-iSocial Relations in the Understanding of PhysicalEvents among the Eastern KorekoreM. F. C. BourdillonDepartment of Sociology, University of Rhodesia, SalisburyDurkheim and Mauss (1901-2) were thefirst seriously to consider how the structure ofsociety can influence the way in which peopleunderstand the physical world, but they con-sidered the classification of objects rather thanthe explanation of events. More recently Hortonhas suggested that in African traditional thoughtthe explanatory models of religious beliefscome from the prototype of human society(1964, 99), and has claimed that to traditionalAfricans 'the human scene is the locus parexcellence of order, predictability, regularity'(1967, 65). That there is some relationship be-tween the social structure and the understandingof events, particularly of misfortune, is shownin the changing patterns of explanations foundamong urban Africans (see Mitchell, 1965;Hammond-Tooke, 1970). Horton went on toemphasise the role of beliefs in spirits in pro-viding explanatory theory, an emphasis whichdetracts from the direct relationship betweenthe observed state of social relations and theunderstanding of physical events.This paper examines the relationship be-tween the social order and the understanding ofcertain physical events among the EasternKorekore, a Shona-speaking community in theextreme north-east corner of Rhodesia betweenthe Mozambique border and the Mazoc river.The paper concerns beliefs about and reactionsto sickness and death, and attempts to analysehow the Korekore explain and understand these.The topic is a matter of convenienceŠreactionsto sickness and death are to some extent ritual-ized and are easy to observe Š but it shouldbe noticed that while these physical events haveconsiderable social significance, one should be-ware of applying the analysis to the Korekoreunderstanding of all physical events.ILLUSTRATIVE CASE HISTORYI shall focus my analysis on the illness anddeath of a young man (No. 17 in the genealogy).He was the youngest son of his parents, abouttwenty years old when he died, unmarried andstill living at the home of his parents. Thefamily was poor with no source of cash incomeat the time in question. There were then elevenresidents at the homestead: the parents, foursons (13, 14. 15, 17), two daughters-in-law andthree grandchildren (20, 21, 23).The father closely associated with his latesekuru (5) (in this context, classificatorymother's brother), his mother's brother's son,who in life had a good source of monetaryincome. This sekuru invited the family tolive near his home and helped to financeone of the marriages of the youngest livingson (15). At the former's death, his eldestson (9), the present sekuru, who had a smallerand dwindling source of monetary income,continued this association and the familyin question moved to within fifty yards of61the sekuru's homestead. The boys of the familyhelped their sekuru by herding his cattleand doing other chores. He in turn kept adaughter of the eldest son (19) and often pro-vided food for the boys.*The numerous moves of the homestead siteare of interest. About 1956, at the death of thefather's father (3), the family moved at thesuggestion of the late sekuru about five milesto within a mile of the latter's homestead inorder to be near some relatives. A couple ofyears later they lost a twenty-year-old son (16)and moved to a site about three miles awayon the other side of the sekuru's homestead:they did not consult a diviner about the causeof the death but they do hold that such deathsare always due to witchcraft* They did notbuild huts at the new site and within a yearmoved a mile for no ostensible reason to a suit-able place which they inhabited for about sixyears. They moved again at the death of thefather's mother (2) in 1966 to within a mileof the late sekuru where they were living withthe families of the father's younger brother (8)and of the eldest son (12). At the new site,the father's younger brother lost four children(18) and the son lost one child (22), after whichthey moved in 1968 to their present hamletnext to their sekuru's homestead. In a dozenyears, the family moved five times, and fourof these moves were occasioned by death inthe family.We come now to the sickness and deathof the youngest son (17). In March of one year,he had sores that went septic and would notheal. The family consulted a man with a heal-ing spirit who gave him some medicines anddivined that the illness was- caused by thesekuru's sister (10) who was bewitching him*The late sekuru is in fact the late chief and his sonis the acting chief. The close association between thefamilies of a man and his sekuru is not uncommon inthe area although the financial aspect varies with thefinances of the parties concerned.*Although a case can be made for the classical dis-tinction between witchcraft and sorcery (cf. Evans-Pritchard, 1937, 21, 387) in Korekore thought, theKorekore use one word (huroyi) for both and thedistinction is somewhat vague. For the purposes ofthis paper the distinction is not useful and I use'witchcraft' in a broad sense that does not excludesorcery.because her brother, the sekuru, was payingtoo much attention to the family in questionand not enough to herself and her children.Her brother was maintaining the kinship rela-tionships he had inherited from his father tothe neglect of his personal relationships. Thesores still failed to heal and the young manwent to the nearby mission hospital where hewas treated for three weeks. This state of affairslasted until October, with one of the woundsre-opening and finally healing at the mission.About this time the young man started com-plaining of severe headaches, and when theremedy of certain roots and leaves known tohis father had no effect he asked for Europeanmedicines. In November, after a severe doseof what appeared to be influenza, he consulteda diviner to find out who had caused the illness,it being supposed that his mother's late mother(1) was troubling the family. Divination againpointed to bewitchment by the sekuru's sister.Early in December, he was again very illwith severe headaches and general weaknessand went to consult a local woman known tohave a healing spirit. She claimed that the illnesswas due to a very bad spirit, the spirit of aperson who died very long ago and not at rest.The spirit was in his blood and her healingspirit would need £1 to drive it away (the usualfee being 7s. 6d.). The family raised the moneyand were told that they would have to find ablack chicken into which to transfer the evilspirit before driving it into the bush. Althoughthey found the chicken, the members of thefamily claim that she later said that she couldnot cure the sick man, whereas she claimedthat they changed their minds and did not gofor treatment. In either case, the traditional curewas not carried out and the sick man went in-stead to the mission hospital some twenty milesaway. There his condition improvedspontaneously and he left after two days:according to the hospital records he simplyhad his ears cleaned. He was again very illand returned to the hospital ten days laterwhere, after rallying strongly, he died suddenlyearly in January.Before he died the accused sister stayedsome days with her brother, the sekuru. Sheopenly stated her grievances saying that herbrother was caring for and feeding the familyof their late father's muzukuru (in this context,classificatory sister's son) whereas she wantedv JVVA6214Ł4_j *her own children to stay with her brother.Before she left for her own home some tenmiles away, she said that she would not returnuntil she had heard of a death in the familyconcerned Š a remark regarded as an openthreat of bewitchment. Although on that daythe young man had appeared to be well on themend, the next day he suddenly lapsed anddied.Although everything pointed to witchcrafton the part of the sekuru's sister, the sekuruhimself would not accept this and a week latersummoned a diviner. This diviner is relatedto, and lives with, the sekuru's brother-in-law(II) who is on close terms with the family, anddivines while speaking in a state of possessionby his divining spirit. He had been consultedby the father of the young man before the death,when he had stated that the sickness was beingcaused by the sekuru's sister. Now, whensummoned by the sekuru, he repeated his find-ing together with a few intimate family factsin support of it. In particular, he remarked thatthe sister in question had helped her motherto bewitch her father, and added that the oldman had during his last illness claimed thathe was being bewitched by his wife (4), towhich the son agreed.TYPES OF EXPLANATIONIt the above case, the supposed action ofthe witch does not appear to be entirely dis-similar conceptually from the 'modern' ideaof a malevolent person putting poison intosomeone's food to cause illness and death inhis victim. It is true that according to oneinterpretation the witch had introduced an evilspirit into the blood of her victim, but thespirit was treated as an impersonal object andit could be removed independently of the witchjust as we would treat a poisoned person in-dependently of the agent who introduced thepoison into his food. The difference betweenthe two types of explanation appears to be intheir content rather than in their logic. Theexplanation in terms of witchcraft was based onthe empirical evidence of the open jealousyof the supposed witch, and her threat followedby the sudden and surprising death. That thesickness was in the blood followed from thefact that the septic sores appeared in differentparts of the body and from the final generalweakness.In the case cited, the woman healer sug-gested that the witch operated on her victimby introducing a bad spirit into his blood. Themore usual manner in which witches arethought to operate is to place some harmfulconcoction or poison in the victim's food ordrink or on a path along which the victim islikely to pass. Witches are also believed tokeep certain animals to help them carry outtheir nefarious deeds: an owl alighting on theroof of a hut or a snake in the vicinity of ahamlet immediately rouses fears of a witchat work. If a man is bitten by a snake anddies, it is quite clear that the snake is thephysical cause of the death. People take careto hunt out and kill a snake seen near theirhomestead, and if the snake is killed, the witch'splan is foiled. So far, the witch appears as ahuman agent using her special knowledge tomake use of techniques not available to theordinary person. This requires no peculiar logicto understand.It is true that the Korekore more readilythan ourselves suspect the presence of a per-sonal cause: a personal agent is believed to beinvolved in any premature death through sick-ness. Opinion is divided on whether a manwashed away while crossing a river or killedin a car crash must first have been bewitched.A common view is that when the causes areso clear, the relatives of the deceased wouldnot even bother to consult a diviner, thoughyounger men say that there is always witchcraftoperative in such accidents and that anyonewho urged the deceased to make the journeyresulting in his death would be suspected. Allagree that a very old person dies simply fromold age, though a few say that they wouldconsult a diviner at the death of an aged rela-tive in order to be sure that there is nothingelse. Thus the Eastern Korekore do accept thepossibility of a purely natural death in certaincircumstances, but they generally expect othercauses.The fact cited in the case history that thefamily moved their place of residence fourtimes after death in the family does not neces-sarily indicate some 'mystical' fear of death.This practice is common at the death of anadult, and the reason given by informants isthat the old homestead with the hut of thedeceased has too close an association with thedead relative: they move in order to forget.63The Eastern Korekore do not normally moveresidence at the death of a child, but in thecase of a number of deaths the family doesmove through fear that more will die. In sucha situation the locality is regarded as dangerousto the family, possibly on account of the pre-sence of an enemy, and to abandon such aplace might be regarded as reasonable in anysociety.Even customary precautions against the in-fection of death need not be understood in a'mystical' sense. Thus everyone attending theburial of the young man had to wash his armsand legs in a river before returning to thevillage, and those whose tasks brought theminto contact with the corpse or grave had towash their whole bodies. All had to wash theireyes with certain leaves crushed in water orelse the eyes would 'die', and children were notallowed to see the corpse for the same reason(formerly they would have been shut away inan empty hut). All shoes worn at the grave andimplements used for digging it had to be care-fully washed before returning to the village.There is something about death which is be-lieved to be infectious and precautions mustbe taken to avoid it, yet there is nothing inthe rites mentioned which radically differentiatesthem from the way in which Europeans mighttreat contact with the corpse of a victim ofcholera or bubonic plague (though other Kore-kore rites do require further explanation). Eventhe possibility of blindness from looking at thecorpse need not be understood 'mystically' Šone can become blind from looking at the sun.Certainly the Korekore have some concept ofillness being passed on through contagion. Theoccurrence of venereal diseases, and often ofmeasles, is understood in this way and it ap-pears that the idea is extended to that aspectof death which is regarded as infectious.The use of leaves to protect the eyes isnot altogether dissimilar from the western useof prophylactics. The Korekore understand theirmany traditional protective charms and curingconcoctions in the same way as they under-stand European medicines. People frequentlyswitch from traditional to European medicaltreatment if the former fails to achieve thedesired effect, and vice versa. The fact thatthe traditional medicines are usually scientific-ally unsound does not necessarily imply thatthe people who use them regard them as havingsome magical power any more than the useof scientifically unsound 'natural' remedies bymany Europeans implies that they believe inmagic. It is simply a matter of believing inthe good effect, or possible good effect, of somesubstance without knowing how this effect isachieved.There is an element in the Korekore under-standing of sickness and death which involvesthe explanation of physical events in terms ofphysical or natural causes. The situation be-comes more complex when we move into therealm of personal causes.PERSONAL CAUSESThe first point to establish in a discussion ofpersonal causes is that even in Western societypersonal causality is conceptually distinct fromphysical causality.* Compare the statement,'My finger is moving because electrical impulsesare passing down the nerve cells and contract-ing the muscles,' with 'My finger is movingbecause I so wish' or 'I am moving my finger'.The two explanations are not mutually ex-clusive, but they refer to distinct types ofcauses. In one case the cause is consideredas a chain of prior physical events; in the othercase the cause is a conscious person controllingthe effect without reference to a chain throughtime. Physical causality, or the relation be-tween physical cause and effect is diachronic:personal causality, the relationship betweenpersonal cause and physical effect, is conceivedas synronic or a-chronic and operates at adifferent level.One can extend a belief in personal causalityto belief in causation by spirits. In contem-porary Western society, there are many whoaccept the physical explanations offered by*I do not here wish to enter the philosophical debateabout the verifiability of possible meanings forcausality: I am speaking of common concepts of therelationship between the cause and the effect, arelationship which in some way and to some extentinvolves determination. People believe in differenttypes of relationship, different ways in which thecause is supposed to influence the effect, and theseI call types of causality.64modern science yet still believe in spiritualcauses which control the outcome of empiricalevents, normally in accordance with the lawsof scientific causality but sometimes, possibly,interfering with them. The question we mustnow consider is whether the Korekore under-stand causation by their spirits in this way.The classic description of a similar kindof reasoning is found in Evans-Pritchard'swork on the Azande who consider witchcraftas a concomitant secondary cause ('the secondspear') working on a different plane fromthat of known causes. The author cites anexample in which witchcraft is used to ex-plain why a granary weakened by termitescollapses precisely when certain persons aresitting underneath it (1937, 69f). This isanalogous to the statement made by Korekoreinformants that it was witchcraft that made aman try to cross the flooded Mazoe river againstthe advice of his companions. Yet not all theKorekore look for secondary causes when thephysical cause of death is quite clear, and evenwhen a duality of witchcraft and physical causesis accepted the question remains whether sucha conceptual duality applies when the spiritsare believed to be involved.I shall now examine the interplay betweenthe concepts of physical and spiritual causality,exposed firstly in the reactions of the Korekoreto sickness and secondly in their beliefs aboutthe causation of sickness. I shall then compareexplanations in terms of witchcraft with ex-planations in terms of spirits.When a person first becomes ill, the onlyreaction is to seek treatment by means of tradi-tional or European medicine. If, however, thedisease becomes serious, or if a mild illnesslingers for a week or more, the sick personor his close relatives go to consult a divinerabout the cause of the trouble and the actionthey should take to remedy it.Usually divination on the cause of chronicor serious sickness points to the spirit of a deadperson. Usually it is the spirit of a grandfatheror grandmother asking for beer to be brewedin its honour, but it may be some other spirit,perhaps an alien shave spirit, and it may wantsome other rite performed, or restitution forsome grievance. When the spirit and its desiresare revealed by divination, the sick person ora close relative (always someone who is muzu-kuru Š a classification/ grandchild or sister'schild Š to a family spirit) formally addressesthe spirit stating the result of the divinationand saying that the persons concerned are aboutto take the appropriate action. In the case ofa spirit wanting beer, the muzukuru, whilemaking his invocation holds a calabash con-taining a little grain which he then hangs upover where the sick person is to sleep. Providedthere is some sign of improvement in the morn-ing, the brewing process commences. When thebeer is ready, further rituals involve offeringsto the spirit at the beginning and end of theproceedings and on each occasion the muzukuniinvokes the spirit demanding that it ceasetroubling its descendants. The sick personshould then recover completely if he had notdone so before.Even after the spiritual cause has been re-vealed, most people continue to treat the illnesswith medicines in order to help the spirits toheal the patient: the spiritual cause must beappeased and the physical symptoms must betreated. In such a case, the reactions to illnesssuggest a duality between the spiritual and thephysical worlds working concomitantly.Although common, this dual treatment isby no means universal: many cease all medicaltreatment once they are sure that they knowthe spirit causing the trouble and that theycan appease it. Occasionally people treata mild chronic illness with medicine alonewithout recourse to the spirits. One tradi-tional healer affirmed emphatically that peopleshould either appease the spirit causing theillness or, if there is no spirit, use medicaltreatment (traditional or European), but nevertake both courses of action together; andthough uncommon, this view is not unique. Soin their treatment of sickness most, but notall, Korekore show a belief in dual causationinvolving both physical medicines and con-trolling spirits.Even when dual treatment is used, beliefin a duality of spiritual and physical causes ofsickness does not necessarily follow. In fact,troubles such as venereal disease and pro-tracted labour, the physical causes of whichare believed to be known (in the latter case,illicit sexual intercourse), are not believed tohave any spiritual causes. On the other hand,a case of pneumonia believed to be caused65by a spirit is said to have nothing to do withthe patient having caught a chill while out inthe rain. The Eastern Korekore rarely if everconsider an illness to be caused both by spiritsand by some purely physical phenomenon: it iseither one or the other.*A duality of supposed causes does, however,sometimes occur involving witchcraft and aphysical cause (which we have already men-tioned) or witchcraft and a spiritual cause,either the witch using the spirit to injure thevictim or a displeased spirit giving a witchaccess to the victim. In one case the death ofa child was divined to be due to a spirit ofits mother's family and the ensuing divorcecase of two three-hour sessions ignored anydistinction between the suspected action ofthe spirit and the mother's suspected witchcraft.This does not, however, imply that causationby witchcraft and causation by the spirits areconceptually the same.There is a similarity in that both witchesand the spirits are believed to be consciousagents capable of causing serious illness andboth are frightening when angered. Witchcraftnevertheless differs considerably from theaction of spirits in that the former is essentiallyevil in design and is normally a deliberateaction on the part of a witch, neither of thesepoints applying to the spirits.Witchcraft is always evil in intentionwhereas the action of the spirit is usuallyfriendly. It is true that the Korekore do believein evil spirits which must be driven away Šthe evil spirit in the blood of the young manin the case cited is an example Š but theseare rare and are not generally associated withany particular dead person. It is said thatoccasionally a person who died with a grievancebecomes active as a deadly avenging spirit. Onesuch spirit of a man murdered some eighty*In this paper, I omit any discussion of causation byDedza, or Mwari, the remote high god. In caseswhere there is believed to be no witch or spirit ofa dead person causing illness or death (especially inthe case of the death of a very aged person), in-formants sometimes remark, 'IDedza basi' [It is Godalone]. Generally, people speak little about the highgod and then only in the vaguest terms. It appearsthat they think of him as remotely permitting whatcomes to pass and only very rarely taking any positiveaction: 'You know Dedza when the lightning strikes'.years ago is believed to have caused the deathof the two brothers who plotted his murder,their father and six of their children. It stillclaims the occasional victim among their des-cendants, while the surviving relatives are stilltrying to raise the fine in cattle necessary toappease the spirit. But such spirits becomequite amenable when once appeased, as op-posed to witches who are always dangerous.Generally, witchcraft is associated with deathand the spirits are associated with illnessesthat can easily be cured.The fact that spirits normally make theirpresence felt through causing illness does notnecessarily imply that their influence is regardedas evil. Although some people regard suchrequests by spirits as a nuisance, it is generallysaid that if a spirit is simply asking for beeror some other rite it is a good spirit. It issignificant that the spirits of deceased parentsnever ask for for beer from their children. Themost common requests come from deceasedgrandparents who are regarded with less res-pectful fear and more affection. The spirit isonly regarded as evil when the illness persistsafter the beer has been brewed, and shouldthe illness become serious, especially if it resultsin death, the accompanying work of a witch isnormally assumed. The most common form ofcausation of illness by a spirit is a friendlyspirit requesting beer from its descendants.A more fundamental difference between theactions of the spirits and those of witches liesin the manner in which they are supposed tocause illness. We have remarked the the Kore-kore concept of witchcraft normally involvesdeliberate intention and the use of physicaltechniques. Sickness caused by a spirit, on theother hand, is the concomitant physical pheno-menon of the spirit's displeasure, anger, desire,for beer, etc. I shall illustrate this point witha case in which illness in a small girl was inter-preted as a request for beer by her living pater-nal grandfather. The event was regarded by allas being very unusual, but the fact that thedifferent parties concurred independently intheir explanations of the event makes the oc-currence informative with respect to the presentdiscussion.The sick child's father went to consult adiviner about the illness and was told that itwas caused by his father-in-law who wanted66Hbeer to be brewed for him. Since his father-in-law was alive and relatively young, the child'sfather was not satisfied with this surprisingresult and consulted a second diviner who cameto the same conclusion. The family then ap-proached the father-in-law and the sick childformally addressed her grandfather, telling himthe result of the divination and saying thatthey were about to brew beer for him Š allaccording to the procedures followed when aspirit asks for beer through illness. At firstthe father-in-law refused to have anything todo with it saying that it was nonsense that aliving person should ask for beer in this way.Under the persuasion of the son-in-law's family,and no doubt influenced by the fact that thecondition of the child improved after the in-vocation, he eventually agreed to attend thebeer drink and admitted that he would liketo be honoured in this way (an honour thata man can occasionally expect from hisdaughter's husband). When the day for drinkingthe beer arrived, friends and relatives gatheredand the father-in-law went into a hut withclose relatives of the family where the firstpot of beer was served. The grandchild whohad been ill drew the first calabash, sipped alittle and gave it to her grandfather saying thatthis was the beer he had asked for. He acceptedresponsibility for the illness by taking the beerand demanding that his agnatic relatives beserved first. Although the rites for honouringa living person with beer were followed andthe rituals of setting aside beer for the spirits-were omitted, there was clearly a close con-ceptual association between this case of a livingman asking for beer in the same way. It shouldbe noticed that the son-in-law and his familywere on good terms with the father-in-law andthat neither he nor his child had failed in anyof their duties towards their senior. It shouldalso be noticed that there was no question ofthe father-in-law having done or said anythingto make the child sick and any suggestion ofwitchcraft was emphatically denied by all.**Months later, people were inclined to make remarksabout witchcraft when reminded of the event, itbeing considered that a very old man can ask for beerlike the spirits but that the father-in-law concernedwas too young. Although they claimed that the eventwas very unusual, they do admit that it was possiblewithout the use of witchcraft and this admission issufficient for the purposes of analysing the conceptsinvolved.The child was sick simply on account of thegrandfather's desire to have beer brewed forhim.This type of causality is further illustratedby a case of a woman very ill with pneumonia.Relatives presumed that the illness had beencaused by her brother who had had angrywords with her at a meeting shortly before thesymptoms appeared. Again there was no sug-gestion of witchcraft or sorcery. The illness wasregarded as the physical concomitant of thebrother's anger.During disputes at the chief's court it isoften said that quarrelling brings sickness andI once heard a remark that quarrelling is worsethan witchcraft. Clearly we have here an ideaof causing sickness by a means other than bywitchcraft. Bewitchment arises from a deliberateaction, or at least intention, on the part of thewitch to cause some evil effect, an action whichcan be suspended by the witch and possiblythwarted by the victim. In the cases just cited,the illness is believed to be the necessary physi-cal concomitant of the current social relations.EXPLANATORY ROLE OF SOCIETYCompare the statement, 'The compassneedle is pointing north because that is thedirection of the magnetic field in which it islying', with 'A is sick because B is angry withhim', or 'C is sick because his late grandfatherwants beer.' In the latter statements it is truethat the causal explanation normally precedesthe effect in time; nevertheless, the grudge ordesire must be operative at the time of sicknessfor the explanation to be valid. The suggestionis that for the Korekore the complex of inter-personal relationships, or the social structure,to some extent takes the place of physicalstructures in the understanding of events. Thiswould result in a certain difference in under-standing even where the Korekore conceptsappear at first sight to be similar to ours, andwe now return to examine in this light theexamples cited earlier in this paper.Comparing the Korekore concept of witch-craft with the western concept of crime, wenotice that the latter is considered valid onlywithin a physical spacio-temporal framework:a good alibi exonerates a suspect. The witch,on the other hand, is supposed to be able to67work wherever her (or his) social presence isfelt. The sekuru's sister was jealous of theyoung man wherever he was and could con-sequently kill him while he was still at themission hospital some thirty miles from herhome. The fact that witches are believed some-times to send familiars to achieve their endsdoes not affect this argument since they havecontrol over their familiars just as far as theirsocial influence is felt. It is significant thatphenomena such as mental telepathy, readilyacceptable to more primitive peoples, are pro-blematic to a more scientific mind.The means by which witches obtain theirpower is also understood in terms of the socialstructure. One reputed witch is believed tohave strengthened his medicine hom bypoisoning his daughter-in-law, in keeping withthe common belief that a person can becomestrong through the performance of evil. Oneinformant explained the incest myth, in whicha powerful ancestor of the tribe is said to havegained his position through sexual intercoursewith his sister, by pointing out that the ancestorin question did what others were afraid to do.Because he does what others are afraid to dosocially it is presumed that he can do whatothers are afraid to do physically, and the sameapplies to the witch.In this context, it is significant that themost powerful medicines are considered to bethe secret concoctions of specialists and themost powerful charms a man may have arekept secret from the general public. Althoughmany informants state that the motive for suchsecrecy is simply economic (you cannot sell amedicine if everybody knows how to make it),the power of the medicine or charm is, never-theless, judged with reference to the relativeknowledge of the person who makes or ownsit. If one man knows what others do not, heis presumed to have power that others have not.The converse is also true: if a man can dosomething extraordinary, it is presumed thathe knows of some charm which he is keepingsecret from the community at large. This iswell illustrated by one man who had remark-able success at hunting small buck. Althoughhe openly displayed the whistle with whichhe could summon them and readily explainedhis tactics with respect to weather and windconditions, he was unable to persuade hisfriends that he had no secret medicine to bringhim success. Their focus was not on the physicalreasons for his success but on the fact that hewas conspicuous in the realm of hunting.We come now to the fear of the infectionof death. If this fear is understood with refer-ence to social rather than physical structures,people leave a place at which a number ofdeaths have occurred among their numbers notso much because such a place is regarded asunhealthy in the sense of the physical presenceof a dangerous influence but because it involvessocial contact with some witch. By movinga few miles, they are able to remain in contactwith the relatives whom they trust while re-moving the danger of contact with the witchwho may remain unknown to them.The funeral rites show more clearly therelationship between social phenomena andthe danger of being infected by death. It isrelevant that the implements used for diggingthe grave have to be carefully washed beforebeing carried back to the homestead althoughthey come nowhere near the corpse: a graveis conceptually a place of the dead whether ornot it has yet come into contact with a corpse.At the homestead, the poles left over frommaking the bier*, ashes from the fires at whichmourners sat through the night and certainpersonal possessions of the deceased are col-lected by the ritual burial friend and depositedin the veld some distance from the homesteadalthough none of these come anywhere nearthe corpse. It is a social rather than physicalassociation with death which makes objectssupposedly dangerous.Further, the danger of being infected bydeath varies with a person's position in thesocial structure; and this applies both to thedeceased and to the survivors. The death ofan influential person is more dangerous thanof a common man, and a relative of the de-ceased is more endangered than one who hasno ties of kinship with him. Thus all whoattend any ceremony at the grave of a deceasedchief must wash themselves all over on theirway back to the villages in order to avert*A stretcher like a bier is always associated with death:if a sick person must be moved, he is carried onsomeone's back Š never on a stretcher.68subsequent illness, whereas non-functionariesreturning from the grave of any other adultneed wash only their arms, legs and faces,and many do not bother to wash at all whenreturning from the grave of a small child. Atdeath, only the burial friends, who are notkinsmen of the deceased, may attend to thecorpse and place it in the grave. Distant rela-tives may occasionally help to carry the bierbut close relatives may not have anything todo with the corpse,* and should any touch ithe must take medicines to keep away illness.Also of interest with reference to the questionof infection by death is the belief that shoulda burial friend visit a very sick man (a socialevent), death will shortly follow (the physicalconcomitant of the structure of the social event).The Korekore concept of infectious diseasesalso requires some modification in that thesocial rather than physical aspect of contactis considered to be the cause of infection. Thisis not readily verifiable where (as in the spreadof measles or venereal diseases) the social andthe physical are inseparable. It is, however,significant that the effects of measles are be-lieved to depend on social conditions Š it isbelieved that a child will die of measles if,and only if, the mother has illicit sexual inter-course. Just as the social status of the motheraffects the social status of the child, so theaction of the mother is believed to affect thephysical state of the child. This is further illus-trated by the belief that illicit sexual inter-course on the part of a pregnant woman willresult in a protracted labour. Although thismight be understood in terms of simple physicalcausality coupled with technical ignorance, thefact that the effect of a protracted labour isnot considered to be caused by intercourse*Cf. also Holleman (1953, 21-26). It should bepointed out that rituals of the Eastern Korekorediffer in detail from those described by Holleman.Also the role of the vakwasha in the funerary rites ofthe Eastern Korekore is based on their status asproviders of bride-price cattle rather than on thefact that affines are in less danger of being infectedby death.with the husband suggests that the physical evilis understood in terms of what is socially evil.This is radically distinct from modernpsychological explanations of bodily mal-functioning. Integral to these are the attitudesof the patient which can relate the physicalstate to the moral or social order and resultantworries and anxieties. When, on the other hand,the Korekore regard physical disorder as arisingfrom moral or social disorder, the attitudes ofthe individuals are not considered to be re-levant: the physical is the direct concomitantof the social. Although their conclusions mayoccasionally be in accordance with those ofmodern psychology, their reasoning is different.We conclude this section with a brief remarkabout spirits. At the death of a person, his in-fluence continues to be felt among his associatesand relatives, particularly among his descend-ants. The deceased maintains his position inthe kinship structure of the society he has left.Because they remain part of the social structure,the dead continue to be a principle of explana-tion of physical events, particularly amongtheir descendants who are most closely linkedto them in the social structure. The angryspirit of one who has been wronged is danger-ous, while the spirit of one who is rememberedwith affection is considered good and helpful.The foreign shave spirits are sometimes con-sidered helpful and sometimes simply anuisance.CONCLUSIONMy conclusion is a confirmation of Horton'sobservation that for traditional African peoplessociety is the paradigm of order and regularity.Further, the current relations between membersof society are believed directly to affect, andhence to explain, physical events of significanceto members of that society. This is not to saythat in traditional African society, physicalcauses are unknown or ignored; rather theseare considered trivial and often irrelevant incomparison with social relations between mem-bers of a close-knit community who are inter-dependent for all that is necessary in life, fromeconomic survival to relaxing entertainment.69V -IGENEALOGY CONNECTING PERSONS MENTIONED IN THE CASEHISTORY.late5 sekuru9 10sekuru witchAA15II13 14 15 16 17 wm1819 20 21 22 23- deceased personsACKNOWLEDGEMENTI wish to acknowledge helpful comments from members of the Department of Sociology ofthe University of Rhodesia when I presented an earlier draft of this paper at a research seminar, andthe criticisms by Dr A. O'Hear of Hull University which occasioned considerable revisions. Theopinions expressed remain the responsibility of the author alone.REFERENCESDURKEHIM, E. and MAUSS, M. 1901-2 'De quelques formes primitives de classification', L'Annee Sociologique,76, 1-72 (translated by R. Needham as Primitive Classification, London, Cohen & West, 1963).EVANS-PRITCHARD, E. E. 1937 Witchcraft, Oracles and Magic among the Azande, Oxford, Clarendon Press.HOLLEMAN, J. F. 1953 Accommodating the Spirit amongst Some North-Eastern Shona Tribes, Cape Town,Oxford Univ. Press, Rhodes-Livingstone Papers No 22.HAMMOND-TOOKE, W. D. 1970 'Urbanization and the interpretation of misfortune: A quantitative analysis',Africa, 40, 25-39.HORTON, R. 1964 'Ritual Man in Africa', Africa, 34, 85-104.1987 'African traditional thought and western science', Africa, 37, 50-71, 155-87.MITCHELL, J. C. 1965 'The meaning of misfortune for urban Africans', in African Systems of Thought, ed. M.Fortes and G. Dieterlen, London, Oxford Univ. Press.70