Divination and its Social Contexts Author(s): George K.

Park Reviewed work(s): Source: The Journal of the Royal Anthropological Institute of Great Britain and Ireland, Vol. 93, No. 2 (Jul. - Dec., 1963), pp. 195-209 Published by: Royal Anthropological Institute of Great Britain and Ireland Stable URL: http://www.jstor.org/stable/2844242 . Accessed: 25/05/2012 10:13
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 support@jstor.org.

Royal Anthropological Institute of Great Britain and Ireland is collaborating with JSTOR to digitize, preserve and extend access to The Journal of the Royal Anthropological Institute of Great Britain and Ireland.

http://www.jstor.org

Divinationand its Social Contexts

GEORGE

K. PARK

IN A GOOD MANY SOCIETIES STUDIED BY ANTHROPOLOGISTS divination holds, or has held, a fairlyexalted place; yet the argumentis not oftenput forwardthat the workingof any particular social systemhas hinged in any critical way upon the performancesof its diviners.In a general way, divinersare to be classed with the native herbalistand the shaman as private practitioners an art to which natural science lends littlesupport; of claims to and when it is once assumed that the 'doctor' does not do what he manifestly do-that a diviner does not, in fact,divine-reason would seem to suggestthat on the whole he is likelyto do as much harm as good. In particularwould that reasoningseem to apply forthe many societiesin which the divineris most conspicuouslyemployed in the findingof witches,and where a normal consequence of his fallibleaccusations may be the destruction innocentpersons. of There will be occasion to examine a few such cases in later sections of this paper, where I shall argue that theircruelleraspects tell us less about the nature of divination than about the general problem of order in human societies. But first should like to I present, in the light of some less dramatic facts, an argument for taking divination of seriouslyas a characteristicsocial institution which our understandingneeds to be improved. From the studyof divining,I thinkwe may properlyexpect to gain some inalso exhibiting sightinto the workingof other,and perhaps more reputable, institutions the phenomenon which I shall call 'procedural intervention'.It will be a thesisof this paper that developed systems divinationshould not be regarded as mere excrescences of upon the body politic, doing none ofits work; but that the diviner,with all his peculiar skills and his characteristicparaphernalia, does in a controlledway intervenein and affectthe social process with ratherdefiniteand socially usefulresults.I do not suggest that the consequences of any given act of divination are more likelyto be just than unjust; nor shall I claim to have become convinced,by an examinationofcases, thatwitchdoctorsusually help to rid theirsocietiesof disagreeable deviants.But I shall argue that, quite apart fromsuch considerationsas these,divinationhas as its regular consequence the eliminationofan importantsource ofdisorderin social relationships.
I

Divination is always, I think,associated with a situationwhich, fromthe point of view ofthe clientor instigator, seems to call fordecision upon some plan ofaction whichis not is easily taken. Even the urban addict to fortune-telling probably no exception to that rule; though the point need not be laboured. Typically, divinationis called forin cases of illness and death, and in other life-crises; the corroborationof a marriage-choice in and in individual or collective moves involving some change in social alignmentsor, perhaps, economic condition; and in situationsof loss, calamity,or unresolvedconflict, whetheron a personal or a much larger scale. For each societyin which divination is practisedthereis, to be sure, a proper listofits occasions; and such a listmay say much I95

i96

GEORGE

K. PARK

about the sources of strain in that society-the Chinese diviner in Singapore (Elliott I955, pp. I59 seq.) does not receive the same patternof cases as does the Zulu; nor has among European contact failed to change (by enlargement)the scope of the institution the Plateau Tonga (Colson I958, pp. 79-80). Divination is associated with the sense of danger, and oftenseems to relieveit. But thosewho have regarded the divineronly as a sort of primitivetherapist,relievinghis client of doubt and indecision, or explaining away such phenomena as convulsions,may surelybe accused of understatingthe case. For it is generally true that where convention calls for divination,its omission might have genuinely difficult consequences for the person or group who acted without it. Divination normallyprovides more than the 'psychological release that comes fromthe conviction that subsequent action is in tune with the wishes of supernatural forces' (HerskovitsI 938, II, p. 2 I 7); the associationofdivinationwithsituationsofproblematical action is best explained, afterall, by the factthat it lends to a client'ssubsequent act a peculiar but effective typeoflegitimation. of What I propose is, put mostsimply,that a 'sociological' interpretation divination than the 'psychological' analysis will be found more general, and more satisfactory, which is so much more readily suggestedby the usual circumstancesof actual observation in the field. The ethnographerto-day seldom observes divination in conjunction with what Durkheim called 'collective representations';far more often it is a mere matterof individual purpose-irrelevant, almost, to anythingso grand as the collective welfare.A Yoruba employsIfa divination,forexample, in the selectionof a house site. Stripped of its social context,and taken only fromthe immediate point of view of the actor, divination for such a purpose would seem to have doubtfulvalue. On the one hesitation and hand, as Bascom (I94I, p. 45) observes, 'the elimination of fruitless would seem to enable the individual to concentratehis entireenergy,without indecision distraction,upon the task at hand'. But against this one must weigh a heavy material cost and all the added 'distractions'which, among the Yoruba, that is likelyto entail. readilyapparent Yet the choice of a house site by a Yoruba has, in fact,a special gravity from the point of view of social-structuralanalysis, although perhaps unlikely to be conceived by the actor himself. Where one builds must decide where a parintelligibly ticular familyis to be placed in social space. Thus the inclusivelineage splitswhenever a memberestablishesa compound in a separate town; while withinthe town the assortis ment of households into small, tightly localized, lineage units (omole) similarlydependent upon voluntary choice of residence. Overcrowding and quarrelling within the omole leads to the formationof a secedent group which co-operates over an extended A period in establishinga new compound, and hence a new omole. 'firstchamber' is establishedforthe leader, who is helped and will gradually be followed by 'any other men ofthe omole may care tojoin him' (Bascom I944, pp. i i-i6). who On the testimonyof the ethnographerhimself,then, the choice of a house site is is the principal actor alone. The omole scarcely to be regarded as a decision affecting definedby agnatic descent andby common residencein a given compound. In building a house one musteitherdesertor remain withthat nuclear kinshipunit ofwhich one has been a part; one mustjoin or fail tojoin a particularpartyofsecession; whateverchoice one adopts must be comparativelypermanent. I suggestthat the custom of prescribing divination in such a context is eminentlyunderstandable on the ground of social

DIVINATION

AND

ITS

SOCIAL

CONTEXTS

I97

function.For it is the peculiar propertyof the diviner'srole that he is able, in the public for conscience, to remove the agency and responsibility a decision fromthe actor himself,castingit upon the heavens where it lies beyond cavil and beyond reproach. In the Yoruba case which has been cited, the divinerin effect provides a legitimating sanction upon a process of structuralrealignmentwhich, depending as it does upon a voluntary act, would be difficult indeed to sanctionin any remarkablydifferent manner. Nor is the Yoruba case exceptional, unlessin the factthat one is able to reconstruct, fromthe ethnographer'smaterial, the social contextof a procedure which too oftenis presentedsimplyas an isolated bit of magic, of no social consequence.' In othersocieties where a similarinsightis to be gained, divination appears to play a similar 'structural' role, sanctioningby depersonalizingthe various typesof action which may normallybe required in the process of sortingand resortinglocal living arrangements.Often the divinerplays a part in the cleavage ofgroupsthroughconflict. means ofdivination,a By groundless and merely personal or private accusation may be given publicity as an apparent fact,shorn of the obvious bias of its original sponsor. Thus among the Yao a large measure of voluntarismis involved in the processby which,in a matrilinealand largely uxorilocal society,a headman comes to command a village or hamlet mainly composed, on the male side, of a congeriesof individuals tied to him only by marriage. There are strainstoward virilocal marriage which are counteredby the stress laid upon which increases, the more successfula woman may be in producing matrifiliation, children (Mitchell I956, p. I44). Again, successionwithinthe matrilineageto the position of headman or chiefis competitiveand oftentransitory; depends upon political it ability and upon that peculiar social talent by which a leader is able to avoid serious accusation of such acts as sorcery(Mitchell I 956, p. I58), incest forthe sake of maleficent power (Mitchell I956, pp. I46, I8o), or lack ofdiligencein protecting interests the of the lineage women who hold the balance ofpower in the village. Inevitably,the cycle of alignment,succession,and secessionis moved by accusation and counter-accusation; and in virtuallyeverynew initiativethe divinerplays a part. But Yao divinationdiffers in form and degree of 'objectivity' from the Ifa divination of the Yoruba; and this characteristic difference accords with the more nearly transitory and, apparently,more 'emotional' characterofYao social relationships. Clarke I939) and perhaps needs no special recountinghere. The questions asked of the oracle are oftenkept secret,in the preciseformasked, even fromthe divinerhimself;the divinerrepeats set verses,corresponding the throwof a chance device; recognitionof to the applicabilityof a verseby the clientis the chiefmeans by which subjectiveselection is allowed to enterthe procedure. Mitchell describesYao divinationas follows: 'The typeofdivining usuallyby gourdand object (ndumba). is The divinerfirst all of establishes reason forthe visit,i.e. death, illness,theft, whateverit is. He then the or consults gourd and throws his into his hand a variety symbolic of objects.He builds a storyround theseobjects,and finally pronouncesthat he sees some personwhom he vaguely defines.The consultor then asked to mentionwho this could be. Though is is sorcery supposedto be practised mainlyin orderto gethumanflesh ghoulish for feasts, it is not believedto act haphazard. In otherwords,a sorcerer supposedto selecthis is victims revenge well as formeat. The selective for as processes work,therefore, at when theconsultor seeksto reconcile diviner's the findings withhisown suspicions, thathe are

Ifa divination been carefully has described(Bascom I94I;

I942;

OgunbiyiI940;

I98

GEORGE

K. PARK

his evil, i.e. from own pointof view, one with should selectsomeonewho is obviously for motivation sorcery. and whomhe has been quarrelling, he shouldappraisethepossible may by and the consultor the After has made theselection name is tested thediviner he (Mitchell1956, findings' latersubmita chickento thepoisonordeal to testthe diviner's
P. 153 n.).

It is notable that the formof divination,as described here, allows a degree of flexibility and publicity to which an accused might reasonably object; nor in the cases which Mitchell presentsis there any hint that determinationsof this sort are normally unan challenged by the accused. On the contrary, inconvenientdivinationis oftendoubted and belittled; and it may be tried again elsewhere. In the end, it is not so much the the but choice ofthe divininginstrument thatofthepublic, reflecting alignmentofinterwhose divinationwill prevail,and how far. personal loyalties,which decides
II

It may be objected that in a systemlike that of the Yao, since divination is ultimately part. But public significant overriddenby opinion, the divinercan play no functionally credence goes by degrees; it is not absolute. Moreover, divination is not simply a weapon to be taken in hand by any who wishes to increase his influence; the call upon the divinerrequires a particular sort of occasion, and the diviner must look to his own rules and to his own need for professionalindependence. An important aspect of procedure is just this-that it provides 'resistance' in its divination as institutionalized own rightto any client'sproposal. An interesting but, I think,not all-importantsource of such 'resistance'lies in the employmentof chance or chance-like mechanisms in the rendering of decisions. In Moore's briefbut suggestiveanalysis of scapulimancy among the Naskapi of Labrador (Moore I957) the point was made that the chance elementsin divination may sometimeshave a verypractical effect dispersinghuntingactivities,and the like, so as to in preventover-useof a favouriteresort.The emphasis was here upon the practical consequences of the use of a randomizingdevice in the selectionof certain criticalcourses of action. One is led to conceive of the divineras a sortofspinnerof a wheel offate,which is wiserthan any humanjudge. But thereare objectionsto what may be called the 'probabilitytheory'ofdivination, even when fulljustice be done to it. One is that it puts an incongruousemphasis upon the actual 'objectivity' of the divining process, an emphasis very hard to justifyon mustdig A ethnographictestimony. related objection is the obvious one thatthe theorist materialsto findonly a fewexamples where a random veryhard amongstthe descriptive device, by scatteringchoices, might have some use; while he is surrounded by unplausible cases. (Is it usefulor not to scatteraccusations ofwitchcraft?)Finally, divination in many societiesthrivesin the face of the fact that it possessesno truly'objective' formsat all. Among the Dinka, chance devices are evidentlymere importswith no firm standingand uncontrolledby the safeguardofknown and respectedrules; the divinerof in Power, a free-divinity, hisbody' and to reputeis 'one who is thought have an important who divines by uncontrolledpossession (Lienhardt I96I, pp. 68 seq., 7I). If we are to identify generic functionin divination,then, it must be one of more general applicaa bility.

DIVINATION

AND

ITS

SOCIAL

CONTEXTS

I99

We have consideredthe point that the Yao overridethe decisionsoftheirdivinersin many cases, and yet that the practitionerthrivesamong them. The Azande, who refer major questions to a procedurallyguarded poison oracle forwhich the probabilityof a given outcome in response to a given query is subject to mathematical calculation, frankly, certainlypossess a more 'objective' procedure; and theyhonour its objectivity by relegating other important oracles to lesser places. But complementaryto their respectforthe major oracles is theirdisinclinationto take seriouslyothers,which none the less persistin use. Such is the makama,which, though it is only used to indicate a witch, 'is regarded lightlyby every one and ratherin the nature of a joke' (Evansin PritchardI 937, p. 376). If we now ask whyYao divinationpersists the face ofobvious scepticism,the answer may perhaps be phrased analogously to the drama of makama is divination among the Azande. The instrument a cone of wood, insertedin a tightfitting sheath which is held by the witch-doctorat a seance. No one supposes that the witch-doctorcannot, by the way he insertsthe conical peg, decide whetheror not the man he hands it to is likelyto be able to pull it out. Yet when one who has taken the peg is unable to tug it away, the force of the demonstrationupon the audience is clear has been told to sticktightif the man is a witch 'and has come to enough. The makama spoil divination'; and behold, tug as he may, that man cannot dislodge the peg from the sheath held by the diviner. In thisway (I thinkwe may suppose) the witch-doctor consent in barring a who chooses to manipulate his audience may gain its effective certain person fromthe seance. And so in a great variety of other and less artificial situationsmay an act of divinationbecome the occasion forthe emergenceof consensus of where beforethere was none. That all shall agree, the previous commitments some mustmeet resistancein a momentofunanimity. My point is that divinatoryprocedure, whether 'objective' in quality or merely consensus upon a a constitutes technique for establishingan effective inter-subjective, by ratherparticularproject. If thatproject be conceived in itsentirety the clientpriorto his seeking the corroborationand quasi-religioussanction of the diviner, the requirehis mentto divine yetprovesan obstacle to him. By deflecting plan, divinationmay tame it. But more often,I judge, a client's 'project' is but vaguely formed;he has a grievance or cause foracute anxietybut no clear path beforehim. Then divination may act as a mechanism by taking the matter,as it were, out of the client's hands. decision-making Here a random device, forwhich intelligentalternativeproposals must be framed,is most appropriate. Such was Moore's case among the Naskapi. A band must hunt togetheror its essential unitywould be lost; yet in the contextof Naskapi culture the band had no permanent structure.Scapulimancy was much used by the Naskapi in connexionwiththe broadestimaginable range ofpurposes,and seemsto have permitted As for of fairly freeinterpretation. an instrument achieving a collective decision on the directionof huntingit was probably not particularly'random'; but it was suited by its apparent impartialityand its association with prescience to the task of rendering a decision withexceptionalauthority(Speck, I935, pp. I39 seq.). groups may make no use situated migratory By contrastto the Naskapi, differently at all of divination in determiningroutes, which may be determined by politically established rules of precedence or on economic grounds; for the little-structured organization of the Naskapi is not universal. The appropriatenessor inappropriateness

200

GEORGE

K. PARK

of divination se is thusone question;and the appropriateness a truly per of random deviceis another. Bothquestions requireanalysis thecontext socialstructure of of and conventional thought well as of the immediate as situation divining. of Externalor 'objective' divination maybe calledfor wherethestatus thediviner relatively of is low; but 'objective'divination whichis also trulyrandommay have verylimiteduses. Azande makama divination itsown propercontext, does,in greatmeasure, has as each separateoracle used in thatsociety. Divination 'objective'oraclecharacterizes by the Azande socialsystem clearly prophetic as as divination characterizes Dinka. In both the societies divination rather is clearly stratified, thesensethatthere greater in is attributionofreliability pronouncements one sourcethanto thosefrom to from another. On theother hand,theYao knowno suchuniversal stratification. better For reliability one hasno fairer alternative totakethetrouble going than of farther awayfrom home, where theatmosphere contention be lessheavy.The consequence suchegalitarianism of may of is one ofa generaltentativeness all social arrangements in predicated moralallegaon tionswhichwerelegitimated divination; by and that,indeed,wouldseemas essential to theoperation theYao socialsystem is formal of as procedure theAzande or depth to offaith theDinka. to
III

I have attempted place divination to amongtheimportant institutions theprimitive of worldby comparing operation severaldifferent its in social contexts, and suggesting thatin each case divinatory procedure theeffect stamping has of witha markofspecial a legitimacy particular decision a particular or kindofresponse crisis. to Paradoxically, divination appearsto have a derandomizing function; establishing consensus, renders it action morepredictable and regular.We have thuscome to the pointof regarding divination a practiceclosely as relatedto the problem controlling channelling of and publicopinionand belief;and thuswe comecloseto themoregeneral subjectofritual and itssociological meaning. Ritual occasionspresuppose underlying an that consensus; and by demonstrating consensus a dramatic in fashion they strengthen The 'field' ritual timeand space it. of in is indefinitely extended theborders thesocialworldknown itscongregation. to of to By contrast, 'field'of the diviner's the seance is restricted, centring upon an immediate problem, dissipating distance timerobtheproblem reality. divination and as and of Yet and ritual interplay; is thescaleofdivination nor of alwaysso smallas thesituation the 'seance' suggests. Ritual is employedto solemnizeattitudes divinerand towardthe towardhisparaphernalia amongst public,and evento solemnize own attitudes his his toward professional Thus an Ovimbundu his role. divining-basketto be 'put to bed' is each nightby a procedure The suggesting 'ritual' of psycho-analytic the literature. basketis further apartfrom set ordinary things containing skullof a poisoned by the child,bitsof the corpseof a fameddiviner, animalswho and durableremains from appearedin portentous circumstances (TuckerI 940). The conceived 'power'ofsucha it basket, wouldseem,must equal tothedramatic be effect required itin use. of The same themeof solemnization may be remarked the close associationof in divination withsacrifice. The consultant diviner as mayinvariably prescribe sacrifice, amongtheYoruba; in priestly divination omensare sought through solemn ritual.The

DIVINATION

AND

ITS

SOCIAL

CONTEXTS

201

contextwithinwhich a class of priestlydivinersmightintervenein affairsof state has been suggestedby Sidney Smith in an essay, 'The Practice of Kingship in Early Semitic Kingdoms' (I958, p. 3I): of 'The subjection the individualactionsofthekingto a procedure-theframing a of of or the questioncapable of onlya positive negativeresponse, examination the liveror weretobe interpreted as theflight birdsor thelike,thedecision to howthemanyomens of results-showsthat no kingacted to procurea majority favourableor unfavourable of of by accordingto his ownjudgmentalone, withoutthe possibility interference others. They were themselves well instructed the devicesof divination.... But theycannot in if werein opposition.' measures alwayshaveforced through, thediviners The suggestionof a check-and-balance system,stabilized by a qualitative division of role, is supported by what is known of the importance attributedto divination by the and by records of the severe ritual constraintsto which the Assyriankings themselves, king,at the hands of the priest,was subject. The point upon which I would insist,howbetween omen and public attitudein a inhering ever,is thatofthe necessaryrelationship societywhich so exalts the diviner.The war-lordwho disregardspublished omens cannot but prejudice victoryand, in the case of reversal,invite disaster: such is the plot of many an ancienttale. It is perhaps then a littleless importantto know how such a systemactually worked than to understandhow, of necessity, musthave been supposed to work.The stability it of ritual and structureover centuries,which for ancient Mesopotamia is subject to shared conventionalundercarefuldocumentation,bespeaks a rootednessin universally standings.The favourofthe gods was regularlytestedin public by the kinghimself;and the procedureswere patentlysuch as to lie convincingly beyond public suspicion-they of only to enhance the possesseda genuine drama. If the effect that drama was, directly, verity myth, the resulting of yet sacrednessof mythand ritual-as being the whole basis for the legitimacyof the systemof authorityitself-was such as to provide a terrible safeguardofdue procedure against the depredationsof expedience. It is mostunlikely,I suggest,that a critical militarydecision ever was thrownup to mere chance; or that chance ill-omensever were allowed in factto shake the legitimacyof a king's rule. Such possibilitiesare no more likely than that the Naskapi of Labrador ever went offin a completelywrong directionas the resultof an odd crack in a scapula. The functionof the ritual precautions, then, was precisely indicated by their manifestpurpose of colloquy with the gods. No enterpriseof state could proceed without the drama of suspense by which the eventual approval of the gods was made known, and the factof the directionof such enterprise beings higher yet than the king was demonstrated by to all. The verityand the power of the mythare thus of greaterimportanceto us than the genuineness of the chance or inspirational devices employed, which need have only dramatic truth. If we would correctlyunderstand Yoruba precautions against contaminationof the 'objectivity'of theiroracles, or Zande down-gradingof devices which theybelieve may sometimesbe rigged,we should thinkofsuch precautionsas protective of the essential credibilitypossessed by their more solemn procedures of divinationprocedureswhich,having ceased to be convincing,cease to have value. For while in one social contextthe requisitesof compellingdrama may be music,dancing, and shouting,

202

GEORGE

K. PARK

and the appearance of sudden 'possession' in an immoderatelydecorated diviner,in a quite different context a drama of markedlydifferent quality may be the only appropriate one. For the canons ofpopular drama are not everywhere same. the We began with the point that divination must 'resist' in order to produce convicin viction; and we have been broughtto the point ofconceivingthat resistance dramatic a terms.Where it is used, the genuine randomizing mechanism effects set of bafflesto directmovementtoward a goal, and the drama is that of a maze; but it is functionally equivalent to the merelyritual or emotivedramatizationfoundin othercontexts.If the typesof divination should be named, then, I would call them mechanical, ritual, and emotive; although the three types easily shade into one another in many systemsof divinatorypractice. Thus the drama of 'possession' is not necessarilya frightening one; it may merely serve, like a mechanical device, to establish the apparent presence of otherwise invisible beings. Clairvoyance and mediumship or ventriloquism are in different contexts the functional equivalents of the induced convulsion of the body. Divining bones and similar collectionsof paired and named objects, even though they and even though theybe interbe cast out in public view or cast by the client himself, pretedby a whollyexotericcalculus, asJunod (I927, II, p. 564) made it quite clear that nice application to the case in hand-they do not theymightbe, only dramaticallyresist refusebut parry,reverse,and redirectthe questioning until it finallyculminates in a meaningfulresolution,a denouement which, ideally, suddenly reveals the hidden clue to the turnsof the drama, like the verb in Virgilian verse. Among the Sotho by I935 (Ashton I952, pp. 296-300) the elementof chance, or fate,or directcommunicationby spiritswas merelyassertedin the formalact of casting the bones; the emphasis lay not into theproblems therebut in the skillwithwhich the divinerwas able to achieve insight of his client and to bring what was usually a protractedseance to an impressiveconclusion. Thus the great work of the Sotho divinerwas not in the cure or charm which he finallysold to the client,but in the artificialtask of discoveringwhat the client had come about-by skilful diagnosis he must establish his own extraordinary powers as a practitioner;and thatwould give to his cures theirvalue. So among the Buye it was the lesserdivinerswho employed the more random and emotionallyless impressivedevices; the kilumbu diviner,upon whose word a man might be seized fora witch and submittedto the poison ordeal, was in his body but in strange voice the medium of communication froma spiritworld. The findingof a witch was througha culminativeprocess. It began with an 'objective' seance; and sorcerycould it only be divined when, afterspiritsand ancestorsall alike had been triedand refused, was necessaryto hint at living persons. There followed an impressiveseance with the kilumbu diviner, whose mechanical devices only supplemented and coloured his own charismatic aura. Beyond, there was a sort of 'high oracle' of final recourse, or the poison ordeal (Colle I9I3,I, pp. 38I-92; II, pp. 469 f.). It is my suggestionthat these various formsof trial should be conceived as ranging along a gradient of increasing unimpeachability; and that each item of procedure should be understoodas essentially of toward that end-the dramatic establishment an ostensibly irrevocable contributing judgment.

DIVINATION

AND

ITS IV

SOCIAL

CONTEXTS

203

of There is perhaps a certainlack of logic in a cumulative system trials,each of which is but any one of which may contradict the ritually exalted to the pitch of infallibility, in proceding. But there is no lack of drama here; nor does the same characteristic the developed court systemin the urban societypreventthe lower courtsfrompronouncing finaljudgments. Instead, there is an enhancement of the sense that truth,eventually, must always be found out. A hierarchical systemof guilt-divinationparallels, in the logical frame of its organization, a systemof courts. The drama and seriousnessof procedure increases, as only the 'guilty' pass on to a further trial; suspense holds, for a each succeeding test offers chance of exoneration; only no person or body present for the critical judgment. At the bottom of is invested with apparent responsibility the systemis the lowly consultantdiviner; at the top, perhaps, the ordeal, divination by invisible diviner-that ne plus ultra,as it has so often been pictured, of human irrationality. But ordeals in many societiesare interchangeablewith other formsof guilt-divination. Bourgeois (I956, pp. 260 seq.) reportsfromRwanda and Burundi2techniques of guilt-determination ranging fromaccusation by a jumping grasshopper to a series of ordeals, some of which could manifestly operate only on a 'psychological' basis, and otherswhich as clearlyare insensitive the psychicstateof the accused. Here the social to contextof the ordeal as a trial forsorceryis perhaps epitomized in the fact that torture mustbe used to extractconfession where the signsindicatedguiltbut theaccused denied; and by the further fact that the drama mighteven then turn: fora stoutsubmissionto could resultin pardon by the highestauthority. torture The analytical problem, I suggest,which such facts present is essentiallyone of showing a constant relation between a divining procedure, insusceptibleto any but a procedural criterionof validity, and the public conception of right. As the Yoruba, moving in social space, legitimateshis move by submitting alternativeplans to divination, so a movement of witch-killing, whatever its true determiningconditions,must have legitimation.Moreover, as in the political theory legitimacy, mustdistinguish of we between formaland substantivemanifestations: is in the necessity achieving more of it than formalgrounds forpublic execution that the puzzle of the mortal ordeal may be understood.At the momentof action, I would argue, it is substantivelegitimacywhich is mostessential; the act which is predicated upon the decisionofdivinatory proceedings mustbe the one which,as all concerned now concede, ultimatewisdom would findtrue. But thereis the further that,in the long run, the emotionalconsensuscreated by the fact divining session will fade; so that the legitimacyof the action, in retrospect, will rest upon formal considerations. Both conditions, the formal and the substantive,must therefore met; the latter to preventimmediate disorder,the former preservethe be to charteringmythof constitutedauthority-its claim of embodyingtranscendent justice. The immediate importance of substantive legitimation is well illustrated in the materials reported by Bourgeois, Weeks, and Colle. Where a confessionis required beforeexecution,it is evidentthat the formalindicationofguiltis feltto have insufficient weight.We mustimagine, I think,thatin a segmentary societyno accused is withouthis kin, disposed to support him if the matter is not handled well; and that it is their

204

GEORGE

K. PARK

acquiescence above all which must be publicly demonstratedbeforean exacted death will become an act of the greater collectivity and quite distinctfromaction by a vengeance group. But public torturecan, in certain contexts,bring about a revolutionin group sentiment.One way of handling that emergencywould be to reservethe possibilityof a pardon; another,I think,is to be found where procedure reliesmore fullyon the logic ofthe mortalordeal. A comparisonofBuye and Kongo ordeals is instructive. The importanceof the state ofpublic sentiment, and an indicationofthe logic in the ordeal, emergefromthe Kongo materials of Weeks (I9I4, pp. 263-4). A duly accused witch is given bark poison and may exonerate himself vomitingit in foursuccessivetrials.Normally,on succeeding by in thisway, he is fetedwith great enthusiasm;no taintof the earlierindictmentremains against him. Yet in some cases thereis no immediate reversionof public sentiment.A person 'very obnoxious to the people generally' (a person deserted even by his kin?) is or subjected to a further procedure, a sortofintellectualtorture ordeal. Dazed fromthe of effects the poison, he must none the less show normallyquick powers of discriminathe or tion,identifying species of twig thrownto him, or the type of ant, butterfly, bird pointed out; and a single miss legitimatesexecution on the spot. The poison ordeal is intended,as it were, not so much to determineguilt as to demonstrate Where opinion it. is divided, it may demonstrateeitherguilt or innocence. But where opinion is one, and the poison yet fail to demonstrateits truth,a more certain ordeal may be added, whose formis tinsel but whose decision will validate the popular mandate. The same bark poison which, in fouradministrations, should kill a witch is employed in a singledose in a lessercase; yet the value of the poison is here reversed.Innocence is demonstratedby retainingthe poison unharmed; and vomitingdemonstrates guilt,whose punishmentis short of death. Yet there is a consistencyof logic, when the ordeal is perceived as demonstrationand not merely trial: if the poison is to demonstrateguilt in a capital it offence, cannot fail to kill; ifpoison is to demonstrateinnocence in a less-than-capital it offence, mustbe retainedwithoutkilling.Here is divinationby 'objective' sign. The procedures of Buye (Colle I9I3) law in connexion with the poison ordeal suggesta rather different social context,in which the poison in fact operates as legitimating divination,of the highestorder, in what is not a public execution but a public vengeance-murder.If the accused failed quickly to vomit the poison upon the first trial, he was set upon by the waiting relativesof his supposed victim,who cut offhead and limbs and threw them into 'un grand brasier', probably then to be bought fromthe group by an anthropophage, whose secret societywould obtain fromsuch a body the sense of increased supernatural power. The ceremonial ordeal here recounted was administered,away fromthe village of the deceased, by a ritual specialist of official standing; but it mightbe duplicated on a less imposing,thoughequally mortal,scale in private vengeance trialsadministeredwholly by the vengeance group itself(ii, pp. 8I3 seq.; pp. 469 seq.). The interplayof formalprocedure and direct,vengefulaggression indicates the immediate translationof formallegitimationinto substantiveright.Presumably, the divinatorysystemcould be said to have reduced bloodshed as against a putative vengeance systemrequiring a death fora death. Since Colle remarks(p. 814) that the ceremonial ordeal would be imposed in a given case by the kilumbu diviner we himself, may suppose that he was not unaware of the relation inheringbetween the

DIVINATION

AND

ITS

SOCIAL

CONTEXTS

205

and of formal with socialsituation achieving legitimation thecertainty whichit maybe the to of Divination without diviner the expected resolve problem substantive right. is As of peculiarly suitedto such a context controversy. a public demonstration guilt, of ordealcouldpossess dramatic a or equallyofinnocence, mortal the qualityunequalled we The merepublicity theeventas a mortal of bytheother techniques have examined. it trial,withits apparentacceptanceas such by a greatcrowdsuffering to proceed, appearsto ruleout thepossibility in thepublicconception, actualresult that, the might have been predetermined theofficial by mixer whatthemodern of readermustregard as a merely chemical poison.It was notso regarded theBuye. by Thereare, I think, majorpuzzlesin connexion two withtheordeal.One, however, of is simply artifact themeeting an between believer sceptical it and observer; evaporates is from superficial the of whichsupport whenattention shifted plausibility the beliefs in of how divinatory practiceto the problem understanding such beliefs factbecome in The established reinforced collective and one experience. other puzzleis thetechnical of ofunderstanding peculiarproperties theordealwhichhave made it a widespread the in in phenomenon variouspartsof the worldat varioustimes.I have suggested this as functioned theapical procedure paper thatthepoisonordealhas in severalsocieties and thatI have offered funca in a system guilt-determination; it is in thoseterms of of of The ordeal,likeother forms divination, been prehas tionalanalysis thecustom. in situations as sented an instrument which, concrete by callingforaction,a particular and a particular resolution construction be put upon a problem, may sociallyestabof lishedas theright one. In thecontext guilt-divination, consequence legitimathe is or The of tionofdemandsforpunishment restitution. ordealis characteristic societies in whichthe rightto demand punishment not been successfully has pre-empted by or constituted the authority, wherethatauthority requires dramaofhumansacrifice to the penetrate publicconscience.
v

Divination typical thefolk, is of and notofthecontemporary of urban,form thesocial is life; its distribution also unevenwithinthe folkworld. 'Divination',wroteLowie (I935, p. 255), 'is rightly considered atypicalforAmericanIndians.' His perspective and was thatofthePlainsIndians,possessed a sociallifein whichloss,theft, of illness, different thatwhich characterized has most the of to deathweremetin a manner wholly in of African considered thispaper. As the individualistic societies quest-religion the a in Plains differed from congregational centred the regardfor diametrically religion as suchgroups theCrow common so of ancestry, was therelation man to man amongst in war, in hunting, marriage, in the ceremonial in or Cheyenne whether or distinct, a solemnificationthosepeculiar of commitment the to valueswhichcharacterize tribe's life has known. one weretosetoneself taskofdiscovering If the whatparticular it cluster of human sentiments association witha mightmostprobablybe foundin constant of the between PlainsIndian and BantuAfrican developedsystem divination, contrast to loss and mortal responses physical helplessness, by death,characterological deviance, a of dangerwould offer starting-point obviousmerit.Nor can such a task,I think, be the distribution divinaof ultimately avoidedifwe are to understand ethnographic and thence establish scientific to a of tion, comprehension itsnature.
0
J.R.A.I.

2o6

GEORGE

K.

PARK

It has been the tendencyof thispaper to suggestthat divinationis regularlyassociated, in a comprehensible manner, with certain types of social situation; and it is of perhaps unnecessaryto insistthat, with due regard to the operational fictions 'pure in structuralism' anthropoliticalanalysis, equivalent social situationscannot occur in cultures.That is because actual human situationsarise as much out two widelydifferent as and actor-definition theydo out of such eventsas death or cattleof actor-perception of If theft. divinationis to be known as an importantinstitution the folkworld, and not thoughtcan uncork, I have suggestedthat it as a mere example of what unscientific must be appreciated in its contextof private sentimentand public opinion as these are societies.Guilt-divinationamong the Buye culminated in found in variouslystructured a public vengeance-murderand cannibalism; among the Azande (Evans-Pritchard neithercannibalism nor the mortal ordeal had any such firmplace; rather,by Ig60) which were celebrated in the Buye practice had historicaltimes, the very sentiments and formalistic control: been subjected among the Azande to firm withdanger fraught 'It is apparentthatwhena witchis exposedby theoraclesa situation dignity to are sincetheinjuredman and hiskinsmen angryat an affront their is created, shall thatanother No welfare a neighbour. one acceptslightly by and an attackon their his or ruin his hunting undermine healthout of spiteand jealousy,and Azande would werenot resentment themiftheir who are provedto be injuring certainly assaultwitches 1937, (Evans-Pritchard channelsbacked by politicalauthority' directedinto customary p. 85). The active role of witch is all but wholly imaginaryamongst the Azande; yet persons identifiedas witches abound and are neithercut offfrommost normal social ties nor of by usually killed. The witchis identified formaland 'objective' systems corroborative divination,which accept and rejectproposals at random; and the act by which a witch recalls and cancels his psychic aggression is equally formal. Public opinion is little roused; substantivelegitimacyis not a problem but followsfromgeneral acceptance of the correctness thoseformalproceduresofdivinationwhich are supposed to have been of followed.When substantivelegitimacydoes, rarely,prove a matterofsome concern,the case may be put to a test; although normallynot until the accused dies a natural death, when public examination of entrailsagain puts to proof,by corroborativedivination, the proposal that in life he was indeed a witch. A particular belief,moreover,would to appear specifically discourage the relativesof a deceased frominsistingupon such a test-witchcraftis thought to be inherited,and the taint, in the case of a damaging finding,then falls upon the living. It would be only an already embattled group of kinsmenwho would stand to gain fromraising the issue; in general, the Azande are content with formal and undramatic procedure, so that public feelingfairlyreadily subsides, excuses, and forgets.While all may agree that the witches about them are many, thereis no general agreementas to who theymay be (Evans-Pritchard I937, pp. 23 seq., etpassim). fromthat of the Buye; The social systemof the Azande, I submit,is very different nor is thereany obvious need that we should trace the reason forthisgeneral difference to a source,as it were,in one ofitsparts. The differences political and social organizaof tion are manifestly of important;but so are the differences ecological setting;nor would an analysis betraying preferencefor 'psychological' causes be barren of pertinent

DIVINATION

AND

ITS

SOCIAL

CONTEXTS

207

and to systems divination, itseemsevident of of paperis with evidence. The concern this value. It shouldbe clearthatZande also,have someexplanatory theauthor thatthese, is to is as to of divination suited thecharacter Zande life, Buyedivination suited another; has but divination, have insisted, a logic of its own and is not an inconsequential I propose,then,thatZande ideas about witchcraft institution. may as reasonably We Zande divinatory as practices, we mayput the case theotherway round. follow from was we thatthedramaofBuyedivination notmerely And in similar fashion mayinsist social by a fierce but sortof entertainment, was an instrument whichthe prevailing had of and for appetites, fears, ideas,as reported theBuyeat thebeginning thiscentury, the between I thatwe maybestperceive affinity been shaped.It is in thislight, think, of and ordealand crucifixion, between poisoning theburning witches. or the with in worldhas muchin common procedure, divination thefolk As a legitimating urban complex'in the contemporary what we may call the 'licensing and certifying and death, marriage, as concerned withsuch matters birth, world,whichis similarly and the inoversocial boundaries and whichfunctions connexion in withmovement or insanity. of rights, wellas withcrimeand dangerous irksome as fringement property bureauxhas a Yet a world which cleansedand ordered a vasthierarchy licensing is by of ratification all its of to qualitylackingin the worldwhichmustresort supernatural in minorchangesofstatus.In peasantChina, as exemplified WestTown (Hsu I948, a in of was no ritualintervention marriage, thesortwhichrequires pp. 85 seq.), there There concerned. solemncommitment thechoiceby thetwoindividuals families or to of a stagein thebetrothal, quietprocedure submitting was, however, a preliminary at which to the 'eightcharacters' a son and prospective of daughter-in-law divination, scheduleof shouldexposean unsuitable the of matchbefore start a long and eventful of essentialto the establishment the match as a duly gift-giving negotiation and society, which in was thatof a seclusionist contract. The social context acknowledged in of quality, centring notions family symbolic and putative publicity a peculiarly had of and expression from timetotimein theform which required objective dignity honour, and to feasts, processions, gifts the livingand the dead. The elaborately ceremonial was feltto be genuine,even thougha submission a marriage-choice divination of to incomfoundthe 'characters' final:ifone diviner negative result was not necessarily Divination simply to another. go of on patible, father theboymight, consideration, the it thenprovided morethana formalistic no ratification thechoice;yetin itscontext of a test had function meaning. was during betrothal pre-eminently The negotiation and resources ofthestatus theboy'sfamily; economic the of and as suchit strained family's to the utmost. launching a peculiar, of the Divination hereprecededand legitimated ceremonial mustpair itself thepubliceyewithanother in by splurge whichthefamily in size family, demonstrating own meansin thenegotiated ofitsgifts its and unrelated and kind, as wouldprobably returned dowry. be of money onlya portion which character-itwas More usually, divination WestTown had an evenmoreroutine in proceedings each pointuntilthe ausat employed onlyto hold and checkceremonial on its divination betrays, analysis, peculiar picioushourwas met.Yet evenin thisform a the stretch ceremony, greater in usefulness; the longer, West Town, one might for was one'sdemonstrable particularly, think, I illuminates importance. That proposition the use of divination burial. Predetermining saving forthe cost of an old and in

2o8

GEORGE

K. PARK

an man's deathcould becomea majorpreoccupation his family, for whichrecognized impending of its public and heavenlystanding(Hsu I948, p. I33). The extent test of mourning, feasting, economicoutlayfora procession and mustbe wisely decided. In theevent muchdepended uponthelength the dead man's stay,as a corpse, of itself, withhisfamily; thiswas a sign,not onlyof the family's for regardforthe man, but also of its regardforthe house whosefortune had guided in life.Thus it was, I he suggest, thatdivination mustdetermine hourwhenthecoffin willbe nailedon 'the lid thecoffin, hourwhenthecoffin be removed the will from house,and theday when the thecoffin be lowered of will intothepitforburial' (Hsu I948, p. I56). But thesystem divinedonly was divination nothereone whichleft matter chance.Poorfamilies the to burial amongst well-to-do the hour,not the date; while an important the mightbe The of delayedfor or twomonths. diviner one whosettheworth a family thustiming by he itsfuneral from sharedthe universal exemption reproach a peculiarinstitution: of not was themidwife decision, of knowing itsauthor. have had in the I have attempted sketch meaningwhichdivination to the might because it has usuallybeen treatedonly as a context Chinesefamily of institutions and because I have found it or minorpart of an intellectual criteriological system in of illustrative themanner which divinatory legitimation paralleland supplement may in its working mostformal is a the ritualsanction.Legitimation a processwhereby is particular act, event, establishment declaredto be an exampleof a class already or defined the presupposed in norms a society;and whereformal of onlyis legitimation is concerned movement simply, in a lowercourt law,from the as of 'norm'to 'case'. But proas in a higher courtthemovement a tendency reverse, the 'case' from has to and as viding precedent comesto modify 'norm', I think should the so we regard divination a instrument social control.Earlier anthropologists of were not sufficiently two-edged of impressed, perhaps,withthe airiness 'norms'in the preliterate society:Sumner's in as wereforces themselves, real as Durkheim's mores Society mechanically-conceived or as the Churchto earlierunsympathetic historians mediaevallife.Yet Durkheim of and the had the insight perceive to thatSocietyis the creation ceremony, to stress of in of interaction theceremonial lateranthropologists importance emotional gathering; but have come to see thatthemores notlive on ofthemselves mustbe recreated do by in and of ritual,ceremony, the act of constituted authority the experience each indihas and thetorture heretics of thattheordealsofwitches vidual; and history perceived mayhave been morethanthemereapplication secure'norms'to 'cases'-that such of in and dramatic reveal us a crisis the'securitythenorms to of cruel themselves. phenomena
NOTES
1 The readeracquaintedwithsome oftheolderpublished sourcescitedin thispaper will appreciate the difficulty reconstructing, of frommaterialsorganizedonly as description and in keepingwith a layman'sinterests, effective the social context 'sorcery' of and itsallied arts.I have been led, however, to makeuse ofaccounts from before onlyshortly or after authoritative Europeancontact;for divination was thenmoreimportantly concerned withsocial control thannow.Junod (I927, II, p. 571) was preparedto ratetheart of'astragalomancy' veryhighly indeed: 'All theelements Nativelifeare represented.... of It is a resume their of wholesocial order, all their of institutions.' he wenton to deploreitsuse, as a But bulwarkof backwardness; and I thinkfewlater observers have been in a positionto note so high a of development theart.

DIVINATION

AND

ITS

SOCIAL

CONTEXTS

209

2 The precise whether explicit, reported Bourgeois rarely by is reference thematerials of ethnographic of of context thesystem torture as toplace, ethnic group,or timelevel.For thisreasononlytheimmediate practices contained he describes clear; and thesamewouldapplyto all therichcatalogueofdivinatory is in hiswork.

REFERENCES
ASHTON, HUGH BASCOM, WILLIAM BASCOM, WILLIAM

BASCOM, WILLIAM R. 1944. The Sociological Role of the Yoruba BOURGEOIS, R. 1956. Banyarwanda etBarundi. Brussels.

1952. The Basuto. London. R. 1941. The Sanctions of Ifa Divination. J.R. ANTHROP. R. 1942. Ifa Divination. MAN, 1942, 21.

INST., 71,

pp. 43-54.
part 2.

Cult-group. Amer. 46, Anthrop., I,

London. Cults ELLIOTT, ALANJ. A. 1955. Chinese Spirit-medium inSingapore. Oxford. EVANS-PRITCHARD, E. E. 1937. Witchcraft, Oracles Magicamong Azande. and the EVANS-PRITCHARD, E. E. I960. Zande Cannibalism. J.R. ANTHROP. INST., 90, pp. 238-58. HERSKOVITS, M. J. 1938. Dahomey, vols. New York. 2 Hsu, FRANCIS L. K. 1948. Under Ancestors' the Shadow. New York. JUNOD, HENRI A. 1927. TheLife a South of African Tribe, vols. London. 2 I 96 I. Divinity Experience. LIENHARDT, GODFREY and Oxford. LowIE, ROBERT H. 1935. The CrowIndians.New York. MITCHELL, J. CLYDE 1956. TheYao Village. Manchester. OMAR KHAYYAM MOORE, 1957. Divination-a New Perspective. Amer. Anthrop., pp. 69-74. 59, OGUNBIYI, Modes Divination. T. A. J. 1940. Yoruba Oracles their and of Lagos.
SMITH,

CLARKE,J. D. 1939. Ifa Divination. J.R. ANTHROP. INST., 69, pp. 235-56. COLLE, P. 1913. LesBaluba. Brussels. COLSON, ELIZABETH 1958. Marriage the and Family Manchester. among PlateauTonga. the

Ritual, Kingship. and Oxford. SPECK, FRANK G. 1935. Naskapi. Norman, Oklahoma.
LEONA STUKEY

SYDNEY

1958. The Practice of Kingship in Early Semitic Kingdoms, in S. H. Hooke (ed.), Myth, 1940. Divining Baskets of the Ovimbundu. J.R.
ANTHROP. INST.,

TUCKER,

WEEKS, JOHN

H. 1914. Among Primitive the Bakongo. London.

70, pp. 171-201 .