Sacred Texts  Pacific  Index  Previous 



1 This article contains part of the results of ethnographical work in British New Guinea carried on in connection with the Robert Mond Travelling Studentship (University of London), and the Constance Hutchinson Scholarship of the London School of Economics (University of London), with assistance from the Commonwealth Department of External Affairs, Melbourne.

The writer spent some ten months, May, 1915-March, 1916, at Omarakana and the neighboring villages of Kiriwina (Trobriand Islands), where he lived among the natives in a tent. By October, 1915, he had acquired sufficient knowledge of the Kiriwinian language to be able to dispense with the services of an interpreter.

The writer desires to acknowledge the assistance he has received from Mr. Atlee Hunt, Secretary to the Commonwealth Department of External Affairs, and from Dr. C. G. Seligman, Professor of Ethnology in the University of London. The unfailing kindness and encouragement of Dr. Seligman have been of the greatest assistance throughout,

p. 256

and his work, The Melanesians of British New Guinea, provided a solid foundation on which to base the present investigations. Sir Baldwin Spencer, K.C.M.G., has been kind enough to read parts of the MS. and to give the writer his valuable advice on several important points.

2 For an account of Kiriwina sociology, cf. Seligman's work, The Melanesians of British New Guinea, chaps. xlix-lii, pp. 660-707, and chap. lix for a description of the mortuary practices. Prof. Seligman gives also an outline of the native beliefs concerning an afterlife (chap. lv), and his data, which were collected in a different locality of the district, will be quoted hereafter.

3 Seligman, op. cit., p. 733.

4 Cf. below, where the various versions are discussed. The nature of the baloma and kosi, and the material of which they are built, so to speak--whether shadow or reflection or body--will also be dealt with there. It may suffice here to say that the baloma are certainly considered to retain exactly the likeness of the living individual.

5 I have been struck by the enormous difference in this respect obtaining between the Northern Massim and the Mailu, a tribe on the south coast of New Guinea, which I visited during a six months stay in Papua in 1914-15. The Mailu people are conspicuously afraid of darkness. When, towards the end of my stay, I visited Woodlark Island, the natives there, who belong to the same group as the Kiriwinians (a group called by Seligman the Northern Massim), differed so obviously in that respect from the Mailu that I was struck with this the first evening, which I spent in the village of Dikoias. Cf. "The Natives of Mailu: Preliminary Results of the Robert Mond Research Work in British New Guinea," Trans. Roy. Soc. South Australia, vol. xxxix, 1915.

p. 257

6 Cf. C. G. Seligman, op. cit., chap. xlvii, where similar maleficent women from another district (Southern Massim) are described. I do not dwell here in detail on the beliefs about the mulukuausi, but I am under the impression that the natives are not quite sure whether it is a kind of "sending" or "double" that leaves the body of the witch or whether she goes out herself on her errand in an invisible form. Cf. also "The Natives of Mailu," p. 653, and footnote on p. 648.

7 The preliminary burial, as well as burying in the middle of the village, has recently been suppressed by Government.

8 It must be noted that the grave was in olden days situated right in the middle of the village, and that a close vigil was kept over it, having, among other motives, that of protecting the corpse from these female ghouls. Now that the grave is outside the village the vigil has had to be abandoned, and the mulukuausi can prey on the corpse as they like. There seems to be an association between the mulukuausi and the high trees on which they like to perch, so that the present site of burial , placed as it is right among the high trees of the grove (weika) surrounding each village, is specially odious to the natives.

9 This well is situated not far from the shore, in the raiboag, the elevated, stony, wooded coral ridge, which runs in a ring round almost all the smaller islands of the archipelago and the greater part of the large island Boiowa. All the stones and the well here mentioned are real and can be seen by mortals.

10 This effect of the Gilala water was explained by one of my informants only; the others did not know the object of this ablution, though all affirmed its existence.

11 This is a contradiction of the statement that the

p. 258

[paragraph continues] baloma assemble round the new arrival and help him in wailing. See below, VIII, the remarks about such inherent inconsistencies.

12 The natives strictly distinguish between the vaigu'a (valuable possessions) and gugu'a (the other less valuable ornaments and objects of use). The main objects classified as vaigu'a will be enumerated in this article later on.

13 In practice the corpse is most carefully stripped of all valuables just before burial, and I saw even small shell earrings being extracted from the ear lobes, articles which the natives would not hesitate to sell for half a stick of tobacco (three farthings). On one occasion, when a small boy had been buried in my presence, and a very small and poor belt of kaloma (shell discs) was left on the body by mistake, there was great consternation and a serious discussion whether the body ought to be unearthed.

14 During my stay one young man committed suicide in the lo'u manner in a neighboring village. Though I saw the corpse a few hours after death, and was present at the wailing and burial and all the mortuary ceremonies, it was only after a few months that I learned he had committed suicide, and I never could learn his motive. The Rev. E. S. Johns, the head of the Methodist Mission in the Trobriands, informs me that be used at times to register as many as two suicides a week (through poison) in Kavataria, a group of large villages situated in the immediate neighborhood of the Mission station. Mr. Johns tells me that suicides occur in epidemics, and that they have been fostered by the discovery by the natives of the white man's power to counteract the poison. The aim of the suicide is to punish the survivors, or some of them.

15 This poison is prepared from the roots of a cultivated vine; its action is not very rapid, and if emetics be properly administered in time life is usually saved.

p. 259

16 There seems to be some possibility of death by old age, especially in the case of very insignificant old men and women. Several times, when I was asking of what a man had died, I received the answer, "He was very old and weak and he just died." But when I asked about M'tabalu, a very old and decrepit man, the chief of Kasana'i, whether he was going to die soon, I was told that, if no silami (evil spell) were thrown on him, there was no reason why he should not go on living. Again, it must be remembered that a silami is a private thing, not to be talked about except with intimate friends. It must be emphasized that the "ignorance of natural death" is the general typical attitude expressed in custom and reflected in such legal and moral institutions as exist, rather than some kind of absolute apodictic statement, excluding any contradictions or uncertainties.

17 Seligman, op. cit., p. 733.

18 The distinction between rank and authority is important in Kiriwinian sociology. The members, of the Tabalu section of the Malasi clan have the highest rank, The head of this clan wields authority over the village of Omarakana and, in a way, over a great portion of the main island and some adjacent islands. Whether he will retain this authority after death in Tuma seemed doubtful to To'uluwa, the present chief of Omarakana. But there was not the slightest doubt that he and all the other Tabalu, as well as everyone else, would retain their respective rank and. their membership in clan and subclan. To understand this, cf. the excellent account of the Trobriand social system, in Seligman, op. cit., chaps. xlix-liii.

19 In order to understand this statement the reader must be acquainted with the social system of the Kiriwinians (see Seligman, loc. cit.). There is a very close connection between every village and a certain section of a clan. Usually, but not always, this section is descended from an

p. 260

ancestor, who came out of the ground in that locality. In any case, the head of this section is always said to be the master or owner of the land (tolipuaipucaia, from toli, a prefix denoting mastership, ownership, and puaipuaia, ground, soil, land).

20 This wooing in Tuma, as described to me by my informants, corresponds to the manner in which people mate on certain occasions called katuyausi. The katuyausi are expeditions of amorous adventure, in which the unmarried girls of a village go en bloc to another village and there sleep with the youths of that village. Any single male who fancies one of the girl guests gives her (through an intermediary) some small present (a comb, some shell discs or turtle shell rings), which is handed over with the words "kam paku." If accepted, the two belong to each other for the night. Such expeditions, though well established and sanctioned by custom, are strongly resented by the young men of the village from which the katuyausi starts, and they end as a rule in a sound thrashing administered by the male to the female youth of the village.

21 A "lifetime" is undoubtedly a much less definite period to the natives than it is to ourselves.

22 Another center is the Island of Kaileula.

23 Seligman, op. cit., p. 734.

24 Similar songs have also been brought by other people from Tuma.

25 To judge leniently such "inconsistencies" of native belief, it is sufficient to remember that we meet the same difficulties in our own ideas about ghosts and spirits. No one who believes in ghosts or spirits ever doubts that they can speak, and even act; they can rap on tables or with table legs, lift objects, etc.

p. 261

26 op. cit., p. 733.

27 op. cit., p. 679.

28 "Ceremonial in the narrower sense," as opposed to the mere uttering of the spell over a certain object.

29 For instance, the Mailu on the south coast. See Trans. Roy. Soc. South Australia, vol. xxxix, p. 696.

30 In this short, purely descriptive account of harvesting, I have purposely avoided sociological technicalities. The complex system of mutual gardening duties is an extremely interesting feature of Kiriwinian social economics. It will be described in another place.

31 In this and other instances I do not dwell upon such sociological details as do not bear directly upon the subject of this article.

32 The calendar arrangements in the Trobriand Islands are complicated by the fact that there are four districts, each of which places the beginning of the year, i.e., the end of the milamala moon--at a different time. Thus in Kitava, an island to the east of the main island of the group, the milamala is celebrated some time in June or July. Then follow in July and August the southern and western districts of Bwoiowa, the main island, and some islands to the west (Kaileula and others). After which the milamala takes place in August or September in the central and eastern districts of the main island, in what is called by the natives Kiriwina, and last there follows Vakuta, the island to the south of Bwoiowa, where the milamala takes place in September or October. Thus the feast, and with it the whole calendar, is shifted over the space of four moons in one district. It seems that the dates of the garden activities also vary, keeping time with the calendar. This was stated by the natives with emphasis, but I found during

p. 262

the year I was in Bwoiowa that the gardens were more advanced in Kiriwina than in the western district, though the latter is one moon in advance of the former.

The dates of the moons axe fixed by the position of the stars, in which astronomical art the natives of Wawela, a village lying on the beach in the southern half of the main island, excel. The Rev. M. K. Gilmour told me that the appearance of the palolo, the marine annelid Eunice viridis, which takes place on the reef near Vakuta, is a very important factor in regulating the native calendar, in fact, in doubtful cases it decides the question. This worm appears on certain days towards the full moon, falling early in November or late in October, and this is the milamala time of Vakuta. In Kiriwina the natives told me, however, that they thoroughly rely on the astronomical knowledge of the Wawela men.

33 No tokaikaya were made in Omarakana or Olivilevi during the milamala I saw in those villages. The custom is on the decline, and the erection of a tokaikaya necessitates a considerable amount of labor and trouble. I saw one in the village of Gumilababa, where there lives a chief of the highest rank (Mitakata, a guya'u of the tabalu rank).

34 How far, besides and behind this professed aim, vanity and the aesthetic motive are at work in prompting such displays, cannot be discussed here.

35 This is one of the innumerable food distributions (generic name sagali) which are connected with almost every feature of social life in the Trobriands. It is usually one clan (or two clans) that arranges the sagali, and other clans receive the food. Thus in the katukuala the Malasi clan first distribute the food and the lukulabuta, the lukuasisiga and the lukubia receive it. After a few days another katukuala is held, with the inverse social grouping. The dual arrangements of the clans varies according to

p. 263

the district. In Omarakana the Malasi are so preponderant that they form one moiety for themselves, the three remaining clans forming the other. It is impossible to enter here into the detailed examination of the social mechanism and of the other features of the sagali.

36 Of course, the chiefs have as much pig as they require before giving any to the tokay. But it is characteristic that the privileges of the chief have much more to do with the liberty to give than with the liberty to consume. Vanity is a stronger passion than greed--though perhaps this reflection does not express the whole truth of the matter!

37 Thus dancing in general is inaugurated by the initiation of the drums (katuvivisa kasausa'u), which is connected with the katukuala. The kaidebu have to be initiated separately by a katuvivisa kaidebu.

38 There are names for each day about full moon. Thus the day (and night) of the full moon are called Yapila or Kaitaulo. One day before, Yamkevila; two days, Ulakaiwa. The day after full moon, Valaita; the following one, Woulo. The ioba takes place on the night of Woulo.

39 The drums of the Kiriwinians consist of: (1) the large drum (normal size of New Guinea drum) called kasausa'u or kupi (this latter word being an obscene synonym for the glans penis): and (2) the small drum, about one third the size of the larger, called katunenia. All their drumbeats are a combination of the two drums, the kupi and the katunenia leading each its separate voice.

40 There are two main types of dance in Boiowa. The circular dances, where the orchestra (the drums and the singers) stands in the middle, and the performers go round them in a circle, always in the opposite direction to the hands of a watch. These dances are again subdivided into: (1) bisila (pandanus streamer) dances with slow movement

p. 264

[paragraph continues] (2) kitatuva (two bunches of leaves), with a quick movement; and (3) kaidebu (wooden painted shield), dances with the same quick movement. In the bisila dances women can take part (very exceptionally), and all the performers wear women's petticoats. The second group of dances are the kasawaga, where only three men dance, always in imitation of animal movements, though these are very conventionalized and unrealistic. These dances are not circular, there are no songs (as a rule) to accompany them; the orchestra consists of five kupi drums and one katunenia.

41 When a village is in mourning (bola), and drums are taboo, the ioba is performed by means of a conch shell (ta'uio)--but it must not be omitted even under such circumstances.

42 The dread of "leading questions," as expressed over and over again in all instructions for ethnographical field work is, according to my experience, one of the most misleading prejudices. "Leading questions" are dangerous with a new informant, for the first half hour or couple of hours, at the outside, of your work with him. But any work done with a new, and consequently bewildered, informant is not worth being recorded. The informant must know that you want from him exact and detailed statements of fact. A good informant, after a few days, will contradict and correct you even if you make a lapsus linguae, and to think of any danger from leading questions in such a case is absolutely groundless. Again, real ethnographical work moves much more in statement of actual details, details which, as a rule, can be checked by observation--where again there is in no case any danger from leading questions. The only case where direct questioning is necessary, where it is the only instrument of the ethnographist, is when he wants to know what is the interpretation of a ceremony, what is his informant's opinion about some state of things; then leading questions are absolutely necessary. You might ask a native, "What is your interpretation of such and such a ceremony?"

p. 265

and wait for years before getting your answer (even if you know how to word it in native language). You would more or less solicit the native to take up your attitude, and look at things in ethnographical perspective. Again, when dealing with facts that are just out of range of immediate observation, like customs of war, and some of the obsolete technological objects, it is absolutely impossible to work without leading questions, if many important features are not to be omitted, and as there is no earthly reason to avoid this type of questioning, it is directly erroneous to brand the leading questions. Ethnological inquiry and judicial examination are essentially different, in that in the latter the witness has usually to express his personal, individual opinion, or to relate his impressions, both of which can be easily modified by suggestion: whereas in ethnological inquiry the informant is expected to give such eminently crystallized and solidified items of knowledge as an outline of certain customary activities, or a belief or a statement of traditional opinion. In such cases a leading question is dangerous only when dealing with a lazy, ignorant, or unscrupulous informant--in which case the best thing is to discard him altogether.

43 A characteristic fact to illustrate this statement is furnished by a Scotchman, who has been living for years among the natives as trader and pearl buyer. He has in no way lost the "caste" and dignity of the white man, in fact, he is an extremely kind, hospitable gentleman; nevertheless, be has assumed certain native peculiarities and habits such as the chewing of areca nut, a habit seldom adopted by white men. He is also married to a Kiriwinian. In order to make his garden thrive, he uses the help of a native towosi (garden wizard) from the next village, and that is the reason, my informants told me, that his garden is always considerably better than that of any other white man.

44 The broad generalities given about Kiriwinian garden magic are, of course, not to be taken even as an outline of

p. 266

this magic, which, it is hoped, will be described in another paper.

45 It should be remembered that each village has its own system of garden magic, intimately connected with that village, and transmitted in the maternal line. The membership in a village community is also transmitted in the female line.

46 I cannot deal here in detail with this rule, to which there are many apparent exceptions. This will be done in another place. The statement in the text ought to be amended: "hereditary in the female line in the long run." For instance, very often a father gives the magic to his son, who practices it during his lifetime, but this son cannot pass it on to his son unless he has married a girl of his father's clan, so that his son belongs to the original clan again. Cross-cousin marriage, prompted by this and similar motives, is fairly frequent, and is considered distinctly desirable.

47 Thus, for instance, the kainagola, one of the most powerful silami (evil spells), is associated with a myth localized in the villages of Ba'u and Buoitalu. Again, certain canoe-building magic, called wa'iugo, contains references to a myth, the scene of which is the island of Kitava. Many other examples could be adduced.

48 The native name for a subclan is dala, cf. Seligman, op. cit., p. 678, where the form dalela is dala, with the pronominal suffix of the third person, "his family." The author gives there the names of dala belonging to the four clans. They include the most important dala, but there are many others. As Professor Seligman says, the members of each dala trace their origin to a common ancestor. Such an ancestor emerged originally out of a hole in the earth in a given locality. And, as a rule, the dala lives in, or near, that locality--very often the "hole is in the grove surrounding

p. 267

the village, or even right in the village. These holes, called "houses" (buala), are, at present, either waterholes or stone heaps, or small shallow cavities. The hole mentioned by Professor Seligman on p. 679 is the one out of which emerged several of the most aristocratic dala. But this is an exception, the rule being one buala, one dala.

49 I am almost certain that it is an archaic form, connected with vitu, a prefix expressing causation. Thus, "to show the way," "to explain," vitu loki, is composed of vitu, to "cause," and loki, to "go there." There are a number of such causative prefixes in Kiriwinian, each possessing a different shade of meaning. In this place, of course, they cannot be discussed.

50 This is an example of the above-mentioned exceptions to the matrilineal descent of certain magical formulae. Iowana, father of Bagido'u, was the son of a tabalu (i.e., of one of the family who "own" Omarakana). His father, Puraiasi, gave him the magic, and as Iowana married Kadu Bulami, his cousin, a tabalu woman, he could transmit the magic to his son, Bagido'u, and the office of towosi (garden magician) thus returned to the tabalu subclan.

51 As a matter of fact, this system is imported from another village, Luebila, situated on the northern coast. Hence its name, kailuebila. It contains only one or two references to some places near that village, but it was not known in Omarakana whether those places were sacred or not.

52 Ovavavala is an archaic form of the name Ovavavile.

53 Bom' is abbreviated from boma, which means taboo. Likuliku is an expression for earthquake, which is an important item in the magical vocabulary.

54 Kamkuam, eat; kami, the personal prefix of the second

p. 268

person plural, used with food; nunumuaia, plural of numuaia, old woman. The two personal names of the baloma old women are remarkable for beginning with ili, very likely derived from ilia-fish. Bualita means sea. It seems thus possible that they are some mythical persons, associated with fishing, concerning whom the tradition has been lost. But such guesses have little charm, and still less value, in the opinion of the present writer.

55 The first name is that of a woman; iaegulo means "I"; Iamuana is said to have been the mother of Umnalibu. Here also the name is suggestive of some connection with the spell, which is said over the libu plant. The last name but one, Taigala, means, literally, his ear, but here it was said to stand for a bili baloma name.

56 It must be stated that several of these formulae have not been translated in a satisfactory manner. It was often impossible to secure the help of the man who recited the spell. Several spells were collected during short visits to outlying villages. In several cases the man was too old or too stupid to help in the, from the native point of view, extremely difficult and puzzling task of translating the archaic and condensed formula, and of commenting upon all its obscurities. And, as a rule, it is no use asking anyone but the original owner to translate or comment upon any formula. I have been able, however, from my knowledge of the "colloquial" language to grasp the general meaning of almost all the formulae.

57 The full discussion of this subject must be deferred to another place. It is interesting that in a certain class of silami (evil spells) there is a direct invocation to a being, tokuay (a wood spirit living in trees), to come and perform the evil. And everybody agrees that it is the tokuay who is the u'ula (basis, reason, cause) of the silami, that he enters the body and produces disastrous internal disorders.

p. 269

58 All these general statements must be regarded as preliminary, they will be supported by proper documents in the proper place.

59 Compare those data with the above-discussed "ignorance of natural death." In this ignorance there ought to be distinguished: (1) the ignorance of the necessity of death, of the life coming to an end; (2) ignorance of the natural causes of sickness as we conceive it. Only the second ignorance seems to be quite prevalent, the action of evil sorcerers being always assumed, except, perhaps, in the above-mentioned cases of very old and insignificant folk.

60 Suma is the root for pregnancy; nasusuma, a pregnant woman; isume, she becomes pregnant. There is no term denoting conception, as distinguished from pregnancy. The general meaning of suma is "take," "take possession of."

61 I am using here the expression "spirit child" as a terminus technicus. This is the term used by Spencer and Gillen to denote analogous beings in Australia, where this type of reincarnation was first discovered. How far the Kiriwinian facts are ethnographically or psychologically connected with those described by Spencer and Gillen will not be discussed in this place.

62 This information was obtained from a woman on the west coast. I think the woman belonged to the village of Kavataria. Mr. G. Auerbach, a pearl buyer, who resides in Sinaketa, a coastal village on the southern half of the island, told me that there are some stones there, to which a woman who wants to become enceinte may have recourse. My informant was unable to tell me whether this was ceremonial or not.

63 There is a remarkable rule which compels the woman to perform all sorts of practices in order to have her skin quite light after childbirth; she keeps in the house, she has

p. 270

to wear the saikeulo over her shoulders, she washes with hot water, and frequently puts coconut cream on her skin. The degree of lightness of skin thus achieved is remarkable. The above described ceremony is a kind of magical inauguration of the period when she will have to keep her skin light.

64 A genealogy shows the relationship in an instant--



The black discs represent males, the rings, females.

65 The majority of my informants were equally positive about the rule that a baloma of the veiola must convey the child. But I have come across one or two dissenting opinions, affirming that the father's mother may bring the child. It was said by one man that if the child resembles the mother it has been brought by some of her veiola; if it resembles the father it has been brought by his mother. But this opinion may be my informant's private speculation.

66 In which, nota bene, by the baloma who "gave" the child, the natives mean either the original baloma, who has become the child, or the baloma who brought the waiwaia.

67 The sexual freedom of unmarried girls is complete. They begin intercourse with the other sex very early, at the age of six to eight years. They change their lovers as often as they please, until they feel inclined to marry. Then a girl settles down to a protracted and, more or less, exclusive intrigue with one man, who, after a time, usually becomes her husband. Illegitimate children are by no means rare,

p. 271

cf. the excellent description of sexual life and marriage among the Southern Massim, who, in this respect, resemble the Kiriwinians to a great extent, in Seligman, op. cit., xxxviii, p. 499, and the short but correct account there given of the same subject among the Northern Massim (including the Trobriand Islanders), chap. liii, P. 708.

68 A white settlement in the east end of New Guinea.

69 As I do not want to criticize particular views, so much as to add some data bearing on this problem, I shall not note any statements, especially from those authors whose opinions appear to me to be untenable. The probability of a "non-recognition in early times of the physical relation between father and child" was first suggested by Mr. E. S. Hartland (The Legend of Perseus, 1894-96), and the discoveries of Spencer and Gillen brilliantly confirmed his views. Mr. Hartland has subsequently devoted the most exhaustive inquiry extant to this problem (Primitive Paternity). Sir J. G. Frazer has also given the support of his illustrious opinion to the view that ignorance of physical fatherhood was universal among early mankind (Totemism and Exogamy).

70 The Melanesians of British New Guinea, p. 84.

71 Trans. of Roy. Soc. South Australia, vol. xxxix, 1915, p. 562.

72 I Use Professor Seligman's terminology, based on his classification of the Papuasians, op. cit., pp. 1-8.

73 Cf. Seligman, op. cit., passim; also chap. xlix.

74 My own notes taken among the Mailu, and the conclusion I drew from them, are typical of such a failure. As other instances, may be quoted the denial by Strehlow and von Leonhardi of the discoveries of Spencer and Gillen; a

p. 272

denial which, if the argument of von Leonhardi be carefully read, and the data given by Strehlow examined, turns out to be only a futile controversy based upon inadequate premises, and, in fact, completely confirms the original discoveries of Spencer and Gillen. Here the explanation lies in the insufficient mental training of the observer (Strehlow). You can no more expect good all-round ethnographical work from an untrained observer than you can expect a good geological statement from a miner, or hydrodynamic theory from a diver. It is not enough to have the facts right in front of one, the faculty to deal with them must be there. But lack of training and mental capacity is not the only cause of failure. In the excellent book about the natives of New Guinea (Goodenough Bay on the N.E. coast), written by the Rev. H. Newton, now Bishop of Carpentaria, than whom none could be better equipped to understand the native mind and to grasp native customs, we read the following statement: "There may be races as ignorant [of the causal relation between connection and pregnancy] as is implied [in Spencer and Gillen's statement]; it is difficult to imagine such, when marital infidelity appears to be so severely punished everywhere, and when the responsibility of the father for the child is recognized, if only to a small extent." (In Far New Guinea, p. 194.) Thus, an excellent observer (such as the present Bishop of Carpentaria undoubtedly is), living for years among the natives, knowing their language, has to imagine a state of things which exists fully and completely all round him. And his arguments for denying this state (everywhere, not only among his tribe) is that marital jealousy and recognition of fatherhood both exist (a recognition, which again is not known in the tribe in question, on the physical side)! As if there were the slightest logical nexus between jealousy (a pure instinct) and ideas about conception; or, again, between these latter and the social ties of the family! I have taken this statement for criticism, just because it is found in one of the very best ethnographical books which we have about South Sea natives. But I wish to add that my criticism

p. 273

is in a way unfair, because Mr. Newton, as a missionary, could scarcely discuss with the natives all the details of the question, and also because Mr. Newton gives the reader fully to understand that he has not inquired into the question directly, and candidly states the reasons for his doubts. I have quoted the statement, nevertheless, in order to show the many technical difficulties which are connected with the obtaining of accurate information on this subject, and the many gaps through which errors can leak into our knowledge.

75 My experience in the field has persuaded me of the complete futility of the theories which attribute to the savage a different type of mind and different logical faculties. The native is not "prelogical" in his beliefs, he is alogical, for belief or dogmatic thinking does not obey the law of logic among savages any more than among ourselves.

76 To test this sociological principle on civilized instances; when we say that the "Roman Catholics believe in the infallibility of the Pope," we are correct only in so far as we mean that this is the orthodox belief, enjoined on all members of that church. The Roman Catholic Polish peasant knows as much about this dogma as about the Infinitesimal Calculus. And if it were proposed to study the Christian religion, not as a doctrine, but as a sociological reality (a study which, as far as I am aware, has never yet been attempted), all the remarks in this paragraph would apply, mutatis mutandis, to any civilized community with the same strength as to the "savages' of Kiriwina.

77 1 am purposely not using the term "collective ideas," introduced by Professor Durkheim and his school, to denote a conception, which in their hands, more especially in the writings of Hubert and Mauss, has proved extremely fertile. In the first place, I am not able to judge whether the above analysis would really cover what that school denotes by "collective ideas." Remarkably enough, there does not seem

p. 274

to be anywhere a clear, candid statement of what they mean by "collective idea," nothing approaching a definition. It is obvious that in this discussion, and in general, I am under a great obligation to these writers. But I am afraid that I am entirely out of touch with Professor Durkheim's philosophical basis of sociology. It seems to me that this philosophy involves the metaphysical postulate of a "collective soul," which, for me, is untenable. Moreover, whatever discussion might be carried on as to the theoretical value of the conception of a "collective soul," in all practical sociological investigations one would be left hopelessly in the lurch by it. In the field, when studying a native or civilized community, one has to do with the whole aggregate of individual souls, and the methods and theoretical conceptions have to be framed exclusively with this multiplex material in view. The postulate of a collective consciousness is barren and absolutely useless for an ethnographical observer.