horton, robin. on the rationality of conversion. part i

7/28/2019 Horton, Robin. on the Rationality of Conversion. Part I

http://slidepdf.com/reader/full/horton-robin-on-the-rationality-of-conversion-part-i 1/18

International African Institute

On the Rationality of Conversion. Part IAuthor(s): Robin HortonReviewed work(s):Source: Africa: Journal of the International African Institute, Vol. 45, No. 3 (1975), pp. 219-235Published by: Cambridge University Press on behalf of the International African Institute

Stable URL: http://www.jstor.org/stable/1159632 .

Accessed: 12/04/2012 18:16

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Cambridge University Press and International African Institute are collaborating with JSTOR to digitize,

preserve and extend access to Africa: Journal of the International African Institute.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=cup

http://www.jstor.org/action/showPublisher?publisherCode=iai

http://www.jstor.org/stable/1159632?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/stable/1159632?origin=JSTOR-pdf

http://www.jstor.org/action/showPublisher?publisherCode=iai

http://www.jstor.org/action/showPublisher?publisherCode=cup



AFRICAJOURNAL OF THE INTERNATIONAL AFRICAN INSTITUTE

JOURNAL DE L'INSTITUT INTERNATIONAL AFRICAIN

Volume45 No. 3, 1975

ON THE RATIONALITY OF CONVERSION'

PART I

ROBIN HORTON

INTRODUCTION

TN a recentarticlein this journalI sketched the outline of a theorywhich I sug-gested might helpus to make senseof a largebut puzzlingaccumulation f data

relatingto the 'conversion'of Africanpeoplesto Islam andChristianity.The theoryevokeda vigorous criticalresponse,alsoin this journal, rom the IslamistHumphreyFisher.2 n the presentpaperI shallreplyto Fisher'scriticism.In the process,I hopeto show that the theorycan be used to make sense of a rangeof data not carefullyconsidered n my previousarticle.

Beforegetting down to a detailedtreatmentof Fisher'sobjections,let me restate

brieflythe ideasthatevoked them.I start with the ideaof a 'basic'Africancosmologywhich has a two-tierstructure,

the firsttierbeing that of the lesserspiritsand the second that of the supremebeing.The lesserspiritsunderpinevents andprocesses n the microcosmof the local com-

munity and its environment,whilst the supremebeing underpinsevents and pro-cessesin the macrocosm-i.e. in the world as a whole. As the microcosmformspart

I A number of people have provided help and

inspiration during the writing of this essay. I amparticularlyindebted to Michael Crowder and Mur-

ray Last, both of whom gave me long written

critiques of an earlierdraft,and saved me from com-

mitting to print the numerous howlers it contained.

I am also indebted to JamilAbun-Nasr, A. E. Afigbo,Tunji Oloruntimehin, and John Peel, for critical

reading of the second draft and for numerous stimu-

lating discussions. None of these friends and col-

leagues, however, is in any way responsible for themain trend of my argument, or for any remainingfactualerrors.Finally, I am gratefulto Professor J. 0.

Sodipo, head of the Department of Philosophy and

Religious Studies, University of Ife, for a depart-mental grant toward the costs of publication.

2 Horton, 1971: 85-Io8; Fisher, I973: 27-40.

'Africa', theJournal f theInternationalAfricanInstitute,s publishedytheInstitute, ut

exceptwhere therwisetatedhewritersofthearticlesreresponsibleforheopinionsxpressed.Issued uarterly. InternationalfricanInstitute, 97y. All rights eserved.

Q

I

I



ON THE RATIONALITY OF CONVERSION

of the macrocosm, so the lesser spirits are thought of, either as manifestations of the

supreme being, or as entities ultimately deriving their power from him.3

Where the way of life is dominated by subsistence farming and commerce is

poorly developed, the social relations of the people of a particular area are likely to

be largely confined by the boundaries of their microcosm. They will be aware of the

wider world; but they will be aware of it as an arena which does not directly concern

them. Given the assumptions of the 'basic' cosmology about the respective spheres of

concern of lesser spirits and supreme being, it follows that this sort of situation is

likely to favour a religious life in which a great deal of attention is paid to the lesser

spirits (underpinners of the microcosm), whilst very little attention is paid to the

supreme being (underpinner of the macrocosm). However, where there is a develop-ment of factors making for wider communication (for instance, a development of

long-distance trade), the social life of those involved will no longer be so stronglyconfined

by

the boundaries of their microcosm.Many

of theirrelationships,

indeed,will cut dramaticallyacross these boundaries. In this situation, given the same 'basic'

cosmology, religious life is likely to take a somewhat different form. Less attention

will be paid to the spirits, and more to the supreme being.This scheme, I go on to argue, provides us with the basis for understanding, in

any given case, the outcome of exposure to Islam and/or Christianity. In the first

situation, where microcosmic boundaries are still strong and where attention to the

lesser spirits consequently overshadows attention paid to the supreme being, such

exposure is unlikely to be followed by any dramatic changes. In the second situation,where microcosmic boundaries are weakening, and where many people are con-

sequently gropingtoward a more elaborate definition of the

supreme beingand a

more developed cult of this being, exposure is likely to have more positive sequels.Even in this second situation, however, acceptance of Islamic or Christian patternswill be highly selective. Just what is accepted and what rejected will be largelydetermined by the structure of the 'basic' cosmology, and by the limits which this

structure imposes on the cosmology's potential for response to social change. It

is this view of the reaction to Islam and Christianity which leads me to talk of them

as catalysts for changes that were 'in the air' anyway.For the sake of brevity, I shall refer to this set of ideas as the Intellectualist Theory.

However, I hope that the more bigoted type of sociologist will note that it is also verymuch a

sociological theory.MISCELLANEOUS CRITICISMS

In what follows I shall start by dealing with those of Fisher's criticisms which

require short, sharp, and essentially destructive answers. Having cleared them out3 In my earlier paper, the idea of a 'basic' two- can religious life (Evans-Pritchard,I956; Middleton,

tiered cosmology looked all too much like some I960;Lienhardt, 196I). Later,after a period of inten-

purely hypothetical construct, fruitful perhaps for sive field-work in the eastern Niger Delta, I con-the explanation of conversion phenomena, but with cluded that the two-tiered scheme was essential to anno direct empirical grounding. In this respect my understanding of the Kalabariworld-view (Horton,presentation did not do justice to it. In fact, I first I962). Finally, as I read more widely, it became clear

played about with a scheme of this kind during an that this scheme was applicable to numerous less

attempt to make sense of some material on religious careful and systematic descriptions of other African

ideas in northern Iboland (Horton, 1956). Not long cosmologies. Far from being a hypothetical con-afterward,I found varying formulations of the same struct, then, the scheme is the product of a slow andidea in three now classic monographs on East Afri- painful process of induction.

220




of the way, I shall go on to deal with two points that require longer and more con-

structive responses.Fisher criticizes me, in the first instance, for holding that 'what may appear, for

example, as Muslim or Christian in Africa today is in reality the development of that

which was inherent in the traditional faiths'. This proposition he finds so odd that he

feels compelled to move straight on to an examination of my motives for advancingit, and to do so even before getting down to consider the question of evidence. He

diagnoses a sentimental, over-romantic attitude to the traditional cultures. He also

goes on to warn of the political implications of such romanticism, suggesting that it

encourages, amongst other things, the extinction of parliamentary democracy and

the growth of tyranny.4Now I agree with Fisher on the necessity for ideological diagnosis in the ap-

praisal of interpretations of African religious life. For this is an area of study in which

the

buzzing

of bees in scholarly bonnets can be deafening, and in which everyoneinvolved should be prepared to act as a bee-detector. I also hold that indignantdenial, however sincere, is no way to refute an ideological diagnosis: for bees in the

bonnet, like body odour, are not often perceptible to the owner. The diagnost,however, must show that his assertions are not grossly inconsistent with parts of his

victim's work other than that which happens to have attracted his attention. And

on this count Fisher's diagnosis is way off mark. For one look at other things I have

written in this area will show that, if I have any bees in my bonnet, they are much the

same anti-sentimentalist, anti-romantic, libertarian bees as Fisher seems to carryaround in his. (Though perhaps they don't buzz so loud !)5

Inany

case, what is sovery

odd about theproposition

to which Fisherobjects

?

First of all, it is surely a cardinal principle of modern historiography that one does

not treat any human group as a tabularasa automatically registering the imprint of

external cultural influences. Rather, one treats it as the locus of thought-patterns and

values that determine rather closely which of these influences will be accepted and

which rejected. To anyone who agrees with this, my proposition must seem a gooddeal less odd than it does to Fisher. Secondly, the pattern of facts that we have before

us in this instance is in itself a very peculiar one. Given the same Muslim or Christian

stimulus, some people remain unmoved whilst others respond. And amongst those

that respond, many do so by throwing up forms of religious life that bear the most

tenuous relation to the formspresented

in theoriginal

stimulus. Here, it stares one in

the face that the crucial variables are not the external influences (Islam, Christianity),but the pre-existing thought-patterns and values, and the pre-existing socio-

economic matrix. Anyone who cannot see this needs his bonnet examined !6

4 Fisher, 1973: 28-9. the maximum possible mileage out of their early5 See for instance: Horton, I967a: 284-7; i967b: training. And this has meant making maximum use

I55-87; I970: 87-I15; 1973: 249-305. of theological developments in the core areas as key6 Failure to see this point, or at least to act on it, factors in accounting for religious changes on the

is a fault of many scholars dealing with conversion African periphery. Perhaps the anthropologist, with

to Islam and/or Christianity. One reason, I would his usually considerable acquaintance with the in-

guess, is that many such scholars received their digenous religions, andhis usually poor acquaintance

original training in the theologies of the two world with core-areaIslamic or Christian theology, is, on

religions as generated in their core areas: i.e. the the same principle, the best person to redress theMiddle East, the Mediterranean, and Western balance. A hopeful sign of a fresh start is in fact to

Europe. Like the rest of us, they have tried to get be found in a volume of essays recently edited by

22I




Fisher also criticizes the assumption, implicit in my approach, that the traditional

religions had a potential for radical and enduring change before the advent of the

world religions and of literacy. In these circumstances, he says, significant changes'seem

rarelyto have

got permanentlyoff the

ground'. Indeed,far from the

responsesto Islam and Christianity having been 'in the air' anyway, it was only with the advent

of these two religions that such changes became possible at all.7 In other words, the

indigenous religions were essentially static, and it took Islam and Christianity to

infuse some degree of dynamism into the religious scene.

This last proposition is very odd indeed. What evidence can Fisher possibly have

for it? It is perfectly true that many African peoples see their cosmologies as having a

timeless, absolute status; but this is no ground at all for concluding that such cos-

mologies really arefrozen for all eternity.8 Again, it is perfectly true that, for most of

the indigenous cosmologies, we have very little direct evidence concerning their

past history.But

althoughthis means we have little direct evidence that

theyare the

products of radical and enduring change, it also means we have little direct evidence

that they have remained static over a long period. When it comes to indirect evidence,

moreover, we have more positive indications. Thus linguistic evidence indicates

pretty clearly that whole clusters of present-day societies have come into being

through the expansion, fission, and dispersal of single ancestral groups. This is

probably true, for instance, of the Mande- and Bantu-speaking peoples. Now in each

of such clusters at the present day we find a considerable variety of religious systems;and in each member of a given cluster we find the religious system closely articulated

with the social and environmental setting in which it flourishes. Given the evidence

as to origins, the religious varietyshown

byeach cluster can

onlybe

interpretedas the result of divergences from an original unity-divergences arising from

adaptations to differing social and environmental circumstances. Whatever the

dearth of direct evidence, then, indirect evidence points quite strongly to changerather than stasis as the salient feature of the pre-Islamic, pre-Christian religiousscene.

Yet another of Fisher's criticisms is to the effect that dissolution of microcosmic

boundaries, whether in the presence or in the absence of stimulus from Islam and

Christianity, often results in changes quite other than development of the concept and

cult of the supreme being. As an instance of such changes, he cites the growth of

witch-finding movements.9 To this, I would reply that nowhere in my originalpaper did I suggest that increased attention to the supreme being was likely to be

the onlyeffect of the weakening of microcosmic boundaries. There may well, for all

I know, be other effects such as an increase in witchcraft accusations and a develop-ment of witch-finding cults. What I do maintain, and I don't think Fisher has yet

produced the evidence to refute me here, is that where microcosmic boundaries are

two historians (Ranger and Kimambo, 1972). In I have tried to take account of the criticisms in the

these essays one finds a determined attempt to present paper. 7 Fisher, I973: 29.

explore the pre-existing sociocultural matrix, and to 8 For an excellent discussion of the mechanisms

ask whyexternal religious influences had the effects whereby, in pre-literatecultures, a steady current of

they did. Although the introduction to this book ideational change can coexist with a total unaware-

contains appreciative comments on the ideas about ness of such change on the part of those involved,conversion put forward in my earlier paper, it also see: Goody and Watt, I963.

contains some constructive criticisms of these ideas. 9 Fisher, I973: 29.

222



ON THE RATIONALITY OF CONVERSION 223

weakened, whatever the other consequences,increasedattention to the supremebeing is likely to be thereas well.

The lastcriticismwhich I shalltry to dispatch n a summarymannerconcernsmyderogatoryreference o

'ComparativeReligionists'.Here, I think, we are

talkingat

cross-purposes.Fisherassumesthat I am criticizing'ComparativeReligionists'for

making comparisonsbetween religions-e.g. between Islam and Christianity.o1n

fact, I have no objection to such comparisons,having made them myself in myoriginalpaperandbeing about to do so againin the presentpaper.I do havestrongobjections to the methodological and theoretical assumptionsof 'ComparativeReligionists',but these have nothing to do with their determination o be com-

parative.As to what these objectionsare, I shallsaymore at the end of this paper.So much for those of Fisher'scriticismsthat can be disposedof brieflyand de-

structively.Now for the two criticismsthat requirea more positive approach.Thefirst of these is that the Intellectualist

Theoryfails to accountfor the

wayin which

the position of the supremebeingvaries as betweenone indigenousreligious systemand another.The second is thattheTheorydoes not account or theresultsof African

exposureto Islam.In my originalpaper,which was primarily oncernedwith 'con-version' to Christianity, gave ratherinadequateattentionto these topics. In whatfollows, I hope to show that the IntellectualistTheorycanincreaseour insightintoboth.

THE SUPREME BEING IN THE INDIGENOUS RELIGIONS

Even in areasof Africawith little or no exposure o Muslim or Christiannfluence,

attentionpaid to the supreme being is by no meansuniformlyminimal.The con-ventionalstereotypeof the Africandeus tiosuss certainly o befoundin a great manypre-Islamic,pre-Christianultures.But a muchmore developedcult of the supremebeing is to be found in a great many others.Now on my premises,the deusotiosusshould be found where microcosmicboundariesare strong and confining,and themoredevelopeddoctrineand cult of the supremebeingshouldbe foundwheremanysocialrelationships gnorethese boundaries.Fisher,however,suggeststhatthis is notin fact the case.Accordingto him: 'Godappearswhen he ought to lurkin thewings,and loiters when he should be gainingin importance.',

In thepresentstate of ourknowledge,I amnot certain hatthispointof contention

can be fully resolved.There are threemajordifficulties n our way.Firstof all,manymonographs hatgive anadequateaccountof socialorganizationgive an inadequateaccount of religious life, whilst others that give a ratherfullaccount of religiontell us little about socialorganization.

Secondly,we meet gravedifficultieswhen we try to discover how farpresent-daybelief and cult relatingto the supreme being are part of the indigenous religiousheritage,and how far they are the results of Islamicand/or Christianreligiousin-fluence.Ironically,much of thetroubleherestemsfrom a factwhich lendssupport othe IntellectualistTheory: the fact that Africanpeoplescharacteristicallyee Islamicand Christiandeasof God as referring o their own supremebeing, and drawfreely

on such ideas in the process of elaboratingtheir conceptionof this being. It may

O1 Fisher, 1973: 29-30." Fisher, I973: 29-




even be true that, the more the initial attention paid to the supreme being, the more

use subsequently made of Islamic and/or Christian belief and cult, and hence the

more the difficulties for the scholar trying to get some idea of the initial situation.i2

In most cases, of course, one can in practice see certain elements that are clearly indi-

genous, and certain elements that are as clearly Islamic or Christian.But there remains

a very considerable middle ground where one cannot be at all sure.

Thirdly, the whole topic of the position of the supreme being in indigenous belief

and cult is so loaded with emotion that it is sometimes difficult to know where fact

ends and wishful thinking begins. African scholars who are committed Christians

seem particularly obsessed with the need to show that, all over the continent, there is

an indigenous concept of the supreme being which is continuous in its essentials with

the Christian concept. Those who question the wisdom of such scholars in decidingthe issue in advance of the evidence are accused of trying to belittle the indigenous

religions.

And if

theyare Westerners

theyare accused of worse !I3

With obstacles such as these in the way, it is little wonder that a thorough socio-

logical-cum-phenomenological study of the idea of a supreme being in the indigenous

12 This point was brought home to me duringfield-work in a remote village in the eastern NigerDelta. This village was remarkablefor the extent towhich it seemed to have retained the pre-Christian

cosmology and cult organization of the area. Butthere was one aspect of pre-Christianreligious life

which hadfallen into abeyance,and that was the cultof the supreme being. Indeed, the fact that such acult had ever existed might well have escapedme had

it not been for references in earlier literature andlong-disused relics in one or two houses whichelders identified as the impedimenta of the cult.Once questioned about these relics, some of the oldmen talked at length about the cult associated withthem: a cult which had been very much alive in their

youth. One of them, asked why he had dropped thiscult whilst remainingactively involved with the cultsof the various lesser spirits associated with the vil-

lage, answered very promptly: 'Nowadays we prayto the supreme being in church.'

13 This attitude comes out most clearlyin the workof Idowu, who is also one of the most forceful andinfluential

spokesmenfor

'ReligiousStudies' in

Africa. In a programmaticarticle in the first issue ofOrita(the University of Ibadan journal of ReligiousStudies), Idowu makes much of the need to avoid

preconceptions. Yet, in a field where most of theserious research,analysis, and interpretationremainsto be done, he ends up by recommending for the

guidance of all workers in the field the followingaxiom:

'Africa recognizes only one God, the Supreme,Universal God. Even though she has picturesof Himwhich are of various shades, calls Him by variousnames and approaches Him in various ways, Henevertheless remains one and the same God, the

Creator of all the end of the earth'(Idowu, I967: 12).He goes on to accuse 'foreign investigators' of

'deliberate refusal . . . to accept that Africans are

entitled to a place with the supreme God as they are'

(Idowu, I967: II).In his recent book, these attitudes are even more

obtrusive. Thus at one point he says:'The mind of the Western scholar, or investigator,

or still more to the point, theorist, has played a trick

on him: he has rejected on the ground of prejudiceand emotional resistance to truth the fact that the

same God who is Lord of the Universe is the one

whose revelation is apprehended universally andtherefore by the "primitive peoples" in their own

way, and he immediatelyfalls, consequently, into the

trap of makingGod in his own image by thinkingthat, "I look down upon these people as an offensive

scum of humanity: it follows, therefore,that the God

whom I worship or, at least, who is regarded as the

God of my glorious and incomparableculture, mustbe of the same mind as I-He can have no time for

an excrescence of their kind!"' (Idowu, I973: 6I).

Perhaps only another African scholar can effec-

tively bell a cat of this kind. Fortunately, such a

scholar has emerged, in the person of Kenyan poet,

literary critic,and

anthropologistOkot

p'Bitek.Though conceding nothing to Idowu in his scornfor the attempts of European scholars to interpretAfrican religious life, he is as withering in his scornfor their African counterparts.All readers of Idowu

should, I think, ponder the following passage in

p'Bitek's recent book on the subject:'Like Danquah, Mbiti, Idowu, Busia, Abraham,

Kenyatta, Senghor and the missionaries, modernWestern Christian anthropologists are intellectual

smugglers. They are busy introducing Greek meta-

physical conceptions into African religious thought.The African deities of the books, clothed with theattributes of the Christian God, are, in the main,

creations of the students of African religion. Theyare beyond all recognition to the ordinary Africanin the countryside' (p'Bitek, I970).

224




religious traditionsremainsa project for the future. Indeed, it would perhapsbe

surprising f such a projectever got underway.However, if the means of getting a clear-cutdecisionon the issue betweenFisher

andmyselfarejustnot to hand,therearenone the less somepointers,some straws n

the wind. Thus if we turn our attentionto the handful of culturesfor which thereare good descriptionsboth of social organizationand of religion, the indications

seem, by andlarge,to supportthe IntellectualistTheory.Let us start by saying boo to the 'God has shown himself to everyone' brigade,

and by acceptingthat as much objectivityand good faithare shown by those whodescribe certain peoples as showing nil or minimal interest in a supreme being, as bythose who describe other peoples as paying a great deal of attention to such a being.If we take this attitude and press on regardless, we find a whole spectrum of cos-

mologies, ranging from those which include almost nothing in the way of reference

to a supremebeing,

to those which are almost dominated by the

concept

and cult of

this being.A few peoples have been reported as having no concept of a supreme being. The

most circumstantial f such reportsis Wilson's on the Nyakyusaof East Africa.'4

Nyakyusa,significantly,are subsistence armerswhose social relationsareconfinedalmost entirelywithin the bounds of their local microcosm.

A whole list of peoplesarereportedas havinga religiouslife largelyconcentrated

upon the lesser spirits, with minimal developmentof the concept of a supremebeing, and only the most infrequentapproachesto him. Lodagaa,Tallensi, Tiv,Ndembu, andLugbaracome to mindin this context.15Once again,all areprimarilysubsistencefarmers,and their social relations are confinedrather

stronglybytheir

microcosmicboundaries.From the point of view of the IntellectualistTheory, so

far, so good.

Many more peoples than is commonlyrealizedoccupy a middle position; their

religious life still primarily ocused on the lesserspirits,but with a more elaborate

concept of the supremebeing, and with more frequent approaches o this being.Examplesarethe Ibo, Ijo, Fon, Ashanti, Edo, and Yorubaof West Africa,and the

I4 For a very full account of Nyakyusa religiouslife, see Monica Wilson's two classic volumes (Wil-son, 1957, 1959). For discussion of the lack of anyconcept of a supreme being in the pre-Christian

cosmology,see

especially: Wilson, 1959: I54-223.Even in these earliermonographic volumes, Wilsonis clearly impressed by the possibility of a connectionbetween the extreme physical and social isolation ofthe Nyakyusa people and the lack of any concept ofa supreme being in Nyakyusa cosmology. In her

case too, a hunch originally derived from field-workin a particular area seems to have set her thinkingabout a more general correlation between degree ofcontact with the wider world and degree of develop-ment of a concept of a supreme being. In the event,afterworking quite independently, we published our

general statements at the same time (Wilson, I971;

Horton, I97I). If there is a significant difference

between these statements, it is that Wilson's inter-pretation of the correlation is more functionalist,whilst mine is more intellectualist. Personally, I feel

that intellectualist explanation is methodologicallysounder than functionalist; but that is another story.More important, perhaps, is the inadequacy of bothour original formulations when it comes to explain-

ingthe dramatic variations in extent of

developmentof the concept of a supreme being which we find in

settings uninfluenced by the world religions. I tryto remedy this inadequacy in the present paper.

Is For Lodagaa, see: Goody, I962: 209, 37I, 373;

Goody, 1972: 14-33. For Tallensi, see: scattered

references in: Fortes, I945; Fortes, 1948; Fortes,

1959. For Tiv, see: Bohannan and Bohannan, 1953:

8I-93. For Ndembu, see: Turner, V., I953; Turner,

V., I962. The latter is a rather curious work. In itTurner tries to elevate the cult of one of the lesser

spirits to the status of a cult of the supreme being.From the data he himself supplies, however, it is

quite clear that the Ndembu supreme being-

Nzambi-is of the conventional otiose type. ForLugbara, see: Middleton, I960: 28-32, 85-I28,

230-69.

225



226 ON THE RATIONALITY OF CONVERSION

Shona of East Africa.I6 What most if not all of these peoples have in common is an

economy in which farming (and, in the case of the Ijo, fishing) within more or less

fixed territories is supplemented by considerable trade beyond the boundaries of

these territories. Here again, the situation seems to be much as the Intellectualist

Theory would predict. For, given the premises of the Theory, the balance between

relations which are confined within microcosmic boundaries (i.e. those involved in

farming) and relations which transcend these boundaries (i.e. those involved in

trade) must encourage a religious life in which a good deal of attention is paid, not

only to the lesser spirits, but also to the supreme being.One difference is worth noting within this set of examples. Amongst some peoples

such as Ibo and Ijo, the cult of the supreme being is practised with equal enthusiasm

by all adult members of society. Amongst others, such as Fon, Ashanti, Edo, Yoruba,and Shona, the cult, although it may be practised by ordinary people, attains its

greatest importanceand elaboration in the hands of the rulers.17This too is what the

Theory would lead us to expect. For in the less centralized societies such as those of

Ibo and Ijo, most adult members of the population not only farm but engage in

trade. Hence there are no great differences in degree of involvement with the wider

world, and no corresponding religious differences. In the more centralized polities,however, ordinary people do the bulk of the farming, and are more confined by the

boundaries of their local microcosms. Chiefs and royalty, on the other hand, are more

involved, directly or indirectly, with matters like long-distance trade and inter-

state diplomacy which imply relations that transcend such boundaries. Here, then,there are considerable differences in degree of involvement with the wider world,and

correspondingly largedifferences in

religiouslife.

At the far end of our spectrum, we have a set of peoples who pay considerable

I6 For Ibo, see: Meek, I937: esp. 20-87; Horton,

I956; Arinze, I970. For KalabariIjo, see: Horton,

I962; Horton, I965. For Fon, see: Herskovitz, 1938,

esp. vol. ii: Ioi-28, 289-308; Mercier, 1954: 210-34;

Argyle, 1966: I75-80. For Ashanti, see: Rattray,

1923: 94, I39-44, I47-8, 178-9, 323; Evans, 1950:

252-4. Despite Rattray'sremark about the 'remote-

ness' of the Sky God, everything else he says on this

subject makes it clearthat there was a well-developedcult of the supreme being. Evans's account lends

circumstantial supportto this view. For

Edo,see:

Bradbury, 1957: 52-60.Data on the Yoruba are copious and often con-

flicting. For some representativeviews, see: Idowu,1962; Morton-Williams, I964; Verger, I966;Morton-Williams, 1967: esp. 58-6o; Peel, 1973: esp.I6-17. Much earlier writing on Yoruba religion isthe work of Yoruba clergymen whose parti pris is

scarcely concealed. In this respect, Idowu is the

latest of a long line. However, although he greatlyweakens his own credibility by his methodologicaldogmatism, he does make a case for the relatively

great importance of the supreme being in pre-Islamic, pre-ChristianYoruba thinking. More doubt-

ful is the extent of cult. Idowu makes a case for agreat deal of informal prayer.And he also providesapparentlycircumstantialinformation on formalized

offerings (Idowu, 1962: I40-3). Peel points to testi-

mony from an early missionary visitor which seems

to confirm the latter (Peel, 1973: 17). By contrast,

Morton-Williams, writing of Oyo, talks of informal

prayer but denies formal cult (Morton-Williams,

1964: 246). On the other hand, he talks of a royalfestival for the 'Sky and spirits in the sky' (Morton-

Williams, I967: 6o). Also interesting in this context

are his scattered references to the cult of the indivi.

dual's orunor 'spirit double'. The etymology of this

word, pluswhat he

saysabout the nature of its

referent,make one wonder if one is not dealing here

with the counterpart of the Ibo ci, the Kalabari so,and the Anago se, all of which seem best regardedas

emanations of the supreme being in special relation

to the individual. Be all this as it may, perhaps it is

now too late to resolve the controversy over the

characterand extent of the indigenous Yoruba cult

of the supreme being. Like other African peoples

struggling toward the further development of this

cult, Yoruba may well have obscured its indigenouscore with a welter of imported additions, some

Islamic, some Christian.For Shona, see: Daneel, 1970.

17 On this, see passages just cited from Herskovitz,Argyle, Rattray, Evans, Bradbury, Idowu, Morton-

Williams, Daneel.




attention to the lesser spirits, but in whose religious life the supreme being none the

less looms very large indeed. Prominent amongst such peoples are the Nuer and

Dinka pastoralists of the southern Sudan, the Pygmies of the Ituri Forest, and the

Bushmen of the Kalahari Desert.

In this set, let us look first at the Nuer and the Dinka. Fisher hails them as cases

in which God has appeared when he should be lurking in the wings.i8 Thinking

casually, one might be inclined to agree with him; for here, surely, are two more lots

of subsistence farmers, whose religious life ought, according to the Theory, to give

very little place to the concept and cult of the supreme being. A closer look, however,reveals that the situation is by no means as simple as this. For Nuer and Dinka

farming is first and foremost cattle-herding; and it involves a pattern of trans-

humance seldom paralleled amongst the tillers of the soil. In the course of this

transhumance, the individual cattle-herder not only leaves the microcosm physicallyto wander in the world outside. He also forms

many relationships

which breach its

boundaries socially. Thus, for part of the year, he lives with fellow members of the

lineage-cluster which forms his primary point of reference, and which may be said

to constitute his microcosm. Yet at other times of the year he finds himself livingin other places, often with people from lineages other than his own. He may even

change his primary point of reference several times during his life.'9 Hence, over

the years, he is not only constantly breaking out of his microcosm and returning,he is also, albeit more rarely, exchanging one microcosm for another. This is the sort

of situation in which, on the basis of the Intellectualist Theory, one would expect the

cult of the supreme being to be important.With

Pygmyand Bushman hunters, we have an

emphasison the

supreme beingwhich exceeds even that found among Nuer and Dinka. But what was the case with

the cattle-herders is even more the case with the hunters. For not only do the huntingbands of these peoples rove about a good deal. Their composition is constantly

changing as individuals depart at will to join other bands or come in at will from

elsewhere.20 Here microcosmic boundaries, either territorial or social, are very weak.

So once again, this is the sort of situation in which the Intellectualist Theory would

predict a strong emphasis on the supreme being.So far, then, the evidence seems to corroborate the Intellectualist Theory. Admit-

tedly, what I have just given could be described by the unkind as an impressionistic

sight-seeing trip. Certainly,it is no substitute for an exhaustive and

statistically sophis-ticated cross-cultural survey. But, as I said earlier, the latter remains a project for the

future. Meanwhile, such cross-cultural evidence as we do have is complemented bysome impressive pieces of intra-cultural evidence.

A starting-point here is provided by Jack Goody's recent book on the religiouslife of the Lodagaa of northern Ghana.21Goody stresses the dynamic character of the

IS Fisher, 1973: 29. see: Meillassoux, I973: 187-203. For more detailed19 For accounts of Nuer and Dinka social life material on the Bushmen, see: Marshall, 1960: 325-

which bring out these features very clearly, see: 55; Marshall, 1962: 22I-52. For a summary of

Evans-Pritchard, 1940; Lienhardt, 1958: 97-I34. pioneering work on the social and religious life of

For accounts of Nuer and Dinka religion, see: the Pygmies, see: Schebesta, I940. For a more up-to-

Evans-Pritchard, 1956; Lienhardt, 196I. date study of the Pygmies of a particulararea, see:20 For a vivid portrayalof these features in the life Turnbull, I965.

of African hunting-and-gathering peoples generally, 2I Goody, 1972: esp. I4-33.




balance between lesser spirits and supreme being in Lodagaa cosmology. For most

people in this culture, the supreme being is a remote figure with whom they have no

direct dealings. But there are signs that, even in the absence of intervention from

agents

of the world

religions,

this attitude can

easily change.

As

examples

of such

signs Goody cites the cult of 'Little God' (God's manifestation on earth?) which

swept Lodagaa country in the nineteen-fifties, and the far reaching speculation about

the position of God in the cosmos which is to be found in the myth of the Bagre

society.22He suggests that such ripeness for change is a feature of many other West

African cosmologies.

Unfortunately for present purposes, Goody makes no suggestions as to the social

context of increasing Lodagaa attention to the supreme being. But one can hardly

help recalling that the period he refers to is one in which Lodagaa were being dragged

slowly but surely into the wider society of modern Ghana. In any case, the virtue of

his remarks on thistopic

is thatthey

stimulate us to look around for otherexamplesof the same process. And in such other examples as I have been able to turn up, the

link between social and ideational change is clear.

Perhaps the most impressive example of this kind is provided by the Ibo peoples of

south-eastern Nigeria. The central Ibo, though lacking in state political institutions,have long supplemented farming with a fair amount of inter-communal trade; and it

is not surprising that in their indigenous religious tradition, cults of the lesser spiritsare supplemented by a cult of the supreme being. None the less, over most of Ibo-

land, it is the lesser spirits that have taken up the greater part of people's religious

energies. About three hundred years ago, however, a strikingly different situation

beganto

developin the south-east corner of the area. The Aro, members of a

modestlysized village-group, became more and more deeply involved in the long-distancecommerce stimulated by the Atlantic slave dealers, and eventually succeeded in

establishing a trading network whose tentacles extended, not only through much of

Iboland itself, but also through parts of the neighbouring Ibibio, Ekoi, and Ijocountries. By the beginning of the nineteenth century the Aro, though they still

looked to their ancestral village-group as their home base, had established colonies

all over the area penetrated by this network. Over the years, moreover, their com-

mercial prestige took on political overtones. As agents of a lucrative trade in slaves

and other commodities, they had free passage throughout the area, even when there

were seriousquarrels

between communitiesalong

their routes. Hencepeople

who

wished to travel far from their home villages came to look upon them as the natural

providers of safe-conduct. Again, their position made them ideal arbitrators in dis-

putes between differentautonomous communities. And they grew so confident in this

22Goody hasapassageon this which indicatesaline of peoples like the Lodagaa into (and out of) the

of thought remarkablysimilar to mine. Thus he says: world religions like Islam and Christianity. The'It is this ambiguous position of God, omnipotent Lodagaa invariably identify Naangminwith the Allah

but powerless, that adds one dynamic aspect to the of the Muslims and the Jehovah of the Christians.

religion of the Lodagaa, as well as of many other There is no problem here; it is only his role that

peoples of the Western Sudan. For given this situa- differs in these religions, partly because they are

tion, there is always the possibility of drawing God literate, and are therefore more concerned withback into the human situation. This is how I inter- explanation,andpartlybecausethey are more univer-

pret the argument of part of the Black Bagre, as well salistic, and are therefore less concerned with par-as the phenomenon of the "little God" to which I ticularistic features like ancestors, totems, and localhave alreadyreferred,and the continuing movement shrines or Baals' (Goody, I972: 32).

228




role that sometimes, where their verdict was ignored, they called on the Abam, a

related group who were not averse to hiring themselves out as mercenary warriors,to come and punish the offenders by sacking their villages and seizing them as slaves.

In short, though themselves lacking developed state institutions, the Aro created

a network of communications and a sphere of relative peace and order that covered

a very large area of what is now the south-eastern corner of Nigeria.

Inseparable from these commercial and political developments were others of a

religious nature. Most notable was the development of a cult of the supreme beingfar more elaborate than anything to be found elsewhere in Iboland. The centre of

the cult was the great oracle-shrine at the home base of Aro-Chukwu. Here came

people from all over Iboland and beyond, seeking resolution of their disputes or

answers to questions about their fortunes. Here too came all those who had been

instructed by diviners to make major offerings to the supreme being. The Aro

claimed that their commercial and

political prowess

was

directly underpinnedby the

supreme being; and the mass of Ibo and Ibibio villagers came to accept this claim.

To ordinary people they were Umu Chukwu: 'Children of God'.23

The setting of these remarkabledevelopments was one from which both Muslim and

Christian influences were completely absent.24 Yet in them, we can see a beautifully

unambiguous link between the breaking down of the boundaries dividing innumer-

able local microcosms and the dramatic elaboration of the cult of the supreme being.A second example is provided by Lienhardt in his book on the religion of the

Dinka. In Dinkaland generally, as we have seen, the cult of the supreme being is

of considerable importance. From time to time, however, there arise situations in

which the cult becomes notmerely important

but central. Such situations are created

by prophetic leaders, who bring together numerous mutually antagonistic groups,and who establish peace and order over huge areas where they did not obtain before.

One of the greatest of these leaders was a man called Arianhdit, who proclaimed his

mission around i918, and who established an area of influence which stretched way

beyond the confines of Dinkaland. Although misunderstandings eventually led to

conflict between Arianhdit's followers and the colonial government, he is said to

have approved of the latter's efforts to set up better communications and a wider

sphere of orderly social life, and to have prophesied that these efforts would cul-

minate in the development of a great nation state in which all would enjoy peace.True to the

premisesof a

cosmologyin which all events that

clearlytranscend the

boundaries of local groups are thought to be the direct work of the supreme being,Dinka saw both Arianhdit and the early colonial government as manifestations of

divine activity. Arianhdit himself claimed to be possessed by the supreme being, and

the latter's cult was central in the religious life of his followers.25

23 Most of the published information about the 24 Aro can have had only the most indirect com-social and religious life of the Aro is to be found mercial links with Islamized groups to the north.scatteredthrough the pages of C. K. Meek's pioneer- Nor were they in direct contact with Europeaning book on the Ibo (Meek, 1937: 2o-I, 44-5, 48, slavers on the coast to the south. In any case, we

213-17, 224-5, 247). Over the last few years, how- have no evidence of European attempts to propagateever, intensive field-work on this group has been Christian deas on this partof the coast until half-waycarried out by Miss F. Ekejiuba and Dr. A. Afigbo, through the nineteenth century.

both of the University of Nigeria, Nsukka. When 25 On Arianhdit and his movement, see: Lien-their work is published, we should have a much hardt, I96I: 76-8I, 156-8.fuller picture of this fascinating people.




Here, we cannot be so completely sure about the absence of Muslim and Christian

influences as we could in the case of the Aro. Dinka were certainly aware of the

activities of the Mahdi.26In earlier times they must have been at least dimly aware of

the existence of the great Christian kingdom of Nubia to the north of them.27And

at about the time of Arianhdit's emergence, the first Christian missions were cominginto the area.28None the less, it would appear from Lienhardt's exposition that, bythe time of these developments, Dinka had had little or no effective contact with the

agents of either of the two world religions, and that Arianhdit's movement was

essentially indigenous in its inspiration. Once again, then, we seem to be con-

fronted, in circumstances in which the influence of the world religions can be dis-

counted, with an unambiguous link between dissolution of microcosmic boundaries

and increased attention to the supreme being.A third example is taken from Gwassa's recent account of the Maji Maji revolt.

Maji Majiwas a massive

attemptto throw off colonial rule which took

placein

German East Africa during the years I905-7. It involved over twenty different

ethnic groups and an area of over ioo,ooo square miles.

Resentment of German rule had been growing amongst the peoples of this regionsince the beginning of the century. But although they were linked by certain cultural

continuities, their political fragmentation made any concerted expression of this

resentment very difficult. Along with the political fragmentation went a corre-

sponding religious fragmentation. Each clan unit maintained a cult of its own

ancestors, whilst people drawn from a number of clans might be habituesof the cult

of a powerful nature spirit associated with some remarkable geographical feature of

their area. There was,throughout

theregion,

a more or less uniformconcept

of the

supreme being. But this being was thought of as having little direct concern with

the affairs of men, and received little ritual attention. In short, the region was one

of small, fairly strongly bounded microcosms, with a religious life correspondinglyfocused on the lesser spirits rather than on the supreme being.

In this situation, there emerged a prophetic leader called Kinjikitile. Although he

was a recent immigrant to the village where he lived, he seems to have hailed from

somewhere within the region, and to have had a wide knowledge of its various cults.

Kinjikitile's ministry began with possession by one of the powerful nature spiritswhich, though locally based, none the less appealed to people of several different

clans. And it was with the voice of thisspirit

that hebegan

to talk topeople

about

uniting to make war upon the Germans. Later, however, he was commanded and

possessed by the supreme being himself. In this capacity he announced that the

supreme being was going to bring back to life the ancestors of all Africans, who

would then fight as one to drive the Germans out.

Before long, delegations began to come in to Kinjikitile's village from all over

the area. They were sent away with instructions about the uprising, with religious

26 Lienhardt, I96I: 72-4, 164-5. desperate measures.27 Wilson, I971: 6. Not seeing how prominence of 28 Lienhardtgives no information on the establish-

the supreme being in pre-Islamic, pre-Christian ment or impactof the missions. According to Evans-

settings can be explained in terms of contemporary Pritchard,however (Evans-Pritchard,I956: 48-9 n.),

social organization, she is forced into ad hoc nvoca- the firstmission in the areawas the AmericanPresby-tions of Islamic and Christian influence. As I suggest terian, founded on the Sobat River in 913.

here, however, there is really no need for such

230




advisers given them by the prophet, and with holy water to neutralize German

bullets. In the event, the leaders of the various war-parties lost their patience and rose

before Kinjikitile had given his command. The prophet, caught off his guard, was

capturedand

hanged by

the Germans at the very beginning of the rising. None the

less, the peoples of the religion, fired by his ideas, succeeded in waging a concerted

war for nearly two more years.29

Kinjikitile brought people to see that the Germans were a scourge that transcended

the innumerable inter-communal boundaries, and that, as such, they could only be

fought by a mass movement whose members were willing to forget about these

boundaries. It is in this context that his religious innovations are best understood.

Thus on the one hand, there was his promise that the ancestors would return and

join the fight en masse.This was not just a promise of added strength in the trials

to come. It was a promise to put into reverse the process that had created the an-

cestors in the first

place;

a

promise,

in other words, to

liquidate

the ancestors and

their cult. Such a promise was nothing less than a commitment to remove the

spiritual underpinnings of the clan microcosms. On the other hand, there was his

emphasis on direct inspiration by the supreme being. And with this, of course, he

provided a spiritual underpinning for his followers' new involvement in the wider

world.

This is another example in which we cannot entirely rule out the possibility of

Muslim or Christian influence. Gwassa, however, argues convincingly that Kin-

jikitile's ideas were a straightforward development of the indigenous cosmology of

the region, and that his rituals were built up out of an indigenous symbolic vocabu-

lary.Once more, then, it seems

possibleto discount the influence of the world

religions; and once more, we are faced with an unambiguous link between micro-

cosmic dissolution and a switch of attention from lesser spirits to supreme being.

My last example is drawn from Middleton's description of Yakan-a cult which

flourished among the Lugbara of East Africa between i890 and I920.30In the nineteenth century Lugbara were subsistence farmers, living in a large

number of tiny, lineage-based communities under the authority of elders. The

members of each community guarded its autonomy jealously; and their social rela-

tions were largely confined by its boundaries. They regarded the community as an

oasis of morality, order, and stability in a desert of sub-human chaos. They saw all

changeas

comingfrom the world outside, and saw it as bad

bydefinition.

Lugbara had evolved a subtle cosmological scheme which made sense of their

predicament and gave them some feeling of control over it. Two categories of lesser

spirits guided the fortunes of each little community: ancestors and adro.Ancestors

were the guarantors of communal prosperity, order, and authority. Adro were

associated with the uncultivated bush around the area of settlement, and hence with

natural phenomena in the strict sense. They were also associated with the vicious,

self-seeking and innovative side of men; with their untamed or 'natural' side, as

Middleton puts it. Thus whilst ancestors were forces of culture, adrowere forces of

nature in the widest sense. Again, whilst ancestors were forces of stability, adrowere

forces ofchange.

Hence adro,though they

werepart

of thespiritual complement

of

the microcosm, were also a constant threat to its integrity. They were, moreover, a

29 Gwassa, 1972: 202-17. 30 Middleton, 1960: 258-70; Middleton, 1963: 8o-xo8.




link with the chaotic and evil world outside, whose denizens were seen as controlled

entirely by spirits of a similar kind. Adro, then, were a sort of spiritual fifth column;the means whereby the chaos of the world outside could at any time erupt into the

microcosm.

'Behind' both ancestors and adrostood Adroa, the supreme being, who had created

the entire world, and who both sustained it and initiated all change. In some con-

texts, Adroa was associated equally with ancestors and adro; for he was the creator

and sustainer of both. In other contexts, however, he was associated with adro

rather than with the ancestors. For although he was the ultimate originator of all

change in the world, the adrospirits were the proximate means whereby such changewas brought into the microcosm. Hence in any situation of change, Adroa and adro

were thought to be closely linked.

Lugbara religious life was dominated by rituals performed for the two categoriesof lesser spirits. Direct approach to the supreme being was relatively rare.

This was the setting in which Yakan began to develop. Of its early vicissitudes,from about 890 to 1914, we know relatively little, except that it seems to have been

a movement connected with a series of social dislocations and natural disasters. Our

detailed knowledge dates from I914, when, after a period of quiescence, it began to

flourish mightily not only in Lugbaraland but in many neighbouring areas. The

causes of this resurgence seem to have been, on the one hand the beginning of serious

colonial administration, with all its interference in the life of the little communities,

and, on the other hand, a series of devastating epidemics affecting both men and

cattle.3I

Theresurgence

of Yakan in

Lugbaralandwas led

by

a

prophetcalled Rembe,

who hailed from the neighbouring Kakwa country. Like Kinjikitile among the Bantu

to the south-east, Rembe convinced a great many Lugbara that both the Europeansand the epidemics were scourges which transcended the boundaries of the numerous

little village communities, and that, as such, they could only be warded off by a

response which ignored these boundaries. Appointing numerous deputies, he set

up an organization covering a large area of which Lugbaraland was only a part; an

organization which totally ignored both existing authorities and existing inter-

communal boundaries.

Rembe was seen, in the first instance, as inspired by Dede, an adrospirit. From the

poolof this

spirit,he and his

deputies dispenseda water which was believed to be

proof both against European bullets and against the various epidemics. But Rembe

was also thought to be directly inspired by the supreme being; and it was with the

aid of this being that he proposed to bring the ancestors to life again to join the

struggle.Rembe and his followers were undoubtedly preparing for concerted resistance to

31 An association between acceleratedsocial change here another way in which weakening of micro-and major epidemics is suggested by many accounts cosmic boundaries can bring about increased atten-of this period in African history. Can it be that a tion to the supreme being. For epidemics obviouslyrapid increase in movements across microcosmic transcending microcosmic boundaries are, in theboundaries places the population of any given area traditional world-views, characteristicallyattributed

in contact with a series of 'new' disease variants for to the supreme being; and the larger they loom onwhich it has not had the necessary time to acquire a the scene, the more significant will a people's deal-

degree of resistance? If this is so, then we may have ings with this being become.

232



234 ON THE RATIONALITY OF CONVERSION

themselves made of it. And what they made of it, as Middleton shows very clearly,can

only be understood in terms of the premises of their own pre-existing cosmology.Here, then, we have a movement for which Islam clearly provided the trigger;

but a movement which the most ardent of Islamists would be reluctant to describe in

terms of 'conversion'. Once again, the nub of the matter seems to be that, given the

structure of their own indigenous cosmology, people found the most reasonable

adjustment to the weakening of microcosmic boundaries to be one which included

(admittedly this time amongst other things) an increased emphasis on the presenceand cult of the supreme being.

These four intra-cultural examples not only provide us with unambiguous in-

stances of the link between weakening of microcosmic boundaries and increased

attention to the supreme being, they also show us that this link derives from rational

adaptation

of

existing

beliefs to a

changed

social situation. Hence they lend addi-

tional significance to the general correlation suggested by my rather impressionisticcross-cultural 'tour'. At the same time, the cross-cultural evidence points to the

probability that the kind of adaptive change which features in my four examples is

one that has occurred very widely in the past. Between them, then, the two kinds of

evidence cited suggest that the Intellectualist Theory is a rather good guide to the

understanding of religious variation and change in settings where the influence of

Islam and Christianity can be discounted.

At this point, perhaps, it would be wise to sound a note of caution. For the In-

tellectualist Theory deals with only one factor in the field of African religious change;albeit a most

importantfactor. And where

manyfactors are at work, as

theyundoubtedly are in this field, some of them will almost certainly operate in such a

way as to undo the effects of others. Hence, even if the Theory is correct so far as

the factor it deals with is concerned, one cannot expect that things will alwaysgo the

way it says they should. One can only mutter 'ceteris paribus' and keep one's fingerscrossed. This, however, is true of any fruitful historical generalization. So, given the

caveat, I believe we are none the less justified in regarding the Theory as a source of

genuine insight into the dynamics of the indigenous religions.This conclusion has crucial implications for the validity of my interpretation of

'conversion' to Islam and Christianity. For one of the salient features of this inter-

pretationis the

propositionthat African

responsesto the world

religionsare

responsesthat were 'in the air' anyway. In other words, they are responses which, given the

appropriate economic and social background conditions, would most likely have

occurred in some recognizable form even in the absence of the world religions. And

the only way of validating this proposition is to do what I have done here: i.e. show

that, given a marked weakening of microcosmic boundaries, some of the principal

developments associated with 'conversion' commonly did take place even in the

absence of significant influence from either Islam or Christianity.In this context, what I have said so far represents a clearing of the ground; a

necessary preliminary to dealing with Fisher's main criticism of the Intellectualist

Theory,which is that it fails to make sense of the African

responseto Islam. Rebuttal

of this criticism will be my principal task in the second part of this paper.




REFERENCES

ARGYLE, J. I966. The Fon of Dahomey,Oxford.

ARINZE, F. I970. Sacrificen Ibo Religion, badan.

BOHANNAN, L. and P. I953. The Tiv of CentralNigeria, London.

BRADBURY,R.

1957.TheBenin

KingdomEthnographic Surveyof

Africa,W.

Africa,Pt.

XIII),London.

DANEEL,M. I970. The Godof theMatopoHills, The Hague.EVANS, . I950. 'The Akan Doctrine of God' in Smith, E. (ed.), African Ideasof God,London.

EVANS-PRITCHARD,. 1940. TheNuer, Oxford.- 1956. Nuer Religion,Oxford.

FISHER, H. I973. 'Conversion Reconsidered: Some Historical Aspects of Religious Conversion in Black

Africa', Africa, xliii. I.

FORTES,M. 1945. Dynamicsof Clanshipamong heTallensi,London.- 948. The Webof Kinshipamong he Tallensi,London.

1959. Oedipus ndJob, Cambridge.GOODY, J. I962. Death, PropertyandtheAncestors,London.-- I972. TheMythof theBagre,Oxford.

GOODY, J., and WArr, I. I963. 'The Consequences of Literacy', Comp.Studies n Society ndHistory,v, No. 3.

GWASSA, G. 1972. 'Kinjikitile and the Ideology of Maji-Maji'in Ranger, T., and Kimambo, I. (eds.), The

HistoricalStudyof African Religion,Berkeley.HERSKOVITZ,. I938. Dahomey,2 vols., New York.

HORTON, . I956. 'God, man, and the land in a northern Ibo village group', Africa, xxvi. 17-28.1962. 'The KalabariWorld-View: an Outline and an Interpretation', Africa, xxxii. 3.

I965. KalabariSculpture,Lagos.

1967a. 'Philosophy' in Crowder, M. (ed.), Africa andthe WiderWorld,London.- 967b. 'African Traditional Thought and Western Science', Africa, xxxvii. x-2.

1970. 'The Romantic Illusion: Roger Bastide on Africa and the West', Odu,N.S.,no. 3.I971. 'African Conversion', Africa, xli. 2.

I973. 'Levy-Bruhl, Durkheim and the Scientific Revolution' in Finnegan, R., and Horton, R. (eds.),Modesof Thought,London.

IDOWU, . I962. Olodumare:God n YorubaBelief,London.

x967. 'The Study of Religion', Orita, i. I.

- 1973. African TraditionalReligion,London.LIENHARDT, G. 1958. 'The Western Dinka' in Middleton, J., and Tait, D. (eds.), TribesWithoutRulers,

London.

-- 96I. DivinityandExperience: heReligionof theDinka, Oxford.

MARSHALL,. I960. 'IKung Bushman Bands', Africa, xxx. 4.-- 962. ' IKung Bushman Religious Beliefs', Africa, xxxii. 3.MEEK, C. K. 1937. Law andAuthority n a NigerianTribe,Oxford.

MEILLASSOUX, C. 1973. 'On the Mode of Production of the Hunting Band' in Alexandre, P. (ed.), French

Perspectivesn African Studies,London.

MERCIER, P. 1954. 'The Fon of Dahomey' in Forde, D. (ed.), African Worlds,London.

MIDDLETON,J. 1960. LugbaraReligion,Oxford.

-- 963. 'The Yakan or Allah Water Cult among the Lugbara',J. Roy. Anthrop.Inst., lciii, Pt. II.

MORTON-WILLIAMS,. I964. 'An Outline of the Cosmology and Cult Organization of the Oyo Yoruba',

Africa, xxxiv. 3.967. 'The Yoruba Kingdom of Oyo' in Forde, D., and Kaberry, P. (eds.), WestAfrican Kingdomsn

the NinteeenthCentury,London.

P'BITEK, 0. I970. African Religionsn Western cholarship,Nairobi.

PEEL, J. I973. 'Explaining Religious Change in Two African Societies', Unpublished Paper read to the

University Association for the Sociology of Religion, London.

RANGER, ., and KIMAMBO, I. 1972. TheHistoricalStudyof African Religions,Berkeley.RATTRAY,R. I923. Ashanti, Oxford.

SCHEBESTA,P. I940. Les Pygmees,Paris.

TURNBULL,. 1965. WaywardServants,ondon.

TURNER,V. I953. LundaRites and CeremoniesRhodes-Livingstone Papers No. Io), Livingstone.- 162. Chihamba:he WhiteSpirit (Rhodes-Livingstone Papers No. 33), Manchester.

VERGER,P. I966. 'The Yoruba High God', Odu,ii. 2.

WILSON,M. 1957. Ritualsof Kinshipamong heNyakyusa,London.

I959. CommunalRitualsof theNyakyusa,London.

I97I. Religionand the Transformationf Society,Cambridge.

R

235

horton, robin. on the rationality of conversion. part i

Documents