which alone was aimed at), but normally implies the effectiveness of other wants and needs besides. Thei r satisfaction is attained, as it were, automatic• ally, wit h the satisfaction of the conscious want, and may be embodied in the very conventional form whic h the satisfaction must take. Radcliffe- Brown has shown that the paramount ubiquitous need of this kin d is that inherent in the structure of society itself, of integration and stability. Now , concrete evaluation may in certain cases ru n counter to that need, whic h woul d be relegated to the ideological validity of theoretical wants. These social needs are rarely present in the minds of individuals—except in those of far-sighted individuals acting as execu• tive organs of their group. Rather the student of society hypothesizes their existence; for to hi m they reveal their existence and effective-
1 Goodfellow assigns the study of ‘wants’ to economics proper, and the study of ‘needs’ to a separate discipline, which he calls Social Economics (op. cit., P. 275)-
Z
338 ECONOMIC BALANCE: MEANS AN D WANTS
ness indirectly, through certain significant changes in the stability of social life.
Wit h these methodological viewpoints in mind, we may now attempt to build up our scale of values. We cannot hope, of course, to make it exact in every grade or detail; but we can trace its outline which gives us the extremes of the scale—its maximum and minimum points—and, between them, certain well-defined marks. I propose to adopt a three• fold approach.
(i ) We find the ‘minimum point’ of our scale in the demands upon resources which refer to the satisfaction of the physiological necessities of life—’primary needs’, in Malinowski’s terminology—above all, nutrition and shelter. Wit h Malinowski, we realize that these needs never exist as such in society, that is, in the form of undisguised elementary impulses, but already appear in a certain conventionalized form, shaped and elaborated by culture.1 Closely linked with these wants is another group which we may call, with Malinowski, ‘derived’ wants, in the sense that they refer to the maintenance of the productive organization by means of which the ‘primary needs’ are satisfied.2
(2) On this second stage we shall examine certain more or less fixed demands upon resources which are involved in institutions and customs and which are accepted with little variation, irrespective of status or wealth, throughout the society. Typical instances of these demands are bride-price and taxation.
(3) We study the wants and forms of satisfaction apparent in the conduct of the recognized elite of the society. They define the ‘maxi• mum point’ of our scale, and we may take them to represent the ideals and generally desirable things—the things ‘one ought to be able to afford’—in the opinion of the society at large.3 I must add that this last assumption is justified only if we can prove that no rift runs across the ideological framework of the society separating the desires and tastes typical of the various social strata. We must reckon with the possibility of individuals of the lower social strata being well content with their
‘station’; or of tastes and desires of higher social strata showing a caste• like exclusiveness. We shall see later that one of the main characteristics of Nupe society lies in the absence of a self-limitation of the former kind; the latter kind of rift, on the other hand, is negligible, and is restricted mainly to the prerogatives of royalty, for example, to the traditional in • signia of rulership, the rules of dress and food of the Nupe kings, &c. In modern Nupe even these caste-like prerogatives have mostly dis• appeared.
1 I cannot argue here the question whether or not sexual satisfaction should be included among these ‘primary needs’; I have put sexual desires in the second group of wants mainly because they are subject to a much more complete cul• tural ‘elaboration’ and institutionalization.
2 It will be seen that this definition is narrower than Malinowski’s definition
of ‘derived needs’, which can, in fact, be extended over the whole field of culture.
3 In this assumption, which is justified by the stratified structure of Nupe society, we differ from Dr . R. W. Firth, who states that ‘there is no final organiza• tion in primitive societies in the sense of a single individual or ruling group of individuals who are in a position to dictate all choices’ (Primitive Polynesian Economy, p. 24).
ECONOMIC BALANCE: MEANS AN D WANTS 339
Bu t we must reckon also wit h another ideological rift i n tastes and desires, whic h woul d affect all three stages of our investigation, namely a rif t whic h might be involved in the complex and composite nature of Nupe culture. It woul d separate, not social strata, but different sections of the population, different cultural groups withi n Nupe society. A rif t of this kin d does, in fact, exist wit h respect to the different cultural standards valid in tow n and country, in the Mohammedanized capital and the pagan or semi-pagan districts.
STANDARDS OF LIVING : NECESSITIES OF LIFE
Th e extent to whic h the economy of an individual or a family group is
able to incorporate into its ‘ balance’ wants placed lower or higher on the
scale of cultural values gives us what is generally called the ‘standard of
living’ . We shall find that it is possible to arrange these varying standards
in certain well-defined groups or grades, and to embody them in a com•
prehensive economic stratification of the society.
I t wil l be best to begin wit h concrete illustrations of the economic balance achieved in Nupe households. Ou t of fifty household budgets whic h I have recorded I propose to select five representative examples, taken from different localities and differently organized household groups, in whic h we can study the main variations that occur. Al l these examples refer to peasant households. Unfortunately I found it im • possible to work out budgets of this kin d also for craftsmen and ‘free professions’ (wit h two exceptions to be quoted later). Thei r extremely flui d economy presented great difficulties; to reduce, in their case, in • come and expenditure to exact figures woul d have demanded much more time than I was able to devote to this task. However, just because these types of economy are more fluid and more easily adjustable to varying wants, they are of less interest to us than the more restricted and more rigi d economy of the agriculturist.
Le t me say a few words about the method by whic h I arrived at these records of Nupe household economy. I have examined households in Bida, Doko, Lemu , and Kutigi . Th e figures referring to the total farm- yield , and to the proportion of the produce kept for consumption or set aside for sale, were ascertained by inquiry . Th e nature of Nupe agri• culture—the series of successive harvests, and the widely varying arrangements wit h regard to storage and sale of produce—precluded any more direct approach (e.g. by measuring the yield or the stores). How • ever, there were important safeguards: first, as we have seen, the Nupe farmers are well acquainted wit h the use of standard measures, and
indeed themselves use them to calculate their yield.1 Moreover, after a
number of preliminary inquiries (i n whic h I was helped by reliable informants) I could already for m an opinion as to likely averages and individua l deviations. Finally, whenever any dubious figures emerged I could verify them by independent inquiries of other household members. In this respect the tendency of the Nupe farmers to keep track of the’ contributions of the various family members to the com• mo n store proved of considerable assistance.
1 See p. 315. In the charts below the terms ‘load’ and ‘bundle’ are used as equivalents.
34° ECONOMIC BALANCE: MEANS AND WANTS
Household A
Tunga near Bida. Labour unit: one man. Family of four
Kept for
Crops and Total household Additional
trees yield consumption Sold income
Bulr.-millet S loads All Mat-weaving, for
yd. apiece.
Late millet 3 ,. Total earnings 5s.
Cassava j ‘?
Locust-beans 5 loads All, for is. 3d.
Income fro m farm produce
Tota l income
Mone y expenditure: Ta x
s. d.
6 3
11 3
9 6
Household B
Bida. Labou r unit : father and boy. Family of three
Sorghum
Beans
Rice
6 „ ,, All, for’25.
31 gourds 10 gAoiui rds 21 gourds for
relations for tax.
Ya m 6 loads
Locust-beans
3s. 6d.
Sold the whole tree for fell• ing to the Bida brick• works, for
- 6d.
Income from farm produce . Tota l income (incl . money loan) Money expenditure: Ta x
ECONOMIC BALANCE: MEANS AN D WANTS 341
Household C
Tunga near Bida. Labour unit : father (old) and married son (invalid) . Famil y o f fiv e
Kept for
Crops and Total household Additional trees yield consumption Sold income
Bulr.-mille t Bad crop N o yield Sold 2 young goats for 35. each, in order to obtain the rest of the tax.
Late mille t 10 loads 9 loads, for
9′. ; 1 load given as dza- nkd to Maku
Sorghum 5 .. 1 load 3 loads, for
3s.; 1 load
given as dza-
nkd to Maku
Beans 7 ,, All , for s\d.
Locust-beans 9 „ All , for 25.
Income fro m farm produce
Tota l income
Mone y expenditure: Ta x
s. d.
15 5 i
2 1 5 ‘
2 1 o
Household D
Kutigi . Labou r unit : father and tw o sons, one of them married. Family of seven
Kept for
Crops and Total household Additional trees yield consumption Sold income
Bulr.-mille t 20 loads Al l Famil y head: occa•
Late mille t 15 „ 13 loads 2 loads,
1s. 4d.
Sorghum 2 0 „ 10 ,, 10 loads,
7s. 6d.
for sional hunting .
Hides sold for 2d.
for to 4d. apiece, total earnings 1s. 2d.
Ya m 7 t> Al l Sons: mat-weaving, Cassava ? ,, for 6d. apiece; total Beans 2 loads . 4 earnings approx. Ground-beans 1 load 105. p.a. (i.e. buca Sw. potatoes little income). Each son Ground-nuts 5 loads Onl y seed Rest, for 35. 6d. contributes 35. to • Bananas ? Al l wards the family Locust-beans 5 loads All , for is. 3d. tax.
Palm-kernels ? All , for £ 1 or
Kola-nuts 5 loads
25s.
All , for 6s.
3d.
._
£ s. d.
Income from i arm productt . 2 4 10
Tota l income [incl. contritmtion from so tis) . . 2 12 1
Mone y expenc iture : Ta x
Egbeywork .
. 1 13 o
6 8
Fina l instalment of bride-price for younger son 10 o
342 ECONOMIC BALANCE: MEANS AN D WANTS
Household E
Lemu . Labou r unit : father and four sons, tw o o f the m married . Famil y o f ten
Kept for
Crops and Total household Additional trees yield consumption Sold income
Bulr.-mille t 50 loads 20 loads 30 loads for
15′-
Lat e mille t 30 „ 10 „ 20 loads for
10s.
Sorghum 40 ,, 20 „ 20 loads for
- 6d.
Famil y head: trade in rams and sheep. Tota l earnings ap- prox. 10s. p.a.
Sons: buca farm – work . Tota l earn•
Beans 6 „ Al l ings approx. 4s. Ya m ? A littl e 8 loads for 4s. p.a.
Cassava ? Eaten at once
Mone y expenditure: Ta x . . . . .
£ d.
• 3 6 approx. 4 o
. 2 o
Egbe wor k . . . . . . 6
Name-givin g ceremony for grandchild o
Firs t instalment of bride-price for younger son o
Th e construction of these household budgets is designed to show the main division of resources into those which are utilized in kind, i.e. for food consumption, and those utilized for exchange, i.e. for the acquisi• tion of commodities and the satisfaction of other, non-nutritive, wants. We have said that the latitude in satisfying wants beyond the ‘ minimum level’ gives us the standards of living and economic stratification.
Now, it is characteristic that Nupe society, which is so conscious of its class stratification, is equally conscious of its economic stratification and the varying standards of living. Even the principle of the ‘minimum point’ of the scale of values, the existence of a minimum cost of living, is clearly understood and, what is more important, generally accepted. I have worked out questionnaires referring to this minimum standard of living and collected answers from twenty men in Bida and the villages, of different classes and different degree of wealth. They all were in agreement on the essential points. Expressed in agricultural produce, i.e. staple food crops, calculated for a small family of four, the minimum annual cost of living amounts to 12 to 15 bundles of grain (of any variety) or 90 loads of yam, if the household relies on one of the two staple food crops; or, if both are combined, 6 to 8 bundles of grain and 40 to 50 loads of yam. Expressed in money, 1½d. per day is the lowest cost of living per adult. But this food, as my informants added, will hardly be nutritive, it would only ‘make the belly swell’. Wit h more nutritive food, the lowest cost of living is 2d. per day. Even this is not yet a high standard, for it excludes na ma re na, things ‘that taste nice’, e.g. sweet• meats, savouries, and suchlike. For a family of four or five I was given
ECONOMIC BALANCE: MEANS AN D WANTS 343 the following daily ‘menu’ corresponding to a standard of living slightly
above the minimum:
Morning meal: corn-water or gruel with ‘ bread’ (fura) . . . 1½d. Midday meal: cassava, beans, or rice . . . . . . 3d. Evening meal: corn porridge . . . . . . 1½d.
Tota l cost 6d.
If fats, shea-butter, or palm-oil are added in form of sauce, the bill in • creases by another i\d.; meat or fish, the addition of which would raise the meal to a much higher ‘ standard of living’, cost 6 d. a day extra. For example:
orning meal: corn-gruel and ground-nut cakes . 1½d-2d. idday meal: corn porridge . . . . . 1½d.
palm-oil, greens, and spices 1-2’/ .
fish 3d.
/ening meal: r i c e . . . . . . Ad.
palm-oil, &c. (partly left over from midday meal) 0.3d
. 2-3d.
1s. 1.3d-1s. 3.8d.
Calculating the cost of food for a family of this size for a week and taking into account that the morning meal often consists of food left over from the previous day, and that meat, fish, and certain savouries are only bought occasionally, we arrive at the sum of 8s. 9d. per week, or 1s. 3d, per day.
These figures refer to the cost of living in Bida, where the majority of the population have to buy their whole food. But in the village, food other than the staple crops—including fats, fish, and meat—has also to be bought for money. On the other hand, cost of living in the villages is slightly cheaper, as is shown in the following example, which is again calculated for four or five persons, and corresponds to a fairly high stan• dard, comparable to the elaborate Bida ‘menu’ just quoted:
Morning meal: corn-gruel and ground-nut cakes . . . . 0.8d . Midday meal: corn-porridge and fish, with palm-oil, greens, and
s p i c e s . . . . . . . . . 4 d .
Th e standards of living in the village are also lower than in Bida: an average standard in the village would correspond to a low standard in Bida. Dried fish rather than meat is the main item on the menu, meat being eaten not more than once a week or, in poorer households, once in several weeks. Rice, which appears fairly regularly on the Bida menu, is rare in the, village; sorghum-beer, on the other hand, which may replace a regular meal in the village, does not figure on the menu of Bida house• holds. But if the very high standard of living characteristic of the Bida upper classes does not occur in the peasant households, the same is true of the other extreme, the very low standard of living; for although I have found many households in or near Bida which live on the very minimum level, these very low standards seem absent in the village.
Let me give a few concrete examples to illustrate the food economy in
344 ECONOMIC BALANCE: MEANS AN D WANTS
typical Nupe households. They are taken from four households in Bida (Nos. i and 2) and in the country (Nos. 3 and 4). ‘Menu’ No. 1 is that of a wealthy noble household, No. 2 that of an average trader’s household, No. 3 that of a prosperous peasant household in Kutigi (household D in the list of household budgets given above), and No. 4 that of an average peasant household in the same village. All four examples were obtained about the same time, the end of May and beginning of June. B = morn• ing meal (i.e. ‘breakfast’), M — midday meal, E — evening meal.
No. 1
19-5-1936.
B. Corn-water and corn-cakes. M . Rice and fish .
E. Rice and meat with palm-oil, greens, and spices.
20.5.1936.
B. Corn-gruel.
M. Rice and meat with shea-butter, onions, and spices.
E. Corn-porridge and fish, with palm-oil, greens, and spices.
21.5.1936.
B. Corn-gruel and bean-cakes. M. Rice and chicken.
E. Corn-porridge and fish, with greens and spices.
22.5.1936.
B. Corn-water and bean-cakes. M . Rice and fish .
E. Rice and fish, with palm-oil, greens, and spices.
23.5-I936.
B. Corn-water and bean-cakes.
M. Rice and fish, with palm-oil, greens, and spices.
E: Rice and meat, with palm-oil, greens, and spices.
24.5.1936.
B. Corn-water and corn-cakes.
M. Corn-porridge and fish, with palm-oil, greens, and spices. E. Rice and meat, with palm-oil.
25.5.I936.
B. Corn-gruel.
M. Corn-porridge and fish, with palm-oil, greens, and spices. E. Rice and meat, with palm-oil.
18.5.1936.
No. 2
B. Corn-water and ground-nut cakes.
M. Corn-porridge and fish, with palm-oil, greens, and spices. E. Rice and meat, with palm-oil.
I9.5.I936.
B. Corn-water and ground-nut cakes. M. Cassava with palm-oil.
E. Corn-porridge and fish, with palm-oil, greens, and spices.
20.5.1936.
B. Beans cooked in thick palm-oil, with salt and pepper. M. Boiled cassava, with palm-oil.
E. Corn-porridge and fish, with shea-butter, greens, and spices.
21.5.1936.
B. Bean-cakes.
M. Corn-porridge and fish, with shea-butter, &c.—left over from previous day.
E. Rice and meat, with palm-oil, greens, and spices.
22.5.1936.
ECONOMI C BALANCE : MEAN S AN D WANT S 345
B. Corn-water and corn-cakes. M . Corn-porridge .
E. Rice and meat, wit h palm-oi l and spices.
23.5.1936.
B. Corn-water and corn-cakes. M . Rice and fish , wit h palm-oil .
E. Corn-porridg e and fish, wit h palm-oil , greens, and spices.
24.5.1936.
B . Ho t corn-water and corn-cakes. M . Sweet potatoes and palm-oil .
E. Cor n porridge and fish, wit h palm-oil , greens, and spices.
Not e that ground-nu t cakes, thoug h eaten repeatedly for the mornin g meal i n this household, are absent i n Men u No . 1 . I n the latter household i t was explained to me that ground-nuts were ‘ou t of season’, and no longer good for makin g cakes—a culinary distinctio n whic h the less well-to-d o (and less sophisticated) households do not follow .
No. 3
3151936 .
B. Corn-grue l and beans.
M . Ya m and beans, wit h palm-oil , salt, and pepper.
E. Corn-porridg e and fish, wit h palm-oil , greens, and spices. .
1.6.1936.
B. Corn-grue l and beans.
M . Ya m and beans, wit h palm-oil .
E. Corn-porridg e and meat, wit h palm-oil , greens, and spices.
2.6.1936.
B. Corn-grue l and beans.
M . Corn-porridg e and meat, fro m previous day.
E. Rice and fish, wit h palm-oil , greens, and spices.
3.6.1936.
B . Corn-gruel .
M . Boiled yam wit h palm-oil .
E. Corn-porridge and fish, wit h palm-oil , greens, and spices.
4.6.1936.
B. Corn-grue l and beans.
M . Beans wit h palm-oil , salt, and pepper.
E. Corn-porridge and fish, wit h palm-oil , greens, and spices.
5.6.1936.
B . Corn-gruel .
M . Corn-porridge and fish , wit h palm-oil , greens, and spices. E. Rice and fish, wit h palm-oil , greens, and spices.
6.6.1936.
B . Corn-gruel .
M . Boiled yam, wit h palm-oil .
E. Corn-porridge and fish, wit h palm-oil , greens, and spices.
No. 4
13-6.1936.
B. Corn-water and ground-nu t cakes. M . Cassava.
E. Porridge and fish, wit h palm-oi l (no greens).
14.6.1936.
B . Corn-porridge , fro m previous day.
M . Corn-porridge, left over fro m mornin g meal. E. Fresh corn-porridge wit h greens (no fish).
346 ECONOMIC BALANCE: MEANS AN D WANTS
15.6.1936.
B. Corn-porridge and very little fish (bought for 0.1d.), with palm-oil, greens, and spices.
M . Sorghum beer.
E. Corn-water and ground-nut cakes.
16.6.1936.
B. Nothing.
M . Raw cassava.
E. Corn-porridge and fish, with palm-oil, greens, and spices.
17.6.1936.
B. Corn-gruel.
M. Corn-gruel and ground-nut cakes.
E. Corn-porridge and fish, with palm-oil, greens, and spices.
18.6.1936.
B. Ho t com-water and ground-nut cakes. M . Corn-porridge.
E. Corn-porridge, left over from midday meal.
19.6.1936.
B. Cold corn-water and corn-cakes. M. Cassava boiled in water.
E. Corn-porridge and palm-oil (no fish).
In examining the budgets from the viewpoint of food economy, we
find that in this one domain of wants they are already representative of
widely varying standards of living. Measuring the produce available for
food consumption in the various households against the standard of 12
to 15 bundles (loads) of grain per family of four, or 3 to 4 bundles of
grain per head, we obtain the following relations:
Household A: family of four—13 loads, or over 3 loads per head.
Household B: family of three—12 loads (also yam), or over 4 loads per head.
Household C: family of four—1 load, or ¼ load per head.
Household D: family of seven—43 loads of grain (also yam), or over 6 loads per
head.
Household E: family of ten—50 loads (also yam), or over 5 loads per head.
In other words, we find poverty, i.e. ‘insufficiency of means’, a certain marginal sufficiency, and also ample and prosperous, standards of living.
We can correlate the prosperity in this sense with certain other social factors.1 Th e most prosperous households are found in Trans-Kaduna and in the country north of Bida (e.g. Lemu) where land is fertile and plentiful. Households with a large labour-unit are invariably more prosperous than households with a small labour-unit. Th e budget which expresses real poverty and insufficiency (C) comes from Bida, and from a small labour group whose efficiency was abnormally reduced: one of the two men who formed the labour unit was old and past work in age, and the other an invalid. I have spoken on different occasions of the detrimental effects upon production of the spreading dissolution of the large labour-unit. That a situation such as this could occur, that this prima facie inefficient productive arrangement which leads an old man and
1 It will be understood that in this as in similar general statements I am referring not only to the selected examples, representative though they are, but to the whole material which I have examined.
ECONOMIC BALANCE: MEANS AN D WANTS 347 an invalid to work on their own, could at all be tolerated, reveals most
conspicuously the inherent weakness in present-day labour organization.
We turn now to the examination of the money surplus achieved in
the various households. Here we find the same wide variations as in the
examination of the output which was retained for food. But the two
ranges of variation do not run parallel; which means that the Nupe
peasant does not divide his above-minimum-output equally between
nutritive wants and wants demanding money surplus. He may accumu•
late food-supplies up to what we may regard as the maximum at the cost
of the money surplus (as in D) ; or he may go below the maximum in
food-supplies in order to obtain a higher money surplus (as in E) ; or he
may be forced to fall even below the minimum standard of food-supplies
for the sake of a certain necessary money surplus (as in C). In order to
understand the full significance of this we must anticipate here another
demand upon resources which, though not expressing ‘primary needs’,
possesses in native economy the same urgency and inevitability—tax.
This demand was, in fact, already embodied in the budgets which we
have drawn up. For when measuring Sufficiency of means’ with regard
to the primary necessities of life we measured not ability to reach an
adequate level of production, but ability to retain an adequate supply
of produce after this other rigid demand had been met. Normally, every
farmer would retain at least the necessary minimum supplies, which is
easily achieved in the ample husbandry of prosperous households, but
may prove impossible in the households which keep close to the margin.
Here the free planning of the balance between essential food stores
and produce for sale may be restricted by the necessity of providing
the money ‘surplus’ for tax, even (as in C) to the degree of reducing
food-supplies below the minimum level. Only after deducting the tax
can we thus speak of ‘real’ surplus, in the sense of money resources
that can be applied freely to wants, and only with respect to this final
surplus can we speak strictly of sufficiency and prosperity.
Wit h regard to surplus in this sense the correlation between large labour-units and high degree of ‘sufficiency’ is no longer true. In house• hold D the family head could secure a money surplus only by claiming contributions from the buca work of his sons—against the rules of efako collaboration. In a great many cases (not recorded here) large labour- units, though able to produce a large surplus output, could not secure a corresponding money surplus. Th e explanation lies in the low market- price which ordinary agricultural produce commands. If large quanti• ties have to be sold at once—as they must be where a large sum of money (e.g. for tax) has to be found by a single producer (i.e. the head of the labour-unit)—the price will be even more unfavourable. Only where the production comprises commercial crops, or generally crops com• manding a high market-price, will it be possible to combine both aspects of surplus. We find, on the other hand, that households in which a handi• craft is practised in addition to farm-work attain a higher measure of prosperity than households of farmers pure and simple. A profitable handicraft can procure ready money for certain urgent expenses, above all, tax, and thus free the household from the necessity of selling a too
348 ECONOMIC BALANCE: MEANS AN D WANTS
large proportion of the produce, or selling it in a hurry, at unfavourable prices.
From nutritive wants we proceed to the wants involved in the ‘primary
need’ of shelter. Th e upkeep or building of houses represents a want
which, occurring at much longer and more irregular intervals, does
not figure conspicuously in the annual household budget. In the village,
upkeep and building of houses is mainly a question of co-operative
labour, involving little or no money outlay. Smaller house repairs,
above all thatching, become necessary every three to five years; more
complete repairs, such as rebuilding of huts or of the compound wall,
are reckoned to occur every ten to twelve years. Th e small repairs are
carried out by the household group, and take rarely more than two or
three days. For larger repairs the help of neighbours and relations must
be enlisted on egbe scale; they last much longer, according to the body of
workers that can be mustered, and the work may often have to be inter•
rupted for a day or two if the workers have other tasks in hand. Most
building work in the village is left till the beginning of the rains, when
the ground (from which the clay for building is dug) is soft, the grass
for thatching not too dry (such grass is said to last less long), and when
the people are still free from more exigent farm-work. We may add that
the building of more elaborate houses of the kind that is found in the
town and also in certain villages entails the employment of expert labour,
and thus money expenditure, which varies widely with the style and
scale of the work.
Finally, we add the ‘derived wants’ involved in agricultural produc• tion. Under this heading we include repairs and purchase of tools and the financing of egbe work. In the same category falls the payment of a fixed land rent in kind which obtains in the area round Bida (as in our example C). I have only partly been able to include these wants in the household budgets. However, the widespread efforts to economize, for example, on the cost of egbe work, or the neglected appearance of many of the sumptuous Bida houses, reveal the difficulties which many house• holds find to-day in coping with these wants.
STANDARDS OF LIVING : CULTURAL REQUIREMENTS
The next step in our inquiry is to ascertain the demands embodied in cultural standards which the Nupe peasant has to meet out of the sur• plus (if surplus there be) derived from his production. We can group these demands under three headings, (i ) Demands implied in kinship duties. In so far as the household or kinship group coincides with the labour-unit these demands overlap with those we listed under
‘maintenance of labour’, (2) Demands implied in duties towards the community—religious, political, or social. (3) Certain personal re• quirements which are regarded, by convention, as essential, and even indispensable.
Under the first heading we list the expenses of kinship ceremonies and feasts—the name-giving ceremony for a new-born child, wedding feasts, and burial and funeral ceremonies. These expenses are borne primarily
ECONOMIC BALANCE: MEANS AN D WANTS 349 by the family head in whose house the ceremony is performed; but he
can also count on contributions from the various relations and friends
who attend the ceremony as guests of the house. We have noted the
modern tendency among peasants to replace the traditional gifts in kin d
wit h contributions in money, whic h are customary in Bida. These ex•
penses vary widely wit h wealth and status. Th e minimu m costs—
whic h interest us here—amount to about 35. to 55. in the name-giving
ceremony and 8s. to 10s. in marriage and funeral ceremonies. In the case
of junio r family members this group of kinship expenses also includes
certain occasional gifts in kin d to elder relatives, especially the family
head. In one Doko household (not quoted among our examples) a young
farmer who worked on his own gave his whole produce of beans (worth
zs.) to his father and elder brother. In another household, in Bida, a
man shared his crop of maize wit h his elder brother, who was the family
head. Family heads, on the other hand, have to provide for the costs of
education of their children or younger brothers (so far as it involves
expenditure), e.g. the apprenticeship in the workshop of a craftsman, or
the attendance at a Koran school or the Government school. They must
bear the expenses that might be caused by illness—the purchase of
medicines or of the services of diviner or barber-doctor. ‘Pocket-
money’, finally, for children or junio r family members is, as we have
seen, supplied in the form of buca work, which they are encouraged to
take up.
Bride-price
Th e largest item under this heading is the payment of bride-price, whic h devolves, in traditional practice at least, on the head of the family section that is united in the labour-unit. I have pointed out already that the bride-price also involves the accumulation of wealth, and com• paratively long-range planning and saving.
Th e bride-price is always reckoned in money, although it may be paid partly in kind ; the gifts in kind , however, which may form part of the bride-price, such as cloth and beads, all have a definite money value, and must be bought for money. Th e Nupe bride-price, called ewo yawo,
‘money of marriage’, comprises a series of payments of varying amount, each of whic h is known by a special name: we have, first, the gift of the Greeting; we have further two gifts called Small Thanks and Great Thanks, respectively, the latter which represents bride-price proper, being sometimes referred to by the modern Hausa-Mohammedan term sadeiki; then there is the Money for Henna, to buy the henna wit h whic h the bride wil l adorn herself for the marriage ceremony, the Money for Pots, wit h which pots and cooking utensils wil l be bought, and finally a small gift to the bridesmaids who accompany the bride when she enters her husband’s house for the first time. In the country these last two gifts- are a comparatively recent introduction, having originated and spread fro m Bida. In addition to these various ‘part payments’ there is an annual gift of produce whic h is presented by the father of the bride• groom to the father of the bride during the period between the formal betrothal and the final marriage ceremony. Th e whole bride-price is
350 ECONOMIC BALANCE : MEANS AN D WANTS
rarely paid over at once; as a rule the various part-payments are made in the course of three to five years, in the order jus t given. Th e first gift, the Greeting, is meant to seal the betrothal and the agreement of the parents of bride and bridegroom. Th e various gifts listed after the Great Thanks are left til l immediately before the marriage ceremony. Onl y the tw o largest ‘instalments’, Small Thanks and Great Thanks, can be paid partly in the for m of gifts of cloth or beads; all other gifts must be in money. Le t me give a concrete illustration. Th e bride-price of N. , of Kutigi , who married in 1934, consisted of the following payments, all of whic h were in money:
£ s. d.
Greeting . . . . . . . . . 1 6
Small Thanks . . . . . . . . 1 0 o
Great Thanks . . . . . . . . 3 0 0
Money for Henna . . . . . . . 5 0
Money for Pots . . . . . . . . 1 0 o
Gifts t o bridesmaids . . . . . . . 2 6
Four annual gifts of food, to the total value of. . . 10 o
Total £41 9 o
Bride-price in Nupe has undergone great changes. Fro m £3 in pre- Britis h days it has risen later to £10, £15, and even £20. Recently it has fallen again to a more modest sum, between £5 and £10. Th e height of the bride-price has, in fact, caused the Administration to intervene in
1934, and to lay down a maximu m amount of £7. On the whole, the people welcomed the reduction. But it has not been adopted uniformly ; I have found a bride-price of £9 and £10 stil l being paid two years after the law was enacted. In very poor households, on the other hand, the reduction has merely academic significance, as these people may not be able to afford even that amount. My informations were that originally the bride-price represented a fixed sum—gba shita, 120,000 cowries (£3 )
—whic h di d not vary with , nor depend on, prosperity or social status, as it does to-day. Similarly every item of the bride-price was fixed tradi • tionally, the amounts in cowries being frequently used as alternative terms for some of these payments. Thi s is the series of payments in the traditional Nupe bride-price:
Gba (lit. 2,000 cowries—a preliminary gift absorbed to-day in the
£ s.
Total £ 3 6
Wit h the rise, and subsequent fall, in the bride-price its amount was made dependent upon the varying standards of living . It is significant that the large items in the bride-price, Small and Great Thanks, have remained comparatively unchanged; the changes affected chiefly the smaller gifts, new ones being included, or increased or decreased in amount. No w a large bride-price, whic h comprises all these various gifts, is stil l the ideal. But certain new ways have been found to harmonize the ideal wit h lower standards
ECONOMIC BALANCE: MEANS AN D WANT S 351 of living. To-day a distinction which did not exist in former times
is universally recognized, between bride-price entailed in ordinary
marriage and bride-price paid in the case of yawo dengi, marriage in the
family: the latter commands a lower bride-price, approximately £3 , i.e.
the original uniform bride-price. In every case in which I examined the
relation of ‘ marriage in the family’ to outside marriage my informants
emphasized that there were more family marriages to-day than formerly
because it had become so difficult to pay the high bride-price. We see,
then, that the bride-price was not simply reduced in amount uniformly;
rather a certain context was introduced—invented almost—in which
a lower bride-price could be legitimately accepted. Th e lower bride-
price, which would run counter to traditional cultural standards, was
thus given a certain qualified sanction by the society. However, even the
low bride-price cannot always be provided by the family head. And here
we find another adjustment to the lower standards of living, namely, the
spreading practice to pay bride-price only for the eldest son, while
younger sons must either contribute to, or bear themselves, the cost of
their bride-price. The traditional duty of the family head is expressed
only in a symbolic transaction: the young man who has to contribute his
own bride-price in many cases still hands over the amount to the family
head, who will then take all the necessary steps and carry through all the
negotiations.
To the bride-price we must add its counterpart, the dowry which the bride brings into marriage and which is paid by her father. Though paid over in one, immediately before the marriage ceremony, the dowry like the bride-price also implies accumulation of wealth over a long period. Th e amount of the dowry used to correspond exactly to the low, tradi• tional bride-price. Although it fell short of the exorbitant bride-price of recent years, it has been adjusted to the level of a moderately high bride-price (as represented in our example), which proves clearly the underlying intention to balance it against the bride-price even to a degree of mathematical exactitude. Unlike the bride-price, the dowry does not consist in money payments, but in various gifts and contribu• tions in kind. It is reckoned to provide the young bride with cooking utensils, pots, and clothes, sufficient to last her for the first four or five years of the marriage; it also provides food—very excellent food—for the newly wedded couple, so that for the first month or two they are indepen• dent of the earnings of the husband or contributions by his family. This is the dowry that corresponds to the example of bride-price quoted above:
5 pieces of cloth, worth between 55. and ys. each, representing a value of 1 10
Tw o months’ food-supply, worth . . . . . . 1 1 0
Total ~£i~S
Let us for the moment interrupt this description and examine the economic and sociological significance of the changes in bride-price. Th e question of the sociological significance or ‘function’ of bride-price
353 ECONOMIC BALANCE: MEANS AND WANTS
has greatly engaged the attention of anthropologists. Yet the various answers which have been found, showing no little disagreement, mostly disregard the fact that institutions never subserve one single purpose, but rather combine (‘amalgamate’ as Malinowski puts it) a number of social purposes in their structure. The overt purpose of bride-price is clearly that of securing to a man the various legal rights over a woman which are embodied in marriage. Marriage marks the end of certain rights held over the woman by her own kin, above all, her parents, and their replacement by the new rights vested in the husband and his kin. Bride-price in this sense is adequately defined as ‘an indemnification for
. . . a . . . surrender of rights’.1 However, as the bride-price is not supplied and received by bridegroom and bride themselves but by their parents or other senior family members, the transaction also has the two• fold social significance of actuating existing kinship bonds (between the bridegroom and his relations who provide the bride-price for him, and similarly between the bride and her family), and of confirming new kin• ship bonds which stretch, not between the two individuals, but between their families and kinship groups. The transaction of bride-price in this sense has the significance of a contract, to which the two families are the guarantors. The same applies to the parallel transaction of the dowry. Finally, a bride-price, or dowry, of considerable dimensions necessitates the accumulation of wealth and careful economic planning. Bride-price can thus also be said to provide a stimulus for production and economic enterprise in general—one of the motives which every society provides, and needs, to impel it to ‘produce, to muster, to count, and publicize its wealth’.2 This last aspect of the purposefulness of bride-price is, of course, not present in the conscious thought of the people themselves. It expresses a purpose or need read into the facts by the student of society. The first two purposes of bride-price, on the other hand, repre• sent conscious motives of conduct, and the degree of their consciousness becomes evident in the changes which bride-price has undergone in Nupe.
Now, in the traditional Nupe bride-price the aspect of’indemnification’ is greatly overshadowed by the other two aspects. As we have seen, wealth was not ‘given’ but merely exchanged in the strict sense of the word, the bride-price being balanced exactly against the dowry. When bride-price increased disproportionately, far outstripping the dowry, the element of indemnification and payment as such came to the foreground. The people themselves link this surprising rise with the new economic and political development under British rule: ‘Prices were high, and everyone had money.’ To some extent this was no doubt due to an in• creased demand, the greater prosperity showing itself in a rising scale of polygamy. Partly also the motive of ‘indemnification’ gained a new, additional, significance, that of display for the sake of prestige. But the higher bride-price also showed the effectiveness of the third, economic, factor: for in the highly stimulated economic system the traditional stimulus embodied in the bride-price had to be increased in order to keep pace with the new, higher, level of production and expenditure.
1 A. R. Radcliffe-Brown, in Iowa Law Review, xx, 1935, p. 295.
2 D . M . Goodfellow, op. cit., p . 132.
ECONOMIC BALANCE : MEANS AN D WANT S 353
Thi s interdependence of bride-price and level of production also ex• plains the limite d success, among the more prosperous sections of the population, of the attempts of the Administration to fix the bride-price at a lower and, as it were, arbitrary figure.
Indeed the development of bride-price under the economic stress of the last decade shows how firmly the new ‘want’ of higher bride-price had struck root. Th e solution whic h the people themselves worked out left the ‘indemnification’ and the ‘economic’ aspect of the bride-price intact; it achieved the necessary adjustment by reassessing the ‘socio• logical’ or kinship aspect. Th e contribution of the family to the bride- price of its junio r members was reduced to the symbolic transaction of the family head acting on behalf of the bridegroom who must himself provide the bride-price; above all, the new rules allowed a reduction in the bride-price in family marriages, i.e. marriages in which the marriage ties do not establish new, but utilize already existing kinship bonds.
We may mention here certain marriage rules customary among the upper classes of Nupe in whic h an outwardly similar result is achieved by a somewhat different process of reasoning. We have learned that bride-price is kept low in class endogamous marriages among the nobility ; a man of status may also marry his daughter to one of his bara, and himself pu t up the bride-price, so that in effect no exchange of wealth takes place; and there is, finally, the custom among the upper classes to marry their daughters wit h a very low, nominal, bride-price to Mallams—the remission of the bride-price being regarded as a pious act. In these cases the bride-price is kept artificially low in order to ensure that the women of the aristocracy should all be married to men of the same class, or should marry certain men who m they woul d be unable to marry if the bride-price were high : the political and religious gain balances the material sacrifice.
Th e institution of bride-price in Nupe gives us an exceptionally clear picture of the multitudinous structure of institutions. It also illustrates the discriminating action of economic evaluation, which , out of the number of social purposes subserved in the institution, fastens upon certain specific purposes for transformation into concrete economic wants. Yet those that are not thus transformed need not be purposes whic h have lost their significance entirely, but may reveal it indirectly, in the for m of ‘theoretical values’. We recall the disintegration of the institution of bride-price in present-day Bida, in the case of those young labourers who spend their money among other things on ‘women’ in • stead of saving it for bride-price. Le t us realize that in a society like Nupe, in whic h traditional morality forbids pre-marital sexual inter• course, the ‘indemnification’ aspect of bride-price also implies the pur• chase of legitimate sexual satisfaction. Now , these young labourers have discovered that to-day sexual satisfaction can be bought more cheaply, and withou t incurrin g the obligation of marriage. Short-sightedly, they perceive only this one need and disregard the other aspects of marriage and bride-price—the need of founding a family, of establishing new kinship bonds, and of ensuring continuity of the group. The y admit these other needs—in theory; yet these needs are present, concretely, in
A a
354 ECONOMIC BALANCE: MEANS AN D WANTS
the minds of the older generation (though it is unable to enforce its authority), and they will be present, we may assume, also in the minds of the young men themselves when they reach a more mature age and understanding.
Th e economic demands which we must list under the second heading— demands implied in duties towards the community—are comparatively light. In village economy they are embodied in the regular contribu• tions to village rituals in money or kind, and in occasional gifts to the chief (mostly gifts of kola-nuts or produce, worth between 3d. and 6d.). To give an example of the former: the total cost of a village ritual of moderate dimensions comes to IOS., which sum would be divided among, say, twenty households. But rituals may also be celebrated both more lavishly, and on a much smaller scale, with a cost between 35. and
6s.l In Bida we have again certain expenses entailed in religious obliga• tions—alms to Mallams, the traditional killing of a ram on the day of the Sallah or, occasionally, labour and money spent on the building of one of the small mosques in which neighbours and relatives worship.2
Logically, tax should have been listed in this group of expenses.
The last group of ‘cultural requirements’—personal wants—is the most fluid and the most difficult to outline. It follows most closely economic stratification and the varying standards of living. As we shall see later, almost every want in this group is seen against the standards set by the social elite of the country. In certain respects, however, we can attempt to define certain minimum wants even in this field. Thus every family head is expected to possess, in addition to his simple working clothes, at least one ‘good’ dress, a dark-blue or white gown, in which he would appear at meetings in the house of the chief and at various cere• monial occasions. Similarly he should own a white turban in addition to the straw hat or cap which he wears on the farm, ornamented sandals of leather, and a stick or carved star! to carry about; he should wear a heavy glass bangle round the arm, possibly a finger-ring and various charms of the kind that are sold by Hausa traders on the markets of the country.3
A major failure to achieve these standards of personal requirements is regarded as a sign of poverty and disgrace. Often have I listened to old men complaining about the sad state of affairs which compelled them to be content with a shabby cloth or cap when gown and turban were more fitting to their age and status.
But here we must take into account that cultural ‘rift’ between urban and rural standards of which I have spoken earlier in this chapter. Many wants which either do not exist, or do not exist in the same urgency, in the village are considered essential in Bida. Here the closeness of the social elite, both in the literal sense of the word and in the sense of exist• ing social contacts, raises the standards of living. In Bida, where one
1 See my article on the Gunnu, p. 99, n. 2.
2 Th e building itself is done by the people, the roofing and thatching by the professional builders of Bida. Th e costs were, in a concrete case, ios., and were divided among five households.
3 These charms consist of a piece of paper supposed to be inscribed with a
verse of the Koran or a Mohammedan formula of blessing, which is sewn in leather, and worn round the arm or neck.
ECONOMIC BALANCE: MEANS AN D WANTS 355 can change one’s profession and try new means of livelihood, these
‘minimum’ wants can be studied most profitably in the individuals who have abandoned their original occupation in order to better their stan• dard of living. When discussing with labourers the reasons for their change of profession, I was given this as a typical reply: ‘We are Bida people, not peasants; we are not satisfied to live as peasants do. We are not content with one gown and one cap; we want to go to the evening market and spend money; we want to make presents at the feasts in the houses of our friends, and we also want women.’ Over half of these men were unmarried. Although their wages were comparatively high (75. a week), they could not save enough for the bride-price—’our money disappears at once’, they said.
Th e examples of household budgets from Bida which I have given above must then be interpreted with these viewpoints in mind. I will add here two more examples, of typically urban households, i.e. non- farming households, one of a weaver, and the other of a hat-maker, in Bida. Th e scope and money surplus of these budgets does not differ essentially from that of the peasant households. But even where the surplus seems larger, in figures, it is in effect more restricted. Note that the more prosperous of the two men, the weaver, is as yet unmarried. Th e less fortunate of the two, the hat-maker, had moreover given up hat- making this same year, and had become a labourer. It is not a paradox to say that the poor in Bida are poorer than the poor in the rural districts.
Household F
Bida. Profession: weaver, aged 30, unmarried
Output
48 pieces of cloth p.a., sold for 4s. to 4s. 6d.
Income
Profit per cloth
8d. to 1s. 2d.
Additional income
Farmin g on garden-plots close to the house, producin g about a quarter of the annual food consumption.
Tota l earnings . . £2. 85. p.a.
Mojie y expenditure: £ s. d. ow n tax . . . . . • . . . . n o contributio n t o tax o f father (shared wit h brother) . . . 2 6 contributio n t o costs o f food and clothing o f father . . . 4 0 minimu m cost o f livin g for himself . . . . . . 8 0
Total £i~~5 6
Household G
Bida. Profession: hat-maker, aged 35, one wife
Output Income Additional income
Approx . 50 hats p.a., sold for Sd. each
Profit 5d. per hat. Borrowed 2$. from wife
Tota l earnings . . £1
Mone y expenditure: £ s. d. tax . . . . . . 9 6 minimu m cost of livin g . . . 12 o
Total £1 ~~6
356 ECONOMIC BALANCE: MEANS AN D WANTS
It is impossible and unnecessary to reduce to £ s. d. the multitudinous forms of expenditure whic h fall under the heading of ‘cultural require• ments’. We have already seen that most often the surplus whic h can be derived fro m the main occupation is not large enough to cover them all . Th e various attempts to reduce or avoid expenditure, or to discover additional sources of income, are convincing evidence of this. To us, the most essential question is the selection and order of preference of expenses on whic h the hard-pressed household decides to economize. Individua l differences play, of course, a certain part. In the quarter of the Bida beadworkers, for example, I saw among a number of dilapidated and neglected houses one whose excellent condition and nicely orna• mented katamba spoke of the care whic h the inhabitants devoted to its upkeep. Th e financial position of the different families in the quarter was the same, yet one, to use the familiar Nupe expression, had ‘made an effort’ while the others had let things drift . Nevertheless, individua l variations appear to be of littl e account in the general picture. Th e main differences occur, not between individuals, but between groups, that is, between the urban and rural population. I have just spoken of the young men of Bida who wil l forgo marriage (involving the bride-price) rather than the pleasures of town-life ; they wil l spend money on dresses and feasts rather than contribute to the cost of the repairs of the family home. Perhaps the most striking illustration of this attitude was the utterly neglected state of the house of the guild-head of the carpenters and builders. Althoug h the family numbered eight adult men apart from the ol d family head (two of who m had jobs in the workshops), and although house repairs were their profession, the men di d not devote more than the absolutely necessary time and labour to the upkeep of the house, keeping the sleeping-huts in repair but letting compound wal l and entrance hut go to rack and ruin . Wit h a littl e money in hand, most Bida men wil l insist on a sumptuous marriage feast or naming ceremony for chil d or grandchild rather than save it for other expenses. I have twice attended such lavish feasts in neglected houses, in families whose income I knew to be small and precarious. But even the men wit h large and secure incomes wil l live above their means. Certain highly paid officials of the N.A . were heavily in debt to local traders and the Euro • pean stores, having spent lavishly on dresses, luxuries, gifts to friends, and various forms of entertainment.
If in Bida expenditure on ostentatious display and on social obliga• tions conforming to the standards of livin g which spread downward from the top of the social scale takes precedence of the more sober de• mands of bride-price, maintenance of the house, or saving in general, the villagers take the opposite view. In a small pagan village, for ex• ample, an annual ceremony had been postponed for three months for lack of money; in another case the same ceremony had been performed for three years in succession on a very small scale, owing to ‘bad times’. Economies of this kin d occur even in prosperous households: in the family whose budget I have quoted above under E the marriage feast of the eldest son had been postponed for one year so that it could be celebrated together wit h the naming ceremony of the first child , thus
ECONOMIC BALANCE : MEANS AN D WANTS 357 halving the costs of the festivity. In many essentially pagan villages
Mohammedan burials are now becoming the fashion partly because, as
I have reasons to believe, they are cheaper :han the more elaborate pagan
funeral feasts. Ar e then traditional customs such weak incentives, and
regarded as of so littl e importance, that they are easily modified and
even abandoned? Or shall we interpret this lack of appreciation as
being due to an eminently reasonable attitude of the Nupe peasant?
Th e answer is that the Nup e peasant or craftsman has littl e or no alter•
native choice. He is faced wit h two rigid, inescapable, demands which ,
between them, eliminate nearly all freedom of choice: one is mainten•
ance of labour, involving food-supplies, clothing, and, to some extent at
least, payment of bride-price for the labour-group on the co-operation
of whic h the livelihood of the family depends; the otrjer, even more
rigid , demand is tax. Th e surrender of all these cultural wants is a
result, not of ‘ reasonableness’ or free choice, but essentially of external pressure.1 Tha t they stil l represent valid Values’, capable of determin• in g social behaviour, is shown in that radical solution to secure their satisfaction individually if it proved impossible to secure them for the whole group, even at the risk of destroying the labour-unit.
Th e part played by tax in this failure of the Nupe household to adjust itself to the various cultural requirements was shown clearly in our analysis of the household budgets. We have seen that the balance could be maintained withou t difficulty only where a home industry (which, as we have seen, is carried out as individual work) could be relied on as a subsidiary means of the family income. Administrative Officers are wel l aware of this fact. Indeed, some are of the opinion that additional handicrafts or similar occupations, for example, work as temporary labourers, should be encouraged as a means of procuring’money for tax withou t deranging the general scheme of household economy. Tw o Administrative Officers, of long experience, have pointed out to me that among the Hausa it is an established custom for young peasants to travel durin g the dry season and seek work of this kind , whic h woul d later enable them to pay their tax withou t difficulty. I am not qualified to dis• cuss the fiscal implications of such a policy. But I may be allowed to point out its probable sociological and psychological effects, namely, that the encouragement of temporary labour of individuals must further, and indeed favour, that break-up of the labour- and family-unit the beginnings of whic h are already visible.
STANDARDS OF LIVING : STATUS AND PRESTIGE
We tur n to the highest standard of living , that accepted by the upper classes of Nupe. Thei r needs and demands, the ‘things they do’ wit h their wealth, can again be grouped under the three headings whic h
1 It must be said that certain Administrative Officers regard native festivities which fall in the time when tax is being collected as definitely conflicting with the prompt payment of tax. I have already mentioned (see p. 64) that in Bida Emirate all festivities (which, incidentally, happened to include certain harvest and fertility rites) were forbidden for a certain period until tax arrears had been
paid.
358 ECONOMIC BALANCE: MEANS AN D WANTS
characterize the standards of living of the lower social strata—with their meaning significantly altered. For ‘maintenance of labour’ read increase of capital and wealth; for ‘demands of the community’ dictates of status and prestige. Th e latter pervade the whole field of expenditure. Ex• penditure of wealth in any form, whatever its primary and avowed pur• pose, represents expenditure for the sake of status and prestige at the same time.
Investment of capital in human labour—slaves, pawns, or clients— also enhances the prestige of the man who can afford such investment and the life of leisure which it entails. Th e lavish scale of kinship feasts similarly reflects social status. Presents on a generous scale, to the king or influential noblemen, or liberal hospitality as befits the man of sub• stance, are at the same time practical means of winning the favour of social superiors and securing a reliable political following. In the house of a Bida nobleman of moderately high standing, for example, six bara who lived with him, and five bara who lived outside, in their own houses, were daily guests at at least one meal—a hospitality which very nearly doubled household expenses.1
In a society in which marriage is by bride-price, and by high bride- price at that, the marriage arrangements must also become an index of wealth and status. Wealth and status are infallibly reflected in the number of wives a man possesses. We have seen that, while small peasants or small craftsmen and traders are content with one or two wives, no Bida man of standing would have less than four, and many even considerably more (whether legitimate wives or concubines). We will not exaggerate the share of prestige and status in this ‘investment’: the practice in the houses of the nobility to ‘pension off’ ageing wives and replace them by young attractive girls shows the contribution of more concrete motives also. On the other hand, wealthy Bida families will pride themselves in marrying off their sons at an early age, four to five years before they reach what is normally regarded as marriageable age. This is clearly done for the sake of prestige of the parents rather than for the benefit of the boy-husband, who, for example, may still be a pupil in the Government school (I have known of two such).
This twofold significance of economic incentives among the elite of Nupe society is no less clearly pronounced in expenditure of a more personal nature. ‘Clothes make the man’ in Nupe as elsewhere; in the possession of arms one sees not only fighting equipment, and in the possession of a horse not only a means of transport, but symbols of rank and status. I will conclude this list with the description of one cultural incentive which carries the concept of status and prestige into a rather remote domain of social life—travelling. Travelling forms indeed part of the life of the Nupe gentleman, and the young men of the higher nobility, especially the princes of the royal house, are sent, like their brothers in Victorian England, on a traditional Grand Tour. It takes them through the northern Emirates, to the cities and courts of famous
1 Th e total consumption at the midday meal was approximately 5 lb. of porridge, made with fish and palm-oil sauce. Th e share of the family, including the family head, his wives, and his children, was 3 lb., that of the bara 2 lb.
ECONOMIC BALANCE: MEANS AN D WANTS 359 rulers, with whom they stay as honoured guests, and whom they will
repay, one day, by similar hospitality extended to their young relations.
(I have mentioned, in a different context, how I met two princes from
Sokoto whose Grand Tour had taken them to the court of the Etsu Nupe
in Bida.) Th e noble tourists enjoy their travels and enjoy—as they put
it—’seeing the world’. But this incentive of travelling for travelling’s
sake with which their education and cultural background inculcate them
also serves, indirectly, the political purpose of fostering personal con•
tacts and ties of friendship between the ruling families of Nigeria and,
more specially, the widespread Fulani empire.
We must now ask ourselves to what extent these wants and incentives apply to the society at large and determine (more or less) the economic decisions of the people. We must ask, more particularly, whether there do not exist ‘upper limits’ by which the economic ambitions of the various classes are bounded—whether by choice or of necessity. Such limits exist in the form of objects and values which have come to be regarded as fixed symbols of political status and which, like status itself, are excluded from free economic choice, being attainable only by the ladder of social promotion. As fixed symbols of status, much as a uni• form or decoration in our society, the glazed turban of the nobility, or the right to carry a silver sword, cannot be coveted as such. This limita• tion has also been accepted with respect to the recent innovation, the clock in the glass case carried by the body-servant of a noble master: the people outside the ranks of the nobility have not attempted—as well they might—to possess themselves of this new emblem of rank.
But though promotion to political status and acquisition of its para• phernalia represent things that ‘cannot be bought’ (we shall discuss certain exceptions later), most of the features characteristic of the social life on these more exalted planes are, in theory, of the nature of universal economic wants. Beautiful clothes, costly turbans, and the possession of a horse are widely desired and can be acquired by anyone who can afford the expense. Even wants which are so far beyond the means or labour which commoners and peasants can muster that for this reason alone they seem unlikely to become concrete ambitions outside the ilite of the country, represent yet common wants and ideals. Thus we find the ornate high-gabled square or round entrance hut which is typical of the large houses of the Bida nobility also in certain villages, put up by chiefs or prosperous family heads. Beautifully ornamented houses of a more modest type, moreover, are found everywhere, in every section or stratum of the population. Now, one of the most convincing proofs of the universality of these economic wants is the absence of all unfavour• able comments, especially in the form of ridicule, with respect to indi• viduals reaching out for goals not normally attained in their station of life. This is a significant illustration: in the small village of Doko I observed one house built in an unusually sumptuous style—with coloured earthenware plates inlaid in the walls of the katamba. It be• longed to a,musician, an ordinary ‘commoner’, who had been very successful in his profession. The people who told me about the house and its owner admitted that he was neither a titled elder nor in any way
36 0 ECONOMIC BALANCE : MEANS AN D WANTS
a person of consequence, bu t di d not find it at all out of place that he should have buil t himself such a pretentious house. We shall also hear later that in the Bida age- grade associations young men of low birt h can even taste the thril l of having special drum-signals and songs composed for them like the members of the nobility .
Polygamy is another characteristic instance. Th e large-scale poly• gamy of the Nupe aristocracy remains beyond the reach of all other classes. Polygamy as it exists in the houses of the Nupe kings must even be regarded as a royal prerogative and thus as a fixed symbol of status. Yet polygamy on a more modest scale represents another universal want, the effectiveness of whic h can be studied in every stratum of society, and when peasants tel l you of the ‘good ol d times’ when their chiefs and elders were able to marry large numbers of wives, they ex• press the effectiveness of this incentive.1
We may also mention the demands of hospitality, whic h are recog• nized among the poor as well as the rich . Inabilit y to fulfil these demands is regarded as a definite deprivation. Even in small peasant households wil l you find ‘permanent’ guests—persons who have been adopted into the household and may even be supported by their hosts when they are too old , or too ill , to gain their ow n living . Significantly, they were described to me as ke bar a na, ‘a kin d of bar a’.
One of the common ideals of the people of Nupe demands closer in • vestigation. Th e reliance on employed labour by the privileged sections of the society establishes a separate leisured class, whic h lives by the labours of others and gains its livelihood to a large extent from the ‘interest’ borne by their human capital. Evidently, ‘leisure’ is a relative concept. In a society where the large majority works (say) ten hours a day, a group whic h can gain its livelihood by workin g only half the time repre• sents already, comparatively speaking, a ‘leisured class’. Thu s for Nupe , as for every society, the definition of the leisured class must remain fluid. Th e well-to-do traders, the ‘intelligentsia’ of the Mallams, the ric h landowners, and the well-paid N. A. officials, all are, in different measure, members of the leisured class compared wit h the hard-working farmers, craftsmen, or manual labourers. Le t us understand the ful l impor t of this fact. Man y forms of enjoyment depend upon leisure: recreation and aesthetic pleasures; among women, the possibility of keeping young and attractive. The y all for m part of the standard of livin g whic h is typical of the highest social stratum of Nupe. Again, they become, as such, also incentives for Nupe society at large, and affect the evaluation of specific occupations. We have studied their effectiveness in such reactions as the break-up of the family labour-unit under the influence of what we have called the ‘new values and temptations’ spreading from the town ; we shall discover an even more conspicuous instance in the spreading of the Malla m profession.
1 It is impossible to test the correctness of their sometimes rather extravagant claims. Yet the psychological reality of the incentive is unaffected by such questions of historical truth.
ECONOMIC BALANCE : MEANS AN D WANT S 361
WANTS AND CULTURAL ASSIMILATION
Le t me repeat that all these wants and incentives whic h are realized in the social elite also represent common incentives of the society at large, and of every social stratum, unattainable though they may appear in actuality. The y remain ideals’, and the inability to realize them is regarded as a failure, a shortcoming in the socio-economic system. There exist, however, certain exceptions: instances of that divergence, or rift , in the appreciation of certain wants and incentives whic h divides tow n and country. Thoug h few in number, they occur in widely different domains of social life—nutrition , recreation, sex morality.
Certain instances concern the for m of a want, the conventionalized taste in whic h it appears, rather than its essence. I have spoken in an earlier context of the different taste in intoxicants typical of tow n and country. Th e Bida people prefer palm-wine to the sorghum-beer whic h is drunk commonly in the village. Palm-wine is the native drink in the oil-palm countries of Southern Nigeria ; we also find it , side by side wit h sorghum-beer, in certain riverain districts of Nupe where the oil-pal m is cultivated. Bida itself imports palm-wine from these dis• tricts and the south. Th e adoption of this, as it were, alien drin k by the people of Bida was explained to me as being due to the demands of Islam, whic h forbids fermented drinks like sorghum-beer, but permits palm-wine—at least in the eyes of the Nupe followers of the Prophet. I do not know how far this vindication of palm-wine wil l pass muster wit h Mohammedan scholars. But I thin k we may accept the negative feature of this interpretation, namely, the condemnation on religious grounds of sorghum-beer, which , as we shall see presently, represents a drink closely associated wit h pagan religious rites. Or I migh t quote the instance of the different styles of dance and music appreciated in Bida and in the peasant districts: in the tow n we find the dance of pro• fessional women-dancers who are watched by an appreciative audience; in the village the dance of peasant men and women who are performers and public in one.
In other tastes and wants tow n and country differ fundamentally. Le t me quote tw o striking examples: the attitude towards sexual licence (about whic h I need say no more here),1 and the attitude towards travel• ling . What I have said above about the appreciation of travelling in the social elite of the country could be extended to apply to the whole of Bida. To have travelled widely, to have ‘seen the world’ , helps to establish the social standing of the townsman. Th e interest in travels even outweighs other, similar, interests. Whe n I was collecting Nup e folk-lore I met wit h almost complete failure in Bida, although the villages had proved a most fruitfu l field. An d this was the explanation whic h my Bida informants gave me : ‘Yo u must go to the villages to hear folk-tales. We Bida people have forgotten them all. No one wants them now. Why ? Because when we tell stories, in the evenings, we like to hear about tfavels, and about other towns; we listen to the people who
1 See Chaps. IX , XIX .
36a ECONOMIC BALANCE: MEANS AN D WANTS
can tell us about Kano, Zaria, or Lagos.’ Th e widespread familiarity with travelling, indeed the love of travelling, which characterizes the Bida people, has no counterpart among the peasants.1 We have heard that even journeys to another district are rare and of little appeal to the villagers.
How are we to explain this negative form of culture contact ? Wh y did cultural assimilation stop short of these few traits while so many other incentives and wants are spreading from the capital to the rural districts ? When we examined, in a previous chapter, the different sex morality in town and country, we discovered its close functional interrelation with other social facts. Behind the divergent attitude we perceived the effectiveness of political and economic factors. This also points out the answer of the present problem. Sex morality can change only together with those other social determinants. Let me recall the main facts. Negative facts, such as the absence in the village of the ‘ double morality’ of the capital, and also the absence of a large floating population which proved one of the incentives for the creation of semi-professional prosti• tution in the town. Positive facts, such as the more closely knit village life, which forbids the open and condoned matrimonial laxity character• istic of Bida; the efforts of peasant family heads to provide bride-price for an early marriage of the men, and thus to maintain the labour-unit; the conditions of labour in peasant households which tend to preclude sexual promiscuity among the women. The ‘negative’ culture contact which we are observing does not, then, represent a failure of assimilation in certain specific domains of culture, but the inertia of a cultural whole which cannot suffer changes in these domains without undergoing far- reaching changes itself.
Th e same is true of the incentive of travelling. In Bida, travelling serves the political interests of the ruling group; it is helpful to the pro• ductive enterprises of the town population; and it follows the political interests converging in the capital of the kingdom. None of these factors exists in the village. On the contrary, productive organization, political and religious life, move in a narrow compass beyond which the com• munity affords no protection or promise of attachment.
In the case of the other, minor, instances of cultural divergence, which refer only to the form which a particular satisfaction takes, one might argue that the essential similarity of the wants themselves must tend to retard a more complete assimilation. But here again we must view the situation under the aspect of the cultural whble and the close integration within the whole even of apparently ephemeral traits. Th e appreciation of a certain kind of drink or a certain style of dance is indeed built firmly into the framework of village culture. Thus all religious rituals involve a libation of beer. The temptation to replace it with palm-wine would have to break down, not only an existing nutritive habit, but also, partly, the spiritual background against which the people (consciously or un• consciously) must regard their traditional beverage. Th e alternative solution of admitting palm-wine, and having different drinks for ritual and profane purposes, would involve an equally radical change in other
1 See p. 39.
ECONOMIC BALANCE: MEANS AN D WANTS 363 respects: not only in the productive arrangements which, in these circum•
stances, would no longer supply the ritual beverage as a matter of course,
but also in the psychological attitude, which would have to admit a ritual
beverage largely emptied of what RadclifTe-Brown calls ‘social value’, i.e.
which would be nothing but a ritual beverage, completely removed from
the habits of ordinary life. Th e village dance, again, is an expression of
group co-operation and coactivity in the recreational sphere. Like the
traditional occasions themselves on which it is performed—kinship
feasts, egbe work, religious ceremonies, and so forth—the dance mobi•
lizes and intensifies group sentiments. We might even go so far as to say
that the unitary nature of the group which is thus mobilized forbids a
division into neutral performer and aloof lookers-on. Th e correlation
between professionalized performance and differentiation within the
group is indeed clearly visible in Bida. An aloof aesthetic appreciation
of professional dancers and musicians appears to represent the only form
of entertainment acceptable to a society in which class prejudices and
motives of status and prestige dominate all gatherings and festivities.
Class prejudice has fostered a conception of dignity set against public
exhibitions of this kind. Th e explanation is, I believe, that it would
obliterate, even though only temporarily, the demarcation of rank and
status. It is, in fact, quite inconceivable for a man of status to join in a
dance, even in his own house, for example, at a family feast—to dance,
that is, together with, and in front of, people who must invariably in•
clude persons of lesser station. Yet in the villages old men, even titled
elders, commonly join in the dances of the younger folk, without danger
to their dignity. To assimilate, then, the village dance to the perform•
ances which appeal to the ‘sophisticated’ people of Bida would mean
assimilating the complete social organization.
For the sake of the theoretical argument I have stated the case of this negative form of culture contact a little more strongly than is warranted by actual conditions. Throughout this book we have found evidence of the gradual closing of the rift, and of the advancing assimilation even in these domains of culture. But indeed our argument is supported by this development. For the assimilation does not advance by the narrow channels of confined, specific, cultural changes. Rather these changes are carried by the tide of a far-reaching cultural adjustment which is transforming the whole productive system and the organization of labour as well as the social life at large in the Nupe village. The breaking up of the family labour-unit, and the appearance, in the country, of professional Bida musicians as spokesmen of what I have called the
‘ propaganda for the kingdom’, are two convincing instances.
WEALTH AND STATUS
Hitherto we have been using the terms ‘status’ and ‘prestige’ as mean• ing one and the same thing. It becomes important now to distinguish between them. By ‘status’ we understand a more or less fixed place on the social ladder; ‘prestige’ represents a more fluid form of social recog• nition, gained by certain services or deeds in accordance with the
364 ECONOMIC BALANCE : MEANS AN D WANT S
accepted code of society. Status in Nupe is determined by descent or formal promotion to rank and office. One of the accepted and most im • portant means by whic h prestige can be gained is wealth. Thi s raises the question of the closeness of the relation between status, and prestige gained through wealth. In other words, we must ask to what extent prestige gained through wealth tends to facilitate the attainment of status, and may even come to be regarded as its equivalent, or, con• versely, to what extent hig h social status can be reached and maintained withou t adequate financial support.
Under traditional conditions social status and wealth could hardly become separated as political rank implie d far-reaching economic privi • leges—the acquisition of fiefs and land, the chances of booty, profits fro m slave-trade, and so forth . Even assuming that considerable wealth could be accumulated also by groups outside the recognized social Slite, such as traders and craftsmen, it is clear that their highest income could not rival the resources vested in a privileged political position. In present- day Nup e traditional political ranks and wealth are no longer inseparably linked. Hig h social status does not invariably guarantee the demands of prestige also. Th e members of the elite, no longer in a position to com• bine the satisfaction of the more concrete wants wit h that of the de• mands of prestige, wil l have to choose between the different forms o f expenditure. Yet here we find, rather surprisingly, that the expenses for the sake of prestige can be very considerably reduced withou t affecting the conception of status. Th e fixed symbols of status, the glazed turban and the silver sword, are either comparatively cheap or are heirlooms. Th e house, also inherited, may have to be left in disrepair; bu t its lay• out woul d stil l betray the status of its owner. Polygamy can be main • tained to a certain extent thanks to the practice of marriage in the family, wit h low bride-price, whic h is customary among the elite. Horses, new dresses, ornaments, lavish hospitality—they all have to go. Bu t travel• lin g and leisure need not disappear. Th e former is satisfied by claiming hospitality from relatives or friends in distant parts, or even from social inferiors, who woul d not thin k of refusing such flattering requests. Th e latter manifests itself in the choice of profession of impecunious noble• men : very few have taken to farming or trie d to learn a craft; theirs are mostly the unskilled ‘free’ professions of the leisured class. Thu s even on the decline, the social ilite secures the satisfaction of certain wants whic h to the lower orders of the society are permanently unattainable. Nup e society buil t up a strong defence of the prerogatives of status in the for m of class-obligations and the utilization of inheritance. Wit h its help the social elite of the country can hope to weather at least temporary diffi • culties and set-backs.
Tha t this defence of class prerogatives has not yet lost its strength is evident in the hesitant attitude of the people towards accepting the new evaluation of wealth as independent of traditional status. Th e issue is by no means clear; it is confused by various concomitant factors, specially by the introduction of the new official ranks and positions under the N.A . some of whic h are vested in the old rulin g class, while others have been bestowed on ‘new men’ from outside the traditional ilite. But
ECONOMIC BALANCE: MEANS AN D WANTS 365 there is convincing evidence to show that high social status still attaches
to traditional ranks, though they may no longer command their former
financial reward, rather than to wealth alone. Evidently this is only a
transitional solution. In a society in which high status is so closely
identified with a definite standard of living the separation of the two can•
not be maintained indefinitely, at least not without some radical change in
the whole concept of the elite. Theoretically, had their ranks not been
divided by the appointment of some of their members to modern official
posts, the Nupe aristocracy could have chosen the diehard solution of
reinterpreting the whole concept of status and making it independent of
economic advantages. As it is, the old elite must either win back its
economic supremacy (which is impossible, though the attempts have
never been given up), or else admit the new nobility of wealth into its
ranks. We can observe this new conception taking shape. Th e respect
for the traditional values is only strong enough to maintain an external
continuity—the continuity of the symbols of status. And the ‘new men’
are the first to submit to this rule which allows them to legitimatize their
position: they buy the titles which are the only passports to acceptance
among the social ilite. Here then, we have a final, and most modern, use
of capital investment, for the sake of social advancement in its purely
abstract and symbolic form.
CHAPTER XXI
ECONOMICS AND SOCIAL STRUCTURE
ECONOMI C MOBILIT Y
H E lack o f tangible data makes i t impossible t o formulat e any theor y about the extent t o whic h economic mobility , i.e. movemen t be • twee n the strata, was a reality in pre-Europea n days. We can onl y mak e a few statements based on social an d historical facts the economic sig• nificance of whic h we can, wit h some certainty, assess. Economi c better • men t in every society depends on th e possibility of increasing one’s hol d over resources an d means o f production . Th e Fulan i conquest, an d the exploitation o f lan d whic h followed it , restricted, for one par t o f Nup e a t least, mobilit y wit h regard t o th e mai n resource, land . Th e restriction could, however, be circumvente d by means of political allegiance, i.e. by becomin g the client of an influential landowner. Ther e is also evidence tha t certain peasants an d craftsmen have owne d slaves an d take n pawns, thus acquirin g interest-bearing capital, an d so rising above th e average level of a peasant’s or artisan’s livelihood. In the rulin g classes an d amon g thei r dependants careers involvin g increased economic opportu • nities wer e th e norma l thing , whethe r the y wer e expressed in a share in booty an d slave-trade or in rewards to warriors an d political followers ‘for services rendered’ . Th e reverse process, loss of economic opportunities , an d in th e extrem e case th e loss of all possessions, was no less marked . I n the life o f the peasant, such changes occurred i n the course o f wars an d raids involvin g the destruction of homesteads, farms, an d stores. Th e peasants i n Nup e kingdo m proper, thoug h safe fro m slave-raids, wer e still exposed to similar attacks: punitiv e expeditions which , directed against a rebellious feudal lord , wer e fought on thei r land , or th e visitation o f armies whic h marche d throug h an d foraged fro m thei r farms. I n th e rulin g group economic downfal l could mea n onl y one thing :
political disgrace, often accompanied by confiscation of thei r property .
To-da y the risks o f wa r an d raids have disappeared, together wit h the exploitation an d one-sided source of income whic h the y represented. Thu s the ne w rul e has to some extent levelled dow n the old economic inequalities. Clearl y this levelling-dow n cannot have been equally wel • come t o every section o f th e population . Le t m e quote here th e ver y fran k and , i n fact, typica l statement o f one o f m y most honest informants , a n influential village elder o f Mokwa . H e admitte d the great benefits o f th e present rule , thanks to whic h the tdlaka, th e poor man , enjoys safety fro m wars, raids, an d political abuses. Bu t the n he added (rathe r heatedly i n fact) : ‘Bu t yo u have turne d u s all int o tdlaka.’ Thi s view , comin g as it does no t fro m a membe r of the rulin g class bu t fro m a typica l Nup e peasant, makes it necessary to review mor e closely th e economic effects o f the ne w order.1 Al l ou r data ten d t o show that , although trad e
1 It incidentally supplies further evidence of the fact that a rise to superior economic position also occurred in the peasant classes.
ECONOMICS AN D SOCIAL STRUCTURE 367 and commerce have greatly benefited from the new development, the
position of the peasants and craftsmen has not improved to the extent of
enabling them to accumulate reserves and increase their resources to any
considerable extent. Th e introduction of commercial crops, the cultiva•
tio n of whic h is officially encouraged, represents a policy designed to
open to the native cultivator the way to economic betterment. Yet the
insecurity of these new resources may defeat their own ends. A more
recent move to improve the economic position of the farmers was the
launching of a policy to introduce mixed farming into Bida Emirate.
Mixe d farming, whic h had been adopted some years previously in
certain northern Emirates, was started in Nupe towards the end of my
second expedition. At that time only the demonstration farms of the
N.A . and certain stretches of land belonging to the Emir , all worked by
wage-labour, had been put under the new cultivation. Whil e it is thus
too early to describe actual results, it is possible to outline, from the
study of the underlying principles, certain probable effects of this agri•
cultural policy on the economic and social system of the country. The y
are sufficiently far-reaching to warrant here a more extensive examina-‘ tion. 1
Thi s policy envisages a gradual transformation of the whole of Norther n Nigeria into arable country cultivated on Western lines, on the basis of individua l landholdings. Th e workin g team should consist of one man and a boy. Th e 2 acres per head whic h have hitherto corre• sponded to an individua l worker woul d be extended to 12, and eventually to 20 acres. Th e manure gained from the draught cattle woul d do away wit h the necessity of shifting cultivation and woul d allow practi• cally unlimite d cultivation of the same stretch of land. A four-field scheme of rotation is suggested, wit h one field manured, and the follow• in g crops grown on the different fields: sorghum and millet (on the manured field); cotton ; ground-nuts; cassava and small crops. Th e most prosperous farmers in a district are to be persuaded to take up mixe d farming first, thus giving an example which , it is hoped, many others wil l soon follow. Th e total cost of the draught cattle (two bul • locks and one cow) and plough is £7 for an iron , and £5 for a wooden, plough. Farmers who are unable to buy the equipment outright wil l be given an advance on future profits of £1 to £1. 10s., or can pay an
‘entrance fee’ of approximately that amount, being expected to pay the rest, viz. repay the advance, in four yearly instalments which include a hirin g charge of 5 per cent. Th e greatly increased yield should make these payments easy: i t has been calculated that the farmer wil l be able to increase his production eightfold by the new method.
Th e scheme, admirable though its intentions are, may easily be wrecked if certain inherent dangers and difficulties are not carefully con• sidered. Le t us take the example of the most prosperous and progressive farmers of Lem u whose landholdings we have examined in an earlier chapter (see p. 212). We found that the total land under cultivation in the case of a workin g group of five and three men was 9½ and 6 acres,
1 I am greatly indebted to the Superintendent of Agriculture in Bida for having allowed me access to the departmental files on mixed farming.
368 ECONOMICS AN D SOCIAL STRUCTURE
respectively. Th e larger group possessed one, the smaller two, fallow fields of the same size as the field under cultivation. Their total land- holdings, including both cultivated and fallow fields, thus amounted to
18 and 19 acres. Now, under the new scheme one individual would take under cultivation very nearly the whole landholdings of the family group. The other adult members of the family will have to obtain new land or (only too probably) they will give up farming. In Trans-Kaduna, with its large land supplies, new land might be found with no great difficulty from vacant or virgin village land. But even under these favourable conditions the fourfold increase of landholdings entailed in this development must lead, if not to actual land shortage, to serious competition for land. For one thing, farm-land for ploughing should lie closer together than it does now. The Nupe farmers will soon find it profitable to exchange their scattered ½ acre or 1 acre fields for larger and compact areas, an exchange which, involving a discriminative apprecia• tion of land, must invariably lead to land purchase and land sale. In Cis-Kaduna, with its limited land resources, land sales are, as we have seen, already practised on a small scale. Here, in the country of the big landowners, land purchase and land sale will increase rapidly under the new scheme.
Th e redistribution of land resources will go hand in hand with far- reaching changes in the economic stratification of the country. Th e prosperous peasant who was able to take advantage of the new scheme will in a few years’ time command considerable resources. No doubt, the obvious way to invest the newly won capital will be in more ploughs and more land.1 Th e more unfortunate farmers who, for reasons of poverty or lack of enterprise, are not able to avail themselves early of the new scheme will be hopelessly outdistanced by the rapid development. Th e increased agricultural production, of which a half will be for in • ternal consumption, must lower the price of food crops in the country. Th e small farmer, who relies on food crops more than on commercial crops, will be compelled to sell his produce at a price so low that he may be unable to pay his tax or to maintain his labour costs. Sooner or later he will succumb to the temptation to sell or let his land to those who, having benefited from the new scheme, will be anxious to acquire new land, and he will be forced eventually into the position of a landless labourer on the farms of the new landowning class. Th e Emir, on whose estate mixed farming has been established first, will be foremost among the new landlords of the country.
Th e levelling-down of economic inequality which accompanied the establishment of British rule in Nigeria will thus be replaced by a new, much more defined stratification. Poverty in the present productive system of the country is due, as we have seen, mainly to inadequate labour (disregarding here the factors of land shortage and tax), that is, to such factors as age, illness, misfortune, or ill-planned labour-
1 This is not a mere hypothesis. Th e tendency to accumulate land is already noticeable in the northern Emirates where mixed farming was introduced earlier, and has led more than once to official attempts to keep landed property within certain limits.
ECONOMICS AN D SOCIAL STRUCTURE 369 organization, all of which—i n theory at least—can easily be remedied.
Poverty i n the new scheme wil l be due to a fundamentally different in •
adequacy, the inabilit y to secure the necessary means of production. It
is a permanent, and even progressive, inadequacy, whic h must get worse
wit h the continued process of production. For the same process that
leads to accumulation of capital (i n the modern sense of the word) in the
hands o f the owner o f the adequate means o f production wil l render
more and more remote the chances of their acquisition by the people
who are as yet withou t them . Economic inequality in its most extreme
form , latifundia existing side by side wit h a landless proletariat, wil l be
the final result.
These grave consequences can be avoided, and the valuable elements
in the new policy saved, I think , by slowing-up the transformation of the
agricultural system of the country unti l a corresponding transformation
of the labour organization can be achieved. Th e increase in the cost of
the means of production must be met by widening the basis of the co•
operative uni t whic h makes itself responsible for the maintenance of
labour in this sense into co-operative societies. Co-operative societies,
whic h could buy the new means of production from common funds, and
woul d jointl y own ploughs and cattle, woul d enable also the less pros•
perous farmer to profit from the technical improvements. A co-operative
society whic h Would include, for example, all the owners of adjacent
fields woul d also be able to utilize ploughing to the fullest advan•
tage withou t incurrin g the danger of large-scale land-exchange. Th e
increase in the extent of cultivation woul d be gradual and well controlled,
and the traditional labour-unit not rendered useless overnight; it woul d
stil l be useful for supplementary farm-work (i n the traditional fashion) at
times when plough and cattle are employed on the field of a fellow member of the farm co-operative.1
RISK AND INSURANCE
Every productive organization must include in its calculations the factor of risk and insurance against risk. Certain risks are over and above the normal in that they are due to accidents outside the produc• tive system itself, and beyond the range of foresight whic h the producers
1 When I discussed this question in a lecture given at the Summer School for Colonial Studies in Oxford, in 1937 (see the Summary of Lectures), I was attacked by a member of the audience for attempting to ‘boost poverty’. Now, this is a serious charge. However, I hope to have made it clear that my attempt was not towards fostering poverty or pampering backwardness, but towards avoiding a development which would create, and make permanent, the most dangerous form of poverty and economic dependence. To organize the less prosperous so that they should not be left behind in economic progress does in no way imply putting obstacles in the way of the more enterprising. ‘Th e rich’, as the chairman, who came to my assistance in the discussion, put it, ‘will look after themselves.’ It is interesting to note in this connexion that when mixed farming was discussed at the Residents’ Conference in 1936 one Resident suggested that it might be feasible to utilize the Hausa gaya (collective farm- work) for the new agricultural organization. I presume that he, too, had in mind a co-operative organization of the kind suggested above.
Bb
372 ECONOMICS AN D SOCIAL STRUCTURE
request to be considered out of his turn . He can thus put his hands at once on a sum of money for which, normally, he woul d have had to save for weeks or months. In this, then, lies the benefit of the dashi. Th e administration of the dashi association is entrusted to a man of good reputation, for example, a well-known trader or the head of a craft-guild. People who decide to form a dashi request such a man to undertake its organization. He wil l summon the first, and only, general meeting of the group, at whic h all the important points are settled, and after• wards arrange all matters on his own. People who learn from neigh• bours or friends that a certain dashi is being formed are, up to the date of the first subscription, free to joi n it ; after that, while the dashi is running, no new members are accepted. Th e ‘chairman’ of the associa• tion , the Etsu Dashi (Kin g of the dashi) as the Nupe call him , is respon• sible for the collection of the subscriptions and the regular paying out of the fund. He does not himself pay into it , nor does he receive any money payment for his trouble. He is recompensed by being allowed to borrow from the fund occasionally for himself and his own business provided that the general arrangement on which the group had agreed, and the duration of the dashi, are not altered. Sometimes he also receives gifts from members who desire an early share without being able to show serious reasons for this request. When money is needed, for example, for a funeral, in the case of illness, or for the payment of bride-price, no gifts or bribes are necessary to secure immediate consideration. Dashi chairmen who have the reputation of being good organizers are in great demand, and maybe continuously engaged in organizing and supervising dashi associations, often several at the same time. Women are regarded as specially able organizers: one of the most successful dashi ‘chairmen’
I met in Bida was a woman ; she was actually livin g in Doko, where she had lived since her marriage, but organized a dashi in Bida, sending her messenger, a young girl , every Friday into Bida to collect the sub• scriptions.
Although the benefits of the dashi embrace every variety of risk and expense, its practice is limited to the professions whic h depend on regular money earnings and money investments. It is practised by traders and brokers, both men and women, by certain groups of crafts• men, the wives of farmers, canoe-men (who are enabled to buy their canoes wit h the help of the dashi), labourers, and officials. Th e school• boys in Bida Middl e School have a little ‘pocket-money’ dashi of their own, the subscription being idf., paid every Friday. Th e farmers, although familiar wit h the dashi, do not practise it , nor have they apparently ever done so. Clearly their stores of produce are regarded as sufficient reserve, and their labour organization as sufficient ‘insurance’.
Th e number of members and the amount of the subscription vary widely. Th e sharp distinctions of social and economic stratification are visible also in the organization of the dashi. A well-to-do trader or nobleman woul d never joi n one of the petty dashi associations of small craftsmen or tradesmen in whic h the contribution is a few pennies. The y woul d form a small dashi of their own, wit h a high subscription, even if they have to be content wit h five members only. Durin g my stay
ECONOMICS AN D SOCIAL STRUCTURE 373 in Nupe I found the following dashi associations in practice: dashi
associations formed by women, the wives of farmers in Doko, Kacha,
and one or two other villages, in whic h the weekly subscription was 3d.:
they numbered between twelve and thirt y members; a dashi among the
Bida barbers, wit h the subscription of \d. weekly, numbering ten
members (I was told , incidentally, that formerly their dashi was organ•
ized on a much larger scale). Another dashi, numbering twenty-four
members, was formed by N.A . scribes; here the subscription was high,
in accordance wit h their salary, 10s. per month. Th e teachers of Bida
Middl e School also formed a dashi, numbering twelve members, wit h a
subscription of 7s. 6d. per month. Eight members of this group had
formed a dashi wit h a subscription of £1 per month the previous year.
Several dashi associations have disappeared owing to the economic stress.
Th e labourers at the Bida workshops had practised the dashi til l five
years ago, but had given it up when their wages were lowered. Th e
beadworkers, who had also practised the dashi formerly, had abandoned
it owing to the decline of their business. I received the same informa•
tio n among carpenters, blacksmiths, and butchers. Th e canoe-men in
Jebba, finally, who used to organize their dashi at irregular times, had
not had a dashi for several years.
Returning to the general discussion of ‘insurance’: as the large co• operative unit in agriculture and home industry is disappearing, and, wit h it , the ‘insurance’ which it provided, one may wonder whether this must not bring the other, individual, insurance arrangements more to the foreground. It is thus rather surprising that the practice of the dashi, far from increasing, is disappearing. Th e people, as we have heard, explain this as being due to the general economic decline. Th e bead- workers said that they had no money to spare, and the barber^ and labourers di d not think it wort h their while to organize a dashi wit h a small subscription. Borrowing, on the other hand, especially the borrow• ing by husbands from their wives, is markedly on the increase. Disputes over debts of this kin d are frequent in divorce proceedings in Bida. Th e examination of household budgets also reveals a considerable number of instances (five out of fifty) in which husbands borrowed money, chiefly for tax, from their more prosperous wives; and even in remote places, such as fishing villages on islands in the Niger, where the women are by no means the enterprising traders they are in Bida or larger villages, husbands borrowed up to 8s. and 9s. from their wives. Tha t under a con• tinued economic stress borrowing may spread and increase further, and eventually lead to the introduction of professional money-lending, is not improbable. Borrowing in Nupe, which has once already reached a peak (in pawning), might well , in a different form, grow again to such dimensions.
It is clearly a sign of the growing instability of the native economic system that the solid and comprehensive insurance against risks whic h is vested in the co-operative labour-unit is disappearing, and that its place is taken by the most restricted, most temporary, form of insurance
—borrowing., Once again, we encounter social and economic repercus• sions of the dissolution of the large labour-unit. If previously we have seen that this development has adversely affected certain aspects of
374 ECONOMICS AN D SOCIAL STRUCTURE
production, the present case demonstrates forcibly that it affects the very stability of the economic system of the country. We cannot claim to be able to describe exactly how the labour-unit was working under traditional conditions, and to state categorically that the scheme of in • surance entailed in it operated without difficulty or friction. But on the evidence which we possess it is reasonable to assume that under certain conditions, with a smaller taxation, and protected from severe, incalcul• able, market fluctuations, the large co-operative group represented—or would represent—an eminently stable and successful labour-unit. Th e process of transformation cannot be halted. Where the large labour- unit still exists, it may, given the conditions just mentioned, be pre• served ; where it has disappeared, it cannot be revived. Our conclusions point to one solution only: to the creation of new institutions and labour organizations which can be made to fulfil, under existing conditions, those tasks which fell formerly to the labour organization that has dis• appeared—above all, the task of insurance. Such institutions can be found in co-operative organization, saving societies, co-operative market• ing, and suchlike—organizations which in other parts of Africa have amply proved their efficiency.1
ECONOMIC RELATIONS AND THE GROUP
• Throughout this discussion we have dealt with family and kinship group purely in their economic aspect. We have interpreted the mechanisms which maintain the working-unit and thus the co-operation in its human substratum, the family group, in terms of economic rela• tions only. Against this treatment of the problem one might argue that it sees only one side of it, and that the kinship and family structure must be taken into account also as an entity sui generis, which, within the framework of economic relations, exercises its own specific influences. One might argue, in particular, that as the Nupe efako-unit is based on the family, kinship ties and kinship sentiments as such supply the bind• ing forces of the labour group—’ maintain labour’, in our terminology— and that the binding forces which we have recognized, namely, com• monness of material interests and benefits accruing from co-operation, are in reality mutual kinship obligations.
Th e double nature of economic obligations in a productive unit which happens to be a kinship-unit at the same time cannot be denied. Th e influence of kinship factors can, in fact, be made evident by com• paring the labour-unit based on kinship, with similar labour-units in which kinship plays no part. While, in the former, co-operation is achieved by mutual obligations such as payment of bride-price, and supply of clothing, food, and shelter, labour of the latter order must be recompensed in the form of money wages. Th e effectiveness of kinship
1 It has been said that our intention should always be to graft these new co• operative organizations on to existing, indigenous, organizations. Th e scope and the possibilities of such a scheme must not be exaggerated. Although the tasks with which the new organization is to be charged may have formed part of the traditional system, the organization itself may have to be newly created, and may, in fact (as in our case), have to replace a failing indigenous organization, rendered inadequate by modern conditions.
ECONOMICS AN D SOCIAL STRUCTURE 375 sentiments in co-operation was also implie d in various statements made
earlier in the discussion of economics. We have seen that the craft-
guilds plan their labour organization on kinship lines, and reinforce the
labour group by what we have called ‘professional endogamy’. But let
us realize that in doing this they do not appeal to abstract kinship ties or
‘ bonds of blood’, but merely duplicate the common labour interests by
corresponding family interests, namely, the concrete interests and forms
of social interdependence entailed in intermarriage. Another instance of
the influence of kinship ties became apparent in the fact that to-day sons
woul d stil l agree to stay in the efakd, while brother’s sons decline to do
so. Ye t we have also seen that even sons desire under certain conditions
to be released fro m the efako-unit. In the case of individual crafts and
trade, kinship obligations, even of sons to fathers, are no longer recog•
nized. Th e dissolution of the large labour-unit, moreover, has made it
abundantly clear that its size and coherence is determined predomi•
nantly by economic interests and relations. Kinshi p bonds and senti•
ments are not strong enough to maintain a labour group once its
economic usefulness is impaired. Generally speaking, then, kinship
bonds, in order to supply a substratum for productive co-operation,
must have the support of economic incentives—material benefits and
forms of compensation. Where the latter fail, the kinship bonds them •
selves wil l fail i n this task.
Thoug h we cannot thus admit the effectiveness of kinship factors as such (at least not in a decisive measure) upon the labour organization, the reverse is no doubt true. Th e labour organization exercises a two • fold influence upon the structure of the family and kinship group: (1) through contributing to the co-operative mechanisms withi n the kinship group, and (2) through accepting the responsibility for certain tasks whic h are vita l to the continuity and stability of the kinship group.
(1) Th e integration of the kinship group, that is, the ‘internal adapta• tion ‘ of the social uni t (Radcliffe-Brown), is largely determined by the scope and closeness of contacts and forms of co-operation existing in the group. Co-operation in the same labour-unit thus represents one of the most important integrative factors in this sense. We remember also that the labour-unit coincides wit h the household-unit, and that the individuals who agree to work together also ‘keep house’, eat, and live together. Th e constant contacts and the lasting mutual adjustment entailed in this productive arrangement buil d up strongly coherent units. Th e reduction in size, and the splitting-up into small individual units, of the large labour-unit, must detract essentially from the integration of the family group.
(2) Speaking generally, the large labour group, led by ol d and ex• perienced men, is more far-sighted in its economic planning than the small one-man or two-men unit . It allows for certain group interests whic h the individual worker wil l ignore, or discount as of lesser impor • tance ; i t can assess necessities and wants more objectively, wit h a view t o the fundamental needs of the group rather than immediate individual satisfaction. Thu s the family head wil l place foremost the necessity of finding the bride-price for the junio r family members in order to
376 ECONOMICS AN D SOCIAL STRUCTURE
establish them socially as ful l members of the community, i.e. as married men, and, indirectly, to assure the continuity of the family group. We have seen that even if the family head is unable to produce the necessary amount himself, and the young men have themselves to contribute, the family head woul d keep and save the money for them. Th e young men, left to themselves, are tempted to squander money on things whic h appear to them to be of greater urgency, and to disregard, to their own detriment as wel l as to the detriment of the group, the more distant necessity of marriage. As in the case of the Bida labourers, their money
‘goes at once’, and they remain unmarried even a considerable time after having reached marriageable age. In the extreme case a family head woul d even pawn the labour and liberty of a younger family member who is less essential to the group for the sake of keeping in freedom the elder, more important family member, or enabling hi m to found a family. We can go even farther than this. Older men, tending to uphold tradition, wil l plan labour and expenditure so that there should be money, food, as wel l as time available for family feasts or communal religious ceremonies. Thei r efforts thus serve (unconsciously) the interests of group integra• tio n at large, whic h the younger generation, not so fully steeped in tradi • tional ways of thinking , may hesitate to pu t before more immediate attractions. Finally, by submitting to the authority of the older men in productive co-operation, the labour-unit also sustains their position of authority in a wider sense, as leaders of the community. We can say, then, that the large labour-unit is committed to the maintenance of the structure and the traditional values of the society, and that wit h its dissolution an important agency of social integration and continuity disappears.
Th e changes whic h the gunnu ritua l has undergone recently are a con• vincing instance of this kind . In its traditional for m the gunnu con• tained a certain most stirring scene whic h took place on the eve of the ritual , and whic h concerned the young men of the village especially. The y were herded together in a certain place in the uninhabited bush, where they had to spend the night ; they were to wait for a terrifying apparition, the spirit of the ndako gboyd, who woul d question them about their moral conduct in the preceding year. If they were found guilt y of having behaved disrespectfully towards their elders, they were flogged by their comrades at the command of the spirit. I have quoted already the article in whic h I described in detail the ritual and its effectiveness as a mechanism of social integration in the family group as wel l as the community.1 But even from this short description we can gather the psychological importance of a ritua l whic h uses such impressive stimul i to press home moral teaching, and to bin d together periodically, in a common experience, the growing generation. Now , although the gunnu is stil l performed, the ‘bush night ‘ has been abandoned in many villages, and the explanation given was everywhere that the young men had re• fused to be subjected to the ordeal of this inquisition and punishment. Th e young men themselves were outspoken about the reasons of their refusal: ‘W e are now workin g as independently as the ol d men, we are paying tax for ourselves—why should we be flogged and threatened as if
1 See pp. 25, 63.
ECONOMICS AN D SOCIAL STRUCTURE 377 we were small boys?’ In their ow n words they confirm my argument
that the weakening of productive co-operation entails a weakening also
of other forms and bonds of group unity .
We may for the moment reflect on the fact that this progressive weak• ening of group integration occurs in a society wit h decreasing popula• tion . Partly this development underlines the strength and urgency of the forces whic h are workin g for the dissolution of the large labour-unit, and whic h here foster the tendency to split and reduce in size a working – uni t whic h is already, as it were, automatically declining. But it also reveals the significant fact that a loosening integration and progressive subdivision of units is not invariably correlated wit h an increase in the size of groups, and the inherent difficulty of maintaining cohesion.1
Th e present discussion has transcended the domain of economics. Where previously we have attempted to demonstrate the effects of changes in the structure of the labour-unit withi n the range of economic relations, i.e. on productive organization and the stability of the economic system, we are now studying the significance of these changes for the organization and stability of the social group. We can trace the influence of the changing economic organization upon social structure through wider and wider circles. Take the institution of the Nupe market. We have seen that the peasant attendance of the local village markets coincides wit h a local group characterized by regular social contacts, intermarriage and ‘ common living ‘ in general. Th e market meeting mobilizes the local community, adds to the common interests, and forms an important focus for these contacts and forms of inter• course. Lik e the labour-unit on a smaller scale, it contributes essentially to social integration. Thi s is no longer true of professional trade, whic h is carried on across community boundaries, and remains independent of established group-ties. Lik e the modern labour-unit, professional trade accepts solely the dictate of the impersonal profit motive.
Expanding trade connexions of this kin d are only possible in a social system whic h provides safety and protection outside the narrow boun• daries of the local community. The y depend, in other words, on the large-scale political organization of the state. Lik e the economic relations whic h take place withi n its confines, the integrative mechanism of the state is essentially impersonal, and based on abstract principles rather than forms of co-adaptation growing from ‘common living’ . Th e trans• formation in the group substratum of economic relations is thus deter• mined by the same concepts whic h also determine the transformation of the largest group unit , from community to state. Technically, the ex• pansion of trade and commerce, production and exchange, woul d have been impossible withou t the corresponding expansion of political con• trol . It would , perhaps, be going too far to say that spiritually also the former woul d have been impossible withou t the latter—without the conceptual framework whic h the latter provided. But we are justified in saying that the two forms of expansion reflect, materially as well as spiritually, two aspects of the same social process.
1 This latter view was expressed, in the form of a ‘social law’, by Professor
Radcliffe-Brown.
CHAPTE R X X I I
EDUCATIO N FOR CITIZENSHI P
AS education in the widest sense must be described all efforts, organized
/ \ o r otherwise, whic h ten d t o impar t t o individuals habitual forms o f acting an d thinking . Understoo d in this sense, every social institution involves educational activities by means of whic h individuals are led to acquire the behaviour-patterns that make u p th e institution . Thu s our description of economic life has involved th e description of certain educational techniques—the trainin g i n productive methods, the impart • in g of economic knowledge, or of concepts of propert y and wealth . Bu t most societies also operate wit h mor e formalize d methods of education, specific educational institutions whic h make it thei r business to trai n the rising generation in forms of acting and thinkin g regarded as essential by th e society. W e shall deal i n this chapter wit h these formalize d educa• tiona l institutions, o f whic h there are tw o i n Nupe : the institution o f the Kora n school, concerned wit h Mohammeda n religious teaching, an d th e institutio n o f age-grade associations. Th e shortest definition o f a n educational institution i s b y th e ‘type’ whic h i t i s mean t t o produce. I n this sense we can say tha t Mohammeda n religious education produces th e Mallam , the Mohammeda n scholar, and th e age-grade associations the zoon politikon as understood in Nupe . We shall see that in certain respects bot h types of education can be identified wit h motives an d aims underlyin g the political system, an d thus wit h institutions devoted to wha t I have called in the heading of this chapter ‘Educatio n for Citizen • ship’.
MALLA M SCHOOLS
Ther e are man y degrees and varieties o f Mohammeda n learning. Th e titl e Dallam’— md n i n Nupe—whic h one encounters s o often among th e Nupe , is by no means a safe guide. Fo r the ma n wh o has gone throug h a few years’ study of the Koran , having learned by heart a few sura; th e sand-diviner o r charm-seller wh o writes ou t verses fro m the Kora n an d sells the m as amulets ; the ma n wh o has acquired real learning an d scholarship, an d wh o can read an d writ e i n Arabi c an d possibly i n Hausa—the y all are called Mallam . Thoug h the titl e is used indiscri • minately, no Nup e ignores the essential difference between scholarship an d th e thi n veneer o f semi-learning. Th e Nup e distinguish tw o grades of learning : th e lower grade is th e study of the Kora n (kurani), whic h consists mostly in th e memorizin g of prayers an d whol e pages an d chapters fro m the Koran ; the higher grade is called kpikpe, knowledge, or litafi (Hausa) , book, an d consists in the study of Arabi c and the care• fu l reading o f the Kora n an d the better-know n commentaries. Th e most advanced scholars woul d study, in additio n to these books, also the historical records of th e Nigeria n Emirate s whic h have been compile d (i n Hausa , writte n in Arabi c script) by .scholars of forme r days. Onl y a student of ‘ knowledge’ can clai m to be a ‘real’ Mallam . Whil e there are
EDUCATIO N FOR CITIZENSHI P 379 hundreds of Mallams of the lower grade, there are few real scholars, and
they are well known and highly esteemed. These scholars will occupy
that most influential position which a Mohammedan culture accords to the men of learning.1 Their advice will be sought by friends and strangers, in matters spiritual or profane.2 Th e important political and religious offices of judge, scribe, and Imam at the town mosque are open to them. Some become attached to a noble household as ‘ private chaplain’. Royal princes and noblemen will seek their friendship, give them their daughters in marriage without bride-price, regarding marriage with a Mallam as sddaka, pious alms. Such is, the Nupe say, the cinwa katu, the Greatness of Learning. To some extent, however, the high reputation of the learned profession is reflected also in the persons of lesser erudition. Learning is, after all, relative. And a Mallam of very inferior scholarship judged by Bida standards, may be revered as a fount of knowledge in a
small half-pagan village. These Mallams in the villages are frequently supported by the people and receive annual gifts of crops as dzanka, tithe.
The Mallams are, above all, teachers. They teach both children and adults, the former the Koran and the latter ‘knowledge’. Every Mallam may, besides, be called in to perform a marriage ceremony or to officiate at a funeral—services for which they are remunerated according to their status and reputation. A few are well-to-do and own land, which is worked by their sons and pupils or, in some cases, by wage-labour. But most Mallams practise, in addition to teaching, some other occupation, which may or may not be associated with their scholarly profession, to supplement the meagre income of a Koran teacher. Some write out and illuminate pages of the Koran, or copy Arabic manuscripts, for sale. You can buy a sheet of this sort, very beautifully illuminated, for $d. on the Bida market; more ambitious scribes would sell a whole copy of the Koran and receive for it—the work of five months—£1. Students of
‘ knowledge’ eagerly buy these copies and use them in their studies. Other Mallams, again, embroider gowns, make straw hats or saddle• cloths, or do tailoring.
Th e size and standard of the Koran schools varies as widely as the status and standard of learning of the Mallams. You find Schools’ with two little boys for pupils, and these close relations of the teacher; and you find others—the schools of famous Bida scholars—the classrooms of which (as a rule the katamba in the Mallam’s house) are crowded with pupils. Th e teachers are very jealous of the reputation of their school. When I was a regular visitor to Bida Koran schools the Mallam of one which I had not yet visited came to me complaining that I had left out his school, and quoted as reference a Bida Mallam of very high standing who was a special friend of mine. In the large Koran schools the Mallam himself teaches only the adult students of ‘knowledge’, while the elemen• tary teaching is carried out by his sons or certain advanced pupils.
1 Bida learning has a great reputation even outside Nupe; Mallams from Bida have been given important posts in other countries, and several Bida Koran teachers have established flourishing schools in Ilorin.
2 One aspect of this spiritual advice is divining, which, as we have seen, plays
an important part even in such political activities as the election of a high officer of state or the preparation for war.
380 EDUCATIO N FOR CITIZENSHI P
Certain progressive Bida Mallams include the teaching of Hausa and elementary arithmetic in their syllabus, and many adults, traders and other native business men, attend their classes to acquire knowledge so invaluable to their trade.1 In certain Bida schools you will find young girls being taught together with boys. They come from the families of the nobility, which, conscientious Mohammedans, insist on elementary religious instruction being extended also to women. Girls, however, are only taught the Koran, and do not study ‘knowledge’.
Adult students, specially those who study the Koran or ‘knowledge’ for their edification, following another profession besides, visit the house of the teacher only for their lessons, which take place mostly in the even• ings, once or twice a week, or even at longer intervals, according to the leisure which their occupation allows them.2 Young pupils who live in the neighbourhood attend their classes daily while living with their family. Pupils whose family live some distance away will stay in the house of the teacher as members of his household.
Th e teacher of the Koran may not demand special payment for the tuition; according to Mohammedan principles religious instruction must be free—ebo albarka, ‘for the sake of blessing’. Th e wealthy Mallam, in fact, scorns all such payments as contrary to the spirit of religion. Th e poorer Mallam depends on what the family of the pupil offers him as sddaka, alms: a penny or twopence every two months, or a bundle of grain (worth gd. or is.) once a year. Even these small gifts are very irregular; in the school of a Bida Mallam of good reputation, whose pupils numbered thirteen, half of the parents had sent nothing for over a year. Sometimes it happens that a boy’s father dies while the boy is still studying, and his relations are unwilling to pay for his education. In this case the Mallam will continue to teach the orphan, let him live in his house and give him food and clothing, again ebo albarka only. Koran teachers in the country are worse off than their colleagues in Bida. In Kacha a Mallam who taught Hausa and arithmetic in addition to the Koran and ‘knowledge’, and who had eight young and fifteen adult students, received no payment from his students except an occasional gift of kola-nuts. If he asked for money, the students would indignantly threaten to leave and to seek another teacher. But both poor and wealthy Mallams receive a final gift from the parents of the student who has completed his studies, either money (in Bida), between 5$. and 10s. > or (in the village) grain or a goat. Th e teachers can also obtain a certain indirect remuneration by letting their pupils work for them. School- work does not take up the whole day, nor every day in the week. In their
1 Th e Education Department in Bida has arranged special evening classes for these enterprising Koran school teachers, and gives them also certain practical assistance, for example, by supplying them with blackboards. This scheme of adult education is of considerable value in the spreading of elementary education among that section of the population which has not had the advantage of educa• tion in the Government schools.
2 Students in a Koran school in Kacha, who were canoe-men by profession,
attended the school whenever they returned to Kacha from their voyages, inter• rupting their classes for weeks and months at a time when they were out on a canoe expedition.
EDUCATIO N FOR CITIZENSHI P 381 spare time the boys wil l help i n the house or on the farm, or sometimes
in the handicrafts whic h their teachers practise.
Th e students of ‘knowledge’ pay nothing for the tuition . But they have studied the Kora n wit h the same Malla m under who m they are now studying ‘knowledge’; and they wil l send their children again to his school, or to the school of his sons. Thu s reputations are buil t up and fir m links established between families of Kora n teachers and families whic h patronize their school. There are Malla m families in Bida whic h have been teaching three generations of the same families. For the sons of famous Mallams wil l follow the profession of their father. Man y also intermarry wit h other Malla m families. The y carry on the reputation for scholarship won by their fathers and families, and thus help to buil d up the dynasties of Mallams, of whic h there are many in Bida and Nupe.
As teaching implies so meagre and irregular an income one migh t ask how it happens that certain Mallams become well-to-do, and even joi n the wealthy landowning class. These Mallams do not come from the families of the hereditary or office nobility . Althoug h many men of this class study the Kora n and ‘knowledge’, and proudly call themselves Mallams, they do not adopt the teaching profession, but remain amateurs, noble dilettanti. Sometimes a Malla m who in the past attained one of the official positions open to Mohammedan scholars has laid the foundation of the wealth of the family. Another has benefited from the close association as the ‘guide, philosopher, and friend’ of an influential personage—a clientship of more exalted order—and received land and wealth through the help of his patron. It is, for example, customary for all marriage services in the royal family—a most lucrative office—to be held by the Malla m who is attached to the house of the Shaba; or at all funeral services of the royal house the Liman, the head of the Moham • medan clergy, wil l officiate. However, wealthy Mallams are rare. Mos t Mohammedan scholars are poor, or at least unable to maintain them• selves by their religious and educational work alone. But poverty is no impediment to scholarly reputation; nor does it detract from the honour of the profession. In fact, ordinary people wil l never call in a title d or even wealthy Malla m to perform a marriage or funeral service; they in • variably prefer a Malla m of the talaka class. A ‘great Malla m is too weighty’, they explain; they want a man who m they can offer a small sadaka as fee.
Amon g the Mallams of the tow n a certain loose economic co-opera• tio n exists, whic h comes into play at all such ceremonial occasions. At large marriage ceremonies or funerals several Mallams may be called in to assist the Malla m who is officiating. The y wil l also divide part o f the money whic h they receive for their services among their other col• leagues in the same tow n quarter. Or the family which celebrates the ceremony wil l send sadaka to all the Mallams in the neighbourhood. Thi s practice of communal benefits does not involve any closer or more formalized co-operation among the Mallams. Th e wide diversity be• tween them wit h regard to social status, wealth, and standard of learning forbids any more strongly pronounced appreciation of their common• ness beyond that fluid unity embodied i n the fact that they all work wit h
382 EDUCATIO N FOR CITIZENSHI P
their brains, and represent, in a society of peasants, craftsmen, and traders, the intelligentsia. This will become even clearer when we examine the motley rising generation of Mallams.
Koran teaching begins as a rule in early youth, when the child is 6 or even 5 years old. Th e teaching of the Koran itself is reckoned to take five to ten years. Some pupils learn quicker than others, and in one school I found a boy of 13 already studying ‘knowledge’, while one of 17 was still worrying his way through the Koran. But on the whole this elementary stage is considered mere routine work which most students can master in the given time. Th e teaching of ‘knowledge1, on the other hand, depends entirely on talent and individual inclination. Pupils may be eager enough to try, but will give up later when they discover that
‘they have no head for it’ (a literal translation of the Nupe phrase). Nor can the time taken by these higher studies be fixed precisely. Te n to fifteen years is an average estimate—at least, after fifteen years’ study one is reckoned to have become ‘fully a Mallam’. But only prospective teachers will take this advanced course systematically, and continue their study year after year. Th e dilettanti pursue their studies much more irregularly and leisurely, and students of 35 or more may often be seen in the schools of the teachers of ‘knowledge’.
Th e Mallam schools receive pupils from every class and every walk of life. I have compiled some figures to illustrate the social origin of the pupils of Bida Koran schools:
Father’s
profession Mallams Traders Farmers
Guild- crafts• men
Indivi• dual crafts•
men Nobility
Clerks, messengers,
&c. Total
Studying : Koran 49
‘Knowledge’ 28
16 3 0
4 2
6 2 0 22
2 0 8
8 151
10 72
That the sons of Mallams are in a majority is not surprising, nor that individual craftsmen and members of the nobility should come next in number. Of the twenty-eight sons of Mallams who were studying
‘knowledge’ six were going to adopt a different profession (of messengers, clerks, labourers, or servants to Europeans). Of the forty-four other students of ‘knowledge’ sixteen were going to become Koran teachers either in Bida or in the villages. These sixteen prospective Mallams in • cluded, incidentally, the two farmers’ sons. None of the sons of guild- craftsmen took up teaching as a profession: a proof of the strong organization of these ‘closed professions’. Th e large number of farmers who send their boys to Bida Koran schools may appear surprising; and even more so that a certain number of them should take up the study of the Koran professionally (for among the boys who are at present study• ing the Koran a few are likely to become Koran teachers). One example may serve: a boy of 13, the only son of a farmer, was studying to become a Mallam. He was well aware that at the death of his father the land would be lost to his family. Nevertheless he would not return to farm- work. His father had made him a Mallam, and that was all there was to it. Again and again, in similar cases, I was told: ‘We are no longer strong enough to work on the land’, or ‘After all, is the profession of a Mallam
A BIDA MALLA M AND ONE OF HIS PUPILS
AN AGE-GRADE ASSOCIATION DANCE
EDUCATIO N FOR CITIZENSHIP 383 not as good as that of a farmer?1 There is no evidence, incidentally, that physical disability, an ailment whic h migh t impair their working
capacity, is a reason for sending boys to Malla m schools. Th e choice of
the Malla m profession is, i n the full sense o f the word , a free choice of profession.1
Th e Malla m schools both in Bida and in the peasant districts are steadily increasing. Th e Nigerian Census mentions over 1,000 Koran schools in Niger Province, most of whic h are found in Nupe Emirate. I have estimated that Bida town possesses over 100 Kora n schools, wit h over 1,000 pupils (out of a population of 24,000, wit h 5,000 male children). Kacha has fifteen Kora n schools, Kutigi , a much more primitiv e and semi-pagan village, six, of whic h two already give higher instruction. Thi s spreading of Mohammedan education leaves no doubt that parents increasingly desire their children to receive the benefits of Mohammedan learning. We have to accept this as a general trend in modern Nupe society, not so much towards true education—which few of the boys wil l ever attain—as towards the higher social status—the status of a leisured-class intelligentsia—which the position of a Mallam , rightly or wrongly, suggests.
Earlier in this book I have referred to the profession of Mallams as one of the mechanisms of social mobility . I have also called Mohammedan education, by means of whic h the religion and, indirectly, the whole spiritual background of the rulin g classes is diffused throughout the country, one of the ‘bindin g forces’ of the Nupe state. Here we have found concrete evidence to support these statements. Le t me summarize the main conclusions whic h we can draw from the present discussion. First, the profession of Mallams comprises contrasts of wealth and rank, whic h occur nowhere else in the framework of a single profession. Th e only common feature is that of a life without manual work, and of a certain social status raised above that of the ordinary commoner. Secondly, the Malla m profession is the only profession whic h we dis• covered in Nupe whic h is not ruled by the motive of material gains. On the contrary, people wil l sacrifice even the security of work on the land for the meagre and insecure income that goes wit h the rise in status and the attainment of the life of the leisured class.
AGE-GRADE ASSOCIATIONS
Th e Nupe name for age-grade association is ena gbarufuzi, lit . ‘society’ or ‘association of the young men’.2 Although nearly every Nupe boy or
1 I had wondered whether there was any relation between the professional choice and the size of the family, whether perhaps in large families a son would be more likely to study for Mallam than in small families, where every member would be needed for farm-work. Then again, I thought it possible that in these days of the breaking-up of the family-unit orphans would be placed more readily in Mallam schools. Neither correlation, however, is true—as is shown in these figures: of 32 students of farming stock 9 were only sons, 10 were one of two sons, 9 had brothers who also studied the Koran, and 4 were orphans.
2 Informants who have picked up a few words of pidgin-English translated
ena as ‘company’.
384 EDUCATIO N FOR CITIZENSHI P
young man belongs to one or the other age-grade association, the age- grades are not a compulsory organization. No coercion is used to induce individuals to joi n the association, but failure to do so woul d be regarded as unusual, and under certain circumstances even suspicious. Th e con• stitution of the ena gbarufuz’i is based essentially on the agreement of a group of individuals who, wit h clear knowledge of the purpose and the rules of the organization, for m themselves into an age-grade association. Sociologically, then, our translation of ena as association is fully justified.
Th e description of the age-grade system of Nupe is complicated by the large number of variations that occur. There exist, above all, certain essential differences between the age-grade association in Bida and in the village; we shall, in fact, have to contrast the two throughout this dis• cussion. Th e rules governing the internal organization of the age- grades are elastic: much more so in Bida than in the village. In the village association, too, mino r differences occur, sometimes even between certain local associations in the same village. Thus , the age-factor itself is not rigidly defined, and the entrance age and age-limits of the different grades vary widely. No r is the duration of the single age-grade fixed universally or permanently. In the village we find a more rigi d arrange• ment: in Kutigi , for example, all age-grades last six years, in Jebba four years. In Bida the ter m of the age-grade association varies from four to ten years, and the duration may vary in different grades, in accordance wit h the wishes of the members.
In the village each grade is represented by one ena only, and age- grade associations ascend as a rule en bloc from grade to grade. Changes in the membership, so far as they occur, as the ena advances to higher grades, take the for m of a slight decrease caused by death or emigration. In Bida each grade is represented by a number of associations, and pro • spective members have the choice of alternative associations. Th e rules of membership are more flexible in Bida. One may skip a grade—Vest’, as the people say—one may start in a higher instead of the bottom grade, or one may leave the association before it advances to the highest grade, even before its ter m comes to an end. In two villages age-grade associa• tions of whic h I examined the membership, whic h was 17 and 18 in the highest grade, had remained almost unchanged through the three grades of whic h the age-grade association is composed. In Bida there is much coming and going. In two ena, for example, the highest grade of whic h numbered 26 and 30 members, respectively, only 7 and 15, respectively, had gone through all three grades, 16 and 10 through two grades, and
3 and 5 had joined the highest grade withou t having passed through the lower grades.
Th e most common and, in the eyes of the people, normal type of age- grade association comprises three grades, whic h are numbered fro m the top downward: the senior grade is called First Grade, the middle grade, Second, and the junio r grade, Thir d Grade or ena dzakangizi, Children’s Grade. We shall deal, in what follows, wit h this most typical three- grade organization. Th e average age in the senior grade is 20 to 30, in the second 15 to 20, and in the junio r grade 10 to 15. Th e age-limit for
EDUCATIO N FOR CITIZENSHI P 385 membership is about 30. It is interesting to note that the fact of marriage
is of no significance—the highest grade comprises both married and
unmarried men. Th e sign that a man is past the age of membership of
‘young men societies’ is the -appearance of the external symptom of
manhood, a beard—za na de nukpayi na da ena be a, ‘a man who has a
beard no longer joins the association’. In certain small communities the
age-grade system may be limited to the two higher grades; in other
communities a fourth, junior, grade is added, called ena wawagizi,
Association of the Small Children. This lowest grade is not taken
seriously: it is regarded as mere dzodzo, play—of the children-playing-
at-adults type—and shares with the other grades only their superficial
external characteristics. Th e junior, third, grade is also in some measure
regarded as less serious and significant than the higher grades, and as
essentially a preparatory stage. Certain important activities of the
higher grades are, as we shall see, of necessity excluded, or reduced to
insignificant proportions, in associations which are composed of young
boys. In one respect, however, the junior age-grade makes its own
specific contribution to the age-grade system; it is the only one which
has its counterpart also in the other sex. Th e highest age-grade, too,
possesses certain specific features, in the form of certain age-grade ranks
which are not bestowed on members of junior groups.
As ena membership is defined by age, and not by generation, it happens occasionally that, in large families, the young men of different generations but of approximately equal age belong to the same ena, and that young men of the same generation are members of different age- grade associations. Th e following genealogical diagram will illustrate a typical case. Th e letters S, M, and J indicate membership of the senior, middle, and junior age-grade, respectively; FH means family head, and a cross a deceased family member.
Th e separation from family and homelife is indeed one of the most characteristic features of the ena gbarufuzi. Th e people attach con• siderable weight to the fact that the bonds and forms of obligation which are implied in the ena are of an entirely different, and specific, nature. Even small boys, the youngest members of the junior ena, will tell you with an air of importance that the ena has nothing to do with family, home, or parents. Th e father ? ‘ Yes, we tell him, and we ask him some• times to help out with money or food for a feast which we arrange. But he has no say in our ena.* Fathers will confirm this. In Kutigi parents were quite surprised, and laughed uproariously, when I told them about the titles which had just been distributed in a newly started junior grade; the boys had gone with the times and introduced titles which did not exist in the days of their fathers* ena. But for my information the parents
c c
386 EDUCATION FOR CITIZENSHIP
would have remained unaware of these changes, which were regarded as a purely ‘internal’ matter of the age-grade association.
Let us now see how an age-grade association starts its life. We will begin with the Bida ena. Here the procedure is the same both in the foundation of a new ena and in its promotion to a higher grade. A number of boys or young men who have become well acquainted with one another in various walks of life, possibly when participating in a previous ena, decide to form themselves into an age-grade association. They meet in the house of one of them, a group of only six or seven members at first, and discuss how many more members they can muster. Each of the founders of the prospective ena then enlists other friends and acquaintances of the same age, till eventually the membership reaches the required number, twenty or twenty-five. The first thing now is to find a ‘captain’ and a ‘president’ for the new association, or, in native terminology, an Etsu and Ndakotsu (lit. Grandfather of the Etsu). The Etsu is a member of the age-group itself, as a rule one of the original founders, who is elected by the group.1 The Ndakotsu must be of senior age: in the junior grades he will be a member of the next higher grade, and in the senior grade an adult who has ‘passed’, go ga, i.e. already gone through all the grades. Under their Etsu the members of the prospective ena approach a certain young man whom they regard as a suitable candidate, and request him to accept the position of presi• dent in the new association. Every member of the Nupe age-grades is assigned a certain rank which defines his place on the miniature social scale of the age-grade association for the duration of the ena. It is the first official duty of the newly elected Ndakotsu and Etsu to call a general meeting in the house of the Ndakotsu at which the distribution of these ena ranks is to take place. The elastic rules which govern duration and membership of the Bida ena imply also that no fixed rules can be laid down with regard to the time when associations usually start or ‘change over’ to higher grades. Moreover, in these associations in which only a small nucleus of members may remain faithful to the ena throughout its three grades long waits are likely to occur between the termination of one ena and the founding of a new association of a higher grade.
In the village the procedure governing the foundation and promotion of the ena, although the same in principle, differs in certain significant features from that obtaining in Bida. To begin with, the time when age- grade associations are started and promoted to higher grades is fixed permanently: it falls towards the end of the dry season, an arrangement which gives the new association time to be fairly launched when the agricultural season begins, and ready to undertake the tasks which, in a farming community, fall to the age-grade associations. The enrolment of members for the new ena is much less selective and arbitrary. I n a small community where most boys of approximately the same age know each other, and normally meet in work or play, there is no need to canvass the new ena among friends and acquaintances. Nor is there any special search for captain and president of the new association. The village associations have, in fact, their Ndakotsu and Etsu appointed for
1 Sometimes the captain of the second grade bears a different title, Sokyara.
EDUCATIO N FOR CITIZENSHI P 387 the m in advance. Shortly before the existing associations are dissolved
previous to their reorganization on a higher grade, the outgoing presi•
dents of the different grades appoint, as a last official act of their ter m of
office, one of the members of their group as ‘ Ndakotsu-elcct’ of the next
junio r group, i.e. the group which , in the new term , wil l take the place of
the outgoing group. Th e Ndakotsu-elect wil l then take charge of the
organization or reorganization of the new ena, and also of the appoint•
ment of a captain for the new association.
In these age-grade associations whic h rise en bloc to higher grades there is no gap between the termination of the ol d and the foundation of the new ena. Th e association whic h is about to be dissolved and promoted stages a last public appearance in a great feast, wit h music and dancing, to whic h all friends and relations are invited. A few days later the members of the ena meet again for the preliminary discussion in the house of the new Ndakotsu, for the distribution of the age-grade ranks whic h wil l be assigned to them in the new ena. Whe n all ranks have been satisfactorily allotted the group arranges another large festivity, whic h is to announce publicly its reorganization and to inaugurate the new term . I had the good fortune to be in Kutig i in 1936, the year when the six years’ term of the local ena came to an end and the general ‘change• over’ took place. It fell in the middle of March ; the last public meeting of the ol d association was on the 7th of March , the session in the house of the Ndakotsu took place five days later, and the inaugurative meeting of the group in its new for m as a senior grade on the 2nd of April .
Now , what are the qualifications for the position of a Ndakotsu and Etsu, and according to what principles are the ena ranks distributed? For the election of the Ndakotsu a well-established social position, good reputation, intelligence, and also a pleasant nature and, if possible, an open hand, are regarded as essential qualifications. A satisfactory financial position is essential because the Ndakotsu wil l be expected to entertain and to organize parties—and to do it in style. Mos t of these conditions also apply to the election of the group captain. Th e members of a newly founded Bida ena explained to me: ‘W e made Mohamadu our Etsu because he is a nice fellow. We all like him . He has good con• nexions ; you see he is invited to the court ; besides, he has plenty of money, and both his father and his mother are rich . He is also ready to spend it when we have a feast and want drummers and dancers.’ On the other hand, a very ambitious young man who had been Etsu of an ena (of the second grade) and who hoped to be made Etsu again in a new, senior, association whose captain was resigning shortly, complained to me that his chances were slight, for he had not much money and could not afford the expenses whic h he woul d be expected to meet. It is implied in the nature of these qualifications that the Etsu is not always the oldest member in the group. In one ena whic h I analysed the Etsu was 20 when the ena started, and there were three members older than he. But as a rule he is of higher social position, and often better educated, than the rest of the members. Thu s in a Bida ena whic h consisted almost entirely of tailors and embroiderers the captain was a weaver, the son of the Bida guild-head. In another group of a similar composition the
388 EDUCATIO N FOR CITIZENSHI P
captain was an expert cap-maker, who made the expensive, fashionable embroidered caps. And in a group the members of which came mostly from the classes of the nobility, the captain was the son of a late Emir. It has become the fashion for captains of Bida age-grade associations to adopt an additional, allegoric or symbolic, title, which would also give the ena its name. Bida associations compete with each other in the in• vention of high-sounding names, which should evince the refinement and selectness of the ena. Thus one Etsu was known as Etsu Yimani,
‘King Faith’, as the ena over which he presided prided itself of the piety of its members; another captain called himself Etsu Hankuri, ‘King Forgiveness’, for more or less the same reason; a third Etsu Turu, ‘King of the Europeans’, a title which was meant to bear witness to the fact that this ena was organized on ‘European lines’ (to be explained presently).
Th e financial qualifications and those of status are of less importance in the village, where inequality of wealth and status are much less strongly marked, although even in the village ena the captain may be of a somewhat higher social position than the rest of the members, e.g. the son or brother of a village elder of high rank. Th e symbolic names of age-grade associations are absent in the village, where age-grade associa• tions are neither as numerous as in Bida nor as interested in advertising their refinement and selectness.
Th e allotment of ranks in a newly founded association of the youngest grade is largely a formality—although the popularity and influence of the young boys among their age-mates will to some extent be taken into account. We have seen that these junior age-grades are generally regarded as representing a preparatory stage rather than a fully fledged ena. In the higher grades the distribution of ranks becomes much more serious. Change of rank during the lifetime of the ena occurs only when a member dies or withdraws prematurely, especially if he was holding one of the higher ena ranks, which must be represented. But when the ena termin• ates and the members form themselves into a new group, a complete reshuffle takes place. As the different grades use largely the same rank- system this redistribution of ranks at the end of an ena makes conspicu• ous the promotion, non-promotion, or degradation—as the case may be. These changes in the holding of ena ranks reflect faithfully the success or failure of individual members in the previous association, the capability shown in the performance of group activities, and the popu• larity enjoyed among comrades and superiors. Th e eager competition notwithstanding, the task of distributing ranks justly and to everybody’s satisfaction does not, as a rule, take very long. I have seen these discus• sions carried through in one or two afternoons. But then, the general discussion is only a last, formal, step. Mostly, the two leaders of the group have decided the distribution at least of the more important ranks long before the first general meeting. Th e Ndakotsu of a prospective village ena, for example, knew a few months before the group was actually called into being exactly whom he was going to invest with the three highest ranks. As a rule the authority of captain and president is sufficiently strong to carry the vote. But instances of disagreement and
EDUCATIO N FOR CITIZENSHI P 389 long fruitless discussions also occur, both in town and village, and some•
times a compromise has to be effected by introducing a special title
(with the word Tsowa before any rank, equivalent to ‘vice-‘ or ‘sub-‘),
which will satisfy a refractory member without upsetting the arrange•
ment at large. Severe degradation, on the other hand, is rare, as it can
be easily circumvented in these associations which are not compulsory,
and cannot force members to accept their status. This is a typical in•
stance: A very ambitious and vain young man of Bida who had held
high ranks in previous ena, but was given a very low rank in a senior
grade, resented this treatment as unjust; he could, however, do nothing
except to show his disdain by disinteresting himself in the association,
and staying away from all their meetings. Actually he never told me of
his shameful treatment when we discussed his association with the ena;
I learned of it only by chance when; examining a certain association, I
came across his name on the list of members.
The village ena comprises one rank which can be won only in strict competition. We have spoken of this rank already (p. 250): it carries the title Sode, and is bestowed on the best farmer of the group. Sometimes there is also a second farmer’s rank, won in the same way for which, rather incongruously, the title Maiyaki, ‘Minister of War’, has been chosen. These ranks are taken only in the highest and middle grade, and are won in open competition at the first of the more important egbe occasions of the season, frequently at the egbe performed on the farms of the village chief. Traditionally, the title Sode ranked next to that of group captain in importance and precedence. To-day it has been de• graded to a lower place on the scale of ranks. Of course, neither the title nor the particular qualification for which it is bestowed is recognized in Bida. Th e rank of Sode is also the only rank that may change hands (at least in theory) during the term of the ena. Although this happens very rarely (I have not met with any actual instance), all people agree that if at one of the occasions of collective farm-work an appointed Sode proved himself to be inferior in skill to another member of the ena, he would have to surrender his title.
Age-grade ranks are largely copied from the political rank-system obtaining in the community to which the ena belongs. It is not a very exact copy—although the ena tends to arrange its ranks according to the order of precedence which they follow in the political hierarchy. In Bida the distinctions between the different classes of political ranks are not observed, and military and civil ranks, office titles and titles of the royal nobility, are bestowed indiscriminately. Th e village ena remains more faithful to the rank-system of the community, and would, for example, adopt its typical, possibly uncommon, titles of elders. But as the membership of the ena—especially of the junior grades—may be considerably larger than the number of titled village elders, ranks from outside the traditional rank-system may have to be introduced, prefer• ably from the rank-system of the capital. Th e younger generation is indeed always ready and eager to adopt such innovations. Rank-systems of heterogeneous nature, and rather arbitrary precedence are the result. However, we must not forget that the rank-systems of many villages are
390 EDUCATIO N FOR CITIZENSHI P
themselves similarly heterogeneous and arbitrary, including Bida titles among its traditional village ranks.
Th e system of precedence involved in ena ranks and grades is taken very seriously, and the rigid rules of etiquette which govern it are enacted with all solemnity. Much of the ena business takes place, as we shall see, within the single ena; but when members of different grades meet—for example, at family feasts in the house of a member, or at col• lective farm-work—submissive or condescending behaviour marks their relation towards each other. Members of a junior grade will never dance when members of senior grades are present, except with their special permission and at their request; in collective farm-work the junior grades work at the back of the team, at less important and less con• spicuous tasks. Age-grade etiquette is even more emphatic within the individual association. At communal meals the ‘subalterns’ will wait till their superiors have eaten; at ena feasts subaltern and senior ranks will never dance at the same time; when greeting a superior, the age-grade member of lower rank will squat down and bend his head, exactly as prescribed for intercourse between superiors and inferiors in the social scale of ‘real life’.
In the village ena it is only in general terms that the ranks indicate social position on the rank-scale of the association. In the town ena the ranks are sometimes chosen so as to indicate and do justice to the special gifts, or the special social position outside the ena, which characterize individual members. In this the organization of the town ena reflects the more highly differentiated social structure of urban society. Thus a Bida ena, composed largely of illiterate members, assigned to a young man who had been educated in the Government Middle School the title Naibly a rank of the Mohammedan clergy which can be held only by scholars. Th e rank-list of another ena included the rank Boroti, lit. Head of the Boror6, its holder being the son of the late Emir’s Regent-delegate’ for the Boror6 Fulani. Some Bida age-grade ranks indicate a certain official capacity held by a member in the ena. Thus I found a rank Dzufa (lit. pocket) bestowed on a young man who acted as ‘purser’ of the group. Al l purchases and payments for ena purposes were made through him ; when, for example, a member decided to give a party for the group he would place his money contribution in the hands of the Etsu, who would then hand it over to the ‘ purser’ as the member responsible for financial arrangements. In one ena I found the rather curious rank Mataushe (Hausa, lit. ‘the man who rubs’, i.e. ‘masseur’), by which name a certain slave or body-servant of the Nupe king was known whose special duty it was to massage the feet of his master. In like manner it was the duty of this ‘office holder’ in the age-grade association to massage the feet of his Etsu at meetings of the group when his master felt tired.
One group of ranks typical of the Bida ena deserves special attention, namely the ranks which are organized on ‘European lines’ and copy, not ranks and offices of Nupe society, but ranks, offices, and even names of Europeans with whom the Nupe had come into contact. Such titles are often amusing and even puzzling. In one ena the head of the group was
EDUCATIO N FOR CITIZENSHI P 391 not called Etsu but Gomna Ture, European Governor. Other ranks in
the same ena were Kantoma (Station Magistrate), Joji (District Officer),
Likita (Doctor); there was also the rank Dupienne, derived from the
name of Mr . Dupigny, a former Resident of Niger Province, a rank
Borti (Major Burton), Mallam Smi (Mr . Smith, a former Superintendent
of Education), and Jamsi Dogo (lit. Long James, a title which referred to
a former clerk in the Divisional Office); it was, not unnaturally, the
lowest title on the rank-list. Another ena organized on modern lines was
divided into two sections, a ‘native’ and a ‘European’ section, the former
consisting of Nupe titles, the latter of Europeanized names and ranks,
such as Gomna, Joji, Likita, Rezdet (Resident), Kantoma, Misisi (Mrs.) ,
Hafissa (Officer), and Karamin Joji (Assistant District Officer). In the
normal business of the ena these two sections behaved differently, and
were bound by different rules of etiquette, native and ‘European’, which
meant that the members of the ‘European section’ had to greet each
other in European fashion, by lifting their caps, and had to eat at a table
with fork and spoon. I could not help laughing when these rules were
explained to me; but the head of the group pointed out, rather indig•
nantly, that this was the most natural thing in the world; after all, ‘a
Governor does not eat with his hands’.
Th e seriousness with which these distinctions of ranks are observed, the realism with which this playing at society a la Europeenne is enacted, forbids us to treat it merely as a childish and meaningless make-belief. Psychologically, it has the significance of a substitute for thrills and achievements which normal life cannot offer. Th e inclusion of titles and forms of conduct copied from the British officials in the country is part of that ‘realism’ which marks age-grade etiquette and precedence: it does justice to the new political order of the society, and partly also to the fact that high social position in present-day native society implies intimacy with the Administration and high British officials. If it were asked why the Bida ena chose as symbols for social gradation ranks and offices so far removed from their ordinary life, I should point out that there is less difference than one might assume between the two models which the ena could choose—the model of European and the model of native society. To the average Bida man the highest ranks and offices of his own society are no less remote and inaccessible than the office of a District Officer or Governor. We must add that the choice of a particular rank-system for the ena is in no way determined by the social stratum from which it is recruited. Bida age-grade associations of talaka and saraki alike adopt the highest ranks both of traditional and British society. Only for a small minority—the few ena of the ruling class—may these ranks and offices, and the whole exalted social life which they depict, one day become reality. For the large majority they represent an infinitely remote, imaginary social existence.
Here we are led to the wider problem of the relation between the world of pretence built up in the age-grade association and the values of ‘real’, that is, adult, social life. It can best be approached through a comparison of village and town ena. In the village association the social life enacted in the ena, and its prototype in adult life, stand in the relation of potentiality
392 EDUCATIO N FOR CITIZENSHI P
and reality. Th e prototypes of ena ranks are all (at least theoretically) within the reach of the individual members. Every one of the young men who have gone through the ena may find when he reaches the suit• able age that elimination by death or illness of senior relatives has singled him out for the position of a family head, and thus for a nusa rank. Indeed, the promotion which accompanied his age-grade career will serve as an important qualification in the bestowal of the real rank by chiefs and elders. In certain respects all ena ranks of the senior grade retain their validity outside the ena, and after its expiry. Th e code of age-grade etiquette expires with the age-grade association. Not so the symbols of age-grade achievement: a man will be called by his rank rather than his name years after the ena has come to an end, often even for life—especially if he had held a very high rank. Me n who have once been elected as Etsu or Ndakotsu, even when they are not, or not yet, appointed to a nusa rank, will occupy a position of some distinction in their native village. Not only are they addressed and known by their ena titles, but village elders will pay attention to their opinion, and will entrust them with responsible tasks—in the organization of a ceremony, or of important collective work carried on outside the framework of the age-grade association. Ena ranks will never be confused with the political ranks of the village; they remain tici nya ena, ranks of the age- grade association. But although still only models of the ‘real thing’, they are ‘real’ in the sense that they indicate the varying qualification of individuals for access to the final social advancement. In the village ena, then, and similarly in the few age-grade associations in Bida which are recruited from the ruling classes, the ena rank-system anticipates a future social status; it introduces social concepts, and practises forms of con• duct the effectiveness of which will be proved in adult life.
None of this is true of the average Bida association. Although here, again, the ranks attained in the senior grade accompany a man through life, they can claim no significance beyond that accorded to them by the rules of the game. A game indeed it is. These ranks let you enjoy fictitious contacts with a world of rank and power which you can never hope to enter in reality. Their upward trend, their incentive of competi• tion, imply, not preparation or anticipation, but pretence, imitation—
‘substitution’.
Let us examine now the composition of the age-grade associations with regard to the social groups and strata which they absorb. In small villages a single age-grade association includes the youths of the whole village, and may even go beyond it, including the ‘dependencies’ of the village, the small tunga settlements, or linking a number of tunga settlements with one another. In the scattered hamlets round Bida the boys and young men of three to four hamlets join in a common ena. In very large villages (e.g. Kutigi) the age-grade is subdivided into a number of associations, each embracing one large, or two or three small, neighbouring, village efu. Th e age-grade itself, however, stretches across the whole village community, and the different associations con• form to the same pattern with regard to the length of the term, and the time and year of the ‘change-over’ from grade to grade. In Bida the sub-
EDUCATIO N FOR CITIZENSHI P 393
division goes much farther, and we find a large number of disconnected associations, which are no longer co-ordinated or conform to a common pattern. I estimate the number of age-grade associations in Bida at
- Th e individual ena is, again, bound up with locality, and includes boys and young men from the same part of the town. But here two new factors enter: social differentiation, and divided political allegiance. Th e very fact that the Bida ena finds its members among friends and close acquaintances already expresses the tendency to make it socially homogeneous, and to limit its membership to individuals from the same social stratum, or the same professional group. Saraki£i and talakazi never mix in the same ena. Nor do prosperous traders or Mallams join the same ena as peasants or small craftsmen; nor does, finally, the youth of the craft-guilds intermingle—glass-workers, blacksmiths, weavers, all have their separate age-grade associations. Th e limitation with regard to locality and close acquaintances is less rigid in the highest grades than in the junior grades. For as in the course of years the membership decreases, some members having died, others having left the ena or perhaps the town, it sometimes becomes necessary to open the member• ship of the higest grade to people from other parts of the town. Yet no ena would accept members from beyond the boundary of the town ekpa. Th e political solidarity of the three dynastic divisions of the town is thus reflected in, and even, as we shall see, actively supported by, the age- grade associations of the capital.
Here once more we gain insight into the social significance of the Nupe age-grade association: it underlines and fosters the solidarity of the existing social groupings. This means something different in the village and in the town. In the former it refers to the village community at large, and we find the age-grade associations stretching across it, and across the large kinship units and ‘houses’ into which the village society is divided. We can liken the age-grade system to a horizontal plane cutting across the separate structures of the extended families, with their wide base, tapering off to the apex of the one official representative of the family group, the titled elder. Th e age-grade association must thus counteract all possible separatist tendencies of individual kinship groups and sustain the large-scale integration on which the existence of the village as a social unit depends. In large villages this integrative scheme is in some measure narrowed down, owing to the size of the groups that can be absorbed effectively in a single age-grade association. It assumes, in fact, a somewhat different aspect, in uniting the extended kinship units within themselves—in the form of the large efu group—that is, in holding together the widely expanded structure which is to support the apex of a common official representative.
In the town the existing social grouping is created and held together by political and class interests. Here again, the age-grade association fosters the unity of a group which is to submit, in ‘real’ life, to a com• mon head. But he no longer represents the ‘apex’ of an organically grown structure; as the ena is largely the product of a free, voluntary association, the group loyalty which it fosters is towards the freely chosen leader of a local political faction.
394 EDUCATIO N FOR CITIZENSHI P
Th e integrative tendencies in the ena are expressed in every one of the typical age-grade activities. In the village they revolve round the three centres of social life : work, recreation, religion. In the Bida ena the first is reduced to insignificant dimensions; the second is raised to almost paramount importance; the thir d remains on the whole unchanged, being only transferred to a different plane—Mohammedan religion.
I have already spoken of the role whic h the ena plays in organizing collective farm-work in the village. We need only add that these activities concern the higher age-grades more than the junio r groups, the members of whic h are not yet skilled farmers. It is commonly agreed that every young man who has gone through the three grades leaves the ena a fully trained farmer. No t everyone can become a Sode; bu t the success of the team counts for much. Here the Ndakotsu wil l prove his capability as organizer. He wil l pu t his pride int o making his team the most efficient—all means being counted as fair. An ol d man, stil l known as Ndakotsu, tol d me how he used to help his group of workers to defeat all the other teams by means of a charm whic h ‘made their hoes go twice as fast.’
Under recreational aims we must understand both recreation in the narrow sense—age-grade feasts and ‘parties’—and the conventional festivities wit h whic h the Nup e celebrate kinship events: naming cere• monies, weddings, and funerals. Th e recreational activities in the narrow sense concern the individua l association; in the celebration of kinship events to whic h all young men who are related to the family are invited, members of different grades frequently join . Th e Ndakotsu, Etsu, and also the members of higher rank are expected to entertain the group i n their houses at fairly regular intervals. There wil l be food, drinks, kola-nuts, and often drummers and dancers. In Bida the mem• bers of the ena meet at least once a mont h at such parties. Smaller festivities are attended only by the ena members themselves; to the larger meetings, especially those whic h mark the beginning and end of the ena term , friends and relations of both sexes are also invited . At one of the festivities wit h whic h the inauguration of the new age-grade ter m was celebrated in Kutig i I counted wel l over a hundred guests.
Le t me take you to such an age-grade feast in Bida. It isliel d in the courtyard of the house. It starts in the afternoon or evening and lasts far into the night. Th e guests sit and squat in a closely packed circle, leaving a small space in the centre for drummers and dancers. On one side the host is sitting together wit h Etsu and Ndakotsu; as a rule they only sit and watch the performance; bu t the guests are ready at.every moment to exchange the role of spectators for that of performers. There exist special ena gbarufuzi dances, performed at these occasions: they consist of a slow, graceful forward-and-backward movement of a row of dancers, five or six young men, who dance side by side, the arms stretched out and the body bent slightly forward, setting their feet in small, measured, steps. Group follows group in the dancing arena. Individua l dancers leap into the ‘arena’, and perform a short solo or
EDUCATIO N FOR CITIZENSHI P 395
pas de deux of leaps and cart-wheels which, when executed skilfully, will be greeted with applause or, when less successful, with good-natured laughter and banter. There are special drum-signals for every higher ena rank which are sounded when the young men take the scene and join in the dance. Rich age-grade associations, or certain high rank-holders in the association, even have special dances and songs composed for them, which are played, not only at these meetings of the ena, but also whenever the ena members appear elsewhere at a public gathering. Drummers and woman dancer would walk up to the young man, and the dancer would recite the song which she composed in his honour.
Moonlight makes all things good to look at.
The maba bids the townsfolk farewell for the night. Give us of thy knowledge,
Ibrahim Ganleyi,
Thou grandson of wealth.1
Drum-signals and songs are not the only symbols of age-grade status. At every such festivity one of the drummers or some self-appointed
‘ master of ceremonies’ will assume the role of an announcer. He is called maba; announcer and jester in one, he must be witty and have the gift of the gab. He collects the small gifts of money which guests make to drummers and dancers; he announces the name and rank of the donor and recites the virtues of his ancestors and famous relations (im- pressiveness counting more than truthfulness), adding such praises or witticisms as seem appropriate to the rank of the donor and the amount of the gift. We see again how closely Bida age-grade etiquette follows its model—the Nupe nobility; it copies every detail of its symbols of status and wealth, even such features as the specific interpretation of dignity forbidding persons of high rank to join in the common dance.
It is easily understood that in the village, along with the model, the copy must be absent also. At the gatherings of the village ena profes• sional dancers and musicians are rarely summoned. There is no maba and no extolling the feats and virtues of celebrated ancestors. Age- grades and age-grade ranks ‘own’ their special drum-signals and dances; but they are associated with the group, and not with certain individuals who can afford to pay for the distinction.
Th e religious events in which age-grade associations participate con• cern all grades. In the pagan ceremony especially the different grades are allotted different tasks, which imply in varying measure admission to the esoteric ritual activities. The main religious ceremony of pagan Nupe, the gunnu, makes the ena members the protagonists of its ritual drama and dances. Another ceremony, which occurs only in the Benu
1 This song, which was composed by the zaworo group of dancers (see p. 302), in the honour of the young man called Ibrahim Ganleyi, was recorded at an age- grade feast in the house of a well-known Bida Mallam. It begins in the fashion of all Nupe songs with the quotation of a well-known proverb. The term maba which occurs in the second line is explained in the description that follows.
396 EDUCATIO N FOR CITIZENSHI P
sub-tribe of Kutigi , Enagi, and Dabba, the gant even dramatizes the age- grade structure itself, its system of ranks, and the promotion from grade to grade. Th e gani assigns different ceremonial duties to the different grades, and involves certain symbolic tests of manhood of progressive severity whic h the different age-grades, one after the other, must under• go. Th e ceremony can, in fact, be regarded as a ritual of age-grade initia • tion . However, I must leave the description of this ceremony, closely associated though it is wit h the conception of age-grades, for a future occasion.
Le t us tur n to the Mohammedan religious ceremonies. At the cele• bration of the Sallah the age-grade associations play a much less specific role. The y arrange a small festivity in the house of their group captains on the eve of the Sallah, at which—as in all houses of Bida and Moham • medan Nup e on that evening—a ram is kille d and eaten. On the follow• in g morning the various ena of the tow n meet in the market to joi n the thousands o f people who wil l follow the royal procession to the mosque and round the tow n walls, and gather later in front of the Emir’ s house, listening to drummers and musicians, and watching the display of the horsemen and the coming and going of the royal guests.
Th e celebration of the Mohammedan Ne w Year, on the other hand, called navu in Nup e (lit . torches), represents the climax of the public appearances of the Bida ena. Th e navu is essentially a festival of youth . Its ritua l acts, whic h appear to express, symbolically, the death and renewal of the year, are carried out by the young folk of Bida. On the eve of the Ne w Year, at sunset, all the age-grade associations meet in the houses of their captains, and then march to the market of their tow n quarter, where all the associations of the ekpa forgather. The y carry burnin g torches of grass in their hands, and are clad in simple, scanty garments. Three long processions, one in each ekpa, move slowly through the streets of Bida, down to the arm of the river flowing through that part of the town . Th e boys sing and dance while they walk, and thro w their torches high up in the air, catching them as they fall. Arrive d at the river, they take off their garments, and, clad only in a loin • cloth, wade into the water. There is no atmosphere of ritua l or ceremony about it ; they splash and play in the water, dance, wrestle, sing, and laugh at the clumsiness of those who drop and extinguish their torches. Al l the time the torch-play goes on ; they play wit h the torches like balls, throwin g and catching them. It is a dangerous game: I saw many lucky escapes, and quite a few incidents not so lucky.
Girls and young women, too, go to the river to bathe on the navu, and the sexual attraction adds its stimulus to the general excitement. Th e navu is, in fact, the only occasion on whic h I have seen the Nup e indulge in open, undisguised, sexual play. Torches are throw n into the water to ligh t up the spot where girls are bathing almost naked; boys accost girls in the water, and couples disappear in the dark of the river- banks. Bu t the navu is in more than one respect a period of licence. Indeed, more important than this amorous play is the licence whic h was extended til l about fifteen years ago to all acts o f violence and assault that migh t occur on navu night. An d such liberty of action was not
EDUCATION FOR CITIZENSHIP 397 limited to acts of violence committed unwittingly, in the course of the
torch throwing and playful fighting. Th e torch processions invariably
led to serious organized fights between the three ena groups from the
three divisions of the town. On their way back from the river one
ena would ambush another, and try to invade its territory’. A fierce fight
would ensue, lasting the whole night, fought with torches, sticks, stones,
and even formerly (as my older informants maintained) with swords and
guns. Serious wounds and loss of eyesight were a common feature of these
‘ battles of youth’, which were fought with all the violence of heated
party feuds. For party feuds they were—inspired by the political rivalry
that is embodied in the threefold dynastic division of the kingdom. Let
us reflect that the navu provides the only occasion at which all the
scattered age-grade groups of Bida are united in common action. Th e
unity which they achieve is typical of Bida society—a unity based on
partisanship and political factions. Th e acts of violence committed in
the navu fights led the Administration to intervene, and the fights were
forbidden under Etsu Bello. Yet the memory is still vividly alive.
Younger and older men who have once participated in a fight still
proudly recount their feats which helped their side to win. But even
young boys, novices at the navu, are aware of the traditional significance
of the ceremony. At the navu of 1936 schoolboys from the Middle
School had been encouraged by their teachers to send, for the first time,
an official contingent to the navu procession and to take their place
among the age-grade groups. Th e boys did not anticipate a ‘proper’
navu; for, as they explained to the director of the school, if they were to
behave according to traditional standards, ‘the yan doka (police) would
get nasty’.
Certain important activities and obligations embodied in the age- grade association outlive the lifetime of the ena and accompany the in• dividual through his whole adult life. Th e celebration of kinship events with the co-operation of age-grade associations is a first instance. All members of an ena are required to appear at the kinship feasts celebrated in the house of one of the ena members, and this duty applies both to comrades and ex-comrades. When a man celebrates the naming cere• mony of a new-born child, or the wedding of himself or a close relative, or when he buries his father or elder brother, he will expect all his old comrades to come, to bring their congratulations or condolences, and to join the family guests. Th e Nupe greatly appreciate kinship festivities on an imposing scale, with a large—and generous—attendance. Age- grade loyalty is thus enlisted to make the festivities a success and to add to the prestige of the host. Th e readiness of the ex-members of an ena to fulfil these social obligations towards an old comrade depends, how• ever, on the rank which he had held in the ena, and on the popularity which he enjoyed among his fellow members. If the age-grade associa• tion is thus designed to contribute to the social prestige of its ex- members in adult life, this dependence on the goodwill of one’s comrades must, in anticipation, act as an important controlling influence upon behaviour in the age-grade association itself.
There exists also a certain clearly formulated code of conduct which
398 EDUCATIO N FOR CITIZENSHI P
lays down the rules of behaviour for ena members, and which is again valid for life. These rules demand friendliness towards one another, and avoidance of all quarrels; mutual assistance in work (though not with regard to loans); respect towards one’s superiors in the group; and respect also for the sanctity of marriage among all fellow members. Th e significance of this last rule, which applies to brides as well as wives, in age-grade associations the lifetime of which comprises the years when the adolescents first awake to the reality of sex, when they court and finally marry, need not be emphasized especially. Yet only in the village can the rule be regarded as tending to eliminate quarrels and feuds which must disrupt the community. In Bida it is in effect abetting matri• monial laxity outside the small group to which this code applies. Th e narrow political and class loyalty of the Bida ena has thus its counter• part in the no less narrow esprit de corps with its qualified respect for matrimonial rights.
In the case of a breach of the age-grade code the culprit is summoned before the Etsu and Ndakotsu, who have the right to punish him. A first, light offence may entail only a warning. Repeated offences will be punished more severely. If a young man shows himself lazy in egbe work or reluctant to fulfil the obligations of collective farm-work, it is regarded as sufficient punishment to order him to work in the back row, among the youngest and most inexperienced ena members. In the case of more serious offences the penalty may take the form of social ostra• cism : the Etsu and Ndakotsu would rule that no member of the ena must attend festivities in the house of the offender. This sentence, I was told, soon brings the culprit to heel; he will kneel down in front of the Etsu and ask him for forgiveness, protesting that he would henceforth mend his ways. In certain age-grade associations, especially in Bida, fines are inflicted in the case of graver offences against the ena code. Expulsion is never practised, not even as a penalty of the gravest of all age-grade offences—misconduct with the bride or wife of a comrade. However, my inquiries into the occurrence of this offence were invariably answered with a blunt denial. Whether or not it was founded on fact I have no means of deciding. Quite possibly the esprit de corps of the ena forbade admission of so shameful a charge. Th e rather too sweeping denial with which my questions were met makes it appear not unlikely: ‘We are like brothers,’ many of my informants said, ‘who would steal a brother’s wife?’
In one respect, at least, this statement is undoubtedly an exaggeration. Age-grade comrades are not ‘like brothers’. Th e intimate relation between age-grade comrades which this simile suggests is not true to facts. Th e relationship existing between friends, on the other hand, is often likened to the relationship between brothers, and does in certain respects resemble it. But age-grade comradeship as all informants agreed, is sharply distinguished from friendship, and the mutual obliga• tions of friendship in no way coincide with those implied in ena member• ship. Th e subject is worth pursuing.
A man (or woman) will never have more than one friend. Th e friend may come from a different locality, a different profession, even a differ-
EDUCATIO N FOR CITIZENSHI P 399 ent tribe. I have seen friendship between a Nupe and a Yoruba in Jebba,
and between a Kutigi farmer and a konu weaver, between a Bida tailor
and a peasant from Patigi. Only social class remains an effective barrier
even in friendship. Unlike age-grade comrades, friends will help each
other with money. Friends may also exchange their children, that is,
adopt each other’s children, and marry them to their own children—as
is also done by men who stand to each other in the relationship of
classificatory brothers. This practice has the aim, as the people say,
‘to increase friendship’. Th e essential difference between age-grade
fellowship and friendship is thus clear: if the former fashions sentiments
of ‘belonging together’ on the basis of similarities in outlook and in•
terest, i.e. similarities involved in common locality, tribal origin, com•
mon occupation, the latter unites individuals qua individuals, in a union
which is essentially new, and initiates rather than follows existing bonds
and sentiments.
It remains to discuss the ena nyentsugizi, the age-grade associations of the girls, and the relation between the parallel age-grades of the sexes. I have said already that the girl’s ena is limited to the youngest age- grade. As explanation of this different organization among the girls the people argue there can be no senior grades among girls, since the approximate age limit of the junior grade, 15 to 16 years, coincides with marriageable age. Unlike boys, the girls can no longer carry on in the ena once they are betrothed or married; their duties towards husband and family cannot be reconciled with age-grade activities. This ex• planation is borne out by the characteristic relations that unite the parallel ena of boys and girls. They both take exactly the same titles; and a boy and a girl who bear the same title enter a special, close, re• lationship : the boy calls the girl nna, mother, and the girl the boy nda, father. Girls’ and boys’ groups meet at the various feasts which age- grade associations arrange. It is the duty of the girls to prepare food and beer, the money being contributed by the boys, and to serve it at the ena feast. Th e Etsu of the boy’s ena presents the girls with kola-nuts. Th e girls will sing and dance, and the boys will watch, and then, in their turn, the boys will perform before the audience of the girls. They all behave with utmost dignity, and even severity: the ena is quite obviously not the place for flirts and amorous play of the kind one can observe between boys and girls who meet in the street or market. Th e symbol• ism of the names by which girls and boys address each other—names which are never used by lovers or by husband and wife—emphasizes the ‘respectable’, non-sexual, nature of this relationship, even to the extent of excluding any reference to legitimate sex relations— marriage. It is supposed to happen occasionally that boys and girls who were brought together in the ena marry afterwards. I have not come across an actual case. Th e relationship between them, while the ena lasts, is definitely meant to be in the nature of comradeship, and must never lead to sexual relations. Yet once the girl is married, it may be difficult to maintain this aloofness. Th e danger of jealousy on the part
of the husband, and possibly the temptation of carrying comradeship with the other sex to a more intimate relationship, forbid a continuation
400 EDUCATIO N FOR CITIZENSHI P
of the girls’ ena. So much for the interpretation of the ena nyentsugizi on the lines of the Nupe argument.
But if for women the age-grade career comes to an end with marriage, while for men marriage represents neither a limit nor even a conspicuous grade in the age-grade career, this reflects also the different sociological significance that marriage has for women and men. For women, it is the end of one life and the beginning of another. It means change of home and family bonds, and initiation into new aims and duties. For men, marriage—a first marriage—does not even coincide with the periodical readjustments in the labour-unit, i.e. the release from the efako group which, as we have seen, takes place as a rule when a man has two wives or grown-up children. Th e social career of the men—achievement of economic independence, promotion to office, and responsible position in the community—only begins to fulfil itself long after their first mar• riage ; that of the women is meant to enter its last phase in marriage. Indeed we have seen that when the woman claims the right of an in• dependent, continued, social career, she must buy it with the sacrifice of the marriage union.
As regards the institution of the parallel age-grades its practical value seems to be that it prepares the ground for the first experiences of sex relations. Or rather, it aims at circumventing, and dulling, this un• settling first experience. Enabling the sexes to meet in the critical age, between 13 and 16, as it were on neutral ground, openly and respectably, it tends to remove some of the secrecy and unhealthy curiosity that is part of the mental transition from the self-contained existence of early youth to the new awareness of the polarity of sex. Our study of adult sex morality has left no doubt about the restricted effectiveness of this native ‘ co-educational’ scheme. It must fail because it is, like all faulty education, in the nature of a palliative. It breaks the adolescent, we might say, gently, to the moral code of adulthood, but it adopts this code in all its dangerous narrowness and its rigid exclusion of amorous experiments and pre-marital sex relations. If age-grade discipline paves the way for adult sex morality, it does so also with respect to its repressions and inhibitions, and that final sweeping reaction against conventional morality which we could study in the adult life of the Nupe people.
To summarize. Th e age-grade associations include the adolescent male (and to lesser extent female) population of Nupe from early youth up to the threshold of adulthood. Th e most vital years in individual development are thus brought under the influence of this institution. We have studied the decisive part which it plays in moulding the person• ality of its members and in defining their attitude towards tasks and problems which they are to face in adult life. Th e essence of the institu• tion of age-grades is, then, preparation—a twofold preparation. Its first aspect is the fashioning in the framework of the age-grades of sentiments of solidarity which are to bind the group at large.
This involves an important theoretical problem which has greatly engaged the attention of modern sociologists and social psychologists, namely, the possibility of transferring co-operation and solidarity
EDUCATIO N FOR CITIZENSHI P 401 achieved in one context of social life to other social contexts.1 Our data allow us to make the following observation on this point : co-operation
and solidarity fashioned in the age-grades appear capable of being
transferred to other social contexts because of two conditions whic h they
satisfy. Co-operation and solidarity are not left to grow at random and,
as it were, unconsciously; rather they are of a specific, ad hoc, type ; they
clearly formulate their purpose and operate in the sphere of distinct
awareness. Moreover, age-grade co-operation is actuated in circum •
stances of psychologically impressive nature, i.e. in the context of
religious experiences. Bu t age-grade integration also admits its own
limitation in that it avoids attempting to bridge over the deepest existing
social gulfs, of local community and social class.
Th e second aspect of the preparation embodied in the age-grade system is the training of the adolescent, through discipline and certain concrete activities, in the tasks and duties whic h await hi m in adult life, and in the values and incentives whic h he wil l have to make his own. In the village this preparation represents a first step towards a final social career; in Bida it mostly implants in the min d of the adolescent values and incentives, as it were, in abstracto, withou t promising equally their fulfilment i n later life. I t teaches people who wil l never share i n direct• in g the fate of their country to appreciate, and make their own, the motives and values whic h govern the conduct of those who do. It im • parts, if not claims to responsible citizenship, yet its spirit.
THE MEANING OF NUPE LIFE
If we compare Nupe religious teaching wit h the teaching implied in the age-grade system we find that the two educational institutions sup• plement each other in many respects and that, between them, they take in the whole field of ‘education for citizenship’. Th e Malla m school offers promotion to one social plane; the age-grade association practises promotion in the framework of a complex scale of steps and grades. Th e promotion afforded by religious education is ‘real’, in the sense that it involves social privileges in adult life ; that practised in the age-grades is. fictitious, even vicarious, and concerns mental readiness rather than concrete achievements. Religious education, finally,, creates a loosely organized professional group and at the same time fosters a much wider psychological unity, whic h embraces the whole of Nupe society, across triba l and class boundaries. Th e age-grades create an intensive solidarity, but only in a very narrow framework, and give triba l and class bound• aries a wide berth. Common to both religious education and the age- grades is the social upward trend, the motive of rise and promotion. Thi s upward trend appears indeed as the paramount motive in Nupe cultural life. If it is possible to reduce a whole culture to a single formula, to a fundamental unitary ethos, we find it in this leitmotif of
1 Cf. F. C. Bartlett, in The Study of Society, 1939, p. 43 : ‘A specially interest• in g question is : if cooperation is established in one respect, when and how far does this help individua l members of the groups concerned to cooperate in other respects ?’
D d
402 EDUCATIO N FOR CITIZENSHI P
Nup e education. No aspect of culture affords closer access to the para• mount cultural motives and values. Fo r education not only teaches openly the specific aims and conceptions of life accepted by the society, bu t reveals them also indirectly, in the motives and conceptions whic h it utilizes to call fort h the desired response.
Is, then, the reverse equally true? is behaviour that does not conform to this conception of life met wit h public disapproval ? One migh t in • deed ask if Nup e culture di d not reserve some legitimate place withi n its framework for individuals to who m its dominant trend was uncongenial, and who represent what some anthropologists have called ‘deviants’ from the type of personality whic h the society regards as so highly desir• able.1 I must admit that I have found no conspicuous instances of such
‘deviant’ types. But we can guess the public response towards such social or cultural ‘failures’ from certain occasional incidents: for an attitude that belies the upward trend, and this dynamic aspect of self- advancement, is indeed regarded as a ‘failure’ ; such individuals woul d be dismissed by the people wit h a mental shrug—’u de kokari a\ ‘he makes no effort’. As the Nupe state undertakes to govern the moral life of the people and to co-ordinate rigidl y their conceptions of right and wrong , so its education attempts a complete unification of types of personality.
Yet successful suppression in normal life may only mean that the
‘deviant’ type is forced into abnormality and wil l reveal itself in psycho-
pathological cases. I have admitted in the first chapter the scantiness of
data concerning mental disorders among the Nupe. But so far as they
exist they seem to bear out the view that mental disorders in Nupe are
largely symptoms of the failure of individuals to adjust themselves to the
demands of the social environment. I have mentioned that the large
majority of psycho-pathological cases that came to my notice presented
the clinical picture of paranoid schizophrenia, wit h marked delusions
and persecution-mania. No w modern psychiatry has shown that schizo•
phrenia is essentially the result of inadequacy to a social environment.
Indeed, in the few cases whic h I could examine more closely features of
the social environment conspicuously entered the clinical picture. One
man was under the delusion that he had been persecuted by a person of
very high rank and robbed, through his intrigues, of his home and
family ; another maintained that intrigues against hi m at the court had
cost hi m his land ; a thir d case was that of an old man who was under the
delusion that his younger brother (whom he admitted to be the more
clever and successful of the two) was out to kil l him . I n creating these
very specific delusions the psycho-pathological individua l escapes fro m
the consciousness of his ow n inadequacy. Thoug h distorted and purely
negative, the fears and anxieties whic h make up the delusion are the same
fears and anxieties that accompany the successful mental adjustment to
the ethos of Nupe life—the anxiety to ‘get on’ in the world , not to be left
behind in the race for rank and position, or the fear to lose the support of
the ‘powers that be’, on whose influence one’s career and safety depend.
Th e border-line between normality and abnormality is that between
1 Ruth Benedict, Patterns of Culture, and Margaret Mead, Sex and Tempera•
ment.
EDUCATIO N FOR CITIZENSHI P 403 anxiety and unceasing efforts to master reality or, conversely, the open
admission of inability to live up to the exigent standards of the social ethos,
and that complete withdrawal from reality and surrender to subjective
emotional and thought-life which we classify as pathological.
Have we, as sociologists, to leave off here, and to regard the leitmotif of Nupe culture as an ultimate, irreducible, fact—as would be the view of certain modern schools of anthropology—or can we hope to trace it farther, and to correlate it, as we could most cultural traits, with other, more fundamental, social facts ?
Not having examined in this book the whole field of culture, we can only venture on a tentative answer. But I believe that such a correlation is possible. Th e upward trend in social life is entailed in the class structure of the Nupe state; in the existence of a social elite which must both renew and widen its ranks by admission of selected groups from outside its sphere, and build its strength on a population whose motives and ways of thought are in conformity with its own aims. Th e particular form which the upward trend takes, that complex social ladder of many steps and grades, is part of the ‘admission of selected groups’; it is ex• plained by the necessity of controlling and canalizing the upward trend, lest it should endanger the position of the ruling group itself. The ethos of Nupe society does thus not represent a unique, irrational, and, as it were, autonomous cultural trend, but the product of an interdependence of phenomena which is open to rational interpretation. Behind it we perceive the public approval of forms of behaviour which are essential to the integrity of the class-organized state.
In reducing the ethos of Nupe culture to necessities involved in the social structure we have not merely reduced one unknown quantity to another. Th e class-state biiilt upon, and expanding by, conquest is not a unique phenomenon, but a typical and regular occurrence in the history of society which we have learned to correlate with specific environmental and historical happenings. These themselves fall outside our range of vision: the environmental factors which made the river Niger the hotbed of powerful military states; environmental changes which have most probably started the tribal movements resulting eventually in conquests and the formation of states with an alien ruling class; political develop• ments in the north of Nigeria which resulted in the Fulani conquests of the central kingdoms; or the historical changes in civilization which gradually led up to greater political concentration and the evolution of centralized states. I do not mean to defend an absolute determinism of this kind. I will not deny that Nupe culture may have added specific features of its own to this general development; it may well be that certain autonomous psychological determinants, not entailed in these external conditions, have played an important part in moulding the structure of the Nupe state and the psychological traits and motives which it has made its own. I have worked out elsewhere the principle of
‘alternative solutions’ of which a cultural adjustment to environmental and historical demands may admit.1 In these alternative solutions we
1 ‘Th e Typological Approach to Culture’, Character and Personality, v, No. 4,
Th e Duke University Press, 1937.
D d 2
40 4 EDUCATIO N FO R CITIZENSHI P
discover th e effectiveness of such autonomous psychological tendencies. Th e specific ethos o f Nup e cultur e ma y indeed embod y such a n ‘alter• native solution’. Bu t the extent to whic h this is true , the extent to whic h the autonomous psychological forces limi t or qualify the effectiveness of historical an d environmenta l factors (whic h we mus t regard as sufficient to explain the evolution of the social structure as such), can only be ascertained on the basis of comparative studies of similar , class-organized, societies. Thi s is, once more , outside the scope of th e present investigation.
A P P E N D I X I BIBLIOGRAPH Y
T H E following list comprises the works o f early travellers and writers o n
Nigeri a and Nupe , arranged in chronological order, some of whose data
were invaluable for the historical descriptions and reconstructions of this
book.
IB N BATUT A (1353) ‘ describing the various countries on the Nige r (whic h he knew as the Nile) , mentions one called Youfi or Noufi, south of Moul i (Mur i ?), as ‘one of the most importan t countries of the Sudan’, and its ‘sovereign as one of the greatest kings of the area’. ‘N o whit e ma n [Arab?] enters Youfi, for the negroes there woul d kil l hi m as soon as he arrived. ‘ Thes e remarks have been interpreted by some authors as refer• rin g to Nupe , or Nouffie as it was called by certain early travellers. Bu t it is very doubtful whether this can be true . Ib n Batuta gives the same name Noufi also to a certain tribe in East Africa, some ‘six weeks’ march ‘ fro m the sea. Moreover , Leo Africanus (1520) makes no mention of Nupe .
Commande r CLAPPERTON , Second Expedition into the Interior of Africa,
1829.
RICHAR D and JOH N LANDER , Journal of an Expedition to explore the
Course and Termination of the Niger, 1832.
MACGREGO R LAIR D and A . K . R . OLDFIELD , Narrative of an Expedition
into the Interior of Africa, 1837.
Rev. J . F . SCHOE N and SAMUE L CROWTHER , Journals of an Expedition up
the Niger, 1842.
HENR Y BARTH , Travels in North and Central Africa, vols, i i and v,
1849-50 .
A . C . G . HASTINGS , The Voyage of the Day spring, 1857.
Dr . W . B . BAIKIE , R.N. , Notes on a Journey from Bida in Nupe to
Kano in Haussa, 1862. Journal Roya l Geographical Society, 1867.
A . F . MOCKLER-FERRYMAN , Up the Niger, 1892.
Canon C . G . ROBINSON , Nigeria, our latest Protectorate, 1900.
Lad y LUGARD , A Tropical Dependency, 1906.
A P P E N D I X I I
TRIBA L MARK S O F NUP E SUB-TRIBE S
BEN I sub-tribe: Traditiona l marks:
Moder n marks
(i.e. Bida marks):
BEN U sub-tribe: Traditiona l marks:
Moder n marks:
*NUP E ZAM ‘ o f Kutigi : Traditiona l marks:
Moder n marks:
KUSOP A sub-tribe: Traditiona l marks:
Moder n marks: identical wit h Bida marks. DIB O sub-tribe:
GBEDEG I sub-tribe:
4o6 TRIBAL MARKS OF NUPE SUB-TRIBES KYBDY E sub-tribe:
EBAG I sub-tribe: Traditional marks:
Modern marks: very mixed. Have adopted Bida and Kutigi marks.
BASSANG E sub-tribe: Traditional marks:
Modern marks: and also Bida marks.
KAKANDAtribe :
Modern marks: identical with Bida marks.
APPENDI X II I HISTORICA L CHART OF NUPE RULERS
A.D .
1463 ? Birth of Tsoede.
1493 Tsoede taken as slave to Iddah.
1523 Tsoede’s flight from Iddah.
1531 Tsoede established himself as king of Nupe in Nupeko.
1591 Tsoede’s death.
1591-1600 Etsu Shaba.
1600-25 Etsu Zavunla.
1625-70 Etsu Jigba.
1670-9 Etsu Mamma Wari.
1679-1700 Etsu Abdu Waliyi, foundation of the capital Jima.
1700-9 Etsu Alyu.
1710-13 Etsu Ganamace or Saci.
1713-17 Etsu Ibrahim.
1717-21 Etsu Idirisu.
1721-42 Etsu Tsado.
1742-6 Etsu Abubakari Kolo.
1746-59 Etsu Jibiri, the first Mohammedan king of Nupe. He was driven from his kingdom after thirteen years’ reign. Died in exile in Kutigi, where his grave can still be seen.
1759-67 Etsu Ma’azu’s first reign. Power of Jima at its height. He resigned the throne and left Nupe to live in Yauri.
1767-77 Etsu Majiya I.
HISTORICA L CHAR T O F NUP E RULER S 407
1777- 8 Etsu Ilyasu . He is said to have disappeared suddenly.
1778-95 Etsu Ma’azu’ s second reign, after having been requested by
the people t o retur n t o Nupe . Malla m Dend o (Manko) , the
Fulan i emissary, appeared i n Nupe .
1795 Etsu Alikol o Tankari . Reigned only for eight months, and was
deposed by the people.
1795- 6 £b u Mamma . Foundatio n o f the capital Raba. Divisio n o f
the kingdom .
1796-180 5 Etsu Jimada, reigning a t Jima. Wa r wit h Majiy a I I . Jimada was
kille d in battle at Jengi (Ragada?).
1796-181 0 Etsu Majiy a I I , reigning a t Raba.
1806 Majiya’ s wa r against the Fulani . Siege of Ilori n an d defeat of
Nup e armies.
1810? Th e Fulani , supported b y Idirisu , son o f Jimada, conquered
Rabd. Majiy a fled to Zuguma . Idiris u (Yisa) installed as
shadow-king o f Nup e i n Adam a Lelu , near Egga, b y Fulani .
1810(13) – Etsu Idiris u (Yisa), reigning at Adam a Lelu .
30
1830 Idirisu’s rebellion against the Fulani . He was killed in the
battle. Malla m Dend o recalled Majiy a fro m his exile and made hi m shadow-king o f Nup e i n Zuguma .
1830- 4 Etsu Majiya’ s second reign, in Zuguma .
1831 Peat h o f Malla m Dendo .
1836 Rebellion and death of Tsado , Majiya’ s son. Usma n Zaki , son of Malla m Dendo , was crowned first Fulan i Etsu of Nupe .
1838 First, rebellion of Masaba, Malla m Dendo’s youngest son,
against his brother. Established himself as rival kin g of
Nup e i n Lade .
1841 Masaba’s second rebellion. Usma n Zak i resigned the throne.
1847 Rebellion of the Hausa general Uma r Bahaushe against his overlords. Masaba fled to Ilorin .
1850 Fal l of Bida and defeat of Uma r Bahaushe by Fulan i armies
under Umar u Majigi . Usma n Zaki’s second reign.
APPENDIX IV
GENEALOGY OF THE FULANI EMIRS OF NUPE
APPENDI X V
NOTES ON TH E NUPE CALENDAR
T H E calendar of the Nup e peasant is an agricultural calendar. It is based on lunar months, and starts wit h the first rains, i.e. roughly in April . Fro m the beginning of the rains the people count twelve, thirteen, and sometimes fourteen months unti l the new rains break. Th e last mont h of the old year (i.e. the 13th or 14th month) is then identified wit h the first mont h of the new year. Th e year is divided into six seasons, each com• prising two months, as follows:
gbama first rains
zuzukd rainy season
malikd (cf. the Hausa marka)—heavy rains
sabakd time of growth
gbdfere cold season
banagu hot season, or small-pox season.
Frequently, however, the Nup e peasant refers to the different months of the year by their number only, speaking of ‘first month’, or ‘second month’, and so on. Th e name zuzukd is also used to denote the whole rainy season, and the whole dry season is called yikere, which means, literally, harvest.
Thi s calendar is the pre-Islamic, original, Nup e calendar, whic h is rapidly going out of use, specially in Bida and the larger, semi-Moham • medan, villages, being superseded by a Nup e adaptation of the Moham • medan calendar. Th e latter, also operating wit h twelve lunar months, may coincide wit h the Nup e peasant calendar in the years when the Moham • medan Ne w Year falls in the neighbourhood of April . But whe n the two do not coincide, and differ by a mont h or more, the time reckoning of the peasants in the semi-Mohammedan districts is throw n into serious con• fusion.1 Th e Mohammedans, incidentally, unlike the pagan peasants, use the names of the months rather than the numbers. Th e Nupe-Moham – medan names of the months are as below. It wil l be seen that the idea of the original Nup e calendar, wit h its twi n months grouped in six seasons, lias been preserved in the new nomenclature:
- Etswa navu—month of the torches (so called after the custom of throwing fire-brands during the celebration of the Mohammeda n Ne w Year) .
- Etswa bawa gani—month preceding the gani (a Nup e ceremony).
- Etswa gani—month of the gani.
- Etswa togaya—’twin month ‘ (taken over fro m Hausa).
- Etswa togaya guba—second twi n month . These two months are believed to be unlucky, and no important events, such as a wedding or naming-ceremony, wil l be celebrated during them.
- Etswa bawa azu za gbakd—month preceding the fast of the old men
(i.e. first part of Ramadan).
1 In June 1936, travelling from Kutigi to Jebba, I found that the Moham• medan ized people of Kutigi were counting the second month, while the pagan Jebba Islanders maintained that it was the fourth.
41 0 NOTE S O N TH E NUP E CALENDA R
- Etswa azu za gbako—month of the fast of the old men .
- Etswa bawa azu—month preceding the (general) fast (i.e. Ramadan) .
- Etswa azu—month of the (general) fast.
- Etswa Salagi—month of the small ‘Sallah’ (the first Mohammeda n
Id festival).
11 . Etswa bawa Salako—month preceding the great ‘ Sallah’.
12 Etswa Salako—month of the great ‘Saltan’ (the second Id festival).
INDE X
Administration, British: and Bida town, 160; economic and political development under, 352, 368; and land-tenure, 192; and Etsu Mo – hamadu, 84; and Mohammedan law,
172; officials, 391 ; and village chiefs,
64; and ward-heads, 160.
— Native: vi, 52, 55, 58, 60, 64-5 ; and bride-price, 350, 353; and cotton-growing, 318; and crown- land, 182; and expenses, 162, 164; and mixed farming, 367; and ndako gboya, 142; officials, 52, 148, 157,
163-4, 175-8, 251, 356, 373, 391 ; Public Works Department, 43, 175,
268, 304, 326; and tax-revenue, 162, and the Shaba, 93-5 ; and women ranks, 148.
Administrative Officer, 357, and the courts, 174, 176; and village chiefs,
52, 64, 176.
Adultery, 56, 152, 154-5, 168.
Agaie, ix, and Bida, 89, 115; birth rate,
6; and Emir of Gwandu, 89, 115; Emirate of, 13, 79, 159, 189; Etsu of,
115, 152, 201-2 ; Malla m Baba and,
78-9 ; Masaba and, 82 ; population,
8, 11, 13.
Age-grade associations, 16, 37, 50,
248-51 , 309, 360, 383-401. Agricultural calendar, 213-16, 225,
252, 386, 411 ; — department, 238,
249, 367.
Agriculture: organization of labour,
241-56; production, 205-40; see Crops, Farming, Farm-land, Manure.
ajele, 115-16, 119.
Alkali, as bara, 125; at Court, 92, 129,
331 ; in court, 101, 117, 126-7, 166,
170, 172, 174-7, 191, 199; heredi•
tary office, 99-102; office purchased,
98 ; his rank, 97, 124-5; Etsu Saidu and, 95; training, 171 ; and village chief, 57; see also Liman.
Ancestors, 32-3, 36, 44, 53, 66, 75,
130; see also Tsoede.
Anthropology, iii, vii, 72, 155, 335,
352, 403.
Apprenticeship, 257, 266, 272, 278,
282, 299-302, 349.
Arabic, 23, 59, 115, 132, 378-9. Armies, private, 40, 105, 123.
Army, Nupe, 77, 80, 84, 98-9, 109-11,
130.
aro, 182, 191 ; see also ‘Borrowing’ of land.
asiri, 78; see Dendo, Magic.
Atta of Idah (Eda), 75.
Baboons and monkeys, 184, 222. Badeggi, and Bida, 43, 81 ; District
Head, 95, 159, 181; labourers, 308; motor road and, 81 , 140, 178; popu• lation, 11 ; pottery, 296; and rail• way, 175, 178, 305; and tribal land,
185.
Baikie, W. B., 82.
bara, 123-7, 358-60, 371 ; and District Heads, 160; drummers, 302; and land, 204; and Shaba Usman, 120; ex-slaves, 155; soldiers, 105-7, no ,
112, 118, 196; see also Clientship.
Barber-doctors, 10, 42, 103, 298-301,
325, 349, 371-3.
Baro, 1, 43, 138, 175, 296, 334. Barrenness, see gunnu.
Basa Kom , 16.
Bassa Nge, viii, 12-14, 82-3. Bataci, 19, 22-5, 79, 109.
Bead-workers, 41 , 103, 257, 282-5,
291, 373.
‘Beating the ground’, 140, 148, 255. Bello, Etsu, 42, 84, 96, 147; and com•
munal farm-work, 199; genealogy,
410; his motor car, 139-41 ; and ndakd gboya, 90; sold titles, 98, 108. Beni, and Bida, 25, 40, 44, 78, 84, 109,
114, 116, 135, 139; blacksmiths,
259, 264-8 ; confederacy, 25-6, 69,
73, 82, 119, 121 ; country, 21-2, 37,
181, 205, 212-13; core of the Nupe kingdom, 76, 115, 159, 192, 259; and eggs, 204; emigrants from, 21 ; and private fiefs, 113-14; and the number four, 47 ; and Gbari, 21 ; land, 89, 212; Etsu Masaba and,
118-19; as a royal domain, 115-16,
119; sub-tribe, 19, 22-4, 65, 407; tribal marks, 407; ‘town-king’ of Bida, 25, 44, 111, 135, 139; and Tsoede, 74 ; villages, 21,39, 53, 116; and war, 104.
Benin, 74-5.
Benu sub-tribe, 19, 20, 23, 63, 83, 100,
108, 189, 407. Benue river, 12. Bicycles, 292, 326.
Bida, birth rate in, 6; dispensaries in,
5; Division, 5, 11, 83, 158, 170, 204,
238; dynastic divisions in (ekpa), 43,
89 , 100 , 102 , 119-21 , 126 , 160-2 ,
201, 259, 269, 278, 283, 299, 393 ; Emirate, 4, 8-13, 40, 52, 59, 92,
41 2 INDE X
142-4, 158-64, 174; European Re•
servation, 43 ; Greater, 119; history,
81-4, 113; Inne r Cit y (ba nin), 40,
119; markets, 16,41-2, 82,148,316; Middl e School, ix , xii , 3, 43, 176-7,
251, 326, 372-3, 390, 397; New , 40,
43 ; population, 10, 13, 40, 205; sex ratio in , 9, 10; Strangers’ in , see Idlemi; temperature, 3, 4; and private tunga, 36-7.
Bida town , 6, 8-10, 16, 40-3, 89, 103,
115, 160, 179, 188, 201, 204 ; Ad •
ministration, 107, 160; Council , 92,
96, 101, 108, 120, 162, 177-9. See also Craft Guilds , Markets.
Bindin g forces, 135-46, 383. Birni n Gwari , 12.
Birt h rate, 6 10, 37, 181. Birt h restriction, 6-8, 154. Blacksmiths, 10, 28, 39-43, 46-7, 54,
74,103, 240, 258-69, 291-2,318-19,
Blindness, 4, 48. Borgu , 12, 231 .
Bornu , and Benu, 20 ; in Bida, 115;
cattle traders, 201 ; settlers of Kutigi ,
3, 63, 188; natron from , 204; and
Raba, 81-2 ; and the slave trade,
104; temperature, 4; titles borrowed from , 97, 99.
bororogi, 71 , 116, 201 , 206, 390. Borrowing , 267, 340, 371-4 ; see also
dashi.
‘Borrowing ‘ of land, 182-5, 190-4,
198, 216.
Brass-workers, 10, 41-3, 74, 102-3,
269-72, 291-4, 318-19.
Bribery , 67, 90, 138, 160, 166-7, 195. Bride-price, 31 , 349-58; Administra • tio n and, 350, 353; efako heads and,
245-7, 258, 374; high , 61 , 151, 358; low , 105, 150, 152, 364; paid by family heads, 375; by father, 289 –
90 ; by king , 284; by mother, 332,
371 ; paid i n money (cowrie shells),
245, 309, 314-15; and pawning,
311-12 ; peasants and, 150, 152,
357, 362; polygamy and, 61 , 151,
153, 358; saving and workin g for,
252, 280-1 , 306, 355-6 ; Tsoede and, 74.
Bridge-building , 288, 308.
British : pre-Britis h conditions, 66,
103, 157, 171, 192, 242; and trade- routes, 40.
Brokers, 42, 102-3, 254, 280, 320, 330,
372.
buca, 184, 189, 197, 241-2, 248, 263-4,
272, 286, 300, 303, 309, 347-9.
Budgets, household, 339-43, 346-8,
355-7. 370, 373. Buka, 25.
Bunu , 40, 74, 82-3, 108, 115. Butchers, 42, 103, 201 , 257.
Calendar, see Agricultura l Calendar. Cameron, Sir D. , 66.
Canoes and canoe men , 10, 160, 229,
236; Kyedye, 23, 74, 164; Lokoja,
10, 86, 320; and taxation, 164; an d trade-routes, 319-22; Tsoede’s canoe, 73-4.
Capital, accumulation of, 369; in •
dustrial, 293; invested, 290, 293,
310-13, 358, 365-8 ; liquid , 202,
328, 332; money-, 61 , 282-4, 290,
309-11 , 328, 332; women’s, 331-2,
351.
‘Capitalists’, 59, 267-8, 284, 371. Cap-makers, 281, 388.
Caravans, 85, 92, 120, 199, 204, 330. Carpenters, 10, 103, 200, 240,257, 356. Cattle, 4, 20, 85, 201-3, 367.
Cece, 191, 198.
Cekpa, 19, 37, 119, 196.
Census (1921), 9, 10; (1930: local),
205; (1931), 4, 6, 9, 10; (1934:
local), 10, 197; Nigerian, 383.
Chad, Lake, 85, 275. Charms, 354, 378.
Chief’s cult , 53-6, 62 ; trees, 187.
Chieftainship, x, 13, 45, 50-6, 73. Childlessness, see gunnu.
tin amana, 110, 122. Circumcision, 300. tire perdue, 74, 271 .
Cis-Kaduna, 11, 181-3, 186, 190-2,
195-6, 368.
Class, social, 117, 127-35, 138, 145,
Clientship, 313, 381 ; as insurance,
371 ; and intermarriage, 136; and land-tenure, 195-7, 252; and pro • tection, 122-3, 125, 143, 162, 176-
- See also bar a. Concubinage, see Polygamy. Conformit y and uniformity , 15-17.
‘Conquered territory’ , 181-3, 190-4,
198.
‘Conservatism’, 237. Contractors, 283-5.
Co-operation, in crafts, 370; in farm • ing , 207, 238, 241-7, 254, 257, 373. See also efako.
Co-operative Societies, proposed, 369. Cotton , 2, 205, 218, 228, 238, 239, 242,
247, 251.
Council of Elders. See Village Elders. Council, Emir’ s (Etsu’s), 92, 102, 158,
177.
Councillors of the Inne r House, 99. Courts : Appeal, 170;Etsu’s, 170, 181,
188; District , 170, 172, 174; Grade
INDE X
413
A, Grade C, 170; High , 170; Native,
Doko, and
Beni confederacy, 25;
57, 170-1, 174-5, 195; Bida, 170-1 . Cowrie shells as currency, see Cur•
rency.
Craft-guilds, 102, 257-85, 382; and insurance, 370; privileges, 130, 163;
‘professional endogamy’ of, 375; and social mobility, 133-4, 382; ex• empted from tax, 119, 163 ; and vil• lage artisans, 136;in the three wards of Bida, 102. See also Guild-heads.
Crafts, individual, 286-97.
‘Crimes of the King’, 166, 172.
Crops, 2-3, 205-6, 214-15; ‘bor•
rowed’, 2; commercial, ix, 225, 247,
367; disposal of, 226-8 ; harvest,
213-14, 220-2 ; interplanting of,
210; rotation of, 184, 208-12, 237,
367; sowing and planting, 213,
216-19; storage, 226-8 ; transplant•
ing, 219-20.
Cross-cousin marriage. See Marriage. Crown land, 182, 199, 200.
Culture contact, xi, 207, 247, 362.
Currency, 310, 314-15, 350. Cushion-makers, 236.
Dabba, 143, 279, 319, 396.
dakdriy 110—11.
Dancers, 301-3, 394, 397; Etsu’s gifts to, 130; professional women, 361,
395; ritual, 16; zaworo group, 302,
395 ; see also Drummers.
Dances, at egbe, 249; village, 131,
361-3. See also Age-grades.
dashi, 371-3.
Dazi, 34-5, 55.
Death penalty, 166-7. Debts, 166.
Dendo, Mallam , 76-80, 88-9, 93, 96,
100, 109, 137, 143, 408-10.
dengi, 27, 32.
Depression, economic (1932-5), 11,
136, 164, 225.
Dibo, Beni villages in, 21 ; chieftain• ship, 51, 53; and number four, 47 ; and the ‘mythical charter’, 76 ; and ndakd gboya, 16, 24; ‘Outside People’, 115; and Sleeping Sickness Commission, 137-8; sub-tribe, 10,
19, 407; titles, 97, 100.
Diseases, 4, 5, 7, 46, 48, 402. Dispensaries, 5, 163.
District Heads, 36, 57-60, 65, 92, 137,
159-64, 174-5, 189, 251.
District Officers, II, 47, 64, 108, 140,
143, 157-64.
Diviners, 52, 73, 76-8, 94, 110-11,
204, 332, 349, 378-9.
Divorce, 152-5, 170, 174, 373. Doctors. See Barber-doctors.
chiefs of, 55-6, 59, 62, 186-7, 247; District Head, 159; farming, 184-6,
205-6, 349; and harvest rite, 53; land-tithe, 60; map of, 35; popula• tion, 11, 34, 37; rank-system, 55,
67, 186; tradition of, 44-5 ; and
tunga, 37.
Dokodza, 259, 263-9. Dokoyiringi, 99, 101, 111-13. Donkeys, 199, 204, 320. Dowry, 284, 332, 351-2.
Drummers, 301-3 ; age-grade associa•
tions and, 387, 394-5; in Bida, 42,
92-5, i n , 119; and clientship, 371 ; and the egbe, 248-50; Hausa, 46 ; in the markets, 324; on the royal estates, 199; and shela, 166; in the village, 38,45,131 ; and village ranks,
47, 49-
Drum-rhythms (take), 95, 130, 139,
144, 249, 303, 360.
Dyeing, 86, 188, 200, 235, 239, 242,
295, 331-
Dynastic power, 89, 118; ranks, 56,
93-7, 120; threefold structure, 33,
88-9, 147, 157, 160, 192, 198.
dzanka, see Tithe.
dzolo, 248, 251 .
Ebagi, 19, 109, 113.
Ebe, 13-16, 19-24, 48, 74, 82, 115,
173.
Economics, 29-34, 89-91 , 152, 180-
204, 314-77.
Eda (Idah), 25, 73~4- Edegi, see Tsoede.
Education, 176, 378-404. See also
Schools.
Edukpd, 96, 116.
efako in crafts, 257-8, 264, 272, 280-4,
289; farm, 189, 281, 370; in free professions, 300, 303; group, 281,
400; -income, 347; -land, 190, 197;
-plots, 184; -unit, 29, 32, 246, 257,
263, 267, 272, 277, 346-9, 357, 360 –
3, 367-77-
efuy 28, 35, 57, 184, 206, 296, 393; in
Bida, 120, 160; -names, 34-6, 120-
1; and tunga, 28, 36; and village elders, 45.
egba, 95, 113, 117, 167.
Egbe, 25.
egbe, 59, 248-51, 254, 301, 348,
363.
egbagiy 283.
Egga, xii, 78-82,86, 117-19,164, 320.
egikata, 123-5.
ekpa (wards), see under Bida: dynastic divisions.
elo, 23, 56.
41 4 INDE X
Embroidery, 61 , 120, 286-9, 293, 318 –
20, 379.
Emigration, 10, 11, 24, 33, 191′ 242. Emirs and Emirates, see under Agaie,
Bida, Fulani, Gwandu, Ilorin, Kontagora, Nupe, Patigi, Zaria.
emttsd, 29, 188.
ena gbarufuzi, see Age-grade associa•
tions.
ena ‘ku, 97-9 ; see War.
ena Maiyaki, 101, 113, 157.
ena manzi, see Mallams.
ena Ndeji, 97, 100, 113, 157. ena wuzi, 97; see Slaves. Enagi, 117, 279, 396.
Endogamy, class, 136, 150-1, 297,
353; among Komi groups, 296; pro•
fessional, 375.
Entrance hut, see katamba.
esozi (Royal Estates), 81 , 90, 92, 198-
200.
Etiquette, 62, 129-30,
Etsu, 39, 51 , 76. Etsu Dashi, 372. Etsu Dazi, 35, 55.
Etsu Dilaliy 102, 330.
Etsu Dogari, 92, 109, 157.
Etsu Dza, 301.
Etsu Nupe, appeals to, 117, 126, 195; and bara, 124, 197; and the Benu rulers of Kutigi, 63; and Bida Tow n Council, 162; his bodyguard, 92,
108-9, 143 ; and bribes, 67; and building land, 201 ; his council, 92,
98, 147, 158, 171, 177, 294; his Court of Law, 67, 170, 181, 188; court routine, 91-3 ; and the craft- guilds, 102, 119, 268, 273-4, 279,
282-4,294,301 ; and District Heads,
159-60; and District Officer, 158-
60, 163; and dynastic structure,
88-9 ; and experimental farms,
116, 200; his gifts, 204, 274,
292; income and expenses, 90-1 ;
European goods, 86, 175, 280, 284,
292, 327-8, 356; stores and trad•
ing companies, 228, 233, 240, 261,
265, 271, 288, 292, 318. Ewe, 75.
Exogamy, 150. Experimental farms, 200. Exports, 180, 228, 270, 320, 370.
Eyagi-Nupe, 20.
ezi, 34, 37.
Face-markings, 14, 22, 407.
Family: ceremonies, 349; extended,
27, 32, 46; group, 45, 51, 59-61 ,
182-4, 189, 241-2, 250, 374-6;
head, 28-31 , 45-7, 56-7, 60-4, 164,
169, 173, 182, 185-90, 196-7, 241 –
7, 250, 253, 281, 349, 353-4, 375-6;
land, 183, 189, 245.
Farming: ‘burning of the farms’, 215,
221, 237; mixed farming, 367-8;
‘for the mouth’, 226, 264; 277, technique, 207-8, 237-9, 367-9.
Farm-land, 182-97, 368. Farm-settlements, see tunga.
‘Father of the Town ‘ (Ndeji), 52-4, 99,
158.
Fertility, 6-9, 141, 154; cults, 238,
357-
I Feudal system, 84, no , 115-23, 134,
157, 166-8, 173, 193, 197-8.
Fiefs, 95, 99, 117-18, 124, 147-8, 157,
181, 193.
Firewood, 200, 277, 321, 326.
Fishing, 1, 22-5, 56, 60, 180, 242,
321-2.
Flute-players, see Musicians.
Food: -crops, 228-30; diet, 7, 19, 202,
204, 229, 234, 250, 342-6, 358;
-habits, 131, 358; preparation of,
229, 233-4, 253; scarcity, 11 ; stor•
age, 226.
Friday, 42, 72-3, 100, 143, 222, 242,
275-7, 280.
and kola nuts, 90, 129; and land, I Fruit-bearing trees, 2, 47, 56, 59, 60,
193-5, ’98, 201 ; and the ‘mythical charter’, 141 -2 ; and the ndako gboya,
90, 141-2; and the nko, 144; offences against, 166; and poly• gamy 107, 360; his residence,
157, 162, 268; and the royal estates, 199; salary, 158; slaves,
107; and the succession, 89, 93 –
4, 158; and war, no , 112; and witchcraft-trial, 127. See also Bello, Dendo, Maliki , Masaba, Moham – adu, Mohamadu Ndayako, Saidu, Umaru Majigi , Usman Zalu.
Etsu Nyenkpa, 52, 60. Etsu Tafie, 158-9. Eunuchs, 107.
187-91, 199, 230-7, 254.
‘Foreigners’, 39.
Fulani, conquest, n , 20-1 , 25, 35,
46-8, 70-7, 84-8, 98, 104, 135, 181,
191, 202, 334, 366, 403, 408; Emirs,
85, 87-8, 90, 115, 142; emissaries,
71 , 78 ; Mohammedanism and, 71 ,
102, 142-3, 150, 169; nobility, 100,
119, 147; pre-Fulani conditions,
85-8, 98, 107, 130, 144, 159, 167,
192, 202; rule, 11,77, 85,113, 117-
18, 143,168-9,172, 359; town-, 71 ; wars, 35, 46, 48, 334. See also bororozi.
Fulfulde, 20, 71 . Funeral rites, 19, 53.
Gana-Gana , 10 , 19. gani, 20 , 23 , 396 . Gara , 16, 40 , 73 . Gbagede , 76 , 83 , 85 .
Gbajibo , 79 , 149 , 322 . Gbak o river , xii , 181 , 211 . Gbanga , 11 , 36 , 81 , 96 , 159 , 319 . Gbara , 20 , 74-7 , 259 , 274 , 319 .
INDE X 415
Homosexuality, 152.
Horses, as booty, 99 ; breeding, 311 ;
census of, 204; Hausa traders and,
320; markets, 42, 120; in Raba, 85; as royal presents, 91, 204; as sym• bols of rank, 61 , 128, 130, 143, 202,
204, 359; Tsoede’s, 74.
Horse-trappings, 270,273.
Gbari , viii , ix ; an d cattle , 202 ; an d gunnu cult , 16 ; an d indigo , 235 ; an d Lemu , 71 ; an d Masaba , 80 , 82 ; a s
‘Outsid e People’ , 115 .
‘Gbar i Slaves’, 21-2 , 70 , 76 , 85 . Gbedegi , an d fishing, 242 ; an d Ku –
sopa, 109 ; an d Kyedye , 24 ;
language, 23 ; an d mamma cult ,
23 ; Masab a and , 82 ; pagans, 25 ;
sub-tribe , 19 ; triba l marks , 407 .
Gbira , 13 , 82-3 , 115 .
Gebba, 44 , 51 , 65 .
Gifts , 60 , 61 , 129 , 163 , 190 . See also
Kola-nuts .
gitsugi, 93-7 , 113, 147 , 158 . See also
Nobility , Roya l Nobility .
Glass-making , 1 , 43 , 102-3 , 140, 240 ,
257-8 , 274-8 , 291 , 296 . Gold-washing , 308 .
Government , British , see Administra •
tion .
Gowns , 62 , 67 , 91-2 , 116 ; embroi • dered , 61 , 120, 287-8 , 379 . See also Sandals, Turban .
Grave-digging , Maste r of, 46 .
Group-behaviour , 132 ; -conscious•
ness, 131 , 150 .
Guild-head , o f beadworkers , 284-5 ;
o f blacksmiths, 261 , 264 , 266 , 293 ;
o f brokers , 102 , 330 ; o f carpenters,
356 ; an d dashiy 301 , 371 ; o f glass-
workers , 278 ; an d th e king , 92 , 294 ;
a s rank-holders , 102 , 294 ; an d taxa •
tion , 163 ; p f traders , 92 ; o f weavers,
279 , 282 , 293 .
gunnu cult , 5, 16, 23-5 , 39 , 45 , 54, 63 ,
141 , 154, 298 , 354 , 376 , 395 .
Gwandu , 77 , 80-1 , 89 , 91 , 109 , 115 .
Hausa , ix ; customs adopte d from , 132 ;
mport s from , 2-3 ; ranks an d titles,
99 ; a s settlers, 40 , 43 ; traders , 275 ,
283-4 , 328 , 354 ; word s an d expres•
sions, 2 , 23 , 59 , 71-2 , 97 , 106 , 125 –
6 , 132 , 140-1 . 147 , 160 , 181 , 229 ,
233 , 235 , 248 , 251 , 329 , 349 , 371 ,
378 , 380 , 390 .
Hereditar y succession: Alkali , 99 ; chieftainship , 51-6 ; Mallams , 381 ; nobility , 98 ; i n priestl y ranks ,
47 , 99 , 381 ; i n lowest professional ranks , 47 ; rank-system , 55 ; titles,
101 , 108 .
Hospitality, 13, 56, 329, 358-60, 364.
Hospitals, 4, 5, 7, 43, 163. House-building, repair of, 348. Human sacrifices, 90, 232.
Hunting, 14,46-7,56, 59,180,247,341. Husband and wife. See Marriage.
Ibadan, ix, 10,11,16, 320-2, 330, 334. Idah, 73-5, 320.
Idirisu, 78-80, 83.
Ilorin, ix, and Mallam Dendo, 78-9 ; Emir of, 13, 83, 113, 164; Koran teachers from Bida in, 379; Etsu Majiya and, 78-9 ; and Masaba, 81 ; and population movement, 10-11 ; taxes, 10-11 ; trade-routes and, 319,
322.
Immigration, 34-6, 75-84. Incest, 165.
Income, of chiefs and elders, 60-2 ;
Etsu, 90—1, 142; of family head, 281 ;
group, 258, 277, 340-2; individual,
241, 245; women’s, 253. Income-tax, 164.
Indigo, 2, 188, 200, 228, 235-6. See also Dyeing.
Indirect Rule, vi, 157, 164. Inheritance, 169, 171-3, 181, 187, 245. Intermarriage, 24, 63, 71 , 136, 266,
325,. 375- Investiture, 94.
Iron-ore workers, 2, 200, 240, 257,
260-7, 291-3.
Irrigation, 3, 208-10, 213-14. Islam. See Mohammedanism.
Japanese manufactures, 273, 281. Jebba Island, ix; and age-grade associa•
tions, 384; and agriculture, 210,252,
322; and Chain of Tsoede, 49, 73;
chieftainship in, 51, 65; and fishing,
22, 24-5, 56, 321-2; Kyedye and,
24-5 ; and Nupe calendar, 411 ; nusa ranks, 48, 51 ; pottery, 296; and the railway, 175, 320; and sacri• fices, 25; trade-routes and, 319-20; wage-labour in, 251.
Jibiri, 76.
Jima, 81 ; once the capital, 36, 76, 274; its population, 11 ; present Etsu former District Head of, 159; and Umaru Majigi, 198, 200.
Jimada, 77-8.
41 6 INDE X
Kabba Province, 83, 319.
Kacha, xii ; conquest of, 79 ; Distric t
Heads, 60, 95, 159; farming in , 60,
251,373; Koran-schools in , 380,383; Kyedye and, 113-14, 121 ; markets,
320-1 , 329; marsh-land plots, 184;
population, 11 ; trade-routes and,
320; traders in , 329, 380; wage- labour in , 60, 251-2.
Kadun a river, Kyedye and, 24 ; marsh• land farms, 211, 230; and rice, 230; Tsoede and, 74. See also Cis- Kaduna, Trans-Kaduna .
Kakanda, viii ; bilingual, 13, 21 ; and the Fulan i conquest, 85-6 ; indepen• dence of, 22, 83 ; language, 12—14; Masaba and, 80, 82 ; as ‘ Outside People’, 21-2, 115; Tsoede and, 74. Kamberi , and the Kusodu, 116; as
‘Outside People’, 115; as ‘slaves’,
139; and Tsoede, 74, 76.
Kamuku , and Masaba, 80 ; as ‘slaves’,
139; and Tsoede, 74.
Kano , 85, 104, 274, 277, 318. Kataerigi, 1 , 21 , 71 .
katamba (entrance hut) , the chief’s, 62 ;
in Etsu’s palace, 93, 157; and the
‘house’, 27, 31 , 38 ; in the village,
359; as a workshop, 258, 283.
‘King-maker’ , the (Sheshi), 52-5, 57,
65, i4i » 159-
‘King’ s cult’ , 90, 141, 146.
‘King’ s Hangmen’ , 73, 87, 167.
‘King’ s peace’, 114, 122. Kingship , 83, 87-91 , 141, 144. Kinship , 27-33, 136, 169, 374-5 ; age-
rade associations and, 393; and ride-price, 352-3 ; and craft-guilds,
278-9, 330; -feasts, 356-8, 394, 397;
and labour-units, 348,374; and land,
186, 198; offences, 165, 168, 171.
kintsozi, 70, 72, 119, 189.
Kola-nuts, 231-3 ; as gifts, 49, 61 , 130,
188, 285, 309, 329, 380, 399; — fro m the king , 91 , 129, 144, 292; royal monopoly, 90, 122.
Kontagora, 6, 13, 20, 82, 109, 115,319.
Komi, see Weavers.
Koran , 142, 172-3, 286, 304, 354,
378-82.
Koran-schools, 349, 383. See also
Mallams .
Kora n teachers, 10, 41 , 305, 379—80.
See also Mallams .
Kpotu, 54, 89, 93-6, no , 116, 137. Kupa , 14, 19, 20, 23, 108-9, 113. Kusopa, ‘Forest Tenders’, 20-3, 234 –
6; Etsu Malik i and, 119; Etsu
Masaba and, 118, 122; Etsu Umar u and, 82 ; revolt against Bida, 82,
109-10; sub-tribe, 19-22 ; triba l
marks, 407; of Yeti , 82, 234. See also Kola-nuts.
Kuta, 25, 116, 159. See also Kyedye Kutigi, ix ; age-grade associations in, 384-7, 392-6; and agriculture,
205, 211, 215, 222-4, 235-6, 239,
265; blacksmiths, 264-5, 269; Bor- nu settlers, 3, 63, 188-9; and cotton- growing, 224, 239; and Konu weavers, 279, 399; its legal fran• chise, 122, 169; market, 296, 316,
321-4; mat-making, 117, 316, 325; and Nupe calendar, 411 ; popula• tion, 11, 34, 321 ; and trade-routes,
296, 319; tunga settlements, 37,
192.
Kyedye, and Etsu Abubakari, 169; canoe-men, 23, 164; dialect, 23 ; fishermen, 23; and Kacha, 113,211 ; and the ketsd ritual, 25, 202; Kuta,
25, 116, 159; Masaba and, 82, 113; and the nddduma ritual, 25; the Niger Company and, 25-6, 83, 113 ; political organization, 24—5, 68,
115-16, 121-2, 169; as a royal do•
main, 115-16; sub-tribe, 9, 22-5,
68; Tsoede and, 74; wars, 83, 109-
10 , 113 , 121 .
Labour, collective, 248-56; mainten•
ance of, 244-56, 267-9, 284, 293,
348, 358, 374; organization, 241-56; permanent, 305; temporary, 305-8; units, see efako; wage-, 60, 91 , 251 –
2, 304, 307-8.
Labozi, and kola-nuts, 90, 122, 129,
231-2. See also Kusopa.
Lade, 80, 274.
Lafiagi, 11, 13, 78, 83, 113. Lagos, 10, 16, 277.
Idlemi (strangers), 13, 16, 40, 43, 49,
119-20, 162, 188, 326.
Land, ‘borrowing’ of, 182-5, 190-4,
198, 216; family land, 182-3, 189-
90 ; hereditary, 182, 195-9; lease and purchase of, 182-3, 190, 192-5; litigation, 173, 181, 191-6; non- agricultural use of, 200-1 ; Ordi • nance Act, 182; possession of, 181 –
3; pressure, 181 ; reserve, 186, 108;
sale and purchase, 193,368; scarcity,
134, 368; tithe, 59, 60; village, 181 – “98. See also Cis-Kaduna, Trans- Kaduna.
Language, 13, 20, 23, 71 . Lapai, 1, 6, 8, 11, 13, 80, 252.
Law, 165-74; changes in, 137; ‘di •
vided’, 167-70, 173; litigation, 56 –
7, 67, 92, 105, 123, 153, 181, 191-6,
310-12, 331, 373′
Leaba, 1, 220, 255, 308, 322.
INDE X 417
Leatherwork, 20, 103, 257, 286-9, 322,
331 .
Legal franchise, 122, 169. Leisure, 246, 360.
Leisured classes, 306-7, 360. Lelu , 148-9.
Lemu , ix , 81 ; agriculture in , 212, 215,
247, 317, 367; diseases, 5; and
Gbari , 71 ; household budgets, 342,
346; and ‘Outside Peoples’, 71 .
Lile, 54-7, 186.
Liman, 54, 94, 97-102, 142-3, 381. Liv e stock, 201-4, 311, 331.
Lokoja, ix , 10, 12, 40, 84-6, 97, 334. Luxur y trades, 270, 282, 288, 319.
Magic , and agriculture, 232, 238;
blacksmith’s, 39 ; chieftainship and,
45, 53, 74 ; hunter’s, 57 ; kingship and, 88 ; kola-nut planters and, 232,
238; Malla m Dendo and, 78 ; the
Sagi and, 148; Tsoede and, 45, 73 –
4, 87 ; woodcutter’s, 239.
Maiyaki , 97-101 , 111-13 , 127-9, 156-
60, 177, 197, 389.
Maku, 93, 159, 197.
Malinowski , B. , xii , 72, 165, 338, 352. Maliki , Etsu, 78, 82, 101, 109, 113,
116, 119, 124, 108. Majigi , Etsu, 88.
Mallams, 102, 134, 360; alms to, 354,
379-81 ; as bara, 123-5, 371 ; and litigants, 172; marriages of, 102,
353, 379; and the N.A. , 176-8; Order of, 97, 99, 157; schools, 142,
378-83, 401 ; ‘tax Mallams’ , 58.
mamma cult , 23. Mankd , 78, 113, 147.
Manuring , 187, 199, 205-6, 215, 221 ,
230, 237, 367. Market-crops, 228-30.
Markets, 38, 204, 229, 254, 316-27,
331, 377- See also Bida.
Marriage, 8, 9, 149-56, 351-3, 379,
385, 398-400; husband and wife,
253-6, 331-2, 371-4 ; of slaves, 105. See also Adultery , Bride-price, Con • cubinage, Dowry , Intermarriage, Polygamy, Sexual Morality .
Marsh-land , 181, 184, 193, 205-14,
218, 222, 230.
Masaba, Etsu, 41 , 78, 80, 82, 86-9, 96,
113, 116, 118, 122, 141, 147, 169,
199, 200, 279, 282, 409-10. Masked rituals, 16, 23, 56.
Mat-making , 200, 221 , 239, 287—8; as bucd work , 242, 246, 340-1 ; a free profession, 279, 286; Kutigi , the centre of, 117, 325; in Lokoja, 10; marketing, 42, 292, 316, 331 ; pro • fits, 225, 288.
Matrilineal descent, x, 31-2, 51 , 54,
149,278. Measures, 315.
Medicine-men, 43, 131, 299, 301. See
also Barber-doctors.
Mercenary soldiers, 77, 80, 84, 98. Messengers, bara as, 123; and District
Heads, 60, 65, 160, 252, 280; and
Divisional Office, 158; and N.A. ,
10, 176-7; royal, 110, 115, 122. Metayer system, 91 , 182, 195-7. Middlemen, 288, 322, 327; women as,
253, 324, 334′
Missionaries, xii, 182, 201 , 237,
3°7-
Mobility, economic, 366-9; social,
133-5, 176-8.
Mohamadu, Etsu, 71 , 83-4, 90, 125,
196, 199, 201, 270.
Mohamadu Ndayako, Etsu, 75, 84, 88,
91-5, 102, 124, 129, 157-8, 191,
194-5, 200, 273-4, 410. Mohammedan almsgiving, 310, 354,
381 ; burials, 19, 310, 357; culture,
85, 132, 178, 286, 378-9; emissaries,
77-80 ; festivals, 59, 91 , 102, 129,
143-4, 160, 203, 266, 273, 290-2,
316, 354, 396, 412; food habits, 67,
203-4; law, 151-3, 169-74, 181 ;
and pagans, 56, 63-7, 122, 132, 142-
3, 146, 167-9, 172, 302, 339, 361,
411 ; preachers, 77-8, 84; propa•
ganda, 142, 146; proselytizing, 67,
143, 146; scholars, viii, 72, 76-8, 99,
102, 130, 361, 381, 390. See also
Alkali, Liman, Mallams. Mohammedanism, 76, 138, 142-3,
394-
Mokwa, viii, ix, xii ; agriculture in, 205,
2ii , 216, 235-7, 247; chieftainship,
50-2, 61 ; ‘chiefs’ trees’ in, 187; District Head, 65, 95, 96, 137;. founding of, 76; missionaries, xii ; population, 11, 34; trade-routes, 51,
296, 319; and Tsoede, 76, 141;
village-ranks in, 46-54, 57, 61 . Money, and bride-price, 245, 349;
-capital, 61, 282-4, 290, 309-11,
328, 332; family head and, 245; in • vested in live stock, 203-4; Nupe economy based on, 194, 309, 314;
-surplus, 347; women and, 332. Money Chief, 52, 58, 60.
Mosques, 82, 88, 92-3, 142, 201. Motor roads, 40, 58, 81 , 140, 178, 270,
319, 323.
Muregi, 24, 83, 321. Musicians, see Drummers.
‘Mythical charter’, see Tsoede. Mythology, 25, 44-5, 53, 84, 138, 141,
236.
418 INDE X
Ndgenu, 93, ’59- Ndgya, 158, 177. Natbi, 177-8.
owners, 366; and social mobility,
134; and taxation, 163, 176, 189,
357; and town, 134-7.
‘Nation’, meaning of, 145.
Native Administration, see Adminis•
tration.
navU, 397.
ndakd gboya (anti-witchcraft society),
16, 24, 36, 87, 90, 141-2, 146, 376.
See also Witchcraft.
Ndeji, 51, 57, 97-100, 110, 113, 117,
125, 129, 157-60, 177. Nigeria, Central, 82, 235.
Nigeria, Government of, 83, 158, 162. Nigeria, Northern, n, 83, 164, 403. Nigeria, Southern, 214, 235.
Nigerian Medical Service, 7.
Nimwoye, 147-8.
nko, 92, 95, 99, 144. 204. Nku , 73-4.
Nna Ku, 112.
Nobility, civil, 120-1, 133, 177, 306;
executions of, 167; military, 110-
13, 133, 177; office, 97-102; royal,
88, 93-7, 113. 147, 158-9; of wealth,
365-
Numerical system, 315.
Nupe language, see Language. Nupeko, 1, 25, 74.
nusazi, see Village Elders.
Ogudu, 32.
Oil-palm, 2, 23, 73, 131, 187, 192, 204,
234, 320. Oshogbo, 10, 322. Ostriches, 85, 202, 311.
‘Outside Peoples’, 71 , 76, 115, 119.
Pagan, as a term of contempt, 106,
131 ; ceremonies, 395; food habits,
67, 204, 361 ; law and, 167-9, 172;
see also Mohammedans.
Palm-wine, 50, 131-2,250,310, 361-2. Parochialism’, 39, 40, 159, 174.
Patigi, ix; banana plantations in, 230;
blacksmiths, 259; Emirate of, 142,
259; Etsu of, 73, 87; Etsu Malik i and, 119; and Hausa titles, 97; and ndakd gboyd, 87; population, II, 13; and rice-growing, 164.
Patronage, see bora, Clientship.
Pawns, 61 , 252, 311-13, 358, 366, 373,
376.
Payi, 80, 82.
Peasants, and bride-price, 150, 152,
357, 362; economy, 230, 308, 347,
367; and farming technique, 237; and financial depression, 136; and land, 195, 198, 368; and money,
311, 347, 349; as N.A . officials, 175; and paid labour, 307-8, 357; and sexual morality, 155, 362; as slave•
Pichi, 39, 57, 261.
Pigs, 203.
Police, 58, 91-2, 107-8, 160, 165, 251,
323, 326.
Polygamy, 9, 61 , 150-6, 352, 358-60,
364.
Population, 5-13, 34, 181, 377. Pottery, 200, 293-7, 32’1 325, 331- Prestige, 61 , 313, 357~64-
Prices, see Currency, markets. Priesthood, 47, 53, 147.
Priest-king-maker, 53. Primogeniture, 88.
Promotion, 48, 51-2, 61 , 93-9, “7 .
124, 133, 176-8, 187.
‘Propaganda’ of the kingdom, 142-6. Prostitution, 152-5, 333, 362.
Raba, sometime capital of Nupe, 1, 76,
80, 107, 113, 119; and cattle, 201 ; and craft-guilds, 119, 259, 274. 279; majiya and, 77-9 ; and Mallam Dendo, 78-9, 88, 137; and Masaba,
80-1 ; slave-market, 103-4.
Radcliffe-Brown, R. A. , xii, 31, 165,
363, 375-7.
Railways, 43, 175, 178, 305, 318-20.
Rani, 88, 137, 159.
Ranks and titles; age-grade ranks, 250,
386-9 ; guild, 282; hereditary, 47-9,
99,181 ; investiture, 94; list, 50; pro•
fessional, 47 ; of promotion type, 48,
54, 66, 88, 93-5, 133; royal, 93~5′, system, 98, 129-33, 389, 392; women’s, 147-9; see also ena, gitsuzi, sarakigi, ticizi, Village Elders, Village Ranks.
Relics, sacred, see Tsoede.
Ritual, chief’s, 53, 62; dances, 16; emblems, 139; king’s, 87, 141, 146. Rituals, annual, 142; fertility, 357; harvest, 53, 64-5 ; Mohammedan,
396; peculiar to sub-tribes, 23-4 ; village, 63, 354; see also elo, gunnu, Magic, ndakd gboyd, Sacrifices.
Riverain culture, 14; districts, 19, 25,
38-9, 121, 204; groups, 24-5. Roads, 41 , 163, 175, 318; see also
Motor roads.
Road-work, 65, 140, 248.
Royal domains, 115-16, 119, 122; estates, see esozi; income, 90—1, 142; Niger Company, 12, 26, 82-3, 86,
113; nobility, see nobility; princes,
41 , 94, 124, 129, 157-8, 162, 167,
198; “princesses, 148.
Rule, British, see Administration.
INDEX 41 9
Sacrifices, 3, 53, 62, 88, 202; annual,
39; human, 74, 90, 232.
Sagi, 147-9-
Saidu, Etsu, 84, 94, 96-8, 124, 151,
195, 197, 270, 410.
Sakpe, viii , 40, 47, 311, 317; chief of,
55, 65 ; and cotton growing , 239; Distric t Head of, 95,’5 9 ‘, and horse- breeding, 311 ; and indigo, 215, 235; and Etsu Masaba, 122, 169; as a royal domain, 122, 169; and silk- cotton trees, 236, 317; and bangles of Tsoede, 53.
Salaries, civi l and militar y officers, 99 ; Etsu’s, 91 , 158; i n Ilorin , i64J\paid by Bida N.A. , 59, 60, 91 , 148, 158,
162, 175-7; Sonya’s, 148; Village
Heads, 59, 60.
Sallah (Great S., Small S.), iv , 143,
144, 273, 274, 292, 316, 396. Salu, 279, 282.
Salutes, 62, 94, 102, 119, 188, 196. Sandals, as insignia of office, 67, 116,
354; see also gowns, turban .
sarakizi, 97-100, 107, 113, 124-6,
149-50, 158, 177, 252, 391-3 ; see also Nobility . [349, 358. Schools, 43, 65-6, 160, 163, 175-6,
Seligman, C. G. , xii , 275. Self-enhancement, 135, 139.
Sewing-machines, 271-2, 289-90, 323. Sexual morality , 151-6, 166, 168, 333,
353-6, 361-2, 376, 396-400.
Shaba, 50, 54-6, 89, 93,95-6,113, 124,
126, 147, 186, 196. Shari, 1, 13, 82, 261.
Shebe (Shebe Woro) , 37, 187. Sheep and goats, 41 , 203-4, 320. sheri’a, 166—7.
Sheshi, see ‘King-maker’ . Shonga, 13, 83, 113. Silversmiths, 41 , 257, 270-2.
Singers, professional women, 132,
302-3.
Slave raids, 85, 104, 119, 130, 142,
366; ranks, 107-8 ; trade, 75, 85, 90,
97, I03-4-
Slavery, abolition of, 91 , 104, 157, 196,
202 .
Slaves, as agricultural workers, 196-9,
252; as booty, 99, 112; as capital,
358; of chiefs, 61 ; court-, 97, 107,
117, 198 ; of craftsmen, 278-9, 293 ; descendants of, 20, 193, 196—9; Gbari , 21-2, 70, 76, 85 ; as land• owners, 197-8; markets, 103-4,
119—20; as police, 91 , 165, 323;
property of, 105; on royal estates,
90-1 , 199; as tribute , 37, 115; in tunga, 36-7; ‘ see also ena wuzi, Konu.
Sleeping Sickness Commission, 5,137. Soap-making, 227, 234.
Social advancement, 150, 176, 366;
class, 127-35; mobility , 134, 176-9,
383 ; see also Status. Sodc, 250, 389, 394. Sokoto, 77, 92, 359.
Sokyara, 99-103, 111, 126-7, 196, 199,
299, 386.
Sonya, 147-9.
Soothsayers, see Diviners .
Sorghum beer, 50, 61 , 67, 131-2, 188,
199, 206, 310, 343, 361.
State, the, 69-72, 121-2, 144-5, 165,
168-9, 377.
Status, 61-2, 95, 97-8, 102, 106, 119-
21 , 131, 148-51, 163, 177-8, 187,
198, 204, 294, 313, 329, 357-65,
388.
Sterility, 6, 8, 19, 154. Strangers, see lalemi.
Straw-hat makers, 125, 200, 257, 286 –
8, 293, 325, 340, 371,379.
Sub-tribes, 9, 10, 19-26, 68, 109, 136,
407.
Succession, see Hereditary, Matri – lineal, Promotion.
Sword , as symbol of rank, 128, 130,
359, 364.
Sword and turban , 123, 143, 354. Sword-smiths, 294.
Tada, 73-4. Tafie, 25, 116.
Tailoring , 125-6, 257, 286, 289, 318 –
19, 322, 325.
take, see Drum-rhythms .
tdlaka (talakazi), 94, 134, 150-2, 158,
176, 366, 381, 391-3-
Taxation , assessment, 164; buca wor k and, 281 ; collection, 58, 64-7, 107,
115-19, 160; defaulters, 64-5, 162,
188; Distric t Heads and, 160-4; efako heads and, 245—7, 258; ex• emptions, 102, 119-20, 163-4, 294; extortion, 160; farmers and, 368; fief-holders and, 118; figures, 10,
117; and household budgets, 347,
354, 357; illegal, 122; in Ilorin , 164;
income-tax, 164; and king’s income,
90 ; and labour-units, 280-1 , 374-6 ; land-tax, 59, 185, 189; legal fran• chise and, 169; money-tax, 59, 116—
17, 192 ;’Outsid e Peoples’and, 115; peasants and, 118-19, 164, 357; poll-tax, 164; receipts, 164-5, 1 76 ; revenue from , 60, 162-4; and royal domains, 116; wage-labour and,
252; wives and, 332, 373.
Theft , 166.
‘Threshing , Master of , 232.
42 0 INDE X
ticiBy 16, 45, 97 ; see also Village
Elders.
Tithe , 59-62, 116, 123, 125, 185,
189-200, 378. Titles , see Ranks.
Tow n and country, 131-7, 155, 354,
361 , 384, 391-5.
‘Town-king’ , 25, 44, 52-5, 60, 116,
139, 159, 192.
Town-walls , 34-6, 40, 91 , 111-12,
or ticizi), 45-51 , 56-63, 66, 118,159,
162;Head , 52, 58-60, 65-6, 162-4,
174, 187; ‘head of the women’ , 148-
9; land, 181-98; political organiza•
tio n of, 44-68, 135; priests, 47, 53-5,
232; ranks and titles, 45-57, 66, 94,
97, 148-9, 159, 186, 267, 392; sex- moralit y in , 155; and strangers, 188; and town , 155,175-6; women i n the,
147-9; see also Chief’s cult , tunga.
248.
Trade , 136-7, 329~34, 377-
Trade-routes, 40, 85, 296, 319-20; see also Caravans, European Stores, Wome n Traders.
Tradin g posts, 25, 83, 175, 182.
Trans-Kaduna , agriculture in , 205; land-tenure in , 181-6, 190, 368; population, 11, 205; revolt of, 143; and sub-tribes, 19.
Travelling , 39, 358-62.
Trees, 208, 239-40; see also fruit-bear•
in g trees.
Triba l marks, 14, 22, 407.
Tribe , 18-22; definition of, 14-17. Tribute , to Bida, 115, 143; to Gwan –
du , 89, 91 ; t o Idah , 73. Trusteeship, 189.
Tsetse fly , 4, 20, 201 .
Tsoede, ancestor-king and ‘cultur e hero’, 19, 21 , 44-5 , 66, 72, 76, 84,
87, 141, 202; bangles of, 53, 56, 74 ;
his canoe, 73-4 ; chain of, 36, 44, 49,
53, 56, 73-4, 141, 167, 172; con•
quests, 19, 74-5 ; and the crafts, 74,
274; dynasty of, 73, 142; grave of,
85 ; and gunnu, 141 ; his history, 72 –
4, 408 ; insignia, 44-5 , 66, 73-4, 87,
141-2 ; the ‘mythica l charter’, 72,
75, 87, 141-3; and ndakd gboya, 36.
tunga, 28, 34-7, 46, 119, 125-6, 184-
5, 192, 392; private, 196-8.
turare, 298, 300.
Turban , 61-2, 67, 94, 97, 116, 128,
354, 359, 364; see also Gown , San•
dals.
Uban Dawaki, 99-101 , 127.
Uma r Bahaushe, 80-2, 89, 139, 409.
Umar u Majigi , 80-2, 86-9, 94-5, 113,
116-18, 143, 147, 198, 200, 282,
409-10.
Usma n Zaki , 42, 79-82, 88-9 , 269,
273, 279, 408-10.
Village, the, 34-68 ; chief, 50-67, 118,
144, 148, 159-64, 174, 185-90, 250;
crafv^men, 267, 294 ; elders (nusazi
wad, 63, 94.
Wage-labour, 60, 91 , 200, 251-2, 280,
367, 379.
Wallace, Si r W. , 83.
War , 108-14, 143, 176; the order of
(ena ‘ku), 97 ; ranks, 99.
Wards, see under Bida : dynastic di •
visions.
Wealth, 151, 363-5, 388.
Weavers, 278-82; as a craft-guild , 103,
257, 282; and efako, 256, 280-1 ; a household budget, 355; Komi-, 20,
86, 188, 235, 279, 282, 295-7,
399; women- , 295-7, 331. Witchcraft , 126-7, 141-2, 148-9, 168,
179; see also ndako gboya.
Women , beating the floors, 140, 148,
225; professional dancers and sing•
ers, 131-2, 302-3, 361 , 395; dyers,
235, 295; fief-holders, 117,148,181 ; inheritance, 181, 331 ; as’middle • men’ , 253,324, 334; and money, 32,
253-6,332; position of, 147-56; pot•
ters, 295-7, 322; ranks and offices,
147-9; and shea-nut trees, 187, 233,
254 ; and taxation, 164; traders, 152-
4,164,201,239,254, 277,296,322-4,
330-4 ; weavers, 295-7 ; witchcraft,
148-9, 187, 200, 225-9, 233-5,
239, 252-6, 330-4, 372.
Wood-cutters, 239-40.
Wuya , 40, 81 , 126, 140, 199, 321.
Yagba, 12, 13, 74, 83, 108-16, 279,
- Yam , 3.
Yeti , ix , 23, 76, 82, 96, 234.
Yoruha , 3, 6, 10-13, 16, 20, 24, 37, 74 –
5, 80-2, 86, 98, 101, 109, 120, 162,
188, 210, 227, 235, 279, 282, 297,
307, 326, 329; see also Konti.
Zaria, 40, 115, 274, 318.
gikinta, 23.
2itako , 13, 14, 23. Zuguma, 13, 20, 79-83. Zungeru , 140.
TRIBA L MA P O F NUP E
BIDA CIT Y