675143

8/13/2019 675143

1/11

Wiley American Anthropological Association http://www.jstor.org/stable/675143.

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

.

WileyandAmerican Anthropological Associationare collaborating with JSTOR to digitize, preserve and

extend access toAmerican Anthropologist.

http://www.jstor.org

This content downloaded from 38.114.81.227 on Wed, 9 Oct 2013 23:08:22 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=blackhttp://www.jstor.org/action/showPublisher?publisherCode=anthrohttp://www.jstor.org/stable/675143?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/675143?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=anthrohttp://www.jstor.org/action/showPublisher?publisherCode=black

8/13/2019 675143

2/11

DiseaseEtiologiesin Non-WesternMedicalSystems

GEORGEM. FOSTERUniversity of California, Berkeley

This paper argues that disease etiology is the key to cross-cultural comparison ofnon-Western medical systems. Two principal etiologies are identified: personalisticand naturalistic. Correlated with personalistic etiologies are the belief that allmisfortune, disease included, is explained in the same way; illness, religion, andmagic are inseparable; the most powerful curers have supernatural and magicalpowers, and their primary role is diagnostic. Correlated with naturalistic etiologiesare the belief that disease causality has nothing to do with other misfortunes;religion and magic are largely unrelated to illness; the principal curers lacksupernatural or magical powers, and their primary role is therapeutic. [disease,religion, and magic; ethnomedicine, medical anthropology, non-Westernmedicalsystems,shamans]

IMPRESSIVE n ethnographicaccounts of non-Westernmedicine is the tendency ofauthorsto generalize rom the particularsof the system(s) within which they have worked.Subconsciously,at least, anthropologists ilter the dataof all exotic systems throughthe lensof belief and practiceof the people they know best. Whether t be causality,diagnosis, henature and role of the curer,or the perceptionof illness within the widersupernatural ndsocial universe, general statements seem strongly influenced by the writers' personalexperiences.Glick,for example,in one of the most interestingof recentgeneralessays,notesthat in many culturesreligionand medicalpracticesare almostinseparable, ndhe addsthatWe must think about how and where 'medicine' fits into 'religion'.... In anethnographyof a religioussystem, where does the descriptionof the medicalsystem belong;and howdoes it relateto the remainder? Glick 1967:33).

Yet in many medicalsystems, as, for example, those characterizingmestizo villagersandurbanites n Latin America,medicinewould have the most minimalrole in an ethnographyof religiousbeliefs and practices. Illness and curing are dealt with largely in nonreligiousterms. In Tzintzuntzan, for example, in many hours of recording deas about originsandcures of illness, not once has religionbeen mentioned-even thoughmost villagers, f asked,wouldcertainlyagreethat illnessultimatelycomes fromGod.

The ethnologist analyzingmedical beliefs and practices in an Africancommunity canscarcelyavoiddealingwith witchcraft,oracles,magic,divining,andpropitiation,all of whicharecategoriesof only modestconcern to the studentof IndianAyurvedicmedicine. In short,there has been all too little dialoguebetweenanthropologistswho have studieddramaticallydifferent non-Westernmedical systems. So striking s the parochialismat times that one istempted to agreewith the medicalsociologist Freidsonwho notes the existence of a verylarge body of sociological and anthropological nformation about popularknowledgeofand attitudes toward health and disease, but finds most of it to be grosslydescriptive.Aside from culturaldesignationsike Mexican,Subanun,andMashona, he writes, there isno method by which the material s orderedsavefor focusingon knowledgeaboutparticularillnesses. Such studies are essentiallycatalogues,often without a classified ndex (Freidson1970:10).773

This content downloaded from 38.114.81.227 on Wed, 9 Oct 2013 23:08:22 PM

All use subject to JSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsp

8/13/2019 675143

3/11

774 AMERICAN ANTHROPOLOGIST [78,1976Yet if we cansuccessfullyclassify kinship,politicaland economic systems,andwitchcraftand sorcerybeliefs, and find the significantbehavioralcorrelatesassociatedwith each, thencertainly we can do the same with medicalsystems. Weare, after all, dealingwith limitedpossibilities in each of these cases. In this paper I am concerned with the cross-culturalpatterningthat underliesnon-Westernmedicalsystems,and with identifyingandexplicatingthe primary independent variable-disease etiology-around which orbit such dependentvariablesas types of curers,the nature of diagnosis,the roles of religionandmagic,andthelike. This is, then, an essay on comparativeethnomedicine,a term Hughesaptly defines asthose beliefs and practicesrelatingto diseasewhicharethe productsof indigenousculturaldevelopment and are not explicitly derived from the conceptual frameworkof modernmedicine (Hughes1968:99).

THE PROBLEMSOF TERMINOLOGYThroughoutmost of anthropology'sbrief history ethnologists have labeledthe institu-

tions of the peoples they have studied as primitive, peasant, or folk, dependingonthe basic societal type concerned.Untilrelativelyrecentlywe investigatedprimitivereligion,primitiveeconomics, primitiveart-and, of course, primitivemedicine. Theseminalwritingsof the ethnologist-physicianAckerknechtduring he 1940s displayno uncertaintyas to whatinterested him: it was primitivemedicine, a pair of words that appeared n the title ofnearlyevery articlehe published(Ackerknecht1971). Caudill, oo, in the firstsurveyof thenew field of medical anthropology spoke unashamedlyof primitivemedicine (Caudill1953).When,following WorldWarII, studiesof peasantcommunitiesbecamefashionable, hesepeoples were describedas possessinga folk culture. Not surprisinglyheirmedicalbeliefsand practiceswerelabeled folk medicine, a frequentsourceof confusionsince the popularmedicine of technologicallycomplexsocietiesalsooften was,andis, so described.In recent years, however, this traditionalterminologyhas come to embarrassus. In arapidly changing world, where yesterday's nonliterate villagersmay be today's cabinetministers in newly independent countries, the word primitive -initially a politeeuphemism for savage -is increasinglyoutmoded. Ackerknecht himself recognizes thischange, for in the 1971 collection of his classic essays most titles have been edited toeliminate the word primitive. Peasant and folk are less sensitivewords,but they tooare being replacedby rural, agrarian, r somethingof the kind. Theextent to which wehave been troubled by terminology is illustrated by the circumlocutionsand quotationmarks found in the major review articles of recent years: popular health culture,indigenous or folk medical roles, nonscientific health practices, native conceptualtraditions about illness, culture specific illness, the vocabularyof Westernscientificmedicine, indigenous medical systems, and the like (e.g., Polgar1962; Scotch 1963;Fabrega1972; Lieban1973).

ETIOLOGY:THE INDEPENDENTVARIABLEYet the greatestshortcomingof our traditionalmedical terminology-at least within theprofession itself-is not that it may denigrate non-Westernpeople, but rather that, byfocusingon societal types it has blindedus to the basic characteristics f the medicalsystemsthemselves.Thereis more than a grainof truth in Freidson'scomments,for manyaccountsare grossly descriptive, with lists of illnesses and treatments taking precedence overinterpretationand synthesis. So where do we startto rectify the situation?Glick(1967:36),I believe, gives us the critical lead when he writes that the most importantfact about anillness in most medicalsystemsis not the underlyingpathologicalprocessbut the underlyingcause. This is such a centralconsideration hat most diagnosesproveto be statementsabout



8/13/2019 675143

4/11

Foster] DISEASE ETIOLOGIES IN NON-WESTERN MEDICAL SYSTEMS 775causation, and most treatments, responses directed against particular causal agents(emphasisadded).A casual survey of the ethnomedical literaturetends to confirm Glick's statement. Inaccount after account we find that the kinds of curers, the mode of diagnosis,curingtechniques,preventiveacts, and the relationshipof all thesevariables o the widersociety ofwhich they area part,derivefrombeliefs about illnesscausality.It is not goingtoo far to saythat, if we are givena clear descriptionof what a people believe to be the causesof illness,we can in broad outline fill in the other elements in that medical system. It thereforelogically follows that the first task of the anthropologistconcernedwith medicalsystemsisto find the simplest taxonomy for causality beliefs. Two basic principles, which I callpersonalisticand naturalistic,seem to me to account for most (but not all) of the etiologiesthat characterizenon-Westernmedicalsystems.While he terms referspecificallyto causalityconcepts, I believe they can convenientlybe used to speak of entire systems, i.e., not onlycauses,but all of the associatedbehavior hat follows from theseviews.A personalisticmedicalsystem is one in which diseaseis explainedas due to the active,purposeful interventionof an agent, who maybe human(a witch or sorcerer),nonhuman(aghost, an ancestor,an evil spirit),or supernaturala deity or otherverypowerfulbeing).Thesick person literally is a victim, the object of aggressionor punishmentdirectedspecificallyagainsthim, for reasons hat concernhim alone. Personalistic ausalityallows little room foraccident or chance;in fact, for some peoples the statementis madeby anthropologistswhohave studied them that all illnessand death are believed to stem from the acts of the agent.Personalisticetiologies are illustrated by beliefs found among the Mano of Liberia,recordedby the physicianHarley,who practicedmedicineamongthemfor 15 years. Deathis unnatural, he writes, resulting rom the intrusionof anoutsideforce, usuallydirectedby some magicalmeans (Harley 1941:7). Similarly,among the Abron of the Ivory Coast,

Peoplesicken and die becausesome power, good or evil, hasactedagainstthem.... Abrondisease theory contains a host of agents which may be responsible for a specificcondition.... These agentscut across the naturaland supernaturalworld.Ordinarypeople,equipped with the proper technical skills, sorcerers,varioussupernatural ntities, such asghosts, bush devils,and witches, or the supreme god Nyame, actingalone or through essergods,all cause disease (Alland1964:714-715).In contrast to personalisticsystems, naturalisticsystems explain illness in impersonal,systemicterms.Disease s thoughtto stem, not fromthe machinationsof an angrybeing,butratherfrom such natural orces or conditions ascold, heat, winds,dampness,and, aboveall,by an upset in the balance of the basic body elements. In naturalisticsystems, healthconforms to an equilibriummodel: when the humors,the yin and yang, or the Ayurvedicdosha are in the balanceappropriate o the ageand conditionof the individual, n his naturaland social environment,health results.Causalityconcepts explain or accountfor the upsetsin this balancethat trigger llness.Contemporarynaturalistic ystems resemble each other in an importanthistoricalsense:the bulk of their explanationsand practices represent simplified and popularized egaciesfrom the great tradition medicine of ancient classicalcivilizations,particularly hose ofGreece and Rome, India, and China. Although equilibrium s expressed in many ways inclassical accounts, contemporary descriptions most frequently deal with the hot-colddichotomy which explains illness as due to excessive heat or cold entering the body.Treatment, ogically, attemptsto restore the properbalancethrough hot and cold foodsand herbs,and other treatmentssuch as poultices that are thoughtto withdrawexcess heator cold from the body.In suggesting that most non-Westernetiologies can be described as personalisticornaturalistic I am, of course, painting with a broad brush. Every anthropologist willimmediately think of examples from his research that appear not to conform to thisclassification.Mosttroublesome,at least at firstglance,arethose illnessesbelievedcausedby



8/13/2019 675143

5/11

776 AMERICAN ANTHROPOLOGIST [78,1976emotional disturbances uchasfright,jealousy, envy, shame,anger,or grief.Fright,orsusto,widespread n Latin America,can be causedby a ghost, a spirit, or an encounter with thedevil; if the agent intended harmto the victim, the etiology is certainlypersonalistic.Butoften accounts of such encounterssuggestchance or accident ratherthanpurposiveaction.And, whenan individual lipsbeside a stream,and fearshe is about to fall into the wateranddrown,the etiology is certainlynaturalistic.The Latin American muina, an indisposition resulting from anger, may reflect adisagreeable nterpersonal pisode,but it is unlikelythat the event wasstagedby an evil doerto bring illness to a victim. In Mexico and Central Americathe knee child's envy andresentment of its new sibling-to-be,still in the mother's womb, gives rise to chipil, thesymptoms of which are apathy, whining, and a desire to cling to the mother'sskirt. Thefoetus can be said, in a narrowsense, to be the causeof the illness,but it is certainlynot anactive agent, nor is it blamed for the result. Since in a majorityof emotionally explainedillnessesit is hardto identify purposiveaction on the part of an agent intent upon causingsickness, I am inclined to view emotional etiologies as more nearly conformingto thenaturalistic han to the personalisticprinciple.Obviously,a dual taxonomy for phenomenaas complex as worldwidebeliefs about causesof illness leavesmanyloose ends. But it mustbe rememberedthat a taxonomy is not an end in itself, something to be polished andadmired;ts value ies rather n the understanding f relationshipsbetweenapparentlydiversephenomena that it makes possible. I hope that the following pageswill illustratehow thepersonalistic-naturalisticlassification,for all its loose ends, throws into sharp perspectivecorrelations in health institutions and health behavior that tend to be overlooked indescriptiveaccounts.Before proceeding, a word of caution is necessary: the two etiologies are rarely if evermutually exclusive as far as their presenceor absencein a particular ociety is concerned.Peoples who invoke personalisticcauses to explain most illness usually recognizesomenatural,or chance, causes. And peoples for whom naturalisticcausespredominatealmostinvariablyexplainsome illness as due to witchcraftor the evil eye. But in spite of obviousoverlapping, he literaturesuggeststhat many, if not most, peoplesarecommittedto one orthe other of these explanatory principlesto account for a majority of illness. When,forexample, we read that in the Venezuelanpeasant villageof El Morro89%of a sampleofreported illnesses are natural n origin, whereasonly 11%are attributed to magicalorsupernatural auses (Suarez1974), it seems reasonable o say that the indigenouscausationsystem of this groupis naturalisticand not personalistic.And, in contrast,whenwe readofthe MelanesianDobuans hat all illnessanddiseaseare attributedto envy, andthat Deathiscaused by witchcraft, sorcery, poisoning, suicide, or actual assault (Fortune 1932:135,150), it is clearthat personalistic ausalitypredominates.Althoughin the presentcontext I am not concernedwith problemsof evolution,I believethe personalisticetiology is the more ancient of the two. At the dawn of humanhistoryitseems highly likely that all illness, as well as other forms of misfortune,was explainedinpersonalistic erms. I see man'sability to depersonalizecausality,in all spheresof thought,including llness,as a majorstep forward n the evolution of culture.

ETIOLOGIES: OMPREHENSIVEND RESTRICTEDWe now turn to the principal dependent variables n medical institutions and healthbehaviorthat correlate with personalisticand naturalisticetiologies. The first thingwe noteis that personalistic medical etiologies are parts of more comprehensive,or general,explanatory systems, while naturalisticetiologies are largely restrictedto illness. In otherwords, in personalistic systems illness is but a special case in the explanation of allmisfortune.Some societies, to quote Horton(1967) haveadopteda personal diom as thebasis of their attempt to understand the world, to account for almost everythingthat



8/13/2019 675143

6/11

Foster] DISEASE ETIOLOGIES IN NON-WESTERN MEDICAL SYSTEMS 777happensin the world, only incidentallyincludingillness. In such societies the samedeities,ghosts,witches,and sorcerers hat send illnessmay blightcrops,cause financialreverses, ourhusband-wife relationships, and produce all manner of other misfortune. To illustrate,Price-Williamstates Thegeneral eatureof illnessamongthe Tiv is that it is interpreted n aframeworkof witchcraft and malevolent forces (1962:123). In common with a greatmany other people, Tiv do not regard illness'or 'disease'asa completely separatecategorydistinct from misfortunesto compoundandfarm,fromrelationshipsbetweenkin, andfromcomplicatedmattersrelating o the controlof land (1962:125).Similarly, the Kaguruof Taznazia believemost misfortunes,howeversmall, are due towitchcraft. Most illness, death, miscarriages,terility,difficult childbirths,poorcrops,sicklylivestock and poultry, loss of articles,bad luck in hunting, and sometimesevenlack of rainarecausedby witches (Beidelman1963:63-64).In contrast,naturalisticetiologies are restrictedto disease as such. Althougha systemicidiom may prevail to account for much of what happens in the world, a humoralor ayin-yangimbalancewhich explainsan illness is not invokedto explaincropfailure,disputesover land, or kin quarreling. n fact, the strikingthing is that while in naturalistic ystemsdiseaseetiologiesarediseasespecific,other areasof misfortune,such as personalquarrels re,not surprisingly,explained in personalistic erms.In Tzintzuntzan,for example, misunder-standings between friends may be due to natural-born rouble makers, who delight inspreadingrumorsand falsehoods.Financialreverses, oo, may be accounted for by badluck,or dishonesty and deceit on the part of false friends. But these explanationsare quitedivorced romillnessetiology, which has its own framework, xclusiveto it.

DISEASE,RELIGION,ANDMAGIC

WhenGlick (1967:32) writes that it is commonknowledgethat in many cultures, deasand practices relating to illness are for the most part inseparablefrom the domain ofreligious beliefs and practices, he is speaking only of those systems with personalisticetiologies. Jansen (1973:34) makes this clear in writing of the Bomvana (Xhosa) thatreligion,medicine and magic are closely interwoven,... being partsof a complex wholewhich finds its religiousdestinationin the well-beingof the tribe.... The Bomvanahimselfdoes not distinguishbetweenhis religion,magicandmedicine. Whencurersaredescribedaspriests and priestesses, as is often the case in Africa (e.g., Warren1974-75:27), we areclearlyin the domain of religion.In contrast, in naturalisticsystems religion and magicplay only the most limited rolesinsofar as we are dealing with etiology, and to the extent that religious rituals are found,they are significantlydifferent in form and concept from religiousritualsin personalisticsystems. For example, in those Latin Americansocieties whose etiological systems arelargely naturalistic, victims of illness sometimes place votive offerings on or nearmiraculous mages of Christ,the VirginMary,or powerful saints,or light votive candlesfor these supernaturalbeings, asking for help. These are certainly religiousacts. But it isimportant to note that in personalistic systems the beings supplicated, and to whompropitiatory offerings are made, are themselves held responsible for the illness. It is toappeasetheir angeror ill will that such offeringsaremade. In contrast,in Catholiccountriesthe beings to whom prayersare raisedand offeringsmade are not viewed as causes of theillness. They are seen as mercifuladvocateswho, if moved, can interveneto help a humansufferer. It should be noted, too, that most of these acts conform to a generalpatterninwhich aid of supernaturalss soughtfor any kindof misfortune,such as financialreversesorthe releaseof a son fromjail, aswell as illness or accident.Thus, there is a significantcontrast in structureand style between the two systems. Insocieties wherepersonalisticetiologies predominate,all causality is generalandcomprehen-sive, andnot specific to illness;but paradoxically,whenritualsupplicationsandsacrificesare



8/13/2019 675143

7/11

778 AMERICAN ANTHROPOLOGIST [78,1976made, usually they are narrowly limited in scope, specific to a particular llness, or toprevent feared illness. In contrast, in societies where naturalisticetiologies predominate,illness causalityis specific to illnessalone, and does not applyto otherkinds of misfortune.But, insofaras religionis a part of curing,it is comprehensive r general,conforming o thesamepatternsthat characterize pious personin the face of anymisfortune.

LEVELSOF CAUSALITYPersonalisticand naturalisticetiologies further differ importantly n that, for the former,it is necessaryto postulate at least two levels of causality:the deity, ghost,witch, or otherbeing on whomultimateresponsibility or illnessrests,and the instrumentor techniqueusedby this being, such as intrusion of a disease object, theft of the soul, possession, orwitchcraft. In the literatureon ethnomedicinethe first level-the being-is often referred oas the efficient cause, while the second level-the instrumentor technique-is referred o asthe instrumental, or immediate cause. A few anthropologistsrecognize three levels of

causation. Goody (1962:209-210), for example, describes both efficient and immediatecauses, to which he adds a final cause, an ancestor or earth shrine that withdraws itsprotection from a personso that he falls victim to a sorcerer.In HondurasPeck(1968:78)recognizesessentiallythe same threelevels:an instrumental ause( i.e., whathas been doneto the patient, or what is used ), an efficient cause ( i.e., who or what has done it to thepatient ), and a final, or ultimate, cause ( i.e., an attemptto answerthe question, 'why didthishappento me at this time?' ).Naturalistic etiologies differ significantly in that levels of causation are much lessapparent; n most cases they tend to be collapsed. Althoughit can be arguedthat a personwho willfully or through carelessnessengages in activities known to upset his bodilyequilibrium s the efficient cause of his illness, in practicethis line of argumenthas littleanalyticalvalue.It was failure to recognize levels of causality that limited the value of Clements'pioneeringstudy of diseaseetiology (1932), a defect first pointed out by Hallowell(1935).This distinction, as we are about to see, is criticalto an understanding f basicdifferences ncuringstrategies ound in the two systems.SHAMANSAND OTHERCURERS

The kinds of curers found in a particularsociety, and the curing acts in which theyengage, stem logically from the etiologies that are recognized.Personalistic ystems, withmultiple levels of causation,logically requirecurers with supernatural nd/ormagicalskills,for the primaryconcern of the patient and his family is not the immediatecauseof illness,but rather Who? and Why? Among the Bomvana Xhosa)Jansen(1973:39) puts it thisway: They are less interested to know: How did it happen? rather than: Who isresponsible? Similarly,in Mali we read that In generalthe Bambarawant to know whythey are ill and not how they got ill (Imperato 1974-75:44). And in the Indianvillagestudied by Dube the Brahminor a local seer is essentialto find out what ancestorspirit isangry,andwhy (Dube 1955:128).The shaman,with his supernaturalpowers, and directcontact with the spiritworld,andthe witch doctor (to use an outmodedterm from the African iterature),with his magicalpowers, both of whom are primarilyconcerned with finding out who, and why, are thelogical responses n personalistic,multiple causality,etiological systems. After the who andwhy have been determined,treatmentfor the immediatecause may be administeredby thesame person, or the task may be turnedover to a lessercurer,perhapsan herbalist.Thus,among the Nyima of the Kordofanmountainsin the Sudan,the shamangoes into a tranceand discoversthe causeandcureof the disease. But he himselfperformsno therapeuticacts;



8/13/2019 675143

8/11

Foster] DISEASE ETIOLOGIES IN NON-WESTERN MEDICAL SYSTEMS 779this is the field of other healing experts, to whom the patient will be referred(Nadel1946:26).Naturalisticetiological systems, with single levels of causation, logically requirea verydifferent type of curer,a doctor in the full senseof the word,a specialist n symptomatictreatment who knows the appropriateherbs,food restrictions,andother forms of treatmentsuch as cupping,massage,poultices, enemas,and the like. The curanderoor the Ayurvedicspecialist is not primarilyconcerned with the who or why, for he and the patient bothusuallyare in completeagreementas to what hashappened.

DIAGNOSISPersonalisticand naturalisticetiological systemsdividealongstill anotheraxis, the natureof diagnosis. In personalisticsystems, as we have just seen, the shamanor witch doctordiagnosesby means of trance, or other divinatorytechniques. Diagnosis-to find out whoand why-is the primaryskill that the patient seeks from his curer. Treatment of the

instrumental ause,while important, s of secondaryconcern.In contrast, in naturalisticsystems diagnosisis of very minor importance,as far as thecurer is concerned. Diagnosis usually is made, not by the curer, but by the patient ormembersof his family. When the patient ceases treatmentwith home remediesand turnstoa professional,he believeshe knows what afflicts hitn. His primaryconcernis treatment ocure him. And how is diagnosisdone by the layman?The answeris simple, pointed outmany years ago by Erasmus 1952:414), specificallyfor Ecuador.Whenan individualwhosediseaseetiology is largelynaturalistic eels unwell,he thinksbackto an earlierexperience, nthe night, the day before, or even a month or ayearearlier, o an event that transpired, r asituation in which he found himself, that is known to cause illness.Did the patientawakenin the morningwith swollen tonsils?Heremembers hat on goingto bed the night beforehecarelessly stepped on the cold tile floor of his bedroom in his bare feet. This, he knows,causes cold to enter his feet and compressthe normalheat of his body into the upperchestand head. He suffers from risenheat. He tells the doctor what is wrong,andmerelyasksfor an appropriate emedy.Does a woman suffer an attackof painfulrheumatism?She remembers hat she had beenironing,therebyheatingherhands andarms,andthat without thinkingshe hadwashed themin cold water. The cold, to her vulnerable superheated arms and hands, caused herdiscomfort. She needs no diviner or shaman to tell her what is wrong. The strikingthingabout a naturalisticsystem is that, in theory at least, the patient can, upon reflection,identify every cause of illness that may afflict him. So powerful is this patterntoday inTzintzuntzan that when people consultmedicaldoctors,theirstandardopeningstatement isDoctor, please give me something for - - - - -, whatever their diagnosismay be.Doctors,traditionalor modern,are viewedas curers,not diagnosticians.To summarize,we may say that in personalisticsystems the primaryrole of the shamanor witch doctor is diagnostic,while in naturalistic ystemsit is therapeutic.

PREVENTIVEMEASURESPreventivemedicine, insofar as it refers to individualhealth-orientedbehavior,can bethought of as a series of dos and don'ts, or shoulds and shouldn'ts. In

contemporaryAmericawe should get an annualphysicalexamination,oureyes andteethchecked regularly,and make sure our immunizationsare up to date when we travelabroad.We should not smoke cigarettes, consume alcohol to excess, breathe polluted air, orengagein a seriesof otheractivitiesknownor believedto be inimicalto health. Ourpersonalpreventivemeasuresare,perhaps,aboutequallydividedbetween the dos and don'ts.In all other societies similar shoulds and shouldn'ts can be identified.Although my



8/13/2019 675143

9/11

780 AMERICAN ANTHROPOLOGIST [78,1976grounds are highly impressionistic,I rather have the feeling that naturalisticetiologiescorrelatepredominantlywith don'ts, while personalistic tiologiescorrelatewith dos. Innaturalistic ystems a personalhealth strategyseems to consist of avoidingthose situationsor not engagingin behavior,known to produce illness. In Tzintzuntzan,and many otherLatin Americancommunities,the prudent person doesn't standon a cold floor in barefeet,doesn't wash handsafter whitewashinga wall, doesn'tgo out into the nightairimmediatelyafterusingthe eyes, anda host of other things.In theory,at least,a hypercautious ndividualshouldbe able to avoid almost all illnessby not doingcertainthings.In contrast, in personalistic ystemsthe basicpersonalhealthstrategyseemsto emphasizethe dos, and especially the need to make sure that one's social networks, with fellowhuman beings, with ancestors, and with deities, are maintained in good workingorder.Although this means avoiding those acts known to arouse resentment- don'ts -itparticularlymeans careful attention being paid to the propitiatoryritualsthat are a god'sdue, to positive demonstrationsto ancestors that they have not been forgotten, and tofriendly acts to neighbors and fellow villagersthat remind them that their good will isvalued. In short, recognizingmajoroverlapping, he primary trategies o maintainhealth inthe two systems are significantly different. Both require thought. But in one-thepersonalistic-time and money are essentialingredients n the maintenanceof health. In theother-the naturalistic-knowledgeof how the system works,and the will to live accordingto its dictates,is the essential hing;this costsverylittle, in eithertime or money.

THELOCUSOF RESPONSIBILITYWith respect to personal responsibility for falling ill, do the two etiological systemsdiffer? To some extent I think they do. In Tzintzuntzan,as pointed out, the exercise ofabsolute care in avoidingdisease-producingituationsshould, in theory, keep one healthy.Hence, illness is prima facie evidence that the patient has been guilty of lack of care.Althoughillness is as frighteningas in any other society, and family membersdo theirbestto help a sick member,thereis often an ambivalent eelingthat includesangerat the patientfor havingfallen ill. I have seen worriedgrowndaughtersosinga night'ssleepas they soughtmedical care for a mother they feared was suffering a heart attack. Whenthe motherconfessed that she had not taken her daily pill to keep her blood pressuredown (andaftershe was back to normal,the crisispast), the daughtersbecamehighly indignantandangryather for causingthem to lose sleep.In personalisticsystems people also know the kinds of behavior-sinsof commissionandomission-that may lead to retaliationby a deity, spirit, or witch. To the extent they can

lead blameless lives they should avoid sickness. But personalisticcausality is far morecomplex than naturalisticcausality,since there are no absolute rules to avoid arousing heenvy of others, for doing just the right amount of ritualto satisfyan ancestor,for knowinghow far one can shade a taboo without actuallybreachingt. Consequently, n suchsystemsone has less control overthe conditionsthat lead to illnessthan in the other, wherethe rulesare clearly stated. Spiro (1967:4) makes this contrast clear among the Burmese.Sincesuffering (includingillness) is the karmic consequenceof one's demeritsaccumulated nearlierincarnations,the responsibilityfor sufferingrests on the shoulders of the suffererhimself. But, says Spiro, to accept this responsibilityis emotionally unsatisfying.On theother hand, if one subscribes to a supernatural-magicalxplanatory system, in which allsufferingcomes from ghosts, demons,witches,andnats, in at least some casesthe sufferer sentirely blameless.He simply happensto be the victim of a witch who, frommalice,choseshim as victim. In other cases he is only inadvertently responsible-he has unwittinglyoffended or neglected a nat who, annoyed by his behavior,punisheshim. Spiro sees thisreasoningas underlyingthe juxtapositionof Buddhismandsupernaturalism-ofpersonalisticandnaturalistic tiologies-in Burma.



8/13/2019 675143

10/11

Foster] DISEASE ETIOLOGIES IN NON-WESTERN MEDICAL SYSTEMS 781SUMMARY

By way of summary the two systems of disease etiology and their correlates may betabularized as follows:System:Causation:Illness:Religion, Magic:Causality:Prevention:Responsibility:

PersonalisticActive agentSpecial case of misfortuneIntimately tied to illnessMultiple levelsPositive actionBeyond patient control

NaturalisticEquilibrium lossUnrelated to othermisfortune

Largely unrelated to illnessSingle levelAvoidanceResides with patient

REFERENCES CITEDAckerknecht, Erwin H.1971 Medicine and Ethnology: Selected Essays. Baltimore: Johns Hopkins Press.Alland, Alexander, Jr.1964 Native Therapists and Western Medical Practitioners among the Abron of the IvoryCoast. Transactions of the New York Academy of Sciences. Vol. 26. Pp. 714-725.Beidelman, T. O.1963 Witchcraft in Ukaguru. In Witchcraft and Sorcery in East Africa. J. Middleton andE. H. Winter, eds. Pp. 57-98. London: Routledge and Kegan Paul.Caudill, William1953 Applied Anthropology in Medicine. In Anthropology Today. A. L. Kroeber, ed.Pp. 771-806. Chicago: University of Chicago Press.Clements, Forrest E.1932 Primitive Concepts of Disease. University of California Publications in AmericanArchaeology and Ethnology, Vol. 32, Part 2. Pp. 185-252.Dube, S. C.1955 Indian Village. London: Routledge and Kegan Paul.Erasmus, Charles J.1952 Changing Folk Beliefs and the Relativity of Empirical Knowledge. SouthwesternJournal of Anthropology 8:411-428.Fabrega, iHoracio,Jr.1972 Medical Anthropology. In Biennial Review of Anthropology: 1971. Bernard J.Siegel, ed. Pp. 167-229. Stanford: Stanford University Press.Fortune, Reo F.1932 Sorcerers of Dobu: The Social Anthropology of the Dobu Islanders of the Western

Pacific. London: George Routledge.Freidson, Eliot1970 Professional Dominance: The Social Structure of Medical Care. New York:Atherton.Glick, Leonard B.1967 Medicine as an Ethnographic Category: The Gimi of the New Guinea Highlands.Ethnology 6:31-56.Goody, Jack1962 Death, Property and the Ancestors: A Study of the Mortuary Customs of theLodagas of West Africa. Stanford: Stanford University Press.Hallowell, A. Irving1935 Primitive Concepts of Disease. American Anthropologist 37:365-368.Harley, George Way1941 Native African Medicine: With Special Reference to Its Practice in the Mano Tribeof Liberia. Cambridge: HarvardUniversity Press.Horton, Robin1967 African Traditional Thought and Western Science. Africa 37:50-71, 155-187.Hughes, Charles C.1968 Ethnomedicine. In International Encyclopedia of the Social Sciences. Vol. 10. Pp.87-93. New York: Free Press/Macmillan.



8/13/2019 675143

11/11

782 AMERICAN ANTHROPOLOGIST [78,1976Imperato, Pascual James1974-75 Traditional Medical Practitioners among the Bambara of Mali and their Role inthe Modern Health-Care-Delivery System. In Traditional Healers: Use and Non-Use inHealth Care Delivery. I. E. Harrison and D. W. Dunlop, eds. Rural Africana 26:41-53.Jansen, G.1973 The Doctor-Patient Relationship in an African Tribal Society. Assen, TheNetherlands: Van Gorcum.Lieban, Richard W.1973 Medical Anthropology. In Handbook of Social and Cultural Anthropology. John J.Honigmann, ed. Pp. 1031-1072. Chicago: Rand McNally.Nadel, S. F.1946 A Study of Shamanism in the Nuba Mountains. Journal of the Royal Anthropolog-ical Institute 76:25-37.Peck, John G.1968 Doctor Medicine and Bush Medicine in Kaukira, Honduras. In Essays on MedicalAnthropology. Thomas Weaver, ed. Pp. 78-87. Southern Anthropological SocietyProceedings, Vol. 1.Polgar, Steven1962 Health and Human Behavior: Areas of Interest Common to the Social and MedicalSciences. Current Anthropology 3:159-205.Price-Williams,D. R.1962 A Case Study of Ideas Concerning Disease among the Tiv. Africa 32:123-131.Scotch, Norman A.1963 Medical Anthropology. In Biennial Review of Anthropology. Bernard J. Siegel, ed.Pp. 30-68. Stanford: Stanford University Press.Spiro, Melford E.1967 Burmese Supernaturalism: A Study in the Explanation and Reduction ofSuffering. Englewood Cliffs: Prentice-Hall.Suarez, Maria Matilde1974 Etiology, Hunger, and Folk Diseases in the Venezuelan Andes. Journal ofAnthropological Research 30:41-54.Warren,Dennis M.1974-75 Bono Traditional Healers. In Traditional Healers: Use and Non-Use in HealthCare Delivery. I. E. Harrison and D. W. Dunlop, eds. Rural Africana 26:25-39.Submitted 10 February 1976Accepted 22 May 1976

675143

Documents