PRIMITIVE CONCEPTS OF DISEASE BY FORREST E. CLEMENTS UNIvnsiTy or CALIFORNIA PUBLICATIONS IN AMRICAN ARCHAZOLOGY AN) ETHNOLOGY Xme 32, N0. 2} PP. 1854252, 4 maps * u . . . . . . . . , . . . . . . . . . . .. . . . . : . . , . - . . X * . * f f . . . . . . . . . . . . . , . . . . :, . .! ' . , \ . . . . , . : ; : . . ' , . . . ' . . . . . : , . . . V : , , , . . . . . , , . . . . . . , ,' - , UNIVERSITY OF CALIPORNIA PRESS BERKELEY CAtIFORN}A - : - -4I932 ABmv ~IoNBU L'An'thro'oogle Ama- AJthropologixt, . :. Amria A&u;lhogpiog,aI4ssoiation Memoi 0-s. - chiV ItropoIogie, A,eri-a > v '1i w ^~~~~~At: eo,A-X,;"b- - :' ' ' ational .:onre- .of? Amrcait ,(Qyd:" , :. ;Znternso*al ournal- of ttiriDn Zi ; - rt-:t -J :Qlwnal: -oAmrn F-olkFLore. - ;t-- .1Jounl of tie Royal A, An icl :nt ;te. Mu ws > eu of th/e 4mr ica -XeyE1ouu:a:ion-- : India N.ot<->es and Monographs.-:/-:0-: ; Peabody Iufeuni (of ard -UHist -- 1 a , rp u 0 ,ers. *;x --:Report-s. ;- ;;0 ...,Public Mnsu (o th (ity) of ilw '-Auke, Rnlti. .,':.:'.1i,t A devs , A cat, ,, Ia.rs, o, urna SmitesonianV Iongtrtutio; 'Aruaii eprt4.A 1 isvelke,u8 OMe o ' X;',tAan.d Zthnolg.., Uni:esit ''0f Pennoalaia (,esy Mtt!,.,, Anhrpo gion t i'on United ta*,oteltoia ndAWuseugmap- ep, ortt Univesity f War*uto, Publicatosi nhoooy Zdeitoris tIi mEtuoIgne 7'A '- -''- .PA . AAA .. . AP.8MP _Q.M *:AJPA,*4- BAE B,. 9 .O..AB GUj ,A FL, : ,}.; 5tttMNH. , . .M *- ' ' A - : , - I3AL ' ,-. 3AFL,$ 4.J * J , . EAZ., i_ A,s .; f -_' li ,4 !','IJ rn; PM X' JE. i,,,? ''-P PMM' . i.R ' ,AP-J ,,..-f ., ...-OK,.''.A r 2.', YXO+kv U,f../S, .?AAEM<,',,,;4t' U&,]tlxi \' '0' ' ' -.' p - ' ;-P-> tIW-PA, A"' ,., PRIMITIVE CONCEPTS OF DISEASE BY FORREST E. CLEMENTS UNIVERSITY OF' CAiEiouNiA PUBLICATiONS IN AMERICAN ARCHAEOLOGY AND ETHNOLOGY Volume 32, No. 2, pp. 185-252, 4 maps Issued February 9, 1932 UNIVERSITY OF OAIFORNI PRSS BERIELEY, CALIFORA CAMBRIDGE UNIVERSITY PRESS LONDON, ENGLAND CONTENTS PAGE Introduction ........................................ 185 Description and classification of disease concepts ........................................ 186 1. Sorcery ........................................ 186 2. Breach of taboo ........................................ 187 3. Disease-object intrusion ........................................ 188 4. Spirit intrusion ........................................ 188 5. Soul loss ........................................ 190 Discussion of concepts ............... 202 Sorcery ................... 202 Breach of taboo ................... 204 Disease-object intrusion ................... 209 Spirit intrusion ................... 216 Possession ................... 224 Soul loss ................... 225 Relative antiquity ................... 240 Summary of conclusions ................... 242 Bibliography ................... 244 TABLES 1. Distribution of the five main disease concepts over the world ........................ 193 2. Association between breach of taboo disease concept and confession method 205 3. Dental diseases due to worms in the teeth ....................................................... 212 4. Association between sucking treatment and disease-object intrusion ............ 213 5. Spirit intrusion ....................................................... 219 6. Soul loss ....................................................... 231 6. Supplement. Other ideas associated with soul loss ............................................ 232 MAPS 1. Geographical distribution of breach of taboo ....................................................... 203 2. Geographical distribution of disease-object intrusion .......................................... 210 3. Geographical distribution of spirit intrusion ....................................................... 217 4. Geographical distribution of soul loss .227 PRIMITIVE CONCEPTS OF DISEASE BY FORREST E. CLEMENTS INTRODUCTION A great deal of information on primitive theories of disease has been published but in widely scattered form. Histories of medicine usually devote a few introductory pages to the subject, while works which may be styled "prehistories" of medicine present a more com- plete but nevertheless unsatisfactory collection of the data. Of these the best studies are Die Medicin der Naturv6lker by M. Bartels, and The Infancy of Medicine by D. McKenzie, the latter being the more recent. One of the earliest attempts at ethnographic treatment of the subject is that of E. B. Tylor in his Researches into the Early History of Mankind. In this work Tylor devoted a brief space to the wide- spread notion that disease is due to the presence in the body of some foreign substance not in itself pathogenic. In view of Tylor's general theoretical position it is rather surprising to find that he was inclined to interpret the occurrences of this concept on the basis of historical connection.' Another approach to the subject by an ethnologist is that of W. H. R. Rivers in Medicine, Magic and Religion. Rivers dealt with a greater number of concepts than Tylor and drew certain conclusions as to historical relationship between various peoples, although prin- cipally with reference to medical practices as such and not to specific concepts of disease. Rivers' information on many areas, especially Asia, was very incomplete. The object of the present study is to offer a scheme of classification for the disease concepts of primitive peoples; to obtain a fairly com- plete geographical distribution of the classified concepts; and to frame certain conclusions as to their relative antiquity, probable origin, and historical connections. Such a study affords a good opportunity for 1 218:281. Citations are to numbers in terminal bibliography, followed by page reference. 186 Univer8ity of California Publications in Am. Aroh. and Ethn. [Vol.32 the application of general methodological principles, thus involving one of the most basic questions of modern theoretical anthropology- the interpretation of cultural similarities and the convertibility of distributions in space into time sequences.2 DESCRIPTION AND CLASSIFICATION OF DISEASE CONCEPTS Primitive concepts of disease are those ideas held by primitive people as to the cause or genesis of siekness. Examining the disease concepts of peoples the world over, we shall find them all similar in one respect: disease is never regarded as a normal thing but always as abnormal even though its etiology may be held natural. Aside from this similarity, by no means as psycho- logically natural as it seems, the concepts will fall roughly into three broad categories: first, natural causes, which include the modern medi- cal theory and all injuries obviously inflicted by material agencies; second, human agency, i.e., disease is considered directly due to the malefic action of some human being, embracing sorcery in all its phases; and third, supernatural agency, i.e., sickness is regarded as due to the action of supernatural factors. All three of these categories occur among both civilized and primi- tive peoples, although the first is by far the most prominent among the civilized peoples of Europe and America today. Even so, traces of the other beliefs still linger, especially among the peasantry of eastern Europe. Among primitive people and, perhaps to a lesser extent, among the nations of antiquity the latter two categories attain the greatest adherence. Closer examination of the concepts of primitive people, however, although confirming our classification, showssthat the three categories are far too broad for practical purposes. Narrowing these down with- out creating artificial classes, primitive disease concepts are of five main types, which are described below. 1. Sorcery Here are grouped all those theories which ascribe sickness either to the manipulations of persons skilled in magic or to the operations of human beings who exercise some control over the supernatural world. 2 I wish to acknowledge hitherto unpublished information and field notes kindly placed at my disposal by Dr. E. M. Loeb, Dr. A. H. Gayton, Dr. R. L. Olson, and Mr. J. M. Garvan. Clements: PrKmitive Coiweptg of Disea8e In the main, magic of this sort is of two types, which have been analyzed by Frazer. The magician who wishes to bring sickness or injury on a person may construct a small image which represents the victim. This image is then transfixed with darts, burned, or otherwise maltreated, all these operations being accompanied by suitable incan- tations. Such measures are supposed to cause the person against whom they are directed to fall ill. This type of magic is usually labeled as imitative. In the second type, the sorcerer obtains some part of his victim's body, such as hair or nail clippings, excrements, or even some article of clothing which has been in intimate contact with his body. These objects are then subjected to the proper magical procedure and the victim is soon stricken with disease. This is called contagiowts magic. In addition to these two forms of magic, however, there is the idea that individuals in direct communication with the supernatural world may cause evil spirits or demons to lodge in a person's body. Again, such persons may steal the victim's soul, thus causing him to fall ill. Witches may also magically "shoot" disease-objects into people. All these theories are included under the term sorcery; and while the term as used here obviously includes culture phenomena which might be niore finely classified, in a paper of limited scope such differ- entiation is impracticable. There is one other theory which might conceivably be classed as sorcery although not included here. This is the idea of the Evil Eye, widespread in the Old World but apparently quite foreign to Ameri- can aborigines. Since there seems to be nothing definite as to the supposed mechanism by which the Evil Eye causes diseases, it has been omitted here.8 2. Breach of Taboo All theories which explain sickness as a punishment sent by the gods for breach of religious prohibitions or social prohibitions having divine sanction, are included under this heading. The breach may be quite unintentional and even unknown to the sufferer but it is none the less regarded as the real cause of his sickness. In such cases confession often serves as a means of treatment. 3 A discussion of the Evil Eye will be found in Hastings' Encyclopaedia of Religion and Ethics, 5:608-615. 187 1932] 188 Univers-ity of Califotrnia Publwication.s n Am. Arch. and Ethn. [Vol. 32 3. Disease-Object Intrusion Under this heading are listed all theories attributing disease to the presence in the body of some malefic foreign substance. This need not be intrinsically pathogenic and usually takes the form of a bit of bone, hair, a pebble, splinter of wood, or even small animals, such as lizards, worms, and insects. In any case, the intruding substance is tangible. Among certain peoples, e.g., the Indians of northern California, most Australian tribes, and various South American aborigines, the disease-object is reported to contain, according to native belief, a spiritual essence which is the real cause of the sickness. All peoples with the disease-object concept have been said really to believe in this essence, the disease-object being used merely to give the spirit tangible form in the primitive mind. Accordingly the real cause would be a spirit in the body, and such diseases should be classed along with those caused by demons who enter the body without the aid of a tangible vehicle. It is not definitely known, however, that the peoples referred to believe in the essence as the prime cause without which intrusive objects would be harmless. Accordingly the criterion here set up for disease-object intrusion is the actual presence in the body of a tangible, supposedly pathogenic substance, whether it is regarded as the embodiment of a spiritual essence or not. 4. Spirit Intrusion This class includes all those etiologies whieh hold that disease is due to the presence in the body of evil spirits, ghosts, or demons. At first sight it might seem that the more familiar term possession would serve to designate this concept. However, as generally used, this has too variable a meaning to be satisfactory in a classification. Thus, the following quotation from Boas may be cited: Belief in obsession is a most characteristic form of belief in the Old World. On the other hand, it seems quite foreign to the beliefs of American tribes. Ideas of personal contact between man and supernatural beings are quite frequent, but obsession and with it the various forms of exorcism do not seem to occur. The spirits may attack man but they do not enter his body.4 Although Boas uses the term obsession in the above quotation, it obviously has the same general connotation as "possession." His 4 20:27. Clenentes: Prmitwe Concepts of Disease statement is not accurate since the belief is found in many parts of South America, in the eastern United States, and even in Greenland, although it is not so prominent in the New World as the Old. Supernatural beings in the body, however, do not always cause sickness. Their hosts may not fall ill in the ordinary sense but become "possessed." Their actions become erratic, they may fall into trances, and their speech at such times is accepted as the utterance of the supernatural being. These persons, being thus in far more direct touch with the supernatural than ordinary mortals, are usually respected as holy men or women and are often consulted as oracles by the populace. This belief is clearly illustrated in Polynesia, where the term for priest literally means "god-box." Madness is also explained in this way by many peoples, the utterances of the insane person being interpreted as the voice of the resident spirit. In parts of East Africa, insanity is called the "madness of the gods." Obviously, these causes are not comparable with the entrance of spirits or demons into the bodies of ordinary mortals to cause actual sickness. In such cases the demon is rarely vocal, his presence becoming known only through the medicine-man's diagnosis. In the treatment of such sickness three methods are used. It may be purely psychic, consisting of incantations, invocations of more powerful demons, and appeals or commands to depart. Secondly, the demon may be extracted by mechanical means such as bleeding, the ingestion by the patient of noxious herbal decoctions, by evil smelling incense, loud and discordant noises. Thirdly, he may be entangled in specially prepared bundles of twigs brushed over the patient's body or cast into an animal. However, these therapeutic measures are rarely used separately among primitive peoples. Mechanical means are generally employed but are regarded as ineffective without the proper verbal conjurations. Yet, in the usual description of such practices the terms "bleed- ings," "exorcism, " " transference," etc., are used indiscriminately. In the hope therefore of systematizing the terminology, the following nomenclature will be adopted: Spirit intrusion includes all cases of disease ascribed to the pres- ence in the body of a supernatural being. Possession is restricted to cases in which the supernatural being speaks through his host. Generally, the only form of sickness it includes is insanity, although this may have other causes, such as the loss of the soul. The criterion of true possession is the belief that 1932] 189 190 University of Califoria Publications in Am. Aroh. and Ethn. [Vol. 32 the voice of the possessed person is really that of the supernatural intruder. Exorcism will be applied only to the purely psychic methods of getting a disease-demon to leave a sick person's body and its meaning as given by the dictionary can be taken literally,-' 'the driving out of evil spirits from persons or places by conjuration." Mechanicat extraction is self-explanatory, applying wherever the patient's body is freed from the intruding spirit by direct surgical, medical, or manipulative means. Transference describes the transferring of the disease demon from the body of the patient to that of another human being, an animal, or even substances sueh as rock crystals and special bundles of leaves or twigs. Disease transference to animals is usually called the "scape- goat" idea, but there seems to be no essential difference between it and other forms. Such a terminology is undoubtedly somewhat artificial but prob- ably not more so than most classifications. In actual practice of course there is a blending of the ideas included. Thus, certain peoples may have true possession as a specialization of the more general spirit intrusion, while others have only ideas classed under the latter term. In the treatment of such diseases, exorcism is nearly always used with one or both of the other methods, although its importance varies greatly over the world. 5. Soutl Loss5 This class includes all theories attributing sickness to loss of the soul. This may be abstracted by ghosts or sorcerers; or, when leaving the body during sleep, it may meet with some accident on its nocturnal ramblings which prevents its return. In either case the unfortunate owner of the soul soon becomes sick and dies unless it is shortly returned. It is not my intention here to deal with primitive methods of treating disease except in so far as is necessary for an understanding of the theories of causation. It will suffice for this purpose to sketch the main therapeutic measures of primitive people in connection with the five main disease concepts. These procedures usually have little actual efficacy, but as Rivers remarks: Their modes of treatment follow directly from their ideas concerning etiology and pathology. From our modern standpoint we are able to see that these ideas 5 The term "soul" as used here refers to the primitive conception of a shadow -or tenuous double and not-to the metaphysical creation of sophisticated theologians. 2Ctements: Primiti-ve Concepts of Disease are wrong. But the important point is that, however wrong may be the beliefs of the Papuan and Melanesian concerning the causation of disease, their practices are the logical consequences of those beliefs.6 This is to say that primitive medical practice is the result of a very simple cause and effect sort of reasoning. Thus, if sickness is due to loss of the soul, the obvious remedy is to find and restore it. It may merely be lost or it may be held captive by some evil spirit or sorcerer. In either case, it is the medicine man's task to find it and bring it back to its owner. If illness is due to evil magic worked against the sufferer, counter-magic must be invoked to discover the malevolent sorcerer and force him to cease his operations. Sickness caused by object intrusion is treated by the extraction of the malefic substance, while the manifest remedy for diseases resulting from spirit intrusion is to oust the intruder. Similarly, if illness is a punishment sent by the gods for a breach of taboo, the obvious corrective is to propitiate the angered god or spirit. Hysteria and epilepsy account for practically all cases of posses- sion. Mental suggestion plays an enormous part in the cures as well as the sicknesses. Perfectly healthy persons who discover that evil sorcery is being worked against them have been known to sicken and die within a few days, so strong was the influence of suggestion. Extracted disease objects are always shown to the patient, who believes they really came from his own body and often recovers under the powerful suggestion that the sickness is gone. Actually of course the doctor merely pretended to extract the object from the patient's body, sleight-of-hand being one of the most important tools in the primitive medical man's bag. Purely physical methods, such as mas- sage, heat applications, and certain items of primitive materia medica, undoubtedly have a direct beneficial action. Finally, the primitive medicine-man is aided by the factor that often cooperates with his civilized brother, viz., the fact that in a large proportion of illnesses the patient will recover either in spite of the treatment or wholly without it. While the classification given above embraces all the important theoties, a few isolated ideas do not fit the classification. Thus, in parts of Indonesia it is believed that a person is born with a certain amount of soul substance which gradually becomes less during life, much like the oil in a lighted lamp, sickness being due to the shrink- age of this substance. Among the Hottentot, disease is thought to be 5 156:51. 1932] 191 192 Unimersity of California Publications in Amn Aroh. and Ethn. [Vol. 32 due to the disarrangement of one 's internal organs. These are restored to their proper positions by massage. However, other theories of disease which fall under the five concept classification also occur among the Hottentot. Another very curious idea found among these people is that a particular disease can be treated successfully only by persons who have previously had that disease. Shallow incisions are made in the patient's flesh, and the bodily dirt of a person who has had the disease is rubbed into the CUt.7 The analogy to early methods of smallpox vaccination is certainly odd and the practice may really be derived from it, as such vaccination was introduced into South Africa by the English. The tabulated bibliographic data and the geographical distribu- tion of the five main disease concepts are given in table 1. Following this are theoretical discussions of each concept, with outline maps showing the plotted distribution of all but sorcery. 7 87:77-78. Clements: Primitive Concepts of Disease TABLE 1 DISTRIBuTION OF THE FIVE MAIN DISEASE CONCEPTS OVER. THE WORLD Disease- Breach Tribes or regions Soul lo0s object Spirit of Sorcery intrusion intrusion taboo I. II NORTH AMERICA Eskimo Baffinland .................. 11: 207 11: 207 Behring strait .................. 142 : 422 142: 435 142 422 142 422 60:193 Central 15: 593 15: 595 Copper .................. 101 :172 101 :173 101 :171 101 :173 Cumberland sound .......... 16: 506 16 :121 Greenland ................. 95: 243 91 : 80 36 :198 36:198 91: 80 91: 80 Herschel island ................ 196 : 342 196: 342 196: 342 Hudson bay, east 217 :196 217 :196 Hudson bay, west 16 :121 16 :121 Labrador 83: 131 83 :131 Point Barrow .................. 141 : 432 141: 423 141 : 422 Point Stivens 196: 360 196: 360 Smith bay 196: 128 Mackenzie Area Canadian Den6 (gener- ally) .................. 140: 209 140: 206 140: 207 140: 208 Chipewyan .................. 82: 729 Yellow Knife .................. 140: 209 Plateau Area Carrier .................. 140 : 209 Kootenay .... 30 :185 Nez Perc6 .................. 193 : 253 Shuswap .................. 82 : 729 208 : 612 208 : 612 208 : 612 Tahltan 51: 113 51: 113 Thompson .................. 207 : 360 207 : 360 207 : 360 North Pacific Coast Bella Coola .................. 13: 37 Chemakum .................. 6 : 201 Chinook .................. 14: 205 14 : 207 14 : 207 14: 207 Haida .................. 202: 42 202: 40 202: 40 6 : 203 Hupa .................. 63: 66 68:63-66 68: 63-66 Klallam .................. 77 : 295 6: 201 73: 293 73: 300 73: 300 193 1932] 194 University of California Ptbluiations in Am. ArcI aid Ethn. [Vol. 32 TABLi 1- (Continued) Diseae- Breach Tribes or regions Soul los object Spirit of Sorcery intrusion intrusion taboo K w akiutl............................ Q uileute.............................. Q uinault.............................. Salish of Oregon.............. Songish................................ T illam ook.......................... T lingit................................ Tsim shian.......................... T w ana.................................. Vancouver island (gen- erally).............................. Y urok.................................. California-Great Basin A chom aw i.......................... C ahuilla.............................. California (generally).... D iegue io............................ Luisefio................................ M odoc.................................. M ohave.............................. M ono.................................... Paiute.................................. Paviotso.............................. Pom o.................................. Salinan................................ Shoshoni (Wind river)... U te........................................ Y okuts................................ Southwest A pache................................ C ochiti................................ N avaho................................ Pim a.................................... Sia........................................ Zufii...................................... Plains A rapaho.............................. C row .................................... D akota................................ 82 : 729 59 :335 148 60: 187 82 : 729 113 : 292 12 : 475 82 : 729 6 : 201 4: 356 92 : 337 46: 83 152 : 219 6: 203 115 : 299 201 122 : 290 128 : 294 66 ? 47 :157 199:144 244: 89 148 148 18: 10 203 : 470 136: 66 169 : 594 115 : 299 92 : 335 115 : 299 192 : 312 152: 216 201 128 : 292 128 : 292 61: 63 131: 184 128 : 292 128 : 292 66 25 : 471 47: 154 166: 261 199: 75 153 : 281 200: 40 118 : 437 127 : 374 244: 89 Gros Ventre ................ 1116 : 276 1 116 : 222 113 : 284 46: 83 152 : 222 201 61:63, 73 66 134:133 1134:154 244: 89 6: 18 153 : 283 127 : 374 148 148 18: 10 208 : 470 169 : 594 115 : 302 192: 315 152: 215 128 : 292 128 : 292 61: 70 131: 184 128: 292 128: 292 66 25: 471 .47: 151 134: 20 166 : 261 199: 75 153: 281 200: 40 127 : 374 244: 89 I I I 195 Clement*: Primitive Co&wept* of Disease TABLkE 1- (Continued) Diase- Breach Tribes or regions Soul loss object Spirit of Sorcery intrusion intrusion taboo I . H idatsa.............................. K ansa.................................. M andan.............................. Ojibway (Plains).............. O m aha................................ O sage.................................. Ponka.................................. Ea8tern Woodland Algonkian.......................... Cree (east)........................ Iroquois.............................. Huron.................................. Menomini............................ Mohegan............................ Nascopie............................ Penobscot.......................... Ojibway (Great Lakes). Sauk and Fox.................... Saulteaux (north)............ Southeast C herokee............................ C hickasaw .......................... C hoctaw .............................. C reek.................................. N atchez.............................. Y uchi.................................. A ntilles................................ Nahua Aztec.................................... Costa Rica........................ G uatem ala........................ H uichol.............................. M aya.................................... Mexico (generally).......... Mexico (southern)............ N icaragua.......................... Tepecano........................... 219 : 436 183: 67 181 :194 88 :157 6: 204 183: 67 149 : 605? 45 : 418 182 : 501 45 : 417 28 : 583 45 : 418 45 : 374 45 : 417 107A:507 181 :182 45: 374 183:63, 76 139 :155 217 : 270? 88:157, 197188:157, 197 204: 62 185 : 292 205: 80 189: 132 25 : 472 71: 40 170 :1-4 195: 64 183:63, 76 205: 81 167: 8 180: 47 34 :155 2 : 333 67: 19 167: 7 229:120 170 :1-4 210 : 60 1210 : 68 80: 615 1 132 : 119 1133 : 206 1133 : 220 132:119-120 45 : 517 45 418 45 : 512 183: 63 139 : 156 107A:513 181 :188 188 :196 182 : 244 88: 157,197 76 : 249 183:63, 76 138 : 337 2 : 294 204 : 62 185 : 292 189: 132 163 : 353 167 : 245 170 :1-4 129 :175 63: 36 210: 72 80 : 615 195: 21 149 : 605 133 : 220 132 :119 1932] f I I I I 196 Univerzity of California Publications 'i Ami. Arch. and Ethn. [Vol. 32 TABLE 1- (Continued) Disease- Breach Tribres or xegions Soul loBss object Spirit of Sorcery intrusion intrusion taboo I S S SOurH AMERICA Chibcha Area (western Co- lombia, Panama, and Ecuador) Aurohuaca.......................... Canelos................................ Cayapa................................ Cherigoto.......................... Chibcha.............................. Choco.................................. Colima................................ Ijca. ;. Jivaros................................ M agdalena river.............. M uzos.................................. Paragoto............................ Pitagoto.............................. Peru.......................................... (general; no specific tribes) Amazon Area.......................... (eastern Colombia, Ven- ezuela, Guiana, Brazil) Bakairi................................ Bororo................................. Carib (generally)........... Coroados........................... Guinau................................ Guiana (generally)......... Karaya............................... Paressi............................... Piaroas......................... Puris................................... Taulipang..................... Tukano............................... Tupi.................................... Venezuela............................ Xingu................................. Yahuna................................ Yekuana.............................. Yuruna................................ Chaco (southeast Boli- via, Paraguay, north- ern Argentine).............. Abipones............................ 5 : 357 213: 35 190 : 315 107 : 473 5: 344 177 :191 147:347-50 178 : 221 178 : 221 177 :191 177 :191 190 : 314 184 : 277 107 :160 I 221: 340 1 222 : 300 111 :192 111: 170 163 :165 221 : 435 1 222 : 300 111 :170 111 :170 111 :170 56: 73 99 : 346-49 170 :160 163 : 350 43 : 249 107 : 473 5 : 344 24 :137 24:137 107 :154 171: 224 56: 73 107 :158 99: 346-49 163:164,182 107 :158 107:158 111 :192 42: 194 107: 158 107: 158 136: 84 107 :156 144: 639 107 : 489 107: 285 107:489,491 213: 35 107 : 278 107 : 281 194: 78 107: 471 S 348 177 : 191 107 : 471 213: 35 107 :160 171 : 242 111 :192 99: 34649 43 : 67 I I I I I 1932] Clement8: Primitive Concepts of Disease 197 TABix 1-(Ccrntinuzed) Disease- Breach Tribes or regions Soul loss object Spirit of Sorcery intrusion intrusion taboo I Ashluslays ................. 146: 54 Chamacoco ................. 107: 6 Choroti ................. 107 :156 Guaycurus ................. 107: 157 Itonama ..... 126 : 547 Lengua ..... 72:127, 135 72:127,135 72:127, 135 84:290 84:290 84:290 Matacos 107: 7 Todas 107: 7 Tonocotes .................. 107 :157 Southern Argentine and Chili Araucanos ................. 120 : 353 120: 353 Patagonians 44: 93 Tierra del Fuego 35 :160 35 :159 Onas . ............. 62: 297 62:301 62:299 Yahgan ... 112:72, 172 112 :175 ASIA Siberia ................... 77: 724 Ainu 7: 290 95:250 7:290 Altaian tribes ...... 40: 282 Amur river ...... 219 : 437 Buriat ...... 168:576-585 40:158,287 40 :158 Chuckchee ...... 103 :102 40: 259 23: 332 Eskimo ...... 22 21:424, 448 Finnic tribes .40:163 Gilyak .197 126 : 548 197 Gold .198: 478 Kirghiz 145A: 45 Koryak .103 :101 103: 28 103: 101 103 :101 40: 149 Mongol and Turkish tribes (generally) ........ 197 Samoyed ................ 40: 163 Teleut and other Turk- ish tribes ................ 198: 488 Tungus ................ 219: 437 104: 156 Turanian Tatars .............. 219: 437 Yakut .103 : 102 198: 483 Yukaghir .104:156 104 : 152 104:140,147 Tibet and Mongolia Buddhist tribes (gener- ally) .219: 437 198 Univerity of CaUfornia Publication ian Am. Aro. and Ethln. [Vol.32 TAB&; 1-(Contiwe) Disease- Breach Tribes or regions soul loss object Spirit of Sorcery intrusion intrusion taboo China .................. 219: 438 77: 578 Lolo of southwest China 60 183 Miao of southwest China 32: 69 32: 71 Southeast Asia Annamese ................. 6: 38 136: 66 Karen of Burma ................ 219: 438 Peninsular Malays .......... 179 : 410 179 : 410 179 : 410 Sakai .................. -53: 186 Semang Pygmies .............. 172 : 264 172 140 Siamese 136: 66 Western Asia Arabs of Mesopotamia.. 77: 724 Arabs (generally) 94: 410 Assyria (ancient) 77: 724 77 : 742 77 : 741 Babylonia (ancient) 77 : 742 77 : 741 77: 568 Jews (ancient) .................. 219 :102 10 77: 755 Persians (ancient) ............ 81 : 398 India Angami Nagas 98 :179 98: 242 Bhutan ... 37:123-25 Bombay . 52: 257 Dravidians 39 : 247 Garos of Assam ................ 154: 104 154:104 154:116 Hindus (generally) .......... 81: 398 Nagas of Manipur 86:136 86 :136 86 : 137 Nayars of Malabar 150: 280 150: 280 150: 280 Northern India (gener- ally) ................. 37: 232 37 :141 37 :141 88:277-78 Rajputana 37:123-25 Southern India 97 : 478 Todas 155: 256 Veddas of Ceylon 175 : 346 INDONESUA Aaru islands .................. 6: 24 136: 84 Ambon .................. 119: 92 6: 18 6 :201 Andaman islands ............... 26:217 26:178 26: 89 26:302 Babar islands 136: 84 Bataks of Sumatra .............. 119: 92 225: 14 225: 14 Borneo .................. 93: 31 39:121 54: 251 93:128 39:117 145: 111 Dyaks .................. 70:134,168 70: 165 70: 165 70: 97 219 : 442 ]Cenen.ts: Primititve Concepts of Disease Disea- Breach Tribes or regions Soul los object Spirit of Sorcery intrusion intrusion taboo I I~ ~ ~ I .. . , Kayan .................. 93: 34 93: 28 Klemantans 93:129 Melanaus 93: 39,117 Punans 93:130-35 Sarawak .............. 161: 273 Formosa .............. 135 147 135: 147 135: 147 Indonesia (generally) .......... 79 : 236 Java .............. 119: 92 Manobo of Mindanao .......... 119:192 64: 593 64:593 Mentawei .................. 124 124(late) 124 124 Minangkabaus of Sumatra 6 : 201 Nias .................. 109: 45 109: 36 6: 38 173: 565 Moluccas ....................... 60:186 Nicobar . ... 110: 302 110: 303 225: 14 Philippines (generally) ...... 117 :180 117 :186 117: 186 Pygmies of the Philippines 65 65 65 Serang .................. 6: 38 6 138 Timur .................. 58 : 417 58: 417 58 :417 Toradjas of Celebes ............ 119A:370, 119A:396 381 Uliase .................. 6 : 201 6: 18 Watubela .................. 6: 38 Wetan .................. 60: 189 MELANESIA 156: 15 33:224-25 33: 203 Banks islands .................. 33: 198 33 :198 33: 198 Fiji .................. 220 : 353 219 : 436 Leper's island .................. 33: 227 Loyalty islands .......... 60: 185 Mota .................. 33 : 227 New Britain .108: 337 New Caledonia .60: 548 New Guinea .................. 143: 516 174: 640 6: 18 219:436 174:640 New Hebrides .................. 33: 208 33: 200 33:194,199 Santa Cruz . .33:197 - Solomon Islands .................. 100: 16 100: 192 100: 256 100: 278 33:209 Trobriands ..... ..... 130:239-40 130:239-40 MICRONESIA Carolines (Yap) ........... 27: 114 Marshall islands ........... 55:121,322, 55: 329 55 : 333 334 I I I I 1932] 199 TABLz 1- (Contikued) 200 University of California Publications in. An. Arch. and Ethn. [Vol. 32 TABIE 1-(Continued) Disease- Breach Tribes or regions Soul loss object Spirit of Sorcery intrusion intrusion taboo POLYNESIA 75: 233 75: 233 75: 233 75: 233 Easter islands .................. 164: 238 164 : 239 Hawaii .................. 75:236 75:236 75: 242 75: 236 50 : 293 Hervey islands .................. 6: 38 74 :171 Maori of New Zealand ....... 9: 88 214: 19 214:19-21 214:19-21 214 : 495 69: 19 9: 83 9: 69 Marquesas .................. 75 : 237 75 : 237 Niue . 123 : 397 123 : 397 Samoa 75: 245 216 :140 Tahiti 50: 349 Tonga 75: 245 AUSTRALIA 191:534-37 191:534-37 Arunta . ............ 74 :115 Dieri . .............. 74:115 Euahlayi .................. 151 : 27 151 : 27 151: 45 151:27, 32 North Central tribes .......... 191:534-37 191:534-37 Southern tribes .................. 219 : 436 96:379-80 96:379-80 196 : 435 Victoria ..................6 : 204 6: 24 Wurunjerri ...... 204 : 206 60 :.183 TASMANIA 223:247-48 162 : 64 AFRICA North Africa Semitic Egypt (modern) ............ 95: 243 Morocco 1 : 23 Siwan (upper Egypt) 1: 23 Hamitic Egypt (ancient) 220 : 354 77 : 750 Tuareg 157: 279 East Africa Hamitic Abyssinians (Galla and Lango) 3: 79 Masai .................. 89 : 308 Sudanese: Nilotic Bateso 160 : 286 160: 285 Kavirondo 160 : 386 160: 285 Nandi .................. 90: 69 90: 51 Shilluk 227 :176 Suk 8: 30 I I I I 1932] Clemnents: Pri'mitive Concepts of Disease 201 TABiE 1- (Con-duded) Disease- Breach Tribes or regions Soul loss object Spirit of Sorcery intrusion intrusion taboo Sudanese: Bantu A-Kamba ................ 85: 97 85: 93 Akikuyu ............... 165:240-41 165: 258 Baganda .............. 159:18, 23 159: 100 159: 101 159:101,288 159:98, 344 Bageshu ............... 160:177 160: 176 Banyankole ............... 160: 128 160 :127 Banyoro ............... 160: 57? 160: 54 Basoga . ......... 160: 223 16: 223 160: 222 Bondei 41: 219? South Africa South Africa (generally) 218 : 280 Basuto ................. 219: 431 Bechuana ................. 77 : 730? Thonga ................. 106 : 341 106 : 438 Vandau .................. 19: 1 ? 19: 3 Zulu .................. 219: 98 29 : 262 Hottentot .................. 77 : 730? Congo and Central Africa Angola .................. 31:145-47, 31:130 31: 173 290 Bakongo .................. 226 : 225 226:227,281 226 : 219 Bushongo .................. 212: 71? 212 : 238 Upper Congo 77 : 730 77 : 730? West Africa: Sudanese Ekoi .................. 206: 17( 206: 407 Ewe ......... 48:96, 106 48: 96 48: 99 Kpelle ......... 228:183 228: 211 228: 296 228: 202 North Guinea ......... 219 : 437 Old Calabar ......... 219 : 430 Tshi ......... 49:150 49:148 49:142 Madagascar ......... 219 : 454 176: 285 176 : 292 EUROPE 219: 442? Bosnia .................. 95 : 239 England 38 : 277 Germany .................. 219 : 442? 211: 61 Greece (ancient) -.- 220: 15 77: 590 Ireland . .............- 136: 65 136: 65 38:277, 278 M oravia .................. 95 : 247 Lapland .................. 215:164-65 Rome (ancient) .................. 219: 103 Sweden .................. 211 : 61 Switzerland .................. 94 : 294 Transylvania .................. 60:182 Slavic peoples .................. 77 : 622 I 202 University of California Publications in Am. Arch, aitu Ethn. [Vol.32 DISCUSSION OF CONCEPTS SORCERY The preceding table shows that belief in disease-causing sorcery is distributed literally to the ends of the earth. Even the Fuegians at the extreme end of South America, one of the most primitive groups alive, believe in evil contagious magic. Likewise the Aus- tralians, also one of the most isolated and backward of primitive peoples, have a well developed system of sorcery, although here the method is either to "shoot" some object magically into the victim or to abstract his soul, which is regarded as resident in the kidney fat. Again, disease is attributed to human agency working through magic by the Andaman islanders, whose culture is probably the crudest found today. In the New World a belief in sorcery, usually in the form of imitative or contagious magic, ranges from the Arctic to Tierra del Fuego. In the Old World the universal occurrence of sorcery and its persistence in Europe until quite recent times are so familiar that copious bibliographic proof seemed unnecessary. The New World data for sorcery are fairly complete, and it will be seen that its distribution in both Americas is continuous or nearly so. For this reason, no map of the geographical range of the concept has been included, as a belief in disease sorcery may be safely credited to all peoples outside the West European culture sphere; and even there it has only recently declined, still lingering among the ignorant. However, the forms of sorcery vary considerably even though the generic idea is universal. This universality may be explained by Bastian's "Elementargedanken" or by historical connection, formal variations being due to several independent, though not parallel, elaborations of the basic idea. The places of origin of these elabora- tions might then serve as centers of diffusion. Again, each form of sorcery might have a single origin, all occurrences of one type going back to that spot. The various types may thus be chronologically separate, or they may all trace back to an archaic culture horizon. At any rate, the subject is sufficiently complex to form a secondary problem and its full treatment is beyond the scope of this paper. 1932] Clements: Primitive Concepts of Disease 203 204 UnAversity of Ca2fornia Publications in Am. Archl. and Ethn. [Vol 32 BREACH OF TABOO The distribution of this concept is given in map 1. A glance will show that its occurrence is rather scattered, although it is found in all the major geographical divisions. No doubt this discontinuity is partly due to insufficient data. Yet, a fairly representative sample of peoples for every area of the world has been studied and the concept is not mentioned except as shown. This distribution appears rather haphazard but in certain regions the concept is of much more importance than in others. Thus, the Yukaghir appear to be the only Siberians holding the belief but it is evidently of little importance, most sickness being ascribed to loss of the soul. In India there is the theory that a breach of taboo may cause demons to lodge in the body, thus producing illness, but the idea se'ems to be feebly developed. Among the Naga of Manipur there is mere mention of this cause of sickness, the important concept being disease-object intrusion. Similarly in western Oceania the concept occurs but does not seem comparable in importance with other theories. The same is true of the Euahlayi of southern Australia and the Ainu of northern Japan. Among the Akikuyu and A-Kamba of East Africa a breach of taboo or failure to perform certain sacred duties angers one 's ancestral ghosts, who will then send sickness. Here again, how- ever, this seems unimportant as a direct cause of disease. The Bakongo of Angola are likewise reported as possessing the concept but it is not accentuated. Among the Ekoi of West Africa, however, it appears to be somewhat more prominent. On the other hand, the belief appears to have been of great impor- tance in ancient Assyria, Babylonia, and Palestine. The same is true of Polynesia, where the extreme development of the general taboo idea leads one to expect corresponding emphasis on the causation of sickness by sin. According to the data, Polynesia and the ancient Near Orient are the only regions in the Old World where the concept attains real importance. In the New World, the idea occurs among most of the Eskimos and, according to Father Morice, is general among the Canadian Dene. In the United States its distribution is extremely sporadic, being recorded for the Hupa, Pomo, Luisefno, Mono, Yokuts, Navaho, Zuii, Crow, Iroquois, and Chickasaw. Among all these tribes, how- ever, the concept is of relatively little importance except among the Eskimos and, to a lesser extent, the Dene. The Iroquois case is, more- Clernents: Primitive Conoepts of Disease over, not strictly a breach of taboo but rather a belief that failure to perform certain duties will cause sickness. In Mexico, Colombia, and Peru, however, there is a great develop- ment of the concept. There are only two other South American occurrences, the Bororo and the Tupi, and among both the notion seems but little developed. Associated with this belief is another idea, namely that treatment is futile unless the patient confesses to his sin. In such cases, the medicine-man usually questions him persistently until some forgotten breach of taboo is recalled. The sufferer then acknowledges his sin and his recovery is supposed to follow as a matter of course. In addition, the angered god is often tempted to relent by means of sacrifice. The occurrence of confession8 with breach of taboo sickness is shown on map 1. Table 2 gives the bibliographic references for this association. This combination of ideas proves to be still more limited in distri- bution than breach of taboo alone. In Africa, confession is recorded only for the Ekoi and the Baganda, who are widely separated. The association does not seem to occur at all in Asia and its presence in Indonesia is confined to Mentawei, where it may be of recent intro- duction. In Polynesia, however, confession is definitely linked with breach of taboo sickness, playing a quite important part in the treatment. TABLE 2 ASSOCIATION BETWEEN BREACH OF TABOO DISEASE CONCEPT AND CONFESSION METHOD OF TREATMENT North America. Baffinland Eskimo, 11:207; Central Eskimo, 15:595; Cum- berland sound, 16:121; Hudson's bay (west coast), 16:147; Smith bay and Dease river, 196:128; Canadian Den6, 140:207; Aztec, 67:19; Guatemala, 229:120; 170:1-4; Maya, 210:68; Mexico generally, 60:615. South America. Aurohuaea of Colombia., 144:639; Ijea of Colombia, 107:489; Peru, 107:489, 491; 213:35. Oceania. Hawaii, 75:242; Maori of New Zealand, 75:242; Polynesia generally, 75:233; Mentawei, 124. Africa. Baganda, 159:101, Ekoi, 206:407. In the New World, confession goes with breach of taboo among the Eskimos of Baffinland, on the west coast of Hudson's bay, on Smith bay, and the Dease river. Father Morice states that this association is common among the Canadian Dene. Finally, nowhere 8 This refers only to confession as a treatment for sickness due to breach of taboo and not to confession in general. 1932] 205 206 Un.iversity of California Publications in Ain. Arch. and Ethn. [Vol. 32 else in America is confession coupled with breach of taboo disease outside the region of highest culture,-Mexico, Colombia, and Peru. The above facts are curiously difficult of interpretation.. Were it a mere question of the breach of taboo disease concept alone, a fairly plausible case might be made out for its independent origin in many places. It obviously depends upon a belief in gods or spirits of a higher order than mere ancestral ghosts. Given this belief, to refer sickness to such divinities seems a simple step which might be taken independently by many peoples. Turning to map 1, we find, not a continuous distribution over the world but certain separated areas. There are two of these in the New World, Middle America and the Arctic, this latter including northwest Canada. In the Old World, Polynesia, Melanesia, parts of Indonesia including the Andaman Islands, and north India form a fairly continuous area of occurrence. Aside from these the concept is found sporadically. The theory here advanced rests on the universality of a belief in spirits or gods who take a direct part in human affairs and are sus- ceptible of placation. On such a basis the idea of taboo-religious prohibition or social prohibition carrying divine sanction-might arise independently in several different regions. As a means for enforcing such taboos the idea of divine punishment for transgression evolves naturally, from which it is an obvious step to ascribe misfortune and sickness to breach of prohibitions. This has evidently taken place independently in Middle America, in Arctic North America, and in southern Asia. The African occurrences may also be independent, but this will be discussed later. Taking the Oceanic data first, the breach-confession complex is most highly developed in Polynesia. It is definitely registered here for Hawaii and New Zealand, and, according to Handy, its occur- rence elsewhere in the area is extremely probable. Outside Polynesia the complex is limited to Mentawei. In Africa the association is reported only for the Baganda and Ekoi. Considering the lack of the taboo idea in ancient Egypt and the very limited occurrence on this continent of its linkage with sickness, I incline to the view of an independent origin of the con- cept in Africa. Alternately, it may have been introduced by mis- sionaries, especially where it is associated with confession, as such religious influence has been rather marked among both the tribes mentioned. Clemnents: Ptimititve Concepts of Diwease As for the Asiatic occurrences, a fairly continuous distribution from northern India eastward through Polynesia will be seen. The western Asiatic occurrence is also quite possibly part of this distri- bution, for the taboo idea was found even among the ancient Sumerians. As many culture traits have traveled the route from Mesopotamia across Persia to India and so into Oceania, this con- jecture has a certain plausibility. However this may be, it is here concluded that the idea of sickness due to breach of taboo originated independently somewhere in southern or western Asia and diffused eastward until it covered most of Polynesia. Here the taboo idea in general has had an extreme development, its increased importance in connection with disease concepts being a direct result. The South Australian occurrence may not be so isolated as it appears for the idea may occur elsewhere here, in which case the Australian instance would be part of the general Oceanic distribution. The concept, how- ever, is not mentioned for other Australian tribes and the Euahlayi may have independently derived the idea from the underlying belief in general taboo. This occurrence will be taken up again. The Ainu concept is possibly related to that of the Yukaghir, though it may be independent. The latter occurrence is regarded as related to the general Arctic area of distribution in North America, especially since the Eskimos of Behring strait share the idea. Here again an independent development is postu.lated for the Eskimos. From its point of origin it diffused eastward to Greenland, westward across Behring strait to the Yukaghir, and southward to the Canadian Dene. The Eskimos have a considerable development of the general taboo idea in connection with the goddess Sedna, hence here as in Polynesia, the association with siclhness appears natural. Turning to Middle America, a practically continuous distribution is found for Mexico, Colombia, and Peru. Here independent origin is-again inferred in view of the elaboration of organized religion in this area. The sporadic occurrences in the United States are too indefinite for discussion except for those in the Southwest, for which a Middle American origin is suggested. The few Brazilian instances are interpreted in the same way. This liberal use of the principle of independent origin is supported by the presence of the breach of taboo theory among the Andaman islanders, the Pygmies of the Malay peninsula, the Philippine Pyg- mies, and the South Australians. These are among the most primitive people in the world today and have been reached by comparatively 1932] 207 208 Uniiversity of California Publications in Am. Arch. and Ethn. [Vol. 3' few waves of diffusion. Accordingly their culture may be rated no only primitive but primeval, thus preserving isolated remnants o: very old forms of culture which once were widely spread. If the theory under discussion were as ancient as its presenc among these marginal peoples indicates, its distribution ought to bf much more general than it actually is. Furthermore, as an elemen in an archafc culture horizon which has been preserved in the mar ginal areas, we could reasonably expect it among other extremel3 isolated and primitive peoples, such as the Puegians and Tasmanians However, the available data on Tasmania, though scanty, do no suggest its presence and its absence in Tierra del Fuego seems fairl3 certain. The extreme antiquity of the concept may thus be doubted, a well as the historical connection of its occurrences, except within th areas discussed above. For confession as a means of treatment, independent origin musl again be invoked. Its association with the theory occurs in each oJ the three main areas, the Oceanic, the Arctic, and the Middle Ameri can. Its occurrence in Mentawei and among two African tribes maJ well be due to missionary influence. However, in the three mair areas this association is regarded as independent and functionallI related to the underlying taboo concept. Thus, in Polynesia there i, a great development of taboo. Polynesia is part of the general souti Asiatic-Oceanian distribution of the breach of taboo disease concept but confession is not associated with the concept elsewhere in th( area. Furthermore, in ancient Mesopotamia, although there was con siderable elaboration of the idea that sickness might be caused b3 breach of taboo, treatment by confession appears not to have beer associated with it. Confession as such did exist and later came tc play a great part in organized Christianity, but apparently was nol thought of as a treatment for disease until that time. We may there. fore reasonably regard confession in Polynesia as much later thar the breach of taboo disease concept itself and as an independeni development from the Polynesian elaboration of taboo. Much the same argument can be applied to Arctic North America Owing to its more limited distribution, confession here can be regarded as somewhat later than its associated concept. However, it traces tc a local origin and is probably derived from the stress placed by the Eskimos on taboos sponsored by the goddess Sedna. Clemei&tt: Primitive Concepts of Disease In Middle America, confession is confined strictly to Mexico, Colombia, and Peru, but the breach of taboo concept of disease has a somewhat wider, though sporadic, distribution. Accordingly con- fession is here again assumed as later than its associated concept and as an independent development functionally derived from the great elaboration of organized religion and divine placation so character- istic of Middle America. In both Peru and Mexico confession of sin was highly developed and a great deal of sickness wqs treated in this way. It has been thought in some quarters that the practice was introduced by the Spanish priests, as it is well known that the Cath- olic idea of sacrament and penitence made a tremendous impression upon the natives. This would cause the practice to spread quickly. However, there is fairly strong evidence to show that confession in Middle America is pre-Columbian, and its indigenous character is supported by the conclusions drawn here. The more likely explanation is that the natives already had a similar system in operation and so found no difficulty in fitting the Catholic idea into the old established culture pattern. The presence of confession where the breach of taboo disease con- cept is most developed seems to support its functional derivation from this concept or the underlying taboo idea. It is conceivable that, were our ethnographic knowledge more com- plete, the face of the distribution map would be considerably changed, which would doubtless radically modify the interpretation here given. DISEASE-OBJECT INTRUSION The distribution of this concept is plotted on map 2, which shows that the concept is almost universal in the New World, being absent only among the eastern Eskimos.9 In the Old World, however, occurrence is by no means so continuous, although the limits are extremely wide. Judging solely from distribution we might postulate the inde- pendent origin of the feature for America, with the usual tentative Middle American center and diffusion outward from it. By this hypothetical process the concept might have spread south to Tierra del Fuego and northward until the entire New World was covered. 9 The diseontinuity of distribution in parts of the eastern Woodland area is owing to lck of data on certain of the tribes which have been long extinct. 1932] 209 210 University of California Publications in AXm. Arch. and Ethn. [Vol. 32 Cltements: Primnitive Concepts of D'iease It could then have crossed Behring strait to the Yukaghir and Gilyak, possibly reaching the Ainu, although the last mentioned case is doubtful since the intrusion is here applied only to diseases of the teeth. From the Gilyak onward the concept is not encountered again before southeastern Asia, where another fairly continuous distribu- tion area appears. These occurrences could be regarded as inde- pendent of the American equivalents but all as historically related to a South Asiatic origin. Similarly the European, West Asiatic, and African distributions might be tentatively connected, but owing to their lack of continuity with the southeastern Asiatic area, they would represent still another independent origin. In support of this hypothesis it might be argued that there is sufficient basis in everyday experience to suggest several independent centers. Splinters, arrowheads, and like objects embedded in the flesh are widespread experiences and might give rise to, the idea that sickness in general is due to some foreign object in the body. This explanation fits the distribution fairly well, but it will be fruitful to examine an alternative hypothesis. If the trait under con- sideration is really independent in these three areas, there is no particular reason for them to possess any similar features in this regard other than the basic idea In fact, any similarity in details would cast great doubt on their mutual independence. If in addition certain other ideas which have no organic connection with the funda- mental concept are nevertheless associated with it, and if such associa- tions occur widely, then historical connection between all such occurrences becomes a practical certainty. By definition the criterion of our concept is the actual presence in the body of a tangible, supposedly pathogenic substance. Now, although the exact nature of this substance varies considerably, such variation is quite haphazard. That is to say, there is no regional dif- ferentiation, the most common intruders being small pebbles, bits of leather, sticks, little bones, hairs, coagulated blood, insects, and even small animals. This common use of the same kinds of disease objects hints vaguely at historical connection between the occurrences, but the case may be made far stronger. In certain areas we find the idea thsat dental diseases are due to worms in the teeth (see table 3). The distribution is quite sporadic. Yet it is a specialization of the general disease-object concept and 1932] 211 212 Univer8ity of California Publications in Am. Aroh. and Ethn. [Vol. 32 not one that could plausibly be viewed as arising independently.10 Accordingly its occurrences are all regarded as historically connected and strengthen the case for such connection between the general disease-object occurrences. Furthermore, as the idea was current in ancient Rome and still earlier in ancient Assyria, a reasonable degree of antiquity is assured, the wide distribution indicating an origin considerably more ancient. TABLE 3 DENTAL DisEAsEs DuE TO WORMS IN THE TEETH North America. Omaha, 45:417; Ponka, 45:417. Asia. Ainu, 77:724; Arabs of Mesopotamia, 77:724; Assyria (ancient), 77:724. Oceania. Banks islands, 33:193; Maori of New Zealand, 214:19. Africa. Madagascar, 176:285. Europe. Europe (generally), 136:66; Rome (ancient), 136:66. To strengthen our argument further the therapy employed in such cases must be discussed. A considerable variation in modes of treat- ment might be expected if the concept were really independent in various areas. Such variation, however, does not occur. In fact, the uniformity of the therapeutic methods is a most striking phenomenon. Obviously the disease-object must be extracted, but this is done in a peculiar way. Sometimes it is removed by massage, but by far the most usual method consists in the medicine man's sucking the object out. Now suction as opposed to other forms of extraction is certainly not organically necessary to the disease-object intrusion concept. Furthermore, this treatment itself involves a whole complex of ideas. Obviously the real cause of the sickness is not a disease object in the body; its apparent presence involves trickery on the part of the doctor, who usually carries an assortment of disease objects about with him. When treating a patient, he sucks over the afflicted spot but either surreptitiously introduces a disease object into his mouth 10 It is conceivable that observations of nerves clinging to extracted teeth could cause this idea to arise independently but this does not seem probable. In the first place, even with modern methods of 'extraction the nerve is usually broken off and is not often seen clinging to the tooth. Even when a piece of the nerve does adhere to the tooth it is not easily distinguished from the bloody bits of other tissue which cling to the root. Primitive methods of extraction would almost invaria-bly break the nerve. Furthermore, extraction of teeth is quite uncommon among primitive people. The Australians furnish an exception as the knocking out of an upper incisor is frequently a part of their initiation rites. If the idea that worms get into the teeth to cause toothache were really spontaneously derived from observations of nerves clinging to the extracted teeth, the belief ought certainly to be found in Australia, but it does not seem to occur there. Clemnents: Primitive Coneepts of Ti8se or palms it. At the proper time this object is exhibited to the patient and his relatives as though it had actually been extracted from the body of the sufferer. Its removal by massage involves the same sort of trickery. All this implies no mean skill in sleight-of-hand, and while conjuring in general might independently develop, this par- ticular form of legerdemain in association with the idea of disease- object intrusion is surely too complex and unnatural a phenomenon to have originated more than once. Table 4 gives the distribution of this association. It occurs all over the world in every area where the disease concept itself occurs, with the exception of Europe. For the latter there are no data, so the presence or absence of suction cannot be determined. As this treatment involves a fairly complex linkage of ideas and furthermore has no conceivable organic relation to its associated disease concept, the association of the two must be regarded as an historical adhesion. That is to say, this mode of treatment probably originated about the same time and in the same place as the disease-object concept and the two became functionally"' associated. Once linked, they diffiused together simultaneously. TABLE 4 ASSOCIATION BETWEEN SUCEKING TREATMENT AN" DISEASE-OBJECT INTRUSION North America. Plateau, Den6, 140:2206; Kootenay, 30:185; Thompson, 207:360. Northwest coast, Hupa, 68:66; Klallam, 73:298; Quinault, 148; Yurok, 115:299. California-Great Basin, Cahuilla, 92:335; California (generally), 115:299; Luiseflo, 152:216; Mono, 201; Paiute, 128:292; Paviotso, 128:292; Salinan, 131:184; Shoshoni, 128:292; Ute, 128:292; Yokuts, 66. Southwest, Apache, 25:471; Cochiti, 47:154; Zufni, 200:40. Plains, Arapaho, 118:437; Crow, 127:374; Gros Ventre, 166:222; Omaha, 45:417; Ponka, 45:417. Eastern Wood- land, Huron, 107A:507. Southeast, Natchez, 205:80; Antilles, 25:472; 71:40. Mexieo, Mexico (southern), 195:64; Tepecano, 132:119. South America. Chibeha, Cayapa, 5:535; Cherigoto, 177:191; Choco, 147:347; Colima, 178:221; Muzos, 178:221; Paragoto, 1777:191; Pitagoto, 177:191. Peru, Amazon-Orinoco, Carib (generally), 111:199; Taulipang, 111:199; Venezuela, 163:350; Guiana (generally), 111:199; 163:181. Chaco, Abipones, 43:249; Guayacurus, 107:157; Lengua, 84:290; Tonocotes, 107:157. Tierra del Fuego, Onas, 35:160; Yahgan, 112:175. Asia. Peninsular Malay, 179:410; Nagas of Manipur, 86:136. Oceania. Borneo, 39:121; Nias, 109:35; Pygmies of Philippines, 65; Toradjas of Celebes, l19A:396. Bank~ islands, 33:198; New Guinea, 174:640. Australia. Australia (generally), 191: 534-37; Euahlayi, 151:27; North Central tribes, 191:534-37; Southern tribes, 96:379-80. Africa. Baganda, 159:100; Upper Congo, 77:730; Kpelle, 228:211. 11 The term functional as used here indicates traits which are not organically or causally connected but which, once associated, function together and thus form a trait complex. 1932] 213 214 University of California Pubtications in Am. Arch, and Ethn. [Vol. 32 Thus it seems most reasonable to view all occurrences of the disease-object intrusion concept as tracing back to a single origin in conformity with Tylor's original idea regarding it.12 Assuming this to be true, a considerable antiquity for, the idea is fairly certain. In the Old World it is extremely widespread, and in America it seems practically universal, extending to the very tip of the southern continent. Generally pan-American traits are regarded as extremely old, probably dating back to the migrations across Behring strait from Asia toward the end of the Pleistocene. Since the evidence against an earlier population of the New World is very strong, all culture in America is probably post-Pleistocene, or little more than ten thousand years old.13 The first immigrants brought their culture with them and carried it with them as they and their descendents spread southward to the extreme end of the southern continent. This proto-American culture developed during the Old World Palaeolithic and is therefore of Pleistocene age. Elements of universal distribution in the western hemisphere are held to belong to this horizon; accordingly the disease-object intrusion concept may reasonably be assigned to it. Though this yields an antiquity of at least ten thousand years, a considerably greater age may reasonably be inferred. Its occurrence at the geographical limits of the Old World, together with the fact that most archaic diffusion between the hemispheres has taken place in an easterly direction, strongly implies its Old World origin. Now the concept with other archaic elements was probably carried across Behring strait toward the end of the Pleistocene. Yet the diffusion into northeastern Asia must have taken a comparatively long time, so the actual origin of the concept can be viewed as considerably earlier than the American migrations. It seems probable therefore that disease-object intrusion is a genuine Palaeolithic trait. It is true that contrary to this interpretation the concept does not occur continuously in certain parts of the world. This discontinuity, however, can be reasonably explained. The areas where the concept is absent are all rather widely separated. Thus, the Eskimos of Baffin- land and Greenland appear to lack the idea. Across Behring strait it is found among the Koryak and Gilyak, but not elsewhere in Siberia or central Asia, nor in India and China. In southeastern Asia, Indonesia, Australia, and part of Melanesia it occurs but is absent in 12 218:281. 18 Recent finds of supposedly Pleistocene artifacts in New Mexico are too uncertain at present to be discussed here. Clements: Primitive Concepts of Di8euse Polynesia except in New Zealand, where it is limited to the notion that worms in the teeth cause toothache. Its distribution in Africa is sporadic and, with the exception of Ireland, its European form is again the specialized idea of worms causing toothache. The evidence for historical connection makes us assume that this distribution was once more continuous, but that as the idea is extremely old, either later disease concepts overlaid and submerged it in certain areas, or contemporary concepts were elaborated at the expense of intrusion. This interpretation finds almost perfect confirmation in the distribu- tion of the gaps. Thus the only vacant area in America is the eastern Eskimo, and it is precisely here that breach of taboo and soul loss are of great importance. In Middle America the concept occurs but its importance is by no means so great as in other parts of this con- tinent. In keeping with the elaboration of religion and the increasing emphasis on breach of taboo and spirit intrusion the disease-object concept seems to be in process of atrophy. This argument is sup- ported by the Amazonian data, for while disease-object intrusion is fairly important, spirit intrusion accompanies it and seems to enjoy an equal, if not superior, status. In most of North America, however, spirit intrusion is relatively feeble, disease-object intrusion being the main theory of sickness. In the Old World, the Siberian gap seems to be directly connected with the great development of the soul-loss idea. In China and India spirit intrusion has been greatly elaborated, evidently crowding out disease-object intrusion. Again, in Polynesia great emphasis is placed on breach of taboo, apparently with the same effect on the older theory as among the eastern Eskimos. In Africa spirit intrusion is the most dominant disease concept, the natives waging an eternal battle with ghosts and demons. Here then, as in India and China, spirit intru- sion has lessened the prestige of the disease-object theory and caused its gradual decay. The "displacement" hypothesis advanced above finds some slip- port in the European data. Here there are numerous references to the former existence of the disease-object concept. According to McKenzie,"4 the belief that worms in the teeth are the cause of tooth- ache is, or was recently, general over Europe. In Ireland, until quite lately, sickness in cattle was believed due to malefic objects, such as oddly shaped stones, or old arrowheads getting into the animal's body; and these "elf-bolts" were extracted by wizards.'5 Likewise European 14136:66. 15 218:280. 1932] 215 216 University of California Pu.blicationm in Am. Arch. a,nd Ethn. [Vol. 32 folklore contains scattered references to small animals, such as mice, getting into the body. We may safely conclude that these are survivals from a time when a more general credence was placed in the whole disease-object theory, this either being submerged by later beliefs diffusing in, or decaying through the elaboration of some other concept, probably spirit intrusion. Thus, while in many parts of the world disease-object intrusion exists side by side with other concepts, those regions possessing an elaboration of some other concept either exhibit the disease-object idea in atrophied form, or entirely lack it. In other words, there is a fairly high negative correlation between disease-object intrusion and any other concept (except sorcery) when that concept has undergone elaboration. The above interpretation seems quite in consonance with the extreme antiquity of the disease-object idea, for if it were less old its occurrence would probably be more continuous. Furthermore, the coincidence of its areas of absence with those where other disease concepts are elaborated is too close to be accidental. Two customs connected with disease-object intrusion are of inter- est. Among the Arapaho-" and the Choctaw'7 in the United States the object is extracted by bleeding. The skin is scarified over the afflicted spot and a horn cup applied. 'This cup is sucked by the doctor to create a vacuum, and bleeding ensues. The doctor has meanwhile introduced an object into the cup and when this is taken off the object is found in the blood as though it had actually been sucked out of the body. This same treatment is found among the Baganda of Africa.'8 The idea that the disease object can be invisibly transferred into another person, an animal, or even into a bunch of specially prepared twigs is found among the Creek'9 and the Yuchi20 in the southeastern United States and among the Solomon islanders in Melanesia. SPIRIT INTRUSION The distribution of this concept is plotted on map 3.- It seems to be continuous for Europe, Africa, all southern Asia, and most of Oceania, but more sporadic in the New World, although the various occurrences axe so close together that connection between them seems 16118:439. 17 204:62. 18 159:100. '19 186:121. 20189:132. 21100:192. Clements: Primritive Concepts of lDiseg e 1932] 217 218 Univer8ity of California Publications in Am. Arch. and Ethn. [Vol. 32 reasonable. The American distribution links up with the northeast Siberian occurrences and these again, although seemingly separated from other Old World areas, are probably historically related to them. Differences between the American concept and its Old World form, moreover, are fairly numerous and furnish some basis for viewing the two hemispheres as independent in this regard. However, data on spirit intrusion in America are vague and scanty, this prob- ably being the reason its presence in the New World has been hitherto overlooked by Americanists. Also definite information regarding spirit intrusion in the Old World is not so plentiful as might be thought. Furthermore, spirit intrusion has many ramifications into religion, magic, medicine, and kindred phases of culture. The detailed study of its cultural contexts is impossible here and the idea has accordingly had to be artificially isolated. Such a process of cutting out has severed many of its associations with other elements of culture. In the first place, this theory of disease obviously depends upon the universal belief in spiritual beings. If Tylor's view of animism is accepted as an explanation of the universal credence in spiritual beings, such a belief may have arisen independently many times. However this may be, given this universal belief, the supposition that spirits may work good or evil seems an obvious one and may likewise be an "elementary idea." Similarly we might account for the widespread belief that spirits may actually enter the body and cause sickness. This last idea, however, is really not so "natural" as it seems. In the first place, it is not clear how ordinary experience could give rise to the notion of an intruding spirit; and furthermore there is no obvious reason for attributing sickness to the presence in the body of such a foreign spirit. Moreover, if the idea of spirit intru- sion as a cause of sickness were spontaneously derived from generic belief in supernatural beings, the concept should occur among all peoples. This, however, is not the case. The spirit intrusion disease concept seems quite lacking in Aus- tralia and Tasmania. It does not occur among the Negritos of the Malay peninsula nor among the Philippine Pygmies. Its occurrence in the Andaman islands is quite doubtful, and the same may be said for Tierra del Fuego, where it is only vaguely mentioned for the Ona and not at all for the Yahgan. Other areas of absence shown on map 3 are not so definite and will not be considered here except to remark that the Siberian absence is probably c6rrelated with the great development of the soul-loss concept. Inspirational possession Clements: Primitive Concepts of Disease of shamans is marked in Siberia, so the idea of spirit intrusion as a cause of disease may exist here, although in atrophied form. On the other hand, such absences may be reasonably accounted for if the disease concept of spirit intrusion is regarded as a complex which was only invented once or twice and diffused from its point of origin. The peoples mentioned above are all extremely primitive. Owing to their long isolation they have been reached by few waves of diffusion, thus preserving a primeval culture relatively uninfluenced except by later local developments. Now if spirit intrusion was diffused after this isolation, its absence is explained much more reasonably than by some psychological peculiarity of theirs preventing spontaneous derivation of this disease theory. Moreover, it has been argued that the whole spirit intrusion concept is not one obviously derived from experience even granting a generic belief in spirits. Its wide distribution, then, is again more plausibly explained if its occurrences are regarded as historically connected. The case is strengthened by the distribution of certain associated ideas listed in table 5. This shows that spirit intrusion as a cause of insanity is found once in America,-among the eastern Cree. It crops up again in Borneo and in Africa. These widely separated occurrences are conceivably independent but, in view of the argument advanced above, this does not seem likely. New World data on the subject are very scant and other American instances may well exist. Furthermore, such an explanation of insanity was current in Europe during medieval times. Thus the linkage of insanity with spirit intrusion is more common than is indicated in table 5, possibly even in America, and the historical connection between its occurrences is at least possible. TABLE 5 SPIRIT INTRUSION Exorcism North America.. Eskimo, Point Barrow, 141:422. Plateau area, Tahltan, 51:113. Nahua, Maya, 210:72; Tepecano, 132:119. South America. Chibcha area (western Colombia, Panama, and Ecuador), Caya.pa, 5:346; Ijea., 24:134; Magdalena river, 24:134. Peru (general; no specific tribes), Carib (generally), 56:73; Guiana (generally), 163:182. Asia. China, 77:578. Southeast Asia, Peninsular Malays, 179:451. Western Asia, Assyria (ancient), 77:742; I3abylonia (ancient), 77:742; Jews (ancient), 10. India, Bhutan, 37:123-25; Bombay, 52:257; no-rthern India (generally), 37:141, 169; Rajputana, 37:123-25. Indonesia. Dyaks, 7:165; Klemantans, 93:129; Punans, 93:130-35. 1932] 219 220 University of California Publications in Am. Arch. and Ethn. [Vol. 32 Polynesia. 75:233; Maori of New Zealand, 69:4. Africa. North Africa, Semitic, Morocco, 1:23; Siwan (upper Egypt), 1:23. Hamitic, Tuareg, 157:279. East Africa, Hamitic, Abyssinians (Galla and Lango), 102:221. Bantu, Baganda., 159:101; Banyoro, 160:55; Basoga, 160:223. South Africa, Thonga, 106:439 (recent). Congo and Central Africa, Bakongo, 226:227, 281. Sudanese, Tshi, 49:150. Europe, Germany, 211:61; Lapland, 215:164; Sweden, 211:61. Mechanical Extraction North America. Eskimo, Point Barrow, 141:422. Eastern Woodland, Cree (east), 183:76. Plateau area, Tahltan, 51:113. South America. Chibcha area (western Colombia, Panama, and Ecuador), Cayapa, 5:346. Peru (general; no specifie tribes), 107:158 (bleeding); Coroados, 107:158 (bleeding); Guiana (generally), 163:182; Karaya, 107:158 (bleeding); Paressi, 107:158 (bleeding); Xingu, 107:158 (bleeding); Yahuna, 107:158 (bleeding). Chaco (southeast Bolivia., Paraguay, northern Argentine), 107:156 (bleeding); Ashluslays, 146:54 (bleeding); Choroti, 107:156 (bleeding). South- ern Argentine and Chile, Patagonians, 44:93 (bleeding). Polynesia. 75:233. Africa. East Africa Hanmitic, Abyssinians (Galla and Lango), 102:221. Bantu, Banyankole, 160:128. Transference to Animals Asia. Western Asia, Jews (ancient), 10. India, Bombay, 52:257; Northern India (generally), 37:141, 169. Africa. East Africa, Nilotic, Shilluk, 227:119. Bantu, Baganda, 159:342; Banyoro, 160:55; Basoga, 160:223. South Africa, Bechuana., 77:730; Hottentot, 77:730. Congo and Central Africa, upper Congo, 77:730. Europe. England, 136:84; Ireland, 136:85; Rome (ancient), 136:85. Transferenoe to Plants and Inanimate Objects North America. Nahua., Maya, 210:72; Tepecano, 132:120. South America. Chibcha area (western Colombia, Panama, and Ecuador), Ijea, 24:134; Magdalena river, 24:134. Peru, Yuruna, 135:84. Asia. India, northern India (generally), 37:164. Indonesia. Aaru islands, 136:84; Babar islands, 136:84; Borneo, 54:251; Punans, 93:130-35. Polynesia. Maori of New Zealand, 214:19-21. Africa. East Africa, Nilotic, Ba.teso, 160:286; Kavirondo, 160:286. Bantu, Bageshu, 160:177; Banyankole, 160:128. Europe. England, 136:84; Ireland, 37:164. Insanity North America. Eastern Woodland, Cree (east), 183:78. Indonesia. Kayam, 93:39, 117; Melanaus, 93:39, 117. Africa. North Africa, Hamitic, Tuareg, 157:278. South Africa, Thonga, 106 :438. Clements: Primitive Concepts of Disexe In the treatment of sickness due to spirit intrusion, exorcism is nearly always employed in both hemispheres. Mechanical extraction may be regarded as more important, as well as transference of the demon, but verbal conjuration is essential for the success of these other methods. Now, exorcism is surely not causally connected with the spirit intrusion concept. That is to say, there is no reason why the practices of exoreism should arise spontaneously from the concept of spirit intrusion. The association is purely functional, comparable to the adhesion of the sucking treatment to the disease object concept. Accordingly, the prevalence of the association between exorcism and spirit intrusion must be regarded as fairly strong evidence for the historical connection between all occurrences of this disease concept. Another idea associated with spirit intrusion is that of transfer- ence. Thus, the intruding demon may be cast into other persons, animals, plants, or specially prepared substances. Here again the relation is not necessary but accidental. The idea of transference occurs widely and is likewise to be regarded as a purely functional adhesion to the spirit-intrusion concept. In nearly every case it is combined with exorcism. Here then is a three-element association. The chances of such a linkage of ideas occurring very often are remote, hence its frequency strongly supports the hypothesis of his- torical connection. Judged by its distribution the spirit intrusion idea must be rather old. Its occurrence in America as far south as Patagonia implies a fairly remote time of entrance into the western hemisphere, although not necessarily dating back to the first migrations. Likewise its absence in the marginal regions indicates a later than truly archaic horizon. In the New World the concept is far more vague than in the Old, its importance in American etiology being feeble outside of Mexico and parts of South America, where it has undergone elabora- tion. Its origin then may tentatively be regarded as Old World post-Pleistocene,-definitely later than the disease-object theory. Although the evidence for historical connection between all occur- rences of spirit intrusion is strong, certain differences between the Old and New Worlds remain to be considered, although they do not invalidate the diffusion theory. In both hemispheres transference occurs, but in the Old World the demon, though occasionally trans- ferred to plants or inanimate objects, is far oftener cast into an animal, while in America he is transferred to bundles of twigs or special stones and only very rarely, if ever, to animals. In Costa 1932] 221 222 University of Caltifornia Publications in Am. Arch. and Ethn. [Vol. 32 Rica22 and Guatemala23 animals are held over the patient, but it is not stated that these are cases of demon transference. On the whole, transference of the disease demon to animals is probably foreign to America and the common procedure in the Old World. The lack of this feature in the New World seems connected with the comparative lack of domesticated species,-the only ones used for the purpose in the Old World. Now it has been seen that transfer- ence is common to both hemispheres and with exorcism widely accom- panies spirit intrusion. I suggest that these three traits became functionally associated, diffused together, and are all part of the same culture horizon. Their wide distribution implies considerable, although post-Pleistocene, antiquity, an age greater than that of ani- mal domestication. It is therefore argued that this disease concept with its adhesions of exorcism and transference originated sometime after the close of the Palaeolithic but before the Full Neolithic and began to diffuse over the world. After animal domestication began in the Old World Full Neolithic, the transference idea was applied to the domesticated species and this also diffused. Being a late development, however, its range was considerably less than the older generic idea of demon transference and remained confined to the Old World. Even here, however, animal transference is most prominent in regions where the use of domestic animals is more important, while, elsewhere, transference to plants and inanimate objects is usual. Demon trans- ference to animals is thus possibly a later development of the generic idea and connected wvith animal domestication in a purely fortuitous way. Another difference between the two hemispheres lies in the great use of bleeding as a means of removing intruding spirits in South America. Bleeding is widely practiced in the Old World but not specifically to extract disease demons from the body. In America the idea apparently does not occur north of Mexico and is probably a South American development. In Africa, contrary to general belief, spirit intrusion cause of disease is not particularly developed. True enough, it occurs widely, and spirits or ghosts figure as the prime cause of sickness. However, most illness is ascribed to attacks by spirits, and not specifically to their presence inside the body. This with the feebleness of the con- cept in most of America and its only relative importance in certain 22 180:47. 23 229:120. Cleten-ts: Primitive Coneepts of Disease other parts of the world, leads to the important generalization that spirit intrusion is most elaborated in the regions of highest culture. Thus in America this concept of disease attains real importance only in Middle America, and even here it functions feebly in com- parison with Europe and Asia. In the Old World it is highly elab- orated in China, India, western Asia, Egypt, and Europe. Elsewhere it occurs, but only as equal or inferior in standing to other concepts. In the areas just mentioned, however, it is by far the most important theory of sickness, and it is surely no coincidence that these should also be areas of most complex culture. Furthermore, exorcism in con- nection with mechanical extraction, while used widely, is most empha- sized in the higher cultures, enacting merely a supplementary role among simpler peoples. The importance of physical treatment in the regions of higher culture seems to be relatively feeble. This emphasis on exorcism among the higher cultures of the Old World might be explained as an .elaboration of the older and more widespread use of incantations with other modes of treatment. Reliance on exorcism alone might then have diffused among these cultures. However, the American data hint at a different interpreta- tion. In the New World incantations are widely used but emphasis on exorcism at the expense of other therapeutic methods seems con- fined to Mexico. Here there was a well defined procedure for casting out demons by conjuration and transference.24 As this development of exorcism seems confined to Mexico, its recent independent elabora- tion in America is probable. Moreover, Mexico was the center of greatest religious development in the western hemisphere, while Egypt, western Asia, and India were such centers in the Old World. Taking these data as a whole, it seems probable that the develop- ment of organized religion is functionally connected with an elabora- tion of the spirit-intrusion concept and the emphasis on exorcism as a means of treatment. Such elaboration may have taken place in only one of the Old World centers and diffused thence to the others, or it may have been independent in each. At any rate, the Mexican development, though not so marked as the Old World phase, seems a genuine parallel and strengthens the hypothesis that organized religion tends to elaborate a pre-existing belief in spirit intrusion and to emphasize the purely exorcistic phases of treatment. 24 210:72. 1932] 223 224 University of California Publications in Am. Arch. and Ethn. [Vol. 32 POSSESSION Possession as defined (p. 189) is rarely regarded as causing illness other than insanity. Generally possessed persons are believed inspired by the gods and nearly always have a religious function. Possession, in this sense, exists widely in the Old World except in the most marginal areas. It was strongly developed in the nations of antiquity and it plays a great part in Oceanic religion. The belief was universal in Europe; it exists over most of Africa, throughout India, and China, and functions importantly in Siberian shamanism. So far as is known, it does not occur among the Pygmy peoples nor in Australia. In the New World, possession occurs in the North Pacific Coast area and as an adjunct of Eskimos, this distribution being continuous with that of Siberia. The notion is not reported elsewhere for North America and for South America only a single case has been noted. On the basis of this distribution, possession may be held to be a more recent development than spirit intrusion and as a relatively late diffusion into North America from Siberia. In origin the idea may be derived from the underlying belief in spirit intrusion and the coincidence of its areas of elaboration with the regions of organized religion points again to the probability that the latter tends to elaborate a pre-existing belief in spirit intrusion. The isolated South American case occurs in Venezuela. Here the doctor sends his own soul out to bring back the spirits of deceased shamans. These spirits enter the doctor's body and speak through him, giving advice, answering questions, and behaving in the tradi- tional manner of spirits.25 This case is conceivably due to Negro influence which has been rather marked in this part of South America since the introduction of black slaves. However, the description given by Koch-GrUnberg ha-s a wholly un-African flavor, so that the trait. is probably indigenous. It fits perfectly in the aboriginal culture pat- tern and bears no marks of foreign sources. Quite likely, then, here is an independent development of possession, seemingly derived from the underlying belief in spirit intrusion.28 The psychological factors which form the true basis of inspira- tional seizures probably occur universally. Given an established doc- trine regarding gods, demons, and spirit intrusion, possession has possibly developed independently several times. The fact, however, 25 111:196-198. 26 125. Clements: Primitive Concepts of Disease that its distribution is practically continuous and that the belief nearly always involves clairvoyance and telesthesia seems to give more weight to an explanation by diffusion from a single source for the Old World and Arctic North America. SOUL LOSS The geographical distribution of this disease theory is plotted on map 4. A mere glimpse will show the enormous extent of the concept, reaching to the limits of the inhabited earth. Its occurrence, nevertheless, is not universal. Although Rivers27 denies the soul loss idea of disease to Asia, a very respectable distribution is found in northern India, Tibet, neigh- boring Mongolia, and all of southeastern Asia. This conception was likewise current among the ancient Persians and probably among the Jews of biblical times. Furthermore it occurs very widely in Siberia, being by far the most important theory of disease. Its elaboration here is remarkable; the Buriat, on whom unusually full information is available, have eight of the commonly associated ten ideas, the two associations lacking being excluded by their nature, as a glance at table 6 will show. Soul loss is likewise of great importance among most of the Eskimos and in the North Pacific Coast area. It occurs widely else- where in both hemispheres, but the Siberian-Eskimo;.North Pacific Coast distribution is undoubtedly the area of greatest elaboration, and serves to connect the two hemispheres across Behring strait. Previous to the present survey, only a few sporadic occurrences were known elsewhere in America, whence the tendency to regard soul loss as a Siberian development only recently diffused into the New World and overlaying other concepts. Search of the literature has, however, revealed a much wider range and this implies a different interpretation. As to the distribution as such, certain facts must be discussed which are not obvious from the map. In the first place, although the dotted areas indicate that the presence of the concept is uncertain, its actual occurrence there is probable. Thus, in Germany the soul is believed to quit the body during sleep in the form of a mouse or bumblebee.28 In neighboring Switzerland29 and also in Transylvania30 it is likewise said to leave the body at night in the form of a mouse, sickness resulting if it does not return. According to Tylor,8' a belief 27156:79. 2894:179. 2994:294. 30 60:182. 31 219:442. 1932] 225 226 University of California Publications in Am. Arch. and Ethn. [Vol. 32 that sleeping persons should not be disturbed was general over Europe not long ago, owing to the fear that a returning soul might not be able to find its body. Again, sneezing salutations are common in Europe82 and while their implication of soul loss is not strong, a definite belief that a sneeze indicates the soul's departure is widely linked with such salutations elsewhere in the world. These facts, though not conclusive, certainly hint strongly at a general European occurrence of the soul-loss concept, which has lately died out while certain associated ideas survive in tradition and myth. In Africa the dotted area on the upper Congo applies to the Bushongo. Although there is no definite mention of soul loss as a cause of sickness there, they do believe that the soul can, leave the body during sleep, dreams being the experiences of the wandering soul.33 This notion is very widely associated with the soul theory of sickness. The Bushongo also hold that ghosts can steal upon sleeping people and squeeze their hearts, thus causing death.34 This looks suspiciously like soul abduction. Furthermore, the neighboring Bakongo on the west and the Baganda on the east definitely possess the soul-loss concept. Here again the supplementary facts strongly hint at its presence among the Bushongo, although it is not specifically recorded. In southeast Africa the dotted area refers to the Vandau, who have the dream wandering idea and are, moreover, in close contact with the Thonga and the Basuto, both of whom definitely hold the soul-loss theory.85 Its presence among the Vandau thus seems quite probable. Although not definite, soul loss in ancient Palestine is implied, for Tylor states that Jacob prayed for the soul not to leave the body on sneezing as before.36 In view of the prevalent association between sneezing and soul loss where this theory of sickness is known to obtain, its occurrence among the ancient Jews is likely. Information on the Tasmanians is very scanty; however, it is known that they greatly disliked to mention an absent person by name, believing that this might injure him.37 Many primitive people regard the name as a second self and are reluctant to divulge their names lest persons learning them might call away their souls or other- wise harm them. A similar belief may have existed in Tasmania. 32 219:101. 33212:196. 34 212:71. 35 19:1. 36219:102. 87223:244. Clementg: Primitive C&noepts of Di8emee 1932] 227 228 University of California Publicatioms in Am. Arch.. and Ethn. [Vol. 32 These people also believed that death never occurred before sunset, the soul quitting the body but remaining near it until the sun had gone down.38 Although the facts are far from satisfactory, they do indicate the soul-loss concept here as more than a mere possibility. While soul loss is not definitely mentioned for Nicaragua, a certain kind of witch is said to assault sleeping people by applying the mouth to the navels of their victims and sucking so as to cause death.-9 As certain other peoples locate the soul in the abdomen, this is possibly a case of soul abduction. The Tukano in the northwest Amazon district may share the soul- loss concept, as they are in close contact with tribes where its pres- ence is definite. Moreover, the Tukano identify the heart with the soul, an association widely found among such peoples.40 Turning to definite cases, the geographical limits of the theory prove to be extremely wide, taking in such primitive peoples as the Tierra del Fuegians, the Easter islanders, the Pygmies of the Philip- pines, the Semang Pygmies of the Malay peninsula, the Australians, and the Andaman islanders. It occurs sporadically but widely in Africa and is present in Madagascar. Its presence in Europe is defi- nite for Switzerland and Transylvania, while its wider distribution here seems probable. In Asia its occurrence among the biblical Jews is likely and it is definitely recorded for the ancient Persians. North- ern India, Tibet, and part of Mongolia also share the concept, which is general throughout Oceania, reaching over to Easter island. It is highly developed in Siberia, extending across Behring strait and covering most of Arctic North America.4' Continuous with this Arctic area is its distribution through western Canada and the Southwest, and generally about the Great Lakes region. Absence of soul loss seems fairly certain for northeastern Canada, most of the Plains area, and the general region east of the Mississippi. It does not occur in Mexico or Central America, with the doubtful exception of Nicaragua. In South America, however, it is found in Ecuador, Peru, Guiana, and Venezuela, parts of the Amazon area, the Chaco, and Tierra del Fuego. Such a distribution, including many of the most isolated and primitive peoples, admits of several interpretations. Soul loss obvi- 38 223:247-248. 39 149:605. 40111:171. 41 Information on the natives of the Aleutian islands would be very desirable, but the Aleut have been so thoroughly Christianized that all traces of their former shamanistic beliefs have disappeared. The only source of information on their early customs known to me is the work of the Russian missionary, Veniaminoff, but this work is not at present available. Clements: Primitive Concepts of Disease ously depends upon a belief in the soul, the universality of which has been previously mentioned. Various explanations of this universal belief in souls and spirits have been advanced but that of Tylor is probably the most satisfactory. The primitive notion of the soul is that of a double or ethereal image of the body, not actually immaterial but merely less grossly corporeal or more tenuous than the real body and also capable of surviving after the death of the material body. According to Tylor, primitive man arrived at such ideas through inferences from observed facts and on the evidence of his own senses. The phenomenon of death is a striking one, and early man must have been led to reflect on the difference between a living body and a corpse. He would thus be led to assume some tenuous substance or life essence in the living body lacking in the dead, whence the notion of a vital principle. Moreover, in dreams and visions the sleeper sees human shapes which differ from those of conscious experience. They may be persons who are really dead or who have long been absent. Again, they may resemble friends known to be far away. When the sleeper wakes and reflects on his dreams he will infer that he has really somehow been away or has been visited by other persons. However, since his body has apparently not moved and since the persons in his dream are known to be either dead or far away, it obviously could not be the actual persons who appeared in the dream but phantoms who are real and resemble their fleshy bodies. In this way the idea of ethereal bodily images would arise. The life essence is apparently separable from the body during sleep, coma, and death, while the ethereal image can quit the body any time to wander about, sometimes appearing to persons at a distance. Tylor submits that primitive man combines these two notions into the idea of a ghost-soul whose presence in the body, except for short periods, is essential for life and health. According to Tylor the belief in a ghost-soul forms one of Bastian's "Elementargedanken,"-ideas occurring universally and due to the fundamental psychological similarity of mankind. Because of this likeness, similar experiences would always give rise to similar ideas. Although Bastian did not otherwise attempt to account for these universal notions, Tylor's theory of animism essays a psychological explanation for one of them. What Tylor really implies is that this belief in a soul has been spontaneously developed by many different peoples and probably at several different periods. If this view is tentatively accepted, a 1932] 229 230 Univuersity of California Publications in Am. Areh and Ethn. [Vol. 32 plausible argument for a similar independent development of the soul-loss concept can be advanced. Thus, there are two main ideas which are widely distributed. First, there is a soul whose presence in the body except for relatively short periods is essential for life. Second, the soul may leave the body during sleep. Given these two ideas, the soul-loss concept of sickness will follow automatically if accidents can befall the wandering soul and prevent its return. This is not only obvious but easily derived from experience, for mishaps of various kinds are a matter of common occurrence. Accord- ingly it would be plausible to explain the world-wide distribution of this theory of disease by independent derivation from pre-existing conceptions of the soul. Its occurrence among isolated peoples adds weight to this hypothesis. The soul-loss concept extends to the limits of the earth but its distribution is not continuous, but quite sporadic in certain areas. Such discontinuities seem to suggest independent origin, and except for other facts such an explanation might reason- ably be adopted. However, the supplementary data put a different complexion on the problem. They involve a principle utilized in the sections on intrusion. When dealing with single culture traits, espe- cially if of a simple nature, it is often difficult to determine whether similar traits in different areas developed independently there or were invented once and diffused elsewhere. Diffusion takes time, and this enables culture historians to determine a relative chroiology. Other things equal, the older the trait, the wider will be its geographical range. By plotting distributions on a map relative time positions can thus be inferred. This principle must be used with caution but seems to hold fairly well except for certain instances of wide but speedy spread due to modern rapid transportation. Diffusionist theories have been carried to great lengths in certain quarters. Elliot Smith and Perry derive most elements of higher culture from Egypt, while the German diffusionists explain all simi- larities by historical connection. Sane use of the principle, however, does not deny the possibility of independent origin: similar traits, if widely separated, are treated as parallels; continuously distributed, they are traced to a single origin. But this method is successful only if the trait has really been dif- fused from a single origin; and such historical connection is often difficult to establish, particularly if the trait is simple. This raises a further point of method which has already been applied to adhesions. A simple trait may conceivably have spontaneously arisen many times. Clemrents: Primitive ()oncept& of Dieae23 If, however, certain other traits not organically related to it are nevertheless regularly linked with it, such association cannot be rea- sonably regarded as independent. That is to say, the linkage of sev- eral unrelated ideas in a particular form may occur once in the history of the world, but the chance of recurrence is most remote. Accordingly the presence of similar traits similarly but not causally connected with a central trait is proof of historical connection. Paral- lelists could prove their point only by establishing not only inde- pendent traits separated in space but also similar initial steps leading up to the two terminal similarities. TABLE 6 SoUL Loss Sneezing North America. Eskimo, Cumberland sound, 16:506. South America. Amazon area, Guinau, 111:170; Guiana (generally), 163:182; Taulipang, 111:170; Yekuana, 111:170. Asia. Siberia., Buriat, 40:158. Southeast Asia, Peninsular Malays, 179:533. Western Asia, Jews (ancient), 219:102; Persians (ancient), 81:398. India, Hindus (generally), 81:398; northern India (generally), 37:240 ?. Indonesia. Ambon, 119:92; Bataks of Sumatra, 119:92; Formosa, 135:147; Indonesia (generally), 79:398; Java, 119:92; Manobo of Mindanao, 119:92; Toradjas of Celebes, 119:92; 119A:381. Melanesia. Banks islands, 33:226; Leper's island, 33:227; Mota, 33:227; New Guinea, 81:398; New Hebrides, 33:227. Polynesia. Maori of New Zealand, 214:144. Africa. East Africa, Hamitic, Baganda, 159:23 T. Congo and Central Africa, Ba.kongo, 226:277. West Africa, Sudanese, Ewe, 48.:96; Tshi, 49:203. Dream WandeTing North America. Eskimo, Greenland, 36:185. North Pacific Coast, Chemakum, 6:201; Klallam, 6:201; Quinault, 148; Twana, 6:201. California-Great Basin, Cahuilla, 92:337; Pomo, 122:290; Yokuts, 66. Eastern Woodland, Algonkian, 219:434; Ojibway (Great Lakes), 219:443. South America. Chibeha area, Cayapa, 5:345. Amazon area, Bakairi, 221:340; C(arib (generally), 111:172; Guinau, 111:170; Guiana (generally), 99:346-49; Paressi, 221:435; Taulipang, 111:170; Yekuana, 111:170. Chaco, Lengua, 72:127, 135. Asia. Siberia, Buriat, 40:287; Gilyak, 40:272; Yukaghir, 104:157. South- east Asia, Karen of Burma, 219:438; Peninsular Malays, 179:48; Semang Pygmies, 172:140. India, northern India (generally), 37:231. Indonesia. Andaman islands, 26:217; Dyaks, 70:161; Kayam, 93:34; Formosa, 135:147; Mentawei, 124; Philippines (generally), 117:180; Pygmies of the Philippines, 65. Melanesia. 33:249; 33:266; New Guinea, 143:518; Solomon islands, 100:16. Micronesia.. Carolines (Yap), 27:114. . .. 1932] 231 232 University of California Publi-cations in Am. Arch. and Ethn. [Vol. 32 Polynesia. Easter island, 164:238; Hawaii, 75:236; Hervey islands, 75:58; Maori of New Zealand, 214:208. Australia. Euahlayi, 151:27; southern tribes, 96:434; Wurunjerri, 209:206. Africa. East Africa, Hamitic, Masai, 89:308. South Africa, Thonga, 106:340; Vandau, 19:1. Congo and Central Africa, Bakongo, 226:282; Bushongo, 212:196. West Africa, Sudanese, Ekoi, 206:231; Ewe, 48:102; Kpelle, 228:183; North Guinea, 219:437; Tshi, 49:151. Europe. 219:442; Germany, 94:179; Transylvania, 60:182. Head as Port North Ameriea. Mackenzie area, Canadian Den6 (generally), 140:209; Yellow Knife, 140:209. Plateau area, Carrier, 140:209; Shuswap, 208:612; Thompson, 208:612. North Pacific Coast, Quinault, 148; Salish of Oregon, 60:187; Tsim- shian, 12:560; 60:189; Vancouver island (generally, 169:594. California-Great Basin, Cahuilla, 92:340; Pomo, 122:290. South America. Amazon area (eastern Colombia, Venezuela, Guiana, Brazil), Carib (generally), 111:172. Asia. Siberia, Buriat, 168:576-85; Yukaghir, 104:156. Southeast Asia, Karen of Burma, 219:437; Semang Pygmies, 172:246. Indonesia. Dyaks, 70:168; Kayam, 93:31; Mentawei, 124; Nias, 109:45; 173:565; Pygmies of the Philippines, 65; Toradjas of Celebes, 119A:381; 60:186. Africa. South Africa, Thonga, 105:46. West Africa, Sudanese, Kpelle, 228:183. Madagascar, 219:454. Identified wth Heart North America. North Pacifie Coast, Salish of Oregon, 60:187. California- Great Basin, Modoe, 6:204. Southwest, Cochiti, 47:157; Sia, 199:144. Nahua, Nicaragua, 219:432. South America. Amazon area (eastern Colombia, Venezuela, Guiana, Brazil), Carib (generally), 219:431; Tukano, 111:171. Asia. Siberia, Buriat, 168:576-85. Polynesia. Tonga, 219:431. Africa. South Africa, Basuto, 219:431; Thonga, 105:46. West Africa, Sudanese, Kpelle, 228:300. Associated with Children North America. Plateau area, Nez Perc6, 193:253. California-Great Basin, Luisefio, 46:83. South America. Peru (general; no specific tribes), 213:34. Asia.. Siberia, Buriat, 168:576-85. Melanesia. Banks islands, 33:184, 198; Solomon islands, 33:209. Africa. South Africa, Thonga, 105:46. Europe. Transylvania, 60:182. TABLE 6 (SUPPLEMENT) OTrHER IDEAS AssocTAT WITH THE Sour Loss Soul Loss Due to Fright North America. Quinault, 140. South America. Peru, 213:34. Asia. Buriat, 168:576-85; Koryak, 40:269; Garos of Assam, 154:104. Oceania. Nias, 109:45;Torajda of Celebes, 6:201. Clements: Prmitive Coepts of Disese Captive Soul Eaten by Captor North America. Chinook, 14:207. Asia. Buriat, 168:576-85; Chuckehee, 103:102; Yakut, 103:102. Oceania. Manobo of Mindanao, 65; Nias, 109:45; Trobriands, 130:240. Africa. Angola, 21:173. Liver as Seat of Soul Asia. Buriat, 168:576-85; Chuckehee, 103:119; Koryak, 103:119; Yakut, 103:119. Oceania. Borneo, 79:236; Mentawei, 124; Torajda of Celebes, 79:236. Australia, 220:354. Kidney Fat as Seat of Soul Oceania. Andaman islands, 26:166. Australia, Arunta, 74:115; Dieri, 74:115; southern tribes, 96:337; Victoria, 6:204. Gall Blader as Seat of Soul North America. Ojibway of Great Lakes, 6:204. Returning to soul loss, an explanation by diffusion is justified only if supported by the association principle. The concept is actually linked with a number of ideas registered in table 6. All the peoples listed there have the concept, and the table gives the correlates for each group as well as the proper bibliographic references. As these associated ideas are merely indicated by catchwords in the table headings, a brief definition follows: Sneezing.-This refers to the idea that a sneeze is a sign either of the soul's abandonment of the body or its return after an absence, usually the former; its prolonged absence resulting in sickness. Dream wandering.-This is the belief that the soul leaves the body during sleep. It may meet with an accident or be waylaid by evil spirits who will imprison it. The owner becomes ill and dies if the soul remains absent. Head as port.-This refers to the belief that the top of the head, the fontanelle in children, is the point of exit for the soul. When returning an absent soul the medicine man reintroduces it into the body through this point. Identified with heart.-Many peoples identify the soul with the heart, even calling both by the same term. Evil spirits or sorcerers may steal a person's heart although the actual organ is obviously present. It is not the material heart that is regarded as absent but the soul bearing the same name. Associated with children.-Very often children are thought to be most susceptible to soul loss, hence are carefully guarded from things which would cause the soul to leave, such as sneezing or sudden fright. 1932] 233 234 University of Calhfornia Publications in Am. Arh. and Ethn. [Vol. 32 Among the Thonga of South Africa, when an infant sneezes or hic- coughs the mother will blow upon the fontanelle to "make his heart stand still," while often the fontanelle is covered with wax.'2 In the supplement to table 6 other associated ideas are given, as follows: Soul loss due to fright.-This refers to the belief that a sudden shock or fright will cause the soul to quit the body, which becomes ill unless it soon returns. This idea is often associated with children. Soutl cannibalism.-Many tribes, particularly those of Siberia, believe that cannibal spirits kidnap the souls of the living and eat them. Souls may be trapped, caged, and even fattened for the spiritual table. The unfortunate owner falls ill but does not die until his cap- tured soul is actually eaten. Sometimes a skillful shaman may release it from captivity. Liver as seat of soul.-This refers to the belief that the soul is resident in the liver, which cannibal spirits regard as particularly succulent, spiritual palates evidently relishing the ethereal taste of soul substance. Kidney fat as seat of soul.-This resembles the preceding except that the soul is regarded as being resident in the thick deposits of fat behind the kidneys. Gall bladder as seat of soul-.This is similar to the two preceding beliefs, the heading being self-explanatory. Of all the peoples listed, only the Great Lakes Ojibway place the soul in the gall bladder. This may well be a local development, as the gall bladder is a conspicuous organ, its yellowish green color contrasting sharply with the dark brownish red of the surrounding liver. A queer looking organ might easily be interpreted as the seat of the soul: even in fairly recent times the pineal gland was regarded as the seat of the soul by many European physicians. The distribution of the above ideas is given in table 6. Many of the tribes exhibit several of these associated ideas. Thus, the Buriat of Siberia have sneezing, dream wandering, head as port, identity with heart, special susceptibility of children, soul loss due to fright, captive soul eaten by captor, and the liver as seat of the soul. The Yukaghir associate dream wandering and the fontanelle as port of exit. Both Chukehee and Yakut believe that the soul is devoured by its captors and consider the liver its seat, as do the Koryak, who also connect fright with soul loss. The Gilyak have dream wandering. 42 105:46. Cleinents: Primitive Concepts of Disease23 The Karen of Burma have the same idea together with the head as port, while the Peninsular Malays associate both sneezing and dread. wandering with soul loss. Of particular significance is, the complex of soul loss, dream wandering, and the head as port among the Semang. The Garos of Assam link soul loss with sudden fright. Sneezing was associated with soul loss in ancient Persia and also probably among the ancient Jews. This complex likewise occurs widely in India together with dream wandering. The Andaman islanders have dream wandering and locate the soul in the kidney fat. This la.tter belief is widespread in Australia, although some of the tribes also connect the liver with the soul. In Indonesia sneezing goes with soul loss in Ambon, Sumatra., Java, and generally over the area. This association also occurs in Formosa along with the dream wandering idea while the Manobo of Mindanao have both the sneezing association and the idea that the captured soul is eaten. In Mentawei dream wandering occurs with the head as port and the belief that the soul is resident in the liver. In Borneo the liver likewise figures as the seat of the soul, while the Dyak and the Kayan have both dream wandering and the head as port. In Nias this latter idea recurs, while soul loss appears with fright and spirit cannibals. The Torajda of Celebes couple both sneezing and fright with soul loss and rega.rd the head as port, the liver as the seat of the soul. In the Trobriands the captive soul is eaten, while in the Philippines dream wandering is the common association, also being present among the Pygmies there, who likewise conceive the head as port. In Melanesia dream wandering commonly appears with soul loss, being found in the Banks islands, Leper islands, Mota, the Solomon islands, New Guinea, and generally over the New Hebrides. In New Guinea, soul loss is also associated with sneezing, and the Solomon islanders believe that children are especially susceptible. Dream wandering also occurs in the Caroline islands. In Polynesia dream wandering is widely combined with soul loss, as is signified by its presence in Easter island, Hawaii, the Hervey islands, and New Zealand. The New Zealanders add sneezing, while in Tonga the soul is identified with the heart. In Australia dream wandering usually accompanies soul loss, and the soul is generally located in either the kidney fat or the liver. In Africa the Masai have dream wandering, while the Baganda probably associate sneezing with soul loss. The Basuto identify the 1932] 235 236 Unwmersity of California PubUcation8 in Am. Aroh. and Ethn. [Vol. 32 heart with the soul, as do the Thonga, who likewise regard the head as port, associate soul loss with dream wandering, and believe in children's susceptibility. The Bakongo and the Bushongo are credited with dream wandering. In West Africa this idea is found among the Ekoi, Ewe, Kpelle, Tshi, and generally in north Guinea. Both the Ewe and Tshi couple sneezing with soul loss, while the Kpelle identify the soul with the heart and regard the head as port. This latter association also occurs in Madagascar. In Angola the captured soul is believed to be eaten by the captor. In Europe dream wandering seems to be general, but in Transylvania children are regarded .as especially liable to soul loss. These facts of table 6 have been embodied in the text for the reader's convenience and also to illustrate forcibly the wide occur- rence of whole groups of associated ideas. So far only the Old World data have been recapitulated. They undoubtedly indicate historical connection between all occurrences of the soul-loss concept in that hemisphere. Furthermore, its elaboration in Siberia and the occurrence of eight out of the ten associated ideas among the Buriat, strongly suggest a Siberian origin for the concept. Turning to similar associations in the New World, the Eskimos of Cumberland Sound associate sneezing with loss of the soul, and the dream wandering idea occurs in Greenland. The Canadian Dene generally view the head as the port of exit, which association is also found among the Yellow Knife, Carrier, Shuswap, and Thompson Indians. The Nez Perce consider children especially susceptible, while the Chemakum, Klallam, Quinault, and Twana share the dream wan- dering association. The Quinault likewise regard the head as port and link soul loss with sudden fright. The Chinook believe that cap- tured souls are eaten and the Salish of Oregon identify the heart with the soul. The Salish also have the "head as port" motive, which likewise occurs among the Tsimshian and on Vancouver island. Farther south, the Cahuilla of California associate dream wandering with soul loss and share the "head as port" idea. The Luise-no believe that children are more susceptible to soul loss than adults. The Pomo and Yokuts both possess the association of dream wandering, the for- mer also regarding the head as the port. The Modoc, Sia, and Cochiti all identify the heart with the soul. Farther east, the Algonkians and the Great Lakes Ojibway combine dream wandering with soul loss, the Ojibway placing the soul in the gall bladder. This may or may not be related to the Siberian theory that the soul resides in the liver. Clements: Primitive Conwepts of Disase2 Through Mexico clear to Colombia the soul-loss concept does not occur, with the doubtful exception of Nicaragua, where soul and heart are identified. The above data clearly indicate connection between all the North American occurrences and their historical relationship with the Old World distribution. Owing to the Mexican and Central American break in the distribution, the South American occurrences might be treated as independent. However, the associations again indicate historical connection with the rest of the world. The Cayapa of Ecuador link dream wandering with soul loss, while the Peruvians consider children especially susceptible and also associate soul loss with sudden fright. In Venezuela, Guiana, and Brazil, many associations occur. The Bakairi connect dream wander- ing and soul loss. Most of the Carib share the dream wandering idea, also identify the soul with the heart, and regard the head as the port of exit. The Guinau and the Taulipang combine sneezing with soul loss as well as dream wandering. Soul loss with sneezing is found generally over Guiana together with dream wandering, and this holds also for the Yekuana. The Tukano identify the heart with the soul, and the Paressi conceive dreams as the experiences of the wandering soul. Farther south, the Lengua associate dream wandering with soul loss and this association is probably general over the Chaco. The soul- loss concept of disease does not occur south of the Chaco until Tierra del Fuego is reached, but none of the above associations have been dis- covered among the Fuegians. From Koppers' account it seems prob- able that the captured soul is returned to the body through the back of the neck.43 In the light of these data it seems almost certain that all cases of soul loss are historically connected, the evidence being as conclusive as indirect methods permit. Accepting a common origin, the question of age remains. Accord- ing to the principle that widely diffused traits having a common origin are quite old, soul loss must trace back to a hoary antiquity. Its presence at the extreme limits of the earth argues for a universal distribution, notwithstanding present discontinuities. It has been seen, however, that in spite of such gaps soul loss all over the world is the same. Its occurrence among the Andaman islanders, the- Semang, the Philippine Pygmies, and the Australians likewise implies a remote time of origin. If all instances are historically connected, then its diffusion to these most primitive and isolated peoples must 43112:72. 1932] 237 238 University of California Publications in Am. Arch. and Ethn. [Vol. 32 have taken place prior to their isolation. Again, its presence in Tierra del Fuego indicates it as one of the archaic culture elements brought into the New World by the proto-Americans who first migrated across Behring strait. Since these first migrations probably occurred at the end of the Pleistocene, soul loss must have been diffused as far as northeastern Asia by this time. Now, the marginal peoples mentioned above have a culture relatively uninfluenced since the end of the Palaeolithic, at any rate, one strikingly similar to the late Palaeo- lithic and Azilian culture of western Europe, with very few later elements. Soul loss thus probably belongs to the same horizon, hence traces back at least to the beginning of the Recent geological period. However, a still greater antiquity may reasonably be urged. The concept evidently had a considerable range in the Old World prior to the American migrations and the isolation of our marginal cultures. Such diffusion must have taken considerable time, even if Siberia be the point of origin. Therefore the concept seems to have actually originated in the late Pleistocene, probably sometime during the Upper Palaeolithic. Color is given to this remote origin by fairly strong evidence that Palaeolithic man had some notion of the soul. Burials dating from the Mousterian period indicate funerary observances of a sort and Upper Palaeolithic burials show evidences of rather elaborate cere- monies. Such rites imply a belief in something which survives after death. Thus, if a soul concept did originate sometime in the Palaeo- lithic it is reasonable to assume, in the light of the other evidence for extreme antiquity, that the soul-loss concept of sickness developed shortly afterward. As this concept seems definitely to belong to a very old horizon, it may have formed part of a complex which diffused as a whole. The principle of the diffusion of large blocks of culture has been utilized in Germany by Graebner and Schmidt, who term such com- plexes "Kulturkreise." Their school explains all similarities by the diffusion of these "Kulturkreise" and has evolved a relative chron- ology, different complexes being held to have diffused at different times and to have been preserved in various parts of the world. Criticism of this pan-diffusionistic theory has no place here, but its chronology may be briefly examined with reference to our theme. Schmidt and Graebner identify the Tasmanians with the archaic stratum of Oceania, which in all other parts of this area has been buried beneath other layers. Cle]ento: Primitive Concepts of Di8ewe The next oldest "Kulturkreis," typified by certain present day Australians, is called the " Old Australian." Elements of this stratum are found by Schmidt in parts of Oceania, in Tierra del Fuego, and generally among Pygmy groups. He likewise considers the Arctic culture of North America as a geographical adaptation of the "Old Australian" type. In South America the natives of the Argentine Chaco, the Ges, the Puri, the Coroados of Brazil, and the early Araucanians are grouped with the Fuegians, all these cultures belonging to the general "Urkulturkreis." Schmidt's other "Kulturkreise" are all later, apply to peoples of more complex cultures, and are not pertinent to the discussion. According to table 1, soul loss occurs in several of the regions assigned to the "Urkulturkreis." Thus, in the Old World the south- ern Australians, the Andamanese, the Semang, the Philippine Negritos, and possibly the Tasmanians share the concept, while in America it occurs among the Eskimos, the Lengua of the Argentine Chaco, and the Fuegians. We might therefore reasonably include the concept among the elements of the "Urkulturkreis," were it not for the fact that it occurs among much higher cultures. Thus, ancient Persia, West Africa, Tibet, southeast India, the North Pacific Coast, and Peru, to name only a few, are all on a far higher plane. Naturally, these cultures are not comparable as to complexity, but they all share the soul-loss concept. Of course, soul loss may conceivably be unusually tenacious, surviving among all these peoples from a primeval culture period. This supposition, however, seems far-fetched unless other elements of the "Urkulturkreis" should also have survived among these peoples, which does not seem to have been the case. While soul loss undoubtedly belongs to a very early culture horizon and actually has survived among several of the higher peoples, it does not, then, fit in very well with the German scheme. Neverthless, its coincidence with the marginal regions of primeval culture is remarkable. As to its point of origin, very little can be said. The concept is most highly elaborated in Siberia, which may indicate Siberia as the center of origin. The presence here of so many of the associated ideas, particularly among the Buriat, adds weight to this conjecture. How- ever, the task of determining the ultimate source of such an old and widely distributed trait is almost hopeless. There have been complex movements of culture since the Palaeolithic, with many transient 1932] 239 240 UniTersity of CaUfornia Publicationts im Am. Aroh. and Ethn. [Vol. 32 centers of development. The archaic culture has atrophied in many places and been heavily overlaid in others. If the soul loss concept really belongs to this early horizon, its origin is buried in the obscurity of tremendous age. RELATIVE ANTIQUITY It remains to synthesize briefly the preceding conclusions from the standpoint of relative chronology. Breach of taboo as a cause of sickness was considered independent in several areas, which are widely separated and relatively circum- scribed, particularly in America. Hence breach of taboo must be recent in comparison with other disease concepts. Its development in Middle America can hardly be much older than the beginning of organized religion there, while its Arctic occurrence is likewise recent. Its wider distribution here, however, implies an origin earlier than the Middle American, although this is no more than conjecture. If the Old World instances outside Siberia all belong to one area, its larger extent would indicate a- still greater age for the concept here. Its elaboration in Polynesia appears to be considerably more recent than its actual time of origin. As to sorcery, only very general statements may be made. Uni- versal culture traits held to be historically connected are generally traced back to at least the end of the Pleistocene. Objects interpreted as "rods of command" are found in European deposits dating from the Magdalenian culture period, and if this interpretation is correct, ideas of magic must have been in existence during Magdalenian times. As these represent the last period of the European Palaeolithic, the close of which coincided with that of the geological Pleistocene, the idea of human control over nature by magical means may be a Mag- dalenian invention. The generic idea of sorcery may then be extremely old. If this idea has diffused all over the world from a single origin, the connection of its various forms with sickness may be adaptations of the idea to various coexisting theories of disease. What is needed, however, is a study of the distribution of the various forms of sorcery. These may prove to be unequal in age, i.e., certain forms may be worldwide, while others appear to be derived from them. Spirit intrusion is concluded to be definitely later than the disease- object theory of sickness, and its time of origin is tentatively placed between the end of the Pleistocene and the beginning of the Full Neolithic in western Asia. Ctements: Primitive Conwepts of Disease Both disease-object intrusion and soul loss have been assigned to the Palaeolithic. Differentiation in the time between two traits as old as this is extremely difficult, and except for actual artifacts can never be much more than mere conjecture. The limits of the distribution of these concepts are extremely wide, soul loss occurring among all the marginal peoples upon which data are available while disease- object intrusion does not. True, it is found in Tierra del Fuego and generally over Australia, but it does not occur among the Andamanese or the Semang. The Philippine Pygmies. however, possess the con- cept and its presence is possible in Tasmania, where scarification and bleeding was practiced to "let the pain out."44 Disease-object intrusion seems to have been crowded out by soul loss in Siberia. This would imply a greater age for the disease-object concept were it not for the fact that soul loss itself seems to have been displaced by object intrusion in most of the United States. Obviously, no conclusion can be reached from the distribution data, and we must fall back on another type of argument. The idea of the soul upon which the soul-loss concept rests is after all of a more subtle sort than the disease-object theory. This latter is both simple and easily derived from and supported by common experi- ence. Moreover, it demands only the simplest sort of cause and effect reasoning. On the other hand, the idea of the soul, even if originating in the way Tylor assumes, certainly demands more complex mental processes than does the idea of the disease object. It thus seems a legitimate conjecture that while both concepts are Palaeolithic, disease-object intrusion is considerably the older idea. The final conclusions as to the relative antiquity of the various dis- ease theories are indicated in the following arrangement of the con- cepts in a chronological series. The list reads down from the latest development. Breach of Taboo (Mexico) Breach of Taboo (Arctic) Breach of Taboo (South Asia and Oceania) Spirit Intrusion Soul Loss Disease-Object Intrusion 44 162:64. 1932] 241 242 University of California Publioations in Am. Aroh. and Ethn. [Vol. 32 SUMMARY OF CONCLUSIONS Breach of Taboo This concept is the latest of those considered and has probably originated independently in Middle America, in the Arctic region, and in southern Asia. It is associated with treatment by confession in the regions of greatest elaboration. This association seems at least partly causal, elaboration of the disease concept spontaneously giving rise to the confession treatment. Spirit Intrusion All occurrences are historically connected. Spirit intrusion seems to have originated considerably later than either disease-object intru- sion or soul loss but had evidently undergone considerable diffusion before the Full Neolithic of western Asia. It is generally connected with exorcism and demon transference. Transference to animals is confined to the Old World and probably became associated with spirit intrusion after animal domestication in the Full Neolithic. Exorcism occurs widely together with other methods of extracting demons but becomes emphasized under the influence of organized religion. This in turn tends to elaborate and strengthen the spirit- intrusion concept of disease. True possession is a differentiation of the general spirit-intrusion idea and occurs widely in the Old World, although absent among the more primitive peoples. Its Arctic and Northwest Coast occurrences are continuous with the Old World dis- tribution. Because of the universality of the psychological bases for inspirational seizures, possession may have originated spontaneously many times. Conversely, its fairly continuous spread together with similar associations seems to give greater weight to its distribution by diffusion. Soul Loss This concept as found over the earth traces to a single Old World origin, possibly in Siberia. It was probably derived from a pre- existing belief in the soul but became associated with other ideas which bear a functional relation to it. Its time of origin seems to have been the upper Palaeolithic, but somewhat later than disease-object intru- sion. Stability of the generic belief in a soul probably accounts for the persistence of the soul-loss-disease concept in higher cultures. 1932] Clements: Primitive Conoepts of D'isease 243 Disease-Object Intrusion All occurrences are historica,lly connected, probably tracing back to an Old World Palaeolithic origin. Distribution was probably once continuous or nearly so, but in certain areas later disease concepts overlaid and submerged it. Elsewhere possibly contemporary theories were elaborated at its expense, resulting in its gradual atrophy. The concept itself assumes its greatest importance in western North America and in Australia. It is probably the earliest of all theories of disease. 244 University of Cca2fornia Publications in Amn Arch. and Ethn. [Vol. 32 BIBLIOGRAPHY 1. ABD ALLAH, M. M. 1917. Siwan customs. Harva,rd African Studies, vol. 1. Cambridge. 2. ADAm, J. 1848. History of the American Indians, in Kingsborough's Antiquities, vol. 8. London. 3. ANDREE, R. 1878. Ethnographische Parallelen. Stuttgart. 4. ANGuo, JAIME DR. 1926. The background of the religious feeling in a primitive tribe. AA 28, no. 2. 5. BARRETT, S. A. The Cayapa Indians of Eucadlor. MAIH-F-INM, 40, no. 2. 6. BARITELS, M. 1893. Die Medicin der Naturv8lker. Leipzig. 7. BATCHELOR, J. 1901. The Ainu and their folklore. London. 8. BEECH, M. H. 1911. The Suk. Oxford. 9. BEST, E. 1901. Maori magic. New Zealand Institute Transactions, 34. 10. BIBLE, THm Mark, 5:8-14. 11. BILBY, J. 1923. Among unknown Eskimo. London. 12. BOAS, F. 1916. Tsimshian mythology. BAE-R 31. 13. 1898. Mythology of the Bella Coola. AMNH-M 2. 14. 1894. Chinook texts. BAE-B 20. 15. 1888. The central Eskimo. BAE-R 6. 16. 1901. The Eskimo. AMNH-B 15. 17. 1893. The Chinook Indians-Doctrines of souls and disease. JAFL 6. 18. 1923. Notes on the Tillamook. UO-PAAE 20. 19. 1920-21. Der Seelenglaube der Vandau. ZE 1. 20. 1925. America and the Old World. ICA 21. Goteborg. 21. BoGoRAs, W. 1913. The Eskimo of Siberia. AMNH-M 12, no. 3. 22. 1910. Chuckehee texts. AMNH-M 12, no. 1. 23. The Chuckehee. AMN1H-M 7. 24. BOLINDER, GuSTAF. 1925. Die Indianer der tropischen Schneegebirge. Stuttgart. 25. BOURKE, J. G. 1892. Medicine men of the Apache. BAE-R 9. 26. BROWN, A. R. , 1922. The Andaman islanders. Cambridge. 27. BROWN, J. M. 1927. Peoples and problems of the Pacific, vol. 1. London. Clements: Prkimitive Concepts of Disewse 28. BUSHNELL, D. I. The Choctaw of Bayou Laeomb. BAE-B 48. 29. CALLAWAY, C. 1884. The religion of the Amazulu. FL 15. 30. CHAMBERLAIN, A. F. 1905. The Kootenay. Archaeological report of the Toronto Museum. 31. CLARDGiE, G. C. 1922. Wild bush tribes of tropical Africa. London. 32. CLARRE, S. R. 1911. Among the tribes in southwest China. London. 33. CODRINwON, R. H. 1891. The Melanesians. Oxford. 34. CONVERSE, H. M. 1908. Myths and legends of the New York State Iroquois. N. Y. State Mus. Bull. 125. 35. COOPER, J. M. 1917. The Tribes of Tierra del Fuego. BAE-B 63. 36. CRANZ, D. 1820. History of Greenland, vol. 1. London. 37. CaOOKE, W. 1896. The popular religion and folklore of northern India, vol. 1. Westminster. 38. Ibid., vol. 2. 39. 1919. NuditJ in India in custom and burial. JRAI 49. 40. CzAPLIcKA, M. A. 1914. Aboriginal Siberia. Oxford. 41. DALE, G. 1896. The Bondei. JRAI 25. 42. DEPONs, F. 1806. A voyage to the eastern part of Terra Firma. New York. 43. DOBRIZHOFFER, M. 1882. An account of the Abipones, vol. 2. London. 44. D ORBIGNY, ALCIDE. 1839. Voyages dans 1'Amerique meridionale, vol. 2. Paris. 45. DORsEY, J. 0. 1894. Siouan cults. BAE-R 11. 46. DIJBois, C. G. 1908. The religion of the Luisefno Indians of southern California. UO-PAAE 8, no. 3. 47. DUMAREST, NoEL. 1919. Notes on Cochiti, New Mexico. AAA-M 6, pt. 3. 48. ELLIS, A. B. 1890. The Ewe-speaking peoples of the Slave Coast of West Africa. London. 49. 1887. The Tahi-speaking peoples of the Gold Coast of West Africa. London. 50. ELLIS, W. 1831. Polynesian researches. London. 51. EMmoNS, G. T. 1911. The Tahltan Indians. UPM-AP 4, no. 1. 52. ENTHOVEN, R. E. 1924. Bombay folklore. Oxford. 1932] 245 246 University of Califorvia PublicatiWons in. Am. Arch. and Ethn. [Vol. 32 53. EvANs, I. H. N. 1918. Some Sakai beliefs and customs. JRAI 48. 54. 1922. Among primitive peoples in Borneo. London. 55. ERDLAND, P. A. 1914. Die Marshall-Insulaner. A 2, no. 1. 56. FARABEE, W. C. 1924. The central Carib. UPM-AP 10. 57. FLETCHER, A. C. 1911. The Omaha. BAE-R 27. 58. FORBES, H. 0. 1884. On some of the tribes of the island of Timur. JRAI 8. 59. FRACHTENBERG, LEo. 1920. The esehatology of the Quileute Indians. AA 22, no. 4. 60. FRAZER, SIR JAMES. 1922. The golden bough. New York. One vol. ed. 61. FREELAND, L. S. 1923. Pomo doctors and poisoners. UC-PAAE 20. 62. GALLARDO, CARLOS R 1910. Los Onas. Buenos Aires. 63. GANN, T. W. 1918. The Maya Indians of Yucatan and British Honduras. BAE-B 64. 64. GARVAN, J. M. 1927. A study of the material and sociological culture of the Manobo of eastern Mindanao. AA 29, no. 4. 65. 1928. Unpublished manuscript. 66. GAYTON, A. H. 1928. Unpublished manuscript on Yokuts Indians. 67. GEFsTE, A. 1910. La medecine et la botanique des anciens mexicains. Imprimerie Polyglotte Vaticane. Rome. 68. GODDARD, P. E. 1903. Life and culture of the Hupa. UC-PAAE 1, no. 1. 69. GOLDIE, W. H. 1904. Maori medical lore. New Zealand Institute Transactions 37. 70. GOMES, E. H. 1911. Seventeen years among the Sea Dyaks of Borneo. Philadelphia.. 71. GowER, C. D. 1927. The northern and southern affiliations of Antillean culture. AAA-M 35. 72. GRuIBE, B. W. 1911. An unknown people in an unknown land. London. 73. GUNTHER, ERNA. 1927. Klallam ethnography. UW-PA 1, no. 5. 74. 1925. Handbonk to the ethnographical collections of the British Museum. Oxford University Press. 75. HANDY, E. S. C. 1927. Polynesian religion. Bishop Museum Bulletin 34. Honolulu. 76. HAwuNGwTON, M. R. 1924. Sacred bundles of the Sauk and Fox. UPM-AP 4, no. 2. 77. HASTINGS, J. 1916. Encyclopedia of religion and ethics, vo. 4. New York. 78. Ibid., vol. 5. Clements: Primitive Conoepts of Disease 79. Ibid., vol. 7. 80. Ibid., vol. 8. 81. Ibid., vol. 9. 82. Ibid., Vol. 11. 83. HAWKES, E. W. The Labrador Eskimo. Mem. 91, Canadian Geological Survey, Anthr. ser. 14. 84. HAWTRRY, S. H. C. 1901. The Lengua Indians of the Paraguayan Chaco. JRAI 31. 85. HOBLEY, C. W. 1910. Ethnology of the A-Kamba and other East African tribes. Cam- bridge. 86. HODSON, T. C. 1911. The Naga tribes of Manipur. London. 87. HonRNN:, A. W. 1918. Rites of transition and the conception of !Nau among the Hot- tentot. Harvard African Studies, vol. 2. Cambridge. 88. HOFFMAN, W. J. 1891. The Midewiwin of the Ojibway. BAE-R 7. 89. HoLLIS, A. C. 1905. The Masai. Oxford. 90. 1909. The Nandi. Oxford. 91. HOLM, G. W. 1914. The Ammassalik Eskimo. Meddelser om Gronland 39. Copen- hagen. 92. Hoopan, L. 1920. The Cahuilla Indians. UC-PAAE 16, no. 6. 93. HOSE, CHARLES and McDO-UGALL, W. 1912. The pagan tribes of Borneo, vol. 2. London. 94. HOVORKA, 0. V., and KRONFELD, A. 1908. Vergleichende Volksmedizin, vol. 1. Stuttgart. 95. Ibid., vol. 2. 96. HOWITT, A. W. 1904. The native tribes of southeast Australia. London. 97. HUTCHINSON, W. 1913. Customs of the world, vol. 1. London. 98. HUTTON, J. H. 1921. The Angami Nagas. London. 99. Im THuRN, P. 1882. Among the Indians of Guiana. London. 100. IvErs, W. G. 1927. Melanesians of the southeast Solomon Islands. London. 101. JENNESS, D. 1913-18. The Copper Eskimo. Report Canadian Arctic Exped., 12. 102. JENNINGS, W. and ADDISON, C. 1905. With the Abyssinians in Somaliland. London. 103. JOCHELSON, W. The Koryak. 104. 1924. The Yukaghir and the Ukaghirized Tungus. AMNH-M 13, no. 2. 105. JUNOD, H. A. 1913. The life of a South African tribe, vol. 1. London. 106. Ibid., vol. 2. 247 1932] 248 University of California Publioations in Am.. Arch. and Ethn. [Vol. 32 107. K.ARSTEN, R. 1926. The civilization of the South American Indians. New York. 107A. KENTON, E. (Editor). 1927. The Indians of North America, vol. 1. (Extracts from the Jesuit Relations and Allied Volumeg.) New York. 108. KLENTITSCHEN, P. A. 1906. Die Kiistenbewohner der Gazellehalbinsel. Miinster. 109. KLFLEWEG DE ZWAAN, J. P. 1913. Die Insel Nias bei Sumatra Die Heilkunde der Nias, vol. 1. Haag. 110. KLOSS, C. BODEN. 1903. In the Andamans and Nicobars. London. 111. KoCH-GRUNBGo, THwiDoR. 1923. Vom Roroima zum Orinoco, vol. 3. Stuttgart. 112. KoPPERs, W. 1924. Unter Feuerland-Indianern. Stuttgart. 113. KRAUSE, A. 1885. Die Tlinkit Indianer. Jena. 144. KRoEBaa, A. L. 1923. Anthropology. New York. 115. 1917. Elements of culture in native California. UC-PAAE 13, no. 8. 116. 1908. Ethnology of the Gros Ventre. AMNH-AP 1, no. 3. 117. 1919. People of the Philippines. New York. 118. 1902. The Arapaho. AMNH-B 18. 119. KnuRJT, ALB. C. 1906. Het Anisme in den Indischen Archipel. Gravenhiige. 119A. KRIUIJT, A. C. and ADRIAI, N. 1912. De Bare's sprekende Torodjas von Celebes, vol. 1. Batavia. 120. LATCHAM, R. E. 1909. Ethnology of the Araucanos. JRAI 39. 121. LEWIs, A. B. 1906. Tribes of the Columbia River. AAA-M 1, pt. 2. 122. LoEB, E. M. 1926. Pomo Folkways. UC-PAAE 19, no. 2. 123. 1924. The shaman of Niue. AA 26, no. 3. 124. 1928. Mentawei social organization. AA 30:408. 125. 1929. Shaman and Seer. AA 31, no. 1. 126. LowiE, R. H. 1925. On the historical connection between certain Old World and New World beliefs. ICA 21, pt. 2. Goteborg. 127. 1922. The religion of the Crow Indians. AMNE-AP 25, no. 2. 128. 1924. Notes on Shoshonean ethnography. AMNH-AP 20, no. 3. 129. LUMHOLTZ, C. 1900. Symbolism of the Huichol Indians. AMNH-M 3, no. 1. 130. MALINOWSKI, B. 1922. Argonauts of the Western Pacifie. London. 131. MASON, J. A. 1912. The etlnology of the SaJinan Indians. UC-PAAE 10, no. 4. 132. 1918. Tepecano prayers. IJAL 1, no. 2. 133. 1922. The chief singer of the Tepecano, in E. C. Parsons (ed.), Ameri- can Indian Life. New York. Clements: Primitive Concepts of Di.se 2e 134. MATTHEWS, W. 1902. The night chant-A Navaho ceremony. AMNH-M 6. 135. MCGOVERN, J. 1922. Among the Head Hunters of Formosa. London. 136. MCKENZIE, D. 1927. The infancy of medicine. London. 137. MERKER, M. 1904. Die Masai. Berlin. 138. MOONEY, J. 1891. Sacred formulae of the Cherokee. BAE-R 7. 139. MORGAN, L. H. 1922. The league of the Hode-no-sau-nee. New York. 140. MoRicE, FATHER. 1905. The Canadian Dene. Annual Arch. Report .Toronto Provincial Museum. 141. MURDOCK, J. 1892. The Alaska Eskimo. BAE-R 9. 142. NELSON, E. W. 1899. The Eskimo about Behring Strait. BAE-R 18. 143. NEUHAUS, R. 1911. Deutsch Neu Guinea. Berlin. 144. NICHOLAS, F. C. 1901. The aborigines of Santa Maria, Columbia. AA 3. 145. NIEUWENHUIS, A. W. 1907. Quer durch Borneo. Leiden. 145A. NIORADZE, G. 1925. Der Schamanismus bei den Sibirischen Volkern. Stuttgart. 146. NORDENSKI6LD, E. 1912. Indianerleben. Leipzig. 147. 1927. The Choco Indians of Colombia and Panama. Discovery 8:95, London. 148. OLSON, RONALD. Life and culture of the Quinault Indians. MS to be published in UW-PA. 149. OVIEDO Y VALDEZ, GONzALES FERNANDEZ. 1535. Historia general y natural de las Indias. Real academia de la historia, vol. 4. Madrid. 150. PANIKAR, K. M. 1918. Some aspects of Nayar life. JRAI 48. 151. PARKER, K. L. 1905. The Euahlayi tribe. London. 152. PARKMAN, P. S. 1908. The culture of the Luisefno Indians. UC-PAAE 8, no. 3. 153. PARSONS, E. C. 1917. Notes on Zufii. AAA-M 4. 154. PLAYFAR, A. 1909. The Garos. London. 155. RrvERs, W. H. R. 1906. The Todas. London. 156. 1924. Medicine, magic and religion. London. 157. RODD, F. R. 1926. People of the Veil. London. 1932] 249 250 University of CalUfomrnia Publioations in Am. Aroh. and Ethn. [Vol. 32 158. RoscoE, J. 1909. Notes on the Bageshu. JRAI 39. 159. 1911. The Baganda. London. 160. 1915. The Northern Bantu. Cambridge. 161. ROTH, L. 1896. Natives of Sarawak and British North Borneo, vol. 1. London. 162. 1899. The aborigines of Tasmania. Halifax (England). 163. ROTH, W. E. 1915. The animism and folklore of the Guiana Indians. BAE-R 30. 164. ROUTLEDGE, K. 1919. The mystery of Easter Island. London. 165. ROUTLGE, W. S. 1910. With a prehistoric people. London. 166. RussELL, F. The Pima Indians. BAE-R 26. 167. SAHAGUN, BERNARDINO DE. 1829. Historia general de las cosas de nueva espafia, vol. 1, bk. 1. Mexico City. 167A. Ibid., vol. 3, bk. 10. 168. SANDSCREJEW, GARMA. 1927. Weltanschauung und Schamanismus der Alaren-Buryaten. A 22. 169. SAPIR, E. The Vaneouver Island Indians, in Hastings' Encyclopedia of Religion and Ethics, vol. 12. 170. SAPPER, KARL. 1926. Das Heilwesen bei den Indianern Guatemalas. Mitteilungen der deutschen Amerikanistenvereinigung. Heft 1, Vortriige ge- halten auf der deutschen Anthropologtagung, Halle A. S. 1925. Wiirzburg. 171. SAVAGE-LANDoR, H. 1913. Across unknown South America, vol. 1. New York. 172. SCHEBESTA, P. 1927. Bei den Urwaldzwergen von Malaya. Leipzig. 173. ScHRoEDan, E. E. W. 1917. Gs. Nias. Leiden. 174. SELIGMANN, C. G. 1910. The Melanesians of British New Guiana. Cambridge. 175. SELIGMANN, C. G. and B. Z. 1911. The Veddas. Cambridge. 176. SIBREE, J. 1880. The great African island. London. 177. SIMoN, PEDRo. 1882. Noticias historiales de las conquistas de tierra firme en las indias occidentales. Edicion hecha sobre de la cuenca de 1626, pt. 1. Bogota. 178. Ibid., pt. 3. 179. SKEAT, W. W. 1900. Malay magic. London. 180. SKiNNEa, A. 1920. Notes on the Bribri of Costa Rica. MAIHF-INM 6, no. 3. 181. 1915. Menomini Indians. AMNH-AP 13. 1932] Clements: Priimitive Concepts of Disease 251 182. 1912-16. Political and ceremonial organization of the Plains-Ojibway. AMNH-AP 2. 183. 1911. The Eastern Cree and Northern Saulteaux. AMNH-AP 9, no. 1. 184. SKINNER, JOSEPH. 1805. The present state of Peru. London. 185. SPECK, F. G. 1907. Outlines of aboriginal culture in the southeastern United States. AA 9, no. 2. 186. 1907. Creek Indians of Taskagi Town. AAA-M 2, no. 2. 187. 1919. Penobscot shamanism. AAA-M 6. 188. 1909. Notes on the Mohegan and Niantic Indians. AMNH-AP 3. 189. 1909. Ethnology of the Yuchi Indians. UPM-AP 1. 190. SPENCE, LEWIS. 1913. The myths of Peru and Mexico. London. 191. SPENCER, B. and GLLAN, F. T. 1904. The northern tribes of central Australia. London. 192. SPIER, LESLiE. 1923. Southern Dieguenlo customs. UC-PAAE 20. 193. SPINDEN, H. J. 1908. The Nez Perce Indians. AAA-M 2, no. 3. 194. STADE, HANS. 1874. The captivity of Hans Stade. Hakluyt Society, vol. 51. London. 195. STARR, F. 1902. Notes upon the ethnography of southern Mexico. Proceedings of the Davenport Academy of Sciences, 9. Davenport, Iowa. 196. STEFANSSON, V. 1914. The Stefansson-Anderson Arctic expedition of the American Museum. AMNE-AP 14, no. 1. 197. STERNBEEG, L. 1925. Conversation with R. H. Lowie. 198. 1925. Divine election in primitive religion. ICA 21, pt. 2. Goteborg. 199. STEVENSON, M. C. The Sia. BAE-R 11. 200. 1915. Ethnobotany of the Zunli Indians. BAE-R 30. 201. STEWARD, J. H. 1928. Unpublished notes on the Eastern Mono and Northern Paiute. 202. SWANTON, J. R. 1909. Ethnology of the Haida. AMNH-M 8, no. 1. 203. 1908. Social conditions, beliefs and linguistic relationship of the Tlingit Indians. BAE-R 26. 204. 1918. An early account of the Choctaw Indians. AAA-M 5. 205. 1911. The Natchez. BAE-B 43. 206. TALBOT, P. M. 1912. In the shadow of the bush. London. 207. TEIT, J. 1900. The Thompson Indians of British Columbia. AMNH-M 2, no. 4. 208. 1909. The Shuswap. AMNH-M 4, no. 7. 209. THOMAS, N. W. 1906. Natives of Australia. London. 210. THOMPSON, J. ERIC. 1927. The civilization of the Mayas. Field Mus. Leaflets in Anthro. 25. 252 University of Califoria Publications in Am. Arch. and Ethn. [Vol. 32 211. THORPE, B. 1851. Northern mythology, vol. 2. London. 212. TORDAY, E. 1925. On the trail of the Bushongo. Philadelphia. 213. ToRi4io MEJIA XESSPE, M. 1923. Algunas costumbres y creencias de los indigenas. Folklore Peru- ano, Lima. 214. TREGmAR, E. 1904. The Maori race, Wanganoi, New Zealand. 215. TUii, JOHANN. 1912. Das Buch des Lappen. Frankfort. 216. TuRNER, G. 1884. Samoa. London. 217. TuRNER, L. M. 1894. The Hudson Bay Eskimo. BAE-R 11. 218. TYLOR, E. B. 1870. Early history of mankind. London. 219. 1920. Primitive culture, vol. 1. London. 220. 1898. Anthropology. New York. 221. VON DEN STEINEN, KARL. 1894. Unter den Naturvilkern Zentral-Brasilens. Berlin. 222. 1897. Unter den Naturvolkern Zentral-Brasilens. Berlin. 223. WALKER, J. B. 1914. Early Tasmania. Hobart. 224. WALKER, J. R. 1917. The sun dance and other ceremonies of the Dakota. AMNH-AP 16, no. 2. 225. WARNECK, IL. J. 1909. Die Religion der Batak. Gbttingen. 226. WEEKS, J. H. 1914. Among the primitive Bakongo. London. 227. WESTER1MANN, DIEDRICH. 1912. The Shilluk people. Philadelphia. 228. 1921. Die Kpelle. Gottingen. 229. XIMENEZ, FRANCISCO. 1926. Las historias del origen de los Indios. San Salvador.